{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 544,
  "global_step": 5438,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000367816091954023,
      "grad_norm": 2.7641916275024414,
      "learning_rate": 0.0,
      "loss": 7.1805,
      "num_input_tokens_seen": 4624,
      "step": 1,
      "train_runtime": 2.7035,
      "train_tokens_per_second": 1710.397
    },
    {
      "epoch": 0.000735632183908046,
      "grad_norm": 2.057875156402588,
      "learning_rate": 4e-05,
      "loss": 5.892,
      "num_input_tokens_seen": 12848,
      "step": 2,
      "train_runtime": 4.3199,
      "train_tokens_per_second": 2974.117
    },
    {
      "epoch": 0.001103448275862069,
      "grad_norm": 2.612283706665039,
      "learning_rate": 8e-05,
      "loss": 6.9995,
      "num_input_tokens_seen": 18240,
      "step": 3,
      "train_runtime": 5.2159,
      "train_tokens_per_second": 3496.988
    },
    {
      "epoch": 0.001471264367816092,
      "grad_norm": 3.118577480316162,
      "learning_rate": 0.00012,
      "loss": 6.7426,
      "num_input_tokens_seen": 23120,
      "step": 4,
      "train_runtime": 6.0224,
      "train_tokens_per_second": 3838.994
    },
    {
      "epoch": 0.0018390804597701149,
      "grad_norm": 2.8236703872680664,
      "learning_rate": 0.00016,
      "loss": 5.1427,
      "num_input_tokens_seen": 31184,
      "step": 5,
      "train_runtime": 7.2549,
      "train_tokens_per_second": 4298.314
    },
    {
      "epoch": 0.002206896551724138,
      "grad_norm": 3.1930270195007324,
      "learning_rate": 0.0002,
      "loss": 4.6294,
      "num_input_tokens_seen": 37328,
      "step": 6,
      "train_runtime": 8.2321,
      "train_tokens_per_second": 4534.427
    },
    {
      "epoch": 0.002574712643678161,
      "grad_norm": 3.055657386779785,
      "learning_rate": 0.00019996318792563962,
      "loss": 4.1374,
      "num_input_tokens_seen": 44592,
      "step": 7,
      "train_runtime": 9.3409,
      "train_tokens_per_second": 4773.86
    },
    {
      "epoch": 0.002942528735632184,
      "grad_norm": 2.088937759399414,
      "learning_rate": 0.00019992637585127923,
      "loss": 3.3767,
      "num_input_tokens_seen": 50960,
      "step": 8,
      "train_runtime": 10.36,
      "train_tokens_per_second": 4918.932
    },
    {
      "epoch": 0.003310344827586207,
      "grad_norm": 2.6767897605895996,
      "learning_rate": 0.00019988956377691884,
      "loss": 3.3144,
      "num_input_tokens_seen": 56304,
      "step": 9,
      "train_runtime": 11.2362,
      "train_tokens_per_second": 5010.951
    },
    {
      "epoch": 0.0036781609195402297,
      "grad_norm": 2.8294262886047363,
      "learning_rate": 0.00019985275170255845,
      "loss": 2.8192,
      "num_input_tokens_seen": 60576,
      "step": 10,
      "train_runtime": 11.961,
      "train_tokens_per_second": 5064.469
    },
    {
      "epoch": 0.004045977011494253,
      "grad_norm": 1.635740876197815,
      "learning_rate": 0.00019981593962819808,
      "loss": 2.564,
      "num_input_tokens_seen": 67552,
      "step": 11,
      "train_runtime": 13.0349,
      "train_tokens_per_second": 5182.388
    },
    {
      "epoch": 0.004413793103448276,
      "grad_norm": 1.3069396018981934,
      "learning_rate": 0.00019977912755383766,
      "loss": 2.4683,
      "num_input_tokens_seen": 75648,
      "step": 12,
      "train_runtime": 14.261,
      "train_tokens_per_second": 5304.541
    },
    {
      "epoch": 0.0047816091954022985,
      "grad_norm": 1.1834617853164673,
      "learning_rate": 0.00019974231547947727,
      "loss": 2.5426,
      "num_input_tokens_seen": 80176,
      "step": 13,
      "train_runtime": 15.0271,
      "train_tokens_per_second": 5335.427
    },
    {
      "epoch": 0.005149425287356322,
      "grad_norm": 0.8042702078819275,
      "learning_rate": 0.00019970550340511688,
      "loss": 2.2478,
      "num_input_tokens_seen": 88976,
      "step": 14,
      "train_runtime": 16.366,
      "train_tokens_per_second": 5436.65
    },
    {
      "epoch": 0.005517241379310344,
      "grad_norm": 0.9168936610221863,
      "learning_rate": 0.0001996686913307565,
      "loss": 2.4822,
      "num_input_tokens_seen": 93824,
      "step": 15,
      "train_runtime": 17.1795,
      "train_tokens_per_second": 5461.393
    },
    {
      "epoch": 0.005885057471264368,
      "grad_norm": 0.8179077506065369,
      "learning_rate": 0.0001996318792563961,
      "loss": 2.2916,
      "num_input_tokens_seen": 100336,
      "step": 16,
      "train_runtime": 18.1928,
      "train_tokens_per_second": 5515.154
    },
    {
      "epoch": 0.006252873563218391,
      "grad_norm": 0.748948872089386,
      "learning_rate": 0.0001995950671820357,
      "loss": 1.9101,
      "num_input_tokens_seen": 105328,
      "step": 17,
      "train_runtime": 18.9957,
      "train_tokens_per_second": 5544.827
    },
    {
      "epoch": 0.006620689655172414,
      "grad_norm": 0.6820124387741089,
      "learning_rate": 0.00019955825510767532,
      "loss": 2.2968,
      "num_input_tokens_seen": 110640,
      "step": 18,
      "train_runtime": 19.8653,
      "train_tokens_per_second": 5569.503
    },
    {
      "epoch": 0.006988505747126437,
      "grad_norm": 0.6636161208152771,
      "learning_rate": 0.00019952144303331495,
      "loss": 2.2185,
      "num_input_tokens_seen": 118448,
      "step": 19,
      "train_runtime": 21.1195,
      "train_tokens_per_second": 5608.458
    },
    {
      "epoch": 0.0073563218390804595,
      "grad_norm": 0.6526411175727844,
      "learning_rate": 0.00019948463095895456,
      "loss": 2.2731,
      "num_input_tokens_seen": 125648,
      "step": 20,
      "train_runtime": 22.2501,
      "train_tokens_per_second": 5647.084
    },
    {
      "epoch": 0.007724137931034483,
      "grad_norm": 0.6273838877677917,
      "learning_rate": 0.00019944781888459417,
      "loss": 1.9624,
      "num_input_tokens_seen": 136128,
      "step": 21,
      "train_runtime": 24.5859,
      "train_tokens_per_second": 5536.825
    },
    {
      "epoch": 0.008091954022988505,
      "grad_norm": 0.7814621925354004,
      "learning_rate": 0.00019941100681023378,
      "loss": 2.2268,
      "num_input_tokens_seen": 141008,
      "step": 22,
      "train_runtime": 25.4363,
      "train_tokens_per_second": 5543.573
    },
    {
      "epoch": 0.00845977011494253,
      "grad_norm": 0.6532659530639648,
      "learning_rate": 0.0001993741947358734,
      "loss": 2.2661,
      "num_input_tokens_seen": 148560,
      "step": 23,
      "train_runtime": 26.615,
      "train_tokens_per_second": 5581.812
    },
    {
      "epoch": 0.008827586206896552,
      "grad_norm": 0.7463673949241638,
      "learning_rate": 0.000199337382661513,
      "loss": 2.0873,
      "num_input_tokens_seen": 153424,
      "step": 24,
      "train_runtime": 27.4236,
      "train_tokens_per_second": 5594.593
    },
    {
      "epoch": 0.009195402298850575,
      "grad_norm": 0.7072988748550415,
      "learning_rate": 0.00019930057058715258,
      "loss": 2.0608,
      "num_input_tokens_seen": 158096,
      "step": 25,
      "train_runtime": 28.234,
      "train_tokens_per_second": 5599.498
    },
    {
      "epoch": 0.009563218390804597,
      "grad_norm": 0.7270534634590149,
      "learning_rate": 0.0001992637585127922,
      "loss": 2.1637,
      "num_input_tokens_seen": 162704,
      "step": 26,
      "train_runtime": 29.018,
      "train_tokens_per_second": 5607.012
    },
    {
      "epoch": 0.009931034482758621,
      "grad_norm": 0.7840702533721924,
      "learning_rate": 0.0001992269464384318,
      "loss": 2.3241,
      "num_input_tokens_seen": 167680,
      "step": 27,
      "train_runtime": 29.8571,
      "train_tokens_per_second": 5616.093
    },
    {
      "epoch": 0.010298850574712644,
      "grad_norm": 0.6798800826072693,
      "learning_rate": 0.00019919013436407143,
      "loss": 2.3167,
      "num_input_tokens_seen": 176560,
      "step": 28,
      "train_runtime": 31.2144,
      "train_tokens_per_second": 5656.365
    },
    {
      "epoch": 0.010666666666666666,
      "grad_norm": 0.7969369888305664,
      "learning_rate": 0.00019915332228971104,
      "loss": 2.2511,
      "num_input_tokens_seen": 182032,
      "step": 29,
      "train_runtime": 32.0914,
      "train_tokens_per_second": 5672.299
    },
    {
      "epoch": 0.011034482758620689,
      "grad_norm": 0.6256918907165527,
      "learning_rate": 0.00019911651021535065,
      "loss": 2.0108,
      "num_input_tokens_seen": 187344,
      "step": 30,
      "train_runtime": 32.9732,
      "train_tokens_per_second": 5681.703
    },
    {
      "epoch": 0.011402298850574713,
      "grad_norm": 0.576696515083313,
      "learning_rate": 0.00019907969814099026,
      "loss": 2.055,
      "num_input_tokens_seen": 196384,
      "step": 31,
      "train_runtime": 34.9768,
      "train_tokens_per_second": 5614.7
    },
    {
      "epoch": 0.011770114942528736,
      "grad_norm": 0.7484147548675537,
      "learning_rate": 0.00019904288606662987,
      "loss": 2.115,
      "num_input_tokens_seen": 204208,
      "step": 32,
      "train_runtime": 36.2152,
      "train_tokens_per_second": 5638.743
    },
    {
      "epoch": 0.012137931034482758,
      "grad_norm": 0.6978005170822144,
      "learning_rate": 0.00019900607399226948,
      "loss": 2.0565,
      "num_input_tokens_seen": 209888,
      "step": 33,
      "train_runtime": 37.113,
      "train_tokens_per_second": 5655.372
    },
    {
      "epoch": 0.012505747126436782,
      "grad_norm": 0.7106545567512512,
      "learning_rate": 0.00019896926191790909,
      "loss": 1.9952,
      "num_input_tokens_seen": 216192,
      "step": 34,
      "train_runtime": 38.1015,
      "train_tokens_per_second": 5674.114
    },
    {
      "epoch": 0.012873563218390805,
      "grad_norm": 0.7270675897598267,
      "learning_rate": 0.0001989324498435487,
      "loss": 2.1925,
      "num_input_tokens_seen": 222464,
      "step": 35,
      "train_runtime": 39.0922,
      "train_tokens_per_second": 5690.747
    },
    {
      "epoch": 0.013241379310344827,
      "grad_norm": 0.5558647513389587,
      "learning_rate": 0.0001988956377691883,
      "loss": 2.0255,
      "num_input_tokens_seen": 229888,
      "step": 36,
      "train_runtime": 40.2696,
      "train_tokens_per_second": 5708.723
    },
    {
      "epoch": 0.01360919540229885,
      "grad_norm": 0.6663904190063477,
      "learning_rate": 0.0001988588256948279,
      "loss": 2.0809,
      "num_input_tokens_seen": 235456,
      "step": 37,
      "train_runtime": 41.1822,
      "train_tokens_per_second": 5717.42
    },
    {
      "epoch": 0.013977011494252874,
      "grad_norm": 0.6171004772186279,
      "learning_rate": 0.00019882201362046752,
      "loss": 2.0906,
      "num_input_tokens_seen": 240944,
      "step": 38,
      "train_runtime": 42.0899,
      "train_tokens_per_second": 5724.502
    },
    {
      "epoch": 0.014344827586206896,
      "grad_norm": 0.6836889982223511,
      "learning_rate": 0.00019878520154610713,
      "loss": 2.1421,
      "num_input_tokens_seen": 246128,
      "step": 39,
      "train_runtime": 42.9646,
      "train_tokens_per_second": 5728.628
    },
    {
      "epoch": 0.014712643678160919,
      "grad_norm": 0.6907501816749573,
      "learning_rate": 0.00019874838947174674,
      "loss": 2.049,
      "num_input_tokens_seen": 253152,
      "step": 40,
      "train_runtime": 44.0632,
      "train_tokens_per_second": 5745.199
    },
    {
      "epoch": 0.015080459770114943,
      "grad_norm": 0.791257917881012,
      "learning_rate": 0.00019871157739738635,
      "loss": 2.1161,
      "num_input_tokens_seen": 257744,
      "step": 41,
      "train_runtime": 44.852,
      "train_tokens_per_second": 5746.546
    },
    {
      "epoch": 0.015448275862068966,
      "grad_norm": 0.6082923412322998,
      "learning_rate": 0.00019867476532302596,
      "loss": 2.0632,
      "num_input_tokens_seen": 266240,
      "step": 42,
      "train_runtime": 46.1241,
      "train_tokens_per_second": 5772.257
    },
    {
      "epoch": 0.01581609195402299,
      "grad_norm": 0.660675585269928,
      "learning_rate": 0.00019863795324866557,
      "loss": 2.0848,
      "num_input_tokens_seen": 270912,
      "step": 43,
      "train_runtime": 46.9415,
      "train_tokens_per_second": 5771.27
    },
    {
      "epoch": 0.01618390804597701,
      "grad_norm": 0.7294403910636902,
      "learning_rate": 0.00019860114117430517,
      "loss": 2.0272,
      "num_input_tokens_seen": 275680,
      "step": 44,
      "train_runtime": 47.7241,
      "train_tokens_per_second": 5776.539
    },
    {
      "epoch": 0.016551724137931035,
      "grad_norm": 0.6849520206451416,
      "learning_rate": 0.0001985643290999448,
      "loss": 2.0366,
      "num_input_tokens_seen": 281616,
      "step": 45,
      "train_runtime": 48.6737,
      "train_tokens_per_second": 5785.796
    },
    {
      "epoch": 0.01691954022988506,
      "grad_norm": 0.67539381980896,
      "learning_rate": 0.00019852751702558442,
      "loss": 2.1385,
      "num_input_tokens_seen": 287168,
      "step": 46,
      "train_runtime": 49.5918,
      "train_tokens_per_second": 5790.637
    },
    {
      "epoch": 0.01728735632183908,
      "grad_norm": 0.7245020270347595,
      "learning_rate": 0.00019849070495122403,
      "loss": 2.189,
      "num_input_tokens_seen": 292832,
      "step": 47,
      "train_runtime": 50.4942,
      "train_tokens_per_second": 5799.325
    },
    {
      "epoch": 0.017655172413793104,
      "grad_norm": 0.6003301739692688,
      "learning_rate": 0.0001984538928768636,
      "loss": 1.8756,
      "num_input_tokens_seen": 300880,
      "step": 48,
      "train_runtime": 51.7412,
      "train_tokens_per_second": 5815.098
    },
    {
      "epoch": 0.018022988505747125,
      "grad_norm": 0.758628785610199,
      "learning_rate": 0.00019841708080250322,
      "loss": 2.1104,
      "num_input_tokens_seen": 305552,
      "step": 49,
      "train_runtime": 52.5593,
      "train_tokens_per_second": 5813.476
    },
    {
      "epoch": 0.01839080459770115,
      "grad_norm": 0.6878520846366882,
      "learning_rate": 0.00019838026872814283,
      "loss": 1.9863,
      "num_input_tokens_seen": 310800,
      "step": 50,
      "train_runtime": 53.4344,
      "train_tokens_per_second": 5816.477
    },
    {
      "epoch": 0.018758620689655173,
      "grad_norm": 0.671724259853363,
      "learning_rate": 0.00019834345665378244,
      "loss": 2.1907,
      "num_input_tokens_seen": 316768,
      "step": 51,
      "train_runtime": 54.3971,
      "train_tokens_per_second": 5823.251
    },
    {
      "epoch": 0.019126436781609194,
      "grad_norm": 0.6084379553794861,
      "learning_rate": 0.00019830664457942205,
      "loss": 1.9718,
      "num_input_tokens_seen": 324432,
      "step": 52,
      "train_runtime": 55.5928,
      "train_tokens_per_second": 5835.864
    },
    {
      "epoch": 0.01949425287356322,
      "grad_norm": 0.6924793124198914,
      "learning_rate": 0.00019826983250506165,
      "loss": 2.1719,
      "num_input_tokens_seen": 329984,
      "step": 53,
      "train_runtime": 56.4761,
      "train_tokens_per_second": 5842.892
    },
    {
      "epoch": 0.019862068965517243,
      "grad_norm": 0.7155401706695557,
      "learning_rate": 0.0001982330204307013,
      "loss": 2.0417,
      "num_input_tokens_seen": 336144,
      "step": 54,
      "train_runtime": 57.4735,
      "train_tokens_per_second": 5848.68
    },
    {
      "epoch": 0.020229885057471263,
      "grad_norm": 0.7476499080657959,
      "learning_rate": 0.0001981962083563409,
      "loss": 2.0562,
      "num_input_tokens_seen": 340672,
      "step": 55,
      "train_runtime": 58.2548,
      "train_tokens_per_second": 5847.96
    },
    {
      "epoch": 0.020597701149425288,
      "grad_norm": 0.680925726890564,
      "learning_rate": 0.0001981593962819805,
      "loss": 2.0042,
      "num_input_tokens_seen": 346432,
      "step": 56,
      "train_runtime": 59.1757,
      "train_tokens_per_second": 5854.297
    },
    {
      "epoch": 0.020965517241379312,
      "grad_norm": 0.6965209245681763,
      "learning_rate": 0.00019812258420762012,
      "loss": 1.9252,
      "num_input_tokens_seen": 352000,
      "step": 57,
      "train_runtime": 60.0969,
      "train_tokens_per_second": 5857.208
    },
    {
      "epoch": 0.021333333333333333,
      "grad_norm": 0.7026582956314087,
      "learning_rate": 0.00019808577213325973,
      "loss": 1.9766,
      "num_input_tokens_seen": 359616,
      "step": 58,
      "train_runtime": 61.3006,
      "train_tokens_per_second": 5866.44
    },
    {
      "epoch": 0.021701149425287357,
      "grad_norm": 0.7206132411956787,
      "learning_rate": 0.00019804896005889934,
      "loss": 2.1241,
      "num_input_tokens_seen": 365184,
      "step": 59,
      "train_runtime": 62.2128,
      "train_tokens_per_second": 5869.918
    },
    {
      "epoch": 0.022068965517241378,
      "grad_norm": 0.8915102481842041,
      "learning_rate": 0.00019801214798453894,
      "loss": 2.1134,
      "num_input_tokens_seen": 371232,
      "step": 60,
      "train_runtime": 63.179,
      "train_tokens_per_second": 5875.872
    },
    {
      "epoch": 0.022436781609195402,
      "grad_norm": 0.7137648463249207,
      "learning_rate": 0.00019797533591017853,
      "loss": 2.1075,
      "num_input_tokens_seen": 381120,
      "step": 61,
      "train_runtime": 65.413,
      "train_tokens_per_second": 5826.362
    },
    {
      "epoch": 0.022804597701149426,
      "grad_norm": 0.7164263129234314,
      "learning_rate": 0.00019793852383581816,
      "loss": 2.222,
      "num_input_tokens_seen": 386944,
      "step": 62,
      "train_runtime": 66.3496,
      "train_tokens_per_second": 5831.897
    },
    {
      "epoch": 0.023172413793103447,
      "grad_norm": 0.7986876368522644,
      "learning_rate": 0.00019790171176145777,
      "loss": 2.1947,
      "num_input_tokens_seen": 393200,
      "step": 63,
      "train_runtime": 67.3526,
      "train_tokens_per_second": 5837.933
    },
    {
      "epoch": 0.02354022988505747,
      "grad_norm": 0.6848318576812744,
      "learning_rate": 0.00019786489968709738,
      "loss": 2.0857,
      "num_input_tokens_seen": 399840,
      "step": 64,
      "train_runtime": 68.3923,
      "train_tokens_per_second": 5846.27
    },
    {
      "epoch": 0.023908045977011495,
      "grad_norm": 0.6817879676818848,
      "learning_rate": 0.000197828087612737,
      "loss": 2.2208,
      "num_input_tokens_seen": 405040,
      "step": 65,
      "train_runtime": 69.268,
      "train_tokens_per_second": 5847.432
    },
    {
      "epoch": 0.024275862068965516,
      "grad_norm": 0.6265236735343933,
      "learning_rate": 0.0001977912755383766,
      "loss": 2.0757,
      "num_input_tokens_seen": 412400,
      "step": 66,
      "train_runtime": 70.4133,
      "train_tokens_per_second": 5856.85
    },
    {
      "epoch": 0.02464367816091954,
      "grad_norm": 0.6848111152648926,
      "learning_rate": 0.0001977544634640162,
      "loss": 2.108,
      "num_input_tokens_seen": 418752,
      "step": 67,
      "train_runtime": 71.4046,
      "train_tokens_per_second": 5864.495
    },
    {
      "epoch": 0.025011494252873565,
      "grad_norm": 0.8280056715011597,
      "learning_rate": 0.00019771765138965582,
      "loss": 2.1599,
      "num_input_tokens_seen": 423712,
      "step": 68,
      "train_runtime": 72.213,
      "train_tokens_per_second": 5867.534
    },
    {
      "epoch": 0.025379310344827585,
      "grad_norm": 0.7020395994186401,
      "learning_rate": 0.00019768083931529542,
      "loss": 1.9425,
      "num_input_tokens_seen": 429712,
      "step": 69,
      "train_runtime": 73.2099,
      "train_tokens_per_second": 5869.591
    },
    {
      "epoch": 0.02574712643678161,
      "grad_norm": 0.6710706353187561,
      "learning_rate": 0.00019764402724093503,
      "loss": 2.0704,
      "num_input_tokens_seen": 435968,
      "step": 70,
      "train_runtime": 74.201,
      "train_tokens_per_second": 5875.5
    },
    {
      "epoch": 0.026114942528735634,
      "grad_norm": 0.7658275961875916,
      "learning_rate": 0.00019760721516657464,
      "loss": 2.2412,
      "num_input_tokens_seen": 441104,
      "step": 71,
      "train_runtime": 75.0628,
      "train_tokens_per_second": 5876.468
    },
    {
      "epoch": 0.026482758620689655,
      "grad_norm": 0.7414282560348511,
      "learning_rate": 0.00019757040309221425,
      "loss": 1.953,
      "num_input_tokens_seen": 446656,
      "step": 72,
      "train_runtime": 75.9771,
      "train_tokens_per_second": 5878.826
    },
    {
      "epoch": 0.02685057471264368,
      "grad_norm": 0.6625975966453552,
      "learning_rate": 0.00019753359101785386,
      "loss": 1.9609,
      "num_input_tokens_seen": 451696,
      "step": 73,
      "train_runtime": 76.8217,
      "train_tokens_per_second": 5879.798
    },
    {
      "epoch": 0.0272183908045977,
      "grad_norm": 0.7489330768585205,
      "learning_rate": 0.00019749677894349347,
      "loss": 1.977,
      "num_input_tokens_seen": 456176,
      "step": 74,
      "train_runtime": 77.5958,
      "train_tokens_per_second": 5878.876
    },
    {
      "epoch": 0.027586206896551724,
      "grad_norm": 0.721580982208252,
      "learning_rate": 0.00019745996686913308,
      "loss": 2.113,
      "num_input_tokens_seen": 461136,
      "step": 75,
      "train_runtime": 78.4053,
      "train_tokens_per_second": 5881.442
    },
    {
      "epoch": 0.027954022988505748,
      "grad_norm": 0.6909099817276001,
      "learning_rate": 0.00019742315479477269,
      "loss": 1.8897,
      "num_input_tokens_seen": 465632,
      "step": 76,
      "train_runtime": 79.1659,
      "train_tokens_per_second": 5881.721
    },
    {
      "epoch": 0.02832183908045977,
      "grad_norm": 0.6312218308448792,
      "learning_rate": 0.0001973863427204123,
      "loss": 2.0388,
      "num_input_tokens_seen": 471472,
      "step": 77,
      "train_runtime": 80.0953,
      "train_tokens_per_second": 5886.385
    },
    {
      "epoch": 0.028689655172413793,
      "grad_norm": 0.7163512706756592,
      "learning_rate": 0.0001973495306460519,
      "loss": 1.9092,
      "num_input_tokens_seen": 476880,
      "step": 78,
      "train_runtime": 80.9825,
      "train_tokens_per_second": 5888.683
    },
    {
      "epoch": 0.029057471264367817,
      "grad_norm": 0.6447576880455017,
      "learning_rate": 0.00019731271857169154,
      "loss": 2.0245,
      "num_input_tokens_seen": 486816,
      "step": 79,
      "train_runtime": 82.4537,
      "train_tokens_per_second": 5904.116
    },
    {
      "epoch": 0.029425287356321838,
      "grad_norm": 0.6254683136940002,
      "learning_rate": 0.00019727590649733115,
      "loss": 2.0485,
      "num_input_tokens_seen": 493136,
      "step": 80,
      "train_runtime": 83.4795,
      "train_tokens_per_second": 5907.271
    },
    {
      "epoch": 0.029793103448275862,
      "grad_norm": 0.6712125539779663,
      "learning_rate": 0.00019723909442297076,
      "loss": 2.0567,
      "num_input_tokens_seen": 498016,
      "step": 81,
      "train_runtime": 84.3014,
      "train_tokens_per_second": 5907.567
    },
    {
      "epoch": 0.030160919540229886,
      "grad_norm": 0.7119662761688232,
      "learning_rate": 0.00019720228234861037,
      "loss": 1.9589,
      "num_input_tokens_seen": 504736,
      "step": 82,
      "train_runtime": 85.3856,
      "train_tokens_per_second": 5911.256
    },
    {
      "epoch": 0.030528735632183907,
      "grad_norm": 0.7559812664985657,
      "learning_rate": 0.00019716547027424998,
      "loss": 2.3758,
      "num_input_tokens_seen": 509776,
      "step": 83,
      "train_runtime": 86.2249,
      "train_tokens_per_second": 5912.165
    },
    {
      "epoch": 0.03089655172413793,
      "grad_norm": 0.7353950142860413,
      "learning_rate": 0.00019712865819988958,
      "loss": 1.8818,
      "num_input_tokens_seen": 515584,
      "step": 84,
      "train_runtime": 87.1635,
      "train_tokens_per_second": 5915.134
    },
    {
      "epoch": 0.03126436781609195,
      "grad_norm": 0.676337718963623,
      "learning_rate": 0.00019709184612552917,
      "loss": 1.8061,
      "num_input_tokens_seen": 521408,
      "step": 85,
      "train_runtime": 88.1383,
      "train_tokens_per_second": 5915.793
    },
    {
      "epoch": 0.03163218390804598,
      "grad_norm": 0.6870601177215576,
      "learning_rate": 0.00019705503405116878,
      "loss": 1.8463,
      "num_input_tokens_seen": 527504,
      "step": 86,
      "train_runtime": 89.1419,
      "train_tokens_per_second": 5917.579
    },
    {
      "epoch": 0.032,
      "grad_norm": 0.6499676704406738,
      "learning_rate": 0.00019701822197680838,
      "loss": 1.9626,
      "num_input_tokens_seen": 534000,
      "step": 87,
      "train_runtime": 90.2038,
      "train_tokens_per_second": 5919.928
    },
    {
      "epoch": 0.03236781609195402,
      "grad_norm": 0.7987402081489563,
      "learning_rate": 0.00019698140990244802,
      "loss": 2.0343,
      "num_input_tokens_seen": 539200,
      "step": 88,
      "train_runtime": 91.0746,
      "train_tokens_per_second": 5920.421
    },
    {
      "epoch": 0.03273563218390805,
      "grad_norm": 0.706471860408783,
      "learning_rate": 0.00019694459782808763,
      "loss": 2.2647,
      "num_input_tokens_seen": 545168,
      "step": 89,
      "train_runtime": 92.0414,
      "train_tokens_per_second": 5923.077
    },
    {
      "epoch": 0.03310344827586207,
      "grad_norm": 0.6408177018165588,
      "learning_rate": 0.00019690778575372724,
      "loss": 1.9102,
      "num_input_tokens_seen": 550832,
      "step": 90,
      "train_runtime": 92.9689,
      "train_tokens_per_second": 5924.908
    },
    {
      "epoch": 0.03347126436781609,
      "grad_norm": 0.6255334615707397,
      "learning_rate": 0.00019687097367936685,
      "loss": 1.8175,
      "num_input_tokens_seen": 557872,
      "step": 91,
      "train_runtime": 94.6874,
      "train_tokens_per_second": 5891.721
    },
    {
      "epoch": 0.03383908045977012,
      "grad_norm": 0.7798760533332825,
      "learning_rate": 0.00019683416160500646,
      "loss": 2.1608,
      "num_input_tokens_seen": 562304,
      "step": 92,
      "train_runtime": 95.4545,
      "train_tokens_per_second": 5890.806
    },
    {
      "epoch": 0.03420689655172414,
      "grad_norm": 0.7144827246665955,
      "learning_rate": 0.00019679734953064606,
      "loss": 1.9687,
      "num_input_tokens_seen": 567520,
      "step": 93,
      "train_runtime": 96.3045,
      "train_tokens_per_second": 5892.973
    },
    {
      "epoch": 0.03457471264367816,
      "grad_norm": 0.7443671226501465,
      "learning_rate": 0.00019676053745628567,
      "loss": 2.0866,
      "num_input_tokens_seen": 575472,
      "step": 94,
      "train_runtime": 97.564,
      "train_tokens_per_second": 5898.406
    },
    {
      "epoch": 0.03494252873563218,
      "grad_norm": 0.7238397598266602,
      "learning_rate": 0.00019672372538192528,
      "loss": 1.8491,
      "num_input_tokens_seen": 582480,
      "step": 95,
      "train_runtime": 98.6733,
      "train_tokens_per_second": 5903.114
    },
    {
      "epoch": 0.03531034482758621,
      "grad_norm": 0.6610636115074158,
      "learning_rate": 0.0001966869133075649,
      "loss": 1.9805,
      "num_input_tokens_seen": 587536,
      "step": 96,
      "train_runtime": 99.5254,
      "train_tokens_per_second": 5903.378
    },
    {
      "epoch": 0.03567816091954023,
      "grad_norm": 0.7728227376937866,
      "learning_rate": 0.0001966501012332045,
      "loss": 2.0613,
      "num_input_tokens_seen": 592624,
      "step": 97,
      "train_runtime": 100.3693,
      "train_tokens_per_second": 5904.437
    },
    {
      "epoch": 0.03604597701149425,
      "grad_norm": 0.7076001763343811,
      "learning_rate": 0.0001966132891588441,
      "loss": 1.9572,
      "num_input_tokens_seen": 597536,
      "step": 98,
      "train_runtime": 101.2043,
      "train_tokens_per_second": 5904.255
    },
    {
      "epoch": 0.03641379310344828,
      "grad_norm": 0.7417423129081726,
      "learning_rate": 0.00019657647708448372,
      "loss": 2.0314,
      "num_input_tokens_seen": 601744,
      "step": 99,
      "train_runtime": 101.9606,
      "train_tokens_per_second": 5901.733
    },
    {
      "epoch": 0.0367816091954023,
      "grad_norm": 0.7208441495895386,
      "learning_rate": 0.00019653966501012333,
      "loss": 2.0895,
      "num_input_tokens_seen": 611232,
      "step": 100,
      "train_runtime": 103.4121,
      "train_tokens_per_second": 5910.644
    },
    {
      "epoch": 0.03714942528735632,
      "grad_norm": 0.7312285900115967,
      "learning_rate": 0.00019650285293576294,
      "loss": 1.9561,
      "num_input_tokens_seen": 617136,
      "step": 101,
      "train_runtime": 104.3597,
      "train_tokens_per_second": 5913.55
    },
    {
      "epoch": 0.03751724137931035,
      "grad_norm": 0.7724730372428894,
      "learning_rate": 0.00019646604086140254,
      "loss": 1.9849,
      "num_input_tokens_seen": 621312,
      "step": 102,
      "train_runtime": 105.0845,
      "train_tokens_per_second": 5912.498
    },
    {
      "epoch": 0.03788505747126437,
      "grad_norm": 0.7689005732536316,
      "learning_rate": 0.00019642922878704215,
      "loss": 2.0727,
      "num_input_tokens_seen": 625456,
      "step": 103,
      "train_runtime": 105.8097,
      "train_tokens_per_second": 5911.143
    },
    {
      "epoch": 0.03825287356321839,
      "grad_norm": 0.7934838533401489,
      "learning_rate": 0.00019639241671268176,
      "loss": 1.981,
      "num_input_tokens_seen": 630224,
      "step": 104,
      "train_runtime": 106.6147,
      "train_tokens_per_second": 5911.232
    },
    {
      "epoch": 0.038620689655172416,
      "grad_norm": 0.7310318350791931,
      "learning_rate": 0.0001963556046383214,
      "loss": 1.8631,
      "num_input_tokens_seen": 635408,
      "step": 105,
      "train_runtime": 107.4963,
      "train_tokens_per_second": 5910.973
    },
    {
      "epoch": 0.03898850574712644,
      "grad_norm": 0.7392246127128601,
      "learning_rate": 0.000196318792563961,
      "loss": 1.9564,
      "num_input_tokens_seen": 639856,
      "step": 106,
      "train_runtime": 108.2494,
      "train_tokens_per_second": 5910.943
    },
    {
      "epoch": 0.03935632183908046,
      "grad_norm": 0.8037939071655273,
      "learning_rate": 0.00019628198048960062,
      "loss": 2.111,
      "num_input_tokens_seen": 644816,
      "step": 107,
      "train_runtime": 109.1141,
      "train_tokens_per_second": 5909.557
    },
    {
      "epoch": 0.039724137931034485,
      "grad_norm": 0.7811733484268188,
      "learning_rate": 0.0001962451684152402,
      "loss": 1.9755,
      "num_input_tokens_seen": 648992,
      "step": 108,
      "train_runtime": 109.8421,
      "train_tokens_per_second": 5908.406
    },
    {
      "epoch": 0.040091954022988506,
      "grad_norm": 0.6561762094497681,
      "learning_rate": 0.0001962083563408798,
      "loss": 2.0295,
      "num_input_tokens_seen": 655824,
      "step": 109,
      "train_runtime": 110.9528,
      "train_tokens_per_second": 5910.839
    },
    {
      "epoch": 0.04045977011494253,
      "grad_norm": 0.7730290293693542,
      "learning_rate": 0.00019617154426651942,
      "loss": 2.0266,
      "num_input_tokens_seen": 661232,
      "step": 110,
      "train_runtime": 111.8435,
      "train_tokens_per_second": 5912.118
    },
    {
      "epoch": 0.040827586206896554,
      "grad_norm": 0.7397712469100952,
      "learning_rate": 0.00019613473219215902,
      "loss": 1.903,
      "num_input_tokens_seen": 668048,
      "step": 111,
      "train_runtime": 112.9125,
      "train_tokens_per_second": 5916.509
    },
    {
      "epoch": 0.041195402298850575,
      "grad_norm": 0.775788426399231,
      "learning_rate": 0.00019609792011779863,
      "loss": 2.0736,
      "num_input_tokens_seen": 672672,
      "step": 112,
      "train_runtime": 113.6835,
      "train_tokens_per_second": 5917.059
    },
    {
      "epoch": 0.041563218390804596,
      "grad_norm": 0.7797473669052124,
      "learning_rate": 0.00019606110804343824,
      "loss": 1.9446,
      "num_input_tokens_seen": 678000,
      "step": 113,
      "train_runtime": 114.5807,
      "train_tokens_per_second": 5917.227
    },
    {
      "epoch": 0.041931034482758624,
      "grad_norm": 0.6599826216697693,
      "learning_rate": 0.00019602429596907788,
      "loss": 1.8504,
      "num_input_tokens_seen": 683504,
      "step": 114,
      "train_runtime": 115.49,
      "train_tokens_per_second": 5918.295
    },
    {
      "epoch": 0.042298850574712644,
      "grad_norm": 0.7596398591995239,
      "learning_rate": 0.0001959874838947175,
      "loss": 1.9072,
      "num_input_tokens_seen": 687792,
      "step": 115,
      "train_runtime": 116.2558,
      "train_tokens_per_second": 5916.193
    },
    {
      "epoch": 0.042666666666666665,
      "grad_norm": 0.7828064560890198,
      "learning_rate": 0.0001959506718203571,
      "loss": 1.9661,
      "num_input_tokens_seen": 694304,
      "step": 116,
      "train_runtime": 117.278,
      "train_tokens_per_second": 5920.156
    },
    {
      "epoch": 0.04303448275862069,
      "grad_norm": 0.7471738457679749,
      "learning_rate": 0.0001959138597459967,
      "loss": 2.0649,
      "num_input_tokens_seen": 698976,
      "step": 117,
      "train_runtime": 118.0951,
      "train_tokens_per_second": 5918.756
    },
    {
      "epoch": 0.043402298850574714,
      "grad_norm": 0.8071371912956238,
      "learning_rate": 0.00019587704767163631,
      "loss": 2.0733,
      "num_input_tokens_seen": 704432,
      "step": 118,
      "train_runtime": 118.9873,
      "train_tokens_per_second": 5920.227
    },
    {
      "epoch": 0.043770114942528734,
      "grad_norm": 0.7780758738517761,
      "learning_rate": 0.00019584023559727592,
      "loss": 2.0317,
      "num_input_tokens_seen": 709296,
      "step": 119,
      "train_runtime": 119.8229,
      "train_tokens_per_second": 5919.536
    },
    {
      "epoch": 0.044137931034482755,
      "grad_norm": 0.7223494648933411,
      "learning_rate": 0.00019580342352291553,
      "loss": 1.9139,
      "num_input_tokens_seen": 716320,
      "step": 120,
      "train_runtime": 120.9563,
      "train_tokens_per_second": 5922.14
    },
    {
      "epoch": 0.04450574712643678,
      "grad_norm": 0.7625210285186768,
      "learning_rate": 0.0001957666114485551,
      "loss": 1.9894,
      "num_input_tokens_seen": 720848,
      "step": 121,
      "train_runtime": 122.3193,
      "train_tokens_per_second": 5893.168
    },
    {
      "epoch": 0.044873563218390804,
      "grad_norm": 0.6506988406181335,
      "learning_rate": 0.00019572979937419475,
      "loss": 2.0348,
      "num_input_tokens_seen": 727888,
      "step": 122,
      "train_runtime": 123.4414,
      "train_tokens_per_second": 5896.629
    },
    {
      "epoch": 0.045241379310344824,
      "grad_norm": 0.7588003873825073,
      "learning_rate": 0.00019569298729983436,
      "loss": 2.0239,
      "num_input_tokens_seen": 733184,
      "step": 123,
      "train_runtime": 124.3282,
      "train_tokens_per_second": 5897.164
    },
    {
      "epoch": 0.04560919540229885,
      "grad_norm": 0.7006298303604126,
      "learning_rate": 0.00019565617522547397,
      "loss": 2.0033,
      "num_input_tokens_seen": 738528,
      "step": 124,
      "train_runtime": 125.2185,
      "train_tokens_per_second": 5897.915
    },
    {
      "epoch": 0.04597701149425287,
      "grad_norm": 0.7602880597114563,
      "learning_rate": 0.00019561936315111358,
      "loss": 2.1163,
      "num_input_tokens_seen": 745008,
      "step": 125,
      "train_runtime": 126.252,
      "train_tokens_per_second": 5900.958
    },
    {
      "epoch": 0.046344827586206894,
      "grad_norm": 0.704163134098053,
      "learning_rate": 0.00019558255107675319,
      "loss": 1.9013,
      "num_input_tokens_seen": 750800,
      "step": 126,
      "train_runtime": 127.2374,
      "train_tokens_per_second": 5900.779
    },
    {
      "epoch": 0.04671264367816092,
      "grad_norm": 0.812581479549408,
      "learning_rate": 0.0001955457390023928,
      "loss": 1.8615,
      "num_input_tokens_seen": 756416,
      "step": 127,
      "train_runtime": 128.152,
      "train_tokens_per_second": 5902.491
    },
    {
      "epoch": 0.04708045977011494,
      "grad_norm": 0.7117610573768616,
      "learning_rate": 0.0001955089269280324,
      "loss": 1.8595,
      "num_input_tokens_seen": 766512,
      "step": 128,
      "train_runtime": 129.6708,
      "train_tokens_per_second": 5911.217
    },
    {
      "epoch": 0.04744827586206896,
      "grad_norm": 0.6793472170829773,
      "learning_rate": 0.000195472114853672,
      "loss": 1.6775,
      "num_input_tokens_seen": 771776,
      "step": 129,
      "train_runtime": 130.5364,
      "train_tokens_per_second": 5912.344
    },
    {
      "epoch": 0.04781609195402299,
      "grad_norm": 0.7294008135795593,
      "learning_rate": 0.00019543530277931162,
      "loss": 2.1307,
      "num_input_tokens_seen": 777696,
      "step": 130,
      "train_runtime": 131.4895,
      "train_tokens_per_second": 5914.509
    },
    {
      "epoch": 0.04818390804597701,
      "grad_norm": 0.7181180715560913,
      "learning_rate": 0.00019539849070495123,
      "loss": 1.6838,
      "num_input_tokens_seen": 782928,
      "step": 131,
      "train_runtime": 132.3444,
      "train_tokens_per_second": 5915.836
    },
    {
      "epoch": 0.04855172413793103,
      "grad_norm": 0.824769914150238,
      "learning_rate": 0.00019536167863059084,
      "loss": 2.1886,
      "num_input_tokens_seen": 788032,
      "step": 132,
      "train_runtime": 133.2162,
      "train_tokens_per_second": 5915.437
    },
    {
      "epoch": 0.04891954022988506,
      "grad_norm": 1.0359686613082886,
      "learning_rate": 0.00019532486655623045,
      "loss": 2.1412,
      "num_input_tokens_seen": 792992,
      "step": 133,
      "train_runtime": 134.0748,
      "train_tokens_per_second": 5914.551
    },
    {
      "epoch": 0.04928735632183908,
      "grad_norm": 0.7491018176078796,
      "learning_rate": 0.00019528805448187006,
      "loss": 2.0047,
      "num_input_tokens_seen": 798208,
      "step": 134,
      "train_runtime": 134.9399,
      "train_tokens_per_second": 5915.287
    },
    {
      "epoch": 0.0496551724137931,
      "grad_norm": 0.7163192629814148,
      "learning_rate": 0.00019525124240750967,
      "loss": 1.9637,
      "num_input_tokens_seen": 804240,
      "step": 135,
      "train_runtime": 135.9313,
      "train_tokens_per_second": 5916.517
    },
    {
      "epoch": 0.05002298850574713,
      "grad_norm": 0.7383331060409546,
      "learning_rate": 0.00019521443033314927,
      "loss": 1.8713,
      "num_input_tokens_seen": 809808,
      "step": 136,
      "train_runtime": 136.8463,
      "train_tokens_per_second": 5917.645
    },
    {
      "epoch": 0.05039080459770115,
      "grad_norm": 0.856783390045166,
      "learning_rate": 0.00019517761825878888,
      "loss": 2.0877,
      "num_input_tokens_seen": 814768,
      "step": 137,
      "train_runtime": 137.6868,
      "train_tokens_per_second": 5917.548
    },
    {
      "epoch": 0.05075862068965517,
      "grad_norm": 0.743935763835907,
      "learning_rate": 0.0001951408061844285,
      "loss": 2.071,
      "num_input_tokens_seen": 819120,
      "step": 138,
      "train_runtime": 138.428,
      "train_tokens_per_second": 5917.298
    },
    {
      "epoch": 0.0511264367816092,
      "grad_norm": 0.8687429428100586,
      "learning_rate": 0.00019510399411006813,
      "loss": 1.7694,
      "num_input_tokens_seen": 823904,
      "step": 139,
      "train_runtime": 139.2622,
      "train_tokens_per_second": 5916.206
    },
    {
      "epoch": 0.05149425287356322,
      "grad_norm": 0.8015475273132324,
      "learning_rate": 0.00019506718203570774,
      "loss": 2.1534,
      "num_input_tokens_seen": 829264,
      "step": 140,
      "train_runtime": 140.1519,
      "train_tokens_per_second": 5916.894
    },
    {
      "epoch": 0.05186206896551724,
      "grad_norm": 0.7176015973091125,
      "learning_rate": 0.00019503036996134735,
      "loss": 1.8552,
      "num_input_tokens_seen": 837184,
      "step": 141,
      "train_runtime": 141.4017,
      "train_tokens_per_second": 5920.608
    },
    {
      "epoch": 0.05222988505747127,
      "grad_norm": 0.808400571346283,
      "learning_rate": 0.00019499355788698695,
      "loss": 1.8303,
      "num_input_tokens_seen": 843648,
      "step": 142,
      "train_runtime": 142.4251,
      "train_tokens_per_second": 5923.451
    },
    {
      "epoch": 0.05259770114942529,
      "grad_norm": 0.7893350124359131,
      "learning_rate": 0.00019495674581262656,
      "loss": 1.9399,
      "num_input_tokens_seen": 848464,
      "step": 143,
      "train_runtime": 143.254,
      "train_tokens_per_second": 5922.796
    },
    {
      "epoch": 0.05296551724137931,
      "grad_norm": 0.7189632058143616,
      "learning_rate": 0.00019491993373826615,
      "loss": 1.9882,
      "num_input_tokens_seen": 853712,
      "step": 144,
      "train_runtime": 144.1225,
      "train_tokens_per_second": 5923.517
    },
    {
      "epoch": 0.05333333333333334,
      "grad_norm": 0.7173884510993958,
      "learning_rate": 0.00019488312166390575,
      "loss": 1.9626,
      "num_input_tokens_seen": 862640,
      "step": 145,
      "train_runtime": 145.4933,
      "train_tokens_per_second": 5929.07
    },
    {
      "epoch": 0.05370114942528736,
      "grad_norm": 0.700886070728302,
      "learning_rate": 0.00019484630958954536,
      "loss": 2.0403,
      "num_input_tokens_seen": 868240,
      "step": 146,
      "train_runtime": 146.4209,
      "train_tokens_per_second": 5929.754
    },
    {
      "epoch": 0.05406896551724138,
      "grad_norm": 0.7500973343849182,
      "learning_rate": 0.00019480949751518497,
      "loss": 1.9662,
      "num_input_tokens_seen": 873792,
      "step": 147,
      "train_runtime": 147.3459,
      "train_tokens_per_second": 5930.209
    },
    {
      "epoch": 0.0544367816091954,
      "grad_norm": 0.688825249671936,
      "learning_rate": 0.0001947726854408246,
      "loss": 1.9987,
      "num_input_tokens_seen": 878896,
      "step": 148,
      "train_runtime": 148.2093,
      "train_tokens_per_second": 5930.098
    },
    {
      "epoch": 0.05480459770114943,
      "grad_norm": 0.7477114200592041,
      "learning_rate": 0.00019473587336646422,
      "loss": 1.8754,
      "num_input_tokens_seen": 884880,
      "step": 149,
      "train_runtime": 149.1824,
      "train_tokens_per_second": 5931.532
    },
    {
      "epoch": 0.05517241379310345,
      "grad_norm": 0.6594926118850708,
      "learning_rate": 0.00019469906129210383,
      "loss": 1.8852,
      "num_input_tokens_seen": 892704,
      "step": 150,
      "train_runtime": 150.4013,
      "train_tokens_per_second": 5935.481
    },
    {
      "epoch": 0.05554022988505747,
      "grad_norm": 0.8141273260116577,
      "learning_rate": 0.00019466224921774343,
      "loss": 2.0667,
      "num_input_tokens_seen": 897776,
      "step": 151,
      "train_runtime": 151.7296,
      "train_tokens_per_second": 5916.947
    },
    {
      "epoch": 0.055908045977011496,
      "grad_norm": 0.7371804118156433,
      "learning_rate": 0.00019462543714338304,
      "loss": 1.9458,
      "num_input_tokens_seen": 902800,
      "step": 152,
      "train_runtime": 152.5903,
      "train_tokens_per_second": 5916.497
    },
    {
      "epoch": 0.05627586206896552,
      "grad_norm": 0.8326572775840759,
      "learning_rate": 0.00019458862506902265,
      "loss": 1.8437,
      "num_input_tokens_seen": 908064,
      "step": 153,
      "train_runtime": 153.4696,
      "train_tokens_per_second": 5916.899
    },
    {
      "epoch": 0.05664367816091954,
      "grad_norm": 0.6677982211112976,
      "learning_rate": 0.00019455181299466226,
      "loss": 1.8823,
      "num_input_tokens_seen": 914304,
      "step": 154,
      "train_runtime": 154.4962,
      "train_tokens_per_second": 5917.969
    },
    {
      "epoch": 0.057011494252873565,
      "grad_norm": 0.8135290145874023,
      "learning_rate": 0.00019451500092030187,
      "loss": 1.8449,
      "num_input_tokens_seen": 919136,
      "step": 155,
      "train_runtime": 155.2979,
      "train_tokens_per_second": 5918.536
    },
    {
      "epoch": 0.057379310344827586,
      "grad_norm": 0.846254825592041,
      "learning_rate": 0.00019447818884594148,
      "loss": 1.8222,
      "num_input_tokens_seen": 923792,
      "step": 156,
      "train_runtime": 156.0922,
      "train_tokens_per_second": 5918.246
    },
    {
      "epoch": 0.05774712643678161,
      "grad_norm": 0.7704864740371704,
      "learning_rate": 0.0001944413767715811,
      "loss": 2.081,
      "num_input_tokens_seen": 928592,
      "step": 157,
      "train_runtime": 156.9028,
      "train_tokens_per_second": 5918.261
    },
    {
      "epoch": 0.058114942528735634,
      "grad_norm": 0.7155718803405762,
      "learning_rate": 0.0001944045646972207,
      "loss": 1.7299,
      "num_input_tokens_seen": 933488,
      "step": 158,
      "train_runtime": 157.7431,
      "train_tokens_per_second": 5917.772
    },
    {
      "epoch": 0.058482758620689655,
      "grad_norm": 0.8601603507995605,
      "learning_rate": 0.0001943677526228603,
      "loss": 1.9165,
      "num_input_tokens_seen": 938368,
      "step": 159,
      "train_runtime": 158.5841,
      "train_tokens_per_second": 5917.165
    },
    {
      "epoch": 0.058850574712643676,
      "grad_norm": 0.6630919575691223,
      "learning_rate": 0.00019433094054849991,
      "loss": 1.8437,
      "num_input_tokens_seen": 944416,
      "step": 160,
      "train_runtime": 159.5758,
      "train_tokens_per_second": 5918.292
    },
    {
      "epoch": 0.059218390804597704,
      "grad_norm": 0.7675873041152954,
      "learning_rate": 0.00019429412847413952,
      "loss": 1.8833,
      "num_input_tokens_seen": 950832,
      "step": 161,
      "train_runtime": 160.6078,
      "train_tokens_per_second": 5920.21
    },
    {
      "epoch": 0.059586206896551724,
      "grad_norm": 0.6628847122192383,
      "learning_rate": 0.00019425731639977913,
      "loss": 1.8991,
      "num_input_tokens_seen": 957472,
      "step": 162,
      "train_runtime": 161.6768,
      "train_tokens_per_second": 5922.136
    },
    {
      "epoch": 0.059954022988505745,
      "grad_norm": 0.7538667917251587,
      "learning_rate": 0.00019422050432541874,
      "loss": 1.9682,
      "num_input_tokens_seen": 962592,
      "step": 163,
      "train_runtime": 162.527,
      "train_tokens_per_second": 5922.66
    },
    {
      "epoch": 0.06032183908045977,
      "grad_norm": 0.7532102465629578,
      "learning_rate": 0.00019418369225105835,
      "loss": 2.1817,
      "num_input_tokens_seen": 967392,
      "step": 164,
      "train_runtime": 163.3328,
      "train_tokens_per_second": 5922.827
    },
    {
      "epoch": 0.060689655172413794,
      "grad_norm": 0.7017723321914673,
      "learning_rate": 0.00019414688017669799,
      "loss": 1.6279,
      "num_input_tokens_seen": 977776,
      "step": 165,
      "train_runtime": 164.8869,
      "train_tokens_per_second": 5929.979
    },
    {
      "epoch": 0.061057471264367814,
      "grad_norm": 0.7645009160041809,
      "learning_rate": 0.0001941100681023376,
      "loss": 2.0955,
      "num_input_tokens_seen": 982480,
      "step": 166,
      "train_runtime": 165.6833,
      "train_tokens_per_second": 5929.866
    },
    {
      "epoch": 0.06142528735632184,
      "grad_norm": 0.7056939005851746,
      "learning_rate": 0.0001940732560279772,
      "loss": 2.0229,
      "num_input_tokens_seen": 986656,
      "step": 167,
      "train_runtime": 166.4108,
      "train_tokens_per_second": 5929.038
    },
    {
      "epoch": 0.06179310344827586,
      "grad_norm": 0.6746269464492798,
      "learning_rate": 0.00019403644395361679,
      "loss": 1.9518,
      "num_input_tokens_seen": 992192,
      "step": 168,
      "train_runtime": 167.3108,
      "train_tokens_per_second": 5930.232
    },
    {
      "epoch": 0.062160919540229884,
      "grad_norm": 0.833311915397644,
      "learning_rate": 0.0001939996318792564,
      "loss": 2.1163,
      "num_input_tokens_seen": 998560,
      "step": 169,
      "train_runtime": 168.3248,
      "train_tokens_per_second": 5932.339
    },
    {
      "epoch": 0.0625287356321839,
      "grad_norm": 0.6848273277282715,
      "learning_rate": 0.000193962819804896,
      "loss": 1.7791,
      "num_input_tokens_seen": 1008480,
      "step": 170,
      "train_runtime": 169.8266,
      "train_tokens_per_second": 5938.292
    },
    {
      "epoch": 0.06289655172413793,
      "grad_norm": 0.7359018325805664,
      "learning_rate": 0.0001939260077305356,
      "loss": 2.0557,
      "num_input_tokens_seen": 1013920,
      "step": 171,
      "train_runtime": 170.74,
      "train_tokens_per_second": 5938.385
    },
    {
      "epoch": 0.06326436781609196,
      "grad_norm": 0.7058089375495911,
      "learning_rate": 0.00019388919565617522,
      "loss": 1.9924,
      "num_input_tokens_seen": 1019136,
      "step": 172,
      "train_runtime": 171.6074,
      "train_tokens_per_second": 5938.765
    },
    {
      "epoch": 0.06363218390804598,
      "grad_norm": 0.76641845703125,
      "learning_rate": 0.00019385238358181483,
      "loss": 1.9809,
      "num_input_tokens_seen": 1026608,
      "step": 173,
      "train_runtime": 172.7726,
      "train_tokens_per_second": 5941.96
    },
    {
      "epoch": 0.064,
      "grad_norm": 0.7802574634552002,
      "learning_rate": 0.00019381557150745447,
      "loss": 2.0284,
      "num_input_tokens_seen": 1031312,
      "step": 174,
      "train_runtime": 173.5671,
      "train_tokens_per_second": 5941.865
    },
    {
      "epoch": 0.06436781609195402,
      "grad_norm": 0.6622971892356873,
      "learning_rate": 0.00019377875943309408,
      "loss": 1.9194,
      "num_input_tokens_seen": 1039200,
      "step": 175,
      "train_runtime": 174.8053,
      "train_tokens_per_second": 5944.901
    },
    {
      "epoch": 0.06473563218390804,
      "grad_norm": 0.7032092213630676,
      "learning_rate": 0.00019374194735873368,
      "loss": 2.0711,
      "num_input_tokens_seen": 1046608,
      "step": 176,
      "train_runtime": 175.9545,
      "train_tokens_per_second": 5948.174
    },
    {
      "epoch": 0.06510344827586206,
      "grad_norm": 0.7073813080787659,
      "learning_rate": 0.0001937051352843733,
      "loss": 2.0348,
      "num_input_tokens_seen": 1051536,
      "step": 177,
      "train_runtime": 176.7741,
      "train_tokens_per_second": 5948.474
    },
    {
      "epoch": 0.0654712643678161,
      "grad_norm": 0.8278349041938782,
      "learning_rate": 0.0001936683232100129,
      "loss": 2.0159,
      "num_input_tokens_seen": 1057920,
      "step": 178,
      "train_runtime": 177.8032,
      "train_tokens_per_second": 5949.949
    },
    {
      "epoch": 0.06583908045977012,
      "grad_norm": 0.6394902467727661,
      "learning_rate": 0.0001936315111356525,
      "loss": 1.7842,
      "num_input_tokens_seen": 1065664,
      "step": 179,
      "train_runtime": 179.0201,
      "train_tokens_per_second": 5952.762
    },
    {
      "epoch": 0.06620689655172414,
      "grad_norm": 0.7177083492279053,
      "learning_rate": 0.00019359469906129212,
      "loss": 2.049,
      "num_input_tokens_seen": 1070560,
      "step": 180,
      "train_runtime": 179.8403,
      "train_tokens_per_second": 5952.836
    },
    {
      "epoch": 0.06657471264367816,
      "grad_norm": 0.5524939298629761,
      "learning_rate": 0.0001935578869869317,
      "loss": 1.59,
      "num_input_tokens_seen": 1079584,
      "step": 181,
      "train_runtime": 181.7961,
      "train_tokens_per_second": 5938.432
    },
    {
      "epoch": 0.06694252873563218,
      "grad_norm": 0.650891125202179,
      "learning_rate": 0.00019352107491257134,
      "loss": 1.7681,
      "num_input_tokens_seen": 1085744,
      "step": 182,
      "train_runtime": 182.7904,
      "train_tokens_per_second": 5939.83
    },
    {
      "epoch": 0.0673103448275862,
      "grad_norm": 0.7197988033294678,
      "learning_rate": 0.00019348426283821095,
      "loss": 2.0104,
      "num_input_tokens_seen": 1092928,
      "step": 183,
      "train_runtime": 183.9451,
      "train_tokens_per_second": 5941.598
    },
    {
      "epoch": 0.06767816091954024,
      "grad_norm": 0.7250876426696777,
      "learning_rate": 0.00019344745076385056,
      "loss": 2.012,
      "num_input_tokens_seen": 1098544,
      "step": 184,
      "train_runtime": 184.8766,
      "train_tokens_per_second": 5942.04
    },
    {
      "epoch": 0.06804597701149426,
      "grad_norm": 0.8138119578361511,
      "learning_rate": 0.00019341063868949016,
      "loss": 1.8514,
      "num_input_tokens_seen": 1103024,
      "step": 185,
      "train_runtime": 185.6527,
      "train_tokens_per_second": 5941.331
    },
    {
      "epoch": 0.06841379310344828,
      "grad_norm": 0.7372742295265198,
      "learning_rate": 0.00019337382661512977,
      "loss": 2.0076,
      "num_input_tokens_seen": 1109296,
      "step": 186,
      "train_runtime": 186.6716,
      "train_tokens_per_second": 5942.501
    },
    {
      "epoch": 0.0687816091954023,
      "grad_norm": 0.7455793023109436,
      "learning_rate": 0.00019333701454076938,
      "loss": 1.8861,
      "num_input_tokens_seen": 1115472,
      "step": 187,
      "train_runtime": 187.6639,
      "train_tokens_per_second": 5943.988
    },
    {
      "epoch": 0.06914942528735632,
      "grad_norm": 0.6509843468666077,
      "learning_rate": 0.000193300202466409,
      "loss": 1.8759,
      "num_input_tokens_seen": 1122800,
      "step": 188,
      "train_runtime": 188.8104,
      "train_tokens_per_second": 5946.706
    },
    {
      "epoch": 0.06951724137931034,
      "grad_norm": 0.7092277407646179,
      "learning_rate": 0.0001932633903920486,
      "loss": 1.9346,
      "num_input_tokens_seen": 1128288,
      "step": 189,
      "train_runtime": 189.7066,
      "train_tokens_per_second": 5947.541
    },
    {
      "epoch": 0.06988505747126436,
      "grad_norm": 0.7340949177742004,
      "learning_rate": 0.0001932265783176882,
      "loss": 1.9018,
      "num_input_tokens_seen": 1133280,
      "step": 190,
      "train_runtime": 190.5675,
      "train_tokens_per_second": 5946.87
    },
    {
      "epoch": 0.0702528735632184,
      "grad_norm": 0.7532691359519958,
      "learning_rate": 0.00019318976624332782,
      "loss": 1.8199,
      "num_input_tokens_seen": 1139216,
      "step": 191,
      "train_runtime": 191.5455,
      "train_tokens_per_second": 5947.497
    },
    {
      "epoch": 0.07062068965517242,
      "grad_norm": 0.7443405985832214,
      "learning_rate": 0.00019315295416896743,
      "loss": 1.9553,
      "num_input_tokens_seen": 1145456,
      "step": 192,
      "train_runtime": 192.5722,
      "train_tokens_per_second": 5948.189
    },
    {
      "epoch": 0.07098850574712644,
      "grad_norm": 0.6958128809928894,
      "learning_rate": 0.00019311614209460704,
      "loss": 1.9752,
      "num_input_tokens_seen": 1154736,
      "step": 193,
      "train_runtime": 193.9743,
      "train_tokens_per_second": 5953.035
    },
    {
      "epoch": 0.07135632183908046,
      "grad_norm": 0.7440369129180908,
      "learning_rate": 0.00019307933002024664,
      "loss": 1.9428,
      "num_input_tokens_seen": 1159824,
      "step": 194,
      "train_runtime": 194.8327,
      "train_tokens_per_second": 5952.922
    },
    {
      "epoch": 0.07172413793103448,
      "grad_norm": 0.7526656985282898,
      "learning_rate": 0.00019304251794588625,
      "loss": 1.8749,
      "num_input_tokens_seen": 1165600,
      "step": 195,
      "train_runtime": 195.773,
      "train_tokens_per_second": 5953.833
    },
    {
      "epoch": 0.0720919540229885,
      "grad_norm": 0.6941686272621155,
      "learning_rate": 0.00019300570587152586,
      "loss": 1.8185,
      "num_input_tokens_seen": 1172352,
      "step": 196,
      "train_runtime": 196.8749,
      "train_tokens_per_second": 5954.808
    },
    {
      "epoch": 0.07245977011494253,
      "grad_norm": 0.6862158179283142,
      "learning_rate": 0.00019296889379716547,
      "loss": 1.9434,
      "num_input_tokens_seen": 1178640,
      "step": 197,
      "train_runtime": 197.8842,
      "train_tokens_per_second": 5956.212
    },
    {
      "epoch": 0.07282758620689656,
      "grad_norm": 0.6635754108428955,
      "learning_rate": 0.00019293208172280508,
      "loss": 1.8859,
      "num_input_tokens_seen": 1182816,
      "step": 198,
      "train_runtime": 198.6144,
      "train_tokens_per_second": 5955.338
    },
    {
      "epoch": 0.07319540229885058,
      "grad_norm": 0.7505982518196106,
      "learning_rate": 0.00019289526964844472,
      "loss": 2.1551,
      "num_input_tokens_seen": 1188160,
      "step": 199,
      "train_runtime": 199.5198,
      "train_tokens_per_second": 5955.099
    },
    {
      "epoch": 0.0735632183908046,
      "grad_norm": 0.9316896200180054,
      "learning_rate": 0.00019285845757408432,
      "loss": 1.9084,
      "num_input_tokens_seen": 1192640,
      "step": 200,
      "train_runtime": 200.2805,
      "train_tokens_per_second": 5954.849
    },
    {
      "epoch": 0.07393103448275862,
      "grad_norm": 0.6875803470611572,
      "learning_rate": 0.00019282164549972393,
      "loss": 1.9004,
      "num_input_tokens_seen": 1199184,
      "step": 201,
      "train_runtime": 201.3469,
      "train_tokens_per_second": 5955.812
    },
    {
      "epoch": 0.07429885057471264,
      "grad_norm": 0.7336552739143372,
      "learning_rate": 0.00019278483342536354,
      "loss": 2.0354,
      "num_input_tokens_seen": 1204128,
      "step": 202,
      "train_runtime": 202.1979,
      "train_tokens_per_second": 5955.196
    },
    {
      "epoch": 0.07466666666666667,
      "grad_norm": 0.7187710404396057,
      "learning_rate": 0.00019274802135100315,
      "loss": 2.0457,
      "num_input_tokens_seen": 1210640,
      "step": 203,
      "train_runtime": 203.2341,
      "train_tokens_per_second": 5956.875
    },
    {
      "epoch": 0.0750344827586207,
      "grad_norm": 0.7940569519996643,
      "learning_rate": 0.00019271120927664273,
      "loss": 1.814,
      "num_input_tokens_seen": 1215328,
      "step": 204,
      "train_runtime": 204.0195,
      "train_tokens_per_second": 5956.922
    },
    {
      "epoch": 0.07540229885057471,
      "grad_norm": 0.7493606209754944,
      "learning_rate": 0.00019267439720228234,
      "loss": 2.0379,
      "num_input_tokens_seen": 1220416,
      "step": 205,
      "train_runtime": 204.8887,
      "train_tokens_per_second": 5956.482
    },
    {
      "epoch": 0.07577011494252874,
      "grad_norm": 0.7162241339683533,
      "learning_rate": 0.00019263758512792195,
      "loss": 1.8618,
      "num_input_tokens_seen": 1225584,
      "step": 206,
      "train_runtime": 205.754,
      "train_tokens_per_second": 5956.549
    },
    {
      "epoch": 0.07613793103448276,
      "grad_norm": 0.7061337232589722,
      "learning_rate": 0.00019260077305356156,
      "loss": 1.7301,
      "num_input_tokens_seen": 1234432,
      "step": 207,
      "train_runtime": 207.1037,
      "train_tokens_per_second": 5960.454
    },
    {
      "epoch": 0.07650574712643678,
      "grad_norm": 0.7702038884162903,
      "learning_rate": 0.0001925639609792012,
      "loss": 2.174,
      "num_input_tokens_seen": 1240592,
      "step": 208,
      "train_runtime": 208.1051,
      "train_tokens_per_second": 5961.371
    },
    {
      "epoch": 0.07687356321839081,
      "grad_norm": 0.7636705636978149,
      "learning_rate": 0.0001925271489048408,
      "loss": 1.867,
      "num_input_tokens_seen": 1246032,
      "step": 209,
      "train_runtime": 209.0143,
      "train_tokens_per_second": 5961.467
    },
    {
      "epoch": 0.07724137931034483,
      "grad_norm": 0.8681410551071167,
      "learning_rate": 0.0001924903368304804,
      "loss": 1.9139,
      "num_input_tokens_seen": 1251264,
      "step": 210,
      "train_runtime": 209.8934,
      "train_tokens_per_second": 5961.426
    },
    {
      "epoch": 0.07760919540229885,
      "grad_norm": 0.7371796369552612,
      "learning_rate": 0.00019245352475612002,
      "loss": 1.9159,
      "num_input_tokens_seen": 1256928,
      "step": 211,
      "train_runtime": 211.6387,
      "train_tokens_per_second": 5939.028
    },
    {
      "epoch": 0.07797701149425287,
      "grad_norm": 0.797585666179657,
      "learning_rate": 0.00019241671268175963,
      "loss": 2.0173,
      "num_input_tokens_seen": 1264112,
      "step": 212,
      "train_runtime": 212.7609,
      "train_tokens_per_second": 5941.467
    },
    {
      "epoch": 0.0783448275862069,
      "grad_norm": 0.7724104523658752,
      "learning_rate": 0.00019237990060739924,
      "loss": 1.6428,
      "num_input_tokens_seen": 1268048,
      "step": 213,
      "train_runtime": 213.4497,
      "train_tokens_per_second": 5940.736
    },
    {
      "epoch": 0.07871264367816092,
      "grad_norm": 0.8864970207214355,
      "learning_rate": 0.00019234308853303885,
      "loss": 1.7023,
      "num_input_tokens_seen": 1272768,
      "step": 214,
      "train_runtime": 214.2451,
      "train_tokens_per_second": 5940.71
    },
    {
      "epoch": 0.07908045977011494,
      "grad_norm": 0.7212643027305603,
      "learning_rate": 0.00019230627645867846,
      "loss": 1.9303,
      "num_input_tokens_seen": 1279456,
      "step": 215,
      "train_runtime": 215.3088,
      "train_tokens_per_second": 5942.424
    },
    {
      "epoch": 0.07944827586206897,
      "grad_norm": 0.822053849697113,
      "learning_rate": 0.00019226946438431807,
      "loss": 2.0435,
      "num_input_tokens_seen": 1284064,
      "step": 216,
      "train_runtime": 216.1007,
      "train_tokens_per_second": 5941.972
    },
    {
      "epoch": 0.07981609195402299,
      "grad_norm": 0.8493820428848267,
      "learning_rate": 0.00019223265230995768,
      "loss": 1.9627,
      "num_input_tokens_seen": 1288400,
      "step": 217,
      "train_runtime": 216.8378,
      "train_tokens_per_second": 5941.769
    },
    {
      "epoch": 0.08018390804597701,
      "grad_norm": 0.7004702687263489,
      "learning_rate": 0.00019219584023559728,
      "loss": 1.8237,
      "num_input_tokens_seen": 1295120,
      "step": 218,
      "train_runtime": 217.9149,
      "train_tokens_per_second": 5943.237
    },
    {
      "epoch": 0.08055172413793103,
      "grad_norm": 0.7256537675857544,
      "learning_rate": 0.0001921590281612369,
      "loss": 1.7867,
      "num_input_tokens_seen": 1299584,
      "step": 219,
      "train_runtime": 218.6966,
      "train_tokens_per_second": 5942.407
    },
    {
      "epoch": 0.08091954022988505,
      "grad_norm": 0.710449755191803,
      "learning_rate": 0.0001921222160868765,
      "loss": 1.8448,
      "num_input_tokens_seen": 1305504,
      "step": 220,
      "train_runtime": 219.6432,
      "train_tokens_per_second": 5943.748
    },
    {
      "epoch": 0.08128735632183907,
      "grad_norm": 0.8347249627113342,
      "learning_rate": 0.0001920854040125161,
      "loss": 1.7269,
      "num_input_tokens_seen": 1312032,
      "step": 221,
      "train_runtime": 220.6707,
      "train_tokens_per_second": 5945.656
    },
    {
      "epoch": 0.08165517241379311,
      "grad_norm": 0.6923952698707581,
      "learning_rate": 0.00019204859193815572,
      "loss": 1.7396,
      "num_input_tokens_seen": 1317600,
      "step": 222,
      "train_runtime": 221.5963,
      "train_tokens_per_second": 5945.947
    },
    {
      "epoch": 0.08202298850574713,
      "grad_norm": 0.8048452734947205,
      "learning_rate": 0.00019201177986379533,
      "loss": 1.9067,
      "num_input_tokens_seen": 1322128,
      "step": 223,
      "train_runtime": 222.3803,
      "train_tokens_per_second": 5945.346
    },
    {
      "epoch": 0.08239080459770115,
      "grad_norm": 0.8158103823661804,
      "learning_rate": 0.00019197496778943494,
      "loss": 2.1227,
      "num_input_tokens_seen": 1328480,
      "step": 224,
      "train_runtime": 223.4103,
      "train_tokens_per_second": 5946.369
    },
    {
      "epoch": 0.08275862068965517,
      "grad_norm": 0.7361687421798706,
      "learning_rate": 0.00019193815571507457,
      "loss": 1.8807,
      "num_input_tokens_seen": 1333520,
      "step": 225,
      "train_runtime": 224.2625,
      "train_tokens_per_second": 5946.245
    },
    {
      "epoch": 0.08312643678160919,
      "grad_norm": 0.6840609312057495,
      "learning_rate": 0.00019190134364071418,
      "loss": 1.9136,
      "num_input_tokens_seen": 1339040,
      "step": 226,
      "train_runtime": 225.1582,
      "train_tokens_per_second": 5947.107
    },
    {
      "epoch": 0.08349425287356321,
      "grad_norm": 0.7369843125343323,
      "learning_rate": 0.00019186453156635376,
      "loss": 1.9659,
      "num_input_tokens_seen": 1344880,
      "step": 227,
      "train_runtime": 226.1295,
      "train_tokens_per_second": 5947.389
    },
    {
      "epoch": 0.08386206896551725,
      "grad_norm": 0.785943329334259,
      "learning_rate": 0.00019182771949199337,
      "loss": 1.7901,
      "num_input_tokens_seen": 1349840,
      "step": 228,
      "train_runtime": 226.9633,
      "train_tokens_per_second": 5947.393
    },
    {
      "epoch": 0.08422988505747127,
      "grad_norm": 0.6510745882987976,
      "learning_rate": 0.00019179090741763298,
      "loss": 1.8959,
      "num_input_tokens_seen": 1357520,
      "step": 229,
      "train_runtime": 228.1892,
      "train_tokens_per_second": 5949.098
    },
    {
      "epoch": 0.08459770114942529,
      "grad_norm": 0.847619891166687,
      "learning_rate": 0.0001917540953432726,
      "loss": 1.9112,
      "num_input_tokens_seen": 1362224,
      "step": 230,
      "train_runtime": 228.9873,
      "train_tokens_per_second": 5948.905
    },
    {
      "epoch": 0.08496551724137931,
      "grad_norm": 0.7419125437736511,
      "learning_rate": 0.0001917172832689122,
      "loss": 2.0395,
      "num_input_tokens_seen": 1368336,
      "step": 231,
      "train_runtime": 229.9886,
      "train_tokens_per_second": 5949.581
    },
    {
      "epoch": 0.08533333333333333,
      "grad_norm": 0.7783828973770142,
      "learning_rate": 0.0001916804711945518,
      "loss": 1.9454,
      "num_input_tokens_seen": 1372864,
      "step": 232,
      "train_runtime": 230.7706,
      "train_tokens_per_second": 5949.041
    },
    {
      "epoch": 0.08570114942528735,
      "grad_norm": 0.7647214531898499,
      "learning_rate": 0.00019164365912019142,
      "loss": 1.8955,
      "num_input_tokens_seen": 1377808,
      "step": 233,
      "train_runtime": 231.5597,
      "train_tokens_per_second": 5950.121
    },
    {
      "epoch": 0.08606896551724139,
      "grad_norm": 0.7165390253067017,
      "learning_rate": 0.00019160684704583105,
      "loss": 1.9889,
      "num_input_tokens_seen": 1384784,
      "step": 234,
      "train_runtime": 232.6514,
      "train_tokens_per_second": 5952.185
    },
    {
      "epoch": 0.0864367816091954,
      "grad_norm": 0.8087878823280334,
      "learning_rate": 0.00019157003497147066,
      "loss": 1.8181,
      "num_input_tokens_seen": 1391072,
      "step": 235,
      "train_runtime": 233.6636,
      "train_tokens_per_second": 5953.31
    },
    {
      "epoch": 0.08680459770114943,
      "grad_norm": 0.8888574838638306,
      "learning_rate": 0.00019153322289711027,
      "loss": 1.9668,
      "num_input_tokens_seen": 1396048,
      "step": 236,
      "train_runtime": 234.508,
      "train_tokens_per_second": 5953.094
    },
    {
      "epoch": 0.08717241379310345,
      "grad_norm": 0.9070819616317749,
      "learning_rate": 0.00019149641082274988,
      "loss": 1.9223,
      "num_input_tokens_seen": 1401600,
      "step": 237,
      "train_runtime": 235.423,
      "train_tokens_per_second": 5953.539
    },
    {
      "epoch": 0.08754022988505747,
      "grad_norm": 0.7615693211555481,
      "learning_rate": 0.0001914595987483895,
      "loss": 2.067,
      "num_input_tokens_seen": 1407728,
      "step": 238,
      "train_runtime": 236.4278,
      "train_tokens_per_second": 5954.155
    },
    {
      "epoch": 0.08790804597701149,
      "grad_norm": 0.7873780131340027,
      "learning_rate": 0.0001914227866740291,
      "loss": 1.9651,
      "num_input_tokens_seen": 1413600,
      "step": 239,
      "train_runtime": 237.4136,
      "train_tokens_per_second": 5954.165
    },
    {
      "epoch": 0.08827586206896551,
      "grad_norm": 0.8089304566383362,
      "learning_rate": 0.0001913859745996687,
      "loss": 1.7522,
      "num_input_tokens_seen": 1419696,
      "step": 240,
      "train_runtime": 238.4135,
      "train_tokens_per_second": 5954.763
    },
    {
      "epoch": 0.08864367816091955,
      "grad_norm": 0.7701553106307983,
      "learning_rate": 0.0001913491625253083,
      "loss": 1.7723,
      "num_input_tokens_seen": 1425312,
      "step": 241,
      "train_runtime": 239.8515,
      "train_tokens_per_second": 5942.476
    },
    {
      "epoch": 0.08901149425287357,
      "grad_norm": 0.626757800579071,
      "learning_rate": 0.00019131235045094792,
      "loss": 1.8497,
      "num_input_tokens_seen": 1435088,
      "step": 242,
      "train_runtime": 241.3315,
      "train_tokens_per_second": 5946.543
    },
    {
      "epoch": 0.08937931034482759,
      "grad_norm": 0.8239079713821411,
      "learning_rate": 0.00019127553837658753,
      "loss": 1.9828,
      "num_input_tokens_seen": 1440080,
      "step": 243,
      "train_runtime": 242.1764,
      "train_tokens_per_second": 5946.408
    },
    {
      "epoch": 0.08974712643678161,
      "grad_norm": 0.7541573643684387,
      "learning_rate": 0.00019123872630222714,
      "loss": 1.9666,
      "num_input_tokens_seen": 1445360,
      "step": 244,
      "train_runtime": 243.0461,
      "train_tokens_per_second": 5946.856
    },
    {
      "epoch": 0.09011494252873563,
      "grad_norm": 0.7623248100280762,
      "learning_rate": 0.00019120191422786675,
      "loss": 2.2626,
      "num_input_tokens_seen": 1451968,
      "step": 245,
      "train_runtime": 244.1103,
      "train_tokens_per_second": 5948.001
    },
    {
      "epoch": 0.09048275862068965,
      "grad_norm": 0.6623736619949341,
      "learning_rate": 0.00019116510215350636,
      "loss": 1.9304,
      "num_input_tokens_seen": 1456608,
      "step": 246,
      "train_runtime": 244.8875,
      "train_tokens_per_second": 5948.071
    },
    {
      "epoch": 0.09085057471264368,
      "grad_norm": 0.7321152687072754,
      "learning_rate": 0.00019112829007914597,
      "loss": 2.1734,
      "num_input_tokens_seen": 1460976,
      "step": 247,
      "train_runtime": 245.6377,
      "train_tokens_per_second": 5947.687
    },
    {
      "epoch": 0.0912183908045977,
      "grad_norm": 0.6610857844352722,
      "learning_rate": 0.00019109147800478558,
      "loss": 2.0194,
      "num_input_tokens_seen": 1466416,
      "step": 248,
      "train_runtime": 246.5502,
      "train_tokens_per_second": 5947.737
    },
    {
      "epoch": 0.09158620689655173,
      "grad_norm": 0.7229866981506348,
      "learning_rate": 0.0001910546659304252,
      "loss": 2.0629,
      "num_input_tokens_seen": 1471728,
      "step": 249,
      "train_runtime": 247.4438,
      "train_tokens_per_second": 5947.726
    },
    {
      "epoch": 0.09195402298850575,
      "grad_norm": 0.6359601020812988,
      "learning_rate": 0.0001910178538560648,
      "loss": 1.8498,
      "num_input_tokens_seen": 1477264,
      "step": 250,
      "train_runtime": 248.366,
      "train_tokens_per_second": 5947.932
    },
    {
      "epoch": 0.09232183908045977,
      "grad_norm": 0.7383832335472107,
      "learning_rate": 0.0001909810417817044,
      "loss": 2.0568,
      "num_input_tokens_seen": 1482000,
      "step": 251,
      "train_runtime": 249.2068,
      "train_tokens_per_second": 5946.868
    },
    {
      "epoch": 0.09268965517241379,
      "grad_norm": 0.7328740358352661,
      "learning_rate": 0.00019094422970734401,
      "loss": 2.0613,
      "num_input_tokens_seen": 1486800,
      "step": 252,
      "train_runtime": 250.0313,
      "train_tokens_per_second": 5946.455
    },
    {
      "epoch": 0.09305747126436782,
      "grad_norm": 0.6441663503646851,
      "learning_rate": 0.00019090741763298362,
      "loss": 1.7576,
      "num_input_tokens_seen": 1492976,
      "step": 253,
      "train_runtime": 251.0282,
      "train_tokens_per_second": 5947.444
    },
    {
      "epoch": 0.09342528735632184,
      "grad_norm": 0.721956193447113,
      "learning_rate": 0.00019087060555862323,
      "loss": 2.0332,
      "num_input_tokens_seen": 1499360,
      "step": 254,
      "train_runtime": 252.0726,
      "train_tokens_per_second": 5948.128
    },
    {
      "epoch": 0.09379310344827586,
      "grad_norm": 0.7011234760284424,
      "learning_rate": 0.00019083379348426284,
      "loss": 1.9336,
      "num_input_tokens_seen": 1504304,
      "step": 255,
      "train_runtime": 252.9219,
      "train_tokens_per_second": 5947.701
    },
    {
      "epoch": 0.09416091954022988,
      "grad_norm": 0.7434554100036621,
      "learning_rate": 0.00019079698140990245,
      "loss": 1.7389,
      "num_input_tokens_seen": 1513584,
      "step": 256,
      "train_runtime": 254.3211,
      "train_tokens_per_second": 5951.468
    },
    {
      "epoch": 0.0945287356321839,
      "grad_norm": 0.7105264663696289,
      "learning_rate": 0.00019076016933554206,
      "loss": 1.8841,
      "num_input_tokens_seen": 1520432,
      "step": 257,
      "train_runtime": 255.4117,
      "train_tokens_per_second": 5952.867
    },
    {
      "epoch": 0.09489655172413793,
      "grad_norm": 0.7543395757675171,
      "learning_rate": 0.00019072335726118167,
      "loss": 1.7332,
      "num_input_tokens_seen": 1525280,
      "step": 258,
      "train_runtime": 256.2414,
      "train_tokens_per_second": 5952.512
    },
    {
      "epoch": 0.09526436781609196,
      "grad_norm": 0.7514438033103943,
      "learning_rate": 0.0001906865451868213,
      "loss": 1.8843,
      "num_input_tokens_seen": 1530400,
      "step": 259,
      "train_runtime": 257.1177,
      "train_tokens_per_second": 5952.138
    },
    {
      "epoch": 0.09563218390804598,
      "grad_norm": 0.7670447826385498,
      "learning_rate": 0.0001906497331124609,
      "loss": 1.9526,
      "num_input_tokens_seen": 1535216,
      "step": 260,
      "train_runtime": 257.936,
      "train_tokens_per_second": 5951.927
    },
    {
      "epoch": 0.096,
      "grad_norm": 0.6435094475746155,
      "learning_rate": 0.00019061292103810052,
      "loss": 1.9275,
      "num_input_tokens_seen": 1545712,
      "step": 261,
      "train_runtime": 259.5147,
      "train_tokens_per_second": 5956.164
    },
    {
      "epoch": 0.09636781609195402,
      "grad_norm": 0.8275080919265747,
      "learning_rate": 0.00019057610896374013,
      "loss": 1.7604,
      "num_input_tokens_seen": 1550496,
      "step": 262,
      "train_runtime": 260.3058,
      "train_tokens_per_second": 5956.439
    },
    {
      "epoch": 0.09673563218390804,
      "grad_norm": 0.7310609221458435,
      "learning_rate": 0.00019053929688937974,
      "loss": 2.0309,
      "num_input_tokens_seen": 1555312,
      "step": 263,
      "train_runtime": 261.1423,
      "train_tokens_per_second": 5955.803
    },
    {
      "epoch": 0.09710344827586206,
      "grad_norm": 0.7967656254768372,
      "learning_rate": 0.00019050248481501932,
      "loss": 1.7773,
      "num_input_tokens_seen": 1561456,
      "step": 264,
      "train_runtime": 262.1296,
      "train_tokens_per_second": 5956.809
    },
    {
      "epoch": 0.09747126436781609,
      "grad_norm": 0.73561030626297,
      "learning_rate": 0.00019046567274065893,
      "loss": 2.0359,
      "num_input_tokens_seen": 1566480,
      "step": 265,
      "train_runtime": 262.96,
      "train_tokens_per_second": 5957.105
    },
    {
      "epoch": 0.09783908045977012,
      "grad_norm": 0.8052525520324707,
      "learning_rate": 0.00019042886066629854,
      "loss": 1.9693,
      "num_input_tokens_seen": 1571760,
      "step": 266,
      "train_runtime": 263.8268,
      "train_tokens_per_second": 5957.546
    },
    {
      "epoch": 0.09820689655172414,
      "grad_norm": 0.7762439846992493,
      "learning_rate": 0.00019039204859193815,
      "loss": 1.9535,
      "num_input_tokens_seen": 1577888,
      "step": 267,
      "train_runtime": 264.8283,
      "train_tokens_per_second": 5958.155
    },
    {
      "epoch": 0.09857471264367816,
      "grad_norm": 0.8418145179748535,
      "learning_rate": 0.00019035523651757778,
      "loss": 1.8523,
      "num_input_tokens_seen": 1582432,
      "step": 268,
      "train_runtime": 265.5949,
      "train_tokens_per_second": 5958.066
    },
    {
      "epoch": 0.09894252873563218,
      "grad_norm": 0.7509084343910217,
      "learning_rate": 0.0001903184244432174,
      "loss": 1.7253,
      "num_input_tokens_seen": 1587728,
      "step": 269,
      "train_runtime": 266.4938,
      "train_tokens_per_second": 5957.843
    },
    {
      "epoch": 0.0993103448275862,
      "grad_norm": 0.7126884460449219,
      "learning_rate": 0.000190281612368857,
      "loss": 1.8486,
      "num_input_tokens_seen": 1594560,
      "step": 270,
      "train_runtime": 267.5912,
      "train_tokens_per_second": 5958.939
    },
    {
      "epoch": 0.09967816091954022,
      "grad_norm": 0.737308919429779,
      "learning_rate": 0.0001902448002944966,
      "loss": 2.015,
      "num_input_tokens_seen": 1599696,
      "step": 271,
      "train_runtime": 268.9544,
      "train_tokens_per_second": 5947.834
    },
    {
      "epoch": 0.10004597701149426,
      "grad_norm": 0.760443389415741,
      "learning_rate": 0.00019020798822013622,
      "loss": 2.2514,
      "num_input_tokens_seen": 1605232,
      "step": 272,
      "train_runtime": 269.8642,
      "train_tokens_per_second": 5948.296
    },
    {
      "epoch": 0.10041379310344828,
      "grad_norm": 0.7480552196502686,
      "learning_rate": 0.00019017117614577583,
      "loss": 1.7752,
      "num_input_tokens_seen": 1610304,
      "step": 273,
      "train_runtime": 270.7357,
      "train_tokens_per_second": 5947.882
    },
    {
      "epoch": 0.1007816091954023,
      "grad_norm": 0.6984736323356628,
      "learning_rate": 0.00019013436407141544,
      "loss": 1.8319,
      "num_input_tokens_seen": 1616640,
      "step": 274,
      "train_runtime": 271.7384,
      "train_tokens_per_second": 5949.252
    },
    {
      "epoch": 0.10114942528735632,
      "grad_norm": 0.8335543870925903,
      "learning_rate": 0.00019009755199705505,
      "loss": 1.9431,
      "num_input_tokens_seen": 1622208,
      "step": 275,
      "train_runtime": 272.6471,
      "train_tokens_per_second": 5949.845
    },
    {
      "epoch": 0.10151724137931034,
      "grad_norm": 0.710389256477356,
      "learning_rate": 0.00019006073992269465,
      "loss": 1.9058,
      "num_input_tokens_seen": 1627584,
      "step": 276,
      "train_runtime": 273.5563,
      "train_tokens_per_second": 5949.723
    },
    {
      "epoch": 0.10188505747126436,
      "grad_norm": 0.6866905689239502,
      "learning_rate": 0.00019002392784833426,
      "loss": 2.0803,
      "num_input_tokens_seen": 1633776,
      "step": 277,
      "train_runtime": 274.5595,
      "train_tokens_per_second": 5950.535
    },
    {
      "epoch": 0.1022528735632184,
      "grad_norm": 0.6699697971343994,
      "learning_rate": 0.00018998711577397387,
      "loss": 1.7394,
      "num_input_tokens_seen": 1642944,
      "step": 278,
      "train_runtime": 275.9506,
      "train_tokens_per_second": 5953.761
    },
    {
      "epoch": 0.10262068965517242,
      "grad_norm": 0.7261507511138916,
      "learning_rate": 0.00018995030369961348,
      "loss": 1.9211,
      "num_input_tokens_seen": 1647952,
      "step": 279,
      "train_runtime": 276.7905,
      "train_tokens_per_second": 5953.789
    },
    {
      "epoch": 0.10298850574712644,
      "grad_norm": 0.6739513278007507,
      "learning_rate": 0.0001899134916252531,
      "loss": 1.9313,
      "num_input_tokens_seen": 1653136,
      "step": 280,
      "train_runtime": 277.6571,
      "train_tokens_per_second": 5953.875
    },
    {
      "epoch": 0.10335632183908046,
      "grad_norm": 0.6748112440109253,
      "learning_rate": 0.0001898766795508927,
      "loss": 1.8026,
      "num_input_tokens_seen": 1662784,
      "step": 281,
      "train_runtime": 279.1249,
      "train_tokens_per_second": 5957.132
    },
    {
      "epoch": 0.10372413793103448,
      "grad_norm": 0.7566770315170288,
      "learning_rate": 0.0001898398674765323,
      "loss": 1.7318,
      "num_input_tokens_seen": 1669056,
      "step": 282,
      "train_runtime": 280.1709,
      "train_tokens_per_second": 5957.279
    },
    {
      "epoch": 0.1040919540229885,
      "grad_norm": 0.6678112745285034,
      "learning_rate": 0.00018980305540217192,
      "loss": 1.8015,
      "num_input_tokens_seen": 1677040,
      "step": 283,
      "train_runtime": 281.4366,
      "train_tokens_per_second": 5958.855
    },
    {
      "epoch": 0.10445977011494254,
      "grad_norm": 0.7495002746582031,
      "learning_rate": 0.00018976624332781153,
      "loss": 2.0167,
      "num_input_tokens_seen": 1683840,
      "step": 284,
      "train_runtime": 282.5108,
      "train_tokens_per_second": 5960.268
    },
    {
      "epoch": 0.10482758620689656,
      "grad_norm": 0.7781804800033569,
      "learning_rate": 0.00018972943125345116,
      "loss": 1.9407,
      "num_input_tokens_seen": 1689840,
      "step": 285,
      "train_runtime": 283.4778,
      "train_tokens_per_second": 5961.101
    },
    {
      "epoch": 0.10519540229885058,
      "grad_norm": 0.605254054069519,
      "learning_rate": 0.00018969261917909077,
      "loss": 1.8986,
      "num_input_tokens_seen": 1697088,
      "step": 286,
      "train_runtime": 284.5868,
      "train_tokens_per_second": 5963.341
    },
    {
      "epoch": 0.1055632183908046,
      "grad_norm": 0.7675965428352356,
      "learning_rate": 0.00018965580710473035,
      "loss": 2.0143,
      "num_input_tokens_seen": 1702080,
      "step": 287,
      "train_runtime": 285.4211,
      "train_tokens_per_second": 5963.398
    },
    {
      "epoch": 0.10593103448275862,
      "grad_norm": 0.7370654940605164,
      "learning_rate": 0.00018961899503036996,
      "loss": 1.8864,
      "num_input_tokens_seen": 1708352,
      "step": 288,
      "train_runtime": 286.4182,
      "train_tokens_per_second": 5964.537
    },
    {
      "epoch": 0.10629885057471264,
      "grad_norm": 0.7976446747779846,
      "learning_rate": 0.00018958218295600957,
      "loss": 1.9853,
      "num_input_tokens_seen": 1718384,
      "step": 289,
      "train_runtime": 287.9519,
      "train_tokens_per_second": 5967.608
    },
    {
      "epoch": 0.10666666666666667,
      "grad_norm": 0.7801894545555115,
      "learning_rate": 0.00018954537088164918,
      "loss": 1.9481,
      "num_input_tokens_seen": 1723824,
      "step": 290,
      "train_runtime": 288.8585,
      "train_tokens_per_second": 5967.712
    },
    {
      "epoch": 0.1070344827586207,
      "grad_norm": 0.8280543684959412,
      "learning_rate": 0.0001895085588072888,
      "loss": 1.8806,
      "num_input_tokens_seen": 1730512,
      "step": 291,
      "train_runtime": 289.9123,
      "train_tokens_per_second": 5969.088
    },
    {
      "epoch": 0.10740229885057472,
      "grad_norm": 0.7578046917915344,
      "learning_rate": 0.0001894717467329284,
      "loss": 1.8189,
      "num_input_tokens_seen": 1739856,
      "step": 292,
      "train_runtime": 291.3373,
      "train_tokens_per_second": 5971.963
    },
    {
      "epoch": 0.10777011494252874,
      "grad_norm": 0.7137064933776855,
      "learning_rate": 0.000189434934658568,
      "loss": 1.8415,
      "num_input_tokens_seen": 1744496,
      "step": 293,
      "train_runtime": 292.1212,
      "train_tokens_per_second": 5971.824
    },
    {
      "epoch": 0.10813793103448276,
      "grad_norm": 0.6923521161079407,
      "learning_rate": 0.00018939812258420764,
      "loss": 1.9076,
      "num_input_tokens_seen": 1749936,
      "step": 294,
      "train_runtime": 293.0484,
      "train_tokens_per_second": 5971.492
    },
    {
      "epoch": 0.10850574712643678,
      "grad_norm": 0.8342931866645813,
      "learning_rate": 0.00018936131050984725,
      "loss": 1.8256,
      "num_input_tokens_seen": 1754480,
      "step": 295,
      "train_runtime": 293.8334,
      "train_tokens_per_second": 5971.002
    },
    {
      "epoch": 0.1088735632183908,
      "grad_norm": 0.7755042910575867,
      "learning_rate": 0.00018932449843548686,
      "loss": 1.8483,
      "num_input_tokens_seen": 1760672,
      "step": 296,
      "train_runtime": 294.8565,
      "train_tokens_per_second": 5971.284
    },
    {
      "epoch": 0.10924137931034483,
      "grad_norm": 0.8064647316932678,
      "learning_rate": 0.00018928768636112647,
      "loss": 1.9537,
      "num_input_tokens_seen": 1766752,
      "step": 297,
      "train_runtime": 295.8338,
      "train_tokens_per_second": 5972.11
    },
    {
      "epoch": 0.10960919540229885,
      "grad_norm": 0.6643753051757812,
      "learning_rate": 0.00018925087428676608,
      "loss": 1.9009,
      "num_input_tokens_seen": 1773728,
      "step": 298,
      "train_runtime": 296.9491,
      "train_tokens_per_second": 5973.171
    },
    {
      "epoch": 0.10997701149425287,
      "grad_norm": 0.7302203178405762,
      "learning_rate": 0.00018921406221240569,
      "loss": 2.1203,
      "num_input_tokens_seen": 1779744,
      "step": 299,
      "train_runtime": 297.9294,
      "train_tokens_per_second": 5973.71
    },
    {
      "epoch": 0.1103448275862069,
      "grad_norm": 0.9114261269569397,
      "learning_rate": 0.00018917725013804527,
      "loss": 1.9024,
      "num_input_tokens_seen": 1786064,
      "step": 300,
      "train_runtime": 298.9549,
      "train_tokens_per_second": 5974.36
    },
    {
      "epoch": 0.11071264367816092,
      "grad_norm": 0.6636778712272644,
      "learning_rate": 0.00018914043806368488,
      "loss": 1.7655,
      "num_input_tokens_seen": 1790288,
      "step": 301,
      "train_runtime": 300.4691,
      "train_tokens_per_second": 5958.31
    },
    {
      "epoch": 0.11108045977011494,
      "grad_norm": 0.7007011771202087,
      "learning_rate": 0.0001891036259893245,
      "loss": 1.8956,
      "num_input_tokens_seen": 1797184,
      "step": 302,
      "train_runtime": 301.5972,
      "train_tokens_per_second": 5958.888
    },
    {
      "epoch": 0.11144827586206897,
      "grad_norm": 0.6551647782325745,
      "learning_rate": 0.00018906681391496412,
      "loss": 1.8594,
      "num_input_tokens_seen": 1802448,
      "step": 303,
      "train_runtime": 302.5017,
      "train_tokens_per_second": 5958.472
    },
    {
      "epoch": 0.11181609195402299,
      "grad_norm": 0.7097108960151672,
      "learning_rate": 0.00018903000184060373,
      "loss": 1.7435,
      "num_input_tokens_seen": 1808064,
      "step": 304,
      "train_runtime": 303.4323,
      "train_tokens_per_second": 5958.706
    },
    {
      "epoch": 0.11218390804597701,
      "grad_norm": 0.7346459627151489,
      "learning_rate": 0.00018899318976624334,
      "loss": 1.8105,
      "num_input_tokens_seen": 1816320,
      "step": 305,
      "train_runtime": 304.6947,
      "train_tokens_per_second": 5961.115
    },
    {
      "epoch": 0.11255172413793103,
      "grad_norm": 0.6427913308143616,
      "learning_rate": 0.00018895637769188295,
      "loss": 1.7099,
      "num_input_tokens_seen": 1827840,
      "step": 306,
      "train_runtime": 306.3925,
      "train_tokens_per_second": 5965.68
    },
    {
      "epoch": 0.11291954022988505,
      "grad_norm": 0.7342320680618286,
      "learning_rate": 0.00018891956561752256,
      "loss": 1.7789,
      "num_input_tokens_seen": 1837168,
      "step": 307,
      "train_runtime": 307.8332,
      "train_tokens_per_second": 5968.062
    },
    {
      "epoch": 0.11328735632183908,
      "grad_norm": 0.6813055872917175,
      "learning_rate": 0.00018888275354316217,
      "loss": 1.774,
      "num_input_tokens_seen": 1846208,
      "step": 308,
      "train_runtime": 309.2113,
      "train_tokens_per_second": 5970.7
    },
    {
      "epoch": 0.11365517241379311,
      "grad_norm": 0.7257830500602722,
      "learning_rate": 0.00018884594146880177,
      "loss": 1.8381,
      "num_input_tokens_seen": 1850944,
      "step": 309,
      "train_runtime": 310.0094,
      "train_tokens_per_second": 5970.605
    },
    {
      "epoch": 0.11402298850574713,
      "grad_norm": 0.7832248210906982,
      "learning_rate": 0.00018880912939444138,
      "loss": 1.7781,
      "num_input_tokens_seen": 1857984,
      "step": 310,
      "train_runtime": 311.1421,
      "train_tokens_per_second": 5971.496
    },
    {
      "epoch": 0.11439080459770115,
      "grad_norm": 0.6972246170043945,
      "learning_rate": 0.000188772317320081,
      "loss": 1.7533,
      "num_input_tokens_seen": 1862880,
      "step": 311,
      "train_runtime": 311.9947,
      "train_tokens_per_second": 5970.871
    },
    {
      "epoch": 0.11475862068965517,
      "grad_norm": 0.7305742502212524,
      "learning_rate": 0.0001887355052457206,
      "loss": 1.6188,
      "num_input_tokens_seen": 1868720,
      "step": 312,
      "train_runtime": 312.9403,
      "train_tokens_per_second": 5971.49
    },
    {
      "epoch": 0.11512643678160919,
      "grad_norm": 0.7020468711853027,
      "learning_rate": 0.0001886986931713602,
      "loss": 1.7287,
      "num_input_tokens_seen": 1878432,
      "step": 313,
      "train_runtime": 314.4184,
      "train_tokens_per_second": 5974.307
    },
    {
      "epoch": 0.11549425287356321,
      "grad_norm": 0.6499274969100952,
      "learning_rate": 0.00018866188109699982,
      "loss": 1.9282,
      "num_input_tokens_seen": 1885712,
      "step": 314,
      "train_runtime": 315.5883,
      "train_tokens_per_second": 5975.228
    },
    {
      "epoch": 0.11586206896551725,
      "grad_norm": 0.6274926662445068,
      "learning_rate": 0.00018862506902263943,
      "loss": 1.8809,
      "num_input_tokens_seen": 1890864,
      "step": 315,
      "train_runtime": 316.4599,
      "train_tokens_per_second": 5975.051
    },
    {
      "epoch": 0.11622988505747127,
      "grad_norm": 0.6690942049026489,
      "learning_rate": 0.00018858825694827904,
      "loss": 1.8814,
      "num_input_tokens_seen": 1897008,
      "step": 316,
      "train_runtime": 317.4438,
      "train_tokens_per_second": 5975.886
    },
    {
      "epoch": 0.11659770114942529,
      "grad_norm": 0.6185339689254761,
      "learning_rate": 0.00018855144487391865,
      "loss": 1.749,
      "num_input_tokens_seen": 1903200,
      "step": 317,
      "train_runtime": 318.458,
      "train_tokens_per_second": 5976.298
    },
    {
      "epoch": 0.11696551724137931,
      "grad_norm": 0.7599213123321533,
      "learning_rate": 0.00018851463279955826,
      "loss": 1.9214,
      "num_input_tokens_seen": 1908896,
      "step": 318,
      "train_runtime": 319.3718,
      "train_tokens_per_second": 5977.034
    },
    {
      "epoch": 0.11733333333333333,
      "grad_norm": 0.748734176158905,
      "learning_rate": 0.0001884778207251979,
      "loss": 1.6331,
      "num_input_tokens_seen": 1915056,
      "step": 319,
      "train_runtime": 320.3784,
      "train_tokens_per_second": 5977.482
    },
    {
      "epoch": 0.11770114942528735,
      "grad_norm": 0.6842741370201111,
      "learning_rate": 0.0001884410086508375,
      "loss": 1.5177,
      "num_input_tokens_seen": 1920400,
      "step": 320,
      "train_runtime": 321.2694,
      "train_tokens_per_second": 5977.538
    },
    {
      "epoch": 0.11806896551724137,
      "grad_norm": 0.8277127742767334,
      "learning_rate": 0.0001884041965764771,
      "loss": 1.9366,
      "num_input_tokens_seen": 1926160,
      "step": 321,
      "train_runtime": 322.2376,
      "train_tokens_per_second": 5977.453
    },
    {
      "epoch": 0.11843678160919541,
      "grad_norm": 0.6963765025138855,
      "learning_rate": 0.00018836738450211672,
      "loss": 1.8562,
      "num_input_tokens_seen": 1931856,
      "step": 322,
      "train_runtime": 323.1614,
      "train_tokens_per_second": 5977.991
    },
    {
      "epoch": 0.11880459770114943,
      "grad_norm": 0.8006249070167542,
      "learning_rate": 0.00018833057242775633,
      "loss": 1.7196,
      "num_input_tokens_seen": 1936448,
      "step": 323,
      "train_runtime": 323.9565,
      "train_tokens_per_second": 5977.493
    },
    {
      "epoch": 0.11917241379310345,
      "grad_norm": 0.7537630796432495,
      "learning_rate": 0.0001882937603533959,
      "loss": 1.9581,
      "num_input_tokens_seen": 1942464,
      "step": 324,
      "train_runtime": 324.9406,
      "train_tokens_per_second": 5977.905
    },
    {
      "epoch": 0.11954022988505747,
      "grad_norm": 0.6968724727630615,
      "learning_rate": 0.00018825694827903552,
      "loss": 1.8721,
      "num_input_tokens_seen": 1948464,
      "step": 325,
      "train_runtime": 325.9105,
      "train_tokens_per_second": 5978.525
    },
    {
      "epoch": 0.11990804597701149,
      "grad_norm": 0.6964346766471863,
      "learning_rate": 0.00018822013620467513,
      "loss": 1.941,
      "num_input_tokens_seen": 1954944,
      "step": 326,
      "train_runtime": 326.9522,
      "train_tokens_per_second": 5979.296
    },
    {
      "epoch": 0.12027586206896551,
      "grad_norm": 0.8223045468330383,
      "learning_rate": 0.00018818332413031474,
      "loss": 2.0361,
      "num_input_tokens_seen": 1962848,
      "step": 327,
      "train_runtime": 328.1729,
      "train_tokens_per_second": 5981.14
    },
    {
      "epoch": 0.12064367816091955,
      "grad_norm": 0.7109692692756653,
      "learning_rate": 0.00018814651205595437,
      "loss": 1.8854,
      "num_input_tokens_seen": 1967952,
      "step": 328,
      "train_runtime": 329.0374,
      "train_tokens_per_second": 5980.937
    },
    {
      "epoch": 0.12101149425287357,
      "grad_norm": 0.7180686593055725,
      "learning_rate": 0.00018810969998159398,
      "loss": 1.8343,
      "num_input_tokens_seen": 1972928,
      "step": 329,
      "train_runtime": 329.8873,
      "train_tokens_per_second": 5980.613
    },
    {
      "epoch": 0.12137931034482759,
      "grad_norm": 0.6951878070831299,
      "learning_rate": 0.0001880728879072336,
      "loss": 1.8489,
      "num_input_tokens_seen": 1978048,
      "step": 330,
      "train_runtime": 330.7614,
      "train_tokens_per_second": 5980.287
    },
    {
      "epoch": 0.12174712643678161,
      "grad_norm": 0.7157859802246094,
      "learning_rate": 0.0001880360758328732,
      "loss": 1.8287,
      "num_input_tokens_seen": 1982960,
      "step": 331,
      "train_runtime": 332.1025,
      "train_tokens_per_second": 5970.928
    },
    {
      "epoch": 0.12211494252873563,
      "grad_norm": 0.8075045347213745,
      "learning_rate": 0.0001879992637585128,
      "loss": 1.8563,
      "num_input_tokens_seen": 1987360,
      "step": 332,
      "train_runtime": 332.8569,
      "train_tokens_per_second": 5970.614
    },
    {
      "epoch": 0.12248275862068965,
      "grad_norm": 0.5464359521865845,
      "learning_rate": 0.00018796245168415242,
      "loss": 1.5142,
      "num_input_tokens_seen": 1999376,
      "step": 333,
      "train_runtime": 335.0336,
      "train_tokens_per_second": 5967.687
    },
    {
      "epoch": 0.12285057471264368,
      "grad_norm": 0.7989075779914856,
      "learning_rate": 0.00018792563960979202,
      "loss": 2.0805,
      "num_input_tokens_seen": 2004368,
      "step": 334,
      "train_runtime": 335.8894,
      "train_tokens_per_second": 5967.346
    },
    {
      "epoch": 0.1232183908045977,
      "grad_norm": 0.7454589009284973,
      "learning_rate": 0.00018788882753543163,
      "loss": 1.8748,
      "num_input_tokens_seen": 2008912,
      "step": 335,
      "train_runtime": 336.6859,
      "train_tokens_per_second": 5966.724
    },
    {
      "epoch": 0.12358620689655173,
      "grad_norm": 0.7054830193519592,
      "learning_rate": 0.00018785201546107124,
      "loss": 1.8777,
      "num_input_tokens_seen": 2014752,
      "step": 336,
      "train_runtime": 337.6222,
      "train_tokens_per_second": 5967.475
    },
    {
      "epoch": 0.12395402298850575,
      "grad_norm": 0.6879221796989441,
      "learning_rate": 0.00018781520338671085,
      "loss": 1.6602,
      "num_input_tokens_seen": 2020816,
      "step": 337,
      "train_runtime": 338.6147,
      "train_tokens_per_second": 5967.893
    },
    {
      "epoch": 0.12432183908045977,
      "grad_norm": 0.6789718866348267,
      "learning_rate": 0.00018777839131235046,
      "loss": 1.8764,
      "num_input_tokens_seen": 2027536,
      "step": 338,
      "train_runtime": 339.6712,
      "train_tokens_per_second": 5969.113
    },
    {
      "epoch": 0.12468965517241379,
      "grad_norm": 0.7514356374740601,
      "learning_rate": 0.00018774157923799007,
      "loss": 1.7888,
      "num_input_tokens_seen": 2034096,
      "step": 339,
      "train_runtime": 340.7086,
      "train_tokens_per_second": 5970.192
    },
    {
      "epoch": 0.1250574712643678,
      "grad_norm": 0.7880290150642395,
      "learning_rate": 0.00018770476716362968,
      "loss": 1.8084,
      "num_input_tokens_seen": 2038784,
      "step": 340,
      "train_runtime": 341.5257,
      "train_tokens_per_second": 5969.636
    },
    {
      "epoch": 0.12542528735632183,
      "grad_norm": 0.6867057085037231,
      "learning_rate": 0.0001876679550892693,
      "loss": 1.7311,
      "num_input_tokens_seen": 2043248,
      "step": 341,
      "train_runtime": 342.3003,
      "train_tokens_per_second": 5969.167
    },
    {
      "epoch": 0.12579310344827585,
      "grad_norm": 0.7414432168006897,
      "learning_rate": 0.0001876311430149089,
      "loss": 1.8801,
      "num_input_tokens_seen": 2052016,
      "step": 342,
      "train_runtime": 343.6221,
      "train_tokens_per_second": 5971.723
    },
    {
      "epoch": 0.1261609195402299,
      "grad_norm": 0.7141637206077576,
      "learning_rate": 0.0001875943309405485,
      "loss": 1.9534,
      "num_input_tokens_seen": 2057392,
      "step": 343,
      "train_runtime": 344.5139,
      "train_tokens_per_second": 5971.87
    },
    {
      "epoch": 0.12652873563218392,
      "grad_norm": 0.7349096536636353,
      "learning_rate": 0.0001875575188661881,
      "loss": 1.9322,
      "num_input_tokens_seen": 2062096,
      "step": 344,
      "train_runtime": 345.3398,
      "train_tokens_per_second": 5971.209
    },
    {
      "epoch": 0.12689655172413794,
      "grad_norm": 0.7744662761688232,
      "learning_rate": 0.00018752070679182775,
      "loss": 2.0181,
      "num_input_tokens_seen": 2067424,
      "step": 345,
      "train_runtime": 346.1893,
      "train_tokens_per_second": 5971.947
    },
    {
      "epoch": 0.12726436781609196,
      "grad_norm": 0.7406309843063354,
      "learning_rate": 0.00018748389471746736,
      "loss": 1.7036,
      "num_input_tokens_seen": 2072560,
      "step": 346,
      "train_runtime": 347.0615,
      "train_tokens_per_second": 5971.738
    },
    {
      "epoch": 0.12763218390804598,
      "grad_norm": 0.6880764365196228,
      "learning_rate": 0.00018744708264310694,
      "loss": 1.8413,
      "num_input_tokens_seen": 2077360,
      "step": 347,
      "train_runtime": 347.8817,
      "train_tokens_per_second": 5971.455
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.7967108488082886,
      "learning_rate": 0.00018741027056874655,
      "loss": 2.0652,
      "num_input_tokens_seen": 2082464,
      "step": 348,
      "train_runtime": 348.7529,
      "train_tokens_per_second": 5971.174
    },
    {
      "epoch": 0.12836781609195402,
      "grad_norm": 0.6809425354003906,
      "learning_rate": 0.00018737345849438616,
      "loss": 2.0862,
      "num_input_tokens_seen": 2087648,
      "step": 349,
      "train_runtime": 349.6336,
      "train_tokens_per_second": 5970.959
    },
    {
      "epoch": 0.12873563218390804,
      "grad_norm": 0.7168233394622803,
      "learning_rate": 0.00018733664642002577,
      "loss": 2.0956,
      "num_input_tokens_seen": 2093328,
      "step": 350,
      "train_runtime": 350.5688,
      "train_tokens_per_second": 5971.233
    },
    {
      "epoch": 0.12910344827586206,
      "grad_norm": 0.7352126836776733,
      "learning_rate": 0.00018729983434566538,
      "loss": 1.6114,
      "num_input_tokens_seen": 2098816,
      "step": 351,
      "train_runtime": 351.4873,
      "train_tokens_per_second": 5971.243
    },
    {
      "epoch": 0.12947126436781609,
      "grad_norm": 0.6451615691184998,
      "learning_rate": 0.00018726302227130498,
      "loss": 1.6803,
      "num_input_tokens_seen": 2106368,
      "step": 352,
      "train_runtime": 352.6819,
      "train_tokens_per_second": 5972.43
    },
    {
      "epoch": 0.1298390804597701,
      "grad_norm": 0.744642972946167,
      "learning_rate": 0.0001872262101969446,
      "loss": 2.0963,
      "num_input_tokens_seen": 2111472,
      "step": 353,
      "train_runtime": 353.5399,
      "train_tokens_per_second": 5972.373
    },
    {
      "epoch": 0.13020689655172413,
      "grad_norm": 0.7254801988601685,
      "learning_rate": 0.00018718939812258423,
      "loss": 2.0037,
      "num_input_tokens_seen": 2117248,
      "step": 354,
      "train_runtime": 354.4882,
      "train_tokens_per_second": 5972.69
    },
    {
      "epoch": 0.13057471264367815,
      "grad_norm": 0.6664026975631714,
      "learning_rate": 0.00018715258604822384,
      "loss": 1.5922,
      "num_input_tokens_seen": 2125056,
      "step": 355,
      "train_runtime": 355.7031,
      "train_tokens_per_second": 5974.241
    },
    {
      "epoch": 0.1309425287356322,
      "grad_norm": 0.7322072386741638,
      "learning_rate": 0.00018711577397386345,
      "loss": 1.9118,
      "num_input_tokens_seen": 2133520,
      "step": 356,
      "train_runtime": 357.0289,
      "train_tokens_per_second": 5975.762
    },
    {
      "epoch": 0.13131034482758622,
      "grad_norm": 0.7217851281166077,
      "learning_rate": 0.00018707896189950306,
      "loss": 1.6786,
      "num_input_tokens_seen": 2140592,
      "step": 357,
      "train_runtime": 358.1497,
      "train_tokens_per_second": 5976.809
    },
    {
      "epoch": 0.13167816091954024,
      "grad_norm": 0.7309592366218567,
      "learning_rate": 0.00018704214982514266,
      "loss": 1.8584,
      "num_input_tokens_seen": 2144640,
      "step": 358,
      "train_runtime": 358.8743,
      "train_tokens_per_second": 5976.02
    },
    {
      "epoch": 0.13204597701149426,
      "grad_norm": 0.6250271797180176,
      "learning_rate": 0.00018700533775078227,
      "loss": 1.75,
      "num_input_tokens_seen": 2151248,
      "step": 359,
      "train_runtime": 359.9582,
      "train_tokens_per_second": 5976.383
    },
    {
      "epoch": 0.13241379310344828,
      "grad_norm": 0.5960398316383362,
      "learning_rate": 0.00018696852567642186,
      "loss": 1.5959,
      "num_input_tokens_seen": 2157760,
      "step": 360,
      "train_runtime": 360.9964,
      "train_tokens_per_second": 5977.234
    },
    {
      "epoch": 0.1327816091954023,
      "grad_norm": 0.709035336971283,
      "learning_rate": 0.00018693171360206146,
      "loss": 1.9184,
      "num_input_tokens_seen": 2164032,
      "step": 361,
      "train_runtime": 362.5187,
      "train_tokens_per_second": 5969.436
    },
    {
      "epoch": 0.13314942528735632,
      "grad_norm": 0.6583025455474854,
      "learning_rate": 0.0001868949015277011,
      "loss": 1.8613,
      "num_input_tokens_seen": 2168768,
      "step": 362,
      "train_runtime": 363.3445,
      "train_tokens_per_second": 5968.903
    },
    {
      "epoch": 0.13351724137931034,
      "grad_norm": 0.7209729552268982,
      "learning_rate": 0.0001868580894533407,
      "loss": 1.9978,
      "num_input_tokens_seen": 2175744,
      "step": 363,
      "train_runtime": 364.4787,
      "train_tokens_per_second": 5969.469
    },
    {
      "epoch": 0.13388505747126436,
      "grad_norm": 0.686311662197113,
      "learning_rate": 0.00018682127737898032,
      "loss": 1.8192,
      "num_input_tokens_seen": 2181200,
      "step": 364,
      "train_runtime": 365.3909,
      "train_tokens_per_second": 5969.498
    },
    {
      "epoch": 0.13425287356321838,
      "grad_norm": 0.7123120427131653,
      "learning_rate": 0.00018678446530461993,
      "loss": 1.8503,
      "num_input_tokens_seen": 2187680,
      "step": 365,
      "train_runtime": 366.4415,
      "train_tokens_per_second": 5970.066
    },
    {
      "epoch": 0.1346206896551724,
      "grad_norm": 0.7308655977249146,
      "learning_rate": 0.00018674765323025954,
      "loss": 1.8202,
      "num_input_tokens_seen": 2192928,
      "step": 366,
      "train_runtime": 367.3112,
      "train_tokens_per_second": 5970.218
    },
    {
      "epoch": 0.13498850574712642,
      "grad_norm": 0.7211245894432068,
      "learning_rate": 0.00018671084115589914,
      "loss": 1.8473,
      "num_input_tokens_seen": 2200880,
      "step": 367,
      "train_runtime": 368.5491,
      "train_tokens_per_second": 5971.742
    },
    {
      "epoch": 0.13535632183908047,
      "grad_norm": 0.7467523217201233,
      "learning_rate": 0.00018667402908153875,
      "loss": 1.8847,
      "num_input_tokens_seen": 2206432,
      "step": 368,
      "train_runtime": 369.4775,
      "train_tokens_per_second": 5971.763
    },
    {
      "epoch": 0.1357241379310345,
      "grad_norm": 0.741328239440918,
      "learning_rate": 0.00018663721700717836,
      "loss": 1.7566,
      "num_input_tokens_seen": 2215552,
      "step": 369,
      "train_runtime": 370.8853,
      "train_tokens_per_second": 5973.685
    },
    {
      "epoch": 0.13609195402298851,
      "grad_norm": 0.693181574344635,
      "learning_rate": 0.00018660040493281797,
      "loss": 1.7464,
      "num_input_tokens_seen": 2221344,
      "step": 370,
      "train_runtime": 371.8289,
      "train_tokens_per_second": 5974.102
    },
    {
      "epoch": 0.13645977011494254,
      "grad_norm": 0.7232048511505127,
      "learning_rate": 0.00018656359285845758,
      "loss": 1.8445,
      "num_input_tokens_seen": 2228256,
      "step": 371,
      "train_runtime": 372.9361,
      "train_tokens_per_second": 5974.9
    },
    {
      "epoch": 0.13682758620689656,
      "grad_norm": 0.936562716960907,
      "learning_rate": 0.0001865267807840972,
      "loss": 2.0189,
      "num_input_tokens_seen": 2233728,
      "step": 372,
      "train_runtime": 373.851,
      "train_tokens_per_second": 5974.915
    },
    {
      "epoch": 0.13719540229885058,
      "grad_norm": 0.7300688624382019,
      "learning_rate": 0.0001864899687097368,
      "loss": 2.0485,
      "num_input_tokens_seen": 2239248,
      "step": 373,
      "train_runtime": 374.7594,
      "train_tokens_per_second": 5975.162
    },
    {
      "epoch": 0.1375632183908046,
      "grad_norm": 0.7056641578674316,
      "learning_rate": 0.0001864531566353764,
      "loss": 1.9858,
      "num_input_tokens_seen": 2244416,
      "step": 374,
      "train_runtime": 375.6114,
      "train_tokens_per_second": 5975.367
    },
    {
      "epoch": 0.13793103448275862,
      "grad_norm": 0.7921895980834961,
      "learning_rate": 0.00018641634456101602,
      "loss": 1.6793,
      "num_input_tokens_seen": 2249888,
      "step": 375,
      "train_runtime": 376.5335,
      "train_tokens_per_second": 5975.267
    },
    {
      "epoch": 0.13829885057471264,
      "grad_norm": 0.6739850044250488,
      "learning_rate": 0.00018637953248665562,
      "loss": 1.7377,
      "num_input_tokens_seen": 2258544,
      "step": 376,
      "train_runtime": 377.8526,
      "train_tokens_per_second": 5977.316
    },
    {
      "epoch": 0.13866666666666666,
      "grad_norm": 0.7935346364974976,
      "learning_rate": 0.00018634272041229523,
      "loss": 1.8909,
      "num_input_tokens_seen": 2264816,
      "step": 377,
      "train_runtime": 378.8602,
      "train_tokens_per_second": 5977.973
    },
    {
      "epoch": 0.13903448275862068,
      "grad_norm": 0.6490564346313477,
      "learning_rate": 0.00018630590833793484,
      "loss": 1.8605,
      "num_input_tokens_seen": 2276096,
      "step": 378,
      "train_runtime": 380.5525,
      "train_tokens_per_second": 5981.031
    },
    {
      "epoch": 0.1394022988505747,
      "grad_norm": 0.8517224788665771,
      "learning_rate": 0.00018626909626357448,
      "loss": 2.0638,
      "num_input_tokens_seen": 2281888,
      "step": 379,
      "train_runtime": 381.4745,
      "train_tokens_per_second": 5981.758
    },
    {
      "epoch": 0.13977011494252872,
      "grad_norm": 0.6885678768157959,
      "learning_rate": 0.0001862322841892141,
      "loss": 1.8711,
      "num_input_tokens_seen": 2289072,
      "step": 380,
      "train_runtime": 382.6107,
      "train_tokens_per_second": 5982.77
    },
    {
      "epoch": 0.14013793103448277,
      "grad_norm": 0.6883024573326111,
      "learning_rate": 0.0001861954721148537,
      "loss": 1.9096,
      "num_input_tokens_seen": 2295104,
      "step": 381,
      "train_runtime": 383.6013,
      "train_tokens_per_second": 5983.046
    },
    {
      "epoch": 0.1405057471264368,
      "grad_norm": 0.7881075739860535,
      "learning_rate": 0.0001861586600404933,
      "loss": 1.8993,
      "num_input_tokens_seen": 2299872,
      "step": 382,
      "train_runtime": 384.3935,
      "train_tokens_per_second": 5983.119
    },
    {
      "epoch": 0.1408735632183908,
      "grad_norm": 0.7175195813179016,
      "learning_rate": 0.0001861218479661329,
      "loss": 1.9201,
      "num_input_tokens_seen": 2304720,
      "step": 383,
      "train_runtime": 385.2356,
      "train_tokens_per_second": 5982.624
    },
    {
      "epoch": 0.14124137931034483,
      "grad_norm": 0.7776727676391602,
      "learning_rate": 0.0001860850358917725,
      "loss": 1.735,
      "num_input_tokens_seen": 2309680,
      "step": 384,
      "train_runtime": 386.0757,
      "train_tokens_per_second": 5982.454
    },
    {
      "epoch": 0.14160919540229885,
      "grad_norm": 0.8024613261222839,
      "learning_rate": 0.0001860482238174121,
      "loss": 1.8909,
      "num_input_tokens_seen": 2315600,
      "step": 385,
      "train_runtime": 387.064,
      "train_tokens_per_second": 5982.473
    },
    {
      "epoch": 0.14197701149425287,
      "grad_norm": 0.7118456363677979,
      "learning_rate": 0.00018601141174305171,
      "loss": 2.0387,
      "num_input_tokens_seen": 2321328,
      "step": 386,
      "train_runtime": 388.0032,
      "train_tokens_per_second": 5982.755
    },
    {
      "epoch": 0.1423448275862069,
      "grad_norm": 0.8724689483642578,
      "learning_rate": 0.00018597459966869132,
      "loss": 2.1292,
      "num_input_tokens_seen": 2326464,
      "step": 387,
      "train_runtime": 388.875,
      "train_tokens_per_second": 5982.549
    },
    {
      "epoch": 0.14271264367816092,
      "grad_norm": 0.8295997381210327,
      "learning_rate": 0.00018593778759433096,
      "loss": 1.9022,
      "num_input_tokens_seen": 2332688,
      "step": 388,
      "train_runtime": 389.9018,
      "train_tokens_per_second": 5982.758
    },
    {
      "epoch": 0.14308045977011494,
      "grad_norm": 0.7838996052742004,
      "learning_rate": 0.00018590097551997057,
      "loss": 1.7828,
      "num_input_tokens_seen": 2338704,
      "step": 389,
      "train_runtime": 390.8694,
      "train_tokens_per_second": 5983.339
    },
    {
      "epoch": 0.14344827586206896,
      "grad_norm": 0.7227248549461365,
      "learning_rate": 0.00018586416344561018,
      "loss": 1.853,
      "num_input_tokens_seen": 2344688,
      "step": 390,
      "train_runtime": 391.858,
      "train_tokens_per_second": 5983.515
    },
    {
      "epoch": 0.14381609195402298,
      "grad_norm": 0.6935778260231018,
      "learning_rate": 0.00018582735137124979,
      "loss": 2.0029,
      "num_input_tokens_seen": 2348976,
      "step": 391,
      "train_runtime": 393.137,
      "train_tokens_per_second": 5974.955
    },
    {
      "epoch": 0.144183908045977,
      "grad_norm": 0.7656466960906982,
      "learning_rate": 0.0001857905392968894,
      "loss": 1.8044,
      "num_input_tokens_seen": 2355600,
      "step": 392,
      "train_runtime": 394.2132,
      "train_tokens_per_second": 5975.446
    },
    {
      "epoch": 0.14455172413793105,
      "grad_norm": 0.6530233025550842,
      "learning_rate": 0.000185753727222529,
      "loss": 1.7168,
      "num_input_tokens_seen": 2360704,
      "step": 393,
      "train_runtime": 395.0927,
      "train_tokens_per_second": 5975.064
    },
    {
      "epoch": 0.14491954022988507,
      "grad_norm": 0.8266283869743347,
      "learning_rate": 0.0001857169151481686,
      "loss": 1.851,
      "num_input_tokens_seen": 2365904,
      "step": 394,
      "train_runtime": 395.962,
      "train_tokens_per_second": 5975.079
    },
    {
      "epoch": 0.1452873563218391,
      "grad_norm": 0.7002015709877014,
      "learning_rate": 0.00018568010307380822,
      "loss": 1.9675,
      "num_input_tokens_seen": 2370992,
      "step": 395,
      "train_runtime": 396.7961,
      "train_tokens_per_second": 5975.34
    },
    {
      "epoch": 0.1456551724137931,
      "grad_norm": 0.6839689612388611,
      "learning_rate": 0.00018564329099944783,
      "loss": 1.8971,
      "num_input_tokens_seen": 2376944,
      "step": 396,
      "train_runtime": 397.7684,
      "train_tokens_per_second": 5975.699
    },
    {
      "epoch": 0.14602298850574713,
      "grad_norm": 0.6042901277542114,
      "learning_rate": 0.00018560647892508744,
      "loss": 1.7526,
      "num_input_tokens_seen": 2389056,
      "step": 397,
      "train_runtime": 399.5638,
      "train_tokens_per_second": 5979.16
    },
    {
      "epoch": 0.14639080459770115,
      "grad_norm": 0.7558072805404663,
      "learning_rate": 0.00018556966685072705,
      "loss": 2.0548,
      "num_input_tokens_seen": 2394864,
      "step": 398,
      "train_runtime": 400.5058,
      "train_tokens_per_second": 5979.599
    },
    {
      "epoch": 0.14675862068965517,
      "grad_norm": 0.9416640996932983,
      "learning_rate": 0.00018553285477636666,
      "loss": 1.8636,
      "num_input_tokens_seen": 2400224,
      "step": 399,
      "train_runtime": 401.397,
      "train_tokens_per_second": 5979.676
    },
    {
      "epoch": 0.1471264367816092,
      "grad_norm": 0.8565811514854431,
      "learning_rate": 0.00018549604270200627,
      "loss": 2.034,
      "num_input_tokens_seen": 2404736,
      "step": 400,
      "train_runtime": 402.2011,
      "train_tokens_per_second": 5978.939
    },
    {
      "epoch": 0.14749425287356321,
      "grad_norm": 0.7060405611991882,
      "learning_rate": 0.00018545923062764587,
      "loss": 2.0218,
      "num_input_tokens_seen": 2412416,
      "step": 401,
      "train_runtime": 403.3895,
      "train_tokens_per_second": 5980.364
    },
    {
      "epoch": 0.14786206896551723,
      "grad_norm": 0.6839686036109924,
      "learning_rate": 0.00018542241855328548,
      "loss": 1.8888,
      "num_input_tokens_seen": 2417072,
      "step": 402,
      "train_runtime": 404.1697,
      "train_tokens_per_second": 5980.339
    },
    {
      "epoch": 0.14822988505747126,
      "grad_norm": 0.6669468879699707,
      "learning_rate": 0.0001853856064789251,
      "loss": 1.5507,
      "num_input_tokens_seen": 2427840,
      "step": 403,
      "train_runtime": 405.8133,
      "train_tokens_per_second": 5982.652
    },
    {
      "epoch": 0.14859770114942528,
      "grad_norm": 0.6962823867797852,
      "learning_rate": 0.0001853487944045647,
      "loss": 1.9593,
      "num_input_tokens_seen": 2434480,
      "step": 404,
      "train_runtime": 406.8635,
      "train_tokens_per_second": 5983.53
    },
    {
      "epoch": 0.1489655172413793,
      "grad_norm": 0.834463357925415,
      "learning_rate": 0.00018531198233020434,
      "loss": 2.0281,
      "num_input_tokens_seen": 2439696,
      "step": 405,
      "train_runtime": 407.7248,
      "train_tokens_per_second": 5983.683
    },
    {
      "epoch": 0.14933333333333335,
      "grad_norm": 0.7330803275108337,
      "learning_rate": 0.00018527517025584395,
      "loss": 1.9421,
      "num_input_tokens_seen": 2447360,
      "step": 406,
      "train_runtime": 408.9428,
      "train_tokens_per_second": 5984.603
    },
    {
      "epoch": 0.14970114942528737,
      "grad_norm": 0.793417751789093,
      "learning_rate": 0.00018523835818148353,
      "loss": 2.0883,
      "num_input_tokens_seen": 2452256,
      "step": 407,
      "train_runtime": 409.7899,
      "train_tokens_per_second": 5984.178
    },
    {
      "epoch": 0.1500689655172414,
      "grad_norm": 0.6819415092468262,
      "learning_rate": 0.00018520154610712314,
      "loss": 1.8594,
      "num_input_tokens_seen": 2457360,
      "step": 408,
      "train_runtime": 410.6425,
      "train_tokens_per_second": 5984.183
    },
    {
      "epoch": 0.1504367816091954,
      "grad_norm": 0.6674189567565918,
      "learning_rate": 0.00018516473403276275,
      "loss": 1.829,
      "num_input_tokens_seen": 2463264,
      "step": 409,
      "train_runtime": 411.6104,
      "train_tokens_per_second": 5984.455
    },
    {
      "epoch": 0.15080459770114943,
      "grad_norm": 0.7914707660675049,
      "learning_rate": 0.00018512792195840235,
      "loss": 1.8782,
      "num_input_tokens_seen": 2468912,
      "step": 410,
      "train_runtime": 412.518,
      "train_tokens_per_second": 5984.98
    },
    {
      "epoch": 0.15117241379310345,
      "grad_norm": 0.7480087876319885,
      "learning_rate": 0.00018509110988404196,
      "loss": 1.8196,
      "num_input_tokens_seen": 2474960,
      "step": 411,
      "train_runtime": 413.5005,
      "train_tokens_per_second": 5985.386
    },
    {
      "epoch": 0.15154022988505747,
      "grad_norm": 0.8891626596450806,
      "learning_rate": 0.00018505429780968157,
      "loss": 1.9407,
      "num_input_tokens_seen": 2480304,
      "step": 412,
      "train_runtime": 414.3766,
      "train_tokens_per_second": 5985.628
    },
    {
      "epoch": 0.1519080459770115,
      "grad_norm": 0.7251510620117188,
      "learning_rate": 0.00018501748573532118,
      "loss": 1.7448,
      "num_input_tokens_seen": 2485568,
      "step": 413,
      "train_runtime": 415.2665,
      "train_tokens_per_second": 5985.477
    },
    {
      "epoch": 0.1522758620689655,
      "grad_norm": 0.7945650219917297,
      "learning_rate": 0.00018498067366096082,
      "loss": 1.9516,
      "num_input_tokens_seen": 2493104,
      "step": 414,
      "train_runtime": 416.4634,
      "train_tokens_per_second": 5986.37
    },
    {
      "epoch": 0.15264367816091953,
      "grad_norm": 0.7043851017951965,
      "learning_rate": 0.00018494386158660043,
      "loss": 1.9806,
      "num_input_tokens_seen": 2498912,
      "step": 415,
      "train_runtime": 417.4214,
      "train_tokens_per_second": 5986.545
    },
    {
      "epoch": 0.15301149425287355,
      "grad_norm": 0.6988221406936646,
      "learning_rate": 0.00018490704951224003,
      "loss": 1.8911,
      "num_input_tokens_seen": 2506976,
      "step": 416,
      "train_runtime": 418.6712,
      "train_tokens_per_second": 5987.935
    },
    {
      "epoch": 0.15337931034482757,
      "grad_norm": 0.617727518081665,
      "learning_rate": 0.00018487023743787964,
      "loss": 1.6414,
      "num_input_tokens_seen": 2513920,
      "step": 417,
      "train_runtime": 419.7861,
      "train_tokens_per_second": 5988.574
    },
    {
      "epoch": 0.15374712643678162,
      "grad_norm": 0.7416211366653442,
      "learning_rate": 0.00018483342536351925,
      "loss": 1.7568,
      "num_input_tokens_seen": 2519328,
      "step": 418,
      "train_runtime": 420.6898,
      "train_tokens_per_second": 5988.565
    },
    {
      "epoch": 0.15411494252873564,
      "grad_norm": 0.7337747812271118,
      "learning_rate": 0.00018479661328915886,
      "loss": 1.8018,
      "num_input_tokens_seen": 2524432,
      "step": 419,
      "train_runtime": 421.5648,
      "train_tokens_per_second": 5988.242
    },
    {
      "epoch": 0.15448275862068966,
      "grad_norm": 0.6360253691673279,
      "learning_rate": 0.00018475980121479844,
      "loss": 1.7843,
      "num_input_tokens_seen": 2537488,
      "step": 420,
      "train_runtime": 423.4871,
      "train_tokens_per_second": 5991.89
    },
    {
      "epoch": 0.15485057471264368,
      "grad_norm": 0.7357736825942993,
      "learning_rate": 0.00018472298914043805,
      "loss": 1.9194,
      "num_input_tokens_seen": 2542048,
      "step": 421,
      "train_runtime": 424.7362,
      "train_tokens_per_second": 5985.004
    },
    {
      "epoch": 0.1552183908045977,
      "grad_norm": 0.793707013130188,
      "learning_rate": 0.0001846861770660777,
      "loss": 1.849,
      "num_input_tokens_seen": 2549024,
      "step": 422,
      "train_runtime": 425.827,
      "train_tokens_per_second": 5986.055
    },
    {
      "epoch": 0.15558620689655173,
      "grad_norm": 0.6722812056541443,
      "learning_rate": 0.0001846493649917173,
      "loss": 1.7081,
      "num_input_tokens_seen": 2553696,
      "step": 423,
      "train_runtime": 426.6429,
      "train_tokens_per_second": 5985.559
    },
    {
      "epoch": 0.15595402298850575,
      "grad_norm": 0.6561516523361206,
      "learning_rate": 0.0001846125529173569,
      "loss": 1.8133,
      "num_input_tokens_seen": 2559408,
      "step": 424,
      "train_runtime": 427.5789,
      "train_tokens_per_second": 5985.815
    },
    {
      "epoch": 0.15632183908045977,
      "grad_norm": 0.7355641722679138,
      "learning_rate": 0.00018457574084299651,
      "loss": 1.783,
      "num_input_tokens_seen": 2565840,
      "step": 425,
      "train_runtime": 428.609,
      "train_tokens_per_second": 5986.435
    },
    {
      "epoch": 0.1566896551724138,
      "grad_norm": 0.8073827028274536,
      "learning_rate": 0.00018453892876863612,
      "loss": 1.9642,
      "num_input_tokens_seen": 2571600,
      "step": 426,
      "train_runtime": 429.5708,
      "train_tokens_per_second": 5986.44
    },
    {
      "epoch": 0.1570574712643678,
      "grad_norm": 0.6649340987205505,
      "learning_rate": 0.00018450211669427573,
      "loss": 1.9262,
      "num_input_tokens_seen": 2576816,
      "step": 427,
      "train_runtime": 430.4384,
      "train_tokens_per_second": 5986.491
    },
    {
      "epoch": 0.15742528735632183,
      "grad_norm": 0.6613366603851318,
      "learning_rate": 0.00018446530461991534,
      "loss": 1.7773,
      "num_input_tokens_seen": 2586304,
      "step": 428,
      "train_runtime": 431.8826,
      "train_tokens_per_second": 5988.442
    },
    {
      "epoch": 0.15779310344827585,
      "grad_norm": 0.7264322638511658,
      "learning_rate": 0.00018442849254555495,
      "loss": 1.7111,
      "num_input_tokens_seen": 2592512,
      "step": 429,
      "train_runtime": 432.9044,
      "train_tokens_per_second": 5988.648
    },
    {
      "epoch": 0.15816091954022987,
      "grad_norm": 0.6878933906555176,
      "learning_rate": 0.00018439168047119456,
      "loss": 1.997,
      "num_input_tokens_seen": 2597856,
      "step": 430,
      "train_runtime": 433.7964,
      "train_tokens_per_second": 5988.653
    },
    {
      "epoch": 0.15852873563218392,
      "grad_norm": 0.7570738792419434,
      "learning_rate": 0.00018435486839683417,
      "loss": 2.0054,
      "num_input_tokens_seen": 2603952,
      "step": 431,
      "train_runtime": 434.7809,
      "train_tokens_per_second": 5989.113
    },
    {
      "epoch": 0.15889655172413794,
      "grad_norm": 0.6946471929550171,
      "learning_rate": 0.00018431805632247378,
      "loss": 1.9308,
      "num_input_tokens_seen": 2610672,
      "step": 432,
      "train_runtime": 435.8553,
      "train_tokens_per_second": 5989.768
    },
    {
      "epoch": 0.15926436781609196,
      "grad_norm": 0.7897321581840515,
      "learning_rate": 0.00018428124424811339,
      "loss": 2.1253,
      "num_input_tokens_seen": 2615520,
      "step": 433,
      "train_runtime": 436.6891,
      "train_tokens_per_second": 5989.433
    },
    {
      "epoch": 0.15963218390804598,
      "grad_norm": 0.6874866485595703,
      "learning_rate": 0.000184244432173753,
      "loss": 1.8993,
      "num_input_tokens_seen": 2622336,
      "step": 434,
      "train_runtime": 437.7709,
      "train_tokens_per_second": 5990.202
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7127911448478699,
      "learning_rate": 0.0001842076200993926,
      "loss": 1.9341,
      "num_input_tokens_seen": 2626576,
      "step": 435,
      "train_runtime": 438.5356,
      "train_tokens_per_second": 5989.424
    },
    {
      "epoch": 0.16036781609195402,
      "grad_norm": 0.7237369418144226,
      "learning_rate": 0.0001841708080250322,
      "loss": 1.8596,
      "num_input_tokens_seen": 2631056,
      "step": 436,
      "train_runtime": 439.3382,
      "train_tokens_per_second": 5988.68
    },
    {
      "epoch": 0.16073563218390804,
      "grad_norm": 0.7314378023147583,
      "learning_rate": 0.00018413399595067182,
      "loss": 1.9974,
      "num_input_tokens_seen": 2636208,
      "step": 437,
      "train_runtime": 440.196,
      "train_tokens_per_second": 5988.714
    },
    {
      "epoch": 0.16110344827586207,
      "grad_norm": 0.6894383430480957,
      "learning_rate": 0.00018409718387631143,
      "loss": 1.7693,
      "num_input_tokens_seen": 2641376,
      "step": 438,
      "train_runtime": 441.0401,
      "train_tokens_per_second": 5988.97
    },
    {
      "epoch": 0.1614712643678161,
      "grad_norm": 0.7083792090415955,
      "learning_rate": 0.00018406037180195107,
      "loss": 1.7195,
      "num_input_tokens_seen": 2646512,
      "step": 439,
      "train_runtime": 441.9162,
      "train_tokens_per_second": 5988.719
    },
    {
      "epoch": 0.1618390804597701,
      "grad_norm": 0.7953554391860962,
      "learning_rate": 0.00018402355972759068,
      "loss": 1.851,
      "num_input_tokens_seen": 2653584,
      "step": 440,
      "train_runtime": 443.05,
      "train_tokens_per_second": 5989.355
    },
    {
      "epoch": 0.16220689655172413,
      "grad_norm": 0.7280542254447937,
      "learning_rate": 0.00018398674765323028,
      "loss": 1.7869,
      "num_input_tokens_seen": 2660112,
      "step": 441,
      "train_runtime": 444.088,
      "train_tokens_per_second": 5990.056
    },
    {
      "epoch": 0.16257471264367815,
      "grad_norm": 0.7719009518623352,
      "learning_rate": 0.0001839499355788699,
      "loss": 1.8121,
      "num_input_tokens_seen": 2666112,
      "step": 442,
      "train_runtime": 445.0742,
      "train_tokens_per_second": 5990.264
    },
    {
      "epoch": 0.1629425287356322,
      "grad_norm": 0.7539399266242981,
      "learning_rate": 0.00018391312350450947,
      "loss": 1.6833,
      "num_input_tokens_seen": 2671152,
      "step": 443,
      "train_runtime": 445.9296,
      "train_tokens_per_second": 5990.075
    },
    {
      "epoch": 0.16331034482758622,
      "grad_norm": 0.7633247375488281,
      "learning_rate": 0.00018387631143014908,
      "loss": 1.8458,
      "num_input_tokens_seen": 2676720,
      "step": 444,
      "train_runtime": 446.8475,
      "train_tokens_per_second": 5990.232
    },
    {
      "epoch": 0.16367816091954024,
      "grad_norm": 0.8550528287887573,
      "learning_rate": 0.0001838394993557887,
      "loss": 2.0313,
      "num_input_tokens_seen": 2681552,
      "step": 445,
      "train_runtime": 447.6588,
      "train_tokens_per_second": 5990.17
    },
    {
      "epoch": 0.16404597701149426,
      "grad_norm": 0.9032266139984131,
      "learning_rate": 0.0001838026872814283,
      "loss": 1.9196,
      "num_input_tokens_seen": 2686128,
      "step": 446,
      "train_runtime": 448.4734,
      "train_tokens_per_second": 5989.492
    },
    {
      "epoch": 0.16441379310344828,
      "grad_norm": 0.7678905129432678,
      "learning_rate": 0.0001837658752070679,
      "loss": 1.8884,
      "num_input_tokens_seen": 2690288,
      "step": 447,
      "train_runtime": 449.1985,
      "train_tokens_per_second": 5989.085
    },
    {
      "epoch": 0.1647816091954023,
      "grad_norm": 0.6993767619132996,
      "learning_rate": 0.00018372906313270755,
      "loss": 1.8937,
      "num_input_tokens_seen": 2695008,
      "step": 448,
      "train_runtime": 450.0075,
      "train_tokens_per_second": 5988.807
    },
    {
      "epoch": 0.16514942528735632,
      "grad_norm": 0.6627291440963745,
      "learning_rate": 0.00018369225105834716,
      "loss": 1.9193,
      "num_input_tokens_seen": 2702896,
      "step": 449,
      "train_runtime": 451.234,
      "train_tokens_per_second": 5990.009
    },
    {
      "epoch": 0.16551724137931034,
      "grad_norm": 0.7582489848136902,
      "learning_rate": 0.00018365543898398676,
      "loss": 1.9296,
      "num_input_tokens_seen": 2708224,
      "step": 450,
      "train_runtime": 452.1299,
      "train_tokens_per_second": 5989.925
    },
    {
      "epoch": 0.16588505747126436,
      "grad_norm": 0.6843869686126709,
      "learning_rate": 0.00018361862690962637,
      "loss": 1.717,
      "num_input_tokens_seen": 2717376,
      "step": 451,
      "train_runtime": 453.9767,
      "train_tokens_per_second": 5985.717
    },
    {
      "epoch": 0.16625287356321838,
      "grad_norm": 0.7444376349449158,
      "learning_rate": 0.00018358181483526598,
      "loss": 2.0068,
      "num_input_tokens_seen": 2722432,
      "step": 452,
      "train_runtime": 454.8415,
      "train_tokens_per_second": 5985.452
    },
    {
      "epoch": 0.1666206896551724,
      "grad_norm": 0.6823301315307617,
      "learning_rate": 0.0001835450027609056,
      "loss": 1.9761,
      "num_input_tokens_seen": 2728016,
      "step": 453,
      "train_runtime": 455.7643,
      "train_tokens_per_second": 5985.585
    },
    {
      "epoch": 0.16698850574712643,
      "grad_norm": 0.7306808829307556,
      "learning_rate": 0.0001835081906865452,
      "loss": 1.8948,
      "num_input_tokens_seen": 2734208,
      "step": 454,
      "train_runtime": 456.769,
      "train_tokens_per_second": 5985.976
    },
    {
      "epoch": 0.16735632183908045,
      "grad_norm": 0.6250959634780884,
      "learning_rate": 0.0001834713786121848,
      "loss": 1.8936,
      "num_input_tokens_seen": 2743024,
      "step": 455,
      "train_runtime": 458.1282,
      "train_tokens_per_second": 5987.459
    },
    {
      "epoch": 0.1677241379310345,
      "grad_norm": 0.7861682772636414,
      "learning_rate": 0.00018343456653782442,
      "loss": 1.8531,
      "num_input_tokens_seen": 2747328,
      "step": 456,
      "train_runtime": 458.9087,
      "train_tokens_per_second": 5986.655
    },
    {
      "epoch": 0.16809195402298852,
      "grad_norm": 0.6755948662757874,
      "learning_rate": 0.00018339775446346403,
      "loss": 1.8501,
      "num_input_tokens_seen": 2754192,
      "step": 457,
      "train_runtime": 460.0253,
      "train_tokens_per_second": 5987.044
    },
    {
      "epoch": 0.16845977011494254,
      "grad_norm": 0.642189621925354,
      "learning_rate": 0.00018336094238910364,
      "loss": 1.7918,
      "num_input_tokens_seen": 2759072,
      "step": 458,
      "train_runtime": 460.8482,
      "train_tokens_per_second": 5986.944
    },
    {
      "epoch": 0.16882758620689656,
      "grad_norm": 0.7578986287117004,
      "learning_rate": 0.00018332413031474324,
      "loss": 1.9248,
      "num_input_tokens_seen": 2765824,
      "step": 459,
      "train_runtime": 461.9587,
      "train_tokens_per_second": 5987.168
    },
    {
      "epoch": 0.16919540229885058,
      "grad_norm": 0.662889838218689,
      "learning_rate": 0.00018328731824038285,
      "loss": 1.8313,
      "num_input_tokens_seen": 2771872,
      "step": 460,
      "train_runtime": 462.9587,
      "train_tokens_per_second": 5987.299
    },
    {
      "epoch": 0.1695632183908046,
      "grad_norm": 0.77558434009552,
      "learning_rate": 0.00018325050616602246,
      "loss": 1.8102,
      "num_input_tokens_seen": 2777584,
      "step": 461,
      "train_runtime": 463.8951,
      "train_tokens_per_second": 5987.527
    },
    {
      "epoch": 0.16993103448275862,
      "grad_norm": 0.709566056728363,
      "learning_rate": 0.00018321369409166207,
      "loss": 1.8505,
      "num_input_tokens_seen": 2784224,
      "step": 462,
      "train_runtime": 464.9616,
      "train_tokens_per_second": 5988.074
    },
    {
      "epoch": 0.17029885057471264,
      "grad_norm": 0.6982594728469849,
      "learning_rate": 0.00018317688201730168,
      "loss": 1.9501,
      "num_input_tokens_seen": 2788768,
      "step": 463,
      "train_runtime": 465.7452,
      "train_tokens_per_second": 5987.754
    },
    {
      "epoch": 0.17066666666666666,
      "grad_norm": 0.7560883164405823,
      "learning_rate": 0.0001831400699429413,
      "loss": 1.7646,
      "num_input_tokens_seen": 2796416,
      "step": 464,
      "train_runtime": 466.9698,
      "train_tokens_per_second": 5988.43
    },
    {
      "epoch": 0.17103448275862068,
      "grad_norm": 0.7397001385688782,
      "learning_rate": 0.00018310325786858092,
      "loss": 1.7924,
      "num_input_tokens_seen": 2801376,
      "step": 465,
      "train_runtime": 467.8232,
      "train_tokens_per_second": 5988.108
    },
    {
      "epoch": 0.1714022988505747,
      "grad_norm": 0.7177740335464478,
      "learning_rate": 0.00018306644579422053,
      "loss": 1.861,
      "num_input_tokens_seen": 2808784,
      "step": 466,
      "train_runtime": 468.9776,
      "train_tokens_per_second": 5989.165
    },
    {
      "epoch": 0.17177011494252872,
      "grad_norm": 0.7760906219482422,
      "learning_rate": 0.00018302963371986012,
      "loss": 2.0148,
      "num_input_tokens_seen": 2814064,
      "step": 467,
      "train_runtime": 469.8607,
      "train_tokens_per_second": 5989.145
    },
    {
      "epoch": 0.17213793103448277,
      "grad_norm": 0.7281026840209961,
      "learning_rate": 0.00018299282164549972,
      "loss": 1.9006,
      "num_input_tokens_seen": 2819952,
      "step": 468,
      "train_runtime": 470.8276,
      "train_tokens_per_second": 5989.351
    },
    {
      "epoch": 0.1725057471264368,
      "grad_norm": 0.7086067795753479,
      "learning_rate": 0.00018295600957113933,
      "loss": 1.788,
      "num_input_tokens_seen": 2825024,
      "step": 469,
      "train_runtime": 471.675,
      "train_tokens_per_second": 5989.344
    },
    {
      "epoch": 0.1728735632183908,
      "grad_norm": 0.6965607404708862,
      "learning_rate": 0.00018291919749677894,
      "loss": 1.6768,
      "num_input_tokens_seen": 2831344,
      "step": 470,
      "train_runtime": 472.7036,
      "train_tokens_per_second": 5989.682
    },
    {
      "epoch": 0.17324137931034483,
      "grad_norm": 0.7752671837806702,
      "learning_rate": 0.00018288238542241855,
      "loss": 2.0327,
      "num_input_tokens_seen": 2835632,
      "step": 471,
      "train_runtime": 473.4686,
      "train_tokens_per_second": 5989.06
    },
    {
      "epoch": 0.17360919540229885,
      "grad_norm": 0.7533464431762695,
      "learning_rate": 0.00018284557334805816,
      "loss": 1.8942,
      "num_input_tokens_seen": 2842752,
      "step": 472,
      "train_runtime": 474.5932,
      "train_tokens_per_second": 5989.871
    },
    {
      "epoch": 0.17397701149425288,
      "grad_norm": 0.7007566690444946,
      "learning_rate": 0.00018280876127369777,
      "loss": 1.9851,
      "num_input_tokens_seen": 2848624,
      "step": 473,
      "train_runtime": 475.5724,
      "train_tokens_per_second": 5989.885
    },
    {
      "epoch": 0.1743448275862069,
      "grad_norm": 0.6253156065940857,
      "learning_rate": 0.0001827719491993374,
      "loss": 2.0184,
      "num_input_tokens_seen": 2856752,
      "step": 474,
      "train_runtime": 476.8209,
      "train_tokens_per_second": 5991.248
    },
    {
      "epoch": 0.17471264367816092,
      "grad_norm": 0.7340561747550964,
      "learning_rate": 0.00018273513712497701,
      "loss": 2.0343,
      "num_input_tokens_seen": 2864240,
      "step": 475,
      "train_runtime": 477.9774,
      "train_tokens_per_second": 5992.417
    },
    {
      "epoch": 0.17508045977011494,
      "grad_norm": 0.7108561992645264,
      "learning_rate": 0.00018269832505061662,
      "loss": 1.8435,
      "num_input_tokens_seen": 2872704,
      "step": 476,
      "train_runtime": 479.3059,
      "train_tokens_per_second": 5993.467
    },
    {
      "epoch": 0.17544827586206896,
      "grad_norm": 0.7289217114448547,
      "learning_rate": 0.00018266151297625623,
      "loss": 1.9543,
      "num_input_tokens_seen": 2877712,
      "step": 477,
      "train_runtime": 480.1488,
      "train_tokens_per_second": 5993.375
    },
    {
      "epoch": 0.17581609195402298,
      "grad_norm": 0.7695021033287048,
      "learning_rate": 0.00018262470090189584,
      "loss": 2.0223,
      "num_input_tokens_seen": 2882560,
      "step": 478,
      "train_runtime": 480.9781,
      "train_tokens_per_second": 5993.121
    },
    {
      "epoch": 0.176183908045977,
      "grad_norm": 0.7470912933349609,
      "learning_rate": 0.00018258788882753545,
      "loss": 1.9365,
      "num_input_tokens_seen": 2887808,
      "step": 479,
      "train_runtime": 481.8695,
      "train_tokens_per_second": 5992.925
    },
    {
      "epoch": 0.17655172413793102,
      "grad_norm": 0.6362117528915405,
      "learning_rate": 0.00018255107675317503,
      "loss": 1.9628,
      "num_input_tokens_seen": 2893648,
      "step": 480,
      "train_runtime": 482.8393,
      "train_tokens_per_second": 5992.984
    },
    {
      "epoch": 0.17691954022988507,
      "grad_norm": 0.7495032548904419,
      "learning_rate": 0.00018251426467881464,
      "loss": 1.8733,
      "num_input_tokens_seen": 2898912,
      "step": 481,
      "train_runtime": 484.1694,
      "train_tokens_per_second": 5987.392
    },
    {
      "epoch": 0.1772873563218391,
      "grad_norm": 0.7536237239837646,
      "learning_rate": 0.00018247745260445428,
      "loss": 1.8017,
      "num_input_tokens_seen": 2905424,
      "step": 482,
      "train_runtime": 485.2276,
      "train_tokens_per_second": 5987.755
    },
    {
      "epoch": 0.1776551724137931,
      "grad_norm": 0.756716251373291,
      "learning_rate": 0.00018244064053009388,
      "loss": 1.7773,
      "num_input_tokens_seen": 2910784,
      "step": 483,
      "train_runtime": 486.1397,
      "train_tokens_per_second": 5987.546
    },
    {
      "epoch": 0.17802298850574713,
      "grad_norm": 0.67543625831604,
      "learning_rate": 0.0001824038284557335,
      "loss": 1.8956,
      "num_input_tokens_seen": 2918240,
      "step": 484,
      "train_runtime": 487.3526,
      "train_tokens_per_second": 5987.944
    },
    {
      "epoch": 0.17839080459770115,
      "grad_norm": 0.7353869676589966,
      "learning_rate": 0.0001823670163813731,
      "loss": 2.063,
      "num_input_tokens_seen": 2923216,
      "step": 485,
      "train_runtime": 488.1997,
      "train_tokens_per_second": 5987.747
    },
    {
      "epoch": 0.17875862068965517,
      "grad_norm": 0.7567138075828552,
      "learning_rate": 0.0001823302043070127,
      "loss": 1.8763,
      "num_input_tokens_seen": 2928080,
      "step": 486,
      "train_runtime": 489.0459,
      "train_tokens_per_second": 5987.332
    },
    {
      "epoch": 0.1791264367816092,
      "grad_norm": 0.6213117837905884,
      "learning_rate": 0.00018229339223265232,
      "loss": 1.8701,
      "num_input_tokens_seen": 2934672,
      "step": 487,
      "train_runtime": 490.1448,
      "train_tokens_per_second": 5987.357
    },
    {
      "epoch": 0.17949425287356321,
      "grad_norm": 0.6765648126602173,
      "learning_rate": 0.00018225658015829193,
      "loss": 1.775,
      "num_input_tokens_seen": 2939760,
      "step": 488,
      "train_runtime": 491.0053,
      "train_tokens_per_second": 5987.226
    },
    {
      "epoch": 0.17986206896551724,
      "grad_norm": 0.6079659461975098,
      "learning_rate": 0.00018221976808393154,
      "loss": 1.7171,
      "num_input_tokens_seen": 2946608,
      "step": 489,
      "train_runtime": 492.0979,
      "train_tokens_per_second": 5987.85
    },
    {
      "epoch": 0.18022988505747126,
      "grad_norm": 0.8331005573272705,
      "learning_rate": 0.00018218295600957115,
      "loss": 2.0085,
      "num_input_tokens_seen": 2951200,
      "step": 490,
      "train_runtime": 492.882,
      "train_tokens_per_second": 5987.64
    },
    {
      "epoch": 0.18059770114942528,
      "grad_norm": 0.7557847499847412,
      "learning_rate": 0.00018214614393521076,
      "loss": 1.8781,
      "num_input_tokens_seen": 2958240,
      "step": 491,
      "train_runtime": 493.9884,
      "train_tokens_per_second": 5988.48
    },
    {
      "epoch": 0.1809655172413793,
      "grad_norm": 0.7792819142341614,
      "learning_rate": 0.00018210933186085036,
      "loss": 1.9078,
      "num_input_tokens_seen": 2963456,
      "step": 492,
      "train_runtime": 494.8557,
      "train_tokens_per_second": 5988.525
    },
    {
      "epoch": 0.18133333333333335,
      "grad_norm": 0.707869291305542,
      "learning_rate": 0.00018207251978648997,
      "loss": 1.6466,
      "num_input_tokens_seen": 2969584,
      "step": 493,
      "train_runtime": 495.8812,
      "train_tokens_per_second": 5988.499
    },
    {
      "epoch": 0.18170114942528737,
      "grad_norm": 0.6605796813964844,
      "learning_rate": 0.00018203570771212958,
      "loss": 1.901,
      "num_input_tokens_seen": 2976560,
      "step": 494,
      "train_runtime": 496.9954,
      "train_tokens_per_second": 5989.11
    },
    {
      "epoch": 0.1820689655172414,
      "grad_norm": 0.7089316844940186,
      "learning_rate": 0.0001819988956377692,
      "loss": 1.8349,
      "num_input_tokens_seen": 2982864,
      "step": 495,
      "train_runtime": 498.0108,
      "train_tokens_per_second": 5989.556
    },
    {
      "epoch": 0.1824367816091954,
      "grad_norm": 0.7388765215873718,
      "learning_rate": 0.0001819620835634088,
      "loss": 1.8852,
      "num_input_tokens_seen": 2988384,
      "step": 496,
      "train_runtime": 498.9228,
      "train_tokens_per_second": 5989.672
    },
    {
      "epoch": 0.18280459770114943,
      "grad_norm": 0.719870924949646,
      "learning_rate": 0.0001819252714890484,
      "loss": 1.8616,
      "num_input_tokens_seen": 2995088,
      "step": 497,
      "train_runtime": 499.9974,
      "train_tokens_per_second": 5990.207
    },
    {
      "epoch": 0.18317241379310345,
      "grad_norm": 0.7674807906150818,
      "learning_rate": 0.00018188845941468802,
      "loss": 1.9816,
      "num_input_tokens_seen": 3000768,
      "step": 498,
      "train_runtime": 500.9315,
      "train_tokens_per_second": 5990.375
    },
    {
      "epoch": 0.18354022988505747,
      "grad_norm": 0.7047696113586426,
      "learning_rate": 0.00018185164734032765,
      "loss": 1.9084,
      "num_input_tokens_seen": 3005760,
      "step": 499,
      "train_runtime": 501.7951,
      "train_tokens_per_second": 5990.015
    },
    {
      "epoch": 0.1839080459770115,
      "grad_norm": 0.7106406688690186,
      "learning_rate": 0.00018181483526596726,
      "loss": 1.7949,
      "num_input_tokens_seen": 3010752,
      "step": 500,
      "train_runtime": 502.6392,
      "train_tokens_per_second": 5989.887
    },
    {
      "epoch": 0.1842758620689655,
      "grad_norm": 0.7400830388069153,
      "learning_rate": 0.00018177802319160687,
      "loss": 1.8798,
      "num_input_tokens_seen": 3015744,
      "step": 501,
      "train_runtime": 503.4964,
      "train_tokens_per_second": 5989.604
    },
    {
      "epoch": 0.18464367816091953,
      "grad_norm": 0.7215652465820312,
      "learning_rate": 0.00018174121111724648,
      "loss": 1.7335,
      "num_input_tokens_seen": 3021968,
      "step": 502,
      "train_runtime": 504.5099,
      "train_tokens_per_second": 5989.908
    },
    {
      "epoch": 0.18501149425287355,
      "grad_norm": 0.6986531615257263,
      "learning_rate": 0.00018170439904288606,
      "loss": 1.797,
      "num_input_tokens_seen": 3027664,
      "step": 503,
      "train_runtime": 505.4588,
      "train_tokens_per_second": 5989.932
    },
    {
      "epoch": 0.18537931034482757,
      "grad_norm": 0.7286319732666016,
      "learning_rate": 0.00018166758696852567,
      "loss": 1.8748,
      "num_input_tokens_seen": 3033808,
      "step": 504,
      "train_runtime": 506.4403,
      "train_tokens_per_second": 5990.455
    },
    {
      "epoch": 0.1857471264367816,
      "grad_norm": 0.8388171195983887,
      "learning_rate": 0.00018163077489416528,
      "loss": 1.9862,
      "num_input_tokens_seen": 3038304,
      "step": 505,
      "train_runtime": 507.2061,
      "train_tokens_per_second": 5990.275
    },
    {
      "epoch": 0.18611494252873564,
      "grad_norm": 0.7551229000091553,
      "learning_rate": 0.0001815939628198049,
      "loss": 1.691,
      "num_input_tokens_seen": 3043040,
      "step": 506,
      "train_runtime": 508.0147,
      "train_tokens_per_second": 5990.063
    },
    {
      "epoch": 0.18648275862068966,
      "grad_norm": 0.736970067024231,
      "learning_rate": 0.0001815571507454445,
      "loss": 1.7491,
      "num_input_tokens_seen": 3048480,
      "step": 507,
      "train_runtime": 508.9133,
      "train_tokens_per_second": 5990.175
    },
    {
      "epoch": 0.18685057471264369,
      "grad_norm": 0.7858811020851135,
      "learning_rate": 0.00018152033867108413,
      "loss": 1.8674,
      "num_input_tokens_seen": 3052992,
      "step": 508,
      "train_runtime": 509.718,
      "train_tokens_per_second": 5989.57
    },
    {
      "epoch": 0.1872183908045977,
      "grad_norm": 0.7412512898445129,
      "learning_rate": 0.00018148352659672374,
      "loss": 1.8044,
      "num_input_tokens_seen": 3060416,
      "step": 509,
      "train_runtime": 510.9081,
      "train_tokens_per_second": 5990.15
    },
    {
      "epoch": 0.18758620689655173,
      "grad_norm": 0.7265469431877136,
      "learning_rate": 0.00018144671452236335,
      "loss": 1.8035,
      "num_input_tokens_seen": 3066096,
      "step": 510,
      "train_runtime": 511.8326,
      "train_tokens_per_second": 5990.427
    },
    {
      "epoch": 0.18795402298850575,
      "grad_norm": 0.8583095669746399,
      "learning_rate": 0.00018140990244800296,
      "loss": 1.9065,
      "num_input_tokens_seen": 3071008,
      "step": 511,
      "train_runtime": 513.1792,
      "train_tokens_per_second": 5984.28
    },
    {
      "epoch": 0.18832183908045977,
      "grad_norm": 0.7504838109016418,
      "learning_rate": 0.00018137309037364257,
      "loss": 1.8447,
      "num_input_tokens_seen": 3075664,
      "step": 512,
      "train_runtime": 513.9811,
      "train_tokens_per_second": 5984.002
    },
    {
      "epoch": 0.1886896551724138,
      "grad_norm": 0.8324645161628723,
      "learning_rate": 0.00018133627829928218,
      "loss": 1.8945,
      "num_input_tokens_seen": 3082672,
      "step": 513,
      "train_runtime": 515.13,
      "train_tokens_per_second": 5984.261
    },
    {
      "epoch": 0.1890574712643678,
      "grad_norm": 0.7464407086372375,
      "learning_rate": 0.0001812994662249218,
      "loss": 1.7967,
      "num_input_tokens_seen": 3087760,
      "step": 514,
      "train_runtime": 515.9755,
      "train_tokens_per_second": 5984.315
    },
    {
      "epoch": 0.18942528735632183,
      "grad_norm": 0.7011414170265198,
      "learning_rate": 0.0001812626541505614,
      "loss": 1.7975,
      "num_input_tokens_seen": 3093904,
      "step": 515,
      "train_runtime": 517.0043,
      "train_tokens_per_second": 5984.291
    },
    {
      "epoch": 0.18979310344827585,
      "grad_norm": 0.7123543620109558,
      "learning_rate": 0.000181225842076201,
      "loss": 1.8877,
      "num_input_tokens_seen": 3098864,
      "step": 516,
      "train_runtime": 517.8234,
      "train_tokens_per_second": 5984.404
    },
    {
      "epoch": 0.19016091954022987,
      "grad_norm": 0.7099828720092773,
      "learning_rate": 0.00018118903000184061,
      "loss": 1.738,
      "num_input_tokens_seen": 3103376,
      "step": 517,
      "train_runtime": 518.628,
      "train_tokens_per_second": 5983.819
    },
    {
      "epoch": 0.19052873563218392,
      "grad_norm": 0.6124518513679504,
      "learning_rate": 0.00018115221792748022,
      "loss": 1.7652,
      "num_input_tokens_seen": 3111616,
      "step": 518,
      "train_runtime": 519.9095,
      "train_tokens_per_second": 5984.919
    },
    {
      "epoch": 0.19089655172413794,
      "grad_norm": 0.7913989424705505,
      "learning_rate": 0.00018111540585311983,
      "loss": 1.8922,
      "num_input_tokens_seen": 3117072,
      "step": 519,
      "train_runtime": 520.8267,
      "train_tokens_per_second": 5984.854
    },
    {
      "epoch": 0.19126436781609196,
      "grad_norm": 0.7353897094726562,
      "learning_rate": 0.00018107859377875944,
      "loss": 1.8105,
      "num_input_tokens_seen": 3125648,
      "step": 520,
      "train_runtime": 522.1645,
      "train_tokens_per_second": 5985.945
    },
    {
      "epoch": 0.19163218390804598,
      "grad_norm": 0.7589907646179199,
      "learning_rate": 0.00018104178170439905,
      "loss": 1.8123,
      "num_input_tokens_seen": 3130208,
      "step": 521,
      "train_runtime": 522.9583,
      "train_tokens_per_second": 5985.579
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.7847319841384888,
      "learning_rate": 0.00018100496963003866,
      "loss": 1.9434,
      "num_input_tokens_seen": 3133936,
      "step": 522,
      "train_runtime": 523.6086,
      "train_tokens_per_second": 5985.265
    },
    {
      "epoch": 0.19236781609195402,
      "grad_norm": 0.7331429719924927,
      "learning_rate": 0.00018096815755567827,
      "loss": 1.8921,
      "num_input_tokens_seen": 3139952,
      "step": 523,
      "train_runtime": 524.6068,
      "train_tokens_per_second": 5985.344
    },
    {
      "epoch": 0.19273563218390805,
      "grad_norm": 0.717780590057373,
      "learning_rate": 0.00018093134548131788,
      "loss": 1.9207,
      "num_input_tokens_seen": 3146464,
      "step": 524,
      "train_runtime": 525.6592,
      "train_tokens_per_second": 5985.749
    },
    {
      "epoch": 0.19310344827586207,
      "grad_norm": 0.7325490117073059,
      "learning_rate": 0.0001808945334069575,
      "loss": 2.1358,
      "num_input_tokens_seen": 3151056,
      "step": 525,
      "train_runtime": 526.4328,
      "train_tokens_per_second": 5985.676
    },
    {
      "epoch": 0.1934712643678161,
      "grad_norm": 0.7939004302024841,
      "learning_rate": 0.0001808577213325971,
      "loss": 1.9071,
      "num_input_tokens_seen": 3155664,
      "step": 526,
      "train_runtime": 527.2572,
      "train_tokens_per_second": 5985.056
    },
    {
      "epoch": 0.1938390804597701,
      "grad_norm": 0.7526847720146179,
      "learning_rate": 0.0001808209092582367,
      "loss": 1.7522,
      "num_input_tokens_seen": 3161520,
      "step": 527,
      "train_runtime": 528.2049,
      "train_tokens_per_second": 5985.405
    },
    {
      "epoch": 0.19420689655172413,
      "grad_norm": 0.7588212490081787,
      "learning_rate": 0.0001807840971838763,
      "loss": 1.7751,
      "num_input_tokens_seen": 3166736,
      "step": 528,
      "train_runtime": 529.0726,
      "train_tokens_per_second": 5985.447
    },
    {
      "epoch": 0.19457471264367815,
      "grad_norm": 0.7636683583259583,
      "learning_rate": 0.00018074728510951592,
      "loss": 1.7831,
      "num_input_tokens_seen": 3174368,
      "step": 529,
      "train_runtime": 530.2866,
      "train_tokens_per_second": 5986.136
    },
    {
      "epoch": 0.19494252873563217,
      "grad_norm": 0.8051586747169495,
      "learning_rate": 0.00018071047303515553,
      "loss": 1.848,
      "num_input_tokens_seen": 3180432,
      "step": 530,
      "train_runtime": 531.2842,
      "train_tokens_per_second": 5986.31
    },
    {
      "epoch": 0.19531034482758622,
      "grad_norm": 0.6396377086639404,
      "learning_rate": 0.00018067366096079514,
      "loss": 1.8452,
      "num_input_tokens_seen": 3188816,
      "step": 531,
      "train_runtime": 532.5598,
      "train_tokens_per_second": 5987.715
    },
    {
      "epoch": 0.19567816091954024,
      "grad_norm": 0.7458609938621521,
      "learning_rate": 0.00018063684888643475,
      "loss": 1.9903,
      "num_input_tokens_seen": 3193312,
      "step": 532,
      "train_runtime": 533.3496,
      "train_tokens_per_second": 5987.277
    },
    {
      "epoch": 0.19604597701149426,
      "grad_norm": 0.739069938659668,
      "learning_rate": 0.00018060003681207436,
      "loss": 1.8407,
      "num_input_tokens_seen": 3200224,
      "step": 533,
      "train_runtime": 534.4622,
      "train_tokens_per_second": 5987.746
    },
    {
      "epoch": 0.19641379310344828,
      "grad_norm": 0.7824617624282837,
      "learning_rate": 0.000180563224737714,
      "loss": 1.8743,
      "num_input_tokens_seen": 3208832,
      "step": 534,
      "train_runtime": 535.7856,
      "train_tokens_per_second": 5989.023
    },
    {
      "epoch": 0.1967816091954023,
      "grad_norm": 0.746059238910675,
      "learning_rate": 0.0001805264126633536,
      "loss": 1.719,
      "num_input_tokens_seen": 3216208,
      "step": 535,
      "train_runtime": 536.9608,
      "train_tokens_per_second": 5989.652
    },
    {
      "epoch": 0.19714942528735632,
      "grad_norm": 0.6949732899665833,
      "learning_rate": 0.0001804896005889932,
      "loss": 1.8434,
      "num_input_tokens_seen": 3226928,
      "step": 536,
      "train_runtime": 538.5816,
      "train_tokens_per_second": 5991.53
    },
    {
      "epoch": 0.19751724137931034,
      "grad_norm": 0.6921513080596924,
      "learning_rate": 0.00018045278851463282,
      "loss": 1.6434,
      "num_input_tokens_seen": 3231440,
      "step": 537,
      "train_runtime": 539.3344,
      "train_tokens_per_second": 5991.533
    },
    {
      "epoch": 0.19788505747126436,
      "grad_norm": 0.7133778929710388,
      "learning_rate": 0.00018041597644027243,
      "loss": 1.9168,
      "num_input_tokens_seen": 3238544,
      "step": 538,
      "train_runtime": 540.4574,
      "train_tokens_per_second": 5992.228
    },
    {
      "epoch": 0.19825287356321838,
      "grad_norm": 0.8469346165657043,
      "learning_rate": 0.000180379164365912,
      "loss": 1.8581,
      "num_input_tokens_seen": 3243568,
      "step": 539,
      "train_runtime": 541.3338,
      "train_tokens_per_second": 5991.807
    },
    {
      "epoch": 0.1986206896551724,
      "grad_norm": 0.7539319396018982,
      "learning_rate": 0.00018034235229155162,
      "loss": 1.6721,
      "num_input_tokens_seen": 3247744,
      "step": 540,
      "train_runtime": 542.1184,
      "train_tokens_per_second": 5990.839
    },
    {
      "epoch": 0.19898850574712643,
      "grad_norm": 0.8973172307014465,
      "learning_rate": 0.00018030554021719123,
      "loss": 1.9919,
      "num_input_tokens_seen": 3251760,
      "step": 541,
      "train_runtime": 543.3016,
      "train_tokens_per_second": 5985.184
    },
    {
      "epoch": 0.19935632183908045,
      "grad_norm": 0.6971712708473206,
      "learning_rate": 0.00018026872814283086,
      "loss": 1.9209,
      "num_input_tokens_seen": 3257168,
      "step": 542,
      "train_runtime": 544.1947,
      "train_tokens_per_second": 5985.299
    },
    {
      "epoch": 0.1997241379310345,
      "grad_norm": 0.7133026123046875,
      "learning_rate": 0.00018023191606847047,
      "loss": 1.7659,
      "num_input_tokens_seen": 3263088,
      "step": 543,
      "train_runtime": 545.1676,
      "train_tokens_per_second": 5985.477
    },
    {
      "epoch": 0.20009195402298852,
      "grad_norm": 0.7758365273475647,
      "learning_rate": 0.00018019510399411008,
      "loss": 1.992,
      "num_input_tokens_seen": 3268144,
      "step": 544,
      "train_runtime": 546.0315,
      "train_tokens_per_second": 5985.267
    },
    {
      "epoch": 0.20009195402298852,
      "eval_loss": 1.8736172914505005,
      "eval_runtime": 4.8041,
      "eval_samples_per_second": 208.156,
      "eval_steps_per_second": 13.114,
      "num_input_tokens_seen": 3268144,
      "step": 544
    },
    {
      "epoch": 0.20045977011494254,
      "grad_norm": 0.7243053913116455,
      "learning_rate": 0.0001801582919197497,
      "loss": 1.9157,
      "num_input_tokens_seen": 3275216,
      "step": 545,
      "train_runtime": 551.9591,
      "train_tokens_per_second": 5933.802
    },
    {
      "epoch": 0.20082758620689656,
      "grad_norm": 0.6808502674102783,
      "learning_rate": 0.0001801214798453893,
      "loss": 1.6435,
      "num_input_tokens_seen": 3281888,
      "step": 546,
      "train_runtime": 553.0054,
      "train_tokens_per_second": 5934.64
    },
    {
      "epoch": 0.20119540229885058,
      "grad_norm": 0.7165328860282898,
      "learning_rate": 0.0001800846677710289,
      "loss": 1.8499,
      "num_input_tokens_seen": 3286448,
      "step": 547,
      "train_runtime": 553.7982,
      "train_tokens_per_second": 5934.378
    },
    {
      "epoch": 0.2015632183908046,
      "grad_norm": 0.7008447051048279,
      "learning_rate": 0.00018004785569666852,
      "loss": 2.0328,
      "num_input_tokens_seen": 3291344,
      "step": 548,
      "train_runtime": 554.642,
      "train_tokens_per_second": 5934.178
    },
    {
      "epoch": 0.20193103448275862,
      "grad_norm": 0.6276164650917053,
      "learning_rate": 0.00018001104362230813,
      "loss": 1.7145,
      "num_input_tokens_seen": 3296880,
      "step": 549,
      "train_runtime": 555.5703,
      "train_tokens_per_second": 5934.227
    },
    {
      "epoch": 0.20229885057471264,
      "grad_norm": 0.6550085544586182,
      "learning_rate": 0.00017997423154794773,
      "loss": 1.9304,
      "num_input_tokens_seen": 3303008,
      "step": 550,
      "train_runtime": 556.568,
      "train_tokens_per_second": 5934.599
    },
    {
      "epoch": 0.20266666666666666,
      "grad_norm": 0.6337085962295532,
      "learning_rate": 0.00017993741947358734,
      "loss": 1.7112,
      "num_input_tokens_seen": 3308576,
      "step": 551,
      "train_runtime": 557.4912,
      "train_tokens_per_second": 5934.76
    },
    {
      "epoch": 0.20303448275862068,
      "grad_norm": 0.6780610084533691,
      "learning_rate": 0.00017990060739922695,
      "loss": 1.7727,
      "num_input_tokens_seen": 3313584,
      "step": 552,
      "train_runtime": 558.3271,
      "train_tokens_per_second": 5934.843
    },
    {
      "epoch": 0.2034022988505747,
      "grad_norm": 0.6733942031860352,
      "learning_rate": 0.00017986379532486656,
      "loss": 1.9891,
      "num_input_tokens_seen": 3321920,
      "step": 553,
      "train_runtime": 559.6137,
      "train_tokens_per_second": 5936.095
    },
    {
      "epoch": 0.20377011494252872,
      "grad_norm": 0.7534530758857727,
      "learning_rate": 0.00017982698325050617,
      "loss": 2.0452,
      "num_input_tokens_seen": 3326640,
      "step": 554,
      "train_runtime": 560.4157,
      "train_tokens_per_second": 5936.022
    },
    {
      "epoch": 0.20413793103448277,
      "grad_norm": 0.6698739528656006,
      "learning_rate": 0.00017979017117614578,
      "loss": 1.8734,
      "num_input_tokens_seen": 3332016,
      "step": 555,
      "train_runtime": 561.3054,
      "train_tokens_per_second": 5936.191
    },
    {
      "epoch": 0.2045057471264368,
      "grad_norm": 0.7130821347236633,
      "learning_rate": 0.0001797533591017854,
      "loss": 1.9783,
      "num_input_tokens_seen": 3336976,
      "step": 556,
      "train_runtime": 562.1583,
      "train_tokens_per_second": 5936.008
    },
    {
      "epoch": 0.20487356321839081,
      "grad_norm": 0.682594895362854,
      "learning_rate": 0.000179716547027425,
      "loss": 1.6873,
      "num_input_tokens_seen": 3344016,
      "step": 557,
      "train_runtime": 563.2984,
      "train_tokens_per_second": 5936.491
    },
    {
      "epoch": 0.20524137931034483,
      "grad_norm": 0.7581687569618225,
      "learning_rate": 0.0001796797349530646,
      "loss": 1.8726,
      "num_input_tokens_seen": 3348432,
      "step": 558,
      "train_runtime": 564.0586,
      "train_tokens_per_second": 5936.319
    },
    {
      "epoch": 0.20560919540229886,
      "grad_norm": 0.6403698325157166,
      "learning_rate": 0.00017964292287870424,
      "loss": 1.7022,
      "num_input_tokens_seen": 3356416,
      "step": 559,
      "train_runtime": 565.3042,
      "train_tokens_per_second": 5937.362
    },
    {
      "epoch": 0.20597701149425288,
      "grad_norm": 0.7418535947799683,
      "learning_rate": 0.00017960611080434385,
      "loss": 1.8055,
      "num_input_tokens_seen": 3360304,
      "step": 560,
      "train_runtime": 565.9934,
      "train_tokens_per_second": 5937.002
    },
    {
      "epoch": 0.2063448275862069,
      "grad_norm": 0.6513658761978149,
      "learning_rate": 0.00017956929872998346,
      "loss": 1.9174,
      "num_input_tokens_seen": 3369008,
      "step": 561,
      "train_runtime": 567.3447,
      "train_tokens_per_second": 5938.203
    },
    {
      "epoch": 0.20671264367816092,
      "grad_norm": 0.6533668041229248,
      "learning_rate": 0.00017953248665562307,
      "loss": 1.7832,
      "num_input_tokens_seen": 3382272,
      "step": 562,
      "train_runtime": 569.315,
      "train_tokens_per_second": 5940.95
    },
    {
      "epoch": 0.20708045977011494,
      "grad_norm": 0.7822544574737549,
      "learning_rate": 0.00017949567458126265,
      "loss": 1.6444,
      "num_input_tokens_seen": 3387600,
      "step": 563,
      "train_runtime": 570.2116,
      "train_tokens_per_second": 5940.953
    },
    {
      "epoch": 0.20744827586206896,
      "grad_norm": 0.8226087689399719,
      "learning_rate": 0.00017945886250690226,
      "loss": 1.7219,
      "num_input_tokens_seen": 3394480,
      "step": 564,
      "train_runtime": 571.3165,
      "train_tokens_per_second": 5941.505
    },
    {
      "epoch": 0.20781609195402298,
      "grad_norm": 0.7007325887680054,
      "learning_rate": 0.00017942205043254187,
      "loss": 1.8154,
      "num_input_tokens_seen": 3400144,
      "step": 565,
      "train_runtime": 572.2224,
      "train_tokens_per_second": 5941.997
    },
    {
      "epoch": 0.208183908045977,
      "grad_norm": 0.7794937491416931,
      "learning_rate": 0.00017938523835818148,
      "loss": 1.8811,
      "num_input_tokens_seen": 3406160,
      "step": 566,
      "train_runtime": 573.1857,
      "train_tokens_per_second": 5942.507
    },
    {
      "epoch": 0.20855172413793102,
      "grad_norm": 0.8454579710960388,
      "learning_rate": 0.00017934842628382109,
      "loss": 1.9583,
      "num_input_tokens_seen": 3410656,
      "step": 567,
      "train_runtime": 573.9627,
      "train_tokens_per_second": 5942.295
    },
    {
      "epoch": 0.20891954022988507,
      "grad_norm": 0.7634625434875488,
      "learning_rate": 0.00017931161420946072,
      "loss": 1.848,
      "num_input_tokens_seen": 3415088,
      "step": 568,
      "train_runtime": 574.7397,
      "train_tokens_per_second": 5941.973
    },
    {
      "epoch": 0.2092873563218391,
      "grad_norm": 0.6991872787475586,
      "learning_rate": 0.00017927480213510033,
      "loss": 1.9338,
      "num_input_tokens_seen": 3424400,
      "step": 569,
      "train_runtime": 576.1774,
      "train_tokens_per_second": 5943.308
    },
    {
      "epoch": 0.2096551724137931,
      "grad_norm": 0.6553771495819092,
      "learning_rate": 0.00017923799006073994,
      "loss": 1.8144,
      "num_input_tokens_seen": 3432064,
      "step": 570,
      "train_runtime": 577.3636,
      "train_tokens_per_second": 5944.372
    },
    {
      "epoch": 0.21002298850574713,
      "grad_norm": 0.7230982184410095,
      "learning_rate": 0.00017920117798637955,
      "loss": 2.0145,
      "num_input_tokens_seen": 3440432,
      "step": 571,
      "train_runtime": 579.2037,
      "train_tokens_per_second": 5939.935
    },
    {
      "epoch": 0.21039080459770115,
      "grad_norm": 0.7031193375587463,
      "learning_rate": 0.00017916436591201916,
      "loss": 1.9469,
      "num_input_tokens_seen": 3445792,
      "step": 572,
      "train_runtime": 580.1046,
      "train_tokens_per_second": 5939.95
    },
    {
      "epoch": 0.21075862068965517,
      "grad_norm": 0.7555750608444214,
      "learning_rate": 0.00017912755383765877,
      "loss": 1.9087,
      "num_input_tokens_seen": 3451088,
      "step": 573,
      "train_runtime": 580.9869,
      "train_tokens_per_second": 5940.045
    },
    {
      "epoch": 0.2111264367816092,
      "grad_norm": 0.6957426071166992,
      "learning_rate": 0.00017909074176329838,
      "loss": 1.722,
      "num_input_tokens_seen": 3456240,
      "step": 574,
      "train_runtime": 581.8388,
      "train_tokens_per_second": 5940.202
    },
    {
      "epoch": 0.21149425287356322,
      "grad_norm": 0.8073890209197998,
      "learning_rate": 0.00017905392968893798,
      "loss": 2.0494,
      "num_input_tokens_seen": 3461072,
      "step": 575,
      "train_runtime": 582.6506,
      "train_tokens_per_second": 5940.219
    },
    {
      "epoch": 0.21186206896551724,
      "grad_norm": 0.6882850527763367,
      "learning_rate": 0.0001790171176145776,
      "loss": 1.9438,
      "num_input_tokens_seen": 3466352,
      "step": 576,
      "train_runtime": 583.525,
      "train_tokens_per_second": 5940.366
    },
    {
      "epoch": 0.21222988505747126,
      "grad_norm": 0.6792072653770447,
      "learning_rate": 0.0001789803055402172,
      "loss": 1.8989,
      "num_input_tokens_seen": 3472784,
      "step": 577,
      "train_runtime": 584.5789,
      "train_tokens_per_second": 5940.659
    },
    {
      "epoch": 0.21259770114942528,
      "grad_norm": 0.7088136076927185,
      "learning_rate": 0.0001789434934658568,
      "loss": 1.9281,
      "num_input_tokens_seen": 3477808,
      "step": 578,
      "train_runtime": 585.4312,
      "train_tokens_per_second": 5940.592
    },
    {
      "epoch": 0.2129655172413793,
      "grad_norm": 0.7313692569732666,
      "learning_rate": 0.00017890668139149642,
      "loss": 1.7956,
      "num_input_tokens_seen": 3483072,
      "step": 579,
      "train_runtime": 586.3013,
      "train_tokens_per_second": 5940.754
    },
    {
      "epoch": 0.21333333333333335,
      "grad_norm": 0.7447689771652222,
      "learning_rate": 0.00017886986931713603,
      "loss": 1.8087,
      "num_input_tokens_seen": 3487808,
      "step": 580,
      "train_runtime": 587.1009,
      "train_tokens_per_second": 5940.73
    },
    {
      "epoch": 0.21370114942528737,
      "grad_norm": 0.666411280632019,
      "learning_rate": 0.00017883305724277564,
      "loss": 1.8288,
      "num_input_tokens_seen": 3493264,
      "step": 581,
      "train_runtime": 588.0008,
      "train_tokens_per_second": 5940.917
    },
    {
      "epoch": 0.2140689655172414,
      "grad_norm": 0.6953850984573364,
      "learning_rate": 0.00017879624516841525,
      "loss": 1.9789,
      "num_input_tokens_seen": 3500064,
      "step": 582,
      "train_runtime": 589.0704,
      "train_tokens_per_second": 5941.673
    },
    {
      "epoch": 0.2144367816091954,
      "grad_norm": 0.7432466149330139,
      "learning_rate": 0.00017875943309405486,
      "loss": 1.9108,
      "num_input_tokens_seen": 3504528,
      "step": 583,
      "train_runtime": 589.8489,
      "train_tokens_per_second": 5941.4
    },
    {
      "epoch": 0.21480459770114943,
      "grad_norm": 0.6990666389465332,
      "learning_rate": 0.00017872262101969446,
      "loss": 1.8579,
      "num_input_tokens_seen": 3509536,
      "step": 584,
      "train_runtime": 590.6963,
      "train_tokens_per_second": 5941.354
    },
    {
      "epoch": 0.21517241379310345,
      "grad_norm": 0.7038804888725281,
      "learning_rate": 0.0001786858089453341,
      "loss": 1.8356,
      "num_input_tokens_seen": 3517008,
      "step": 585,
      "train_runtime": 591.8795,
      "train_tokens_per_second": 5942.102
    },
    {
      "epoch": 0.21554022988505747,
      "grad_norm": 0.6550593972206116,
      "learning_rate": 0.00017864899687097368,
      "loss": 1.9013,
      "num_input_tokens_seen": 3525216,
      "step": 586,
      "train_runtime": 593.1798,
      "train_tokens_per_second": 5942.913
    },
    {
      "epoch": 0.2159080459770115,
      "grad_norm": 0.6644894480705261,
      "learning_rate": 0.0001786121847966133,
      "loss": 1.9321,
      "num_input_tokens_seen": 3531472,
      "step": 587,
      "train_runtime": 594.2065,
      "train_tokens_per_second": 5943.173
    },
    {
      "epoch": 0.2162758620689655,
      "grad_norm": 0.7219209671020508,
      "learning_rate": 0.0001785753727222529,
      "loss": 1.7653,
      "num_input_tokens_seen": 3536448,
      "step": 588,
      "train_runtime": 595.0394,
      "train_tokens_per_second": 5943.216
    },
    {
      "epoch": 0.21664367816091953,
      "grad_norm": 0.684423565864563,
      "learning_rate": 0.0001785385606478925,
      "loss": 1.8095,
      "num_input_tokens_seen": 3541744,
      "step": 589,
      "train_runtime": 595.922,
      "train_tokens_per_second": 5943.302
    },
    {
      "epoch": 0.21701149425287355,
      "grad_norm": 0.7029337882995605,
      "learning_rate": 0.00017850174857353212,
      "loss": 1.8254,
      "num_input_tokens_seen": 3547120,
      "step": 590,
      "train_runtime": 596.8062,
      "train_tokens_per_second": 5943.504
    },
    {
      "epoch": 0.21737931034482758,
      "grad_norm": 0.71671462059021,
      "learning_rate": 0.00017846493649917173,
      "loss": 1.6723,
      "num_input_tokens_seen": 3553872,
      "step": 591,
      "train_runtime": 597.8796,
      "train_tokens_per_second": 5944.126
    },
    {
      "epoch": 0.2177471264367816,
      "grad_norm": 0.6957038640975952,
      "learning_rate": 0.00017842812442481134,
      "loss": 1.7098,
      "num_input_tokens_seen": 3561120,
      "step": 592,
      "train_runtime": 599.0471,
      "train_tokens_per_second": 5944.641
    },
    {
      "epoch": 0.21811494252873564,
      "grad_norm": 0.7224963307380676,
      "learning_rate": 0.00017839131235045094,
      "loss": 1.9218,
      "num_input_tokens_seen": 3566112,
      "step": 593,
      "train_runtime": 599.8729,
      "train_tokens_per_second": 5944.779
    },
    {
      "epoch": 0.21848275862068967,
      "grad_norm": 0.6886675357818604,
      "learning_rate": 0.00017835450027609058,
      "loss": 1.8008,
      "num_input_tokens_seen": 3574928,
      "step": 594,
      "train_runtime": 601.2434,
      "train_tokens_per_second": 5945.891
    },
    {
      "epoch": 0.2188505747126437,
      "grad_norm": 0.7517166137695312,
      "learning_rate": 0.0001783176882017302,
      "loss": 1.7053,
      "num_input_tokens_seen": 3579280,
      "step": 595,
      "train_runtime": 601.9943,
      "train_tokens_per_second": 5945.704
    },
    {
      "epoch": 0.2192183908045977,
      "grad_norm": 0.735989511013031,
      "learning_rate": 0.0001782808761273698,
      "loss": 1.7164,
      "num_input_tokens_seen": 3585120,
      "step": 596,
      "train_runtime": 602.9376,
      "train_tokens_per_second": 5946.088
    },
    {
      "epoch": 0.21958620689655173,
      "grad_norm": 0.7238099575042725,
      "learning_rate": 0.0001782440640530094,
      "loss": 1.8123,
      "num_input_tokens_seen": 3592304,
      "step": 597,
      "train_runtime": 604.0617,
      "train_tokens_per_second": 5946.916
    },
    {
      "epoch": 0.21995402298850575,
      "grad_norm": 0.8470680713653564,
      "learning_rate": 0.00017820725197864902,
      "loss": 1.9266,
      "num_input_tokens_seen": 3598336,
      "step": 598,
      "train_runtime": 605.0367,
      "train_tokens_per_second": 5947.302
    },
    {
      "epoch": 0.22032183908045977,
      "grad_norm": 0.7912116050720215,
      "learning_rate": 0.0001781704399042886,
      "loss": 1.831,
      "num_input_tokens_seen": 3602992,
      "step": 599,
      "train_runtime": 605.8589,
      "train_tokens_per_second": 5946.916
    },
    {
      "epoch": 0.2206896551724138,
      "grad_norm": 0.7828996181488037,
      "learning_rate": 0.0001781336278299282,
      "loss": 1.8715,
      "num_input_tokens_seen": 3608048,
      "step": 600,
      "train_runtime": 606.7243,
      "train_tokens_per_second": 5946.767
    },
    {
      "epoch": 0.2210574712643678,
      "grad_norm": 0.7932271957397461,
      "learning_rate": 0.00017809681575556782,
      "loss": 2.1117,
      "num_input_tokens_seen": 3612880,
      "step": 601,
      "train_runtime": 608.0032,
      "train_tokens_per_second": 5942.205
    },
    {
      "epoch": 0.22142528735632183,
      "grad_norm": 0.7785124778747559,
      "learning_rate": 0.00017806000368120745,
      "loss": 1.9367,
      "num_input_tokens_seen": 3618320,
      "step": 602,
      "train_runtime": 608.885,
      "train_tokens_per_second": 5942.534
    },
    {
      "epoch": 0.22179310344827585,
      "grad_norm": 0.8122251629829407,
      "learning_rate": 0.00017802319160684706,
      "loss": 1.6962,
      "num_input_tokens_seen": 3623488,
      "step": 603,
      "train_runtime": 609.7605,
      "train_tokens_per_second": 5942.478
    },
    {
      "epoch": 0.22216091954022987,
      "grad_norm": 0.6948246955871582,
      "learning_rate": 0.00017798637953248667,
      "loss": 1.733,
      "num_input_tokens_seen": 3629104,
      "step": 604,
      "train_runtime": 610.6906,
      "train_tokens_per_second": 5942.623
    },
    {
      "epoch": 0.22252873563218392,
      "grad_norm": 0.7401772141456604,
      "learning_rate": 0.00017794956745812628,
      "loss": 2.0811,
      "num_input_tokens_seen": 3633664,
      "step": 605,
      "train_runtime": 611.5031,
      "train_tokens_per_second": 5942.184
    },
    {
      "epoch": 0.22289655172413794,
      "grad_norm": 0.8130412697792053,
      "learning_rate": 0.0001779127553837659,
      "loss": 2.0618,
      "num_input_tokens_seen": 3638208,
      "step": 606,
      "train_runtime": 612.2873,
      "train_tokens_per_second": 5941.995
    },
    {
      "epoch": 0.22326436781609196,
      "grad_norm": 0.8034175038337708,
      "learning_rate": 0.0001778759433094055,
      "loss": 1.8695,
      "num_input_tokens_seen": 3642576,
      "step": 607,
      "train_runtime": 613.0453,
      "train_tokens_per_second": 5941.773
    },
    {
      "epoch": 0.22363218390804598,
      "grad_norm": 0.7706856727600098,
      "learning_rate": 0.0001778391312350451,
      "loss": 1.8332,
      "num_input_tokens_seen": 3647024,
      "step": 608,
      "train_runtime": 613.8286,
      "train_tokens_per_second": 5941.437
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.8016795516014099,
      "learning_rate": 0.00017780231916068471,
      "loss": 1.8524,
      "num_input_tokens_seen": 3651744,
      "step": 609,
      "train_runtime": 614.6363,
      "train_tokens_per_second": 5941.309
    },
    {
      "epoch": 0.22436781609195403,
      "grad_norm": 0.6998735666275024,
      "learning_rate": 0.00017776550708632432,
      "loss": 1.8093,
      "num_input_tokens_seen": 3658720,
      "step": 610,
      "train_runtime": 615.7086,
      "train_tokens_per_second": 5942.291
    },
    {
      "epoch": 0.22473563218390805,
      "grad_norm": 0.7392911314964294,
      "learning_rate": 0.00017772869501196393,
      "loss": 1.9444,
      "num_input_tokens_seen": 3663408,
      "step": 611,
      "train_runtime": 616.5343,
      "train_tokens_per_second": 5941.937
    },
    {
      "epoch": 0.22510344827586207,
      "grad_norm": 0.7386958599090576,
      "learning_rate": 0.00017769188293760354,
      "loss": 1.7245,
      "num_input_tokens_seen": 3669232,
      "step": 612,
      "train_runtime": 617.4985,
      "train_tokens_per_second": 5942.091
    },
    {
      "epoch": 0.2254712643678161,
      "grad_norm": 0.8053456544876099,
      "learning_rate": 0.00017765507086324315,
      "loss": 1.8809,
      "num_input_tokens_seen": 3675056,
      "step": 613,
      "train_runtime": 618.4492,
      "train_tokens_per_second": 5942.374
    },
    {
      "epoch": 0.2258390804597701,
      "grad_norm": 0.6959522366523743,
      "learning_rate": 0.00017761825878888276,
      "loss": 1.7003,
      "num_input_tokens_seen": 3680336,
      "step": 614,
      "train_runtime": 619.3484,
      "train_tokens_per_second": 5942.271
    },
    {
      "epoch": 0.22620689655172413,
      "grad_norm": 0.8027039766311646,
      "learning_rate": 0.00017758144671452237,
      "loss": 1.6945,
      "num_input_tokens_seen": 3685568,
      "step": 615,
      "train_runtime": 620.2209,
      "train_tokens_per_second": 5942.348
    },
    {
      "epoch": 0.22657471264367815,
      "grad_norm": 0.6943402290344238,
      "learning_rate": 0.00017754463464016198,
      "loss": 1.7537,
      "num_input_tokens_seen": 3691744,
      "step": 616,
      "train_runtime": 621.2084,
      "train_tokens_per_second": 5942.843
    },
    {
      "epoch": 0.22694252873563217,
      "grad_norm": 0.6399091482162476,
      "learning_rate": 0.00017750782256580158,
      "loss": 1.6842,
      "num_input_tokens_seen": 3702768,
      "step": 617,
      "train_runtime": 622.8754,
      "train_tokens_per_second": 5944.637
    },
    {
      "epoch": 0.22731034482758622,
      "grad_norm": 0.7171786427497864,
      "learning_rate": 0.0001774710104914412,
      "loss": 1.7311,
      "num_input_tokens_seen": 3709216,
      "step": 618,
      "train_runtime": 623.9598,
      "train_tokens_per_second": 5944.639
    },
    {
      "epoch": 0.22767816091954024,
      "grad_norm": 0.7073584198951721,
      "learning_rate": 0.00017743419841708083,
      "loss": 1.8136,
      "num_input_tokens_seen": 3714112,
      "step": 619,
      "train_runtime": 624.7929,
      "train_tokens_per_second": 5944.549
    },
    {
      "epoch": 0.22804597701149426,
      "grad_norm": 0.6660651564598083,
      "learning_rate": 0.00017739738634272044,
      "loss": 1.8246,
      "num_input_tokens_seen": 3723104,
      "step": 620,
      "train_runtime": 626.1956,
      "train_tokens_per_second": 5945.593
    },
    {
      "epoch": 0.22841379310344828,
      "grad_norm": 0.7154439091682434,
      "learning_rate": 0.00017736057426836005,
      "loss": 1.7245,
      "num_input_tokens_seen": 3730000,
      "step": 621,
      "train_runtime": 627.3042,
      "train_tokens_per_second": 5946.079
    },
    {
      "epoch": 0.2287816091954023,
      "grad_norm": 0.7616729736328125,
      "learning_rate": 0.00017732376219399966,
      "loss": 2.0513,
      "num_input_tokens_seen": 3736912,
      "step": 622,
      "train_runtime": 628.3838,
      "train_tokens_per_second": 5946.863
    },
    {
      "epoch": 0.22914942528735632,
      "grad_norm": 0.7419039607048035,
      "learning_rate": 0.00017728695011963924,
      "loss": 1.7101,
      "num_input_tokens_seen": 3741840,
      "step": 623,
      "train_runtime": 629.1809,
      "train_tokens_per_second": 5947.161
    },
    {
      "epoch": 0.22951724137931034,
      "grad_norm": 0.710760772228241,
      "learning_rate": 0.00017725013804527885,
      "loss": 1.9318,
      "num_input_tokens_seen": 3747200,
      "step": 624,
      "train_runtime": 630.0671,
      "train_tokens_per_second": 5947.304
    },
    {
      "epoch": 0.22988505747126436,
      "grad_norm": 0.649756908416748,
      "learning_rate": 0.00017721332597091846,
      "loss": 2.0162,
      "num_input_tokens_seen": 3752016,
      "step": 625,
      "train_runtime": 630.9201,
      "train_tokens_per_second": 5946.895
    },
    {
      "epoch": 0.23025287356321839,
      "grad_norm": 0.7205575108528137,
      "learning_rate": 0.00017717651389655806,
      "loss": 1.82,
      "num_input_tokens_seen": 3757104,
      "step": 626,
      "train_runtime": 631.7768,
      "train_tokens_per_second": 5946.885
    },
    {
      "epoch": 0.2306206896551724,
      "grad_norm": 0.7039726376533508,
      "learning_rate": 0.00017713970182219767,
      "loss": 1.9755,
      "num_input_tokens_seen": 3765232,
      "step": 627,
      "train_runtime": 633.0555,
      "train_tokens_per_second": 5947.712
    },
    {
      "epoch": 0.23098850574712643,
      "grad_norm": 0.6161988973617554,
      "learning_rate": 0.0001771028897478373,
      "loss": 1.7607,
      "num_input_tokens_seen": 3771248,
      "step": 628,
      "train_runtime": 634.038,
      "train_tokens_per_second": 5947.984
    },
    {
      "epoch": 0.23135632183908045,
      "grad_norm": 0.6338945031166077,
      "learning_rate": 0.00017706607767347692,
      "loss": 1.6776,
      "num_input_tokens_seen": 3778848,
      "step": 629,
      "train_runtime": 635.2454,
      "train_tokens_per_second": 5948.642
    },
    {
      "epoch": 0.2317241379310345,
      "grad_norm": 0.6846767663955688,
      "learning_rate": 0.00017702926559911653,
      "loss": 1.8294,
      "num_input_tokens_seen": 3783184,
      "step": 630,
      "train_runtime": 636.0073,
      "train_tokens_per_second": 5948.334
    },
    {
      "epoch": 0.23209195402298852,
      "grad_norm": 0.6843883991241455,
      "learning_rate": 0.00017699245352475614,
      "loss": 2.0506,
      "num_input_tokens_seen": 3788352,
      "step": 631,
      "train_runtime": 637.3439,
      "train_tokens_per_second": 5943.968
    },
    {
      "epoch": 0.23245977011494254,
      "grad_norm": 0.632727861404419,
      "learning_rate": 0.00017695564145039575,
      "loss": 1.7141,
      "num_input_tokens_seen": 3794576,
      "step": 632,
      "train_runtime": 638.3541,
      "train_tokens_per_second": 5944.312
    },
    {
      "epoch": 0.23282758620689656,
      "grad_norm": 0.6620173454284668,
      "learning_rate": 0.00017691882937603535,
      "loss": 1.793,
      "num_input_tokens_seen": 3799728,
      "step": 633,
      "train_runtime": 639.2227,
      "train_tokens_per_second": 5944.295
    },
    {
      "epoch": 0.23319540229885058,
      "grad_norm": 0.7021250128746033,
      "learning_rate": 0.00017688201730167496,
      "loss": 1.8166,
      "num_input_tokens_seen": 3804928,
      "step": 634,
      "train_runtime": 640.0866,
      "train_tokens_per_second": 5944.396
    },
    {
      "epoch": 0.2335632183908046,
      "grad_norm": 0.8043453097343445,
      "learning_rate": 0.00017684520522731457,
      "loss": 1.7254,
      "num_input_tokens_seen": 3811344,
      "step": 635,
      "train_runtime": 641.1146,
      "train_tokens_per_second": 5944.871
    },
    {
      "epoch": 0.23393103448275862,
      "grad_norm": 0.7299250364303589,
      "learning_rate": 0.00017680839315295418,
      "loss": 1.995,
      "num_input_tokens_seen": 3818048,
      "step": 636,
      "train_runtime": 642.1857,
      "train_tokens_per_second": 5945.395
    },
    {
      "epoch": 0.23429885057471264,
      "grad_norm": 0.673685610294342,
      "learning_rate": 0.0001767715810785938,
      "loss": 1.8463,
      "num_input_tokens_seen": 3824112,
      "step": 637,
      "train_runtime": 643.1493,
      "train_tokens_per_second": 5945.917
    },
    {
      "epoch": 0.23466666666666666,
      "grad_norm": 0.6168797016143799,
      "learning_rate": 0.0001767347690042334,
      "loss": 1.8969,
      "num_input_tokens_seen": 3832528,
      "step": 638,
      "train_runtime": 644.4422,
      "train_tokens_per_second": 5947.047
    },
    {
      "epoch": 0.23503448275862068,
      "grad_norm": 0.7642756104469299,
      "learning_rate": 0.000176697956929873,
      "loss": 2.0022,
      "num_input_tokens_seen": 3837584,
      "step": 639,
      "train_runtime": 645.2897,
      "train_tokens_per_second": 5947.072
    },
    {
      "epoch": 0.2354022988505747,
      "grad_norm": 0.7426885962486267,
      "learning_rate": 0.00017666114485551262,
      "loss": 1.8024,
      "num_input_tokens_seen": 3843184,
      "step": 640,
      "train_runtime": 646.2141,
      "train_tokens_per_second": 5947.23
    },
    {
      "epoch": 0.23577011494252872,
      "grad_norm": 0.7234979867935181,
      "learning_rate": 0.00017662433278115223,
      "loss": 1.7453,
      "num_input_tokens_seen": 3850192,
      "step": 641,
      "train_runtime": 647.366,
      "train_tokens_per_second": 5947.473
    },
    {
      "epoch": 0.23613793103448275,
      "grad_norm": 0.6388785243034363,
      "learning_rate": 0.00017658752070679183,
      "loss": 1.7032,
      "num_input_tokens_seen": 3858592,
      "step": 642,
      "train_runtime": 648.6702,
      "train_tokens_per_second": 5948.465
    },
    {
      "epoch": 0.2365057471264368,
      "grad_norm": 0.6985915303230286,
      "learning_rate": 0.00017655070863243144,
      "loss": 1.8575,
      "num_input_tokens_seen": 3866288,
      "step": 643,
      "train_runtime": 649.8596,
      "train_tokens_per_second": 5949.42
    },
    {
      "epoch": 0.23687356321839081,
      "grad_norm": 0.674976646900177,
      "learning_rate": 0.00017651389655807105,
      "loss": 1.7333,
      "num_input_tokens_seen": 3873104,
      "step": 644,
      "train_runtime": 650.9387,
      "train_tokens_per_second": 5950.029
    },
    {
      "epoch": 0.23724137931034484,
      "grad_norm": 0.6618773937225342,
      "learning_rate": 0.0001764770844837107,
      "loss": 1.7685,
      "num_input_tokens_seen": 3878384,
      "step": 645,
      "train_runtime": 651.8213,
      "train_tokens_per_second": 5950.073
    },
    {
      "epoch": 0.23760919540229886,
      "grad_norm": 0.7591787576675415,
      "learning_rate": 0.00017644027240935027,
      "loss": 1.9053,
      "num_input_tokens_seen": 3884592,
      "step": 646,
      "train_runtime": 652.8453,
      "train_tokens_per_second": 5950.249
    },
    {
      "epoch": 0.23797701149425288,
      "grad_norm": 0.7597478032112122,
      "learning_rate": 0.00017640346033498988,
      "loss": 1.8146,
      "num_input_tokens_seen": 3889872,
      "step": 647,
      "train_runtime": 653.7209,
      "train_tokens_per_second": 5950.356
    },
    {
      "epoch": 0.2383448275862069,
      "grad_norm": 0.7259320616722107,
      "learning_rate": 0.0001763666482606295,
      "loss": 1.8141,
      "num_input_tokens_seen": 3894512,
      "step": 648,
      "train_runtime": 654.5088,
      "train_tokens_per_second": 5950.282
    },
    {
      "epoch": 0.23871264367816092,
      "grad_norm": 0.713681697845459,
      "learning_rate": 0.0001763298361862691,
      "loss": 1.7502,
      "num_input_tokens_seen": 3900832,
      "step": 649,
      "train_runtime": 655.5372,
      "train_tokens_per_second": 5950.588
    },
    {
      "epoch": 0.23908045977011494,
      "grad_norm": 0.6367166042327881,
      "learning_rate": 0.0001762930241119087,
      "loss": 1.7876,
      "num_input_tokens_seen": 3909696,
      "step": 650,
      "train_runtime": 656.901,
      "train_tokens_per_second": 5951.728
    },
    {
      "epoch": 0.23944827586206896,
      "grad_norm": 0.8463271856307983,
      "learning_rate": 0.00017625621203754831,
      "loss": 1.9971,
      "num_input_tokens_seen": 3914608,
      "step": 651,
      "train_runtime": 657.741,
      "train_tokens_per_second": 5951.595
    },
    {
      "epoch": 0.23981609195402298,
      "grad_norm": 0.7609216570854187,
      "learning_rate": 0.00017621939996318792,
      "loss": 1.8,
      "num_input_tokens_seen": 3923392,
      "step": 652,
      "train_runtime": 659.0891,
      "train_tokens_per_second": 5952.749
    },
    {
      "epoch": 0.240183908045977,
      "grad_norm": 0.8769307136535645,
      "learning_rate": 0.00017618258788882756,
      "loss": 1.8835,
      "num_input_tokens_seen": 3928576,
      "step": 653,
      "train_runtime": 659.9505,
      "train_tokens_per_second": 5952.835
    },
    {
      "epoch": 0.24055172413793102,
      "grad_norm": 0.8489215970039368,
      "learning_rate": 0.00017614577581446717,
      "loss": 1.7344,
      "num_input_tokens_seen": 3933984,
      "step": 654,
      "train_runtime": 660.8341,
      "train_tokens_per_second": 5953.059
    },
    {
      "epoch": 0.24091954022988507,
      "grad_norm": 0.7326881289482117,
      "learning_rate": 0.00017610896374010678,
      "loss": 1.7455,
      "num_input_tokens_seen": 3942272,
      "step": 655,
      "train_runtime": 662.1217,
      "train_tokens_per_second": 5953.999
    },
    {
      "epoch": 0.2412873563218391,
      "grad_norm": 0.8228838443756104,
      "learning_rate": 0.00017607215166574639,
      "loss": 1.9441,
      "num_input_tokens_seen": 3946848,
      "step": 656,
      "train_runtime": 662.9264,
      "train_tokens_per_second": 5953.674
    },
    {
      "epoch": 0.2416551724137931,
      "grad_norm": 0.6861847043037415,
      "learning_rate": 0.000176035339591386,
      "loss": 1.7707,
      "num_input_tokens_seen": 3951456,
      "step": 657,
      "train_runtime": 663.7259,
      "train_tokens_per_second": 5953.446
    },
    {
      "epoch": 0.24202298850574713,
      "grad_norm": 0.6617019176483154,
      "learning_rate": 0.0001759985275170256,
      "loss": 1.6957,
      "num_input_tokens_seen": 3958720,
      "step": 658,
      "train_runtime": 664.8682,
      "train_tokens_per_second": 5954.142
    },
    {
      "epoch": 0.24239080459770115,
      "grad_norm": 0.6823275089263916,
      "learning_rate": 0.00017596171544266519,
      "loss": 1.934,
      "num_input_tokens_seen": 3968432,
      "step": 659,
      "train_runtime": 666.3434,
      "train_tokens_per_second": 5955.535
    },
    {
      "epoch": 0.24275862068965517,
      "grad_norm": 0.7465804219245911,
      "learning_rate": 0.0001759249033683048,
      "loss": 1.8067,
      "num_input_tokens_seen": 3973152,
      "step": 660,
      "train_runtime": 667.1499,
      "train_tokens_per_second": 5955.411
    },
    {
      "epoch": 0.2431264367816092,
      "grad_norm": 0.7939881086349487,
      "learning_rate": 0.0001758880912939444,
      "loss": 2.0186,
      "num_input_tokens_seen": 3978160,
      "step": 661,
      "train_runtime": 668.4655,
      "train_tokens_per_second": 5951.182
    },
    {
      "epoch": 0.24349425287356322,
      "grad_norm": 0.6984440684318542,
      "learning_rate": 0.00017585127921958404,
      "loss": 1.667,
      "num_input_tokens_seen": 3983360,
      "step": 662,
      "train_runtime": 669.3371,
      "train_tokens_per_second": 5951.202
    },
    {
      "epoch": 0.24386206896551724,
      "grad_norm": 0.6744182109832764,
      "learning_rate": 0.00017581446714522365,
      "loss": 1.7551,
      "num_input_tokens_seen": 3988992,
      "step": 663,
      "train_runtime": 670.2541,
      "train_tokens_per_second": 5951.462
    },
    {
      "epoch": 0.24422988505747126,
      "grad_norm": 0.7228152751922607,
      "learning_rate": 0.00017577765507086326,
      "loss": 1.6449,
      "num_input_tokens_seen": 3994160,
      "step": 664,
      "train_runtime": 671.132,
      "train_tokens_per_second": 5951.377
    },
    {
      "epoch": 0.24459770114942528,
      "grad_norm": 0.7169431447982788,
      "learning_rate": 0.00017574084299650287,
      "loss": 2.0453,
      "num_input_tokens_seen": 3999456,
      "step": 665,
      "train_runtime": 672.0239,
      "train_tokens_per_second": 5951.36
    },
    {
      "epoch": 0.2449655172413793,
      "grad_norm": 0.7038352489471436,
      "learning_rate": 0.00017570403092214247,
      "loss": 1.8064,
      "num_input_tokens_seen": 4004864,
      "step": 666,
      "train_runtime": 672.9381,
      "train_tokens_per_second": 5951.311
    },
    {
      "epoch": 0.24533333333333332,
      "grad_norm": 0.6899456977844238,
      "learning_rate": 0.00017566721884778208,
      "loss": 1.7173,
      "num_input_tokens_seen": 4010624,
      "step": 667,
      "train_runtime": 673.8709,
      "train_tokens_per_second": 5951.621
    },
    {
      "epoch": 0.24570114942528737,
      "grad_norm": 0.6827689409255981,
      "learning_rate": 0.0001756304067734217,
      "loss": 1.9384,
      "num_input_tokens_seen": 4016112,
      "step": 668,
      "train_runtime": 674.7888,
      "train_tokens_per_second": 5951.658
    },
    {
      "epoch": 0.2460689655172414,
      "grad_norm": 0.6630880832672119,
      "learning_rate": 0.0001755935946990613,
      "loss": 1.7599,
      "num_input_tokens_seen": 4022160,
      "step": 669,
      "train_runtime": 675.7794,
      "train_tokens_per_second": 5951.883
    },
    {
      "epoch": 0.2464367816091954,
      "grad_norm": 0.7457953095436096,
      "learning_rate": 0.0001755567826247009,
      "loss": 1.7006,
      "num_input_tokens_seen": 4031584,
      "step": 670,
      "train_runtime": 677.2077,
      "train_tokens_per_second": 5953.246
    },
    {
      "epoch": 0.24680459770114943,
      "grad_norm": 0.6746439933776855,
      "learning_rate": 0.00017551997055034052,
      "loss": 2.0363,
      "num_input_tokens_seen": 4036304,
      "step": 671,
      "train_runtime": 678.0284,
      "train_tokens_per_second": 5953.001
    },
    {
      "epoch": 0.24717241379310345,
      "grad_norm": 0.756391167640686,
      "learning_rate": 0.00017548315847598013,
      "loss": 1.7493,
      "num_input_tokens_seen": 4041232,
      "step": 672,
      "train_runtime": 678.8637,
      "train_tokens_per_second": 5952.936
    },
    {
      "epoch": 0.24754022988505747,
      "grad_norm": 0.6909932494163513,
      "learning_rate": 0.00017544634640161974,
      "loss": 1.6926,
      "num_input_tokens_seen": 4046496,
      "step": 673,
      "train_runtime": 679.729,
      "train_tokens_per_second": 5953.102
    },
    {
      "epoch": 0.2479080459770115,
      "grad_norm": 0.7453582286834717,
      "learning_rate": 0.00017540953432725935,
      "loss": 1.7609,
      "num_input_tokens_seen": 4051168,
      "step": 674,
      "train_runtime": 680.5451,
      "train_tokens_per_second": 5952.828
    },
    {
      "epoch": 0.2482758620689655,
      "grad_norm": 0.8183576464653015,
      "learning_rate": 0.00017537272225289895,
      "loss": 1.7621,
      "num_input_tokens_seen": 4055536,
      "step": 675,
      "train_runtime": 681.315,
      "train_tokens_per_second": 5952.513
    },
    {
      "epoch": 0.24864367816091953,
      "grad_norm": 0.6562678217887878,
      "learning_rate": 0.00017533591017853856,
      "loss": 1.9175,
      "num_input_tokens_seen": 4061200,
      "step": 676,
      "train_runtime": 682.2424,
      "train_tokens_per_second": 5952.723
    },
    {
      "epoch": 0.24901149425287356,
      "grad_norm": 0.6592198014259338,
      "learning_rate": 0.00017529909810417817,
      "loss": 1.7407,
      "num_input_tokens_seen": 4067232,
      "step": 677,
      "train_runtime": 683.2577,
      "train_tokens_per_second": 5952.706
    },
    {
      "epoch": 0.24937931034482758,
      "grad_norm": 0.7017592191696167,
      "learning_rate": 0.00017526228602981778,
      "loss": 2.0132,
      "num_input_tokens_seen": 4073168,
      "step": 678,
      "train_runtime": 684.2171,
      "train_tokens_per_second": 5953.034
    },
    {
      "epoch": 0.2497471264367816,
      "grad_norm": 0.7664760947227478,
      "learning_rate": 0.00017522547395545742,
      "loss": 1.7824,
      "num_input_tokens_seen": 4078976,
      "step": 679,
      "train_runtime": 685.1825,
      "train_tokens_per_second": 5953.124
    },
    {
      "epoch": 0.2501149425287356,
      "grad_norm": 0.7167187333106995,
      "learning_rate": 0.00017518866188109703,
      "loss": 1.8287,
      "num_input_tokens_seen": 4084016,
      "step": 680,
      "train_runtime": 686.0382,
      "train_tokens_per_second": 5953.044
    },
    {
      "epoch": 0.25048275862068964,
      "grad_norm": 0.7765895128250122,
      "learning_rate": 0.00017515184980673664,
      "loss": 1.8548,
      "num_input_tokens_seen": 4088368,
      "step": 681,
      "train_runtime": 686.8084,
      "train_tokens_per_second": 5952.705
    },
    {
      "epoch": 0.25085057471264366,
      "grad_norm": 0.7921008467674255,
      "learning_rate": 0.00017511503773237622,
      "loss": 1.8748,
      "num_input_tokens_seen": 4093440,
      "step": 682,
      "train_runtime": 687.6535,
      "train_tokens_per_second": 5952.765
    },
    {
      "epoch": 0.2512183908045977,
      "grad_norm": 0.7032368183135986,
      "learning_rate": 0.00017507822565801583,
      "loss": 1.9732,
      "num_input_tokens_seen": 4098272,
      "step": 683,
      "train_runtime": 688.4666,
      "train_tokens_per_second": 5952.754
    },
    {
      "epoch": 0.2515862068965517,
      "grad_norm": 0.6406055092811584,
      "learning_rate": 0.00017504141358365543,
      "loss": 1.7464,
      "num_input_tokens_seen": 4105584,
      "step": 684,
      "train_runtime": 689.6071,
      "train_tokens_per_second": 5953.512
    },
    {
      "epoch": 0.2519540229885057,
      "grad_norm": 0.7085546851158142,
      "learning_rate": 0.00017500460150929504,
      "loss": 1.8475,
      "num_input_tokens_seen": 4110640,
      "step": 685,
      "train_runtime": 690.48,
      "train_tokens_per_second": 5953.307
    },
    {
      "epoch": 0.2523218390804598,
      "grad_norm": 0.6811719536781311,
      "learning_rate": 0.00017496778943493465,
      "loss": 1.9318,
      "num_input_tokens_seen": 4116544,
      "step": 686,
      "train_runtime": 691.4348,
      "train_tokens_per_second": 5953.625
    },
    {
      "epoch": 0.2526896551724138,
      "grad_norm": 0.8263639211654663,
      "learning_rate": 0.00017493097736057426,
      "loss": 1.9458,
      "num_input_tokens_seen": 4122624,
      "step": 687,
      "train_runtime": 692.4211,
      "train_tokens_per_second": 5953.926
    },
    {
      "epoch": 0.25305747126436784,
      "grad_norm": 0.5855770707130432,
      "learning_rate": 0.0001748941652862139,
      "loss": 1.7027,
      "num_input_tokens_seen": 4130608,
      "step": 688,
      "train_runtime": 693.7108,
      "train_tokens_per_second": 5954.366
    },
    {
      "epoch": 0.25342528735632186,
      "grad_norm": 0.6426771879196167,
      "learning_rate": 0.0001748573532118535,
      "loss": 1.7845,
      "num_input_tokens_seen": 4139504,
      "step": 689,
      "train_runtime": 695.0823,
      "train_tokens_per_second": 5955.416
    },
    {
      "epoch": 0.2537931034482759,
      "grad_norm": 0.6769132018089294,
      "learning_rate": 0.00017482054113749312,
      "loss": 1.7137,
      "num_input_tokens_seen": 4144176,
      "step": 690,
      "train_runtime": 695.9074,
      "train_tokens_per_second": 5955.068
    },
    {
      "epoch": 0.2541609195402299,
      "grad_norm": 0.7636768221855164,
      "learning_rate": 0.00017478372906313272,
      "loss": 1.9653,
      "num_input_tokens_seen": 4151456,
      "step": 691,
      "train_runtime": 697.567,
      "train_tokens_per_second": 5951.337
    },
    {
      "epoch": 0.2545287356321839,
      "grad_norm": 0.796562910079956,
      "learning_rate": 0.00017474691698877233,
      "loss": 1.6889,
      "num_input_tokens_seen": 4156288,
      "step": 692,
      "train_runtime": 698.3876,
      "train_tokens_per_second": 5951.262
    },
    {
      "epoch": 0.25489655172413794,
      "grad_norm": 0.7037853598594666,
      "learning_rate": 0.00017471010491441194,
      "loss": 1.65,
      "num_input_tokens_seen": 4165248,
      "step": 693,
      "train_runtime": 699.7729,
      "train_tokens_per_second": 5952.286
    },
    {
      "epoch": 0.25526436781609196,
      "grad_norm": 0.7308499813079834,
      "learning_rate": 0.00017467329284005155,
      "loss": 1.7507,
      "num_input_tokens_seen": 4170512,
      "step": 694,
      "train_runtime": 700.6424,
      "train_tokens_per_second": 5952.412
    },
    {
      "epoch": 0.255632183908046,
      "grad_norm": 0.7657448649406433,
      "learning_rate": 0.00017463648076569113,
      "loss": 1.908,
      "num_input_tokens_seen": 4175328,
      "step": 695,
      "train_runtime": 701.4732,
      "train_tokens_per_second": 5952.227
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.8205829858779907,
      "learning_rate": 0.00017459966869133077,
      "loss": 1.8358,
      "num_input_tokens_seen": 4180928,
      "step": 696,
      "train_runtime": 702.384,
      "train_tokens_per_second": 5952.482
    },
    {
      "epoch": 0.256367816091954,
      "grad_norm": 0.7902839779853821,
      "learning_rate": 0.00017456285661697038,
      "loss": 1.7706,
      "num_input_tokens_seen": 4186256,
      "step": 697,
      "train_runtime": 703.2841,
      "train_tokens_per_second": 5952.439
    },
    {
      "epoch": 0.25673563218390805,
      "grad_norm": 0.7034733891487122,
      "learning_rate": 0.00017452604454260999,
      "loss": 1.7407,
      "num_input_tokens_seen": 4191856,
      "step": 698,
      "train_runtime": 704.2155,
      "train_tokens_per_second": 5952.519
    },
    {
      "epoch": 0.25710344827586207,
      "grad_norm": 0.6806679368019104,
      "learning_rate": 0.0001744892324682496,
      "loss": 1.7302,
      "num_input_tokens_seen": 4197824,
      "step": 699,
      "train_runtime": 705.205,
      "train_tokens_per_second": 5952.63
    },
    {
      "epoch": 0.2574712643678161,
      "grad_norm": 0.8030396699905396,
      "learning_rate": 0.0001744524203938892,
      "loss": 1.9087,
      "num_input_tokens_seen": 4202304,
      "step": 700,
      "train_runtime": 706.0103,
      "train_tokens_per_second": 5952.185
    },
    {
      "epoch": 0.2578390804597701,
      "grad_norm": 0.7756878137588501,
      "learning_rate": 0.0001744156083195288,
      "loss": 1.9351,
      "num_input_tokens_seen": 4208848,
      "step": 701,
      "train_runtime": 707.0844,
      "train_tokens_per_second": 5952.399
    },
    {
      "epoch": 0.25820689655172413,
      "grad_norm": 0.7415671944618225,
      "learning_rate": 0.00017437879624516842,
      "loss": 1.9063,
      "num_input_tokens_seen": 4213776,
      "step": 702,
      "train_runtime": 707.9274,
      "train_tokens_per_second": 5952.271
    },
    {
      "epoch": 0.25857471264367815,
      "grad_norm": 0.8117818832397461,
      "learning_rate": 0.00017434198417080803,
      "loss": 2.0566,
      "num_input_tokens_seen": 4219024,
      "step": 703,
      "train_runtime": 708.8186,
      "train_tokens_per_second": 5952.191
    },
    {
      "epoch": 0.25894252873563217,
      "grad_norm": 0.6948313117027283,
      "learning_rate": 0.00017430517209644764,
      "loss": 1.8695,
      "num_input_tokens_seen": 4223392,
      "step": 704,
      "train_runtime": 709.5814,
      "train_tokens_per_second": 5951.949
    },
    {
      "epoch": 0.2593103448275862,
      "grad_norm": 0.6284450888633728,
      "learning_rate": 0.00017426836002208728,
      "loss": 1.4643,
      "num_input_tokens_seen": 4228720,
      "step": 705,
      "train_runtime": 710.4713,
      "train_tokens_per_second": 5951.993
    },
    {
      "epoch": 0.2596781609195402,
      "grad_norm": 0.5065260529518127,
      "learning_rate": 0.00017423154794772686,
      "loss": 1.4118,
      "num_input_tokens_seen": 4243920,
      "step": 706,
      "train_runtime": 714.2472,
      "train_tokens_per_second": 5941.809
    },
    {
      "epoch": 0.26004597701149423,
      "grad_norm": 0.6901469230651855,
      "learning_rate": 0.00017419473587336647,
      "loss": 1.6742,
      "num_input_tokens_seen": 4249312,
      "step": 707,
      "train_runtime": 715.1463,
      "train_tokens_per_second": 5941.878
    },
    {
      "epoch": 0.26041379310344825,
      "grad_norm": 0.7213504314422607,
      "learning_rate": 0.00017415792379900608,
      "loss": 2.0069,
      "num_input_tokens_seen": 4253968,
      "step": 708,
      "train_runtime": 715.9359,
      "train_tokens_per_second": 5941.828
    },
    {
      "epoch": 0.2607816091954023,
      "grad_norm": 0.6717199683189392,
      "learning_rate": 0.00017412111172464568,
      "loss": 1.6791,
      "num_input_tokens_seen": 4261216,
      "step": 709,
      "train_runtime": 717.1007,
      "train_tokens_per_second": 5942.284
    },
    {
      "epoch": 0.2611494252873563,
      "grad_norm": 0.6818445920944214,
      "learning_rate": 0.0001740842996502853,
      "loss": 1.962,
      "num_input_tokens_seen": 4265776,
      "step": 710,
      "train_runtime": 717.9092,
      "train_tokens_per_second": 5941.944
    },
    {
      "epoch": 0.2615172413793104,
      "grad_norm": 0.6660798788070679,
      "learning_rate": 0.0001740474875759249,
      "loss": 1.8531,
      "num_input_tokens_seen": 4273168,
      "step": 711,
      "train_runtime": 719.0833,
      "train_tokens_per_second": 5942.521
    },
    {
      "epoch": 0.2618850574712644,
      "grad_norm": 0.656058132648468,
      "learning_rate": 0.0001740106755015645,
      "loss": 1.6675,
      "num_input_tokens_seen": 4281264,
      "step": 712,
      "train_runtime": 720.3405,
      "train_tokens_per_second": 5943.389
    },
    {
      "epoch": 0.2622528735632184,
      "grad_norm": 0.6193675994873047,
      "learning_rate": 0.00017397386342720415,
      "loss": 1.793,
      "num_input_tokens_seen": 4286816,
      "step": 713,
      "train_runtime": 721.25,
      "train_tokens_per_second": 5943.592
    },
    {
      "epoch": 0.26262068965517243,
      "grad_norm": 0.7687742114067078,
      "learning_rate": 0.00017393705135284376,
      "loss": 1.7398,
      "num_input_tokens_seen": 4291392,
      "step": 714,
      "train_runtime": 722.0391,
      "train_tokens_per_second": 5943.434
    },
    {
      "epoch": 0.26298850574712646,
      "grad_norm": 0.6832330226898193,
      "learning_rate": 0.00017390023927848336,
      "loss": 1.8785,
      "num_input_tokens_seen": 4297248,
      "step": 715,
      "train_runtime": 722.9793,
      "train_tokens_per_second": 5943.805
    },
    {
      "epoch": 0.2633563218390805,
      "grad_norm": 0.6707386374473572,
      "learning_rate": 0.00017386342720412297,
      "loss": 1.5973,
      "num_input_tokens_seen": 4301632,
      "step": 716,
      "train_runtime": 723.7293,
      "train_tokens_per_second": 5943.703
    },
    {
      "epoch": 0.2637241379310345,
      "grad_norm": 0.6461231112480164,
      "learning_rate": 0.00017382661512976258,
      "loss": 1.7142,
      "num_input_tokens_seen": 4308608,
      "step": 717,
      "train_runtime": 724.8146,
      "train_tokens_per_second": 5944.428
    },
    {
      "epoch": 0.2640919540229885,
      "grad_norm": 0.7029957175254822,
      "learning_rate": 0.0001737898030554022,
      "loss": 1.9736,
      "num_input_tokens_seen": 4315872,
      "step": 718,
      "train_runtime": 725.9703,
      "train_tokens_per_second": 5944.97
    },
    {
      "epoch": 0.26445977011494254,
      "grad_norm": 0.7732644081115723,
      "learning_rate": 0.00017375299098104177,
      "loss": 1.9659,
      "num_input_tokens_seen": 4321856,
      "step": 719,
      "train_runtime": 726.9372,
      "train_tokens_per_second": 5945.295
    },
    {
      "epoch": 0.26482758620689656,
      "grad_norm": 0.5895112752914429,
      "learning_rate": 0.00017371617890668138,
      "loss": 1.5641,
      "num_input_tokens_seen": 4330720,
      "step": 720,
      "train_runtime": 728.3081,
      "train_tokens_per_second": 5946.274
    },
    {
      "epoch": 0.2651954022988506,
      "grad_norm": 0.7460876107215881,
      "learning_rate": 0.000173679366832321,
      "loss": 1.7145,
      "num_input_tokens_seen": 4336208,
      "step": 721,
      "train_runtime": 729.7049,
      "train_tokens_per_second": 5942.413
    },
    {
      "epoch": 0.2655632183908046,
      "grad_norm": 0.7500301003456116,
      "learning_rate": 0.00017364255475796063,
      "loss": 1.9684,
      "num_input_tokens_seen": 4341056,
      "step": 722,
      "train_runtime": 730.5126,
      "train_tokens_per_second": 5942.479
    },
    {
      "epoch": 0.2659310344827586,
      "grad_norm": 0.808393657207489,
      "learning_rate": 0.00017360574268360024,
      "loss": 1.8444,
      "num_input_tokens_seen": 4346256,
      "step": 723,
      "train_runtime": 731.3951,
      "train_tokens_per_second": 5942.419
    },
    {
      "epoch": 0.26629885057471264,
      "grad_norm": 0.7359159588813782,
      "learning_rate": 0.00017356893060923984,
      "loss": 1.9504,
      "num_input_tokens_seen": 4351632,
      "step": 724,
      "train_runtime": 732.3094,
      "train_tokens_per_second": 5942.341
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 0.715316116809845,
      "learning_rate": 0.00017353211853487945,
      "loss": 1.8642,
      "num_input_tokens_seen": 4356512,
      "step": 725,
      "train_runtime": 733.1466,
      "train_tokens_per_second": 5942.211
    },
    {
      "epoch": 0.2670344827586207,
      "grad_norm": 0.7524330615997314,
      "learning_rate": 0.00017349530646051906,
      "loss": 1.9275,
      "num_input_tokens_seen": 4362688,
      "step": 726,
      "train_runtime": 734.1415,
      "train_tokens_per_second": 5942.571
    },
    {
      "epoch": 0.2674022988505747,
      "grad_norm": 0.6848141551017761,
      "learning_rate": 0.00017345849438615867,
      "loss": 1.7619,
      "num_input_tokens_seen": 4367984,
      "step": 727,
      "train_runtime": 735.0438,
      "train_tokens_per_second": 5942.481
    },
    {
      "epoch": 0.2677701149425287,
      "grad_norm": 0.7257616519927979,
      "learning_rate": 0.00017342168231179828,
      "loss": 1.7384,
      "num_input_tokens_seen": 4374112,
      "step": 728,
      "train_runtime": 736.0324,
      "train_tokens_per_second": 5942.826
    },
    {
      "epoch": 0.26813793103448275,
      "grad_norm": 0.7487912178039551,
      "learning_rate": 0.0001733848702374379,
      "loss": 1.8576,
      "num_input_tokens_seen": 4378880,
      "step": 729,
      "train_runtime": 736.8682,
      "train_tokens_per_second": 5942.555
    },
    {
      "epoch": 0.26850574712643677,
      "grad_norm": 0.7400490045547485,
      "learning_rate": 0.0001733480581630775,
      "loss": 1.8275,
      "num_input_tokens_seen": 4384432,
      "step": 730,
      "train_runtime": 737.7872,
      "train_tokens_per_second": 5942.679
    },
    {
      "epoch": 0.2688735632183908,
      "grad_norm": 0.7326444983482361,
      "learning_rate": 0.0001733112460887171,
      "loss": 1.9786,
      "num_input_tokens_seen": 4388800,
      "step": 731,
      "train_runtime": 738.5374,
      "train_tokens_per_second": 5942.556
    },
    {
      "epoch": 0.2692413793103448,
      "grad_norm": 0.6822432279586792,
      "learning_rate": 0.00017327443401435672,
      "loss": 1.7764,
      "num_input_tokens_seen": 4393968,
      "step": 732,
      "train_runtime": 739.4289,
      "train_tokens_per_second": 5942.381
    },
    {
      "epoch": 0.26960919540229883,
      "grad_norm": 0.7900007367134094,
      "learning_rate": 0.00017323762193999632,
      "loss": 1.9678,
      "num_input_tokens_seen": 4400048,
      "step": 733,
      "train_runtime": 740.4284,
      "train_tokens_per_second": 5942.571
    },
    {
      "epoch": 0.26997701149425285,
      "grad_norm": 0.7073249220848083,
      "learning_rate": 0.00017320080986563593,
      "loss": 1.7936,
      "num_input_tokens_seen": 4404960,
      "step": 734,
      "train_runtime": 741.2879,
      "train_tokens_per_second": 5942.307
    },
    {
      "epoch": 0.27034482758620687,
      "grad_norm": 0.7599587440490723,
      "learning_rate": 0.00017316399779127554,
      "loss": 1.8764,
      "num_input_tokens_seen": 4409920,
      "step": 735,
      "train_runtime": 742.1208,
      "train_tokens_per_second": 5942.321
    },
    {
      "epoch": 0.27071264367816095,
      "grad_norm": 0.7730521559715271,
      "learning_rate": 0.00017312718571691515,
      "loss": 1.7448,
      "num_input_tokens_seen": 4414752,
      "step": 736,
      "train_runtime": 742.9667,
      "train_tokens_per_second": 5942.059
    },
    {
      "epoch": 0.27108045977011497,
      "grad_norm": 0.7778670191764832,
      "learning_rate": 0.00017309037364255476,
      "loss": 1.9152,
      "num_input_tokens_seen": 4420416,
      "step": 737,
      "train_runtime": 743.9134,
      "train_tokens_per_second": 5942.111
    },
    {
      "epoch": 0.271448275862069,
      "grad_norm": 0.7406691312789917,
      "learning_rate": 0.00017305356156819437,
      "loss": 1.7868,
      "num_input_tokens_seen": 4426352,
      "step": 738,
      "train_runtime": 744.8944,
      "train_tokens_per_second": 5942.255
    },
    {
      "epoch": 0.271816091954023,
      "grad_norm": 0.72142094373703,
      "learning_rate": 0.000173016749493834,
      "loss": 1.919,
      "num_input_tokens_seen": 4431264,
      "step": 739,
      "train_runtime": 745.7425,
      "train_tokens_per_second": 5942.084
    },
    {
      "epoch": 0.27218390804597703,
      "grad_norm": 0.7652568221092224,
      "learning_rate": 0.00017297993741947361,
      "loss": 1.9263,
      "num_input_tokens_seen": 4438400,
      "step": 740,
      "train_runtime": 746.8502,
      "train_tokens_per_second": 5942.825
    },
    {
      "epoch": 0.27255172413793105,
      "grad_norm": 0.9410164952278137,
      "learning_rate": 0.00017294312534511322,
      "loss": 1.78,
      "num_input_tokens_seen": 4444064,
      "step": 741,
      "train_runtime": 747.7511,
      "train_tokens_per_second": 5943.24
    },
    {
      "epoch": 0.27291954022988507,
      "grad_norm": 0.7142931818962097,
      "learning_rate": 0.0001729063132707528,
      "loss": 1.7084,
      "num_input_tokens_seen": 4448624,
      "step": 742,
      "train_runtime": 748.5382,
      "train_tokens_per_second": 5943.082
    },
    {
      "epoch": 0.2732873563218391,
      "grad_norm": 0.8017459511756897,
      "learning_rate": 0.00017286950119639241,
      "loss": 1.65,
      "num_input_tokens_seen": 4453440,
      "step": 743,
      "train_runtime": 749.3396,
      "train_tokens_per_second": 5943.153
    },
    {
      "epoch": 0.2736551724137931,
      "grad_norm": 0.7462841272354126,
      "learning_rate": 0.00017283268912203202,
      "loss": 1.9961,
      "num_input_tokens_seen": 4457808,
      "step": 744,
      "train_runtime": 750.1167,
      "train_tokens_per_second": 5942.82
    },
    {
      "epoch": 0.27402298850574713,
      "grad_norm": 0.7649838328361511,
      "learning_rate": 0.00017279587704767163,
      "loss": 1.6771,
      "num_input_tokens_seen": 4463056,
      "step": 745,
      "train_runtime": 750.9803,
      "train_tokens_per_second": 5942.973
    },
    {
      "epoch": 0.27439080459770115,
      "grad_norm": 0.6718977093696594,
      "learning_rate": 0.00017275906497331124,
      "loss": 1.7933,
      "num_input_tokens_seen": 4469776,
      "step": 746,
      "train_runtime": 752.0665,
      "train_tokens_per_second": 5943.326
    },
    {
      "epoch": 0.2747586206896552,
      "grad_norm": 0.6736374497413635,
      "learning_rate": 0.00017272225289895085,
      "loss": 1.8177,
      "num_input_tokens_seen": 4477584,
      "step": 747,
      "train_runtime": 753.2957,
      "train_tokens_per_second": 5943.993
    },
    {
      "epoch": 0.2751264367816092,
      "grad_norm": 0.8791332244873047,
      "learning_rate": 0.00017268544082459049,
      "loss": 1.9928,
      "num_input_tokens_seen": 4485040,
      "step": 748,
      "train_runtime": 754.491,
      "train_tokens_per_second": 5944.458
    },
    {
      "epoch": 0.2754942528735632,
      "grad_norm": 0.7467145323753357,
      "learning_rate": 0.0001726486287502301,
      "loss": 1.8763,
      "num_input_tokens_seen": 4491648,
      "step": 749,
      "train_runtime": 755.5528,
      "train_tokens_per_second": 5944.85
    },
    {
      "epoch": 0.27586206896551724,
      "grad_norm": 0.7980650067329407,
      "learning_rate": 0.0001726118166758697,
      "loss": 1.9774,
      "num_input_tokens_seen": 4495520,
      "step": 750,
      "train_runtime": 756.2171,
      "train_tokens_per_second": 5944.748
    },
    {
      "epoch": 0.27622988505747126,
      "grad_norm": 0.7379633188247681,
      "learning_rate": 0.0001725750046015093,
      "loss": 1.7208,
      "num_input_tokens_seen": 4503536,
      "step": 751,
      "train_runtime": 757.961,
      "train_tokens_per_second": 5941.646
    },
    {
      "epoch": 0.2765977011494253,
      "grad_norm": 0.6900970339775085,
      "learning_rate": 0.00017253819252714892,
      "loss": 1.7793,
      "num_input_tokens_seen": 4508544,
      "step": 752,
      "train_runtime": 758.8041,
      "train_tokens_per_second": 5941.644
    },
    {
      "epoch": 0.2769655172413793,
      "grad_norm": 0.7257111668586731,
      "learning_rate": 0.00017250138045278853,
      "loss": 1.8201,
      "num_input_tokens_seen": 4514080,
      "step": 753,
      "train_runtime": 759.7166,
      "train_tokens_per_second": 5941.795
    },
    {
      "epoch": 0.2773333333333333,
      "grad_norm": 0.7619184255599976,
      "learning_rate": 0.00017246456837842814,
      "loss": 1.8787,
      "num_input_tokens_seen": 4519264,
      "step": 754,
      "train_runtime": 760.589,
      "train_tokens_per_second": 5941.795
    },
    {
      "epoch": 0.27770114942528734,
      "grad_norm": 0.7413751482963562,
      "learning_rate": 0.00017242775630406772,
      "loss": 1.6194,
      "num_input_tokens_seen": 4523792,
      "step": 755,
      "train_runtime": 761.3815,
      "train_tokens_per_second": 5941.558
    },
    {
      "epoch": 0.27806896551724136,
      "grad_norm": 0.7797419428825378,
      "learning_rate": 0.00017239094422970736,
      "loss": 1.7882,
      "num_input_tokens_seen": 4529664,
      "step": 756,
      "train_runtime": 762.342,
      "train_tokens_per_second": 5941.774
    },
    {
      "epoch": 0.2784367816091954,
      "grad_norm": 0.7607048153877258,
      "learning_rate": 0.00017235413215534697,
      "loss": 1.812,
      "num_input_tokens_seen": 4535472,
      "step": 757,
      "train_runtime": 763.272,
      "train_tokens_per_second": 5942.144
    },
    {
      "epoch": 0.2788045977011494,
      "grad_norm": 0.6602902412414551,
      "learning_rate": 0.00017231732008098657,
      "loss": 1.7064,
      "num_input_tokens_seen": 4539904,
      "step": 758,
      "train_runtime": 764.0453,
      "train_tokens_per_second": 5941.931
    },
    {
      "epoch": 0.2791724137931034,
      "grad_norm": 0.7725996971130371,
      "learning_rate": 0.00017228050800662618,
      "loss": 1.8908,
      "num_input_tokens_seen": 4545040,
      "step": 759,
      "train_runtime": 764.9229,
      "train_tokens_per_second": 5941.827
    },
    {
      "epoch": 0.27954022988505745,
      "grad_norm": 0.7709255218505859,
      "learning_rate": 0.0001722436959322658,
      "loss": 1.8516,
      "num_input_tokens_seen": 4550032,
      "step": 760,
      "train_runtime": 765.7739,
      "train_tokens_per_second": 5941.743
    },
    {
      "epoch": 0.2799080459770115,
      "grad_norm": 0.744976282119751,
      "learning_rate": 0.0001722068838579054,
      "loss": 1.7581,
      "num_input_tokens_seen": 4554784,
      "step": 761,
      "train_runtime": 766.6043,
      "train_tokens_per_second": 5941.506
    },
    {
      "epoch": 0.28027586206896554,
      "grad_norm": 0.7730522155761719,
      "learning_rate": 0.000172170071783545,
      "loss": 1.7988,
      "num_input_tokens_seen": 4560528,
      "step": 762,
      "train_runtime": 767.5414,
      "train_tokens_per_second": 5941.736
    },
    {
      "epoch": 0.28064367816091956,
      "grad_norm": 0.7115368247032166,
      "learning_rate": 0.00017213325970918462,
      "loss": 1.9102,
      "num_input_tokens_seen": 4565328,
      "step": 763,
      "train_runtime": 768.3794,
      "train_tokens_per_second": 5941.503
    },
    {
      "epoch": 0.2810114942528736,
      "grad_norm": 0.6897196173667908,
      "learning_rate": 0.00017209644763482423,
      "loss": 1.6402,
      "num_input_tokens_seen": 4571824,
      "step": 764,
      "train_runtime": 769.4388,
      "train_tokens_per_second": 5941.764
    },
    {
      "epoch": 0.2813793103448276,
      "grad_norm": 0.8004782795906067,
      "learning_rate": 0.00017205963556046384,
      "loss": 2.0911,
      "num_input_tokens_seen": 4576768,
      "step": 765,
      "train_runtime": 770.2905,
      "train_tokens_per_second": 5941.613
    },
    {
      "epoch": 0.2817471264367816,
      "grad_norm": 0.7114158272743225,
      "learning_rate": 0.00017202282348610345,
      "loss": 1.9612,
      "num_input_tokens_seen": 4584576,
      "step": 766,
      "train_runtime": 771.5067,
      "train_tokens_per_second": 5942.367
    },
    {
      "epoch": 0.28211494252873565,
      "grad_norm": 0.7624607086181641,
      "learning_rate": 0.00017198601141174305,
      "loss": 1.834,
      "num_input_tokens_seen": 4589968,
      "step": 767,
      "train_runtime": 772.4216,
      "train_tokens_per_second": 5942.309
    },
    {
      "epoch": 0.28248275862068967,
      "grad_norm": 0.7275245785713196,
      "learning_rate": 0.00017194919933738266,
      "loss": 1.7461,
      "num_input_tokens_seen": 4596208,
      "step": 768,
      "train_runtime": 773.4588,
      "train_tokens_per_second": 5942.408
    },
    {
      "epoch": 0.2828505747126437,
      "grad_norm": 0.7166267037391663,
      "learning_rate": 0.00017191238726302227,
      "loss": 1.7845,
      "num_input_tokens_seen": 4601552,
      "step": 769,
      "train_runtime": 774.3621,
      "train_tokens_per_second": 5942.378
    },
    {
      "epoch": 0.2832183908045977,
      "grad_norm": 0.6857694983482361,
      "learning_rate": 0.00017187557518866188,
      "loss": 1.8959,
      "num_input_tokens_seen": 4606320,
      "step": 770,
      "train_runtime": 775.2022,
      "train_tokens_per_second": 5942.088
    },
    {
      "epoch": 0.28358620689655173,
      "grad_norm": 0.7865335941314697,
      "learning_rate": 0.0001718387631143015,
      "loss": 1.9186,
      "num_input_tokens_seen": 4610576,
      "step": 771,
      "train_runtime": 775.9339,
      "train_tokens_per_second": 5941.97
    },
    {
      "epoch": 0.28395402298850575,
      "grad_norm": 0.6964818835258484,
      "learning_rate": 0.0001718019510399411,
      "loss": 1.7916,
      "num_input_tokens_seen": 4619344,
      "step": 772,
      "train_runtime": 777.269,
      "train_tokens_per_second": 5943.044
    },
    {
      "epoch": 0.28432183908045977,
      "grad_norm": 0.7649399042129517,
      "learning_rate": 0.00017176513896558073,
      "loss": 1.8707,
      "num_input_tokens_seen": 4624352,
      "step": 773,
      "train_runtime": 778.1065,
      "train_tokens_per_second": 5943.083
    },
    {
      "epoch": 0.2846896551724138,
      "grad_norm": 0.7283154726028442,
      "learning_rate": 0.00017172832689122034,
      "loss": 1.7958,
      "num_input_tokens_seen": 4634224,
      "step": 774,
      "train_runtime": 779.6155,
      "train_tokens_per_second": 5944.243
    },
    {
      "epoch": 0.2850574712643678,
      "grad_norm": 0.7610809803009033,
      "learning_rate": 0.00017169151481685995,
      "loss": 1.8795,
      "num_input_tokens_seen": 4639072,
      "step": 775,
      "train_runtime": 780.4551,
      "train_tokens_per_second": 5944.06
    },
    {
      "epoch": 0.28542528735632183,
      "grad_norm": 0.7797113060951233,
      "learning_rate": 0.00017165470274249956,
      "loss": 1.9561,
      "num_input_tokens_seen": 4646432,
      "step": 776,
      "train_runtime": 781.618,
      "train_tokens_per_second": 5944.633
    },
    {
      "epoch": 0.28579310344827585,
      "grad_norm": 0.749673068523407,
      "learning_rate": 0.00017161789066813917,
      "loss": 1.8177,
      "num_input_tokens_seen": 4650320,
      "step": 777,
      "train_runtime": 782.3051,
      "train_tokens_per_second": 5944.381
    },
    {
      "epoch": 0.2861609195402299,
      "grad_norm": 0.6641862988471985,
      "learning_rate": 0.00017158107859377878,
      "loss": 1.7552,
      "num_input_tokens_seen": 4656768,
      "step": 778,
      "train_runtime": 783.373,
      "train_tokens_per_second": 5944.509
    },
    {
      "epoch": 0.2865287356321839,
      "grad_norm": 0.7170074582099915,
      "learning_rate": 0.00017154426651941836,
      "loss": 1.8002,
      "num_input_tokens_seen": 4662272,
      "step": 779,
      "train_runtime": 784.2918,
      "train_tokens_per_second": 5944.563
    },
    {
      "epoch": 0.2868965517241379,
      "grad_norm": 0.7169869542121887,
      "learning_rate": 0.00017150745444505797,
      "loss": 1.6444,
      "num_input_tokens_seen": 4666544,
      "step": 780,
      "train_runtime": 785.0604,
      "train_tokens_per_second": 5944.185
    },
    {
      "epoch": 0.28726436781609194,
      "grad_norm": 0.6579683423042297,
      "learning_rate": 0.00017147064237069758,
      "loss": 1.6995,
      "num_input_tokens_seen": 4672016,
      "step": 781,
      "train_runtime": 786.4483,
      "train_tokens_per_second": 5940.653
    },
    {
      "epoch": 0.28763218390804596,
      "grad_norm": 0.7201480269432068,
      "learning_rate": 0.00017143383029633721,
      "loss": 1.9416,
      "num_input_tokens_seen": 4677088,
      "step": 782,
      "train_runtime": 787.3115,
      "train_tokens_per_second": 5940.582
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.7196846604347229,
      "learning_rate": 0.00017139701822197682,
      "loss": 1.7708,
      "num_input_tokens_seen": 4682448,
      "step": 783,
      "train_runtime": 788.2175,
      "train_tokens_per_second": 5940.553
    },
    {
      "epoch": 0.288367816091954,
      "grad_norm": 0.7448524832725525,
      "learning_rate": 0.00017136020614761643,
      "loss": 1.9324,
      "num_input_tokens_seen": 4687888,
      "step": 784,
      "train_runtime": 789.1208,
      "train_tokens_per_second": 5940.647
    },
    {
      "epoch": 0.288735632183908,
      "grad_norm": 0.6804062128067017,
      "learning_rate": 0.00017132339407325604,
      "loss": 1.6276,
      "num_input_tokens_seen": 4692800,
      "step": 785,
      "train_runtime": 789.9745,
      "train_tokens_per_second": 5940.445
    },
    {
      "epoch": 0.2891034482758621,
      "grad_norm": 0.7315980195999146,
      "learning_rate": 0.00017128658199889565,
      "loss": 1.7298,
      "num_input_tokens_seen": 4697968,
      "step": 786,
      "train_runtime": 790.835,
      "train_tokens_per_second": 5940.516
    },
    {
      "epoch": 0.2894712643678161,
      "grad_norm": 0.8079400062561035,
      "learning_rate": 0.00017124976992453526,
      "loss": 1.7128,
      "num_input_tokens_seen": 4704944,
      "step": 787,
      "train_runtime": 791.9792,
      "train_tokens_per_second": 5940.742
    },
    {
      "epoch": 0.28983908045977014,
      "grad_norm": 0.80372554063797,
      "learning_rate": 0.00017121295785017487,
      "loss": 1.8274,
      "num_input_tokens_seen": 4711152,
      "step": 788,
      "train_runtime": 792.991,
      "train_tokens_per_second": 5940.991
    },
    {
      "epoch": 0.29020689655172416,
      "grad_norm": 0.7757269144058228,
      "learning_rate": 0.00017117614577581448,
      "loss": 2.0107,
      "num_input_tokens_seen": 4716448,
      "step": 789,
      "train_runtime": 793.8848,
      "train_tokens_per_second": 5940.973
    },
    {
      "epoch": 0.2905747126436782,
      "grad_norm": 0.712533712387085,
      "learning_rate": 0.00017113933370145409,
      "loss": 1.8902,
      "num_input_tokens_seen": 4724096,
      "step": 790,
      "train_runtime": 795.0921,
      "train_tokens_per_second": 5941.571
    },
    {
      "epoch": 0.2909425287356322,
      "grad_norm": 0.6622461676597595,
      "learning_rate": 0.0001711025216270937,
      "loss": 1.8504,
      "num_input_tokens_seen": 4729712,
      "step": 791,
      "train_runtime": 796.0393,
      "train_tokens_per_second": 5941.556
    },
    {
      "epoch": 0.2913103448275862,
      "grad_norm": 0.8801368474960327,
      "learning_rate": 0.0001710657095527333,
      "loss": 2.1268,
      "num_input_tokens_seen": 4734576,
      "step": 792,
      "train_runtime": 796.8737,
      "train_tokens_per_second": 5941.438
    },
    {
      "epoch": 0.29167816091954024,
      "grad_norm": 0.7598217725753784,
      "learning_rate": 0.0001710288974783729,
      "loss": 1.8334,
      "num_input_tokens_seen": 4740880,
      "step": 793,
      "train_runtime": 797.8697,
      "train_tokens_per_second": 5941.923
    },
    {
      "epoch": 0.29204597701149426,
      "grad_norm": 0.8002123236656189,
      "learning_rate": 0.00017099208540401252,
      "loss": 1.8155,
      "num_input_tokens_seen": 4746560,
      "step": 794,
      "train_runtime": 798.8056,
      "train_tokens_per_second": 5942.071
    },
    {
      "epoch": 0.2924137931034483,
      "grad_norm": 0.87678062915802,
      "learning_rate": 0.00017095527332965213,
      "loss": 1.7733,
      "num_input_tokens_seen": 4754416,
      "step": 795,
      "train_runtime": 800.0321,
      "train_tokens_per_second": 5942.782
    },
    {
      "epoch": 0.2927816091954023,
      "grad_norm": 0.7613138556480408,
      "learning_rate": 0.00017091846125529174,
      "loss": 1.7808,
      "num_input_tokens_seen": 4759264,
      "step": 796,
      "train_runtime": 800.8724,
      "train_tokens_per_second": 5942.599
    },
    {
      "epoch": 0.2931494252873563,
      "grad_norm": 0.6966599822044373,
      "learning_rate": 0.00017088164918093135,
      "loss": 1.8375,
      "num_input_tokens_seen": 4765280,
      "step": 797,
      "train_runtime": 801.8498,
      "train_tokens_per_second": 5942.859
    },
    {
      "epoch": 0.29351724137931035,
      "grad_norm": 0.7675111293792725,
      "learning_rate": 0.00017084483710657096,
      "loss": 1.9598,
      "num_input_tokens_seen": 4770608,
      "step": 798,
      "train_runtime": 802.7226,
      "train_tokens_per_second": 5943.035
    },
    {
      "epoch": 0.29388505747126437,
      "grad_norm": 0.7366241812705994,
      "learning_rate": 0.0001708080250322106,
      "loss": 1.6794,
      "num_input_tokens_seen": 4777328,
      "step": 799,
      "train_runtime": 803.8293,
      "train_tokens_per_second": 5943.212
    },
    {
      "epoch": 0.2942528735632184,
      "grad_norm": 0.812473714351654,
      "learning_rate": 0.0001707712129578502,
      "loss": 1.9036,
      "num_input_tokens_seen": 4782784,
      "step": 800,
      "train_runtime": 804.7709,
      "train_tokens_per_second": 5943.038
    },
    {
      "epoch": 0.2946206896551724,
      "grad_norm": 0.7038021683692932,
      "learning_rate": 0.0001707344008834898,
      "loss": 1.8064,
      "num_input_tokens_seen": 4790080,
      "step": 801,
      "train_runtime": 805.9535,
      "train_tokens_per_second": 5943.37
    },
    {
      "epoch": 0.29498850574712643,
      "grad_norm": 0.6473475098609924,
      "learning_rate": 0.0001706975888091294,
      "loss": 1.7695,
      "num_input_tokens_seen": 4797008,
      "step": 802,
      "train_runtime": 807.0647,
      "train_tokens_per_second": 5943.772
    },
    {
      "epoch": 0.29535632183908045,
      "grad_norm": 0.6907509565353394,
      "learning_rate": 0.000170660776734769,
      "loss": 1.7874,
      "num_input_tokens_seen": 4802064,
      "step": 803,
      "train_runtime": 807.8913,
      "train_tokens_per_second": 5943.948
    },
    {
      "epoch": 0.29572413793103447,
      "grad_norm": 0.7124350666999817,
      "learning_rate": 0.0001706239646604086,
      "loss": 1.9215,
      "num_input_tokens_seen": 4806672,
      "step": 804,
      "train_runtime": 808.7056,
      "train_tokens_per_second": 5943.661
    },
    {
      "epoch": 0.2960919540229885,
      "grad_norm": 0.747387170791626,
      "learning_rate": 0.00017058715258604822,
      "loss": 1.916,
      "num_input_tokens_seen": 4813040,
      "step": 805,
      "train_runtime": 809.7316,
      "train_tokens_per_second": 5943.994
    },
    {
      "epoch": 0.2964597701149425,
      "grad_norm": 0.7358576059341431,
      "learning_rate": 0.00017055034051168783,
      "loss": 1.7624,
      "num_input_tokens_seen": 4818416,
      "step": 806,
      "train_runtime": 810.6289,
      "train_tokens_per_second": 5944.047
    },
    {
      "epoch": 0.29682758620689653,
      "grad_norm": 0.7139478325843811,
      "learning_rate": 0.00017051352843732744,
      "loss": 1.8146,
      "num_input_tokens_seen": 4824656,
      "step": 807,
      "train_runtime": 811.6399,
      "train_tokens_per_second": 5944.33
    },
    {
      "epoch": 0.29719540229885055,
      "grad_norm": 0.7675763964653015,
      "learning_rate": 0.00017047671636296707,
      "loss": 2.0133,
      "num_input_tokens_seen": 4829056,
      "step": 808,
      "train_runtime": 812.3882,
      "train_tokens_per_second": 5944.271
    },
    {
      "epoch": 0.2975632183908046,
      "grad_norm": 0.6864924430847168,
      "learning_rate": 0.00017043990428860668,
      "loss": 1.8765,
      "num_input_tokens_seen": 4833952,
      "step": 809,
      "train_runtime": 813.2253,
      "train_tokens_per_second": 5944.173
    },
    {
      "epoch": 0.2979310344827586,
      "grad_norm": 0.5202162265777588,
      "learning_rate": 0.0001704030922142463,
      "loss": 1.4446,
      "num_input_tokens_seen": 4855168,
      "step": 810,
      "train_runtime": 816.3312,
      "train_tokens_per_second": 5947.547
    },
    {
      "epoch": 0.29829885057471267,
      "grad_norm": 0.7769670486450195,
      "learning_rate": 0.0001703662801398859,
      "loss": 1.7737,
      "num_input_tokens_seen": 4860800,
      "step": 811,
      "train_runtime": 817.762,
      "train_tokens_per_second": 5944.028
    },
    {
      "epoch": 0.2986666666666667,
      "grad_norm": 0.6722753047943115,
      "learning_rate": 0.0001703294680655255,
      "loss": 1.867,
      "num_input_tokens_seen": 4866144,
      "step": 812,
      "train_runtime": 818.6679,
      "train_tokens_per_second": 5943.978
    },
    {
      "epoch": 0.2990344827586207,
      "grad_norm": 0.675029456615448,
      "learning_rate": 0.00017029265599116512,
      "loss": 1.6928,
      "num_input_tokens_seen": 4870896,
      "step": 813,
      "train_runtime": 819.4896,
      "train_tokens_per_second": 5943.817
    },
    {
      "epoch": 0.29940229885057473,
      "grad_norm": 0.716803252696991,
      "learning_rate": 0.00017025584391680473,
      "loss": 1.7346,
      "num_input_tokens_seen": 4878736,
      "step": 814,
      "train_runtime": 820.7103,
      "train_tokens_per_second": 5944.529
    },
    {
      "epoch": 0.29977011494252875,
      "grad_norm": 0.7450390458106995,
      "learning_rate": 0.0001702190318424443,
      "loss": 1.7828,
      "num_input_tokens_seen": 4884576,
      "step": 815,
      "train_runtime": 821.6753,
      "train_tokens_per_second": 5944.655
    },
    {
      "epoch": 0.3001379310344828,
      "grad_norm": 0.6824129223823547,
      "learning_rate": 0.00017018221976808394,
      "loss": 1.7726,
      "num_input_tokens_seen": 4890496,
      "step": 816,
      "train_runtime": 822.6567,
      "train_tokens_per_second": 5944.759
    },
    {
      "epoch": 0.3005057471264368,
      "grad_norm": 0.6666625142097473,
      "learning_rate": 0.00017014540769372355,
      "loss": 1.6826,
      "num_input_tokens_seen": 4902432,
      "step": 817,
      "train_runtime": 824.4318,
      "train_tokens_per_second": 5946.438
    },
    {
      "epoch": 0.3008735632183908,
      "grad_norm": 0.7873861789703369,
      "learning_rate": 0.00017010859561936316,
      "loss": 1.749,
      "num_input_tokens_seen": 4909008,
      "step": 818,
      "train_runtime": 825.4817,
      "train_tokens_per_second": 5946.841
    },
    {
      "epoch": 0.30124137931034484,
      "grad_norm": 0.7500694990158081,
      "learning_rate": 0.00017007178354500277,
      "loss": 1.7682,
      "num_input_tokens_seen": 4918288,
      "step": 819,
      "train_runtime": 826.8888,
      "train_tokens_per_second": 5947.944
    },
    {
      "epoch": 0.30160919540229886,
      "grad_norm": 0.726616621017456,
      "learning_rate": 0.00017003497147064238,
      "loss": 1.8452,
      "num_input_tokens_seen": 4923248,
      "step": 820,
      "train_runtime": 827.7506,
      "train_tokens_per_second": 5947.743
    },
    {
      "epoch": 0.3019770114942529,
      "grad_norm": 0.6488540768623352,
      "learning_rate": 0.000169998159396282,
      "loss": 1.7898,
      "num_input_tokens_seen": 4929840,
      "step": 821,
      "train_runtime": 828.7956,
      "train_tokens_per_second": 5948.198
    },
    {
      "epoch": 0.3023448275862069,
      "grad_norm": 0.6533604860305786,
      "learning_rate": 0.0001699613473219216,
      "loss": 1.8328,
      "num_input_tokens_seen": 4937088,
      "step": 822,
      "train_runtime": 829.9508,
      "train_tokens_per_second": 5948.651
    },
    {
      "epoch": 0.3027126436781609,
      "grad_norm": 0.7379357814788818,
      "learning_rate": 0.0001699245352475612,
      "loss": 1.9164,
      "num_input_tokens_seen": 4944816,
      "step": 823,
      "train_runtime": 831.1306,
      "train_tokens_per_second": 5949.506
    },
    {
      "epoch": 0.30308045977011494,
      "grad_norm": 0.6995763182640076,
      "learning_rate": 0.00016988772317320082,
      "loss": 1.731,
      "num_input_tokens_seen": 4949472,
      "step": 824,
      "train_runtime": 831.9533,
      "train_tokens_per_second": 5949.219
    },
    {
      "epoch": 0.30344827586206896,
      "grad_norm": 0.7361706495285034,
      "learning_rate": 0.00016985091109884042,
      "loss": 2.0509,
      "num_input_tokens_seen": 4954384,
      "step": 825,
      "train_runtime": 832.8051,
      "train_tokens_per_second": 5949.032
    },
    {
      "epoch": 0.303816091954023,
      "grad_norm": 0.6709271669387817,
      "learning_rate": 0.00016981409902448003,
      "loss": 1.766,
      "num_input_tokens_seen": 4960112,
      "step": 826,
      "train_runtime": 833.7595,
      "train_tokens_per_second": 5949.092
    },
    {
      "epoch": 0.304183908045977,
      "grad_norm": 0.6723732352256775,
      "learning_rate": 0.00016977728695011964,
      "loss": 1.6995,
      "num_input_tokens_seen": 4969952,
      "step": 827,
      "train_runtime": 835.2765,
      "train_tokens_per_second": 5950.068
    },
    {
      "epoch": 0.304551724137931,
      "grad_norm": 0.726488471031189,
      "learning_rate": 0.00016974047487575925,
      "loss": 1.9395,
      "num_input_tokens_seen": 4974688,
      "step": 828,
      "train_runtime": 836.0869,
      "train_tokens_per_second": 5949.965
    },
    {
      "epoch": 0.30491954022988504,
      "grad_norm": 0.6773337721824646,
      "learning_rate": 0.00016970366280139886,
      "loss": 1.8526,
      "num_input_tokens_seen": 4979744,
      "step": 829,
      "train_runtime": 836.9521,
      "train_tokens_per_second": 5949.855
    },
    {
      "epoch": 0.30528735632183907,
      "grad_norm": 0.6449316740036011,
      "learning_rate": 0.00016966685072703847,
      "loss": 1.7464,
      "num_input_tokens_seen": 4985424,
      "step": 830,
      "train_runtime": 837.8923,
      "train_tokens_per_second": 5949.958
    },
    {
      "epoch": 0.3056551724137931,
      "grad_norm": 0.6937980055809021,
      "learning_rate": 0.00016963003865267808,
      "loss": 1.8806,
      "num_input_tokens_seen": 4991264,
      "step": 831,
      "train_runtime": 838.8458,
      "train_tokens_per_second": 5950.157
    },
    {
      "epoch": 0.3060229885057471,
      "grad_norm": 0.6817101836204529,
      "learning_rate": 0.00016959322657831769,
      "loss": 1.7833,
      "num_input_tokens_seen": 4996576,
      "step": 832,
      "train_runtime": 839.745,
      "train_tokens_per_second": 5950.111
    },
    {
      "epoch": 0.3063908045977011,
      "grad_norm": 0.7633034586906433,
      "learning_rate": 0.00016955641450395732,
      "loss": 1.9476,
      "num_input_tokens_seen": 5002288,
      "step": 833,
      "train_runtime": 840.6896,
      "train_tokens_per_second": 5950.22
    },
    {
      "epoch": 0.30675862068965515,
      "grad_norm": 0.7699156403541565,
      "learning_rate": 0.00016951960242959693,
      "loss": 1.8108,
      "num_input_tokens_seen": 5006688,
      "step": 834,
      "train_runtime": 841.4499,
      "train_tokens_per_second": 5950.073
    },
    {
      "epoch": 0.30712643678160917,
      "grad_norm": 0.7813035845756531,
      "learning_rate": 0.00016948279035523654,
      "loss": 1.882,
      "num_input_tokens_seen": 5012448,
      "step": 835,
      "train_runtime": 842.3948,
      "train_tokens_per_second": 5950.236
    },
    {
      "epoch": 0.30749425287356325,
      "grad_norm": 0.7035843133926392,
      "learning_rate": 0.00016944597828087615,
      "loss": 1.794,
      "num_input_tokens_seen": 5017168,
      "step": 836,
      "train_runtime": 843.2245,
      "train_tokens_per_second": 5949.979
    },
    {
      "epoch": 0.30786206896551727,
      "grad_norm": 0.6002615690231323,
      "learning_rate": 0.00016940916620651576,
      "loss": 1.7264,
      "num_input_tokens_seen": 5029808,
      "step": 837,
      "train_runtime": 845.0832,
      "train_tokens_per_second": 5951.849
    },
    {
      "epoch": 0.3082298850574713,
      "grad_norm": 0.7122713327407837,
      "learning_rate": 0.00016937235413215534,
      "loss": 1.7069,
      "num_input_tokens_seen": 5035536,
      "step": 838,
      "train_runtime": 846.0098,
      "train_tokens_per_second": 5952.101
    },
    {
      "epoch": 0.3085977011494253,
      "grad_norm": 0.6311381459236145,
      "learning_rate": 0.00016933554205779495,
      "loss": 1.7592,
      "num_input_tokens_seen": 5042128,
      "step": 839,
      "train_runtime": 847.0835,
      "train_tokens_per_second": 5952.338
    },
    {
      "epoch": 0.30896551724137933,
      "grad_norm": 0.7819897532463074,
      "learning_rate": 0.00016929872998343456,
      "loss": 1.906,
      "num_input_tokens_seen": 5047008,
      "step": 840,
      "train_runtime": 847.9315,
      "train_tokens_per_second": 5952.141
    },
    {
      "epoch": 0.30933333333333335,
      "grad_norm": 0.7180947661399841,
      "learning_rate": 0.00016926191790907417,
      "loss": 1.6724,
      "num_input_tokens_seen": 5051680,
      "step": 841,
      "train_runtime": 849.2043,
      "train_tokens_per_second": 5948.722
    },
    {
      "epoch": 0.30970114942528737,
      "grad_norm": 0.6616408824920654,
      "learning_rate": 0.0001692251058347138,
      "loss": 1.654,
      "num_input_tokens_seen": 5057120,
      "step": 842,
      "train_runtime": 850.1041,
      "train_tokens_per_second": 5948.825
    },
    {
      "epoch": 0.3100689655172414,
      "grad_norm": 0.7178268432617188,
      "learning_rate": 0.0001691882937603534,
      "loss": 1.5172,
      "num_input_tokens_seen": 5063904,
      "step": 843,
      "train_runtime": 851.211,
      "train_tokens_per_second": 5949.059
    },
    {
      "epoch": 0.3104367816091954,
      "grad_norm": 0.6756206750869751,
      "learning_rate": 0.00016915148168599302,
      "loss": 1.7465,
      "num_input_tokens_seen": 5070496,
      "step": 844,
      "train_runtime": 852.2595,
      "train_tokens_per_second": 5949.474
    },
    {
      "epoch": 0.31080459770114943,
      "grad_norm": 0.7164249420166016,
      "learning_rate": 0.00016911466961163263,
      "loss": 1.7599,
      "num_input_tokens_seen": 5074640,
      "step": 845,
      "train_runtime": 852.9907,
      "train_tokens_per_second": 5949.233
    },
    {
      "epoch": 0.31117241379310345,
      "grad_norm": 0.6614233255386353,
      "learning_rate": 0.00016907785753727224,
      "loss": 1.6321,
      "num_input_tokens_seen": 5082512,
      "step": 846,
      "train_runtime": 854.2473,
      "train_tokens_per_second": 5949.697
    },
    {
      "epoch": 0.3115402298850575,
      "grad_norm": 0.8350978493690491,
      "learning_rate": 0.00016904104546291185,
      "loss": 1.8342,
      "num_input_tokens_seen": 5088032,
      "step": 847,
      "train_runtime": 855.1359,
      "train_tokens_per_second": 5949.969
    },
    {
      "epoch": 0.3119080459770115,
      "grad_norm": 0.8022600412368774,
      "learning_rate": 0.00016900423338855146,
      "loss": 1.7741,
      "num_input_tokens_seen": 5093280,
      "step": 848,
      "train_runtime": 856.0052,
      "train_tokens_per_second": 5950.058
    },
    {
      "epoch": 0.3122758620689655,
      "grad_norm": 0.7320350408554077,
      "learning_rate": 0.00016896742131419106,
      "loss": 1.7976,
      "num_input_tokens_seen": 5098016,
      "step": 849,
      "train_runtime": 856.8151,
      "train_tokens_per_second": 5949.961
    },
    {
      "epoch": 0.31264367816091954,
      "grad_norm": 0.6889047622680664,
      "learning_rate": 0.00016893060923983067,
      "loss": 1.9103,
      "num_input_tokens_seen": 5107280,
      "step": 850,
      "train_runtime": 858.2192,
      "train_tokens_per_second": 5951.02
    },
    {
      "epoch": 0.31301149425287356,
      "grad_norm": 0.6495490670204163,
      "learning_rate": 0.00016889379716547028,
      "loss": 1.7081,
      "num_input_tokens_seen": 5113680,
      "step": 851,
      "train_runtime": 859.2414,
      "train_tokens_per_second": 5951.389
    },
    {
      "epoch": 0.3133793103448276,
      "grad_norm": 0.7501242756843567,
      "learning_rate": 0.0001688569850911099,
      "loss": 1.667,
      "num_input_tokens_seen": 5124304,
      "step": 852,
      "train_runtime": 860.8138,
      "train_tokens_per_second": 5952.86
    },
    {
      "epoch": 0.3137471264367816,
      "grad_norm": 0.7219048142433167,
      "learning_rate": 0.0001688201730167495,
      "loss": 1.7866,
      "num_input_tokens_seen": 5128912,
      "step": 853,
      "train_runtime": 861.5859,
      "train_tokens_per_second": 5952.873
    },
    {
      "epoch": 0.3141149425287356,
      "grad_norm": 0.7248823046684265,
      "learning_rate": 0.0001687833609423891,
      "loss": 1.6947,
      "num_input_tokens_seen": 5135216,
      "step": 854,
      "train_runtime": 862.5766,
      "train_tokens_per_second": 5953.345
    },
    {
      "epoch": 0.31448275862068964,
      "grad_norm": 0.7637900114059448,
      "learning_rate": 0.00016874654886802872,
      "loss": 1.702,
      "num_input_tokens_seen": 5142848,
      "step": 855,
      "train_runtime": 863.7986,
      "train_tokens_per_second": 5953.758
    },
    {
      "epoch": 0.31485057471264366,
      "grad_norm": 0.7177774906158447,
      "learning_rate": 0.00016870973679366833,
      "loss": 1.6115,
      "num_input_tokens_seen": 5149696,
      "step": 856,
      "train_runtime": 864.906,
      "train_tokens_per_second": 5954.053
    },
    {
      "epoch": 0.3152183908045977,
      "grad_norm": 0.7024462223052979,
      "learning_rate": 0.00016867292471930794,
      "loss": 1.7505,
      "num_input_tokens_seen": 5156448,
      "step": 857,
      "train_runtime": 865.9797,
      "train_tokens_per_second": 5954.467
    },
    {
      "epoch": 0.3155862068965517,
      "grad_norm": 0.746577799320221,
      "learning_rate": 0.00016863611264494754,
      "loss": 1.7835,
      "num_input_tokens_seen": 5166160,
      "step": 858,
      "train_runtime": 867.4706,
      "train_tokens_per_second": 5955.429
    },
    {
      "epoch": 0.3159540229885057,
      "grad_norm": 0.6877517700195312,
      "learning_rate": 0.00016859930057058718,
      "loss": 1.8222,
      "num_input_tokens_seen": 5171568,
      "step": 859,
      "train_runtime": 868.361,
      "train_tokens_per_second": 5955.55
    },
    {
      "epoch": 0.31632183908045974,
      "grad_norm": 0.6079294085502625,
      "learning_rate": 0.0001685624884962268,
      "loss": 1.6799,
      "num_input_tokens_seen": 5179808,
      "step": 860,
      "train_runtime": 869.6778,
      "train_tokens_per_second": 5956.008
    },
    {
      "epoch": 0.3166896551724138,
      "grad_norm": 0.7623908519744873,
      "learning_rate": 0.0001685256764218664,
      "loss": 1.9252,
      "num_input_tokens_seen": 5184240,
      "step": 861,
      "train_runtime": 870.4561,
      "train_tokens_per_second": 5955.774
    },
    {
      "epoch": 0.31705747126436784,
      "grad_norm": 0.7584787011146545,
      "learning_rate": 0.00016848886434750598,
      "loss": 1.7388,
      "num_input_tokens_seen": 5189696,
      "step": 862,
      "train_runtime": 871.3552,
      "train_tokens_per_second": 5955.89
    },
    {
      "epoch": 0.31742528735632186,
      "grad_norm": 0.6920429468154907,
      "learning_rate": 0.0001684520522731456,
      "loss": 1.8104,
      "num_input_tokens_seen": 5195968,
      "step": 863,
      "train_runtime": 872.3708,
      "train_tokens_per_second": 5956.146
    },
    {
      "epoch": 0.3177931034482759,
      "grad_norm": 0.7273485064506531,
      "learning_rate": 0.0001684152401987852,
      "loss": 1.8228,
      "num_input_tokens_seen": 5200336,
      "step": 864,
      "train_runtime": 873.164,
      "train_tokens_per_second": 5955.738
    },
    {
      "epoch": 0.3181609195402299,
      "grad_norm": 0.7210944890975952,
      "learning_rate": 0.0001683784281244248,
      "loss": 1.7069,
      "num_input_tokens_seen": 5206896,
      "step": 865,
      "train_runtime": 874.2379,
      "train_tokens_per_second": 5955.926
    },
    {
      "epoch": 0.3185287356321839,
      "grad_norm": 0.7385914325714111,
      "learning_rate": 0.00016834161605006442,
      "loss": 1.8228,
      "num_input_tokens_seen": 5211792,
      "step": 866,
      "train_runtime": 875.0804,
      "train_tokens_per_second": 5955.787
    },
    {
      "epoch": 0.31889655172413794,
      "grad_norm": 0.7037887573242188,
      "learning_rate": 0.00016830480397570402,
      "loss": 1.815,
      "num_input_tokens_seen": 5216656,
      "step": 867,
      "train_runtime": 875.9222,
      "train_tokens_per_second": 5955.616
    },
    {
      "epoch": 0.31926436781609197,
      "grad_norm": 0.6879228949546814,
      "learning_rate": 0.00016826799190134366,
      "loss": 1.9361,
      "num_input_tokens_seen": 5222384,
      "step": 868,
      "train_runtime": 876.8561,
      "train_tokens_per_second": 5955.805
    },
    {
      "epoch": 0.319632183908046,
      "grad_norm": 0.8687586784362793,
      "learning_rate": 0.00016823117982698327,
      "loss": 1.9659,
      "num_input_tokens_seen": 5226736,
      "step": 869,
      "train_runtime": 877.6014,
      "train_tokens_per_second": 5955.706
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.7334696054458618,
      "learning_rate": 0.00016819436775262288,
      "loss": 1.7086,
      "num_input_tokens_seen": 5232064,
      "step": 870,
      "train_runtime": 878.5051,
      "train_tokens_per_second": 5955.645
    },
    {
      "epoch": 0.320367816091954,
      "grad_norm": 0.7504156827926636,
      "learning_rate": 0.0001681575556782625,
      "loss": 1.7866,
      "num_input_tokens_seen": 5238832,
      "step": 871,
      "train_runtime": 880.0728,
      "train_tokens_per_second": 5952.726
    },
    {
      "epoch": 0.32073563218390805,
      "grad_norm": 0.6457074880599976,
      "learning_rate": 0.0001681207436039021,
      "loss": 1.7832,
      "num_input_tokens_seen": 5243856,
      "step": 872,
      "train_runtime": 880.9248,
      "train_tokens_per_second": 5952.672
    },
    {
      "epoch": 0.32110344827586207,
      "grad_norm": 0.71819007396698,
      "learning_rate": 0.0001680839315295417,
      "loss": 1.8124,
      "num_input_tokens_seen": 5249360,
      "step": 873,
      "train_runtime": 881.8479,
      "train_tokens_per_second": 5952.682
    },
    {
      "epoch": 0.3214712643678161,
      "grad_norm": 0.7726781368255615,
      "learning_rate": 0.00016804711945518131,
      "loss": 1.7085,
      "num_input_tokens_seen": 5254688,
      "step": 874,
      "train_runtime": 882.7371,
      "train_tokens_per_second": 5952.721
    },
    {
      "epoch": 0.3218390804597701,
      "grad_norm": 0.7883817553520203,
      "learning_rate": 0.0001680103073808209,
      "loss": 1.8612,
      "num_input_tokens_seen": 5260144,
      "step": 875,
      "train_runtime": 883.652,
      "train_tokens_per_second": 5952.733
    },
    {
      "epoch": 0.32220689655172413,
      "grad_norm": 0.6984350681304932,
      "learning_rate": 0.00016797349530646053,
      "loss": 1.9082,
      "num_input_tokens_seen": 5266336,
      "step": 876,
      "train_runtime": 884.6729,
      "train_tokens_per_second": 5952.862
    },
    {
      "epoch": 0.32257471264367815,
      "grad_norm": 0.7587144374847412,
      "learning_rate": 0.00016793668323210014,
      "loss": 1.7635,
      "num_input_tokens_seen": 5272704,
      "step": 877,
      "train_runtime": 885.7196,
      "train_tokens_per_second": 5953.017
    },
    {
      "epoch": 0.3229425287356322,
      "grad_norm": 0.7405463457107544,
      "learning_rate": 0.00016789987115773975,
      "loss": 1.8575,
      "num_input_tokens_seen": 5277872,
      "step": 878,
      "train_runtime": 886.5928,
      "train_tokens_per_second": 5952.983
    },
    {
      "epoch": 0.3233103448275862,
      "grad_norm": 0.787229061126709,
      "learning_rate": 0.00016786305908337936,
      "loss": 1.8151,
      "num_input_tokens_seen": 5284176,
      "step": 879,
      "train_runtime": 887.6247,
      "train_tokens_per_second": 5953.165
    },
    {
      "epoch": 0.3236781609195402,
      "grad_norm": 0.7123964428901672,
      "learning_rate": 0.00016782624700901897,
      "loss": 1.8004,
      "num_input_tokens_seen": 5290176,
      "step": 880,
      "train_runtime": 888.6192,
      "train_tokens_per_second": 5953.254
    },
    {
      "epoch": 0.32404597701149424,
      "grad_norm": 0.7722557187080383,
      "learning_rate": 0.00016778943493465858,
      "loss": 1.8734,
      "num_input_tokens_seen": 5295392,
      "step": 881,
      "train_runtime": 889.4947,
      "train_tokens_per_second": 5953.259
    },
    {
      "epoch": 0.32441379310344826,
      "grad_norm": 0.720944344997406,
      "learning_rate": 0.00016775262286029819,
      "loss": 1.8406,
      "num_input_tokens_seen": 5299904,
      "step": 882,
      "train_runtime": 890.2712,
      "train_tokens_per_second": 5953.135
    },
    {
      "epoch": 0.3247816091954023,
      "grad_norm": 0.6755884289741516,
      "learning_rate": 0.0001677158107859378,
      "loss": 1.6772,
      "num_input_tokens_seen": 5304080,
      "step": 883,
      "train_runtime": 890.979,
      "train_tokens_per_second": 5953.092
    },
    {
      "epoch": 0.3251494252873563,
      "grad_norm": 0.7546120285987854,
      "learning_rate": 0.0001676789987115774,
      "loss": 1.9367,
      "num_input_tokens_seen": 5310944,
      "step": 884,
      "train_runtime": 892.0462,
      "train_tokens_per_second": 5953.665
    },
    {
      "epoch": 0.3255172413793103,
      "grad_norm": 0.7448853254318237,
      "learning_rate": 0.000167642186637217,
      "loss": 1.6476,
      "num_input_tokens_seen": 5316112,
      "step": 885,
      "train_runtime": 892.9141,
      "train_tokens_per_second": 5953.666
    },
    {
      "epoch": 0.3258850574712644,
      "grad_norm": 0.7519321441650391,
      "learning_rate": 0.00016760537456285662,
      "loss": 1.7657,
      "num_input_tokens_seen": 5322256,
      "step": 886,
      "train_runtime": 893.8944,
      "train_tokens_per_second": 5954.01
    },
    {
      "epoch": 0.3262528735632184,
      "grad_norm": 0.7763766050338745,
      "learning_rate": 0.00016756856248849623,
      "loss": 1.8403,
      "num_input_tokens_seen": 5327264,
      "step": 887,
      "train_runtime": 894.7249,
      "train_tokens_per_second": 5954.081
    },
    {
      "epoch": 0.32662068965517244,
      "grad_norm": 0.6998291611671448,
      "learning_rate": 0.00016753175041413584,
      "loss": 1.7858,
      "num_input_tokens_seen": 5333024,
      "step": 888,
      "train_runtime": 895.6699,
      "train_tokens_per_second": 5954.229
    },
    {
      "epoch": 0.32698850574712646,
      "grad_norm": 0.6765996217727661,
      "learning_rate": 0.00016749493833977545,
      "loss": 1.6686,
      "num_input_tokens_seen": 5337936,
      "step": 889,
      "train_runtime": 896.5134,
      "train_tokens_per_second": 5954.106
    },
    {
      "epoch": 0.3273563218390805,
      "grad_norm": 0.7559351325035095,
      "learning_rate": 0.00016745812626541506,
      "loss": 1.6607,
      "num_input_tokens_seen": 5345296,
      "step": 890,
      "train_runtime": 897.6994,
      "train_tokens_per_second": 5954.438
    },
    {
      "epoch": 0.3277241379310345,
      "grad_norm": 0.767314076423645,
      "learning_rate": 0.00016742131419105467,
      "loss": 1.9345,
      "num_input_tokens_seen": 5350448,
      "step": 891,
      "train_runtime": 898.5485,
      "train_tokens_per_second": 5954.546
    },
    {
      "epoch": 0.3280919540229885,
      "grad_norm": 0.7825886011123657,
      "learning_rate": 0.00016738450211669427,
      "loss": 1.9034,
      "num_input_tokens_seen": 5355408,
      "step": 892,
      "train_runtime": 899.3921,
      "train_tokens_per_second": 5954.475
    },
    {
      "epoch": 0.32845977011494254,
      "grad_norm": 0.7350146770477295,
      "learning_rate": 0.0001673476900423339,
      "loss": 1.7683,
      "num_input_tokens_seen": 5361728,
      "step": 893,
      "train_runtime": 900.3859,
      "train_tokens_per_second": 5954.922
    },
    {
      "epoch": 0.32882758620689656,
      "grad_norm": 0.6977401971817017,
      "learning_rate": 0.00016731087796797352,
      "loss": 1.7708,
      "num_input_tokens_seen": 5367024,
      "step": 894,
      "train_runtime": 901.2762,
      "train_tokens_per_second": 5954.916
    },
    {
      "epoch": 0.3291954022988506,
      "grad_norm": 0.6419262886047363,
      "learning_rate": 0.00016727406589361313,
      "loss": 1.7043,
      "num_input_tokens_seen": 5374144,
      "step": 895,
      "train_runtime": 902.4095,
      "train_tokens_per_second": 5955.327
    },
    {
      "epoch": 0.3295632183908046,
      "grad_norm": 0.7703659534454346,
      "learning_rate": 0.00016723725381925274,
      "loss": 1.6718,
      "num_input_tokens_seen": 5381264,
      "step": 896,
      "train_runtime": 903.5123,
      "train_tokens_per_second": 5955.939
    },
    {
      "epoch": 0.3299310344827586,
      "grad_norm": 0.732337474822998,
      "learning_rate": 0.00016720044174489235,
      "loss": 1.8742,
      "num_input_tokens_seen": 5386256,
      "step": 897,
      "train_runtime": 904.3424,
      "train_tokens_per_second": 5955.992
    },
    {
      "epoch": 0.33029885057471264,
      "grad_norm": 0.6581416130065918,
      "learning_rate": 0.00016716362967053193,
      "loss": 1.7263,
      "num_input_tokens_seen": 5395312,
      "step": 898,
      "train_runtime": 905.7289,
      "train_tokens_per_second": 5956.873
    },
    {
      "epoch": 0.33066666666666666,
      "grad_norm": 0.781947910785675,
      "learning_rate": 0.00016712681759617154,
      "loss": 1.8114,
      "num_input_tokens_seen": 5400048,
      "step": 899,
      "train_runtime": 906.5289,
      "train_tokens_per_second": 5956.84
    },
    {
      "epoch": 0.3310344827586207,
      "grad_norm": 0.7370207905769348,
      "learning_rate": 0.00016709000552181115,
      "loss": 1.7365,
      "num_input_tokens_seen": 5404992,
      "step": 900,
      "train_runtime": 907.3548,
      "train_tokens_per_second": 5956.867
    },
    {
      "epoch": 0.3314022988505747,
      "grad_norm": 0.680095374584198,
      "learning_rate": 0.00016705319344745075,
      "loss": 1.5573,
      "num_input_tokens_seen": 5410816,
      "step": 901,
      "train_runtime": 908.8003,
      "train_tokens_per_second": 5953.801
    },
    {
      "epoch": 0.3317701149425287,
      "grad_norm": 0.7197843790054321,
      "learning_rate": 0.0001670163813730904,
      "loss": 1.7531,
      "num_input_tokens_seen": 5417584,
      "step": 902,
      "train_runtime": 909.8755,
      "train_tokens_per_second": 5954.203
    },
    {
      "epoch": 0.33213793103448275,
      "grad_norm": 0.7075902223587036,
      "learning_rate": 0.00016697956929873,
      "loss": 1.6958,
      "num_input_tokens_seen": 5422144,
      "step": 903,
      "train_runtime": 910.6749,
      "train_tokens_per_second": 5953.984
    },
    {
      "epoch": 0.33250574712643677,
      "grad_norm": 0.7745205163955688,
      "learning_rate": 0.0001669427572243696,
      "loss": 1.7477,
      "num_input_tokens_seen": 5428128,
      "step": 904,
      "train_runtime": 911.651,
      "train_tokens_per_second": 5954.173
    },
    {
      "epoch": 0.3328735632183908,
      "grad_norm": 0.6813750267028809,
      "learning_rate": 0.00016690594515000922,
      "loss": 1.5535,
      "num_input_tokens_seen": 5432720,
      "step": 905,
      "train_runtime": 912.4664,
      "train_tokens_per_second": 5953.885
    },
    {
      "epoch": 0.3332413793103448,
      "grad_norm": 0.6669746041297913,
      "learning_rate": 0.00016686913307564883,
      "loss": 1.8188,
      "num_input_tokens_seen": 5438880,
      "step": 906,
      "train_runtime": 913.4526,
      "train_tokens_per_second": 5954.2
    },
    {
      "epoch": 0.33360919540229883,
      "grad_norm": 0.7980788350105286,
      "learning_rate": 0.00016683232100128843,
      "loss": 1.8031,
      "num_input_tokens_seen": 5445488,
      "step": 907,
      "train_runtime": 914.5392,
      "train_tokens_per_second": 5954.351
    },
    {
      "epoch": 0.33397701149425285,
      "grad_norm": 0.8281051516532898,
      "learning_rate": 0.00016679550892692804,
      "loss": 1.9154,
      "num_input_tokens_seen": 5450240,
      "step": 908,
      "train_runtime": 915.349,
      "train_tokens_per_second": 5954.275
    },
    {
      "epoch": 0.33434482758620687,
      "grad_norm": 0.8023442625999451,
      "learning_rate": 0.00016675869685256765,
      "loss": 1.8548,
      "num_input_tokens_seen": 5456560,
      "step": 909,
      "train_runtime": 916.3711,
      "train_tokens_per_second": 5954.531
    },
    {
      "epoch": 0.3347126436781609,
      "grad_norm": 0.7418462038040161,
      "learning_rate": 0.00016672188477820726,
      "loss": 1.7166,
      "num_input_tokens_seen": 5463120,
      "step": 910,
      "train_runtime": 917.4244,
      "train_tokens_per_second": 5954.845
    },
    {
      "epoch": 0.33508045977011497,
      "grad_norm": 0.761345624923706,
      "learning_rate": 0.00016668507270384687,
      "loss": 1.9705,
      "num_input_tokens_seen": 5469792,
      "step": 911,
      "train_runtime": 918.496,
      "train_tokens_per_second": 5955.161
    },
    {
      "epoch": 0.335448275862069,
      "grad_norm": 0.7353963255882263,
      "learning_rate": 0.00016664826062948648,
      "loss": 1.7924,
      "num_input_tokens_seen": 5475952,
      "step": 912,
      "train_runtime": 919.5159,
      "train_tokens_per_second": 5955.255
    },
    {
      "epoch": 0.335816091954023,
      "grad_norm": 0.7788954377174377,
      "learning_rate": 0.0001666114485551261,
      "loss": 1.6776,
      "num_input_tokens_seen": 5483680,
      "step": 913,
      "train_runtime": 920.729,
      "train_tokens_per_second": 5955.802
    },
    {
      "epoch": 0.33618390804597703,
      "grad_norm": 0.7629323601722717,
      "learning_rate": 0.0001665746364807657,
      "loss": 1.8075,
      "num_input_tokens_seen": 5491152,
      "step": 914,
      "train_runtime": 921.9071,
      "train_tokens_per_second": 5956.297
    },
    {
      "epoch": 0.33655172413793105,
      "grad_norm": 0.7548069953918457,
      "learning_rate": 0.0001665378244064053,
      "loss": 1.8454,
      "num_input_tokens_seen": 5496560,
      "step": 915,
      "train_runtime": 922.7855,
      "train_tokens_per_second": 5956.487
    },
    {
      "epoch": 0.3369195402298851,
      "grad_norm": 0.7721388936042786,
      "learning_rate": 0.00016650101233204491,
      "loss": 1.9724,
      "num_input_tokens_seen": 5501232,
      "step": 916,
      "train_runtime": 923.5841,
      "train_tokens_per_second": 5956.395
    },
    {
      "epoch": 0.3372873563218391,
      "grad_norm": 0.6715081930160522,
      "learning_rate": 0.00016646420025768452,
      "loss": 1.7975,
      "num_input_tokens_seen": 5509088,
      "step": 917,
      "train_runtime": 924.827,
      "train_tokens_per_second": 5956.885
    },
    {
      "epoch": 0.3376551724137931,
      "grad_norm": 0.7529799342155457,
      "learning_rate": 0.00016642738818332413,
      "loss": 1.8351,
      "num_input_tokens_seen": 5515488,
      "step": 918,
      "train_runtime": 925.8782,
      "train_tokens_per_second": 5957.034
    },
    {
      "epoch": 0.33802298850574714,
      "grad_norm": 0.7663313150405884,
      "learning_rate": 0.00016639057610896377,
      "loss": 1.9124,
      "num_input_tokens_seen": 5523888,
      "step": 919,
      "train_runtime": 927.1774,
      "train_tokens_per_second": 5957.747
    },
    {
      "epoch": 0.33839080459770116,
      "grad_norm": 0.8396819829940796,
      "learning_rate": 0.00016635376403460338,
      "loss": 1.933,
      "num_input_tokens_seen": 5531200,
      "step": 920,
      "train_runtime": 928.3283,
      "train_tokens_per_second": 5958.237
    },
    {
      "epoch": 0.3387586206896552,
      "grad_norm": 0.6740185618400574,
      "learning_rate": 0.00016631695196024296,
      "loss": 1.6992,
      "num_input_tokens_seen": 5537104,
      "step": 921,
      "train_runtime": 929.283,
      "train_tokens_per_second": 5958.469
    },
    {
      "epoch": 0.3391264367816092,
      "grad_norm": 0.6395431756973267,
      "learning_rate": 0.00016628013988588257,
      "loss": 1.7998,
      "num_input_tokens_seen": 5543520,
      "step": 922,
      "train_runtime": 930.3256,
      "train_tokens_per_second": 5958.688
    },
    {
      "epoch": 0.3394942528735632,
      "grad_norm": 0.7490308284759521,
      "learning_rate": 0.00016624332781152218,
      "loss": 1.9367,
      "num_input_tokens_seen": 5547968,
      "step": 923,
      "train_runtime": 931.0858,
      "train_tokens_per_second": 5958.6
    },
    {
      "epoch": 0.33986206896551724,
      "grad_norm": 0.7609663009643555,
      "learning_rate": 0.00016620651573716179,
      "loss": 1.846,
      "num_input_tokens_seen": 5552720,
      "step": 924,
      "train_runtime": 931.8949,
      "train_tokens_per_second": 5958.526
    },
    {
      "epoch": 0.34022988505747126,
      "grad_norm": 0.754848301410675,
      "learning_rate": 0.0001661697036628014,
      "loss": 1.8017,
      "num_input_tokens_seen": 5557760,
      "step": 925,
      "train_runtime": 932.7539,
      "train_tokens_per_second": 5958.442
    },
    {
      "epoch": 0.3405977011494253,
      "grad_norm": 0.7673571109771729,
      "learning_rate": 0.000166132891588441,
      "loss": 1.8847,
      "num_input_tokens_seen": 5562048,
      "step": 926,
      "train_runtime": 933.5201,
      "train_tokens_per_second": 5958.145
    },
    {
      "epoch": 0.3409655172413793,
      "grad_norm": 0.6410609483718872,
      "learning_rate": 0.0001660960795140806,
      "loss": 1.7325,
      "num_input_tokens_seen": 5568832,
      "step": 927,
      "train_runtime": 934.607,
      "train_tokens_per_second": 5958.474
    },
    {
      "epoch": 0.3413333333333333,
      "grad_norm": 0.739958643913269,
      "learning_rate": 0.00016605926743972025,
      "loss": 1.8312,
      "num_input_tokens_seen": 5573952,
      "step": 928,
      "train_runtime": 935.4746,
      "train_tokens_per_second": 5958.422
    },
    {
      "epoch": 0.34170114942528734,
      "grad_norm": 0.8044548630714417,
      "learning_rate": 0.00016602245536535986,
      "loss": 2.0177,
      "num_input_tokens_seen": 5580032,
      "step": 929,
      "train_runtime": 936.4628,
      "train_tokens_per_second": 5958.627
    },
    {
      "epoch": 0.34206896551724136,
      "grad_norm": 0.7180975675582886,
      "learning_rate": 0.00016598564329099947,
      "loss": 1.8069,
      "num_input_tokens_seen": 5584432,
      "step": 930,
      "train_runtime": 937.2357,
      "train_tokens_per_second": 5958.407
    },
    {
      "epoch": 0.3424367816091954,
      "grad_norm": 0.7389205694198608,
      "learning_rate": 0.00016594883121663908,
      "loss": 1.9265,
      "num_input_tokens_seen": 5592256,
      "step": 931,
      "train_runtime": 938.9232,
      "train_tokens_per_second": 5956.032
    },
    {
      "epoch": 0.3428045977011494,
      "grad_norm": 0.7436507344245911,
      "learning_rate": 0.00016591201914227868,
      "loss": 1.9659,
      "num_input_tokens_seen": 5602304,
      "step": 932,
      "train_runtime": 940.4709,
      "train_tokens_per_second": 5956.913
    },
    {
      "epoch": 0.3431724137931034,
      "grad_norm": 0.736650824546814,
      "learning_rate": 0.0001658752070679183,
      "loss": 1.6487,
      "num_input_tokens_seen": 5606864,
      "step": 933,
      "train_runtime": 941.2631,
      "train_tokens_per_second": 5956.744
    },
    {
      "epoch": 0.34354022988505745,
      "grad_norm": 0.77228844165802,
      "learning_rate": 0.0001658383949935579,
      "loss": 1.8334,
      "num_input_tokens_seen": 5612064,
      "step": 934,
      "train_runtime": 942.1342,
      "train_tokens_per_second": 5956.757
    },
    {
      "epoch": 0.34390804597701147,
      "grad_norm": 0.7864047884941101,
      "learning_rate": 0.00016580158291919748,
      "loss": 1.8269,
      "num_input_tokens_seen": 5616816,
      "step": 935,
      "train_runtime": 942.9496,
      "train_tokens_per_second": 5956.645
    },
    {
      "epoch": 0.34427586206896554,
      "grad_norm": 0.7177388668060303,
      "learning_rate": 0.00016576477084483712,
      "loss": 1.6359,
      "num_input_tokens_seen": 5622864,
      "step": 936,
      "train_runtime": 943.9504,
      "train_tokens_per_second": 5956.736
    },
    {
      "epoch": 0.34464367816091956,
      "grad_norm": 0.7560311555862427,
      "learning_rate": 0.00016572795877047673,
      "loss": 2.1609,
      "num_input_tokens_seen": 5627792,
      "step": 937,
      "train_runtime": 944.7675,
      "train_tokens_per_second": 5956.801
    },
    {
      "epoch": 0.3450114942528736,
      "grad_norm": 0.7919855117797852,
      "learning_rate": 0.00016569114669611634,
      "loss": 1.908,
      "num_input_tokens_seen": 5634768,
      "step": 938,
      "train_runtime": 945.8646,
      "train_tokens_per_second": 5957.267
    },
    {
      "epoch": 0.3453793103448276,
      "grad_norm": 0.7530859708786011,
      "learning_rate": 0.00016565433462175595,
      "loss": 1.7035,
      "num_input_tokens_seen": 5640656,
      "step": 939,
      "train_runtime": 946.854,
      "train_tokens_per_second": 5957.261
    },
    {
      "epoch": 0.3457471264367816,
      "grad_norm": 0.7020770907402039,
      "learning_rate": 0.00016561752254739556,
      "loss": 1.7462,
      "num_input_tokens_seen": 5646960,
      "step": 940,
      "train_runtime": 947.8881,
      "train_tokens_per_second": 5957.412
    },
    {
      "epoch": 0.34611494252873565,
      "grad_norm": 0.8409958481788635,
      "learning_rate": 0.00016558071047303516,
      "loss": 1.669,
      "num_input_tokens_seen": 5652736,
      "step": 941,
      "train_runtime": 948.8596,
      "train_tokens_per_second": 5957.4
    },
    {
      "epoch": 0.34648275862068967,
      "grad_norm": 0.7040790915489197,
      "learning_rate": 0.00016554389839867477,
      "loss": 1.8176,
      "num_input_tokens_seen": 5657824,
      "step": 942,
      "train_runtime": 949.6983,
      "train_tokens_per_second": 5957.496
    },
    {
      "epoch": 0.3468505747126437,
      "grad_norm": 0.7028451561927795,
      "learning_rate": 0.00016550708632431438,
      "loss": 1.8359,
      "num_input_tokens_seen": 5663568,
      "step": 943,
      "train_runtime": 950.6592,
      "train_tokens_per_second": 5957.517
    },
    {
      "epoch": 0.3472183908045977,
      "grad_norm": 0.6852657198905945,
      "learning_rate": 0.000165470274249954,
      "loss": 1.694,
      "num_input_tokens_seen": 5669200,
      "step": 944,
      "train_runtime": 951.5858,
      "train_tokens_per_second": 5957.634
    },
    {
      "epoch": 0.34758620689655173,
      "grad_norm": 0.7269863486289978,
      "learning_rate": 0.0001654334621755936,
      "loss": 1.6913,
      "num_input_tokens_seen": 5675344,
      "step": 945,
      "train_runtime": 952.5801,
      "train_tokens_per_second": 5957.865
    },
    {
      "epoch": 0.34795402298850575,
      "grad_norm": 0.7116245031356812,
      "learning_rate": 0.0001653966501012332,
      "loss": 1.7932,
      "num_input_tokens_seen": 5680464,
      "step": 946,
      "train_runtime": 953.4462,
      "train_tokens_per_second": 5957.823
    },
    {
      "epoch": 0.34832183908045977,
      "grad_norm": 0.780099630355835,
      "learning_rate": 0.00016535983802687282,
      "loss": 1.8072,
      "num_input_tokens_seen": 5686448,
      "step": 947,
      "train_runtime": 954.415,
      "train_tokens_per_second": 5958.045
    },
    {
      "epoch": 0.3486896551724138,
      "grad_norm": 0.710581362247467,
      "learning_rate": 0.00016532302595251243,
      "loss": 1.5974,
      "num_input_tokens_seen": 5694752,
      "step": 948,
      "train_runtime": 955.6773,
      "train_tokens_per_second": 5958.865
    },
    {
      "epoch": 0.3490574712643678,
      "grad_norm": 0.7463304400444031,
      "learning_rate": 0.00016528621387815204,
      "loss": 1.7301,
      "num_input_tokens_seen": 5704592,
      "step": 949,
      "train_runtime": 957.1568,
      "train_tokens_per_second": 5959.935
    },
    {
      "epoch": 0.34942528735632183,
      "grad_norm": 0.8333390951156616,
      "learning_rate": 0.00016524940180379164,
      "loss": 1.8625,
      "num_input_tokens_seen": 5713168,
      "step": 950,
      "train_runtime": 958.4767,
      "train_tokens_per_second": 5960.675
    },
    {
      "epoch": 0.34979310344827585,
      "grad_norm": 0.674447238445282,
      "learning_rate": 0.00016521258972943125,
      "loss": 1.6663,
      "num_input_tokens_seen": 5718144,
      "step": 951,
      "train_runtime": 959.303,
      "train_tokens_per_second": 5960.728
    },
    {
      "epoch": 0.3501609195402299,
      "grad_norm": 0.672986626625061,
      "learning_rate": 0.00016517577765507086,
      "loss": 1.6585,
      "num_input_tokens_seen": 5724848,
      "step": 952,
      "train_runtime": 960.3875,
      "train_tokens_per_second": 5960.977
    },
    {
      "epoch": 0.3505287356321839,
      "grad_norm": 0.7748245000839233,
      "learning_rate": 0.0001651389655807105,
      "loss": 1.7803,
      "num_input_tokens_seen": 5729328,
      "step": 953,
      "train_runtime": 961.1743,
      "train_tokens_per_second": 5960.759
    },
    {
      "epoch": 0.3508965517241379,
      "grad_norm": 0.631734311580658,
      "learning_rate": 0.0001651021535063501,
      "loss": 1.757,
      "num_input_tokens_seen": 5737296,
      "step": 954,
      "train_runtime": 962.4372,
      "train_tokens_per_second": 5961.216
    },
    {
      "epoch": 0.35126436781609194,
      "grad_norm": 0.7916156053543091,
      "learning_rate": 0.00016506534143198972,
      "loss": 1.9381,
      "num_input_tokens_seen": 5744896,
      "step": 955,
      "train_runtime": 963.6028,
      "train_tokens_per_second": 5961.892
    },
    {
      "epoch": 0.35163218390804596,
      "grad_norm": 0.6629148721694946,
      "learning_rate": 0.00016502852935762932,
      "loss": 1.6705,
      "num_input_tokens_seen": 5749744,
      "step": 956,
      "train_runtime": 964.4178,
      "train_tokens_per_second": 5961.881
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.6603271961212158,
      "learning_rate": 0.00016499171728326893,
      "loss": 1.7175,
      "num_input_tokens_seen": 5755152,
      "step": 957,
      "train_runtime": 965.3238,
      "train_tokens_per_second": 5961.887
    },
    {
      "epoch": 0.352367816091954,
      "grad_norm": 0.770685613155365,
      "learning_rate": 0.00016495490520890852,
      "loss": 1.8274,
      "num_input_tokens_seen": 5760752,
      "step": 958,
      "train_runtime": 966.2246,
      "train_tokens_per_second": 5962.125
    },
    {
      "epoch": 0.352735632183908,
      "grad_norm": 0.6816652417182922,
      "learning_rate": 0.00016491809313454812,
      "loss": 1.834,
      "num_input_tokens_seen": 5771312,
      "step": 959,
      "train_runtime": 967.8065,
      "train_tokens_per_second": 5963.291
    },
    {
      "epoch": 0.35310344827586204,
      "grad_norm": 0.7133073806762695,
      "learning_rate": 0.00016488128106018773,
      "loss": 1.7701,
      "num_input_tokens_seen": 5776176,
      "step": 960,
      "train_runtime": 968.641,
      "train_tokens_per_second": 5963.175
    },
    {
      "epoch": 0.3534712643678161,
      "grad_norm": 0.678919792175293,
      "learning_rate": 0.00016484446898582734,
      "loss": 1.7831,
      "num_input_tokens_seen": 5782464,
      "step": 961,
      "train_runtime": 970.1425,
      "train_tokens_per_second": 5960.428
    },
    {
      "epoch": 0.35383908045977014,
      "grad_norm": 0.6276996731758118,
      "learning_rate": 0.00016480765691146698,
      "loss": 1.7643,
      "num_input_tokens_seen": 5787088,
      "step": 962,
      "train_runtime": 970.9439,
      "train_tokens_per_second": 5960.27
    },
    {
      "epoch": 0.35420689655172416,
      "grad_norm": 0.7291053533554077,
      "learning_rate": 0.0001647708448371066,
      "loss": 1.6757,
      "num_input_tokens_seen": 5791952,
      "step": 963,
      "train_runtime": 971.7866,
      "train_tokens_per_second": 5960.107
    },
    {
      "epoch": 0.3545747126436782,
      "grad_norm": 0.6977639198303223,
      "learning_rate": 0.0001647340327627462,
      "loss": 1.6786,
      "num_input_tokens_seen": 5800352,
      "step": 964,
      "train_runtime": 973.1064,
      "train_tokens_per_second": 5960.655
    },
    {
      "epoch": 0.3549425287356322,
      "grad_norm": 0.668307900428772,
      "learning_rate": 0.0001646972206883858,
      "loss": 1.6556,
      "num_input_tokens_seen": 5806704,
      "step": 965,
      "train_runtime": 974.1432,
      "train_tokens_per_second": 5960.832
    },
    {
      "epoch": 0.3553103448275862,
      "grad_norm": 0.7223288416862488,
      "learning_rate": 0.0001646604086140254,
      "loss": 1.8902,
      "num_input_tokens_seen": 5812288,
      "step": 966,
      "train_runtime": 975.0824,
      "train_tokens_per_second": 5960.817
    },
    {
      "epoch": 0.35567816091954024,
      "grad_norm": 0.6939303278923035,
      "learning_rate": 0.00016462359653966502,
      "loss": 1.9322,
      "num_input_tokens_seen": 5818080,
      "step": 967,
      "train_runtime": 976.059,
      "train_tokens_per_second": 5960.787
    },
    {
      "epoch": 0.35604597701149426,
      "grad_norm": 0.7478364706039429,
      "learning_rate": 0.00016458678446530463,
      "loss": 1.7346,
      "num_input_tokens_seen": 5823968,
      "step": 968,
      "train_runtime": 977.0237,
      "train_tokens_per_second": 5960.928
    },
    {
      "epoch": 0.3564137931034483,
      "grad_norm": 0.7032122611999512,
      "learning_rate": 0.00016454997239094424,
      "loss": 1.9145,
      "num_input_tokens_seen": 5830320,
      "step": 969,
      "train_runtime": 978.0622,
      "train_tokens_per_second": 5961.093
    },
    {
      "epoch": 0.3567816091954023,
      "grad_norm": 0.6438683271408081,
      "learning_rate": 0.00016451316031658385,
      "loss": 1.7318,
      "num_input_tokens_seen": 5837904,
      "step": 970,
      "train_runtime": 979.2637,
      "train_tokens_per_second": 5961.524
    },
    {
      "epoch": 0.3571494252873563,
      "grad_norm": 0.651461124420166,
      "learning_rate": 0.00016447634824222346,
      "loss": 1.7959,
      "num_input_tokens_seen": 5844608,
      "step": 971,
      "train_runtime": 980.3301,
      "train_tokens_per_second": 5961.878
    },
    {
      "epoch": 0.35751724137931035,
      "grad_norm": 0.7091881036758423,
      "learning_rate": 0.00016443953616786307,
      "loss": 1.9192,
      "num_input_tokens_seen": 5855760,
      "step": 972,
      "train_runtime": 982.0103,
      "train_tokens_per_second": 5963.033
    },
    {
      "epoch": 0.35788505747126437,
      "grad_norm": 0.7543753385543823,
      "learning_rate": 0.00016440272409350268,
      "loss": 1.6492,
      "num_input_tokens_seen": 5862384,
      "step": 973,
      "train_runtime": 983.0877,
      "train_tokens_per_second": 5963.236
    },
    {
      "epoch": 0.3582528735632184,
      "grad_norm": 0.7147246599197388,
      "learning_rate": 0.00016436591201914228,
      "loss": 1.8111,
      "num_input_tokens_seen": 5867616,
      "step": 974,
      "train_runtime": 983.9649,
      "train_tokens_per_second": 5963.237
    },
    {
      "epoch": 0.3586206896551724,
      "grad_norm": 0.6903164386749268,
      "learning_rate": 0.0001643290999447819,
      "loss": 1.8092,
      "num_input_tokens_seen": 5872832,
      "step": 975,
      "train_runtime": 984.8438,
      "train_tokens_per_second": 5963.212
    },
    {
      "epoch": 0.35898850574712643,
      "grad_norm": 0.7116182446479797,
      "learning_rate": 0.0001642922878704215,
      "loss": 1.8203,
      "num_input_tokens_seen": 5879408,
      "step": 976,
      "train_runtime": 985.923,
      "train_tokens_per_second": 5963.354
    },
    {
      "epoch": 0.35935632183908045,
      "grad_norm": 0.7367668151855469,
      "learning_rate": 0.0001642554757960611,
      "loss": 1.7829,
      "num_input_tokens_seen": 5885088,
      "step": 977,
      "train_runtime": 986.8655,
      "train_tokens_per_second": 5963.415
    },
    {
      "epoch": 0.35972413793103447,
      "grad_norm": 0.6934012174606323,
      "learning_rate": 0.00016421866372170072,
      "loss": 1.7207,
      "num_input_tokens_seen": 5890448,
      "step": 978,
      "train_runtime": 987.7483,
      "train_tokens_per_second": 5963.511
    },
    {
      "epoch": 0.3600919540229885,
      "grad_norm": 0.7513874173164368,
      "learning_rate": 0.00016418185164734036,
      "loss": 1.721,
      "num_input_tokens_seen": 5894912,
      "step": 979,
      "train_runtime": 988.5242,
      "train_tokens_per_second": 5963.346
    },
    {
      "epoch": 0.3604597701149425,
      "grad_norm": 0.6775432825088501,
      "learning_rate": 0.00016414503957297997,
      "loss": 1.6022,
      "num_input_tokens_seen": 5902224,
      "step": 980,
      "train_runtime": 989.6712,
      "train_tokens_per_second": 5963.823
    },
    {
      "epoch": 0.36082758620689653,
      "grad_norm": 0.7553426623344421,
      "learning_rate": 0.00016410822749861955,
      "loss": 1.6472,
      "num_input_tokens_seen": 5908128,
      "step": 981,
      "train_runtime": 990.6623,
      "train_tokens_per_second": 5963.816
    },
    {
      "epoch": 0.36119540229885055,
      "grad_norm": 0.700955867767334,
      "learning_rate": 0.00016407141542425916,
      "loss": 1.6952,
      "num_input_tokens_seen": 5914800,
      "step": 982,
      "train_runtime": 991.7303,
      "train_tokens_per_second": 5964.121
    },
    {
      "epoch": 0.3615632183908046,
      "grad_norm": 0.7294247150421143,
      "learning_rate": 0.00016403460334989876,
      "loss": 1.8799,
      "num_input_tokens_seen": 5919712,
      "step": 983,
      "train_runtime": 992.5735,
      "train_tokens_per_second": 5964.004
    },
    {
      "epoch": 0.3619310344827586,
      "grad_norm": 0.7663295269012451,
      "learning_rate": 0.00016399779127553837,
      "loss": 1.8785,
      "num_input_tokens_seen": 5925024,
      "step": 984,
      "train_runtime": 993.4661,
      "train_tokens_per_second": 5963.992
    },
    {
      "epoch": 0.3622988505747126,
      "grad_norm": 0.6929524540901184,
      "learning_rate": 0.00016396097920117798,
      "loss": 1.8012,
      "num_input_tokens_seen": 5934000,
      "step": 985,
      "train_runtime": 994.8297,
      "train_tokens_per_second": 5964.84
    },
    {
      "epoch": 0.3626666666666667,
      "grad_norm": 0.720958948135376,
      "learning_rate": 0.0001639241671268176,
      "loss": 1.9708,
      "num_input_tokens_seen": 5939616,
      "step": 986,
      "train_runtime": 995.7591,
      "train_tokens_per_second": 5964.913
    },
    {
      "epoch": 0.3630344827586207,
      "grad_norm": 0.7467935085296631,
      "learning_rate": 0.0001638873550524572,
      "loss": 1.8607,
      "num_input_tokens_seen": 5945424,
      "step": 987,
      "train_runtime": 996.6987,
      "train_tokens_per_second": 5965.117
    },
    {
      "epoch": 0.36340229885057473,
      "grad_norm": 0.632853090763092,
      "learning_rate": 0.00016385054297809684,
      "loss": 1.6344,
      "num_input_tokens_seen": 5955056,
      "step": 988,
      "train_runtime": 998.1631,
      "train_tokens_per_second": 5966.015
    },
    {
      "epoch": 0.36377011494252876,
      "grad_norm": 0.7575991153717041,
      "learning_rate": 0.00016381373090373645,
      "loss": 1.8875,
      "num_input_tokens_seen": 5960352,
      "step": 989,
      "train_runtime": 999.0453,
      "train_tokens_per_second": 5966.048
    },
    {
      "epoch": 0.3641379310344828,
      "grad_norm": 0.6999156475067139,
      "learning_rate": 0.00016377691882937605,
      "loss": 1.8642,
      "num_input_tokens_seen": 5969136,
      "step": 990,
      "train_runtime": 1000.3921,
      "train_tokens_per_second": 5966.797
    },
    {
      "epoch": 0.3645057471264368,
      "grad_norm": 0.7687408328056335,
      "learning_rate": 0.00016374010675501566,
      "loss": 1.6813,
      "num_input_tokens_seen": 5974336,
      "step": 991,
      "train_runtime": 1001.7004,
      "train_tokens_per_second": 5964.195
    },
    {
      "epoch": 0.3648735632183908,
      "grad_norm": 0.7835000157356262,
      "learning_rate": 0.00016370329468065527,
      "loss": 1.8687,
      "num_input_tokens_seen": 5980176,
      "step": 992,
      "train_runtime": 1002.6559,
      "train_tokens_per_second": 5964.335
    },
    {
      "epoch": 0.36524137931034484,
      "grad_norm": 0.6699005365371704,
      "learning_rate": 0.00016366648260629488,
      "loss": 1.7947,
      "num_input_tokens_seen": 5986688,
      "step": 993,
      "train_runtime": 1003.7194,
      "train_tokens_per_second": 5964.504
    },
    {
      "epoch": 0.36560919540229886,
      "grad_norm": 0.7430959939956665,
      "learning_rate": 0.00016362967053193446,
      "loss": 1.8256,
      "num_input_tokens_seen": 5992688,
      "step": 994,
      "train_runtime": 1004.7154,
      "train_tokens_per_second": 5964.563
    },
    {
      "epoch": 0.3659770114942529,
      "grad_norm": 0.7016071677207947,
      "learning_rate": 0.00016359285845757407,
      "loss": 1.9459,
      "num_input_tokens_seen": 5999520,
      "step": 995,
      "train_runtime": 1005.784,
      "train_tokens_per_second": 5965.019
    },
    {
      "epoch": 0.3663448275862069,
      "grad_norm": 0.8595346212387085,
      "learning_rate": 0.0001635560463832137,
      "loss": 1.9452,
      "num_input_tokens_seen": 6005120,
      "step": 996,
      "train_runtime": 1006.6982,
      "train_tokens_per_second": 5965.164
    },
    {
      "epoch": 0.3667126436781609,
      "grad_norm": 0.8124039769172668,
      "learning_rate": 0.00016351923430885332,
      "loss": 1.7245,
      "num_input_tokens_seen": 6010576,
      "step": 997,
      "train_runtime": 1007.621,
      "train_tokens_per_second": 5965.116
    },
    {
      "epoch": 0.36708045977011494,
      "grad_norm": 0.7314801216125488,
      "learning_rate": 0.00016348242223449293,
      "loss": 1.7136,
      "num_input_tokens_seen": 6021728,
      "step": 998,
      "train_runtime": 1009.2858,
      "train_tokens_per_second": 5966.326
    },
    {
      "epoch": 0.36744827586206896,
      "grad_norm": 0.7391095161437988,
      "learning_rate": 0.00016344561016013253,
      "loss": 1.7556,
      "num_input_tokens_seen": 6026816,
      "step": 999,
      "train_runtime": 1010.1599,
      "train_tokens_per_second": 5966.2
    },
    {
      "epoch": 0.367816091954023,
      "grad_norm": 0.78415846824646,
      "learning_rate": 0.00016340879808577214,
      "loss": 1.6615,
      "num_input_tokens_seen": 6033104,
      "step": 1000,
      "train_runtime": 1011.1874,
      "train_tokens_per_second": 5966.356
    },
    {
      "epoch": 0.368183908045977,
      "grad_norm": 0.7598666548728943,
      "learning_rate": 0.00016337198601141175,
      "loss": 1.9224,
      "num_input_tokens_seen": 6038704,
      "step": 1001,
      "train_runtime": 1012.1416,
      "train_tokens_per_second": 5966.264
    },
    {
      "epoch": 0.368551724137931,
      "grad_norm": 0.7282189726829529,
      "learning_rate": 0.00016333517393705136,
      "loss": 1.8227,
      "num_input_tokens_seen": 6045744,
      "step": 1002,
      "train_runtime": 1013.2677,
      "train_tokens_per_second": 5966.581
    },
    {
      "epoch": 0.36891954022988505,
      "grad_norm": 0.7118775844573975,
      "learning_rate": 0.00016329836186269097,
      "loss": 1.6633,
      "num_input_tokens_seen": 6055312,
      "step": 1003,
      "train_runtime": 1014.7263,
      "train_tokens_per_second": 5967.434
    },
    {
      "epoch": 0.36928735632183907,
      "grad_norm": 0.746175229549408,
      "learning_rate": 0.00016326154978833058,
      "loss": 1.7093,
      "num_input_tokens_seen": 6063968,
      "step": 1004,
      "train_runtime": 1016.0781,
      "train_tokens_per_second": 5968.014
    },
    {
      "epoch": 0.3696551724137931,
      "grad_norm": 0.7311099767684937,
      "learning_rate": 0.0001632247377139702,
      "loss": 1.7717,
      "num_input_tokens_seen": 6071408,
      "step": 1005,
      "train_runtime": 1017.2374,
      "train_tokens_per_second": 5968.526
    },
    {
      "epoch": 0.3700229885057471,
      "grad_norm": 0.6983187794685364,
      "learning_rate": 0.0001631879256396098,
      "loss": 1.7268,
      "num_input_tokens_seen": 6076272,
      "step": 1006,
      "train_runtime": 1018.0566,
      "train_tokens_per_second": 5968.501
    },
    {
      "epoch": 0.37039080459770113,
      "grad_norm": 0.735249400138855,
      "learning_rate": 0.0001631511135652494,
      "loss": 1.9884,
      "num_input_tokens_seen": 6083232,
      "step": 1007,
      "train_runtime": 1019.144,
      "train_tokens_per_second": 5968.962
    },
    {
      "epoch": 0.37075862068965515,
      "grad_norm": 0.7385590076446533,
      "learning_rate": 0.00016311430149088901,
      "loss": 1.7056,
      "num_input_tokens_seen": 6089696,
      "step": 1008,
      "train_runtime": 1020.167,
      "train_tokens_per_second": 5969.313
    },
    {
      "epoch": 0.37112643678160917,
      "grad_norm": 0.8687577843666077,
      "learning_rate": 0.00016307748941652862,
      "loss": 1.9122,
      "num_input_tokens_seen": 6094128,
      "step": 1009,
      "train_runtime": 1020.9534,
      "train_tokens_per_second": 5969.056
    },
    {
      "epoch": 0.3714942528735632,
      "grad_norm": 0.7456827163696289,
      "learning_rate": 0.00016304067734216823,
      "loss": 1.748,
      "num_input_tokens_seen": 6098944,
      "step": 1010,
      "train_runtime": 1021.7695,
      "train_tokens_per_second": 5969.002
    },
    {
      "epoch": 0.37186206896551727,
      "grad_norm": 0.8254635334014893,
      "learning_rate": 0.00016300386526780784,
      "loss": 1.744,
      "num_input_tokens_seen": 6103408,
      "step": 1011,
      "train_runtime": 1022.5507,
      "train_tokens_per_second": 5968.807
    },
    {
      "epoch": 0.3722298850574713,
      "grad_norm": 0.7087537050247192,
      "learning_rate": 0.00016296705319344745,
      "loss": 1.818,
      "num_input_tokens_seen": 6109232,
      "step": 1012,
      "train_runtime": 1023.5187,
      "train_tokens_per_second": 5968.852
    },
    {
      "epoch": 0.3725977011494253,
      "grad_norm": 0.7285974621772766,
      "learning_rate": 0.00016293024111908709,
      "loss": 1.841,
      "num_input_tokens_seen": 6114352,
      "step": 1013,
      "train_runtime": 1024.3703,
      "train_tokens_per_second": 5968.888
    },
    {
      "epoch": 0.37296551724137933,
      "grad_norm": 0.7073532342910767,
      "learning_rate": 0.0001628934290447267,
      "loss": 1.905,
      "num_input_tokens_seen": 6119776,
      "step": 1014,
      "train_runtime": 1025.2843,
      "train_tokens_per_second": 5968.857
    },
    {
      "epoch": 0.37333333333333335,
      "grad_norm": 0.7369471788406372,
      "learning_rate": 0.0001628566169703663,
      "loss": 2.0135,
      "num_input_tokens_seen": 6125360,
      "step": 1015,
      "train_runtime": 1026.2065,
      "train_tokens_per_second": 5968.935
    },
    {
      "epoch": 0.37370114942528737,
      "grad_norm": 0.7469699382781982,
      "learning_rate": 0.0001628198048960059,
      "loss": 1.7684,
      "num_input_tokens_seen": 6131184,
      "step": 1016,
      "train_runtime": 1027.1512,
      "train_tokens_per_second": 5969.115
    },
    {
      "epoch": 0.3740689655172414,
      "grad_norm": 0.7768290042877197,
      "learning_rate": 0.00016278299282164552,
      "loss": 1.7137,
      "num_input_tokens_seen": 6136256,
      "step": 1017,
      "train_runtime": 1027.9902,
      "train_tokens_per_second": 5969.177
    },
    {
      "epoch": 0.3744367816091954,
      "grad_norm": 0.7599640488624573,
      "learning_rate": 0.0001627461807472851,
      "loss": 1.8537,
      "num_input_tokens_seen": 6142896,
      "step": 1018,
      "train_runtime": 1029.051,
      "train_tokens_per_second": 5969.477
    },
    {
      "epoch": 0.37480459770114943,
      "grad_norm": 0.6762050986289978,
      "learning_rate": 0.0001627093686729247,
      "loss": 1.6498,
      "num_input_tokens_seen": 6148720,
      "step": 1019,
      "train_runtime": 1030.0245,
      "train_tokens_per_second": 5969.489
    },
    {
      "epoch": 0.37517241379310345,
      "grad_norm": 0.7139620184898376,
      "learning_rate": 0.00016267255659856432,
      "loss": 1.9684,
      "num_input_tokens_seen": 6154080,
      "step": 1020,
      "train_runtime": 1030.8916,
      "train_tokens_per_second": 5969.668
    },
    {
      "epoch": 0.3755402298850575,
      "grad_norm": 0.7365214228630066,
      "learning_rate": 0.00016263574452420393,
      "loss": 1.7081,
      "num_input_tokens_seen": 6159360,
      "step": 1021,
      "train_runtime": 1032.3052,
      "train_tokens_per_second": 5966.608
    },
    {
      "epoch": 0.3759080459770115,
      "grad_norm": 0.7719625234603882,
      "learning_rate": 0.00016259893244984357,
      "loss": 1.7645,
      "num_input_tokens_seen": 6164704,
      "step": 1022,
      "train_runtime": 1033.183,
      "train_tokens_per_second": 5966.711
    },
    {
      "epoch": 0.3762758620689655,
      "grad_norm": 0.7170722484588623,
      "learning_rate": 0.00016256212037548317,
      "loss": 1.727,
      "num_input_tokens_seen": 6170592,
      "step": 1023,
      "train_runtime": 1034.1279,
      "train_tokens_per_second": 5966.952
    },
    {
      "epoch": 0.37664367816091954,
      "grad_norm": 0.7749509215354919,
      "learning_rate": 0.00016252530830112278,
      "loss": 1.9907,
      "num_input_tokens_seen": 6176160,
      "step": 1024,
      "train_runtime": 1035.0477,
      "train_tokens_per_second": 5967.029
    },
    {
      "epoch": 0.37701149425287356,
      "grad_norm": 0.6887675523757935,
      "learning_rate": 0.0001624884962267624,
      "loss": 1.8601,
      "num_input_tokens_seen": 6181280,
      "step": 1025,
      "train_runtime": 1035.9168,
      "train_tokens_per_second": 5966.966
    },
    {
      "epoch": 0.3773793103448276,
      "grad_norm": 0.7674352526664734,
      "learning_rate": 0.000162451684152402,
      "loss": 1.6161,
      "num_input_tokens_seen": 6187424,
      "step": 1026,
      "train_runtime": 1036.888,
      "train_tokens_per_second": 5967.302
    },
    {
      "epoch": 0.3777471264367816,
      "grad_norm": 0.6757673025131226,
      "learning_rate": 0.0001624148720780416,
      "loss": 1.7524,
      "num_input_tokens_seen": 6193600,
      "step": 1027,
      "train_runtime": 1037.9082,
      "train_tokens_per_second": 5967.387
    },
    {
      "epoch": 0.3781149425287356,
      "grad_norm": 0.6949533820152283,
      "learning_rate": 0.00016237806000368122,
      "loss": 1.6835,
      "num_input_tokens_seen": 6199632,
      "step": 1028,
      "train_runtime": 1038.9101,
      "train_tokens_per_second": 5967.438
    },
    {
      "epoch": 0.37848275862068964,
      "grad_norm": 0.7543976902961731,
      "learning_rate": 0.00016234124792932083,
      "loss": 1.6057,
      "num_input_tokens_seen": 6205520,
      "step": 1029,
      "train_runtime": 1039.8887,
      "train_tokens_per_second": 5967.485
    },
    {
      "epoch": 0.37885057471264366,
      "grad_norm": 0.7911177277565002,
      "learning_rate": 0.00016230443585496044,
      "loss": 1.8486,
      "num_input_tokens_seen": 6211808,
      "step": 1030,
      "train_runtime": 1040.9373,
      "train_tokens_per_second": 5967.514
    },
    {
      "epoch": 0.3792183908045977,
      "grad_norm": 0.7119301557540894,
      "learning_rate": 0.00016226762378060005,
      "loss": 1.8367,
      "num_input_tokens_seen": 6218032,
      "step": 1031,
      "train_runtime": 1041.9436,
      "train_tokens_per_second": 5967.724
    },
    {
      "epoch": 0.3795862068965517,
      "grad_norm": 0.6604345440864563,
      "learning_rate": 0.00016223081170623965,
      "loss": 1.6414,
      "num_input_tokens_seen": 6228688,
      "step": 1032,
      "train_runtime": 1043.5337,
      "train_tokens_per_second": 5968.842
    },
    {
      "epoch": 0.3799540229885057,
      "grad_norm": 0.7565374970436096,
      "learning_rate": 0.00016219399963187926,
      "loss": 1.7643,
      "num_input_tokens_seen": 6234592,
      "step": 1033,
      "train_runtime": 1044.5181,
      "train_tokens_per_second": 5968.869
    },
    {
      "epoch": 0.38032183908045974,
      "grad_norm": 0.6912949085235596,
      "learning_rate": 0.00016215718755751887,
      "loss": 1.7294,
      "num_input_tokens_seen": 6239568,
      "step": 1034,
      "train_runtime": 1045.3656,
      "train_tokens_per_second": 5968.79
    },
    {
      "epoch": 0.38068965517241377,
      "grad_norm": 0.7250675559043884,
      "learning_rate": 0.00016212037548315848,
      "loss": 1.8777,
      "num_input_tokens_seen": 6245168,
      "step": 1035,
      "train_runtime": 1046.2958,
      "train_tokens_per_second": 5968.836
    },
    {
      "epoch": 0.38105747126436784,
      "grad_norm": 0.7936033010482788,
      "learning_rate": 0.0001620835634087981,
      "loss": 1.6993,
      "num_input_tokens_seen": 6249888,
      "step": 1036,
      "train_runtime": 1047.0848,
      "train_tokens_per_second": 5968.846
    },
    {
      "epoch": 0.38142528735632186,
      "grad_norm": 0.7826706767082214,
      "learning_rate": 0.0001620467513344377,
      "loss": 1.8342,
      "num_input_tokens_seen": 6254544,
      "step": 1037,
      "train_runtime": 1047.8984,
      "train_tokens_per_second": 5968.655
    },
    {
      "epoch": 0.3817931034482759,
      "grad_norm": 0.7239696383476257,
      "learning_rate": 0.0001620099392600773,
      "loss": 1.8242,
      "num_input_tokens_seen": 6260256,
      "step": 1038,
      "train_runtime": 1048.8168,
      "train_tokens_per_second": 5968.875
    },
    {
      "epoch": 0.3821609195402299,
      "grad_norm": 0.7919307947158813,
      "learning_rate": 0.00016197312718571694,
      "loss": 2.0298,
      "num_input_tokens_seen": 6265024,
      "step": 1039,
      "train_runtime": 1049.6387,
      "train_tokens_per_second": 5968.744
    },
    {
      "epoch": 0.3825287356321839,
      "grad_norm": 0.7248693704605103,
      "learning_rate": 0.00016193631511135655,
      "loss": 1.8705,
      "num_input_tokens_seen": 6269536,
      "step": 1040,
      "train_runtime": 1050.4258,
      "train_tokens_per_second": 5968.567
    },
    {
      "epoch": 0.38289655172413795,
      "grad_norm": 0.7681882977485657,
      "learning_rate": 0.00016189950303699613,
      "loss": 1.8041,
      "num_input_tokens_seen": 6273520,
      "step": 1041,
      "train_runtime": 1051.1485,
      "train_tokens_per_second": 5968.253
    },
    {
      "epoch": 0.38326436781609197,
      "grad_norm": 0.7542139887809753,
      "learning_rate": 0.00016186269096263574,
      "loss": 1.9104,
      "num_input_tokens_seen": 6279296,
      "step": 1042,
      "train_runtime": 1052.0979,
      "train_tokens_per_second": 5968.357
    },
    {
      "epoch": 0.383632183908046,
      "grad_norm": 0.6792402267456055,
      "learning_rate": 0.00016182587888827535,
      "loss": 1.6552,
      "num_input_tokens_seen": 6284688,
      "step": 1043,
      "train_runtime": 1052.9885,
      "train_tokens_per_second": 5968.43
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.7532481551170349,
      "learning_rate": 0.00016178906681391496,
      "loss": 1.8242,
      "num_input_tokens_seen": 6290480,
      "step": 1044,
      "train_runtime": 1053.9224,
      "train_tokens_per_second": 5968.637
    },
    {
      "epoch": 0.38436781609195403,
      "grad_norm": 0.8112294673919678,
      "learning_rate": 0.00016175225473955457,
      "loss": 1.8354,
      "num_input_tokens_seen": 6295744,
      "step": 1045,
      "train_runtime": 1054.7935,
      "train_tokens_per_second": 5968.698
    },
    {
      "epoch": 0.38473563218390805,
      "grad_norm": 0.7810726165771484,
      "learning_rate": 0.00016171544266519418,
      "loss": 1.8834,
      "num_input_tokens_seen": 6300080,
      "step": 1046,
      "train_runtime": 1055.5504,
      "train_tokens_per_second": 5968.526
    },
    {
      "epoch": 0.38510344827586207,
      "grad_norm": 0.6948085427284241,
      "learning_rate": 0.0001616786305908338,
      "loss": 1.6368,
      "num_input_tokens_seen": 6306080,
      "step": 1047,
      "train_runtime": 1056.5191,
      "train_tokens_per_second": 5968.733
    },
    {
      "epoch": 0.3854712643678161,
      "grad_norm": 0.7175804972648621,
      "learning_rate": 0.00016164181851647342,
      "loss": 1.7445,
      "num_input_tokens_seen": 6310176,
      "step": 1048,
      "train_runtime": 1057.2421,
      "train_tokens_per_second": 5968.525
    },
    {
      "epoch": 0.3858390804597701,
      "grad_norm": 0.6663169860839844,
      "learning_rate": 0.00016160500644211303,
      "loss": 1.7449,
      "num_input_tokens_seen": 6314912,
      "step": 1049,
      "train_runtime": 1058.0475,
      "train_tokens_per_second": 5968.458
    },
    {
      "epoch": 0.38620689655172413,
      "grad_norm": 0.7119675278663635,
      "learning_rate": 0.00016156819436775264,
      "loss": 1.6852,
      "num_input_tokens_seen": 6320112,
      "step": 1050,
      "train_runtime": 1058.9321,
      "train_tokens_per_second": 5968.383
    },
    {
      "epoch": 0.38657471264367815,
      "grad_norm": 0.6928601264953613,
      "learning_rate": 0.00016153138229339225,
      "loss": 1.6927,
      "num_input_tokens_seen": 6328496,
      "step": 1051,
      "train_runtime": 1060.7128,
      "train_tokens_per_second": 5966.267
    },
    {
      "epoch": 0.3869425287356322,
      "grad_norm": 0.6850346922874451,
      "learning_rate": 0.00016149457021903186,
      "loss": 1.6651,
      "num_input_tokens_seen": 6335520,
      "step": 1052,
      "train_runtime": 1061.8283,
      "train_tokens_per_second": 5966.614
    },
    {
      "epoch": 0.3873103448275862,
      "grad_norm": 0.6681268215179443,
      "learning_rate": 0.00016145775814467147,
      "loss": 1.6906,
      "num_input_tokens_seen": 6344368,
      "step": 1053,
      "train_runtime": 1063.2088,
      "train_tokens_per_second": 5967.189
    },
    {
      "epoch": 0.3876781609195402,
      "grad_norm": 0.8159458041191101,
      "learning_rate": 0.00016142094607031105,
      "loss": 1.9324,
      "num_input_tokens_seen": 6350448,
      "step": 1054,
      "train_runtime": 1064.1692,
      "train_tokens_per_second": 5967.517
    },
    {
      "epoch": 0.38804597701149424,
      "grad_norm": 0.8756287097930908,
      "learning_rate": 0.00016138413399595066,
      "loss": 1.9897,
      "num_input_tokens_seen": 6355520,
      "step": 1055,
      "train_runtime": 1065.041,
      "train_tokens_per_second": 5967.394
    },
    {
      "epoch": 0.38841379310344826,
      "grad_norm": 0.7280504107475281,
      "learning_rate": 0.0001613473219215903,
      "loss": 1.8268,
      "num_input_tokens_seen": 6361424,
      "step": 1056,
      "train_runtime": 1065.9996,
      "train_tokens_per_second": 5967.567
    },
    {
      "epoch": 0.3887816091954023,
      "grad_norm": 0.7619026303291321,
      "learning_rate": 0.0001613105098472299,
      "loss": 1.7893,
      "num_input_tokens_seen": 6367280,
      "step": 1057,
      "train_runtime": 1066.9694,
      "train_tokens_per_second": 5967.631
    },
    {
      "epoch": 0.3891494252873563,
      "grad_norm": 0.7361872792243958,
      "learning_rate": 0.0001612736977728695,
      "loss": 1.7469,
      "num_input_tokens_seen": 6371952,
      "step": 1058,
      "train_runtime": 1067.7463,
      "train_tokens_per_second": 5967.665
    },
    {
      "epoch": 0.3895172413793103,
      "grad_norm": 0.7800519466400146,
      "learning_rate": 0.00016123688569850912,
      "loss": 1.8479,
      "num_input_tokens_seen": 6376320,
      "step": 1059,
      "train_runtime": 1068.5102,
      "train_tokens_per_second": 5967.486
    },
    {
      "epoch": 0.38988505747126434,
      "grad_norm": 0.7623835802078247,
      "learning_rate": 0.00016120007362414873,
      "loss": 1.9095,
      "num_input_tokens_seen": 6381056,
      "step": 1060,
      "train_runtime": 1069.3484,
      "train_tokens_per_second": 5967.238
    },
    {
      "epoch": 0.3902528735632184,
      "grad_norm": 0.8020111322402954,
      "learning_rate": 0.00016116326154978834,
      "loss": 1.7293,
      "num_input_tokens_seen": 6386144,
      "step": 1061,
      "train_runtime": 1070.2159,
      "train_tokens_per_second": 5967.155
    },
    {
      "epoch": 0.39062068965517244,
      "grad_norm": 0.7152970433235168,
      "learning_rate": 0.00016112644947542795,
      "loss": 1.8283,
      "num_input_tokens_seen": 6391712,
      "step": 1062,
      "train_runtime": 1071.1167,
      "train_tokens_per_second": 5967.335
    },
    {
      "epoch": 0.39098850574712646,
      "grad_norm": 0.7725966572761536,
      "learning_rate": 0.00016108963740106756,
      "loss": 2.0763,
      "num_input_tokens_seen": 6397264,
      "step": 1063,
      "train_runtime": 1072.0502,
      "train_tokens_per_second": 5967.318
    },
    {
      "epoch": 0.3913563218390805,
      "grad_norm": 0.7412512302398682,
      "learning_rate": 0.00016105282532670717,
      "loss": 1.907,
      "num_input_tokens_seen": 6404256,
      "step": 1064,
      "train_runtime": 1073.1526,
      "train_tokens_per_second": 5967.703
    },
    {
      "epoch": 0.3917241379310345,
      "grad_norm": 0.7671145796775818,
      "learning_rate": 0.00016101601325234678,
      "loss": 1.8035,
      "num_input_tokens_seen": 6411408,
      "step": 1065,
      "train_runtime": 1074.269,
      "train_tokens_per_second": 5968.159
    },
    {
      "epoch": 0.3920919540229885,
      "grad_norm": 0.8039507269859314,
      "learning_rate": 0.00016097920117798638,
      "loss": 1.8796,
      "num_input_tokens_seen": 6417072,
      "step": 1066,
      "train_runtime": 1075.2024,
      "train_tokens_per_second": 5968.246
    },
    {
      "epoch": 0.39245977011494254,
      "grad_norm": 0.7322057485580444,
      "learning_rate": 0.000160942389103626,
      "loss": 1.749,
      "num_input_tokens_seen": 6422688,
      "step": 1067,
      "train_runtime": 1076.1319,
      "train_tokens_per_second": 5968.31
    },
    {
      "epoch": 0.39282758620689656,
      "grad_norm": 0.6617266535758972,
      "learning_rate": 0.0001609055770292656,
      "loss": 1.7367,
      "num_input_tokens_seen": 6427840,
      "step": 1068,
      "train_runtime": 1076.9937,
      "train_tokens_per_second": 5968.317
    },
    {
      "epoch": 0.3931954022988506,
      "grad_norm": 0.7284842729568481,
      "learning_rate": 0.0001608687649549052,
      "loss": 1.7965,
      "num_input_tokens_seen": 6433408,
      "step": 1069,
      "train_runtime": 1077.9227,
      "train_tokens_per_second": 5968.339
    },
    {
      "epoch": 0.3935632183908046,
      "grad_norm": 0.6676843762397766,
      "learning_rate": 0.00016083195288054482,
      "loss": 1.8287,
      "num_input_tokens_seen": 6438688,
      "step": 1070,
      "train_runtime": 1078.8253,
      "train_tokens_per_second": 5968.24
    },
    {
      "epoch": 0.3939310344827586,
      "grad_norm": 0.6924183368682861,
      "learning_rate": 0.00016079514080618443,
      "loss": 1.7203,
      "num_input_tokens_seen": 6444512,
      "step": 1071,
      "train_runtime": 1079.7658,
      "train_tokens_per_second": 5968.435
    },
    {
      "epoch": 0.39429885057471264,
      "grad_norm": 0.6839929223060608,
      "learning_rate": 0.00016075832873182404,
      "loss": 1.6592,
      "num_input_tokens_seen": 6452320,
      "step": 1072,
      "train_runtime": 1081.0073,
      "train_tokens_per_second": 5968.803
    },
    {
      "epoch": 0.39466666666666667,
      "grad_norm": 0.6901578307151794,
      "learning_rate": 0.00016072151665746367,
      "loss": 1.6968,
      "num_input_tokens_seen": 6457808,
      "step": 1073,
      "train_runtime": 1081.9116,
      "train_tokens_per_second": 5968.887
    },
    {
      "epoch": 0.3950344827586207,
      "grad_norm": 0.6019689440727234,
      "learning_rate": 0.00016068470458310328,
      "loss": 1.5991,
      "num_input_tokens_seen": 6466832,
      "step": 1074,
      "train_runtime": 1083.3273,
      "train_tokens_per_second": 5969.416
    },
    {
      "epoch": 0.3954022988505747,
      "grad_norm": 0.7549545168876648,
      "learning_rate": 0.0001606478925087429,
      "loss": 1.7192,
      "num_input_tokens_seen": 6473312,
      "step": 1075,
      "train_runtime": 1084.396,
      "train_tokens_per_second": 5969.509
    },
    {
      "epoch": 0.39577011494252873,
      "grad_norm": 0.6868123412132263,
      "learning_rate": 0.0001606110804343825,
      "loss": 1.7076,
      "num_input_tokens_seen": 6479952,
      "step": 1076,
      "train_runtime": 1085.4428,
      "train_tokens_per_second": 5969.87
    },
    {
      "epoch": 0.39613793103448275,
      "grad_norm": 0.723551332950592,
      "learning_rate": 0.00016057426836002208,
      "loss": 1.9025,
      "num_input_tokens_seen": 6484832,
      "step": 1077,
      "train_runtime": 1086.2811,
      "train_tokens_per_second": 5969.755
    },
    {
      "epoch": 0.39650574712643677,
      "grad_norm": 0.7306846380233765,
      "learning_rate": 0.0001605374562856617,
      "loss": 1.8355,
      "num_input_tokens_seen": 6489952,
      "step": 1078,
      "train_runtime": 1087.1472,
      "train_tokens_per_second": 5969.708
    },
    {
      "epoch": 0.3968735632183908,
      "grad_norm": 0.8239873647689819,
      "learning_rate": 0.0001605006442113013,
      "loss": 1.9526,
      "num_input_tokens_seen": 6495088,
      "step": 1079,
      "train_runtime": 1088.0027,
      "train_tokens_per_second": 5969.735
    },
    {
      "epoch": 0.3972413793103448,
      "grad_norm": 0.6149707436561584,
      "learning_rate": 0.0001604638321369409,
      "loss": 1.7006,
      "num_input_tokens_seen": 6502224,
      "step": 1080,
      "train_runtime": 1089.1528,
      "train_tokens_per_second": 5969.983
    },
    {
      "epoch": 0.39760919540229883,
      "grad_norm": 0.7983827590942383,
      "learning_rate": 0.00016042702006258052,
      "loss": 1.6715,
      "num_input_tokens_seen": 6509520,
      "step": 1081,
      "train_runtime": 1090.7871,
      "train_tokens_per_second": 5967.727
    },
    {
      "epoch": 0.39797701149425285,
      "grad_norm": 0.7142229676246643,
      "learning_rate": 0.00016039020798822015,
      "loss": 1.7298,
      "num_input_tokens_seen": 6515424,
      "step": 1082,
      "train_runtime": 1091.7569,
      "train_tokens_per_second": 5967.834
    },
    {
      "epoch": 0.3983448275862069,
      "grad_norm": 0.6892955899238586,
      "learning_rate": 0.00016035339591385976,
      "loss": 1.7871,
      "num_input_tokens_seen": 6521200,
      "step": 1083,
      "train_runtime": 1092.7271,
      "train_tokens_per_second": 5967.821
    },
    {
      "epoch": 0.3987126436781609,
      "grad_norm": 0.7192055583000183,
      "learning_rate": 0.00016031658383949937,
      "loss": 1.8794,
      "num_input_tokens_seen": 6526944,
      "step": 1084,
      "train_runtime": 1093.6546,
      "train_tokens_per_second": 5968.012
    },
    {
      "epoch": 0.3990804597701149,
      "grad_norm": 0.7291635870933533,
      "learning_rate": 0.00016027977176513898,
      "loss": 1.8011,
      "num_input_tokens_seen": 6532528,
      "step": 1085,
      "train_runtime": 1094.5757,
      "train_tokens_per_second": 5968.091
    },
    {
      "epoch": 0.399448275862069,
      "grad_norm": 0.8123181462287903,
      "learning_rate": 0.0001602429596907786,
      "loss": 1.7672,
      "num_input_tokens_seen": 6538192,
      "step": 1086,
      "train_runtime": 1095.4815,
      "train_tokens_per_second": 5968.327
    },
    {
      "epoch": 0.399816091954023,
      "grad_norm": 0.7394521832466125,
      "learning_rate": 0.0001602061476164182,
      "loss": 1.5855,
      "num_input_tokens_seen": 6543408,
      "step": 1087,
      "train_runtime": 1096.3395,
      "train_tokens_per_second": 5968.414
    },
    {
      "epoch": 0.40018390804597703,
      "grad_norm": 0.7159187197685242,
      "learning_rate": 0.0001601693355420578,
      "loss": 1.7338,
      "num_input_tokens_seen": 6547760,
      "step": 1088,
      "train_runtime": 1097.0872,
      "train_tokens_per_second": 5968.313
    },
    {
      "epoch": 0.40018390804597703,
      "eval_loss": 1.8306975364685059,
      "eval_runtime": 4.7834,
      "eval_samples_per_second": 209.057,
      "eval_steps_per_second": 13.171,
      "num_input_tokens_seen": 6547760,
      "step": 1088
    },
    {
      "epoch": 0.40055172413793105,
      "grad_norm": 0.7736802697181702,
      "learning_rate": 0.00016013252346769742,
      "loss": 1.8055,
      "num_input_tokens_seen": 6552768,
      "step": 1089,
      "train_runtime": 1102.6923,
      "train_tokens_per_second": 5942.517
    },
    {
      "epoch": 0.4009195402298851,
      "grad_norm": 0.6520863175392151,
      "learning_rate": 0.00016009571139333702,
      "loss": 1.7862,
      "num_input_tokens_seen": 6564096,
      "step": 1090,
      "train_runtime": 1104.3888,
      "train_tokens_per_second": 5943.646
    },
    {
      "epoch": 0.4012873563218391,
      "grad_norm": 0.7434893250465393,
      "learning_rate": 0.00016005889931897663,
      "loss": 1.8366,
      "num_input_tokens_seen": 6568880,
      "step": 1091,
      "train_runtime": 1105.2165,
      "train_tokens_per_second": 5943.523
    },
    {
      "epoch": 0.4016551724137931,
      "grad_norm": 0.8334950804710388,
      "learning_rate": 0.00016002208724461624,
      "loss": 1.8302,
      "num_input_tokens_seen": 6574672,
      "step": 1092,
      "train_runtime": 1106.1654,
      "train_tokens_per_second": 5943.661
    },
    {
      "epoch": 0.40202298850574714,
      "grad_norm": 0.7248815298080444,
      "learning_rate": 0.00015998527517025585,
      "loss": 1.8074,
      "num_input_tokens_seen": 6580928,
      "step": 1093,
      "train_runtime": 1107.199,
      "train_tokens_per_second": 5943.763
    },
    {
      "epoch": 0.40239080459770116,
      "grad_norm": 0.7552504539489746,
      "learning_rate": 0.00015994846309589546,
      "loss": 1.8307,
      "num_input_tokens_seen": 6586688,
      "step": 1094,
      "train_runtime": 1108.1567,
      "train_tokens_per_second": 5943.824
    },
    {
      "epoch": 0.4027586206896552,
      "grad_norm": 0.769370436668396,
      "learning_rate": 0.00015991165102153507,
      "loss": 1.8785,
      "num_input_tokens_seen": 6591680,
      "step": 1095,
      "train_runtime": 1109.0078,
      "train_tokens_per_second": 5943.764
    },
    {
      "epoch": 0.4031264367816092,
      "grad_norm": 0.7354795932769775,
      "learning_rate": 0.00015987483894717468,
      "loss": 1.6408,
      "num_input_tokens_seen": 6597776,
      "step": 1096,
      "train_runtime": 1109.9971,
      "train_tokens_per_second": 5943.958
    },
    {
      "epoch": 0.4034942528735632,
      "grad_norm": 0.6922746896743774,
      "learning_rate": 0.0001598380268728143,
      "loss": 1.5582,
      "num_input_tokens_seen": 6604368,
      "step": 1097,
      "train_runtime": 1111.0385,
      "train_tokens_per_second": 5944.32
    },
    {
      "epoch": 0.40386206896551724,
      "grad_norm": 0.7460592985153198,
      "learning_rate": 0.0001598012147984539,
      "loss": 1.7108,
      "num_input_tokens_seen": 6611648,
      "step": 1098,
      "train_runtime": 1112.1915,
      "train_tokens_per_second": 5944.703
    },
    {
      "epoch": 0.40422988505747126,
      "grad_norm": 0.7240821123123169,
      "learning_rate": 0.00015976440272409353,
      "loss": 1.9362,
      "num_input_tokens_seen": 6616320,
      "step": 1099,
      "train_runtime": 1112.9966,
      "train_tokens_per_second": 5944.6
    },
    {
      "epoch": 0.4045977011494253,
      "grad_norm": 0.6838738322257996,
      "learning_rate": 0.00015972759064973314,
      "loss": 1.7338,
      "num_input_tokens_seen": 6622736,
      "step": 1100,
      "train_runtime": 1114.0388,
      "train_tokens_per_second": 5944.798
    },
    {
      "epoch": 0.4049655172413793,
      "grad_norm": 0.7672039270401001,
      "learning_rate": 0.00015969077857537272,
      "loss": 1.7559,
      "num_input_tokens_seen": 6628288,
      "step": 1101,
      "train_runtime": 1114.9584,
      "train_tokens_per_second": 5944.875
    },
    {
      "epoch": 0.4053333333333333,
      "grad_norm": 0.6691837310791016,
      "learning_rate": 0.00015965396650101233,
      "loss": 1.8892,
      "num_input_tokens_seen": 6633152,
      "step": 1102,
      "train_runtime": 1115.7792,
      "train_tokens_per_second": 5944.861
    },
    {
      "epoch": 0.40570114942528734,
      "grad_norm": 0.6882882118225098,
      "learning_rate": 0.00015961715442665194,
      "loss": 1.5012,
      "num_input_tokens_seen": 6638368,
      "step": 1103,
      "train_runtime": 1116.6535,
      "train_tokens_per_second": 5944.877
    },
    {
      "epoch": 0.40606896551724136,
      "grad_norm": 0.7536081671714783,
      "learning_rate": 0.00015958034235229155,
      "loss": 1.859,
      "num_input_tokens_seen": 6643360,
      "step": 1104,
      "train_runtime": 1117.5014,
      "train_tokens_per_second": 5944.833
    },
    {
      "epoch": 0.4064367816091954,
      "grad_norm": 0.7524173259735107,
      "learning_rate": 0.00015954353027793116,
      "loss": 1.9479,
      "num_input_tokens_seen": 6649440,
      "step": 1105,
      "train_runtime": 1118.4891,
      "train_tokens_per_second": 5945.02
    },
    {
      "epoch": 0.4068045977011494,
      "grad_norm": 0.7337230443954468,
      "learning_rate": 0.00015950671820357077,
      "loss": 1.6217,
      "num_input_tokens_seen": 6654864,
      "step": 1106,
      "train_runtime": 1119.3704,
      "train_tokens_per_second": 5945.185
    },
    {
      "epoch": 0.4071724137931034,
      "grad_norm": 0.7187566161155701,
      "learning_rate": 0.00015946990612921038,
      "loss": 1.8053,
      "num_input_tokens_seen": 6659904,
      "step": 1107,
      "train_runtime": 1120.2256,
      "train_tokens_per_second": 5945.146
    },
    {
      "epoch": 0.40754022988505745,
      "grad_norm": 0.7533746361732483,
      "learning_rate": 0.00015943309405485,
      "loss": 1.9118,
      "num_input_tokens_seen": 6665152,
      "step": 1108,
      "train_runtime": 1121.095,
      "train_tokens_per_second": 5945.216
    },
    {
      "epoch": 0.40790804597701147,
      "grad_norm": 0.6869662404060364,
      "learning_rate": 0.00015939628198048962,
      "loss": 1.7787,
      "num_input_tokens_seen": 6671776,
      "step": 1109,
      "train_runtime": 1122.1691,
      "train_tokens_per_second": 5945.428
    },
    {
      "epoch": 0.40827586206896554,
      "grad_norm": 0.6175126433372498,
      "learning_rate": 0.00015935946990612923,
      "loss": 1.6665,
      "num_input_tokens_seen": 6678448,
      "step": 1110,
      "train_runtime": 1123.2485,
      "train_tokens_per_second": 5945.655
    },
    {
      "epoch": 0.40864367816091957,
      "grad_norm": 0.6928731799125671,
      "learning_rate": 0.00015932265783176884,
      "loss": 1.7474,
      "num_input_tokens_seen": 6682832,
      "step": 1111,
      "train_runtime": 1124.5231,
      "train_tokens_per_second": 5942.814
    },
    {
      "epoch": 0.4090114942528736,
      "grad_norm": 0.7784749269485474,
      "learning_rate": 0.00015928584575740845,
      "loss": 1.7744,
      "num_input_tokens_seen": 6691584,
      "step": 1112,
      "train_runtime": 1125.8423,
      "train_tokens_per_second": 5943.625
    },
    {
      "epoch": 0.4093793103448276,
      "grad_norm": 0.7535203099250793,
      "learning_rate": 0.00015924903368304806,
      "loss": 1.9865,
      "num_input_tokens_seen": 6696480,
      "step": 1113,
      "train_runtime": 1126.6933,
      "train_tokens_per_second": 5943.481
    },
    {
      "epoch": 0.40974712643678163,
      "grad_norm": 0.742264449596405,
      "learning_rate": 0.00015921222160868764,
      "loss": 1.7038,
      "num_input_tokens_seen": 6703424,
      "step": 1114,
      "train_runtime": 1127.7687,
      "train_tokens_per_second": 5943.971
    },
    {
      "epoch": 0.41011494252873565,
      "grad_norm": 0.6702042818069458,
      "learning_rate": 0.00015917540953432725,
      "loss": 1.7514,
      "num_input_tokens_seen": 6709552,
      "step": 1115,
      "train_runtime": 1128.7816,
      "train_tokens_per_second": 5944.066
    },
    {
      "epoch": 0.41048275862068967,
      "grad_norm": 0.7077691555023193,
      "learning_rate": 0.00015913859745996688,
      "loss": 1.6815,
      "num_input_tokens_seen": 6715024,
      "step": 1116,
      "train_runtime": 1129.6984,
      "train_tokens_per_second": 5944.086
    },
    {
      "epoch": 0.4108505747126437,
      "grad_norm": 0.721027672290802,
      "learning_rate": 0.0001591017853856065,
      "loss": 1.8057,
      "num_input_tokens_seen": 6722704,
      "step": 1117,
      "train_runtime": 1130.9179,
      "train_tokens_per_second": 5944.467
    },
    {
      "epoch": 0.4112183908045977,
      "grad_norm": 0.7421275973320007,
      "learning_rate": 0.0001590649733112461,
      "loss": 1.891,
      "num_input_tokens_seen": 6728304,
      "step": 1118,
      "train_runtime": 1131.8472,
      "train_tokens_per_second": 5944.534
    },
    {
      "epoch": 0.41158620689655173,
      "grad_norm": 0.7358080744743347,
      "learning_rate": 0.0001590281612368857,
      "loss": 2.0234,
      "num_input_tokens_seen": 6732848,
      "step": 1119,
      "train_runtime": 1132.641,
      "train_tokens_per_second": 5944.38
    },
    {
      "epoch": 0.41195402298850575,
      "grad_norm": 0.7264527082443237,
      "learning_rate": 0.00015899134916252532,
      "loss": 1.7717,
      "num_input_tokens_seen": 6739312,
      "step": 1120,
      "train_runtime": 1133.6756,
      "train_tokens_per_second": 5944.656
    },
    {
      "epoch": 0.4123218390804598,
      "grad_norm": 0.7438011169433594,
      "learning_rate": 0.00015895453708816493,
      "loss": 1.7899,
      "num_input_tokens_seen": 6744288,
      "step": 1121,
      "train_runtime": 1134.5032,
      "train_tokens_per_second": 5944.706
    },
    {
      "epoch": 0.4126896551724138,
      "grad_norm": 0.661270797252655,
      "learning_rate": 0.00015891772501380454,
      "loss": 1.6328,
      "num_input_tokens_seen": 6749808,
      "step": 1122,
      "train_runtime": 1135.4009,
      "train_tokens_per_second": 5944.867
    },
    {
      "epoch": 0.4130574712643678,
      "grad_norm": 0.7114384174346924,
      "learning_rate": 0.00015888091293944415,
      "loss": 1.8099,
      "num_input_tokens_seen": 6759184,
      "step": 1123,
      "train_runtime": 1136.8282,
      "train_tokens_per_second": 5945.651
    },
    {
      "epoch": 0.41342528735632184,
      "grad_norm": 0.6370521187782288,
      "learning_rate": 0.00015884410086508375,
      "loss": 1.8412,
      "num_input_tokens_seen": 6766272,
      "step": 1124,
      "train_runtime": 1137.9547,
      "train_tokens_per_second": 5945.994
    },
    {
      "epoch": 0.41379310344827586,
      "grad_norm": 0.7143312096595764,
      "learning_rate": 0.00015880728879072336,
      "loss": 1.7608,
      "num_input_tokens_seen": 6775456,
      "step": 1125,
      "train_runtime": 1139.335,
      "train_tokens_per_second": 5946.851
    },
    {
      "epoch": 0.4141609195402299,
      "grad_norm": 0.7388086915016174,
      "learning_rate": 0.00015877047671636297,
      "loss": 1.7346,
      "num_input_tokens_seen": 6781984,
      "step": 1126,
      "train_runtime": 1140.381,
      "train_tokens_per_second": 5947.121
    },
    {
      "epoch": 0.4145287356321839,
      "grad_norm": 0.7524212002754211,
      "learning_rate": 0.00015873366464200258,
      "loss": 1.6919,
      "num_input_tokens_seen": 6787872,
      "step": 1127,
      "train_runtime": 1141.332,
      "train_tokens_per_second": 5947.325
    },
    {
      "epoch": 0.4148965517241379,
      "grad_norm": 0.7130899429321289,
      "learning_rate": 0.0001586968525676422,
      "loss": 1.806,
      "num_input_tokens_seen": 6793824,
      "step": 1128,
      "train_runtime": 1142.2931,
      "train_tokens_per_second": 5947.531
    },
    {
      "epoch": 0.41526436781609194,
      "grad_norm": 0.7975853681564331,
      "learning_rate": 0.0001586600404932818,
      "loss": 1.7579,
      "num_input_tokens_seen": 6800096,
      "step": 1129,
      "train_runtime": 1143.2902,
      "train_tokens_per_second": 5947.83
    },
    {
      "epoch": 0.41563218390804596,
      "grad_norm": 0.8366739749908447,
      "learning_rate": 0.0001586232284189214,
      "loss": 1.8777,
      "num_input_tokens_seen": 6805184,
      "step": 1130,
      "train_runtime": 1144.1525,
      "train_tokens_per_second": 5947.795
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.7377239465713501,
      "learning_rate": 0.00015858641634456102,
      "loss": 1.802,
      "num_input_tokens_seen": 6811600,
      "step": 1131,
      "train_runtime": 1145.2068,
      "train_tokens_per_second": 5947.922
    },
    {
      "epoch": 0.416367816091954,
      "grad_norm": 0.7626973986625671,
      "learning_rate": 0.00015854960427020063,
      "loss": 1.6195,
      "num_input_tokens_seen": 6817296,
      "step": 1132,
      "train_runtime": 1146.1491,
      "train_tokens_per_second": 5948.001
    },
    {
      "epoch": 0.416735632183908,
      "grad_norm": 0.6945323348045349,
      "learning_rate": 0.00015851279219584026,
      "loss": 1.7649,
      "num_input_tokens_seen": 6822800,
      "step": 1133,
      "train_runtime": 1147.0662,
      "train_tokens_per_second": 5948.044
    },
    {
      "epoch": 0.41710344827586204,
      "grad_norm": 0.7901187539100647,
      "learning_rate": 0.00015847598012147987,
      "loss": 1.9968,
      "num_input_tokens_seen": 6827472,
      "step": 1134,
      "train_runtime": 1147.8702,
      "train_tokens_per_second": 5947.948
    },
    {
      "epoch": 0.4174712643678161,
      "grad_norm": 0.6932904124259949,
      "learning_rate": 0.00015843916804711948,
      "loss": 1.5601,
      "num_input_tokens_seen": 6833984,
      "step": 1135,
      "train_runtime": 1148.9271,
      "train_tokens_per_second": 5948.144
    },
    {
      "epoch": 0.41783908045977014,
      "grad_norm": 0.6883950233459473,
      "learning_rate": 0.0001584023559727591,
      "loss": 1.6218,
      "num_input_tokens_seen": 6842144,
      "step": 1136,
      "train_runtime": 1150.2003,
      "train_tokens_per_second": 5948.654
    },
    {
      "epoch": 0.41820689655172416,
      "grad_norm": 0.7180618643760681,
      "learning_rate": 0.00015836554389839867,
      "loss": 1.7734,
      "num_input_tokens_seen": 6847104,
      "step": 1137,
      "train_runtime": 1151.0378,
      "train_tokens_per_second": 5948.635
    },
    {
      "epoch": 0.4185747126436782,
      "grad_norm": 0.7424864172935486,
      "learning_rate": 0.00015832873182403828,
      "loss": 1.6873,
      "num_input_tokens_seen": 6852384,
      "step": 1138,
      "train_runtime": 1151.9092,
      "train_tokens_per_second": 5948.719
    },
    {
      "epoch": 0.4189425287356322,
      "grad_norm": 0.7944812178611755,
      "learning_rate": 0.0001582919197496779,
      "loss": 1.8155,
      "num_input_tokens_seen": 6856832,
      "step": 1139,
      "train_runtime": 1152.6833,
      "train_tokens_per_second": 5948.583
    },
    {
      "epoch": 0.4193103448275862,
      "grad_norm": 0.6998050212860107,
      "learning_rate": 0.0001582551076753175,
      "loss": 1.7667,
      "num_input_tokens_seen": 6863344,
      "step": 1140,
      "train_runtime": 1153.7205,
      "train_tokens_per_second": 5948.88
    },
    {
      "epoch": 0.41967816091954024,
      "grad_norm": 0.6973651051521301,
      "learning_rate": 0.0001582182956009571,
      "loss": 1.7329,
      "num_input_tokens_seen": 6868304,
      "step": 1141,
      "train_runtime": 1155.0697,
      "train_tokens_per_second": 5946.225
    },
    {
      "epoch": 0.42004597701149426,
      "grad_norm": 0.630289614200592,
      "learning_rate": 0.00015818148352659674,
      "loss": 1.6378,
      "num_input_tokens_seen": 6876000,
      "step": 1142,
      "train_runtime": 1156.3029,
      "train_tokens_per_second": 5946.539
    },
    {
      "epoch": 0.4204137931034483,
      "grad_norm": 0.7176012396812439,
      "learning_rate": 0.00015814467145223635,
      "loss": 1.8841,
      "num_input_tokens_seen": 6880944,
      "step": 1143,
      "train_runtime": 1157.1493,
      "train_tokens_per_second": 5946.462
    },
    {
      "epoch": 0.4207816091954023,
      "grad_norm": 0.745722234249115,
      "learning_rate": 0.00015810785937787596,
      "loss": 1.8377,
      "num_input_tokens_seen": 6886272,
      "step": 1144,
      "train_runtime": 1158.0254,
      "train_tokens_per_second": 5946.564
    },
    {
      "epoch": 0.4211494252873563,
      "grad_norm": 0.7454776167869568,
      "learning_rate": 0.00015807104730351557,
      "loss": 1.712,
      "num_input_tokens_seen": 6892960,
      "step": 1145,
      "train_runtime": 1159.0973,
      "train_tokens_per_second": 5946.835
    },
    {
      "epoch": 0.42151724137931035,
      "grad_norm": 0.7079422473907471,
      "learning_rate": 0.00015803423522915518,
      "loss": 1.8538,
      "num_input_tokens_seen": 6897536,
      "step": 1146,
      "train_runtime": 1159.8874,
      "train_tokens_per_second": 5946.729
    },
    {
      "epoch": 0.42188505747126437,
      "grad_norm": 0.6785594820976257,
      "learning_rate": 0.00015799742315479479,
      "loss": 2.0154,
      "num_input_tokens_seen": 6904400,
      "step": 1147,
      "train_runtime": 1160.9871,
      "train_tokens_per_second": 5947.008
    },
    {
      "epoch": 0.4222528735632184,
      "grad_norm": 0.7744385600090027,
      "learning_rate": 0.0001579606110804344,
      "loss": 1.8655,
      "num_input_tokens_seen": 6909200,
      "step": 1148,
      "train_runtime": 1161.8083,
      "train_tokens_per_second": 5946.936
    },
    {
      "epoch": 0.4226206896551724,
      "grad_norm": 0.7387827634811401,
      "learning_rate": 0.000157923799006074,
      "loss": 1.5813,
      "num_input_tokens_seen": 6917760,
      "step": 1149,
      "train_runtime": 1163.1436,
      "train_tokens_per_second": 5947.469
    },
    {
      "epoch": 0.42298850574712643,
      "grad_norm": 0.7304776310920715,
      "learning_rate": 0.00015788698693171359,
      "loss": 1.7869,
      "num_input_tokens_seen": 6922528,
      "step": 1150,
      "train_runtime": 1163.9539,
      "train_tokens_per_second": 5947.424
    },
    {
      "epoch": 0.42335632183908045,
      "grad_norm": 0.7163875699043274,
      "learning_rate": 0.00015785017485735322,
      "loss": 1.7904,
      "num_input_tokens_seen": 6928816,
      "step": 1151,
      "train_runtime": 1164.957,
      "train_tokens_per_second": 5947.701
    },
    {
      "epoch": 0.4237241379310345,
      "grad_norm": 0.7661460041999817,
      "learning_rate": 0.00015781336278299283,
      "loss": 1.7905,
      "num_input_tokens_seen": 6934768,
      "step": 1152,
      "train_runtime": 1165.9195,
      "train_tokens_per_second": 5947.896
    },
    {
      "epoch": 0.4240919540229885,
      "grad_norm": 0.7766834497451782,
      "learning_rate": 0.00015777655070863244,
      "loss": 1.8882,
      "num_input_tokens_seen": 6941296,
      "step": 1153,
      "train_runtime": 1166.9715,
      "train_tokens_per_second": 5948.128
    },
    {
      "epoch": 0.4244597701149425,
      "grad_norm": 0.7434842586517334,
      "learning_rate": 0.00015773973863427205,
      "loss": 1.8711,
      "num_input_tokens_seen": 6946032,
      "step": 1154,
      "train_runtime": 1167.8031,
      "train_tokens_per_second": 5947.948
    },
    {
      "epoch": 0.42482758620689653,
      "grad_norm": 0.6683353781700134,
      "learning_rate": 0.00015770292655991166,
      "loss": 1.728,
      "num_input_tokens_seen": 6951536,
      "step": 1155,
      "train_runtime": 1168.7244,
      "train_tokens_per_second": 5947.969
    },
    {
      "epoch": 0.42519540229885056,
      "grad_norm": 0.7363720536231995,
      "learning_rate": 0.00015766611448555127,
      "loss": 1.9358,
      "num_input_tokens_seen": 6956624,
      "step": 1156,
      "train_runtime": 1169.5912,
      "train_tokens_per_second": 5947.911
    },
    {
      "epoch": 0.4255632183908046,
      "grad_norm": 0.8716381192207336,
      "learning_rate": 0.00015762930241119087,
      "loss": 1.9163,
      "num_input_tokens_seen": 6961200,
      "step": 1157,
      "train_runtime": 1170.3755,
      "train_tokens_per_second": 5947.835
    },
    {
      "epoch": 0.4259310344827586,
      "grad_norm": 0.7376092076301575,
      "learning_rate": 0.00015759249033683048,
      "loss": 1.8922,
      "num_input_tokens_seen": 6969904,
      "step": 1158,
      "train_runtime": 1171.7204,
      "train_tokens_per_second": 5948.436
    },
    {
      "epoch": 0.4262988505747126,
      "grad_norm": 0.7594170570373535,
      "learning_rate": 0.00015755567826247012,
      "loss": 1.7216,
      "num_input_tokens_seen": 6978448,
      "step": 1159,
      "train_runtime": 1173.0555,
      "train_tokens_per_second": 5948.949
    },
    {
      "epoch": 0.4266666666666667,
      "grad_norm": 0.6948701739311218,
      "learning_rate": 0.00015751886618810973,
      "loss": 1.8632,
      "num_input_tokens_seen": 6983824,
      "step": 1160,
      "train_runtime": 1173.9271,
      "train_tokens_per_second": 5949.112
    },
    {
      "epoch": 0.4270344827586207,
      "grad_norm": 0.7659474015235901,
      "learning_rate": 0.0001574820541137493,
      "loss": 1.8775,
      "num_input_tokens_seen": 6988672,
      "step": 1161,
      "train_runtime": 1174.742,
      "train_tokens_per_second": 5949.112
    },
    {
      "epoch": 0.42740229885057474,
      "grad_norm": 0.7262035608291626,
      "learning_rate": 0.00015744524203938892,
      "loss": 1.9307,
      "num_input_tokens_seen": 6994112,
      "step": 1162,
      "train_runtime": 1175.6527,
      "train_tokens_per_second": 5949.131
    },
    {
      "epoch": 0.42777011494252876,
      "grad_norm": 0.7716181874275208,
      "learning_rate": 0.00015740842996502853,
      "loss": 1.7776,
      "num_input_tokens_seen": 7000016,
      "step": 1163,
      "train_runtime": 1176.6114,
      "train_tokens_per_second": 5949.302
    },
    {
      "epoch": 0.4281379310344828,
      "grad_norm": 0.7339833378791809,
      "learning_rate": 0.00015737161789066814,
      "loss": 1.5997,
      "num_input_tokens_seen": 7004448,
      "step": 1164,
      "train_runtime": 1177.3954,
      "train_tokens_per_second": 5949.104
    },
    {
      "epoch": 0.4285057471264368,
      "grad_norm": 0.7201822996139526,
      "learning_rate": 0.00015733480581630775,
      "loss": 1.7873,
      "num_input_tokens_seen": 7009872,
      "step": 1165,
      "train_runtime": 1178.324,
      "train_tokens_per_second": 5949.019
    },
    {
      "epoch": 0.4288735632183908,
      "grad_norm": 0.6710172295570374,
      "learning_rate": 0.00015729799374194735,
      "loss": 1.6479,
      "num_input_tokens_seen": 7016064,
      "step": 1166,
      "train_runtime": 1179.3121,
      "train_tokens_per_second": 5949.285
    },
    {
      "epoch": 0.42924137931034484,
      "grad_norm": 0.6840474605560303,
      "learning_rate": 0.00015726118166758696,
      "loss": 1.7431,
      "num_input_tokens_seen": 7022592,
      "step": 1167,
      "train_runtime": 1180.3673,
      "train_tokens_per_second": 5949.497
    },
    {
      "epoch": 0.42960919540229886,
      "grad_norm": 0.7766196727752686,
      "learning_rate": 0.0001572243695932266,
      "loss": 1.8158,
      "num_input_tokens_seen": 7027392,
      "step": 1168,
      "train_runtime": 1181.1975,
      "train_tokens_per_second": 5949.379
    },
    {
      "epoch": 0.4299770114942529,
      "grad_norm": 0.7490355968475342,
      "learning_rate": 0.0001571875575188662,
      "loss": 1.6519,
      "num_input_tokens_seen": 7033136,
      "step": 1169,
      "train_runtime": 1182.1302,
      "train_tokens_per_second": 5949.544
    },
    {
      "epoch": 0.4303448275862069,
      "grad_norm": 0.6676395535469055,
      "learning_rate": 0.00015715074544450582,
      "loss": 1.723,
      "num_input_tokens_seen": 7040672,
      "step": 1170,
      "train_runtime": 1183.3206,
      "train_tokens_per_second": 5949.927
    },
    {
      "epoch": 0.4307126436781609,
      "grad_norm": 0.7282621264457703,
      "learning_rate": 0.00015711393337014543,
      "loss": 1.6321,
      "num_input_tokens_seen": 7050512,
      "step": 1171,
      "train_runtime": 1185.3117,
      "train_tokens_per_second": 5948.235
    },
    {
      "epoch": 0.43108045977011494,
      "grad_norm": 0.7242358326911926,
      "learning_rate": 0.00015707712129578504,
      "loss": 1.742,
      "num_input_tokens_seen": 7057584,
      "step": 1172,
      "train_runtime": 1186.4304,
      "train_tokens_per_second": 5948.587
    },
    {
      "epoch": 0.43144827586206896,
      "grad_norm": 0.6883956789970398,
      "learning_rate": 0.00015704030922142464,
      "loss": 1.8834,
      "num_input_tokens_seen": 7062576,
      "step": 1173,
      "train_runtime": 1187.2793,
      "train_tokens_per_second": 5948.538
    },
    {
      "epoch": 0.431816091954023,
      "grad_norm": 0.8219404220581055,
      "learning_rate": 0.00015700349714706423,
      "loss": 1.92,
      "num_input_tokens_seen": 7068176,
      "step": 1174,
      "train_runtime": 1188.2348,
      "train_tokens_per_second": 5948.467
    },
    {
      "epoch": 0.432183908045977,
      "grad_norm": 0.7037672996520996,
      "learning_rate": 0.00015696668507270383,
      "loss": 1.6727,
      "num_input_tokens_seen": 7076592,
      "step": 1175,
      "train_runtime": 1189.5587,
      "train_tokens_per_second": 5948.922
    },
    {
      "epoch": 0.432551724137931,
      "grad_norm": 0.7427977323532104,
      "learning_rate": 0.00015692987299834347,
      "loss": 1.9977,
      "num_input_tokens_seen": 7081856,
      "step": 1176,
      "train_runtime": 1190.4593,
      "train_tokens_per_second": 5948.843
    },
    {
      "epoch": 0.43291954022988505,
      "grad_norm": 0.7989617586135864,
      "learning_rate": 0.00015689306092398308,
      "loss": 1.9313,
      "num_input_tokens_seen": 7086752,
      "step": 1177,
      "train_runtime": 1191.2925,
      "train_tokens_per_second": 5948.792
    },
    {
      "epoch": 0.43328735632183907,
      "grad_norm": 0.6868744492530823,
      "learning_rate": 0.0001568562488496227,
      "loss": 1.773,
      "num_input_tokens_seen": 7093040,
      "step": 1178,
      "train_runtime": 1192.3264,
      "train_tokens_per_second": 5948.908
    },
    {
      "epoch": 0.4336551724137931,
      "grad_norm": 0.6020117402076721,
      "learning_rate": 0.0001568194367752623,
      "loss": 1.7855,
      "num_input_tokens_seen": 7100992,
      "step": 1179,
      "train_runtime": 1193.5809,
      "train_tokens_per_second": 5949.318
    },
    {
      "epoch": 0.4340229885057471,
      "grad_norm": 0.7079392671585083,
      "learning_rate": 0.0001567826247009019,
      "loss": 1.794,
      "num_input_tokens_seen": 7105536,
      "step": 1180,
      "train_runtime": 1194.3955,
      "train_tokens_per_second": 5949.064
    },
    {
      "epoch": 0.43439080459770113,
      "grad_norm": 0.7041069269180298,
      "learning_rate": 0.00015674581262654152,
      "loss": 1.563,
      "num_input_tokens_seen": 7111936,
      "step": 1181,
      "train_runtime": 1195.4164,
      "train_tokens_per_second": 5949.338
    },
    {
      "epoch": 0.43475862068965515,
      "grad_norm": 0.7512297034263611,
      "learning_rate": 0.00015670900055218112,
      "loss": 1.9535,
      "num_input_tokens_seen": 7120448,
      "step": 1182,
      "train_runtime": 1196.7516,
      "train_tokens_per_second": 5949.813
    },
    {
      "epoch": 0.43512643678160917,
      "grad_norm": 0.6693354845046997,
      "learning_rate": 0.00015667218847782073,
      "loss": 1.5134,
      "num_input_tokens_seen": 7124464,
      "step": 1183,
      "train_runtime": 1197.4808,
      "train_tokens_per_second": 5949.543
    },
    {
      "epoch": 0.4354942528735632,
      "grad_norm": 0.6892227530479431,
      "learning_rate": 0.00015663537640346034,
      "loss": 1.8114,
      "num_input_tokens_seen": 7130464,
      "step": 1184,
      "train_runtime": 1198.4515,
      "train_tokens_per_second": 5949.731
    },
    {
      "epoch": 0.43586206896551727,
      "grad_norm": 0.6748680472373962,
      "learning_rate": 0.00015659856432909995,
      "loss": 1.6574,
      "num_input_tokens_seen": 7137408,
      "step": 1185,
      "train_runtime": 1199.5737,
      "train_tokens_per_second": 5949.954
    },
    {
      "epoch": 0.4362298850574713,
      "grad_norm": 0.754496693611145,
      "learning_rate": 0.00015656175225473956,
      "loss": 1.7061,
      "num_input_tokens_seen": 7143168,
      "step": 1186,
      "train_runtime": 1200.5203,
      "train_tokens_per_second": 5950.06
    },
    {
      "epoch": 0.4365977011494253,
      "grad_norm": 0.8417204022407532,
      "learning_rate": 0.00015652494018037917,
      "loss": 1.9689,
      "num_input_tokens_seen": 7149392,
      "step": 1187,
      "train_runtime": 1201.5352,
      "train_tokens_per_second": 5950.214
    },
    {
      "epoch": 0.43696551724137933,
      "grad_norm": 0.7532899379730225,
      "learning_rate": 0.00015648812810601878,
      "loss": 1.7614,
      "num_input_tokens_seen": 7154256,
      "step": 1188,
      "train_runtime": 1202.3806,
      "train_tokens_per_second": 5950.076
    },
    {
      "epoch": 0.43733333333333335,
      "grad_norm": 0.7637757062911987,
      "learning_rate": 0.00015645131603165839,
      "loss": 1.7781,
      "num_input_tokens_seen": 7159488,
      "step": 1189,
      "train_runtime": 1203.2554,
      "train_tokens_per_second": 5950.098
    },
    {
      "epoch": 0.4377011494252874,
      "grad_norm": 0.8735445737838745,
      "learning_rate": 0.000156414503957298,
      "loss": 1.8605,
      "num_input_tokens_seen": 7165008,
      "step": 1190,
      "train_runtime": 1204.1588,
      "train_tokens_per_second": 5950.218
    },
    {
      "epoch": 0.4380689655172414,
      "grad_norm": 0.72004634141922,
      "learning_rate": 0.0001563776918829376,
      "loss": 1.6518,
      "num_input_tokens_seen": 7170848,
      "step": 1191,
      "train_runtime": 1205.1289,
      "train_tokens_per_second": 5950.275
    },
    {
      "epoch": 0.4384367816091954,
      "grad_norm": 0.784618079662323,
      "learning_rate": 0.0001563408798085772,
      "loss": 1.715,
      "num_input_tokens_seen": 7176080,
      "step": 1192,
      "train_runtime": 1205.9955,
      "train_tokens_per_second": 5950.337
    },
    {
      "epoch": 0.43880459770114943,
      "grad_norm": 0.7545187473297119,
      "learning_rate": 0.00015630406773421685,
      "loss": 1.8402,
      "num_input_tokens_seen": 7181872,
      "step": 1193,
      "train_runtime": 1206.9579,
      "train_tokens_per_second": 5950.391
    },
    {
      "epoch": 0.43917241379310346,
      "grad_norm": 0.765870213508606,
      "learning_rate": 0.00015626725565985646,
      "loss": 1.826,
      "num_input_tokens_seen": 7187712,
      "step": 1194,
      "train_runtime": 1207.9153,
      "train_tokens_per_second": 5950.51
    },
    {
      "epoch": 0.4395402298850575,
      "grad_norm": 0.8489959239959717,
      "learning_rate": 0.00015623044358549607,
      "loss": 1.6647,
      "num_input_tokens_seen": 7194144,
      "step": 1195,
      "train_runtime": 1208.9361,
      "train_tokens_per_second": 5950.806
    },
    {
      "epoch": 0.4399080459770115,
      "grad_norm": 0.7975943088531494,
      "learning_rate": 0.00015619363151113568,
      "loss": 2.0866,
      "num_input_tokens_seen": 7198864,
      "step": 1196,
      "train_runtime": 1209.7679,
      "train_tokens_per_second": 5950.616
    },
    {
      "epoch": 0.4402758620689655,
      "grad_norm": 0.7624548077583313,
      "learning_rate": 0.00015615681943677526,
      "loss": 1.7441,
      "num_input_tokens_seen": 7205072,
      "step": 1197,
      "train_runtime": 1210.7784,
      "train_tokens_per_second": 5950.777
    },
    {
      "epoch": 0.44064367816091954,
      "grad_norm": 0.7394261956214905,
      "learning_rate": 0.00015612000736241487,
      "loss": 1.7958,
      "num_input_tokens_seen": 7210240,
      "step": 1198,
      "train_runtime": 1211.6565,
      "train_tokens_per_second": 5950.73
    },
    {
      "epoch": 0.44101149425287356,
      "grad_norm": 0.7611791491508484,
      "learning_rate": 0.00015608319528805448,
      "loss": 1.8613,
      "num_input_tokens_seen": 7215296,
      "step": 1199,
      "train_runtime": 1212.5166,
      "train_tokens_per_second": 5950.678
    },
    {
      "epoch": 0.4413793103448276,
      "grad_norm": 0.6911492943763733,
      "learning_rate": 0.00015604638321369408,
      "loss": 1.7601,
      "num_input_tokens_seen": 7225104,
      "step": 1200,
      "train_runtime": 1214.0236,
      "train_tokens_per_second": 5951.37
    },
    {
      "epoch": 0.4417471264367816,
      "grad_norm": 0.7862719893455505,
      "learning_rate": 0.0001560095711393337,
      "loss": 1.753,
      "num_input_tokens_seen": 7231952,
      "step": 1201,
      "train_runtime": 1215.6237,
      "train_tokens_per_second": 5949.17
    },
    {
      "epoch": 0.4421149425287356,
      "grad_norm": 0.7533947825431824,
      "learning_rate": 0.00015597275906497333,
      "loss": 1.8045,
      "num_input_tokens_seen": 7237680,
      "step": 1202,
      "train_runtime": 1216.5692,
      "train_tokens_per_second": 5949.255
    },
    {
      "epoch": 0.44248275862068964,
      "grad_norm": 0.7339320182800293,
      "learning_rate": 0.00015593594699061294,
      "loss": 1.9346,
      "num_input_tokens_seen": 7243376,
      "step": 1203,
      "train_runtime": 1217.5381,
      "train_tokens_per_second": 5949.199
    },
    {
      "epoch": 0.44285057471264366,
      "grad_norm": 0.6699645519256592,
      "learning_rate": 0.00015589913491625255,
      "loss": 1.6096,
      "num_input_tokens_seen": 7249152,
      "step": 1204,
      "train_runtime": 1218.4953,
      "train_tokens_per_second": 5949.266
    },
    {
      "epoch": 0.4432183908045977,
      "grad_norm": 0.7506150007247925,
      "learning_rate": 0.00015586232284189216,
      "loss": 1.7413,
      "num_input_tokens_seen": 7255808,
      "step": 1205,
      "train_runtime": 1219.5746,
      "train_tokens_per_second": 5949.458
    },
    {
      "epoch": 0.4435862068965517,
      "grad_norm": 0.7534520030021667,
      "learning_rate": 0.00015582551076753176,
      "loss": 1.8262,
      "num_input_tokens_seen": 7260432,
      "step": 1206,
      "train_runtime": 1220.3457,
      "train_tokens_per_second": 5949.488
    },
    {
      "epoch": 0.4439540229885057,
      "grad_norm": 0.7158861756324768,
      "learning_rate": 0.00015578869869317137,
      "loss": 1.9487,
      "num_input_tokens_seen": 7265024,
      "step": 1207,
      "train_runtime": 1221.1488,
      "train_tokens_per_second": 5949.336
    },
    {
      "epoch": 0.44432183908045975,
      "grad_norm": 0.7745321989059448,
      "learning_rate": 0.00015575188661881098,
      "loss": 1.7816,
      "num_input_tokens_seen": 7270432,
      "step": 1208,
      "train_runtime": 1222.0681,
      "train_tokens_per_second": 5949.286
    },
    {
      "epoch": 0.44468965517241377,
      "grad_norm": 0.743904709815979,
      "learning_rate": 0.0001557150745444506,
      "loss": 1.8334,
      "num_input_tokens_seen": 7275296,
      "step": 1209,
      "train_runtime": 1222.8908,
      "train_tokens_per_second": 5949.261
    },
    {
      "epoch": 0.44505747126436784,
      "grad_norm": 0.7275137305259705,
      "learning_rate": 0.00015567826247009017,
      "loss": 1.6622,
      "num_input_tokens_seen": 7281088,
      "step": 1210,
      "train_runtime": 1223.8488,
      "train_tokens_per_second": 5949.336
    },
    {
      "epoch": 0.44542528735632186,
      "grad_norm": 0.7486593723297119,
      "learning_rate": 0.0001556414503957298,
      "loss": 1.7694,
      "num_input_tokens_seen": 7292704,
      "step": 1211,
      "train_runtime": 1225.6052,
      "train_tokens_per_second": 5950.288
    },
    {
      "epoch": 0.4457931034482759,
      "grad_norm": 0.7362422943115234,
      "learning_rate": 0.00015560463832136942,
      "loss": 1.8745,
      "num_input_tokens_seen": 7297536,
      "step": 1212,
      "train_runtime": 1226.4411,
      "train_tokens_per_second": 5950.173
    },
    {
      "epoch": 0.4461609195402299,
      "grad_norm": 0.658754825592041,
      "learning_rate": 0.00015556782624700903,
      "loss": 1.6796,
      "num_input_tokens_seen": 7304240,
      "step": 1213,
      "train_runtime": 1227.5015,
      "train_tokens_per_second": 5950.494
    },
    {
      "epoch": 0.4465287356321839,
      "grad_norm": 0.7258212566375732,
      "learning_rate": 0.00015553101417264864,
      "loss": 1.8287,
      "num_input_tokens_seen": 7310704,
      "step": 1214,
      "train_runtime": 1228.5276,
      "train_tokens_per_second": 5950.786
    },
    {
      "epoch": 0.44689655172413795,
      "grad_norm": 0.8535351753234863,
      "learning_rate": 0.00015549420209828824,
      "loss": 1.8639,
      "num_input_tokens_seen": 7314928,
      "step": 1215,
      "train_runtime": 1229.2914,
      "train_tokens_per_second": 5950.524
    },
    {
      "epoch": 0.44726436781609197,
      "grad_norm": 0.7198713421821594,
      "learning_rate": 0.00015545739002392785,
      "loss": 1.7436,
      "num_input_tokens_seen": 7319600,
      "step": 1216,
      "train_runtime": 1230.0666,
      "train_tokens_per_second": 5950.572
    },
    {
      "epoch": 0.447632183908046,
      "grad_norm": 0.8227468132972717,
      "learning_rate": 0.00015542057794956746,
      "loss": 1.8286,
      "num_input_tokens_seen": 7327216,
      "step": 1217,
      "train_runtime": 1231.2571,
      "train_tokens_per_second": 5951.004
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.7701994776725769,
      "learning_rate": 0.00015538376587520707,
      "loss": 1.7154,
      "num_input_tokens_seen": 7333424,
      "step": 1218,
      "train_runtime": 1232.2851,
      "train_tokens_per_second": 5951.078
    },
    {
      "epoch": 0.44836781609195403,
      "grad_norm": 0.7645250558853149,
      "learning_rate": 0.0001553469538008467,
      "loss": 1.6711,
      "num_input_tokens_seen": 7339216,
      "step": 1219,
      "train_runtime": 1233.2616,
      "train_tokens_per_second": 5951.062
    },
    {
      "epoch": 0.44873563218390805,
      "grad_norm": 0.6568512916564941,
      "learning_rate": 0.0001553101417264863,
      "loss": 1.7855,
      "num_input_tokens_seen": 7347520,
      "step": 1220,
      "train_runtime": 1234.5633,
      "train_tokens_per_second": 5951.513
    },
    {
      "epoch": 0.44910344827586207,
      "grad_norm": 0.7112115621566772,
      "learning_rate": 0.0001552733296521259,
      "loss": 1.9502,
      "num_input_tokens_seen": 7352064,
      "step": 1221,
      "train_runtime": 1235.3536,
      "train_tokens_per_second": 5951.384
    },
    {
      "epoch": 0.4494712643678161,
      "grad_norm": 0.7758644819259644,
      "learning_rate": 0.0001552365175777655,
      "loss": 1.874,
      "num_input_tokens_seen": 7356480,
      "step": 1222,
      "train_runtime": 1236.1272,
      "train_tokens_per_second": 5951.232
    },
    {
      "epoch": 0.4498390804597701,
      "grad_norm": 0.7849825620651245,
      "learning_rate": 0.00015519970550340512,
      "loss": 1.7783,
      "num_input_tokens_seen": 7361136,
      "step": 1223,
      "train_runtime": 1236.9524,
      "train_tokens_per_second": 5951.026
    },
    {
      "epoch": 0.45020689655172413,
      "grad_norm": 0.799024760723114,
      "learning_rate": 0.00015516289342904472,
      "loss": 1.7842,
      "num_input_tokens_seen": 7365520,
      "step": 1224,
      "train_runtime": 1237.6999,
      "train_tokens_per_second": 5950.974
    },
    {
      "epoch": 0.45057471264367815,
      "grad_norm": 0.7010101079940796,
      "learning_rate": 0.00015512608135468433,
      "loss": 1.7123,
      "num_input_tokens_seen": 7375312,
      "step": 1225,
      "train_runtime": 1239.2089,
      "train_tokens_per_second": 5951.629
    },
    {
      "epoch": 0.4509425287356322,
      "grad_norm": 0.7189601063728333,
      "learning_rate": 0.00015508926928032394,
      "loss": 1.9222,
      "num_input_tokens_seen": 7380256,
      "step": 1226,
      "train_runtime": 1240.0504,
      "train_tokens_per_second": 5951.578
    },
    {
      "epoch": 0.4513103448275862,
      "grad_norm": 0.7187652587890625,
      "learning_rate": 0.00015505245720596355,
      "loss": 1.6839,
      "num_input_tokens_seen": 7385632,
      "step": 1227,
      "train_runtime": 1240.9305,
      "train_tokens_per_second": 5951.689
    },
    {
      "epoch": 0.4516781609195402,
      "grad_norm": 0.8691778779029846,
      "learning_rate": 0.0001550156451316032,
      "loss": 1.8125,
      "num_input_tokens_seen": 7390368,
      "step": 1228,
      "train_runtime": 1241.7399,
      "train_tokens_per_second": 5951.623
    },
    {
      "epoch": 0.45204597701149424,
      "grad_norm": 0.7041910886764526,
      "learning_rate": 0.0001549788330572428,
      "loss": 1.9237,
      "num_input_tokens_seen": 7395872,
      "step": 1229,
      "train_runtime": 1242.659,
      "train_tokens_per_second": 5951.65
    },
    {
      "epoch": 0.45241379310344826,
      "grad_norm": 0.7146864533424377,
      "learning_rate": 0.0001549420209828824,
      "loss": 1.6705,
      "num_input_tokens_seen": 7403504,
      "step": 1230,
      "train_runtime": 1243.835,
      "train_tokens_per_second": 5952.159
    },
    {
      "epoch": 0.4527816091954023,
      "grad_norm": 0.7021743655204773,
      "learning_rate": 0.00015490520890852201,
      "loss": 1.5599,
      "num_input_tokens_seen": 7409776,
      "step": 1231,
      "train_runtime": 1245.342,
      "train_tokens_per_second": 5949.993
    },
    {
      "epoch": 0.4531494252873563,
      "grad_norm": 0.7877197265625,
      "learning_rate": 0.00015486839683416162,
      "loss": 2.0735,
      "num_input_tokens_seen": 7414736,
      "step": 1232,
      "train_runtime": 1246.1915,
      "train_tokens_per_second": 5949.917
    },
    {
      "epoch": 0.4535172413793103,
      "grad_norm": 0.8014360070228577,
      "learning_rate": 0.00015483158475980123,
      "loss": 1.9159,
      "num_input_tokens_seen": 7424320,
      "step": 1233,
      "train_runtime": 1247.6426,
      "train_tokens_per_second": 5950.679
    },
    {
      "epoch": 0.45388505747126434,
      "grad_norm": 0.7110464572906494,
      "learning_rate": 0.0001547947726854408,
      "loss": 1.8617,
      "num_input_tokens_seen": 7429328,
      "step": 1234,
      "train_runtime": 1248.4905,
      "train_tokens_per_second": 5950.649
    },
    {
      "epoch": 0.4542528735632184,
      "grad_norm": 0.7932233810424805,
      "learning_rate": 0.00015475796061108042,
      "loss": 2.0679,
      "num_input_tokens_seen": 7433936,
      "step": 1235,
      "train_runtime": 1249.2881,
      "train_tokens_per_second": 5950.538
    },
    {
      "epoch": 0.45462068965517244,
      "grad_norm": 0.641743004322052,
      "learning_rate": 0.00015472114853672006,
      "loss": 1.7498,
      "num_input_tokens_seen": 7441360,
      "step": 1236,
      "train_runtime": 1250.4701,
      "train_tokens_per_second": 5950.85
    },
    {
      "epoch": 0.45498850574712646,
      "grad_norm": 0.6838722825050354,
      "learning_rate": 0.00015468433646235967,
      "loss": 1.9534,
      "num_input_tokens_seen": 7446112,
      "step": 1237,
      "train_runtime": 1251.2728,
      "train_tokens_per_second": 5950.83
    },
    {
      "epoch": 0.4553563218390805,
      "grad_norm": 0.6989526152610779,
      "learning_rate": 0.00015464752438799928,
      "loss": 1.7391,
      "num_input_tokens_seen": 7452720,
      "step": 1238,
      "train_runtime": 1252.3217,
      "train_tokens_per_second": 5951.123
    },
    {
      "epoch": 0.4557241379310345,
      "grad_norm": 0.7069482803344727,
      "learning_rate": 0.00015461071231363889,
      "loss": 1.6438,
      "num_input_tokens_seen": 7460480,
      "step": 1239,
      "train_runtime": 1253.5605,
      "train_tokens_per_second": 5951.432
    },
    {
      "epoch": 0.4560919540229885,
      "grad_norm": 0.8050363063812256,
      "learning_rate": 0.0001545739002392785,
      "loss": 1.8958,
      "num_input_tokens_seen": 7465232,
      "step": 1240,
      "train_runtime": 1254.3919,
      "train_tokens_per_second": 5951.276
    },
    {
      "epoch": 0.45645977011494254,
      "grad_norm": 0.7380297780036926,
      "learning_rate": 0.0001545370881649181,
      "loss": 1.8833,
      "num_input_tokens_seen": 7472992,
      "step": 1241,
      "train_runtime": 1255.6188,
      "train_tokens_per_second": 5951.641
    },
    {
      "epoch": 0.45682758620689656,
      "grad_norm": 0.7242332100868225,
      "learning_rate": 0.0001545002760905577,
      "loss": 1.7404,
      "num_input_tokens_seen": 7477664,
      "step": 1242,
      "train_runtime": 1256.4113,
      "train_tokens_per_second": 5951.605
    },
    {
      "epoch": 0.4571954022988506,
      "grad_norm": 0.7481887936592102,
      "learning_rate": 0.00015446346401619732,
      "loss": 2.011,
      "num_input_tokens_seen": 7482560,
      "step": 1243,
      "train_runtime": 1257.2501,
      "train_tokens_per_second": 5951.529
    },
    {
      "epoch": 0.4575632183908046,
      "grad_norm": 0.6763362288475037,
      "learning_rate": 0.00015442665194183693,
      "loss": 1.7012,
      "num_input_tokens_seen": 7489152,
      "step": 1244,
      "train_runtime": 1258.3283,
      "train_tokens_per_second": 5951.668
    },
    {
      "epoch": 0.4579310344827586,
      "grad_norm": 0.7744765281677246,
      "learning_rate": 0.00015438983986747654,
      "loss": 1.8941,
      "num_input_tokens_seen": 7496928,
      "step": 1245,
      "train_runtime": 1259.5245,
      "train_tokens_per_second": 5952.189
    },
    {
      "epoch": 0.45829885057471265,
      "grad_norm": 0.6633315682411194,
      "learning_rate": 0.00015435302779311615,
      "loss": 1.6114,
      "num_input_tokens_seen": 7503776,
      "step": 1246,
      "train_runtime": 1260.6242,
      "train_tokens_per_second": 5952.429
    },
    {
      "epoch": 0.45866666666666667,
      "grad_norm": 0.7456023097038269,
      "learning_rate": 0.00015431621571875576,
      "loss": 1.7191,
      "num_input_tokens_seen": 7507968,
      "step": 1247,
      "train_runtime": 1261.355,
      "train_tokens_per_second": 5952.304
    },
    {
      "epoch": 0.4590344827586207,
      "grad_norm": 0.7540046572685242,
      "learning_rate": 0.00015427940364439537,
      "loss": 1.797,
      "num_input_tokens_seen": 7512928,
      "step": 1248,
      "train_runtime": 1262.2075,
      "train_tokens_per_second": 5952.213
    },
    {
      "epoch": 0.4594022988505747,
      "grad_norm": 0.6673805713653564,
      "learning_rate": 0.00015424259157003497,
      "loss": 1.7042,
      "num_input_tokens_seen": 7520000,
      "step": 1249,
      "train_runtime": 1263.3601,
      "train_tokens_per_second": 5952.381
    },
    {
      "epoch": 0.45977011494252873,
      "grad_norm": 0.7562255859375,
      "learning_rate": 0.00015420577949567458,
      "loss": 1.915,
      "num_input_tokens_seen": 7526640,
      "step": 1250,
      "train_runtime": 1264.4259,
      "train_tokens_per_second": 5952.615
    },
    {
      "epoch": 0.46013793103448275,
      "grad_norm": 0.7367727160453796,
      "learning_rate": 0.0001541689674213142,
      "loss": 1.9246,
      "num_input_tokens_seen": 7533120,
      "step": 1251,
      "train_runtime": 1265.5062,
      "train_tokens_per_second": 5952.654
    },
    {
      "epoch": 0.46050574712643677,
      "grad_norm": 0.7032127380371094,
      "learning_rate": 0.0001541321553469538,
      "loss": 1.7504,
      "num_input_tokens_seen": 7541296,
      "step": 1252,
      "train_runtime": 1266.8118,
      "train_tokens_per_second": 5952.973
    },
    {
      "epoch": 0.4608735632183908,
      "grad_norm": 0.7035644054412842,
      "learning_rate": 0.00015409534327259344,
      "loss": 1.8283,
      "num_input_tokens_seen": 7546064,
      "step": 1253,
      "train_runtime": 1267.6159,
      "train_tokens_per_second": 5952.958
    },
    {
      "epoch": 0.4612413793103448,
      "grad_norm": 0.7468724250793457,
      "learning_rate": 0.00015405853119823305,
      "loss": 1.8595,
      "num_input_tokens_seen": 7551648,
      "step": 1254,
      "train_runtime": 1268.5227,
      "train_tokens_per_second": 5953.104
    },
    {
      "epoch": 0.46160919540229883,
      "grad_norm": 0.7778677940368652,
      "learning_rate": 0.00015402171912387265,
      "loss": 1.8202,
      "num_input_tokens_seen": 7558864,
      "step": 1255,
      "train_runtime": 1269.6841,
      "train_tokens_per_second": 5953.342
    },
    {
      "epoch": 0.46197701149425285,
      "grad_norm": 0.6948017477989197,
      "learning_rate": 0.00015398490704951226,
      "loss": 1.8597,
      "num_input_tokens_seen": 7563968,
      "step": 1256,
      "train_runtime": 1270.5369,
      "train_tokens_per_second": 5953.364
    },
    {
      "epoch": 0.4623448275862069,
      "grad_norm": 0.7547342777252197,
      "learning_rate": 0.00015394809497515185,
      "loss": 1.7094,
      "num_input_tokens_seen": 7568784,
      "step": 1257,
      "train_runtime": 1271.3599,
      "train_tokens_per_second": 5953.298
    },
    {
      "epoch": 0.4627126436781609,
      "grad_norm": 0.7779310941696167,
      "learning_rate": 0.00015391128290079145,
      "loss": 1.8208,
      "num_input_tokens_seen": 7573456,
      "step": 1258,
      "train_runtime": 1272.1845,
      "train_tokens_per_second": 5953.111
    },
    {
      "epoch": 0.4630804597701149,
      "grad_norm": 0.736392617225647,
      "learning_rate": 0.00015387447082643106,
      "loss": 1.6906,
      "num_input_tokens_seen": 7578064,
      "step": 1259,
      "train_runtime": 1272.9577,
      "train_tokens_per_second": 5953.115
    },
    {
      "epoch": 0.463448275862069,
      "grad_norm": 0.6372681260108948,
      "learning_rate": 0.00015383765875207067,
      "loss": 1.8011,
      "num_input_tokens_seen": 7583200,
      "step": 1260,
      "train_runtime": 1273.8091,
      "train_tokens_per_second": 5953.168
    },
    {
      "epoch": 0.463816091954023,
      "grad_norm": 0.7548249363899231,
      "learning_rate": 0.00015380084667771028,
      "loss": 1.9357,
      "num_input_tokens_seen": 7593792,
      "step": 1261,
      "train_runtime": 1275.9045,
      "train_tokens_per_second": 5951.693
    },
    {
      "epoch": 0.46418390804597703,
      "grad_norm": 0.7687066793441772,
      "learning_rate": 0.00015376403460334992,
      "loss": 1.7145,
      "num_input_tokens_seen": 7600304,
      "step": 1262,
      "train_runtime": 1276.9643,
      "train_tokens_per_second": 5951.853
    },
    {
      "epoch": 0.46455172413793105,
      "grad_norm": 0.6674917936325073,
      "learning_rate": 0.00015372722252898953,
      "loss": 1.6502,
      "num_input_tokens_seen": 7607616,
      "step": 1263,
      "train_runtime": 1278.1101,
      "train_tokens_per_second": 5952.238
    },
    {
      "epoch": 0.4649195402298851,
      "grad_norm": 0.7151092886924744,
      "learning_rate": 0.00015369041045462913,
      "loss": 1.8285,
      "num_input_tokens_seen": 7612368,
      "step": 1264,
      "train_runtime": 1278.9395,
      "train_tokens_per_second": 5952.094
    },
    {
      "epoch": 0.4652873563218391,
      "grad_norm": 0.7578483819961548,
      "learning_rate": 0.00015365359838026874,
      "loss": 1.8126,
      "num_input_tokens_seen": 7617424,
      "step": 1265,
      "train_runtime": 1279.8069,
      "train_tokens_per_second": 5952.011
    },
    {
      "epoch": 0.4656551724137931,
      "grad_norm": 0.7614312767982483,
      "learning_rate": 0.00015361678630590835,
      "loss": 1.7758,
      "num_input_tokens_seen": 7625232,
      "step": 1266,
      "train_runtime": 1281.0303,
      "train_tokens_per_second": 5952.421
    },
    {
      "epoch": 0.46602298850574714,
      "grad_norm": 0.7852227687835693,
      "learning_rate": 0.00015357997423154796,
      "loss": 1.5556,
      "num_input_tokens_seen": 7631440,
      "step": 1267,
      "train_runtime": 1282.0452,
      "train_tokens_per_second": 5952.552
    },
    {
      "epoch": 0.46639080459770116,
      "grad_norm": 0.7623286843299866,
      "learning_rate": 0.00015354316215718757,
      "loss": 2.0483,
      "num_input_tokens_seen": 7636320,
      "step": 1268,
      "train_runtime": 1282.8609,
      "train_tokens_per_second": 5952.57
    },
    {
      "epoch": 0.4667586206896552,
      "grad_norm": 0.7269994616508484,
      "learning_rate": 0.00015350635008282718,
      "loss": 1.8532,
      "num_input_tokens_seen": 7641104,
      "step": 1269,
      "train_runtime": 1283.694,
      "train_tokens_per_second": 5952.434
    },
    {
      "epoch": 0.4671264367816092,
      "grad_norm": 0.7663736343383789,
      "learning_rate": 0.0001534695380084668,
      "loss": 1.7431,
      "num_input_tokens_seen": 7645760,
      "step": 1270,
      "train_runtime": 1284.4938,
      "train_tokens_per_second": 5952.353
    },
    {
      "epoch": 0.4674942528735632,
      "grad_norm": 0.6806995868682861,
      "learning_rate": 0.0001534327259341064,
      "loss": 1.7246,
      "num_input_tokens_seen": 7650608,
      "step": 1271,
      "train_runtime": 1285.3396,
      "train_tokens_per_second": 5952.207
    },
    {
      "epoch": 0.46786206896551724,
      "grad_norm": 0.7199662923812866,
      "learning_rate": 0.000153395913859746,
      "loss": 1.7299,
      "num_input_tokens_seen": 7657392,
      "step": 1272,
      "train_runtime": 1286.441,
      "train_tokens_per_second": 5952.385
    },
    {
      "epoch": 0.46822988505747126,
      "grad_norm": 0.7160378694534302,
      "learning_rate": 0.00015335910178538561,
      "loss": 1.7186,
      "num_input_tokens_seen": 7662368,
      "step": 1273,
      "train_runtime": 1287.2697,
      "train_tokens_per_second": 5952.418
    },
    {
      "epoch": 0.4685977011494253,
      "grad_norm": 0.7738577723503113,
      "learning_rate": 0.00015332228971102522,
      "loss": 1.8797,
      "num_input_tokens_seen": 7667728,
      "step": 1274,
      "train_runtime": 1288.1874,
      "train_tokens_per_second": 5952.339
    },
    {
      "epoch": 0.4689655172413793,
      "grad_norm": 0.8060719966888428,
      "learning_rate": 0.00015328547763666483,
      "loss": 1.8157,
      "num_input_tokens_seen": 7672352,
      "step": 1275,
      "train_runtime": 1288.987,
      "train_tokens_per_second": 5952.234
    },
    {
      "epoch": 0.4693333333333333,
      "grad_norm": 0.7524076700210571,
      "learning_rate": 0.00015324866556230444,
      "loss": 1.8068,
      "num_input_tokens_seen": 7676912,
      "step": 1276,
      "train_runtime": 1289.7771,
      "train_tokens_per_second": 5952.123
    },
    {
      "epoch": 0.46970114942528735,
      "grad_norm": 0.7522364258766174,
      "learning_rate": 0.00015321185348794405,
      "loss": 1.6173,
      "num_input_tokens_seen": 7682224,
      "step": 1277,
      "train_runtime": 1290.6579,
      "train_tokens_per_second": 5952.177
    },
    {
      "epoch": 0.47006896551724137,
      "grad_norm": 0.6947898864746094,
      "learning_rate": 0.00015317504141358366,
      "loss": 1.7595,
      "num_input_tokens_seen": 7689296,
      "step": 1278,
      "train_runtime": 1291.8022,
      "train_tokens_per_second": 5952.379
    },
    {
      "epoch": 0.4704367816091954,
      "grad_norm": 0.7850360870361328,
      "learning_rate": 0.0001531382293392233,
      "loss": 1.7286,
      "num_input_tokens_seen": 7693856,
      "step": 1279,
      "train_runtime": 1292.5775,
      "train_tokens_per_second": 5952.336
    },
    {
      "epoch": 0.4708045977011494,
      "grad_norm": 0.7709944844245911,
      "learning_rate": 0.00015310141726486288,
      "loss": 1.9462,
      "num_input_tokens_seen": 7701376,
      "step": 1280,
      "train_runtime": 1293.7759,
      "train_tokens_per_second": 5952.635
    },
    {
      "epoch": 0.47117241379310343,
      "grad_norm": 0.7349576354026794,
      "learning_rate": 0.00015306460519050249,
      "loss": 1.8251,
      "num_input_tokens_seen": 7707600,
      "step": 1281,
      "train_runtime": 1294.793,
      "train_tokens_per_second": 5952.766
    },
    {
      "epoch": 0.47154022988505745,
      "grad_norm": 0.8141226768493652,
      "learning_rate": 0.0001530277931161421,
      "loss": 2.045,
      "num_input_tokens_seen": 7713360,
      "step": 1282,
      "train_runtime": 1295.7657,
      "train_tokens_per_second": 5952.743
    },
    {
      "epoch": 0.47190804597701147,
      "grad_norm": 0.7103748917579651,
      "learning_rate": 0.0001529909810417817,
      "loss": 1.8578,
      "num_input_tokens_seen": 7718976,
      "step": 1283,
      "train_runtime": 1296.7129,
      "train_tokens_per_second": 5952.725
    },
    {
      "epoch": 0.4722758620689655,
      "grad_norm": 0.7216552495956421,
      "learning_rate": 0.0001529541689674213,
      "loss": 1.7909,
      "num_input_tokens_seen": 7725968,
      "step": 1284,
      "train_runtime": 1297.8497,
      "train_tokens_per_second": 5952.899
    },
    {
      "epoch": 0.47264367816091957,
      "grad_norm": 0.7933657765388489,
      "learning_rate": 0.00015291735689306092,
      "loss": 1.6847,
      "num_input_tokens_seen": 7731856,
      "step": 1285,
      "train_runtime": 1298.8148,
      "train_tokens_per_second": 5953.009
    },
    {
      "epoch": 0.4730114942528736,
      "grad_norm": 0.7735483050346375,
      "learning_rate": 0.00015288054481870053,
      "loss": 1.857,
      "num_input_tokens_seen": 7736144,
      "step": 1286,
      "train_runtime": 1299.5438,
      "train_tokens_per_second": 5952.969
    },
    {
      "epoch": 0.4733793103448276,
      "grad_norm": 0.8700661063194275,
      "learning_rate": 0.00015284373274434014,
      "loss": 2.1009,
      "num_input_tokens_seen": 7740864,
      "step": 1287,
      "train_runtime": 1300.3149,
      "train_tokens_per_second": 5953.069
    },
    {
      "epoch": 0.47374712643678163,
      "grad_norm": 0.7280611991882324,
      "learning_rate": 0.00015280692066997977,
      "loss": 1.8547,
      "num_input_tokens_seen": 7746336,
      "step": 1288,
      "train_runtime": 1301.2355,
      "train_tokens_per_second": 5953.062
    },
    {
      "epoch": 0.47411494252873565,
      "grad_norm": 0.6782256364822388,
      "learning_rate": 0.00015277010859561938,
      "loss": 1.6634,
      "num_input_tokens_seen": 7751888,
      "step": 1289,
      "train_runtime": 1302.13,
      "train_tokens_per_second": 5953.236
    },
    {
      "epoch": 0.47448275862068967,
      "grad_norm": 0.7127692699432373,
      "learning_rate": 0.000152733296521259,
      "loss": 1.6083,
      "num_input_tokens_seen": 7757120,
      "step": 1290,
      "train_runtime": 1303.0133,
      "train_tokens_per_second": 5953.216
    },
    {
      "epoch": 0.4748505747126437,
      "grad_norm": 0.7067503333091736,
      "learning_rate": 0.0001526964844468986,
      "loss": 1.7024,
      "num_input_tokens_seen": 7764352,
      "step": 1291,
      "train_runtime": 1304.6325,
      "train_tokens_per_second": 5951.371
    },
    {
      "epoch": 0.4752183908045977,
      "grad_norm": 0.6711677312850952,
      "learning_rate": 0.0001526596723725382,
      "loss": 1.627,
      "num_input_tokens_seen": 7769200,
      "step": 1292,
      "train_runtime": 1305.4533,
      "train_tokens_per_second": 5951.343
    },
    {
      "epoch": 0.47558620689655173,
      "grad_norm": 0.7733351588249207,
      "learning_rate": 0.0001526228602981778,
      "loss": 1.8773,
      "num_input_tokens_seen": 7773072,
      "step": 1293,
      "train_runtime": 1306.1194,
      "train_tokens_per_second": 5951.272
    },
    {
      "epoch": 0.47595402298850575,
      "grad_norm": 0.6531370282173157,
      "learning_rate": 0.0001525860482238174,
      "loss": 1.8501,
      "num_input_tokens_seen": 7779600,
      "step": 1294,
      "train_runtime": 1307.2063,
      "train_tokens_per_second": 5951.318
    },
    {
      "epoch": 0.4763218390804598,
      "grad_norm": 0.6826766133308411,
      "learning_rate": 0.000152549236149457,
      "loss": 1.7617,
      "num_input_tokens_seen": 7784288,
      "step": 1295,
      "train_runtime": 1308.0051,
      "train_tokens_per_second": 5951.267
    },
    {
      "epoch": 0.4766896551724138,
      "grad_norm": 0.7375050783157349,
      "learning_rate": 0.00015251242407509665,
      "loss": 1.6421,
      "num_input_tokens_seen": 7790512,
      "step": 1296,
      "train_runtime": 1309.0152,
      "train_tokens_per_second": 5951.43
    },
    {
      "epoch": 0.4770574712643678,
      "grad_norm": 0.7035674452781677,
      "learning_rate": 0.00015247561200073625,
      "loss": 1.8262,
      "num_input_tokens_seen": 7796176,
      "step": 1297,
      "train_runtime": 1309.9483,
      "train_tokens_per_second": 5951.514
    },
    {
      "epoch": 0.47742528735632184,
      "grad_norm": 0.7339024543762207,
      "learning_rate": 0.00015243879992637586,
      "loss": 1.6042,
      "num_input_tokens_seen": 7800928,
      "step": 1298,
      "train_runtime": 1310.7584,
      "train_tokens_per_second": 5951.461
    },
    {
      "epoch": 0.47779310344827586,
      "grad_norm": 0.753303587436676,
      "learning_rate": 0.00015240198785201547,
      "loss": 1.936,
      "num_input_tokens_seen": 7807536,
      "step": 1299,
      "train_runtime": 1311.8357,
      "train_tokens_per_second": 5951.611
    },
    {
      "epoch": 0.4781609195402299,
      "grad_norm": 0.6756641864776611,
      "learning_rate": 0.00015236517577765508,
      "loss": 1.7588,
      "num_input_tokens_seen": 7812896,
      "step": 1300,
      "train_runtime": 1312.7304,
      "train_tokens_per_second": 5951.638
    },
    {
      "epoch": 0.4785287356321839,
      "grad_norm": 0.8002522587776184,
      "learning_rate": 0.0001523283637032947,
      "loss": 1.8177,
      "num_input_tokens_seen": 7818576,
      "step": 1301,
      "train_runtime": 1313.6618,
      "train_tokens_per_second": 5951.742
    },
    {
      "epoch": 0.4788965517241379,
      "grad_norm": 0.752668023109436,
      "learning_rate": 0.0001522915516289343,
      "loss": 1.6623,
      "num_input_tokens_seen": 7823008,
      "step": 1302,
      "train_runtime": 1314.4388,
      "train_tokens_per_second": 5951.595
    },
    {
      "epoch": 0.47926436781609194,
      "grad_norm": 0.7155261635780334,
      "learning_rate": 0.0001522547395545739,
      "loss": 1.922,
      "num_input_tokens_seen": 7827744,
      "step": 1303,
      "train_runtime": 1315.2469,
      "train_tokens_per_second": 5951.54
    },
    {
      "epoch": 0.47963218390804596,
      "grad_norm": 0.7759922742843628,
      "learning_rate": 0.00015221792748021352,
      "loss": 1.8296,
      "num_input_tokens_seen": 7834736,
      "step": 1304,
      "train_runtime": 1316.3564,
      "train_tokens_per_second": 5951.835
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.6895577311515808,
      "learning_rate": 0.00015218111540585313,
      "loss": 1.57,
      "num_input_tokens_seen": 7842624,
      "step": 1305,
      "train_runtime": 1317.5822,
      "train_tokens_per_second": 5952.285
    },
    {
      "epoch": 0.480367816091954,
      "grad_norm": 0.8040077686309814,
      "learning_rate": 0.00015214430333149274,
      "loss": 1.6446,
      "num_input_tokens_seen": 7847312,
      "step": 1306,
      "train_runtime": 1318.4024,
      "train_tokens_per_second": 5952.137
    },
    {
      "epoch": 0.480735632183908,
      "grad_norm": 0.7056049704551697,
      "learning_rate": 0.00015210749125713234,
      "loss": 1.7206,
      "num_input_tokens_seen": 7854928,
      "step": 1307,
      "train_runtime": 1319.6195,
      "train_tokens_per_second": 5952.419
    },
    {
      "epoch": 0.48110344827586204,
      "grad_norm": 0.6856859922409058,
      "learning_rate": 0.00015207067918277195,
      "loss": 1.8034,
      "num_input_tokens_seen": 7864432,
      "step": 1308,
      "train_runtime": 1321.0915,
      "train_tokens_per_second": 5952.981
    },
    {
      "epoch": 0.48147126436781607,
      "grad_norm": 0.7165193557739258,
      "learning_rate": 0.00015203386710841156,
      "loss": 1.6175,
      "num_input_tokens_seen": 7870768,
      "step": 1309,
      "train_runtime": 1322.1183,
      "train_tokens_per_second": 5953.149
    },
    {
      "epoch": 0.48183908045977014,
      "grad_norm": 0.7350046634674072,
      "learning_rate": 0.00015199705503405117,
      "loss": 1.8635,
      "num_input_tokens_seen": 7878960,
      "step": 1310,
      "train_runtime": 1323.3977,
      "train_tokens_per_second": 5953.585
    },
    {
      "epoch": 0.48220689655172416,
      "grad_norm": 0.773739218711853,
      "learning_rate": 0.00015196024295969078,
      "loss": 1.8664,
      "num_input_tokens_seen": 7884144,
      "step": 1311,
      "train_runtime": 1324.2627,
      "train_tokens_per_second": 5953.61
    },
    {
      "epoch": 0.4825747126436782,
      "grad_norm": 0.7551418542861938,
      "learning_rate": 0.0001519234308853304,
      "loss": 1.9343,
      "num_input_tokens_seen": 7889232,
      "step": 1312,
      "train_runtime": 1325.12,
      "train_tokens_per_second": 5953.598
    },
    {
      "epoch": 0.4829425287356322,
      "grad_norm": 0.7108722925186157,
      "learning_rate": 0.00015188661881097002,
      "loss": 1.5528,
      "num_input_tokens_seen": 7894256,
      "step": 1313,
      "train_runtime": 1325.9796,
      "train_tokens_per_second": 5953.528
    },
    {
      "epoch": 0.4833103448275862,
      "grad_norm": 0.7087517976760864,
      "learning_rate": 0.00015184980673660963,
      "loss": 1.7848,
      "num_input_tokens_seen": 7898944,
      "step": 1314,
      "train_runtime": 1326.7818,
      "train_tokens_per_second": 5953.461
    },
    {
      "epoch": 0.48367816091954025,
      "grad_norm": 0.6742291450500488,
      "learning_rate": 0.00015181299466224924,
      "loss": 1.7007,
      "num_input_tokens_seen": 7904640,
      "step": 1315,
      "train_runtime": 1327.7155,
      "train_tokens_per_second": 5953.564
    },
    {
      "epoch": 0.48404597701149427,
      "grad_norm": 0.8570288419723511,
      "learning_rate": 0.00015177618258788885,
      "loss": 1.642,
      "num_input_tokens_seen": 7912192,
      "step": 1316,
      "train_runtime": 1328.9099,
      "train_tokens_per_second": 5953.896
    },
    {
      "epoch": 0.4844137931034483,
      "grad_norm": 0.725561797618866,
      "learning_rate": 0.00015173937051352843,
      "loss": 1.7728,
      "num_input_tokens_seen": 7917408,
      "step": 1317,
      "train_runtime": 1329.7929,
      "train_tokens_per_second": 5953.865
    },
    {
      "epoch": 0.4847816091954023,
      "grad_norm": 0.7222111821174622,
      "learning_rate": 0.00015170255843916804,
      "loss": 1.8547,
      "num_input_tokens_seen": 7922656,
      "step": 1318,
      "train_runtime": 1330.6758,
      "train_tokens_per_second": 5953.859
    },
    {
      "epoch": 0.48514942528735633,
      "grad_norm": 0.7159226536750793,
      "learning_rate": 0.00015166574636480765,
      "loss": 1.9004,
      "num_input_tokens_seen": 7927904,
      "step": 1319,
      "train_runtime": 1331.5501,
      "train_tokens_per_second": 5953.891
    },
    {
      "epoch": 0.48551724137931035,
      "grad_norm": 0.8363777995109558,
      "learning_rate": 0.00015162893429044726,
      "loss": 1.867,
      "num_input_tokens_seen": 7933184,
      "step": 1320,
      "train_runtime": 1332.4598,
      "train_tokens_per_second": 5953.789
    },
    {
      "epoch": 0.48588505747126437,
      "grad_norm": 0.7550817131996155,
      "learning_rate": 0.00015159212221608687,
      "loss": 1.8652,
      "num_input_tokens_seen": 7938112,
      "step": 1321,
      "train_runtime": 1333.811,
      "train_tokens_per_second": 5951.452
    },
    {
      "epoch": 0.4862528735632184,
      "grad_norm": 0.7746520042419434,
      "learning_rate": 0.0001515553101417265,
      "loss": 2.1084,
      "num_input_tokens_seen": 7943120,
      "step": 1322,
      "train_runtime": 1334.6604,
      "train_tokens_per_second": 5951.417
    },
    {
      "epoch": 0.4866206896551724,
      "grad_norm": 0.7697386145591736,
      "learning_rate": 0.0001515184980673661,
      "loss": 1.8643,
      "num_input_tokens_seen": 7947568,
      "step": 1323,
      "train_runtime": 1335.4402,
      "train_tokens_per_second": 5951.272
    },
    {
      "epoch": 0.48698850574712643,
      "grad_norm": 0.6934213042259216,
      "learning_rate": 0.00015148168599300572,
      "loss": 1.6074,
      "num_input_tokens_seen": 7954176,
      "step": 1324,
      "train_runtime": 1336.5095,
      "train_tokens_per_second": 5951.455
    },
    {
      "epoch": 0.48735632183908045,
      "grad_norm": 0.7284837365150452,
      "learning_rate": 0.00015144487391864533,
      "loss": 1.7163,
      "num_input_tokens_seen": 7960496,
      "step": 1325,
      "train_runtime": 1337.5432,
      "train_tokens_per_second": 5951.58
    },
    {
      "epoch": 0.4877241379310345,
      "grad_norm": 0.8141064047813416,
      "learning_rate": 0.00015140806184428494,
      "loss": 1.9707,
      "num_input_tokens_seen": 7965568,
      "step": 1326,
      "train_runtime": 1338.4014,
      "train_tokens_per_second": 5951.554
    },
    {
      "epoch": 0.4880919540229885,
      "grad_norm": 0.781174898147583,
      "learning_rate": 0.00015137124976992455,
      "loss": 2.0404,
      "num_input_tokens_seen": 7970704,
      "step": 1327,
      "train_runtime": 1339.2559,
      "train_tokens_per_second": 5951.592
    },
    {
      "epoch": 0.4884597701149425,
      "grad_norm": 0.7122456431388855,
      "learning_rate": 0.00015133443769556416,
      "loss": 1.6171,
      "num_input_tokens_seen": 7976000,
      "step": 1328,
      "train_runtime": 1340.1517,
      "train_tokens_per_second": 5951.565
    },
    {
      "epoch": 0.48882758620689654,
      "grad_norm": 0.7224553227424622,
      "learning_rate": 0.00015129762562120377,
      "loss": 1.9309,
      "num_input_tokens_seen": 7981520,
      "step": 1329,
      "train_runtime": 1341.0704,
      "train_tokens_per_second": 5951.604
    },
    {
      "epoch": 0.48919540229885056,
      "grad_norm": 0.7628961801528931,
      "learning_rate": 0.00015126081354684338,
      "loss": 1.8542,
      "num_input_tokens_seen": 7987776,
      "step": 1330,
      "train_runtime": 1342.0553,
      "train_tokens_per_second": 5951.898
    },
    {
      "epoch": 0.4895632183908046,
      "grad_norm": 0.7004364728927612,
      "learning_rate": 0.00015122400147248298,
      "loss": 1.7703,
      "num_input_tokens_seen": 7994288,
      "step": 1331,
      "train_runtime": 1343.0854,
      "train_tokens_per_second": 5952.181
    },
    {
      "epoch": 0.4899310344827586,
      "grad_norm": 0.80036860704422,
      "learning_rate": 0.0001511871893981226,
      "loss": 1.6565,
      "num_input_tokens_seen": 7999984,
      "step": 1332,
      "train_runtime": 1344.0172,
      "train_tokens_per_second": 5952.293
    },
    {
      "epoch": 0.4902988505747126,
      "grad_norm": 0.7471275329589844,
      "learning_rate": 0.0001511503773237622,
      "loss": 1.844,
      "num_input_tokens_seen": 8005888,
      "step": 1333,
      "train_runtime": 1344.9827,
      "train_tokens_per_second": 5952.41
    },
    {
      "epoch": 0.49066666666666664,
      "grad_norm": 0.6946808099746704,
      "learning_rate": 0.0001511135652494018,
      "loss": 1.9624,
      "num_input_tokens_seen": 8011120,
      "step": 1334,
      "train_runtime": 1345.8456,
      "train_tokens_per_second": 5952.481
    },
    {
      "epoch": 0.4910344827586207,
      "grad_norm": 0.76739102602005,
      "learning_rate": 0.00015107675317504142,
      "loss": 2.1191,
      "num_input_tokens_seen": 8015840,
      "step": 1335,
      "train_runtime": 1346.6438,
      "train_tokens_per_second": 5952.457
    },
    {
      "epoch": 0.49140229885057474,
      "grad_norm": 0.6828115582466125,
      "learning_rate": 0.00015103994110068103,
      "loss": 1.679,
      "num_input_tokens_seen": 8021408,
      "step": 1336,
      "train_runtime": 1347.5491,
      "train_tokens_per_second": 5952.591
    },
    {
      "epoch": 0.49177011494252876,
      "grad_norm": 0.6785314679145813,
      "learning_rate": 0.00015100312902632064,
      "loss": 1.8206,
      "num_input_tokens_seen": 8027680,
      "step": 1337,
      "train_runtime": 1348.5725,
      "train_tokens_per_second": 5952.724
    },
    {
      "epoch": 0.4921379310344828,
      "grad_norm": 0.7338114380836487,
      "learning_rate": 0.00015096631695196025,
      "loss": 1.8427,
      "num_input_tokens_seen": 8031952,
      "step": 1338,
      "train_runtime": 1349.3344,
      "train_tokens_per_second": 5952.529
    },
    {
      "epoch": 0.4925057471264368,
      "grad_norm": 0.7458033561706543,
      "learning_rate": 0.00015092950487759988,
      "loss": 1.815,
      "num_input_tokens_seen": 8038240,
      "step": 1339,
      "train_runtime": 1350.3655,
      "train_tokens_per_second": 5952.64
    },
    {
      "epoch": 0.4928735632183908,
      "grad_norm": 0.8153566122055054,
      "learning_rate": 0.00015089269280323946,
      "loss": 1.8637,
      "num_input_tokens_seen": 8043472,
      "step": 1340,
      "train_runtime": 1351.254,
      "train_tokens_per_second": 5952.598
    },
    {
      "epoch": 0.49324137931034484,
      "grad_norm": 0.850492537021637,
      "learning_rate": 0.00015085588072887907,
      "loss": 1.6821,
      "num_input_tokens_seen": 8052160,
      "step": 1341,
      "train_runtime": 1352.6096,
      "train_tokens_per_second": 5953.055
    },
    {
      "epoch": 0.49360919540229886,
      "grad_norm": 0.7156057953834534,
      "learning_rate": 0.00015081906865451868,
      "loss": 1.8425,
      "num_input_tokens_seen": 8058032,
      "step": 1342,
      "train_runtime": 1353.5981,
      "train_tokens_per_second": 5953.046
    },
    {
      "epoch": 0.4939770114942529,
      "grad_norm": 0.7298743724822998,
      "learning_rate": 0.0001507822565801583,
      "loss": 1.9257,
      "num_input_tokens_seen": 8063792,
      "step": 1343,
      "train_runtime": 1354.5703,
      "train_tokens_per_second": 5953.026
    },
    {
      "epoch": 0.4943448275862069,
      "grad_norm": 0.7520971298217773,
      "learning_rate": 0.0001507454445057979,
      "loss": 1.7358,
      "num_input_tokens_seen": 8072512,
      "step": 1344,
      "train_runtime": 1355.8972,
      "train_tokens_per_second": 5953.631
    },
    {
      "epoch": 0.4947126436781609,
      "grad_norm": 0.7416859269142151,
      "learning_rate": 0.0001507086324314375,
      "loss": 1.7749,
      "num_input_tokens_seen": 8078416,
      "step": 1345,
      "train_runtime": 1356.8677,
      "train_tokens_per_second": 5953.724
    },
    {
      "epoch": 0.49508045977011494,
      "grad_norm": 0.7004967927932739,
      "learning_rate": 0.00015067182035707712,
      "loss": 1.7735,
      "num_input_tokens_seen": 8084160,
      "step": 1346,
      "train_runtime": 1357.8386,
      "train_tokens_per_second": 5953.697
    },
    {
      "epoch": 0.49544827586206897,
      "grad_norm": 0.7175948023796082,
      "learning_rate": 0.00015063500828271673,
      "loss": 1.8509,
      "num_input_tokens_seen": 8089664,
      "step": 1347,
      "train_runtime": 1358.7614,
      "train_tokens_per_second": 5953.705
    },
    {
      "epoch": 0.495816091954023,
      "grad_norm": 0.7371270060539246,
      "learning_rate": 0.00015059819620835636,
      "loss": 1.8303,
      "num_input_tokens_seen": 8094336,
      "step": 1348,
      "train_runtime": 1359.5875,
      "train_tokens_per_second": 5953.523
    },
    {
      "epoch": 0.496183908045977,
      "grad_norm": 0.7381982207298279,
      "learning_rate": 0.00015056138413399597,
      "loss": 1.9887,
      "num_input_tokens_seen": 8099520,
      "step": 1349,
      "train_runtime": 1360.4582,
      "train_tokens_per_second": 5953.523
    },
    {
      "epoch": 0.496551724137931,
      "grad_norm": 0.7753965258598328,
      "learning_rate": 0.00015052457205963558,
      "loss": 1.823,
      "num_input_tokens_seen": 8103952,
      "step": 1350,
      "train_runtime": 1361.255,
      "train_tokens_per_second": 5953.295
    },
    {
      "epoch": 0.49691954022988505,
      "grad_norm": 0.689374566078186,
      "learning_rate": 0.0001504877599852752,
      "loss": 1.7141,
      "num_input_tokens_seen": 8108976,
      "step": 1351,
      "train_runtime": 1362.6403,
      "train_tokens_per_second": 5950.929
    },
    {
      "epoch": 0.49728735632183907,
      "grad_norm": 0.7397502064704895,
      "learning_rate": 0.0001504509479109148,
      "loss": 1.7321,
      "num_input_tokens_seen": 8114048,
      "step": 1352,
      "train_runtime": 1363.4943,
      "train_tokens_per_second": 5950.922
    },
    {
      "epoch": 0.4976551724137931,
      "grad_norm": 0.8675876259803772,
      "learning_rate": 0.00015041413583655438,
      "loss": 1.8876,
      "num_input_tokens_seen": 8118896,
      "step": 1353,
      "train_runtime": 1364.3335,
      "train_tokens_per_second": 5950.815
    },
    {
      "epoch": 0.4980229885057471,
      "grad_norm": 0.6720005869865417,
      "learning_rate": 0.000150377323762194,
      "loss": 1.6809,
      "num_input_tokens_seen": 8126496,
      "step": 1354,
      "train_runtime": 1365.5222,
      "train_tokens_per_second": 5951.2
    },
    {
      "epoch": 0.49839080459770113,
      "grad_norm": 0.743710458278656,
      "learning_rate": 0.0001503405116878336,
      "loss": 1.7912,
      "num_input_tokens_seen": 8132800,
      "step": 1355,
      "train_runtime": 1366.5561,
      "train_tokens_per_second": 5951.311
    },
    {
      "epoch": 0.49875862068965515,
      "grad_norm": 0.5289599895477295,
      "learning_rate": 0.00015030369961347323,
      "loss": 1.3647,
      "num_input_tokens_seen": 8147296,
      "step": 1356,
      "train_runtime": 1368.6841,
      "train_tokens_per_second": 5952.649
    },
    {
      "epoch": 0.4991264367816092,
      "grad_norm": 0.7412967085838318,
      "learning_rate": 0.00015026688753911284,
      "loss": 1.6776,
      "num_input_tokens_seen": 8151552,
      "step": 1357,
      "train_runtime": 1369.4391,
      "train_tokens_per_second": 5952.475
    },
    {
      "epoch": 0.4994942528735632,
      "grad_norm": 0.7483311891555786,
      "learning_rate": 0.00015023007546475245,
      "loss": 1.6162,
      "num_input_tokens_seen": 8158032,
      "step": 1358,
      "train_runtime": 1370.4953,
      "train_tokens_per_second": 5952.616
    },
    {
      "epoch": 0.4998620689655172,
      "grad_norm": 0.7232673168182373,
      "learning_rate": 0.00015019326339039206,
      "loss": 1.6481,
      "num_input_tokens_seen": 8162416,
      "step": 1359,
      "train_runtime": 1371.2699,
      "train_tokens_per_second": 5952.45
    },
    {
      "epoch": 0.5002298850574712,
      "grad_norm": 0.6927854418754578,
      "learning_rate": 0.00015015645131603167,
      "loss": 1.6787,
      "num_input_tokens_seen": 8168432,
      "step": 1360,
      "train_runtime": 1372.2369,
      "train_tokens_per_second": 5952.64
    },
    {
      "epoch": 0.5005977011494253,
      "grad_norm": 0.7753093838691711,
      "learning_rate": 0.00015011963924167128,
      "loss": 1.8056,
      "num_input_tokens_seen": 8173232,
      "step": 1361,
      "train_runtime": 1373.0682,
      "train_tokens_per_second": 5952.532
    },
    {
      "epoch": 0.5009655172413793,
      "grad_norm": 0.8077579736709595,
      "learning_rate": 0.0001500828271673109,
      "loss": 1.974,
      "num_input_tokens_seen": 8179136,
      "step": 1362,
      "train_runtime": 1374.0547,
      "train_tokens_per_second": 5952.555
    },
    {
      "epoch": 0.5013333333333333,
      "grad_norm": 0.8416463136672974,
      "learning_rate": 0.0001500460150929505,
      "loss": 1.8059,
      "num_input_tokens_seen": 8183824,
      "step": 1363,
      "train_runtime": 1374.8755,
      "train_tokens_per_second": 5952.411
    },
    {
      "epoch": 0.5017011494252873,
      "grad_norm": 0.6934306621551514,
      "learning_rate": 0.0001500092030185901,
      "loss": 1.6866,
      "num_input_tokens_seen": 8190320,
      "step": 1364,
      "train_runtime": 1375.9421,
      "train_tokens_per_second": 5952.518
    },
    {
      "epoch": 0.5020689655172413,
      "grad_norm": 0.6957128643989563,
      "learning_rate": 0.00014997239094422971,
      "loss": 1.8055,
      "num_input_tokens_seen": 8195856,
      "step": 1365,
      "train_runtime": 1376.8374,
      "train_tokens_per_second": 5952.668
    },
    {
      "epoch": 0.5024367816091954,
      "grad_norm": 0.681606113910675,
      "learning_rate": 0.00014993557886986932,
      "loss": 1.8427,
      "num_input_tokens_seen": 8201664,
      "step": 1366,
      "train_runtime": 1377.8184,
      "train_tokens_per_second": 5952.645
    },
    {
      "epoch": 0.5028045977011494,
      "grad_norm": 0.7329640984535217,
      "learning_rate": 0.00014989876679550893,
      "loss": 1.6902,
      "num_input_tokens_seen": 8206368,
      "step": 1367,
      "train_runtime": 1378.6255,
      "train_tokens_per_second": 5952.572
    },
    {
      "epoch": 0.5031724137931034,
      "grad_norm": 0.767595648765564,
      "learning_rate": 0.00014986195472114854,
      "loss": 1.7588,
      "num_input_tokens_seen": 8210448,
      "step": 1368,
      "train_runtime": 1379.354,
      "train_tokens_per_second": 5952.387
    },
    {
      "epoch": 0.5035402298850574,
      "grad_norm": 0.7514697909355164,
      "learning_rate": 0.00014982514264678815,
      "loss": 1.9968,
      "num_input_tokens_seen": 8216192,
      "step": 1369,
      "train_runtime": 1380.2934,
      "train_tokens_per_second": 5952.497
    },
    {
      "epoch": 0.5039080459770114,
      "grad_norm": 0.668262779712677,
      "learning_rate": 0.00014978833057242776,
      "loss": 1.6642,
      "num_input_tokens_seen": 8223328,
      "step": 1370,
      "train_runtime": 1381.4292,
      "train_tokens_per_second": 5952.768
    },
    {
      "epoch": 0.5042758620689655,
      "grad_norm": 0.7377183437347412,
      "learning_rate": 0.00014975151849806737,
      "loss": 2.0848,
      "num_input_tokens_seen": 8228240,
      "step": 1371,
      "train_runtime": 1382.2631,
      "train_tokens_per_second": 5952.731
    },
    {
      "epoch": 0.5046436781609196,
      "grad_norm": 0.7200998067855835,
      "learning_rate": 0.00014971470642370698,
      "loss": 1.6967,
      "num_input_tokens_seen": 8235536,
      "step": 1372,
      "train_runtime": 1383.4245,
      "train_tokens_per_second": 5953.007
    },
    {
      "epoch": 0.5050114942528736,
      "grad_norm": 0.8013726472854614,
      "learning_rate": 0.0001496778943493466,
      "loss": 1.8624,
      "num_input_tokens_seen": 8240064,
      "step": 1373,
      "train_runtime": 1384.2108,
      "train_tokens_per_second": 5952.897
    },
    {
      "epoch": 0.5053793103448276,
      "grad_norm": 0.791252851486206,
      "learning_rate": 0.00014964108227498622,
      "loss": 1.9451,
      "num_input_tokens_seen": 8244928,
      "step": 1374,
      "train_runtime": 1385.0554,
      "train_tokens_per_second": 5952.779
    },
    {
      "epoch": 0.5057471264367817,
      "grad_norm": 0.7399970293045044,
      "learning_rate": 0.00014960427020062583,
      "loss": 1.5378,
      "num_input_tokens_seen": 8250576,
      "step": 1375,
      "train_runtime": 1385.9858,
      "train_tokens_per_second": 5952.857
    },
    {
      "epoch": 0.5061149425287357,
      "grad_norm": 0.7669424414634705,
      "learning_rate": 0.0001495674581262654,
      "loss": 1.5796,
      "num_input_tokens_seen": 8258176,
      "step": 1376,
      "train_runtime": 1387.1818,
      "train_tokens_per_second": 5953.204
    },
    {
      "epoch": 0.5064827586206897,
      "grad_norm": 0.7926695346832275,
      "learning_rate": 0.00014953064605190502,
      "loss": 1.8736,
      "num_input_tokens_seen": 8264544,
      "step": 1377,
      "train_runtime": 1388.1821,
      "train_tokens_per_second": 5953.501
    },
    {
      "epoch": 0.5068505747126437,
      "grad_norm": 0.7332091927528381,
      "learning_rate": 0.00014949383397754463,
      "loss": 1.6914,
      "num_input_tokens_seen": 8272928,
      "step": 1378,
      "train_runtime": 1389.4938,
      "train_tokens_per_second": 5953.915
    },
    {
      "epoch": 0.5072183908045977,
      "grad_norm": 0.7481079697608948,
      "learning_rate": 0.00014945702190318424,
      "loss": 1.9171,
      "num_input_tokens_seen": 8279392,
      "step": 1379,
      "train_runtime": 1390.5524,
      "train_tokens_per_second": 5954.031
    },
    {
      "epoch": 0.5075862068965518,
      "grad_norm": 0.857302725315094,
      "learning_rate": 0.00014942020982882385,
      "loss": 1.8295,
      "num_input_tokens_seen": 8285536,
      "step": 1380,
      "train_runtime": 1391.5582,
      "train_tokens_per_second": 5954.143
    },
    {
      "epoch": 0.5079540229885058,
      "grad_norm": 0.7441621422767639,
      "learning_rate": 0.00014938339775446346,
      "loss": 1.7978,
      "num_input_tokens_seen": 8290704,
      "step": 1381,
      "train_runtime": 1392.9524,
      "train_tokens_per_second": 5951.893
    },
    {
      "epoch": 0.5083218390804598,
      "grad_norm": 0.6255395412445068,
      "learning_rate": 0.0001493465856801031,
      "loss": 1.6786,
      "num_input_tokens_seen": 8297440,
      "step": 1382,
      "train_runtime": 1394.0449,
      "train_tokens_per_second": 5952.061
    },
    {
      "epoch": 0.5086896551724138,
      "grad_norm": 0.7366942167282104,
      "learning_rate": 0.0001493097736057427,
      "loss": 1.5836,
      "num_input_tokens_seen": 8305760,
      "step": 1383,
      "train_runtime": 1395.3208,
      "train_tokens_per_second": 5952.581
    },
    {
      "epoch": 0.5090574712643678,
      "grad_norm": 0.7217180728912354,
      "learning_rate": 0.0001492729615313823,
      "loss": 1.9036,
      "num_input_tokens_seen": 8311712,
      "step": 1384,
      "train_runtime": 1396.2982,
      "train_tokens_per_second": 5952.677
    },
    {
      "epoch": 0.5094252873563219,
      "grad_norm": 0.8432775735855103,
      "learning_rate": 0.00014923614945702192,
      "loss": 1.8468,
      "num_input_tokens_seen": 8316960,
      "step": 1385,
      "train_runtime": 1397.1639,
      "train_tokens_per_second": 5952.745
    },
    {
      "epoch": 0.5097931034482759,
      "grad_norm": 0.6771023273468018,
      "learning_rate": 0.00014919933738266153,
      "loss": 1.8409,
      "num_input_tokens_seen": 8324048,
      "step": 1386,
      "train_runtime": 1398.2843,
      "train_tokens_per_second": 5953.044
    },
    {
      "epoch": 0.5101609195402299,
      "grad_norm": 0.6785527467727661,
      "learning_rate": 0.00014916252530830114,
      "loss": 1.6691,
      "num_input_tokens_seen": 8330400,
      "step": 1387,
      "train_runtime": 1399.3143,
      "train_tokens_per_second": 5953.202
    },
    {
      "epoch": 0.5105287356321839,
      "grad_norm": 0.725290834903717,
      "learning_rate": 0.00014912571323394075,
      "loss": 1.7793,
      "num_input_tokens_seen": 8338928,
      "step": 1388,
      "train_runtime": 1400.6484,
      "train_tokens_per_second": 5953.62
    },
    {
      "epoch": 0.510896551724138,
      "grad_norm": 0.7309304475784302,
      "learning_rate": 0.00014908890115958035,
      "loss": 1.6867,
      "num_input_tokens_seen": 8344576,
      "step": 1389,
      "train_runtime": 1401.6118,
      "train_tokens_per_second": 5953.557
    },
    {
      "epoch": 0.511264367816092,
      "grad_norm": 0.7281897664070129,
      "learning_rate": 0.00014905208908521996,
      "loss": 1.6497,
      "num_input_tokens_seen": 8351248,
      "step": 1390,
      "train_runtime": 1402.6846,
      "train_tokens_per_second": 5953.76
    },
    {
      "epoch": 0.511632183908046,
      "grad_norm": 0.7505301237106323,
      "learning_rate": 0.00014901527701085957,
      "loss": 1.7723,
      "num_input_tokens_seen": 8356624,
      "step": 1391,
      "train_runtime": 1403.562,
      "train_tokens_per_second": 5953.869
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.7180195450782776,
      "learning_rate": 0.00014897846493649918,
      "loss": 1.6918,
      "num_input_tokens_seen": 8363600,
      "step": 1392,
      "train_runtime": 1404.6948,
      "train_tokens_per_second": 5954.034
    },
    {
      "epoch": 0.512367816091954,
      "grad_norm": 0.8358976244926453,
      "learning_rate": 0.0001489416528621388,
      "loss": 1.9481,
      "num_input_tokens_seen": 8368768,
      "step": 1393,
      "train_runtime": 1405.5573,
      "train_tokens_per_second": 5954.057
    },
    {
      "epoch": 0.512735632183908,
      "grad_norm": 0.6864331364631653,
      "learning_rate": 0.0001489048407877784,
      "loss": 1.8447,
      "num_input_tokens_seen": 8377136,
      "step": 1394,
      "train_runtime": 1406.8409,
      "train_tokens_per_second": 5954.573
    },
    {
      "epoch": 0.5131034482758621,
      "grad_norm": 0.680988609790802,
      "learning_rate": 0.000148868028713418,
      "loss": 1.7178,
      "num_input_tokens_seen": 8382224,
      "step": 1395,
      "train_runtime": 1407.711,
      "train_tokens_per_second": 5954.506
    },
    {
      "epoch": 0.5134712643678161,
      "grad_norm": 0.6796252131462097,
      "learning_rate": 0.00014883121663905762,
      "loss": 1.6514,
      "num_input_tokens_seen": 8386800,
      "step": 1396,
      "train_runtime": 1408.5091,
      "train_tokens_per_second": 5954.381
    },
    {
      "epoch": 0.5138390804597701,
      "grad_norm": 0.7472183704376221,
      "learning_rate": 0.00014879440456469723,
      "loss": 1.6846,
      "num_input_tokens_seen": 8391968,
      "step": 1397,
      "train_runtime": 1409.3813,
      "train_tokens_per_second": 5954.363
    },
    {
      "epoch": 0.5142068965517241,
      "grad_norm": 0.7881629467010498,
      "learning_rate": 0.00014875759249033683,
      "loss": 1.8198,
      "num_input_tokens_seen": 8398384,
      "step": 1398,
      "train_runtime": 1410.4282,
      "train_tokens_per_second": 5954.493
    },
    {
      "epoch": 0.5145747126436782,
      "grad_norm": 0.7668195962905884,
      "learning_rate": 0.00014872078041597647,
      "loss": 1.8793,
      "num_input_tokens_seen": 8403120,
      "step": 1399,
      "train_runtime": 1411.2595,
      "train_tokens_per_second": 5954.341
    },
    {
      "epoch": 0.5149425287356322,
      "grad_norm": 0.8129005432128906,
      "learning_rate": 0.00014868396834161605,
      "loss": 1.8726,
      "num_input_tokens_seen": 8408352,
      "step": 1400,
      "train_runtime": 1412.1596,
      "train_tokens_per_second": 5954.25
    },
    {
      "epoch": 0.5153103448275862,
      "grad_norm": 0.7139448523521423,
      "learning_rate": 0.00014864715626725566,
      "loss": 1.6119,
      "num_input_tokens_seen": 8413056,
      "step": 1401,
      "train_runtime": 1412.9822,
      "train_tokens_per_second": 5954.113
    },
    {
      "epoch": 0.5156781609195402,
      "grad_norm": 0.5423973798751831,
      "learning_rate": 0.00014861034419289527,
      "loss": 1.5718,
      "num_input_tokens_seen": 8424096,
      "step": 1402,
      "train_runtime": 1414.6656,
      "train_tokens_per_second": 5954.832
    },
    {
      "epoch": 0.5160459770114942,
      "grad_norm": 0.6507823467254639,
      "learning_rate": 0.00014857353211853488,
      "loss": 1.7399,
      "num_input_tokens_seen": 8433392,
      "step": 1403,
      "train_runtime": 1416.0559,
      "train_tokens_per_second": 5955.55
    },
    {
      "epoch": 0.5164137931034483,
      "grad_norm": 0.6087756752967834,
      "learning_rate": 0.0001485367200441745,
      "loss": 1.5716,
      "num_input_tokens_seen": 8440848,
      "step": 1404,
      "train_runtime": 1417.2555,
      "train_tokens_per_second": 5955.77
    },
    {
      "epoch": 0.5167816091954023,
      "grad_norm": 0.7732454538345337,
      "learning_rate": 0.0001484999079698141,
      "loss": 1.793,
      "num_input_tokens_seen": 8446464,
      "step": 1405,
      "train_runtime": 1418.1905,
      "train_tokens_per_second": 5955.803
    },
    {
      "epoch": 0.5171494252873563,
      "grad_norm": 0.718552827835083,
      "learning_rate": 0.0001484630958954537,
      "loss": 1.7763,
      "num_input_tokens_seen": 8451248,
      "step": 1406,
      "train_runtime": 1419.0042,
      "train_tokens_per_second": 5955.76
    },
    {
      "epoch": 0.5175172413793103,
      "grad_norm": 0.7031474113464355,
      "learning_rate": 0.00014842628382109331,
      "loss": 1.6331,
      "num_input_tokens_seen": 8457248,
      "step": 1407,
      "train_runtime": 1419.9806,
      "train_tokens_per_second": 5955.89
    },
    {
      "epoch": 0.5178850574712643,
      "grad_norm": 0.701958954334259,
      "learning_rate": 0.00014838947174673295,
      "loss": 1.8663,
      "num_input_tokens_seen": 8463760,
      "step": 1408,
      "train_runtime": 1421.0375,
      "train_tokens_per_second": 5956.043
    },
    {
      "epoch": 0.5182528735632184,
      "grad_norm": 0.6807476878166199,
      "learning_rate": 0.00014835265967237256,
      "loss": 1.7774,
      "num_input_tokens_seen": 8472128,
      "step": 1409,
      "train_runtime": 1422.3573,
      "train_tokens_per_second": 5956.399
    },
    {
      "epoch": 0.5186206896551724,
      "grad_norm": 0.7757298946380615,
      "learning_rate": 0.00014831584759801217,
      "loss": 1.9072,
      "num_input_tokens_seen": 8477072,
      "step": 1410,
      "train_runtime": 1423.2046,
      "train_tokens_per_second": 5956.327
    },
    {
      "epoch": 0.5189885057471264,
      "grad_norm": 0.6979550123214722,
      "learning_rate": 0.00014827903552365178,
      "loss": 1.6075,
      "num_input_tokens_seen": 8482992,
      "step": 1411,
      "train_runtime": 1424.7002,
      "train_tokens_per_second": 5954.23
    },
    {
      "epoch": 0.5193563218390804,
      "grad_norm": 0.6153293251991272,
      "learning_rate": 0.00014824222344929139,
      "loss": 1.5254,
      "num_input_tokens_seen": 8490352,
      "step": 1412,
      "train_runtime": 1425.8888,
      "train_tokens_per_second": 5954.428
    },
    {
      "epoch": 0.5197241379310344,
      "grad_norm": 0.7379152178764343,
      "learning_rate": 0.00014820541137493097,
      "loss": 1.7054,
      "num_input_tokens_seen": 8496352,
      "step": 1413,
      "train_runtime": 1426.8805,
      "train_tokens_per_second": 5954.495
    },
    {
      "epoch": 0.5200919540229885,
      "grad_norm": 0.6838432550430298,
      "learning_rate": 0.00014816859930057058,
      "loss": 1.6971,
      "num_input_tokens_seen": 8503712,
      "step": 1414,
      "train_runtime": 1428.0408,
      "train_tokens_per_second": 5954.81
    },
    {
      "epoch": 0.5204597701149425,
      "grad_norm": 0.7826948761940002,
      "learning_rate": 0.00014813178722621019,
      "loss": 1.9042,
      "num_input_tokens_seen": 8507872,
      "step": 1415,
      "train_runtime": 1428.7718,
      "train_tokens_per_second": 5954.675
    },
    {
      "epoch": 0.5208275862068965,
      "grad_norm": 0.8368574976921082,
      "learning_rate": 0.00014809497515184982,
      "loss": 1.7309,
      "num_input_tokens_seen": 8512864,
      "step": 1416,
      "train_runtime": 1429.6138,
      "train_tokens_per_second": 5954.66
    },
    {
      "epoch": 0.5211954022988505,
      "grad_norm": 0.7584893107414246,
      "learning_rate": 0.00014805816307748943,
      "loss": 1.666,
      "num_input_tokens_seen": 8521168,
      "step": 1417,
      "train_runtime": 1430.9217,
      "train_tokens_per_second": 5955.021
    },
    {
      "epoch": 0.5215632183908046,
      "grad_norm": 0.7287805676460266,
      "learning_rate": 0.00014802135100312904,
      "loss": 1.6451,
      "num_input_tokens_seen": 8528288,
      "step": 1418,
      "train_runtime": 1432.0732,
      "train_tokens_per_second": 5955.204
    },
    {
      "epoch": 0.5219310344827586,
      "grad_norm": 0.7619727849960327,
      "learning_rate": 0.00014798453892876865,
      "loss": 1.6289,
      "num_input_tokens_seen": 8533984,
      "step": 1419,
      "train_runtime": 1433.0352,
      "train_tokens_per_second": 5955.181
    },
    {
      "epoch": 0.5222988505747126,
      "grad_norm": 0.6437995433807373,
      "learning_rate": 0.00014794772685440826,
      "loss": 1.8128,
      "num_input_tokens_seen": 8542608,
      "step": 1420,
      "train_runtime": 1434.3722,
      "train_tokens_per_second": 5955.642
    },
    {
      "epoch": 0.5226666666666666,
      "grad_norm": 0.7566819787025452,
      "learning_rate": 0.00014791091478004787,
      "loss": 1.841,
      "num_input_tokens_seen": 8547696,
      "step": 1421,
      "train_runtime": 1435.2206,
      "train_tokens_per_second": 5955.667
    },
    {
      "epoch": 0.5230344827586207,
      "grad_norm": 0.8183180689811707,
      "learning_rate": 0.00014787410270568747,
      "loss": 1.7804,
      "num_input_tokens_seen": 8552624,
      "step": 1422,
      "train_runtime": 1436.0804,
      "train_tokens_per_second": 5955.533
    },
    {
      "epoch": 0.5234022988505748,
      "grad_norm": 0.7553067207336426,
      "learning_rate": 0.00014783729063132708,
      "loss": 1.8143,
      "num_input_tokens_seen": 8556944,
      "step": 1423,
      "train_runtime": 1436.8532,
      "train_tokens_per_second": 5955.336
    },
    {
      "epoch": 0.5237701149425288,
      "grad_norm": 0.7072001695632935,
      "learning_rate": 0.0001478004785569667,
      "loss": 1.7463,
      "num_input_tokens_seen": 8562816,
      "step": 1424,
      "train_runtime": 1437.8544,
      "train_tokens_per_second": 5955.273
    },
    {
      "epoch": 0.5241379310344828,
      "grad_norm": 0.8855698704719543,
      "learning_rate": 0.0001477636664826063,
      "loss": 1.9317,
      "num_input_tokens_seen": 8568272,
      "step": 1425,
      "train_runtime": 1438.7734,
      "train_tokens_per_second": 5955.261
    },
    {
      "epoch": 0.5245057471264368,
      "grad_norm": 0.7797963619232178,
      "learning_rate": 0.0001477268544082459,
      "loss": 1.7609,
      "num_input_tokens_seen": 8575680,
      "step": 1426,
      "train_runtime": 1439.9866,
      "train_tokens_per_second": 5955.389
    },
    {
      "epoch": 0.5248735632183908,
      "grad_norm": 0.7859804034233093,
      "learning_rate": 0.00014769004233388552,
      "loss": 1.7861,
      "num_input_tokens_seen": 8580288,
      "step": 1427,
      "train_runtime": 1440.7872,
      "train_tokens_per_second": 5955.278
    },
    {
      "epoch": 0.5252413793103449,
      "grad_norm": 0.8036105036735535,
      "learning_rate": 0.00014765323025952513,
      "loss": 1.7659,
      "num_input_tokens_seen": 8584864,
      "step": 1428,
      "train_runtime": 1441.5938,
      "train_tokens_per_second": 5955.12
    },
    {
      "epoch": 0.5256091954022989,
      "grad_norm": 0.7197122573852539,
      "learning_rate": 0.00014761641818516474,
      "loss": 1.6921,
      "num_input_tokens_seen": 8589312,
      "step": 1429,
      "train_runtime": 1442.3718,
      "train_tokens_per_second": 5954.992
    },
    {
      "epoch": 0.5259770114942529,
      "grad_norm": 0.711627721786499,
      "learning_rate": 0.00014757960611080435,
      "loss": 1.7333,
      "num_input_tokens_seen": 8595312,
      "step": 1430,
      "train_runtime": 1443.3454,
      "train_tokens_per_second": 5955.132
    },
    {
      "epoch": 0.5263448275862069,
      "grad_norm": 0.7196218371391296,
      "learning_rate": 0.00014754279403644395,
      "loss": 1.7495,
      "num_input_tokens_seen": 8599984,
      "step": 1431,
      "train_runtime": 1444.1471,
      "train_tokens_per_second": 5955.061
    },
    {
      "epoch": 0.526712643678161,
      "grad_norm": 0.8260558843612671,
      "learning_rate": 0.00014750598196208356,
      "loss": 1.8117,
      "num_input_tokens_seen": 8605232,
      "step": 1432,
      "train_runtime": 1445.0251,
      "train_tokens_per_second": 5955.074
    },
    {
      "epoch": 0.527080459770115,
      "grad_norm": 0.7250815629959106,
      "learning_rate": 0.0001474691698877232,
      "loss": 1.753,
      "num_input_tokens_seen": 8612912,
      "step": 1433,
      "train_runtime": 1446.2358,
      "train_tokens_per_second": 5955.4
    },
    {
      "epoch": 0.527448275862069,
      "grad_norm": 0.7463332414627075,
      "learning_rate": 0.0001474323578133628,
      "loss": 1.8104,
      "num_input_tokens_seen": 8618336,
      "step": 1434,
      "train_runtime": 1447.1439,
      "train_tokens_per_second": 5955.411
    },
    {
      "epoch": 0.527816091954023,
      "grad_norm": 0.7387081980705261,
      "learning_rate": 0.00014739554573900242,
      "loss": 1.6921,
      "num_input_tokens_seen": 8624736,
      "step": 1435,
      "train_runtime": 1448.1832,
      "train_tokens_per_second": 5955.556
    },
    {
      "epoch": 0.528183908045977,
      "grad_norm": 0.7554429769515991,
      "learning_rate": 0.000147358733664642,
      "loss": 1.7071,
      "num_input_tokens_seen": 8631920,
      "step": 1436,
      "train_runtime": 1449.3473,
      "train_tokens_per_second": 5955.729
    },
    {
      "epoch": 0.5285517241379311,
      "grad_norm": 0.8548153638839722,
      "learning_rate": 0.0001473219215902816,
      "loss": 2.0403,
      "num_input_tokens_seen": 8637696,
      "step": 1437,
      "train_runtime": 1450.2901,
      "train_tokens_per_second": 5955.84
    },
    {
      "epoch": 0.5289195402298851,
      "grad_norm": 0.7452429533004761,
      "learning_rate": 0.00014728510951592122,
      "loss": 1.7235,
      "num_input_tokens_seen": 8642608,
      "step": 1438,
      "train_runtime": 1451.1141,
      "train_tokens_per_second": 5955.843
    },
    {
      "epoch": 0.5292873563218391,
      "grad_norm": 0.7534968256950378,
      "learning_rate": 0.00014724829744156083,
      "loss": 1.9277,
      "num_input_tokens_seen": 8648720,
      "step": 1439,
      "train_runtime": 1452.142,
      "train_tokens_per_second": 5955.836
    },
    {
      "epoch": 0.5296551724137931,
      "grad_norm": 0.7616642713546753,
      "learning_rate": 0.00014721148536720043,
      "loss": 1.7343,
      "num_input_tokens_seen": 8653360,
      "step": 1440,
      "train_runtime": 1452.95,
      "train_tokens_per_second": 5955.718
    },
    {
      "epoch": 0.5300229885057471,
      "grad_norm": 0.6579344868659973,
      "learning_rate": 0.00014717467329284004,
      "loss": 1.6678,
      "num_input_tokens_seen": 8661280,
      "step": 1441,
      "train_runtime": 1454.7228,
      "train_tokens_per_second": 5953.904
    },
    {
      "epoch": 0.5303908045977012,
      "grad_norm": 0.7073023319244385,
      "learning_rate": 0.00014713786121847968,
      "loss": 2.0388,
      "num_input_tokens_seen": 8667312,
      "step": 1442,
      "train_runtime": 1455.7077,
      "train_tokens_per_second": 5954.019
    },
    {
      "epoch": 0.5307586206896552,
      "grad_norm": 0.695111870765686,
      "learning_rate": 0.0001471010491441193,
      "loss": 2.0914,
      "num_input_tokens_seen": 8676160,
      "step": 1443,
      "train_runtime": 1457.0471,
      "train_tokens_per_second": 5954.619
    },
    {
      "epoch": 0.5311264367816092,
      "grad_norm": 0.722130298614502,
      "learning_rate": 0.0001470642370697589,
      "loss": 1.5904,
      "num_input_tokens_seen": 8680816,
      "step": 1444,
      "train_runtime": 1457.8518,
      "train_tokens_per_second": 5954.526
    },
    {
      "epoch": 0.5314942528735632,
      "grad_norm": 0.7415043711662292,
      "learning_rate": 0.0001470274249953985,
      "loss": 1.9128,
      "num_input_tokens_seen": 8686400,
      "step": 1445,
      "train_runtime": 1458.7828,
      "train_tokens_per_second": 5954.553
    },
    {
      "epoch": 0.5318620689655172,
      "grad_norm": 0.8077269196510315,
      "learning_rate": 0.00014699061292103812,
      "loss": 1.7098,
      "num_input_tokens_seen": 8691696,
      "step": 1446,
      "train_runtime": 1459.6599,
      "train_tokens_per_second": 5954.603
    },
    {
      "epoch": 0.5322298850574713,
      "grad_norm": 0.6825448274612427,
      "learning_rate": 0.00014695380084667772,
      "loss": 1.83,
      "num_input_tokens_seen": 8698976,
      "step": 1447,
      "train_runtime": 1460.8417,
      "train_tokens_per_second": 5954.77
    },
    {
      "epoch": 0.5325977011494253,
      "grad_norm": 0.8043240308761597,
      "learning_rate": 0.00014691698877231733,
      "loss": 1.6199,
      "num_input_tokens_seen": 8706016,
      "step": 1448,
      "train_runtime": 1461.9492,
      "train_tokens_per_second": 5955.074
    },
    {
      "epoch": 0.5329655172413793,
      "grad_norm": 0.6993316411972046,
      "learning_rate": 0.00014688017669795692,
      "loss": 1.6032,
      "num_input_tokens_seen": 8711872,
      "step": 1449,
      "train_runtime": 1462.9355,
      "train_tokens_per_second": 5955.062
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.6930384635925293,
      "learning_rate": 0.00014684336462359655,
      "loss": 1.7395,
      "num_input_tokens_seen": 8718848,
      "step": 1450,
      "train_runtime": 1464.0365,
      "train_tokens_per_second": 5955.349
    },
    {
      "epoch": 0.5337011494252873,
      "grad_norm": 0.7121605277061462,
      "learning_rate": 0.00014680655254923616,
      "loss": 1.7851,
      "num_input_tokens_seen": 8725440,
      "step": 1451,
      "train_runtime": 1465.0913,
      "train_tokens_per_second": 5955.56
    },
    {
      "epoch": 0.5340689655172414,
      "grad_norm": 0.7211445569992065,
      "learning_rate": 0.00014676974047487577,
      "loss": 1.8139,
      "num_input_tokens_seen": 8730224,
      "step": 1452,
      "train_runtime": 1465.9078,
      "train_tokens_per_second": 5955.507
    },
    {
      "epoch": 0.5344367816091954,
      "grad_norm": 0.7330206632614136,
      "learning_rate": 0.00014673292840051538,
      "loss": 1.7346,
      "num_input_tokens_seen": 8735536,
      "step": 1453,
      "train_runtime": 1466.7838,
      "train_tokens_per_second": 5955.572
    },
    {
      "epoch": 0.5348045977011494,
      "grad_norm": 0.7850086092948914,
      "learning_rate": 0.000146696116326155,
      "loss": 1.7919,
      "num_input_tokens_seen": 8740368,
      "step": 1454,
      "train_runtime": 1467.6293,
      "train_tokens_per_second": 5955.433
    },
    {
      "epoch": 0.5351724137931034,
      "grad_norm": 0.7872830629348755,
      "learning_rate": 0.0001466593042517946,
      "loss": 2.0429,
      "num_input_tokens_seen": 8745520,
      "step": 1455,
      "train_runtime": 1468.5106,
      "train_tokens_per_second": 5955.367
    },
    {
      "epoch": 0.5355402298850575,
      "grad_norm": 0.743243932723999,
      "learning_rate": 0.0001466224921774342,
      "loss": 1.9007,
      "num_input_tokens_seen": 8750992,
      "step": 1456,
      "train_runtime": 1469.4036,
      "train_tokens_per_second": 5955.472
    },
    {
      "epoch": 0.5359080459770115,
      "grad_norm": 0.7098945379257202,
      "learning_rate": 0.0001465856801030738,
      "loss": 1.9738,
      "num_input_tokens_seen": 8756368,
      "step": 1457,
      "train_runtime": 1470.3048,
      "train_tokens_per_second": 5955.478
    },
    {
      "epoch": 0.5362758620689655,
      "grad_norm": 0.7724363207817078,
      "learning_rate": 0.00014654886802871342,
      "loss": 1.8807,
      "num_input_tokens_seen": 8761376,
      "step": 1458,
      "train_runtime": 1471.161,
      "train_tokens_per_second": 5955.416
    },
    {
      "epoch": 0.5366436781609195,
      "grad_norm": 0.7502484321594238,
      "learning_rate": 0.00014651205595435306,
      "loss": 1.6915,
      "num_input_tokens_seen": 8766192,
      "step": 1459,
      "train_runtime": 1472.01,
      "train_tokens_per_second": 5955.253
    },
    {
      "epoch": 0.5370114942528735,
      "grad_norm": 0.6478996276855469,
      "learning_rate": 0.00014647524387999264,
      "loss": 1.7703,
      "num_input_tokens_seen": 8772624,
      "step": 1460,
      "train_runtime": 1473.0389,
      "train_tokens_per_second": 5955.46
    },
    {
      "epoch": 0.5373793103448276,
      "grad_norm": 0.7149221301078796,
      "learning_rate": 0.00014643843180563225,
      "loss": 1.6999,
      "num_input_tokens_seen": 8777968,
      "step": 1461,
      "train_runtime": 1473.912,
      "train_tokens_per_second": 5955.558
    },
    {
      "epoch": 0.5377471264367816,
      "grad_norm": 0.7612764835357666,
      "learning_rate": 0.00014640161973127186,
      "loss": 1.7736,
      "num_input_tokens_seen": 8783584,
      "step": 1462,
      "train_runtime": 1474.8196,
      "train_tokens_per_second": 5955.701
    },
    {
      "epoch": 0.5381149425287356,
      "grad_norm": 0.8812617659568787,
      "learning_rate": 0.00014636480765691147,
      "loss": 1.9525,
      "num_input_tokens_seen": 8789696,
      "step": 1463,
      "train_runtime": 1475.7964,
      "train_tokens_per_second": 5955.9
    },
    {
      "epoch": 0.5384827586206896,
      "grad_norm": 0.7720367312431335,
      "learning_rate": 0.00014632799558255108,
      "loss": 1.836,
      "num_input_tokens_seen": 8796368,
      "step": 1464,
      "train_runtime": 1476.8689,
      "train_tokens_per_second": 5956.093
    },
    {
      "epoch": 0.5388505747126436,
      "grad_norm": 0.802435040473938,
      "learning_rate": 0.00014629118350819068,
      "loss": 1.8637,
      "num_input_tokens_seen": 8804656,
      "step": 1465,
      "train_runtime": 1478.1706,
      "train_tokens_per_second": 5956.454
    },
    {
      "epoch": 0.5392183908045977,
      "grad_norm": 0.7168628573417664,
      "learning_rate": 0.0001462543714338303,
      "loss": 1.6763,
      "num_input_tokens_seen": 8809744,
      "step": 1466,
      "train_runtime": 1479.0385,
      "train_tokens_per_second": 5956.399
    },
    {
      "epoch": 0.5395862068965517,
      "grad_norm": 0.7387890815734863,
      "learning_rate": 0.0001462175593594699,
      "loss": 1.7679,
      "num_input_tokens_seen": 8817200,
      "step": 1467,
      "train_runtime": 1480.2136,
      "train_tokens_per_second": 5956.708
    },
    {
      "epoch": 0.5399540229885057,
      "grad_norm": 0.7312313914299011,
      "learning_rate": 0.00014618074728510954,
      "loss": 1.7397,
      "num_input_tokens_seen": 8822592,
      "step": 1468,
      "train_runtime": 1481.1268,
      "train_tokens_per_second": 5956.676
    },
    {
      "epoch": 0.5403218390804597,
      "grad_norm": 0.7393628358840942,
      "learning_rate": 0.00014614393521074915,
      "loss": 1.6317,
      "num_input_tokens_seen": 8829520,
      "step": 1469,
      "train_runtime": 1482.2097,
      "train_tokens_per_second": 5956.998
    },
    {
      "epoch": 0.5406896551724137,
      "grad_norm": 0.7658875584602356,
      "learning_rate": 0.00014610712313638876,
      "loss": 1.7367,
      "num_input_tokens_seen": 8834592,
      "step": 1470,
      "train_runtime": 1483.0665,
      "train_tokens_per_second": 5956.976
    },
    {
      "epoch": 0.5410574712643678,
      "grad_norm": 0.8385458588600159,
      "learning_rate": 0.00014607031106202836,
      "loss": 1.7231,
      "num_input_tokens_seen": 8839808,
      "step": 1471,
      "train_runtime": 1484.4497,
      "train_tokens_per_second": 5954.939
    },
    {
      "epoch": 0.5414252873563219,
      "grad_norm": 0.8299917578697205,
      "learning_rate": 0.00014603349898766797,
      "loss": 1.8165,
      "num_input_tokens_seen": 8845632,
      "step": 1472,
      "train_runtime": 1485.4174,
      "train_tokens_per_second": 5954.981
    },
    {
      "epoch": 0.5417931034482759,
      "grad_norm": 0.7017998695373535,
      "learning_rate": 0.00014599668691330756,
      "loss": 1.7511,
      "num_input_tokens_seen": 8851744,
      "step": 1473,
      "train_runtime": 1486.4067,
      "train_tokens_per_second": 5955.129
    },
    {
      "epoch": 0.5421609195402299,
      "grad_norm": 0.832878589630127,
      "learning_rate": 0.00014595987483894716,
      "loss": 1.9765,
      "num_input_tokens_seen": 8856320,
      "step": 1474,
      "train_runtime": 1487.2134,
      "train_tokens_per_second": 5954.976
    },
    {
      "epoch": 0.542528735632184,
      "grad_norm": 0.6871554851531982,
      "learning_rate": 0.00014592306276458677,
      "loss": 1.6499,
      "num_input_tokens_seen": 8861040,
      "step": 1475,
      "train_runtime": 1488.0334,
      "train_tokens_per_second": 5954.866
    },
    {
      "epoch": 0.542896551724138,
      "grad_norm": 0.7615302205085754,
      "learning_rate": 0.0001458862506902264,
      "loss": 1.7343,
      "num_input_tokens_seen": 8865904,
      "step": 1476,
      "train_runtime": 1488.8609,
      "train_tokens_per_second": 5954.823
    },
    {
      "epoch": 0.543264367816092,
      "grad_norm": 0.8005667924880981,
      "learning_rate": 0.00014584943861586602,
      "loss": 1.9106,
      "num_input_tokens_seen": 8870800,
      "step": 1477,
      "train_runtime": 1489.693,
      "train_tokens_per_second": 5954.784
    },
    {
      "epoch": 0.543632183908046,
      "grad_norm": 0.8917537331581116,
      "learning_rate": 0.00014581262654150563,
      "loss": 1.834,
      "num_input_tokens_seen": 8879760,
      "step": 1478,
      "train_runtime": 1491.0717,
      "train_tokens_per_second": 5955.287
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.716110110282898,
      "learning_rate": 0.00014577581446714524,
      "loss": 1.8432,
      "num_input_tokens_seen": 8886960,
      "step": 1479,
      "train_runtime": 1492.2273,
      "train_tokens_per_second": 5955.5
    },
    {
      "epoch": 0.5443678160919541,
      "grad_norm": 0.736108660697937,
      "learning_rate": 0.00014573900239278484,
      "loss": 1.6274,
      "num_input_tokens_seen": 8894736,
      "step": 1480,
      "train_runtime": 1493.4512,
      "train_tokens_per_second": 5955.827
    },
    {
      "epoch": 0.5447356321839081,
      "grad_norm": 0.7554885745048523,
      "learning_rate": 0.00014570219031842445,
      "loss": 1.8776,
      "num_input_tokens_seen": 8900208,
      "step": 1481,
      "train_runtime": 1494.3712,
      "train_tokens_per_second": 5955.821
    },
    {
      "epoch": 0.5451034482758621,
      "grad_norm": 0.7288740277290344,
      "learning_rate": 0.00014566537824406406,
      "loss": 1.7232,
      "num_input_tokens_seen": 8906448,
      "step": 1482,
      "train_runtime": 1495.3899,
      "train_tokens_per_second": 5955.937
    },
    {
      "epoch": 0.5454712643678161,
      "grad_norm": 0.6896114349365234,
      "learning_rate": 0.00014562856616970367,
      "loss": 1.6443,
      "num_input_tokens_seen": 8912032,
      "step": 1483,
      "train_runtime": 1496.3142,
      "train_tokens_per_second": 5955.99
    },
    {
      "epoch": 0.5458390804597701,
      "grad_norm": 0.7094660997390747,
      "learning_rate": 0.00014559175409534328,
      "loss": 1.6302,
      "num_input_tokens_seen": 8917136,
      "step": 1484,
      "train_runtime": 1497.1773,
      "train_tokens_per_second": 5955.965
    },
    {
      "epoch": 0.5462068965517242,
      "grad_norm": 0.7645075917243958,
      "learning_rate": 0.0001455549420209829,
      "loss": 1.8246,
      "num_input_tokens_seen": 8925280,
      "step": 1485,
      "train_runtime": 1498.4548,
      "train_tokens_per_second": 5956.323
    },
    {
      "epoch": 0.5465747126436782,
      "grad_norm": 0.7458211183547974,
      "learning_rate": 0.0001455181299466225,
      "loss": 1.9777,
      "num_input_tokens_seen": 8929664,
      "step": 1486,
      "train_runtime": 1499.2019,
      "train_tokens_per_second": 5956.279
    },
    {
      "epoch": 0.5469425287356322,
      "grad_norm": 0.7357628345489502,
      "learning_rate": 0.0001454813178722621,
      "loss": 1.7962,
      "num_input_tokens_seen": 8934400,
      "step": 1487,
      "train_runtime": 1500.0131,
      "train_tokens_per_second": 5956.215
    },
    {
      "epoch": 0.5473103448275862,
      "grad_norm": 0.7921798825263977,
      "learning_rate": 0.00014544450579790172,
      "loss": 1.9475,
      "num_input_tokens_seen": 8939488,
      "step": 1488,
      "train_runtime": 1500.8562,
      "train_tokens_per_second": 5956.259
    },
    {
      "epoch": 0.5476781609195402,
      "grad_norm": 0.6897305846214294,
      "learning_rate": 0.00014540769372354132,
      "loss": 1.6873,
      "num_input_tokens_seen": 8945168,
      "step": 1489,
      "train_runtime": 1501.7927,
      "train_tokens_per_second": 5956.327
    },
    {
      "epoch": 0.5480459770114943,
      "grad_norm": 0.7136052846908569,
      "learning_rate": 0.00014537088164918093,
      "loss": 1.7006,
      "num_input_tokens_seen": 8951696,
      "step": 1490,
      "train_runtime": 1502.8463,
      "train_tokens_per_second": 5956.495
    },
    {
      "epoch": 0.5484137931034483,
      "grad_norm": 0.716546356678009,
      "learning_rate": 0.00014533406957482054,
      "loss": 1.8145,
      "num_input_tokens_seen": 8956768,
      "step": 1491,
      "train_runtime": 1503.6865,
      "train_tokens_per_second": 5956.54
    },
    {
      "epoch": 0.5487816091954023,
      "grad_norm": 0.7017339468002319,
      "learning_rate": 0.00014529725750046015,
      "loss": 1.7564,
      "num_input_tokens_seen": 8962176,
      "step": 1492,
      "train_runtime": 1504.5732,
      "train_tokens_per_second": 5956.623
    },
    {
      "epoch": 0.5491494252873563,
      "grad_norm": 0.7172056436538696,
      "learning_rate": 0.0001452604454260998,
      "loss": 1.6274,
      "num_input_tokens_seen": 8966976,
      "step": 1493,
      "train_runtime": 1505.3678,
      "train_tokens_per_second": 5956.668
    },
    {
      "epoch": 0.5495172413793104,
      "grad_norm": 0.719150960445404,
      "learning_rate": 0.0001452236333517394,
      "loss": 1.6852,
      "num_input_tokens_seen": 8972080,
      "step": 1494,
      "train_runtime": 1506.2439,
      "train_tokens_per_second": 5956.592
    },
    {
      "epoch": 0.5498850574712644,
      "grad_norm": 0.6909351348876953,
      "learning_rate": 0.000145186821277379,
      "loss": 1.6893,
      "num_input_tokens_seen": 8977792,
      "step": 1495,
      "train_runtime": 1507.1743,
      "train_tokens_per_second": 5956.705
    },
    {
      "epoch": 0.5502528735632184,
      "grad_norm": 0.7258477807044983,
      "learning_rate": 0.0001451500092030186,
      "loss": 1.8736,
      "num_input_tokens_seen": 8982656,
      "step": 1496,
      "train_runtime": 1507.9864,
      "train_tokens_per_second": 5956.722
    },
    {
      "epoch": 0.5506206896551724,
      "grad_norm": 0.7304890155792236,
      "learning_rate": 0.0001451131971286582,
      "loss": 1.9225,
      "num_input_tokens_seen": 8988592,
      "step": 1497,
      "train_runtime": 1508.9593,
      "train_tokens_per_second": 5956.816
    },
    {
      "epoch": 0.5509885057471264,
      "grad_norm": 0.7762201428413391,
      "learning_rate": 0.0001450763850542978,
      "loss": 1.8553,
      "num_input_tokens_seen": 8993248,
      "step": 1498,
      "train_runtime": 1509.7696,
      "train_tokens_per_second": 5956.702
    },
    {
      "epoch": 0.5513563218390805,
      "grad_norm": 0.6422848105430603,
      "learning_rate": 0.00014503957297993741,
      "loss": 1.7149,
      "num_input_tokens_seen": 8999744,
      "step": 1499,
      "train_runtime": 1510.7929,
      "train_tokens_per_second": 5956.968
    },
    {
      "epoch": 0.5517241379310345,
      "grad_norm": 0.7266050577163696,
      "learning_rate": 0.00014500276090557702,
      "loss": 1.8602,
      "num_input_tokens_seen": 9004432,
      "step": 1500,
      "train_runtime": 1511.6212,
      "train_tokens_per_second": 5956.805
    },
    {
      "epoch": 0.5520919540229885,
      "grad_norm": 0.721544623374939,
      "learning_rate": 0.00014496594883121663,
      "loss": 1.5666,
      "num_input_tokens_seen": 9009488,
      "step": 1501,
      "train_runtime": 1513.0064,
      "train_tokens_per_second": 5954.693
    },
    {
      "epoch": 0.5524597701149425,
      "grad_norm": 0.8202746510505676,
      "learning_rate": 0.00014492913675685627,
      "loss": 1.834,
      "num_input_tokens_seen": 9014608,
      "step": 1502,
      "train_runtime": 1513.855,
      "train_tokens_per_second": 5954.737
    },
    {
      "epoch": 0.5528275862068965,
      "grad_norm": 0.6941748261451721,
      "learning_rate": 0.00014489232468249588,
      "loss": 1.6427,
      "num_input_tokens_seen": 9020256,
      "step": 1503,
      "train_runtime": 1514.786,
      "train_tokens_per_second": 5954.806
    },
    {
      "epoch": 0.5531954022988506,
      "grad_norm": 0.7136889696121216,
      "learning_rate": 0.00014485551260813549,
      "loss": 1.7417,
      "num_input_tokens_seen": 9026224,
      "step": 1504,
      "train_runtime": 1515.7735,
      "train_tokens_per_second": 5954.863
    },
    {
      "epoch": 0.5535632183908046,
      "grad_norm": 0.7266203761100769,
      "learning_rate": 0.0001448187005337751,
      "loss": 1.7972,
      "num_input_tokens_seen": 9032240,
      "step": 1505,
      "train_runtime": 1516.7763,
      "train_tokens_per_second": 5954.893
    },
    {
      "epoch": 0.5539310344827586,
      "grad_norm": 0.6732388138771057,
      "learning_rate": 0.0001447818884594147,
      "loss": 1.7354,
      "num_input_tokens_seen": 9037376,
      "step": 1506,
      "train_runtime": 1517.6366,
      "train_tokens_per_second": 5954.901
    },
    {
      "epoch": 0.5542988505747126,
      "grad_norm": 0.7037345170974731,
      "learning_rate": 0.0001447450763850543,
      "loss": 1.7045,
      "num_input_tokens_seen": 9045904,
      "step": 1507,
      "train_runtime": 1518.9749,
      "train_tokens_per_second": 5955.269
    },
    {
      "epoch": 0.5546666666666666,
      "grad_norm": 0.7394117712974548,
      "learning_rate": 0.00014470826431069392,
      "loss": 1.8442,
      "num_input_tokens_seen": 9051856,
      "step": 1508,
      "train_runtime": 1519.9305,
      "train_tokens_per_second": 5955.441
    },
    {
      "epoch": 0.5550344827586207,
      "grad_norm": 0.7001480460166931,
      "learning_rate": 0.0001446714522363335,
      "loss": 1.7071,
      "num_input_tokens_seen": 9056784,
      "step": 1509,
      "train_runtime": 1520.7739,
      "train_tokens_per_second": 5955.378
    },
    {
      "epoch": 0.5554022988505747,
      "grad_norm": 0.8510955572128296,
      "learning_rate": 0.00014463464016197314,
      "loss": 1.885,
      "num_input_tokens_seen": 9061984,
      "step": 1510,
      "train_runtime": 1521.6431,
      "train_tokens_per_second": 5955.394
    },
    {
      "epoch": 0.5557701149425287,
      "grad_norm": 0.7102080583572388,
      "learning_rate": 0.00014459782808761275,
      "loss": 1.7542,
      "num_input_tokens_seen": 9067248,
      "step": 1511,
      "train_runtime": 1522.52,
      "train_tokens_per_second": 5955.421
    },
    {
      "epoch": 0.5561379310344827,
      "grad_norm": 0.7432495355606079,
      "learning_rate": 0.00014456101601325236,
      "loss": 1.7311,
      "num_input_tokens_seen": 9074384,
      "step": 1512,
      "train_runtime": 1523.6388,
      "train_tokens_per_second": 5955.732
    },
    {
      "epoch": 0.5565057471264367,
      "grad_norm": 0.7131504416465759,
      "learning_rate": 0.00014452420393889197,
      "loss": 1.8028,
      "num_input_tokens_seen": 9081296,
      "step": 1513,
      "train_runtime": 1524.7709,
      "train_tokens_per_second": 5955.843
    },
    {
      "epoch": 0.5568735632183908,
      "grad_norm": 0.7579213976860046,
      "learning_rate": 0.00014448739186453157,
      "loss": 1.7905,
      "num_input_tokens_seen": 9086800,
      "step": 1514,
      "train_runtime": 1525.6932,
      "train_tokens_per_second": 5955.85
    },
    {
      "epoch": 0.5572413793103448,
      "grad_norm": 0.7163369059562683,
      "learning_rate": 0.00014445057979017118,
      "loss": 1.785,
      "num_input_tokens_seen": 9091984,
      "step": 1515,
      "train_runtime": 1526.5762,
      "train_tokens_per_second": 5955.801
    },
    {
      "epoch": 0.5576091954022988,
      "grad_norm": 0.7238402962684631,
      "learning_rate": 0.0001444137677158108,
      "loss": 1.8952,
      "num_input_tokens_seen": 9096720,
      "step": 1516,
      "train_runtime": 1527.403,
      "train_tokens_per_second": 5955.678
    },
    {
      "epoch": 0.5579770114942528,
      "grad_norm": 0.6420456767082214,
      "learning_rate": 0.0001443769556414504,
      "loss": 1.6005,
      "num_input_tokens_seen": 9103920,
      "step": 1517,
      "train_runtime": 1528.547,
      "train_tokens_per_second": 5955.931
    },
    {
      "epoch": 0.5583448275862068,
      "grad_norm": 0.7540795803070068,
      "learning_rate": 0.00014434014356709,
      "loss": 1.764,
      "num_input_tokens_seen": 9110816,
      "step": 1518,
      "train_runtime": 1529.6538,
      "train_tokens_per_second": 5956.129
    },
    {
      "epoch": 0.5587126436781609,
      "grad_norm": 0.7519972920417786,
      "learning_rate": 0.00014430333149272962,
      "loss": 1.6031,
      "num_input_tokens_seen": 9117616,
      "step": 1519,
      "train_runtime": 1530.7302,
      "train_tokens_per_second": 5956.383
    },
    {
      "epoch": 0.5590804597701149,
      "grad_norm": 0.9189901947975159,
      "learning_rate": 0.00014426651941836923,
      "loss": 1.9835,
      "num_input_tokens_seen": 9122400,
      "step": 1520,
      "train_runtime": 1531.5378,
      "train_tokens_per_second": 5956.366
    },
    {
      "epoch": 0.5594482758620689,
      "grad_norm": 0.7206523418426514,
      "learning_rate": 0.00014422970734400884,
      "loss": 1.8126,
      "num_input_tokens_seen": 9128368,
      "step": 1521,
      "train_runtime": 1532.5308,
      "train_tokens_per_second": 5956.401
    },
    {
      "epoch": 0.559816091954023,
      "grad_norm": 0.7543719410896301,
      "learning_rate": 0.00014419289526964845,
      "loss": 1.7646,
      "num_input_tokens_seen": 9134016,
      "step": 1522,
      "train_runtime": 1533.4732,
      "train_tokens_per_second": 5956.424
    },
    {
      "epoch": 0.5601839080459771,
      "grad_norm": 0.7814950346946716,
      "learning_rate": 0.00014415608319528805,
      "loss": 1.8127,
      "num_input_tokens_seen": 9139088,
      "step": 1523,
      "train_runtime": 1534.3196,
      "train_tokens_per_second": 5956.443
    },
    {
      "epoch": 0.5605517241379311,
      "grad_norm": 0.736058235168457,
      "learning_rate": 0.00014411927112092766,
      "loss": 1.7082,
      "num_input_tokens_seen": 9145120,
      "step": 1524,
      "train_runtime": 1535.308,
      "train_tokens_per_second": 5956.538
    },
    {
      "epoch": 0.5609195402298851,
      "grad_norm": 0.7190161347389221,
      "learning_rate": 0.00014408245904656727,
      "loss": 1.8945,
      "num_input_tokens_seen": 9154080,
      "step": 1525,
      "train_runtime": 1536.7074,
      "train_tokens_per_second": 5956.944
    },
    {
      "epoch": 0.5612873563218391,
      "grad_norm": 0.7549131512641907,
      "learning_rate": 0.00014404564697220688,
      "loss": 1.8791,
      "num_input_tokens_seen": 9159280,
      "step": 1526,
      "train_runtime": 1537.5611,
      "train_tokens_per_second": 5957.018
    },
    {
      "epoch": 0.5616551724137931,
      "grad_norm": 0.6511595249176025,
      "learning_rate": 0.0001440088348978465,
      "loss": 1.7814,
      "num_input_tokens_seen": 9165888,
      "step": 1527,
      "train_runtime": 1538.6242,
      "train_tokens_per_second": 5957.197
    },
    {
      "epoch": 0.5620229885057472,
      "grad_norm": 0.7519131898880005,
      "learning_rate": 0.00014397202282348613,
      "loss": 1.6841,
      "num_input_tokens_seen": 9172544,
      "step": 1528,
      "train_runtime": 1539.6694,
      "train_tokens_per_second": 5957.476
    },
    {
      "epoch": 0.5623908045977012,
      "grad_norm": 0.6885892152786255,
      "learning_rate": 0.00014393521074912573,
      "loss": 1.7104,
      "num_input_tokens_seen": 9178688,
      "step": 1529,
      "train_runtime": 1540.6876,
      "train_tokens_per_second": 5957.527
    },
    {
      "epoch": 0.5627586206896552,
      "grad_norm": 0.7921882271766663,
      "learning_rate": 0.00014389839867476534,
      "loss": 1.8299,
      "num_input_tokens_seen": 9183648,
      "step": 1530,
      "train_runtime": 1541.5326,
      "train_tokens_per_second": 5957.479
    },
    {
      "epoch": 0.5631264367816092,
      "grad_norm": 0.7679993510246277,
      "learning_rate": 0.00014386158660040495,
      "loss": 1.5701,
      "num_input_tokens_seen": 9190976,
      "step": 1531,
      "train_runtime": 1543.1608,
      "train_tokens_per_second": 5955.942
    },
    {
      "epoch": 0.5634942528735633,
      "grad_norm": 0.789773166179657,
      "learning_rate": 0.00014382477452604453,
      "loss": 1.9069,
      "num_input_tokens_seen": 9196048,
      "step": 1532,
      "train_runtime": 1544.0384,
      "train_tokens_per_second": 5955.842
    },
    {
      "epoch": 0.5638620689655173,
      "grad_norm": 0.8108073472976685,
      "learning_rate": 0.00014378796245168414,
      "loss": 1.6931,
      "num_input_tokens_seen": 9201056,
      "step": 1533,
      "train_runtime": 1544.874,
      "train_tokens_per_second": 5955.862
    },
    {
      "epoch": 0.5642298850574713,
      "grad_norm": 0.7268442511558533,
      "learning_rate": 0.00014375115037732375,
      "loss": 1.7064,
      "num_input_tokens_seen": 9206480,
      "step": 1534,
      "train_runtime": 1545.7775,
      "train_tokens_per_second": 5955.889
    },
    {
      "epoch": 0.5645977011494253,
      "grad_norm": 0.6699070930480957,
      "learning_rate": 0.00014371433830296336,
      "loss": 1.7518,
      "num_input_tokens_seen": 9212976,
      "step": 1535,
      "train_runtime": 1546.8221,
      "train_tokens_per_second": 5956.067
    },
    {
      "epoch": 0.5649655172413793,
      "grad_norm": 0.7682430744171143,
      "learning_rate": 0.000143677526228603,
      "loss": 1.6584,
      "num_input_tokens_seen": 9219744,
      "step": 1536,
      "train_runtime": 1547.9016,
      "train_tokens_per_second": 5956.286
    },
    {
      "epoch": 0.5653333333333334,
      "grad_norm": 0.6585272550582886,
      "learning_rate": 0.0001436407141542426,
      "loss": 1.738,
      "num_input_tokens_seen": 9226000,
      "step": 1537,
      "train_runtime": 1548.9294,
      "train_tokens_per_second": 5956.372
    },
    {
      "epoch": 0.5657011494252874,
      "grad_norm": 0.7462416291236877,
      "learning_rate": 0.00014360390207988221,
      "loss": 1.605,
      "num_input_tokens_seen": 9232496,
      "step": 1538,
      "train_runtime": 1550.0037,
      "train_tokens_per_second": 5956.435
    },
    {
      "epoch": 0.5660689655172414,
      "grad_norm": 0.7598044872283936,
      "learning_rate": 0.00014356709000552182,
      "loss": 1.6289,
      "num_input_tokens_seen": 9238400,
      "step": 1539,
      "train_runtime": 1550.9758,
      "train_tokens_per_second": 5956.508
    },
    {
      "epoch": 0.5664367816091954,
      "grad_norm": 0.7205649018287659,
      "learning_rate": 0.00014353027793116143,
      "loss": 1.7779,
      "num_input_tokens_seen": 9243728,
      "step": 1540,
      "train_runtime": 1551.8721,
      "train_tokens_per_second": 5956.501
    },
    {
      "epoch": 0.5668045977011494,
      "grad_norm": 0.6952161192893982,
      "learning_rate": 0.00014349346585680104,
      "loss": 1.6689,
      "num_input_tokens_seen": 9250256,
      "step": 1541,
      "train_runtime": 1552.9207,
      "train_tokens_per_second": 5956.683
    },
    {
      "epoch": 0.5671724137931035,
      "grad_norm": 0.7485373020172119,
      "learning_rate": 0.00014345665378244065,
      "loss": 1.8296,
      "num_input_tokens_seen": 9255984,
      "step": 1542,
      "train_runtime": 1553.8545,
      "train_tokens_per_second": 5956.789
    },
    {
      "epoch": 0.5675402298850575,
      "grad_norm": 0.7204800248146057,
      "learning_rate": 0.00014341984170808026,
      "loss": 1.6307,
      "num_input_tokens_seen": 9260464,
      "step": 1543,
      "train_runtime": 1554.6314,
      "train_tokens_per_second": 5956.694
    },
    {
      "epoch": 0.5679080459770115,
      "grad_norm": 0.7552061676979065,
      "learning_rate": 0.00014338302963371987,
      "loss": 1.8105,
      "num_input_tokens_seen": 9266272,
      "step": 1544,
      "train_runtime": 1555.5878,
      "train_tokens_per_second": 5956.766
    },
    {
      "epoch": 0.5682758620689655,
      "grad_norm": 0.7821717262268066,
      "learning_rate": 0.00014334621755935948,
      "loss": 1.9491,
      "num_input_tokens_seen": 9273712,
      "step": 1545,
      "train_runtime": 1556.7733,
      "train_tokens_per_second": 5957.009
    },
    {
      "epoch": 0.5686436781609195,
      "grad_norm": 0.7952330708503723,
      "learning_rate": 0.00014330940548499909,
      "loss": 1.884,
      "num_input_tokens_seen": 9279344,
      "step": 1546,
      "train_runtime": 1557.7073,
      "train_tokens_per_second": 5957.052
    },
    {
      "epoch": 0.5690114942528736,
      "grad_norm": 0.8070065379142761,
      "learning_rate": 0.0001432725934106387,
      "loss": 1.9077,
      "num_input_tokens_seen": 9283488,
      "step": 1547,
      "train_runtime": 1558.4422,
      "train_tokens_per_second": 5956.902
    },
    {
      "epoch": 0.5693793103448276,
      "grad_norm": 0.77339106798172,
      "learning_rate": 0.0001432357813362783,
      "loss": 1.8652,
      "num_input_tokens_seen": 9288368,
      "step": 1548,
      "train_runtime": 1559.2853,
      "train_tokens_per_second": 5956.811
    },
    {
      "epoch": 0.5697471264367816,
      "grad_norm": 0.7823001146316528,
      "learning_rate": 0.0001431989692619179,
      "loss": 1.7667,
      "num_input_tokens_seen": 9293184,
      "step": 1549,
      "train_runtime": 1560.1223,
      "train_tokens_per_second": 5956.702
    },
    {
      "epoch": 0.5701149425287356,
      "grad_norm": 0.6910526156425476,
      "learning_rate": 0.00014316215718755752,
      "loss": 1.5904,
      "num_input_tokens_seen": 9297824,
      "step": 1550,
      "train_runtime": 1560.9223,
      "train_tokens_per_second": 5956.622
    },
    {
      "epoch": 0.5704827586206896,
      "grad_norm": 0.7571379542350769,
      "learning_rate": 0.00014312534511319713,
      "loss": 1.7796,
      "num_input_tokens_seen": 9304640,
      "step": 1551,
      "train_runtime": 1562.0164,
      "train_tokens_per_second": 5956.813
    },
    {
      "epoch": 0.5708505747126437,
      "grad_norm": 0.7574077844619751,
      "learning_rate": 0.00014308853303883674,
      "loss": 1.78,
      "num_input_tokens_seen": 9309376,
      "step": 1552,
      "train_runtime": 1562.8363,
      "train_tokens_per_second": 5956.719
    },
    {
      "epoch": 0.5712183908045977,
      "grad_norm": 0.659858763217926,
      "learning_rate": 0.00014305172096447638,
      "loss": 1.7382,
      "num_input_tokens_seen": 9315296,
      "step": 1553,
      "train_runtime": 1563.8107,
      "train_tokens_per_second": 5956.793
    },
    {
      "epoch": 0.5715862068965517,
      "grad_norm": 0.7899975776672363,
      "learning_rate": 0.00014301490889011598,
      "loss": 1.7475,
      "num_input_tokens_seen": 9322784,
      "step": 1554,
      "train_runtime": 1564.9729,
      "train_tokens_per_second": 5957.153
    },
    {
      "epoch": 0.5719540229885057,
      "grad_norm": 0.8226660490036011,
      "learning_rate": 0.0001429780968157556,
      "loss": 1.781,
      "num_input_tokens_seen": 9329168,
      "step": 1555,
      "train_runtime": 1565.9869,
      "train_tokens_per_second": 5957.373
    },
    {
      "epoch": 0.5723218390804597,
      "grad_norm": 0.7388761639595032,
      "learning_rate": 0.00014294128474139517,
      "loss": 1.7917,
      "num_input_tokens_seen": 9335648,
      "step": 1556,
      "train_runtime": 1567.0516,
      "train_tokens_per_second": 5957.461
    },
    {
      "epoch": 0.5726896551724138,
      "grad_norm": 0.7576459050178528,
      "learning_rate": 0.00014290447266703478,
      "loss": 1.7311,
      "num_input_tokens_seen": 9342512,
      "step": 1557,
      "train_runtime": 1568.1657,
      "train_tokens_per_second": 5957.605
    },
    {
      "epoch": 0.5730574712643678,
      "grad_norm": 0.6782925724983215,
      "learning_rate": 0.0001428676605926744,
      "loss": 1.91,
      "num_input_tokens_seen": 9348256,
      "step": 1558,
      "train_runtime": 1569.1158,
      "train_tokens_per_second": 5957.658
    },
    {
      "epoch": 0.5734252873563218,
      "grad_norm": 0.7342045903205872,
      "learning_rate": 0.000142830848518314,
      "loss": 1.7617,
      "num_input_tokens_seen": 9360976,
      "step": 1559,
      "train_runtime": 1571.017,
      "train_tokens_per_second": 5958.545
    },
    {
      "epoch": 0.5737931034482758,
      "grad_norm": 0.7207436561584473,
      "learning_rate": 0.0001427940364439536,
      "loss": 1.8567,
      "num_input_tokens_seen": 9370688,
      "step": 1560,
      "train_runtime": 1572.4884,
      "train_tokens_per_second": 5959.146
    },
    {
      "epoch": 0.5741609195402299,
      "grad_norm": 0.7182794809341431,
      "learning_rate": 0.00014275722436959322,
      "loss": 1.9529,
      "num_input_tokens_seen": 9377184,
      "step": 1561,
      "train_runtime": 1574.0252,
      "train_tokens_per_second": 5957.455
    },
    {
      "epoch": 0.5745287356321839,
      "grad_norm": 0.7070382237434387,
      "learning_rate": 0.00014272041229523286,
      "loss": 1.6306,
      "num_input_tokens_seen": 9383136,
      "step": 1562,
      "train_runtime": 1575.0102,
      "train_tokens_per_second": 5957.508
    },
    {
      "epoch": 0.5748965517241379,
      "grad_norm": 0.7206614017486572,
      "learning_rate": 0.00014268360022087246,
      "loss": 1.5928,
      "num_input_tokens_seen": 9388704,
      "step": 1563,
      "train_runtime": 1575.9126,
      "train_tokens_per_second": 5957.63
    },
    {
      "epoch": 0.5752643678160919,
      "grad_norm": 0.7111892700195312,
      "learning_rate": 0.00014264678814651207,
      "loss": 1.8206,
      "num_input_tokens_seen": 9394832,
      "step": 1564,
      "train_runtime": 1576.8991,
      "train_tokens_per_second": 5957.789
    },
    {
      "epoch": 0.5756321839080459,
      "grad_norm": 0.7477349638938904,
      "learning_rate": 0.00014260997607215168,
      "loss": 1.7513,
      "num_input_tokens_seen": 9400608,
      "step": 1565,
      "train_runtime": 1577.8661,
      "train_tokens_per_second": 5957.798
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.8348462581634521,
      "learning_rate": 0.0001425731639977913,
      "loss": 1.8731,
      "num_input_tokens_seen": 9406448,
      "step": 1566,
      "train_runtime": 1578.8117,
      "train_tokens_per_second": 5957.929
    },
    {
      "epoch": 0.576367816091954,
      "grad_norm": 0.7648438811302185,
      "learning_rate": 0.0001425363519234309,
      "loss": 1.893,
      "num_input_tokens_seen": 9411600,
      "step": 1567,
      "train_runtime": 1579.6747,
      "train_tokens_per_second": 5957.936
    },
    {
      "epoch": 0.576735632183908,
      "grad_norm": 0.7014155983924866,
      "learning_rate": 0.0001424995398490705,
      "loss": 1.7945,
      "num_input_tokens_seen": 9417312,
      "step": 1568,
      "train_runtime": 1580.6089,
      "train_tokens_per_second": 5958.028
    },
    {
      "epoch": 0.577103448275862,
      "grad_norm": 0.7736544013023376,
      "learning_rate": 0.0001424627277747101,
      "loss": 1.7166,
      "num_input_tokens_seen": 9422848,
      "step": 1569,
      "train_runtime": 1581.532,
      "train_tokens_per_second": 5958.051
    },
    {
      "epoch": 0.577471264367816,
      "grad_norm": 0.8276627659797668,
      "learning_rate": 0.00014242591570034973,
      "loss": 1.6446,
      "num_input_tokens_seen": 9428288,
      "step": 1570,
      "train_runtime": 1582.4178,
      "train_tokens_per_second": 5958.153
    },
    {
      "epoch": 0.5778390804597701,
      "grad_norm": 0.7937006950378418,
      "learning_rate": 0.00014238910362598934,
      "loss": 1.7648,
      "num_input_tokens_seen": 9433824,
      "step": 1571,
      "train_runtime": 1583.3446,
      "train_tokens_per_second": 5958.162
    },
    {
      "epoch": 0.5782068965517242,
      "grad_norm": 0.743634819984436,
      "learning_rate": 0.00014235229155162894,
      "loss": 1.765,
      "num_input_tokens_seen": 9441504,
      "step": 1572,
      "train_runtime": 1584.5482,
      "train_tokens_per_second": 5958.483
    },
    {
      "epoch": 0.5785747126436782,
      "grad_norm": 0.7494282126426697,
      "learning_rate": 0.00014231547947726855,
      "loss": 1.673,
      "num_input_tokens_seen": 9450128,
      "step": 1573,
      "train_runtime": 1585.8436,
      "train_tokens_per_second": 5959.054
    },
    {
      "epoch": 0.5789425287356322,
      "grad_norm": 0.7202593088150024,
      "learning_rate": 0.00014227866740290816,
      "loss": 1.682,
      "num_input_tokens_seen": 9456912,
      "step": 1574,
      "train_runtime": 1586.9365,
      "train_tokens_per_second": 5959.225
    },
    {
      "epoch": 0.5793103448275863,
      "grad_norm": 0.6323804259300232,
      "learning_rate": 0.00014224185532854777,
      "loss": 1.7759,
      "num_input_tokens_seen": 9466832,
      "step": 1575,
      "train_runtime": 1588.4478,
      "train_tokens_per_second": 5959.8
    },
    {
      "epoch": 0.5796781609195403,
      "grad_norm": 0.5556498169898987,
      "learning_rate": 0.00014220504325418738,
      "loss": 1.8375,
      "num_input_tokens_seen": 9478992,
      "step": 1576,
      "train_runtime": 1590.229,
      "train_tokens_per_second": 5960.772
    },
    {
      "epoch": 0.5800459770114943,
      "grad_norm": 0.740181565284729,
      "learning_rate": 0.000142168231179827,
      "loss": 1.5583,
      "num_input_tokens_seen": 9483216,
      "step": 1577,
      "train_runtime": 1590.9894,
      "train_tokens_per_second": 5960.578
    },
    {
      "epoch": 0.5804137931034483,
      "grad_norm": 0.7218865752220154,
      "learning_rate": 0.0001421314191054666,
      "loss": 1.672,
      "num_input_tokens_seen": 9489568,
      "step": 1578,
      "train_runtime": 1592.0334,
      "train_tokens_per_second": 5960.659
    },
    {
      "epoch": 0.5807816091954023,
      "grad_norm": 0.7338588833808899,
      "learning_rate": 0.0001420946070311062,
      "loss": 1.8931,
      "num_input_tokens_seen": 9496256,
      "step": 1579,
      "train_runtime": 1593.1098,
      "train_tokens_per_second": 5960.83
    },
    {
      "epoch": 0.5811494252873564,
      "grad_norm": 0.7787767052650452,
      "learning_rate": 0.00014205779495674582,
      "loss": 1.5449,
      "num_input_tokens_seen": 9501968,
      "step": 1580,
      "train_runtime": 1594.0508,
      "train_tokens_per_second": 5960.894
    },
    {
      "epoch": 0.5815172413793104,
      "grad_norm": 0.5992093682289124,
      "learning_rate": 0.00014202098288238542,
      "loss": 1.5074,
      "num_input_tokens_seen": 9512912,
      "step": 1581,
      "train_runtime": 1595.726,
      "train_tokens_per_second": 5961.494
    },
    {
      "epoch": 0.5818850574712644,
      "grad_norm": 0.7882895469665527,
      "learning_rate": 0.00014198417080802503,
      "loss": 1.8337,
      "num_input_tokens_seen": 9519696,
      "step": 1582,
      "train_runtime": 1596.7808,
      "train_tokens_per_second": 5961.805
    },
    {
      "epoch": 0.5822528735632184,
      "grad_norm": 0.7843197584152222,
      "learning_rate": 0.00014194735873366464,
      "loss": 1.7564,
      "num_input_tokens_seen": 9528960,
      "step": 1583,
      "train_runtime": 1598.1963,
      "train_tokens_per_second": 5962.322
    },
    {
      "epoch": 0.5826206896551724,
      "grad_norm": 0.6911957263946533,
      "learning_rate": 0.00014191054665930425,
      "loss": 1.8201,
      "num_input_tokens_seen": 9534896,
      "step": 1584,
      "train_runtime": 1599.1925,
      "train_tokens_per_second": 5962.319
    },
    {
      "epoch": 0.5829885057471265,
      "grad_norm": 0.6987912058830261,
      "learning_rate": 0.00014187373458494386,
      "loss": 1.7853,
      "num_input_tokens_seen": 9542608,
      "step": 1585,
      "train_runtime": 1600.4144,
      "train_tokens_per_second": 5962.586
    },
    {
      "epoch": 0.5833563218390805,
      "grad_norm": 0.8136776089668274,
      "learning_rate": 0.00014183692251058347,
      "loss": 1.9193,
      "num_input_tokens_seen": 9546944,
      "step": 1586,
      "train_runtime": 1601.1843,
      "train_tokens_per_second": 5962.427
    },
    {
      "epoch": 0.5837241379310345,
      "grad_norm": 0.6825462579727173,
      "learning_rate": 0.00014180011043622308,
      "loss": 1.6683,
      "num_input_tokens_seen": 9552080,
      "step": 1587,
      "train_runtime": 1602.0215,
      "train_tokens_per_second": 5962.517
    },
    {
      "epoch": 0.5840919540229885,
      "grad_norm": 0.7725069522857666,
      "learning_rate": 0.00014176329836186271,
      "loss": 1.5494,
      "num_input_tokens_seen": 9561296,
      "step": 1588,
      "train_runtime": 1603.4466,
      "train_tokens_per_second": 5962.965
    },
    {
      "epoch": 0.5844597701149425,
      "grad_norm": 0.6888120174407959,
      "learning_rate": 0.00014172648628750232,
      "loss": 1.8348,
      "num_input_tokens_seen": 9566608,
      "step": 1589,
      "train_runtime": 1604.3389,
      "train_tokens_per_second": 5962.96
    },
    {
      "epoch": 0.5848275862068966,
      "grad_norm": 0.6902614831924438,
      "learning_rate": 0.00014168967421314193,
      "loss": 1.7664,
      "num_input_tokens_seen": 9573504,
      "step": 1590,
      "train_runtime": 1605.4505,
      "train_tokens_per_second": 5963.126
    },
    {
      "epoch": 0.5851954022988506,
      "grad_norm": 0.6990094184875488,
      "learning_rate": 0.00014165286213878154,
      "loss": 1.8354,
      "num_input_tokens_seen": 9578512,
      "step": 1591,
      "train_runtime": 1606.7769,
      "train_tokens_per_second": 5961.321
    },
    {
      "epoch": 0.5855632183908046,
      "grad_norm": 0.7648910880088806,
      "learning_rate": 0.00014161605006442112,
      "loss": 1.938,
      "num_input_tokens_seen": 9584144,
      "step": 1592,
      "train_runtime": 1607.6993,
      "train_tokens_per_second": 5961.403
    },
    {
      "epoch": 0.5859310344827586,
      "grad_norm": 0.7185373306274414,
      "learning_rate": 0.00014157923799006073,
      "loss": 1.7699,
      "num_input_tokens_seen": 9589712,
      "step": 1593,
      "train_runtime": 1608.6148,
      "train_tokens_per_second": 5961.472
    },
    {
      "epoch": 0.5862988505747126,
      "grad_norm": 0.7881008982658386,
      "learning_rate": 0.00014154242591570034,
      "loss": 1.8258,
      "num_input_tokens_seen": 9593984,
      "step": 1594,
      "train_runtime": 1609.4041,
      "train_tokens_per_second": 5961.203
    },
    {
      "epoch": 0.5866666666666667,
      "grad_norm": 0.7478007674217224,
      "learning_rate": 0.00014150561384133995,
      "loss": 1.8598,
      "num_input_tokens_seen": 9599808,
      "step": 1595,
      "train_runtime": 1610.3605,
      "train_tokens_per_second": 5961.279
    },
    {
      "epoch": 0.5870344827586207,
      "grad_norm": 0.7251457571983337,
      "learning_rate": 0.00014146880176697958,
      "loss": 1.9524,
      "num_input_tokens_seen": 9606384,
      "step": 1596,
      "train_runtime": 1611.4233,
      "train_tokens_per_second": 5961.428
    },
    {
      "epoch": 0.5874022988505747,
      "grad_norm": 0.6985070705413818,
      "learning_rate": 0.0001414319896926192,
      "loss": 1.5909,
      "num_input_tokens_seen": 9613808,
      "step": 1597,
      "train_runtime": 1612.6153,
      "train_tokens_per_second": 5961.625
    },
    {
      "epoch": 0.5877701149425287,
      "grad_norm": 0.723919689655304,
      "learning_rate": 0.0001413951776182588,
      "loss": 1.7206,
      "num_input_tokens_seen": 9620704,
      "step": 1598,
      "train_runtime": 1613.7181,
      "train_tokens_per_second": 5961.825
    },
    {
      "epoch": 0.5881379310344828,
      "grad_norm": 0.6995579600334167,
      "learning_rate": 0.0001413583655438984,
      "loss": 1.8156,
      "num_input_tokens_seen": 9625760,
      "step": 1599,
      "train_runtime": 1614.5891,
      "train_tokens_per_second": 5961.74
    },
    {
      "epoch": 0.5885057471264368,
      "grad_norm": 0.7042402625083923,
      "learning_rate": 0.00014132155346953802,
      "loss": 1.7435,
      "num_input_tokens_seen": 9633712,
      "step": 1600,
      "train_runtime": 1615.8097,
      "train_tokens_per_second": 5962.158
    },
    {
      "epoch": 0.5888735632183908,
      "grad_norm": 0.7484963536262512,
      "learning_rate": 0.00014128474139517763,
      "loss": 1.8737,
      "num_input_tokens_seen": 9638352,
      "step": 1601,
      "train_runtime": 1616.6114,
      "train_tokens_per_second": 5962.071
    },
    {
      "epoch": 0.5892413793103448,
      "grad_norm": 0.7516323328018188,
      "learning_rate": 0.00014124792932081724,
      "loss": 1.7346,
      "num_input_tokens_seen": 9643600,
      "step": 1602,
      "train_runtime": 1617.5021,
      "train_tokens_per_second": 5962.032
    },
    {
      "epoch": 0.5896091954022988,
      "grad_norm": 0.7752770185470581,
      "learning_rate": 0.00014121111724645685,
      "loss": 1.9747,
      "num_input_tokens_seen": 9650528,
      "step": 1603,
      "train_runtime": 1618.6587,
      "train_tokens_per_second": 5962.052
    },
    {
      "epoch": 0.5899770114942529,
      "grad_norm": 0.7031365036964417,
      "learning_rate": 0.00014117430517209646,
      "loss": 1.8129,
      "num_input_tokens_seen": 9655072,
      "step": 1604,
      "train_runtime": 1619.4675,
      "train_tokens_per_second": 5961.881
    },
    {
      "epoch": 0.5903448275862069,
      "grad_norm": 0.724492073059082,
      "learning_rate": 0.00014113749309773606,
      "loss": 1.6754,
      "num_input_tokens_seen": 9660656,
      "step": 1605,
      "train_runtime": 1620.3947,
      "train_tokens_per_second": 5961.915
    },
    {
      "epoch": 0.5907126436781609,
      "grad_norm": 0.7145460844039917,
      "learning_rate": 0.00014110068102337567,
      "loss": 1.6511,
      "num_input_tokens_seen": 9667888,
      "step": 1606,
      "train_runtime": 1621.5382,
      "train_tokens_per_second": 5962.171
    },
    {
      "epoch": 0.5910804597701149,
      "grad_norm": 0.7644027471542358,
      "learning_rate": 0.00014106386894901528,
      "loss": 1.9032,
      "num_input_tokens_seen": 9673024,
      "step": 1607,
      "train_runtime": 1622.3767,
      "train_tokens_per_second": 5962.255
    },
    {
      "epoch": 0.5914482758620689,
      "grad_norm": 0.7568883299827576,
      "learning_rate": 0.0001410270568746549,
      "loss": 1.968,
      "num_input_tokens_seen": 9678736,
      "step": 1608,
      "train_runtime": 1623.2874,
      "train_tokens_per_second": 5962.429
    },
    {
      "epoch": 0.591816091954023,
      "grad_norm": 0.7393959760665894,
      "learning_rate": 0.0001409902448002945,
      "loss": 1.7143,
      "num_input_tokens_seen": 9684064,
      "step": 1609,
      "train_runtime": 1624.2223,
      "train_tokens_per_second": 5962.278
    },
    {
      "epoch": 0.592183908045977,
      "grad_norm": 0.7215691208839417,
      "learning_rate": 0.0001409534327259341,
      "loss": 1.5933,
      "num_input_tokens_seen": 9689936,
      "step": 1610,
      "train_runtime": 1625.2071,
      "train_tokens_per_second": 5962.278
    },
    {
      "epoch": 0.592551724137931,
      "grad_norm": 0.6725964546203613,
      "learning_rate": 0.00014091662065157372,
      "loss": 1.5474,
      "num_input_tokens_seen": 9697424,
      "step": 1611,
      "train_runtime": 1626.3965,
      "train_tokens_per_second": 5962.521
    },
    {
      "epoch": 0.592919540229885,
      "grad_norm": 0.7662464380264282,
      "learning_rate": 0.00014087980857721333,
      "loss": 1.6828,
      "num_input_tokens_seen": 9704816,
      "step": 1612,
      "train_runtime": 1627.5835,
      "train_tokens_per_second": 5962.715
    },
    {
      "epoch": 0.593287356321839,
      "grad_norm": 0.7117657661437988,
      "learning_rate": 0.00014084299650285296,
      "loss": 1.8546,
      "num_input_tokens_seen": 9711792,
      "step": 1613,
      "train_runtime": 1628.678,
      "train_tokens_per_second": 5962.991
    },
    {
      "epoch": 0.5936551724137931,
      "grad_norm": 0.7615235447883606,
      "learning_rate": 0.00014080618442849257,
      "loss": 1.7151,
      "num_input_tokens_seen": 9718368,
      "step": 1614,
      "train_runtime": 1629.7387,
      "train_tokens_per_second": 5963.145
    },
    {
      "epoch": 0.5940229885057471,
      "grad_norm": 0.7365219593048096,
      "learning_rate": 0.00014076937235413218,
      "loss": 1.6078,
      "num_input_tokens_seen": 9723696,
      "step": 1615,
      "train_runtime": 1630.6211,
      "train_tokens_per_second": 5963.185
    },
    {
      "epoch": 0.5943908045977011,
      "grad_norm": 0.7892609238624573,
      "learning_rate": 0.00014073256027977176,
      "loss": 1.7772,
      "num_input_tokens_seen": 9729600,
      "step": 1616,
      "train_runtime": 1631.5826,
      "train_tokens_per_second": 5963.29
    },
    {
      "epoch": 0.5947586206896551,
      "grad_norm": 0.6700955629348755,
      "learning_rate": 0.00014069574820541137,
      "loss": 1.9247,
      "num_input_tokens_seen": 9735536,
      "step": 1617,
      "train_runtime": 1632.5541,
      "train_tokens_per_second": 5963.377
    },
    {
      "epoch": 0.5951264367816091,
      "grad_norm": 0.7988229990005493,
      "learning_rate": 0.00014065893613105098,
      "loss": 1.7863,
      "num_input_tokens_seen": 9741680,
      "step": 1618,
      "train_runtime": 1633.5474,
      "train_tokens_per_second": 5963.512
    },
    {
      "epoch": 0.5954942528735632,
      "grad_norm": 0.7733209133148193,
      "learning_rate": 0.0001406221240566906,
      "loss": 1.6866,
      "num_input_tokens_seen": 9745968,
      "step": 1619,
      "train_runtime": 1634.2738,
      "train_tokens_per_second": 5963.485
    },
    {
      "epoch": 0.5958620689655172,
      "grad_norm": 0.6917933821678162,
      "learning_rate": 0.0001405853119823302,
      "loss": 1.8153,
      "num_input_tokens_seen": 9751280,
      "step": 1620,
      "train_runtime": 1635.1591,
      "train_tokens_per_second": 5963.505
    },
    {
      "epoch": 0.5962298850574712,
      "grad_norm": 0.6422238349914551,
      "learning_rate": 0.0001405484999079698,
      "loss": 1.6872,
      "num_input_tokens_seen": 9757744,
      "step": 1621,
      "train_runtime": 1636.7361,
      "train_tokens_per_second": 5961.709
    },
    {
      "epoch": 0.5965977011494253,
      "grad_norm": 0.7320050001144409,
      "learning_rate": 0.00014051168783360944,
      "loss": 1.7059,
      "num_input_tokens_seen": 9762208,
      "step": 1622,
      "train_runtime": 1637.4913,
      "train_tokens_per_second": 5961.685
    },
    {
      "epoch": 0.5969655172413794,
      "grad_norm": 0.7011889219284058,
      "learning_rate": 0.00014047487575924905,
      "loss": 1.7788,
      "num_input_tokens_seen": 9770480,
      "step": 1623,
      "train_runtime": 1638.7633,
      "train_tokens_per_second": 5962.106
    },
    {
      "epoch": 0.5973333333333334,
      "grad_norm": 0.763520359992981,
      "learning_rate": 0.00014043806368488866,
      "loss": 1.9754,
      "num_input_tokens_seen": 9777408,
      "step": 1624,
      "train_runtime": 1639.8716,
      "train_tokens_per_second": 5962.301
    },
    {
      "epoch": 0.5977011494252874,
      "grad_norm": 0.6897074580192566,
      "learning_rate": 0.00014040125161052827,
      "loss": 1.7607,
      "num_input_tokens_seen": 9783312,
      "step": 1625,
      "train_runtime": 1640.8268,
      "train_tokens_per_second": 5962.428
    },
    {
      "epoch": 0.5980689655172414,
      "grad_norm": 0.781053364276886,
      "learning_rate": 0.00014036443953616788,
      "loss": 1.7596,
      "num_input_tokens_seen": 9788368,
      "step": 1626,
      "train_runtime": 1641.6773,
      "train_tokens_per_second": 5962.419
    },
    {
      "epoch": 0.5984367816091954,
      "grad_norm": 0.8557214736938477,
      "learning_rate": 0.0001403276274618075,
      "loss": 1.7264,
      "num_input_tokens_seen": 9794640,
      "step": 1627,
      "train_runtime": 1642.6856,
      "train_tokens_per_second": 5962.577
    },
    {
      "epoch": 0.5988045977011495,
      "grad_norm": 0.8663838505744934,
      "learning_rate": 0.0001402908153874471,
      "loss": 1.6593,
      "num_input_tokens_seen": 9800224,
      "step": 1628,
      "train_runtime": 1643.6044,
      "train_tokens_per_second": 5962.642
    },
    {
      "epoch": 0.5991724137931035,
      "grad_norm": 0.7625656723976135,
      "learning_rate": 0.00014025400331308668,
      "loss": 1.7317,
      "num_input_tokens_seen": 9805936,
      "step": 1629,
      "train_runtime": 1644.5428,
      "train_tokens_per_second": 5962.713
    },
    {
      "epoch": 0.5995402298850575,
      "grad_norm": 0.7845171093940735,
      "learning_rate": 0.00014021719123872631,
      "loss": 1.9655,
      "num_input_tokens_seen": 9811808,
      "step": 1630,
      "train_runtime": 1645.5052,
      "train_tokens_per_second": 5962.794
    },
    {
      "epoch": 0.5999080459770115,
      "grad_norm": 0.7410557270050049,
      "learning_rate": 0.00014018037916436592,
      "loss": 1.8309,
      "num_input_tokens_seen": 9816880,
      "step": 1631,
      "train_runtime": 1646.3463,
      "train_tokens_per_second": 5962.828
    },
    {
      "epoch": 0.6002758620689655,
      "grad_norm": 0.7271944880485535,
      "learning_rate": 0.00014014356709000553,
      "loss": 1.7485,
      "num_input_tokens_seen": 9825328,
      "step": 1632,
      "train_runtime": 1647.6708,
      "train_tokens_per_second": 5963.162
    },
    {
      "epoch": 0.6002758620689655,
      "eval_loss": 1.798221468925476,
      "eval_runtime": 4.7872,
      "eval_samples_per_second": 208.892,
      "eval_steps_per_second": 13.16,
      "num_input_tokens_seen": 9825328,
      "step": 1632
    },
    {
      "epoch": 0.6006436781609196,
      "grad_norm": 0.7300437688827515,
      "learning_rate": 0.00014010675501564514,
      "loss": 1.609,
      "num_input_tokens_seen": 9832016,
      "step": 1633,
      "train_runtime": 1653.5168,
      "train_tokens_per_second": 5946.124
    },
    {
      "epoch": 0.6010114942528736,
      "grad_norm": 0.7482008934020996,
      "learning_rate": 0.00014006994294128475,
      "loss": 1.8593,
      "num_input_tokens_seen": 9838368,
      "step": 1634,
      "train_runtime": 1654.5654,
      "train_tokens_per_second": 5946.195
    },
    {
      "epoch": 0.6013793103448276,
      "grad_norm": 0.7498264908790588,
      "learning_rate": 0.00014003313086692436,
      "loss": 1.8512,
      "num_input_tokens_seen": 9848592,
      "step": 1635,
      "train_runtime": 1656.1239,
      "train_tokens_per_second": 5946.772
    },
    {
      "epoch": 0.6017471264367816,
      "grad_norm": 0.6734253764152527,
      "learning_rate": 0.00013999631879256397,
      "loss": 1.8363,
      "num_input_tokens_seen": 9856048,
      "step": 1636,
      "train_runtime": 1657.2965,
      "train_tokens_per_second": 5947.064
    },
    {
      "epoch": 0.6021149425287357,
      "grad_norm": 0.7074384093284607,
      "learning_rate": 0.00013995950671820358,
      "loss": 1.7518,
      "num_input_tokens_seen": 9862304,
      "step": 1637,
      "train_runtime": 1658.3,
      "train_tokens_per_second": 5947.237
    },
    {
      "epoch": 0.6024827586206897,
      "grad_norm": 0.6251190900802612,
      "learning_rate": 0.00013992269464384319,
      "loss": 1.6581,
      "num_input_tokens_seen": 9882384,
      "step": 1638,
      "train_runtime": 1661.1535,
      "train_tokens_per_second": 5949.11
    },
    {
      "epoch": 0.6028505747126437,
      "grad_norm": 0.7110159397125244,
      "learning_rate": 0.0001398858825694828,
      "loss": 1.9105,
      "num_input_tokens_seen": 9887760,
      "step": 1639,
      "train_runtime": 1662.0336,
      "train_tokens_per_second": 5949.194
    },
    {
      "epoch": 0.6032183908045977,
      "grad_norm": 0.8627580404281616,
      "learning_rate": 0.0001398490704951224,
      "loss": 1.6832,
      "num_input_tokens_seen": 9893584,
      "step": 1640,
      "train_runtime": 1662.9878,
      "train_tokens_per_second": 5949.282
    },
    {
      "epoch": 0.6035862068965517,
      "grad_norm": 0.7155346274375916,
      "learning_rate": 0.000139812258420762,
      "loss": 1.7047,
      "num_input_tokens_seen": 9898928,
      "step": 1641,
      "train_runtime": 1663.8888,
      "train_tokens_per_second": 5949.273
    },
    {
      "epoch": 0.6039540229885058,
      "grad_norm": 0.8029899001121521,
      "learning_rate": 0.00013977544634640162,
      "loss": 2.0038,
      "num_input_tokens_seen": 9903856,
      "step": 1642,
      "train_runtime": 1664.7325,
      "train_tokens_per_second": 5949.218
    },
    {
      "epoch": 0.6043218390804598,
      "grad_norm": 0.7011070847511292,
      "learning_rate": 0.00013973863427204123,
      "loss": 1.7154,
      "num_input_tokens_seen": 9909568,
      "step": 1643,
      "train_runtime": 1665.6688,
      "train_tokens_per_second": 5949.303
    },
    {
      "epoch": 0.6046896551724138,
      "grad_norm": 0.8248561024665833,
      "learning_rate": 0.00013970182219768084,
      "loss": 1.9599,
      "num_input_tokens_seen": 9916320,
      "step": 1644,
      "train_runtime": 1666.7481,
      "train_tokens_per_second": 5949.501
    },
    {
      "epoch": 0.6050574712643678,
      "grad_norm": 0.692307710647583,
      "learning_rate": 0.00013966501012332045,
      "loss": 1.7467,
      "num_input_tokens_seen": 9922192,
      "step": 1645,
      "train_runtime": 1667.7328,
      "train_tokens_per_second": 5949.51
    },
    {
      "epoch": 0.6054252873563218,
      "grad_norm": 0.7132775783538818,
      "learning_rate": 0.00013962819804896006,
      "loss": 1.736,
      "num_input_tokens_seen": 9935904,
      "step": 1646,
      "train_runtime": 1669.7171,
      "train_tokens_per_second": 5950.651
    },
    {
      "epoch": 0.6057931034482759,
      "grad_norm": 0.7879961133003235,
      "learning_rate": 0.00013959138597459967,
      "loss": 1.8888,
      "num_input_tokens_seen": 9941536,
      "step": 1647,
      "train_runtime": 1670.6597,
      "train_tokens_per_second": 5950.665
    },
    {
      "epoch": 0.6061609195402299,
      "grad_norm": 0.8213357329368591,
      "learning_rate": 0.0001395545739002393,
      "loss": 1.8003,
      "num_input_tokens_seen": 9946304,
      "step": 1648,
      "train_runtime": 1671.4905,
      "train_tokens_per_second": 5950.56
    },
    {
      "epoch": 0.6065287356321839,
      "grad_norm": 0.7963126301765442,
      "learning_rate": 0.0001395177618258789,
      "loss": 1.7383,
      "num_input_tokens_seen": 9951376,
      "step": 1649,
      "train_runtime": 1672.3473,
      "train_tokens_per_second": 5950.544
    },
    {
      "epoch": 0.6068965517241379,
      "grad_norm": 0.7608811855316162,
      "learning_rate": 0.00013948094975151852,
      "loss": 1.9336,
      "num_input_tokens_seen": 9958048,
      "step": 1650,
      "train_runtime": 1673.4209,
      "train_tokens_per_second": 5950.713
    },
    {
      "epoch": 0.6072643678160919,
      "grad_norm": 0.8075297474861145,
      "learning_rate": 0.00013944413767715813,
      "loss": 1.8932,
      "num_input_tokens_seen": 9962608,
      "step": 1651,
      "train_runtime": 1674.7239,
      "train_tokens_per_second": 5948.806
    },
    {
      "epoch": 0.607632183908046,
      "grad_norm": 0.7269268035888672,
      "learning_rate": 0.0001394073256027977,
      "loss": 1.7587,
      "num_input_tokens_seen": 9969648,
      "step": 1652,
      "train_runtime": 1675.8584,
      "train_tokens_per_second": 5948.98
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.7436178922653198,
      "learning_rate": 0.00013937051352843732,
      "loss": 1.9583,
      "num_input_tokens_seen": 9974880,
      "step": 1653,
      "train_runtime": 1676.7301,
      "train_tokens_per_second": 5949.008
    },
    {
      "epoch": 0.608367816091954,
      "grad_norm": 0.6937170028686523,
      "learning_rate": 0.00013933370145407693,
      "loss": 1.736,
      "num_input_tokens_seen": 9980448,
      "step": 1654,
      "train_runtime": 1677.6689,
      "train_tokens_per_second": 5948.998
    },
    {
      "epoch": 0.608735632183908,
      "grad_norm": 0.7714893817901611,
      "learning_rate": 0.00013929688937971654,
      "loss": 1.652,
      "num_input_tokens_seen": 9984864,
      "step": 1655,
      "train_runtime": 1678.4509,
      "train_tokens_per_second": 5948.857
    },
    {
      "epoch": 0.609103448275862,
      "grad_norm": 0.7302749156951904,
      "learning_rate": 0.00013926007730535617,
      "loss": 1.8014,
      "num_input_tokens_seen": 9990048,
      "step": 1656,
      "train_runtime": 1679.3369,
      "train_tokens_per_second": 5948.805
    },
    {
      "epoch": 0.6094712643678161,
      "grad_norm": 0.7291936278343201,
      "learning_rate": 0.00013922326523099578,
      "loss": 1.7244,
      "num_input_tokens_seen": 10000960,
      "step": 1657,
      "train_runtime": 1680.9856,
      "train_tokens_per_second": 5949.462
    },
    {
      "epoch": 0.6098390804597701,
      "grad_norm": 0.7018803954124451,
      "learning_rate": 0.0001391864531566354,
      "loss": 1.807,
      "num_input_tokens_seen": 10006240,
      "step": 1658,
      "train_runtime": 1681.8686,
      "train_tokens_per_second": 5949.478
    },
    {
      "epoch": 0.6102068965517241,
      "grad_norm": 0.7114816904067993,
      "learning_rate": 0.000139149641082275,
      "loss": 1.6622,
      "num_input_tokens_seen": 10012336,
      "step": 1659,
      "train_runtime": 1682.859,
      "train_tokens_per_second": 5949.599
    },
    {
      "epoch": 0.6105747126436781,
      "grad_norm": 0.7551282644271851,
      "learning_rate": 0.0001391128290079146,
      "loss": 1.8038,
      "num_input_tokens_seen": 10017248,
      "step": 1660,
      "train_runtime": 1683.6945,
      "train_tokens_per_second": 5949.564
    },
    {
      "epoch": 0.6109425287356322,
      "grad_norm": 0.7015314102172852,
      "learning_rate": 0.00013907601693355422,
      "loss": 1.7449,
      "num_input_tokens_seen": 10024160,
      "step": 1661,
      "train_runtime": 1684.8099,
      "train_tokens_per_second": 5949.728
    },
    {
      "epoch": 0.6113103448275862,
      "grad_norm": 0.709729790687561,
      "learning_rate": 0.00013903920485919383,
      "loss": 1.7064,
      "num_input_tokens_seen": 10029104,
      "step": 1662,
      "train_runtime": 1685.6684,
      "train_tokens_per_second": 5949.631
    },
    {
      "epoch": 0.6116781609195402,
      "grad_norm": 0.6485025882720947,
      "learning_rate": 0.00013900239278483343,
      "loss": 1.7677,
      "num_input_tokens_seen": 10036672,
      "step": 1663,
      "train_runtime": 1686.8448,
      "train_tokens_per_second": 5949.968
    },
    {
      "epoch": 0.6120459770114942,
      "grad_norm": 0.7688760757446289,
      "learning_rate": 0.00013896558071047304,
      "loss": 1.7656,
      "num_input_tokens_seen": 10041584,
      "step": 1664,
      "train_runtime": 1687.697,
      "train_tokens_per_second": 5949.874
    },
    {
      "epoch": 0.6124137931034482,
      "grad_norm": 0.7552171945571899,
      "learning_rate": 0.00013892876863611265,
      "loss": 1.8779,
      "num_input_tokens_seen": 10048672,
      "step": 1665,
      "train_runtime": 1688.8181,
      "train_tokens_per_second": 5950.121
    },
    {
      "epoch": 0.6127816091954023,
      "grad_norm": 0.770167350769043,
      "learning_rate": 0.00013889195656175226,
      "loss": 1.7685,
      "num_input_tokens_seen": 10054176,
      "step": 1666,
      "train_runtime": 1689.7469,
      "train_tokens_per_second": 5950.108
    },
    {
      "epoch": 0.6131494252873563,
      "grad_norm": 0.6745887398719788,
      "learning_rate": 0.00013885514448739187,
      "loss": 1.6688,
      "num_input_tokens_seen": 10061808,
      "step": 1667,
      "train_runtime": 1690.9469,
      "train_tokens_per_second": 5950.399
    },
    {
      "epoch": 0.6135172413793103,
      "grad_norm": 0.757714569568634,
      "learning_rate": 0.00013881833241303148,
      "loss": 1.7799,
      "num_input_tokens_seen": 10066816,
      "step": 1668,
      "train_runtime": 1691.7745,
      "train_tokens_per_second": 5950.448
    },
    {
      "epoch": 0.6138850574712643,
      "grad_norm": 0.723790168762207,
      "learning_rate": 0.0001387815203386711,
      "loss": 1.781,
      "num_input_tokens_seen": 10072032,
      "step": 1669,
      "train_runtime": 1692.6537,
      "train_tokens_per_second": 5950.439
    },
    {
      "epoch": 0.6142528735632183,
      "grad_norm": 0.8631724715232849,
      "learning_rate": 0.0001387447082643107,
      "loss": 1.8296,
      "num_input_tokens_seen": 10077136,
      "step": 1670,
      "train_runtime": 1693.4843,
      "train_tokens_per_second": 5950.534
    },
    {
      "epoch": 0.6146206896551724,
      "grad_norm": 0.6833540201187134,
      "learning_rate": 0.0001387078961899503,
      "loss": 1.7611,
      "num_input_tokens_seen": 10084496,
      "step": 1671,
      "train_runtime": 1694.6263,
      "train_tokens_per_second": 5950.867
    },
    {
      "epoch": 0.6149885057471265,
      "grad_norm": 0.7217798829078674,
      "learning_rate": 0.00013867108411558991,
      "loss": 1.6531,
      "num_input_tokens_seen": 10089664,
      "step": 1672,
      "train_runtime": 1695.4952,
      "train_tokens_per_second": 5950.866
    },
    {
      "epoch": 0.6153563218390805,
      "grad_norm": 0.7099449634552002,
      "learning_rate": 0.00013863427204122955,
      "loss": 1.6828,
      "num_input_tokens_seen": 10094672,
      "step": 1673,
      "train_runtime": 1696.3242,
      "train_tokens_per_second": 5950.909
    },
    {
      "epoch": 0.6157241379310345,
      "grad_norm": 0.746695339679718,
      "learning_rate": 0.00013859745996686916,
      "loss": 1.9705,
      "num_input_tokens_seen": 10100064,
      "step": 1674,
      "train_runtime": 1697.2443,
      "train_tokens_per_second": 5950.861
    },
    {
      "epoch": 0.6160919540229886,
      "grad_norm": 0.8037787079811096,
      "learning_rate": 0.00013856064789250874,
      "loss": 1.9572,
      "num_input_tokens_seen": 10104512,
      "step": 1675,
      "train_runtime": 1698.0032,
      "train_tokens_per_second": 5950.82
    },
    {
      "epoch": 0.6164597701149426,
      "grad_norm": 0.7081107497215271,
      "learning_rate": 0.00013852383581814835,
      "loss": 1.5542,
      "num_input_tokens_seen": 10114976,
      "step": 1676,
      "train_runtime": 1699.5581,
      "train_tokens_per_second": 5951.533
    },
    {
      "epoch": 0.6168275862068966,
      "grad_norm": 0.7685478329658508,
      "learning_rate": 0.00013848702374378796,
      "loss": 1.9161,
      "num_input_tokens_seen": 10119472,
      "step": 1677,
      "train_runtime": 1700.3736,
      "train_tokens_per_second": 5951.323
    },
    {
      "epoch": 0.6171954022988506,
      "grad_norm": 0.8034707307815552,
      "learning_rate": 0.00013845021166942757,
      "loss": 1.8704,
      "num_input_tokens_seen": 10125040,
      "step": 1678,
      "train_runtime": 1701.285,
      "train_tokens_per_second": 5951.407
    },
    {
      "epoch": 0.6175632183908046,
      "grad_norm": 0.6806014180183411,
      "learning_rate": 0.00013841339959506718,
      "loss": 1.7043,
      "num_input_tokens_seen": 10132832,
      "step": 1679,
      "train_runtime": 1702.4977,
      "train_tokens_per_second": 5951.745
    },
    {
      "epoch": 0.6179310344827587,
      "grad_norm": 0.7518782615661621,
      "learning_rate": 0.00013837658752070679,
      "loss": 1.8486,
      "num_input_tokens_seen": 10138928,
      "step": 1680,
      "train_runtime": 1703.4918,
      "train_tokens_per_second": 5951.85
    },
    {
      "epoch": 0.6182988505747127,
      "grad_norm": 0.7308506369590759,
      "learning_rate": 0.0001383397754463464,
      "loss": 1.8203,
      "num_input_tokens_seen": 10145648,
      "step": 1681,
      "train_runtime": 1705.0308,
      "train_tokens_per_second": 5950.419
    },
    {
      "epoch": 0.6186666666666667,
      "grad_norm": 0.830508828163147,
      "learning_rate": 0.00013830296337198603,
      "loss": 1.6679,
      "num_input_tokens_seen": 10150160,
      "step": 1682,
      "train_runtime": 1705.8125,
      "train_tokens_per_second": 5950.338
    },
    {
      "epoch": 0.6190344827586207,
      "grad_norm": 0.8149420022964478,
      "learning_rate": 0.00013826615129762564,
      "loss": 1.8565,
      "num_input_tokens_seen": 10155296,
      "step": 1683,
      "train_runtime": 1706.6886,
      "train_tokens_per_second": 5950.292
    },
    {
      "epoch": 0.6194022988505747,
      "grad_norm": 0.7269219160079956,
      "learning_rate": 0.00013822933922326525,
      "loss": 1.7568,
      "num_input_tokens_seen": 10160512,
      "step": 1684,
      "train_runtime": 1707.5746,
      "train_tokens_per_second": 5950.26
    },
    {
      "epoch": 0.6197701149425288,
      "grad_norm": 0.7423215508460999,
      "learning_rate": 0.00013819252714890486,
      "loss": 1.7513,
      "num_input_tokens_seen": 10165344,
      "step": 1685,
      "train_runtime": 1708.4175,
      "train_tokens_per_second": 5950.152
    },
    {
      "epoch": 0.6201379310344828,
      "grad_norm": 0.8427226543426514,
      "learning_rate": 0.00013815571507454447,
      "loss": 1.9456,
      "num_input_tokens_seen": 10170464,
      "step": 1686,
      "train_runtime": 1709.2529,
      "train_tokens_per_second": 5950.239
    },
    {
      "epoch": 0.6205057471264368,
      "grad_norm": 0.7710717916488647,
      "learning_rate": 0.00013811890300018408,
      "loss": 1.7877,
      "num_input_tokens_seen": 10174880,
      "step": 1687,
      "train_runtime": 1710.0111,
      "train_tokens_per_second": 5950.184
    },
    {
      "epoch": 0.6208735632183908,
      "grad_norm": 0.7333926558494568,
      "learning_rate": 0.00013808209092582366,
      "loss": 1.8212,
      "num_input_tokens_seen": 10180720,
      "step": 1688,
      "train_runtime": 1710.9644,
      "train_tokens_per_second": 5950.281
    },
    {
      "epoch": 0.6212413793103448,
      "grad_norm": 0.749277651309967,
      "learning_rate": 0.00013804527885146327,
      "loss": 1.5334,
      "num_input_tokens_seen": 10190208,
      "step": 1689,
      "train_runtime": 1712.4203,
      "train_tokens_per_second": 5950.763
    },
    {
      "epoch": 0.6216091954022989,
      "grad_norm": 0.7796466946601868,
      "learning_rate": 0.0001380084667771029,
      "loss": 1.6282,
      "num_input_tokens_seen": 10196304,
      "step": 1690,
      "train_runtime": 1713.4181,
      "train_tokens_per_second": 5950.856
    },
    {
      "epoch": 0.6219770114942529,
      "grad_norm": 0.8130245208740234,
      "learning_rate": 0.0001379716547027425,
      "loss": 1.7571,
      "num_input_tokens_seen": 10201280,
      "step": 1691,
      "train_runtime": 1714.253,
      "train_tokens_per_second": 5950.86
    },
    {
      "epoch": 0.6223448275862069,
      "grad_norm": 0.7046164274215698,
      "learning_rate": 0.00013793484262838212,
      "loss": 1.6075,
      "num_input_tokens_seen": 10206544,
      "step": 1692,
      "train_runtime": 1715.0916,
      "train_tokens_per_second": 5951.02
    },
    {
      "epoch": 0.6227126436781609,
      "grad_norm": 0.7137652039527893,
      "learning_rate": 0.00013789803055402173,
      "loss": 1.7459,
      "num_input_tokens_seen": 10211584,
      "step": 1693,
      "train_runtime": 1715.9287,
      "train_tokens_per_second": 5951.054
    },
    {
      "epoch": 0.623080459770115,
      "grad_norm": 0.7368882894515991,
      "learning_rate": 0.00013786121847966134,
      "loss": 1.7421,
      "num_input_tokens_seen": 10215968,
      "step": 1694,
      "train_runtime": 1716.7015,
      "train_tokens_per_second": 5950.929
    },
    {
      "epoch": 0.623448275862069,
      "grad_norm": 0.7959756851196289,
      "learning_rate": 0.00013782440640530095,
      "loss": 2.035,
      "num_input_tokens_seen": 10221456,
      "step": 1695,
      "train_runtime": 1717.6204,
      "train_tokens_per_second": 5950.94
    },
    {
      "epoch": 0.623816091954023,
      "grad_norm": 0.8156048059463501,
      "learning_rate": 0.00013778759433094056,
      "loss": 1.7757,
      "num_input_tokens_seen": 10229632,
      "step": 1696,
      "train_runtime": 1718.8915,
      "train_tokens_per_second": 5951.296
    },
    {
      "epoch": 0.624183908045977,
      "grad_norm": 0.7234564423561096,
      "learning_rate": 0.00013775078225658016,
      "loss": 1.6543,
      "num_input_tokens_seen": 10235488,
      "step": 1697,
      "train_runtime": 1719.8438,
      "train_tokens_per_second": 5951.406
    },
    {
      "epoch": 0.624551724137931,
      "grad_norm": 0.7619455456733704,
      "learning_rate": 0.00013771397018221977,
      "loss": 1.648,
      "num_input_tokens_seen": 10243344,
      "step": 1698,
      "train_runtime": 1721.0519,
      "train_tokens_per_second": 5951.793
    },
    {
      "epoch": 0.624919540229885,
      "grad_norm": 0.8125543594360352,
      "learning_rate": 0.00013767715810785938,
      "loss": 1.99,
      "num_input_tokens_seen": 10248144,
      "step": 1699,
      "train_runtime": 1721.8566,
      "train_tokens_per_second": 5951.799
    },
    {
      "epoch": 0.6252873563218391,
      "grad_norm": 0.7408743500709534,
      "learning_rate": 0.000137640346033499,
      "loss": 1.6796,
      "num_input_tokens_seen": 10252880,
      "step": 1700,
      "train_runtime": 1722.6699,
      "train_tokens_per_second": 5951.738
    },
    {
      "epoch": 0.6256551724137931,
      "grad_norm": 0.8177375793457031,
      "learning_rate": 0.0001376035339591386,
      "loss": 1.7792,
      "num_input_tokens_seen": 10257168,
      "step": 1701,
      "train_runtime": 1723.411,
      "train_tokens_per_second": 5951.667
    },
    {
      "epoch": 0.6260229885057471,
      "grad_norm": 0.7161668539047241,
      "learning_rate": 0.0001375667218847782,
      "loss": 1.6625,
      "num_input_tokens_seen": 10265856,
      "step": 1702,
      "train_runtime": 1724.7605,
      "train_tokens_per_second": 5952.047
    },
    {
      "epoch": 0.6263908045977011,
      "grad_norm": 0.8035367727279663,
      "learning_rate": 0.00013752990981041782,
      "loss": 1.8391,
      "num_input_tokens_seen": 10271216,
      "step": 1703,
      "train_runtime": 1725.6463,
      "train_tokens_per_second": 5952.098
    },
    {
      "epoch": 0.6267586206896552,
      "grad_norm": 0.7855319976806641,
      "learning_rate": 0.00013749309773605743,
      "loss": 1.801,
      "num_input_tokens_seen": 10277328,
      "step": 1704,
      "train_runtime": 1726.6384,
      "train_tokens_per_second": 5952.218
    },
    {
      "epoch": 0.6271264367816092,
      "grad_norm": 0.7613573670387268,
      "learning_rate": 0.00013745628566169704,
      "loss": 1.7867,
      "num_input_tokens_seen": 10284176,
      "step": 1705,
      "train_runtime": 1727.7258,
      "train_tokens_per_second": 5952.435
    },
    {
      "epoch": 0.6274942528735632,
      "grad_norm": 0.6075987219810486,
      "learning_rate": 0.00013741947358733664,
      "loss": 1.5675,
      "num_input_tokens_seen": 10294688,
      "step": 1706,
      "train_runtime": 1729.2841,
      "train_tokens_per_second": 5953.15
    },
    {
      "epoch": 0.6278620689655172,
      "grad_norm": 0.817808985710144,
      "learning_rate": 0.00013738266151297625,
      "loss": 1.8666,
      "num_input_tokens_seen": 10299344,
      "step": 1707,
      "train_runtime": 1730.1032,
      "train_tokens_per_second": 5953.023
    },
    {
      "epoch": 0.6282298850574712,
      "grad_norm": 0.7143842577934265,
      "learning_rate": 0.0001373458494386159,
      "loss": 1.6343,
      "num_input_tokens_seen": 10305600,
      "step": 1708,
      "train_runtime": 1731.1297,
      "train_tokens_per_second": 5953.107
    },
    {
      "epoch": 0.6285977011494253,
      "grad_norm": 0.7010509967803955,
      "learning_rate": 0.0001373090373642555,
      "loss": 1.794,
      "num_input_tokens_seen": 10310512,
      "step": 1709,
      "train_runtime": 1731.968,
      "train_tokens_per_second": 5953.061
    },
    {
      "epoch": 0.6289655172413793,
      "grad_norm": 0.6494395136833191,
      "learning_rate": 0.0001372722252898951,
      "loss": 1.5411,
      "num_input_tokens_seen": 10316128,
      "step": 1710,
      "train_runtime": 1732.884,
      "train_tokens_per_second": 5953.156
    },
    {
      "epoch": 0.6293333333333333,
      "grad_norm": 0.7045603394508362,
      "learning_rate": 0.00013723541321553472,
      "loss": 1.6267,
      "num_input_tokens_seen": 10321728,
      "step": 1711,
      "train_runtime": 1734.3244,
      "train_tokens_per_second": 5951.44
    },
    {
      "epoch": 0.6297011494252873,
      "grad_norm": 0.7397974133491516,
      "learning_rate": 0.0001371986011411743,
      "loss": 1.7847,
      "num_input_tokens_seen": 10328672,
      "step": 1712,
      "train_runtime": 1735.413,
      "train_tokens_per_second": 5951.708
    },
    {
      "epoch": 0.6300689655172413,
      "grad_norm": 0.7169364094734192,
      "learning_rate": 0.0001371617890668139,
      "loss": 1.7564,
      "num_input_tokens_seen": 10334720,
      "step": 1713,
      "train_runtime": 1736.3962,
      "train_tokens_per_second": 5951.821
    },
    {
      "epoch": 0.6304367816091954,
      "grad_norm": 0.7214260697364807,
      "learning_rate": 0.00013712497699245352,
      "loss": 1.7275,
      "num_input_tokens_seen": 10339904,
      "step": 1714,
      "train_runtime": 1737.2741,
      "train_tokens_per_second": 5951.798
    },
    {
      "epoch": 0.6308045977011494,
      "grad_norm": 0.6682936549186707,
      "learning_rate": 0.00013708816491809312,
      "loss": 1.7087,
      "num_input_tokens_seen": 10346256,
      "step": 1715,
      "train_runtime": 1738.3112,
      "train_tokens_per_second": 5951.901
    },
    {
      "epoch": 0.6311724137931034,
      "grad_norm": 0.7426211833953857,
      "learning_rate": 0.00013705135284373276,
      "loss": 1.7421,
      "num_input_tokens_seen": 10352192,
      "step": 1716,
      "train_runtime": 1739.2825,
      "train_tokens_per_second": 5951.99
    },
    {
      "epoch": 0.6315402298850574,
      "grad_norm": 0.6931152939796448,
      "learning_rate": 0.00013701454076937237,
      "loss": 1.6883,
      "num_input_tokens_seen": 10356544,
      "step": 1717,
      "train_runtime": 1740.0553,
      "train_tokens_per_second": 5951.848
    },
    {
      "epoch": 0.6319080459770114,
      "grad_norm": 0.7280430793762207,
      "learning_rate": 0.00013697772869501198,
      "loss": 1.8632,
      "num_input_tokens_seen": 10363024,
      "step": 1718,
      "train_runtime": 1741.0983,
      "train_tokens_per_second": 5952.004
    },
    {
      "epoch": 0.6322758620689655,
      "grad_norm": 0.6989037394523621,
      "learning_rate": 0.0001369409166206516,
      "loss": 1.7903,
      "num_input_tokens_seen": 10368976,
      "step": 1719,
      "train_runtime": 1742.0536,
      "train_tokens_per_second": 5952.157
    },
    {
      "epoch": 0.6326436781609195,
      "grad_norm": 0.8285207152366638,
      "learning_rate": 0.0001369041045462912,
      "loss": 1.876,
      "num_input_tokens_seen": 10373504,
      "step": 1720,
      "train_runtime": 1742.8326,
      "train_tokens_per_second": 5952.094
    },
    {
      "epoch": 0.6330114942528735,
      "grad_norm": 0.7388542294502258,
      "learning_rate": 0.0001368672924719308,
      "loss": 1.8336,
      "num_input_tokens_seen": 10378272,
      "step": 1721,
      "train_runtime": 1743.6549,
      "train_tokens_per_second": 5952.022
    },
    {
      "epoch": 0.6333793103448276,
      "grad_norm": 0.7335883378982544,
      "learning_rate": 0.00013683048039757041,
      "loss": 1.7633,
      "num_input_tokens_seen": 10384320,
      "step": 1722,
      "train_runtime": 1744.6271,
      "train_tokens_per_second": 5952.172
    },
    {
      "epoch": 0.6337471264367817,
      "grad_norm": 0.7219555377960205,
      "learning_rate": 0.00013679366832321002,
      "loss": 1.9432,
      "num_input_tokens_seen": 10388832,
      "step": 1723,
      "train_runtime": 1745.4345,
      "train_tokens_per_second": 5952.003
    },
    {
      "epoch": 0.6341149425287357,
      "grad_norm": 0.7718085646629333,
      "learning_rate": 0.00013675685624884963,
      "loss": 1.6726,
      "num_input_tokens_seen": 10394112,
      "step": 1724,
      "train_runtime": 1746.3262,
      "train_tokens_per_second": 5951.988
    },
    {
      "epoch": 0.6344827586206897,
      "grad_norm": 0.7376514077186584,
      "learning_rate": 0.00013672004417448924,
      "loss": 1.849,
      "num_input_tokens_seen": 10401728,
      "step": 1725,
      "train_runtime": 1747.5016,
      "train_tokens_per_second": 5952.342
    },
    {
      "epoch": 0.6348505747126437,
      "grad_norm": 0.753606379032135,
      "learning_rate": 0.00013668323210012885,
      "loss": 1.5803,
      "num_input_tokens_seen": 10412208,
      "step": 1726,
      "train_runtime": 1749.0547,
      "train_tokens_per_second": 5953.049
    },
    {
      "epoch": 0.6352183908045977,
      "grad_norm": 0.8534219264984131,
      "learning_rate": 0.00013664642002576846,
      "loss": 1.8085,
      "num_input_tokens_seen": 10419376,
      "step": 1727,
      "train_runtime": 1750.2152,
      "train_tokens_per_second": 5953.197
    },
    {
      "epoch": 0.6355862068965518,
      "grad_norm": 0.6987608671188354,
      "learning_rate": 0.00013660960795140807,
      "loss": 1.749,
      "num_input_tokens_seen": 10427104,
      "step": 1728,
      "train_runtime": 1751.4127,
      "train_tokens_per_second": 5953.539
    },
    {
      "epoch": 0.6359540229885058,
      "grad_norm": 0.8042802810668945,
      "learning_rate": 0.00013657279587704768,
      "loss": 1.7879,
      "num_input_tokens_seen": 10434736,
      "step": 1729,
      "train_runtime": 1752.6227,
      "train_tokens_per_second": 5953.783
    },
    {
      "epoch": 0.6363218390804598,
      "grad_norm": 0.6358485817909241,
      "learning_rate": 0.00013653598380268728,
      "loss": 1.5264,
      "num_input_tokens_seen": 10444912,
      "step": 1730,
      "train_runtime": 1754.1559,
      "train_tokens_per_second": 5954.381
    },
    {
      "epoch": 0.6366896551724138,
      "grad_norm": 0.7460195422172546,
      "learning_rate": 0.0001364991717283269,
      "loss": 1.7442,
      "num_input_tokens_seen": 10450368,
      "step": 1731,
      "train_runtime": 1755.0487,
      "train_tokens_per_second": 5954.46
    },
    {
      "epoch": 0.6370574712643678,
      "grad_norm": 0.7388708591461182,
      "learning_rate": 0.0001364623596539665,
      "loss": 1.7638,
      "num_input_tokens_seen": 10455792,
      "step": 1732,
      "train_runtime": 1755.9778,
      "train_tokens_per_second": 5954.399
    },
    {
      "epoch": 0.6374252873563219,
      "grad_norm": 0.6434199810028076,
      "learning_rate": 0.00013642554757960614,
      "loss": 1.6617,
      "num_input_tokens_seen": 10465408,
      "step": 1733,
      "train_runtime": 1757.4856,
      "train_tokens_per_second": 5954.762
    },
    {
      "epoch": 0.6377931034482759,
      "grad_norm": 0.6927239894866943,
      "learning_rate": 0.00013638873550524575,
      "loss": 1.8211,
      "num_input_tokens_seen": 10472256,
      "step": 1734,
      "train_runtime": 1758.5554,
      "train_tokens_per_second": 5955.033
    },
    {
      "epoch": 0.6381609195402299,
      "grad_norm": 0.7508346438407898,
      "learning_rate": 0.00013635192343088533,
      "loss": 1.5389,
      "num_input_tokens_seen": 10476848,
      "step": 1735,
      "train_runtime": 1759.3473,
      "train_tokens_per_second": 5954.963
    },
    {
      "epoch": 0.6385287356321839,
      "grad_norm": 0.6954201459884644,
      "learning_rate": 0.00013631511135652494,
      "loss": 1.749,
      "num_input_tokens_seen": 10482048,
      "step": 1736,
      "train_runtime": 1760.2107,
      "train_tokens_per_second": 5954.996
    },
    {
      "epoch": 0.638896551724138,
      "grad_norm": 0.7972614169120789,
      "learning_rate": 0.00013627829928216455,
      "loss": 1.9329,
      "num_input_tokens_seen": 10487616,
      "step": 1737,
      "train_runtime": 1761.129,
      "train_tokens_per_second": 5955.053
    },
    {
      "epoch": 0.639264367816092,
      "grad_norm": 0.727857768535614,
      "learning_rate": 0.00013624148720780416,
      "loss": 1.7875,
      "num_input_tokens_seen": 10499600,
      "step": 1738,
      "train_runtime": 1762.8783,
      "train_tokens_per_second": 5955.942
    },
    {
      "epoch": 0.639632183908046,
      "grad_norm": 0.788093626499176,
      "learning_rate": 0.00013620467513344376,
      "loss": 1.8738,
      "num_input_tokens_seen": 10503808,
      "step": 1739,
      "train_runtime": 1763.6169,
      "train_tokens_per_second": 5955.833
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.746026873588562,
      "learning_rate": 0.00013616786305908337,
      "loss": 1.5872,
      "num_input_tokens_seen": 10509504,
      "step": 1740,
      "train_runtime": 1764.5378,
      "train_tokens_per_second": 5955.953
    },
    {
      "epoch": 0.640367816091954,
      "grad_norm": 0.8277755379676819,
      "learning_rate": 0.00013613105098472298,
      "loss": 1.9395,
      "num_input_tokens_seen": 10514608,
      "step": 1741,
      "train_runtime": 1765.9185,
      "train_tokens_per_second": 5954.186
    },
    {
      "epoch": 0.640735632183908,
      "grad_norm": 0.8165010809898376,
      "learning_rate": 0.00013609423891036262,
      "loss": 1.8424,
      "num_input_tokens_seen": 10520272,
      "step": 1742,
      "train_runtime": 1766.8628,
      "train_tokens_per_second": 5954.21
    },
    {
      "epoch": 0.6411034482758621,
      "grad_norm": 0.7696282267570496,
      "learning_rate": 0.00013605742683600223,
      "loss": 1.8617,
      "num_input_tokens_seen": 10525856,
      "step": 1743,
      "train_runtime": 1767.7774,
      "train_tokens_per_second": 5954.288
    },
    {
      "epoch": 0.6414712643678161,
      "grad_norm": 0.7378665804862976,
      "learning_rate": 0.00013602061476164184,
      "loss": 1.7307,
      "num_input_tokens_seen": 10531584,
      "step": 1744,
      "train_runtime": 1768.7411,
      "train_tokens_per_second": 5954.283
    },
    {
      "epoch": 0.6418390804597701,
      "grad_norm": 0.7567402720451355,
      "learning_rate": 0.00013598380268728145,
      "loss": 1.7956,
      "num_input_tokens_seen": 10540384,
      "step": 1745,
      "train_runtime": 1770.0833,
      "train_tokens_per_second": 5954.739
    },
    {
      "epoch": 0.6422068965517241,
      "grad_norm": 0.8224642276763916,
      "learning_rate": 0.00013594699061292105,
      "loss": 1.819,
      "num_input_tokens_seen": 10545552,
      "step": 1746,
      "train_runtime": 1770.9408,
      "train_tokens_per_second": 5954.774
    },
    {
      "epoch": 0.6425747126436782,
      "grad_norm": 0.9020531177520752,
      "learning_rate": 0.00013591017853856066,
      "loss": 1.9135,
      "num_input_tokens_seen": 10550592,
      "step": 1747,
      "train_runtime": 1771.8163,
      "train_tokens_per_second": 5954.676
    },
    {
      "epoch": 0.6429425287356322,
      "grad_norm": 0.7469763159751892,
      "learning_rate": 0.00013587336646420024,
      "loss": 1.9463,
      "num_input_tokens_seen": 10556592,
      "step": 1748,
      "train_runtime": 1772.8111,
      "train_tokens_per_second": 5954.719
    },
    {
      "epoch": 0.6433103448275862,
      "grad_norm": 0.7308046221733093,
      "learning_rate": 0.00013583655438983985,
      "loss": 1.9843,
      "num_input_tokens_seen": 10561792,
      "step": 1749,
      "train_runtime": 1773.7013,
      "train_tokens_per_second": 5954.662
    },
    {
      "epoch": 0.6436781609195402,
      "grad_norm": 0.7411881685256958,
      "learning_rate": 0.0001357997423154795,
      "loss": 1.9861,
      "num_input_tokens_seen": 10566144,
      "step": 1750,
      "train_runtime": 1774.4737,
      "train_tokens_per_second": 5954.523
    },
    {
      "epoch": 0.6440459770114942,
      "grad_norm": 0.7587552070617676,
      "learning_rate": 0.0001357629302411191,
      "loss": 1.9987,
      "num_input_tokens_seen": 10571184,
      "step": 1751,
      "train_runtime": 1775.333,
      "train_tokens_per_second": 5954.48
    },
    {
      "epoch": 0.6444137931034483,
      "grad_norm": 0.7099258303642273,
      "learning_rate": 0.0001357261181667587,
      "loss": 1.7047,
      "num_input_tokens_seen": 10576848,
      "step": 1752,
      "train_runtime": 1776.2653,
      "train_tokens_per_second": 5954.543
    },
    {
      "epoch": 0.6447816091954023,
      "grad_norm": 0.7215557098388672,
      "learning_rate": 0.00013568930609239832,
      "loss": 1.7609,
      "num_input_tokens_seen": 10581920,
      "step": 1753,
      "train_runtime": 1777.142,
      "train_tokens_per_second": 5954.459
    },
    {
      "epoch": 0.6451494252873563,
      "grad_norm": 0.7627148628234863,
      "learning_rate": 0.00013565249401803793,
      "loss": 1.8373,
      "num_input_tokens_seen": 10586704,
      "step": 1754,
      "train_runtime": 1777.9699,
      "train_tokens_per_second": 5954.377
    },
    {
      "epoch": 0.6455172413793103,
      "grad_norm": 0.6375512480735779,
      "learning_rate": 0.00013561568194367753,
      "loss": 1.6086,
      "num_input_tokens_seen": 10594096,
      "step": 1755,
      "train_runtime": 1779.1234,
      "train_tokens_per_second": 5954.672
    },
    {
      "epoch": 0.6458850574712643,
      "grad_norm": 0.7750902771949768,
      "learning_rate": 0.00013557886986931714,
      "loss": 1.8008,
      "num_input_tokens_seen": 10598464,
      "step": 1756,
      "train_runtime": 1779.8568,
      "train_tokens_per_second": 5954.672
    },
    {
      "epoch": 0.6462528735632184,
      "grad_norm": 0.7399452328681946,
      "learning_rate": 0.00013554205779495675,
      "loss": 1.716,
      "num_input_tokens_seen": 10604304,
      "step": 1757,
      "train_runtime": 1780.8424,
      "train_tokens_per_second": 5954.656
    },
    {
      "epoch": 0.6466206896551724,
      "grad_norm": 0.7232922315597534,
      "learning_rate": 0.00013550524572059636,
      "loss": 1.6881,
      "num_input_tokens_seen": 10610480,
      "step": 1758,
      "train_runtime": 1781.8633,
      "train_tokens_per_second": 5954.71
    },
    {
      "epoch": 0.6469885057471264,
      "grad_norm": 0.6953706741333008,
      "learning_rate": 0.00013546843364623597,
      "loss": 1.6719,
      "num_input_tokens_seen": 10616560,
      "step": 1759,
      "train_runtime": 1782.8585,
      "train_tokens_per_second": 5954.797
    },
    {
      "epoch": 0.6473563218390804,
      "grad_norm": 0.714076578617096,
      "learning_rate": 0.00013543162157187558,
      "loss": 1.7592,
      "num_input_tokens_seen": 10623328,
      "step": 1760,
      "train_runtime": 1783.9605,
      "train_tokens_per_second": 5954.912
    },
    {
      "epoch": 0.6477241379310344,
      "grad_norm": 0.7438048720359802,
      "learning_rate": 0.0001353948094975152,
      "loss": 1.7207,
      "num_input_tokens_seen": 10628864,
      "step": 1761,
      "train_runtime": 1784.8848,
      "train_tokens_per_second": 5954.93
    },
    {
      "epoch": 0.6480919540229885,
      "grad_norm": 0.791041374206543,
      "learning_rate": 0.0001353579974231548,
      "loss": 1.7115,
      "num_input_tokens_seen": 10636384,
      "step": 1762,
      "train_runtime": 1786.0825,
      "train_tokens_per_second": 5955.147
    },
    {
      "epoch": 0.6484597701149425,
      "grad_norm": 0.7846536040306091,
      "learning_rate": 0.0001353211853487944,
      "loss": 1.8612,
      "num_input_tokens_seen": 10642704,
      "step": 1763,
      "train_runtime": 1787.1234,
      "train_tokens_per_second": 5955.215
    },
    {
      "epoch": 0.6488275862068965,
      "grad_norm": 0.7329516410827637,
      "learning_rate": 0.00013528437327443401,
      "loss": 1.625,
      "num_input_tokens_seen": 10648160,
      "step": 1764,
      "train_runtime": 1788.0154,
      "train_tokens_per_second": 5955.296
    },
    {
      "epoch": 0.6491954022988505,
      "grad_norm": 0.714447021484375,
      "learning_rate": 0.00013524756120007362,
      "loss": 1.766,
      "num_input_tokens_seen": 10654272,
      "step": 1765,
      "train_runtime": 1789.0111,
      "train_tokens_per_second": 5955.397
    },
    {
      "epoch": 0.6495632183908046,
      "grad_norm": 0.7463008165359497,
      "learning_rate": 0.00013521074912571323,
      "loss": 1.8246,
      "num_input_tokens_seen": 10660864,
      "step": 1766,
      "train_runtime": 1790.1083,
      "train_tokens_per_second": 5955.43
    },
    {
      "epoch": 0.6499310344827586,
      "grad_norm": 0.8064841032028198,
      "learning_rate": 0.00013517393705135284,
      "loss": 1.8307,
      "num_input_tokens_seen": 10665504,
      "step": 1767,
      "train_runtime": 1790.9236,
      "train_tokens_per_second": 5955.309
    },
    {
      "epoch": 0.6502988505747126,
      "grad_norm": 0.7825813293457031,
      "learning_rate": 0.00013513712497699248,
      "loss": 1.7428,
      "num_input_tokens_seen": 10669872,
      "step": 1768,
      "train_runtime": 1791.6774,
      "train_tokens_per_second": 5955.242
    },
    {
      "epoch": 0.6506666666666666,
      "grad_norm": 0.7944249510765076,
      "learning_rate": 0.00013510031290263209,
      "loss": 1.8493,
      "num_input_tokens_seen": 10674368,
      "step": 1769,
      "train_runtime": 1792.482,
      "train_tokens_per_second": 5955.077
    },
    {
      "epoch": 0.6510344827586206,
      "grad_norm": 0.760125994682312,
      "learning_rate": 0.0001350635008282717,
      "loss": 1.7274,
      "num_input_tokens_seen": 10679600,
      "step": 1770,
      "train_runtime": 1793.3581,
      "train_tokens_per_second": 5955.085
    },
    {
      "epoch": 0.6514022988505748,
      "grad_norm": 0.7194905281066895,
      "learning_rate": 0.0001350266887539113,
      "loss": 1.7112,
      "num_input_tokens_seen": 10687776,
      "step": 1771,
      "train_runtime": 1795.1084,
      "train_tokens_per_second": 5953.833
    },
    {
      "epoch": 0.6517701149425288,
      "grad_norm": 0.770473062992096,
      "learning_rate": 0.00013498987667955089,
      "loss": 1.6892,
      "num_input_tokens_seen": 10694480,
      "step": 1772,
      "train_runtime": 1796.1946,
      "train_tokens_per_second": 5953.965
    },
    {
      "epoch": 0.6521379310344828,
      "grad_norm": 0.7285348773002625,
      "learning_rate": 0.0001349530646051905,
      "loss": 1.723,
      "num_input_tokens_seen": 10700432,
      "step": 1773,
      "train_runtime": 1797.1615,
      "train_tokens_per_second": 5954.074
    },
    {
      "epoch": 0.6525057471264368,
      "grad_norm": 0.6952595710754395,
      "learning_rate": 0.0001349162525308301,
      "loss": 1.7812,
      "num_input_tokens_seen": 10705728,
      "step": 1774,
      "train_runtime": 1798.0341,
      "train_tokens_per_second": 5954.129
    },
    {
      "epoch": 0.6528735632183909,
      "grad_norm": 0.7674087285995483,
      "learning_rate": 0.0001348794404564697,
      "loss": 1.6874,
      "num_input_tokens_seen": 10711552,
      "step": 1775,
      "train_runtime": 1799.0317,
      "train_tokens_per_second": 5954.065
    },
    {
      "epoch": 0.6532413793103449,
      "grad_norm": 0.7821603417396545,
      "learning_rate": 0.00013484262838210935,
      "loss": 1.7043,
      "num_input_tokens_seen": 10720400,
      "step": 1776,
      "train_runtime": 1800.4336,
      "train_tokens_per_second": 5954.343
    },
    {
      "epoch": 0.6536091954022989,
      "grad_norm": 0.7462404370307922,
      "learning_rate": 0.00013480581630774896,
      "loss": 1.7187,
      "num_input_tokens_seen": 10725504,
      "step": 1777,
      "train_runtime": 1801.2868,
      "train_tokens_per_second": 5954.357
    },
    {
      "epoch": 0.6539770114942529,
      "grad_norm": 0.7450446486473083,
      "learning_rate": 0.00013476900423338857,
      "loss": 1.9282,
      "num_input_tokens_seen": 10731584,
      "step": 1778,
      "train_runtime": 1802.295,
      "train_tokens_per_second": 5954.399
    },
    {
      "epoch": 0.6543448275862069,
      "grad_norm": 0.7468043565750122,
      "learning_rate": 0.00013473219215902817,
      "loss": 1.9602,
      "num_input_tokens_seen": 10737184,
      "step": 1779,
      "train_runtime": 1803.2264,
      "train_tokens_per_second": 5954.429
    },
    {
      "epoch": 0.654712643678161,
      "grad_norm": 0.6669394969940186,
      "learning_rate": 0.00013469538008466778,
      "loss": 1.6997,
      "num_input_tokens_seen": 10744272,
      "step": 1780,
      "train_runtime": 1804.3842,
      "train_tokens_per_second": 5954.537
    },
    {
      "epoch": 0.655080459770115,
      "grad_norm": 0.7069559097290039,
      "learning_rate": 0.0001346585680103074,
      "loss": 1.7087,
      "num_input_tokens_seen": 10749104,
      "step": 1781,
      "train_runtime": 1805.1987,
      "train_tokens_per_second": 5954.527
    },
    {
      "epoch": 0.655448275862069,
      "grad_norm": 0.725911021232605,
      "learning_rate": 0.000134621755935947,
      "loss": 1.6093,
      "num_input_tokens_seen": 10755392,
      "step": 1782,
      "train_runtime": 1806.2113,
      "train_tokens_per_second": 5954.67
    },
    {
      "epoch": 0.655816091954023,
      "grad_norm": 0.7561075091362,
      "learning_rate": 0.0001345849438615866,
      "loss": 1.536,
      "num_input_tokens_seen": 10762416,
      "step": 1783,
      "train_runtime": 1807.3389,
      "train_tokens_per_second": 5954.841
    },
    {
      "epoch": 0.656183908045977,
      "grad_norm": 0.7972083687782288,
      "learning_rate": 0.00013454813178722622,
      "loss": 1.8664,
      "num_input_tokens_seen": 10766848,
      "step": 1784,
      "train_runtime": 1808.094,
      "train_tokens_per_second": 5954.805
    },
    {
      "epoch": 0.6565517241379311,
      "grad_norm": 0.7465891242027283,
      "learning_rate": 0.00013451131971286583,
      "loss": 1.8208,
      "num_input_tokens_seen": 10772032,
      "step": 1785,
      "train_runtime": 1808.9644,
      "train_tokens_per_second": 5954.806
    },
    {
      "epoch": 0.6569195402298851,
      "grad_norm": 0.7285373210906982,
      "learning_rate": 0.00013447450763850544,
      "loss": 1.8164,
      "num_input_tokens_seen": 10777360,
      "step": 1786,
      "train_runtime": 1809.8471,
      "train_tokens_per_second": 5954.846
    },
    {
      "epoch": 0.6572873563218391,
      "grad_norm": 0.7597459554672241,
      "learning_rate": 0.00013443769556414505,
      "loss": 1.5547,
      "num_input_tokens_seen": 10786320,
      "step": 1787,
      "train_runtime": 1811.2173,
      "train_tokens_per_second": 5955.288
    },
    {
      "epoch": 0.6576551724137931,
      "grad_norm": 0.7373406887054443,
      "learning_rate": 0.00013440088348978465,
      "loss": 1.7168,
      "num_input_tokens_seen": 10791888,
      "step": 1788,
      "train_runtime": 1812.1421,
      "train_tokens_per_second": 5955.321
    },
    {
      "epoch": 0.6580229885057471,
      "grad_norm": 0.8384848237037659,
      "learning_rate": 0.00013436407141542426,
      "loss": 1.6351,
      "num_input_tokens_seen": 10800544,
      "step": 1789,
      "train_runtime": 1813.4966,
      "train_tokens_per_second": 5955.646
    },
    {
      "epoch": 0.6583908045977012,
      "grad_norm": 0.7638373970985413,
      "learning_rate": 0.00013432725934106387,
      "loss": 1.6756,
      "num_input_tokens_seen": 10806704,
      "step": 1790,
      "train_runtime": 1814.5051,
      "train_tokens_per_second": 5955.731
    },
    {
      "epoch": 0.6587586206896552,
      "grad_norm": 0.8199154138565063,
      "learning_rate": 0.00013429044726670348,
      "loss": 1.9613,
      "num_input_tokens_seen": 10811824,
      "step": 1791,
      "train_runtime": 1815.3818,
      "train_tokens_per_second": 5955.675
    },
    {
      "epoch": 0.6591264367816092,
      "grad_norm": 0.6468129754066467,
      "learning_rate": 0.0001342536351923431,
      "loss": 1.5419,
      "num_input_tokens_seen": 10819584,
      "step": 1792,
      "train_runtime": 1816.5835,
      "train_tokens_per_second": 5956.007
    },
    {
      "epoch": 0.6594942528735632,
      "grad_norm": 0.7387782335281372,
      "learning_rate": 0.00013421682311798273,
      "loss": 1.7476,
      "num_input_tokens_seen": 10826496,
      "step": 1793,
      "train_runtime": 1817.6847,
      "train_tokens_per_second": 5956.201
    },
    {
      "epoch": 0.6598620689655172,
      "grad_norm": 0.7090125679969788,
      "learning_rate": 0.00013418001104362234,
      "loss": 1.6913,
      "num_input_tokens_seen": 10835280,
      "step": 1794,
      "train_runtime": 1819.0353,
      "train_tokens_per_second": 5956.608
    },
    {
      "epoch": 0.6602298850574713,
      "grad_norm": 0.6886354088783264,
      "learning_rate": 0.00013414319896926192,
      "loss": 1.5162,
      "num_input_tokens_seen": 10845792,
      "step": 1795,
      "train_runtime": 1820.6393,
      "train_tokens_per_second": 5957.134
    },
    {
      "epoch": 0.6605977011494253,
      "grad_norm": 0.7396478652954102,
      "learning_rate": 0.00013410638689490153,
      "loss": 1.7201,
      "num_input_tokens_seen": 10853200,
      "step": 1796,
      "train_runtime": 1821.8114,
      "train_tokens_per_second": 5957.367
    },
    {
      "epoch": 0.6609655172413793,
      "grad_norm": 0.7231253385543823,
      "learning_rate": 0.00013406957482054113,
      "loss": 1.7426,
      "num_input_tokens_seen": 10858336,
      "step": 1797,
      "train_runtime": 1822.6624,
      "train_tokens_per_second": 5957.404
    },
    {
      "epoch": 0.6613333333333333,
      "grad_norm": 0.7091934084892273,
      "learning_rate": 0.00013403276274618074,
      "loss": 1.6118,
      "num_input_tokens_seen": 10866416,
      "step": 1798,
      "train_runtime": 1823.9454,
      "train_tokens_per_second": 5957.643
    },
    {
      "epoch": 0.6617011494252873,
      "grad_norm": 0.7346258163452148,
      "learning_rate": 0.00013399595067182035,
      "loss": 1.8568,
      "num_input_tokens_seen": 10871264,
      "step": 1799,
      "train_runtime": 1824.7824,
      "train_tokens_per_second": 5957.567
    },
    {
      "epoch": 0.6620689655172414,
      "grad_norm": 0.7573279142379761,
      "learning_rate": 0.00013395913859745996,
      "loss": 1.7545,
      "num_input_tokens_seen": 10875296,
      "step": 1800,
      "train_runtime": 1825.5415,
      "train_tokens_per_second": 5957.299
    },
    {
      "epoch": 0.6624367816091954,
      "grad_norm": 0.7126784324645996,
      "learning_rate": 0.00013392232652309957,
      "loss": 1.7663,
      "num_input_tokens_seen": 10880112,
      "step": 1801,
      "train_runtime": 1826.7989,
      "train_tokens_per_second": 5955.835
    },
    {
      "epoch": 0.6628045977011494,
      "grad_norm": 0.7735157608985901,
      "learning_rate": 0.0001338855144487392,
      "loss": 1.7039,
      "num_input_tokens_seen": 10886160,
      "step": 1802,
      "train_runtime": 1827.8044,
      "train_tokens_per_second": 5955.867
    },
    {
      "epoch": 0.6631724137931034,
      "grad_norm": 0.6788606643676758,
      "learning_rate": 0.00013384870237437882,
      "loss": 1.6366,
      "num_input_tokens_seen": 10891728,
      "step": 1803,
      "train_runtime": 1828.7318,
      "train_tokens_per_second": 5955.891
    },
    {
      "epoch": 0.6635402298850575,
      "grad_norm": 0.7734730243682861,
      "learning_rate": 0.00013381189030001842,
      "loss": 1.7109,
      "num_input_tokens_seen": 10898304,
      "step": 1804,
      "train_runtime": 1829.7889,
      "train_tokens_per_second": 5956.044
    },
    {
      "epoch": 0.6639080459770115,
      "grad_norm": 0.7669985294342041,
      "learning_rate": 0.00013377507822565803,
      "loss": 1.8261,
      "num_input_tokens_seen": 10902800,
      "step": 1805,
      "train_runtime": 1830.5775,
      "train_tokens_per_second": 5955.935
    },
    {
      "epoch": 0.6642758620689655,
      "grad_norm": 0.8388978838920593,
      "learning_rate": 0.00013373826615129764,
      "loss": 1.7974,
      "num_input_tokens_seen": 10909616,
      "step": 1806,
      "train_runtime": 1831.6497,
      "train_tokens_per_second": 5956.169
    },
    {
      "epoch": 0.6646436781609195,
      "grad_norm": 0.7415384650230408,
      "learning_rate": 0.00013370145407693725,
      "loss": 1.7101,
      "num_input_tokens_seen": 10916032,
      "step": 1807,
      "train_runtime": 1832.6885,
      "train_tokens_per_second": 5956.294
    },
    {
      "epoch": 0.6650114942528735,
      "grad_norm": 0.7248274683952332,
      "learning_rate": 0.00013366464200257683,
      "loss": 1.8749,
      "num_input_tokens_seen": 10921952,
      "step": 1808,
      "train_runtime": 1833.6336,
      "train_tokens_per_second": 5956.453
    },
    {
      "epoch": 0.6653793103448276,
      "grad_norm": 0.7237791419029236,
      "learning_rate": 0.00013362782992821644,
      "loss": 1.7853,
      "num_input_tokens_seen": 10927472,
      "step": 1809,
      "train_runtime": 1834.5498,
      "train_tokens_per_second": 5956.487
    },
    {
      "epoch": 0.6657471264367816,
      "grad_norm": 0.7164573669433594,
      "learning_rate": 0.00013359101785385608,
      "loss": 1.8934,
      "num_input_tokens_seen": 10932752,
      "step": 1810,
      "train_runtime": 1835.4394,
      "train_tokens_per_second": 5956.477
    },
    {
      "epoch": 0.6661149425287356,
      "grad_norm": 0.7893067598342896,
      "learning_rate": 0.00013355420577949569,
      "loss": 1.6218,
      "num_input_tokens_seen": 10941600,
      "step": 1811,
      "train_runtime": 1836.8188,
      "train_tokens_per_second": 5956.82
    },
    {
      "epoch": 0.6664827586206896,
      "grad_norm": 0.8066549897193909,
      "learning_rate": 0.0001335173937051353,
      "loss": 1.6749,
      "num_input_tokens_seen": 10946992,
      "step": 1812,
      "train_runtime": 1837.7,
      "train_tokens_per_second": 5956.898
    },
    {
      "epoch": 0.6668505747126436,
      "grad_norm": 0.6956750750541687,
      "learning_rate": 0.0001334805816307749,
      "loss": 1.5983,
      "num_input_tokens_seen": 10953584,
      "step": 1813,
      "train_runtime": 1838.7707,
      "train_tokens_per_second": 5957.015
    },
    {
      "epoch": 0.6672183908045977,
      "grad_norm": 0.7238765954971313,
      "learning_rate": 0.0001334437695564145,
      "loss": 1.645,
      "num_input_tokens_seen": 10958688,
      "step": 1814,
      "train_runtime": 1839.6198,
      "train_tokens_per_second": 5957.04
    },
    {
      "epoch": 0.6675862068965517,
      "grad_norm": 0.8044825792312622,
      "learning_rate": 0.00013340695748205412,
      "loss": 1.7582,
      "num_input_tokens_seen": 10963728,
      "step": 1815,
      "train_runtime": 1840.465,
      "train_tokens_per_second": 5957.042
    },
    {
      "epoch": 0.6679540229885057,
      "grad_norm": 0.7360802888870239,
      "learning_rate": 0.00013337014540769373,
      "loss": 1.8713,
      "num_input_tokens_seen": 10971888,
      "step": 1816,
      "train_runtime": 1841.7483,
      "train_tokens_per_second": 5957.322
    },
    {
      "epoch": 0.6683218390804597,
      "grad_norm": 0.6782446503639221,
      "learning_rate": 0.00013333333333333334,
      "loss": 1.6668,
      "num_input_tokens_seen": 10977072,
      "step": 1817,
      "train_runtime": 1842.6362,
      "train_tokens_per_second": 5957.265
    },
    {
      "epoch": 0.6686896551724137,
      "grad_norm": 0.7192976474761963,
      "learning_rate": 0.00013329652125897295,
      "loss": 1.9386,
      "num_input_tokens_seen": 10982016,
      "step": 1818,
      "train_runtime": 1843.4671,
      "train_tokens_per_second": 5957.262
    },
    {
      "epoch": 0.6690574712643678,
      "grad_norm": 0.7062948346138,
      "learning_rate": 0.00013325970918461256,
      "loss": 1.8731,
      "num_input_tokens_seen": 10987360,
      "step": 1819,
      "train_runtime": 1844.3595,
      "train_tokens_per_second": 5957.277
    },
    {
      "epoch": 0.6694252873563218,
      "grad_norm": 0.7209194302558899,
      "learning_rate": 0.00013322289711025217,
      "loss": 1.6957,
      "num_input_tokens_seen": 10992128,
      "step": 1820,
      "train_runtime": 1845.181,
      "train_tokens_per_second": 5957.209
    },
    {
      "epoch": 0.6697931034482759,
      "grad_norm": 0.7385466694831848,
      "learning_rate": 0.00013318608503589178,
      "loss": 1.8945,
      "num_input_tokens_seen": 10997120,
      "step": 1821,
      "train_runtime": 1846.0394,
      "train_tokens_per_second": 5957.143
    },
    {
      "epoch": 0.6701609195402299,
      "grad_norm": 0.7902348041534424,
      "learning_rate": 0.00013314927296153138,
      "loss": 1.6632,
      "num_input_tokens_seen": 11002160,
      "step": 1822,
      "train_runtime": 1846.8926,
      "train_tokens_per_second": 5957.12
    },
    {
      "epoch": 0.670528735632184,
      "grad_norm": 0.7628754377365112,
      "learning_rate": 0.000133112460887171,
      "loss": 1.6967,
      "num_input_tokens_seen": 11009808,
      "step": 1823,
      "train_runtime": 1848.11,
      "train_tokens_per_second": 5957.334
    },
    {
      "epoch": 0.670896551724138,
      "grad_norm": 0.7581140398979187,
      "learning_rate": 0.0001330756488128106,
      "loss": 1.847,
      "num_input_tokens_seen": 11015088,
      "step": 1824,
      "train_runtime": 1849.0022,
      "train_tokens_per_second": 5957.315
    },
    {
      "epoch": 0.671264367816092,
      "grad_norm": 0.6395372748374939,
      "learning_rate": 0.0001330388367384502,
      "loss": 1.7134,
      "num_input_tokens_seen": 11022160,
      "step": 1825,
      "train_runtime": 1850.1106,
      "train_tokens_per_second": 5957.568
    },
    {
      "epoch": 0.671632183908046,
      "grad_norm": 0.8300169706344604,
      "learning_rate": 0.00013300202466408982,
      "loss": 1.8971,
      "num_input_tokens_seen": 11027040,
      "step": 1826,
      "train_runtime": 1850.9352,
      "train_tokens_per_second": 5957.551
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.7367687225341797,
      "learning_rate": 0.00013296521258972946,
      "loss": 1.7755,
      "num_input_tokens_seen": 11034128,
      "step": 1827,
      "train_runtime": 1852.0761,
      "train_tokens_per_second": 5957.708
    },
    {
      "epoch": 0.6723678160919541,
      "grad_norm": 0.7843712568283081,
      "learning_rate": 0.00013292840051536906,
      "loss": 1.7056,
      "num_input_tokens_seen": 11038624,
      "step": 1828,
      "train_runtime": 1852.8793,
      "train_tokens_per_second": 5957.552
    },
    {
      "epoch": 0.6727356321839081,
      "grad_norm": 0.746307909488678,
      "learning_rate": 0.00013289158844100867,
      "loss": 1.8213,
      "num_input_tokens_seen": 11043632,
      "step": 1829,
      "train_runtime": 1853.7329,
      "train_tokens_per_second": 5957.51
    },
    {
      "epoch": 0.6731034482758621,
      "grad_norm": 0.7821405529975891,
      "learning_rate": 0.00013285477636664828,
      "loss": 2.0339,
      "num_input_tokens_seen": 11048640,
      "step": 1830,
      "train_runtime": 1854.5729,
      "train_tokens_per_second": 5957.512
    },
    {
      "epoch": 0.6734712643678161,
      "grad_norm": 0.7343426942825317,
      "learning_rate": 0.00013281796429228786,
      "loss": 1.6836,
      "num_input_tokens_seen": 11055120,
      "step": 1831,
      "train_runtime": 1856.183,
      "train_tokens_per_second": 5955.835
    },
    {
      "epoch": 0.6738390804597701,
      "grad_norm": 0.7531905770301819,
      "learning_rate": 0.00013278115221792747,
      "loss": 1.7754,
      "num_input_tokens_seen": 11059568,
      "step": 1832,
      "train_runtime": 1856.9851,
      "train_tokens_per_second": 5955.658
    },
    {
      "epoch": 0.6742068965517242,
      "grad_norm": 0.7625032067298889,
      "learning_rate": 0.00013274434014356708,
      "loss": 1.8033,
      "num_input_tokens_seen": 11064608,
      "step": 1833,
      "train_runtime": 1857.852,
      "train_tokens_per_second": 5955.592
    },
    {
      "epoch": 0.6745747126436782,
      "grad_norm": 0.7211398482322693,
      "learning_rate": 0.0001327075280692067,
      "loss": 1.9395,
      "num_input_tokens_seen": 11069888,
      "step": 1834,
      "train_runtime": 1858.723,
      "train_tokens_per_second": 5955.642
    },
    {
      "epoch": 0.6749425287356322,
      "grad_norm": 0.9047591686248779,
      "learning_rate": 0.0001326707159948463,
      "loss": 1.8467,
      "num_input_tokens_seen": 11074624,
      "step": 1835,
      "train_runtime": 1859.5346,
      "train_tokens_per_second": 5955.589
    },
    {
      "epoch": 0.6753103448275862,
      "grad_norm": 0.7916085720062256,
      "learning_rate": 0.00013263390392048594,
      "loss": 1.8939,
      "num_input_tokens_seen": 11079216,
      "step": 1836,
      "train_runtime": 1860.3121,
      "train_tokens_per_second": 5955.569
    },
    {
      "epoch": 0.6756781609195402,
      "grad_norm": 0.8396905660629272,
      "learning_rate": 0.00013259709184612554,
      "loss": 1.8422,
      "num_input_tokens_seen": 11083840,
      "step": 1837,
      "train_runtime": 1861.1096,
      "train_tokens_per_second": 5955.501
    },
    {
      "epoch": 0.6760459770114943,
      "grad_norm": 0.7099091410636902,
      "learning_rate": 0.00013256027977176515,
      "loss": 1.7967,
      "num_input_tokens_seen": 11089040,
      "step": 1838,
      "train_runtime": 1862.0071,
      "train_tokens_per_second": 5955.423
    },
    {
      "epoch": 0.6764137931034483,
      "grad_norm": 0.731762707233429,
      "learning_rate": 0.00013252346769740476,
      "loss": 1.6724,
      "num_input_tokens_seen": 11098848,
      "step": 1839,
      "train_runtime": 1863.5256,
      "train_tokens_per_second": 5955.833
    },
    {
      "epoch": 0.6767816091954023,
      "grad_norm": 0.7560499310493469,
      "learning_rate": 0.00013248665562304437,
      "loss": 1.7258,
      "num_input_tokens_seen": 11105088,
      "step": 1840,
      "train_runtime": 1864.5461,
      "train_tokens_per_second": 5955.92
    },
    {
      "epoch": 0.6771494252873563,
      "grad_norm": 0.7681247591972351,
      "learning_rate": 0.00013244984354868398,
      "loss": 1.7083,
      "num_input_tokens_seen": 11110912,
      "step": 1841,
      "train_runtime": 1865.4978,
      "train_tokens_per_second": 5956.004
    },
    {
      "epoch": 0.6775172413793104,
      "grad_norm": 0.8783697485923767,
      "learning_rate": 0.0001324130314743236,
      "loss": 1.8156,
      "num_input_tokens_seen": 11115904,
      "step": 1842,
      "train_runtime": 1866.3432,
      "train_tokens_per_second": 5955.981
    },
    {
      "epoch": 0.6778850574712644,
      "grad_norm": 0.6927594542503357,
      "learning_rate": 0.0001323762193999632,
      "loss": 1.7041,
      "num_input_tokens_seen": 11120800,
      "step": 1843,
      "train_runtime": 1867.2007,
      "train_tokens_per_second": 5955.868
    },
    {
      "epoch": 0.6782528735632184,
      "grad_norm": 0.7401049733161926,
      "learning_rate": 0.00013233940732560278,
      "loss": 1.7515,
      "num_input_tokens_seen": 11126272,
      "step": 1844,
      "train_runtime": 1868.1105,
      "train_tokens_per_second": 5955.896
    },
    {
      "epoch": 0.6786206896551724,
      "grad_norm": 0.7136857509613037,
      "learning_rate": 0.00013230259525124242,
      "loss": 1.7286,
      "num_input_tokens_seen": 11132528,
      "step": 1845,
      "train_runtime": 1869.1201,
      "train_tokens_per_second": 5956.026
    },
    {
      "epoch": 0.6789885057471264,
      "grad_norm": 0.7986119985580444,
      "learning_rate": 0.00013226578317688202,
      "loss": 2.0149,
      "num_input_tokens_seen": 11139376,
      "step": 1846,
      "train_runtime": 1870.2233,
      "train_tokens_per_second": 5956.174
    },
    {
      "epoch": 0.6793563218390805,
      "grad_norm": 0.7753489017486572,
      "learning_rate": 0.00013222897110252163,
      "loss": 1.7657,
      "num_input_tokens_seen": 11144416,
      "step": 1847,
      "train_runtime": 1871.0785,
      "train_tokens_per_second": 5956.146
    },
    {
      "epoch": 0.6797241379310345,
      "grad_norm": 0.7110564708709717,
      "learning_rate": 0.00013219215902816124,
      "loss": 1.6737,
      "num_input_tokens_seen": 11150544,
      "step": 1848,
      "train_runtime": 1872.1052,
      "train_tokens_per_second": 5956.153
    },
    {
      "epoch": 0.6800919540229885,
      "grad_norm": 0.7208719849586487,
      "learning_rate": 0.00013215534695380085,
      "loss": 1.817,
      "num_input_tokens_seen": 11155776,
      "step": 1849,
      "train_runtime": 1872.9985,
      "train_tokens_per_second": 5956.105
    },
    {
      "epoch": 0.6804597701149425,
      "grad_norm": 0.7724442481994629,
      "learning_rate": 0.00013211853487944046,
      "loss": 1.7954,
      "num_input_tokens_seen": 11161056,
      "step": 1850,
      "train_runtime": 1873.8924,
      "train_tokens_per_second": 5956.082
    },
    {
      "epoch": 0.6808275862068965,
      "grad_norm": 0.7539997100830078,
      "learning_rate": 0.00013208172280508007,
      "loss": 1.6489,
      "num_input_tokens_seen": 11168368,
      "step": 1851,
      "train_runtime": 1875.0539,
      "train_tokens_per_second": 5956.292
    },
    {
      "epoch": 0.6811954022988506,
      "grad_norm": 0.7062416076660156,
      "learning_rate": 0.00013204491073071968,
      "loss": 1.4751,
      "num_input_tokens_seen": 11176256,
      "step": 1852,
      "train_runtime": 1876.2653,
      "train_tokens_per_second": 5956.65
    },
    {
      "epoch": 0.6815632183908046,
      "grad_norm": 0.7522175312042236,
      "learning_rate": 0.00013200809865635931,
      "loss": 1.8941,
      "num_input_tokens_seen": 11182608,
      "step": 1853,
      "train_runtime": 1877.2662,
      "train_tokens_per_second": 5956.858
    },
    {
      "epoch": 0.6819310344827586,
      "grad_norm": 0.7194035649299622,
      "learning_rate": 0.00013197128658199892,
      "loss": 1.9232,
      "num_input_tokens_seen": 11187392,
      "step": 1854,
      "train_runtime": 1878.0975,
      "train_tokens_per_second": 5956.768
    },
    {
      "epoch": 0.6822988505747126,
      "grad_norm": 0.7332926392555237,
      "learning_rate": 0.0001319344745076385,
      "loss": 1.6695,
      "num_input_tokens_seen": 11193280,
      "step": 1855,
      "train_runtime": 1879.0632,
      "train_tokens_per_second": 5956.841
    },
    {
      "epoch": 0.6826666666666666,
      "grad_norm": 0.7417240738868713,
      "learning_rate": 0.00013189766243327811,
      "loss": 1.7371,
      "num_input_tokens_seen": 11198624,
      "step": 1856,
      "train_runtime": 1879.9562,
      "train_tokens_per_second": 5956.854
    },
    {
      "epoch": 0.6830344827586207,
      "grad_norm": 0.7554106712341309,
      "learning_rate": 0.00013186085035891772,
      "loss": 1.8963,
      "num_input_tokens_seen": 11203424,
      "step": 1857,
      "train_runtime": 1880.787,
      "train_tokens_per_second": 5956.774
    },
    {
      "epoch": 0.6834022988505747,
      "grad_norm": 0.7874431014060974,
      "learning_rate": 0.00013182403828455733,
      "loss": 1.9356,
      "num_input_tokens_seen": 11208112,
      "step": 1858,
      "train_runtime": 1881.5802,
      "train_tokens_per_second": 5956.755
    },
    {
      "epoch": 0.6837701149425287,
      "grad_norm": 0.7526617050170898,
      "learning_rate": 0.00013178722621019694,
      "loss": 1.6181,
      "num_input_tokens_seen": 11213488,
      "step": 1859,
      "train_runtime": 1882.4626,
      "train_tokens_per_second": 5956.819
    },
    {
      "epoch": 0.6841379310344827,
      "grad_norm": 0.732140839099884,
      "learning_rate": 0.00013175041413583655,
      "loss": 1.6019,
      "num_input_tokens_seen": 11219840,
      "step": 1860,
      "train_runtime": 1883.4777,
      "train_tokens_per_second": 5956.981
    },
    {
      "epoch": 0.6845057471264367,
      "grad_norm": 0.8277818560600281,
      "learning_rate": 0.00013171360206147616,
      "loss": 1.594,
      "num_input_tokens_seen": 11224816,
      "step": 1861,
      "train_runtime": 1884.867,
      "train_tokens_per_second": 5955.23
    },
    {
      "epoch": 0.6848735632183908,
      "grad_norm": 0.8288012742996216,
      "learning_rate": 0.0001316767899871158,
      "loss": 1.6824,
      "num_input_tokens_seen": 11232544,
      "step": 1862,
      "train_runtime": 1886.0893,
      "train_tokens_per_second": 5955.468
    },
    {
      "epoch": 0.6852413793103448,
      "grad_norm": 0.7560052275657654,
      "learning_rate": 0.0001316399779127554,
      "loss": 1.7055,
      "num_input_tokens_seen": 11238240,
      "step": 1863,
      "train_runtime": 1887.0228,
      "train_tokens_per_second": 5955.54
    },
    {
      "epoch": 0.6856091954022988,
      "grad_norm": 0.6578978896141052,
      "learning_rate": 0.000131603165838395,
      "loss": 1.6784,
      "num_input_tokens_seen": 11247264,
      "step": 1864,
      "train_runtime": 1888.3915,
      "train_tokens_per_second": 5956.002
    },
    {
      "epoch": 0.6859770114942528,
      "grad_norm": 0.7248901724815369,
      "learning_rate": 0.00013156635376403462,
      "loss": 1.6512,
      "num_input_tokens_seen": 11252624,
      "step": 1865,
      "train_runtime": 1889.2707,
      "train_tokens_per_second": 5956.068
    },
    {
      "epoch": 0.6863448275862069,
      "grad_norm": 0.7533191442489624,
      "learning_rate": 0.00013152954168967423,
      "loss": 1.9194,
      "num_input_tokens_seen": 11258496,
      "step": 1866,
      "train_runtime": 1890.2607,
      "train_tokens_per_second": 5956.055
    },
    {
      "epoch": 0.6867126436781609,
      "grad_norm": 0.7918365597724915,
      "learning_rate": 0.00013149272961531384,
      "loss": 1.8506,
      "num_input_tokens_seen": 11263488,
      "step": 1867,
      "train_runtime": 1891.1111,
      "train_tokens_per_second": 5956.016
    },
    {
      "epoch": 0.6870804597701149,
      "grad_norm": 0.7057754397392273,
      "learning_rate": 0.00013145591754095342,
      "loss": 1.561,
      "num_input_tokens_seen": 11270160,
      "step": 1868,
      "train_runtime": 1892.1857,
      "train_tokens_per_second": 5956.16
    },
    {
      "epoch": 0.6874482758620689,
      "grad_norm": 0.8185402750968933,
      "learning_rate": 0.00013141910546659303,
      "loss": 2.1177,
      "num_input_tokens_seen": 11278208,
      "step": 1869,
      "train_runtime": 1893.4343,
      "train_tokens_per_second": 5956.482
    },
    {
      "epoch": 0.6878160919540229,
      "grad_norm": 0.7990933656692505,
      "learning_rate": 0.00013138229339223267,
      "loss": 1.7334,
      "num_input_tokens_seen": 11285120,
      "step": 1870,
      "train_runtime": 1894.5252,
      "train_tokens_per_second": 5956.701
    },
    {
      "epoch": 0.6881839080459771,
      "grad_norm": 0.761961817741394,
      "learning_rate": 0.00013134548131787227,
      "loss": 1.6831,
      "num_input_tokens_seen": 11290672,
      "step": 1871,
      "train_runtime": 1895.4421,
      "train_tokens_per_second": 5956.748
    },
    {
      "epoch": 0.6885517241379311,
      "grad_norm": 0.738158106803894,
      "learning_rate": 0.00013130866924351188,
      "loss": 1.8655,
      "num_input_tokens_seen": 11294960,
      "step": 1872,
      "train_runtime": 1896.2095,
      "train_tokens_per_second": 5956.599
    },
    {
      "epoch": 0.6889195402298851,
      "grad_norm": 0.7615065574645996,
      "learning_rate": 0.0001312718571691515,
      "loss": 1.6124,
      "num_input_tokens_seen": 11300096,
      "step": 1873,
      "train_runtime": 1897.0776,
      "train_tokens_per_second": 5956.581
    },
    {
      "epoch": 0.6892873563218391,
      "grad_norm": 0.6883339285850525,
      "learning_rate": 0.0001312350450947911,
      "loss": 1.6175,
      "num_input_tokens_seen": 11306480,
      "step": 1874,
      "train_runtime": 1898.1007,
      "train_tokens_per_second": 5956.734
    },
    {
      "epoch": 0.6896551724137931,
      "grad_norm": 0.6871680617332458,
      "learning_rate": 0.0001311982330204307,
      "loss": 1.6909,
      "num_input_tokens_seen": 11312192,
      "step": 1875,
      "train_runtime": 1899.0291,
      "train_tokens_per_second": 5956.829
    },
    {
      "epoch": 0.6900229885057472,
      "grad_norm": 0.7546426653862,
      "learning_rate": 0.00013116142094607032,
      "loss": 1.8858,
      "num_input_tokens_seen": 11317920,
      "step": 1876,
      "train_runtime": 1899.9744,
      "train_tokens_per_second": 5956.88
    },
    {
      "epoch": 0.6903908045977012,
      "grad_norm": 0.7262266278266907,
      "learning_rate": 0.00013112460887170993,
      "loss": 1.6922,
      "num_input_tokens_seen": 11325136,
      "step": 1877,
      "train_runtime": 1901.1451,
      "train_tokens_per_second": 5957.008
    },
    {
      "epoch": 0.6907586206896552,
      "grad_norm": 0.7512328624725342,
      "learning_rate": 0.00013108779679734954,
      "loss": 1.8147,
      "num_input_tokens_seen": 11329936,
      "step": 1878,
      "train_runtime": 1901.9565,
      "train_tokens_per_second": 5956.99
    },
    {
      "epoch": 0.6911264367816092,
      "grad_norm": 0.7658445239067078,
      "learning_rate": 0.00013105098472298915,
      "loss": 1.7388,
      "num_input_tokens_seen": 11335920,
      "step": 1879,
      "train_runtime": 1902.92,
      "train_tokens_per_second": 5957.118
    },
    {
      "epoch": 0.6914942528735633,
      "grad_norm": 0.7307431101799011,
      "learning_rate": 0.00013101417264862875,
      "loss": 1.8446,
      "num_input_tokens_seen": 11340384,
      "step": 1880,
      "train_runtime": 1903.7032,
      "train_tokens_per_second": 5957.013
    },
    {
      "epoch": 0.6918620689655173,
      "grad_norm": 0.7976140975952148,
      "learning_rate": 0.00013097736057426836,
      "loss": 1.84,
      "num_input_tokens_seen": 11344672,
      "step": 1881,
      "train_runtime": 1904.4681,
      "train_tokens_per_second": 5956.872
    },
    {
      "epoch": 0.6922298850574713,
      "grad_norm": 0.8744789361953735,
      "learning_rate": 0.00013094054849990797,
      "loss": 1.7794,
      "num_input_tokens_seen": 11350240,
      "step": 1882,
      "train_runtime": 1905.4273,
      "train_tokens_per_second": 5956.795
    },
    {
      "epoch": 0.6925977011494253,
      "grad_norm": 0.6834043264389038,
      "learning_rate": 0.00013090373642554758,
      "loss": 1.5589,
      "num_input_tokens_seen": 11356704,
      "step": 1883,
      "train_runtime": 1906.4917,
      "train_tokens_per_second": 5956.86
    },
    {
      "epoch": 0.6929655172413793,
      "grad_norm": 0.744666337966919,
      "learning_rate": 0.0001308669243511872,
      "loss": 1.7072,
      "num_input_tokens_seen": 11363824,
      "step": 1884,
      "train_runtime": 1907.6162,
      "train_tokens_per_second": 5957.081
    },
    {
      "epoch": 0.6933333333333334,
      "grad_norm": 0.7436920404434204,
      "learning_rate": 0.0001308301122768268,
      "loss": 1.8886,
      "num_input_tokens_seen": 11368656,
      "step": 1885,
      "train_runtime": 1908.4117,
      "train_tokens_per_second": 5957.13
    },
    {
      "epoch": 0.6937011494252874,
      "grad_norm": 0.8020740747451782,
      "learning_rate": 0.0001307933002024664,
      "loss": 1.7037,
      "num_input_tokens_seen": 11376048,
      "step": 1886,
      "train_runtime": 1909.57,
      "train_tokens_per_second": 5957.387
    },
    {
      "epoch": 0.6940689655172414,
      "grad_norm": 0.8908138275146484,
      "learning_rate": 0.00013075648812810604,
      "loss": 1.9804,
      "num_input_tokens_seen": 11380272,
      "step": 1887,
      "train_runtime": 1910.3019,
      "train_tokens_per_second": 5957.316
    },
    {
      "epoch": 0.6944367816091954,
      "grad_norm": 0.664467453956604,
      "learning_rate": 0.00013071967605374565,
      "loss": 1.6017,
      "num_input_tokens_seen": 11386096,
      "step": 1888,
      "train_runtime": 1911.2454,
      "train_tokens_per_second": 5957.422
    },
    {
      "epoch": 0.6948045977011494,
      "grad_norm": 0.7617599368095398,
      "learning_rate": 0.00013068286397938526,
      "loss": 1.655,
      "num_input_tokens_seen": 11392608,
      "step": 1889,
      "train_runtime": 1912.2834,
      "train_tokens_per_second": 5957.594
    },
    {
      "epoch": 0.6951724137931035,
      "grad_norm": 0.7180255055427551,
      "learning_rate": 0.00013064605190502487,
      "loss": 1.5084,
      "num_input_tokens_seen": 11398736,
      "step": 1890,
      "train_runtime": 1913.2593,
      "train_tokens_per_second": 5957.758
    },
    {
      "epoch": 0.6955402298850575,
      "grad_norm": 0.7027455568313599,
      "learning_rate": 0.00013060923983066445,
      "loss": 1.5955,
      "num_input_tokens_seen": 11407920,
      "step": 1891,
      "train_runtime": 1915.1687,
      "train_tokens_per_second": 5956.614
    },
    {
      "epoch": 0.6959080459770115,
      "grad_norm": 0.8807379007339478,
      "learning_rate": 0.00013057242775630406,
      "loss": 1.9017,
      "num_input_tokens_seen": 11415568,
      "step": 1892,
      "train_runtime": 1916.3593,
      "train_tokens_per_second": 5956.904
    },
    {
      "epoch": 0.6962758620689655,
      "grad_norm": 0.7481083273887634,
      "learning_rate": 0.00013053561568194367,
      "loss": 1.6372,
      "num_input_tokens_seen": 11421568,
      "step": 1893,
      "train_runtime": 1917.3738,
      "train_tokens_per_second": 5956.881
    },
    {
      "epoch": 0.6966436781609195,
      "grad_norm": 0.6700345873832703,
      "learning_rate": 0.00013049880360758328,
      "loss": 1.6483,
      "num_input_tokens_seen": 11428208,
      "step": 1894,
      "train_runtime": 1918.4374,
      "train_tokens_per_second": 5957.04
    },
    {
      "epoch": 0.6970114942528736,
      "grad_norm": 0.7736784815788269,
      "learning_rate": 0.0001304619915332229,
      "loss": 1.6346,
      "num_input_tokens_seen": 11434000,
      "step": 1895,
      "train_runtime": 1919.3645,
      "train_tokens_per_second": 5957.18
    },
    {
      "epoch": 0.6973793103448276,
      "grad_norm": 0.7341681122779846,
      "learning_rate": 0.00013042517945886252,
      "loss": 1.7157,
      "num_input_tokens_seen": 11438944,
      "step": 1896,
      "train_runtime": 1920.2291,
      "train_tokens_per_second": 5957.073
    },
    {
      "epoch": 0.6977471264367816,
      "grad_norm": 0.7348836064338684,
      "learning_rate": 0.00013038836738450213,
      "loss": 1.5479,
      "num_input_tokens_seen": 11445344,
      "step": 1897,
      "train_runtime": 1921.2621,
      "train_tokens_per_second": 5957.201
    },
    {
      "epoch": 0.6981149425287356,
      "grad_norm": 0.7838300466537476,
      "learning_rate": 0.00013035155531014174,
      "loss": 1.6346,
      "num_input_tokens_seen": 11451840,
      "step": 1898,
      "train_runtime": 1922.2957,
      "train_tokens_per_second": 5957.377
    },
    {
      "epoch": 0.6984827586206896,
      "grad_norm": 0.7902573943138123,
      "learning_rate": 0.00013031474323578135,
      "loss": 1.7418,
      "num_input_tokens_seen": 11457312,
      "step": 1899,
      "train_runtime": 1923.1835,
      "train_tokens_per_second": 5957.472
    },
    {
      "epoch": 0.6988505747126437,
      "grad_norm": 0.6877724528312683,
      "learning_rate": 0.00013027793116142096,
      "loss": 1.6553,
      "num_input_tokens_seen": 11462464,
      "step": 1900,
      "train_runtime": 1924.0756,
      "train_tokens_per_second": 5957.388
    },
    {
      "epoch": 0.6992183908045977,
      "grad_norm": 0.8958332538604736,
      "learning_rate": 0.00013024111908706057,
      "loss": 1.6905,
      "num_input_tokens_seen": 11467136,
      "step": 1901,
      "train_runtime": 1924.875,
      "train_tokens_per_second": 5957.341
    },
    {
      "epoch": 0.6995862068965517,
      "grad_norm": 0.7542179226875305,
      "learning_rate": 0.00013020430701270018,
      "loss": 1.5782,
      "num_input_tokens_seen": 11471520,
      "step": 1902,
      "train_runtime": 1925.6698,
      "train_tokens_per_second": 5957.158
    },
    {
      "epoch": 0.6999540229885057,
      "grad_norm": 0.7624781727790833,
      "learning_rate": 0.00013016749493833979,
      "loss": 1.6267,
      "num_input_tokens_seen": 11477936,
      "step": 1903,
      "train_runtime": 1926.6957,
      "train_tokens_per_second": 5957.316
    },
    {
      "epoch": 0.7003218390804598,
      "grad_norm": 0.7376978993415833,
      "learning_rate": 0.00013013068286397937,
      "loss": 1.686,
      "num_input_tokens_seen": 11486512,
      "step": 1904,
      "train_runtime": 1928.0176,
      "train_tokens_per_second": 5957.68
    },
    {
      "epoch": 0.7006896551724138,
      "grad_norm": 0.7072744965553284,
      "learning_rate": 0.000130093870789619,
      "loss": 1.5617,
      "num_input_tokens_seen": 11492464,
      "step": 1905,
      "train_runtime": 1929.0047,
      "train_tokens_per_second": 5957.717
    },
    {
      "epoch": 0.7010574712643678,
      "grad_norm": 0.7275297045707703,
      "learning_rate": 0.0001300570587152586,
      "loss": 1.8489,
      "num_input_tokens_seen": 11500048,
      "step": 1906,
      "train_runtime": 1930.1843,
      "train_tokens_per_second": 5958.005
    },
    {
      "epoch": 0.7014252873563218,
      "grad_norm": 0.7295718193054199,
      "learning_rate": 0.00013002024664089822,
      "loss": 1.7177,
      "num_input_tokens_seen": 11508528,
      "step": 1907,
      "train_runtime": 1931.491,
      "train_tokens_per_second": 5958.365
    },
    {
      "epoch": 0.7017931034482758,
      "grad_norm": 0.7888707518577576,
      "learning_rate": 0.00012998343456653783,
      "loss": 1.8384,
      "num_input_tokens_seen": 11513488,
      "step": 1908,
      "train_runtime": 1932.342,
      "train_tokens_per_second": 5958.308
    },
    {
      "epoch": 0.7021609195402299,
      "grad_norm": 0.7353121638298035,
      "learning_rate": 0.00012994662249217744,
      "loss": 1.5916,
      "num_input_tokens_seen": 11519328,
      "step": 1909,
      "train_runtime": 1933.2922,
      "train_tokens_per_second": 5958.4
    },
    {
      "epoch": 0.7025287356321839,
      "grad_norm": 0.7697341442108154,
      "learning_rate": 0.00012990981041781705,
      "loss": 1.8178,
      "num_input_tokens_seen": 11525088,
      "step": 1910,
      "train_runtime": 1934.2349,
      "train_tokens_per_second": 5958.474
    },
    {
      "epoch": 0.7028965517241379,
      "grad_norm": 0.6874209642410278,
      "learning_rate": 0.00012987299834345666,
      "loss": 1.944,
      "num_input_tokens_seen": 11532992,
      "step": 1911,
      "train_runtime": 1935.4912,
      "train_tokens_per_second": 5958.69
    },
    {
      "epoch": 0.7032643678160919,
      "grad_norm": 0.6833820939064026,
      "learning_rate": 0.00012983618626909627,
      "loss": 1.4793,
      "num_input_tokens_seen": 11538496,
      "step": 1912,
      "train_runtime": 1936.4092,
      "train_tokens_per_second": 5958.707
    },
    {
      "epoch": 0.7036321839080459,
      "grad_norm": 0.7527967095375061,
      "learning_rate": 0.0001297993741947359,
      "loss": 1.7455,
      "num_input_tokens_seen": 11544544,
      "step": 1913,
      "train_runtime": 1937.3877,
      "train_tokens_per_second": 5958.82
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.8245985507965088,
      "learning_rate": 0.00012976256212037548,
      "loss": 1.6053,
      "num_input_tokens_seen": 11549968,
      "step": 1914,
      "train_runtime": 1938.2868,
      "train_tokens_per_second": 5958.854
    },
    {
      "epoch": 0.704367816091954,
      "grad_norm": 0.7384008765220642,
      "learning_rate": 0.0001297257500460151,
      "loss": 1.8279,
      "num_input_tokens_seen": 11555328,
      "step": 1915,
      "train_runtime": 1939.1646,
      "train_tokens_per_second": 5958.921
    },
    {
      "epoch": 0.704735632183908,
      "grad_norm": 0.7503043413162231,
      "learning_rate": 0.0001296889379716547,
      "loss": 1.792,
      "num_input_tokens_seen": 11563120,
      "step": 1916,
      "train_runtime": 1940.3784,
      "train_tokens_per_second": 5959.209
    },
    {
      "epoch": 0.705103448275862,
      "grad_norm": 0.6756961345672607,
      "learning_rate": 0.0001296521258972943,
      "loss": 1.6234,
      "num_input_tokens_seen": 11569088,
      "step": 1917,
      "train_runtime": 1941.3312,
      "train_tokens_per_second": 5959.358
    },
    {
      "epoch": 0.705471264367816,
      "grad_norm": 0.792194128036499,
      "learning_rate": 0.00012961531382293392,
      "loss": 1.8971,
      "num_input_tokens_seen": 11574256,
      "step": 1918,
      "train_runtime": 1942.2034,
      "train_tokens_per_second": 5959.343
    },
    {
      "epoch": 0.7058390804597701,
      "grad_norm": 0.6884828209877014,
      "learning_rate": 0.00012957850174857353,
      "loss": 1.728,
      "num_input_tokens_seen": 11582288,
      "step": 1919,
      "train_runtime": 1943.4348,
      "train_tokens_per_second": 5959.7
    },
    {
      "epoch": 0.7062068965517241,
      "grad_norm": 0.6793946027755737,
      "learning_rate": 0.00012954168967421314,
      "loss": 1.5603,
      "num_input_tokens_seen": 11590112,
      "step": 1920,
      "train_runtime": 1944.6731,
      "train_tokens_per_second": 5959.928
    },
    {
      "epoch": 0.7065747126436782,
      "grad_norm": 0.7928634285926819,
      "learning_rate": 0.00012950487759985275,
      "loss": 1.7447,
      "num_input_tokens_seen": 11596208,
      "step": 1921,
      "train_runtime": 1946.1896,
      "train_tokens_per_second": 5958.416
    },
    {
      "epoch": 0.7069425287356322,
      "grad_norm": 0.8594651818275452,
      "learning_rate": 0.00012946806552549238,
      "loss": 1.9975,
      "num_input_tokens_seen": 11601296,
      "step": 1922,
      "train_runtime": 1947.0531,
      "train_tokens_per_second": 5958.387
    },
    {
      "epoch": 0.7073103448275863,
      "grad_norm": 0.8485431671142578,
      "learning_rate": 0.000129431253451132,
      "loss": 1.713,
      "num_input_tokens_seen": 11607616,
      "step": 1923,
      "train_runtime": 1948.0885,
      "train_tokens_per_second": 5958.464
    },
    {
      "epoch": 0.7076781609195403,
      "grad_norm": 0.7600086331367493,
      "learning_rate": 0.0001293944413767716,
      "loss": 1.8426,
      "num_input_tokens_seen": 11612528,
      "step": 1924,
      "train_runtime": 1948.9202,
      "train_tokens_per_second": 5958.442
    },
    {
      "epoch": 0.7080459770114943,
      "grad_norm": 0.8211314678192139,
      "learning_rate": 0.0001293576293024112,
      "loss": 1.8949,
      "num_input_tokens_seen": 11617152,
      "step": 1925,
      "train_runtime": 1949.7238,
      "train_tokens_per_second": 5958.358
    },
    {
      "epoch": 0.7084137931034483,
      "grad_norm": 0.615080714225769,
      "learning_rate": 0.00012932081722805082,
      "loss": 1.7059,
      "num_input_tokens_seen": 11626176,
      "step": 1926,
      "train_runtime": 1951.102,
      "train_tokens_per_second": 5958.774
    },
    {
      "epoch": 0.7087816091954023,
      "grad_norm": 0.7907627820968628,
      "learning_rate": 0.00012928400515369043,
      "loss": 1.6856,
      "num_input_tokens_seen": 11630752,
      "step": 1927,
      "train_runtime": 1951.8731,
      "train_tokens_per_second": 5958.764
    },
    {
      "epoch": 0.7091494252873564,
      "grad_norm": 0.7222896814346313,
      "learning_rate": 0.00012924719307933,
      "loss": 1.5589,
      "num_input_tokens_seen": 11635424,
      "step": 1928,
      "train_runtime": 1952.7017,
      "train_tokens_per_second": 5958.628
    },
    {
      "epoch": 0.7095172413793104,
      "grad_norm": 0.718788743019104,
      "learning_rate": 0.00012921038100496962,
      "loss": 1.5662,
      "num_input_tokens_seen": 11645120,
      "step": 1929,
      "train_runtime": 1954.1959,
      "train_tokens_per_second": 5959.034
    },
    {
      "epoch": 0.7098850574712644,
      "grad_norm": 0.7751200199127197,
      "learning_rate": 0.00012917356893060925,
      "loss": 1.6999,
      "num_input_tokens_seen": 11650368,
      "step": 1930,
      "train_runtime": 1955.0553,
      "train_tokens_per_second": 5959.099
    },
    {
      "epoch": 0.7102528735632184,
      "grad_norm": 0.8169651627540588,
      "learning_rate": 0.00012913675685624886,
      "loss": 1.6364,
      "num_input_tokens_seen": 11655040,
      "step": 1931,
      "train_runtime": 1955.8653,
      "train_tokens_per_second": 5959.02
    },
    {
      "epoch": 0.7106206896551724,
      "grad_norm": 0.6965715885162354,
      "learning_rate": 0.00012909994478188847,
      "loss": 1.648,
      "num_input_tokens_seen": 11662224,
      "step": 1932,
      "train_runtime": 1957.0072,
      "train_tokens_per_second": 5959.214
    },
    {
      "epoch": 0.7109885057471265,
      "grad_norm": 0.8124786615371704,
      "learning_rate": 0.00012906313270752808,
      "loss": 1.81,
      "num_input_tokens_seen": 11668608,
      "step": 1933,
      "train_runtime": 1958.0443,
      "train_tokens_per_second": 5959.318
    },
    {
      "epoch": 0.7113563218390805,
      "grad_norm": 0.7508975863456726,
      "learning_rate": 0.0001290263206331677,
      "loss": 1.6221,
      "num_input_tokens_seen": 11673376,
      "step": 1934,
      "train_runtime": 1958.8372,
      "train_tokens_per_second": 5959.339
    },
    {
      "epoch": 0.7117241379310345,
      "grad_norm": 0.79662024974823,
      "learning_rate": 0.0001289895085588073,
      "loss": 1.9828,
      "num_input_tokens_seen": 11677600,
      "step": 1935,
      "train_runtime": 1959.5708,
      "train_tokens_per_second": 5959.264
    },
    {
      "epoch": 0.7120919540229885,
      "grad_norm": 0.7343766093254089,
      "learning_rate": 0.0001289526964844469,
      "loss": 1.4957,
      "num_input_tokens_seen": 11684336,
      "step": 1936,
      "train_runtime": 1960.6409,
      "train_tokens_per_second": 5959.447
    },
    {
      "epoch": 0.7124597701149425,
      "grad_norm": 0.7641183733940125,
      "learning_rate": 0.00012891588441008652,
      "loss": 1.6359,
      "num_input_tokens_seen": 11689232,
      "step": 1937,
      "train_runtime": 1961.4639,
      "train_tokens_per_second": 5959.443
    },
    {
      "epoch": 0.7128275862068966,
      "grad_norm": 0.7237861752510071,
      "learning_rate": 0.00012887907233572612,
      "loss": 1.7829,
      "num_input_tokens_seen": 11693904,
      "step": 1938,
      "train_runtime": 1962.2659,
      "train_tokens_per_second": 5959.388
    },
    {
      "epoch": 0.7131954022988506,
      "grad_norm": 0.7369078397750854,
      "learning_rate": 0.00012884226026136573,
      "loss": 1.6614,
      "num_input_tokens_seen": 11700976,
      "step": 1939,
      "train_runtime": 1963.3925,
      "train_tokens_per_second": 5959.57
    },
    {
      "epoch": 0.7135632183908046,
      "grad_norm": 0.8303338289260864,
      "learning_rate": 0.00012880544818700534,
      "loss": 1.8742,
      "num_input_tokens_seen": 11706000,
      "step": 1940,
      "train_runtime": 1964.2572,
      "train_tokens_per_second": 5959.505
    },
    {
      "epoch": 0.7139310344827586,
      "grad_norm": 0.8005943298339844,
      "learning_rate": 0.00012876863611264495,
      "loss": 1.6352,
      "num_input_tokens_seen": 11711152,
      "step": 1941,
      "train_runtime": 1965.1197,
      "train_tokens_per_second": 5959.511
    },
    {
      "epoch": 0.7142988505747127,
      "grad_norm": 0.6863134503364563,
      "learning_rate": 0.00012873182403828456,
      "loss": 1.7554,
      "num_input_tokens_seen": 11716992,
      "step": 1942,
      "train_runtime": 1966.0904,
      "train_tokens_per_second": 5959.539
    },
    {
      "epoch": 0.7146666666666667,
      "grad_norm": 0.6974879503250122,
      "learning_rate": 0.00012869501196392417,
      "loss": 1.7619,
      "num_input_tokens_seen": 11725152,
      "step": 1943,
      "train_runtime": 1967.3838,
      "train_tokens_per_second": 5959.768
    },
    {
      "epoch": 0.7150344827586207,
      "grad_norm": 0.6026696562767029,
      "learning_rate": 0.00012865819988956378,
      "loss": 1.5532,
      "num_input_tokens_seen": 11734640,
      "step": 1944,
      "train_runtime": 1968.8399,
      "train_tokens_per_second": 5960.18
    },
    {
      "epoch": 0.7154022988505747,
      "grad_norm": 0.8114022016525269,
      "learning_rate": 0.00012862138781520339,
      "loss": 1.9169,
      "num_input_tokens_seen": 11738784,
      "step": 1945,
      "train_runtime": 1969.5843,
      "train_tokens_per_second": 5960.031
    },
    {
      "epoch": 0.7157701149425287,
      "grad_norm": 0.7067683339118958,
      "learning_rate": 0.000128584575740843,
      "loss": 1.7911,
      "num_input_tokens_seen": 11743680,
      "step": 1946,
      "train_runtime": 1970.4193,
      "train_tokens_per_second": 5959.99
    },
    {
      "epoch": 0.7161379310344828,
      "grad_norm": 0.7846788167953491,
      "learning_rate": 0.00012854776366648263,
      "loss": 1.7656,
      "num_input_tokens_seen": 11749936,
      "step": 1947,
      "train_runtime": 1971.4395,
      "train_tokens_per_second": 5960.079
    },
    {
      "epoch": 0.7165057471264368,
      "grad_norm": 0.7177066802978516,
      "learning_rate": 0.00012851095159212224,
      "loss": 1.6964,
      "num_input_tokens_seen": 11754080,
      "step": 1948,
      "train_runtime": 1972.1799,
      "train_tokens_per_second": 5959.943
    },
    {
      "epoch": 0.7168735632183908,
      "grad_norm": 0.7986196279525757,
      "learning_rate": 0.00012847413951776185,
      "loss": 1.952,
      "num_input_tokens_seen": 11759520,
      "step": 1949,
      "train_runtime": 1973.0671,
      "train_tokens_per_second": 5960.02
    },
    {
      "epoch": 0.7172413793103448,
      "grad_norm": 0.7428293228149414,
      "learning_rate": 0.00012843732744340146,
      "loss": 1.4596,
      "num_input_tokens_seen": 11766064,
      "step": 1950,
      "train_runtime": 1974.1306,
      "train_tokens_per_second": 5960.124
    },
    {
      "epoch": 0.7176091954022988,
      "grad_norm": 0.6595841646194458,
      "learning_rate": 0.00012840051536904104,
      "loss": 1.6227,
      "num_input_tokens_seen": 11773344,
      "step": 1951,
      "train_runtime": 1975.7998,
      "train_tokens_per_second": 5958.774
    },
    {
      "epoch": 0.7179770114942529,
      "grad_norm": 0.6061815023422241,
      "learning_rate": 0.00012836370329468065,
      "loss": 1.588,
      "num_input_tokens_seen": 11780144,
      "step": 1952,
      "train_runtime": 1976.9078,
      "train_tokens_per_second": 5958.874
    },
    {
      "epoch": 0.7183448275862069,
      "grad_norm": 0.6141793727874756,
      "learning_rate": 0.00012832689122032026,
      "loss": 1.4424,
      "num_input_tokens_seen": 11787712,
      "step": 1953,
      "train_runtime": 1978.1196,
      "train_tokens_per_second": 5959.049
    },
    {
      "epoch": 0.7187126436781609,
      "grad_norm": 0.7297881245613098,
      "learning_rate": 0.00012829007914595987,
      "loss": 1.8247,
      "num_input_tokens_seen": 11792032,
      "step": 1954,
      "train_runtime": 1978.8867,
      "train_tokens_per_second": 5958.922
    },
    {
      "epoch": 0.7190804597701149,
      "grad_norm": 0.8033981323242188,
      "learning_rate": 0.00012825326707159948,
      "loss": 1.855,
      "num_input_tokens_seen": 11797456,
      "step": 1955,
      "train_runtime": 1979.8012,
      "train_tokens_per_second": 5958.91
    },
    {
      "epoch": 0.7194482758620689,
      "grad_norm": 0.6842999458312988,
      "learning_rate": 0.0001282164549972391,
      "loss": 1.4373,
      "num_input_tokens_seen": 11803648,
      "step": 1956,
      "train_runtime": 1980.7901,
      "train_tokens_per_second": 5959.061
    },
    {
      "epoch": 0.719816091954023,
      "grad_norm": 0.7530986070632935,
      "learning_rate": 0.00012817964292287872,
      "loss": 1.9147,
      "num_input_tokens_seen": 11808464,
      "step": 1957,
      "train_runtime": 1981.5987,
      "train_tokens_per_second": 5959.059
    },
    {
      "epoch": 0.720183908045977,
      "grad_norm": 0.8145410418510437,
      "learning_rate": 0.00012814283084851833,
      "loss": 1.8343,
      "num_input_tokens_seen": 11813840,
      "step": 1958,
      "train_runtime": 1982.4734,
      "train_tokens_per_second": 5959.142
    },
    {
      "epoch": 0.720551724137931,
      "grad_norm": 0.754804790019989,
      "learning_rate": 0.00012810601877415794,
      "loss": 1.6074,
      "num_input_tokens_seen": 11819872,
      "step": 1959,
      "train_runtime": 1983.461,
      "train_tokens_per_second": 5959.216
    },
    {
      "epoch": 0.720919540229885,
      "grad_norm": 0.6832389831542969,
      "learning_rate": 0.00012806920669979755,
      "loss": 1.4985,
      "num_input_tokens_seen": 11826448,
      "step": 1960,
      "train_runtime": 1984.5438,
      "train_tokens_per_second": 5959.278
    },
    {
      "epoch": 0.721287356321839,
      "grad_norm": 0.8029295206069946,
      "learning_rate": 0.00012803239462543716,
      "loss": 1.6099,
      "num_input_tokens_seen": 11832944,
      "step": 1961,
      "train_runtime": 1985.6032,
      "train_tokens_per_second": 5959.37
    },
    {
      "epoch": 0.7216551724137931,
      "grad_norm": 0.8071450591087341,
      "learning_rate": 0.00012799558255107676,
      "loss": 2.0379,
      "num_input_tokens_seen": 11837680,
      "step": 1962,
      "train_runtime": 1986.3993,
      "train_tokens_per_second": 5959.366
    },
    {
      "epoch": 0.7220229885057471,
      "grad_norm": 0.7568224668502808,
      "learning_rate": 0.00012795877047671637,
      "loss": 1.6085,
      "num_input_tokens_seen": 11842832,
      "step": 1963,
      "train_runtime": 1987.2801,
      "train_tokens_per_second": 5959.317
    },
    {
      "epoch": 0.7223908045977011,
      "grad_norm": 0.7475882768630981,
      "learning_rate": 0.00012792195840235596,
      "loss": 1.8751,
      "num_input_tokens_seen": 11847632,
      "step": 1964,
      "train_runtime": 1988.1143,
      "train_tokens_per_second": 5959.231
    },
    {
      "epoch": 0.7227586206896551,
      "grad_norm": 0.7568109631538391,
      "learning_rate": 0.0001278851463279956,
      "loss": 1.6683,
      "num_input_tokens_seen": 11852496,
      "step": 1965,
      "train_runtime": 1988.9588,
      "train_tokens_per_second": 5959.146
    },
    {
      "epoch": 0.7231264367816091,
      "grad_norm": 0.8110637664794922,
      "learning_rate": 0.0001278483342536352,
      "loss": 1.7528,
      "num_input_tokens_seen": 11857392,
      "step": 1966,
      "train_runtime": 1989.7776,
      "train_tokens_per_second": 5959.154
    },
    {
      "epoch": 0.7234942528735632,
      "grad_norm": 0.7403090596199036,
      "learning_rate": 0.0001278115221792748,
      "loss": 1.8209,
      "num_input_tokens_seen": 11863360,
      "step": 1967,
      "train_runtime": 1990.7431,
      "train_tokens_per_second": 5959.262
    },
    {
      "epoch": 0.7238620689655172,
      "grad_norm": 0.8106614947319031,
      "learning_rate": 0.00012777471010491442,
      "loss": 1.817,
      "num_input_tokens_seen": 11868656,
      "step": 1968,
      "train_runtime": 1991.6255,
      "train_tokens_per_second": 5959.281
    },
    {
      "epoch": 0.7242298850574712,
      "grad_norm": 0.7311268448829651,
      "learning_rate": 0.00012773789803055403,
      "loss": 1.5936,
      "num_input_tokens_seen": 11878192,
      "step": 1969,
      "train_runtime": 1993.0743,
      "train_tokens_per_second": 5959.734
    },
    {
      "epoch": 0.7245977011494252,
      "grad_norm": 0.8467444777488708,
      "learning_rate": 0.00012770108595619364,
      "loss": 1.7505,
      "num_input_tokens_seen": 11883472,
      "step": 1970,
      "train_runtime": 1993.959,
      "train_tokens_per_second": 5959.737
    },
    {
      "epoch": 0.7249655172413794,
      "grad_norm": 0.7418035864830017,
      "learning_rate": 0.00012766427388183324,
      "loss": 1.5061,
      "num_input_tokens_seen": 11893072,
      "step": 1971,
      "train_runtime": 1995.438,
      "train_tokens_per_second": 5960.131
    },
    {
      "epoch": 0.7253333333333334,
      "grad_norm": 0.8001358509063721,
      "learning_rate": 0.00012762746180747285,
      "loss": 1.8678,
      "num_input_tokens_seen": 11897568,
      "step": 1972,
      "train_runtime": 1996.224,
      "train_tokens_per_second": 5960.036
    },
    {
      "epoch": 0.7257011494252874,
      "grad_norm": 0.7486362457275391,
      "learning_rate": 0.0001275906497331125,
      "loss": 1.639,
      "num_input_tokens_seen": 11904384,
      "step": 1973,
      "train_runtime": 1997.3442,
      "train_tokens_per_second": 5960.106
    },
    {
      "epoch": 0.7260689655172414,
      "grad_norm": 0.7364412546157837,
      "learning_rate": 0.00012755383765875207,
      "loss": 1.5826,
      "num_input_tokens_seen": 11909536,
      "step": 1974,
      "train_runtime": 1998.2081,
      "train_tokens_per_second": 5960.108
    },
    {
      "epoch": 0.7264367816091954,
      "grad_norm": 0.9112614393234253,
      "learning_rate": 0.00012751702558439168,
      "loss": 1.7799,
      "num_input_tokens_seen": 11913888,
      "step": 1975,
      "train_runtime": 1998.9613,
      "train_tokens_per_second": 5960.039
    },
    {
      "epoch": 0.7268045977011495,
      "grad_norm": 0.8161247372627258,
      "learning_rate": 0.0001274802135100313,
      "loss": 1.8897,
      "num_input_tokens_seen": 11919296,
      "step": 1976,
      "train_runtime": 1999.8523,
      "train_tokens_per_second": 5960.088
    },
    {
      "epoch": 0.7271724137931035,
      "grad_norm": 0.7600432634353638,
      "learning_rate": 0.0001274434014356709,
      "loss": 1.7352,
      "num_input_tokens_seen": 11926032,
      "step": 1977,
      "train_runtime": 2000.9307,
      "train_tokens_per_second": 5960.242
    },
    {
      "epoch": 0.7275402298850575,
      "grad_norm": 0.7857576608657837,
      "learning_rate": 0.0001274065893613105,
      "loss": 2.3608,
      "num_input_tokens_seen": 11931232,
      "step": 1978,
      "train_runtime": 2001.788,
      "train_tokens_per_second": 5960.288
    },
    {
      "epoch": 0.7279080459770115,
      "grad_norm": 0.8168516755104065,
      "learning_rate": 0.00012736977728695012,
      "loss": 1.8073,
      "num_input_tokens_seen": 11936000,
      "step": 1979,
      "train_runtime": 2002.6203,
      "train_tokens_per_second": 5960.191
    },
    {
      "epoch": 0.7282758620689656,
      "grad_norm": 0.748710572719574,
      "learning_rate": 0.00012733296521258972,
      "loss": 1.666,
      "num_input_tokens_seen": 11941696,
      "step": 1980,
      "train_runtime": 2003.548,
      "train_tokens_per_second": 5960.274
    },
    {
      "epoch": 0.7286436781609196,
      "grad_norm": 0.6982470750808716,
      "learning_rate": 0.00012729615313822933,
      "loss": 1.8343,
      "num_input_tokens_seen": 11948992,
      "step": 1981,
      "train_runtime": 2005.1802,
      "train_tokens_per_second": 5959.061
    },
    {
      "epoch": 0.7290114942528736,
      "grad_norm": 0.9832120537757874,
      "learning_rate": 0.00012725934106386897,
      "loss": 1.8208,
      "num_input_tokens_seen": 11953600,
      "step": 1982,
      "train_runtime": 2005.9987,
      "train_tokens_per_second": 5958.927
    },
    {
      "epoch": 0.7293793103448276,
      "grad_norm": 0.7828192710876465,
      "learning_rate": 0.00012722252898950858,
      "loss": 1.613,
      "num_input_tokens_seen": 11961408,
      "step": 1983,
      "train_runtime": 2007.2128,
      "train_tokens_per_second": 5959.213
    },
    {
      "epoch": 0.7297471264367816,
      "grad_norm": 0.7174785137176514,
      "learning_rate": 0.0001271857169151482,
      "loss": 1.7013,
      "num_input_tokens_seen": 11966896,
      "step": 1984,
      "train_runtime": 2008.1356,
      "train_tokens_per_second": 5959.207
    },
    {
      "epoch": 0.7301149425287357,
      "grad_norm": 0.7602179050445557,
      "learning_rate": 0.0001271489048407878,
      "loss": 1.6543,
      "num_input_tokens_seen": 11972416,
      "step": 1985,
      "train_runtime": 2009.0761,
      "train_tokens_per_second": 5959.165
    },
    {
      "epoch": 0.7304827586206897,
      "grad_norm": 0.8048250675201416,
      "learning_rate": 0.0001271120927664274,
      "loss": 1.8423,
      "num_input_tokens_seen": 11977072,
      "step": 1986,
      "train_runtime": 2009.8775,
      "train_tokens_per_second": 5959.105
    },
    {
      "epoch": 0.7308505747126437,
      "grad_norm": 0.8352282643318176,
      "learning_rate": 0.000127075280692067,
      "loss": 1.8687,
      "num_input_tokens_seen": 11982752,
      "step": 1987,
      "train_runtime": 2010.8086,
      "train_tokens_per_second": 5959.171
    },
    {
      "epoch": 0.7312183908045977,
      "grad_norm": 0.7257148027420044,
      "learning_rate": 0.0001270384686177066,
      "loss": 1.8688,
      "num_input_tokens_seen": 11987664,
      "step": 1988,
      "train_runtime": 2011.6541,
      "train_tokens_per_second": 5959.108
    },
    {
      "epoch": 0.7315862068965517,
      "grad_norm": 0.7688496708869934,
      "learning_rate": 0.0001270016565433462,
      "loss": 1.8043,
      "num_input_tokens_seen": 11992656,
      "step": 1989,
      "train_runtime": 2012.5131,
      "train_tokens_per_second": 5959.045
    },
    {
      "epoch": 0.7319540229885058,
      "grad_norm": 0.7292170524597168,
      "learning_rate": 0.00012696484446898584,
      "loss": 2.0214,
      "num_input_tokens_seen": 12000000,
      "step": 1990,
      "train_runtime": 2013.6972,
      "train_tokens_per_second": 5959.188
    },
    {
      "epoch": 0.7323218390804598,
      "grad_norm": 0.6529267430305481,
      "learning_rate": 0.00012692803239462545,
      "loss": 1.6034,
      "num_input_tokens_seen": 12005648,
      "step": 1991,
      "train_runtime": 2014.6269,
      "train_tokens_per_second": 5959.241
    },
    {
      "epoch": 0.7326896551724138,
      "grad_norm": 0.703542947769165,
      "learning_rate": 0.00012689122032026506,
      "loss": 1.8618,
      "num_input_tokens_seen": 12011552,
      "step": 1992,
      "train_runtime": 2015.5873,
      "train_tokens_per_second": 5959.331
    },
    {
      "epoch": 0.7330574712643678,
      "grad_norm": 0.7824158668518066,
      "learning_rate": 0.00012685440824590467,
      "loss": 1.8234,
      "num_input_tokens_seen": 12016400,
      "step": 1993,
      "train_runtime": 2016.4292,
      "train_tokens_per_second": 5959.247
    },
    {
      "epoch": 0.7334252873563218,
      "grad_norm": 0.7852529883384705,
      "learning_rate": 0.00012681759617154428,
      "loss": 1.9107,
      "num_input_tokens_seen": 12021936,
      "step": 1994,
      "train_runtime": 2017.3475,
      "train_tokens_per_second": 5959.279
    },
    {
      "epoch": 0.7337931034482759,
      "grad_norm": 0.6936224102973938,
      "learning_rate": 0.00012678078409718389,
      "loss": 1.9709,
      "num_input_tokens_seen": 12028400,
      "step": 1995,
      "train_runtime": 2018.4101,
      "train_tokens_per_second": 5959.344
    },
    {
      "epoch": 0.7341609195402299,
      "grad_norm": 0.744227945804596,
      "learning_rate": 0.0001267439720228235,
      "loss": 1.6797,
      "num_input_tokens_seen": 12036320,
      "step": 1996,
      "train_runtime": 2019.6516,
      "train_tokens_per_second": 5959.602
    },
    {
      "epoch": 0.7345287356321839,
      "grad_norm": 0.7873906493186951,
      "learning_rate": 0.0001267071599484631,
      "loss": 1.6767,
      "num_input_tokens_seen": 12041424,
      "step": 1997,
      "train_runtime": 2020.49,
      "train_tokens_per_second": 5959.655
    },
    {
      "epoch": 0.7348965517241379,
      "grad_norm": 0.644244909286499,
      "learning_rate": 0.0001266703478741027,
      "loss": 1.7367,
      "num_input_tokens_seen": 12047792,
      "step": 1998,
      "train_runtime": 2021.5059,
      "train_tokens_per_second": 5959.81
    },
    {
      "epoch": 0.735264367816092,
      "grad_norm": 0.7175727486610413,
      "learning_rate": 0.00012663353579974232,
      "loss": 1.847,
      "num_input_tokens_seen": 12054928,
      "step": 1999,
      "train_runtime": 2022.655,
      "train_tokens_per_second": 5959.953
    },
    {
      "epoch": 0.735632183908046,
      "grad_norm": 0.6197425127029419,
      "learning_rate": 0.00012659672372538193,
      "loss": 1.8297,
      "num_input_tokens_seen": 12061664,
      "step": 2000,
      "train_runtime": 2023.742,
      "train_tokens_per_second": 5960.08
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.7840290069580078,
      "learning_rate": 0.00012655991165102154,
      "loss": 1.8551,
      "num_input_tokens_seen": 12066368,
      "step": 2001,
      "train_runtime": 2024.5934,
      "train_tokens_per_second": 5959.897
    },
    {
      "epoch": 0.736367816091954,
      "grad_norm": 0.6831390857696533,
      "learning_rate": 0.00012652309957666115,
      "loss": 1.703,
      "num_input_tokens_seen": 12072608,
      "step": 2002,
      "train_runtime": 2025.605,
      "train_tokens_per_second": 5960.001
    },
    {
      "epoch": 0.736735632183908,
      "grad_norm": 0.7383309602737427,
      "learning_rate": 0.00012648628750230076,
      "loss": 1.7188,
      "num_input_tokens_seen": 12082176,
      "step": 2003,
      "train_runtime": 2027.0885,
      "train_tokens_per_second": 5960.359
    },
    {
      "epoch": 0.737103448275862,
      "grad_norm": 0.6271421909332275,
      "learning_rate": 0.00012644947542794037,
      "loss": 1.5585,
      "num_input_tokens_seen": 12087824,
      "step": 2004,
      "train_runtime": 2028.0339,
      "train_tokens_per_second": 5960.366
    },
    {
      "epoch": 0.7374712643678161,
      "grad_norm": 0.7438689470291138,
      "learning_rate": 0.00012641266335357997,
      "loss": 1.5752,
      "num_input_tokens_seen": 12094032,
      "step": 2005,
      "train_runtime": 2029.0501,
      "train_tokens_per_second": 5960.44
    },
    {
      "epoch": 0.7378390804597701,
      "grad_norm": 0.7262468934059143,
      "learning_rate": 0.00012637585127921958,
      "loss": 1.7402,
      "num_input_tokens_seen": 12098864,
      "step": 2006,
      "train_runtime": 2029.8621,
      "train_tokens_per_second": 5960.436
    },
    {
      "epoch": 0.7382068965517241,
      "grad_norm": 0.7145463228225708,
      "learning_rate": 0.00012633903920485922,
      "loss": 1.8283,
      "num_input_tokens_seen": 12105168,
      "step": 2007,
      "train_runtime": 2030.8815,
      "train_tokens_per_second": 5960.549
    },
    {
      "epoch": 0.7385747126436781,
      "grad_norm": 0.7067509293556213,
      "learning_rate": 0.00012630222713049883,
      "loss": 1.5464,
      "num_input_tokens_seen": 12113472,
      "step": 2008,
      "train_runtime": 2032.185,
      "train_tokens_per_second": 5960.812
    },
    {
      "epoch": 0.7389425287356322,
      "grad_norm": 0.7784838080406189,
      "learning_rate": 0.00012626541505613844,
      "loss": 1.693,
      "num_input_tokens_seen": 12118912,
      "step": 2009,
      "train_runtime": 2033.0862,
      "train_tokens_per_second": 5960.845
    },
    {
      "epoch": 0.7393103448275862,
      "grad_norm": 0.8181419968605042,
      "learning_rate": 0.00012622860298177805,
      "loss": 1.7745,
      "num_input_tokens_seen": 12123984,
      "step": 2010,
      "train_runtime": 2033.9546,
      "train_tokens_per_second": 5960.794
    },
    {
      "epoch": 0.7396781609195402,
      "grad_norm": 0.7721001505851746,
      "learning_rate": 0.00012619179090741763,
      "loss": 1.67,
      "num_input_tokens_seen": 12128640,
      "step": 2011,
      "train_runtime": 2035.2192,
      "train_tokens_per_second": 5959.378
    },
    {
      "epoch": 0.7400459770114942,
      "grad_norm": 0.8139654994010925,
      "learning_rate": 0.00012615497883305724,
      "loss": 1.7246,
      "num_input_tokens_seen": 12134400,
      "step": 2012,
      "train_runtime": 2036.1615,
      "train_tokens_per_second": 5959.449
    },
    {
      "epoch": 0.7404137931034482,
      "grad_norm": 0.8119025826454163,
      "learning_rate": 0.00012611816675869685,
      "loss": 1.9269,
      "num_input_tokens_seen": 12138720,
      "step": 2013,
      "train_runtime": 2036.9013,
      "train_tokens_per_second": 5959.405
    },
    {
      "epoch": 0.7407816091954023,
      "grad_norm": 0.8293254971504211,
      "learning_rate": 0.00012608135468433645,
      "loss": 1.7921,
      "num_input_tokens_seen": 12143136,
      "step": 2014,
      "train_runtime": 2037.6792,
      "train_tokens_per_second": 5959.297
    },
    {
      "epoch": 0.7411494252873563,
      "grad_norm": 0.7966312766075134,
      "learning_rate": 0.00012604454260997606,
      "loss": 1.8548,
      "num_input_tokens_seen": 12148224,
      "step": 2015,
      "train_runtime": 2038.5389,
      "train_tokens_per_second": 5959.28
    },
    {
      "epoch": 0.7415172413793103,
      "grad_norm": 0.7699953317642212,
      "learning_rate": 0.0001260077305356157,
      "loss": 2.0045,
      "num_input_tokens_seen": 12154496,
      "step": 2016,
      "train_runtime": 2039.571,
      "train_tokens_per_second": 5959.339
    },
    {
      "epoch": 0.7418850574712643,
      "grad_norm": 0.7591118216514587,
      "learning_rate": 0.0001259709184612553,
      "loss": 1.8046,
      "num_input_tokens_seen": 12161296,
      "step": 2017,
      "train_runtime": 2040.6988,
      "train_tokens_per_second": 5959.378
    },
    {
      "epoch": 0.7422528735632183,
      "grad_norm": 0.7616413831710815,
      "learning_rate": 0.00012593410638689492,
      "loss": 1.9069,
      "num_input_tokens_seen": 12167056,
      "step": 2018,
      "train_runtime": 2041.6362,
      "train_tokens_per_second": 5959.463
    },
    {
      "epoch": 0.7426206896551724,
      "grad_norm": 0.7836195230484009,
      "learning_rate": 0.00012589729431253453,
      "loss": 1.8534,
      "num_input_tokens_seen": 12172720,
      "step": 2019,
      "train_runtime": 2042.5845,
      "train_tokens_per_second": 5959.469
    },
    {
      "epoch": 0.7429885057471264,
      "grad_norm": 0.7529764175415039,
      "learning_rate": 0.00012586048223817413,
      "loss": 1.7127,
      "num_input_tokens_seen": 12182560,
      "step": 2020,
      "train_runtime": 2044.0916,
      "train_tokens_per_second": 5959.89
    },
    {
      "epoch": 0.7433563218390805,
      "grad_norm": 0.6801377534866333,
      "learning_rate": 0.00012582367016381374,
      "loss": 1.4794,
      "num_input_tokens_seen": 12191040,
      "step": 2021,
      "train_runtime": 2045.3843,
      "train_tokens_per_second": 5960.269
    },
    {
      "epoch": 0.7437241379310345,
      "grad_norm": 0.721573531627655,
      "learning_rate": 0.00012578685808945335,
      "loss": 1.6391,
      "num_input_tokens_seen": 12197312,
      "step": 2022,
      "train_runtime": 2046.4215,
      "train_tokens_per_second": 5960.313
    },
    {
      "epoch": 0.7440919540229886,
      "grad_norm": 0.6895381808280945,
      "learning_rate": 0.00012575004601509296,
      "loss": 1.8129,
      "num_input_tokens_seen": 12202880,
      "step": 2023,
      "train_runtime": 2047.3434,
      "train_tokens_per_second": 5960.348
    },
    {
      "epoch": 0.7444597701149426,
      "grad_norm": 0.7531507611274719,
      "learning_rate": 0.00012571323394073254,
      "loss": 1.721,
      "num_input_tokens_seen": 12207936,
      "step": 2024,
      "train_runtime": 2048.1965,
      "train_tokens_per_second": 5960.334
    },
    {
      "epoch": 0.7448275862068966,
      "grad_norm": 0.7214846014976501,
      "learning_rate": 0.00012567642186637218,
      "loss": 1.7397,
      "num_input_tokens_seen": 12214976,
      "step": 2025,
      "train_runtime": 2049.3395,
      "train_tokens_per_second": 5960.445
    },
    {
      "epoch": 0.7451954022988506,
      "grad_norm": 0.7114675045013428,
      "learning_rate": 0.0001256396097920118,
      "loss": 1.6025,
      "num_input_tokens_seen": 12221200,
      "step": 2026,
      "train_runtime": 2050.3778,
      "train_tokens_per_second": 5960.462
    },
    {
      "epoch": 0.7455632183908046,
      "grad_norm": 0.8285331130027771,
      "learning_rate": 0.0001256027977176514,
      "loss": 1.9333,
      "num_input_tokens_seen": 12227776,
      "step": 2027,
      "train_runtime": 2051.4502,
      "train_tokens_per_second": 5960.552
    },
    {
      "epoch": 0.7459310344827587,
      "grad_norm": 0.7888330817222595,
      "learning_rate": 0.000125565985643291,
      "loss": 2.0145,
      "num_input_tokens_seen": 12232592,
      "step": 2028,
      "train_runtime": 2052.2486,
      "train_tokens_per_second": 5960.58
    },
    {
      "epoch": 0.7462988505747127,
      "grad_norm": 0.7405606508255005,
      "learning_rate": 0.00012552917356893061,
      "loss": 1.6829,
      "num_input_tokens_seen": 12239440,
      "step": 2029,
      "train_runtime": 2053.3713,
      "train_tokens_per_second": 5960.656
    },
    {
      "epoch": 0.7466666666666667,
      "grad_norm": 0.7686442732810974,
      "learning_rate": 0.00012549236149457022,
      "loss": 1.8015,
      "num_input_tokens_seen": 12244496,
      "step": 2030,
      "train_runtime": 2054.2121,
      "train_tokens_per_second": 5960.677
    },
    {
      "epoch": 0.7470344827586207,
      "grad_norm": 0.8337616324424744,
      "learning_rate": 0.00012545554942020983,
      "loss": 1.9287,
      "num_input_tokens_seen": 12248944,
      "step": 2031,
      "train_runtime": 2054.9783,
      "train_tokens_per_second": 5960.62
    },
    {
      "epoch": 0.7474022988505747,
      "grad_norm": 0.9073026776313782,
      "learning_rate": 0.00012541873734584944,
      "loss": 1.8618,
      "num_input_tokens_seen": 12256256,
      "step": 2032,
      "train_runtime": 2056.1338,
      "train_tokens_per_second": 5960.826
    },
    {
      "epoch": 0.7477701149425288,
      "grad_norm": 0.7766493558883667,
      "learning_rate": 0.00012538192527148908,
      "loss": 1.8335,
      "num_input_tokens_seen": 12260656,
      "step": 2033,
      "train_runtime": 2056.889,
      "train_tokens_per_second": 5960.777
    },
    {
      "epoch": 0.7481379310344828,
      "grad_norm": 0.7368164658546448,
      "learning_rate": 0.00012534511319712866,
      "loss": 1.9069,
      "num_input_tokens_seen": 12266432,
      "step": 2034,
      "train_runtime": 2057.8409,
      "train_tokens_per_second": 5960.826
    },
    {
      "epoch": 0.7485057471264368,
      "grad_norm": 0.7862488031387329,
      "learning_rate": 0.00012530830112276827,
      "loss": 1.8104,
      "num_input_tokens_seen": 12272448,
      "step": 2035,
      "train_runtime": 2058.8031,
      "train_tokens_per_second": 5960.962
    },
    {
      "epoch": 0.7488735632183908,
      "grad_norm": 0.7069132924079895,
      "learning_rate": 0.00012527148904840788,
      "loss": 1.715,
      "num_input_tokens_seen": 12277568,
      "step": 2036,
      "train_runtime": 2059.6927,
      "train_tokens_per_second": 5960.874
    },
    {
      "epoch": 0.7492413793103448,
      "grad_norm": 0.910798192024231,
      "learning_rate": 0.00012523467697404749,
      "loss": 1.8036,
      "num_input_tokens_seen": 12282144,
      "step": 2037,
      "train_runtime": 2060.4815,
      "train_tokens_per_second": 5960.813
    },
    {
      "epoch": 0.7496091954022989,
      "grad_norm": 0.7396913170814514,
      "learning_rate": 0.0001251978648996871,
      "loss": 1.6484,
      "num_input_tokens_seen": 12287152,
      "step": 2038,
      "train_runtime": 2061.3257,
      "train_tokens_per_second": 5960.801
    },
    {
      "epoch": 0.7499770114942529,
      "grad_norm": 0.7725208401679993,
      "learning_rate": 0.0001251610528253267,
      "loss": 1.7524,
      "num_input_tokens_seen": 12293136,
      "step": 2039,
      "train_runtime": 2062.3164,
      "train_tokens_per_second": 5960.839
    },
    {
      "epoch": 0.7503448275862069,
      "grad_norm": 0.6665602922439575,
      "learning_rate": 0.0001251242407509663,
      "loss": 1.6469,
      "num_input_tokens_seen": 12301552,
      "step": 2040,
      "train_runtime": 2063.6231,
      "train_tokens_per_second": 5961.143
    },
    {
      "epoch": 0.7507126436781609,
      "grad_norm": 0.7234289050102234,
      "learning_rate": 0.00012508742867660592,
      "loss": 1.5413,
      "num_input_tokens_seen": 12307536,
      "step": 2041,
      "train_runtime": 2065.1515,
      "train_tokens_per_second": 5959.629
    },
    {
      "epoch": 0.751080459770115,
      "grad_norm": 0.76047283411026,
      "learning_rate": 0.00012505061660224556,
      "loss": 1.7646,
      "num_input_tokens_seen": 12312768,
      "step": 2042,
      "train_runtime": 2066.0224,
      "train_tokens_per_second": 5959.649
    },
    {
      "epoch": 0.751448275862069,
      "grad_norm": 0.7163563966751099,
      "learning_rate": 0.00012501380452788517,
      "loss": 1.5964,
      "num_input_tokens_seen": 12319168,
      "step": 2043,
      "train_runtime": 2067.0654,
      "train_tokens_per_second": 5959.738
    },
    {
      "epoch": 0.751816091954023,
      "grad_norm": 0.8260768055915833,
      "learning_rate": 0.00012497699245352478,
      "loss": 1.8702,
      "num_input_tokens_seen": 12323648,
      "step": 2044,
      "train_runtime": 2067.8419,
      "train_tokens_per_second": 5959.666
    },
    {
      "epoch": 0.752183908045977,
      "grad_norm": 0.6744913458824158,
      "learning_rate": 0.00012494018037916438,
      "loss": 1.7112,
      "num_input_tokens_seen": 12331184,
      "step": 2045,
      "train_runtime": 2069.0505,
      "train_tokens_per_second": 5959.828
    },
    {
      "epoch": 0.752551724137931,
      "grad_norm": 0.7469673156738281,
      "learning_rate": 0.000124903368304804,
      "loss": 1.5905,
      "num_input_tokens_seen": 12338928,
      "step": 2046,
      "train_runtime": 2070.2474,
      "train_tokens_per_second": 5960.122
    },
    {
      "epoch": 0.752919540229885,
      "grad_norm": 0.7135205268859863,
      "learning_rate": 0.00012486655623044357,
      "loss": 1.7349,
      "num_input_tokens_seen": 12343824,
      "step": 2047,
      "train_runtime": 2071.0714,
      "train_tokens_per_second": 5960.115
    },
    {
      "epoch": 0.7532873563218391,
      "grad_norm": 0.7622992396354675,
      "learning_rate": 0.00012482974415608318,
      "loss": 1.6711,
      "num_input_tokens_seen": 12348448,
      "step": 2048,
      "train_runtime": 2071.8711,
      "train_tokens_per_second": 5960.046
    },
    {
      "epoch": 0.7536551724137931,
      "grad_norm": 0.7797056436538696,
      "learning_rate": 0.0001247929320817228,
      "loss": 1.8897,
      "num_input_tokens_seen": 12354400,
      "step": 2049,
      "train_runtime": 2072.8584,
      "train_tokens_per_second": 5960.079
    },
    {
      "epoch": 0.7540229885057471,
      "grad_norm": 0.7252853512763977,
      "learning_rate": 0.00012475612000736243,
      "loss": 1.6295,
      "num_input_tokens_seen": 12360608,
      "step": 2050,
      "train_runtime": 2073.8961,
      "train_tokens_per_second": 5960.09
    },
    {
      "epoch": 0.7543908045977011,
      "grad_norm": 0.6995819807052612,
      "learning_rate": 0.00012471930793300204,
      "loss": 1.7527,
      "num_input_tokens_seen": 12367920,
      "step": 2051,
      "train_runtime": 2075.0452,
      "train_tokens_per_second": 5960.314
    },
    {
      "epoch": 0.7547586206896552,
      "grad_norm": 0.7781098484992981,
      "learning_rate": 0.00012468249585864165,
      "loss": 1.687,
      "num_input_tokens_seen": 12374560,
      "step": 2052,
      "train_runtime": 2076.1121,
      "train_tokens_per_second": 5960.449
    },
    {
      "epoch": 0.7551264367816092,
      "grad_norm": 0.6926596760749817,
      "learning_rate": 0.00012464568378428126,
      "loss": 1.6659,
      "num_input_tokens_seen": 12384880,
      "step": 2053,
      "train_runtime": 2077.6468,
      "train_tokens_per_second": 5961.013
    },
    {
      "epoch": 0.7554942528735632,
      "grad_norm": 0.8198338747024536,
      "learning_rate": 0.00012460887170992086,
      "loss": 1.7728,
      "num_input_tokens_seen": 12392736,
      "step": 2054,
      "train_runtime": 2078.8786,
      "train_tokens_per_second": 5961.26
    },
    {
      "epoch": 0.7558620689655172,
      "grad_norm": 0.7452566027641296,
      "learning_rate": 0.00012457205963556047,
      "loss": 1.8164,
      "num_input_tokens_seen": 12402096,
      "step": 2055,
      "train_runtime": 2080.2999,
      "train_tokens_per_second": 5961.687
    },
    {
      "epoch": 0.7562298850574712,
      "grad_norm": 0.7812275886535645,
      "learning_rate": 0.00012453524756120008,
      "loss": 1.7473,
      "num_input_tokens_seen": 12407424,
      "step": 2056,
      "train_runtime": 2081.2145,
      "train_tokens_per_second": 5961.627
    },
    {
      "epoch": 0.7565977011494253,
      "grad_norm": 0.8216825127601624,
      "learning_rate": 0.0001244984354868397,
      "loss": 1.7814,
      "num_input_tokens_seen": 12414384,
      "step": 2057,
      "train_runtime": 2082.3231,
      "train_tokens_per_second": 5961.795
    },
    {
      "epoch": 0.7569655172413793,
      "grad_norm": 0.7050502896308899,
      "learning_rate": 0.0001244616234124793,
      "loss": 1.574,
      "num_input_tokens_seen": 12423104,
      "step": 2058,
      "train_runtime": 2083.6656,
      "train_tokens_per_second": 5962.139
    },
    {
      "epoch": 0.7573333333333333,
      "grad_norm": 0.7369188070297241,
      "learning_rate": 0.0001244248113381189,
      "loss": 1.7756,
      "num_input_tokens_seen": 12427824,
      "step": 2059,
      "train_runtime": 2084.4768,
      "train_tokens_per_second": 5962.083
    },
    {
      "epoch": 0.7577011494252873,
      "grad_norm": 0.7803534269332886,
      "learning_rate": 0.00012438799926375852,
      "loss": 1.8057,
      "num_input_tokens_seen": 12433264,
      "step": 2060,
      "train_runtime": 2085.3858,
      "train_tokens_per_second": 5962.093
    },
    {
      "epoch": 0.7580689655172413,
      "grad_norm": 0.7051107287406921,
      "learning_rate": 0.00012435118718939813,
      "loss": 1.4501,
      "num_input_tokens_seen": 12441856,
      "step": 2061,
      "train_runtime": 2086.6988,
      "train_tokens_per_second": 5962.459
    },
    {
      "epoch": 0.7584367816091954,
      "grad_norm": 0.8329285383224487,
      "learning_rate": 0.00012431437511503774,
      "loss": 1.7789,
      "num_input_tokens_seen": 12446368,
      "step": 2062,
      "train_runtime": 2087.4846,
      "train_tokens_per_second": 5962.376
    },
    {
      "epoch": 0.7588045977011494,
      "grad_norm": 0.7532010674476624,
      "learning_rate": 0.00012427756304067734,
      "loss": 1.5868,
      "num_input_tokens_seen": 12451728,
      "step": 2063,
      "train_runtime": 2088.3826,
      "train_tokens_per_second": 5962.379
    },
    {
      "epoch": 0.7591724137931034,
      "grad_norm": 0.744062602519989,
      "learning_rate": 0.00012424075096631695,
      "loss": 1.7547,
      "num_input_tokens_seen": 12458624,
      "step": 2064,
      "train_runtime": 2089.5056,
      "train_tokens_per_second": 5962.475
    },
    {
      "epoch": 0.7595402298850574,
      "grad_norm": 0.749465823173523,
      "learning_rate": 0.00012420393889195656,
      "loss": 1.7925,
      "num_input_tokens_seen": 12466032,
      "step": 2065,
      "train_runtime": 2090.6747,
      "train_tokens_per_second": 5962.684
    },
    {
      "epoch": 0.7599080459770114,
      "grad_norm": 0.7444941997528076,
      "learning_rate": 0.00012416712681759617,
      "loss": 1.5725,
      "num_input_tokens_seen": 12472576,
      "step": 2066,
      "train_runtime": 2091.7511,
      "train_tokens_per_second": 5962.744
    },
    {
      "epoch": 0.7602758620689655,
      "grad_norm": 0.7577369809150696,
      "learning_rate": 0.0001241303147432358,
      "loss": 1.6854,
      "num_input_tokens_seen": 12479296,
      "step": 2067,
      "train_runtime": 2092.8088,
      "train_tokens_per_second": 5962.941
    },
    {
      "epoch": 0.7606436781609195,
      "grad_norm": 0.8165404200553894,
      "learning_rate": 0.00012409350266887542,
      "loss": 1.6778,
      "num_input_tokens_seen": 12488464,
      "step": 2068,
      "train_runtime": 2094.1946,
      "train_tokens_per_second": 5963.373
    },
    {
      "epoch": 0.7610114942528735,
      "grad_norm": 0.7348002791404724,
      "learning_rate": 0.00012405669059451502,
      "loss": 1.6465,
      "num_input_tokens_seen": 12493792,
      "step": 2069,
      "train_runtime": 2095.1024,
      "train_tokens_per_second": 5963.332
    },
    {
      "epoch": 0.7613793103448275,
      "grad_norm": 0.8466042280197144,
      "learning_rate": 0.0001240198785201546,
      "loss": 1.83,
      "num_input_tokens_seen": 12498736,
      "step": 2070,
      "train_runtime": 2095.9461,
      "train_tokens_per_second": 5963.291
    },
    {
      "epoch": 0.7617471264367817,
      "grad_norm": 0.7623975276947021,
      "learning_rate": 0.00012398306644579422,
      "loss": 1.7275,
      "num_input_tokens_seen": 12504064,
      "step": 2071,
      "train_runtime": 2097.3246,
      "train_tokens_per_second": 5961.912
    },
    {
      "epoch": 0.7621149425287357,
      "grad_norm": 0.6703829765319824,
      "learning_rate": 0.00012394625437143382,
      "loss": 1.5622,
      "num_input_tokens_seen": 12510992,
      "step": 2072,
      "train_runtime": 2098.417,
      "train_tokens_per_second": 5962.109
    },
    {
      "epoch": 0.7624827586206897,
      "grad_norm": 0.7693400382995605,
      "learning_rate": 0.00012390944229707343,
      "loss": 1.9204,
      "num_input_tokens_seen": 12516480,
      "step": 2073,
      "train_runtime": 2099.3181,
      "train_tokens_per_second": 5962.164
    },
    {
      "epoch": 0.7628505747126437,
      "grad_norm": 0.7057474255561829,
      "learning_rate": 0.00012387263022271304,
      "loss": 1.5577,
      "num_input_tokens_seen": 12520768,
      "step": 2074,
      "train_runtime": 2100.0795,
      "train_tokens_per_second": 5962.045
    },
    {
      "epoch": 0.7632183908045977,
      "grad_norm": 0.6748824119567871,
      "learning_rate": 0.00012383581814835265,
      "loss": 1.7114,
      "num_input_tokens_seen": 12526352,
      "step": 2075,
      "train_runtime": 2100.9763,
      "train_tokens_per_second": 5962.158
    },
    {
      "epoch": 0.7635862068965518,
      "grad_norm": 0.7221955060958862,
      "learning_rate": 0.0001237990060739923,
      "loss": 1.6284,
      "num_input_tokens_seen": 12530928,
      "step": 2076,
      "train_runtime": 2101.7524,
      "train_tokens_per_second": 5962.133
    },
    {
      "epoch": 0.7639540229885058,
      "grad_norm": 0.7431067228317261,
      "learning_rate": 0.0001237621939996319,
      "loss": 1.8156,
      "num_input_tokens_seen": 12536832,
      "step": 2077,
      "train_runtime": 2102.7421,
      "train_tokens_per_second": 5962.135
    },
    {
      "epoch": 0.7643218390804598,
      "grad_norm": 0.6931885480880737,
      "learning_rate": 0.0001237253819252715,
      "loss": 1.7219,
      "num_input_tokens_seen": 12541968,
      "step": 2078,
      "train_runtime": 2103.6022,
      "train_tokens_per_second": 5962.139
    },
    {
      "epoch": 0.7646896551724138,
      "grad_norm": 0.7696422338485718,
      "learning_rate": 0.0001236885698509111,
      "loss": 1.7431,
      "num_input_tokens_seen": 12547920,
      "step": 2079,
      "train_runtime": 2104.5984,
      "train_tokens_per_second": 5962.145
    },
    {
      "epoch": 0.7650574712643678,
      "grad_norm": 0.7124376893043518,
      "learning_rate": 0.00012365175777655072,
      "loss": 1.5987,
      "num_input_tokens_seen": 12554176,
      "step": 2080,
      "train_runtime": 2105.6296,
      "train_tokens_per_second": 5962.196
    },
    {
      "epoch": 0.7654252873563219,
      "grad_norm": 0.7974727749824524,
      "learning_rate": 0.00012361494570219033,
      "loss": 1.7726,
      "num_input_tokens_seen": 12560688,
      "step": 2081,
      "train_runtime": 2106.6475,
      "train_tokens_per_second": 5962.406
    },
    {
      "epoch": 0.7657931034482759,
      "grad_norm": 0.7285959720611572,
      "learning_rate": 0.00012357813362782994,
      "loss": 1.5395,
      "num_input_tokens_seen": 12567616,
      "step": 2082,
      "train_runtime": 2107.7777,
      "train_tokens_per_second": 5962.496
    },
    {
      "epoch": 0.7661609195402299,
      "grad_norm": 0.7363659143447876,
      "learning_rate": 0.00012354132155346955,
      "loss": 1.6736,
      "num_input_tokens_seen": 12574960,
      "step": 2083,
      "train_runtime": 2108.932,
      "train_tokens_per_second": 5962.715
    },
    {
      "epoch": 0.7665287356321839,
      "grad_norm": 0.7323724627494812,
      "learning_rate": 0.00012350450947910913,
      "loss": 1.4709,
      "num_input_tokens_seen": 12585248,
      "step": 2084,
      "train_runtime": 2110.4861,
      "train_tokens_per_second": 5963.199
    },
    {
      "epoch": 0.766896551724138,
      "grad_norm": 0.7748041152954102,
      "learning_rate": 0.00012346769740474877,
      "loss": 1.7863,
      "num_input_tokens_seen": 12591040,
      "step": 2085,
      "train_runtime": 2111.4462,
      "train_tokens_per_second": 5963.23
    },
    {
      "epoch": 0.767264367816092,
      "grad_norm": 0.6988790035247803,
      "learning_rate": 0.00012343088533038838,
      "loss": 1.8432,
      "num_input_tokens_seen": 12599248,
      "step": 2086,
      "train_runtime": 2112.7415,
      "train_tokens_per_second": 5963.459
    },
    {
      "epoch": 0.767632183908046,
      "grad_norm": 0.7449609637260437,
      "learning_rate": 0.00012339407325602798,
      "loss": 1.6399,
      "num_input_tokens_seen": 12606800,
      "step": 2087,
      "train_runtime": 2113.916,
      "train_tokens_per_second": 5963.718
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.9035874605178833,
      "learning_rate": 0.0001233572611816676,
      "loss": 1.7134,
      "num_input_tokens_seen": 12611248,
      "step": 2088,
      "train_runtime": 2114.6979,
      "train_tokens_per_second": 5963.617
    },
    {
      "epoch": 0.768367816091954,
      "grad_norm": 0.7643798589706421,
      "learning_rate": 0.0001233204491073072,
      "loss": 1.7322,
      "num_input_tokens_seen": 12616240,
      "step": 2089,
      "train_runtime": 2115.5549,
      "train_tokens_per_second": 5963.561
    },
    {
      "epoch": 0.7687356321839081,
      "grad_norm": 0.7421737909317017,
      "learning_rate": 0.0001232836370329468,
      "loss": 1.7024,
      "num_input_tokens_seen": 12622592,
      "step": 2090,
      "train_runtime": 2116.5657,
      "train_tokens_per_second": 5963.714
    },
    {
      "epoch": 0.7691034482758621,
      "grad_norm": 0.7030812501907349,
      "learning_rate": 0.00012324682495858642,
      "loss": 1.8837,
      "num_input_tokens_seen": 12629888,
      "step": 2091,
      "train_runtime": 2117.6955,
      "train_tokens_per_second": 5963.977
    },
    {
      "epoch": 0.7694712643678161,
      "grad_norm": 0.7125244140625,
      "learning_rate": 0.00012321001288422603,
      "loss": 1.7142,
      "num_input_tokens_seen": 12634848,
      "step": 2092,
      "train_runtime": 2118.5549,
      "train_tokens_per_second": 5963.899
    },
    {
      "epoch": 0.7698390804597701,
      "grad_norm": 0.7095870971679688,
      "learning_rate": 0.00012317320080986567,
      "loss": 1.7827,
      "num_input_tokens_seen": 12639792,
      "step": 2093,
      "train_runtime": 2119.4017,
      "train_tokens_per_second": 5963.849
    },
    {
      "epoch": 0.7702068965517241,
      "grad_norm": 0.7154878973960876,
      "learning_rate": 0.00012313638873550525,
      "loss": 1.809,
      "num_input_tokens_seen": 12644832,
      "step": 2094,
      "train_runtime": 2120.2578,
      "train_tokens_per_second": 5963.818
    },
    {
      "epoch": 0.7705747126436782,
      "grad_norm": 0.720576822757721,
      "learning_rate": 0.00012309957666114486,
      "loss": 1.7733,
      "num_input_tokens_seen": 12651584,
      "step": 2095,
      "train_runtime": 2121.3524,
      "train_tokens_per_second": 5963.924
    },
    {
      "epoch": 0.7709425287356322,
      "grad_norm": 0.6914993524551392,
      "learning_rate": 0.00012306276458678446,
      "loss": 1.7504,
      "num_input_tokens_seen": 12659728,
      "step": 2096,
      "train_runtime": 2122.6242,
      "train_tokens_per_second": 5964.187
    },
    {
      "epoch": 0.7713103448275862,
      "grad_norm": 0.6382989287376404,
      "learning_rate": 0.00012302595251242407,
      "loss": 1.6161,
      "num_input_tokens_seen": 12667440,
      "step": 2097,
      "train_runtime": 2123.7999,
      "train_tokens_per_second": 5964.517
    },
    {
      "epoch": 0.7716781609195402,
      "grad_norm": 0.7535708546638489,
      "learning_rate": 0.00012298914043806368,
      "loss": 1.6673,
      "num_input_tokens_seen": 12673472,
      "step": 2098,
      "train_runtime": 2124.7815,
      "train_tokens_per_second": 5964.6
    },
    {
      "epoch": 0.7720459770114942,
      "grad_norm": 0.7594273686408997,
      "learning_rate": 0.0001229523283637033,
      "loss": 1.6248,
      "num_input_tokens_seen": 12679872,
      "step": 2099,
      "train_runtime": 2125.8015,
      "train_tokens_per_second": 5964.749
    },
    {
      "epoch": 0.7724137931034483,
      "grad_norm": 0.7400872707366943,
      "learning_rate": 0.0001229155162893429,
      "loss": 1.7397,
      "num_input_tokens_seen": 12686208,
      "step": 2100,
      "train_runtime": 2126.8215,
      "train_tokens_per_second": 5964.867
    },
    {
      "epoch": 0.7727816091954023,
      "grad_norm": 0.7387911677360535,
      "learning_rate": 0.0001228787042149825,
      "loss": 1.5423,
      "num_input_tokens_seen": 12691120,
      "step": 2101,
      "train_runtime": 2128.1152,
      "train_tokens_per_second": 5963.549
    },
    {
      "epoch": 0.7731494252873563,
      "grad_norm": 0.803270161151886,
      "learning_rate": 0.00012284189214062215,
      "loss": 1.6986,
      "num_input_tokens_seen": 12696080,
      "step": 2102,
      "train_runtime": 2128.9782,
      "train_tokens_per_second": 5963.462
    },
    {
      "epoch": 0.7735172413793103,
      "grad_norm": 0.695408046245575,
      "learning_rate": 0.00012280508006626175,
      "loss": 1.5859,
      "num_input_tokens_seen": 12700608,
      "step": 2103,
      "train_runtime": 2129.7188,
      "train_tokens_per_second": 5963.514
    },
    {
      "epoch": 0.7738850574712643,
      "grad_norm": 0.7624605894088745,
      "learning_rate": 0.00012276826799190136,
      "loss": 1.6724,
      "num_input_tokens_seen": 12705856,
      "step": 2104,
      "train_runtime": 2130.6006,
      "train_tokens_per_second": 5963.509
    },
    {
      "epoch": 0.7742528735632184,
      "grad_norm": 0.7603652477264404,
      "learning_rate": 0.00012273145591754097,
      "loss": 1.7543,
      "num_input_tokens_seen": 12712064,
      "step": 2105,
      "train_runtime": 2131.6049,
      "train_tokens_per_second": 5963.612
    },
    {
      "epoch": 0.7746206896551724,
      "grad_norm": 0.6373720169067383,
      "learning_rate": 0.00012269464384318058,
      "loss": 1.6929,
      "num_input_tokens_seen": 12718816,
      "step": 2106,
      "train_runtime": 2132.6744,
      "train_tokens_per_second": 5963.787
    },
    {
      "epoch": 0.7749885057471264,
      "grad_norm": 0.6505287289619446,
      "learning_rate": 0.00012265783176882016,
      "loss": 1.5399,
      "num_input_tokens_seen": 12727232,
      "step": 2107,
      "train_runtime": 2133.9805,
      "train_tokens_per_second": 5964.081
    },
    {
      "epoch": 0.7753563218390804,
      "grad_norm": 0.722303032875061,
      "learning_rate": 0.00012262101969445977,
      "loss": 1.7155,
      "num_input_tokens_seen": 12732752,
      "step": 2108,
      "train_runtime": 2134.9092,
      "train_tokens_per_second": 5964.072
    },
    {
      "epoch": 0.7757241379310345,
      "grad_norm": 0.7534534335136414,
      "learning_rate": 0.00012258420762009938,
      "loss": 1.6947,
      "num_input_tokens_seen": 12737184,
      "step": 2109,
      "train_runtime": 2135.6835,
      "train_tokens_per_second": 5963.985
    },
    {
      "epoch": 0.7760919540229885,
      "grad_norm": 0.6947422623634338,
      "learning_rate": 0.00012254739554573902,
      "loss": 1.5609,
      "num_input_tokens_seen": 12742512,
      "step": 2110,
      "train_runtime": 2136.5908,
      "train_tokens_per_second": 5963.946
    },
    {
      "epoch": 0.7764597701149425,
      "grad_norm": 0.7415337562561035,
      "learning_rate": 0.00012251058347137863,
      "loss": 1.729,
      "num_input_tokens_seen": 12749376,
      "step": 2111,
      "train_runtime": 2137.6753,
      "train_tokens_per_second": 5964.131
    },
    {
      "epoch": 0.7768275862068965,
      "grad_norm": 0.7485516667366028,
      "learning_rate": 0.00012247377139701823,
      "loss": 1.7938,
      "num_input_tokens_seen": 12756128,
      "step": 2112,
      "train_runtime": 2138.7622,
      "train_tokens_per_second": 5964.257
    },
    {
      "epoch": 0.7771954022988505,
      "grad_norm": 0.7775939702987671,
      "learning_rate": 0.00012243695932265784,
      "loss": 1.6705,
      "num_input_tokens_seen": 12763040,
      "step": 2113,
      "train_runtime": 2139.8673,
      "train_tokens_per_second": 5964.407
    },
    {
      "epoch": 0.7775632183908046,
      "grad_norm": 0.7350584268569946,
      "learning_rate": 0.00012240014724829745,
      "loss": 1.5915,
      "num_input_tokens_seen": 12768512,
      "step": 2114,
      "train_runtime": 2140.7666,
      "train_tokens_per_second": 5964.458
    },
    {
      "epoch": 0.7779310344827586,
      "grad_norm": 0.7638656497001648,
      "learning_rate": 0.00012236333517393706,
      "loss": 1.7557,
      "num_input_tokens_seen": 12774624,
      "step": 2115,
      "train_runtime": 2141.7513,
      "train_tokens_per_second": 5964.569
    },
    {
      "epoch": 0.7782988505747126,
      "grad_norm": 0.7217133641242981,
      "learning_rate": 0.00012232652309957667,
      "loss": 1.7415,
      "num_input_tokens_seen": 12781152,
      "step": 2116,
      "train_runtime": 2142.8009,
      "train_tokens_per_second": 5964.694
    },
    {
      "epoch": 0.7786666666666666,
      "grad_norm": 0.8002574443817139,
      "learning_rate": 0.00012228971102521628,
      "loss": 1.6022,
      "num_input_tokens_seen": 12788592,
      "step": 2117,
      "train_runtime": 2143.9867,
      "train_tokens_per_second": 5964.865
    },
    {
      "epoch": 0.7790344827586206,
      "grad_norm": 0.7704402804374695,
      "learning_rate": 0.0001222528989508559,
      "loss": 1.7797,
      "num_input_tokens_seen": 12793296,
      "step": 2118,
      "train_runtime": 2144.7906,
      "train_tokens_per_second": 5964.823
    },
    {
      "epoch": 0.7794022988505747,
      "grad_norm": 0.7450360059738159,
      "learning_rate": 0.0001222160868764955,
      "loss": 1.674,
      "num_input_tokens_seen": 12797744,
      "step": 2119,
      "train_runtime": 2145.5677,
      "train_tokens_per_second": 5964.736
    },
    {
      "epoch": 0.7797701149425287,
      "grad_norm": 0.7535125613212585,
      "learning_rate": 0.0001221792748021351,
      "loss": 1.8459,
      "num_input_tokens_seen": 12805136,
      "step": 2120,
      "train_runtime": 2146.7432,
      "train_tokens_per_second": 5964.913
    },
    {
      "epoch": 0.7801379310344828,
      "grad_norm": 0.7665154933929443,
      "learning_rate": 0.00012214246272777471,
      "loss": 1.7696,
      "num_input_tokens_seen": 12812240,
      "step": 2121,
      "train_runtime": 2147.8888,
      "train_tokens_per_second": 5965.039
    },
    {
      "epoch": 0.7805057471264368,
      "grad_norm": 0.7314074039459229,
      "learning_rate": 0.00012210565065341432,
      "loss": 1.7164,
      "num_input_tokens_seen": 12823360,
      "step": 2122,
      "train_runtime": 2149.5374,
      "train_tokens_per_second": 5965.637
    },
    {
      "epoch": 0.7808735632183909,
      "grad_norm": 0.7371465563774109,
      "learning_rate": 0.00012206883857905393,
      "loss": 1.6773,
      "num_input_tokens_seen": 12828144,
      "step": 2123,
      "train_runtime": 2150.375,
      "train_tokens_per_second": 5965.538
    },
    {
      "epoch": 0.7812413793103449,
      "grad_norm": 0.8078320622444153,
      "learning_rate": 0.00012203202650469354,
      "loss": 1.6974,
      "num_input_tokens_seen": 12835040,
      "step": 2124,
      "train_runtime": 2151.477,
      "train_tokens_per_second": 5965.688
    },
    {
      "epoch": 0.7816091954022989,
      "grad_norm": 0.7091101408004761,
      "learning_rate": 0.00012199521443033315,
      "loss": 1.797,
      "num_input_tokens_seen": 12839616,
      "step": 2125,
      "train_runtime": 2152.272,
      "train_tokens_per_second": 5965.61
    },
    {
      "epoch": 0.7819770114942529,
      "grad_norm": 0.6919854283332825,
      "learning_rate": 0.00012195840235597276,
      "loss": 1.6841,
      "num_input_tokens_seen": 12846720,
      "step": 2126,
      "train_runtime": 2153.4085,
      "train_tokens_per_second": 5965.761
    },
    {
      "epoch": 0.7823448275862069,
      "grad_norm": 0.7704769968986511,
      "learning_rate": 0.00012192159028161238,
      "loss": 1.8583,
      "num_input_tokens_seen": 12851632,
      "step": 2127,
      "train_runtime": 2154.2562,
      "train_tokens_per_second": 5965.693
    },
    {
      "epoch": 0.782712643678161,
      "grad_norm": 0.7886033058166504,
      "learning_rate": 0.00012188477820725199,
      "loss": 1.6769,
      "num_input_tokens_seen": 12859456,
      "step": 2128,
      "train_runtime": 2155.4833,
      "train_tokens_per_second": 5965.927
    },
    {
      "epoch": 0.783080459770115,
      "grad_norm": 0.7637695670127869,
      "learning_rate": 0.0001218479661328916,
      "loss": 1.8165,
      "num_input_tokens_seen": 12866368,
      "step": 2129,
      "train_runtime": 2156.5852,
      "train_tokens_per_second": 5966.084
    },
    {
      "epoch": 0.783448275862069,
      "grad_norm": 0.7374247908592224,
      "learning_rate": 0.00012181115405853121,
      "loss": 1.6218,
      "num_input_tokens_seen": 12871632,
      "step": 2130,
      "train_runtime": 2157.4662,
      "train_tokens_per_second": 5966.087
    },
    {
      "epoch": 0.783816091954023,
      "grad_norm": 0.6884121894836426,
      "learning_rate": 0.00012177434198417082,
      "loss": 1.9006,
      "num_input_tokens_seen": 12878016,
      "step": 2131,
      "train_runtime": 2159.0725,
      "train_tokens_per_second": 5964.606
    },
    {
      "epoch": 0.784183908045977,
      "grad_norm": 0.841325581073761,
      "learning_rate": 0.00012173752990981043,
      "loss": 1.7529,
      "num_input_tokens_seen": 12883808,
      "step": 2132,
      "train_runtime": 2160.0043,
      "train_tokens_per_second": 5964.714
    },
    {
      "epoch": 0.7845517241379311,
      "grad_norm": 0.6641340255737305,
      "learning_rate": 0.00012170071783545003,
      "loss": 1.6923,
      "num_input_tokens_seen": 12891104,
      "step": 2133,
      "train_runtime": 2161.1285,
      "train_tokens_per_second": 5964.987
    },
    {
      "epoch": 0.7849195402298851,
      "grad_norm": 0.8208033442497253,
      "learning_rate": 0.00012166390576108963,
      "loss": 1.5294,
      "num_input_tokens_seen": 12897616,
      "step": 2134,
      "train_runtime": 2162.188,
      "train_tokens_per_second": 5965.076
    },
    {
      "epoch": 0.7852873563218391,
      "grad_norm": 0.7175398468971252,
      "learning_rate": 0.00012162709368672924,
      "loss": 1.7781,
      "num_input_tokens_seen": 12905216,
      "step": 2135,
      "train_runtime": 2163.3935,
      "train_tokens_per_second": 5965.265
    },
    {
      "epoch": 0.7856551724137931,
      "grad_norm": 0.7060106992721558,
      "learning_rate": 0.00012159028161236887,
      "loss": 1.5183,
      "num_input_tokens_seen": 12909872,
      "step": 2136,
      "train_runtime": 2164.1901,
      "train_tokens_per_second": 5965.221
    },
    {
      "epoch": 0.7860229885057471,
      "grad_norm": 0.7808321118354797,
      "learning_rate": 0.00012155346953800848,
      "loss": 1.8642,
      "num_input_tokens_seen": 12915520,
      "step": 2137,
      "train_runtime": 2165.1377,
      "train_tokens_per_second": 5965.219
    },
    {
      "epoch": 0.7863908045977012,
      "grad_norm": 0.7710825800895691,
      "learning_rate": 0.00012151665746364809,
      "loss": 1.7373,
      "num_input_tokens_seen": 12920992,
      "step": 2138,
      "train_runtime": 2166.0458,
      "train_tokens_per_second": 5965.244
    },
    {
      "epoch": 0.7867586206896552,
      "grad_norm": 0.7178471088409424,
      "learning_rate": 0.00012147984538928769,
      "loss": 1.6675,
      "num_input_tokens_seen": 12926112,
      "step": 2139,
      "train_runtime": 2166.902,
      "train_tokens_per_second": 5965.25
    },
    {
      "epoch": 0.7871264367816092,
      "grad_norm": 0.7546533346176147,
      "learning_rate": 0.0001214430333149273,
      "loss": 1.7433,
      "num_input_tokens_seen": 12931424,
      "step": 2140,
      "train_runtime": 2167.7763,
      "train_tokens_per_second": 5965.294
    },
    {
      "epoch": 0.7874942528735632,
      "grad_norm": 0.7169228196144104,
      "learning_rate": 0.0001214062212405669,
      "loss": 1.7944,
      "num_input_tokens_seen": 12939808,
      "step": 2141,
      "train_runtime": 2169.0765,
      "train_tokens_per_second": 5965.584
    },
    {
      "epoch": 0.7878620689655172,
      "grad_norm": 0.7573052644729614,
      "learning_rate": 0.00012136940916620651,
      "loss": 1.6879,
      "num_input_tokens_seen": 12944656,
      "step": 2142,
      "train_runtime": 2169.9196,
      "train_tokens_per_second": 5965.5
    },
    {
      "epoch": 0.7882298850574713,
      "grad_norm": 0.7861320972442627,
      "learning_rate": 0.00012133259709184612,
      "loss": 1.8693,
      "num_input_tokens_seen": 12949776,
      "step": 2143,
      "train_runtime": 2170.7861,
      "train_tokens_per_second": 5965.478
    },
    {
      "epoch": 0.7885977011494253,
      "grad_norm": 0.7815845608711243,
      "learning_rate": 0.00012129578501748573,
      "loss": 2.004,
      "num_input_tokens_seen": 12955040,
      "step": 2144,
      "train_runtime": 2171.6736,
      "train_tokens_per_second": 5965.464
    },
    {
      "epoch": 0.7889655172413793,
      "grad_norm": 0.8198308944702148,
      "learning_rate": 0.00012125897294312535,
      "loss": 1.6251,
      "num_input_tokens_seen": 12963728,
      "step": 2145,
      "train_runtime": 2173.0125,
      "train_tokens_per_second": 5965.786
    },
    {
      "epoch": 0.7893333333333333,
      "grad_norm": 0.769403874874115,
      "learning_rate": 0.00012122216086876496,
      "loss": 1.6499,
      "num_input_tokens_seen": 12971776,
      "step": 2146,
      "train_runtime": 2174.2927,
      "train_tokens_per_second": 5965.975
    },
    {
      "epoch": 0.7897011494252874,
      "grad_norm": 0.7350329160690308,
      "learning_rate": 0.00012118534879440457,
      "loss": 1.6657,
      "num_input_tokens_seen": 12980208,
      "step": 2147,
      "train_runtime": 2175.6096,
      "train_tokens_per_second": 5966.239
    },
    {
      "epoch": 0.7900689655172414,
      "grad_norm": 0.8037062287330627,
      "learning_rate": 0.00012114853672004418,
      "loss": 1.7512,
      "num_input_tokens_seen": 12986576,
      "step": 2148,
      "train_runtime": 2176.6467,
      "train_tokens_per_second": 5966.322
    },
    {
      "epoch": 0.7904367816091954,
      "grad_norm": 0.7732772827148438,
      "learning_rate": 0.00012111172464568379,
      "loss": 1.6079,
      "num_input_tokens_seen": 12991728,
      "step": 2149,
      "train_runtime": 2177.5444,
      "train_tokens_per_second": 5966.229
    },
    {
      "epoch": 0.7908045977011494,
      "grad_norm": 0.7495967149734497,
      "learning_rate": 0.0001210749125713234,
      "loss": 1.8363,
      "num_input_tokens_seen": 12999984,
      "step": 2150,
      "train_runtime": 2178.8418,
      "train_tokens_per_second": 5966.465
    },
    {
      "epoch": 0.7911724137931034,
      "grad_norm": 0.6866952776908875,
      "learning_rate": 0.00012103810049696301,
      "loss": 1.7136,
      "num_input_tokens_seen": 13005744,
      "step": 2151,
      "train_runtime": 2179.7851,
      "train_tokens_per_second": 5966.526
    },
    {
      "epoch": 0.7915402298850575,
      "grad_norm": 0.8390110731124878,
      "learning_rate": 0.0001210012884226026,
      "loss": 1.9442,
      "num_input_tokens_seen": 13012000,
      "step": 2152,
      "train_runtime": 2180.795,
      "train_tokens_per_second": 5966.632
    },
    {
      "epoch": 0.7919080459770115,
      "grad_norm": 0.7232964038848877,
      "learning_rate": 0.00012096447634824224,
      "loss": 1.6981,
      "num_input_tokens_seen": 13019568,
      "step": 2153,
      "train_runtime": 2181.999,
      "train_tokens_per_second": 5966.807
    },
    {
      "epoch": 0.7922758620689655,
      "grad_norm": 0.7241571545600891,
      "learning_rate": 0.00012092766427388185,
      "loss": 1.5719,
      "num_input_tokens_seen": 13030912,
      "step": 2154,
      "train_runtime": 2183.6801,
      "train_tokens_per_second": 5967.409
    },
    {
      "epoch": 0.7926436781609195,
      "grad_norm": 0.7424749732017517,
      "learning_rate": 0.00012089085219952146,
      "loss": 1.6579,
      "num_input_tokens_seen": 13036224,
      "step": 2155,
      "train_runtime": 2184.5661,
      "train_tokens_per_second": 5967.42
    },
    {
      "epoch": 0.7930114942528735,
      "grad_norm": 0.7053840160369873,
      "learning_rate": 0.00012085404012516107,
      "loss": 1.8049,
      "num_input_tokens_seen": 13043360,
      "step": 2156,
      "train_runtime": 2185.6927,
      "train_tokens_per_second": 5967.609
    },
    {
      "epoch": 0.7933793103448276,
      "grad_norm": 0.828298807144165,
      "learning_rate": 0.00012081722805080067,
      "loss": 1.8097,
      "num_input_tokens_seen": 13047968,
      "step": 2157,
      "train_runtime": 2186.509,
      "train_tokens_per_second": 5967.489
    },
    {
      "epoch": 0.7937471264367816,
      "grad_norm": 0.7036006450653076,
      "learning_rate": 0.00012078041597644027,
      "loss": 1.9218,
      "num_input_tokens_seen": 13052880,
      "step": 2158,
      "train_runtime": 2187.3554,
      "train_tokens_per_second": 5967.425
    },
    {
      "epoch": 0.7941149425287356,
      "grad_norm": 0.7244173884391785,
      "learning_rate": 0.00012074360390207988,
      "loss": 1.6771,
      "num_input_tokens_seen": 13060016,
      "step": 2159,
      "train_runtime": 2188.4741,
      "train_tokens_per_second": 5967.636
    },
    {
      "epoch": 0.7944827586206896,
      "grad_norm": 0.780311107635498,
      "learning_rate": 0.00012070679182771949,
      "loss": 1.7898,
      "num_input_tokens_seen": 13065552,
      "step": 2160,
      "train_runtime": 2189.3758,
      "train_tokens_per_second": 5967.706
    },
    {
      "epoch": 0.7948505747126436,
      "grad_norm": 0.770302414894104,
      "learning_rate": 0.0001206699797533591,
      "loss": 1.8771,
      "num_input_tokens_seen": 13070752,
      "step": 2161,
      "train_runtime": 2190.7747,
      "train_tokens_per_second": 5966.269
    },
    {
      "epoch": 0.7952183908045977,
      "grad_norm": 0.7521932125091553,
      "learning_rate": 0.00012063316767899873,
      "loss": 1.8639,
      "num_input_tokens_seen": 13075680,
      "step": 2162,
      "train_runtime": 2191.5995,
      "train_tokens_per_second": 5966.273
    },
    {
      "epoch": 0.7955862068965517,
      "grad_norm": 0.7477977275848389,
      "learning_rate": 0.00012059635560463833,
      "loss": 1.6864,
      "num_input_tokens_seen": 13081232,
      "step": 2163,
      "train_runtime": 2192.544,
      "train_tokens_per_second": 5966.235
    },
    {
      "epoch": 0.7959540229885057,
      "grad_norm": 0.7869067192077637,
      "learning_rate": 0.00012055954353027794,
      "loss": 1.6155,
      "num_input_tokens_seen": 13087344,
      "step": 2164,
      "train_runtime": 2193.5473,
      "train_tokens_per_second": 5966.292
    },
    {
      "epoch": 0.7963218390804597,
      "grad_norm": 0.7853013873100281,
      "learning_rate": 0.00012052273145591755,
      "loss": 1.8206,
      "num_input_tokens_seen": 13094016,
      "step": 2165,
      "train_runtime": 2194.6491,
      "train_tokens_per_second": 5966.337
    },
    {
      "epoch": 0.7966896551724137,
      "grad_norm": 0.7521070241928101,
      "learning_rate": 0.00012048591938155715,
      "loss": 1.6824,
      "num_input_tokens_seen": 13099232,
      "step": 2166,
      "train_runtime": 2195.5045,
      "train_tokens_per_second": 5966.388
    },
    {
      "epoch": 0.7970574712643678,
      "grad_norm": 0.7937947511672974,
      "learning_rate": 0.00012044910730719676,
      "loss": 1.7893,
      "num_input_tokens_seen": 13104288,
      "step": 2167,
      "train_runtime": 2196.3591,
      "train_tokens_per_second": 5966.369
    },
    {
      "epoch": 0.7974252873563218,
      "grad_norm": 0.8112798929214478,
      "learning_rate": 0.00012041229523283637,
      "loss": 1.7948,
      "num_input_tokens_seen": 13112928,
      "step": 2168,
      "train_runtime": 2197.6897,
      "train_tokens_per_second": 5966.688
    },
    {
      "epoch": 0.7977931034482758,
      "grad_norm": 0.6776713728904724,
      "learning_rate": 0.00012037548315847598,
      "loss": 1.6845,
      "num_input_tokens_seen": 13125312,
      "step": 2169,
      "train_runtime": 2199.5096,
      "train_tokens_per_second": 5967.381
    },
    {
      "epoch": 0.7981609195402298,
      "grad_norm": 0.5651150345802307,
      "learning_rate": 0.0001203386710841156,
      "loss": 1.5699,
      "num_input_tokens_seen": 13137456,
      "step": 2170,
      "train_runtime": 2201.3142,
      "train_tokens_per_second": 5968.006
    },
    {
      "epoch": 0.798528735632184,
      "grad_norm": 0.7914756536483765,
      "learning_rate": 0.00012030185900975521,
      "loss": 1.7932,
      "num_input_tokens_seen": 13142816,
      "step": 2171,
      "train_runtime": 2202.1938,
      "train_tokens_per_second": 5968.056
    },
    {
      "epoch": 0.798896551724138,
      "grad_norm": 0.7965108752250671,
      "learning_rate": 0.00012026504693539482,
      "loss": 1.6363,
      "num_input_tokens_seen": 13149216,
      "step": 2172,
      "train_runtime": 2203.2465,
      "train_tokens_per_second": 5968.109
    },
    {
      "epoch": 0.799264367816092,
      "grad_norm": 0.7645534873008728,
      "learning_rate": 0.00012022823486103443,
      "loss": 1.7914,
      "num_input_tokens_seen": 13153952,
      "step": 2173,
      "train_runtime": 2204.074,
      "train_tokens_per_second": 5968.017
    },
    {
      "epoch": 0.799632183908046,
      "grad_norm": 0.8161355257034302,
      "learning_rate": 0.00012019142278667404,
      "loss": 1.9488,
      "num_input_tokens_seen": 13159376,
      "step": 2174,
      "train_runtime": 2204.964,
      "train_tokens_per_second": 5968.068
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.69480299949646,
      "learning_rate": 0.00012015461071231365,
      "loss": 1.6936,
      "num_input_tokens_seen": 13171120,
      "step": 2175,
      "train_runtime": 2206.6956,
      "train_tokens_per_second": 5968.707
    },
    {
      "epoch": 0.8003678160919541,
      "grad_norm": 0.7443561553955078,
      "learning_rate": 0.00012011779863795324,
      "loss": 1.7426,
      "num_input_tokens_seen": 13179040,
      "step": 2176,
      "train_runtime": 2207.929,
      "train_tokens_per_second": 5968.96
    },
    {
      "epoch": 0.8003678160919541,
      "eval_loss": 1.782747507095337,
      "eval_runtime": 4.7692,
      "eval_samples_per_second": 209.678,
      "eval_steps_per_second": 13.21,
      "num_input_tokens_seen": 13179040,
      "step": 2176
    },
    {
      "epoch": 0.8007356321839081,
      "grad_norm": 0.7878211140632629,
      "learning_rate": 0.00012008098656359285,
      "loss": 1.8395,
      "num_input_tokens_seen": 13184048,
      "step": 2177,
      "train_runtime": 2213.5639,
      "train_tokens_per_second": 5956.028
    },
    {
      "epoch": 0.8011034482758621,
      "grad_norm": 0.7659966349601746,
      "learning_rate": 0.00012004417448923246,
      "loss": 1.727,
      "num_input_tokens_seen": 13189664,
      "step": 2178,
      "train_runtime": 2214.516,
      "train_tokens_per_second": 5956.003
    },
    {
      "epoch": 0.8014712643678161,
      "grad_norm": 0.7582219243049622,
      "learning_rate": 0.0001200073624148721,
      "loss": 1.8222,
      "num_input_tokens_seen": 13195456,
      "step": 2179,
      "train_runtime": 2215.4365,
      "train_tokens_per_second": 5956.143
    },
    {
      "epoch": 0.8018390804597701,
      "grad_norm": 0.8086379766464233,
      "learning_rate": 0.0001199705503405117,
      "loss": 1.637,
      "num_input_tokens_seen": 13201664,
      "step": 2180,
      "train_runtime": 2216.4302,
      "train_tokens_per_second": 5956.273
    },
    {
      "epoch": 0.8022068965517242,
      "grad_norm": 0.726866602897644,
      "learning_rate": 0.0001199337382661513,
      "loss": 1.7879,
      "num_input_tokens_seen": 13206528,
      "step": 2181,
      "train_runtime": 2217.2678,
      "train_tokens_per_second": 5956.217
    },
    {
      "epoch": 0.8025747126436782,
      "grad_norm": 0.7773523926734924,
      "learning_rate": 0.00011989692619179091,
      "loss": 1.6477,
      "num_input_tokens_seen": 13214272,
      "step": 2182,
      "train_runtime": 2218.5035,
      "train_tokens_per_second": 5956.39
    },
    {
      "epoch": 0.8029425287356322,
      "grad_norm": 0.8535124063491821,
      "learning_rate": 0.00011986011411743052,
      "loss": 1.7861,
      "num_input_tokens_seen": 13218992,
      "step": 2183,
      "train_runtime": 2219.3368,
      "train_tokens_per_second": 5956.28
    },
    {
      "epoch": 0.8033103448275862,
      "grad_norm": 0.7061176896095276,
      "learning_rate": 0.00011982330204307013,
      "loss": 1.6619,
      "num_input_tokens_seen": 13223696,
      "step": 2184,
      "train_runtime": 2220.1693,
      "train_tokens_per_second": 5956.165
    },
    {
      "epoch": 0.8036781609195403,
      "grad_norm": 0.7682154178619385,
      "learning_rate": 0.00011978648996870974,
      "loss": 1.6372,
      "num_input_tokens_seen": 13228752,
      "step": 2185,
      "train_runtime": 2221.0482,
      "train_tokens_per_second": 5956.085
    },
    {
      "epoch": 0.8040459770114943,
      "grad_norm": 0.7980086207389832,
      "learning_rate": 0.00011974967789434935,
      "loss": 1.9487,
      "num_input_tokens_seen": 13235280,
      "step": 2186,
      "train_runtime": 2222.1041,
      "train_tokens_per_second": 5956.192
    },
    {
      "epoch": 0.8044137931034483,
      "grad_norm": 0.7011343836784363,
      "learning_rate": 0.00011971286581998897,
      "loss": 1.4009,
      "num_input_tokens_seen": 13240352,
      "step": 2187,
      "train_runtime": 2222.9402,
      "train_tokens_per_second": 5956.234
    },
    {
      "epoch": 0.8047816091954023,
      "grad_norm": 0.8399803042411804,
      "learning_rate": 0.00011967605374562858,
      "loss": 2.0242,
      "num_input_tokens_seen": 13246960,
      "step": 2188,
      "train_runtime": 2224.0106,
      "train_tokens_per_second": 5956.339
    },
    {
      "epoch": 0.8051494252873563,
      "grad_norm": 0.7678438425064087,
      "learning_rate": 0.00011963924167126819,
      "loss": 1.694,
      "num_input_tokens_seen": 13254576,
      "step": 2189,
      "train_runtime": 2225.1893,
      "train_tokens_per_second": 5956.606
    },
    {
      "epoch": 0.8055172413793104,
      "grad_norm": 0.694980800151825,
      "learning_rate": 0.0001196024295969078,
      "loss": 1.6687,
      "num_input_tokens_seen": 13263024,
      "step": 2190,
      "train_runtime": 2226.4744,
      "train_tokens_per_second": 5956.962
    },
    {
      "epoch": 0.8058850574712644,
      "grad_norm": 0.7804933190345764,
      "learning_rate": 0.0001195656175225474,
      "loss": 1.5794,
      "num_input_tokens_seen": 13267664,
      "step": 2191,
      "train_runtime": 2227.7249,
      "train_tokens_per_second": 5955.701
    },
    {
      "epoch": 0.8062528735632184,
      "grad_norm": 0.8200054168701172,
      "learning_rate": 0.00011952880544818701,
      "loss": 1.6344,
      "num_input_tokens_seen": 13273072,
      "step": 2192,
      "train_runtime": 2228.625,
      "train_tokens_per_second": 5955.723
    },
    {
      "epoch": 0.8066206896551724,
      "grad_norm": 0.754323422908783,
      "learning_rate": 0.00011949199337382662,
      "loss": 1.8163,
      "num_input_tokens_seen": 13278544,
      "step": 2193,
      "train_runtime": 2229.5405,
      "train_tokens_per_second": 5955.731
    },
    {
      "epoch": 0.8069885057471264,
      "grad_norm": 0.7527128458023071,
      "learning_rate": 0.00011945518129946622,
      "loss": 1.7368,
      "num_input_tokens_seen": 13283280,
      "step": 2194,
      "train_runtime": 2230.3715,
      "train_tokens_per_second": 5955.636
    },
    {
      "epoch": 0.8073563218390805,
      "grad_norm": 0.730532705783844,
      "learning_rate": 0.00011941836922510583,
      "loss": 1.68,
      "num_input_tokens_seen": 13288400,
      "step": 2195,
      "train_runtime": 2231.2546,
      "train_tokens_per_second": 5955.573
    },
    {
      "epoch": 0.8077241379310345,
      "grad_norm": 0.7435619831085205,
      "learning_rate": 0.00011938155715074546,
      "loss": 1.8355,
      "num_input_tokens_seen": 13294944,
      "step": 2196,
      "train_runtime": 2232.3143,
      "train_tokens_per_second": 5955.677
    },
    {
      "epoch": 0.8080919540229885,
      "grad_norm": 0.8014752864837646,
      "learning_rate": 0.00011934474507638507,
      "loss": 1.9065,
      "num_input_tokens_seen": 13300240,
      "step": 2197,
      "train_runtime": 2233.2014,
      "train_tokens_per_second": 5955.683
    },
    {
      "epoch": 0.8084597701149425,
      "grad_norm": 0.9064193367958069,
      "learning_rate": 0.00011930793300202468,
      "loss": 1.7887,
      "num_input_tokens_seen": 13305248,
      "step": 2198,
      "train_runtime": 2234.0727,
      "train_tokens_per_second": 5955.602
    },
    {
      "epoch": 0.8088275862068965,
      "grad_norm": 0.7992514371871948,
      "learning_rate": 0.00011927112092766428,
      "loss": 1.6639,
      "num_input_tokens_seen": 13311728,
      "step": 2199,
      "train_runtime": 2235.098,
      "train_tokens_per_second": 5955.769
    },
    {
      "epoch": 0.8091954022988506,
      "grad_norm": 0.7761949896812439,
      "learning_rate": 0.00011923430885330388,
      "loss": 1.693,
      "num_input_tokens_seen": 13316736,
      "step": 2200,
      "train_runtime": 2235.9207,
      "train_tokens_per_second": 5955.818
    },
    {
      "epoch": 0.8095632183908046,
      "grad_norm": 0.754311740398407,
      "learning_rate": 0.00011919749677894349,
      "loss": 1.6683,
      "num_input_tokens_seen": 13322912,
      "step": 2201,
      "train_runtime": 2236.9148,
      "train_tokens_per_second": 5955.932
    },
    {
      "epoch": 0.8099310344827586,
      "grad_norm": 0.7518620491027832,
      "learning_rate": 0.0001191606847045831,
      "loss": 1.7392,
      "num_input_tokens_seen": 13328640,
      "step": 2202,
      "train_runtime": 2237.8588,
      "train_tokens_per_second": 5955.979
    },
    {
      "epoch": 0.8102988505747126,
      "grad_norm": 0.7505711913108826,
      "learning_rate": 0.00011912387263022271,
      "loss": 1.6373,
      "num_input_tokens_seen": 13336416,
      "step": 2203,
      "train_runtime": 2239.0681,
      "train_tokens_per_second": 5956.235
    },
    {
      "epoch": 0.8106666666666666,
      "grad_norm": 0.7444790601730347,
      "learning_rate": 0.00011908706055586232,
      "loss": 1.7822,
      "num_input_tokens_seen": 13342096,
      "step": 2204,
      "train_runtime": 2239.9991,
      "train_tokens_per_second": 5956.295
    },
    {
      "epoch": 0.8110344827586207,
      "grad_norm": 0.8003002405166626,
      "learning_rate": 0.00011905024848150194,
      "loss": 1.6414,
      "num_input_tokens_seen": 13348368,
      "step": 2205,
      "train_runtime": 2241.0353,
      "train_tokens_per_second": 5956.34
    },
    {
      "epoch": 0.8114022988505747,
      "grad_norm": 0.685434877872467,
      "learning_rate": 0.00011901343640714155,
      "loss": 1.5408,
      "num_input_tokens_seen": 13355424,
      "step": 2206,
      "train_runtime": 2242.1707,
      "train_tokens_per_second": 5956.471
    },
    {
      "epoch": 0.8117701149425287,
      "grad_norm": 0.7814292907714844,
      "learning_rate": 0.00011897662433278116,
      "loss": 1.731,
      "num_input_tokens_seen": 13359808,
      "step": 2207,
      "train_runtime": 2242.9528,
      "train_tokens_per_second": 5956.348
    },
    {
      "epoch": 0.8121379310344827,
      "grad_norm": 0.7573452591896057,
      "learning_rate": 0.00011893981225842077,
      "loss": 1.7434,
      "num_input_tokens_seen": 13366848,
      "step": 2208,
      "train_runtime": 2244.0741,
      "train_tokens_per_second": 5956.509
    },
    {
      "epoch": 0.8125057471264368,
      "grad_norm": 0.8169957399368286,
      "learning_rate": 0.00011890300018406038,
      "loss": 1.7472,
      "num_input_tokens_seen": 13371776,
      "step": 2209,
      "train_runtime": 2244.9226,
      "train_tokens_per_second": 5956.453
    },
    {
      "epoch": 0.8128735632183908,
      "grad_norm": 0.735317051410675,
      "learning_rate": 0.00011886618810969999,
      "loss": 1.6542,
      "num_input_tokens_seen": 13377616,
      "step": 2210,
      "train_runtime": 2245.904,
      "train_tokens_per_second": 5956.45
    },
    {
      "epoch": 0.8132413793103448,
      "grad_norm": 0.7750908732414246,
      "learning_rate": 0.0001188293760353396,
      "loss": 1.7314,
      "num_input_tokens_seen": 13383696,
      "step": 2211,
      "train_runtime": 2246.8816,
      "train_tokens_per_second": 5956.565
    },
    {
      "epoch": 0.8136091954022988,
      "grad_norm": 0.6906810402870178,
      "learning_rate": 0.00011879256396097919,
      "loss": 1.7276,
      "num_input_tokens_seen": 13389056,
      "step": 2212,
      "train_runtime": 2247.7847,
      "train_tokens_per_second": 5956.556
    },
    {
      "epoch": 0.8139770114942528,
      "grad_norm": 0.8501411080360413,
      "learning_rate": 0.00011875575188661883,
      "loss": 1.6156,
      "num_input_tokens_seen": 13394256,
      "step": 2213,
      "train_runtime": 2248.6531,
      "train_tokens_per_second": 5956.568
    },
    {
      "epoch": 0.8143448275862069,
      "grad_norm": 0.7399520874023438,
      "learning_rate": 0.00011871893981225844,
      "loss": 1.7377,
      "num_input_tokens_seen": 13400720,
      "step": 2214,
      "train_runtime": 2249.7217,
      "train_tokens_per_second": 5956.612
    },
    {
      "epoch": 0.8147126436781609,
      "grad_norm": 0.7963682413101196,
      "learning_rate": 0.00011868212773789804,
      "loss": 1.8527,
      "num_input_tokens_seen": 13405792,
      "step": 2215,
      "train_runtime": 2250.5961,
      "train_tokens_per_second": 5956.552
    },
    {
      "epoch": 0.8150804597701149,
      "grad_norm": 0.7445478439331055,
      "learning_rate": 0.00011864531566353765,
      "loss": 1.6572,
      "num_input_tokens_seen": 13413312,
      "step": 2216,
      "train_runtime": 2251.7895,
      "train_tokens_per_second": 5956.734
    },
    {
      "epoch": 0.8154482758620689,
      "grad_norm": 0.785130500793457,
      "learning_rate": 0.00011860850358917725,
      "loss": 1.6997,
      "num_input_tokens_seen": 13418192,
      "step": 2217,
      "train_runtime": 2252.6025,
      "train_tokens_per_second": 5956.751
    },
    {
      "epoch": 0.8158160919540229,
      "grad_norm": 0.682814359664917,
      "learning_rate": 0.00011857169151481686,
      "loss": 1.7225,
      "num_input_tokens_seen": 13426336,
      "step": 2218,
      "train_runtime": 2253.8897,
      "train_tokens_per_second": 5956.962
    },
    {
      "epoch": 0.816183908045977,
      "grad_norm": 0.7894740700721741,
      "learning_rate": 0.00011853487944045647,
      "loss": 1.7328,
      "num_input_tokens_seen": 13432064,
      "step": 2219,
      "train_runtime": 2254.8597,
      "train_tokens_per_second": 5956.94
    },
    {
      "epoch": 0.8165517241379311,
      "grad_norm": 0.7319313287734985,
      "learning_rate": 0.00011849806736609608,
      "loss": 1.7548,
      "num_input_tokens_seen": 13436448,
      "step": 2220,
      "train_runtime": 2255.6346,
      "train_tokens_per_second": 5956.837
    },
    {
      "epoch": 0.8169195402298851,
      "grad_norm": 0.7173370718955994,
      "learning_rate": 0.00011846125529173568,
      "loss": 1.8158,
      "num_input_tokens_seen": 13441664,
      "step": 2221,
      "train_runtime": 2256.9964,
      "train_tokens_per_second": 5955.554
    },
    {
      "epoch": 0.8172873563218391,
      "grad_norm": 0.8172082304954529,
      "learning_rate": 0.00011842444321737532,
      "loss": 1.9257,
      "num_input_tokens_seen": 13447024,
      "step": 2222,
      "train_runtime": 2257.8857,
      "train_tokens_per_second": 5955.582
    },
    {
      "epoch": 0.8176551724137932,
      "grad_norm": 0.7469117641448975,
      "learning_rate": 0.00011838763114301492,
      "loss": 1.7575,
      "num_input_tokens_seen": 13451552,
      "step": 2223,
      "train_runtime": 2258.6704,
      "train_tokens_per_second": 5955.518
    },
    {
      "epoch": 0.8180229885057472,
      "grad_norm": 0.7840341329574585,
      "learning_rate": 0.00011835081906865452,
      "loss": 1.7328,
      "num_input_tokens_seen": 13457200,
      "step": 2224,
      "train_runtime": 2259.6065,
      "train_tokens_per_second": 5955.55
    },
    {
      "epoch": 0.8183908045977012,
      "grad_norm": 0.7672483325004578,
      "learning_rate": 0.00011831400699429413,
      "loss": 1.6439,
      "num_input_tokens_seen": 13462816,
      "step": 2225,
      "train_runtime": 2260.5386,
      "train_tokens_per_second": 5955.579
    },
    {
      "epoch": 0.8187586206896552,
      "grad_norm": 0.759892463684082,
      "learning_rate": 0.00011827719491993374,
      "loss": 1.4372,
      "num_input_tokens_seen": 13468272,
      "step": 2226,
      "train_runtime": 2261.4324,
      "train_tokens_per_second": 5955.638
    },
    {
      "epoch": 0.8191264367816092,
      "grad_norm": 0.8722032308578491,
      "learning_rate": 0.00011824038284557335,
      "loss": 2.0645,
      "num_input_tokens_seen": 13472736,
      "step": 2227,
      "train_runtime": 2262.2421,
      "train_tokens_per_second": 5955.479
    },
    {
      "epoch": 0.8194942528735633,
      "grad_norm": 0.7259359359741211,
      "learning_rate": 0.00011820357077121296,
      "loss": 1.7829,
      "num_input_tokens_seen": 13477296,
      "step": 2228,
      "train_runtime": 2263.0393,
      "train_tokens_per_second": 5955.396
    },
    {
      "epoch": 0.8198620689655173,
      "grad_norm": 0.7328168749809265,
      "learning_rate": 0.00011816675869685257,
      "loss": 1.7047,
      "num_input_tokens_seen": 13482432,
      "step": 2229,
      "train_runtime": 2263.9055,
      "train_tokens_per_second": 5955.386
    },
    {
      "epoch": 0.8202298850574713,
      "grad_norm": 0.7600240111351013,
      "learning_rate": 0.00011812994662249219,
      "loss": 1.6535,
      "num_input_tokens_seen": 13489152,
      "step": 2230,
      "train_runtime": 2264.9988,
      "train_tokens_per_second": 5955.479
    },
    {
      "epoch": 0.8205977011494253,
      "grad_norm": 0.6951624751091003,
      "learning_rate": 0.0001180931345481318,
      "loss": 1.9195,
      "num_input_tokens_seen": 13498592,
      "step": 2231,
      "train_runtime": 2266.4353,
      "train_tokens_per_second": 5955.869
    },
    {
      "epoch": 0.8209655172413793,
      "grad_norm": 0.7410976886749268,
      "learning_rate": 0.00011805632247377141,
      "loss": 1.601,
      "num_input_tokens_seen": 13502768,
      "step": 2232,
      "train_runtime": 2267.1704,
      "train_tokens_per_second": 5955.78
    },
    {
      "epoch": 0.8213333333333334,
      "grad_norm": 0.832591712474823,
      "learning_rate": 0.00011801951039941102,
      "loss": 1.7121,
      "num_input_tokens_seen": 13510432,
      "step": 2233,
      "train_runtime": 2268.3858,
      "train_tokens_per_second": 5955.967
    },
    {
      "epoch": 0.8217011494252874,
      "grad_norm": 0.6923752427101135,
      "learning_rate": 0.00011798269832505063,
      "loss": 1.7457,
      "num_input_tokens_seen": 13516208,
      "step": 2234,
      "train_runtime": 2269.3369,
      "train_tokens_per_second": 5956.017
    },
    {
      "epoch": 0.8220689655172414,
      "grad_norm": 0.8069286346435547,
      "learning_rate": 0.00011794588625069024,
      "loss": 1.7031,
      "num_input_tokens_seen": 13520816,
      "step": 2235,
      "train_runtime": 2270.1437,
      "train_tokens_per_second": 5955.93
    },
    {
      "epoch": 0.8224367816091954,
      "grad_norm": 0.7739611268043518,
      "learning_rate": 0.00011790907417632983,
      "loss": 1.672,
      "num_input_tokens_seen": 13530256,
      "step": 2236,
      "train_runtime": 2271.5866,
      "train_tokens_per_second": 5956.302
    },
    {
      "epoch": 0.8228045977011494,
      "grad_norm": 0.7646172642707825,
      "learning_rate": 0.00011787226210196944,
      "loss": 1.819,
      "num_input_tokens_seen": 13535456,
      "step": 2237,
      "train_runtime": 2272.4355,
      "train_tokens_per_second": 5956.365
    },
    {
      "epoch": 0.8231724137931035,
      "grad_norm": 0.7784832715988159,
      "learning_rate": 0.00011783545002760905,
      "loss": 1.8317,
      "num_input_tokens_seen": 13540256,
      "step": 2238,
      "train_runtime": 2273.2752,
      "train_tokens_per_second": 5956.277
    },
    {
      "epoch": 0.8235402298850575,
      "grad_norm": 0.8190975785255432,
      "learning_rate": 0.00011779863795324869,
      "loss": 1.6273,
      "num_input_tokens_seen": 13544976,
      "step": 2239,
      "train_runtime": 2274.085,
      "train_tokens_per_second": 5956.231
    },
    {
      "epoch": 0.8239080459770115,
      "grad_norm": 0.6801204085350037,
      "learning_rate": 0.0001177618258788883,
      "loss": 1.7859,
      "num_input_tokens_seen": 13553088,
      "step": 2240,
      "train_runtime": 2275.3514,
      "train_tokens_per_second": 5956.481
    },
    {
      "epoch": 0.8242758620689655,
      "grad_norm": 0.7675638198852539,
      "learning_rate": 0.00011772501380452789,
      "loss": 1.6779,
      "num_input_tokens_seen": 13559008,
      "step": 2241,
      "train_runtime": 2276.3182,
      "train_tokens_per_second": 5956.552
    },
    {
      "epoch": 0.8246436781609195,
      "grad_norm": 0.7743769884109497,
      "learning_rate": 0.0001176882017301675,
      "loss": 1.8434,
      "num_input_tokens_seen": 13563792,
      "step": 2242,
      "train_runtime": 2277.1494,
      "train_tokens_per_second": 5956.479
    },
    {
      "epoch": 0.8250114942528736,
      "grad_norm": 0.7585687041282654,
      "learning_rate": 0.00011765138965580711,
      "loss": 1.6791,
      "num_input_tokens_seen": 13568608,
      "step": 2243,
      "train_runtime": 2277.9643,
      "train_tokens_per_second": 5956.462
    },
    {
      "epoch": 0.8253793103448276,
      "grad_norm": 0.7414730191230774,
      "learning_rate": 0.00011761457758144672,
      "loss": 1.8131,
      "num_input_tokens_seen": 13573344,
      "step": 2244,
      "train_runtime": 2278.7933,
      "train_tokens_per_second": 5956.374
    },
    {
      "epoch": 0.8257471264367816,
      "grad_norm": 0.7411015629768372,
      "learning_rate": 0.00011757776550708633,
      "loss": 1.7384,
      "num_input_tokens_seen": 13578160,
      "step": 2245,
      "train_runtime": 2279.6254,
      "train_tokens_per_second": 5956.312
    },
    {
      "epoch": 0.8261149425287356,
      "grad_norm": 0.7856464385986328,
      "learning_rate": 0.00011754095343272593,
      "loss": 1.6141,
      "num_input_tokens_seen": 13582832,
      "step": 2246,
      "train_runtime": 2280.4167,
      "train_tokens_per_second": 5956.294
    },
    {
      "epoch": 0.8264827586206897,
      "grad_norm": 0.7507676482200623,
      "learning_rate": 0.00011750414135836556,
      "loss": 1.446,
      "num_input_tokens_seen": 13588448,
      "step": 2247,
      "train_runtime": 2281.3449,
      "train_tokens_per_second": 5956.332
    },
    {
      "epoch": 0.8268505747126437,
      "grad_norm": 0.7500200271606445,
      "learning_rate": 0.00011746732928400517,
      "loss": 1.6697,
      "num_input_tokens_seen": 13595552,
      "step": 2248,
      "train_runtime": 2282.4981,
      "train_tokens_per_second": 5956.435
    },
    {
      "epoch": 0.8272183908045977,
      "grad_norm": 0.7292554974555969,
      "learning_rate": 0.00011743051720964477,
      "loss": 1.7127,
      "num_input_tokens_seen": 13600432,
      "step": 2249,
      "train_runtime": 2283.324,
      "train_tokens_per_second": 5956.418
    },
    {
      "epoch": 0.8275862068965517,
      "grad_norm": 0.7304660677909851,
      "learning_rate": 0.00011739370513528438,
      "loss": 1.8134,
      "num_input_tokens_seen": 13608592,
      "step": 2250,
      "train_runtime": 2284.6041,
      "train_tokens_per_second": 5956.652
    },
    {
      "epoch": 0.8279540229885057,
      "grad_norm": 0.8456048965454102,
      "learning_rate": 0.00011735689306092399,
      "loss": 1.7913,
      "num_input_tokens_seen": 13614080,
      "step": 2251,
      "train_runtime": 2286.0618,
      "train_tokens_per_second": 5955.254
    },
    {
      "epoch": 0.8283218390804598,
      "grad_norm": 0.8031347393989563,
      "learning_rate": 0.0001173200809865636,
      "loss": 1.8444,
      "num_input_tokens_seen": 13619392,
      "step": 2252,
      "train_runtime": 2286.9267,
      "train_tokens_per_second": 5955.325
    },
    {
      "epoch": 0.8286896551724138,
      "grad_norm": 0.6796359419822693,
      "learning_rate": 0.00011728326891220321,
      "loss": 1.6246,
      "num_input_tokens_seen": 13625664,
      "step": 2253,
      "train_runtime": 2287.9726,
      "train_tokens_per_second": 5955.344
    },
    {
      "epoch": 0.8290574712643678,
      "grad_norm": 0.7418872714042664,
      "learning_rate": 0.0001172464568378428,
      "loss": 1.8719,
      "num_input_tokens_seen": 13630944,
      "step": 2254,
      "train_runtime": 2288.8573,
      "train_tokens_per_second": 5955.349
    },
    {
      "epoch": 0.8294252873563218,
      "grad_norm": 0.8159528970718384,
      "learning_rate": 0.00011720964476348241,
      "loss": 1.8128,
      "num_input_tokens_seen": 13637104,
      "step": 2255,
      "train_runtime": 2289.8756,
      "train_tokens_per_second": 5955.391
    },
    {
      "epoch": 0.8297931034482758,
      "grad_norm": 0.7546332478523254,
      "learning_rate": 0.00011717283268912205,
      "loss": 1.7945,
      "num_input_tokens_seen": 13643456,
      "step": 2256,
      "train_runtime": 2290.8796,
      "train_tokens_per_second": 5955.554
    },
    {
      "epoch": 0.8301609195402299,
      "grad_norm": 0.7049574851989746,
      "learning_rate": 0.00011713602061476166,
      "loss": 1.6575,
      "num_input_tokens_seen": 13649984,
      "step": 2257,
      "train_runtime": 2291.943,
      "train_tokens_per_second": 5955.639
    },
    {
      "epoch": 0.8305287356321839,
      "grad_norm": 0.7260283827781677,
      "learning_rate": 0.00011709920854040127,
      "loss": 1.5161,
      "num_input_tokens_seen": 13658080,
      "step": 2258,
      "train_runtime": 2293.1947,
      "train_tokens_per_second": 5955.918
    },
    {
      "epoch": 0.8308965517241379,
      "grad_norm": 0.7271474599838257,
      "learning_rate": 0.00011706239646604086,
      "loss": 1.6993,
      "num_input_tokens_seen": 13664448,
      "step": 2259,
      "train_runtime": 2294.2081,
      "train_tokens_per_second": 5956.063
    },
    {
      "epoch": 0.8312643678160919,
      "grad_norm": 0.8448441028594971,
      "learning_rate": 0.00011702558439168047,
      "loss": 1.9481,
      "num_input_tokens_seen": 13669040,
      "step": 2260,
      "train_runtime": 2294.9979,
      "train_tokens_per_second": 5956.014
    },
    {
      "epoch": 0.8316321839080459,
      "grad_norm": 0.7977505922317505,
      "learning_rate": 0.00011698877231732008,
      "loss": 1.7906,
      "num_input_tokens_seen": 13673200,
      "step": 2261,
      "train_runtime": 2295.7282,
      "train_tokens_per_second": 5955.931
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.7925207614898682,
      "learning_rate": 0.00011695196024295969,
      "loss": 1.759,
      "num_input_tokens_seen": 13679072,
      "step": 2262,
      "train_runtime": 2296.687,
      "train_tokens_per_second": 5956.002
    },
    {
      "epoch": 0.832367816091954,
      "grad_norm": 0.7455686330795288,
      "learning_rate": 0.0001169151481685993,
      "loss": 1.7642,
      "num_input_tokens_seen": 13684400,
      "step": 2263,
      "train_runtime": 2297.5834,
      "train_tokens_per_second": 5955.997
    },
    {
      "epoch": 0.832735632183908,
      "grad_norm": 0.7316351532936096,
      "learning_rate": 0.00011687833609423891,
      "loss": 1.7993,
      "num_input_tokens_seen": 13691344,
      "step": 2264,
      "train_runtime": 2298.685,
      "train_tokens_per_second": 5956.164
    },
    {
      "epoch": 0.833103448275862,
      "grad_norm": 0.6481342911720276,
      "learning_rate": 0.00011684152401987853,
      "loss": 1.5944,
      "num_input_tokens_seen": 13699920,
      "step": 2265,
      "train_runtime": 2300.0246,
      "train_tokens_per_second": 5956.423
    },
    {
      "epoch": 0.833471264367816,
      "grad_norm": 0.75009685754776,
      "learning_rate": 0.00011680471194551814,
      "loss": 1.8563,
      "num_input_tokens_seen": 13707328,
      "step": 2266,
      "train_runtime": 2301.2178,
      "train_tokens_per_second": 5956.554
    },
    {
      "epoch": 0.8338390804597701,
      "grad_norm": 0.7056087255477905,
      "learning_rate": 0.00011676789987115775,
      "loss": 1.6878,
      "num_input_tokens_seen": 13715136,
      "step": 2267,
      "train_runtime": 2302.4258,
      "train_tokens_per_second": 5956.82
    },
    {
      "epoch": 0.8342068965517241,
      "grad_norm": 0.7421260476112366,
      "learning_rate": 0.00011673108779679736,
      "loss": 1.7236,
      "num_input_tokens_seen": 13720928,
      "step": 2268,
      "train_runtime": 2303.404,
      "train_tokens_per_second": 5956.805
    },
    {
      "epoch": 0.8345747126436781,
      "grad_norm": 0.7184244394302368,
      "learning_rate": 0.00011669427572243697,
      "loss": 1.7688,
      "num_input_tokens_seen": 13726288,
      "step": 2269,
      "train_runtime": 2304.2831,
      "train_tokens_per_second": 5956.858
    },
    {
      "epoch": 0.8349425287356322,
      "grad_norm": 0.7536584138870239,
      "learning_rate": 0.00011665746364807657,
      "loss": 1.7759,
      "num_input_tokens_seen": 13730688,
      "step": 2270,
      "train_runtime": 2305.0536,
      "train_tokens_per_second": 5956.776
    },
    {
      "epoch": 0.8353103448275863,
      "grad_norm": 0.6974349021911621,
      "learning_rate": 0.00011662065157371618,
      "loss": 1.7523,
      "num_input_tokens_seen": 13736784,
      "step": 2271,
      "train_runtime": 2306.0668,
      "train_tokens_per_second": 5956.802
    },
    {
      "epoch": 0.8356781609195403,
      "grad_norm": 0.5871516466140747,
      "learning_rate": 0.00011658383949935578,
      "loss": 1.4242,
      "num_input_tokens_seen": 13743200,
      "step": 2272,
      "train_runtime": 2307.0906,
      "train_tokens_per_second": 5956.94
    },
    {
      "epoch": 0.8360459770114943,
      "grad_norm": 0.7451143264770508,
      "learning_rate": 0.00011654702742499541,
      "loss": 1.7651,
      "num_input_tokens_seen": 13750944,
      "step": 2273,
      "train_runtime": 2308.2738,
      "train_tokens_per_second": 5957.241
    },
    {
      "epoch": 0.8364137931034483,
      "grad_norm": 0.7356095910072327,
      "learning_rate": 0.00011651021535063502,
      "loss": 1.6545,
      "num_input_tokens_seen": 13755184,
      "step": 2274,
      "train_runtime": 2308.9969,
      "train_tokens_per_second": 5957.212
    },
    {
      "epoch": 0.8367816091954023,
      "grad_norm": 0.6590993404388428,
      "learning_rate": 0.00011647340327627463,
      "loss": 1.7812,
      "num_input_tokens_seen": 13762272,
      "step": 2275,
      "train_runtime": 2310.1032,
      "train_tokens_per_second": 5957.427
    },
    {
      "epoch": 0.8371494252873564,
      "grad_norm": 0.8147951364517212,
      "learning_rate": 0.00011643659120191424,
      "loss": 1.7061,
      "num_input_tokens_seen": 13766944,
      "step": 2276,
      "train_runtime": 2310.9298,
      "train_tokens_per_second": 5957.318
    },
    {
      "epoch": 0.8375172413793104,
      "grad_norm": 0.7474108934402466,
      "learning_rate": 0.00011639977912755384,
      "loss": 1.6041,
      "num_input_tokens_seen": 13772400,
      "step": 2277,
      "train_runtime": 2311.8211,
      "train_tokens_per_second": 5957.381
    },
    {
      "epoch": 0.8378850574712644,
      "grad_norm": 0.7573875188827515,
      "learning_rate": 0.00011636296705319345,
      "loss": 1.6255,
      "num_input_tokens_seen": 13777168,
      "step": 2278,
      "train_runtime": 2312.6582,
      "train_tokens_per_second": 5957.287
    },
    {
      "epoch": 0.8382528735632184,
      "grad_norm": 0.6747617125511169,
      "learning_rate": 0.00011632615497883305,
      "loss": 1.4503,
      "num_input_tokens_seen": 13783536,
      "step": 2279,
      "train_runtime": 2313.6899,
      "train_tokens_per_second": 5957.383
    },
    {
      "epoch": 0.8386206896551724,
      "grad_norm": 0.7672241926193237,
      "learning_rate": 0.00011628934290447266,
      "loss": 1.753,
      "num_input_tokens_seen": 13789376,
      "step": 2280,
      "train_runtime": 2314.6685,
      "train_tokens_per_second": 5957.387
    },
    {
      "epoch": 0.8389885057471265,
      "grad_norm": 0.6858580112457275,
      "learning_rate": 0.00011625253083011227,
      "loss": 1.6245,
      "num_input_tokens_seen": 13796144,
      "step": 2281,
      "train_runtime": 2316.2175,
      "train_tokens_per_second": 5956.325
    },
    {
      "epoch": 0.8393563218390805,
      "grad_norm": 0.8250431418418884,
      "learning_rate": 0.0001162157187557519,
      "loss": 1.9339,
      "num_input_tokens_seen": 13801184,
      "step": 2282,
      "train_runtime": 2317.0649,
      "train_tokens_per_second": 5956.322
    },
    {
      "epoch": 0.8397241379310345,
      "grad_norm": 0.7551794648170471,
      "learning_rate": 0.0001161789066813915,
      "loss": 1.7384,
      "num_input_tokens_seen": 13808240,
      "step": 2283,
      "train_runtime": 2318.209,
      "train_tokens_per_second": 5956.426
    },
    {
      "epoch": 0.8400919540229885,
      "grad_norm": 0.7227142453193665,
      "learning_rate": 0.00011614209460703111,
      "loss": 1.7805,
      "num_input_tokens_seen": 13813152,
      "step": 2284,
      "train_runtime": 2319.0509,
      "train_tokens_per_second": 5956.381
    },
    {
      "epoch": 0.8404597701149426,
      "grad_norm": 0.7946128845214844,
      "learning_rate": 0.00011610528253267072,
      "loss": 1.7317,
      "num_input_tokens_seen": 13819584,
      "step": 2285,
      "train_runtime": 2320.0935,
      "train_tokens_per_second": 5956.477
    },
    {
      "epoch": 0.8408275862068966,
      "grad_norm": 0.7335671782493591,
      "learning_rate": 0.00011606847045831033,
      "loss": 1.7084,
      "num_input_tokens_seen": 13824816,
      "step": 2286,
      "train_runtime": 2320.9688,
      "train_tokens_per_second": 5956.485
    },
    {
      "epoch": 0.8411954022988506,
      "grad_norm": 0.7103782296180725,
      "learning_rate": 0.00011603165838394994,
      "loss": 1.6168,
      "num_input_tokens_seen": 13832112,
      "step": 2287,
      "train_runtime": 2322.1484,
      "train_tokens_per_second": 5956.601
    },
    {
      "epoch": 0.8415632183908046,
      "grad_norm": 0.7533090114593506,
      "learning_rate": 0.00011599484630958955,
      "loss": 1.6981,
      "num_input_tokens_seen": 13837856,
      "step": 2288,
      "train_runtime": 2323.0973,
      "train_tokens_per_second": 5956.641
    },
    {
      "epoch": 0.8419310344827586,
      "grad_norm": 0.8349571824073792,
      "learning_rate": 0.00011595803423522916,
      "loss": 1.9,
      "num_input_tokens_seen": 13843472,
      "step": 2289,
      "train_runtime": 2324.029,
      "train_tokens_per_second": 5956.669
    },
    {
      "epoch": 0.8422988505747127,
      "grad_norm": 0.7255327701568604,
      "learning_rate": 0.00011592122216086878,
      "loss": 1.5794,
      "num_input_tokens_seen": 13849456,
      "step": 2290,
      "train_runtime": 2325.0066,
      "train_tokens_per_second": 5956.738
    },
    {
      "epoch": 0.8426666666666667,
      "grad_norm": 0.8030518889427185,
      "learning_rate": 0.00011588441008650839,
      "loss": 1.6444,
      "num_input_tokens_seen": 13854176,
      "step": 2291,
      "train_runtime": 2325.8038,
      "train_tokens_per_second": 5956.726
    },
    {
      "epoch": 0.8430344827586207,
      "grad_norm": 0.8301403522491455,
      "learning_rate": 0.000115847598012148,
      "loss": 1.7434,
      "num_input_tokens_seen": 13859040,
      "step": 2292,
      "train_runtime": 2326.636,
      "train_tokens_per_second": 5956.686
    },
    {
      "epoch": 0.8434022988505747,
      "grad_norm": 0.6595023274421692,
      "learning_rate": 0.0001158107859377876,
      "loss": 1.6479,
      "num_input_tokens_seen": 13866160,
      "step": 2293,
      "train_runtime": 2327.8028,
      "train_tokens_per_second": 5956.759
    },
    {
      "epoch": 0.8437701149425287,
      "grad_norm": 0.795469343662262,
      "learning_rate": 0.00011577397386342721,
      "loss": 1.8107,
      "num_input_tokens_seen": 13871104,
      "step": 2294,
      "train_runtime": 2328.6431,
      "train_tokens_per_second": 5956.733
    },
    {
      "epoch": 0.8441379310344828,
      "grad_norm": 0.811279296875,
      "learning_rate": 0.00011573716178906681,
      "loss": 1.7903,
      "num_input_tokens_seen": 13878176,
      "step": 2295,
      "train_runtime": 2329.7807,
      "train_tokens_per_second": 5956.859
    },
    {
      "epoch": 0.8445057471264368,
      "grad_norm": 0.8911852240562439,
      "learning_rate": 0.00011570034971470642,
      "loss": 1.7742,
      "num_input_tokens_seen": 13882512,
      "step": 2296,
      "train_runtime": 2330.527,
      "train_tokens_per_second": 5956.812
    },
    {
      "epoch": 0.8448735632183908,
      "grad_norm": 0.6991628408432007,
      "learning_rate": 0.00011566353764034603,
      "loss": 1.6731,
      "num_input_tokens_seen": 13888736,
      "step": 2297,
      "train_runtime": 2331.5644,
      "train_tokens_per_second": 5956.831
    },
    {
      "epoch": 0.8452413793103448,
      "grad_norm": 0.9091153740882874,
      "learning_rate": 0.00011562672556598564,
      "loss": 1.8276,
      "num_input_tokens_seen": 13893936,
      "step": 2298,
      "train_runtime": 2332.4512,
      "train_tokens_per_second": 5956.796
    },
    {
      "epoch": 0.8456091954022988,
      "grad_norm": 0.7464302182197571,
      "learning_rate": 0.00011558991349162527,
      "loss": 1.7393,
      "num_input_tokens_seen": 13899088,
      "step": 2299,
      "train_runtime": 2333.3203,
      "train_tokens_per_second": 5956.785
    },
    {
      "epoch": 0.8459770114942529,
      "grad_norm": 0.7169905304908752,
      "learning_rate": 0.00011555310141726488,
      "loss": 1.5769,
      "num_input_tokens_seen": 13905344,
      "step": 2300,
      "train_runtime": 2334.3444,
      "train_tokens_per_second": 5956.852
    },
    {
      "epoch": 0.8463448275862069,
      "grad_norm": 0.7689448595046997,
      "learning_rate": 0.00011551628934290448,
      "loss": 1.8292,
      "num_input_tokens_seen": 13911136,
      "step": 2301,
      "train_runtime": 2335.3171,
      "train_tokens_per_second": 5956.851
    },
    {
      "epoch": 0.8467126436781609,
      "grad_norm": 0.8222184777259827,
      "learning_rate": 0.00011547947726854409,
      "loss": 1.7091,
      "num_input_tokens_seen": 13918912,
      "step": 2302,
      "train_runtime": 2336.5371,
      "train_tokens_per_second": 5957.069
    },
    {
      "epoch": 0.8470804597701149,
      "grad_norm": 0.8501462340354919,
      "learning_rate": 0.0001154426651941837,
      "loss": 1.9229,
      "num_input_tokens_seen": 13923424,
      "step": 2303,
      "train_runtime": 2337.313,
      "train_tokens_per_second": 5957.022
    },
    {
      "epoch": 0.847448275862069,
      "grad_norm": 0.7013068199157715,
      "learning_rate": 0.0001154058531198233,
      "loss": 1.7237,
      "num_input_tokens_seen": 13931792,
      "step": 2304,
      "train_runtime": 2338.5938,
      "train_tokens_per_second": 5957.337
    },
    {
      "epoch": 0.847816091954023,
      "grad_norm": 0.8062384724617004,
      "learning_rate": 0.00011536904104546291,
      "loss": 1.807,
      "num_input_tokens_seen": 13937152,
      "step": 2305,
      "train_runtime": 2339.4781,
      "train_tokens_per_second": 5957.377
    },
    {
      "epoch": 0.848183908045977,
      "grad_norm": 0.7189160585403442,
      "learning_rate": 0.00011533222897110252,
      "loss": 1.7392,
      "num_input_tokens_seen": 13943888,
      "step": 2306,
      "train_runtime": 2340.5742,
      "train_tokens_per_second": 5957.465
    },
    {
      "epoch": 0.848551724137931,
      "grad_norm": 0.6561643481254578,
      "learning_rate": 0.00011529541689674214,
      "loss": 1.4096,
      "num_input_tokens_seen": 13952880,
      "step": 2307,
      "train_runtime": 2341.9742,
      "train_tokens_per_second": 5957.743
    },
    {
      "epoch": 0.848919540229885,
      "grad_norm": 0.7770561575889587,
      "learning_rate": 0.00011525860482238175,
      "loss": 1.871,
      "num_input_tokens_seen": 13960736,
      "step": 2308,
      "train_runtime": 2343.182,
      "train_tokens_per_second": 5958.024
    },
    {
      "epoch": 0.849287356321839,
      "grad_norm": 0.7397069334983826,
      "learning_rate": 0.00011522179274802136,
      "loss": 1.6995,
      "num_input_tokens_seen": 13965824,
      "step": 2309,
      "train_runtime": 2344.0311,
      "train_tokens_per_second": 5958.037
    },
    {
      "epoch": 0.8496551724137931,
      "grad_norm": 0.7244512438774109,
      "learning_rate": 0.00011518498067366097,
      "loss": 1.7351,
      "num_input_tokens_seen": 13972160,
      "step": 2310,
      "train_runtime": 2345.08,
      "train_tokens_per_second": 5958.074
    },
    {
      "epoch": 0.8500229885057471,
      "grad_norm": 0.7044952511787415,
      "learning_rate": 0.00011514816859930058,
      "loss": 1.5918,
      "num_input_tokens_seen": 13976992,
      "step": 2311,
      "train_runtime": 2346.4082,
      "train_tokens_per_second": 5956.761
    },
    {
      "epoch": 0.8503908045977011,
      "grad_norm": 0.8645678758621216,
      "learning_rate": 0.00011511135652494019,
      "loss": 1.7591,
      "num_input_tokens_seen": 13984496,
      "step": 2312,
      "train_runtime": 2347.5906,
      "train_tokens_per_second": 5956.957
    },
    {
      "epoch": 0.8507586206896551,
      "grad_norm": 0.8053102493286133,
      "learning_rate": 0.0001150745444505798,
      "loss": 1.7409,
      "num_input_tokens_seen": 13989728,
      "step": 2313,
      "train_runtime": 2348.4336,
      "train_tokens_per_second": 5957.046
    },
    {
      "epoch": 0.8511264367816092,
      "grad_norm": 0.7906848788261414,
      "learning_rate": 0.00011503773237621939,
      "loss": 1.9027,
      "num_input_tokens_seen": 13994512,
      "step": 2314,
      "train_runtime": 2349.2617,
      "train_tokens_per_second": 5956.983
    },
    {
      "epoch": 0.8514942528735632,
      "grad_norm": 0.7035423517227173,
      "learning_rate": 0.000115000920301859,
      "loss": 1.6731,
      "num_input_tokens_seen": 14001200,
      "step": 2315,
      "train_runtime": 2350.3109,
      "train_tokens_per_second": 5957.169
    },
    {
      "epoch": 0.8518620689655172,
      "grad_norm": 0.7528830766677856,
      "learning_rate": 0.00011496410822749864,
      "loss": 1.7703,
      "num_input_tokens_seen": 14006816,
      "step": 2316,
      "train_runtime": 2351.2203,
      "train_tokens_per_second": 5957.254
    },
    {
      "epoch": 0.8522298850574712,
      "grad_norm": 0.8125345706939697,
      "learning_rate": 0.00011492729615313825,
      "loss": 1.9351,
      "num_input_tokens_seen": 14012544,
      "step": 2317,
      "train_runtime": 2352.1874,
      "train_tokens_per_second": 5957.24
    },
    {
      "epoch": 0.8525977011494252,
      "grad_norm": 0.7331763505935669,
      "learning_rate": 0.00011489048407877786,
      "loss": 1.6909,
      "num_input_tokens_seen": 14018704,
      "step": 2318,
      "train_runtime": 2353.1836,
      "train_tokens_per_second": 5957.335
    },
    {
      "epoch": 0.8529655172413793,
      "grad_norm": 0.8239543437957764,
      "learning_rate": 0.00011485367200441745,
      "loss": 1.8601,
      "num_input_tokens_seen": 14023776,
      "step": 2319,
      "train_runtime": 2354.0342,
      "train_tokens_per_second": 5957.337
    },
    {
      "epoch": 0.8533333333333334,
      "grad_norm": 0.8026068210601807,
      "learning_rate": 0.00011481685993005706,
      "loss": 2.0051,
      "num_input_tokens_seen": 14028384,
      "step": 2320,
      "train_runtime": 2354.8242,
      "train_tokens_per_second": 5957.296
    },
    {
      "epoch": 0.8537011494252874,
      "grad_norm": 0.735466718673706,
      "learning_rate": 0.00011478004785569667,
      "loss": 1.9317,
      "num_input_tokens_seen": 14032704,
      "step": 2321,
      "train_runtime": 2355.5656,
      "train_tokens_per_second": 5957.255
    },
    {
      "epoch": 0.8540689655172414,
      "grad_norm": 0.7632676362991333,
      "learning_rate": 0.00011474323578133628,
      "loss": 1.8079,
      "num_input_tokens_seen": 14037520,
      "step": 2322,
      "train_runtime": 2356.3887,
      "train_tokens_per_second": 5957.218
    },
    {
      "epoch": 0.8544367816091955,
      "grad_norm": 0.772436797618866,
      "learning_rate": 0.00011470642370697589,
      "loss": 1.6746,
      "num_input_tokens_seen": 14044416,
      "step": 2323,
      "train_runtime": 2357.4883,
      "train_tokens_per_second": 5957.364
    },
    {
      "epoch": 0.8548045977011495,
      "grad_norm": 0.7564178109169006,
      "learning_rate": 0.0001146696116326155,
      "loss": 1.7481,
      "num_input_tokens_seen": 14050976,
      "step": 2324,
      "train_runtime": 2358.5274,
      "train_tokens_per_second": 5957.521
    },
    {
      "epoch": 0.8551724137931035,
      "grad_norm": 0.7316631078720093,
      "learning_rate": 0.00011463279955825512,
      "loss": 1.7135,
      "num_input_tokens_seen": 14056912,
      "step": 2325,
      "train_runtime": 2359.5043,
      "train_tokens_per_second": 5957.57
    },
    {
      "epoch": 0.8555402298850575,
      "grad_norm": 0.7063326835632324,
      "learning_rate": 0.00011459598748389473,
      "loss": 1.6468,
      "num_input_tokens_seen": 14063232,
      "step": 2326,
      "train_runtime": 2360.5351,
      "train_tokens_per_second": 5957.646
    },
    {
      "epoch": 0.8559080459770115,
      "grad_norm": 0.798062264919281,
      "learning_rate": 0.00011455917540953434,
      "loss": 1.6813,
      "num_input_tokens_seen": 14069808,
      "step": 2327,
      "train_runtime": 2361.584,
      "train_tokens_per_second": 5957.784
    },
    {
      "epoch": 0.8562758620689656,
      "grad_norm": 0.7643727660179138,
      "learning_rate": 0.00011452236333517394,
      "loss": 1.8075,
      "num_input_tokens_seen": 14076032,
      "step": 2328,
      "train_runtime": 2362.6133,
      "train_tokens_per_second": 5957.823
    },
    {
      "epoch": 0.8566436781609196,
      "grad_norm": 0.8581653833389282,
      "learning_rate": 0.00011448555126081355,
      "loss": 2.06,
      "num_input_tokens_seen": 14081008,
      "step": 2329,
      "train_runtime": 2363.462,
      "train_tokens_per_second": 5957.789
    },
    {
      "epoch": 0.8570114942528736,
      "grad_norm": 0.740445077419281,
      "learning_rate": 0.00011444873918645316,
      "loss": 1.5826,
      "num_input_tokens_seen": 14086496,
      "step": 2330,
      "train_runtime": 2364.3891,
      "train_tokens_per_second": 5957.774
    },
    {
      "epoch": 0.8573793103448276,
      "grad_norm": 0.9068762063980103,
      "learning_rate": 0.00011441192711209277,
      "loss": 1.8779,
      "num_input_tokens_seen": 14095328,
      "step": 2331,
      "train_runtime": 2365.7936,
      "train_tokens_per_second": 5957.97
    },
    {
      "epoch": 0.8577471264367816,
      "grad_norm": 0.7269871830940247,
      "learning_rate": 0.00011437511503773237,
      "loss": 1.7484,
      "num_input_tokens_seen": 14100272,
      "step": 2332,
      "train_runtime": 2366.6589,
      "train_tokens_per_second": 5957.881
    },
    {
      "epoch": 0.8581149425287357,
      "grad_norm": 0.8406885862350464,
      "learning_rate": 0.000114338302963372,
      "loss": 1.7737,
      "num_input_tokens_seen": 14104624,
      "step": 2333,
      "train_runtime": 2367.4101,
      "train_tokens_per_second": 5957.829
    },
    {
      "epoch": 0.8584827586206897,
      "grad_norm": 0.7570016384124756,
      "learning_rate": 0.00011430149088901161,
      "loss": 1.8276,
      "num_input_tokens_seen": 14109840,
      "step": 2334,
      "train_runtime": 2368.3,
      "train_tokens_per_second": 5957.792
    },
    {
      "epoch": 0.8588505747126437,
      "grad_norm": 0.7218365669250488,
      "learning_rate": 0.00011426467881465122,
      "loss": 1.6123,
      "num_input_tokens_seen": 14114784,
      "step": 2335,
      "train_runtime": 2369.1446,
      "train_tokens_per_second": 5957.755
    },
    {
      "epoch": 0.8592183908045977,
      "grad_norm": 0.7858163714408875,
      "learning_rate": 0.00011422786674029083,
      "loss": 1.8551,
      "num_input_tokens_seen": 14119504,
      "step": 2336,
      "train_runtime": 2369.9484,
      "train_tokens_per_second": 5957.726
    },
    {
      "epoch": 0.8595862068965517,
      "grad_norm": 0.7593346238136292,
      "learning_rate": 0.00011419105466593042,
      "loss": 1.824,
      "num_input_tokens_seen": 14124144,
      "step": 2337,
      "train_runtime": 2370.744,
      "train_tokens_per_second": 5957.684
    },
    {
      "epoch": 0.8599540229885058,
      "grad_norm": 0.7639790177345276,
      "learning_rate": 0.00011415424259157003,
      "loss": 1.7448,
      "num_input_tokens_seen": 14134080,
      "step": 2338,
      "train_runtime": 2372.2227,
      "train_tokens_per_second": 5958.159
    },
    {
      "epoch": 0.8603218390804598,
      "grad_norm": 0.8189948797225952,
      "learning_rate": 0.00011411743051720964,
      "loss": 1.6927,
      "num_input_tokens_seen": 14138688,
      "step": 2339,
      "train_runtime": 2373.0139,
      "train_tokens_per_second": 5958.114
    },
    {
      "epoch": 0.8606896551724138,
      "grad_norm": 0.7642329335212708,
      "learning_rate": 0.00011408061844284925,
      "loss": 1.6876,
      "num_input_tokens_seen": 14143600,
      "step": 2340,
      "train_runtime": 2373.8589,
      "train_tokens_per_second": 5958.063
    },
    {
      "epoch": 0.8610574712643678,
      "grad_norm": 0.7678464651107788,
      "learning_rate": 0.00011404380636848886,
      "loss": 1.6466,
      "num_input_tokens_seen": 14151440,
      "step": 2341,
      "train_runtime": 2375.6137,
      "train_tokens_per_second": 5956.962
    },
    {
      "epoch": 0.8614252873563218,
      "grad_norm": 0.7687625885009766,
      "learning_rate": 0.00011400699429412848,
      "loss": 1.6773,
      "num_input_tokens_seen": 14157200,
      "step": 2342,
      "train_runtime": 2376.5428,
      "train_tokens_per_second": 5957.057
    },
    {
      "epoch": 0.8617931034482759,
      "grad_norm": 0.761159360408783,
      "learning_rate": 0.00011397018221976809,
      "loss": 1.8342,
      "num_input_tokens_seen": 14164432,
      "step": 2343,
      "train_runtime": 2377.7258,
      "train_tokens_per_second": 5957.134
    },
    {
      "epoch": 0.8621609195402299,
      "grad_norm": 0.791181743144989,
      "learning_rate": 0.0001139333701454077,
      "loss": 1.6291,
      "num_input_tokens_seen": 14172144,
      "step": 2344,
      "train_runtime": 2378.9399,
      "train_tokens_per_second": 5957.336
    },
    {
      "epoch": 0.8625287356321839,
      "grad_norm": 0.8065009117126465,
      "learning_rate": 0.00011389655807104731,
      "loss": 1.7938,
      "num_input_tokens_seen": 14177360,
      "step": 2345,
      "train_runtime": 2379.8049,
      "train_tokens_per_second": 5957.362
    },
    {
      "epoch": 0.8628965517241379,
      "grad_norm": 0.7265875935554504,
      "learning_rate": 0.00011385974599668692,
      "loss": 1.5831,
      "num_input_tokens_seen": 14184240,
      "step": 2346,
      "train_runtime": 2380.9087,
      "train_tokens_per_second": 5957.49
    },
    {
      "epoch": 0.863264367816092,
      "grad_norm": 0.738709032535553,
      "learning_rate": 0.00011382293392232653,
      "loss": 1.7015,
      "num_input_tokens_seen": 14191344,
      "step": 2347,
      "train_runtime": 2382.0348,
      "train_tokens_per_second": 5957.656
    },
    {
      "epoch": 0.863632183908046,
      "grad_norm": 0.8164908289909363,
      "learning_rate": 0.00011378612184796614,
      "loss": 1.8793,
      "num_input_tokens_seen": 14196016,
      "step": 2348,
      "train_runtime": 2382.8404,
      "train_tokens_per_second": 5957.603
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.7305906414985657,
      "learning_rate": 0.00011374930977360574,
      "loss": 1.7451,
      "num_input_tokens_seen": 14203136,
      "step": 2349,
      "train_runtime": 2383.9815,
      "train_tokens_per_second": 5957.737
    },
    {
      "epoch": 0.864367816091954,
      "grad_norm": 0.7419806718826294,
      "learning_rate": 0.00011371249769924537,
      "loss": 1.7986,
      "num_input_tokens_seen": 14210000,
      "step": 2350,
      "train_runtime": 2385.0887,
      "train_tokens_per_second": 5957.85
    },
    {
      "epoch": 0.864735632183908,
      "grad_norm": 0.76914381980896,
      "learning_rate": 0.00011367568562488498,
      "loss": 1.497,
      "num_input_tokens_seen": 14215392,
      "step": 2351,
      "train_runtime": 2385.9682,
      "train_tokens_per_second": 5957.914
    },
    {
      "epoch": 0.865103448275862,
      "grad_norm": 0.8450071215629578,
      "learning_rate": 0.00011363887355052458,
      "loss": 1.7459,
      "num_input_tokens_seen": 14220784,
      "step": 2352,
      "train_runtime": 2386.8719,
      "train_tokens_per_second": 5957.917
    },
    {
      "epoch": 0.8654712643678161,
      "grad_norm": 0.8300321698188782,
      "learning_rate": 0.0001136020614761642,
      "loss": 1.8838,
      "num_input_tokens_seen": 14226304,
      "step": 2353,
      "train_runtime": 2387.7914,
      "train_tokens_per_second": 5957.934
    },
    {
      "epoch": 0.8658390804597701,
      "grad_norm": 0.8601062297821045,
      "learning_rate": 0.0001135652494018038,
      "loss": 1.6558,
      "num_input_tokens_seen": 14230544,
      "step": 2354,
      "train_runtime": 2388.568,
      "train_tokens_per_second": 5957.772
    },
    {
      "epoch": 0.8662068965517241,
      "grad_norm": 0.7347440719604492,
      "learning_rate": 0.0001135284373274434,
      "loss": 1.624,
      "num_input_tokens_seen": 14236096,
      "step": 2355,
      "train_runtime": 2389.4757,
      "train_tokens_per_second": 5957.833
    },
    {
      "epoch": 0.8665747126436781,
      "grad_norm": 0.7797635793685913,
      "learning_rate": 0.00011349162525308301,
      "loss": 1.6137,
      "num_input_tokens_seen": 14242896,
      "step": 2356,
      "train_runtime": 2390.5773,
      "train_tokens_per_second": 5957.932
    },
    {
      "epoch": 0.8669425287356322,
      "grad_norm": 0.7344309687614441,
      "learning_rate": 0.00011345481317872262,
      "loss": 1.5892,
      "num_input_tokens_seen": 14248368,
      "step": 2357,
      "train_runtime": 2391.5026,
      "train_tokens_per_second": 5957.915
    },
    {
      "epoch": 0.8673103448275862,
      "grad_norm": 0.805598795413971,
      "learning_rate": 0.00011341800110436222,
      "loss": 1.7587,
      "num_input_tokens_seen": 14254880,
      "step": 2358,
      "train_runtime": 2392.5636,
      "train_tokens_per_second": 5957.994
    },
    {
      "epoch": 0.8676781609195402,
      "grad_norm": 0.7437198162078857,
      "learning_rate": 0.00011338118903000186,
      "loss": 1.6393,
      "num_input_tokens_seen": 14261696,
      "step": 2359,
      "train_runtime": 2393.6655,
      "train_tokens_per_second": 5958.099
    },
    {
      "epoch": 0.8680459770114942,
      "grad_norm": 0.7576062679290771,
      "learning_rate": 0.00011334437695564146,
      "loss": 1.8311,
      "num_input_tokens_seen": 14267184,
      "step": 2360,
      "train_runtime": 2394.5848,
      "train_tokens_per_second": 5958.103
    },
    {
      "epoch": 0.8684137931034482,
      "grad_norm": 0.7373422980308533,
      "learning_rate": 0.00011330756488128106,
      "loss": 1.5635,
      "num_input_tokens_seen": 14274368,
      "step": 2361,
      "train_runtime": 2395.7469,
      "train_tokens_per_second": 5958.212
    },
    {
      "epoch": 0.8687816091954023,
      "grad_norm": 0.7066559195518494,
      "learning_rate": 0.00011327075280692067,
      "loss": 1.5493,
      "num_input_tokens_seen": 14280560,
      "step": 2362,
      "train_runtime": 2396.765,
      "train_tokens_per_second": 5958.265
    },
    {
      "epoch": 0.8691494252873563,
      "grad_norm": 0.7652391791343689,
      "learning_rate": 0.00011323394073256028,
      "loss": 1.6412,
      "num_input_tokens_seen": 14287344,
      "step": 2363,
      "train_runtime": 2397.8573,
      "train_tokens_per_second": 5958.38
    },
    {
      "epoch": 0.8695172413793103,
      "grad_norm": 0.741844892501831,
      "learning_rate": 0.00011319712865819989,
      "loss": 1.8602,
      "num_input_tokens_seen": 14291920,
      "step": 2364,
      "train_runtime": 2398.6533,
      "train_tokens_per_second": 5958.31
    },
    {
      "epoch": 0.8698850574712643,
      "grad_norm": 0.7472719550132751,
      "learning_rate": 0.0001131603165838395,
      "loss": 1.6237,
      "num_input_tokens_seen": 14296320,
      "step": 2365,
      "train_runtime": 2399.4563,
      "train_tokens_per_second": 5958.15
    },
    {
      "epoch": 0.8702528735632183,
      "grad_norm": 0.7941020727157593,
      "learning_rate": 0.00011312350450947911,
      "loss": 1.5408,
      "num_input_tokens_seen": 14301664,
      "step": 2366,
      "train_runtime": 2400.3329,
      "train_tokens_per_second": 5958.2
    },
    {
      "epoch": 0.8706206896551724,
      "grad_norm": 0.7930315732955933,
      "learning_rate": 0.00011308669243511873,
      "loss": 1.7939,
      "num_input_tokens_seen": 14308320,
      "step": 2367,
      "train_runtime": 2401.3987,
      "train_tokens_per_second": 5958.328
    },
    {
      "epoch": 0.8709885057471264,
      "grad_norm": 0.7866606116294861,
      "learning_rate": 0.00011304988036075834,
      "loss": 1.7819,
      "num_input_tokens_seen": 14313616,
      "step": 2368,
      "train_runtime": 2402.2818,
      "train_tokens_per_second": 5958.342
    },
    {
      "epoch": 0.8713563218390804,
      "grad_norm": 0.7377987504005432,
      "learning_rate": 0.00011301306828639795,
      "loss": 1.9263,
      "num_input_tokens_seen": 14318256,
      "step": 2369,
      "train_runtime": 2403.1044,
      "train_tokens_per_second": 5958.233
    },
    {
      "epoch": 0.8717241379310345,
      "grad_norm": 0.8972067832946777,
      "learning_rate": 0.00011297625621203756,
      "loss": 1.8571,
      "num_input_tokens_seen": 14324752,
      "step": 2370,
      "train_runtime": 2404.1569,
      "train_tokens_per_second": 5958.327
    },
    {
      "epoch": 0.8720919540229886,
      "grad_norm": 0.8178853392601013,
      "learning_rate": 0.00011293944413767717,
      "loss": 1.8199,
      "num_input_tokens_seen": 14330608,
      "step": 2371,
      "train_runtime": 2405.6503,
      "train_tokens_per_second": 5957.062
    },
    {
      "epoch": 0.8724597701149426,
      "grad_norm": 0.7937817573547363,
      "learning_rate": 0.00011290263206331678,
      "loss": 1.7461,
      "num_input_tokens_seen": 14334576,
      "step": 2372,
      "train_runtime": 2406.3428,
      "train_tokens_per_second": 5956.997
    },
    {
      "epoch": 0.8728275862068966,
      "grad_norm": 0.8121699094772339,
      "learning_rate": 0.00011286581998895637,
      "loss": 1.8924,
      "num_input_tokens_seen": 14340080,
      "step": 2373,
      "train_runtime": 2407.2598,
      "train_tokens_per_second": 5957.014
    },
    {
      "epoch": 0.8731954022988506,
      "grad_norm": 0.7203399538993835,
      "learning_rate": 0.00011282900791459598,
      "loss": 1.8309,
      "num_input_tokens_seen": 14345952,
      "step": 2374,
      "train_runtime": 2408.2133,
      "train_tokens_per_second": 5957.094
    },
    {
      "epoch": 0.8735632183908046,
      "grad_norm": 0.7035325169563293,
      "learning_rate": 0.00011279219584023559,
      "loss": 1.8768,
      "num_input_tokens_seen": 14351712,
      "step": 2375,
      "train_runtime": 2409.1489,
      "train_tokens_per_second": 5957.171
    },
    {
      "epoch": 0.8739310344827587,
      "grad_norm": 0.7170698046684265,
      "learning_rate": 0.00011275538376587523,
      "loss": 1.7413,
      "num_input_tokens_seen": 14356368,
      "step": 2376,
      "train_runtime": 2409.9492,
      "train_tokens_per_second": 5957.125
    },
    {
      "epoch": 0.8742988505747127,
      "grad_norm": 0.7655664682388306,
      "learning_rate": 0.00011271857169151483,
      "loss": 1.7041,
      "num_input_tokens_seen": 14364928,
      "step": 2377,
      "train_runtime": 2411.2495,
      "train_tokens_per_second": 5957.462
    },
    {
      "epoch": 0.8746666666666667,
      "grad_norm": 0.8381845951080322,
      "learning_rate": 0.00011268175961715444,
      "loss": 1.8072,
      "num_input_tokens_seen": 14369744,
      "step": 2378,
      "train_runtime": 2412.0879,
      "train_tokens_per_second": 5957.388
    },
    {
      "epoch": 0.8750344827586207,
      "grad_norm": 0.8350390791893005,
      "learning_rate": 0.00011264494754279404,
      "loss": 1.7421,
      "num_input_tokens_seen": 14374976,
      "step": 2379,
      "train_runtime": 2412.9783,
      "train_tokens_per_second": 5957.358
    },
    {
      "epoch": 0.8754022988505747,
      "grad_norm": 0.8463245034217834,
      "learning_rate": 0.00011260813546843365,
      "loss": 1.8437,
      "num_input_tokens_seen": 14379344,
      "step": 2380,
      "train_runtime": 2413.7322,
      "train_tokens_per_second": 5957.307
    },
    {
      "epoch": 0.8757701149425288,
      "grad_norm": 0.755520761013031,
      "learning_rate": 0.00011257132339407326,
      "loss": 1.7141,
      "num_input_tokens_seen": 14385568,
      "step": 2381,
      "train_runtime": 2414.7521,
      "train_tokens_per_second": 5957.368
    },
    {
      "epoch": 0.8761379310344828,
      "grad_norm": 0.7731467485427856,
      "learning_rate": 0.00011253451131971287,
      "loss": 1.8212,
      "num_input_tokens_seen": 14391168,
      "step": 2382,
      "train_runtime": 2415.6833,
      "train_tokens_per_second": 5957.39
    },
    {
      "epoch": 0.8765057471264368,
      "grad_norm": 0.7874716520309448,
      "learning_rate": 0.00011249769924535247,
      "loss": 1.7065,
      "num_input_tokens_seen": 14397504,
      "step": 2383,
      "train_runtime": 2416.7038,
      "train_tokens_per_second": 5957.496
    },
    {
      "epoch": 0.8768735632183908,
      "grad_norm": 0.8985732793807983,
      "learning_rate": 0.00011246088717099208,
      "loss": 1.5969,
      "num_input_tokens_seen": 14404064,
      "step": 2384,
      "train_runtime": 2417.7451,
      "train_tokens_per_second": 5957.644
    },
    {
      "epoch": 0.8772413793103448,
      "grad_norm": 0.7792977094650269,
      "learning_rate": 0.0001124240750966317,
      "loss": 1.733,
      "num_input_tokens_seen": 14409936,
      "step": 2385,
      "train_runtime": 2418.6848,
      "train_tokens_per_second": 5957.757
    },
    {
      "epoch": 0.8776091954022989,
      "grad_norm": 0.7667710185050964,
      "learning_rate": 0.00011238726302227131,
      "loss": 1.5803,
      "num_input_tokens_seen": 14414976,
      "step": 2386,
      "train_runtime": 2419.5314,
      "train_tokens_per_second": 5957.755
    },
    {
      "epoch": 0.8779770114942529,
      "grad_norm": 0.8061090707778931,
      "learning_rate": 0.00011235045094791092,
      "loss": 1.667,
      "num_input_tokens_seen": 14420544,
      "step": 2387,
      "train_runtime": 2420.4724,
      "train_tokens_per_second": 5957.74
    },
    {
      "epoch": 0.8783448275862069,
      "grad_norm": 0.7755416035652161,
      "learning_rate": 0.00011231363887355053,
      "loss": 1.5153,
      "num_input_tokens_seen": 14429792,
      "step": 2388,
      "train_runtime": 2421.8919,
      "train_tokens_per_second": 5958.066
    },
    {
      "epoch": 0.8787126436781609,
      "grad_norm": 0.9299578666687012,
      "learning_rate": 0.00011227682679919014,
      "loss": 1.8097,
      "num_input_tokens_seen": 14435952,
      "step": 2389,
      "train_runtime": 2422.8789,
      "train_tokens_per_second": 5958.181
    },
    {
      "epoch": 0.879080459770115,
      "grad_norm": 0.768943727016449,
      "learning_rate": 0.00011224001472482975,
      "loss": 1.6817,
      "num_input_tokens_seen": 14442832,
      "step": 2390,
      "train_runtime": 2424.0186,
      "train_tokens_per_second": 5958.218
    },
    {
      "epoch": 0.879448275862069,
      "grad_norm": 0.7778789401054382,
      "learning_rate": 0.00011220320265046936,
      "loss": 1.6514,
      "num_input_tokens_seen": 14448576,
      "step": 2391,
      "train_runtime": 2424.9572,
      "train_tokens_per_second": 5958.281
    },
    {
      "epoch": 0.879816091954023,
      "grad_norm": 0.7198776602745056,
      "learning_rate": 0.00011216639057610895,
      "loss": 1.731,
      "num_input_tokens_seen": 14455328,
      "step": 2392,
      "train_runtime": 2426.0519,
      "train_tokens_per_second": 5958.376
    },
    {
      "epoch": 0.880183908045977,
      "grad_norm": 0.7714976668357849,
      "learning_rate": 0.00011212957850174859,
      "loss": 1.6643,
      "num_input_tokens_seen": 14459680,
      "step": 2393,
      "train_runtime": 2426.8156,
      "train_tokens_per_second": 5958.294
    },
    {
      "epoch": 0.880551724137931,
      "grad_norm": 0.7934778332710266,
      "learning_rate": 0.0001120927664273882,
      "loss": 1.6422,
      "num_input_tokens_seen": 14465808,
      "step": 2394,
      "train_runtime": 2427.8454,
      "train_tokens_per_second": 5958.29
    },
    {
      "epoch": 0.8809195402298851,
      "grad_norm": 0.8033345341682434,
      "learning_rate": 0.00011205595435302781,
      "loss": 1.8191,
      "num_input_tokens_seen": 14470528,
      "step": 2395,
      "train_runtime": 2428.6475,
      "train_tokens_per_second": 5958.266
    },
    {
      "epoch": 0.8812873563218391,
      "grad_norm": 0.7819011211395264,
      "learning_rate": 0.00011201914227866742,
      "loss": 1.7823,
      "num_input_tokens_seen": 14475744,
      "step": 2396,
      "train_runtime": 2429.5037,
      "train_tokens_per_second": 5958.313
    },
    {
      "epoch": 0.8816551724137931,
      "grad_norm": 0.785110354423523,
      "learning_rate": 0.00011198233020430701,
      "loss": 1.6939,
      "num_input_tokens_seen": 14480800,
      "step": 2397,
      "train_runtime": 2430.3699,
      "train_tokens_per_second": 5958.27
    },
    {
      "epoch": 0.8820229885057471,
      "grad_norm": 0.7414306998252869,
      "learning_rate": 0.00011194551812994662,
      "loss": 1.4956,
      "num_input_tokens_seen": 14486560,
      "step": 2398,
      "train_runtime": 2431.3157,
      "train_tokens_per_second": 5958.321
    },
    {
      "epoch": 0.8823908045977011,
      "grad_norm": 0.7555763721466064,
      "learning_rate": 0.00011190870605558623,
      "loss": 1.6408,
      "num_input_tokens_seen": 14492176,
      "step": 2399,
      "train_runtime": 2432.2672,
      "train_tokens_per_second": 5958.299
    },
    {
      "epoch": 0.8827586206896552,
      "grad_norm": 0.7565219402313232,
      "learning_rate": 0.00011187189398122584,
      "loss": 1.7606,
      "num_input_tokens_seen": 14497536,
      "step": 2400,
      "train_runtime": 2433.1439,
      "train_tokens_per_second": 5958.355
    },
    {
      "epoch": 0.8831264367816092,
      "grad_norm": 0.6823695302009583,
      "learning_rate": 0.00011183508190686545,
      "loss": 1.655,
      "num_input_tokens_seen": 14505824,
      "step": 2401,
      "train_runtime": 2434.8916,
      "train_tokens_per_second": 5957.483
    },
    {
      "epoch": 0.8834942528735632,
      "grad_norm": 0.7468881607055664,
      "learning_rate": 0.00011179826983250507,
      "loss": 1.6549,
      "num_input_tokens_seen": 14511232,
      "step": 2402,
      "train_runtime": 2435.7954,
      "train_tokens_per_second": 5957.492
    },
    {
      "epoch": 0.8838620689655172,
      "grad_norm": 0.7681131958961487,
      "learning_rate": 0.00011176145775814468,
      "loss": 1.9565,
      "num_input_tokens_seen": 14516672,
      "step": 2403,
      "train_runtime": 2436.7088,
      "train_tokens_per_second": 5957.491
    },
    {
      "epoch": 0.8842298850574712,
      "grad_norm": 0.8144717812538147,
      "learning_rate": 0.00011172464568378429,
      "loss": 1.832,
      "num_input_tokens_seen": 14521648,
      "step": 2404,
      "train_runtime": 2437.5255,
      "train_tokens_per_second": 5957.537
    },
    {
      "epoch": 0.8845977011494253,
      "grad_norm": 0.7495010495185852,
      "learning_rate": 0.0001116878336094239,
      "loss": 1.7243,
      "num_input_tokens_seen": 14527664,
      "step": 2405,
      "train_runtime": 2438.5025,
      "train_tokens_per_second": 5957.617
    },
    {
      "epoch": 0.8849655172413793,
      "grad_norm": 0.8013268709182739,
      "learning_rate": 0.0001116510215350635,
      "loss": 1.8182,
      "num_input_tokens_seen": 14532464,
      "step": 2406,
      "train_runtime": 2439.3202,
      "train_tokens_per_second": 5957.588
    },
    {
      "epoch": 0.8853333333333333,
      "grad_norm": 0.7739303708076477,
      "learning_rate": 0.00011161420946070311,
      "loss": 1.7566,
      "num_input_tokens_seen": 14537136,
      "step": 2407,
      "train_runtime": 2440.1492,
      "train_tokens_per_second": 5957.478
    },
    {
      "epoch": 0.8857011494252873,
      "grad_norm": 0.7692505121231079,
      "learning_rate": 0.00011157739738634272,
      "loss": 1.7699,
      "num_input_tokens_seen": 14541520,
      "step": 2408,
      "train_runtime": 2440.9293,
      "train_tokens_per_second": 5957.37
    },
    {
      "epoch": 0.8860689655172413,
      "grad_norm": 0.7460946440696716,
      "learning_rate": 0.00011154058531198233,
      "loss": 1.7378,
      "num_input_tokens_seen": 14547936,
      "step": 2409,
      "train_runtime": 2441.9672,
      "train_tokens_per_second": 5957.466
    },
    {
      "epoch": 0.8864367816091954,
      "grad_norm": 0.7677419185638428,
      "learning_rate": 0.00011150377323762195,
      "loss": 1.7723,
      "num_input_tokens_seen": 14552560,
      "step": 2410,
      "train_runtime": 2442.7658,
      "train_tokens_per_second": 5957.411
    },
    {
      "epoch": 0.8868045977011494,
      "grad_norm": 0.7308533191680908,
      "learning_rate": 0.00011146696116326156,
      "loss": 1.7541,
      "num_input_tokens_seen": 14557648,
      "step": 2411,
      "train_runtime": 2443.6304,
      "train_tokens_per_second": 5957.385
    },
    {
      "epoch": 0.8871724137931034,
      "grad_norm": 0.7472432851791382,
      "learning_rate": 0.00011143014908890117,
      "loss": 1.6611,
      "num_input_tokens_seen": 14563088,
      "step": 2412,
      "train_runtime": 2444.5443,
      "train_tokens_per_second": 5957.383
    },
    {
      "epoch": 0.8875402298850574,
      "grad_norm": 0.688766598701477,
      "learning_rate": 0.00011139333701454078,
      "loss": 1.4676,
      "num_input_tokens_seen": 14570096,
      "step": 2413,
      "train_runtime": 2445.6706,
      "train_tokens_per_second": 5957.505
    },
    {
      "epoch": 0.8879080459770115,
      "grad_norm": 0.7346517443656921,
      "learning_rate": 0.00011135652494018039,
      "loss": 1.4383,
      "num_input_tokens_seen": 14577840,
      "step": 2414,
      "train_runtime": 2446.9046,
      "train_tokens_per_second": 5957.666
    },
    {
      "epoch": 0.8882758620689655,
      "grad_norm": 0.7357007265090942,
      "learning_rate": 0.00011131971286581999,
      "loss": 1.7221,
      "num_input_tokens_seen": 14586896,
      "step": 2415,
      "train_runtime": 2448.3271,
      "train_tokens_per_second": 5957.903
    },
    {
      "epoch": 0.8886436781609195,
      "grad_norm": 0.702000617980957,
      "learning_rate": 0.0001112829007914596,
      "loss": 1.5098,
      "num_input_tokens_seen": 14591728,
      "step": 2416,
      "train_runtime": 2449.156,
      "train_tokens_per_second": 5957.86
    },
    {
      "epoch": 0.8890114942528735,
      "grad_norm": 0.8758004307746887,
      "learning_rate": 0.0001112460887170992,
      "loss": 1.7841,
      "num_input_tokens_seen": 14596640,
      "step": 2417,
      "train_runtime": 2449.9939,
      "train_tokens_per_second": 5957.827
    },
    {
      "epoch": 0.8893793103448275,
      "grad_norm": 0.7414559125900269,
      "learning_rate": 0.00011120927664273881,
      "loss": 1.6884,
      "num_input_tokens_seen": 14604672,
      "step": 2418,
      "train_runtime": 2451.2371,
      "train_tokens_per_second": 5958.082
    },
    {
      "epoch": 0.8897471264367816,
      "grad_norm": 0.7551974058151245,
      "learning_rate": 0.00011117246456837845,
      "loss": 1.7889,
      "num_input_tokens_seen": 14612896,
      "step": 2419,
      "train_runtime": 2452.5476,
      "train_tokens_per_second": 5958.252
    },
    {
      "epoch": 0.8901149425287357,
      "grad_norm": 0.6946575045585632,
      "learning_rate": 0.00011113565249401804,
      "loss": 1.7502,
      "num_input_tokens_seen": 14618800,
      "step": 2420,
      "train_runtime": 2453.5058,
      "train_tokens_per_second": 5958.331
    },
    {
      "epoch": 0.8904827586206897,
      "grad_norm": 0.7958633303642273,
      "learning_rate": 0.00011109884041965765,
      "loss": 1.5836,
      "num_input_tokens_seen": 14624800,
      "step": 2421,
      "train_runtime": 2454.4833,
      "train_tokens_per_second": 5958.403
    },
    {
      "epoch": 0.8908505747126437,
      "grad_norm": 0.8049613833427429,
      "learning_rate": 0.00011106202834529726,
      "loss": 1.5905,
      "num_input_tokens_seen": 14630272,
      "step": 2422,
      "train_runtime": 2455.3924,
      "train_tokens_per_second": 5958.425
    },
    {
      "epoch": 0.8912183908045977,
      "grad_norm": 0.6941192746162415,
      "learning_rate": 0.00011102521627093687,
      "loss": 1.6647,
      "num_input_tokens_seen": 14637008,
      "step": 2423,
      "train_runtime": 2456.4795,
      "train_tokens_per_second": 5958.53
    },
    {
      "epoch": 0.8915862068965518,
      "grad_norm": 0.823941171169281,
      "learning_rate": 0.00011098840419657648,
      "loss": 1.7061,
      "num_input_tokens_seen": 14641776,
      "step": 2424,
      "train_runtime": 2457.301,
      "train_tokens_per_second": 5958.479
    },
    {
      "epoch": 0.8919540229885058,
      "grad_norm": 0.7627526521682739,
      "learning_rate": 0.00011095159212221609,
      "loss": 1.7886,
      "num_input_tokens_seen": 14648064,
      "step": 2425,
      "train_runtime": 2458.3115,
      "train_tokens_per_second": 5958.587
    },
    {
      "epoch": 0.8923218390804598,
      "grad_norm": 0.7629676461219788,
      "learning_rate": 0.0001109147800478557,
      "loss": 1.6886,
      "num_input_tokens_seen": 14654272,
      "step": 2426,
      "train_runtime": 2459.3132,
      "train_tokens_per_second": 5958.685
    },
    {
      "epoch": 0.8926896551724138,
      "grad_norm": 0.7743751406669617,
      "learning_rate": 0.00011087796797349532,
      "loss": 1.8038,
      "num_input_tokens_seen": 14659632,
      "step": 2427,
      "train_runtime": 2460.2118,
      "train_tokens_per_second": 5958.687
    },
    {
      "epoch": 0.8930574712643679,
      "grad_norm": 0.7945587038993835,
      "learning_rate": 0.00011084115589913493,
      "loss": 1.831,
      "num_input_tokens_seen": 14664736,
      "step": 2428,
      "train_runtime": 2461.0645,
      "train_tokens_per_second": 5958.696
    },
    {
      "epoch": 0.8934252873563219,
      "grad_norm": 0.678333580493927,
      "learning_rate": 0.00011080434382477454,
      "loss": 1.6323,
      "num_input_tokens_seen": 14671424,
      "step": 2429,
      "train_runtime": 2462.1298,
      "train_tokens_per_second": 5958.834
    },
    {
      "epoch": 0.8937931034482759,
      "grad_norm": 0.7958647608757019,
      "learning_rate": 0.00011076753175041415,
      "loss": 1.876,
      "num_input_tokens_seen": 14676416,
      "step": 2430,
      "train_runtime": 2462.9747,
      "train_tokens_per_second": 5958.817
    },
    {
      "epoch": 0.8941609195402299,
      "grad_norm": 0.7031279802322388,
      "learning_rate": 0.00011073071967605376,
      "loss": 1.6035,
      "num_input_tokens_seen": 14681200,
      "step": 2431,
      "train_runtime": 2464.2554,
      "train_tokens_per_second": 5957.662
    },
    {
      "epoch": 0.8945287356321839,
      "grad_norm": 0.7445070147514343,
      "learning_rate": 0.00011069390760169336,
      "loss": 1.63,
      "num_input_tokens_seen": 14685520,
      "step": 2432,
      "train_runtime": 2465.0009,
      "train_tokens_per_second": 5957.612
    },
    {
      "epoch": 0.894896551724138,
      "grad_norm": 0.7040464878082275,
      "learning_rate": 0.00011065709552733296,
      "loss": 1.747,
      "num_input_tokens_seen": 14693952,
      "step": 2433,
      "train_runtime": 2466.2999,
      "train_tokens_per_second": 5957.893
    },
    {
      "epoch": 0.895264367816092,
      "grad_norm": 0.7695351839065552,
      "learning_rate": 0.00011062028345297257,
      "loss": 1.7753,
      "num_input_tokens_seen": 14699936,
      "step": 2434,
      "train_runtime": 2467.2791,
      "train_tokens_per_second": 5957.954
    },
    {
      "epoch": 0.895632183908046,
      "grad_norm": 0.7717336416244507,
      "learning_rate": 0.00011058347137861218,
      "loss": 1.709,
      "num_input_tokens_seen": 14704640,
      "step": 2435,
      "train_runtime": 2468.0725,
      "train_tokens_per_second": 5957.945
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.7537624835968018,
      "learning_rate": 0.00011054665930425181,
      "loss": 1.9182,
      "num_input_tokens_seen": 14710256,
      "step": 2436,
      "train_runtime": 2469.0109,
      "train_tokens_per_second": 5957.955
    },
    {
      "epoch": 0.896367816091954,
      "grad_norm": 0.7601698637008667,
      "learning_rate": 0.00011050984722989142,
      "loss": 1.6092,
      "num_input_tokens_seen": 14714800,
      "step": 2437,
      "train_runtime": 2469.8044,
      "train_tokens_per_second": 5957.881
    },
    {
      "epoch": 0.8967356321839081,
      "grad_norm": 0.8169834017753601,
      "learning_rate": 0.00011047303515553102,
      "loss": 1.6737,
      "num_input_tokens_seen": 14722096,
      "step": 2438,
      "train_runtime": 2470.9489,
      "train_tokens_per_second": 5958.074
    },
    {
      "epoch": 0.8971034482758621,
      "grad_norm": 0.7955383658409119,
      "learning_rate": 0.00011043622308117063,
      "loss": 1.8738,
      "num_input_tokens_seen": 14726544,
      "step": 2439,
      "train_runtime": 2471.7285,
      "train_tokens_per_second": 5957.994
    },
    {
      "epoch": 0.8974712643678161,
      "grad_norm": 0.8970584273338318,
      "learning_rate": 0.00011039941100681024,
      "loss": 1.8116,
      "num_input_tokens_seen": 14731648,
      "step": 2440,
      "train_runtime": 2472.5688,
      "train_tokens_per_second": 5958.034
    },
    {
      "epoch": 0.8978390804597701,
      "grad_norm": 0.7360374927520752,
      "learning_rate": 0.00011036259893244984,
      "loss": 1.6655,
      "num_input_tokens_seen": 14737824,
      "step": 2441,
      "train_runtime": 2473.5693,
      "train_tokens_per_second": 5958.12
    },
    {
      "epoch": 0.8982068965517241,
      "grad_norm": 0.7386136651039124,
      "learning_rate": 0.00011032578685808945,
      "loss": 1.6351,
      "num_input_tokens_seen": 14742608,
      "step": 2442,
      "train_runtime": 2474.3955,
      "train_tokens_per_second": 5958.064
    },
    {
      "epoch": 0.8985747126436782,
      "grad_norm": 0.8073346614837646,
      "learning_rate": 0.00011028897478372906,
      "loss": 1.7017,
      "num_input_tokens_seen": 14747232,
      "step": 2443,
      "train_runtime": 2475.2134,
      "train_tokens_per_second": 5957.964
    },
    {
      "epoch": 0.8989425287356322,
      "grad_norm": 0.8261056542396545,
      "learning_rate": 0.00011025216270936868,
      "loss": 1.6188,
      "num_input_tokens_seen": 14752192,
      "step": 2444,
      "train_runtime": 2476.0653,
      "train_tokens_per_second": 5957.917
    },
    {
      "epoch": 0.8993103448275862,
      "grad_norm": 0.7028419971466064,
      "learning_rate": 0.0001102153506350083,
      "loss": 1.7697,
      "num_input_tokens_seen": 14763296,
      "step": 2445,
      "train_runtime": 2477.7281,
      "train_tokens_per_second": 5958.4
    },
    {
      "epoch": 0.8996781609195402,
      "grad_norm": 0.744083821773529,
      "learning_rate": 0.0001101785385606479,
      "loss": 1.7576,
      "num_input_tokens_seen": 14768912,
      "step": 2446,
      "train_runtime": 2478.6675,
      "train_tokens_per_second": 5958.408
    },
    {
      "epoch": 0.9000459770114942,
      "grad_norm": 0.7346624135971069,
      "learning_rate": 0.00011014172648628751,
      "loss": 1.5865,
      "num_input_tokens_seen": 14774976,
      "step": 2447,
      "train_runtime": 2479.6505,
      "train_tokens_per_second": 5958.491
    },
    {
      "epoch": 0.9004137931034483,
      "grad_norm": 0.7858444452285767,
      "learning_rate": 0.00011010491441192712,
      "loss": 1.5247,
      "num_input_tokens_seen": 14779616,
      "step": 2448,
      "train_runtime": 2480.4486,
      "train_tokens_per_second": 5958.445
    },
    {
      "epoch": 0.9007816091954023,
      "grad_norm": 0.7909073829650879,
      "learning_rate": 0.00011006810233756673,
      "loss": 1.7006,
      "num_input_tokens_seen": 14787680,
      "step": 2449,
      "train_runtime": 2481.7272,
      "train_tokens_per_second": 5958.624
    },
    {
      "epoch": 0.9011494252873563,
      "grad_norm": 0.7249764204025269,
      "learning_rate": 0.00011003129026320634,
      "loss": 1.6205,
      "num_input_tokens_seen": 14793808,
      "step": 2450,
      "train_runtime": 2482.7139,
      "train_tokens_per_second": 5958.724
    },
    {
      "epoch": 0.9015172413793103,
      "grad_norm": 0.7955921292304993,
      "learning_rate": 0.00010999447818884593,
      "loss": 1.6391,
      "num_input_tokens_seen": 14798368,
      "step": 2451,
      "train_runtime": 2483.5259,
      "train_tokens_per_second": 5958.612
    },
    {
      "epoch": 0.9018850574712644,
      "grad_norm": 0.7814099788665771,
      "learning_rate": 0.00010995766611448554,
      "loss": 1.7179,
      "num_input_tokens_seen": 14802720,
      "step": 2452,
      "train_runtime": 2484.2533,
      "train_tokens_per_second": 5958.619
    },
    {
      "epoch": 0.9022528735632184,
      "grad_norm": 0.7550371885299683,
      "learning_rate": 0.00010992085404012518,
      "loss": 1.6625,
      "num_input_tokens_seen": 14807104,
      "step": 2453,
      "train_runtime": 2485.0039,
      "train_tokens_per_second": 5958.584
    },
    {
      "epoch": 0.9026206896551724,
      "grad_norm": 0.8185312747955322,
      "learning_rate": 0.00010988404196576479,
      "loss": 1.8029,
      "num_input_tokens_seen": 14812960,
      "step": 2454,
      "train_runtime": 2485.9603,
      "train_tokens_per_second": 5958.647
    },
    {
      "epoch": 0.9029885057471264,
      "grad_norm": 0.750242292881012,
      "learning_rate": 0.0001098472298914044,
      "loss": 1.6578,
      "num_input_tokens_seen": 14819360,
      "step": 2455,
      "train_runtime": 2487.0032,
      "train_tokens_per_second": 5958.722
    },
    {
      "epoch": 0.9033563218390804,
      "grad_norm": 0.7446937561035156,
      "learning_rate": 0.000109810417817044,
      "loss": 1.6746,
      "num_input_tokens_seen": 14823744,
      "step": 2456,
      "train_runtime": 2487.7559,
      "train_tokens_per_second": 5958.681
    },
    {
      "epoch": 0.9037241379310345,
      "grad_norm": 0.7383763194084167,
      "learning_rate": 0.0001097736057426836,
      "loss": 1.8532,
      "num_input_tokens_seen": 14829248,
      "step": 2457,
      "train_runtime": 2488.6704,
      "train_tokens_per_second": 5958.703
    },
    {
      "epoch": 0.9040919540229885,
      "grad_norm": 0.7755974531173706,
      "learning_rate": 0.00010973679366832321,
      "loss": 1.6205,
      "num_input_tokens_seen": 14834784,
      "step": 2458,
      "train_runtime": 2489.6018,
      "train_tokens_per_second": 5958.698
    },
    {
      "epoch": 0.9044597701149425,
      "grad_norm": 0.7716488838195801,
      "learning_rate": 0.00010969998159396282,
      "loss": 1.7038,
      "num_input_tokens_seen": 14842784,
      "step": 2459,
      "train_runtime": 2490.8543,
      "train_tokens_per_second": 5958.913
    },
    {
      "epoch": 0.9048275862068965,
      "grad_norm": 0.8037108778953552,
      "learning_rate": 0.00010966316951960243,
      "loss": 1.7376,
      "num_input_tokens_seen": 14849024,
      "step": 2460,
      "train_runtime": 2491.8881,
      "train_tokens_per_second": 5958.945
    },
    {
      "epoch": 0.9051954022988505,
      "grad_norm": 0.8002779483795166,
      "learning_rate": 0.00010962635744524204,
      "loss": 1.8092,
      "num_input_tokens_seen": 14853632,
      "step": 2461,
      "train_runtime": 2493.2241,
      "train_tokens_per_second": 5957.6
    },
    {
      "epoch": 0.9055632183908046,
      "grad_norm": 0.7876670956611633,
      "learning_rate": 0.00010958954537088166,
      "loss": 1.6983,
      "num_input_tokens_seen": 14858096,
      "step": 2462,
      "train_runtime": 2494.0086,
      "train_tokens_per_second": 5957.516
    },
    {
      "epoch": 0.9059310344827586,
      "grad_norm": 0.757910430431366,
      "learning_rate": 0.00010955273329652127,
      "loss": 1.8157,
      "num_input_tokens_seen": 14864064,
      "step": 2463,
      "train_runtime": 2494.9675,
      "train_tokens_per_second": 5957.618
    },
    {
      "epoch": 0.9062988505747126,
      "grad_norm": 0.7434114217758179,
      "learning_rate": 0.00010951592122216088,
      "loss": 1.7639,
      "num_input_tokens_seen": 14869760,
      "step": 2464,
      "train_runtime": 2495.8865,
      "train_tokens_per_second": 5957.707
    },
    {
      "epoch": 0.9066666666666666,
      "grad_norm": 0.8509668111801147,
      "learning_rate": 0.00010947910914780048,
      "loss": 1.5399,
      "num_input_tokens_seen": 14875136,
      "step": 2465,
      "train_runtime": 2496.7698,
      "train_tokens_per_second": 5957.752
    },
    {
      "epoch": 0.9070344827586206,
      "grad_norm": 0.7537459135055542,
      "learning_rate": 0.0001094422970734401,
      "loss": 1.6189,
      "num_input_tokens_seen": 14881568,
      "step": 2466,
      "train_runtime": 2497.8164,
      "train_tokens_per_second": 5957.831
    },
    {
      "epoch": 0.9074022988505747,
      "grad_norm": 0.7488567233085632,
      "learning_rate": 0.0001094054849990797,
      "loss": 1.6997,
      "num_input_tokens_seen": 14886736,
      "step": 2467,
      "train_runtime": 2498.6681,
      "train_tokens_per_second": 5957.869
    },
    {
      "epoch": 0.9077701149425287,
      "grad_norm": 0.7168360352516174,
      "learning_rate": 0.00010936867292471931,
      "loss": 1.5637,
      "num_input_tokens_seen": 14895168,
      "step": 2468,
      "train_runtime": 2499.9755,
      "train_tokens_per_second": 5958.126
    },
    {
      "epoch": 0.9081379310344827,
      "grad_norm": 0.8835721611976624,
      "learning_rate": 0.00010933186085035892,
      "loss": 1.5412,
      "num_input_tokens_seen": 14901216,
      "step": 2469,
      "train_runtime": 2500.9301,
      "train_tokens_per_second": 5958.27
    },
    {
      "epoch": 0.9085057471264368,
      "grad_norm": 0.8334879875183105,
      "learning_rate": 0.00010929504877599854,
      "loss": 1.7239,
      "num_input_tokens_seen": 14907168,
      "step": 2470,
      "train_runtime": 2501.8822,
      "train_tokens_per_second": 5958.381
    },
    {
      "epoch": 0.9088735632183909,
      "grad_norm": 0.7692475914955139,
      "learning_rate": 0.00010925823670163815,
      "loss": 1.6819,
      "num_input_tokens_seen": 14913248,
      "step": 2471,
      "train_runtime": 2502.8624,
      "train_tokens_per_second": 5958.477
    },
    {
      "epoch": 0.9092413793103449,
      "grad_norm": 0.7324846386909485,
      "learning_rate": 0.00010922142462727776,
      "loss": 1.5419,
      "num_input_tokens_seen": 14922784,
      "step": 2472,
      "train_runtime": 2504.3076,
      "train_tokens_per_second": 5958.846
    },
    {
      "epoch": 0.9096091954022989,
      "grad_norm": 0.7073541283607483,
      "learning_rate": 0.00010918461255291737,
      "loss": 1.6384,
      "num_input_tokens_seen": 14927600,
      "step": 2473,
      "train_runtime": 2505.142,
      "train_tokens_per_second": 5958.784
    },
    {
      "epoch": 0.9099770114942529,
      "grad_norm": 0.7420263290405273,
      "learning_rate": 0.00010914780047855698,
      "loss": 1.5424,
      "num_input_tokens_seen": 14934224,
      "step": 2474,
      "train_runtime": 2506.2225,
      "train_tokens_per_second": 5958.858
    },
    {
      "epoch": 0.9103448275862069,
      "grad_norm": 0.8375838398933411,
      "learning_rate": 0.00010911098840419657,
      "loss": 1.7037,
      "num_input_tokens_seen": 14944112,
      "step": 2475,
      "train_runtime": 2507.7242,
      "train_tokens_per_second": 5959.233
    },
    {
      "epoch": 0.910712643678161,
      "grad_norm": 0.8070331811904907,
      "learning_rate": 0.00010907417632983618,
      "loss": 1.61,
      "num_input_tokens_seen": 14948992,
      "step": 2476,
      "train_runtime": 2508.5655,
      "train_tokens_per_second": 5959.179
    },
    {
      "epoch": 0.911080459770115,
      "grad_norm": 0.8206854462623596,
      "learning_rate": 0.00010903736425547579,
      "loss": 1.8077,
      "num_input_tokens_seen": 14955056,
      "step": 2477,
      "train_runtime": 2509.5547,
      "train_tokens_per_second": 5959.247
    },
    {
      "epoch": 0.911448275862069,
      "grad_norm": 0.7671430706977844,
      "learning_rate": 0.0001090005521811154,
      "loss": 1.7379,
      "num_input_tokens_seen": 14960496,
      "step": 2478,
      "train_runtime": 2510.4661,
      "train_tokens_per_second": 5959.25
    },
    {
      "epoch": 0.911816091954023,
      "grad_norm": 0.7520632743835449,
      "learning_rate": 0.00010896374010675504,
      "loss": 1.8196,
      "num_input_tokens_seen": 14965536,
      "step": 2479,
      "train_runtime": 2511.3209,
      "train_tokens_per_second": 5959.229
    },
    {
      "epoch": 0.912183908045977,
      "grad_norm": 0.7265950441360474,
      "learning_rate": 0.00010892692803239463,
      "loss": 1.8204,
      "num_input_tokens_seen": 14971360,
      "step": 2480,
      "train_runtime": 2512.3068,
      "train_tokens_per_second": 5959.209
    },
    {
      "epoch": 0.9125517241379311,
      "grad_norm": 0.7691535949707031,
      "learning_rate": 0.00010889011595803424,
      "loss": 1.7513,
      "num_input_tokens_seen": 14976912,
      "step": 2481,
      "train_runtime": 2513.2256,
      "train_tokens_per_second": 5959.239
    },
    {
      "epoch": 0.9129195402298851,
      "grad_norm": 0.7225045561790466,
      "learning_rate": 0.00010885330388367385,
      "loss": 1.6484,
      "num_input_tokens_seen": 14981504,
      "step": 2482,
      "train_runtime": 2514.0241,
      "train_tokens_per_second": 5959.173
    },
    {
      "epoch": 0.9132873563218391,
      "grad_norm": 0.7425840497016907,
      "learning_rate": 0.00010881649180931346,
      "loss": 1.6489,
      "num_input_tokens_seen": 14988048,
      "step": 2483,
      "train_runtime": 2515.0688,
      "train_tokens_per_second": 5959.299
    },
    {
      "epoch": 0.9136551724137931,
      "grad_norm": 0.8102589845657349,
      "learning_rate": 0.00010877967973495307,
      "loss": 1.7261,
      "num_input_tokens_seen": 14994592,
      "step": 2484,
      "train_runtime": 2516.1357,
      "train_tokens_per_second": 5959.373
    },
    {
      "epoch": 0.9140229885057471,
      "grad_norm": 0.7214230895042419,
      "learning_rate": 0.00010874286766059268,
      "loss": 1.8313,
      "num_input_tokens_seen": 15000832,
      "step": 2485,
      "train_runtime": 2517.1628,
      "train_tokens_per_second": 5959.421
    },
    {
      "epoch": 0.9143908045977012,
      "grad_norm": 0.7304913401603699,
      "learning_rate": 0.00010870605558623228,
      "loss": 1.556,
      "num_input_tokens_seen": 15005520,
      "step": 2486,
      "train_runtime": 2517.9384,
      "train_tokens_per_second": 5959.447
    },
    {
      "epoch": 0.9147586206896552,
      "grad_norm": 0.6967540979385376,
      "learning_rate": 0.00010866924351187191,
      "loss": 1.6727,
      "num_input_tokens_seen": 15010464,
      "step": 2487,
      "train_runtime": 2518.7942,
      "train_tokens_per_second": 5959.385
    },
    {
      "epoch": 0.9151264367816092,
      "grad_norm": 0.7573222517967224,
      "learning_rate": 0.00010863243143751152,
      "loss": 1.5095,
      "num_input_tokens_seen": 15018912,
      "step": 2488,
      "train_runtime": 2520.0803,
      "train_tokens_per_second": 5959.696
    },
    {
      "epoch": 0.9154942528735632,
      "grad_norm": 0.7670751810073853,
      "learning_rate": 0.00010859561936315113,
      "loss": 1.7261,
      "num_input_tokens_seen": 15024560,
      "step": 2489,
      "train_runtime": 2521.0374,
      "train_tokens_per_second": 5959.674
    },
    {
      "epoch": 0.9158620689655173,
      "grad_norm": 0.84031081199646,
      "learning_rate": 0.00010855880728879073,
      "loss": 1.5813,
      "num_input_tokens_seen": 15029136,
      "step": 2490,
      "train_runtime": 2521.8081,
      "train_tokens_per_second": 5959.667
    },
    {
      "epoch": 0.9162298850574713,
      "grad_norm": 0.7741042971611023,
      "learning_rate": 0.00010852199521443034,
      "loss": 1.7828,
      "num_input_tokens_seen": 15033648,
      "step": 2491,
      "train_runtime": 2523.0778,
      "train_tokens_per_second": 5958.456
    },
    {
      "epoch": 0.9165977011494253,
      "grad_norm": 0.8092449307441711,
      "learning_rate": 0.00010848518314006995,
      "loss": 1.9043,
      "num_input_tokens_seen": 15039024,
      "step": 2492,
      "train_runtime": 2523.9892,
      "train_tokens_per_second": 5958.434
    },
    {
      "epoch": 0.9169655172413793,
      "grad_norm": 0.7662644982337952,
      "learning_rate": 0.00010844837106570955,
      "loss": 1.6079,
      "num_input_tokens_seen": 15044224,
      "step": 2493,
      "train_runtime": 2524.8604,
      "train_tokens_per_second": 5958.438
    },
    {
      "epoch": 0.9173333333333333,
      "grad_norm": 0.7252683043479919,
      "learning_rate": 0.00010841155899134916,
      "loss": 1.6034,
      "num_input_tokens_seen": 15050160,
      "step": 2494,
      "train_runtime": 2525.8624,
      "train_tokens_per_second": 5958.424
    },
    {
      "epoch": 0.9177011494252874,
      "grad_norm": 0.7052468657493591,
      "learning_rate": 0.00010837474691698876,
      "loss": 1.6667,
      "num_input_tokens_seen": 15056800,
      "step": 2495,
      "train_runtime": 2526.9288,
      "train_tokens_per_second": 5958.538
    },
    {
      "epoch": 0.9180689655172414,
      "grad_norm": 0.74473637342453,
      "learning_rate": 0.0001083379348426284,
      "loss": 1.5383,
      "num_input_tokens_seen": 15067888,
      "step": 2496,
      "train_runtime": 2528.5974,
      "train_tokens_per_second": 5958.991
    },
    {
      "epoch": 0.9184367816091954,
      "grad_norm": 0.7353703379631042,
      "learning_rate": 0.00010830112276826801,
      "loss": 1.7424,
      "num_input_tokens_seen": 15075024,
      "step": 2497,
      "train_runtime": 2529.7384,
      "train_tokens_per_second": 5959.124
    },
    {
      "epoch": 0.9188045977011494,
      "grad_norm": 0.783466637134552,
      "learning_rate": 0.0001082643106939076,
      "loss": 1.6509,
      "num_input_tokens_seen": 15081808,
      "step": 2498,
      "train_runtime": 2530.8307,
      "train_tokens_per_second": 5959.232
    },
    {
      "epoch": 0.9191724137931034,
      "grad_norm": 0.8483232855796814,
      "learning_rate": 0.00010822749861954721,
      "loss": 1.7336,
      "num_input_tokens_seen": 15086432,
      "step": 2499,
      "train_runtime": 2531.6031,
      "train_tokens_per_second": 5959.24
    },
    {
      "epoch": 0.9195402298850575,
      "grad_norm": 0.7204304337501526,
      "learning_rate": 0.00010819068654518682,
      "loss": 1.76,
      "num_input_tokens_seen": 15092640,
      "step": 2500,
      "train_runtime": 2532.601,
      "train_tokens_per_second": 5959.344
    },
    {
      "epoch": 0.9199080459770115,
      "grad_norm": 0.7520074248313904,
      "learning_rate": 0.00010815387447082643,
      "loss": 1.6571,
      "num_input_tokens_seen": 15097792,
      "step": 2501,
      "train_runtime": 2533.502,
      "train_tokens_per_second": 5959.258
    },
    {
      "epoch": 0.9202758620689655,
      "grad_norm": 0.7919902801513672,
      "learning_rate": 0.00010811706239646604,
      "loss": 1.7254,
      "num_input_tokens_seen": 15105648,
      "step": 2502,
      "train_runtime": 2534.7225,
      "train_tokens_per_second": 5959.488
    },
    {
      "epoch": 0.9206436781609195,
      "grad_norm": 0.7878010272979736,
      "learning_rate": 0.00010808025032210565,
      "loss": 1.7348,
      "num_input_tokens_seen": 15112176,
      "step": 2503,
      "train_runtime": 2535.7762,
      "train_tokens_per_second": 5959.586
    },
    {
      "epoch": 0.9210114942528735,
      "grad_norm": 0.7690539956092834,
      "learning_rate": 0.00010804343824774527,
      "loss": 1.6653,
      "num_input_tokens_seen": 15116336,
      "step": 2504,
      "train_runtime": 2536.5041,
      "train_tokens_per_second": 5959.516
    },
    {
      "epoch": 0.9213793103448276,
      "grad_norm": 0.8095653653144836,
      "learning_rate": 0.00010800662617338488,
      "loss": 1.8966,
      "num_input_tokens_seen": 15122656,
      "step": 2505,
      "train_runtime": 2537.5224,
      "train_tokens_per_second": 5959.615
    },
    {
      "epoch": 0.9217471264367816,
      "grad_norm": 0.8052874803543091,
      "learning_rate": 0.00010796981409902449,
      "loss": 1.9162,
      "num_input_tokens_seen": 15127616,
      "step": 2506,
      "train_runtime": 2538.3589,
      "train_tokens_per_second": 5959.605
    },
    {
      "epoch": 0.9221149425287356,
      "grad_norm": 0.7415193319320679,
      "learning_rate": 0.0001079330020246641,
      "loss": 1.705,
      "num_input_tokens_seen": 15134944,
      "step": 2507,
      "train_runtime": 2539.5313,
      "train_tokens_per_second": 5959.739
    },
    {
      "epoch": 0.9224827586206896,
      "grad_norm": 0.7247180938720703,
      "learning_rate": 0.00010789618995030371,
      "loss": 1.6762,
      "num_input_tokens_seen": 15140960,
      "step": 2508,
      "train_runtime": 2540.5219,
      "train_tokens_per_second": 5959.783
    },
    {
      "epoch": 0.9228505747126436,
      "grad_norm": 0.766548752784729,
      "learning_rate": 0.00010785937787594332,
      "loss": 1.8025,
      "num_input_tokens_seen": 15146048,
      "step": 2509,
      "train_runtime": 2541.3892,
      "train_tokens_per_second": 5959.751
    },
    {
      "epoch": 0.9232183908045977,
      "grad_norm": 0.739657461643219,
      "learning_rate": 0.00010782256580158293,
      "loss": 1.7365,
      "num_input_tokens_seen": 15153216,
      "step": 2510,
      "train_runtime": 2542.5081,
      "train_tokens_per_second": 5959.948
    },
    {
      "epoch": 0.9235862068965517,
      "grad_norm": 0.6670482754707336,
      "learning_rate": 0.00010778575372722252,
      "loss": 1.6166,
      "num_input_tokens_seen": 15160480,
      "step": 2511,
      "train_runtime": 2543.681,
      "train_tokens_per_second": 5960.055
    },
    {
      "epoch": 0.9239540229885057,
      "grad_norm": 0.6904467940330505,
      "learning_rate": 0.00010774894165286213,
      "loss": 1.6364,
      "num_input_tokens_seen": 15167984,
      "step": 2512,
      "train_runtime": 2544.8931,
      "train_tokens_per_second": 5960.165
    },
    {
      "epoch": 0.9243218390804597,
      "grad_norm": 0.8197052478790283,
      "learning_rate": 0.00010771212957850177,
      "loss": 1.7968,
      "num_input_tokens_seen": 15172528,
      "step": 2513,
      "train_runtime": 2545.6902,
      "train_tokens_per_second": 5960.084
    },
    {
      "epoch": 0.9246896551724137,
      "grad_norm": 0.6886665225028992,
      "learning_rate": 0.00010767531750414137,
      "loss": 1.6555,
      "num_input_tokens_seen": 15177216,
      "step": 2514,
      "train_runtime": 2546.5192,
      "train_tokens_per_second": 5959.985
    },
    {
      "epoch": 0.9250574712643678,
      "grad_norm": 0.7456634640693665,
      "learning_rate": 0.00010763850542978098,
      "loss": 1.6618,
      "num_input_tokens_seen": 15183760,
      "step": 2515,
      "train_runtime": 2547.6056,
      "train_tokens_per_second": 5960.012
    },
    {
      "epoch": 0.9254252873563218,
      "grad_norm": 0.7432196736335754,
      "learning_rate": 0.00010760169335542058,
      "loss": 1.712,
      "num_input_tokens_seen": 15191632,
      "step": 2516,
      "train_runtime": 2548.8417,
      "train_tokens_per_second": 5960.21
    },
    {
      "epoch": 0.9257931034482758,
      "grad_norm": 0.8311834335327148,
      "learning_rate": 0.00010756488128106019,
      "loss": 1.7863,
      "num_input_tokens_seen": 15196528,
      "step": 2517,
      "train_runtime": 2549.6778,
      "train_tokens_per_second": 5960.176
    },
    {
      "epoch": 0.9261609195402298,
      "grad_norm": 0.7555972337722778,
      "learning_rate": 0.0001075280692066998,
      "loss": 1.5738,
      "num_input_tokens_seen": 15203424,
      "step": 2518,
      "train_runtime": 2550.7967,
      "train_tokens_per_second": 5960.265
    },
    {
      "epoch": 0.9265287356321839,
      "grad_norm": 0.7590003609657288,
      "learning_rate": 0.0001074912571323394,
      "loss": 1.6935,
      "num_input_tokens_seen": 15208048,
      "step": 2519,
      "train_runtime": 2551.6065,
      "train_tokens_per_second": 5960.185
    },
    {
      "epoch": 0.926896551724138,
      "grad_norm": 0.8056831359863281,
      "learning_rate": 0.00010745444505797901,
      "loss": 1.6413,
      "num_input_tokens_seen": 15214736,
      "step": 2520,
      "train_runtime": 2552.6474,
      "train_tokens_per_second": 5960.375
    },
    {
      "epoch": 0.927264367816092,
      "grad_norm": 0.6924358606338501,
      "learning_rate": 0.00010741763298361862,
      "loss": 1.5707,
      "num_input_tokens_seen": 15219408,
      "step": 2521,
      "train_runtime": 2553.9364,
      "train_tokens_per_second": 5959.196
    },
    {
      "epoch": 0.927632183908046,
      "grad_norm": 0.8808372020721436,
      "learning_rate": 0.00010738082090925825,
      "loss": 2.0587,
      "num_input_tokens_seen": 15224624,
      "step": 2522,
      "train_runtime": 2554.8269,
      "train_tokens_per_second": 5959.161
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.6935523152351379,
      "learning_rate": 0.00010734400883489785,
      "loss": 1.6628,
      "num_input_tokens_seen": 15231632,
      "step": 2523,
      "train_runtime": 2555.9483,
      "train_tokens_per_second": 5959.288
    },
    {
      "epoch": 0.9283678160919541,
      "grad_norm": 0.7480493187904358,
      "learning_rate": 0.00010730719676053746,
      "loss": 1.8916,
      "num_input_tokens_seen": 15236144,
      "step": 2524,
      "train_runtime": 2556.7367,
      "train_tokens_per_second": 5959.215
    },
    {
      "epoch": 0.9287356321839081,
      "grad_norm": 0.7940954566001892,
      "learning_rate": 0.00010727038468617707,
      "loss": 1.7265,
      "num_input_tokens_seen": 15241920,
      "step": 2525,
      "train_runtime": 2557.6615,
      "train_tokens_per_second": 5959.319
    },
    {
      "epoch": 0.9291034482758621,
      "grad_norm": 0.75591641664505,
      "learning_rate": 0.00010723357261181668,
      "loss": 1.8021,
      "num_input_tokens_seen": 15248336,
      "step": 2526,
      "train_runtime": 2558.7215,
      "train_tokens_per_second": 5959.357
    },
    {
      "epoch": 0.9294712643678161,
      "grad_norm": 0.765473484992981,
      "learning_rate": 0.00010719676053745629,
      "loss": 1.7953,
      "num_input_tokens_seen": 15252672,
      "step": 2527,
      "train_runtime": 2559.4795,
      "train_tokens_per_second": 5959.287
    },
    {
      "epoch": 0.9298390804597702,
      "grad_norm": 0.7165312170982361,
      "learning_rate": 0.0001071599484630959,
      "loss": 1.6301,
      "num_input_tokens_seen": 15257920,
      "step": 2528,
      "train_runtime": 2560.3436,
      "train_tokens_per_second": 5959.325
    },
    {
      "epoch": 0.9302068965517242,
      "grad_norm": 0.7552527189254761,
      "learning_rate": 0.0001071231363887355,
      "loss": 1.706,
      "num_input_tokens_seen": 15263920,
      "step": 2529,
      "train_runtime": 2561.3258,
      "train_tokens_per_second": 5959.382
    },
    {
      "epoch": 0.9305747126436782,
      "grad_norm": 0.7574245929718018,
      "learning_rate": 0.00010708632431437513,
      "loss": 1.7196,
      "num_input_tokens_seen": 15269488,
      "step": 2530,
      "train_runtime": 2562.2662,
      "train_tokens_per_second": 5959.368
    },
    {
      "epoch": 0.9309425287356322,
      "grad_norm": 0.6877573132514954,
      "learning_rate": 0.00010704951224001474,
      "loss": 1.5738,
      "num_input_tokens_seen": 15280880,
      "step": 2531,
      "train_runtime": 2563.952,
      "train_tokens_per_second": 5959.893
    },
    {
      "epoch": 0.9313103448275862,
      "grad_norm": 0.7410098910331726,
      "learning_rate": 0.00010701270016565435,
      "loss": 1.7492,
      "num_input_tokens_seen": 15286544,
      "step": 2532,
      "train_runtime": 2564.8864,
      "train_tokens_per_second": 5959.93
    },
    {
      "epoch": 0.9316781609195403,
      "grad_norm": 0.7868102192878723,
      "learning_rate": 0.00010697588809129396,
      "loss": 1.5289,
      "num_input_tokens_seen": 15293232,
      "step": 2533,
      "train_runtime": 2565.9792,
      "train_tokens_per_second": 5959.999
    },
    {
      "epoch": 0.9320459770114943,
      "grad_norm": 0.7302942872047424,
      "learning_rate": 0.00010693907601693357,
      "loss": 1.7216,
      "num_input_tokens_seen": 15298320,
      "step": 2534,
      "train_runtime": 2566.838,
      "train_tokens_per_second": 5959.986
    },
    {
      "epoch": 0.9324137931034483,
      "grad_norm": 0.784263014793396,
      "learning_rate": 0.00010690226394257316,
      "loss": 1.631,
      "num_input_tokens_seen": 15304448,
      "step": 2535,
      "train_runtime": 2567.844,
      "train_tokens_per_second": 5960.038
    },
    {
      "epoch": 0.9327816091954023,
      "grad_norm": 0.649817705154419,
      "learning_rate": 0.00010686545186821277,
      "loss": 1.6417,
      "num_input_tokens_seen": 15313776,
      "step": 2536,
      "train_runtime": 2569.2711,
      "train_tokens_per_second": 5960.358
    },
    {
      "epoch": 0.9331494252873563,
      "grad_norm": 0.8143831491470337,
      "learning_rate": 0.00010682863979385238,
      "loss": 1.8846,
      "num_input_tokens_seen": 15318864,
      "step": 2537,
      "train_runtime": 2570.1268,
      "train_tokens_per_second": 5960.353
    },
    {
      "epoch": 0.9335172413793104,
      "grad_norm": 0.7969774007797241,
      "learning_rate": 0.00010679182771949199,
      "loss": 1.9373,
      "num_input_tokens_seen": 15323216,
      "step": 2538,
      "train_runtime": 2570.8845,
      "train_tokens_per_second": 5960.29
    },
    {
      "epoch": 0.9338850574712644,
      "grad_norm": 0.7657924294471741,
      "learning_rate": 0.00010675501564513162,
      "loss": 1.7736,
      "num_input_tokens_seen": 15329536,
      "step": 2539,
      "train_runtime": 2571.9303,
      "train_tokens_per_second": 5960.323
    },
    {
      "epoch": 0.9342528735632184,
      "grad_norm": 0.7626153230667114,
      "learning_rate": 0.00010671820357077122,
      "loss": 1.6901,
      "num_input_tokens_seen": 15333952,
      "step": 2540,
      "train_runtime": 2572.6821,
      "train_tokens_per_second": 5960.298
    },
    {
      "epoch": 0.9346206896551724,
      "grad_norm": 0.7269489765167236,
      "learning_rate": 0.00010668139149641083,
      "loss": 1.6146,
      "num_input_tokens_seen": 15339040,
      "step": 2541,
      "train_runtime": 2573.5291,
      "train_tokens_per_second": 5960.314
    },
    {
      "epoch": 0.9349885057471264,
      "grad_norm": 0.8552651405334473,
      "learning_rate": 0.00010664457942205044,
      "loss": 1.6695,
      "num_input_tokens_seen": 15344752,
      "step": 2542,
      "train_runtime": 2574.4773,
      "train_tokens_per_second": 5960.337
    },
    {
      "epoch": 0.9353563218390805,
      "grad_norm": 0.7995190024375916,
      "learning_rate": 0.00010660776734769005,
      "loss": 1.7721,
      "num_input_tokens_seen": 15350752,
      "step": 2543,
      "train_runtime": 2575.4699,
      "train_tokens_per_second": 5960.369
    },
    {
      "epoch": 0.9357241379310345,
      "grad_norm": 0.7732382416725159,
      "learning_rate": 0.00010657095527332965,
      "loss": 1.6626,
      "num_input_tokens_seen": 15355936,
      "step": 2544,
      "train_runtime": 2576.3339,
      "train_tokens_per_second": 5960.383
    },
    {
      "epoch": 0.9360919540229885,
      "grad_norm": 0.7307212352752686,
      "learning_rate": 0.00010653414319896926,
      "loss": 1.5954,
      "num_input_tokens_seen": 15361296,
      "step": 2545,
      "train_runtime": 2577.2339,
      "train_tokens_per_second": 5960.381
    },
    {
      "epoch": 0.9364597701149425,
      "grad_norm": 0.8574381470680237,
      "learning_rate": 0.00010649733112460887,
      "loss": 1.6593,
      "num_input_tokens_seen": 15367024,
      "step": 2546,
      "train_runtime": 2578.1668,
      "train_tokens_per_second": 5960.446
    },
    {
      "epoch": 0.9368275862068965,
      "grad_norm": 0.7505806088447571,
      "learning_rate": 0.0001064605190502485,
      "loss": 1.6593,
      "num_input_tokens_seen": 15375536,
      "step": 2547,
      "train_runtime": 2579.4721,
      "train_tokens_per_second": 5960.73
    },
    {
      "epoch": 0.9371954022988506,
      "grad_norm": 0.7568456530570984,
      "learning_rate": 0.0001064237069758881,
      "loss": 1.7068,
      "num_input_tokens_seen": 15381616,
      "step": 2548,
      "train_runtime": 2580.4523,
      "train_tokens_per_second": 5960.822
    },
    {
      "epoch": 0.9375632183908046,
      "grad_norm": 0.7760906219482422,
      "learning_rate": 0.00010638689490152771,
      "loss": 1.6475,
      "num_input_tokens_seen": 15388576,
      "step": 2549,
      "train_runtime": 2581.5703,
      "train_tokens_per_second": 5960.936
    },
    {
      "epoch": 0.9379310344827586,
      "grad_norm": 0.6869860291481018,
      "learning_rate": 0.00010635008282716732,
      "loss": 1.6367,
      "num_input_tokens_seen": 15396304,
      "step": 2550,
      "train_runtime": 2582.806,
      "train_tokens_per_second": 5961.076
    },
    {
      "epoch": 0.9382988505747126,
      "grad_norm": 0.748971700668335,
      "learning_rate": 0.00010631327075280693,
      "loss": 1.6767,
      "num_input_tokens_seen": 15401248,
      "step": 2551,
      "train_runtime": 2584.1357,
      "train_tokens_per_second": 5959.922
    },
    {
      "epoch": 0.9386666666666666,
      "grad_norm": 0.8482131958007812,
      "learning_rate": 0.00010627645867844654,
      "loss": 1.8888,
      "num_input_tokens_seen": 15407504,
      "step": 2552,
      "train_runtime": 2585.164,
      "train_tokens_per_second": 5959.972
    },
    {
      "epoch": 0.9390344827586207,
      "grad_norm": 0.7847908139228821,
      "learning_rate": 0.00010623964660408613,
      "loss": 1.7463,
      "num_input_tokens_seen": 15413056,
      "step": 2553,
      "train_runtime": 2586.0865,
      "train_tokens_per_second": 5959.992
    },
    {
      "epoch": 0.9394022988505747,
      "grad_norm": 0.7715175747871399,
      "learning_rate": 0.00010620283452972574,
      "loss": 1.745,
      "num_input_tokens_seen": 15418528,
      "step": 2554,
      "train_runtime": 2587.0045,
      "train_tokens_per_second": 5959.993
    },
    {
      "epoch": 0.9397701149425287,
      "grad_norm": 0.7222607135772705,
      "learning_rate": 0.00010616602245536535,
      "loss": 1.6166,
      "num_input_tokens_seen": 15427568,
      "step": 2555,
      "train_runtime": 2588.3731,
      "train_tokens_per_second": 5960.334
    },
    {
      "epoch": 0.9401379310344827,
      "grad_norm": 0.6869435906410217,
      "learning_rate": 0.00010612921038100499,
      "loss": 1.6545,
      "num_input_tokens_seen": 15434640,
      "step": 2556,
      "train_runtime": 2589.5271,
      "train_tokens_per_second": 5960.409
    },
    {
      "epoch": 0.9405057471264368,
      "grad_norm": 0.7444092035293579,
      "learning_rate": 0.0001060923983066446,
      "loss": 1.5973,
      "num_input_tokens_seen": 15440432,
      "step": 2557,
      "train_runtime": 2590.4785,
      "train_tokens_per_second": 5960.456
    },
    {
      "epoch": 0.9408735632183908,
      "grad_norm": 0.7215439677238464,
      "learning_rate": 0.00010605558623228419,
      "loss": 1.5527,
      "num_input_tokens_seen": 15448832,
      "step": 2558,
      "train_runtime": 2591.7811,
      "train_tokens_per_second": 5960.701
    },
    {
      "epoch": 0.9412413793103448,
      "grad_norm": 0.7285688519477844,
      "learning_rate": 0.0001060187741579238,
      "loss": 1.6329,
      "num_input_tokens_seen": 15457968,
      "step": 2559,
      "train_runtime": 2593.2118,
      "train_tokens_per_second": 5960.935
    },
    {
      "epoch": 0.9416091954022988,
      "grad_norm": 0.7814379930496216,
      "learning_rate": 0.00010598196208356341,
      "loss": 1.5189,
      "num_input_tokens_seen": 15465248,
      "step": 2560,
      "train_runtime": 2594.3863,
      "train_tokens_per_second": 5961.043
    },
    {
      "epoch": 0.9419770114942528,
      "grad_norm": 0.7479436993598938,
      "learning_rate": 0.00010594515000920302,
      "loss": 1.5687,
      "num_input_tokens_seen": 15472720,
      "step": 2561,
      "train_runtime": 2595.5634,
      "train_tokens_per_second": 5961.218
    },
    {
      "epoch": 0.9423448275862069,
      "grad_norm": 0.7916021943092346,
      "learning_rate": 0.00010590833793484263,
      "loss": 1.7425,
      "num_input_tokens_seen": 15478720,
      "step": 2562,
      "train_runtime": 2596.5569,
      "train_tokens_per_second": 5961.248
    },
    {
      "epoch": 0.9427126436781609,
      "grad_norm": 0.8089719414710999,
      "learning_rate": 0.00010587152586048224,
      "loss": 1.591,
      "num_input_tokens_seen": 15487024,
      "step": 2563,
      "train_runtime": 2597.8165,
      "train_tokens_per_second": 5961.554
    },
    {
      "epoch": 0.9430804597701149,
      "grad_norm": 0.7413442730903625,
      "learning_rate": 0.00010583471378612186,
      "loss": 1.6217,
      "num_input_tokens_seen": 15492496,
      "step": 2564,
      "train_runtime": 2598.7185,
      "train_tokens_per_second": 5961.591
    },
    {
      "epoch": 0.9434482758620689,
      "grad_norm": 0.7030351758003235,
      "learning_rate": 0.00010579790171176147,
      "loss": 1.7155,
      "num_input_tokens_seen": 15497616,
      "step": 2565,
      "train_runtime": 2599.582,
      "train_tokens_per_second": 5961.58
    },
    {
      "epoch": 0.9438160919540229,
      "grad_norm": 0.7779426574707031,
      "learning_rate": 0.00010576108963740108,
      "loss": 1.8197,
      "num_input_tokens_seen": 15502256,
      "step": 2566,
      "train_runtime": 2600.4029,
      "train_tokens_per_second": 5961.482
    },
    {
      "epoch": 0.944183908045977,
      "grad_norm": 0.7877058386802673,
      "learning_rate": 0.00010572427756304069,
      "loss": 1.5917,
      "num_input_tokens_seen": 15508352,
      "step": 2567,
      "train_runtime": 2601.3946,
      "train_tokens_per_second": 5961.553
    },
    {
      "epoch": 0.944551724137931,
      "grad_norm": 0.7855427861213684,
      "learning_rate": 0.0001056874654886803,
      "loss": 1.7309,
      "num_input_tokens_seen": 15512912,
      "step": 2568,
      "train_runtime": 2602.192,
      "train_tokens_per_second": 5961.479
    },
    {
      "epoch": 0.944919540229885,
      "grad_norm": 0.7029417753219604,
      "learning_rate": 0.0001056506534143199,
      "loss": 1.4136,
      "num_input_tokens_seen": 15519264,
      "step": 2569,
      "train_runtime": 2603.2367,
      "train_tokens_per_second": 5961.526
    },
    {
      "epoch": 0.9452873563218391,
      "grad_norm": 0.7505778074264526,
      "learning_rate": 0.00010561384133995951,
      "loss": 1.8622,
      "num_input_tokens_seen": 15524224,
      "step": 2570,
      "train_runtime": 2604.0883,
      "train_tokens_per_second": 5961.482
    },
    {
      "epoch": 0.9456551724137932,
      "grad_norm": 0.766185998916626,
      "learning_rate": 0.00010557702926559911,
      "loss": 1.7902,
      "num_input_tokens_seen": 15530688,
      "step": 2571,
      "train_runtime": 2605.1302,
      "train_tokens_per_second": 5961.578
    },
    {
      "epoch": 0.9460229885057472,
      "grad_norm": 0.7540948390960693,
      "learning_rate": 0.00010554021719123872,
      "loss": 1.9136,
      "num_input_tokens_seen": 15535600,
      "step": 2572,
      "train_runtime": 2605.9731,
      "train_tokens_per_second": 5961.535
    },
    {
      "epoch": 0.9463908045977012,
      "grad_norm": 0.780738890171051,
      "learning_rate": 0.00010550340511687835,
      "loss": 1.5797,
      "num_input_tokens_seen": 15542784,
      "step": 2573,
      "train_runtime": 2607.086,
      "train_tokens_per_second": 5961.746
    },
    {
      "epoch": 0.9467586206896552,
      "grad_norm": 0.7569245100021362,
      "learning_rate": 0.00010546659304251796,
      "loss": 1.7833,
      "num_input_tokens_seen": 15548384,
      "step": 2574,
      "train_runtime": 2607.9889,
      "train_tokens_per_second": 5961.829
    },
    {
      "epoch": 0.9471264367816092,
      "grad_norm": 0.8940502405166626,
      "learning_rate": 0.00010542978096815757,
      "loss": 1.8966,
      "num_input_tokens_seen": 15553600,
      "step": 2575,
      "train_runtime": 2608.8601,
      "train_tokens_per_second": 5961.837
    },
    {
      "epoch": 0.9474942528735633,
      "grad_norm": 0.6905773282051086,
      "learning_rate": 0.00010539296889379717,
      "loss": 1.5211,
      "num_input_tokens_seen": 15558704,
      "step": 2576,
      "train_runtime": 2609.7297,
      "train_tokens_per_second": 5961.807
    },
    {
      "epoch": 0.9478620689655173,
      "grad_norm": 0.7449854016304016,
      "learning_rate": 0.00010535615681943678,
      "loss": 1.6044,
      "num_input_tokens_seen": 15565472,
      "step": 2577,
      "train_runtime": 2610.818,
      "train_tokens_per_second": 5961.914
    },
    {
      "epoch": 0.9482298850574713,
      "grad_norm": 0.8156737089157104,
      "learning_rate": 0.00010531934474507638,
      "loss": 1.7407,
      "num_input_tokens_seen": 15572640,
      "step": 2578,
      "train_runtime": 2611.9509,
      "train_tokens_per_second": 5962.072
    },
    {
      "epoch": 0.9485977011494253,
      "grad_norm": 0.7681053876876831,
      "learning_rate": 0.000105282532670716,
      "loss": 1.7782,
      "num_input_tokens_seen": 15578272,
      "step": 2579,
      "train_runtime": 2612.8764,
      "train_tokens_per_second": 5962.116
    },
    {
      "epoch": 0.9489655172413793,
      "grad_norm": 0.6874492764472961,
      "learning_rate": 0.0001052457205963556,
      "loss": 1.7121,
      "num_input_tokens_seen": 15584480,
      "step": 2580,
      "train_runtime": 2613.9154,
      "train_tokens_per_second": 5962.121
    },
    {
      "epoch": 0.9493333333333334,
      "grad_norm": 0.7125325798988342,
      "learning_rate": 0.00010520890852199521,
      "loss": 1.6029,
      "num_input_tokens_seen": 15589136,
      "step": 2581,
      "train_runtime": 2615.1887,
      "train_tokens_per_second": 5960.998
    },
    {
      "epoch": 0.9497011494252874,
      "grad_norm": 0.7656606435775757,
      "learning_rate": 0.00010517209644763483,
      "loss": 1.7025,
      "num_input_tokens_seen": 15596608,
      "step": 2582,
      "train_runtime": 2616.3702,
      "train_tokens_per_second": 5961.162
    },
    {
      "epoch": 0.9500689655172414,
      "grad_norm": 0.7482950091362,
      "learning_rate": 0.00010513528437327444,
      "loss": 1.6919,
      "num_input_tokens_seen": 15602176,
      "step": 2583,
      "train_runtime": 2617.2695,
      "train_tokens_per_second": 5961.242
    },
    {
      "epoch": 0.9504367816091954,
      "grad_norm": 0.7446708679199219,
      "learning_rate": 0.00010509847229891405,
      "loss": 1.8587,
      "num_input_tokens_seen": 15611088,
      "step": 2584,
      "train_runtime": 2618.631,
      "train_tokens_per_second": 5961.546
    },
    {
      "epoch": 0.9508045977011494,
      "grad_norm": 0.7638434767723083,
      "learning_rate": 0.00010506166022455366,
      "loss": 1.6756,
      "num_input_tokens_seen": 15617600,
      "step": 2585,
      "train_runtime": 2619.6627,
      "train_tokens_per_second": 5961.684
    },
    {
      "epoch": 0.9511724137931035,
      "grad_norm": 0.7562476396560669,
      "learning_rate": 0.00010502484815019327,
      "loss": 1.7775,
      "num_input_tokens_seen": 15623888,
      "step": 2586,
      "train_runtime": 2620.67,
      "train_tokens_per_second": 5961.791
    },
    {
      "epoch": 0.9515402298850575,
      "grad_norm": 0.7628239393234253,
      "learning_rate": 0.00010498803607583288,
      "loss": 1.7905,
      "num_input_tokens_seen": 15629280,
      "step": 2587,
      "train_runtime": 2621.5858,
      "train_tokens_per_second": 5961.766
    },
    {
      "epoch": 0.9519080459770115,
      "grad_norm": 0.7623147964477539,
      "learning_rate": 0.00010495122400147249,
      "loss": 1.9185,
      "num_input_tokens_seen": 15635664,
      "step": 2588,
      "train_runtime": 2622.6236,
      "train_tokens_per_second": 5961.841
    },
    {
      "epoch": 0.9522758620689655,
      "grad_norm": 0.7677077651023865,
      "learning_rate": 0.00010491441192711208,
      "loss": 1.8309,
      "num_input_tokens_seen": 15641088,
      "step": 2589,
      "train_runtime": 2623.5212,
      "train_tokens_per_second": 5961.868
    },
    {
      "epoch": 0.9526436781609195,
      "grad_norm": 0.8027551174163818,
      "learning_rate": 0.00010487759985275172,
      "loss": 1.6273,
      "num_input_tokens_seen": 15645856,
      "step": 2590,
      "train_runtime": 2624.3256,
      "train_tokens_per_second": 5961.858
    },
    {
      "epoch": 0.9530114942528736,
      "grad_norm": 0.7034247517585754,
      "learning_rate": 0.00010484078777839133,
      "loss": 1.6412,
      "num_input_tokens_seen": 15650752,
      "step": 2591,
      "train_runtime": 2625.1697,
      "train_tokens_per_second": 5961.806
    },
    {
      "epoch": 0.9533793103448276,
      "grad_norm": 0.6913555860519409,
      "learning_rate": 0.00010480397570403094,
      "loss": 1.7685,
      "num_input_tokens_seen": 15657296,
      "step": 2592,
      "train_runtime": 2626.2348,
      "train_tokens_per_second": 5961.88
    },
    {
      "epoch": 0.9537471264367816,
      "grad_norm": 0.7643256187438965,
      "learning_rate": 0.00010476716362967054,
      "loss": 1.8081,
      "num_input_tokens_seen": 15662176,
      "step": 2593,
      "train_runtime": 2627.07,
      "train_tokens_per_second": 5961.842
    },
    {
      "epoch": 0.9541149425287356,
      "grad_norm": 0.6745730042457581,
      "learning_rate": 0.00010473035155531014,
      "loss": 1.794,
      "num_input_tokens_seen": 15667568,
      "step": 2594,
      "train_runtime": 2627.9875,
      "train_tokens_per_second": 5961.812
    },
    {
      "epoch": 0.9544827586206897,
      "grad_norm": 0.752946138381958,
      "learning_rate": 0.00010469353948094975,
      "loss": 1.7325,
      "num_input_tokens_seen": 15674240,
      "step": 2595,
      "train_runtime": 2629.0621,
      "train_tokens_per_second": 5961.913
    },
    {
      "epoch": 0.9548505747126437,
      "grad_norm": 0.728472888469696,
      "learning_rate": 0.00010465672740658936,
      "loss": 1.671,
      "num_input_tokens_seen": 15680432,
      "step": 2596,
      "train_runtime": 2630.0762,
      "train_tokens_per_second": 5961.969
    },
    {
      "epoch": 0.9552183908045977,
      "grad_norm": 0.7472367882728577,
      "learning_rate": 0.00010461991533222897,
      "loss": 1.7573,
      "num_input_tokens_seen": 15685760,
      "step": 2597,
      "train_runtime": 2631.0112,
      "train_tokens_per_second": 5961.875
    },
    {
      "epoch": 0.9555862068965517,
      "grad_norm": 0.761685311794281,
      "learning_rate": 0.00010458310325786858,
      "loss": 1.6944,
      "num_input_tokens_seen": 15689984,
      "step": 2598,
      "train_runtime": 2631.7473,
      "train_tokens_per_second": 5961.812
    },
    {
      "epoch": 0.9559540229885057,
      "grad_norm": 0.7775751352310181,
      "learning_rate": 0.0001045462911835082,
      "loss": 1.8871,
      "num_input_tokens_seen": 15695344,
      "step": 2599,
      "train_runtime": 2632.6248,
      "train_tokens_per_second": 5961.861
    },
    {
      "epoch": 0.9563218390804598,
      "grad_norm": 0.8221151828765869,
      "learning_rate": 0.00010450947910914781,
      "loss": 1.76,
      "num_input_tokens_seen": 15701760,
      "step": 2600,
      "train_runtime": 2633.6754,
      "train_tokens_per_second": 5961.919
    },
    {
      "epoch": 0.9566896551724138,
      "grad_norm": 0.789374589920044,
      "learning_rate": 0.00010447266703478742,
      "loss": 1.8581,
      "num_input_tokens_seen": 15707440,
      "step": 2601,
      "train_runtime": 2634.6263,
      "train_tokens_per_second": 5961.923
    },
    {
      "epoch": 0.9570574712643678,
      "grad_norm": 0.7307106256484985,
      "learning_rate": 0.00010443585496042702,
      "loss": 1.5808,
      "num_input_tokens_seen": 15714928,
      "step": 2602,
      "train_runtime": 2635.8214,
      "train_tokens_per_second": 5962.061
    },
    {
      "epoch": 0.9574252873563218,
      "grad_norm": 0.7828757166862488,
      "learning_rate": 0.00010439904288606663,
      "loss": 1.8317,
      "num_input_tokens_seen": 15719760,
      "step": 2603,
      "train_runtime": 2636.6698,
      "train_tokens_per_second": 5961.975
    },
    {
      "epoch": 0.9577931034482758,
      "grad_norm": 0.7009854316711426,
      "learning_rate": 0.00010436223081170624,
      "loss": 1.642,
      "num_input_tokens_seen": 15731584,
      "step": 2604,
      "train_runtime": 2638.4292,
      "train_tokens_per_second": 5962.481
    },
    {
      "epoch": 0.9581609195402299,
      "grad_norm": 0.7014610171318054,
      "learning_rate": 0.00010432541873734585,
      "loss": 1.6235,
      "num_input_tokens_seen": 15738656,
      "step": 2605,
      "train_runtime": 2639.5492,
      "train_tokens_per_second": 5962.63
    },
    {
      "epoch": 0.9585287356321839,
      "grad_norm": 0.7427473664283752,
      "learning_rate": 0.00010428860666298546,
      "loss": 1.766,
      "num_input_tokens_seen": 15745216,
      "step": 2606,
      "train_runtime": 2640.6051,
      "train_tokens_per_second": 5962.73
    },
    {
      "epoch": 0.9588965517241379,
      "grad_norm": 0.702900767326355,
      "learning_rate": 0.00010425179458862508,
      "loss": 1.5759,
      "num_input_tokens_seen": 15751648,
      "step": 2607,
      "train_runtime": 2641.6432,
      "train_tokens_per_second": 5962.822
    },
    {
      "epoch": 0.9592643678160919,
      "grad_norm": 0.707489013671875,
      "learning_rate": 0.00010421498251426469,
      "loss": 1.5724,
      "num_input_tokens_seen": 15756784,
      "step": 2608,
      "train_runtime": 2642.4985,
      "train_tokens_per_second": 5962.835
    },
    {
      "epoch": 0.9596321839080459,
      "grad_norm": 0.7865966558456421,
      "learning_rate": 0.0001041781704399043,
      "loss": 1.6301,
      "num_input_tokens_seen": 15764096,
      "step": 2609,
      "train_runtime": 2643.6271,
      "train_tokens_per_second": 5963.056
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.6945360898971558,
      "learning_rate": 0.00010414135836554391,
      "loss": 1.5704,
      "num_input_tokens_seen": 15772304,
      "step": 2610,
      "train_runtime": 2644.8831,
      "train_tokens_per_second": 5963.328
    },
    {
      "epoch": 0.960367816091954,
      "grad_norm": 0.7937390804290771,
      "learning_rate": 0.00010410454629118352,
      "loss": 1.7341,
      "num_input_tokens_seen": 15776912,
      "step": 2611,
      "train_runtime": 2646.2467,
      "train_tokens_per_second": 5961.996
    },
    {
      "epoch": 0.960735632183908,
      "grad_norm": 0.786213755607605,
      "learning_rate": 0.00010406773421682313,
      "loss": 1.6778,
      "num_input_tokens_seen": 15785968,
      "step": 2612,
      "train_runtime": 2647.6739,
      "train_tokens_per_second": 5962.202
    },
    {
      "epoch": 0.961103448275862,
      "grad_norm": 0.7459332346916199,
      "learning_rate": 0.00010403092214246272,
      "loss": 1.845,
      "num_input_tokens_seen": 15790928,
      "step": 2613,
      "train_runtime": 2648.5297,
      "train_tokens_per_second": 5962.149
    },
    {
      "epoch": 0.961471264367816,
      "grad_norm": 0.7828077673912048,
      "learning_rate": 0.00010399411006810233,
      "loss": 1.8608,
      "num_input_tokens_seen": 15796112,
      "step": 2614,
      "train_runtime": 2649.4005,
      "train_tokens_per_second": 5962.146
    },
    {
      "epoch": 0.9618390804597701,
      "grad_norm": 0.7853997945785522,
      "learning_rate": 0.00010395729799374194,
      "loss": 1.6777,
      "num_input_tokens_seen": 15800560,
      "step": 2615,
      "train_runtime": 2650.1577,
      "train_tokens_per_second": 5962.121
    },
    {
      "epoch": 0.9622068965517241,
      "grad_norm": 0.815669059753418,
      "learning_rate": 0.00010392048591938158,
      "loss": 1.8344,
      "num_input_tokens_seen": 15805104,
      "step": 2616,
      "train_runtime": 2650.9289,
      "train_tokens_per_second": 5962.1
    },
    {
      "epoch": 0.9625747126436781,
      "grad_norm": 0.7245709300041199,
      "learning_rate": 0.00010388367384502119,
      "loss": 1.7529,
      "num_input_tokens_seen": 15811808,
      "step": 2617,
      "train_runtime": 2652.0078,
      "train_tokens_per_second": 5962.203
    },
    {
      "epoch": 0.9629425287356321,
      "grad_norm": 0.792323648929596,
      "learning_rate": 0.00010384686177066078,
      "loss": 1.704,
      "num_input_tokens_seen": 15818512,
      "step": 2618,
      "train_runtime": 2653.08,
      "train_tokens_per_second": 5962.32
    },
    {
      "epoch": 0.9633103448275863,
      "grad_norm": 0.880584180355072,
      "learning_rate": 0.00010381004969630039,
      "loss": 2.0683,
      "num_input_tokens_seen": 15822960,
      "step": 2619,
      "train_runtime": 2653.8374,
      "train_tokens_per_second": 5962.294
    },
    {
      "epoch": 0.9636781609195403,
      "grad_norm": 0.7375814318656921,
      "learning_rate": 0.00010377323762194,
      "loss": 1.5941,
      "num_input_tokens_seen": 15831728,
      "step": 2620,
      "train_runtime": 2655.1664,
      "train_tokens_per_second": 5962.612
    },
    {
      "epoch": 0.9640459770114943,
      "grad_norm": 0.7851631045341492,
      "learning_rate": 0.00010373642554757961,
      "loss": 1.8205,
      "num_input_tokens_seen": 15836816,
      "step": 2621,
      "train_runtime": 2656.0176,
      "train_tokens_per_second": 5962.617
    },
    {
      "epoch": 0.9644137931034483,
      "grad_norm": 0.7137867212295532,
      "learning_rate": 0.00010369961347321922,
      "loss": 1.6866,
      "num_input_tokens_seen": 15842976,
      "step": 2622,
      "train_runtime": 2656.9987,
      "train_tokens_per_second": 5962.734
    },
    {
      "epoch": 0.9647816091954023,
      "grad_norm": 0.7727775573730469,
      "learning_rate": 0.00010366280139885883,
      "loss": 1.6077,
      "num_input_tokens_seen": 15850080,
      "step": 2623,
      "train_runtime": 2658.1403,
      "train_tokens_per_second": 5962.846
    },
    {
      "epoch": 0.9651494252873564,
      "grad_norm": 0.7420926690101624,
      "learning_rate": 0.00010362598932449845,
      "loss": 1.6499,
      "num_input_tokens_seen": 15856640,
      "step": 2624,
      "train_runtime": 2659.191,
      "train_tokens_per_second": 5962.956
    },
    {
      "epoch": 0.9655172413793104,
      "grad_norm": 0.7900503277778625,
      "learning_rate": 0.00010358917725013806,
      "loss": 1.7608,
      "num_input_tokens_seen": 15862512,
      "step": 2625,
      "train_runtime": 2660.1709,
      "train_tokens_per_second": 5962.967
    },
    {
      "epoch": 0.9658850574712644,
      "grad_norm": 0.7923082113265991,
      "learning_rate": 0.00010355236517577767,
      "loss": 1.8159,
      "num_input_tokens_seen": 15870880,
      "step": 2626,
      "train_runtime": 2661.47,
      "train_tokens_per_second": 5963.201
    },
    {
      "epoch": 0.9662528735632184,
      "grad_norm": 0.6829389333724976,
      "learning_rate": 0.00010351555310141727,
      "loss": 1.691,
      "num_input_tokens_seen": 15880576,
      "step": 2627,
      "train_runtime": 2662.9503,
      "train_tokens_per_second": 5963.527
    },
    {
      "epoch": 0.9666206896551724,
      "grad_norm": 0.9263708591461182,
      "learning_rate": 0.00010347874102705688,
      "loss": 1.6139,
      "num_input_tokens_seen": 15885968,
      "step": 2628,
      "train_runtime": 2663.8306,
      "train_tokens_per_second": 5963.58
    },
    {
      "epoch": 0.9669885057471265,
      "grad_norm": 0.8506389260292053,
      "learning_rate": 0.00010344192895269649,
      "loss": 1.7958,
      "num_input_tokens_seen": 15891008,
      "step": 2629,
      "train_runtime": 2664.6889,
      "train_tokens_per_second": 5963.551
    },
    {
      "epoch": 0.9673563218390805,
      "grad_norm": 0.7076355218887329,
      "learning_rate": 0.0001034051168783361,
      "loss": 1.739,
      "num_input_tokens_seen": 15896880,
      "step": 2630,
      "train_runtime": 2665.6396,
      "train_tokens_per_second": 5963.627
    },
    {
      "epoch": 0.9677241379310345,
      "grad_norm": 0.7621473670005798,
      "learning_rate": 0.0001033683048039757,
      "loss": 1.8364,
      "num_input_tokens_seen": 15901072,
      "step": 2631,
      "train_runtime": 2666.3713,
      "train_tokens_per_second": 5963.563
    },
    {
      "epoch": 0.9680919540229885,
      "grad_norm": 0.7550080418586731,
      "learning_rate": 0.0001033314927296153,
      "loss": 1.6171,
      "num_input_tokens_seen": 15907152,
      "step": 2632,
      "train_runtime": 2667.379,
      "train_tokens_per_second": 5963.589
    },
    {
      "epoch": 0.9684597701149426,
      "grad_norm": 0.7544217705726624,
      "learning_rate": 0.00010329468065525494,
      "loss": 1.8811,
      "num_input_tokens_seen": 15915200,
      "step": 2633,
      "train_runtime": 2668.6381,
      "train_tokens_per_second": 5963.791
    },
    {
      "epoch": 0.9688275862068966,
      "grad_norm": 0.7832741737365723,
      "learning_rate": 0.00010325786858089455,
      "loss": 1.5898,
      "num_input_tokens_seen": 15921936,
      "step": 2634,
      "train_runtime": 2669.7265,
      "train_tokens_per_second": 5963.883
    },
    {
      "epoch": 0.9691954022988506,
      "grad_norm": 0.6932991147041321,
      "learning_rate": 0.00010322105650653416,
      "loss": 1.6538,
      "num_input_tokens_seen": 15928800,
      "step": 2635,
      "train_runtime": 2670.8082,
      "train_tokens_per_second": 5964.037
    },
    {
      "epoch": 0.9695632183908046,
      "grad_norm": 0.7697433829307556,
      "learning_rate": 0.00010318424443217375,
      "loss": 1.5751,
      "num_input_tokens_seen": 15937376,
      "step": 2636,
      "train_runtime": 2672.1255,
      "train_tokens_per_second": 5964.307
    },
    {
      "epoch": 0.9699310344827586,
      "grad_norm": 0.7205894589424133,
      "learning_rate": 0.00010314743235781336,
      "loss": 1.6248,
      "num_input_tokens_seen": 15943024,
      "step": 2637,
      "train_runtime": 2673.0484,
      "train_tokens_per_second": 5964.36
    },
    {
      "epoch": 0.9702988505747127,
      "grad_norm": 0.7500474452972412,
      "learning_rate": 0.00010311062028345297,
      "loss": 1.6561,
      "num_input_tokens_seen": 15954080,
      "step": 2638,
      "train_runtime": 2674.722,
      "train_tokens_per_second": 5964.762
    },
    {
      "epoch": 0.9706666666666667,
      "grad_norm": 0.7663933038711548,
      "learning_rate": 0.00010307380820909258,
      "loss": 1.833,
      "num_input_tokens_seen": 15959088,
      "step": 2639,
      "train_runtime": 2675.5563,
      "train_tokens_per_second": 5964.774
    },
    {
      "epoch": 0.9710344827586207,
      "grad_norm": 0.7608526945114136,
      "learning_rate": 0.00010303699613473219,
      "loss": 1.7386,
      "num_input_tokens_seen": 15964128,
      "step": 2640,
      "train_runtime": 2676.3882,
      "train_tokens_per_second": 5964.803
    },
    {
      "epoch": 0.9714022988505747,
      "grad_norm": 0.7225230932235718,
      "learning_rate": 0.0001030001840603718,
      "loss": 1.6109,
      "num_input_tokens_seen": 15969472,
      "step": 2641,
      "train_runtime": 2677.8067,
      "train_tokens_per_second": 5963.639
    },
    {
      "epoch": 0.9717701149425287,
      "grad_norm": 0.7210283279418945,
      "learning_rate": 0.00010296337198601142,
      "loss": 1.5441,
      "num_input_tokens_seen": 15974688,
      "step": 2642,
      "train_runtime": 2678.6548,
      "train_tokens_per_second": 5963.698
    },
    {
      "epoch": 0.9721379310344828,
      "grad_norm": 0.7956779599189758,
      "learning_rate": 0.00010292655991165103,
      "loss": 1.8876,
      "num_input_tokens_seen": 15979536,
      "step": 2643,
      "train_runtime": 2679.5001,
      "train_tokens_per_second": 5963.626
    },
    {
      "epoch": 0.9725057471264368,
      "grad_norm": 0.7508404850959778,
      "learning_rate": 0.00010288974783729064,
      "loss": 1.6356,
      "num_input_tokens_seen": 15985264,
      "step": 2644,
      "train_runtime": 2680.4414,
      "train_tokens_per_second": 5963.669
    },
    {
      "epoch": 0.9728735632183908,
      "grad_norm": 0.7549239993095398,
      "learning_rate": 0.00010285293576293025,
      "loss": 1.6046,
      "num_input_tokens_seen": 15989952,
      "step": 2645,
      "train_runtime": 2681.2477,
      "train_tokens_per_second": 5963.624
    },
    {
      "epoch": 0.9732413793103448,
      "grad_norm": 0.8114033341407776,
      "learning_rate": 0.00010281612368856986,
      "loss": 1.8563,
      "num_input_tokens_seen": 15994800,
      "step": 2646,
      "train_runtime": 2682.0655,
      "train_tokens_per_second": 5963.613
    },
    {
      "epoch": 0.9736091954022988,
      "grad_norm": 0.7232359647750854,
      "learning_rate": 0.00010277931161420947,
      "loss": 1.6082,
      "num_input_tokens_seen": 15999632,
      "step": 2647,
      "train_runtime": 2682.9083,
      "train_tokens_per_second": 5963.54
    },
    {
      "epoch": 0.9739770114942529,
      "grad_norm": 0.792028546333313,
      "learning_rate": 0.00010274249953984907,
      "loss": 1.7005,
      "num_input_tokens_seen": 16006224,
      "step": 2648,
      "train_runtime": 2683.9871,
      "train_tokens_per_second": 5963.599
    },
    {
      "epoch": 0.9743448275862069,
      "grad_norm": 0.8042363524436951,
      "learning_rate": 0.00010270568746548867,
      "loss": 1.6783,
      "num_input_tokens_seen": 16011488,
      "step": 2649,
      "train_runtime": 2684.8653,
      "train_tokens_per_second": 5963.609
    },
    {
      "epoch": 0.9747126436781609,
      "grad_norm": 0.7604905962944031,
      "learning_rate": 0.0001026688753911283,
      "loss": 1.5909,
      "num_input_tokens_seen": 16018384,
      "step": 2650,
      "train_runtime": 2685.9837,
      "train_tokens_per_second": 5963.694
    },
    {
      "epoch": 0.9750804597701149,
      "grad_norm": 0.7024556994438171,
      "learning_rate": 0.00010263206331676791,
      "loss": 1.609,
      "num_input_tokens_seen": 16023616,
      "step": 2651,
      "train_runtime": 2686.8706,
      "train_tokens_per_second": 5963.672
    },
    {
      "epoch": 0.975448275862069,
      "grad_norm": 0.7983068227767944,
      "learning_rate": 0.00010259525124240752,
      "loss": 1.755,
      "num_input_tokens_seen": 16028240,
      "step": 2652,
      "train_runtime": 2687.6906,
      "train_tokens_per_second": 5963.573
    },
    {
      "epoch": 0.975816091954023,
      "grad_norm": 0.8283487558364868,
      "learning_rate": 0.00010255843916804713,
      "loss": 1.7713,
      "num_input_tokens_seen": 16036064,
      "step": 2653,
      "train_runtime": 2688.9164,
      "train_tokens_per_second": 5963.764
    },
    {
      "epoch": 0.976183908045977,
      "grad_norm": 0.7740880250930786,
      "learning_rate": 0.00010252162709368673,
      "loss": 1.7162,
      "num_input_tokens_seen": 16040176,
      "step": 2654,
      "train_runtime": 2689.6435,
      "train_tokens_per_second": 5963.681
    },
    {
      "epoch": 0.976551724137931,
      "grad_norm": 0.8273856043815613,
      "learning_rate": 0.00010248481501932634,
      "loss": 1.8327,
      "num_input_tokens_seen": 16045504,
      "step": 2655,
      "train_runtime": 2690.5483,
      "train_tokens_per_second": 5963.656
    },
    {
      "epoch": 0.976919540229885,
      "grad_norm": 0.6738196015357971,
      "learning_rate": 0.00010244800294496595,
      "loss": 1.7243,
      "num_input_tokens_seen": 16050832,
      "step": 2656,
      "train_runtime": 2691.4469,
      "train_tokens_per_second": 5963.644
    },
    {
      "epoch": 0.977287356321839,
      "grad_norm": 0.7211321592330933,
      "learning_rate": 0.00010241119087060555,
      "loss": 1.6277,
      "num_input_tokens_seen": 16058032,
      "step": 2657,
      "train_runtime": 2692.5648,
      "train_tokens_per_second": 5963.842
    },
    {
      "epoch": 0.9776551724137931,
      "grad_norm": 0.8020983934402466,
      "learning_rate": 0.00010237437879624516,
      "loss": 1.6528,
      "num_input_tokens_seen": 16064752,
      "step": 2658,
      "train_runtime": 2693.6333,
      "train_tokens_per_second": 5963.971
    },
    {
      "epoch": 0.9780229885057471,
      "grad_norm": 0.8549041748046875,
      "learning_rate": 0.00010233756672188479,
      "loss": 1.7419,
      "num_input_tokens_seen": 16070624,
      "step": 2659,
      "train_runtime": 2694.5931,
      "train_tokens_per_second": 5964.026
    },
    {
      "epoch": 0.9783908045977011,
      "grad_norm": 0.7465891242027283,
      "learning_rate": 0.0001023007546475244,
      "loss": 1.5718,
      "num_input_tokens_seen": 16079456,
      "step": 2660,
      "train_runtime": 2695.9589,
      "train_tokens_per_second": 5964.281
    },
    {
      "epoch": 0.9787586206896551,
      "grad_norm": 0.734975278377533,
      "learning_rate": 0.000102263942573164,
      "loss": 1.5642,
      "num_input_tokens_seen": 16087600,
      "step": 2661,
      "train_runtime": 2697.2296,
      "train_tokens_per_second": 5964.49
    },
    {
      "epoch": 0.9791264367816092,
      "grad_norm": 0.7577536106109619,
      "learning_rate": 0.00010222713049880361,
      "loss": 1.8562,
      "num_input_tokens_seen": 16093136,
      "step": 2662,
      "train_runtime": 2698.1667,
      "train_tokens_per_second": 5964.471
    },
    {
      "epoch": 0.9794942528735632,
      "grad_norm": 0.7642478942871094,
      "learning_rate": 0.00010219031842444322,
      "loss": 1.7886,
      "num_input_tokens_seen": 16098784,
      "step": 2663,
      "train_runtime": 2699.1011,
      "train_tokens_per_second": 5964.498
    },
    {
      "epoch": 0.9798620689655172,
      "grad_norm": 0.7987829446792603,
      "learning_rate": 0.00010215350635008283,
      "loss": 1.8555,
      "num_input_tokens_seen": 16105008,
      "step": 2664,
      "train_runtime": 2700.0877,
      "train_tokens_per_second": 5964.624
    },
    {
      "epoch": 0.9802298850574712,
      "grad_norm": 0.7660846710205078,
      "learning_rate": 0.00010211669427572244,
      "loss": 1.6617,
      "num_input_tokens_seen": 16112032,
      "step": 2665,
      "train_runtime": 2701.2047,
      "train_tokens_per_second": 5964.758
    },
    {
      "epoch": 0.9805977011494252,
      "grad_norm": 0.7460705041885376,
      "learning_rate": 0.00010207988220136205,
      "loss": 1.6483,
      "num_input_tokens_seen": 16121664,
      "step": 2666,
      "train_runtime": 2702.6489,
      "train_tokens_per_second": 5965.135
    },
    {
      "epoch": 0.9809655172413793,
      "grad_norm": 0.8566172122955322,
      "learning_rate": 0.00010204307012700167,
      "loss": 1.7779,
      "num_input_tokens_seen": 16126672,
      "step": 2667,
      "train_runtime": 2703.4859,
      "train_tokens_per_second": 5965.14
    },
    {
      "epoch": 0.9813333333333333,
      "grad_norm": 0.7617372870445251,
      "learning_rate": 0.00010200625805264128,
      "loss": 1.8134,
      "num_input_tokens_seen": 16131856,
      "step": 2668,
      "train_runtime": 2704.3445,
      "train_tokens_per_second": 5965.163
    },
    {
      "epoch": 0.9817011494252874,
      "grad_norm": 0.8129032254219055,
      "learning_rate": 0.00010196944597828089,
      "loss": 1.7059,
      "num_input_tokens_seen": 16137744,
      "step": 2669,
      "train_runtime": 2705.3083,
      "train_tokens_per_second": 5965.214
    },
    {
      "epoch": 0.9820689655172414,
      "grad_norm": 0.8158069252967834,
      "learning_rate": 0.0001019326339039205,
      "loss": 1.9423,
      "num_input_tokens_seen": 16142352,
      "step": 2670,
      "train_runtime": 2706.1306,
      "train_tokens_per_second": 5965.105
    },
    {
      "epoch": 0.9824367816091955,
      "grad_norm": 0.7910207509994507,
      "learning_rate": 0.0001018958218295601,
      "loss": 1.6666,
      "num_input_tokens_seen": 16147120,
      "step": 2671,
      "train_runtime": 2707.498,
      "train_tokens_per_second": 5963.853
    },
    {
      "epoch": 0.9828045977011495,
      "grad_norm": 0.7697291374206543,
      "learning_rate": 0.0001018590097551997,
      "loss": 1.7376,
      "num_input_tokens_seen": 16152416,
      "step": 2672,
      "train_runtime": 2708.3828,
      "train_tokens_per_second": 5963.86
    },
    {
      "epoch": 0.9831724137931035,
      "grad_norm": 0.7793175578117371,
      "learning_rate": 0.00010182219768083931,
      "loss": 1.9242,
      "num_input_tokens_seen": 16156976,
      "step": 2673,
      "train_runtime": 2709.1924,
      "train_tokens_per_second": 5963.761
    },
    {
      "epoch": 0.9835402298850575,
      "grad_norm": 0.6621190309524536,
      "learning_rate": 0.00010178538560647892,
      "loss": 1.5626,
      "num_input_tokens_seen": 16162528,
      "step": 2674,
      "train_runtime": 2710.1155,
      "train_tokens_per_second": 5963.778
    },
    {
      "epoch": 0.9839080459770115,
      "grad_norm": 0.6125012636184692,
      "learning_rate": 0.00010174857353211853,
      "loss": 1.4731,
      "num_input_tokens_seen": 16174080,
      "step": 2675,
      "train_runtime": 2711.8501,
      "train_tokens_per_second": 5964.223
    },
    {
      "epoch": 0.9842758620689656,
      "grad_norm": 0.6930036544799805,
      "learning_rate": 0.00010171176145775816,
      "loss": 1.6377,
      "num_input_tokens_seen": 16180304,
      "step": 2676,
      "train_runtime": 2712.8588,
      "train_tokens_per_second": 5964.3
    },
    {
      "epoch": 0.9846436781609196,
      "grad_norm": 0.721152126789093,
      "learning_rate": 0.00010167494938339776,
      "loss": 1.4922,
      "num_input_tokens_seen": 16187296,
      "step": 2677,
      "train_runtime": 2713.998,
      "train_tokens_per_second": 5964.373
    },
    {
      "epoch": 0.9850114942528736,
      "grad_norm": 0.6925070881843567,
      "learning_rate": 0.00010163813730903737,
      "loss": 1.7035,
      "num_input_tokens_seen": 16194480,
      "step": 2678,
      "train_runtime": 2715.1223,
      "train_tokens_per_second": 5964.549
    },
    {
      "epoch": 0.9853793103448276,
      "grad_norm": 0.7697952389717102,
      "learning_rate": 0.00010160132523467698,
      "loss": 1.8778,
      "num_input_tokens_seen": 16199648,
      "step": 2679,
      "train_runtime": 2715.9948,
      "train_tokens_per_second": 5964.536
    },
    {
      "epoch": 0.9857471264367816,
      "grad_norm": 0.729809045791626,
      "learning_rate": 0.00010156451316031659,
      "loss": 1.5232,
      "num_input_tokens_seen": 16204992,
      "step": 2680,
      "train_runtime": 2716.8803,
      "train_tokens_per_second": 5964.559
    },
    {
      "epoch": 0.9861149425287357,
      "grad_norm": 0.7619598507881165,
      "learning_rate": 0.0001015277010859562,
      "loss": 1.7855,
      "num_input_tokens_seen": 16211360,
      "step": 2681,
      "train_runtime": 2717.9292,
      "train_tokens_per_second": 5964.6
    },
    {
      "epoch": 0.9864827586206897,
      "grad_norm": 0.7749170064926147,
      "learning_rate": 0.0001014908890115958,
      "loss": 1.6563,
      "num_input_tokens_seen": 16215712,
      "step": 2682,
      "train_runtime": 2718.6785,
      "train_tokens_per_second": 5964.557
    },
    {
      "epoch": 0.9868505747126437,
      "grad_norm": 0.7425233721733093,
      "learning_rate": 0.00010145407693723541,
      "loss": 1.7613,
      "num_input_tokens_seen": 16220592,
      "step": 2683,
      "train_runtime": 2719.5123,
      "train_tokens_per_second": 5964.522
    },
    {
      "epoch": 0.9872183908045977,
      "grad_norm": 0.7368447780609131,
      "learning_rate": 0.00010141726486287504,
      "loss": 1.6202,
      "num_input_tokens_seen": 16228992,
      "step": 2684,
      "train_runtime": 2720.8321,
      "train_tokens_per_second": 5964.716
    },
    {
      "epoch": 0.9875862068965517,
      "grad_norm": 0.7763941287994385,
      "learning_rate": 0.00010138045278851464,
      "loss": 1.5027,
      "num_input_tokens_seen": 16233184,
      "step": 2685,
      "train_runtime": 2721.5611,
      "train_tokens_per_second": 5964.659
    },
    {
      "epoch": 0.9879540229885058,
      "grad_norm": 0.7982324957847595,
      "learning_rate": 0.00010134364071415425,
      "loss": 1.7941,
      "num_input_tokens_seen": 16237728,
      "step": 2686,
      "train_runtime": 2722.3496,
      "train_tokens_per_second": 5964.601
    },
    {
      "epoch": 0.9883218390804598,
      "grad_norm": 0.7863630652427673,
      "learning_rate": 0.00010130682863979386,
      "loss": 1.6936,
      "num_input_tokens_seen": 16243328,
      "step": 2687,
      "train_runtime": 2723.2512,
      "train_tokens_per_second": 5964.682
    },
    {
      "epoch": 0.9886896551724138,
      "grad_norm": 0.8651296496391296,
      "learning_rate": 0.00010127001656543347,
      "loss": 1.6675,
      "num_input_tokens_seen": 16249632,
      "step": 2688,
      "train_runtime": 2724.2801,
      "train_tokens_per_second": 5964.743
    },
    {
      "epoch": 0.9890574712643678,
      "grad_norm": 0.8110333681106567,
      "learning_rate": 0.00010123320449107308,
      "loss": 1.6978,
      "num_input_tokens_seen": 16255552,
      "step": 2689,
      "train_runtime": 2725.2265,
      "train_tokens_per_second": 5964.844
    },
    {
      "epoch": 0.9894252873563218,
      "grad_norm": 0.8143921494483948,
      "learning_rate": 0.00010119639241671269,
      "loss": 1.6795,
      "num_input_tokens_seen": 16260688,
      "step": 2690,
      "train_runtime": 2726.1166,
      "train_tokens_per_second": 5964.781
    },
    {
      "epoch": 0.9897931034482759,
      "grad_norm": 0.7819022536277771,
      "learning_rate": 0.00010115958034235228,
      "loss": 1.6987,
      "num_input_tokens_seen": 16265776,
      "step": 2691,
      "train_runtime": 2726.9922,
      "train_tokens_per_second": 5964.731
    },
    {
      "epoch": 0.9901609195402299,
      "grad_norm": 0.8476383090019226,
      "learning_rate": 0.00010112276826799189,
      "loss": 1.8078,
      "num_input_tokens_seen": 16270096,
      "step": 2692,
      "train_runtime": 2727.7609,
      "train_tokens_per_second": 5964.634
    },
    {
      "epoch": 0.9905287356321839,
      "grad_norm": 0.8257720470428467,
      "learning_rate": 0.00010108595619363153,
      "loss": 1.6867,
      "num_input_tokens_seen": 16276400,
      "step": 2693,
      "train_runtime": 2728.7838,
      "train_tokens_per_second": 5964.709
    },
    {
      "epoch": 0.9908965517241379,
      "grad_norm": 0.8439770936965942,
      "learning_rate": 0.00010104914411927114,
      "loss": 1.8791,
      "num_input_tokens_seen": 16282608,
      "step": 2694,
      "train_runtime": 2729.8087,
      "train_tokens_per_second": 5964.743
    },
    {
      "epoch": 0.991264367816092,
      "grad_norm": 0.8019626140594482,
      "learning_rate": 0.00010101233204491075,
      "loss": 1.795,
      "num_input_tokens_seen": 16288464,
      "step": 2695,
      "train_runtime": 2730.7785,
      "train_tokens_per_second": 5964.769
    },
    {
      "epoch": 0.991632183908046,
      "grad_norm": 0.7709932923316956,
      "learning_rate": 0.00010097551997055034,
      "loss": 1.6781,
      "num_input_tokens_seen": 16293968,
      "step": 2696,
      "train_runtime": 2731.6941,
      "train_tokens_per_second": 5964.785
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.8245759606361389,
      "learning_rate": 0.00010093870789618995,
      "loss": 1.8092,
      "num_input_tokens_seen": 16299040,
      "step": 2697,
      "train_runtime": 2732.5401,
      "train_tokens_per_second": 5964.794
    },
    {
      "epoch": 0.992367816091954,
      "grad_norm": 0.8198642730712891,
      "learning_rate": 0.00010090189582182956,
      "loss": 1.8803,
      "num_input_tokens_seen": 16303984,
      "step": 2698,
      "train_runtime": 2733.3877,
      "train_tokens_per_second": 5964.754
    },
    {
      "epoch": 0.992735632183908,
      "grad_norm": 0.8194162249565125,
      "learning_rate": 0.00010086508374746917,
      "loss": 1.6704,
      "num_input_tokens_seen": 16309616,
      "step": 2699,
      "train_runtime": 2734.3186,
      "train_tokens_per_second": 5964.783
    },
    {
      "epoch": 0.993103448275862,
      "grad_norm": 0.7265117764472961,
      "learning_rate": 0.00010082827167310878,
      "loss": 1.633,
      "num_input_tokens_seen": 16315472,
      "step": 2700,
      "train_runtime": 2735.2715,
      "train_tokens_per_second": 5964.846
    },
    {
      "epoch": 0.9934712643678161,
      "grad_norm": 0.8403815627098083,
      "learning_rate": 0.00010079145959874839,
      "loss": 1.6689,
      "num_input_tokens_seen": 16321440,
      "step": 2701,
      "train_runtime": 2736.7195,
      "train_tokens_per_second": 5963.87
    },
    {
      "epoch": 0.9938390804597701,
      "grad_norm": 0.722404956817627,
      "learning_rate": 0.00010075464752438801,
      "loss": 1.6407,
      "num_input_tokens_seen": 16327072,
      "step": 2702,
      "train_runtime": 2737.6809,
      "train_tokens_per_second": 5963.833
    },
    {
      "epoch": 0.9942068965517241,
      "grad_norm": 0.7589871883392334,
      "learning_rate": 0.00010071783545002762,
      "loss": 1.7493,
      "num_input_tokens_seen": 16332288,
      "step": 2703,
      "train_runtime": 2738.5263,
      "train_tokens_per_second": 5963.897
    },
    {
      "epoch": 0.9945747126436781,
      "grad_norm": 0.707599401473999,
      "learning_rate": 0.00010068102337566723,
      "loss": 1.8137,
      "num_input_tokens_seen": 16337664,
      "step": 2704,
      "train_runtime": 2739.43,
      "train_tokens_per_second": 5963.892
    },
    {
      "epoch": 0.9949425287356322,
      "grad_norm": 0.7044679522514343,
      "learning_rate": 0.00010064421130130684,
      "loss": 1.5024,
      "num_input_tokens_seen": 16347600,
      "step": 2705,
      "train_runtime": 2740.9504,
      "train_tokens_per_second": 5964.209
    },
    {
      "epoch": 0.9953103448275862,
      "grad_norm": 0.7717908620834351,
      "learning_rate": 0.00010060739922694644,
      "loss": 1.7201,
      "num_input_tokens_seen": 16353184,
      "step": 2706,
      "train_runtime": 2741.8729,
      "train_tokens_per_second": 5964.239
    },
    {
      "epoch": 0.9956781609195402,
      "grad_norm": 0.6623258590698242,
      "learning_rate": 0.00010057058715258605,
      "loss": 1.714,
      "num_input_tokens_seen": 16360256,
      "step": 2707,
      "train_runtime": 2743.0061,
      "train_tokens_per_second": 5964.353
    },
    {
      "epoch": 0.9960459770114942,
      "grad_norm": 0.7541612982749939,
      "learning_rate": 0.00010053377507822566,
      "loss": 1.7051,
      "num_input_tokens_seen": 16367104,
      "step": 2708,
      "train_runtime": 2744.1124,
      "train_tokens_per_second": 5964.444
    },
    {
      "epoch": 0.9964137931034482,
      "grad_norm": 0.7169268727302551,
      "learning_rate": 0.00010049696300386526,
      "loss": 1.5708,
      "num_input_tokens_seen": 16372464,
      "step": 2709,
      "train_runtime": 2745.0082,
      "train_tokens_per_second": 5964.45
    },
    {
      "epoch": 0.9967816091954023,
      "grad_norm": 0.822113573551178,
      "learning_rate": 0.0001004601509295049,
      "loss": 1.7626,
      "num_input_tokens_seen": 16377360,
      "step": 2710,
      "train_runtime": 2745.8413,
      "train_tokens_per_second": 5964.423
    },
    {
      "epoch": 0.9971494252873563,
      "grad_norm": 0.807174026966095,
      "learning_rate": 0.0001004233388551445,
      "loss": 1.8715,
      "num_input_tokens_seen": 16383696,
      "step": 2711,
      "train_runtime": 2746.8574,
      "train_tokens_per_second": 5964.524
    },
    {
      "epoch": 0.9975172413793103,
      "grad_norm": 0.7619220018386841,
      "learning_rate": 0.00010038652678078411,
      "loss": 1.7394,
      "num_input_tokens_seen": 16390272,
      "step": 2712,
      "train_runtime": 2747.932,
      "train_tokens_per_second": 5964.584
    },
    {
      "epoch": 0.9978850574712643,
      "grad_norm": 0.7220108509063721,
      "learning_rate": 0.00010034971470642372,
      "loss": 1.6763,
      "num_input_tokens_seen": 16397728,
      "step": 2713,
      "train_runtime": 2749.1208,
      "train_tokens_per_second": 5964.717
    },
    {
      "epoch": 0.9982528735632183,
      "grad_norm": 0.7697573900222778,
      "learning_rate": 0.00010031290263206332,
      "loss": 1.8442,
      "num_input_tokens_seen": 16403408,
      "step": 2714,
      "train_runtime": 2750.054,
      "train_tokens_per_second": 5964.759
    },
    {
      "epoch": 0.9986206896551724,
      "grad_norm": 0.7933111786842346,
      "learning_rate": 0.00010027609055770292,
      "loss": 1.7856,
      "num_input_tokens_seen": 16408192,
      "step": 2715,
      "train_runtime": 2750.892,
      "train_tokens_per_second": 5964.68
    },
    {
      "epoch": 0.9989885057471264,
      "grad_norm": 0.7672939896583557,
      "learning_rate": 0.00010023927848334253,
      "loss": 1.541,
      "num_input_tokens_seen": 16415936,
      "step": 2716,
      "train_runtime": 2752.1055,
      "train_tokens_per_second": 5964.864
    },
    {
      "epoch": 0.9993563218390804,
      "grad_norm": 0.7983112931251526,
      "learning_rate": 0.00010020246640898214,
      "loss": 1.8038,
      "num_input_tokens_seen": 16421248,
      "step": 2717,
      "train_runtime": 2752.9986,
      "train_tokens_per_second": 5964.859
    },
    {
      "epoch": 0.9997241379310344,
      "grad_norm": 0.7845472097396851,
      "learning_rate": 0.00010016565433462175,
      "loss": 1.6246,
      "num_input_tokens_seen": 16426000,
      "step": 2718,
      "train_runtime": 2753.821,
      "train_tokens_per_second": 5964.803
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.9094229936599731,
      "learning_rate": 0.00010012884226026137,
      "loss": 1.7112,
      "num_input_tokens_seen": 16430656,
      "step": 2719,
      "train_runtime": 2754.5881,
      "train_tokens_per_second": 5964.832
    },
    {
      "epoch": 1.000367816091954,
      "grad_norm": 0.7175472974777222,
      "learning_rate": 0.00010009203018590098,
      "loss": 1.5703,
      "num_input_tokens_seen": 16437632,
      "step": 2720,
      "train_runtime": 2755.7046,
      "train_tokens_per_second": 5964.947
    },
    {
      "epoch": 1.000367816091954,
      "eval_loss": 1.7600165605545044,
      "eval_runtime": 4.7813,
      "eval_samples_per_second": 209.149,
      "eval_steps_per_second": 13.176,
      "num_input_tokens_seen": 16437632,
      "step": 2720
    },
    {
      "epoch": 1.000735632183908,
      "grad_norm": 0.7592592239379883,
      "learning_rate": 0.00010005521811154059,
      "loss": 1.4792,
      "num_input_tokens_seen": 16442400,
      "step": 2721,
      "train_runtime": 2761.2886,
      "train_tokens_per_second": 5954.611
    },
    {
      "epoch": 1.001103448275862,
      "grad_norm": 0.7093924283981323,
      "learning_rate": 0.0001000184060371802,
      "loss": 1.3225,
      "num_input_tokens_seen": 16447312,
      "step": 2722,
      "train_runtime": 2762.1416,
      "train_tokens_per_second": 5954.551
    },
    {
      "epoch": 1.001471264367816,
      "grad_norm": 0.7133712768554688,
      "learning_rate": 9.998159396281981e-05,
      "loss": 1.5349,
      "num_input_tokens_seen": 16452720,
      "step": 2723,
      "train_runtime": 2763.0534,
      "train_tokens_per_second": 5954.543
    },
    {
      "epoch": 1.00183908045977,
      "grad_norm": 0.7522051334381104,
      "learning_rate": 9.994478188845942e-05,
      "loss": 1.7682,
      "num_input_tokens_seen": 16457680,
      "step": 2724,
      "train_runtime": 2763.9061,
      "train_tokens_per_second": 5954.5
    },
    {
      "epoch": 1.0022068965517241,
      "grad_norm": 0.7381378412246704,
      "learning_rate": 9.990796981409904e-05,
      "loss": 1.6991,
      "num_input_tokens_seen": 16462784,
      "step": 2725,
      "train_runtime": 2764.7715,
      "train_tokens_per_second": 5954.483
    },
    {
      "epoch": 1.0025747126436781,
      "grad_norm": 0.7333172559738159,
      "learning_rate": 9.987115773973864e-05,
      "loss": 1.6771,
      "num_input_tokens_seen": 16468544,
      "step": 2726,
      "train_runtime": 2765.7272,
      "train_tokens_per_second": 5954.508
    },
    {
      "epoch": 1.0029425287356322,
      "grad_norm": 0.8070797920227051,
      "learning_rate": 9.983434566537824e-05,
      "loss": 1.5182,
      "num_input_tokens_seen": 16474976,
      "step": 2727,
      "train_runtime": 2766.7678,
      "train_tokens_per_second": 5954.593
    },
    {
      "epoch": 1.0033103448275862,
      "grad_norm": 0.6627984642982483,
      "learning_rate": 9.979753359101785e-05,
      "loss": 1.6668,
      "num_input_tokens_seen": 16481488,
      "step": 2728,
      "train_runtime": 2767.8205,
      "train_tokens_per_second": 5954.681
    },
    {
      "epoch": 1.0036781609195402,
      "grad_norm": 0.7076670527458191,
      "learning_rate": 9.976072151665748e-05,
      "loss": 1.5113,
      "num_input_tokens_seen": 16486960,
      "step": 2729,
      "train_runtime": 2768.7357,
      "train_tokens_per_second": 5954.689
    },
    {
      "epoch": 1.0040459770114942,
      "grad_norm": 0.7460461854934692,
      "learning_rate": 9.972390944229709e-05,
      "loss": 1.5207,
      "num_input_tokens_seen": 16494864,
      "step": 2730,
      "train_runtime": 2769.9809,
      "train_tokens_per_second": 5954.866
    },
    {
      "epoch": 1.0044137931034482,
      "grad_norm": 0.7616891264915466,
      "learning_rate": 9.96870973679367e-05,
      "loss": 1.5562,
      "num_input_tokens_seen": 16501632,
      "step": 2731,
      "train_runtime": 2771.5263,
      "train_tokens_per_second": 5953.987
    },
    {
      "epoch": 1.0047816091954023,
      "grad_norm": 0.7592271566390991,
      "learning_rate": 9.965028529357629e-05,
      "loss": 1.5601,
      "num_input_tokens_seen": 16506176,
      "step": 2732,
      "train_runtime": 2772.3117,
      "train_tokens_per_second": 5953.939
    },
    {
      "epoch": 1.0051494252873563,
      "grad_norm": 0.9049879908561707,
      "learning_rate": 9.96134732192159e-05,
      "loss": 1.757,
      "num_input_tokens_seen": 16510160,
      "step": 2733,
      "train_runtime": 2773.0059,
      "train_tokens_per_second": 5953.886
    },
    {
      "epoch": 1.0055172413793103,
      "grad_norm": 0.6635333895683289,
      "learning_rate": 9.957666114485552e-05,
      "loss": 1.5871,
      "num_input_tokens_seen": 16518736,
      "step": 2734,
      "train_runtime": 2774.3349,
      "train_tokens_per_second": 5954.125
    },
    {
      "epoch": 1.0058850574712643,
      "grad_norm": 0.861168622970581,
      "learning_rate": 9.953984907049513e-05,
      "loss": 1.7289,
      "num_input_tokens_seen": 16526848,
      "step": 2735,
      "train_runtime": 2775.622,
      "train_tokens_per_second": 5954.286
    },
    {
      "epoch": 1.0062528735632184,
      "grad_norm": 0.8384279012680054,
      "learning_rate": 9.950303699613474e-05,
      "loss": 1.4379,
      "num_input_tokens_seen": 16533408,
      "step": 2736,
      "train_runtime": 2776.6356,
      "train_tokens_per_second": 5954.475
    },
    {
      "epoch": 1.0066206896551724,
      "grad_norm": 0.7553403377532959,
      "learning_rate": 9.946622492177435e-05,
      "loss": 1.4498,
      "num_input_tokens_seen": 16539104,
      "step": 2737,
      "train_runtime": 2777.5735,
      "train_tokens_per_second": 5954.515
    },
    {
      "epoch": 1.0069885057471264,
      "grad_norm": 0.7888177633285522,
      "learning_rate": 9.942941284741396e-05,
      "loss": 1.6728,
      "num_input_tokens_seen": 16544336,
      "step": 2738,
      "train_runtime": 2778.4599,
      "train_tokens_per_second": 5954.499
    },
    {
      "epoch": 1.0073563218390804,
      "grad_norm": 0.7468155026435852,
      "learning_rate": 9.939260077305357e-05,
      "loss": 1.577,
      "num_input_tokens_seen": 16551264,
      "step": 2739,
      "train_runtime": 2779.5735,
      "train_tokens_per_second": 5954.606
    },
    {
      "epoch": 1.0077241379310344,
      "grad_norm": 0.742401659488678,
      "learning_rate": 9.935578869869317e-05,
      "loss": 1.5609,
      "num_input_tokens_seen": 16556080,
      "step": 2740,
      "train_runtime": 2780.406,
      "train_tokens_per_second": 5954.555
    },
    {
      "epoch": 1.0080919540229885,
      "grad_norm": 0.7920041084289551,
      "learning_rate": 9.931897662433278e-05,
      "loss": 1.4845,
      "num_input_tokens_seen": 16561616,
      "step": 2741,
      "train_runtime": 2781.3246,
      "train_tokens_per_second": 5954.579
    },
    {
      "epoch": 1.0084597701149425,
      "grad_norm": 0.8238051533699036,
      "learning_rate": 9.92821645499724e-05,
      "loss": 1.7318,
      "num_input_tokens_seen": 16566832,
      "step": 2742,
      "train_runtime": 2782.1996,
      "train_tokens_per_second": 5954.581
    },
    {
      "epoch": 1.0088275862068965,
      "grad_norm": 0.7177878618240356,
      "learning_rate": 9.924535247561201e-05,
      "loss": 1.5797,
      "num_input_tokens_seen": 16574032,
      "step": 2743,
      "train_runtime": 2783.3606,
      "train_tokens_per_second": 5954.684
    },
    {
      "epoch": 1.0091954022988505,
      "grad_norm": 0.7707662582397461,
      "learning_rate": 9.920854040125161e-05,
      "loss": 1.4792,
      "num_input_tokens_seen": 16582032,
      "step": 2744,
      "train_runtime": 2784.6296,
      "train_tokens_per_second": 5954.843
    },
    {
      "epoch": 1.0095632183908045,
      "grad_norm": 0.7912094593048096,
      "learning_rate": 9.917172832689122e-05,
      "loss": 1.7512,
      "num_input_tokens_seen": 16588464,
      "step": 2745,
      "train_runtime": 2785.6728,
      "train_tokens_per_second": 5954.922
    },
    {
      "epoch": 1.0099310344827586,
      "grad_norm": 0.7886949181556702,
      "learning_rate": 9.913491625253083e-05,
      "loss": 1.4886,
      "num_input_tokens_seen": 16594384,
      "step": 2746,
      "train_runtime": 2786.6532,
      "train_tokens_per_second": 5954.951
    },
    {
      "epoch": 1.0102988505747126,
      "grad_norm": 0.8673503994941711,
      "learning_rate": 9.909810417817045e-05,
      "loss": 1.4369,
      "num_input_tokens_seen": 16602416,
      "step": 2747,
      "train_runtime": 2787.922,
      "train_tokens_per_second": 5955.122
    },
    {
      "epoch": 1.0106666666666666,
      "grad_norm": 0.7904326915740967,
      "learning_rate": 9.906129210381006e-05,
      "loss": 1.5817,
      "num_input_tokens_seen": 16607776,
      "step": 2748,
      "train_runtime": 2788.8198,
      "train_tokens_per_second": 5955.127
    },
    {
      "epoch": 1.0110344827586206,
      "grad_norm": 0.781034529209137,
      "learning_rate": 9.902448002944967e-05,
      "loss": 1.5774,
      "num_input_tokens_seen": 16613664,
      "step": 2749,
      "train_runtime": 2789.804,
      "train_tokens_per_second": 5955.137
    },
    {
      "epoch": 1.0114022988505746,
      "grad_norm": 0.7959015369415283,
      "learning_rate": 9.898766795508926e-05,
      "loss": 1.4133,
      "num_input_tokens_seen": 16619056,
      "step": 2750,
      "train_runtime": 2790.7211,
      "train_tokens_per_second": 5955.112
    },
    {
      "epoch": 1.0117701149425287,
      "grad_norm": 0.7907975912094116,
      "learning_rate": 9.895085588072889e-05,
      "loss": 1.4479,
      "num_input_tokens_seen": 16627856,
      "step": 2751,
      "train_runtime": 2792.1191,
      "train_tokens_per_second": 5955.282
    },
    {
      "epoch": 1.0121379310344827,
      "grad_norm": 0.7597295641899109,
      "learning_rate": 9.89140438063685e-05,
      "loss": 1.4861,
      "num_input_tokens_seen": 16633824,
      "step": 2752,
      "train_runtime": 2793.1065,
      "train_tokens_per_second": 5955.313
    },
    {
      "epoch": 1.0125057471264367,
      "grad_norm": 0.7649272680282593,
      "learning_rate": 9.88772317320081e-05,
      "loss": 1.5986,
      "num_input_tokens_seen": 16638640,
      "step": 2753,
      "train_runtime": 2793.9267,
      "train_tokens_per_second": 5955.289
    },
    {
      "epoch": 1.0128735632183907,
      "grad_norm": 0.8077549934387207,
      "learning_rate": 9.884041965764771e-05,
      "loss": 1.4154,
      "num_input_tokens_seen": 16644992,
      "step": 2754,
      "train_runtime": 2794.9765,
      "train_tokens_per_second": 5955.324
    },
    {
      "epoch": 1.0132413793103447,
      "grad_norm": 0.8183664679527283,
      "learning_rate": 9.880360758328732e-05,
      "loss": 1.6345,
      "num_input_tokens_seen": 16650480,
      "step": 2755,
      "train_runtime": 2795.8847,
      "train_tokens_per_second": 5955.353
    },
    {
      "epoch": 1.0136091954022988,
      "grad_norm": 0.7877675890922546,
      "learning_rate": 9.876679550892693e-05,
      "loss": 1.5115,
      "num_input_tokens_seen": 16655632,
      "step": 2756,
      "train_runtime": 2796.7472,
      "train_tokens_per_second": 5955.359
    },
    {
      "epoch": 1.0139770114942528,
      "grad_norm": 0.7440232038497925,
      "learning_rate": 9.872998343456654e-05,
      "loss": 1.399,
      "num_input_tokens_seen": 16661456,
      "step": 2757,
      "train_runtime": 2797.6978,
      "train_tokens_per_second": 5955.417
    },
    {
      "epoch": 1.0143448275862068,
      "grad_norm": 0.80223548412323,
      "learning_rate": 9.869317136020615e-05,
      "loss": 1.558,
      "num_input_tokens_seen": 16665968,
      "step": 2758,
      "train_runtime": 2798.5093,
      "train_tokens_per_second": 5955.302
    },
    {
      "epoch": 1.0147126436781608,
      "grad_norm": 0.8115981817245483,
      "learning_rate": 9.865635928584577e-05,
      "loss": 1.7045,
      "num_input_tokens_seen": 16671952,
      "step": 2759,
      "train_runtime": 2799.4807,
      "train_tokens_per_second": 5955.373
    },
    {
      "epoch": 1.0150804597701149,
      "grad_norm": 0.8773238062858582,
      "learning_rate": 9.861954721148538e-05,
      "loss": 1.437,
      "num_input_tokens_seen": 16676896,
      "step": 2760,
      "train_runtime": 2800.314,
      "train_tokens_per_second": 5955.366
    },
    {
      "epoch": 1.0154482758620689,
      "grad_norm": 0.7974623441696167,
      "learning_rate": 9.858273513712499e-05,
      "loss": 1.6022,
      "num_input_tokens_seen": 16682224,
      "step": 2761,
      "train_runtime": 2801.7062,
      "train_tokens_per_second": 5954.309
    },
    {
      "epoch": 1.015816091954023,
      "grad_norm": 0.8365609645843506,
      "learning_rate": 9.854592306276458e-05,
      "loss": 1.7147,
      "num_input_tokens_seen": 16688048,
      "step": 2762,
      "train_runtime": 2802.6632,
      "train_tokens_per_second": 5954.354
    },
    {
      "epoch": 1.016183908045977,
      "grad_norm": 0.8881251811981201,
      "learning_rate": 9.850911098840419e-05,
      "loss": 1.5766,
      "num_input_tokens_seen": 16696640,
      "step": 2763,
      "train_runtime": 2804.0071,
      "train_tokens_per_second": 5954.564
    },
    {
      "epoch": 1.016551724137931,
      "grad_norm": 0.806018590927124,
      "learning_rate": 9.847229891404381e-05,
      "loss": 1.5516,
      "num_input_tokens_seen": 16701648,
      "step": 2764,
      "train_runtime": 2804.871,
      "train_tokens_per_second": 5954.516
    },
    {
      "epoch": 1.016919540229885,
      "grad_norm": 0.8690130710601807,
      "learning_rate": 9.843548683968342e-05,
      "loss": 1.5414,
      "num_input_tokens_seen": 16709984,
      "step": 2765,
      "train_runtime": 2806.2028,
      "train_tokens_per_second": 5954.66
    },
    {
      "epoch": 1.017287356321839,
      "grad_norm": 0.844491720199585,
      "learning_rate": 9.839867476532303e-05,
      "loss": 1.6738,
      "num_input_tokens_seen": 16715616,
      "step": 2766,
      "train_runtime": 2807.1099,
      "train_tokens_per_second": 5954.742
    },
    {
      "epoch": 1.017655172413793,
      "grad_norm": 0.8478877544403076,
      "learning_rate": 9.836186269096264e-05,
      "loss": 1.5235,
      "num_input_tokens_seen": 16721456,
      "step": 2767,
      "train_runtime": 2808.0622,
      "train_tokens_per_second": 5954.803
    },
    {
      "epoch": 1.018022988505747,
      "grad_norm": 0.7275803685188293,
      "learning_rate": 9.832505061660225e-05,
      "loss": 1.3195,
      "num_input_tokens_seen": 16730800,
      "step": 2768,
      "train_runtime": 2809.4963,
      "train_tokens_per_second": 5955.089
    },
    {
      "epoch": 1.018390804597701,
      "grad_norm": 0.7914295196533203,
      "learning_rate": 9.828823854224186e-05,
      "loss": 1.4319,
      "num_input_tokens_seen": 16736528,
      "step": 2769,
      "train_runtime": 2810.4284,
      "train_tokens_per_second": 5955.152
    },
    {
      "epoch": 1.0187586206896553,
      "grad_norm": 0.8010337948799133,
      "learning_rate": 9.825142646788147e-05,
      "loss": 1.7588,
      "num_input_tokens_seen": 16741472,
      "step": 2770,
      "train_runtime": 2811.2736,
      "train_tokens_per_second": 5955.12
    },
    {
      "epoch": 1.0191264367816093,
      "grad_norm": 0.8199329376220703,
      "learning_rate": 9.821461439352108e-05,
      "loss": 1.5805,
      "num_input_tokens_seen": 16746400,
      "step": 2771,
      "train_runtime": 2812.0898,
      "train_tokens_per_second": 5955.144
    },
    {
      "epoch": 1.0194942528735633,
      "grad_norm": 0.8445271253585815,
      "learning_rate": 9.81778023191607e-05,
      "loss": 1.6346,
      "num_input_tokens_seen": 16751408,
      "step": 2772,
      "train_runtime": 2812.9284,
      "train_tokens_per_second": 5955.149
    },
    {
      "epoch": 1.0198620689655173,
      "grad_norm": 0.7820270657539368,
      "learning_rate": 9.814099024480031e-05,
      "loss": 1.5438,
      "num_input_tokens_seen": 16758688,
      "step": 2773,
      "train_runtime": 2814.1051,
      "train_tokens_per_second": 5955.246
    },
    {
      "epoch": 1.0202298850574714,
      "grad_norm": 0.9145992398262024,
      "learning_rate": 9.81041781704399e-05,
      "loss": 1.5735,
      "num_input_tokens_seen": 16764672,
      "step": 2774,
      "train_runtime": 2815.0655,
      "train_tokens_per_second": 5955.34
    },
    {
      "epoch": 1.0205977011494254,
      "grad_norm": 0.8045191168785095,
      "learning_rate": 9.806736609607951e-05,
      "loss": 1.4078,
      "num_input_tokens_seen": 16770976,
      "step": 2775,
      "train_runtime": 2816.0972,
      "train_tokens_per_second": 5955.397
    },
    {
      "epoch": 1.0209655172413794,
      "grad_norm": 0.7787489891052246,
      "learning_rate": 9.803055402171912e-05,
      "loss": 1.5356,
      "num_input_tokens_seen": 16775568,
      "step": 2776,
      "train_runtime": 2816.8934,
      "train_tokens_per_second": 5955.344
    },
    {
      "epoch": 1.0213333333333334,
      "grad_norm": 0.8159419298171997,
      "learning_rate": 9.799374194735874e-05,
      "loss": 1.5622,
      "num_input_tokens_seen": 16781136,
      "step": 2777,
      "train_runtime": 2817.8314,
      "train_tokens_per_second": 5955.337
    },
    {
      "epoch": 1.0217011494252874,
      "grad_norm": 0.8529191017150879,
      "learning_rate": 9.795692987299835e-05,
      "loss": 1.6036,
      "num_input_tokens_seen": 16786640,
      "step": 2778,
      "train_runtime": 2818.751,
      "train_tokens_per_second": 5955.347
    },
    {
      "epoch": 1.0220689655172415,
      "grad_norm": 0.8145338296890259,
      "learning_rate": 9.792011779863796e-05,
      "loss": 1.553,
      "num_input_tokens_seen": 16796224,
      "step": 2779,
      "train_runtime": 2820.2285,
      "train_tokens_per_second": 5955.625
    },
    {
      "epoch": 1.0224367816091955,
      "grad_norm": 0.7739707827568054,
      "learning_rate": 9.788330572427756e-05,
      "loss": 1.521,
      "num_input_tokens_seen": 16804816,
      "step": 2780,
      "train_runtime": 2821.5699,
      "train_tokens_per_second": 5955.839
    },
    {
      "epoch": 1.0228045977011495,
      "grad_norm": 0.7659571766853333,
      "learning_rate": 9.784649364991718e-05,
      "loss": 1.5963,
      "num_input_tokens_seen": 16809664,
      "step": 2781,
      "train_runtime": 2822.4229,
      "train_tokens_per_second": 5955.757
    },
    {
      "epoch": 1.0231724137931035,
      "grad_norm": 0.8444596529006958,
      "learning_rate": 9.780968157555679e-05,
      "loss": 1.5421,
      "num_input_tokens_seen": 16817488,
      "step": 2782,
      "train_runtime": 2823.6528,
      "train_tokens_per_second": 5955.933
    },
    {
      "epoch": 1.0235402298850576,
      "grad_norm": 0.8246302008628845,
      "learning_rate": 9.77728695011964e-05,
      "loss": 1.6177,
      "num_input_tokens_seen": 16822448,
      "step": 2783,
      "train_runtime": 2824.4988,
      "train_tokens_per_second": 5955.906
    },
    {
      "epoch": 1.0239080459770116,
      "grad_norm": 0.8348226547241211,
      "learning_rate": 9.7736057426836e-05,
      "loss": 1.5066,
      "num_input_tokens_seen": 16829920,
      "step": 2784,
      "train_runtime": 2825.6601,
      "train_tokens_per_second": 5956.102
    },
    {
      "epoch": 1.0242758620689656,
      "grad_norm": 0.8438899517059326,
      "learning_rate": 9.769924535247561e-05,
      "loss": 1.857,
      "num_input_tokens_seen": 16834944,
      "step": 2785,
      "train_runtime": 2826.5049,
      "train_tokens_per_second": 5956.099
    },
    {
      "epoch": 1.0246436781609196,
      "grad_norm": 0.8158878684043884,
      "learning_rate": 9.766243327811522e-05,
      "loss": 1.4434,
      "num_input_tokens_seen": 16841456,
      "step": 2786,
      "train_runtime": 2827.5602,
      "train_tokens_per_second": 5956.179
    },
    {
      "epoch": 1.0250114942528736,
      "grad_norm": 0.8251202702522278,
      "learning_rate": 9.762562120375483e-05,
      "loss": 1.4535,
      "num_input_tokens_seen": 16847216,
      "step": 2787,
      "train_runtime": 2828.521,
      "train_tokens_per_second": 5956.193
    },
    {
      "epoch": 1.0253793103448277,
      "grad_norm": 0.8166171312332153,
      "learning_rate": 9.758880912939444e-05,
      "loss": 1.4599,
      "num_input_tokens_seen": 16854144,
      "step": 2788,
      "train_runtime": 2829.6483,
      "train_tokens_per_second": 5956.268
    },
    {
      "epoch": 1.0257471264367817,
      "grad_norm": 0.876890242099762,
      "learning_rate": 9.755199705503406e-05,
      "loss": 1.4787,
      "num_input_tokens_seen": 16860768,
      "step": 2789,
      "train_runtime": 2830.7132,
      "train_tokens_per_second": 5956.367
    },
    {
      "epoch": 1.0261149425287357,
      "grad_norm": 0.862595796585083,
      "learning_rate": 9.751518498067367e-05,
      "loss": 1.5332,
      "num_input_tokens_seen": 16866464,
      "step": 2790,
      "train_runtime": 2831.6595,
      "train_tokens_per_second": 5956.388
    },
    {
      "epoch": 1.0264827586206897,
      "grad_norm": 0.8052806854248047,
      "learning_rate": 9.747837290631328e-05,
      "loss": 1.6768,
      "num_input_tokens_seen": 16875152,
      "step": 2791,
      "train_runtime": 2833.528,
      "train_tokens_per_second": 5955.527
    },
    {
      "epoch": 1.0268505747126437,
      "grad_norm": 0.8036222457885742,
      "learning_rate": 9.744156083195288e-05,
      "loss": 1.5281,
      "num_input_tokens_seen": 16880832,
      "step": 2792,
      "train_runtime": 2834.4577,
      "train_tokens_per_second": 5955.577
    },
    {
      "epoch": 1.0272183908045978,
      "grad_norm": 0.861149251461029,
      "learning_rate": 9.740474875759249e-05,
      "loss": 1.4774,
      "num_input_tokens_seen": 16886592,
      "step": 2793,
      "train_runtime": 2835.4058,
      "train_tokens_per_second": 5955.617
    },
    {
      "epoch": 1.0275862068965518,
      "grad_norm": 0.8284410834312439,
      "learning_rate": 9.736793668323211e-05,
      "loss": 1.4271,
      "num_input_tokens_seen": 16892400,
      "step": 2794,
      "train_runtime": 2836.3829,
      "train_tokens_per_second": 5955.613
    },
    {
      "epoch": 1.0279540229885058,
      "grad_norm": 0.9155176281929016,
      "learning_rate": 9.733112460887172e-05,
      "loss": 1.4165,
      "num_input_tokens_seen": 16897248,
      "step": 2795,
      "train_runtime": 2837.2065,
      "train_tokens_per_second": 5955.593
    },
    {
      "epoch": 1.0283218390804598,
      "grad_norm": 0.7975917458534241,
      "learning_rate": 9.729431253451133e-05,
      "loss": 1.5142,
      "num_input_tokens_seen": 16902688,
      "step": 2796,
      "train_runtime": 2838.1301,
      "train_tokens_per_second": 5955.572
    },
    {
      "epoch": 1.0286896551724138,
      "grad_norm": 0.7927860021591187,
      "learning_rate": 9.725750046015094e-05,
      "loss": 1.5657,
      "num_input_tokens_seen": 16907584,
      "step": 2797,
      "train_runtime": 2838.9722,
      "train_tokens_per_second": 5955.53
    },
    {
      "epoch": 1.0290574712643679,
      "grad_norm": 0.7859172821044922,
      "learning_rate": 9.722068838579054e-05,
      "loss": 1.4793,
      "num_input_tokens_seen": 16912688,
      "step": 2798,
      "train_runtime": 2839.8292,
      "train_tokens_per_second": 5955.53
    },
    {
      "epoch": 1.0294252873563219,
      "grad_norm": 0.8089045882225037,
      "learning_rate": 9.718387631143015e-05,
      "loss": 1.5247,
      "num_input_tokens_seen": 16917424,
      "step": 2799,
      "train_runtime": 2840.6313,
      "train_tokens_per_second": 5955.516
    },
    {
      "epoch": 1.029793103448276,
      "grad_norm": 0.8845566511154175,
      "learning_rate": 9.714706423706976e-05,
      "loss": 1.638,
      "num_input_tokens_seen": 16921696,
      "step": 2800,
      "train_runtime": 2841.3995,
      "train_tokens_per_second": 5955.409
    },
    {
      "epoch": 1.03016091954023,
      "grad_norm": 0.8561909198760986,
      "learning_rate": 9.711025216270937e-05,
      "loss": 1.4762,
      "num_input_tokens_seen": 16927200,
      "step": 2801,
      "train_runtime": 2842.3262,
      "train_tokens_per_second": 5955.404
    },
    {
      "epoch": 1.030528735632184,
      "grad_norm": 0.7746861577033997,
      "learning_rate": 9.707344008834899e-05,
      "loss": 1.4286,
      "num_input_tokens_seen": 16932768,
      "step": 2802,
      "train_runtime": 2843.2704,
      "train_tokens_per_second": 5955.384
    },
    {
      "epoch": 1.030896551724138,
      "grad_norm": 0.8497796654701233,
      "learning_rate": 9.70366280139886e-05,
      "loss": 1.5589,
      "num_input_tokens_seen": 16937504,
      "step": 2803,
      "train_runtime": 2844.0978,
      "train_tokens_per_second": 5955.317
    },
    {
      "epoch": 1.031264367816092,
      "grad_norm": 0.8362786173820496,
      "learning_rate": 9.69998159396282e-05,
      "loss": 1.549,
      "num_input_tokens_seen": 16942912,
      "step": 2804,
      "train_runtime": 2844.9903,
      "train_tokens_per_second": 5955.35
    },
    {
      "epoch": 1.031632183908046,
      "grad_norm": 0.7372013926506042,
      "learning_rate": 9.69630038652678e-05,
      "loss": 1.2505,
      "num_input_tokens_seen": 16949232,
      "step": 2805,
      "train_runtime": 2846.034,
      "train_tokens_per_second": 5955.386
    },
    {
      "epoch": 1.032,
      "grad_norm": 0.7864677309989929,
      "learning_rate": 9.692619179090742e-05,
      "loss": 1.6084,
      "num_input_tokens_seen": 16955424,
      "step": 2806,
      "train_runtime": 2847.0722,
      "train_tokens_per_second": 5955.39
    },
    {
      "epoch": 1.032367816091954,
      "grad_norm": 0.7990848422050476,
      "learning_rate": 9.688937971654704e-05,
      "loss": 1.6506,
      "num_input_tokens_seen": 16960784,
      "step": 2807,
      "train_runtime": 2847.9678,
      "train_tokens_per_second": 5955.399
    },
    {
      "epoch": 1.032735632183908,
      "grad_norm": 0.8113677501678467,
      "learning_rate": 9.685256764218665e-05,
      "loss": 1.468,
      "num_input_tokens_seen": 16965648,
      "step": 2808,
      "train_runtime": 2848.7954,
      "train_tokens_per_second": 5955.376
    },
    {
      "epoch": 1.033103448275862,
      "grad_norm": 0.77545166015625,
      "learning_rate": 9.681575556782626e-05,
      "loss": 1.5406,
      "num_input_tokens_seen": 16971888,
      "step": 2809,
      "train_runtime": 2849.8163,
      "train_tokens_per_second": 5955.432
    },
    {
      "epoch": 1.0334712643678161,
      "grad_norm": 0.8232600092887878,
      "learning_rate": 9.677894349346585e-05,
      "loss": 1.5528,
      "num_input_tokens_seen": 16976560,
      "step": 2810,
      "train_runtime": 2850.6087,
      "train_tokens_per_second": 5955.416
    },
    {
      "epoch": 1.0338390804597701,
      "grad_norm": 0.8698176741600037,
      "learning_rate": 9.674213141910547e-05,
      "loss": 1.4138,
      "num_input_tokens_seen": 16984000,
      "step": 2811,
      "train_runtime": 2851.7782,
      "train_tokens_per_second": 5955.582
    },
    {
      "epoch": 1.0342068965517242,
      "grad_norm": 0.8614294528961182,
      "learning_rate": 9.670531934474508e-05,
      "loss": 1.4312,
      "num_input_tokens_seen": 16990272,
      "step": 2812,
      "train_runtime": 2852.8114,
      "train_tokens_per_second": 5955.624
    },
    {
      "epoch": 1.0345747126436782,
      "grad_norm": 0.761712908744812,
      "learning_rate": 9.666850727038469e-05,
      "loss": 1.4403,
      "num_input_tokens_seen": 16996192,
      "step": 2813,
      "train_runtime": 2853.8121,
      "train_tokens_per_second": 5955.61
    },
    {
      "epoch": 1.0349425287356322,
      "grad_norm": 0.8359882235527039,
      "learning_rate": 9.66316951960243e-05,
      "loss": 1.5269,
      "num_input_tokens_seen": 17002048,
      "step": 2814,
      "train_runtime": 2854.7539,
      "train_tokens_per_second": 5955.697
    },
    {
      "epoch": 1.0353103448275862,
      "grad_norm": 0.8140478730201721,
      "learning_rate": 9.659488312166391e-05,
      "loss": 1.6261,
      "num_input_tokens_seen": 17007600,
      "step": 2815,
      "train_runtime": 2855.6544,
      "train_tokens_per_second": 5955.763
    },
    {
      "epoch": 1.0356781609195402,
      "grad_norm": 0.8218773007392883,
      "learning_rate": 9.655807104730352e-05,
      "loss": 1.4684,
      "num_input_tokens_seen": 17012032,
      "step": 2816,
      "train_runtime": 2856.4374,
      "train_tokens_per_second": 5955.682
    },
    {
      "epoch": 1.0360459770114943,
      "grad_norm": 0.8206890821456909,
      "learning_rate": 9.652125897294313e-05,
      "loss": 1.3783,
      "num_input_tokens_seen": 17017472,
      "step": 2817,
      "train_runtime": 2857.3232,
      "train_tokens_per_second": 5955.739
    },
    {
      "epoch": 1.0364137931034483,
      "grad_norm": 0.8729467391967773,
      "learning_rate": 9.648444689858274e-05,
      "loss": 1.4988,
      "num_input_tokens_seen": 17022176,
      "step": 2818,
      "train_runtime": 2858.1348,
      "train_tokens_per_second": 5955.694
    },
    {
      "epoch": 1.0367816091954023,
      "grad_norm": 0.8075752854347229,
      "learning_rate": 9.644763482422236e-05,
      "loss": 1.4539,
      "num_input_tokens_seen": 17028976,
      "step": 2819,
      "train_runtime": 2859.2197,
      "train_tokens_per_second": 5955.812
    },
    {
      "epoch": 1.0371494252873563,
      "grad_norm": 0.8630228042602539,
      "learning_rate": 9.641082274986197e-05,
      "loss": 1.4982,
      "num_input_tokens_seen": 17033536,
      "step": 2820,
      "train_runtime": 2860.0137,
      "train_tokens_per_second": 5955.753
    },
    {
      "epoch": 1.0375172413793103,
      "grad_norm": 0.7630841135978699,
      "learning_rate": 9.637401067550158e-05,
      "loss": 1.4693,
      "num_input_tokens_seen": 17040048,
      "step": 2821,
      "train_runtime": 2861.607,
      "train_tokens_per_second": 5954.713
    },
    {
      "epoch": 1.0378850574712644,
      "grad_norm": 0.7785623073577881,
      "learning_rate": 9.633719860114117e-05,
      "loss": 1.4758,
      "num_input_tokens_seen": 17046880,
      "step": 2822,
      "train_runtime": 2862.6959,
      "train_tokens_per_second": 5954.834
    },
    {
      "epoch": 1.0382528735632184,
      "grad_norm": 0.8043609261512756,
      "learning_rate": 9.630038652678078e-05,
      "loss": 1.3458,
      "num_input_tokens_seen": 17053120,
      "step": 2823,
      "train_runtime": 2863.7178,
      "train_tokens_per_second": 5954.888
    },
    {
      "epoch": 1.0386206896551724,
      "grad_norm": 0.8712300658226013,
      "learning_rate": 9.62635744524204e-05,
      "loss": 1.7833,
      "num_input_tokens_seen": 17057648,
      "step": 2824,
      "train_runtime": 2864.5106,
      "train_tokens_per_second": 5954.821
    },
    {
      "epoch": 1.0389885057471264,
      "grad_norm": 0.8355077505111694,
      "learning_rate": 9.622676237806001e-05,
      "loss": 1.7278,
      "num_input_tokens_seen": 17063216,
      "step": 2825,
      "train_runtime": 2865.4378,
      "train_tokens_per_second": 5954.837
    },
    {
      "epoch": 1.0393563218390804,
      "grad_norm": 0.9312095642089844,
      "learning_rate": 9.618995030369962e-05,
      "loss": 1.5908,
      "num_input_tokens_seen": 17068368,
      "step": 2826,
      "train_runtime": 2866.3264,
      "train_tokens_per_second": 5954.789
    },
    {
      "epoch": 1.0397241379310345,
      "grad_norm": 0.8750326037406921,
      "learning_rate": 9.615313822933923e-05,
      "loss": 1.6228,
      "num_input_tokens_seen": 17073504,
      "step": 2827,
      "train_runtime": 2867.2072,
      "train_tokens_per_second": 5954.751
    },
    {
      "epoch": 1.0400919540229885,
      "grad_norm": 0.8155689239501953,
      "learning_rate": 9.611632615497884e-05,
      "loss": 1.4413,
      "num_input_tokens_seen": 17078464,
      "step": 2828,
      "train_runtime": 2868.0506,
      "train_tokens_per_second": 5954.729
    },
    {
      "epoch": 1.0404597701149425,
      "grad_norm": 0.792161226272583,
      "learning_rate": 9.607951408061845e-05,
      "loss": 1.3346,
      "num_input_tokens_seen": 17090560,
      "step": 2829,
      "train_runtime": 2869.8605,
      "train_tokens_per_second": 5955.189
    },
    {
      "epoch": 1.0408275862068965,
      "grad_norm": 0.849923312664032,
      "learning_rate": 9.604270200625806e-05,
      "loss": 1.4588,
      "num_input_tokens_seen": 17096224,
      "step": 2830,
      "train_runtime": 2870.8349,
      "train_tokens_per_second": 5955.14
    },
    {
      "epoch": 1.0411954022988505,
      "grad_norm": 0.8782142996788025,
      "learning_rate": 9.600588993189766e-05,
      "loss": 1.7419,
      "num_input_tokens_seen": 17101136,
      "step": 2831,
      "train_runtime": 2871.6836,
      "train_tokens_per_second": 5955.091
    },
    {
      "epoch": 1.0415632183908046,
      "grad_norm": 0.8202374577522278,
      "learning_rate": 9.596907785753729e-05,
      "loss": 1.3939,
      "num_input_tokens_seen": 17107552,
      "step": 2832,
      "train_runtime": 2872.706,
      "train_tokens_per_second": 5955.205
    },
    {
      "epoch": 1.0419310344827586,
      "grad_norm": 0.8336490988731384,
      "learning_rate": 9.593226578317688e-05,
      "loss": 1.4311,
      "num_input_tokens_seen": 17114784,
      "step": 2833,
      "train_runtime": 2873.8802,
      "train_tokens_per_second": 5955.288
    },
    {
      "epoch": 1.0422988505747126,
      "grad_norm": 0.8728829026222229,
      "learning_rate": 9.589545370881649e-05,
      "loss": 1.4398,
      "num_input_tokens_seen": 17124224,
      "step": 2834,
      "train_runtime": 2875.3225,
      "train_tokens_per_second": 5955.584
    },
    {
      "epoch": 1.0426666666666666,
      "grad_norm": 0.8830198645591736,
      "learning_rate": 9.58586416344561e-05,
      "loss": 1.5632,
      "num_input_tokens_seen": 17129952,
      "step": 2835,
      "train_runtime": 2876.2527,
      "train_tokens_per_second": 5955.649
    },
    {
      "epoch": 1.0430344827586207,
      "grad_norm": 0.7761485576629639,
      "learning_rate": 9.582182956009571e-05,
      "loss": 1.3605,
      "num_input_tokens_seen": 17136592,
      "step": 2836,
      "train_runtime": 2877.3129,
      "train_tokens_per_second": 5955.762
    },
    {
      "epoch": 1.0434022988505747,
      "grad_norm": 0.8355804681777954,
      "learning_rate": 9.578501748573533e-05,
      "loss": 1.4303,
      "num_input_tokens_seen": 17142304,
      "step": 2837,
      "train_runtime": 2878.2516,
      "train_tokens_per_second": 5955.805
    },
    {
      "epoch": 1.0437701149425287,
      "grad_norm": 0.8488525152206421,
      "learning_rate": 9.574820541137494e-05,
      "loss": 1.3209,
      "num_input_tokens_seen": 17147136,
      "step": 2838,
      "train_runtime": 2879.086,
      "train_tokens_per_second": 5955.757
    },
    {
      "epoch": 1.0441379310344827,
      "grad_norm": 0.8953168988227844,
      "learning_rate": 9.571139333701455e-05,
      "loss": 1.4773,
      "num_input_tokens_seen": 17151696,
      "step": 2839,
      "train_runtime": 2879.9045,
      "train_tokens_per_second": 5955.648
    },
    {
      "epoch": 1.0445057471264367,
      "grad_norm": 0.8682263493537903,
      "learning_rate": 9.567458126265414e-05,
      "loss": 1.4159,
      "num_input_tokens_seen": 17157712,
      "step": 2840,
      "train_runtime": 2880.8785,
      "train_tokens_per_second": 5955.722
    },
    {
      "epoch": 1.0448735632183908,
      "grad_norm": 0.8365000486373901,
      "learning_rate": 9.563776918829377e-05,
      "loss": 1.5616,
      "num_input_tokens_seen": 17163152,
      "step": 2841,
      "train_runtime": 2881.8234,
      "train_tokens_per_second": 5955.657
    },
    {
      "epoch": 1.0452413793103448,
      "grad_norm": 0.8236533403396606,
      "learning_rate": 9.560095711393338e-05,
      "loss": 1.5376,
      "num_input_tokens_seen": 17169200,
      "step": 2842,
      "train_runtime": 2882.8304,
      "train_tokens_per_second": 5955.675
    },
    {
      "epoch": 1.0456091954022988,
      "grad_norm": 0.7440382242202759,
      "learning_rate": 9.556414503957298e-05,
      "loss": 1.3429,
      "num_input_tokens_seen": 17178672,
      "step": 2843,
      "train_runtime": 2884.2655,
      "train_tokens_per_second": 5955.995
    },
    {
      "epoch": 1.0459770114942528,
      "grad_norm": 0.748589038848877,
      "learning_rate": 9.55273329652126e-05,
      "loss": 1.3318,
      "num_input_tokens_seen": 17189872,
      "step": 2844,
      "train_runtime": 2885.9277,
      "train_tokens_per_second": 5956.446
    },
    {
      "epoch": 1.0463448275862068,
      "grad_norm": 0.8263586163520813,
      "learning_rate": 9.54905208908522e-05,
      "loss": 1.6037,
      "num_input_tokens_seen": 17195408,
      "step": 2845,
      "train_runtime": 2886.8414,
      "train_tokens_per_second": 5956.478
    },
    {
      "epoch": 1.0467126436781609,
      "grad_norm": 0.9101609587669373,
      "learning_rate": 9.545370881649181e-05,
      "loss": 1.6731,
      "num_input_tokens_seen": 17201008,
      "step": 2846,
      "train_runtime": 2887.7672,
      "train_tokens_per_second": 5956.508
    },
    {
      "epoch": 1.0470804597701149,
      "grad_norm": 0.70653235912323,
      "learning_rate": 9.541689674213142e-05,
      "loss": 1.3396,
      "num_input_tokens_seen": 17207680,
      "step": 2847,
      "train_runtime": 2888.8462,
      "train_tokens_per_second": 5956.593
    },
    {
      "epoch": 1.047448275862069,
      "grad_norm": 0.8281596899032593,
      "learning_rate": 9.538008466777103e-05,
      "loss": 1.4972,
      "num_input_tokens_seen": 17212960,
      "step": 2848,
      "train_runtime": 2889.7149,
      "train_tokens_per_second": 5956.629
    },
    {
      "epoch": 1.047816091954023,
      "grad_norm": 0.8524518013000488,
      "learning_rate": 9.534327259341065e-05,
      "loss": 1.6742,
      "num_input_tokens_seen": 17218752,
      "step": 2849,
      "train_runtime": 2890.6885,
      "train_tokens_per_second": 5956.627
    },
    {
      "epoch": 1.048183908045977,
      "grad_norm": 0.8830690979957581,
      "learning_rate": 9.530646051905026e-05,
      "loss": 1.584,
      "num_input_tokens_seen": 17226480,
      "step": 2850,
      "train_runtime": 2891.9031,
      "train_tokens_per_second": 5956.797
    },
    {
      "epoch": 1.048551724137931,
      "grad_norm": 0.7999176383018494,
      "learning_rate": 9.526964844468987e-05,
      "loss": 1.5444,
      "num_input_tokens_seen": 17231792,
      "step": 2851,
      "train_runtime": 2893.4185,
      "train_tokens_per_second": 5955.513
    },
    {
      "epoch": 1.048919540229885,
      "grad_norm": 0.8545219898223877,
      "learning_rate": 9.523283637032946e-05,
      "loss": 1.451,
      "num_input_tokens_seen": 17237680,
      "step": 2852,
      "train_runtime": 2894.3843,
      "train_tokens_per_second": 5955.56
    },
    {
      "epoch": 1.049287356321839,
      "grad_norm": 0.8693704009056091,
      "learning_rate": 9.519602429596907e-05,
      "loss": 1.5566,
      "num_input_tokens_seen": 17245776,
      "step": 2853,
      "train_runtime": 2895.661,
      "train_tokens_per_second": 5955.73
    },
    {
      "epoch": 1.049655172413793,
      "grad_norm": 0.8018753528594971,
      "learning_rate": 9.51592122216087e-05,
      "loss": 1.5662,
      "num_input_tokens_seen": 17251968,
      "step": 2854,
      "train_runtime": 2896.6672,
      "train_tokens_per_second": 5955.799
    },
    {
      "epoch": 1.050022988505747,
      "grad_norm": 0.9831026792526245,
      "learning_rate": 9.51224001472483e-05,
      "loss": 1.6686,
      "num_input_tokens_seen": 17257488,
      "step": 2855,
      "train_runtime": 2897.5906,
      "train_tokens_per_second": 5955.806
    },
    {
      "epoch": 1.050390804597701,
      "grad_norm": 0.835200846195221,
      "learning_rate": 9.508558807288791e-05,
      "loss": 1.5858,
      "num_input_tokens_seen": 17267792,
      "step": 2856,
      "train_runtime": 2899.1366,
      "train_tokens_per_second": 5956.184
    },
    {
      "epoch": 1.050758620689655,
      "grad_norm": 0.833871603012085,
      "learning_rate": 9.504877599852752e-05,
      "loss": 1.5692,
      "num_input_tokens_seen": 17273760,
      "step": 2857,
      "train_runtime": 2900.1325,
      "train_tokens_per_second": 5956.197
    },
    {
      "epoch": 1.051126436781609,
      "grad_norm": 0.8666744828224182,
      "learning_rate": 9.501196392416713e-05,
      "loss": 1.623,
      "num_input_tokens_seen": 17279552,
      "step": 2858,
      "train_runtime": 2901.0818,
      "train_tokens_per_second": 5956.244
    },
    {
      "epoch": 1.0514942528735631,
      "grad_norm": 0.8661529421806335,
      "learning_rate": 9.497515184980674e-05,
      "loss": 1.4618,
      "num_input_tokens_seen": 17284240,
      "step": 2859,
      "train_runtime": 2901.8634,
      "train_tokens_per_second": 5956.256
    },
    {
      "epoch": 1.0518620689655171,
      "grad_norm": 0.8447166681289673,
      "learning_rate": 9.493833977544635e-05,
      "loss": 1.6758,
      "num_input_tokens_seen": 17291872,
      "step": 2860,
      "train_runtime": 2903.0783,
      "train_tokens_per_second": 5956.392
    },
    {
      "epoch": 1.0522298850574712,
      "grad_norm": 0.9535942077636719,
      "learning_rate": 9.490152770108596e-05,
      "loss": 1.7953,
      "num_input_tokens_seen": 17296400,
      "step": 2861,
      "train_runtime": 2903.8713,
      "train_tokens_per_second": 5956.324
    },
    {
      "epoch": 1.0525977011494252,
      "grad_norm": 0.8083456158638,
      "learning_rate": 9.486471562672558e-05,
      "loss": 1.4004,
      "num_input_tokens_seen": 17304064,
      "step": 2862,
      "train_runtime": 2905.0823,
      "train_tokens_per_second": 5956.48
    },
    {
      "epoch": 1.0529655172413792,
      "grad_norm": 0.9035588502883911,
      "learning_rate": 9.482790355236518e-05,
      "loss": 1.6874,
      "num_input_tokens_seen": 17310432,
      "step": 2863,
      "train_runtime": 2906.1027,
      "train_tokens_per_second": 5956.58
    },
    {
      "epoch": 1.0533333333333332,
      "grad_norm": 0.8191578388214111,
      "learning_rate": 9.479109147800479e-05,
      "loss": 1.4821,
      "num_input_tokens_seen": 17318768,
      "step": 2864,
      "train_runtime": 2907.3731,
      "train_tokens_per_second": 5956.844
    },
    {
      "epoch": 1.0537011494252873,
      "grad_norm": 0.8769479393959045,
      "learning_rate": 9.47542794036444e-05,
      "loss": 1.6053,
      "num_input_tokens_seen": 17324096,
      "step": 2865,
      "train_runtime": 2908.2537,
      "train_tokens_per_second": 5956.872
    },
    {
      "epoch": 1.0540689655172413,
      "grad_norm": 0.8558223843574524,
      "learning_rate": 9.4717467329284e-05,
      "loss": 1.4024,
      "num_input_tokens_seen": 17329248,
      "step": 2866,
      "train_runtime": 2909.119,
      "train_tokens_per_second": 5956.872
    },
    {
      "epoch": 1.0544367816091953,
      "grad_norm": 0.8265432715415955,
      "learning_rate": 9.468065525492363e-05,
      "loss": 1.2791,
      "num_input_tokens_seen": 17336528,
      "step": 2867,
      "train_runtime": 2910.28,
      "train_tokens_per_second": 5956.997
    },
    {
      "epoch": 1.0548045977011493,
      "grad_norm": 0.8245050311088562,
      "learning_rate": 9.464384318056323e-05,
      "loss": 1.3999,
      "num_input_tokens_seen": 17341328,
      "step": 2868,
      "train_runtime": 2911.1046,
      "train_tokens_per_second": 5956.958
    },
    {
      "epoch": 1.0551724137931036,
      "grad_norm": 0.8065523505210876,
      "learning_rate": 9.460703110620284e-05,
      "loss": 1.434,
      "num_input_tokens_seen": 17346736,
      "step": 2869,
      "train_runtime": 2912.0054,
      "train_tokens_per_second": 5956.972
    },
    {
      "epoch": 1.0555402298850574,
      "grad_norm": 0.7951681613922119,
      "learning_rate": 9.457021903184244e-05,
      "loss": 1.3814,
      "num_input_tokens_seen": 17355136,
      "step": 2870,
      "train_runtime": 2913.2957,
      "train_tokens_per_second": 5957.218
    },
    {
      "epoch": 1.0559080459770116,
      "grad_norm": 0.8782454133033752,
      "learning_rate": 9.453340695748206e-05,
      "loss": 1.6197,
      "num_input_tokens_seen": 17362688,
      "step": 2871,
      "train_runtime": 2914.4974,
      "train_tokens_per_second": 5957.352
    },
    {
      "epoch": 1.0562758620689656,
      "grad_norm": 0.8546891808509827,
      "learning_rate": 9.449659488312167e-05,
      "loss": 1.3147,
      "num_input_tokens_seen": 17369264,
      "step": 2872,
      "train_runtime": 2915.5745,
      "train_tokens_per_second": 5957.407
    },
    {
      "epoch": 1.0566436781609196,
      "grad_norm": 0.917632520198822,
      "learning_rate": 9.445978280876128e-05,
      "loss": 1.5459,
      "num_input_tokens_seen": 17373616,
      "step": 2873,
      "train_runtime": 2916.3462,
      "train_tokens_per_second": 5957.323
    },
    {
      "epoch": 1.0570114942528737,
      "grad_norm": 0.9305850863456726,
      "learning_rate": 9.442297073440089e-05,
      "loss": 1.4332,
      "num_input_tokens_seen": 17380176,
      "step": 2874,
      "train_runtime": 2917.3826,
      "train_tokens_per_second": 5957.455
    },
    {
      "epoch": 1.0573793103448277,
      "grad_norm": 0.7726304531097412,
      "learning_rate": 9.43861586600405e-05,
      "loss": 1.3444,
      "num_input_tokens_seen": 17386288,
      "step": 2875,
      "train_runtime": 2918.3697,
      "train_tokens_per_second": 5957.535
    },
    {
      "epoch": 1.0577471264367817,
      "grad_norm": 0.9169855713844299,
      "learning_rate": 9.43493465856801e-05,
      "loss": 1.5653,
      "num_input_tokens_seen": 17392048,
      "step": 2876,
      "train_runtime": 2919.2869,
      "train_tokens_per_second": 5957.636
    },
    {
      "epoch": 1.0581149425287357,
      "grad_norm": 0.8320531249046326,
      "learning_rate": 9.431253451131971e-05,
      "loss": 1.6393,
      "num_input_tokens_seen": 17397712,
      "step": 2877,
      "train_runtime": 2920.2225,
      "train_tokens_per_second": 5957.667
    },
    {
      "epoch": 1.0584827586206897,
      "grad_norm": 0.9271109104156494,
      "learning_rate": 9.427572243695932e-05,
      "loss": 1.5353,
      "num_input_tokens_seen": 17402736,
      "step": 2878,
      "train_runtime": 2921.0878,
      "train_tokens_per_second": 5957.622
    },
    {
      "epoch": 1.0588505747126438,
      "grad_norm": 0.8818534016609192,
      "learning_rate": 9.423891036259895e-05,
      "loss": 1.5844,
      "num_input_tokens_seen": 17408832,
      "step": 2879,
      "train_runtime": 2922.055,
      "train_tokens_per_second": 5957.736
    },
    {
      "epoch": 1.0592183908045978,
      "grad_norm": 0.8145337104797363,
      "learning_rate": 9.420209828823855e-05,
      "loss": 1.6171,
      "num_input_tokens_seen": 17414528,
      "step": 2880,
      "train_runtime": 2922.991,
      "train_tokens_per_second": 5957.777
    },
    {
      "epoch": 1.0595862068965518,
      "grad_norm": 0.8335985541343689,
      "learning_rate": 9.416528621387816e-05,
      "loss": 1.3167,
      "num_input_tokens_seen": 17423856,
      "step": 2881,
      "train_runtime": 2924.8749,
      "train_tokens_per_second": 5957.129
    },
    {
      "epoch": 1.0599540229885058,
      "grad_norm": 0.9051361083984375,
      "learning_rate": 9.412847413951776e-05,
      "loss": 1.5082,
      "num_input_tokens_seen": 17430576,
      "step": 2882,
      "train_runtime": 2925.9497,
      "train_tokens_per_second": 5957.237
    },
    {
      "epoch": 1.0603218390804598,
      "grad_norm": 0.8853550553321838,
      "learning_rate": 9.409166206515737e-05,
      "loss": 1.6005,
      "num_input_tokens_seen": 17437184,
      "step": 2883,
      "train_runtime": 2927.0275,
      "train_tokens_per_second": 5957.301
    },
    {
      "epoch": 1.0606896551724139,
      "grad_norm": 0.7256721258163452,
      "learning_rate": 9.405484999079699e-05,
      "loss": 1.6536,
      "num_input_tokens_seen": 17445136,
      "step": 2884,
      "train_runtime": 2928.2545,
      "train_tokens_per_second": 5957.52
    },
    {
      "epoch": 1.061057471264368,
      "grad_norm": 0.7858527302742004,
      "learning_rate": 9.40180379164366e-05,
      "loss": 1.3077,
      "num_input_tokens_seen": 17455216,
      "step": 2885,
      "train_runtime": 2929.7813,
      "train_tokens_per_second": 5957.856
    },
    {
      "epoch": 1.061425287356322,
      "grad_norm": 0.829235851764679,
      "learning_rate": 9.398122584207621e-05,
      "loss": 1.5469,
      "num_input_tokens_seen": 17462464,
      "step": 2886,
      "train_runtime": 2930.8979,
      "train_tokens_per_second": 5958.06
    },
    {
      "epoch": 1.061793103448276,
      "grad_norm": 0.9107984900474548,
      "learning_rate": 9.394441376771582e-05,
      "loss": 1.5541,
      "num_input_tokens_seen": 17467936,
      "step": 2887,
      "train_runtime": 2931.7968,
      "train_tokens_per_second": 5958.099
    },
    {
      "epoch": 1.06216091954023,
      "grad_norm": 0.8439177870750427,
      "learning_rate": 9.390760169335543e-05,
      "loss": 1.5968,
      "num_input_tokens_seen": 17473248,
      "step": 2888,
      "train_runtime": 2932.6842,
      "train_tokens_per_second": 5958.108
    },
    {
      "epoch": 1.062528735632184,
      "grad_norm": 0.9255894422531128,
      "learning_rate": 9.387078961899503e-05,
      "loss": 1.5437,
      "num_input_tokens_seen": 17480752,
      "step": 2889,
      "train_runtime": 2933.8748,
      "train_tokens_per_second": 5958.248
    },
    {
      "epoch": 1.062896551724138,
      "grad_norm": 0.9633305072784424,
      "learning_rate": 9.383397754463464e-05,
      "loss": 1.2861,
      "num_input_tokens_seen": 17487552,
      "step": 2890,
      "train_runtime": 2934.961,
      "train_tokens_per_second": 5958.359
    },
    {
      "epoch": 1.063264367816092,
      "grad_norm": 0.9214752316474915,
      "learning_rate": 9.379716547027425e-05,
      "loss": 1.5382,
      "num_input_tokens_seen": 17492384,
      "step": 2891,
      "train_runtime": 2935.7757,
      "train_tokens_per_second": 5958.352
    },
    {
      "epoch": 1.063632183908046,
      "grad_norm": 0.9166814684867859,
      "learning_rate": 9.376035339591387e-05,
      "loss": 1.4873,
      "num_input_tokens_seen": 17498720,
      "step": 2892,
      "train_runtime": 2936.8173,
      "train_tokens_per_second": 5958.396
    },
    {
      "epoch": 1.064,
      "grad_norm": 0.8767185211181641,
      "learning_rate": 9.372354132155347e-05,
      "loss": 1.4913,
      "num_input_tokens_seen": 17503808,
      "step": 2893,
      "train_runtime": 2937.7006,
      "train_tokens_per_second": 5958.336
    },
    {
      "epoch": 1.064367816091954,
      "grad_norm": 0.803009569644928,
      "learning_rate": 9.368672924719308e-05,
      "loss": 1.3586,
      "num_input_tokens_seen": 17508864,
      "step": 2894,
      "train_runtime": 2938.5584,
      "train_tokens_per_second": 5958.317
    },
    {
      "epoch": 1.064735632183908,
      "grad_norm": 0.8554779887199402,
      "learning_rate": 9.364991717283269e-05,
      "loss": 1.4783,
      "num_input_tokens_seen": 17515904,
      "step": 2895,
      "train_runtime": 2939.7021,
      "train_tokens_per_second": 5958.394
    },
    {
      "epoch": 1.0651034482758621,
      "grad_norm": 0.8982095718383789,
      "learning_rate": 9.36131050984723e-05,
      "loss": 1.4686,
      "num_input_tokens_seen": 17522416,
      "step": 2896,
      "train_runtime": 2940.7372,
      "train_tokens_per_second": 5958.511
    },
    {
      "epoch": 1.0654712643678161,
      "grad_norm": 0.9042606353759766,
      "learning_rate": 9.357629302411192e-05,
      "loss": 1.4335,
      "num_input_tokens_seen": 17527904,
      "step": 2897,
      "train_runtime": 2941.6553,
      "train_tokens_per_second": 5958.517
    },
    {
      "epoch": 1.0658390804597702,
      "grad_norm": 0.9136054515838623,
      "learning_rate": 9.353948094975153e-05,
      "loss": 1.4951,
      "num_input_tokens_seen": 17532080,
      "step": 2898,
      "train_runtime": 2942.3862,
      "train_tokens_per_second": 5958.456
    },
    {
      "epoch": 1.0662068965517242,
      "grad_norm": 0.911813497543335,
      "learning_rate": 9.350266887539114e-05,
      "loss": 1.6412,
      "num_input_tokens_seen": 17536800,
      "step": 2899,
      "train_runtime": 2943.1875,
      "train_tokens_per_second": 5958.438
    },
    {
      "epoch": 1.0665747126436782,
      "grad_norm": 0.9023128747940063,
      "learning_rate": 9.346585680103073e-05,
      "loss": 1.6356,
      "num_input_tokens_seen": 17541696,
      "step": 2900,
      "train_runtime": 2943.9948,
      "train_tokens_per_second": 5958.467
    },
    {
      "epoch": 1.0669425287356322,
      "grad_norm": 0.8466926217079163,
      "learning_rate": 9.342904472667035e-05,
      "loss": 1.5312,
      "num_input_tokens_seen": 17546400,
      "step": 2901,
      "train_runtime": 2944.7952,
      "train_tokens_per_second": 5958.445
    },
    {
      "epoch": 1.0673103448275862,
      "grad_norm": 0.9492855072021484,
      "learning_rate": 9.339223265230996e-05,
      "loss": 1.6508,
      "num_input_tokens_seen": 17551696,
      "step": 2902,
      "train_runtime": 2945.6983,
      "train_tokens_per_second": 5958.416
    },
    {
      "epoch": 1.0676781609195403,
      "grad_norm": 0.8984745740890503,
      "learning_rate": 9.335542057794957e-05,
      "loss": 1.7771,
      "num_input_tokens_seen": 17556928,
      "step": 2903,
      "train_runtime": 2946.5852,
      "train_tokens_per_second": 5958.398
    },
    {
      "epoch": 1.0680459770114943,
      "grad_norm": 0.8564352989196777,
      "learning_rate": 9.331860850358918e-05,
      "loss": 1.7645,
      "num_input_tokens_seen": 17562528,
      "step": 2904,
      "train_runtime": 2947.5263,
      "train_tokens_per_second": 5958.396
    },
    {
      "epoch": 1.0684137931034483,
      "grad_norm": 0.8706150650978088,
      "learning_rate": 9.328179642922879e-05,
      "loss": 1.3991,
      "num_input_tokens_seen": 17568528,
      "step": 2905,
      "train_runtime": 2948.5063,
      "train_tokens_per_second": 5958.45
    },
    {
      "epoch": 1.0687816091954023,
      "grad_norm": 0.8660470247268677,
      "learning_rate": 9.32449843548684e-05,
      "loss": 1.5146,
      "num_input_tokens_seen": 17576576,
      "step": 2906,
      "train_runtime": 2949.7956,
      "train_tokens_per_second": 5958.574
    },
    {
      "epoch": 1.0691494252873563,
      "grad_norm": 0.8907226324081421,
      "learning_rate": 9.320817228050801e-05,
      "loss": 1.6753,
      "num_input_tokens_seen": 17581760,
      "step": 2907,
      "train_runtime": 2950.6654,
      "train_tokens_per_second": 5958.575
    },
    {
      "epoch": 1.0695172413793104,
      "grad_norm": 0.852679967880249,
      "learning_rate": 9.317136020614762e-05,
      "loss": 1.6204,
      "num_input_tokens_seen": 17588160,
      "step": 2908,
      "train_runtime": 2951.7023,
      "train_tokens_per_second": 5958.65
    },
    {
      "epoch": 1.0698850574712644,
      "grad_norm": 0.8490344882011414,
      "learning_rate": 9.313454813178724e-05,
      "loss": 1.4982,
      "num_input_tokens_seen": 17593680,
      "step": 2909,
      "train_runtime": 2952.6143,
      "train_tokens_per_second": 5958.679
    },
    {
      "epoch": 1.0702528735632184,
      "grad_norm": 0.7942832708358765,
      "learning_rate": 9.309773605742685e-05,
      "loss": 1.5199,
      "num_input_tokens_seen": 17599296,
      "step": 2910,
      "train_runtime": 2953.5516,
      "train_tokens_per_second": 5958.689
    },
    {
      "epoch": 1.0706206896551724,
      "grad_norm": 0.8734784722328186,
      "learning_rate": 9.306092398306644e-05,
      "loss": 1.6302,
      "num_input_tokens_seen": 17605712,
      "step": 2911,
      "train_runtime": 2955.1575,
      "train_tokens_per_second": 5957.622
    },
    {
      "epoch": 1.0709885057471265,
      "grad_norm": 0.931926429271698,
      "learning_rate": 9.302411190870605e-05,
      "loss": 1.5573,
      "num_input_tokens_seen": 17610128,
      "step": 2912,
      "train_runtime": 2955.9247,
      "train_tokens_per_second": 5957.57
    },
    {
      "epoch": 1.0713563218390805,
      "grad_norm": 0.8316265940666199,
      "learning_rate": 9.298729983434566e-05,
      "loss": 1.5155,
      "num_input_tokens_seen": 17614752,
      "step": 2913,
      "train_runtime": 2956.7346,
      "train_tokens_per_second": 5957.502
    },
    {
      "epoch": 1.0717241379310345,
      "grad_norm": 0.8702672719955444,
      "learning_rate": 9.295048775998528e-05,
      "loss": 1.6388,
      "num_input_tokens_seen": 17620544,
      "step": 2914,
      "train_runtime": 2957.7048,
      "train_tokens_per_second": 5957.506
    },
    {
      "epoch": 1.0720919540229885,
      "grad_norm": 0.787736177444458,
      "learning_rate": 9.291367568562489e-05,
      "loss": 1.5138,
      "num_input_tokens_seen": 17625856,
      "step": 2915,
      "train_runtime": 2958.6085,
      "train_tokens_per_second": 5957.482
    },
    {
      "epoch": 1.0724597701149425,
      "grad_norm": 0.9045817255973816,
      "learning_rate": 9.28768636112645e-05,
      "loss": 1.6954,
      "num_input_tokens_seen": 17630016,
      "step": 2916,
      "train_runtime": 2959.3474,
      "train_tokens_per_second": 5957.4
    },
    {
      "epoch": 1.0728275862068966,
      "grad_norm": 0.870911180973053,
      "learning_rate": 9.284005153690411e-05,
      "loss": 1.5238,
      "num_input_tokens_seen": 17635552,
      "step": 2917,
      "train_runtime": 2960.2592,
      "train_tokens_per_second": 5957.435
    },
    {
      "epoch": 1.0731954022988506,
      "grad_norm": 0.8109344840049744,
      "learning_rate": 9.280323946254372e-05,
      "loss": 1.4472,
      "num_input_tokens_seen": 17642512,
      "step": 2918,
      "train_runtime": 2961.3719,
      "train_tokens_per_second": 5957.547
    },
    {
      "epoch": 1.0735632183908046,
      "grad_norm": 0.7967440485954285,
      "learning_rate": 9.276642738818333e-05,
      "loss": 1.6803,
      "num_input_tokens_seen": 17650800,
      "step": 2919,
      "train_runtime": 2962.6772,
      "train_tokens_per_second": 5957.72
    },
    {
      "epoch": 1.0739310344827586,
      "grad_norm": 0.8106200695037842,
      "learning_rate": 9.272961531382294e-05,
      "loss": 1.7842,
      "num_input_tokens_seen": 17657008,
      "step": 2920,
      "train_runtime": 2963.6979,
      "train_tokens_per_second": 5957.762
    },
    {
      "epoch": 1.0742988505747126,
      "grad_norm": 0.9413151144981384,
      "learning_rate": 9.269280323946255e-05,
      "loss": 1.7005,
      "num_input_tokens_seen": 17662768,
      "step": 2921,
      "train_runtime": 2964.6769,
      "train_tokens_per_second": 5957.738
    },
    {
      "epoch": 1.0746666666666667,
      "grad_norm": 0.7692528963088989,
      "learning_rate": 9.265599116510217e-05,
      "loss": 1.5331,
      "num_input_tokens_seen": 17672560,
      "step": 2922,
      "train_runtime": 2966.1636,
      "train_tokens_per_second": 5958.053
    },
    {
      "epoch": 1.0750344827586207,
      "grad_norm": 0.8859119415283203,
      "learning_rate": 9.261917909074176e-05,
      "loss": 1.5103,
      "num_input_tokens_seen": 17679808,
      "step": 2923,
      "train_runtime": 2967.326,
      "train_tokens_per_second": 5958.162
    },
    {
      "epoch": 1.0754022988505747,
      "grad_norm": 0.7270735502243042,
      "learning_rate": 9.258236701638137e-05,
      "loss": 1.3398,
      "num_input_tokens_seen": 17687920,
      "step": 2924,
      "train_runtime": 2968.5694,
      "train_tokens_per_second": 5958.399
    },
    {
      "epoch": 1.0757701149425287,
      "grad_norm": 0.8331897854804993,
      "learning_rate": 9.254555494202098e-05,
      "loss": 1.5113,
      "num_input_tokens_seen": 17694368,
      "step": 2925,
      "train_runtime": 2969.625,
      "train_tokens_per_second": 5958.452
    },
    {
      "epoch": 1.0761379310344827,
      "grad_norm": 0.8930321931838989,
      "learning_rate": 9.250874286766059e-05,
      "loss": 1.557,
      "num_input_tokens_seen": 17699200,
      "step": 2926,
      "train_runtime": 2970.4514,
      "train_tokens_per_second": 5958.421
    },
    {
      "epoch": 1.0765057471264368,
      "grad_norm": 0.7647495865821838,
      "learning_rate": 9.247193079330021e-05,
      "loss": 1.524,
      "num_input_tokens_seen": 17706496,
      "step": 2927,
      "train_runtime": 2971.6297,
      "train_tokens_per_second": 5958.514
    },
    {
      "epoch": 1.0768735632183908,
      "grad_norm": 0.9093469381332397,
      "learning_rate": 9.243511871893982e-05,
      "loss": 1.7038,
      "num_input_tokens_seen": 17711680,
      "step": 2928,
      "train_runtime": 2972.4818,
      "train_tokens_per_second": 5958.549
    },
    {
      "epoch": 1.0772413793103448,
      "grad_norm": 0.8546137809753418,
      "learning_rate": 9.239830664457943e-05,
      "loss": 1.4754,
      "num_input_tokens_seen": 17718112,
      "step": 2929,
      "train_runtime": 2973.5264,
      "train_tokens_per_second": 5958.619
    },
    {
      "epoch": 1.0776091954022988,
      "grad_norm": 0.8514943718910217,
      "learning_rate": 9.236149457021903e-05,
      "loss": 1.5458,
      "num_input_tokens_seen": 17723344,
      "step": 2930,
      "train_runtime": 2974.4143,
      "train_tokens_per_second": 5958.6
    },
    {
      "epoch": 1.0779770114942528,
      "grad_norm": 0.8585690259933472,
      "learning_rate": 9.232468249585865e-05,
      "loss": 1.6278,
      "num_input_tokens_seen": 17728336,
      "step": 2931,
      "train_runtime": 2975.2664,
      "train_tokens_per_second": 5958.571
    },
    {
      "epoch": 1.0783448275862069,
      "grad_norm": 0.8574824929237366,
      "learning_rate": 9.228787042149826e-05,
      "loss": 1.5328,
      "num_input_tokens_seen": 17732896,
      "step": 2932,
      "train_runtime": 2976.0594,
      "train_tokens_per_second": 5958.516
    },
    {
      "epoch": 1.0787126436781609,
      "grad_norm": 0.8984758257865906,
      "learning_rate": 9.225105834713787e-05,
      "loss": 1.5984,
      "num_input_tokens_seen": 17737872,
      "step": 2933,
      "train_runtime": 2976.9175,
      "train_tokens_per_second": 5958.469
    },
    {
      "epoch": 1.079080459770115,
      "grad_norm": 0.838237464427948,
      "learning_rate": 9.221424627277748e-05,
      "loss": 1.5525,
      "num_input_tokens_seen": 17743424,
      "step": 2934,
      "train_runtime": 2977.8497,
      "train_tokens_per_second": 5958.469
    },
    {
      "epoch": 1.079448275862069,
      "grad_norm": 0.7374897003173828,
      "learning_rate": 9.217743419841708e-05,
      "loss": 1.5435,
      "num_input_tokens_seen": 17755552,
      "step": 2935,
      "train_runtime": 2979.6565,
      "train_tokens_per_second": 5958.926
    },
    {
      "epoch": 1.079816091954023,
      "grad_norm": 0.8529292941093445,
      "learning_rate": 9.214062212405669e-05,
      "loss": 1.7153,
      "num_input_tokens_seen": 17760400,
      "step": 2936,
      "train_runtime": 2980.4792,
      "train_tokens_per_second": 5958.908
    },
    {
      "epoch": 1.080183908045977,
      "grad_norm": 0.883734405040741,
      "learning_rate": 9.21038100496963e-05,
      "loss": 1.5194,
      "num_input_tokens_seen": 17766592,
      "step": 2937,
      "train_runtime": 2981.4836,
      "train_tokens_per_second": 5958.977
    },
    {
      "epoch": 1.080551724137931,
      "grad_norm": 0.8351326584815979,
      "learning_rate": 9.206699797533591e-05,
      "loss": 1.4836,
      "num_input_tokens_seen": 17771872,
      "step": 2938,
      "train_runtime": 2982.3758,
      "train_tokens_per_second": 5958.965
    },
    {
      "epoch": 1.080919540229885,
      "grad_norm": 0.8073239922523499,
      "learning_rate": 9.203018590097553e-05,
      "loss": 1.8301,
      "num_input_tokens_seen": 17780256,
      "step": 2939,
      "train_runtime": 2983.6916,
      "train_tokens_per_second": 5959.147
    },
    {
      "epoch": 1.081287356321839,
      "grad_norm": 0.8248845338821411,
      "learning_rate": 9.199337382661514e-05,
      "loss": 1.3707,
      "num_input_tokens_seen": 17786048,
      "step": 2940,
      "train_runtime": 2984.6693,
      "train_tokens_per_second": 5959.135
    },
    {
      "epoch": 1.081655172413793,
      "grad_norm": 0.8870813846588135,
      "learning_rate": 9.195656175225474e-05,
      "loss": 1.5967,
      "num_input_tokens_seen": 17791904,
      "step": 2941,
      "train_runtime": 2986.1007,
      "train_tokens_per_second": 5958.24
    },
    {
      "epoch": 1.082022988505747,
      "grad_norm": 0.8500445485115051,
      "learning_rate": 9.191974967789435e-05,
      "loss": 1.5112,
      "num_input_tokens_seen": 17798736,
      "step": 2942,
      "train_runtime": 2987.217,
      "train_tokens_per_second": 5958.3
    },
    {
      "epoch": 1.082390804597701,
      "grad_norm": 0.8376680612564087,
      "learning_rate": 9.188293760353396e-05,
      "loss": 1.3358,
      "num_input_tokens_seen": 17803552,
      "step": 2943,
      "train_runtime": 2988.0105,
      "train_tokens_per_second": 5958.33
    },
    {
      "epoch": 1.0827586206896551,
      "grad_norm": 0.8387786746025085,
      "learning_rate": 9.184612552917358e-05,
      "loss": 1.4408,
      "num_input_tokens_seen": 17809504,
      "step": 2944,
      "train_runtime": 2989.0129,
      "train_tokens_per_second": 5958.323
    },
    {
      "epoch": 1.0831264367816091,
      "grad_norm": 0.8747749924659729,
      "learning_rate": 9.180931345481319e-05,
      "loss": 1.393,
      "num_input_tokens_seen": 17817632,
      "step": 2945,
      "train_runtime": 2990.2632,
      "train_tokens_per_second": 5958.55
    },
    {
      "epoch": 1.0834942528735632,
      "grad_norm": 0.8513575792312622,
      "learning_rate": 9.17725013804528e-05,
      "loss": 1.4491,
      "num_input_tokens_seen": 17821616,
      "step": 2946,
      "train_runtime": 2990.9583,
      "train_tokens_per_second": 5958.497
    },
    {
      "epoch": 1.0838620689655172,
      "grad_norm": 0.880194365978241,
      "learning_rate": 9.17356893060924e-05,
      "loss": 1.4038,
      "num_input_tokens_seen": 17827616,
      "step": 2947,
      "train_runtime": 2991.9216,
      "train_tokens_per_second": 5958.584
    },
    {
      "epoch": 1.0842298850574712,
      "grad_norm": 0.8144898414611816,
      "learning_rate": 9.169887723173201e-05,
      "loss": 1.4403,
      "num_input_tokens_seen": 17832624,
      "step": 2948,
      "train_runtime": 2992.7686,
      "train_tokens_per_second": 5958.571
    },
    {
      "epoch": 1.0845977011494252,
      "grad_norm": 0.8111633062362671,
      "learning_rate": 9.166206515737162e-05,
      "loss": 1.5983,
      "num_input_tokens_seen": 17838656,
      "step": 2949,
      "train_runtime": 2993.7544,
      "train_tokens_per_second": 5958.624
    },
    {
      "epoch": 1.0849655172413792,
      "grad_norm": 0.8911537528038025,
      "learning_rate": 9.162525308301123e-05,
      "loss": 1.4741,
      "num_input_tokens_seen": 17843664,
      "step": 2950,
      "train_runtime": 2994.6191,
      "train_tokens_per_second": 5958.576
    },
    {
      "epoch": 1.0853333333333333,
      "grad_norm": 0.8204982876777649,
      "learning_rate": 9.158844100865084e-05,
      "loss": 1.5058,
      "num_input_tokens_seen": 17850000,
      "step": 2951,
      "train_runtime": 2995.6509,
      "train_tokens_per_second": 5958.638
    },
    {
      "epoch": 1.0857011494252873,
      "grad_norm": 0.8104643821716309,
      "learning_rate": 9.155162893429046e-05,
      "loss": 1.3813,
      "num_input_tokens_seen": 17854416,
      "step": 2952,
      "train_runtime": 2996.4252,
      "train_tokens_per_second": 5958.572
    },
    {
      "epoch": 1.0860689655172413,
      "grad_norm": 0.9006640911102295,
      "learning_rate": 9.151481685993006e-05,
      "loss": 1.6387,
      "num_input_tokens_seen": 17858816,
      "step": 2953,
      "train_runtime": 2997.1766,
      "train_tokens_per_second": 5958.546
    },
    {
      "epoch": 1.0864367816091953,
      "grad_norm": 0.9306876063346863,
      "learning_rate": 9.147800478556967e-05,
      "loss": 1.6943,
      "num_input_tokens_seen": 17863440,
      "step": 2954,
      "train_runtime": 2998.0012,
      "train_tokens_per_second": 5958.45
    },
    {
      "epoch": 1.0868045977011493,
      "grad_norm": 0.8869686126708984,
      "learning_rate": 9.144119271120928e-05,
      "loss": 1.4693,
      "num_input_tokens_seen": 17868048,
      "step": 2955,
      "train_runtime": 2998.7885,
      "train_tokens_per_second": 5958.422
    },
    {
      "epoch": 1.0871724137931034,
      "grad_norm": 0.8355336785316467,
      "learning_rate": 9.140438063684888e-05,
      "loss": 1.5594,
      "num_input_tokens_seen": 17874000,
      "step": 2956,
      "train_runtime": 2999.75,
      "train_tokens_per_second": 5958.497
    },
    {
      "epoch": 1.0875402298850574,
      "grad_norm": 0.8136854767799377,
      "learning_rate": 9.136756856248851e-05,
      "loss": 1.5336,
      "num_input_tokens_seen": 17880208,
      "step": 2957,
      "train_runtime": 3000.7445,
      "train_tokens_per_second": 5958.591
    },
    {
      "epoch": 1.0879080459770114,
      "grad_norm": 0.9275166988372803,
      "learning_rate": 9.133075648812812e-05,
      "loss": 1.4834,
      "num_input_tokens_seen": 17885648,
      "step": 2958,
      "train_runtime": 3001.6511,
      "train_tokens_per_second": 5958.603
    },
    {
      "epoch": 1.0882758620689654,
      "grad_norm": 0.8196326494216919,
      "learning_rate": 9.129394441376772e-05,
      "loss": 1.436,
      "num_input_tokens_seen": 17890336,
      "step": 2959,
      "train_runtime": 3002.4504,
      "train_tokens_per_second": 5958.578
    },
    {
      "epoch": 1.0886436781609194,
      "grad_norm": 0.8501957654953003,
      "learning_rate": 9.125713233940732e-05,
      "loss": 1.5597,
      "num_input_tokens_seen": 17895296,
      "step": 2960,
      "train_runtime": 3003.3011,
      "train_tokens_per_second": 5958.542
    },
    {
      "epoch": 1.0890114942528735,
      "grad_norm": 0.8026125431060791,
      "learning_rate": 9.122032026504694e-05,
      "loss": 1.3948,
      "num_input_tokens_seen": 17906000,
      "step": 2961,
      "train_runtime": 3004.8807,
      "train_tokens_per_second": 5958.972
    },
    {
      "epoch": 1.0893793103448275,
      "grad_norm": 0.7974033951759338,
      "learning_rate": 9.118350819068655e-05,
      "loss": 1.4652,
      "num_input_tokens_seen": 17911408,
      "step": 2962,
      "train_runtime": 3005.7712,
      "train_tokens_per_second": 5959.006
    },
    {
      "epoch": 1.0897471264367815,
      "grad_norm": 0.8387789130210876,
      "learning_rate": 9.114669611632616e-05,
      "loss": 1.4444,
      "num_input_tokens_seen": 17918560,
      "step": 2963,
      "train_runtime": 3006.8978,
      "train_tokens_per_second": 5959.152
    },
    {
      "epoch": 1.0901149425287355,
      "grad_norm": 0.985151469707489,
      "learning_rate": 9.110988404196577e-05,
      "loss": 1.5851,
      "num_input_tokens_seen": 17923040,
      "step": 2964,
      "train_runtime": 3007.6736,
      "train_tokens_per_second": 5959.104
    },
    {
      "epoch": 1.0904827586206896,
      "grad_norm": 0.8931572437286377,
      "learning_rate": 9.107307196760538e-05,
      "loss": 1.7873,
      "num_input_tokens_seen": 17928576,
      "step": 2965,
      "train_runtime": 3008.5685,
      "train_tokens_per_second": 5959.172
    },
    {
      "epoch": 1.0908505747126438,
      "grad_norm": 0.9616334438323975,
      "learning_rate": 9.103625989324499e-05,
      "loss": 1.6359,
      "num_input_tokens_seen": 17933616,
      "step": 2966,
      "train_runtime": 3009.4456,
      "train_tokens_per_second": 5959.109
    },
    {
      "epoch": 1.0912183908045976,
      "grad_norm": 0.7934318780899048,
      "learning_rate": 9.09994478188846e-05,
      "loss": 1.3704,
      "num_input_tokens_seen": 17942480,
      "step": 2967,
      "train_runtime": 3010.8455,
      "train_tokens_per_second": 5959.283
    },
    {
      "epoch": 1.0915862068965518,
      "grad_norm": 0.8664103746414185,
      "learning_rate": 9.09626357445242e-05,
      "loss": 1.6176,
      "num_input_tokens_seen": 17948304,
      "step": 2968,
      "train_runtime": 3011.807,
      "train_tokens_per_second": 5959.314
    },
    {
      "epoch": 1.0919540229885056,
      "grad_norm": 0.9133998155593872,
      "learning_rate": 9.092582367016383e-05,
      "loss": 1.4887,
      "num_input_tokens_seen": 17952752,
      "step": 2969,
      "train_runtime": 3012.5809,
      "train_tokens_per_second": 5959.26
    },
    {
      "epoch": 1.0923218390804599,
      "grad_norm": 0.8783140182495117,
      "learning_rate": 9.088901159580344e-05,
      "loss": 1.616,
      "num_input_tokens_seen": 17960112,
      "step": 2970,
      "train_runtime": 3013.7252,
      "train_tokens_per_second": 5959.439
    },
    {
      "epoch": 1.0926896551724137,
      "grad_norm": 0.8942212462425232,
      "learning_rate": 9.085219952144303e-05,
      "loss": 1.4825,
      "num_input_tokens_seen": 17965824,
      "step": 2971,
      "train_runtime": 3015.1425,
      "train_tokens_per_second": 5958.532
    },
    {
      "epoch": 1.093057471264368,
      "grad_norm": 0.7843309640884399,
      "learning_rate": 9.081538744708264e-05,
      "loss": 1.4653,
      "num_input_tokens_seen": 17970208,
      "step": 2972,
      "train_runtime": 3015.9106,
      "train_tokens_per_second": 5958.468
    },
    {
      "epoch": 1.093425287356322,
      "grad_norm": 0.9128292798995972,
      "learning_rate": 9.077857537272225e-05,
      "loss": 1.6371,
      "num_input_tokens_seen": 17976752,
      "step": 2973,
      "train_runtime": 3016.949,
      "train_tokens_per_second": 5958.587
    },
    {
      "epoch": 1.093793103448276,
      "grad_norm": 1.0340237617492676,
      "learning_rate": 9.074176329836187e-05,
      "loss": 1.7085,
      "num_input_tokens_seen": 17980912,
      "step": 2974,
      "train_runtime": 3017.6938,
      "train_tokens_per_second": 5958.494
    },
    {
      "epoch": 1.09416091954023,
      "grad_norm": 0.8621233105659485,
      "learning_rate": 9.070495122400148e-05,
      "loss": 1.4981,
      "num_input_tokens_seen": 17987008,
      "step": 2975,
      "train_runtime": 3018.674,
      "train_tokens_per_second": 5958.579
    },
    {
      "epoch": 1.094528735632184,
      "grad_norm": 0.8423202037811279,
      "learning_rate": 9.066813914964109e-05,
      "loss": 1.5849,
      "num_input_tokens_seen": 17992016,
      "step": 2976,
      "train_runtime": 3019.5218,
      "train_tokens_per_second": 5958.565
    },
    {
      "epoch": 1.094896551724138,
      "grad_norm": 0.7866690754890442,
      "learning_rate": 9.06313270752807e-05,
      "loss": 1.4619,
      "num_input_tokens_seen": 18000224,
      "step": 2977,
      "train_runtime": 3020.7904,
      "train_tokens_per_second": 5958.78
    },
    {
      "epoch": 1.095264367816092,
      "grad_norm": 0.8846787810325623,
      "learning_rate": 9.059451500092031e-05,
      "loss": 1.5359,
      "num_input_tokens_seen": 18005392,
      "step": 2978,
      "train_runtime": 3021.6689,
      "train_tokens_per_second": 5958.757
    },
    {
      "epoch": 1.095632183908046,
      "grad_norm": 0.828133761882782,
      "learning_rate": 9.055770292655992e-05,
      "loss": 1.5276,
      "num_input_tokens_seen": 18010320,
      "step": 2979,
      "train_runtime": 3022.5141,
      "train_tokens_per_second": 5958.722
    },
    {
      "epoch": 1.096,
      "grad_norm": 0.8812045454978943,
      "learning_rate": 9.052089085219952e-05,
      "loss": 1.5074,
      "num_input_tokens_seen": 18017584,
      "step": 2980,
      "train_runtime": 3023.6751,
      "train_tokens_per_second": 5958.836
    },
    {
      "epoch": 1.096367816091954,
      "grad_norm": 0.8940513134002686,
      "learning_rate": 9.048407877783913e-05,
      "loss": 1.5522,
      "num_input_tokens_seen": 18022608,
      "step": 2981,
      "train_runtime": 3024.5424,
      "train_tokens_per_second": 5958.788
    },
    {
      "epoch": 1.0967356321839081,
      "grad_norm": 0.8666152954101562,
      "learning_rate": 9.044726670347876e-05,
      "loss": 1.5509,
      "num_input_tokens_seen": 18028928,
      "step": 2982,
      "train_runtime": 3025.5508,
      "train_tokens_per_second": 5958.891
    },
    {
      "epoch": 1.0971034482758621,
      "grad_norm": 0.8712942004203796,
      "learning_rate": 9.041045462911835e-05,
      "loss": 1.5397,
      "num_input_tokens_seen": 18035504,
      "step": 2983,
      "train_runtime": 3026.588,
      "train_tokens_per_second": 5959.022
    },
    {
      "epoch": 1.0974712643678162,
      "grad_norm": 0.9157382845878601,
      "learning_rate": 9.037364255475796e-05,
      "loss": 1.8303,
      "num_input_tokens_seen": 18040240,
      "step": 2984,
      "train_runtime": 3027.4164,
      "train_tokens_per_second": 5958.956
    },
    {
      "epoch": 1.0978390804597702,
      "grad_norm": 0.82842618227005,
      "learning_rate": 9.033683048039757e-05,
      "loss": 1.3537,
      "num_input_tokens_seen": 18045536,
      "step": 2985,
      "train_runtime": 3028.285,
      "train_tokens_per_second": 5958.995
    },
    {
      "epoch": 1.0982068965517242,
      "grad_norm": 0.9429721832275391,
      "learning_rate": 9.030001840603718e-05,
      "loss": 1.5187,
      "num_input_tokens_seen": 18050976,
      "step": 2986,
      "train_runtime": 3029.195,
      "train_tokens_per_second": 5959.001
    },
    {
      "epoch": 1.0985747126436782,
      "grad_norm": 0.8951818346977234,
      "learning_rate": 9.02632063316768e-05,
      "loss": 1.5894,
      "num_input_tokens_seen": 18055808,
      "step": 2987,
      "train_runtime": 3030.02,
      "train_tokens_per_second": 5958.973
    },
    {
      "epoch": 1.0989425287356323,
      "grad_norm": 0.8807364702224731,
      "learning_rate": 9.022639425731641e-05,
      "loss": 1.4075,
      "num_input_tokens_seen": 18065120,
      "step": 2988,
      "train_runtime": 3031.451,
      "train_tokens_per_second": 5959.232
    },
    {
      "epoch": 1.0993103448275863,
      "grad_norm": 0.7793266773223877,
      "learning_rate": 9.0189582182956e-05,
      "loss": 1.4457,
      "num_input_tokens_seen": 18072064,
      "step": 2989,
      "train_runtime": 3032.5382,
      "train_tokens_per_second": 5959.386
    },
    {
      "epoch": 1.0996781609195403,
      "grad_norm": 0.8729456067085266,
      "learning_rate": 9.015277010859561e-05,
      "loss": 1.252,
      "num_input_tokens_seen": 18078096,
      "step": 2990,
      "train_runtime": 3033.5385,
      "train_tokens_per_second": 5959.409
    },
    {
      "epoch": 1.1000459770114943,
      "grad_norm": 0.910948634147644,
      "learning_rate": 9.011595803423524e-05,
      "loss": 1.7483,
      "num_input_tokens_seen": 18083808,
      "step": 2991,
      "train_runtime": 3034.4897,
      "train_tokens_per_second": 5959.423
    },
    {
      "epoch": 1.1004137931034483,
      "grad_norm": 0.9213285446166992,
      "learning_rate": 9.007914595987485e-05,
      "loss": 1.5399,
      "num_input_tokens_seen": 18089536,
      "step": 2992,
      "train_runtime": 3035.4227,
      "train_tokens_per_second": 5959.478
    },
    {
      "epoch": 1.1007816091954024,
      "grad_norm": 0.9003654718399048,
      "learning_rate": 9.004233388551445e-05,
      "loss": 1.4829,
      "num_input_tokens_seen": 18094192,
      "step": 2993,
      "train_runtime": 3036.24,
      "train_tokens_per_second": 5959.408
    },
    {
      "epoch": 1.1011494252873564,
      "grad_norm": 0.9667969942092896,
      "learning_rate": 9.000552181115406e-05,
      "loss": 1.6512,
      "num_input_tokens_seen": 18098800,
      "step": 2994,
      "train_runtime": 3037.0656,
      "train_tokens_per_second": 5959.305
    },
    {
      "epoch": 1.1015172413793104,
      "grad_norm": 0.9871839880943298,
      "learning_rate": 8.996870973679367e-05,
      "loss": 1.6106,
      "num_input_tokens_seen": 18103152,
      "step": 2995,
      "train_runtime": 3037.8376,
      "train_tokens_per_second": 5959.223
    },
    {
      "epoch": 1.1018850574712644,
      "grad_norm": 0.8530781865119934,
      "learning_rate": 8.993189766243328e-05,
      "loss": 1.5053,
      "num_input_tokens_seen": 18109392,
      "step": 2996,
      "train_runtime": 3038.8737,
      "train_tokens_per_second": 5959.245
    },
    {
      "epoch": 1.1022528735632184,
      "grad_norm": 0.8576104640960693,
      "learning_rate": 8.989508558807289e-05,
      "loss": 1.4783,
      "num_input_tokens_seen": 18116496,
      "step": 2997,
      "train_runtime": 3039.9885,
      "train_tokens_per_second": 5959.396
    },
    {
      "epoch": 1.1026206896551725,
      "grad_norm": 0.8468632698059082,
      "learning_rate": 8.98582735137125e-05,
      "loss": 1.3525,
      "num_input_tokens_seen": 18124240,
      "step": 2998,
      "train_runtime": 3041.2022,
      "train_tokens_per_second": 5959.564
    },
    {
      "epoch": 1.1029885057471265,
      "grad_norm": 0.9965177774429321,
      "learning_rate": 8.982146143935212e-05,
      "loss": 1.8205,
      "num_input_tokens_seen": 18129376,
      "step": 2999,
      "train_runtime": 3042.0615,
      "train_tokens_per_second": 5959.569
    },
    {
      "epoch": 1.1033563218390805,
      "grad_norm": 0.9205605387687683,
      "learning_rate": 8.978464936499173e-05,
      "loss": 1.6064,
      "num_input_tokens_seen": 18135536,
      "step": 3000,
      "train_runtime": 3043.0537,
      "train_tokens_per_second": 5959.65
    },
    {
      "epoch": 1.1037241379310345,
      "grad_norm": 0.8596298098564148,
      "learning_rate": 8.974783729063133e-05,
      "loss": 1.5547,
      "num_input_tokens_seen": 18141376,
      "step": 3001,
      "train_runtime": 3044.6021,
      "train_tokens_per_second": 5958.538
    },
    {
      "epoch": 1.1040919540229885,
      "grad_norm": 0.8226555585861206,
      "learning_rate": 8.971102521627093e-05,
      "loss": 1.4079,
      "num_input_tokens_seen": 18148256,
      "step": 3002,
      "train_runtime": 3045.6737,
      "train_tokens_per_second": 5958.7
    },
    {
      "epoch": 1.1044597701149426,
      "grad_norm": 0.9377200603485107,
      "learning_rate": 8.967421314191054e-05,
      "loss": 1.7866,
      "num_input_tokens_seen": 18153024,
      "step": 3003,
      "train_runtime": 3046.4845,
      "train_tokens_per_second": 5958.679
    },
    {
      "epoch": 1.1048275862068966,
      "grad_norm": 0.8783935904502869,
      "learning_rate": 8.963740106755017e-05,
      "loss": 1.5966,
      "num_input_tokens_seen": 18157280,
      "step": 3004,
      "train_runtime": 3047.2454,
      "train_tokens_per_second": 5958.588
    },
    {
      "epoch": 1.1051954022988506,
      "grad_norm": 0.9163652658462524,
      "learning_rate": 8.960058899318977e-05,
      "loss": 1.5184,
      "num_input_tokens_seen": 18163904,
      "step": 3005,
      "train_runtime": 3048.3181,
      "train_tokens_per_second": 5958.664
    },
    {
      "epoch": 1.1055632183908046,
      "grad_norm": 1.082741618156433,
      "learning_rate": 8.956377691882938e-05,
      "loss": 1.6069,
      "num_input_tokens_seen": 18168864,
      "step": 3006,
      "train_runtime": 3049.1353,
      "train_tokens_per_second": 5958.694
    },
    {
      "epoch": 1.1059310344827586,
      "grad_norm": 0.8877405524253845,
      "learning_rate": 8.952696484446899e-05,
      "loss": 1.5197,
      "num_input_tokens_seen": 18174928,
      "step": 3007,
      "train_runtime": 3050.1246,
      "train_tokens_per_second": 5958.749
    },
    {
      "epoch": 1.1062988505747127,
      "grad_norm": 0.8235839009284973,
      "learning_rate": 8.94901527701086e-05,
      "loss": 1.3557,
      "num_input_tokens_seen": 18180848,
      "step": 3008,
      "train_runtime": 3051.0925,
      "train_tokens_per_second": 5958.799
    },
    {
      "epoch": 1.1066666666666667,
      "grad_norm": 0.8531637191772461,
      "learning_rate": 8.945334069574821e-05,
      "loss": 1.5207,
      "num_input_tokens_seen": 18185808,
      "step": 3009,
      "train_runtime": 3051.9449,
      "train_tokens_per_second": 5958.76
    },
    {
      "epoch": 1.1070344827586207,
      "grad_norm": 0.8883745670318604,
      "learning_rate": 8.941652862138782e-05,
      "loss": 1.5447,
      "num_input_tokens_seen": 18192848,
      "step": 3010,
      "train_runtime": 3053.0583,
      "train_tokens_per_second": 5958.893
    },
    {
      "epoch": 1.1074022988505747,
      "grad_norm": 0.8320760726928711,
      "learning_rate": 8.937971654702743e-05,
      "loss": 1.5523,
      "num_input_tokens_seen": 18198352,
      "step": 3011,
      "train_runtime": 3053.9784,
      "train_tokens_per_second": 5958.9
    },
    {
      "epoch": 1.1077701149425287,
      "grad_norm": 0.9060353636741638,
      "learning_rate": 8.934290447266705e-05,
      "loss": 1.714,
      "num_input_tokens_seen": 18203760,
      "step": 3012,
      "train_runtime": 3054.8674,
      "train_tokens_per_second": 5958.936
    },
    {
      "epoch": 1.1081379310344828,
      "grad_norm": 0.8698820471763611,
      "learning_rate": 8.930609239830665e-05,
      "loss": 1.6191,
      "num_input_tokens_seen": 18208160,
      "step": 3013,
      "train_runtime": 3055.6204,
      "train_tokens_per_second": 5958.908
    },
    {
      "epoch": 1.1085057471264368,
      "grad_norm": 0.8704214096069336,
      "learning_rate": 8.926928032394625e-05,
      "loss": 1.5446,
      "num_input_tokens_seen": 18213296,
      "step": 3014,
      "train_runtime": 3056.5084,
      "train_tokens_per_second": 5958.857
    },
    {
      "epoch": 1.1088735632183908,
      "grad_norm": 0.9200130701065063,
      "learning_rate": 8.923246824958586e-05,
      "loss": 1.7013,
      "num_input_tokens_seen": 18220064,
      "step": 3015,
      "train_runtime": 3057.5694,
      "train_tokens_per_second": 5959.003
    },
    {
      "epoch": 1.1092413793103448,
      "grad_norm": 0.9000295996665955,
      "learning_rate": 8.919565617522547e-05,
      "loss": 1.6061,
      "num_input_tokens_seen": 18225168,
      "step": 3016,
      "train_runtime": 3058.437,
      "train_tokens_per_second": 5958.981
    },
    {
      "epoch": 1.1096091954022989,
      "grad_norm": 0.8831059336662292,
      "learning_rate": 8.91588441008651e-05,
      "loss": 1.5282,
      "num_input_tokens_seen": 18229792,
      "step": 3017,
      "train_runtime": 3059.2271,
      "train_tokens_per_second": 5958.954
    },
    {
      "epoch": 1.1099770114942529,
      "grad_norm": 0.9208916425704956,
      "learning_rate": 8.91220320265047e-05,
      "loss": 1.5004,
      "num_input_tokens_seen": 18235072,
      "step": 3018,
      "train_runtime": 3060.1189,
      "train_tokens_per_second": 5958.942
    },
    {
      "epoch": 1.110344827586207,
      "grad_norm": 0.9301306009292603,
      "learning_rate": 8.90852199521443e-05,
      "loss": 1.5404,
      "num_input_tokens_seen": 18240272,
      "step": 3019,
      "train_runtime": 3060.9837,
      "train_tokens_per_second": 5958.958
    },
    {
      "epoch": 1.110712643678161,
      "grad_norm": 0.9371606707572937,
      "learning_rate": 8.904840787778391e-05,
      "loss": 1.473,
      "num_input_tokens_seen": 18244656,
      "step": 3020,
      "train_runtime": 3061.743,
      "train_tokens_per_second": 5958.912
    },
    {
      "epoch": 1.111080459770115,
      "grad_norm": 0.9736334681510925,
      "learning_rate": 8.901159580342353e-05,
      "loss": 1.7267,
      "num_input_tokens_seen": 18248960,
      "step": 3021,
      "train_runtime": 3062.4934,
      "train_tokens_per_second": 5958.857
    },
    {
      "epoch": 1.111448275862069,
      "grad_norm": 0.9131522178649902,
      "learning_rate": 8.897478372906314e-05,
      "loss": 1.4589,
      "num_input_tokens_seen": 18254336,
      "step": 3022,
      "train_runtime": 3063.3766,
      "train_tokens_per_second": 5958.894
    },
    {
      "epoch": 1.111816091954023,
      "grad_norm": 0.8836458921432495,
      "learning_rate": 8.893797165470275e-05,
      "loss": 1.4484,
      "num_input_tokens_seen": 18260448,
      "step": 3023,
      "train_runtime": 3064.3996,
      "train_tokens_per_second": 5958.899
    },
    {
      "epoch": 1.112183908045977,
      "grad_norm": 0.7740640044212341,
      "learning_rate": 8.890115958034236e-05,
      "loss": 1.2408,
      "num_input_tokens_seen": 18268336,
      "step": 3024,
      "train_runtime": 3065.6354,
      "train_tokens_per_second": 5959.07
    },
    {
      "epoch": 1.112551724137931,
      "grad_norm": 0.9201357364654541,
      "learning_rate": 8.886434750598197e-05,
      "loss": 1.601,
      "num_input_tokens_seen": 18273840,
      "step": 3025,
      "train_runtime": 3066.5541,
      "train_tokens_per_second": 5959.08
    },
    {
      "epoch": 1.112919540229885,
      "grad_norm": 0.9534704089164734,
      "learning_rate": 8.882753543162157e-05,
      "loss": 1.4226,
      "num_input_tokens_seen": 18281136,
      "step": 3026,
      "train_runtime": 3067.7171,
      "train_tokens_per_second": 5959.199
    },
    {
      "epoch": 1.113287356321839,
      "grad_norm": 0.8194193243980408,
      "learning_rate": 8.879072335726118e-05,
      "loss": 1.5194,
      "num_input_tokens_seen": 18286432,
      "step": 3027,
      "train_runtime": 3068.6197,
      "train_tokens_per_second": 5959.172
    },
    {
      "epoch": 1.113655172413793,
      "grad_norm": 0.8776971697807312,
      "learning_rate": 8.875391128290079e-05,
      "loss": 1.4382,
      "num_input_tokens_seen": 18290496,
      "step": 3028,
      "train_runtime": 3069.352,
      "train_tokens_per_second": 5959.074
    },
    {
      "epoch": 1.114022988505747,
      "grad_norm": 0.9225988388061523,
      "learning_rate": 8.871709920854041e-05,
      "loss": 1.5206,
      "num_input_tokens_seen": 18295312,
      "step": 3029,
      "train_runtime": 3070.1649,
      "train_tokens_per_second": 5959.065
    },
    {
      "epoch": 1.1143908045977011,
      "grad_norm": 0.9015832543373108,
      "learning_rate": 8.868028713418002e-05,
      "loss": 1.3392,
      "num_input_tokens_seen": 18301984,
      "step": 3030,
      "train_runtime": 3071.234,
      "train_tokens_per_second": 5959.163
    },
    {
      "epoch": 1.1147586206896551,
      "grad_norm": 0.8528132438659668,
      "learning_rate": 8.864347505981962e-05,
      "loss": 1.4663,
      "num_input_tokens_seen": 18308896,
      "step": 3031,
      "train_runtime": 3072.8792,
      "train_tokens_per_second": 5958.222
    },
    {
      "epoch": 1.1151264367816092,
      "grad_norm": 0.917464017868042,
      "learning_rate": 8.860666298545923e-05,
      "loss": 1.6639,
      "num_input_tokens_seen": 18313856,
      "step": 3032,
      "train_runtime": 3073.7292,
      "train_tokens_per_second": 5958.188
    },
    {
      "epoch": 1.1154942528735632,
      "grad_norm": 0.8733834624290466,
      "learning_rate": 8.856985091109884e-05,
      "loss": 1.358,
      "num_input_tokens_seen": 18321152,
      "step": 3033,
      "train_runtime": 3074.8862,
      "train_tokens_per_second": 5958.319
    },
    {
      "epoch": 1.1158620689655172,
      "grad_norm": 0.9368665218353271,
      "learning_rate": 8.853303883673846e-05,
      "loss": 1.6856,
      "num_input_tokens_seen": 18327664,
      "step": 3034,
      "train_runtime": 3075.9424,
      "train_tokens_per_second": 5958.39
    },
    {
      "epoch": 1.1162298850574712,
      "grad_norm": 0.9064751267433167,
      "learning_rate": 8.849622676237807e-05,
      "loss": 1.5717,
      "num_input_tokens_seen": 18332560,
      "step": 3035,
      "train_runtime": 3076.7861,
      "train_tokens_per_second": 5958.347
    },
    {
      "epoch": 1.1165977011494252,
      "grad_norm": 0.967047393321991,
      "learning_rate": 8.845941468801768e-05,
      "loss": 1.6726,
      "num_input_tokens_seen": 18337856,
      "step": 3036,
      "train_runtime": 3077.68,
      "train_tokens_per_second": 5958.337
    },
    {
      "epoch": 1.1169655172413793,
      "grad_norm": 0.8143070340156555,
      "learning_rate": 8.842260261365729e-05,
      "loss": 1.6144,
      "num_input_tokens_seen": 18346224,
      "step": 3037,
      "train_runtime": 3078.9906,
      "train_tokens_per_second": 5958.519
    },
    {
      "epoch": 1.1173333333333333,
      "grad_norm": 0.8116313815116882,
      "learning_rate": 8.83857905392969e-05,
      "loss": 1.3194,
      "num_input_tokens_seen": 18352256,
      "step": 3038,
      "train_runtime": 3079.9892,
      "train_tokens_per_second": 5958.546
    },
    {
      "epoch": 1.1177011494252873,
      "grad_norm": 0.9737692475318909,
      "learning_rate": 8.83489784649365e-05,
      "loss": 1.5578,
      "num_input_tokens_seen": 18358192,
      "step": 3039,
      "train_runtime": 3080.9501,
      "train_tokens_per_second": 5958.614
    },
    {
      "epoch": 1.1180689655172413,
      "grad_norm": 0.8553847670555115,
      "learning_rate": 8.831216639057611e-05,
      "loss": 1.6262,
      "num_input_tokens_seen": 18363520,
      "step": 3040,
      "train_runtime": 3081.8439,
      "train_tokens_per_second": 5958.615
    },
    {
      "epoch": 1.1184367816091954,
      "grad_norm": 0.9897530674934387,
      "learning_rate": 8.827535431621572e-05,
      "loss": 1.6128,
      "num_input_tokens_seen": 18369920,
      "step": 3041,
      "train_runtime": 3082.9058,
      "train_tokens_per_second": 5958.638
    },
    {
      "epoch": 1.1188045977011494,
      "grad_norm": 0.8073896765708923,
      "learning_rate": 8.823854224185534e-05,
      "loss": 1.5645,
      "num_input_tokens_seen": 18376160,
      "step": 3042,
      "train_runtime": 3083.9242,
      "train_tokens_per_second": 5958.694
    },
    {
      "epoch": 1.1191724137931034,
      "grad_norm": 0.8492890000343323,
      "learning_rate": 8.820173016749494e-05,
      "loss": 1.5328,
      "num_input_tokens_seen": 18382176,
      "step": 3043,
      "train_runtime": 3084.9289,
      "train_tokens_per_second": 5958.703
    },
    {
      "epoch": 1.1195402298850574,
      "grad_norm": 0.7779625654220581,
      "learning_rate": 8.816491809313455e-05,
      "loss": 1.3965,
      "num_input_tokens_seen": 18390832,
      "step": 3044,
      "train_runtime": 3086.228,
      "train_tokens_per_second": 5959.0
    },
    {
      "epoch": 1.1199080459770114,
      "grad_norm": 1.0160466432571411,
      "learning_rate": 8.812810601877416e-05,
      "loss": 1.6625,
      "num_input_tokens_seen": 18396944,
      "step": 3045,
      "train_runtime": 3087.2136,
      "train_tokens_per_second": 5959.077
    },
    {
      "epoch": 1.1202758620689655,
      "grad_norm": 0.9516376852989197,
      "learning_rate": 8.809129394441378e-05,
      "loss": 1.7415,
      "num_input_tokens_seen": 18402048,
      "step": 3046,
      "train_runtime": 3088.0721,
      "train_tokens_per_second": 5959.073
    },
    {
      "epoch": 1.1206436781609195,
      "grad_norm": 0.9396638870239258,
      "learning_rate": 8.805448187005339e-05,
      "loss": 1.5369,
      "num_input_tokens_seen": 18406912,
      "step": 3047,
      "train_runtime": 3088.9072,
      "train_tokens_per_second": 5959.037
    },
    {
      "epoch": 1.1210114942528735,
      "grad_norm": 0.8862414956092834,
      "learning_rate": 8.8017669795693e-05,
      "loss": 1.5418,
      "num_input_tokens_seen": 18413600,
      "step": 3048,
      "train_runtime": 3089.9803,
      "train_tokens_per_second": 5959.132
    },
    {
      "epoch": 1.1213793103448275,
      "grad_norm": 0.9260058999061584,
      "learning_rate": 8.798085772133259e-05,
      "loss": 1.3787,
      "num_input_tokens_seen": 18420528,
      "step": 3049,
      "train_runtime": 3091.0747,
      "train_tokens_per_second": 5959.263
    },
    {
      "epoch": 1.1217471264367815,
      "grad_norm": 0.985962450504303,
      "learning_rate": 8.79440456469722e-05,
      "loss": 1.4664,
      "num_input_tokens_seen": 18425632,
      "step": 3050,
      "train_runtime": 3091.9215,
      "train_tokens_per_second": 5959.282
    },
    {
      "epoch": 1.1221149425287356,
      "grad_norm": 0.8389951586723328,
      "learning_rate": 8.790723357261182e-05,
      "loss": 1.5398,
      "num_input_tokens_seen": 18431968,
      "step": 3051,
      "train_runtime": 3092.9666,
      "train_tokens_per_second": 5959.317
    },
    {
      "epoch": 1.1224827586206896,
      "grad_norm": 0.9752079844474792,
      "learning_rate": 8.787042149825143e-05,
      "loss": 1.468,
      "num_input_tokens_seen": 18437408,
      "step": 3052,
      "train_runtime": 3093.8553,
      "train_tokens_per_second": 5959.363
    },
    {
      "epoch": 1.1228505747126436,
      "grad_norm": 0.8355587720870972,
      "learning_rate": 8.783360942389104e-05,
      "loss": 1.4026,
      "num_input_tokens_seen": 18441968,
      "step": 3053,
      "train_runtime": 3094.6498,
      "train_tokens_per_second": 5959.307
    },
    {
      "epoch": 1.1232183908045976,
      "grad_norm": 0.8331617116928101,
      "learning_rate": 8.779679734953065e-05,
      "loss": 1.3438,
      "num_input_tokens_seen": 18448224,
      "step": 3054,
      "train_runtime": 3095.6563,
      "train_tokens_per_second": 5959.39
    },
    {
      "epoch": 1.1235862068965516,
      "grad_norm": 0.8518046736717224,
      "learning_rate": 8.775998527517026e-05,
      "loss": 1.5592,
      "num_input_tokens_seen": 18453248,
      "step": 3055,
      "train_runtime": 3096.5024,
      "train_tokens_per_second": 5959.384
    },
    {
      "epoch": 1.1239540229885057,
      "grad_norm": 0.8022720217704773,
      "learning_rate": 8.772317320080987e-05,
      "loss": 1.502,
      "num_input_tokens_seen": 18460208,
      "step": 3056,
      "train_runtime": 3097.6209,
      "train_tokens_per_second": 5959.479
    },
    {
      "epoch": 1.1243218390804597,
      "grad_norm": 0.9089572429656982,
      "learning_rate": 8.768636112644948e-05,
      "loss": 1.5555,
      "num_input_tokens_seen": 18466528,
      "step": 3057,
      "train_runtime": 3098.6671,
      "train_tokens_per_second": 5959.507
    },
    {
      "epoch": 1.1246896551724137,
      "grad_norm": 0.9539327025413513,
      "learning_rate": 8.764954905208909e-05,
      "loss": 1.5741,
      "num_input_tokens_seen": 18471792,
      "step": 3058,
      "train_runtime": 3099.5244,
      "train_tokens_per_second": 5959.557
    },
    {
      "epoch": 1.1250574712643677,
      "grad_norm": 0.8519731163978577,
      "learning_rate": 8.761273697772871e-05,
      "loss": 1.4594,
      "num_input_tokens_seen": 18477232,
      "step": 3059,
      "train_runtime": 3100.4002,
      "train_tokens_per_second": 5959.628
    },
    {
      "epoch": 1.1254252873563217,
      "grad_norm": 0.8022516369819641,
      "learning_rate": 8.757592490336832e-05,
      "loss": 1.3109,
      "num_input_tokens_seen": 18483440,
      "step": 3060,
      "train_runtime": 3101.4069,
      "train_tokens_per_second": 5959.695
    },
    {
      "epoch": 1.1257931034482758,
      "grad_norm": 0.9718914031982422,
      "learning_rate": 8.753911282900791e-05,
      "loss": 1.5463,
      "num_input_tokens_seen": 18488432,
      "step": 3061,
      "train_runtime": 3102.7425,
      "train_tokens_per_second": 5958.739
    },
    {
      "epoch": 1.1261609195402298,
      "grad_norm": 0.8494237065315247,
      "learning_rate": 8.750230075464752e-05,
      "loss": 1.3881,
      "num_input_tokens_seen": 18495840,
      "step": 3062,
      "train_runtime": 3103.8886,
      "train_tokens_per_second": 5958.925
    },
    {
      "epoch": 1.126528735632184,
      "grad_norm": 0.9133712649345398,
      "learning_rate": 8.746548868028713e-05,
      "loss": 1.6853,
      "num_input_tokens_seen": 18500224,
      "step": 3063,
      "train_runtime": 3104.6616,
      "train_tokens_per_second": 5958.854
    },
    {
      "epoch": 1.1268965517241378,
      "grad_norm": 0.8555759191513062,
      "learning_rate": 8.742867660592675e-05,
      "loss": 1.4305,
      "num_input_tokens_seen": 18510464,
      "step": 3064,
      "train_runtime": 3106.2024,
      "train_tokens_per_second": 5959.194
    },
    {
      "epoch": 1.127264367816092,
      "grad_norm": 0.855987548828125,
      "learning_rate": 8.739186453156636e-05,
      "loss": 1.4913,
      "num_input_tokens_seen": 18517760,
      "step": 3065,
      "train_runtime": 3107.3419,
      "train_tokens_per_second": 5959.357
    },
    {
      "epoch": 1.1276321839080459,
      "grad_norm": 0.8938915729522705,
      "learning_rate": 8.735505245720597e-05,
      "loss": 1.5961,
      "num_input_tokens_seen": 18523232,
      "step": 3066,
      "train_runtime": 3108.2338,
      "train_tokens_per_second": 5959.407
    },
    {
      "epoch": 1.1280000000000001,
      "grad_norm": 0.9515122771263123,
      "learning_rate": 8.731824038284557e-05,
      "loss": 1.5187,
      "num_input_tokens_seen": 18528368,
      "step": 3067,
      "train_runtime": 3109.1039,
      "train_tokens_per_second": 5959.392
    },
    {
      "epoch": 1.128367816091954,
      "grad_norm": 0.8911998867988586,
      "learning_rate": 8.728142830848519e-05,
      "loss": 1.4074,
      "num_input_tokens_seen": 18535184,
      "step": 3068,
      "train_runtime": 3110.2233,
      "train_tokens_per_second": 5959.438
    },
    {
      "epoch": 1.1287356321839082,
      "grad_norm": 0.7230700850486755,
      "learning_rate": 8.72446162341248e-05,
      "loss": 1.4812,
      "num_input_tokens_seen": 18543296,
      "step": 3069,
      "train_runtime": 3111.4991,
      "train_tokens_per_second": 5959.602
    },
    {
      "epoch": 1.129103448275862,
      "grad_norm": 0.8507945537567139,
      "learning_rate": 8.72078041597644e-05,
      "loss": 1.5413,
      "num_input_tokens_seen": 18549344,
      "step": 3070,
      "train_runtime": 3112.4812,
      "train_tokens_per_second": 5959.665
    },
    {
      "epoch": 1.1294712643678162,
      "grad_norm": 0.9214749932289124,
      "learning_rate": 8.717099208540402e-05,
      "loss": 1.6366,
      "num_input_tokens_seen": 18557104,
      "step": 3071,
      "train_runtime": 3113.7089,
      "train_tokens_per_second": 5959.807
    },
    {
      "epoch": 1.12983908045977,
      "grad_norm": 0.859377920627594,
      "learning_rate": 8.713418001104364e-05,
      "loss": 1.4259,
      "num_input_tokens_seen": 18565088,
      "step": 3072,
      "train_runtime": 3114.9288,
      "train_tokens_per_second": 5960.036
    },
    {
      "epoch": 1.1302068965517242,
      "grad_norm": 0.8506039977073669,
      "learning_rate": 8.709736793668323e-05,
      "loss": 1.3779,
      "num_input_tokens_seen": 18571024,
      "step": 3073,
      "train_runtime": 3115.9378,
      "train_tokens_per_second": 5960.011
    },
    {
      "epoch": 1.130574712643678,
      "grad_norm": 0.8265760540962219,
      "learning_rate": 8.706055586232284e-05,
      "loss": 1.4175,
      "num_input_tokens_seen": 18578112,
      "step": 3074,
      "train_runtime": 3117.0484,
      "train_tokens_per_second": 5960.161
    },
    {
      "epoch": 1.1309425287356323,
      "grad_norm": 0.8889214992523193,
      "learning_rate": 8.702374378796245e-05,
      "loss": 1.6282,
      "num_input_tokens_seen": 18582896,
      "step": 3075,
      "train_runtime": 3117.8764,
      "train_tokens_per_second": 5960.113
    },
    {
      "epoch": 1.1313103448275863,
      "grad_norm": 0.9217056632041931,
      "learning_rate": 8.698693171360207e-05,
      "loss": 1.6716,
      "num_input_tokens_seen": 18587984,
      "step": 3076,
      "train_runtime": 3118.7364,
      "train_tokens_per_second": 5960.101
    },
    {
      "epoch": 1.1316781609195403,
      "grad_norm": 0.8785937428474426,
      "learning_rate": 8.695011963924168e-05,
      "loss": 1.5235,
      "num_input_tokens_seen": 18592608,
      "step": 3077,
      "train_runtime": 3119.5382,
      "train_tokens_per_second": 5960.051
    },
    {
      "epoch": 1.1320459770114943,
      "grad_norm": 1.0107359886169434,
      "learning_rate": 8.691330756488129e-05,
      "loss": 1.6961,
      "num_input_tokens_seen": 18598448,
      "step": 3078,
      "train_runtime": 3120.4902,
      "train_tokens_per_second": 5960.105
    },
    {
      "epoch": 1.1324137931034484,
      "grad_norm": 0.879207193851471,
      "learning_rate": 8.687649549052089e-05,
      "loss": 1.52,
      "num_input_tokens_seen": 18604912,
      "step": 3079,
      "train_runtime": 3121.5146,
      "train_tokens_per_second": 5960.219
    },
    {
      "epoch": 1.1327816091954024,
      "grad_norm": 0.8674408197402954,
      "learning_rate": 8.68396834161605e-05,
      "loss": 1.7051,
      "num_input_tokens_seen": 18610064,
      "step": 3080,
      "train_runtime": 3122.3931,
      "train_tokens_per_second": 5960.193
    },
    {
      "epoch": 1.1331494252873564,
      "grad_norm": 0.8419400453567505,
      "learning_rate": 8.680287134180012e-05,
      "loss": 1.3253,
      "num_input_tokens_seen": 18617360,
      "step": 3081,
      "train_runtime": 3123.5533,
      "train_tokens_per_second": 5960.314
    },
    {
      "epoch": 1.1335172413793104,
      "grad_norm": 0.8395043611526489,
      "learning_rate": 8.676605926743973e-05,
      "loss": 1.6573,
      "num_input_tokens_seen": 18622208,
      "step": 3082,
      "train_runtime": 3124.361,
      "train_tokens_per_second": 5960.325
    },
    {
      "epoch": 1.1338850574712644,
      "grad_norm": 0.9543777704238892,
      "learning_rate": 8.672924719307934e-05,
      "loss": 1.5714,
      "num_input_tokens_seen": 18626672,
      "step": 3083,
      "train_runtime": 3125.1369,
      "train_tokens_per_second": 5960.274
    },
    {
      "epoch": 1.1342528735632185,
      "grad_norm": 0.9389495849609375,
      "learning_rate": 8.669243511871894e-05,
      "loss": 1.5084,
      "num_input_tokens_seen": 18632048,
      "step": 3084,
      "train_runtime": 3126.0345,
      "train_tokens_per_second": 5960.282
    },
    {
      "epoch": 1.1346206896551725,
      "grad_norm": 0.9204221367835999,
      "learning_rate": 8.665562304435855e-05,
      "loss": 1.5885,
      "num_input_tokens_seen": 18637168,
      "step": 3085,
      "train_runtime": 3126.9033,
      "train_tokens_per_second": 5960.264
    },
    {
      "epoch": 1.1349885057471265,
      "grad_norm": 0.8767832517623901,
      "learning_rate": 8.661881096999816e-05,
      "loss": 1.5201,
      "num_input_tokens_seen": 18642944,
      "step": 3086,
      "train_runtime": 3127.852,
      "train_tokens_per_second": 5960.302
    },
    {
      "epoch": 1.1353563218390805,
      "grad_norm": 0.8895378708839417,
      "learning_rate": 8.658199889563777e-05,
      "loss": 1.4989,
      "num_input_tokens_seen": 18649344,
      "step": 3087,
      "train_runtime": 3128.9034,
      "train_tokens_per_second": 5960.345
    },
    {
      "epoch": 1.1357241379310345,
      "grad_norm": 0.9417483806610107,
      "learning_rate": 8.654518682127738e-05,
      "loss": 1.6446,
      "num_input_tokens_seen": 18654544,
      "step": 3088,
      "train_runtime": 3129.7597,
      "train_tokens_per_second": 5960.376
    },
    {
      "epoch": 1.1360919540229886,
      "grad_norm": 0.8545156121253967,
      "learning_rate": 8.6508374746917e-05,
      "loss": 1.563,
      "num_input_tokens_seen": 18661552,
      "step": 3089,
      "train_runtime": 3130.8734,
      "train_tokens_per_second": 5960.494
    },
    {
      "epoch": 1.1364597701149426,
      "grad_norm": 0.8982740044593811,
      "learning_rate": 8.647156267255661e-05,
      "loss": 1.5316,
      "num_input_tokens_seen": 18667072,
      "step": 3090,
      "train_runtime": 3131.7973,
      "train_tokens_per_second": 5960.498
    },
    {
      "epoch": 1.1368275862068966,
      "grad_norm": 0.8665157556533813,
      "learning_rate": 8.643475059819621e-05,
      "loss": 1.4421,
      "num_input_tokens_seen": 18671808,
      "step": 3091,
      "train_runtime": 3133.1126,
      "train_tokens_per_second": 5959.508
    },
    {
      "epoch": 1.1371954022988506,
      "grad_norm": 0.8474888205528259,
      "learning_rate": 8.639793852383582e-05,
      "loss": 1.4779,
      "num_input_tokens_seen": 18676992,
      "step": 3092,
      "train_runtime": 3133.9834,
      "train_tokens_per_second": 5959.506
    },
    {
      "epoch": 1.1375632183908047,
      "grad_norm": 0.8338071703910828,
      "learning_rate": 8.636112644947542e-05,
      "loss": 1.3291,
      "num_input_tokens_seen": 18681648,
      "step": 3093,
      "train_runtime": 3134.7528,
      "train_tokens_per_second": 5959.528
    },
    {
      "epoch": 1.1379310344827587,
      "grad_norm": 0.8999766111373901,
      "learning_rate": 8.632431437511505e-05,
      "loss": 1.6534,
      "num_input_tokens_seen": 18688048,
      "step": 3094,
      "train_runtime": 3135.7675,
      "train_tokens_per_second": 5959.641
    },
    {
      "epoch": 1.1382988505747127,
      "grad_norm": 0.883551836013794,
      "learning_rate": 8.628750230075466e-05,
      "loss": 1.6511,
      "num_input_tokens_seen": 18694448,
      "step": 3095,
      "train_runtime": 3136.7852,
      "train_tokens_per_second": 5959.748
    },
    {
      "epoch": 1.1386666666666667,
      "grad_norm": 0.811758279800415,
      "learning_rate": 8.625069022639426e-05,
      "loss": 1.4676,
      "num_input_tokens_seen": 18701344,
      "step": 3096,
      "train_runtime": 3137.8786,
      "train_tokens_per_second": 5959.868
    },
    {
      "epoch": 1.1390344827586207,
      "grad_norm": 0.9775203466415405,
      "learning_rate": 8.621387815203386e-05,
      "loss": 1.7071,
      "num_input_tokens_seen": 18706864,
      "step": 3097,
      "train_runtime": 3138.8017,
      "train_tokens_per_second": 5959.874
    },
    {
      "epoch": 1.1394022988505748,
      "grad_norm": 0.8970943689346313,
      "learning_rate": 8.617706607767348e-05,
      "loss": 1.6993,
      "num_input_tokens_seen": 18711936,
      "step": 3098,
      "train_runtime": 3139.6758,
      "train_tokens_per_second": 5959.831
    },
    {
      "epoch": 1.1397701149425288,
      "grad_norm": 0.7886869311332703,
      "learning_rate": 8.614025400331309e-05,
      "loss": 1.4806,
      "num_input_tokens_seen": 18717632,
      "step": 3099,
      "train_runtime": 3140.638,
      "train_tokens_per_second": 5959.818
    },
    {
      "epoch": 1.1401379310344828,
      "grad_norm": 0.8983283638954163,
      "learning_rate": 8.61034419289527e-05,
      "loss": 1.5685,
      "num_input_tokens_seen": 18726480,
      "step": 3100,
      "train_runtime": 3142.012,
      "train_tokens_per_second": 5960.028
    },
    {
      "epoch": 1.1405057471264368,
      "grad_norm": 0.8518079519271851,
      "learning_rate": 8.606662985459231e-05,
      "loss": 1.7461,
      "num_input_tokens_seen": 18731840,
      "step": 3101,
      "train_runtime": 3142.9133,
      "train_tokens_per_second": 5960.024
    },
    {
      "epoch": 1.1408735632183908,
      "grad_norm": 0.8501981496810913,
      "learning_rate": 8.602981778023192e-05,
      "loss": 1.5845,
      "num_input_tokens_seen": 18739376,
      "step": 3102,
      "train_runtime": 3144.1044,
      "train_tokens_per_second": 5960.163
    },
    {
      "epoch": 1.1412413793103449,
      "grad_norm": 0.9013225436210632,
      "learning_rate": 8.599300570587153e-05,
      "loss": 1.5989,
      "num_input_tokens_seen": 18745408,
      "step": 3103,
      "train_runtime": 3145.0857,
      "train_tokens_per_second": 5960.222
    },
    {
      "epoch": 1.1416091954022989,
      "grad_norm": 0.8777236938476562,
      "learning_rate": 8.595619363151114e-05,
      "loss": 1.4751,
      "num_input_tokens_seen": 18755712,
      "step": 3104,
      "train_runtime": 3146.634,
      "train_tokens_per_second": 5960.563
    },
    {
      "epoch": 1.141977011494253,
      "grad_norm": 0.823156476020813,
      "learning_rate": 8.591938155715074e-05,
      "loss": 1.3572,
      "num_input_tokens_seen": 18762976,
      "step": 3105,
      "train_runtime": 3147.7945,
      "train_tokens_per_second": 5960.674
    },
    {
      "epoch": 1.142344827586207,
      "grad_norm": 0.8540163636207581,
      "learning_rate": 8.588256948279037e-05,
      "loss": 1.5428,
      "num_input_tokens_seen": 18768480,
      "step": 3106,
      "train_runtime": 3148.703,
      "train_tokens_per_second": 5960.702
    },
    {
      "epoch": 1.142712643678161,
      "grad_norm": 0.8515554070472717,
      "learning_rate": 8.584575740842998e-05,
      "loss": 1.443,
      "num_input_tokens_seen": 18775968,
      "step": 3107,
      "train_runtime": 3149.8812,
      "train_tokens_per_second": 5960.85
    },
    {
      "epoch": 1.143080459770115,
      "grad_norm": 0.8569208383560181,
      "learning_rate": 8.580894533406959e-05,
      "loss": 1.2434,
      "num_input_tokens_seen": 18783760,
      "step": 3108,
      "train_runtime": 3151.0979,
      "train_tokens_per_second": 5961.021
    },
    {
      "epoch": 1.143448275862069,
      "grad_norm": 0.9443128108978271,
      "learning_rate": 8.577213325970918e-05,
      "loss": 1.726,
      "num_input_tokens_seen": 18788480,
      "step": 3109,
      "train_runtime": 3151.8997,
      "train_tokens_per_second": 5961.002
    },
    {
      "epoch": 1.143816091954023,
      "grad_norm": 0.9418705105781555,
      "learning_rate": 8.573532118534879e-05,
      "loss": 1.5209,
      "num_input_tokens_seen": 18796000,
      "step": 3110,
      "train_runtime": 3153.0583,
      "train_tokens_per_second": 5961.197
    },
    {
      "epoch": 1.144183908045977,
      "grad_norm": 0.8473823666572571,
      "learning_rate": 8.569850911098841e-05,
      "loss": 1.366,
      "num_input_tokens_seen": 18802192,
      "step": 3111,
      "train_runtime": 3154.0444,
      "train_tokens_per_second": 5961.296
    },
    {
      "epoch": 1.144551724137931,
      "grad_norm": 0.9196009039878845,
      "learning_rate": 8.566169703662802e-05,
      "loss": 1.4962,
      "num_input_tokens_seen": 18808256,
      "step": 3112,
      "train_runtime": 3155.0268,
      "train_tokens_per_second": 5961.362
    },
    {
      "epoch": 1.144919540229885,
      "grad_norm": 0.815726101398468,
      "learning_rate": 8.562488496226763e-05,
      "loss": 1.4066,
      "num_input_tokens_seen": 18815168,
      "step": 3113,
      "train_runtime": 3156.1286,
      "train_tokens_per_second": 5961.471
    },
    {
      "epoch": 1.145287356321839,
      "grad_norm": 0.8924869894981384,
      "learning_rate": 8.558807288790724e-05,
      "loss": 1.4688,
      "num_input_tokens_seen": 18820592,
      "step": 3114,
      "train_runtime": 3157.0382,
      "train_tokens_per_second": 5961.471
    },
    {
      "epoch": 1.145655172413793,
      "grad_norm": 0.941137969493866,
      "learning_rate": 8.555126081354685e-05,
      "loss": 1.6636,
      "num_input_tokens_seen": 18825520,
      "step": 3115,
      "train_runtime": 3157.8432,
      "train_tokens_per_second": 5961.512
    },
    {
      "epoch": 1.1460229885057471,
      "grad_norm": 0.813913106918335,
      "learning_rate": 8.551444873918646e-05,
      "loss": 1.4553,
      "num_input_tokens_seen": 18832304,
      "step": 3116,
      "train_runtime": 3158.9251,
      "train_tokens_per_second": 5961.618
    },
    {
      "epoch": 1.1463908045977012,
      "grad_norm": 0.8313441872596741,
      "learning_rate": 8.547763666482607e-05,
      "loss": 1.3608,
      "num_input_tokens_seen": 18840832,
      "step": 3117,
      "train_runtime": 3160.2374,
      "train_tokens_per_second": 5961.841
    },
    {
      "epoch": 1.1467586206896552,
      "grad_norm": 0.8408114314079285,
      "learning_rate": 8.544082459046567e-05,
      "loss": 1.6147,
      "num_input_tokens_seen": 18847872,
      "step": 3118,
      "train_runtime": 3161.354,
      "train_tokens_per_second": 5961.962
    },
    {
      "epoch": 1.1471264367816092,
      "grad_norm": 0.8338990211486816,
      "learning_rate": 8.54040125161053e-05,
      "loss": 1.4175,
      "num_input_tokens_seen": 18860704,
      "step": 3119,
      "train_runtime": 3163.2252,
      "train_tokens_per_second": 5962.492
    },
    {
      "epoch": 1.1474942528735632,
      "grad_norm": 0.9207831621170044,
      "learning_rate": 8.53672004417449e-05,
      "loss": 1.6247,
      "num_input_tokens_seen": 18865440,
      "step": 3120,
      "train_runtime": 3164.0165,
      "train_tokens_per_second": 5962.497
    },
    {
      "epoch": 1.1478620689655172,
      "grad_norm": 0.9662007689476013,
      "learning_rate": 8.53303883673845e-05,
      "loss": 1.5092,
      "num_input_tokens_seen": 18871680,
      "step": 3121,
      "train_runtime": 3165.5231,
      "train_tokens_per_second": 5961.631
    },
    {
      "epoch": 1.1482298850574713,
      "grad_norm": 0.8944664597511292,
      "learning_rate": 8.529357629302411e-05,
      "loss": 1.6469,
      "num_input_tokens_seen": 18877072,
      "step": 3122,
      "train_runtime": 3166.422,
      "train_tokens_per_second": 5961.641
    },
    {
      "epoch": 1.1485977011494253,
      "grad_norm": 0.9483094215393066,
      "learning_rate": 8.525676421866372e-05,
      "loss": 1.4102,
      "num_input_tokens_seen": 18882896,
      "step": 3123,
      "train_runtime": 3167.3902,
      "train_tokens_per_second": 5961.658
    },
    {
      "epoch": 1.1489655172413793,
      "grad_norm": 0.8910855650901794,
      "learning_rate": 8.521995214430334e-05,
      "loss": 1.6446,
      "num_input_tokens_seen": 18887904,
      "step": 3124,
      "train_runtime": 3168.2024,
      "train_tokens_per_second": 5961.71
    },
    {
      "epoch": 1.1493333333333333,
      "grad_norm": 0.9108201265335083,
      "learning_rate": 8.518314006994295e-05,
      "loss": 1.4592,
      "num_input_tokens_seen": 18892576,
      "step": 3125,
      "train_runtime": 3169.0123,
      "train_tokens_per_second": 5961.661
    },
    {
      "epoch": 1.1497011494252873,
      "grad_norm": 0.9536663889884949,
      "learning_rate": 8.514632799558256e-05,
      "loss": 1.8689,
      "num_input_tokens_seen": 18897296,
      "step": 3126,
      "train_runtime": 3169.8089,
      "train_tokens_per_second": 5961.651
    },
    {
      "epoch": 1.1500689655172414,
      "grad_norm": 0.9189017415046692,
      "learning_rate": 8.510951592122215e-05,
      "loss": 1.6326,
      "num_input_tokens_seen": 18901744,
      "step": 3127,
      "train_runtime": 3170.5744,
      "train_tokens_per_second": 5961.615
    },
    {
      "epoch": 1.1504367816091954,
      "grad_norm": 0.9252599477767944,
      "learning_rate": 8.507270384686178e-05,
      "loss": 1.6476,
      "num_input_tokens_seen": 18907872,
      "step": 3128,
      "train_runtime": 3171.563,
      "train_tokens_per_second": 5961.689
    },
    {
      "epoch": 1.1508045977011494,
      "grad_norm": 0.9546104073524475,
      "learning_rate": 8.503589177250139e-05,
      "loss": 1.642,
      "num_input_tokens_seen": 18912256,
      "step": 3129,
      "train_runtime": 3172.3325,
      "train_tokens_per_second": 5961.625
    },
    {
      "epoch": 1.1511724137931034,
      "grad_norm": 0.9089950919151306,
      "learning_rate": 8.4999079698141e-05,
      "loss": 1.4078,
      "num_input_tokens_seen": 18918768,
      "step": 3130,
      "train_runtime": 3173.3796,
      "train_tokens_per_second": 5961.71
    },
    {
      "epoch": 1.1515402298850574,
      "grad_norm": 0.8216732740402222,
      "learning_rate": 8.49622676237806e-05,
      "loss": 1.389,
      "num_input_tokens_seen": 18924704,
      "step": 3131,
      "train_runtime": 3174.3663,
      "train_tokens_per_second": 5961.727
    },
    {
      "epoch": 1.1519080459770115,
      "grad_norm": 0.9450916647911072,
      "learning_rate": 8.492545554942021e-05,
      "loss": 1.6209,
      "num_input_tokens_seen": 18931856,
      "step": 3132,
      "train_runtime": 3175.4872,
      "train_tokens_per_second": 5961.874
    },
    {
      "epoch": 1.1522758620689655,
      "grad_norm": 0.981218159198761,
      "learning_rate": 8.488864347505982e-05,
      "loss": 1.6775,
      "num_input_tokens_seen": 18935984,
      "step": 3133,
      "train_runtime": 3176.2149,
      "train_tokens_per_second": 5961.808
    },
    {
      "epoch": 1.1526436781609195,
      "grad_norm": 0.8865116238594055,
      "learning_rate": 8.485183140069943e-05,
      "loss": 1.4935,
      "num_input_tokens_seen": 18940912,
      "step": 3134,
      "train_runtime": 3177.0573,
      "train_tokens_per_second": 5961.779
    },
    {
      "epoch": 1.1530114942528735,
      "grad_norm": 0.886468768119812,
      "learning_rate": 8.481501932633904e-05,
      "loss": 1.4288,
      "num_input_tokens_seen": 18947968,
      "step": 3135,
      "train_runtime": 3178.223,
      "train_tokens_per_second": 5961.812
    },
    {
      "epoch": 1.1533793103448275,
      "grad_norm": 0.8873666524887085,
      "learning_rate": 8.477820725197866e-05,
      "loss": 1.5069,
      "num_input_tokens_seen": 18954272,
      "step": 3136,
      "train_runtime": 3179.2383,
      "train_tokens_per_second": 5961.891
    },
    {
      "epoch": 1.1537471264367816,
      "grad_norm": 0.8816490173339844,
      "learning_rate": 8.474139517761827e-05,
      "loss": 1.5566,
      "num_input_tokens_seen": 18959072,
      "step": 3137,
      "train_runtime": 3180.0629,
      "train_tokens_per_second": 5961.854
    },
    {
      "epoch": 1.1541149425287356,
      "grad_norm": 0.8619621992111206,
      "learning_rate": 8.470458310325788e-05,
      "loss": 1.3632,
      "num_input_tokens_seen": 18965120,
      "step": 3138,
      "train_runtime": 3181.0442,
      "train_tokens_per_second": 5961.916
    },
    {
      "epoch": 1.1544827586206896,
      "grad_norm": 0.8746305108070374,
      "learning_rate": 8.466777102889747e-05,
      "loss": 1.5181,
      "num_input_tokens_seen": 18971600,
      "step": 3139,
      "train_runtime": 3182.1005,
      "train_tokens_per_second": 5961.974
    },
    {
      "epoch": 1.1548505747126436,
      "grad_norm": 0.9130301475524902,
      "learning_rate": 8.463095895453708e-05,
      "loss": 1.5407,
      "num_input_tokens_seen": 18979152,
      "step": 3140,
      "train_runtime": 3183.3141,
      "train_tokens_per_second": 5962.073
    },
    {
      "epoch": 1.1552183908045977,
      "grad_norm": 0.8017939925193787,
      "learning_rate": 8.45941468801767e-05,
      "loss": 1.3054,
      "num_input_tokens_seen": 18984144,
      "step": 3141,
      "train_runtime": 3184.177,
      "train_tokens_per_second": 5962.025
    },
    {
      "epoch": 1.1555862068965517,
      "grad_norm": 0.8778960704803467,
      "learning_rate": 8.455733480581631e-05,
      "loss": 1.3331,
      "num_input_tokens_seen": 18992688,
      "step": 3142,
      "train_runtime": 3185.455,
      "train_tokens_per_second": 5962.316
    },
    {
      "epoch": 1.1559540229885057,
      "grad_norm": 0.8500275611877441,
      "learning_rate": 8.452052273145592e-05,
      "loss": 1.6502,
      "num_input_tokens_seen": 19000688,
      "step": 3143,
      "train_runtime": 3186.7304,
      "train_tokens_per_second": 5962.44
    },
    {
      "epoch": 1.1563218390804597,
      "grad_norm": 0.8706491589546204,
      "learning_rate": 8.448371065709553e-05,
      "loss": 1.4937,
      "num_input_tokens_seen": 19006160,
      "step": 3144,
      "train_runtime": 3187.6366,
      "train_tokens_per_second": 5962.461
    },
    {
      "epoch": 1.1566896551724137,
      "grad_norm": 0.9586254954338074,
      "learning_rate": 8.444689858273514e-05,
      "loss": 1.5329,
      "num_input_tokens_seen": 19010720,
      "step": 3145,
      "train_runtime": 3188.4048,
      "train_tokens_per_second": 5962.455
    },
    {
      "epoch": 1.1570574712643678,
      "grad_norm": 0.9184359908103943,
      "learning_rate": 8.441008650837475e-05,
      "loss": 1.5607,
      "num_input_tokens_seen": 19015600,
      "step": 3146,
      "train_runtime": 3189.2566,
      "train_tokens_per_second": 5962.393
    },
    {
      "epoch": 1.1574252873563218,
      "grad_norm": 0.8390594124794006,
      "learning_rate": 8.437327443401436e-05,
      "loss": 1.5284,
      "num_input_tokens_seen": 19021344,
      "step": 3147,
      "train_runtime": 3190.2046,
      "train_tokens_per_second": 5962.421
    },
    {
      "epoch": 1.1577931034482758,
      "grad_norm": 0.8031556606292725,
      "learning_rate": 8.433646235965397e-05,
      "loss": 1.3581,
      "num_input_tokens_seen": 19027728,
      "step": 3148,
      "train_runtime": 3191.2515,
      "train_tokens_per_second": 5962.466
    },
    {
      "epoch": 1.1581609195402298,
      "grad_norm": 0.9039313197135925,
      "learning_rate": 8.429965028529359e-05,
      "loss": 1.5429,
      "num_input_tokens_seen": 19033504,
      "step": 3149,
      "train_runtime": 3192.2037,
      "train_tokens_per_second": 5962.497
    },
    {
      "epoch": 1.1585287356321838,
      "grad_norm": 0.8582308292388916,
      "learning_rate": 8.42628382109332e-05,
      "loss": 1.5543,
      "num_input_tokens_seen": 19039424,
      "step": 3150,
      "train_runtime": 3193.2031,
      "train_tokens_per_second": 5962.485
    },
    {
      "epoch": 1.1588965517241379,
      "grad_norm": 0.8679171204566956,
      "learning_rate": 8.42260261365728e-05,
      "loss": 1.4109,
      "num_input_tokens_seen": 19043952,
      "step": 3151,
      "train_runtime": 3194.4529,
      "train_tokens_per_second": 5961.569
    },
    {
      "epoch": 1.1592643678160919,
      "grad_norm": 0.9401488304138184,
      "learning_rate": 8.41892140622124e-05,
      "loss": 1.6853,
      "num_input_tokens_seen": 19048688,
      "step": 3152,
      "train_runtime": 3195.279,
      "train_tokens_per_second": 5961.51
    },
    {
      "epoch": 1.159632183908046,
      "grad_norm": 0.8548392653465271,
      "learning_rate": 8.415240198785201e-05,
      "loss": 1.4421,
      "num_input_tokens_seen": 19059408,
      "step": 3153,
      "train_runtime": 3196.884,
      "train_tokens_per_second": 5961.87
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.8348060250282288,
      "learning_rate": 8.411558991349163e-05,
      "loss": 1.3781,
      "num_input_tokens_seen": 19066320,
      "step": 3154,
      "train_runtime": 3197.9739,
      "train_tokens_per_second": 5962.0
    },
    {
      "epoch": 1.160367816091954,
      "grad_norm": 0.9685162305831909,
      "learning_rate": 8.407877783913124e-05,
      "loss": 1.6644,
      "num_input_tokens_seen": 19071120,
      "step": 3155,
      "train_runtime": 3198.7732,
      "train_tokens_per_second": 5962.011
    },
    {
      "epoch": 1.160735632183908,
      "grad_norm": 0.9082851409912109,
      "learning_rate": 8.404196576477085e-05,
      "loss": 1.5741,
      "num_input_tokens_seen": 19076144,
      "step": 3156,
      "train_runtime": 3199.6468,
      "train_tokens_per_second": 5961.953
    },
    {
      "epoch": 1.161103448275862,
      "grad_norm": 1.0296233892440796,
      "learning_rate": 8.400515369041045e-05,
      "loss": 1.7934,
      "num_input_tokens_seen": 19081152,
      "step": 3157,
      "train_runtime": 3200.4857,
      "train_tokens_per_second": 5961.955
    },
    {
      "epoch": 1.161471264367816,
      "grad_norm": 0.889957845211029,
      "learning_rate": 8.396834161605007e-05,
      "loss": 1.5453,
      "num_input_tokens_seen": 19089360,
      "step": 3158,
      "train_runtime": 3201.7662,
      "train_tokens_per_second": 5962.134
    },
    {
      "epoch": 1.16183908045977,
      "grad_norm": 0.9897533655166626,
      "learning_rate": 8.393152954168968e-05,
      "loss": 1.7264,
      "num_input_tokens_seen": 19094000,
      "step": 3159,
      "train_runtime": 3202.5681,
      "train_tokens_per_second": 5962.09
    },
    {
      "epoch": 1.162206896551724,
      "grad_norm": 0.8841046094894409,
      "learning_rate": 8.389471746732929e-05,
      "loss": 1.4584,
      "num_input_tokens_seen": 19099952,
      "step": 3160,
      "train_runtime": 3203.5003,
      "train_tokens_per_second": 5962.213
    },
    {
      "epoch": 1.162574712643678,
      "grad_norm": 0.9966636896133423,
      "learning_rate": 8.38579053929689e-05,
      "loss": 1.6434,
      "num_input_tokens_seen": 19104688,
      "step": 3161,
      "train_runtime": 3204.3033,
      "train_tokens_per_second": 5962.197
    },
    {
      "epoch": 1.1629425287356323,
      "grad_norm": 0.8869330883026123,
      "learning_rate": 8.38210933186085e-05,
      "loss": 1.3995,
      "num_input_tokens_seen": 19109680,
      "step": 3162,
      "train_runtime": 3205.1452,
      "train_tokens_per_second": 5962.188
    },
    {
      "epoch": 1.163310344827586,
      "grad_norm": 0.9175199866294861,
      "learning_rate": 8.378428124424811e-05,
      "loss": 1.4857,
      "num_input_tokens_seen": 19118912,
      "step": 3163,
      "train_runtime": 3206.5578,
      "train_tokens_per_second": 5962.441
    },
    {
      "epoch": 1.1636781609195403,
      "grad_norm": 0.9480499625205994,
      "learning_rate": 8.374746916988772e-05,
      "loss": 1.4935,
      "num_input_tokens_seen": 19125344,
      "step": 3164,
      "train_runtime": 3207.5855,
      "train_tokens_per_second": 5962.536
    },
    {
      "epoch": 1.1640459770114941,
      "grad_norm": 0.942071259021759,
      "learning_rate": 8.371065709552733e-05,
      "loss": 1.6727,
      "num_input_tokens_seen": 19131088,
      "step": 3165,
      "train_runtime": 3208.5089,
      "train_tokens_per_second": 5962.61
    },
    {
      "epoch": 1.1644137931034484,
      "grad_norm": 0.8933213949203491,
      "learning_rate": 8.367384502116696e-05,
      "loss": 1.4635,
      "num_input_tokens_seen": 19140320,
      "step": 3166,
      "train_runtime": 3209.9195,
      "train_tokens_per_second": 5962.866
    },
    {
      "epoch": 1.1647816091954022,
      "grad_norm": 0.9491671323776245,
      "learning_rate": 8.363703294680656e-05,
      "loss": 1.4727,
      "num_input_tokens_seen": 19147488,
      "step": 3167,
      "train_runtime": 3211.0507,
      "train_tokens_per_second": 5962.998
    },
    {
      "epoch": 1.1651494252873564,
      "grad_norm": 0.8721309900283813,
      "learning_rate": 8.360022087244617e-05,
      "loss": 1.4922,
      "num_input_tokens_seen": 19153360,
      "step": 3168,
      "train_runtime": 3212.0189,
      "train_tokens_per_second": 5963.028
    },
    {
      "epoch": 1.1655172413793102,
      "grad_norm": 0.9578979015350342,
      "learning_rate": 8.356340879808577e-05,
      "loss": 1.7023,
      "num_input_tokens_seen": 19157600,
      "step": 3169,
      "train_runtime": 3212.7606,
      "train_tokens_per_second": 5962.971
    },
    {
      "epoch": 1.1658850574712645,
      "grad_norm": 0.8974698781967163,
      "learning_rate": 8.352659672372538e-05,
      "loss": 1.5144,
      "num_input_tokens_seen": 19163920,
      "step": 3170,
      "train_runtime": 3213.7974,
      "train_tokens_per_second": 5963.014
    },
    {
      "epoch": 1.1662528735632183,
      "grad_norm": 0.9205195307731628,
      "learning_rate": 8.3489784649365e-05,
      "loss": 1.6333,
      "num_input_tokens_seen": 19171120,
      "step": 3171,
      "train_runtime": 3214.94,
      "train_tokens_per_second": 5963.135
    },
    {
      "epoch": 1.1666206896551725,
      "grad_norm": 0.9501519799232483,
      "learning_rate": 8.345297257500461e-05,
      "loss": 1.458,
      "num_input_tokens_seen": 19175200,
      "step": 3172,
      "train_runtime": 3215.6434,
      "train_tokens_per_second": 5963.099
    },
    {
      "epoch": 1.1669885057471263,
      "grad_norm": 0.9079390168190002,
      "learning_rate": 8.341616050064422e-05,
      "loss": 1.4757,
      "num_input_tokens_seen": 19180704,
      "step": 3173,
      "train_runtime": 3216.5616,
      "train_tokens_per_second": 5963.108
    },
    {
      "epoch": 1.1673563218390806,
      "grad_norm": 0.9358499050140381,
      "learning_rate": 8.337934842628383e-05,
      "loss": 1.5204,
      "num_input_tokens_seen": 19186880,
      "step": 3174,
      "train_runtime": 3217.54,
      "train_tokens_per_second": 5963.214
    },
    {
      "epoch": 1.1677241379310346,
      "grad_norm": 1.000985026359558,
      "learning_rate": 8.334253635192344e-05,
      "loss": 1.5279,
      "num_input_tokens_seen": 19193456,
      "step": 3175,
      "train_runtime": 3218.5911,
      "train_tokens_per_second": 5963.31
    },
    {
      "epoch": 1.1680919540229886,
      "grad_norm": 0.9262905716896057,
      "learning_rate": 8.330572427756304e-05,
      "loss": 1.4546,
      "num_input_tokens_seen": 19200192,
      "step": 3176,
      "train_runtime": 3219.6651,
      "train_tokens_per_second": 5963.413
    },
    {
      "epoch": 1.1684597701149426,
      "grad_norm": 0.8638454079627991,
      "learning_rate": 8.326891220320265e-05,
      "loss": 1.5662,
      "num_input_tokens_seen": 19206608,
      "step": 3177,
      "train_runtime": 3220.7204,
      "train_tokens_per_second": 5963.451
    },
    {
      "epoch": 1.1688275862068966,
      "grad_norm": 0.8458003401756287,
      "learning_rate": 8.323210012884226e-05,
      "loss": 1.5592,
      "num_input_tokens_seen": 19213984,
      "step": 3178,
      "train_runtime": 3221.9136,
      "train_tokens_per_second": 5963.532
    },
    {
      "epoch": 1.1691954022988507,
      "grad_norm": 0.9439873099327087,
      "learning_rate": 8.319528805448188e-05,
      "loss": 1.6025,
      "num_input_tokens_seen": 19218848,
      "step": 3179,
      "train_runtime": 3222.7549,
      "train_tokens_per_second": 5963.484
    },
    {
      "epoch": 1.1695632183908047,
      "grad_norm": 0.9297096729278564,
      "learning_rate": 8.315847598012148e-05,
      "loss": 1.5777,
      "num_input_tokens_seen": 19223664,
      "step": 3180,
      "train_runtime": 3223.5674,
      "train_tokens_per_second": 5963.475
    },
    {
      "epoch": 1.1699310344827587,
      "grad_norm": 0.901832640171051,
      "learning_rate": 8.312166390576109e-05,
      "loss": 1.6016,
      "num_input_tokens_seen": 19228240,
      "step": 3181,
      "train_runtime": 3224.915,
      "train_tokens_per_second": 5962.402
    },
    {
      "epoch": 1.1702988505747127,
      "grad_norm": 0.8717474937438965,
      "learning_rate": 8.30848518314007e-05,
      "loss": 1.5431,
      "num_input_tokens_seen": 19232912,
      "step": 3182,
      "train_runtime": 3225.7097,
      "train_tokens_per_second": 5962.382
    },
    {
      "epoch": 1.1706666666666667,
      "grad_norm": 1.0567877292633057,
      "learning_rate": 8.30480397570403e-05,
      "loss": 1.6796,
      "num_input_tokens_seen": 19237280,
      "step": 3183,
      "train_runtime": 3226.4629,
      "train_tokens_per_second": 5962.343
    },
    {
      "epoch": 1.1710344827586208,
      "grad_norm": 0.858303964138031,
      "learning_rate": 8.301122768267993e-05,
      "loss": 1.5896,
      "num_input_tokens_seen": 19244224,
      "step": 3184,
      "train_runtime": 3227.5823,
      "train_tokens_per_second": 5962.427
    },
    {
      "epoch": 1.1714022988505748,
      "grad_norm": 0.8872172236442566,
      "learning_rate": 8.297441560831954e-05,
      "loss": 1.3913,
      "num_input_tokens_seen": 19250880,
      "step": 3185,
      "train_runtime": 3228.666,
      "train_tokens_per_second": 5962.487
    },
    {
      "epoch": 1.1717701149425288,
      "grad_norm": 0.9927968978881836,
      "learning_rate": 8.293760353395915e-05,
      "loss": 1.6915,
      "num_input_tokens_seen": 19257008,
      "step": 3186,
      "train_runtime": 3229.6287,
      "train_tokens_per_second": 5962.607
    },
    {
      "epoch": 1.1721379310344828,
      "grad_norm": 0.8749542832374573,
      "learning_rate": 8.290079145959874e-05,
      "loss": 1.4406,
      "num_input_tokens_seen": 19261680,
      "step": 3187,
      "train_runtime": 3230.4144,
      "train_tokens_per_second": 5962.603
    },
    {
      "epoch": 1.1725057471264368,
      "grad_norm": 0.9581989049911499,
      "learning_rate": 8.286397938523836e-05,
      "loss": 1.6091,
      "num_input_tokens_seen": 19266368,
      "step": 3188,
      "train_runtime": 3231.2113,
      "train_tokens_per_second": 5962.584
    },
    {
      "epoch": 1.1728735632183909,
      "grad_norm": 0.8969200849533081,
      "learning_rate": 8.282716731087797e-05,
      "loss": 1.4148,
      "num_input_tokens_seen": 19271840,
      "step": 3189,
      "train_runtime": 3232.1195,
      "train_tokens_per_second": 5962.601
    },
    {
      "epoch": 1.173241379310345,
      "grad_norm": 0.9299541711807251,
      "learning_rate": 8.279035523651758e-05,
      "loss": 1.5344,
      "num_input_tokens_seen": 19276944,
      "step": 3190,
      "train_runtime": 3232.9493,
      "train_tokens_per_second": 5962.65
    },
    {
      "epoch": 1.173609195402299,
      "grad_norm": 0.9658132791519165,
      "learning_rate": 8.275354316215719e-05,
      "loss": 1.5201,
      "num_input_tokens_seen": 19282224,
      "step": 3191,
      "train_runtime": 3233.8342,
      "train_tokens_per_second": 5962.651
    },
    {
      "epoch": 1.173977011494253,
      "grad_norm": 0.793863832950592,
      "learning_rate": 8.27167310877968e-05,
      "loss": 1.5155,
      "num_input_tokens_seen": 19291488,
      "step": 3192,
      "train_runtime": 3235.261,
      "train_tokens_per_second": 5962.885
    },
    {
      "epoch": 1.174344827586207,
      "grad_norm": 0.9790344834327698,
      "learning_rate": 8.267991901343641e-05,
      "loss": 1.6852,
      "num_input_tokens_seen": 19296368,
      "step": 3193,
      "train_runtime": 3236.0983,
      "train_tokens_per_second": 5962.85
    },
    {
      "epoch": 1.174712643678161,
      "grad_norm": 0.9354575276374817,
      "learning_rate": 8.264310693907602e-05,
      "loss": 1.4438,
      "num_input_tokens_seen": 19302384,
      "step": 3194,
      "train_runtime": 3237.0986,
      "train_tokens_per_second": 5962.866
    },
    {
      "epoch": 1.175080459770115,
      "grad_norm": 0.9277665019035339,
      "learning_rate": 8.260629486471563e-05,
      "loss": 1.5758,
      "num_input_tokens_seen": 19308688,
      "step": 3195,
      "train_runtime": 3238.1108,
      "train_tokens_per_second": 5962.948
    },
    {
      "epoch": 1.175448275862069,
      "grad_norm": 1.0496010780334473,
      "learning_rate": 8.256948279035525e-05,
      "loss": 1.5927,
      "num_input_tokens_seen": 19314320,
      "step": 3196,
      "train_runtime": 3239.0348,
      "train_tokens_per_second": 5962.986
    },
    {
      "epoch": 1.175816091954023,
      "grad_norm": 1.0036723613739014,
      "learning_rate": 8.253267071599486e-05,
      "loss": 1.6877,
      "num_input_tokens_seen": 19321664,
      "step": 3197,
      "train_runtime": 3240.2072,
      "train_tokens_per_second": 5963.095
    },
    {
      "epoch": 1.176183908045977,
      "grad_norm": 0.9577779769897461,
      "learning_rate": 8.249585864163447e-05,
      "loss": 1.5991,
      "num_input_tokens_seen": 19326288,
      "step": 3198,
      "train_runtime": 3240.9967,
      "train_tokens_per_second": 5963.069
    },
    {
      "epoch": 1.176551724137931,
      "grad_norm": 0.858090341091156,
      "learning_rate": 8.245904656727406e-05,
      "loss": 1.3665,
      "num_input_tokens_seen": 19335088,
      "step": 3199,
      "train_runtime": 3242.3459,
      "train_tokens_per_second": 5963.302
    },
    {
      "epoch": 1.176919540229885,
      "grad_norm": 1.0612149238586426,
      "learning_rate": 8.242223449291367e-05,
      "loss": 1.8193,
      "num_input_tokens_seen": 19339920,
      "step": 3200,
      "train_runtime": 3243.1453,
      "train_tokens_per_second": 5963.322
    },
    {
      "epoch": 1.1772873563218391,
      "grad_norm": 1.0048408508300781,
      "learning_rate": 8.23854224185533e-05,
      "loss": 1.5437,
      "num_input_tokens_seen": 19345280,
      "step": 3201,
      "train_runtime": 3244.0508,
      "train_tokens_per_second": 5963.31
    },
    {
      "epoch": 1.1776551724137931,
      "grad_norm": 0.8895339369773865,
      "learning_rate": 8.23486103441929e-05,
      "loss": 1.5368,
      "num_input_tokens_seen": 19351056,
      "step": 3202,
      "train_runtime": 3245.0114,
      "train_tokens_per_second": 5963.325
    },
    {
      "epoch": 1.1780229885057472,
      "grad_norm": 0.8810693025588989,
      "learning_rate": 8.231179826983251e-05,
      "loss": 1.4941,
      "num_input_tokens_seen": 19355200,
      "step": 3203,
      "train_runtime": 3245.7372,
      "train_tokens_per_second": 5963.268
    },
    {
      "epoch": 1.1783908045977012,
      "grad_norm": 0.8693225979804993,
      "learning_rate": 8.227498619547212e-05,
      "loss": 1.4464,
      "num_input_tokens_seen": 19362144,
      "step": 3204,
      "train_runtime": 3246.8661,
      "train_tokens_per_second": 5963.333
    },
    {
      "epoch": 1.1787586206896552,
      "grad_norm": 1.0319263935089111,
      "learning_rate": 8.223817412111173e-05,
      "loss": 1.6708,
      "num_input_tokens_seen": 19366688,
      "step": 3205,
      "train_runtime": 3247.6717,
      "train_tokens_per_second": 5963.253
    },
    {
      "epoch": 1.1791264367816092,
      "grad_norm": 0.9654633402824402,
      "learning_rate": 8.220136204675134e-05,
      "loss": 1.6914,
      "num_input_tokens_seen": 19371632,
      "step": 3206,
      "train_runtime": 3248.5143,
      "train_tokens_per_second": 5963.228
    },
    {
      "epoch": 1.1794942528735632,
      "grad_norm": 0.9302749633789062,
      "learning_rate": 8.216454997239095e-05,
      "loss": 1.3702,
      "num_input_tokens_seen": 19379280,
      "step": 3207,
      "train_runtime": 3249.7281,
      "train_tokens_per_second": 5963.354
    },
    {
      "epoch": 1.1798620689655173,
      "grad_norm": 0.8170776963233948,
      "learning_rate": 8.212773789803056e-05,
      "loss": 1.5267,
      "num_input_tokens_seen": 19385744,
      "step": 3208,
      "train_runtime": 3250.7753,
      "train_tokens_per_second": 5963.422
    },
    {
      "epoch": 1.1802298850574713,
      "grad_norm": 0.913048505783081,
      "learning_rate": 8.209092582367018e-05,
      "loss": 1.5624,
      "num_input_tokens_seen": 19391312,
      "step": 3209,
      "train_runtime": 3251.6913,
      "train_tokens_per_second": 5963.454
    },
    {
      "epoch": 1.1805977011494253,
      "grad_norm": 0.9876195788383484,
      "learning_rate": 8.205411374930977e-05,
      "loss": 1.4671,
      "num_input_tokens_seen": 19397440,
      "step": 3210,
      "train_runtime": 3252.6969,
      "train_tokens_per_second": 5963.494
    },
    {
      "epoch": 1.1809655172413793,
      "grad_norm": 0.9700420498847961,
      "learning_rate": 8.201730167494938e-05,
      "loss": 1.6127,
      "num_input_tokens_seen": 19401920,
      "step": 3211,
      "train_runtime": 3253.9817,
      "train_tokens_per_second": 5962.517
    },
    {
      "epoch": 1.1813333333333333,
      "grad_norm": 0.740075945854187,
      "learning_rate": 8.198048960058899e-05,
      "loss": 1.4049,
      "num_input_tokens_seen": 19417392,
      "step": 3212,
      "train_runtime": 3256.2541,
      "train_tokens_per_second": 5963.107
    },
    {
      "epoch": 1.1817011494252874,
      "grad_norm": 0.8618344664573669,
      "learning_rate": 8.19436775262286e-05,
      "loss": 1.2838,
      "num_input_tokens_seen": 19423792,
      "step": 3213,
      "train_runtime": 3257.2924,
      "train_tokens_per_second": 5963.171
    },
    {
      "epoch": 1.1820689655172414,
      "grad_norm": 0.9339157342910767,
      "learning_rate": 8.190686545186822e-05,
      "loss": 1.653,
      "num_input_tokens_seen": 19428576,
      "step": 3214,
      "train_runtime": 3258.1024,
      "train_tokens_per_second": 5963.157
    },
    {
      "epoch": 1.1824367816091954,
      "grad_norm": 0.9181302785873413,
      "learning_rate": 8.187005337750783e-05,
      "loss": 1.5645,
      "num_input_tokens_seen": 19433568,
      "step": 3215,
      "train_runtime": 3258.9489,
      "train_tokens_per_second": 5963.14
    },
    {
      "epoch": 1.1828045977011494,
      "grad_norm": 0.8672519326210022,
      "learning_rate": 8.183324130314744e-05,
      "loss": 1.827,
      "num_input_tokens_seen": 19438432,
      "step": 3216,
      "train_runtime": 3259.7928,
      "train_tokens_per_second": 5963.088
    },
    {
      "epoch": 1.1831724137931035,
      "grad_norm": 0.9546900987625122,
      "learning_rate": 8.179642922878704e-05,
      "loss": 1.7126,
      "num_input_tokens_seen": 19443616,
      "step": 3217,
      "train_runtime": 3260.6545,
      "train_tokens_per_second": 5963.102
    },
    {
      "epoch": 1.1835402298850575,
      "grad_norm": 0.8584520816802979,
      "learning_rate": 8.175961715442666e-05,
      "loss": 1.4794,
      "num_input_tokens_seen": 19449088,
      "step": 3218,
      "train_runtime": 3261.566,
      "train_tokens_per_second": 5963.113
    },
    {
      "epoch": 1.1839080459770115,
      "grad_norm": 0.8948041796684265,
      "learning_rate": 8.172280508006627e-05,
      "loss": 1.3936,
      "num_input_tokens_seen": 19454400,
      "step": 3219,
      "train_runtime": 3262.4634,
      "train_tokens_per_second": 5963.101
    },
    {
      "epoch": 1.1842758620689655,
      "grad_norm": 0.9261506199836731,
      "learning_rate": 8.168599300570588e-05,
      "loss": 1.4926,
      "num_input_tokens_seen": 19459792,
      "step": 3220,
      "train_runtime": 3263.3702,
      "train_tokens_per_second": 5963.097
    },
    {
      "epoch": 1.1846436781609195,
      "grad_norm": 0.9783214926719666,
      "learning_rate": 8.164918093134548e-05,
      "loss": 1.5295,
      "num_input_tokens_seen": 19463936,
      "step": 3221,
      "train_runtime": 3264.0771,
      "train_tokens_per_second": 5963.075
    },
    {
      "epoch": 1.1850114942528736,
      "grad_norm": 0.94748854637146,
      "learning_rate": 8.16123688569851e-05,
      "loss": 1.4914,
      "num_input_tokens_seen": 19469328,
      "step": 3222,
      "train_runtime": 3264.9636,
      "train_tokens_per_second": 5963.107
    },
    {
      "epoch": 1.1853793103448276,
      "grad_norm": 0.90382319688797,
      "learning_rate": 8.15755567826247e-05,
      "loss": 1.5486,
      "num_input_tokens_seen": 19474400,
      "step": 3223,
      "train_runtime": 3265.7988,
      "train_tokens_per_second": 5963.135
    },
    {
      "epoch": 1.1857471264367816,
      "grad_norm": 0.9176757335662842,
      "learning_rate": 8.153874470826431e-05,
      "loss": 1.6369,
      "num_input_tokens_seen": 19478688,
      "step": 3224,
      "train_runtime": 3266.5381,
      "train_tokens_per_second": 5963.098
    },
    {
      "epoch": 1.1861149425287356,
      "grad_norm": 0.8889284133911133,
      "learning_rate": 8.150193263390392e-05,
      "loss": 1.5167,
      "num_input_tokens_seen": 19484944,
      "step": 3225,
      "train_runtime": 3267.5445,
      "train_tokens_per_second": 5963.176
    },
    {
      "epoch": 1.1864827586206896,
      "grad_norm": 0.879899263381958,
      "learning_rate": 8.146512055954354e-05,
      "loss": 1.3931,
      "num_input_tokens_seen": 19490464,
      "step": 3226,
      "train_runtime": 3268.4559,
      "train_tokens_per_second": 5963.202
    },
    {
      "epoch": 1.1868505747126437,
      "grad_norm": 0.9044172763824463,
      "learning_rate": 8.142830848518315e-05,
      "loss": 1.4089,
      "num_input_tokens_seen": 19498000,
      "step": 3227,
      "train_runtime": 3269.6555,
      "train_tokens_per_second": 5963.319
    },
    {
      "epoch": 1.1872183908045977,
      "grad_norm": 1.0018138885498047,
      "learning_rate": 8.139149641082276e-05,
      "loss": 1.7576,
      "num_input_tokens_seen": 19503024,
      "step": 3228,
      "train_runtime": 3270.5075,
      "train_tokens_per_second": 5963.302
    },
    {
      "epoch": 1.1875862068965517,
      "grad_norm": 0.9552081227302551,
      "learning_rate": 8.135468433646236e-05,
      "loss": 1.5552,
      "num_input_tokens_seen": 19508976,
      "step": 3229,
      "train_runtime": 3271.4854,
      "train_tokens_per_second": 5963.339
    },
    {
      "epoch": 1.1879540229885057,
      "grad_norm": 1.0562891960144043,
      "learning_rate": 8.131787226210196e-05,
      "loss": 1.4937,
      "num_input_tokens_seen": 19514976,
      "step": 3230,
      "train_runtime": 3272.4508,
      "train_tokens_per_second": 5963.413
    },
    {
      "epoch": 1.1883218390804597,
      "grad_norm": 0.9812391400337219,
      "learning_rate": 8.128106018774159e-05,
      "loss": 1.5314,
      "num_input_tokens_seen": 19521616,
      "step": 3231,
      "train_runtime": 3273.5187,
      "train_tokens_per_second": 5963.496
    },
    {
      "epoch": 1.1886896551724138,
      "grad_norm": 0.9129152297973633,
      "learning_rate": 8.12442481133812e-05,
      "loss": 1.6085,
      "num_input_tokens_seen": 19526848,
      "step": 3232,
      "train_runtime": 3274.3969,
      "train_tokens_per_second": 5963.495
    },
    {
      "epoch": 1.1890574712643678,
      "grad_norm": 0.8810139894485474,
      "learning_rate": 8.12074360390208e-05,
      "loss": 1.4567,
      "num_input_tokens_seen": 19533104,
      "step": 3233,
      "train_runtime": 3275.4096,
      "train_tokens_per_second": 5963.561
    },
    {
      "epoch": 1.1894252873563218,
      "grad_norm": 0.9950277209281921,
      "learning_rate": 8.117062396466041e-05,
      "loss": 1.6301,
      "num_input_tokens_seen": 19538112,
      "step": 3234,
      "train_runtime": 3276.2541,
      "train_tokens_per_second": 5963.552
    },
    {
      "epoch": 1.1897931034482758,
      "grad_norm": 0.8755245208740234,
      "learning_rate": 8.113381189030002e-05,
      "loss": 1.5151,
      "num_input_tokens_seen": 19546416,
      "step": 3235,
      "train_runtime": 3277.5565,
      "train_tokens_per_second": 5963.716
    },
    {
      "epoch": 1.1901609195402298,
      "grad_norm": 0.8667770624160767,
      "learning_rate": 8.109699981593963e-05,
      "loss": 1.5022,
      "num_input_tokens_seen": 19553776,
      "step": 3236,
      "train_runtime": 3278.7231,
      "train_tokens_per_second": 5963.839
    },
    {
      "epoch": 1.1905287356321839,
      "grad_norm": 0.7814840078353882,
      "learning_rate": 8.106018774157924e-05,
      "loss": 1.4018,
      "num_input_tokens_seen": 19563984,
      "step": 3237,
      "train_runtime": 3280.2633,
      "train_tokens_per_second": 5964.15
    },
    {
      "epoch": 1.1908965517241379,
      "grad_norm": 0.8585484623908997,
      "learning_rate": 8.102337566721885e-05,
      "loss": 1.442,
      "num_input_tokens_seen": 19569968,
      "step": 3238,
      "train_runtime": 3281.2562,
      "train_tokens_per_second": 5964.169
    },
    {
      "epoch": 1.191264367816092,
      "grad_norm": 0.8789281249046326,
      "learning_rate": 8.098656359285847e-05,
      "loss": 1.5233,
      "num_input_tokens_seen": 19575920,
      "step": 3239,
      "train_runtime": 3282.2032,
      "train_tokens_per_second": 5964.262
    },
    {
      "epoch": 1.191632183908046,
      "grad_norm": 0.9377002716064453,
      "learning_rate": 8.094975151849807e-05,
      "loss": 1.4642,
      "num_input_tokens_seen": 19581824,
      "step": 3240,
      "train_runtime": 3283.145,
      "train_tokens_per_second": 5964.349
    },
    {
      "epoch": 1.192,
      "grad_norm": 0.8184657096862793,
      "learning_rate": 8.091293944413768e-05,
      "loss": 1.409,
      "num_input_tokens_seen": 19592800,
      "step": 3241,
      "train_runtime": 3285.2853,
      "train_tokens_per_second": 5963.805
    },
    {
      "epoch": 1.192367816091954,
      "grad_norm": 0.8313052654266357,
      "learning_rate": 8.087612736977729e-05,
      "loss": 1.3468,
      "num_input_tokens_seen": 19599104,
      "step": 3242,
      "train_runtime": 3286.3025,
      "train_tokens_per_second": 5963.877
    },
    {
      "epoch": 1.192735632183908,
      "grad_norm": 0.9750943779945374,
      "learning_rate": 8.08393152954169e-05,
      "loss": 1.6016,
      "num_input_tokens_seen": 19603888,
      "step": 3243,
      "train_runtime": 3287.1299,
      "train_tokens_per_second": 5963.831
    },
    {
      "epoch": 1.193103448275862,
      "grad_norm": 1.0952489376068115,
      "learning_rate": 8.080250322105652e-05,
      "loss": 1.7258,
      "num_input_tokens_seen": 19608896,
      "step": 3244,
      "train_runtime": 3287.9824,
      "train_tokens_per_second": 5963.808
    },
    {
      "epoch": 1.193471264367816,
      "grad_norm": 0.8855782747268677,
      "learning_rate": 8.076569114669613e-05,
      "loss": 1.4628,
      "num_input_tokens_seen": 19614032,
      "step": 3245,
      "train_runtime": 3288.8643,
      "train_tokens_per_second": 5963.77
    },
    {
      "epoch": 1.19383908045977,
      "grad_norm": 0.9466802477836609,
      "learning_rate": 8.072887907233573e-05,
      "loss": 1.5911,
      "num_input_tokens_seen": 19619216,
      "step": 3246,
      "train_runtime": 3289.7111,
      "train_tokens_per_second": 5963.811
    },
    {
      "epoch": 1.194206896551724,
      "grad_norm": 0.7806699275970459,
      "learning_rate": 8.069206699797533e-05,
      "loss": 1.504,
      "num_input_tokens_seen": 19629616,
      "step": 3247,
      "train_runtime": 3291.2818,
      "train_tokens_per_second": 5964.125
    },
    {
      "epoch": 1.194574712643678,
      "grad_norm": 0.9241046905517578,
      "learning_rate": 8.065525492361495e-05,
      "loss": 1.2138,
      "num_input_tokens_seen": 19635968,
      "step": 3248,
      "train_runtime": 3292.2749,
      "train_tokens_per_second": 5964.255
    },
    {
      "epoch": 1.1949425287356321,
      "grad_norm": 0.8583694696426392,
      "learning_rate": 8.061844284925456e-05,
      "loss": 1.4875,
      "num_input_tokens_seen": 19643200,
      "step": 3249,
      "train_runtime": 3293.4371,
      "train_tokens_per_second": 5964.347
    },
    {
      "epoch": 1.1953103448275861,
      "grad_norm": 0.8644989728927612,
      "learning_rate": 8.058163077489417e-05,
      "loss": 1.325,
      "num_input_tokens_seen": 19648400,
      "step": 3250,
      "train_runtime": 3294.3087,
      "train_tokens_per_second": 5964.347
    },
    {
      "epoch": 1.1956781609195402,
      "grad_norm": 0.8890516757965088,
      "learning_rate": 8.054481870053378e-05,
      "loss": 1.5347,
      "num_input_tokens_seen": 19652848,
      "step": 3251,
      "train_runtime": 3295.1421,
      "train_tokens_per_second": 5964.188
    },
    {
      "epoch": 1.1960459770114942,
      "grad_norm": 0.7363216280937195,
      "learning_rate": 8.050800662617339e-05,
      "loss": 1.3499,
      "num_input_tokens_seen": 19660448,
      "step": 3252,
      "train_runtime": 3296.3192,
      "train_tokens_per_second": 5964.364
    },
    {
      "epoch": 1.1964137931034482,
      "grad_norm": 0.7804744839668274,
      "learning_rate": 8.0471194551813e-05,
      "loss": 1.3208,
      "num_input_tokens_seen": 19670288,
      "step": 3253,
      "train_runtime": 3297.8119,
      "train_tokens_per_second": 5964.648
    },
    {
      "epoch": 1.1967816091954022,
      "grad_norm": 0.9398249983787537,
      "learning_rate": 8.04343824774526e-05,
      "loss": 1.5689,
      "num_input_tokens_seen": 19675264,
      "step": 3254,
      "train_runtime": 3298.6672,
      "train_tokens_per_second": 5964.61
    },
    {
      "epoch": 1.1971494252873562,
      "grad_norm": 0.9330623149871826,
      "learning_rate": 8.039757040309221e-05,
      "loss": 1.4018,
      "num_input_tokens_seen": 19680784,
      "step": 3255,
      "train_runtime": 3299.5956,
      "train_tokens_per_second": 5964.605
    },
    {
      "epoch": 1.1975172413793103,
      "grad_norm": 0.814540445804596,
      "learning_rate": 8.036075832873184e-05,
      "loss": 1.4019,
      "num_input_tokens_seen": 19686688,
      "step": 3256,
      "train_runtime": 3300.5579,
      "train_tokens_per_second": 5964.655
    },
    {
      "epoch": 1.1978850574712643,
      "grad_norm": 0.9622266888618469,
      "learning_rate": 8.032394625437145e-05,
      "loss": 1.61,
      "num_input_tokens_seen": 19695456,
      "step": 3257,
      "train_runtime": 3301.9167,
      "train_tokens_per_second": 5964.856
    },
    {
      "epoch": 1.1982528735632183,
      "grad_norm": 0.8945811986923218,
      "learning_rate": 8.028713418001104e-05,
      "loss": 1.5055,
      "num_input_tokens_seen": 19701760,
      "step": 3258,
      "train_runtime": 3302.9329,
      "train_tokens_per_second": 5964.929
    },
    {
      "epoch": 1.1986206896551723,
      "grad_norm": 0.8919451832771301,
      "learning_rate": 8.025032210565065e-05,
      "loss": 1.5532,
      "num_input_tokens_seen": 19706656,
      "step": 3259,
      "train_runtime": 3303.7726,
      "train_tokens_per_second": 5964.895
    },
    {
      "epoch": 1.1989885057471263,
      "grad_norm": 1.008229374885559,
      "learning_rate": 8.021351003129026e-05,
      "loss": 1.5664,
      "num_input_tokens_seen": 19711552,
      "step": 3260,
      "train_runtime": 3304.6106,
      "train_tokens_per_second": 5964.864
    },
    {
      "epoch": 1.1993563218390804,
      "grad_norm": 0.8970891237258911,
      "learning_rate": 8.017669795692988e-05,
      "loss": 1.3042,
      "num_input_tokens_seen": 19716848,
      "step": 3261,
      "train_runtime": 3305.4942,
      "train_tokens_per_second": 5964.871
    },
    {
      "epoch": 1.1997241379310344,
      "grad_norm": 1.038142442703247,
      "learning_rate": 8.013988588256949e-05,
      "loss": 1.4882,
      "num_input_tokens_seen": 19723552,
      "step": 3262,
      "train_runtime": 3306.5639,
      "train_tokens_per_second": 5964.969
    },
    {
      "epoch": 1.2000919540229886,
      "grad_norm": 0.958101212978363,
      "learning_rate": 8.01030738082091e-05,
      "loss": 1.3517,
      "num_input_tokens_seen": 19731712,
      "step": 3263,
      "train_runtime": 3307.8666,
      "train_tokens_per_second": 5965.087
    },
    {
      "epoch": 1.2004597701149424,
      "grad_norm": 0.9111717343330383,
      "learning_rate": 8.006626173384871e-05,
      "loss": 1.5232,
      "num_input_tokens_seen": 19737856,
      "step": 3264,
      "train_runtime": 3308.8761,
      "train_tokens_per_second": 5965.124
    },
    {
      "epoch": 1.2004597701149424,
      "eval_loss": 1.7738701105117798,
      "eval_runtime": 4.7827,
      "eval_samples_per_second": 209.086,
      "eval_steps_per_second": 13.172,
      "num_input_tokens_seen": 19737856,
      "step": 3264
    },
    {
      "epoch": 1.2008275862068967,
      "grad_norm": 0.9947562217712402,
      "learning_rate": 8.002944965948832e-05,
      "loss": 1.4272,
      "num_input_tokens_seen": 19746112,
      "step": 3265,
      "train_runtime": 3314.9412,
      "train_tokens_per_second": 5956.701
    },
    {
      "epoch": 1.2011954022988505,
      "grad_norm": 0.9012897610664368,
      "learning_rate": 7.999263758512793e-05,
      "loss": 1.4079,
      "num_input_tokens_seen": 19751040,
      "step": 3266,
      "train_runtime": 3315.7739,
      "train_tokens_per_second": 5956.691
    },
    {
      "epoch": 1.2015632183908047,
      "grad_norm": 0.9524996280670166,
      "learning_rate": 7.995582551076753e-05,
      "loss": 1.2747,
      "num_input_tokens_seen": 19756960,
      "step": 3267,
      "train_runtime": 3316.745,
      "train_tokens_per_second": 5956.732
    },
    {
      "epoch": 1.2019310344827585,
      "grad_norm": 0.9411486387252808,
      "learning_rate": 7.991901343640714e-05,
      "loss": 1.647,
      "num_input_tokens_seen": 19761968,
      "step": 3268,
      "train_runtime": 3317.5922,
      "train_tokens_per_second": 5956.72
    },
    {
      "epoch": 1.2022988505747128,
      "grad_norm": 0.9171606302261353,
      "learning_rate": 7.988220136204677e-05,
      "loss": 1.5055,
      "num_input_tokens_seen": 19766992,
      "step": 3269,
      "train_runtime": 3318.4479,
      "train_tokens_per_second": 5956.698
    },
    {
      "epoch": 1.2026666666666666,
      "grad_norm": 0.9037192463874817,
      "learning_rate": 7.984538928768636e-05,
      "loss": 1.4344,
      "num_input_tokens_seen": 19772752,
      "step": 3270,
      "train_runtime": 3319.3973,
      "train_tokens_per_second": 5956.73
    },
    {
      "epoch": 1.2030344827586208,
      "grad_norm": 0.9305993318557739,
      "learning_rate": 7.980857721332597e-05,
      "loss": 1.505,
      "num_input_tokens_seen": 19778768,
      "step": 3271,
      "train_runtime": 3320.9448,
      "train_tokens_per_second": 5955.765
    },
    {
      "epoch": 1.2034022988505746,
      "grad_norm": 0.8040859699249268,
      "learning_rate": 7.977176513896558e-05,
      "loss": 1.1875,
      "num_input_tokens_seen": 19786816,
      "step": 3272,
      "train_runtime": 3322.1963,
      "train_tokens_per_second": 5955.944
    },
    {
      "epoch": 1.2037701149425288,
      "grad_norm": 0.8995404839515686,
      "learning_rate": 7.973495306460519e-05,
      "loss": 1.4169,
      "num_input_tokens_seen": 19792960,
      "step": 3273,
      "train_runtime": 3323.216,
      "train_tokens_per_second": 5955.965
    },
    {
      "epoch": 1.2041379310344829,
      "grad_norm": 0.9982839822769165,
      "learning_rate": 7.969814099024481e-05,
      "loss": 1.4051,
      "num_input_tokens_seen": 19797424,
      "step": 3274,
      "train_runtime": 3324.0011,
      "train_tokens_per_second": 5955.902
    },
    {
      "epoch": 1.2045057471264369,
      "grad_norm": 0.9728823304176331,
      "learning_rate": 7.966132891588442e-05,
      "loss": 1.5025,
      "num_input_tokens_seen": 19802656,
      "step": 3275,
      "train_runtime": 3324.8707,
      "train_tokens_per_second": 5955.918
    },
    {
      "epoch": 1.204873563218391,
      "grad_norm": 0.9772136211395264,
      "learning_rate": 7.962451684152403e-05,
      "loss": 1.4994,
      "num_input_tokens_seen": 19808768,
      "step": 3276,
      "train_runtime": 3325.8597,
      "train_tokens_per_second": 5955.984
    },
    {
      "epoch": 1.205241379310345,
      "grad_norm": 0.9749705791473389,
      "learning_rate": 7.958770476716362e-05,
      "loss": 1.6572,
      "num_input_tokens_seen": 19813712,
      "step": 3277,
      "train_runtime": 3326.6962,
      "train_tokens_per_second": 5955.973
    },
    {
      "epoch": 1.205609195402299,
      "grad_norm": 0.993110179901123,
      "learning_rate": 7.955089269280325e-05,
      "loss": 1.719,
      "num_input_tokens_seen": 19820544,
      "step": 3278,
      "train_runtime": 3327.792,
      "train_tokens_per_second": 5956.064
    },
    {
      "epoch": 1.205977011494253,
      "grad_norm": 1.0478558540344238,
      "learning_rate": 7.951408061844285e-05,
      "loss": 1.5923,
      "num_input_tokens_seen": 19826768,
      "step": 3279,
      "train_runtime": 3328.8087,
      "train_tokens_per_second": 5956.115
    },
    {
      "epoch": 1.206344827586207,
      "grad_norm": 0.8239120244979858,
      "learning_rate": 7.947726854408246e-05,
      "loss": 1.4741,
      "num_input_tokens_seen": 19834864,
      "step": 3280,
      "train_runtime": 3330.076,
      "train_tokens_per_second": 5956.28
    },
    {
      "epoch": 1.206712643678161,
      "grad_norm": 0.9192394614219666,
      "learning_rate": 7.944045646972207e-05,
      "loss": 1.4886,
      "num_input_tokens_seen": 19841392,
      "step": 3281,
      "train_runtime": 3331.1417,
      "train_tokens_per_second": 5956.334
    },
    {
      "epoch": 1.207080459770115,
      "grad_norm": 0.863332211971283,
      "learning_rate": 7.940364439536168e-05,
      "loss": 1.434,
      "num_input_tokens_seen": 19847728,
      "step": 3282,
      "train_runtime": 3332.1769,
      "train_tokens_per_second": 5956.385
    },
    {
      "epoch": 1.207448275862069,
      "grad_norm": 0.8875561356544495,
      "learning_rate": 7.936683232100129e-05,
      "loss": 1.4211,
      "num_input_tokens_seen": 19855440,
      "step": 3283,
      "train_runtime": 3333.3986,
      "train_tokens_per_second": 5956.515
    },
    {
      "epoch": 1.207816091954023,
      "grad_norm": 0.9878827333450317,
      "learning_rate": 7.93300202466409e-05,
      "loss": 1.6643,
      "num_input_tokens_seen": 19862496,
      "step": 3284,
      "train_runtime": 3334.5049,
      "train_tokens_per_second": 5956.655
    },
    {
      "epoch": 1.208183908045977,
      "grad_norm": 0.9605018496513367,
      "learning_rate": 7.929320817228051e-05,
      "loss": 1.507,
      "num_input_tokens_seen": 19869072,
      "step": 3285,
      "train_runtime": 3335.5712,
      "train_tokens_per_second": 5956.723
    },
    {
      "epoch": 1.208551724137931,
      "grad_norm": 0.8948702216148376,
      "learning_rate": 7.925639609792013e-05,
      "loss": 1.7011,
      "num_input_tokens_seen": 19875024,
      "step": 3286,
      "train_runtime": 3336.538,
      "train_tokens_per_second": 5956.78
    },
    {
      "epoch": 1.2089195402298851,
      "grad_norm": 0.91116863489151,
      "learning_rate": 7.921958402355974e-05,
      "loss": 1.5206,
      "num_input_tokens_seen": 19880000,
      "step": 3287,
      "train_runtime": 3337.3888,
      "train_tokens_per_second": 5956.753
    },
    {
      "epoch": 1.2092873563218391,
      "grad_norm": 0.9455022811889648,
      "learning_rate": 7.918277194919933e-05,
      "loss": 1.5129,
      "num_input_tokens_seen": 19887952,
      "step": 3288,
      "train_runtime": 3338.6477,
      "train_tokens_per_second": 5956.888
    },
    {
      "epoch": 1.2096551724137932,
      "grad_norm": 0.9699750542640686,
      "learning_rate": 7.914595987483894e-05,
      "loss": 1.4118,
      "num_input_tokens_seen": 19894016,
      "step": 3289,
      "train_runtime": 3339.6358,
      "train_tokens_per_second": 5956.942
    },
    {
      "epoch": 1.2100229885057472,
      "grad_norm": 0.9239062666893005,
      "learning_rate": 7.910914780047855e-05,
      "loss": 1.6096,
      "num_input_tokens_seen": 19899472,
      "step": 3290,
      "train_runtime": 3340.5408,
      "train_tokens_per_second": 5956.961
    },
    {
      "epoch": 1.2103908045977012,
      "grad_norm": 0.9467385411262512,
      "learning_rate": 7.907233572611818e-05,
      "loss": 1.4547,
      "num_input_tokens_seen": 19906336,
      "step": 3291,
      "train_runtime": 3341.6436,
      "train_tokens_per_second": 5957.049
    },
    {
      "epoch": 1.2107586206896552,
      "grad_norm": 0.9696337580680847,
      "learning_rate": 7.903552365175778e-05,
      "loss": 1.4063,
      "num_input_tokens_seen": 19912016,
      "step": 3292,
      "train_runtime": 3342.558,
      "train_tokens_per_second": 5957.119
    },
    {
      "epoch": 1.2111264367816093,
      "grad_norm": 0.8752827048301697,
      "learning_rate": 7.899871157739739e-05,
      "loss": 1.5106,
      "num_input_tokens_seen": 19917504,
      "step": 3293,
      "train_runtime": 3343.4682,
      "train_tokens_per_second": 5957.139
    },
    {
      "epoch": 1.2114942528735633,
      "grad_norm": 0.923297643661499,
      "learning_rate": 7.8961899503037e-05,
      "loss": 1.6722,
      "num_input_tokens_seen": 19922128,
      "step": 3294,
      "train_runtime": 3344.2603,
      "train_tokens_per_second": 5957.11
    },
    {
      "epoch": 1.2118620689655173,
      "grad_norm": 1.0017191171646118,
      "learning_rate": 7.892508742867661e-05,
      "loss": 1.5588,
      "num_input_tokens_seen": 19929072,
      "step": 3295,
      "train_runtime": 3345.3518,
      "train_tokens_per_second": 5957.242
    },
    {
      "epoch": 1.2122298850574713,
      "grad_norm": 0.9738130569458008,
      "learning_rate": 7.888827535431622e-05,
      "loss": 1.7527,
      "num_input_tokens_seen": 19933536,
      "step": 3296,
      "train_runtime": 3346.1158,
      "train_tokens_per_second": 5957.216
    },
    {
      "epoch": 1.2125977011494253,
      "grad_norm": 0.9487571120262146,
      "learning_rate": 7.885146327995583e-05,
      "loss": 1.3576,
      "num_input_tokens_seen": 19938496,
      "step": 3297,
      "train_runtime": 3346.9599,
      "train_tokens_per_second": 5957.196
    },
    {
      "epoch": 1.2129655172413794,
      "grad_norm": 0.9088518619537354,
      "learning_rate": 7.881465120559544e-05,
      "loss": 1.635,
      "num_input_tokens_seen": 19943104,
      "step": 3298,
      "train_runtime": 3347.7464,
      "train_tokens_per_second": 5957.173
    },
    {
      "epoch": 1.2133333333333334,
      "grad_norm": 1.029585599899292,
      "learning_rate": 7.877783913123506e-05,
      "loss": 1.3664,
      "num_input_tokens_seen": 19948304,
      "step": 3299,
      "train_runtime": 3348.6192,
      "train_tokens_per_second": 5957.173
    },
    {
      "epoch": 1.2137011494252874,
      "grad_norm": 0.9265835881233215,
      "learning_rate": 7.874102705687466e-05,
      "loss": 1.6813,
      "num_input_tokens_seen": 19954368,
      "step": 3300,
      "train_runtime": 3349.6059,
      "train_tokens_per_second": 5957.229
    },
    {
      "epoch": 1.2140689655172414,
      "grad_norm": 0.877527117729187,
      "learning_rate": 7.870421498251426e-05,
      "loss": 1.4318,
      "num_input_tokens_seen": 19960304,
      "step": 3301,
      "train_runtime": 3351.1194,
      "train_tokens_per_second": 5956.309
    },
    {
      "epoch": 1.2144367816091954,
      "grad_norm": 0.9234224557876587,
      "learning_rate": 7.866740290815387e-05,
      "loss": 1.6335,
      "num_input_tokens_seen": 19965232,
      "step": 3302,
      "train_runtime": 3351.962,
      "train_tokens_per_second": 5956.282
    },
    {
      "epoch": 1.2148045977011495,
      "grad_norm": 0.9050441384315491,
      "learning_rate": 7.863059083379348e-05,
      "loss": 1.6388,
      "num_input_tokens_seen": 19972032,
      "step": 3303,
      "train_runtime": 3353.0618,
      "train_tokens_per_second": 5956.357
    },
    {
      "epoch": 1.2151724137931035,
      "grad_norm": 0.9452304840087891,
      "learning_rate": 7.85937787594331e-05,
      "loss": 1.4423,
      "num_input_tokens_seen": 19977728,
      "step": 3304,
      "train_runtime": 3353.9726,
      "train_tokens_per_second": 5956.437
    },
    {
      "epoch": 1.2155402298850575,
      "grad_norm": 0.9913032650947571,
      "learning_rate": 7.855696668507271e-05,
      "loss": 1.5538,
      "num_input_tokens_seen": 19983664,
      "step": 3305,
      "train_runtime": 3354.9354,
      "train_tokens_per_second": 5956.497
    },
    {
      "epoch": 1.2159080459770115,
      "grad_norm": 0.9770853519439697,
      "learning_rate": 7.852015461071232e-05,
      "loss": 1.506,
      "num_input_tokens_seen": 19989632,
      "step": 3306,
      "train_runtime": 3355.9187,
      "train_tokens_per_second": 5956.53
    },
    {
      "epoch": 1.2162758620689655,
      "grad_norm": 0.8538823127746582,
      "learning_rate": 7.848334253635192e-05,
      "loss": 1.5763,
      "num_input_tokens_seen": 19997840,
      "step": 3307,
      "train_runtime": 3357.1881,
      "train_tokens_per_second": 5956.723
    },
    {
      "epoch": 1.2166436781609196,
      "grad_norm": 0.895853579044342,
      "learning_rate": 7.844653046199154e-05,
      "loss": 1.271,
      "num_input_tokens_seen": 20003536,
      "step": 3308,
      "train_runtime": 3358.1215,
      "train_tokens_per_second": 5956.764
    },
    {
      "epoch": 1.2170114942528736,
      "grad_norm": 0.9088481068611145,
      "learning_rate": 7.840971838763115e-05,
      "loss": 1.5255,
      "num_input_tokens_seen": 20009440,
      "step": 3309,
      "train_runtime": 3359.1074,
      "train_tokens_per_second": 5956.773
    },
    {
      "epoch": 1.2173793103448276,
      "grad_norm": 0.8840969204902649,
      "learning_rate": 7.837290631327076e-05,
      "loss": 1.3131,
      "num_input_tokens_seen": 20015936,
      "step": 3310,
      "train_runtime": 3360.1347,
      "train_tokens_per_second": 5956.885
    },
    {
      "epoch": 1.2177471264367816,
      "grad_norm": 0.9156765341758728,
      "learning_rate": 7.833609423891037e-05,
      "loss": 1.5565,
      "num_input_tokens_seen": 20021232,
      "step": 3311,
      "train_runtime": 3361.0099,
      "train_tokens_per_second": 5956.91
    },
    {
      "epoch": 1.2181149425287356,
      "grad_norm": 0.9763259887695312,
      "learning_rate": 7.829928216454998e-05,
      "loss": 1.5611,
      "num_input_tokens_seen": 20026320,
      "step": 3312,
      "train_runtime": 3361.8524,
      "train_tokens_per_second": 5956.93
    },
    {
      "epoch": 1.2184827586206897,
      "grad_norm": 0.9076117277145386,
      "learning_rate": 7.826247009018958e-05,
      "loss": 1.5586,
      "num_input_tokens_seen": 20032352,
      "step": 3313,
      "train_runtime": 3362.8099,
      "train_tokens_per_second": 5957.028
    },
    {
      "epoch": 1.2188505747126437,
      "grad_norm": 0.9422586560249329,
      "learning_rate": 7.822565801582919e-05,
      "loss": 1.7319,
      "num_input_tokens_seen": 20038032,
      "step": 3314,
      "train_runtime": 3363.7472,
      "train_tokens_per_second": 5957.056
    },
    {
      "epoch": 1.2192183908045977,
      "grad_norm": 0.915673553943634,
      "learning_rate": 7.81888459414688e-05,
      "loss": 1.6474,
      "num_input_tokens_seen": 20044608,
      "step": 3315,
      "train_runtime": 3364.8075,
      "train_tokens_per_second": 5957.134
    },
    {
      "epoch": 1.2195862068965517,
      "grad_norm": 0.9027487635612488,
      "learning_rate": 7.815203386710842e-05,
      "loss": 1.5432,
      "num_input_tokens_seen": 20049632,
      "step": 3316,
      "train_runtime": 3365.6536,
      "train_tokens_per_second": 5957.129
    },
    {
      "epoch": 1.2199540229885057,
      "grad_norm": 0.9269093871116638,
      "learning_rate": 7.811522179274803e-05,
      "loss": 1.4685,
      "num_input_tokens_seen": 20057664,
      "step": 3317,
      "train_runtime": 3366.924,
      "train_tokens_per_second": 5957.267
    },
    {
      "epoch": 1.2203218390804598,
      "grad_norm": 0.9414156079292297,
      "learning_rate": 7.807840971838763e-05,
      "loss": 1.3832,
      "num_input_tokens_seen": 20061456,
      "step": 3318,
      "train_runtime": 3367.6103,
      "train_tokens_per_second": 5957.179
    },
    {
      "epoch": 1.2206896551724138,
      "grad_norm": 0.9017217755317688,
      "learning_rate": 7.804159764402724e-05,
      "loss": 1.3757,
      "num_input_tokens_seen": 20067840,
      "step": 3319,
      "train_runtime": 3368.6296,
      "train_tokens_per_second": 5957.271
    },
    {
      "epoch": 1.2210574712643678,
      "grad_norm": 0.8100501894950867,
      "learning_rate": 7.800478556966685e-05,
      "loss": 1.4528,
      "num_input_tokens_seen": 20074448,
      "step": 3320,
      "train_runtime": 3369.6969,
      "train_tokens_per_second": 5957.345
    },
    {
      "epoch": 1.2214252873563218,
      "grad_norm": 0.9677010774612427,
      "learning_rate": 7.796797349530647e-05,
      "loss": 1.6126,
      "num_input_tokens_seen": 20080608,
      "step": 3321,
      "train_runtime": 3370.6894,
      "train_tokens_per_second": 5957.419
    },
    {
      "epoch": 1.2217931034482759,
      "grad_norm": 0.9213995933532715,
      "learning_rate": 7.793116142094608e-05,
      "loss": 1.4875,
      "num_input_tokens_seen": 20086320,
      "step": 3322,
      "train_runtime": 3371.654,
      "train_tokens_per_second": 5957.409
    },
    {
      "epoch": 1.2221609195402299,
      "grad_norm": 0.862701416015625,
      "learning_rate": 7.789434934658569e-05,
      "loss": 1.3909,
      "num_input_tokens_seen": 20092064,
      "step": 3323,
      "train_runtime": 3372.6159,
      "train_tokens_per_second": 5957.412
    },
    {
      "epoch": 1.222528735632184,
      "grad_norm": 0.9248855710029602,
      "learning_rate": 7.78575372722253e-05,
      "loss": 1.5429,
      "num_input_tokens_seen": 20097008,
      "step": 3324,
      "train_runtime": 3373.4714,
      "train_tokens_per_second": 5957.367
    },
    {
      "epoch": 1.222896551724138,
      "grad_norm": 0.8546602129936218,
      "learning_rate": 7.78207251978649e-05,
      "loss": 1.5184,
      "num_input_tokens_seen": 20105024,
      "step": 3325,
      "train_runtime": 3374.714,
      "train_tokens_per_second": 5957.549
    },
    {
      "epoch": 1.223264367816092,
      "grad_norm": 0.9670182466506958,
      "learning_rate": 7.778391312350451e-05,
      "loss": 1.48,
      "num_input_tokens_seen": 20110864,
      "step": 3326,
      "train_runtime": 3375.6595,
      "train_tokens_per_second": 5957.61
    },
    {
      "epoch": 1.223632183908046,
      "grad_norm": 0.929133951663971,
      "learning_rate": 7.774710104914412e-05,
      "loss": 1.5064,
      "num_input_tokens_seen": 20117952,
      "step": 3327,
      "train_runtime": 3376.7944,
      "train_tokens_per_second": 5957.707
    },
    {
      "epoch": 1.224,
      "grad_norm": 0.7855772972106934,
      "learning_rate": 7.771028897478373e-05,
      "loss": 1.4014,
      "num_input_tokens_seen": 20126416,
      "step": 3328,
      "train_runtime": 3378.1162,
      "train_tokens_per_second": 5957.881
    },
    {
      "epoch": 1.224367816091954,
      "grad_norm": 0.9452453851699829,
      "learning_rate": 7.767347690042335e-05,
      "loss": 1.415,
      "num_input_tokens_seen": 20133056,
      "step": 3329,
      "train_runtime": 3379.1673,
      "train_tokens_per_second": 5957.993
    },
    {
      "epoch": 1.224735632183908,
      "grad_norm": 0.8732830286026001,
      "learning_rate": 7.763666482606295e-05,
      "loss": 1.4347,
      "num_input_tokens_seen": 20140016,
      "step": 3330,
      "train_runtime": 3380.2947,
      "train_tokens_per_second": 5958.065
    },
    {
      "epoch": 1.225103448275862,
      "grad_norm": 0.8743560910224915,
      "learning_rate": 7.759985275170256e-05,
      "loss": 1.613,
      "num_input_tokens_seen": 20145344,
      "step": 3331,
      "train_runtime": 3381.6724,
      "train_tokens_per_second": 5957.213
    },
    {
      "epoch": 1.225471264367816,
      "grad_norm": 0.9391189813613892,
      "learning_rate": 7.756304067734217e-05,
      "loss": 1.4375,
      "num_input_tokens_seen": 20153328,
      "step": 3332,
      "train_runtime": 3382.9303,
      "train_tokens_per_second": 5957.358
    },
    {
      "epoch": 1.22583908045977,
      "grad_norm": 0.9103835821151733,
      "learning_rate": 7.752622860298178e-05,
      "loss": 1.3814,
      "num_input_tokens_seen": 20159488,
      "step": 3333,
      "train_runtime": 3383.9495,
      "train_tokens_per_second": 5957.384
    },
    {
      "epoch": 1.226206896551724,
      "grad_norm": 0.9309260845184326,
      "learning_rate": 7.74894165286214e-05,
      "loss": 1.5395,
      "num_input_tokens_seen": 20164928,
      "step": 3334,
      "train_runtime": 3384.8435,
      "train_tokens_per_second": 5957.418
    },
    {
      "epoch": 1.2265747126436781,
      "grad_norm": 0.8227443695068359,
      "learning_rate": 7.745260445426101e-05,
      "loss": 1.306,
      "num_input_tokens_seen": 20172464,
      "step": 3335,
      "train_runtime": 3386.0466,
      "train_tokens_per_second": 5957.527
    },
    {
      "epoch": 1.2269425287356321,
      "grad_norm": 0.9939490556716919,
      "learning_rate": 7.741579237990062e-05,
      "loss": 1.6273,
      "num_input_tokens_seen": 20177472,
      "step": 3336,
      "train_runtime": 3386.8994,
      "train_tokens_per_second": 5957.506
    },
    {
      "epoch": 1.2273103448275862,
      "grad_norm": 0.9474799633026123,
      "learning_rate": 7.737898030554021e-05,
      "loss": 1.497,
      "num_input_tokens_seen": 20182352,
      "step": 3337,
      "train_runtime": 3387.7387,
      "train_tokens_per_second": 5957.47
    },
    {
      "epoch": 1.2276781609195402,
      "grad_norm": 0.8516086935997009,
      "learning_rate": 7.734216823117983e-05,
      "loss": 1.5328,
      "num_input_tokens_seen": 20188432,
      "step": 3338,
      "train_runtime": 3388.7228,
      "train_tokens_per_second": 5957.534
    },
    {
      "epoch": 1.2280459770114942,
      "grad_norm": 0.8753452301025391,
      "learning_rate": 7.730535615681944e-05,
      "loss": 1.5309,
      "num_input_tokens_seen": 20194128,
      "step": 3339,
      "train_runtime": 3389.6455,
      "train_tokens_per_second": 5957.593
    },
    {
      "epoch": 1.2284137931034482,
      "grad_norm": 0.9226855635643005,
      "learning_rate": 7.726854408245905e-05,
      "loss": 1.46,
      "num_input_tokens_seen": 20200960,
      "step": 3340,
      "train_runtime": 3390.7216,
      "train_tokens_per_second": 5957.717
    },
    {
      "epoch": 1.2287816091954022,
      "grad_norm": 0.8776497840881348,
      "learning_rate": 7.723173200809866e-05,
      "loss": 1.4609,
      "num_input_tokens_seen": 20208640,
      "step": 3341,
      "train_runtime": 3391.9078,
      "train_tokens_per_second": 5957.898
    },
    {
      "epoch": 1.2291494252873563,
      "grad_norm": 0.895530641078949,
      "learning_rate": 7.719491993373827e-05,
      "loss": 1.4375,
      "num_input_tokens_seen": 20214448,
      "step": 3342,
      "train_runtime": 3392.8552,
      "train_tokens_per_second": 5957.946
    },
    {
      "epoch": 1.2295172413793103,
      "grad_norm": 0.9471902847290039,
      "learning_rate": 7.715810785937788e-05,
      "loss": 1.6677,
      "num_input_tokens_seen": 20219600,
      "step": 3343,
      "train_runtime": 3393.7461,
      "train_tokens_per_second": 5957.9
    },
    {
      "epoch": 1.2298850574712643,
      "grad_norm": 0.8841742873191833,
      "learning_rate": 7.712129578501749e-05,
      "loss": 1.5955,
      "num_input_tokens_seen": 20224944,
      "step": 3344,
      "train_runtime": 3394.6294,
      "train_tokens_per_second": 5957.924
    },
    {
      "epoch": 1.2302528735632183,
      "grad_norm": 0.9034956097602844,
      "learning_rate": 7.70844837106571e-05,
      "loss": 1.5876,
      "num_input_tokens_seen": 20229824,
      "step": 3345,
      "train_runtime": 3395.4545,
      "train_tokens_per_second": 5957.913
    },
    {
      "epoch": 1.2306206896551724,
      "grad_norm": 0.9193718433380127,
      "learning_rate": 7.704767163629672e-05,
      "loss": 1.4901,
      "num_input_tokens_seen": 20235920,
      "step": 3346,
      "train_runtime": 3396.4454,
      "train_tokens_per_second": 5957.97
    },
    {
      "epoch": 1.2309885057471264,
      "grad_norm": 0.9795694947242737,
      "learning_rate": 7.701085956193633e-05,
      "loss": 1.628,
      "num_input_tokens_seen": 20240736,
      "step": 3347,
      "train_runtime": 3397.2627,
      "train_tokens_per_second": 5957.954
    },
    {
      "epoch": 1.2313563218390804,
      "grad_norm": 0.9909173250198364,
      "learning_rate": 7.697404748757592e-05,
      "loss": 1.55,
      "num_input_tokens_seen": 20246336,
      "step": 3348,
      "train_runtime": 3398.174,
      "train_tokens_per_second": 5958.005
    },
    {
      "epoch": 1.2317241379310344,
      "grad_norm": 0.9607061147689819,
      "learning_rate": 7.693723541321553e-05,
      "loss": 1.6975,
      "num_input_tokens_seen": 20252016,
      "step": 3349,
      "train_runtime": 3399.0909,
      "train_tokens_per_second": 5958.068
    },
    {
      "epoch": 1.2320919540229884,
      "grad_norm": 0.9655289053916931,
      "learning_rate": 7.690042333885514e-05,
      "loss": 1.4483,
      "num_input_tokens_seen": 20256976,
      "step": 3350,
      "train_runtime": 3399.9044,
      "train_tokens_per_second": 5958.102
    },
    {
      "epoch": 1.2324597701149425,
      "grad_norm": 0.9344881176948547,
      "learning_rate": 7.686361126449476e-05,
      "loss": 1.6377,
      "num_input_tokens_seen": 20262368,
      "step": 3351,
      "train_runtime": 3400.7954,
      "train_tokens_per_second": 5958.126
    },
    {
      "epoch": 1.2328275862068965,
      "grad_norm": 0.988397479057312,
      "learning_rate": 7.682679919013437e-05,
      "loss": 1.5678,
      "num_input_tokens_seen": 20267104,
      "step": 3352,
      "train_runtime": 3401.5778,
      "train_tokens_per_second": 5958.148
    },
    {
      "epoch": 1.2331954022988505,
      "grad_norm": 0.9660370945930481,
      "learning_rate": 7.678998711577398e-05,
      "loss": 1.7162,
      "num_input_tokens_seen": 20272720,
      "step": 3353,
      "train_runtime": 3402.4825,
      "train_tokens_per_second": 5958.214
    },
    {
      "epoch": 1.2335632183908045,
      "grad_norm": 0.774196982383728,
      "learning_rate": 7.675317504141359e-05,
      "loss": 1.535,
      "num_input_tokens_seen": 20279856,
      "step": 3354,
      "train_runtime": 3403.62,
      "train_tokens_per_second": 5958.32
    },
    {
      "epoch": 1.2339310344827585,
      "grad_norm": 0.8448355793952942,
      "learning_rate": 7.67163629670532e-05,
      "loss": 1.3612,
      "num_input_tokens_seen": 20284512,
      "step": 3355,
      "train_runtime": 3404.4459,
      "train_tokens_per_second": 5958.242
    },
    {
      "epoch": 1.2342988505747126,
      "grad_norm": 0.8992934823036194,
      "learning_rate": 7.667955089269281e-05,
      "loss": 1.5434,
      "num_input_tokens_seen": 20289232,
      "step": 3356,
      "train_runtime": 3405.2756,
      "train_tokens_per_second": 5958.176
    },
    {
      "epoch": 1.2346666666666666,
      "grad_norm": 1.0435458421707153,
      "learning_rate": 7.664273881833242e-05,
      "loss": 1.4346,
      "num_input_tokens_seen": 20293312,
      "step": 3357,
      "train_runtime": 3406.0082,
      "train_tokens_per_second": 5958.093
    },
    {
      "epoch": 1.2350344827586206,
      "grad_norm": 0.9971229434013367,
      "learning_rate": 7.660592674397203e-05,
      "loss": 1.715,
      "num_input_tokens_seen": 20297872,
      "step": 3358,
      "train_runtime": 3406.8209,
      "train_tokens_per_second": 5958.01
    },
    {
      "epoch": 1.2354022988505746,
      "grad_norm": 0.9586120843887329,
      "learning_rate": 7.656911466961165e-05,
      "loss": 1.5779,
      "num_input_tokens_seen": 20303328,
      "step": 3359,
      "train_runtime": 3407.7414,
      "train_tokens_per_second": 5958.001
    },
    {
      "epoch": 1.2357701149425286,
      "grad_norm": 0.9365577101707458,
      "learning_rate": 7.653230259525124e-05,
      "loss": 1.3955,
      "num_input_tokens_seen": 20307904,
      "step": 3360,
      "train_runtime": 3408.5291,
      "train_tokens_per_second": 5957.967
    },
    {
      "epoch": 1.2361379310344827,
      "grad_norm": 0.8653137683868408,
      "learning_rate": 7.649549052089085e-05,
      "loss": 1.5332,
      "num_input_tokens_seen": 20313024,
      "step": 3361,
      "train_runtime": 3409.9399,
      "train_tokens_per_second": 5957.004
    },
    {
      "epoch": 1.236505747126437,
      "grad_norm": 0.9239800572395325,
      "learning_rate": 7.645867844653046e-05,
      "loss": 1.5335,
      "num_input_tokens_seen": 20318320,
      "step": 3362,
      "train_runtime": 3410.8195,
      "train_tokens_per_second": 5957.02
    },
    {
      "epoch": 1.2368735632183907,
      "grad_norm": 0.923174262046814,
      "learning_rate": 7.642186637217007e-05,
      "loss": 1.6279,
      "num_input_tokens_seen": 20323024,
      "step": 3363,
      "train_runtime": 3411.6414,
      "train_tokens_per_second": 5956.964
    },
    {
      "epoch": 1.237241379310345,
      "grad_norm": 0.973044753074646,
      "learning_rate": 7.638505429780969e-05,
      "loss": 1.527,
      "num_input_tokens_seen": 20329472,
      "step": 3364,
      "train_runtime": 3412.6722,
      "train_tokens_per_second": 5957.054
    },
    {
      "epoch": 1.2376091954022987,
      "grad_norm": 0.8200168609619141,
      "learning_rate": 7.63482422234493e-05,
      "loss": 1.4998,
      "num_input_tokens_seen": 20336240,
      "step": 3365,
      "train_runtime": 3413.7453,
      "train_tokens_per_second": 5957.164
    },
    {
      "epoch": 1.237977011494253,
      "grad_norm": 0.9200677871704102,
      "learning_rate": 7.63114301490889e-05,
      "loss": 1.4429,
      "num_input_tokens_seen": 20345504,
      "step": 3366,
      "train_runtime": 3415.1698,
      "train_tokens_per_second": 5957.392
    },
    {
      "epoch": 1.2383448275862068,
      "grad_norm": 0.936782717704773,
      "learning_rate": 7.62746180747285e-05,
      "loss": 1.5116,
      "num_input_tokens_seen": 20350368,
      "step": 3367,
      "train_runtime": 3415.9867,
      "train_tokens_per_second": 5957.391
    },
    {
      "epoch": 1.238712643678161,
      "grad_norm": 0.8578950762748718,
      "learning_rate": 7.623780600036813e-05,
      "loss": 1.621,
      "num_input_tokens_seen": 20357280,
      "step": 3368,
      "train_runtime": 3417.0924,
      "train_tokens_per_second": 5957.486
    },
    {
      "epoch": 1.2390804597701148,
      "grad_norm": 0.9597526788711548,
      "learning_rate": 7.620099392600774e-05,
      "loss": 1.6272,
      "num_input_tokens_seen": 20361760,
      "step": 3369,
      "train_runtime": 3417.8722,
      "train_tokens_per_second": 5957.437
    },
    {
      "epoch": 1.239448275862069,
      "grad_norm": 0.9248815178871155,
      "learning_rate": 7.616418185164735e-05,
      "loss": 1.4413,
      "num_input_tokens_seen": 20367568,
      "step": 3370,
      "train_runtime": 3418.8229,
      "train_tokens_per_second": 5957.48
    },
    {
      "epoch": 1.2398160919540229,
      "grad_norm": 0.9244171380996704,
      "learning_rate": 7.612736977728695e-05,
      "loss": 1.3942,
      "num_input_tokens_seen": 20378224,
      "step": 3371,
      "train_runtime": 3420.4181,
      "train_tokens_per_second": 5957.817
    },
    {
      "epoch": 1.2401839080459771,
      "grad_norm": 1.0480886697769165,
      "learning_rate": 7.609055770292656e-05,
      "loss": 1.5966,
      "num_input_tokens_seen": 20386944,
      "step": 3372,
      "train_runtime": 3421.7675,
      "train_tokens_per_second": 5958.016
    },
    {
      "epoch": 1.240551724137931,
      "grad_norm": 0.8595100045204163,
      "learning_rate": 7.605374562856617e-05,
      "loss": 1.4655,
      "num_input_tokens_seen": 20393376,
      "step": 3373,
      "train_runtime": 3422.7904,
      "train_tokens_per_second": 5958.114
    },
    {
      "epoch": 1.2409195402298852,
      "grad_norm": 0.8628688454627991,
      "learning_rate": 7.601693355420578e-05,
      "loss": 1.3638,
      "num_input_tokens_seen": 20399552,
      "step": 3374,
      "train_runtime": 3423.7885,
      "train_tokens_per_second": 5958.181
    },
    {
      "epoch": 1.2412873563218392,
      "grad_norm": 1.0187793970108032,
      "learning_rate": 7.598012147984539e-05,
      "loss": 1.515,
      "num_input_tokens_seen": 20405296,
      "step": 3375,
      "train_runtime": 3424.7322,
      "train_tokens_per_second": 5958.217
    },
    {
      "epoch": 1.2416551724137932,
      "grad_norm": 0.9930346012115479,
      "learning_rate": 7.594330940548501e-05,
      "loss": 1.5144,
      "num_input_tokens_seen": 20412224,
      "step": 3376,
      "train_runtime": 3425.8343,
      "train_tokens_per_second": 5958.322
    },
    {
      "epoch": 1.2420229885057472,
      "grad_norm": 1.0066184997558594,
      "learning_rate": 7.590649733112462e-05,
      "loss": 1.5928,
      "num_input_tokens_seen": 20417072,
      "step": 3377,
      "train_runtime": 3426.6405,
      "train_tokens_per_second": 5958.335
    },
    {
      "epoch": 1.2423908045977012,
      "grad_norm": 0.9400173425674438,
      "learning_rate": 7.586968525676422e-05,
      "loss": 1.3556,
      "num_input_tokens_seen": 20422576,
      "step": 3378,
      "train_runtime": 3427.5715,
      "train_tokens_per_second": 5958.322
    },
    {
      "epoch": 1.2427586206896553,
      "grad_norm": 0.9621667265892029,
      "learning_rate": 7.583287318240383e-05,
      "loss": 1.4669,
      "num_input_tokens_seen": 20427552,
      "step": 3379,
      "train_runtime": 3428.426,
      "train_tokens_per_second": 5958.289
    },
    {
      "epoch": 1.2431264367816093,
      "grad_norm": 0.9765159487724304,
      "learning_rate": 7.579606110804343e-05,
      "loss": 1.3966,
      "num_input_tokens_seen": 20433408,
      "step": 3380,
      "train_runtime": 3429.3912,
      "train_tokens_per_second": 5958.319
    },
    {
      "epoch": 1.2434942528735633,
      "grad_norm": 1.0458918809890747,
      "learning_rate": 7.575924903368306e-05,
      "loss": 1.4234,
      "num_input_tokens_seen": 20441296,
      "step": 3381,
      "train_runtime": 3430.6439,
      "train_tokens_per_second": 5958.443
    },
    {
      "epoch": 1.2438620689655173,
      "grad_norm": 1.007415533065796,
      "learning_rate": 7.572243695932267e-05,
      "loss": 1.6186,
      "num_input_tokens_seen": 20447840,
      "step": 3382,
      "train_runtime": 3431.6697,
      "train_tokens_per_second": 5958.569
    },
    {
      "epoch": 1.2442298850574713,
      "grad_norm": 0.8871510624885559,
      "learning_rate": 7.568562488496227e-05,
      "loss": 1.4455,
      "num_input_tokens_seen": 20453056,
      "step": 3383,
      "train_runtime": 3432.5466,
      "train_tokens_per_second": 5958.566
    },
    {
      "epoch": 1.2445977011494254,
      "grad_norm": 1.016754150390625,
      "learning_rate": 7.564881281060188e-05,
      "loss": 1.6171,
      "num_input_tokens_seen": 20457344,
      "step": 3384,
      "train_runtime": 3433.3161,
      "train_tokens_per_second": 5958.48
    },
    {
      "epoch": 1.2449655172413794,
      "grad_norm": 0.9223800897598267,
      "learning_rate": 7.561200073624149e-05,
      "loss": 1.5945,
      "num_input_tokens_seen": 20463504,
      "step": 3385,
      "train_runtime": 3434.3293,
      "train_tokens_per_second": 5958.515
    },
    {
      "epoch": 1.2453333333333334,
      "grad_norm": 0.9652520418167114,
      "learning_rate": 7.55751886618811e-05,
      "loss": 1.7901,
      "num_input_tokens_seen": 20468960,
      "step": 3386,
      "train_runtime": 3435.2045,
      "train_tokens_per_second": 5958.586
    },
    {
      "epoch": 1.2457011494252874,
      "grad_norm": 0.9440248012542725,
      "learning_rate": 7.553837658752071e-05,
      "loss": 1.5111,
      "num_input_tokens_seen": 20475712,
      "step": 3387,
      "train_runtime": 3436.3015,
      "train_tokens_per_second": 5958.648
    },
    {
      "epoch": 1.2460689655172414,
      "grad_norm": 0.993414044380188,
      "learning_rate": 7.550156451316032e-05,
      "loss": 1.4885,
      "num_input_tokens_seen": 20480352,
      "step": 3388,
      "train_runtime": 3437.0984,
      "train_tokens_per_second": 5958.617
    },
    {
      "epoch": 1.2464367816091955,
      "grad_norm": 0.8374962210655212,
      "learning_rate": 7.546475243879994e-05,
      "loss": 1.4917,
      "num_input_tokens_seen": 20488464,
      "step": 3389,
      "train_runtime": 3438.3627,
      "train_tokens_per_second": 5958.785
    },
    {
      "epoch": 1.2468045977011495,
      "grad_norm": 0.9228314757347107,
      "learning_rate": 7.542794036443954e-05,
      "loss": 1.5263,
      "num_input_tokens_seen": 20494704,
      "step": 3390,
      "train_runtime": 3439.4022,
      "train_tokens_per_second": 5958.798
    },
    {
      "epoch": 1.2471724137931035,
      "grad_norm": 0.8830683827400208,
      "learning_rate": 7.539112829007915e-05,
      "loss": 1.3694,
      "num_input_tokens_seen": 20500816,
      "step": 3391,
      "train_runtime": 3440.9345,
      "train_tokens_per_second": 5957.921
    },
    {
      "epoch": 1.2475402298850575,
      "grad_norm": 0.9051899313926697,
      "learning_rate": 7.535431621571875e-05,
      "loss": 1.5687,
      "num_input_tokens_seen": 20505360,
      "step": 3392,
      "train_runtime": 3441.6919,
      "train_tokens_per_second": 5957.93
    },
    {
      "epoch": 1.2479080459770115,
      "grad_norm": 0.9094662070274353,
      "learning_rate": 7.531750414135836e-05,
      "loss": 1.5989,
      "num_input_tokens_seen": 20511072,
      "step": 3393,
      "train_runtime": 3442.6438,
      "train_tokens_per_second": 5957.942
    },
    {
      "epoch": 1.2482758620689656,
      "grad_norm": 0.8545249104499817,
      "learning_rate": 7.528069206699799e-05,
      "loss": 1.5699,
      "num_input_tokens_seen": 20516688,
      "step": 3394,
      "train_runtime": 3443.5733,
      "train_tokens_per_second": 5957.965
    },
    {
      "epoch": 1.2486436781609196,
      "grad_norm": 0.8795822858810425,
      "learning_rate": 7.52438799926376e-05,
      "loss": 1.3246,
      "num_input_tokens_seen": 20527056,
      "step": 3395,
      "train_runtime": 3445.1291,
      "train_tokens_per_second": 5958.283
    },
    {
      "epoch": 1.2490114942528736,
      "grad_norm": 0.90467369556427,
      "learning_rate": 7.520706791827719e-05,
      "loss": 1.4496,
      "num_input_tokens_seen": 20532960,
      "step": 3396,
      "train_runtime": 3446.0976,
      "train_tokens_per_second": 5958.322
    },
    {
      "epoch": 1.2493793103448276,
      "grad_norm": 0.9017558097839355,
      "learning_rate": 7.51702558439168e-05,
      "loss": 1.6023,
      "num_input_tokens_seen": 20538544,
      "step": 3397,
      "train_runtime": 3447.0216,
      "train_tokens_per_second": 5958.345
    },
    {
      "epoch": 1.2497471264367817,
      "grad_norm": 0.9395778179168701,
      "learning_rate": 7.513344376955642e-05,
      "loss": 1.4888,
      "num_input_tokens_seen": 20544720,
      "step": 3398,
      "train_runtime": 3448.0379,
      "train_tokens_per_second": 5958.38
    },
    {
      "epoch": 1.2501149425287357,
      "grad_norm": 0.8570054173469543,
      "learning_rate": 7.509663169519603e-05,
      "loss": 1.5549,
      "num_input_tokens_seen": 20551024,
      "step": 3399,
      "train_runtime": 3449.0756,
      "train_tokens_per_second": 5958.415
    },
    {
      "epoch": 1.2504827586206897,
      "grad_norm": 0.9774261713027954,
      "learning_rate": 7.505981962083564e-05,
      "loss": 1.5408,
      "num_input_tokens_seen": 20556384,
      "step": 3400,
      "train_runtime": 3449.9732,
      "train_tokens_per_second": 5958.418
    },
    {
      "epoch": 1.2508505747126437,
      "grad_norm": 0.9991695284843445,
      "learning_rate": 7.502300754647525e-05,
      "loss": 1.7331,
      "num_input_tokens_seen": 20561056,
      "step": 3401,
      "train_runtime": 3450.791,
      "train_tokens_per_second": 5958.36
    },
    {
      "epoch": 1.2512183908045977,
      "grad_norm": 0.8814098238945007,
      "learning_rate": 7.498619547211486e-05,
      "loss": 1.3809,
      "num_input_tokens_seen": 20566096,
      "step": 3402,
      "train_runtime": 3451.658,
      "train_tokens_per_second": 5958.324
    },
    {
      "epoch": 1.2515862068965518,
      "grad_norm": 0.9068689942359924,
      "learning_rate": 7.494938339775447e-05,
      "loss": 1.4501,
      "num_input_tokens_seen": 20571120,
      "step": 3403,
      "train_runtime": 3452.5325,
      "train_tokens_per_second": 5958.27
    },
    {
      "epoch": 1.2519540229885058,
      "grad_norm": 0.8637279272079468,
      "learning_rate": 7.491257132339407e-05,
      "loss": 1.3187,
      "num_input_tokens_seen": 20577696,
      "step": 3404,
      "train_runtime": 3453.6066,
      "train_tokens_per_second": 5958.321
    },
    {
      "epoch": 1.2523218390804598,
      "grad_norm": 0.9304091334342957,
      "learning_rate": 7.487575924903368e-05,
      "loss": 1.4497,
      "num_input_tokens_seen": 20582352,
      "step": 3405,
      "train_runtime": 3454.3986,
      "train_tokens_per_second": 5958.303
    },
    {
      "epoch": 1.2526896551724138,
      "grad_norm": 0.9637987017631531,
      "learning_rate": 7.48389471746733e-05,
      "loss": 1.4554,
      "num_input_tokens_seen": 20587408,
      "step": 3406,
      "train_runtime": 3455.2257,
      "train_tokens_per_second": 5958.34
    },
    {
      "epoch": 1.2530574712643678,
      "grad_norm": 0.9325391054153442,
      "learning_rate": 7.480213510031291e-05,
      "loss": 1.5294,
      "num_input_tokens_seen": 20593744,
      "step": 3407,
      "train_runtime": 3456.2635,
      "train_tokens_per_second": 5958.384
    },
    {
      "epoch": 1.2534252873563219,
      "grad_norm": 0.9130005240440369,
      "learning_rate": 7.476532302595251e-05,
      "loss": 1.5144,
      "num_input_tokens_seen": 20601584,
      "step": 3408,
      "train_runtime": 3457.487,
      "train_tokens_per_second": 5958.543
    },
    {
      "epoch": 1.2537931034482759,
      "grad_norm": 0.9999523162841797,
      "learning_rate": 7.472851095159212e-05,
      "loss": 1.5391,
      "num_input_tokens_seen": 20608688,
      "step": 3409,
      "train_runtime": 3458.6463,
      "train_tokens_per_second": 5958.6
    },
    {
      "epoch": 1.25416091954023,
      "grad_norm": 0.9348831176757812,
      "learning_rate": 7.469169887723173e-05,
      "loss": 1.5043,
      "num_input_tokens_seen": 20613488,
      "step": 3410,
      "train_runtime": 3459.4778,
      "train_tokens_per_second": 5958.555
    },
    {
      "epoch": 1.254528735632184,
      "grad_norm": 0.971004843711853,
      "learning_rate": 7.465488680287135e-05,
      "loss": 1.4837,
      "num_input_tokens_seen": 20620208,
      "step": 3411,
      "train_runtime": 3460.5527,
      "train_tokens_per_second": 5958.646
    },
    {
      "epoch": 1.254896551724138,
      "grad_norm": 0.7924219965934753,
      "learning_rate": 7.461807472851096e-05,
      "loss": 1.3498,
      "num_input_tokens_seen": 20630672,
      "step": 3412,
      "train_runtime": 3462.1553,
      "train_tokens_per_second": 5958.91
    },
    {
      "epoch": 1.255264367816092,
      "grad_norm": 0.8895764350891113,
      "learning_rate": 7.458126265415057e-05,
      "loss": 1.4805,
      "num_input_tokens_seen": 20635568,
      "step": 3413,
      "train_runtime": 3462.9937,
      "train_tokens_per_second": 5958.881
    },
    {
      "epoch": 1.255632183908046,
      "grad_norm": 0.9732569456100464,
      "learning_rate": 7.454445057979018e-05,
      "loss": 1.478,
      "num_input_tokens_seen": 20642944,
      "step": 3414,
      "train_runtime": 3464.1752,
      "train_tokens_per_second": 5958.978
    },
    {
      "epoch": 1.256,
      "grad_norm": 0.9345213174819946,
      "learning_rate": 7.450763850542979e-05,
      "loss": 1.5453,
      "num_input_tokens_seen": 20649632,
      "step": 3415,
      "train_runtime": 3465.2542,
      "train_tokens_per_second": 5959.053
    },
    {
      "epoch": 1.256367816091954,
      "grad_norm": 0.9068306684494019,
      "learning_rate": 7.44708264310694e-05,
      "loss": 1.5485,
      "num_input_tokens_seen": 20655520,
      "step": 3416,
      "train_runtime": 3466.2388,
      "train_tokens_per_second": 5959.059
    },
    {
      "epoch": 1.256735632183908,
      "grad_norm": 0.9697032570838928,
      "learning_rate": 7.4434014356709e-05,
      "loss": 1.6231,
      "num_input_tokens_seen": 20661120,
      "step": 3417,
      "train_runtime": 3467.1546,
      "train_tokens_per_second": 5959.1
    },
    {
      "epoch": 1.257103448275862,
      "grad_norm": 0.9295088648796082,
      "learning_rate": 7.439720228234861e-05,
      "loss": 1.4801,
      "num_input_tokens_seen": 20667168,
      "step": 3418,
      "train_runtime": 3468.1497,
      "train_tokens_per_second": 5959.134
    },
    {
      "epoch": 1.257471264367816,
      "grad_norm": 0.9032222628593445,
      "learning_rate": 7.436039020798824e-05,
      "loss": 1.5464,
      "num_input_tokens_seen": 20674848,
      "step": 3419,
      "train_runtime": 3469.3797,
      "train_tokens_per_second": 5959.235
    },
    {
      "epoch": 1.25783908045977,
      "grad_norm": 0.858242928981781,
      "learning_rate": 7.432357813362783e-05,
      "loss": 1.4403,
      "num_input_tokens_seen": 20683024,
      "step": 3420,
      "train_runtime": 3470.6399,
      "train_tokens_per_second": 5959.427
    },
    {
      "epoch": 1.2582068965517241,
      "grad_norm": 0.8810915350914001,
      "learning_rate": 7.428676605926744e-05,
      "loss": 1.4443,
      "num_input_tokens_seen": 20690544,
      "step": 3421,
      "train_runtime": 3472.3821,
      "train_tokens_per_second": 5958.602
    },
    {
      "epoch": 1.2585747126436782,
      "grad_norm": 0.9798994660377502,
      "learning_rate": 7.424995398490705e-05,
      "loss": 1.5493,
      "num_input_tokens_seen": 20696432,
      "step": 3422,
      "train_runtime": 3473.3663,
      "train_tokens_per_second": 5958.609
    },
    {
      "epoch": 1.2589425287356322,
      "grad_norm": 0.9870070815086365,
      "learning_rate": 7.421314191054666e-05,
      "loss": 1.5475,
      "num_input_tokens_seen": 20701104,
      "step": 3423,
      "train_runtime": 3474.1488,
      "train_tokens_per_second": 5958.612
    },
    {
      "epoch": 1.2593103448275862,
      "grad_norm": 0.860073983669281,
      "learning_rate": 7.417632983618628e-05,
      "loss": 1.7247,
      "num_input_tokens_seen": 20708016,
      "step": 3424,
      "train_runtime": 3475.2452,
      "train_tokens_per_second": 5958.721
    },
    {
      "epoch": 1.2596781609195402,
      "grad_norm": 0.8881399035453796,
      "learning_rate": 7.413951776182589e-05,
      "loss": 1.541,
      "num_input_tokens_seen": 20714336,
      "step": 3425,
      "train_runtime": 3476.2573,
      "train_tokens_per_second": 5958.804
    },
    {
      "epoch": 1.2600459770114942,
      "grad_norm": 0.9642382264137268,
      "learning_rate": 7.410270568746548e-05,
      "loss": 1.6147,
      "num_input_tokens_seen": 20719104,
      "step": 3426,
      "train_runtime": 3477.0588,
      "train_tokens_per_second": 5958.802
    },
    {
      "epoch": 1.2604137931034483,
      "grad_norm": 1.0019463300704956,
      "learning_rate": 7.406589361310509e-05,
      "loss": 1.2902,
      "num_input_tokens_seen": 20725072,
      "step": 3427,
      "train_runtime": 3478.0528,
      "train_tokens_per_second": 5958.815
    },
    {
      "epoch": 1.2607816091954023,
      "grad_norm": 0.9634602665901184,
      "learning_rate": 7.402908153874472e-05,
      "loss": 1.4813,
      "num_input_tokens_seen": 20730192,
      "step": 3428,
      "train_runtime": 3478.9122,
      "train_tokens_per_second": 5958.814
    },
    {
      "epoch": 1.2611494252873563,
      "grad_norm": 0.8677273988723755,
      "learning_rate": 7.399226946438432e-05,
      "loss": 1.3573,
      "num_input_tokens_seen": 20734928,
      "step": 3429,
      "train_runtime": 3479.7197,
      "train_tokens_per_second": 5958.793
    },
    {
      "epoch": 1.2615172413793103,
      "grad_norm": 0.9803272485733032,
      "learning_rate": 7.395545739002393e-05,
      "loss": 1.5141,
      "num_input_tokens_seen": 20741504,
      "step": 3430,
      "train_runtime": 3480.7922,
      "train_tokens_per_second": 5958.846
    },
    {
      "epoch": 1.2618850574712643,
      "grad_norm": 0.8590312004089355,
      "learning_rate": 7.391864531566354e-05,
      "loss": 1.4631,
      "num_input_tokens_seen": 20752000,
      "step": 3431,
      "train_runtime": 3482.3606,
      "train_tokens_per_second": 5959.176
    },
    {
      "epoch": 1.2622528735632184,
      "grad_norm": 0.8663723468780518,
      "learning_rate": 7.388183324130315e-05,
      "loss": 1.5808,
      "num_input_tokens_seen": 20758336,
      "step": 3432,
      "train_runtime": 3483.3858,
      "train_tokens_per_second": 5959.241
    },
    {
      "epoch": 1.2626206896551724,
      "grad_norm": 0.9491639733314514,
      "learning_rate": 7.384502116694276e-05,
      "loss": 1.7055,
      "num_input_tokens_seen": 20762832,
      "step": 3433,
      "train_runtime": 3484.198,
      "train_tokens_per_second": 5959.142
    },
    {
      "epoch": 1.2629885057471264,
      "grad_norm": 0.9592745304107666,
      "learning_rate": 7.380820909258237e-05,
      "loss": 1.4115,
      "num_input_tokens_seen": 20771984,
      "step": 3434,
      "train_runtime": 3485.5973,
      "train_tokens_per_second": 5959.376
    },
    {
      "epoch": 1.2633563218390804,
      "grad_norm": 0.9008318185806274,
      "learning_rate": 7.377139701822198e-05,
      "loss": 1.5551,
      "num_input_tokens_seen": 20777104,
      "step": 3435,
      "train_runtime": 3486.4667,
      "train_tokens_per_second": 5959.358
    },
    {
      "epoch": 1.2637241379310344,
      "grad_norm": 0.8908724784851074,
      "learning_rate": 7.37345849438616e-05,
      "loss": 1.5219,
      "num_input_tokens_seen": 20786704,
      "step": 3436,
      "train_runtime": 3487.9341,
      "train_tokens_per_second": 5959.603
    },
    {
      "epoch": 1.2640919540229885,
      "grad_norm": 0.897625744342804,
      "learning_rate": 7.369777286950121e-05,
      "loss": 1.5056,
      "num_input_tokens_seen": 20793200,
      "step": 3437,
      "train_runtime": 3488.964,
      "train_tokens_per_second": 5959.706
    },
    {
      "epoch": 1.2644597701149425,
      "grad_norm": 0.8837428689002991,
      "learning_rate": 7.36609607951408e-05,
      "loss": 1.4936,
      "num_input_tokens_seen": 20798320,
      "step": 3438,
      "train_runtime": 3489.8056,
      "train_tokens_per_second": 5959.736
    },
    {
      "epoch": 1.2648275862068965,
      "grad_norm": 1.0072245597839355,
      "learning_rate": 7.362414872078041e-05,
      "loss": 1.7243,
      "num_input_tokens_seen": 20803824,
      "step": 3439,
      "train_runtime": 3490.7263,
      "train_tokens_per_second": 5959.741
    },
    {
      "epoch": 1.2651954022988505,
      "grad_norm": 0.9121684432029724,
      "learning_rate": 7.358733664642002e-05,
      "loss": 1.575,
      "num_input_tokens_seen": 20808688,
      "step": 3440,
      "train_runtime": 3491.5431,
      "train_tokens_per_second": 5959.74
    },
    {
      "epoch": 1.2655632183908045,
      "grad_norm": 0.9595100283622742,
      "learning_rate": 7.355052457205964e-05,
      "loss": 1.5143,
      "num_input_tokens_seen": 20814464,
      "step": 3441,
      "train_runtime": 3492.4923,
      "train_tokens_per_second": 5959.774
    },
    {
      "epoch": 1.2659310344827586,
      "grad_norm": 0.8965319991111755,
      "learning_rate": 7.351371249769925e-05,
      "loss": 1.4442,
      "num_input_tokens_seen": 20819808,
      "step": 3442,
      "train_runtime": 3493.3964,
      "train_tokens_per_second": 5959.761
    },
    {
      "epoch": 1.2662988505747126,
      "grad_norm": 0.9464142322540283,
      "learning_rate": 7.347690042333886e-05,
      "loss": 1.4589,
      "num_input_tokens_seen": 20825200,
      "step": 3443,
      "train_runtime": 3494.3,
      "train_tokens_per_second": 5959.763
    },
    {
      "epoch": 1.2666666666666666,
      "grad_norm": 0.9292598366737366,
      "learning_rate": 7.344008834897846e-05,
      "loss": 1.6559,
      "num_input_tokens_seen": 20830144,
      "step": 3444,
      "train_runtime": 3495.1356,
      "train_tokens_per_second": 5959.753
    },
    {
      "epoch": 1.2670344827586206,
      "grad_norm": 0.9456391930580139,
      "learning_rate": 7.340327627461808e-05,
      "loss": 1.5098,
      "num_input_tokens_seen": 20836368,
      "step": 3445,
      "train_runtime": 3496.1673,
      "train_tokens_per_second": 5959.774
    },
    {
      "epoch": 1.2674022988505746,
      "grad_norm": 1.042541265487671,
      "learning_rate": 7.336646420025769e-05,
      "loss": 1.6533,
      "num_input_tokens_seen": 20842288,
      "step": 3446,
      "train_runtime": 3497.1368,
      "train_tokens_per_second": 5959.815
    },
    {
      "epoch": 1.2677701149425287,
      "grad_norm": 0.9033098220825195,
      "learning_rate": 7.33296521258973e-05,
      "loss": 1.3759,
      "num_input_tokens_seen": 20849552,
      "step": 3447,
      "train_runtime": 3498.2954,
      "train_tokens_per_second": 5959.917
    },
    {
      "epoch": 1.2681379310344827,
      "grad_norm": 0.9085573554039001,
      "learning_rate": 7.32928400515369e-05,
      "loss": 1.6693,
      "num_input_tokens_seen": 20854976,
      "step": 3448,
      "train_runtime": 3499.2044,
      "train_tokens_per_second": 5959.919
    },
    {
      "epoch": 1.2685057471264367,
      "grad_norm": 0.9716749787330627,
      "learning_rate": 7.325602797717653e-05,
      "loss": 1.574,
      "num_input_tokens_seen": 20859904,
      "step": 3449,
      "train_runtime": 3500.0589,
      "train_tokens_per_second": 5959.872
    },
    {
      "epoch": 1.2688735632183907,
      "grad_norm": 0.9181057810783386,
      "learning_rate": 7.321921590281612e-05,
      "loss": 1.7126,
      "num_input_tokens_seen": 20864576,
      "step": 3450,
      "train_runtime": 3500.8574,
      "train_tokens_per_second": 5959.848
    },
    {
      "epoch": 1.2692413793103448,
      "grad_norm": 1.051499366760254,
      "learning_rate": 7.318240382845573e-05,
      "loss": 1.5253,
      "num_input_tokens_seen": 20871136,
      "step": 3451,
      "train_runtime": 3502.4196,
      "train_tokens_per_second": 5959.062
    },
    {
      "epoch": 1.2696091954022988,
      "grad_norm": 0.9745582342147827,
      "learning_rate": 7.314559175409534e-05,
      "loss": 1.6499,
      "num_input_tokens_seen": 20875504,
      "step": 3452,
      "train_runtime": 3503.1881,
      "train_tokens_per_second": 5959.002
    },
    {
      "epoch": 1.2699770114942528,
      "grad_norm": 0.9093972444534302,
      "learning_rate": 7.310877967973495e-05,
      "loss": 1.4549,
      "num_input_tokens_seen": 20880944,
      "step": 3453,
      "train_runtime": 3504.0929,
      "train_tokens_per_second": 5959.016
    },
    {
      "epoch": 1.2703448275862068,
      "grad_norm": 0.9666886329650879,
      "learning_rate": 7.307196760537457e-05,
      "loss": 1.4058,
      "num_input_tokens_seen": 20885584,
      "step": 3454,
      "train_runtime": 3504.8889,
      "train_tokens_per_second": 5958.986
    },
    {
      "epoch": 1.270712643678161,
      "grad_norm": 0.9249492287635803,
      "learning_rate": 7.303515553101418e-05,
      "loss": 1.5741,
      "num_input_tokens_seen": 20891040,
      "step": 3455,
      "train_runtime": 3505.7531,
      "train_tokens_per_second": 5959.073
    },
    {
      "epoch": 1.2710804597701149,
      "grad_norm": 1.0086640119552612,
      "learning_rate": 7.299834345665378e-05,
      "loss": 1.7149,
      "num_input_tokens_seen": 20897104,
      "step": 3456,
      "train_runtime": 3506.7526,
      "train_tokens_per_second": 5959.104
    },
    {
      "epoch": 1.271448275862069,
      "grad_norm": 0.9152526259422302,
      "learning_rate": 7.296153138229339e-05,
      "loss": 1.7083,
      "num_input_tokens_seen": 20904032,
      "step": 3457,
      "train_runtime": 3507.8665,
      "train_tokens_per_second": 5959.187
    },
    {
      "epoch": 1.271816091954023,
      "grad_norm": 0.8748776316642761,
      "learning_rate": 7.292471930793301e-05,
      "loss": 1.4562,
      "num_input_tokens_seen": 20909312,
      "step": 3458,
      "train_runtime": 3508.7692,
      "train_tokens_per_second": 5959.159
    },
    {
      "epoch": 1.2721839080459771,
      "grad_norm": 0.9453015327453613,
      "learning_rate": 7.288790723357262e-05,
      "loss": 1.5359,
      "num_input_tokens_seen": 20914256,
      "step": 3459,
      "train_runtime": 3509.6246,
      "train_tokens_per_second": 5959.115
    },
    {
      "epoch": 1.272551724137931,
      "grad_norm": 1.0185391902923584,
      "learning_rate": 7.285109515921223e-05,
      "loss": 1.6629,
      "num_input_tokens_seen": 20920272,
      "step": 3460,
      "train_runtime": 3510.6149,
      "train_tokens_per_second": 5959.148
    },
    {
      "epoch": 1.2729195402298852,
      "grad_norm": 0.9400529265403748,
      "learning_rate": 7.281428308485184e-05,
      "loss": 1.6218,
      "num_input_tokens_seen": 20925696,
      "step": 3461,
      "train_runtime": 3511.5017,
      "train_tokens_per_second": 5959.187
    },
    {
      "epoch": 1.273287356321839,
      "grad_norm": 0.952666163444519,
      "learning_rate": 7.277747101049144e-05,
      "loss": 1.7327,
      "num_input_tokens_seen": 20931264,
      "step": 3462,
      "train_runtime": 3512.448,
      "train_tokens_per_second": 5959.167
    },
    {
      "epoch": 1.2736551724137932,
      "grad_norm": 0.9980821013450623,
      "learning_rate": 7.274065893613105e-05,
      "loss": 1.4684,
      "num_input_tokens_seen": 20938320,
      "step": 3463,
      "train_runtime": 3513.5918,
      "train_tokens_per_second": 5959.235
    },
    {
      "epoch": 1.274022988505747,
      "grad_norm": 1.0049513578414917,
      "learning_rate": 7.270384686177066e-05,
      "loss": 1.6166,
      "num_input_tokens_seen": 20943104,
      "step": 3464,
      "train_runtime": 3514.4011,
      "train_tokens_per_second": 5959.224
    },
    {
      "epoch": 1.2743908045977013,
      "grad_norm": 0.9411967992782593,
      "learning_rate": 7.266703478741027e-05,
      "loss": 1.512,
      "num_input_tokens_seen": 20950160,
      "step": 3465,
      "train_runtime": 3515.544,
      "train_tokens_per_second": 5959.294
    },
    {
      "epoch": 1.274758620689655,
      "grad_norm": 0.9308216571807861,
      "learning_rate": 7.26302227130499e-05,
      "loss": 1.4355,
      "num_input_tokens_seen": 20957040,
      "step": 3466,
      "train_runtime": 3516.6574,
      "train_tokens_per_second": 5959.364
    },
    {
      "epoch": 1.2751264367816093,
      "grad_norm": 0.9101845622062683,
      "learning_rate": 7.25934106386895e-05,
      "loss": 1.6134,
      "num_input_tokens_seen": 20961440,
      "step": 3467,
      "train_runtime": 3517.4303,
      "train_tokens_per_second": 5959.305
    },
    {
      "epoch": 1.275494252873563,
      "grad_norm": 0.9455374479293823,
      "learning_rate": 7.25565985643291e-05,
      "loss": 1.5917,
      "num_input_tokens_seen": 20966928,
      "step": 3468,
      "train_runtime": 3518.3414,
      "train_tokens_per_second": 5959.322
    },
    {
      "epoch": 1.2758620689655173,
      "grad_norm": 0.9727440476417542,
      "learning_rate": 7.251978648996871e-05,
      "loss": 1.5776,
      "num_input_tokens_seen": 20972080,
      "step": 3469,
      "train_runtime": 3519.1991,
      "train_tokens_per_second": 5959.333
    },
    {
      "epoch": 1.2762298850574711,
      "grad_norm": 0.8744324445724487,
      "learning_rate": 7.248297441560832e-05,
      "loss": 1.5142,
      "num_input_tokens_seen": 20977840,
      "step": 3470,
      "train_runtime": 3520.1655,
      "train_tokens_per_second": 5959.334
    },
    {
      "epoch": 1.2765977011494254,
      "grad_norm": 0.8911353945732117,
      "learning_rate": 7.244616234124794e-05,
      "loss": 1.5853,
      "num_input_tokens_seen": 20982480,
      "step": 3471,
      "train_runtime": 3520.9811,
      "train_tokens_per_second": 5959.271
    },
    {
      "epoch": 1.2769655172413792,
      "grad_norm": 0.9117625951766968,
      "learning_rate": 7.240935026688755e-05,
      "loss": 1.4741,
      "num_input_tokens_seen": 20987744,
      "step": 3472,
      "train_runtime": 3521.8771,
      "train_tokens_per_second": 5959.249
    },
    {
      "epoch": 1.2773333333333334,
      "grad_norm": 0.8872541785240173,
      "learning_rate": 7.237253819252716e-05,
      "loss": 1.5719,
      "num_input_tokens_seen": 20995120,
      "step": 3473,
      "train_runtime": 3523.045,
      "train_tokens_per_second": 5959.367
    },
    {
      "epoch": 1.2777011494252872,
      "grad_norm": 0.9287063479423523,
      "learning_rate": 7.233572611816675e-05,
      "loss": 1.4267,
      "num_input_tokens_seen": 21001024,
      "step": 3474,
      "train_runtime": 3524.0067,
      "train_tokens_per_second": 5959.417
    },
    {
      "epoch": 1.2780689655172415,
      "grad_norm": 0.9828014373779297,
      "learning_rate": 7.229891404380637e-05,
      "loss": 1.5215,
      "num_input_tokens_seen": 21006048,
      "step": 3475,
      "train_runtime": 3524.8425,
      "train_tokens_per_second": 5959.429
    },
    {
      "epoch": 1.2784367816091953,
      "grad_norm": 0.9956949353218079,
      "learning_rate": 7.226210196944598e-05,
      "loss": 1.6499,
      "num_input_tokens_seen": 21012960,
      "step": 3476,
      "train_runtime": 3525.9487,
      "train_tokens_per_second": 5959.52
    },
    {
      "epoch": 1.2788045977011495,
      "grad_norm": 0.9091798663139343,
      "learning_rate": 7.222528989508559e-05,
      "loss": 1.4731,
      "num_input_tokens_seen": 21020512,
      "step": 3477,
      "train_runtime": 3527.1463,
      "train_tokens_per_second": 5959.637
    },
    {
      "epoch": 1.2791724137931033,
      "grad_norm": 0.9619956612586975,
      "learning_rate": 7.21884778207252e-05,
      "loss": 1.6646,
      "num_input_tokens_seen": 21025312,
      "step": 3478,
      "train_runtime": 3527.9797,
      "train_tokens_per_second": 5959.59
    },
    {
      "epoch": 1.2795402298850576,
      "grad_norm": 0.918745219707489,
      "learning_rate": 7.215166574636481e-05,
      "loss": 1.3209,
      "num_input_tokens_seen": 21035952,
      "step": 3479,
      "train_runtime": 3529.6097,
      "train_tokens_per_second": 5959.852
    },
    {
      "epoch": 1.2799080459770116,
      "grad_norm": 1.0378994941711426,
      "learning_rate": 7.211485367200442e-05,
      "loss": 1.6537,
      "num_input_tokens_seen": 21042192,
      "step": 3480,
      "train_runtime": 3530.663,
      "train_tokens_per_second": 5959.842
    },
    {
      "epoch": 1.2802758620689656,
      "grad_norm": 1.0207245349884033,
      "learning_rate": 7.207804159764403e-05,
      "loss": 1.7408,
      "num_input_tokens_seen": 21047824,
      "step": 3481,
      "train_runtime": 3532.0461,
      "train_tokens_per_second": 5959.102
    },
    {
      "epoch": 1.2806436781609196,
      "grad_norm": 1.0150294303894043,
      "learning_rate": 7.204122952328364e-05,
      "loss": 1.5718,
      "num_input_tokens_seen": 21053152,
      "step": 3482,
      "train_runtime": 3532.9196,
      "train_tokens_per_second": 5959.137
    },
    {
      "epoch": 1.2810114942528736,
      "grad_norm": 0.902804434299469,
      "learning_rate": 7.200441744892324e-05,
      "loss": 1.6311,
      "num_input_tokens_seen": 21058336,
      "step": 3483,
      "train_runtime": 3533.8037,
      "train_tokens_per_second": 5959.113
    },
    {
      "epoch": 1.2813793103448277,
      "grad_norm": 0.9312765598297119,
      "learning_rate": 7.196760537456287e-05,
      "loss": 1.6575,
      "num_input_tokens_seen": 21064336,
      "step": 3484,
      "train_runtime": 3534.7939,
      "train_tokens_per_second": 5959.141
    },
    {
      "epoch": 1.2817471264367817,
      "grad_norm": 1.087038278579712,
      "learning_rate": 7.193079330020248e-05,
      "loss": 1.7055,
      "num_input_tokens_seen": 21070368,
      "step": 3485,
      "train_runtime": 3535.7822,
      "train_tokens_per_second": 5959.182
    },
    {
      "epoch": 1.2821149425287357,
      "grad_norm": 1.0413316488265991,
      "learning_rate": 7.189398122584207e-05,
      "loss": 1.6932,
      "num_input_tokens_seen": 21074656,
      "step": 3486,
      "train_runtime": 3536.548,
      "train_tokens_per_second": 5959.104
    },
    {
      "epoch": 1.2824827586206897,
      "grad_norm": 0.7907637357711792,
      "learning_rate": 7.185716915148168e-05,
      "loss": 1.4059,
      "num_input_tokens_seen": 21083600,
      "step": 3487,
      "train_runtime": 3537.9358,
      "train_tokens_per_second": 5959.294
    },
    {
      "epoch": 1.2828505747126437,
      "grad_norm": 1.0019232034683228,
      "learning_rate": 7.18203570771213e-05,
      "loss": 1.6671,
      "num_input_tokens_seen": 21088480,
      "step": 3488,
      "train_runtime": 3538.7802,
      "train_tokens_per_second": 5959.251
    },
    {
      "epoch": 1.2832183908045978,
      "grad_norm": 0.7934490442276001,
      "learning_rate": 7.178354500276091e-05,
      "loss": 1.2534,
      "num_input_tokens_seen": 21095824,
      "step": 3489,
      "train_runtime": 3539.9356,
      "train_tokens_per_second": 5959.381
    },
    {
      "epoch": 1.2835862068965518,
      "grad_norm": 1.109464168548584,
      "learning_rate": 7.174673292840052e-05,
      "loss": 1.5836,
      "num_input_tokens_seen": 21099856,
      "step": 3490,
      "train_runtime": 3540.6874,
      "train_tokens_per_second": 5959.254
    },
    {
      "epoch": 1.2839540229885058,
      "grad_norm": 0.9460877180099487,
      "learning_rate": 7.170992085404013e-05,
      "loss": 1.4744,
      "num_input_tokens_seen": 21105232,
      "step": 3491,
      "train_runtime": 3541.5702,
      "train_tokens_per_second": 5959.287
    },
    {
      "epoch": 1.2843218390804598,
      "grad_norm": 0.8960955142974854,
      "learning_rate": 7.167310877967974e-05,
      "loss": 1.501,
      "num_input_tokens_seen": 21110128,
      "step": 3492,
      "train_runtime": 3542.4076,
      "train_tokens_per_second": 5959.26
    },
    {
      "epoch": 1.2846896551724138,
      "grad_norm": 0.9716125130653381,
      "learning_rate": 7.163629670531935e-05,
      "loss": 1.7261,
      "num_input_tokens_seen": 21114464,
      "step": 3493,
      "train_runtime": 3543.149,
      "train_tokens_per_second": 5959.237
    },
    {
      "epoch": 1.2850574712643679,
      "grad_norm": 1.0320979356765747,
      "learning_rate": 7.159948463095896e-05,
      "loss": 1.4786,
      "num_input_tokens_seen": 21121264,
      "step": 3494,
      "train_runtime": 3544.2466,
      "train_tokens_per_second": 5959.31
    },
    {
      "epoch": 1.2854252873563219,
      "grad_norm": 0.9457396864891052,
      "learning_rate": 7.156267255659857e-05,
      "loss": 1.5681,
      "num_input_tokens_seen": 21126672,
      "step": 3495,
      "train_runtime": 3545.141,
      "train_tokens_per_second": 5959.332
    },
    {
      "epoch": 1.285793103448276,
      "grad_norm": 0.9831223487854004,
      "learning_rate": 7.152586048223819e-05,
      "loss": 1.6491,
      "num_input_tokens_seen": 21131536,
      "step": 3496,
      "train_runtime": 3545.9768,
      "train_tokens_per_second": 5959.299
    },
    {
      "epoch": 1.28616091954023,
      "grad_norm": 0.9211698770523071,
      "learning_rate": 7.14890484078778e-05,
      "loss": 1.442,
      "num_input_tokens_seen": 21135856,
      "step": 3497,
      "train_runtime": 3546.747,
      "train_tokens_per_second": 5959.223
    },
    {
      "epoch": 1.286528735632184,
      "grad_norm": 0.8640750050544739,
      "learning_rate": 7.145223633351739e-05,
      "loss": 1.5376,
      "num_input_tokens_seen": 21141888,
      "step": 3498,
      "train_runtime": 3547.7259,
      "train_tokens_per_second": 5959.279
    },
    {
      "epoch": 1.286896551724138,
      "grad_norm": 1.0163702964782715,
      "learning_rate": 7.1415424259157e-05,
      "loss": 1.5638,
      "num_input_tokens_seen": 21151488,
      "step": 3499,
      "train_runtime": 3549.2009,
      "train_tokens_per_second": 5959.507
    },
    {
      "epoch": 1.287264367816092,
      "grad_norm": 0.9114499688148499,
      "learning_rate": 7.137861218479661e-05,
      "loss": 1.4073,
      "num_input_tokens_seen": 21159776,
      "step": 3500,
      "train_runtime": 3550.4654,
      "train_tokens_per_second": 5959.719
    },
    {
      "epoch": 1.287632183908046,
      "grad_norm": 0.9567367434501648,
      "learning_rate": 7.134180011043623e-05,
      "loss": 1.657,
      "num_input_tokens_seen": 21164480,
      "step": 3501,
      "train_runtime": 3551.3179,
      "train_tokens_per_second": 5959.613
    },
    {
      "epoch": 1.288,
      "grad_norm": 1.1022112369537354,
      "learning_rate": 7.130498803607584e-05,
      "loss": 1.7188,
      "num_input_tokens_seen": 21169872,
      "step": 3502,
      "train_runtime": 3552.225,
      "train_tokens_per_second": 5959.609
    },
    {
      "epoch": 1.288367816091954,
      "grad_norm": 0.853762686252594,
      "learning_rate": 7.126817596171545e-05,
      "loss": 1.3639,
      "num_input_tokens_seen": 21177264,
      "step": 3503,
      "train_runtime": 3553.3811,
      "train_tokens_per_second": 5959.75
    },
    {
      "epoch": 1.288735632183908,
      "grad_norm": 0.8977413773536682,
      "learning_rate": 7.123136388735505e-05,
      "loss": 1.5133,
      "num_input_tokens_seen": 21182400,
      "step": 3504,
      "train_runtime": 3554.2484,
      "train_tokens_per_second": 5959.741
    },
    {
      "epoch": 1.289103448275862,
      "grad_norm": 0.9821711778640747,
      "learning_rate": 7.119455181299467e-05,
      "loss": 1.4526,
      "num_input_tokens_seen": 21189088,
      "step": 3505,
      "train_runtime": 3555.3378,
      "train_tokens_per_second": 5959.796
    },
    {
      "epoch": 1.2894712643678161,
      "grad_norm": 0.9481285810470581,
      "learning_rate": 7.115773973863428e-05,
      "loss": 1.557,
      "num_input_tokens_seen": 21194432,
      "step": 3506,
      "train_runtime": 3556.2483,
      "train_tokens_per_second": 5959.773
    },
    {
      "epoch": 1.2898390804597701,
      "grad_norm": 0.9218729138374329,
      "learning_rate": 7.112092766427389e-05,
      "loss": 1.5139,
      "num_input_tokens_seen": 21199632,
      "step": 3507,
      "train_runtime": 3557.1185,
      "train_tokens_per_second": 5959.777
    },
    {
      "epoch": 1.2902068965517242,
      "grad_norm": 0.944381058216095,
      "learning_rate": 7.10841155899135e-05,
      "loss": 1.3236,
      "num_input_tokens_seen": 21205104,
      "step": 3508,
      "train_runtime": 3558.0168,
      "train_tokens_per_second": 5959.81
    },
    {
      "epoch": 1.2905747126436782,
      "grad_norm": 0.8969419598579407,
      "learning_rate": 7.10473035155531e-05,
      "loss": 1.4088,
      "num_input_tokens_seen": 21210944,
      "step": 3509,
      "train_runtime": 3558.986,
      "train_tokens_per_second": 5959.828
    },
    {
      "epoch": 1.2909425287356322,
      "grad_norm": 0.8900802135467529,
      "learning_rate": 7.101049144119271e-05,
      "loss": 1.4862,
      "num_input_tokens_seen": 21216640,
      "step": 3510,
      "train_runtime": 3559.9216,
      "train_tokens_per_second": 5959.862
    },
    {
      "epoch": 1.2913103448275862,
      "grad_norm": 0.9737229943275452,
      "learning_rate": 7.097367936683232e-05,
      "loss": 1.6753,
      "num_input_tokens_seen": 21221936,
      "step": 3511,
      "train_runtime": 3561.2479,
      "train_tokens_per_second": 5959.129
    },
    {
      "epoch": 1.2916781609195402,
      "grad_norm": 0.8889402747154236,
      "learning_rate": 7.093686729247193e-05,
      "loss": 1.5952,
      "num_input_tokens_seen": 21226496,
      "step": 3512,
      "train_runtime": 3562.0607,
      "train_tokens_per_second": 5959.049
    },
    {
      "epoch": 1.2920459770114943,
      "grad_norm": 0.9297152757644653,
      "learning_rate": 7.090005521811154e-05,
      "loss": 1.3997,
      "num_input_tokens_seen": 21232624,
      "step": 3513,
      "train_runtime": 3563.0666,
      "train_tokens_per_second": 5959.087
    },
    {
      "epoch": 1.2924137931034483,
      "grad_norm": 0.9345740675926208,
      "learning_rate": 7.086324314375116e-05,
      "loss": 1.6513,
      "num_input_tokens_seen": 21237888,
      "step": 3514,
      "train_runtime": 3563.9465,
      "train_tokens_per_second": 5959.093
    },
    {
      "epoch": 1.2927816091954023,
      "grad_norm": 0.9662009477615356,
      "learning_rate": 7.082643106939077e-05,
      "loss": 1.5845,
      "num_input_tokens_seen": 21242752,
      "step": 3515,
      "train_runtime": 3564.7677,
      "train_tokens_per_second": 5959.085
    },
    {
      "epoch": 1.2931494252873563,
      "grad_norm": 1.0016788244247437,
      "learning_rate": 7.078961899503037e-05,
      "loss": 1.5019,
      "num_input_tokens_seen": 21247344,
      "step": 3516,
      "train_runtime": 3565.5388,
      "train_tokens_per_second": 5959.084
    },
    {
      "epoch": 1.2935172413793103,
      "grad_norm": 0.8903211355209351,
      "learning_rate": 7.075280692066997e-05,
      "loss": 1.4587,
      "num_input_tokens_seen": 21252992,
      "step": 3517,
      "train_runtime": 3566.4798,
      "train_tokens_per_second": 5959.095
    },
    {
      "epoch": 1.2938850574712644,
      "grad_norm": 0.9020424485206604,
      "learning_rate": 7.07159948463096e-05,
      "loss": 1.4961,
      "num_input_tokens_seen": 21260016,
      "step": 3518,
      "train_runtime": 3567.6081,
      "train_tokens_per_second": 5959.179
    },
    {
      "epoch": 1.2942528735632184,
      "grad_norm": 0.9242357611656189,
      "learning_rate": 7.06791827719492e-05,
      "loss": 1.35,
      "num_input_tokens_seen": 21264944,
      "step": 3519,
      "train_runtime": 3568.4149,
      "train_tokens_per_second": 5959.213
    },
    {
      "epoch": 1.2946206896551724,
      "grad_norm": 0.8996397852897644,
      "learning_rate": 7.064237069758881e-05,
      "loss": 1.4686,
      "num_input_tokens_seen": 21275280,
      "step": 3520,
      "train_runtime": 3569.9883,
      "train_tokens_per_second": 5959.482
    },
    {
      "epoch": 1.2949885057471264,
      "grad_norm": 0.9285815358161926,
      "learning_rate": 7.060555862322842e-05,
      "loss": 1.5583,
      "num_input_tokens_seen": 21280704,
      "step": 3521,
      "train_runtime": 3570.8902,
      "train_tokens_per_second": 5959.495
    },
    {
      "epoch": 1.2953563218390804,
      "grad_norm": 0.8893944621086121,
      "learning_rate": 7.056874654886803e-05,
      "loss": 1.4772,
      "num_input_tokens_seen": 21285840,
      "step": 3522,
      "train_runtime": 3571.7523,
      "train_tokens_per_second": 5959.495
    },
    {
      "epoch": 1.2957241379310345,
      "grad_norm": 0.8923265933990479,
      "learning_rate": 7.053193447450764e-05,
      "loss": 1.3956,
      "num_input_tokens_seen": 21293792,
      "step": 3523,
      "train_runtime": 3573.0056,
      "train_tokens_per_second": 5959.63
    },
    {
      "epoch": 1.2960919540229885,
      "grad_norm": 0.9582086801528931,
      "learning_rate": 7.049512240014725e-05,
      "loss": 1.4337,
      "num_input_tokens_seen": 21298336,
      "step": 3524,
      "train_runtime": 3573.7804,
      "train_tokens_per_second": 5959.61
    },
    {
      "epoch": 1.2964597701149425,
      "grad_norm": 1.0114604234695435,
      "learning_rate": 7.045831032578686e-05,
      "loss": 1.5049,
      "num_input_tokens_seen": 21305840,
      "step": 3525,
      "train_runtime": 3574.9771,
      "train_tokens_per_second": 5959.714
    },
    {
      "epoch": 1.2968275862068965,
      "grad_norm": 0.9581440687179565,
      "learning_rate": 7.042149825142648e-05,
      "loss": 1.4544,
      "num_input_tokens_seen": 21310928,
      "step": 3526,
      "train_runtime": 3575.8364,
      "train_tokens_per_second": 5959.704
    },
    {
      "epoch": 1.2971954022988506,
      "grad_norm": 0.9656909704208374,
      "learning_rate": 7.038468617706609e-05,
      "loss": 1.6463,
      "num_input_tokens_seen": 21316368,
      "step": 3527,
      "train_runtime": 3576.7285,
      "train_tokens_per_second": 5959.739
    },
    {
      "epoch": 1.2975632183908046,
      "grad_norm": 0.9722489714622498,
      "learning_rate": 7.034787410270569e-05,
      "loss": 1.5223,
      "num_input_tokens_seen": 21322768,
      "step": 3528,
      "train_runtime": 3577.7989,
      "train_tokens_per_second": 5959.745
    },
    {
      "epoch": 1.2979310344827586,
      "grad_norm": 0.8741322159767151,
      "learning_rate": 7.03110620283453e-05,
      "loss": 1.6018,
      "num_input_tokens_seen": 21329264,
      "step": 3529,
      "train_runtime": 3578.8231,
      "train_tokens_per_second": 5959.854
    },
    {
      "epoch": 1.2982988505747126,
      "grad_norm": 0.8774896264076233,
      "learning_rate": 7.02742499539849e-05,
      "loss": 1.4221,
      "num_input_tokens_seen": 21334624,
      "step": 3530,
      "train_runtime": 3579.7039,
      "train_tokens_per_second": 5959.885
    },
    {
      "epoch": 1.2986666666666666,
      "grad_norm": 0.90668123960495,
      "learning_rate": 7.023743787962453e-05,
      "loss": 1.5109,
      "num_input_tokens_seen": 21339280,
      "step": 3531,
      "train_runtime": 3580.5292,
      "train_tokens_per_second": 5959.812
    },
    {
      "epoch": 1.2990344827586207,
      "grad_norm": 0.9371986985206604,
      "learning_rate": 7.020062580526413e-05,
      "loss": 1.4348,
      "num_input_tokens_seen": 21344512,
      "step": 3532,
      "train_runtime": 3581.4091,
      "train_tokens_per_second": 5959.808
    },
    {
      "epoch": 1.2994022988505747,
      "grad_norm": 0.9359028935432434,
      "learning_rate": 7.016381373090374e-05,
      "loss": 1.4243,
      "num_input_tokens_seen": 21349568,
      "step": 3533,
      "train_runtime": 3582.2542,
      "train_tokens_per_second": 5959.814
    },
    {
      "epoch": 1.2997701149425287,
      "grad_norm": 0.8446172475814819,
      "learning_rate": 7.012700165654334e-05,
      "loss": 1.4644,
      "num_input_tokens_seen": 21355136,
      "step": 3534,
      "train_runtime": 3583.1495,
      "train_tokens_per_second": 5959.879
    },
    {
      "epoch": 1.3001379310344827,
      "grad_norm": 0.8675487637519836,
      "learning_rate": 7.009018958218296e-05,
      "loss": 1.615,
      "num_input_tokens_seen": 21362800,
      "step": 3535,
      "train_runtime": 3584.3378,
      "train_tokens_per_second": 5960.041
    },
    {
      "epoch": 1.3005057471264367,
      "grad_norm": 0.9997730255126953,
      "learning_rate": 7.005337750782257e-05,
      "loss": 1.3972,
      "num_input_tokens_seen": 21368544,
      "step": 3536,
      "train_runtime": 3585.2875,
      "train_tokens_per_second": 5960.064
    },
    {
      "epoch": 1.3008735632183908,
      "grad_norm": 0.9834737777709961,
      "learning_rate": 7.001656543346218e-05,
      "loss": 1.6383,
      "num_input_tokens_seen": 21373328,
      "step": 3537,
      "train_runtime": 3586.0836,
      "train_tokens_per_second": 5960.075
    },
    {
      "epoch": 1.3012413793103448,
      "grad_norm": 0.8357015252113342,
      "learning_rate": 6.997975335910179e-05,
      "loss": 1.5022,
      "num_input_tokens_seen": 21381664,
      "step": 3538,
      "train_runtime": 3587.3661,
      "train_tokens_per_second": 5960.268
    },
    {
      "epoch": 1.3016091954022988,
      "grad_norm": 0.9906278848648071,
      "learning_rate": 6.99429412847414e-05,
      "loss": 1.4367,
      "num_input_tokens_seen": 21388976,
      "step": 3539,
      "train_runtime": 3588.5247,
      "train_tokens_per_second": 5960.381
    },
    {
      "epoch": 1.3019770114942528,
      "grad_norm": 0.891708493232727,
      "learning_rate": 6.9906129210381e-05,
      "loss": 1.4701,
      "num_input_tokens_seen": 21398160,
      "step": 3540,
      "train_runtime": 3589.9164,
      "train_tokens_per_second": 5960.629
    },
    {
      "epoch": 1.3023448275862068,
      "grad_norm": 0.919964611530304,
      "learning_rate": 6.986931713602061e-05,
      "loss": 1.3759,
      "num_input_tokens_seen": 21403472,
      "step": 3541,
      "train_runtime": 3591.2842,
      "train_tokens_per_second": 5959.838
    },
    {
      "epoch": 1.3027126436781609,
      "grad_norm": 0.8878598809242249,
      "learning_rate": 6.983250506166022e-05,
      "loss": 1.7194,
      "num_input_tokens_seen": 21410896,
      "step": 3542,
      "train_runtime": 3592.4723,
      "train_tokens_per_second": 5959.933
    },
    {
      "epoch": 1.3030804597701149,
      "grad_norm": 0.9878170490264893,
      "learning_rate": 6.979569298729983e-05,
      "loss": 1.7097,
      "num_input_tokens_seen": 21415712,
      "step": 3543,
      "train_runtime": 3593.3105,
      "train_tokens_per_second": 5959.884
    },
    {
      "epoch": 1.303448275862069,
      "grad_norm": 1.0171282291412354,
      "learning_rate": 6.975888091293946e-05,
      "loss": 1.4019,
      "num_input_tokens_seen": 21420080,
      "step": 3544,
      "train_runtime": 3594.0673,
      "train_tokens_per_second": 5959.844
    },
    {
      "epoch": 1.303816091954023,
      "grad_norm": 1.0764694213867188,
      "learning_rate": 6.972206883857906e-05,
      "loss": 1.673,
      "num_input_tokens_seen": 21426192,
      "step": 3545,
      "train_runtime": 3595.0949,
      "train_tokens_per_second": 5959.84
    },
    {
      "epoch": 1.304183908045977,
      "grad_norm": 0.8542464971542358,
      "learning_rate": 6.968525676421866e-05,
      "loss": 1.3304,
      "num_input_tokens_seen": 21441504,
      "step": 3546,
      "train_runtime": 3597.304,
      "train_tokens_per_second": 5960.437
    },
    {
      "epoch": 1.304551724137931,
      "grad_norm": 0.9444261789321899,
      "learning_rate": 6.964844468985827e-05,
      "loss": 1.6947,
      "num_input_tokens_seen": 21445888,
      "step": 3547,
      "train_runtime": 3598.0759,
      "train_tokens_per_second": 5960.377
    },
    {
      "epoch": 1.304919540229885,
      "grad_norm": 0.958427369594574,
      "learning_rate": 6.961163261549789e-05,
      "loss": 1.3635,
      "num_input_tokens_seen": 21454608,
      "step": 3548,
      "train_runtime": 3599.4016,
      "train_tokens_per_second": 5960.604
    },
    {
      "epoch": 1.305287356321839,
      "grad_norm": 0.8953106999397278,
      "learning_rate": 6.95748205411375e-05,
      "loss": 1.4,
      "num_input_tokens_seen": 21461008,
      "step": 3549,
      "train_runtime": 3600.4181,
      "train_tokens_per_second": 5960.699
    },
    {
      "epoch": 1.305655172413793,
      "grad_norm": 1.0188813209533691,
      "learning_rate": 6.953800846677711e-05,
      "loss": 1.3752,
      "num_input_tokens_seen": 21467920,
      "step": 3550,
      "train_runtime": 3601.531,
      "train_tokens_per_second": 5960.776
    },
    {
      "epoch": 1.306022988505747,
      "grad_norm": 0.9171669483184814,
      "learning_rate": 6.950119639241672e-05,
      "loss": 1.5849,
      "num_input_tokens_seen": 21472416,
      "step": 3551,
      "train_runtime": 3602.3119,
      "train_tokens_per_second": 5960.732
    },
    {
      "epoch": 1.306390804597701,
      "grad_norm": 0.9831543564796448,
      "learning_rate": 6.946438431805633e-05,
      "loss": 1.5002,
      "num_input_tokens_seen": 21479040,
      "step": 3552,
      "train_runtime": 3603.3969,
      "train_tokens_per_second": 5960.775
    },
    {
      "epoch": 1.306758620689655,
      "grad_norm": 0.8366644978523254,
      "learning_rate": 6.942757224369594e-05,
      "loss": 1.5185,
      "num_input_tokens_seen": 21486784,
      "step": 3553,
      "train_runtime": 3604.6106,
      "train_tokens_per_second": 5960.917
    },
    {
      "epoch": 1.3071264367816091,
      "grad_norm": 0.9007189273834229,
      "learning_rate": 6.939076016933554e-05,
      "loss": 1.5416,
      "num_input_tokens_seen": 21491760,
      "step": 3554,
      "train_runtime": 3605.4519,
      "train_tokens_per_second": 5960.906
    },
    {
      "epoch": 1.3074942528735631,
      "grad_norm": 0.952498733997345,
      "learning_rate": 6.935394809497515e-05,
      "loss": 1.5592,
      "num_input_tokens_seen": 21496864,
      "step": 3555,
      "train_runtime": 3606.2789,
      "train_tokens_per_second": 5960.954
    },
    {
      "epoch": 1.3078620689655174,
      "grad_norm": 0.9148018956184387,
      "learning_rate": 6.931713602061478e-05,
      "loss": 1.3999,
      "num_input_tokens_seen": 21502640,
      "step": 3556,
      "train_runtime": 3607.232,
      "train_tokens_per_second": 5960.981
    },
    {
      "epoch": 1.3082298850574712,
      "grad_norm": 0.9440029263496399,
      "learning_rate": 6.928032394625437e-05,
      "loss": 1.5397,
      "num_input_tokens_seen": 21508736,
      "step": 3557,
      "train_runtime": 3608.2219,
      "train_tokens_per_second": 5961.035
    },
    {
      "epoch": 1.3085977011494254,
      "grad_norm": 0.9105268120765686,
      "learning_rate": 6.924351187189398e-05,
      "loss": 1.4166,
      "num_input_tokens_seen": 21515088,
      "step": 3558,
      "train_runtime": 3609.2621,
      "train_tokens_per_second": 5961.077
    },
    {
      "epoch": 1.3089655172413792,
      "grad_norm": 0.9310047626495361,
      "learning_rate": 6.920669979753359e-05,
      "loss": 1.3953,
      "num_input_tokens_seen": 21521600,
      "step": 3559,
      "train_runtime": 3610.2941,
      "train_tokens_per_second": 5961.176
    },
    {
      "epoch": 1.3093333333333335,
      "grad_norm": 0.9549909234046936,
      "learning_rate": 6.91698877231732e-05,
      "loss": 1.5641,
      "num_input_tokens_seen": 21526384,
      "step": 3560,
      "train_runtime": 3611.108,
      "train_tokens_per_second": 5961.158
    },
    {
      "epoch": 1.3097011494252873,
      "grad_norm": 0.9676443338394165,
      "learning_rate": 6.913307564881282e-05,
      "loss": 1.6549,
      "num_input_tokens_seen": 21530768,
      "step": 3561,
      "train_runtime": 3611.8932,
      "train_tokens_per_second": 5961.075
    },
    {
      "epoch": 1.3100689655172415,
      "grad_norm": 0.9232891201972961,
      "learning_rate": 6.909626357445243e-05,
      "loss": 1.5543,
      "num_input_tokens_seen": 21536848,
      "step": 3562,
      "train_runtime": 3612.9044,
      "train_tokens_per_second": 5961.09
    },
    {
      "epoch": 1.3104367816091953,
      "grad_norm": 0.9708718657493591,
      "learning_rate": 6.905945150009204e-05,
      "loss": 1.4952,
      "num_input_tokens_seen": 21541200,
      "step": 3563,
      "train_runtime": 3613.6287,
      "train_tokens_per_second": 5961.099
    },
    {
      "epoch": 1.3108045977011495,
      "grad_norm": 1.0027772188186646,
      "learning_rate": 6.902263942573163e-05,
      "loss": 1.5742,
      "num_input_tokens_seen": 21548128,
      "step": 3564,
      "train_runtime": 3614.73,
      "train_tokens_per_second": 5961.2
    },
    {
      "epoch": 1.3111724137931033,
      "grad_norm": 0.9488890171051025,
      "learning_rate": 6.898582735137126e-05,
      "loss": 1.4745,
      "num_input_tokens_seen": 21553184,
      "step": 3565,
      "train_runtime": 3615.5819,
      "train_tokens_per_second": 5961.194
    },
    {
      "epoch": 1.3115402298850576,
      "grad_norm": 0.9458577632904053,
      "learning_rate": 6.894901527701086e-05,
      "loss": 1.3461,
      "num_input_tokens_seen": 21558848,
      "step": 3566,
      "train_runtime": 3616.5211,
      "train_tokens_per_second": 5961.212
    },
    {
      "epoch": 1.3119080459770114,
      "grad_norm": 0.9157336354255676,
      "learning_rate": 6.891220320265047e-05,
      "loss": 1.5361,
      "num_input_tokens_seen": 21565360,
      "step": 3567,
      "train_runtime": 3617.5474,
      "train_tokens_per_second": 5961.321
    },
    {
      "epoch": 1.3122758620689656,
      "grad_norm": 0.9321467280387878,
      "learning_rate": 6.887539112829008e-05,
      "loss": 1.44,
      "num_input_tokens_seen": 21570304,
      "step": 3568,
      "train_runtime": 3618.3744,
      "train_tokens_per_second": 5961.324
    },
    {
      "epoch": 1.3126436781609194,
      "grad_norm": 0.9249515533447266,
      "learning_rate": 6.883857905392969e-05,
      "loss": 1.6631,
      "num_input_tokens_seen": 21575376,
      "step": 3569,
      "train_runtime": 3619.2504,
      "train_tokens_per_second": 5961.283
    },
    {
      "epoch": 1.3130114942528737,
      "grad_norm": 0.9264847636222839,
      "learning_rate": 6.88017669795693e-05,
      "loss": 1.4589,
      "num_input_tokens_seen": 21582656,
      "step": 3570,
      "train_runtime": 3620.4294,
      "train_tokens_per_second": 5961.353
    },
    {
      "epoch": 1.3133793103448275,
      "grad_norm": 0.9606678485870361,
      "learning_rate": 6.876495490520891e-05,
      "loss": 1.5251,
      "num_input_tokens_seen": 21590016,
      "step": 3571,
      "train_runtime": 3622.0933,
      "train_tokens_per_second": 5960.646
    },
    {
      "epoch": 1.3137471264367817,
      "grad_norm": 0.955280065536499,
      "learning_rate": 6.872814283084852e-05,
      "loss": 1.4844,
      "num_input_tokens_seen": 21595504,
      "step": 3572,
      "train_runtime": 3623.0297,
      "train_tokens_per_second": 5960.62
    },
    {
      "epoch": 1.3141149425287355,
      "grad_norm": 0.8960136771202087,
      "learning_rate": 6.869133075648813e-05,
      "loss": 1.3609,
      "num_input_tokens_seen": 21602656,
      "step": 3573,
      "train_runtime": 3624.1575,
      "train_tokens_per_second": 5960.739
    },
    {
      "epoch": 1.3144827586206898,
      "grad_norm": 0.9145007729530334,
      "learning_rate": 6.865451868212775e-05,
      "loss": 1.6295,
      "num_input_tokens_seen": 21608560,
      "step": 3574,
      "train_runtime": 3625.1404,
      "train_tokens_per_second": 5960.751
    },
    {
      "epoch": 1.3148505747126435,
      "grad_norm": 0.9274311661720276,
      "learning_rate": 6.861770660776736e-05,
      "loss": 1.4616,
      "num_input_tokens_seen": 21615232,
      "step": 3575,
      "train_runtime": 3626.2065,
      "train_tokens_per_second": 5960.839
    },
    {
      "epoch": 1.3152183908045978,
      "grad_norm": 1.1087721586227417,
      "learning_rate": 6.858089453340695e-05,
      "loss": 1.4991,
      "num_input_tokens_seen": 21620384,
      "step": 3576,
      "train_runtime": 3627.0383,
      "train_tokens_per_second": 5960.892
    },
    {
      "epoch": 1.3155862068965516,
      "grad_norm": 0.8783528208732605,
      "learning_rate": 6.854408245904656e-05,
      "loss": 1.353,
      "num_input_tokens_seen": 21626992,
      "step": 3577,
      "train_runtime": 3628.088,
      "train_tokens_per_second": 5960.989
    },
    {
      "epoch": 1.3159540229885058,
      "grad_norm": 0.9759103059768677,
      "learning_rate": 6.850727038468618e-05,
      "loss": 1.6281,
      "num_input_tokens_seen": 21631824,
      "step": 3578,
      "train_runtime": 3628.9008,
      "train_tokens_per_second": 5960.985
    },
    {
      "epoch": 1.3163218390804596,
      "grad_norm": 1.032320261001587,
      "learning_rate": 6.84704583103258e-05,
      "loss": 1.6755,
      "num_input_tokens_seen": 21636400,
      "step": 3579,
      "train_runtime": 3629.6803,
      "train_tokens_per_second": 5960.966
    },
    {
      "epoch": 1.3166896551724139,
      "grad_norm": 0.9147163033485413,
      "learning_rate": 6.84336462359654e-05,
      "loss": 1.4719,
      "num_input_tokens_seen": 21642672,
      "step": 3580,
      "train_runtime": 3630.7085,
      "train_tokens_per_second": 5961.005
    },
    {
      "epoch": 1.317057471264368,
      "grad_norm": 1.004988431930542,
      "learning_rate": 6.839683416160501e-05,
      "loss": 1.4274,
      "num_input_tokens_seen": 21649264,
      "step": 3581,
      "train_runtime": 3631.7982,
      "train_tokens_per_second": 5961.032
    },
    {
      "epoch": 1.317425287356322,
      "grad_norm": 1.0151127576828003,
      "learning_rate": 6.836002208724462e-05,
      "loss": 1.5953,
      "num_input_tokens_seen": 21654624,
      "step": 3582,
      "train_runtime": 3632.6852,
      "train_tokens_per_second": 5961.052
    },
    {
      "epoch": 1.317793103448276,
      "grad_norm": 0.9770903587341309,
      "learning_rate": 6.832321001288423e-05,
      "loss": 1.4601,
      "num_input_tokens_seen": 21660352,
      "step": 3583,
      "train_runtime": 3633.6143,
      "train_tokens_per_second": 5961.104
    },
    {
      "epoch": 1.31816091954023,
      "grad_norm": 0.9266052842140198,
      "learning_rate": 6.828639793852384e-05,
      "loss": 1.4973,
      "num_input_tokens_seen": 21665968,
      "step": 3584,
      "train_runtime": 3634.5462,
      "train_tokens_per_second": 5961.121
    },
    {
      "epoch": 1.318528735632184,
      "grad_norm": 0.8691406846046448,
      "learning_rate": 6.824958586416345e-05,
      "loss": 1.7214,
      "num_input_tokens_seen": 21672448,
      "step": 3585,
      "train_runtime": 3635.6006,
      "train_tokens_per_second": 5961.174
    },
    {
      "epoch": 1.318896551724138,
      "grad_norm": 0.9937482476234436,
      "learning_rate": 6.821277378980307e-05,
      "loss": 1.3971,
      "num_input_tokens_seen": 21677744,
      "step": 3586,
      "train_runtime": 3636.491,
      "train_tokens_per_second": 5961.171
    },
    {
      "epoch": 1.319264367816092,
      "grad_norm": 1.1415114402770996,
      "learning_rate": 6.817596171544266e-05,
      "loss": 1.6211,
      "num_input_tokens_seen": 21687312,
      "step": 3587,
      "train_runtime": 3637.94,
      "train_tokens_per_second": 5961.427
    },
    {
      "epoch": 1.319632183908046,
      "grad_norm": 0.9889041185379028,
      "learning_rate": 6.813914964108227e-05,
      "loss": 1.4436,
      "num_input_tokens_seen": 21692496,
      "step": 3588,
      "train_runtime": 3638.7972,
      "train_tokens_per_second": 5961.447
    },
    {
      "epoch": 1.32,
      "grad_norm": 1.045847773551941,
      "learning_rate": 6.810233756672188e-05,
      "loss": 1.5134,
      "num_input_tokens_seen": 21697328,
      "step": 3589,
      "train_runtime": 3639.6186,
      "train_tokens_per_second": 5961.429
    },
    {
      "epoch": 1.320367816091954,
      "grad_norm": 0.9165908694267273,
      "learning_rate": 6.806552549236149e-05,
      "loss": 1.5225,
      "num_input_tokens_seen": 21702928,
      "step": 3590,
      "train_runtime": 3640.5434,
      "train_tokens_per_second": 5961.453
    },
    {
      "epoch": 1.320735632183908,
      "grad_norm": 1.0751782655715942,
      "learning_rate": 6.802871341800111e-05,
      "loss": 1.4756,
      "num_input_tokens_seen": 21711568,
      "step": 3591,
      "train_runtime": 3641.8795,
      "train_tokens_per_second": 5961.638
    },
    {
      "epoch": 1.3211034482758621,
      "grad_norm": 1.0258948802947998,
      "learning_rate": 6.799190134364072e-05,
      "loss": 1.6299,
      "num_input_tokens_seen": 21716240,
      "step": 3592,
      "train_runtime": 3642.6783,
      "train_tokens_per_second": 5961.613
    },
    {
      "epoch": 1.3214712643678161,
      "grad_norm": 0.9881296753883362,
      "learning_rate": 6.795508926928033e-05,
      "loss": 1.4815,
      "num_input_tokens_seen": 21721840,
      "step": 3593,
      "train_runtime": 3643.6003,
      "train_tokens_per_second": 5961.642
    },
    {
      "epoch": 1.3218390804597702,
      "grad_norm": 0.9271832704544067,
      "learning_rate": 6.791827719491993e-05,
      "loss": 1.4684,
      "num_input_tokens_seen": 21727680,
      "step": 3594,
      "train_runtime": 3644.5697,
      "train_tokens_per_second": 5961.659
    },
    {
      "epoch": 1.3222068965517242,
      "grad_norm": 0.9892234802246094,
      "learning_rate": 6.788146512055955e-05,
      "loss": 1.6203,
      "num_input_tokens_seen": 21732800,
      "step": 3595,
      "train_runtime": 3645.4517,
      "train_tokens_per_second": 5961.621
    },
    {
      "epoch": 1.3225747126436782,
      "grad_norm": 0.9278784990310669,
      "learning_rate": 6.784465304619916e-05,
      "loss": 1.247,
      "num_input_tokens_seen": 21739424,
      "step": 3596,
      "train_runtime": 3646.5055,
      "train_tokens_per_second": 5961.714
    },
    {
      "epoch": 1.3229425287356322,
      "grad_norm": 0.9651308059692383,
      "learning_rate": 6.780784097183877e-05,
      "loss": 1.6359,
      "num_input_tokens_seen": 21744560,
      "step": 3597,
      "train_runtime": 3647.3596,
      "train_tokens_per_second": 5961.726
    },
    {
      "epoch": 1.3233103448275862,
      "grad_norm": 0.9792284369468689,
      "learning_rate": 6.777102889747838e-05,
      "loss": 1.6924,
      "num_input_tokens_seen": 21751584,
      "step": 3598,
      "train_runtime": 3648.48,
      "train_tokens_per_second": 5961.821
    },
    {
      "epoch": 1.3236781609195403,
      "grad_norm": 0.9538254141807556,
      "learning_rate": 6.773421682311798e-05,
      "loss": 1.4251,
      "num_input_tokens_seen": 21760896,
      "step": 3599,
      "train_runtime": 3649.9267,
      "train_tokens_per_second": 5962.009
    },
    {
      "epoch": 1.3240459770114943,
      "grad_norm": 1.0540046691894531,
      "learning_rate": 6.76974047487576e-05,
      "loss": 1.6302,
      "num_input_tokens_seen": 21765760,
      "step": 3600,
      "train_runtime": 3650.7609,
      "train_tokens_per_second": 5961.979
    },
    {
      "epoch": 1.3244137931034483,
      "grad_norm": 0.9055759310722351,
      "learning_rate": 6.76605926743972e-05,
      "loss": 1.3623,
      "num_input_tokens_seen": 21772736,
      "step": 3601,
      "train_runtime": 3652.7763,
      "train_tokens_per_second": 5960.599
    },
    {
      "epoch": 1.3247816091954023,
      "grad_norm": 0.9791301488876343,
      "learning_rate": 6.762378060003681e-05,
      "loss": 1.7489,
      "num_input_tokens_seen": 21778256,
      "step": 3602,
      "train_runtime": 3653.6762,
      "train_tokens_per_second": 5960.642
    },
    {
      "epoch": 1.3251494252873564,
      "grad_norm": 0.946537971496582,
      "learning_rate": 6.758696852567642e-05,
      "loss": 1.5355,
      "num_input_tokens_seen": 21786032,
      "step": 3603,
      "train_runtime": 3654.9051,
      "train_tokens_per_second": 5960.765
    },
    {
      "epoch": 1.3255172413793104,
      "grad_norm": 0.9573668837547302,
      "learning_rate": 6.755015645131604e-05,
      "loss": 1.4401,
      "num_input_tokens_seen": 21792608,
      "step": 3604,
      "train_runtime": 3655.953,
      "train_tokens_per_second": 5960.856
    },
    {
      "epoch": 1.3258850574712644,
      "grad_norm": 0.9119144678115845,
      "learning_rate": 6.751334437695565e-05,
      "loss": 1.5431,
      "num_input_tokens_seen": 21797040,
      "step": 3605,
      "train_runtime": 3656.7146,
      "train_tokens_per_second": 5960.826
    },
    {
      "epoch": 1.3262528735632184,
      "grad_norm": 1.127028465270996,
      "learning_rate": 6.747653230259525e-05,
      "loss": 1.8795,
      "num_input_tokens_seen": 21801888,
      "step": 3606,
      "train_runtime": 3657.5562,
      "train_tokens_per_second": 5960.78
    },
    {
      "epoch": 1.3266206896551724,
      "grad_norm": 0.8924864530563354,
      "learning_rate": 6.743972022823486e-05,
      "loss": 1.215,
      "num_input_tokens_seen": 21809920,
      "step": 3607,
      "train_runtime": 3658.8111,
      "train_tokens_per_second": 5960.931
    },
    {
      "epoch": 1.3269885057471265,
      "grad_norm": 0.9637953042984009,
      "learning_rate": 6.740290815387448e-05,
      "loss": 1.5693,
      "num_input_tokens_seen": 21817840,
      "step": 3608,
      "train_runtime": 3660.027,
      "train_tokens_per_second": 5961.114
    },
    {
      "epoch": 1.3273563218390805,
      "grad_norm": 0.8713181614875793,
      "learning_rate": 6.736609607951409e-05,
      "loss": 1.5456,
      "num_input_tokens_seen": 21824704,
      "step": 3609,
      "train_runtime": 3661.1169,
      "train_tokens_per_second": 5961.215
    },
    {
      "epoch": 1.3277241379310345,
      "grad_norm": 1.0000393390655518,
      "learning_rate": 6.73292840051537e-05,
      "loss": 1.6462,
      "num_input_tokens_seen": 21829360,
      "step": 3610,
      "train_runtime": 3661.9209,
      "train_tokens_per_second": 5961.177
    },
    {
      "epoch": 1.3280919540229885,
      "grad_norm": 0.9577270746231079,
      "learning_rate": 6.72924719307933e-05,
      "loss": 1.7021,
      "num_input_tokens_seen": 21834288,
      "step": 3611,
      "train_runtime": 3662.7712,
      "train_tokens_per_second": 5961.139
    },
    {
      "epoch": 1.3284597701149425,
      "grad_norm": 1.1305314302444458,
      "learning_rate": 6.725565985643291e-05,
      "loss": 1.5471,
      "num_input_tokens_seen": 21839920,
      "step": 3612,
      "train_runtime": 3663.7165,
      "train_tokens_per_second": 5961.138
    },
    {
      "epoch": 1.3288275862068966,
      "grad_norm": 0.913192868232727,
      "learning_rate": 6.721884778207252e-05,
      "loss": 1.5525,
      "num_input_tokens_seen": 21845136,
      "step": 3613,
      "train_runtime": 3664.6009,
      "train_tokens_per_second": 5961.123
    },
    {
      "epoch": 1.3291954022988506,
      "grad_norm": 0.9686322808265686,
      "learning_rate": 6.718203570771213e-05,
      "loss": 1.4421,
      "num_input_tokens_seen": 21851216,
      "step": 3614,
      "train_runtime": 3665.5821,
      "train_tokens_per_second": 5961.186
    },
    {
      "epoch": 1.3295632183908046,
      "grad_norm": 1.0518219470977783,
      "learning_rate": 6.714522363335174e-05,
      "loss": 1.7396,
      "num_input_tokens_seen": 21856368,
      "step": 3615,
      "train_runtime": 3666.4401,
      "train_tokens_per_second": 5961.196
    },
    {
      "epoch": 1.3299310344827586,
      "grad_norm": 0.9622789621353149,
      "learning_rate": 6.710841155899136e-05,
      "loss": 1.5931,
      "num_input_tokens_seen": 21861296,
      "step": 3616,
      "train_runtime": 3667.2578,
      "train_tokens_per_second": 5961.211
    },
    {
      "epoch": 1.3302988505747126,
      "grad_norm": 0.851626992225647,
      "learning_rate": 6.707159948463096e-05,
      "loss": 1.4147,
      "num_input_tokens_seen": 21868416,
      "step": 3617,
      "train_runtime": 3668.383,
      "train_tokens_per_second": 5961.323
    },
    {
      "epoch": 1.3306666666666667,
      "grad_norm": 0.9522973299026489,
      "learning_rate": 6.703478741027057e-05,
      "loss": 1.5486,
      "num_input_tokens_seen": 21874544,
      "step": 3618,
      "train_runtime": 3669.3964,
      "train_tokens_per_second": 5961.347
    },
    {
      "epoch": 1.3310344827586207,
      "grad_norm": 0.9630358219146729,
      "learning_rate": 6.699797533591018e-05,
      "loss": 1.6862,
      "num_input_tokens_seen": 21878960,
      "step": 3619,
      "train_runtime": 3670.15,
      "train_tokens_per_second": 5961.326
    },
    {
      "epoch": 1.3314022988505747,
      "grad_norm": 0.8978569507598877,
      "learning_rate": 6.696116326154979e-05,
      "loss": 1.4714,
      "num_input_tokens_seen": 21884256,
      "step": 3620,
      "train_runtime": 3671.0358,
      "train_tokens_per_second": 5961.33
    },
    {
      "epoch": 1.3317701149425287,
      "grad_norm": 0.8899297714233398,
      "learning_rate": 6.692435118718941e-05,
      "loss": 1.4217,
      "num_input_tokens_seen": 21889280,
      "step": 3621,
      "train_runtime": 3671.8648,
      "train_tokens_per_second": 5961.352
    },
    {
      "epoch": 1.3321379310344827,
      "grad_norm": 0.9578220844268799,
      "learning_rate": 6.688753911282902e-05,
      "loss": 1.5489,
      "num_input_tokens_seen": 21894656,
      "step": 3622,
      "train_runtime": 3672.8062,
      "train_tokens_per_second": 5961.288
    },
    {
      "epoch": 1.3325057471264368,
      "grad_norm": 0.9276697039604187,
      "learning_rate": 6.685072703846863e-05,
      "loss": 1.4861,
      "num_input_tokens_seen": 21899376,
      "step": 3623,
      "train_runtime": 3673.6428,
      "train_tokens_per_second": 5961.215
    },
    {
      "epoch": 1.3328735632183908,
      "grad_norm": 0.9538954496383667,
      "learning_rate": 6.681391496410822e-05,
      "loss": 1.5598,
      "num_input_tokens_seen": 21905024,
      "step": 3624,
      "train_runtime": 3674.5517,
      "train_tokens_per_second": 5961.278
    },
    {
      "epoch": 1.3332413793103448,
      "grad_norm": 0.9099012613296509,
      "learning_rate": 6.677710288974784e-05,
      "loss": 1.5987,
      "num_input_tokens_seen": 21913040,
      "step": 3625,
      "train_runtime": 3675.7981,
      "train_tokens_per_second": 5961.437
    },
    {
      "epoch": 1.3336091954022988,
      "grad_norm": 0.9048240184783936,
      "learning_rate": 6.674029081538745e-05,
      "loss": 1.6086,
      "num_input_tokens_seen": 21917856,
      "step": 3626,
      "train_runtime": 3676.6371,
      "train_tokens_per_second": 5961.387
    },
    {
      "epoch": 1.3339770114942529,
      "grad_norm": 0.9351340532302856,
      "learning_rate": 6.670347874102706e-05,
      "loss": 1.4364,
      "num_input_tokens_seen": 21922560,
      "step": 3627,
      "train_runtime": 3677.4437,
      "train_tokens_per_second": 5961.358
    },
    {
      "epoch": 1.3343448275862069,
      "grad_norm": 0.9369754791259766,
      "learning_rate": 6.666666666666667e-05,
      "loss": 1.355,
      "num_input_tokens_seen": 21930080,
      "step": 3628,
      "train_runtime": 3678.6422,
      "train_tokens_per_second": 5961.46
    },
    {
      "epoch": 1.334712643678161,
      "grad_norm": 0.967265248298645,
      "learning_rate": 6.662985459230628e-05,
      "loss": 1.4384,
      "num_input_tokens_seen": 21935488,
      "step": 3629,
      "train_runtime": 3679.5302,
      "train_tokens_per_second": 5961.492
    },
    {
      "epoch": 1.335080459770115,
      "grad_norm": 0.872542679309845,
      "learning_rate": 6.659304251794589e-05,
      "loss": 1.335,
      "num_input_tokens_seen": 21941504,
      "step": 3630,
      "train_runtime": 3680.5227,
      "train_tokens_per_second": 5961.518
    },
    {
      "epoch": 1.335448275862069,
      "grad_norm": 0.922120213508606,
      "learning_rate": 6.65562304435855e-05,
      "loss": 1.7397,
      "num_input_tokens_seen": 21946352,
      "step": 3631,
      "train_runtime": 3681.9045,
      "train_tokens_per_second": 5960.598
    },
    {
      "epoch": 1.335816091954023,
      "grad_norm": 0.9233165979385376,
      "learning_rate": 6.65194183692251e-05,
      "loss": 1.5473,
      "num_input_tokens_seen": 21953792,
      "step": 3632,
      "train_runtime": 3683.0954,
      "train_tokens_per_second": 5960.691
    },
    {
      "epoch": 1.336183908045977,
      "grad_norm": 0.9813206791877747,
      "learning_rate": 6.648260629486473e-05,
      "loss": 1.7812,
      "num_input_tokens_seen": 21958320,
      "step": 3633,
      "train_runtime": 3683.9002,
      "train_tokens_per_second": 5960.618
    },
    {
      "epoch": 1.336551724137931,
      "grad_norm": 0.874936580657959,
      "learning_rate": 6.644579422050434e-05,
      "loss": 1.3367,
      "num_input_tokens_seen": 21966848,
      "step": 3634,
      "train_runtime": 3685.2231,
      "train_tokens_per_second": 5960.792
    },
    {
      "epoch": 1.336919540229885,
      "grad_norm": 0.8871551156044006,
      "learning_rate": 6.640898214614393e-05,
      "loss": 1.4398,
      "num_input_tokens_seen": 21974352,
      "step": 3635,
      "train_runtime": 3686.3878,
      "train_tokens_per_second": 5960.944
    },
    {
      "epoch": 1.337287356321839,
      "grad_norm": 0.8556391596794128,
      "learning_rate": 6.637217007178354e-05,
      "loss": 1.4947,
      "num_input_tokens_seen": 21979328,
      "step": 3636,
      "train_runtime": 3687.2122,
      "train_tokens_per_second": 5960.961
    },
    {
      "epoch": 1.337655172413793,
      "grad_norm": 1.025610089302063,
      "learning_rate": 6.633535799742315e-05,
      "loss": 1.7505,
      "num_input_tokens_seen": 21985808,
      "step": 3637,
      "train_runtime": 3688.2614,
      "train_tokens_per_second": 5961.022
    },
    {
      "epoch": 1.338022988505747,
      "grad_norm": 0.9810093641281128,
      "learning_rate": 6.629854592306277e-05,
      "loss": 1.4707,
      "num_input_tokens_seen": 21992528,
      "step": 3638,
      "train_runtime": 3689.3316,
      "train_tokens_per_second": 5961.114
    },
    {
      "epoch": 1.338390804597701,
      "grad_norm": 0.9924654364585876,
      "learning_rate": 6.626173384870238e-05,
      "loss": 1.5178,
      "num_input_tokens_seen": 21998448,
      "step": 3639,
      "train_runtime": 3690.3049,
      "train_tokens_per_second": 5961.147
    },
    {
      "epoch": 1.3387586206896551,
      "grad_norm": 0.8719496130943298,
      "learning_rate": 6.622492177434199e-05,
      "loss": 1.3504,
      "num_input_tokens_seen": 22005936,
      "step": 3640,
      "train_runtime": 3691.4745,
      "train_tokens_per_second": 5961.286
    },
    {
      "epoch": 1.3391264367816091,
      "grad_norm": 0.8840570449829102,
      "learning_rate": 6.61881096999816e-05,
      "loss": 1.6397,
      "num_input_tokens_seen": 22014256,
      "step": 3641,
      "train_runtime": 3692.7687,
      "train_tokens_per_second": 5961.45
    },
    {
      "epoch": 1.3394942528735632,
      "grad_norm": 0.9345259666442871,
      "learning_rate": 6.615129762562121e-05,
      "loss": 1.6824,
      "num_input_tokens_seen": 22021760,
      "step": 3642,
      "train_runtime": 3693.9247,
      "train_tokens_per_second": 5961.616
    },
    {
      "epoch": 1.3398620689655172,
      "grad_norm": 0.8800603747367859,
      "learning_rate": 6.611448555126082e-05,
      "loss": 1.4757,
      "num_input_tokens_seen": 22027232,
      "step": 3643,
      "train_runtime": 3694.8251,
      "train_tokens_per_second": 5961.644
    },
    {
      "epoch": 1.3402298850574712,
      "grad_norm": 1.0404003858566284,
      "learning_rate": 6.607767347690043e-05,
      "loss": 1.6164,
      "num_input_tokens_seen": 22031760,
      "step": 3644,
      "train_runtime": 3695.6109,
      "train_tokens_per_second": 5961.602
    },
    {
      "epoch": 1.3405977011494252,
      "grad_norm": 0.9096168875694275,
      "learning_rate": 6.604086140254003e-05,
      "loss": 1.6111,
      "num_input_tokens_seen": 22038224,
      "step": 3645,
      "train_runtime": 3696.6572,
      "train_tokens_per_second": 5961.663
    },
    {
      "epoch": 1.3409655172413792,
      "grad_norm": 0.8191343545913696,
      "learning_rate": 6.600404932817966e-05,
      "loss": 1.525,
      "num_input_tokens_seen": 22047168,
      "step": 3646,
      "train_runtime": 3698.0279,
      "train_tokens_per_second": 5961.872
    },
    {
      "epoch": 1.3413333333333333,
      "grad_norm": 0.922600269317627,
      "learning_rate": 6.596723725381925e-05,
      "loss": 1.4622,
      "num_input_tokens_seen": 22051824,
      "step": 3647,
      "train_runtime": 3698.817,
      "train_tokens_per_second": 5961.859
    },
    {
      "epoch": 1.3417011494252873,
      "grad_norm": 0.8109387755393982,
      "learning_rate": 6.593042517945886e-05,
      "loss": 1.4115,
      "num_input_tokens_seen": 22060576,
      "step": 3648,
      "train_runtime": 3700.1939,
      "train_tokens_per_second": 5962.005
    },
    {
      "epoch": 1.3420689655172413,
      "grad_norm": 0.9020803570747375,
      "learning_rate": 6.589361310509847e-05,
      "loss": 1.3771,
      "num_input_tokens_seen": 22066608,
      "step": 3649,
      "train_runtime": 3701.1871,
      "train_tokens_per_second": 5962.035
    },
    {
      "epoch": 1.3424367816091953,
      "grad_norm": 1.0523520708084106,
      "learning_rate": 6.585680103073808e-05,
      "loss": 1.6008,
      "num_input_tokens_seen": 22071136,
      "step": 3650,
      "train_runtime": 3701.9872,
      "train_tokens_per_second": 5961.97
    },
    {
      "epoch": 1.3428045977011493,
      "grad_norm": 0.9008994698524475,
      "learning_rate": 6.58199889563777e-05,
      "loss": 1.4083,
      "num_input_tokens_seen": 22076624,
      "step": 3651,
      "train_runtime": 3702.898,
      "train_tokens_per_second": 5961.985
    },
    {
      "epoch": 1.3431724137931034,
      "grad_norm": 0.9648478031158447,
      "learning_rate": 6.578317688201731e-05,
      "loss": 1.6224,
      "num_input_tokens_seen": 22083296,
      "step": 3652,
      "train_runtime": 3703.9538,
      "train_tokens_per_second": 5962.087
    },
    {
      "epoch": 1.3435402298850574,
      "grad_norm": 0.9025352001190186,
      "learning_rate": 6.574636480765692e-05,
      "loss": 1.3237,
      "num_input_tokens_seen": 22091472,
      "step": 3653,
      "train_runtime": 3705.2242,
      "train_tokens_per_second": 5962.25
    },
    {
      "epoch": 1.3439080459770114,
      "grad_norm": 0.9065526127815247,
      "learning_rate": 6.570955273329651e-05,
      "loss": 1.4691,
      "num_input_tokens_seen": 22096976,
      "step": 3654,
      "train_runtime": 3706.1286,
      "train_tokens_per_second": 5962.28
    },
    {
      "epoch": 1.3442758620689657,
      "grad_norm": 0.9897463321685791,
      "learning_rate": 6.567274065893614e-05,
      "loss": 1.5573,
      "num_input_tokens_seen": 22101488,
      "step": 3655,
      "train_runtime": 3706.9331,
      "train_tokens_per_second": 5962.203
    },
    {
      "epoch": 1.3446436781609195,
      "grad_norm": 0.9158437848091125,
      "learning_rate": 6.563592858457575e-05,
      "loss": 1.1949,
      "num_input_tokens_seen": 22113408,
      "step": 3656,
      "train_runtime": 3708.7042,
      "train_tokens_per_second": 5962.57
    },
    {
      "epoch": 1.3450114942528737,
      "grad_norm": 0.9928573966026306,
      "learning_rate": 6.559911651021535e-05,
      "loss": 1.3919,
      "num_input_tokens_seen": 22123456,
      "step": 3657,
      "train_runtime": 3710.2097,
      "train_tokens_per_second": 5962.859
    },
    {
      "epoch": 1.3453793103448275,
      "grad_norm": 0.8950387239456177,
      "learning_rate": 6.556230443585496e-05,
      "loss": 1.6679,
      "num_input_tokens_seen": 22128816,
      "step": 3658,
      "train_runtime": 3711.1151,
      "train_tokens_per_second": 5962.848
    },
    {
      "epoch": 1.3457471264367817,
      "grad_norm": 1.1250476837158203,
      "learning_rate": 6.552549236149457e-05,
      "loss": 1.6202,
      "num_input_tokens_seen": 22134496,
      "step": 3659,
      "train_runtime": 3712.0539,
      "train_tokens_per_second": 5962.87
    },
    {
      "epoch": 1.3461149425287355,
      "grad_norm": 0.9576421976089478,
      "learning_rate": 6.548868028713418e-05,
      "loss": 1.4687,
      "num_input_tokens_seen": 22141824,
      "step": 3660,
      "train_runtime": 3713.2448,
      "train_tokens_per_second": 5962.931
    },
    {
      "epoch": 1.3464827586206898,
      "grad_norm": 0.9883807301521301,
      "learning_rate": 6.545186821277379e-05,
      "loss": 1.6294,
      "num_input_tokens_seen": 22146944,
      "step": 3661,
      "train_runtime": 3714.6739,
      "train_tokens_per_second": 5962.016
    },
    {
      "epoch": 1.3468505747126436,
      "grad_norm": 0.9688007235527039,
      "learning_rate": 6.54150561384134e-05,
      "loss": 1.7011,
      "num_input_tokens_seen": 22151952,
      "step": 3662,
      "train_runtime": 3715.5222,
      "train_tokens_per_second": 5962.002
    },
    {
      "epoch": 1.3472183908045978,
      "grad_norm": 0.9309114217758179,
      "learning_rate": 6.537824406405302e-05,
      "loss": 1.5218,
      "num_input_tokens_seen": 22157680,
      "step": 3663,
      "train_runtime": 3716.4592,
      "train_tokens_per_second": 5962.04
    },
    {
      "epoch": 1.3475862068965516,
      "grad_norm": 0.9676522016525269,
      "learning_rate": 6.534143198969263e-05,
      "loss": 1.4096,
      "num_input_tokens_seen": 22162240,
      "step": 3664,
      "train_runtime": 3717.1989,
      "train_tokens_per_second": 5962.081
    },
    {
      "epoch": 1.3479540229885059,
      "grad_norm": 1.0314520597457886,
      "learning_rate": 6.530461991533223e-05,
      "loss": 1.6148,
      "num_input_tokens_seen": 22167472,
      "step": 3665,
      "train_runtime": 3718.059,
      "train_tokens_per_second": 5962.109
    },
    {
      "epoch": 1.3483218390804597,
      "grad_norm": 0.9597965478897095,
      "learning_rate": 6.526780784097183e-05,
      "loss": 1.7207,
      "num_input_tokens_seen": 22173968,
      "step": 3666,
      "train_runtime": 3719.1156,
      "train_tokens_per_second": 5962.162
    },
    {
      "epoch": 1.348689655172414,
      "grad_norm": 1.004009485244751,
      "learning_rate": 6.523099576661144e-05,
      "loss": 1.3832,
      "num_input_tokens_seen": 22178624,
      "step": 3667,
      "train_runtime": 3719.9082,
      "train_tokens_per_second": 5962.143
    },
    {
      "epoch": 1.3490574712643677,
      "grad_norm": 0.955319881439209,
      "learning_rate": 6.519418369225107e-05,
      "loss": 1.6779,
      "num_input_tokens_seen": 22183008,
      "step": 3668,
      "train_runtime": 3720.6684,
      "train_tokens_per_second": 5962.103
    },
    {
      "epoch": 1.349425287356322,
      "grad_norm": 0.9129797220230103,
      "learning_rate": 6.515737161789068e-05,
      "loss": 1.4714,
      "num_input_tokens_seen": 22194640,
      "step": 3669,
      "train_runtime": 3722.4084,
      "train_tokens_per_second": 5962.441
    },
    {
      "epoch": 1.3497931034482757,
      "grad_norm": 0.8621333241462708,
      "learning_rate": 6.512055954353028e-05,
      "loss": 1.4761,
      "num_input_tokens_seen": 22202496,
      "step": 3670,
      "train_runtime": 3723.6421,
      "train_tokens_per_second": 5962.575
    },
    {
      "epoch": 1.35016091954023,
      "grad_norm": 0.9802247285842896,
      "learning_rate": 6.508374746916989e-05,
      "loss": 1.6187,
      "num_input_tokens_seen": 22206960,
      "step": 3671,
      "train_runtime": 3724.4325,
      "train_tokens_per_second": 5962.508
    },
    {
      "epoch": 1.3505287356321838,
      "grad_norm": 0.9478545188903809,
      "learning_rate": 6.50469353948095e-05,
      "loss": 1.4785,
      "num_input_tokens_seen": 22213264,
      "step": 3672,
      "train_runtime": 3725.4422,
      "train_tokens_per_second": 5962.585
    },
    {
      "epoch": 1.350896551724138,
      "grad_norm": 1.021788477897644,
      "learning_rate": 6.501012332044911e-05,
      "loss": 1.5397,
      "num_input_tokens_seen": 22217888,
      "step": 3673,
      "train_runtime": 3726.2641,
      "train_tokens_per_second": 5962.51
    },
    {
      "epoch": 1.3512643678160918,
      "grad_norm": 0.9838345646858215,
      "learning_rate": 6.497331124608872e-05,
      "loss": 1.53,
      "num_input_tokens_seen": 22223344,
      "step": 3674,
      "train_runtime": 3727.183,
      "train_tokens_per_second": 5962.504
    },
    {
      "epoch": 1.351632183908046,
      "grad_norm": 0.9164777398109436,
      "learning_rate": 6.493649917172833e-05,
      "loss": 1.4674,
      "num_input_tokens_seen": 22228560,
      "step": 3675,
      "train_runtime": 3728.0481,
      "train_tokens_per_second": 5962.52
    },
    {
      "epoch": 1.3519999999999999,
      "grad_norm": 0.8565971851348877,
      "learning_rate": 6.489968709736795e-05,
      "loss": 1.4083,
      "num_input_tokens_seen": 22236192,
      "step": 3676,
      "train_runtime": 3729.2419,
      "train_tokens_per_second": 5962.657
    },
    {
      "epoch": 1.3523678160919541,
      "grad_norm": 0.9358131289482117,
      "learning_rate": 6.486287502300755e-05,
      "loss": 1.5632,
      "num_input_tokens_seen": 22241488,
      "step": 3677,
      "train_runtime": 3730.1418,
      "train_tokens_per_second": 5962.639
    },
    {
      "epoch": 1.352735632183908,
      "grad_norm": 0.945747971534729,
      "learning_rate": 6.482606294864716e-05,
      "loss": 1.4429,
      "num_input_tokens_seen": 22248832,
      "step": 3678,
      "train_runtime": 3731.2897,
      "train_tokens_per_second": 5962.773
    },
    {
      "epoch": 1.3531034482758622,
      "grad_norm": 0.9707575440406799,
      "learning_rate": 6.478925087428676e-05,
      "loss": 1.6877,
      "num_input_tokens_seen": 22259072,
      "step": 3679,
      "train_runtime": 3732.8159,
      "train_tokens_per_second": 5963.078
    },
    {
      "epoch": 1.3534712643678162,
      "grad_norm": 0.870474100112915,
      "learning_rate": 6.475243879992637e-05,
      "loss": 1.4871,
      "num_input_tokens_seen": 22267280,
      "step": 3680,
      "train_runtime": 3734.0859,
      "train_tokens_per_second": 5963.248
    },
    {
      "epoch": 1.3538390804597702,
      "grad_norm": 0.8968935012817383,
      "learning_rate": 6.4715626725566e-05,
      "loss": 1.5029,
      "num_input_tokens_seen": 22272032,
      "step": 3681,
      "train_runtime": 3734.9229,
      "train_tokens_per_second": 5963.184
    },
    {
      "epoch": 1.3542068965517242,
      "grad_norm": 1.1222615242004395,
      "learning_rate": 6.46788146512056e-05,
      "loss": 1.6715,
      "num_input_tokens_seen": 22278448,
      "step": 3682,
      "train_runtime": 3735.9483,
      "train_tokens_per_second": 5963.265
    },
    {
      "epoch": 1.3545747126436782,
      "grad_norm": 0.9637027382850647,
      "learning_rate": 6.464200257684521e-05,
      "loss": 1.6128,
      "num_input_tokens_seen": 22283280,
      "step": 3683,
      "train_runtime": 3736.7621,
      "train_tokens_per_second": 5963.259
    },
    {
      "epoch": 1.3549425287356323,
      "grad_norm": 0.9350388050079346,
      "learning_rate": 6.460519050248481e-05,
      "loss": 1.6421,
      "num_input_tokens_seen": 22289696,
      "step": 3684,
      "train_runtime": 3737.8179,
      "train_tokens_per_second": 5963.291
    },
    {
      "epoch": 1.3553103448275863,
      "grad_norm": 0.9158656597137451,
      "learning_rate": 6.456837842812443e-05,
      "loss": 1.6688,
      "num_input_tokens_seen": 22296272,
      "step": 3685,
      "train_runtime": 3738.9153,
      "train_tokens_per_second": 5963.3
    },
    {
      "epoch": 1.3556781609195403,
      "grad_norm": 0.8514031767845154,
      "learning_rate": 6.453156635376404e-05,
      "loss": 1.4368,
      "num_input_tokens_seen": 22305472,
      "step": 3686,
      "train_runtime": 3740.3086,
      "train_tokens_per_second": 5963.538
    },
    {
      "epoch": 1.3560459770114943,
      "grad_norm": 0.9319664239883423,
      "learning_rate": 6.449475427940365e-05,
      "loss": 1.6136,
      "num_input_tokens_seen": 22311664,
      "step": 3687,
      "train_runtime": 3741.2943,
      "train_tokens_per_second": 5963.622
    },
    {
      "epoch": 1.3564137931034483,
      "grad_norm": 0.914702832698822,
      "learning_rate": 6.445794220504326e-05,
      "loss": 1.5977,
      "num_input_tokens_seen": 22317712,
      "step": 3688,
      "train_runtime": 3742.3073,
      "train_tokens_per_second": 5963.624
    },
    {
      "epoch": 1.3567816091954024,
      "grad_norm": 0.7683727145195007,
      "learning_rate": 6.442113013068287e-05,
      "loss": 1.2976,
      "num_input_tokens_seen": 22329168,
      "step": 3689,
      "train_runtime": 3744.0179,
      "train_tokens_per_second": 5963.959
    },
    {
      "epoch": 1.3571494252873564,
      "grad_norm": 0.9728146195411682,
      "learning_rate": 6.438431805632248e-05,
      "loss": 1.4736,
      "num_input_tokens_seen": 22334752,
      "step": 3690,
      "train_runtime": 3744.9788,
      "train_tokens_per_second": 5963.92
    },
    {
      "epoch": 1.3575172413793104,
      "grad_norm": 0.8961226940155029,
      "learning_rate": 6.434750598196208e-05,
      "loss": 1.4565,
      "num_input_tokens_seen": 22339472,
      "step": 3691,
      "train_runtime": 3746.2865,
      "train_tokens_per_second": 5963.098
    },
    {
      "epoch": 1.3578850574712644,
      "grad_norm": 0.9248019456863403,
      "learning_rate": 6.431069390760169e-05,
      "loss": 1.4495,
      "num_input_tokens_seen": 22345936,
      "step": 3692,
      "train_runtime": 3747.3179,
      "train_tokens_per_second": 5963.181
    },
    {
      "epoch": 1.3582528735632184,
      "grad_norm": 0.9686160683631897,
      "learning_rate": 6.427388183324132e-05,
      "loss": 1.4975,
      "num_input_tokens_seen": 22353504,
      "step": 3693,
      "train_runtime": 3748.5248,
      "train_tokens_per_second": 5963.28
    },
    {
      "epoch": 1.3586206896551725,
      "grad_norm": 0.9702744483947754,
      "learning_rate": 6.423706975888092e-05,
      "loss": 1.717,
      "num_input_tokens_seen": 22359664,
      "step": 3694,
      "train_runtime": 3749.5168,
      "train_tokens_per_second": 5963.345
    },
    {
      "epoch": 1.3589885057471265,
      "grad_norm": 0.8716999888420105,
      "learning_rate": 6.420025768452052e-05,
      "loss": 1.3682,
      "num_input_tokens_seen": 22369984,
      "step": 3695,
      "train_runtime": 3751.1082,
      "train_tokens_per_second": 5963.567
    },
    {
      "epoch": 1.3593563218390805,
      "grad_norm": 0.9872491359710693,
      "learning_rate": 6.416344561016013e-05,
      "loss": 1.5221,
      "num_input_tokens_seen": 22376240,
      "step": 3696,
      "train_runtime": 3752.1485,
      "train_tokens_per_second": 5963.581
    },
    {
      "epoch": 1.3597241379310345,
      "grad_norm": 0.9793648719787598,
      "learning_rate": 6.412663353579974e-05,
      "loss": 1.2984,
      "num_input_tokens_seen": 22383072,
      "step": 3697,
      "train_runtime": 3753.2558,
      "train_tokens_per_second": 5963.641
    },
    {
      "epoch": 1.3600919540229885,
      "grad_norm": 0.9976056814193726,
      "learning_rate": 6.408982146143936e-05,
      "loss": 1.6703,
      "num_input_tokens_seen": 22388144,
      "step": 3698,
      "train_runtime": 3754.1147,
      "train_tokens_per_second": 5963.628
    },
    {
      "epoch": 1.3604597701149426,
      "grad_norm": 0.8496326804161072,
      "learning_rate": 6.405300938707897e-05,
      "loss": 1.3574,
      "num_input_tokens_seen": 22393616,
      "step": 3699,
      "train_runtime": 3755.0069,
      "train_tokens_per_second": 5963.668
    },
    {
      "epoch": 1.3608275862068966,
      "grad_norm": 0.9885788559913635,
      "learning_rate": 6.401619731271858e-05,
      "loss": 1.3272,
      "num_input_tokens_seen": 22399840,
      "step": 3700,
      "train_runtime": 3756.0195,
      "train_tokens_per_second": 5963.718
    },
    {
      "epoch": 1.3611954022988506,
      "grad_norm": 0.9272667765617371,
      "learning_rate": 6.397938523835819e-05,
      "loss": 1.7168,
      "num_input_tokens_seen": 22407200,
      "step": 3701,
      "train_runtime": 3757.164,
      "train_tokens_per_second": 5963.86
    },
    {
      "epoch": 1.3615632183908046,
      "grad_norm": 0.9542990922927856,
      "learning_rate": 6.39425731639978e-05,
      "loss": 1.4067,
      "num_input_tokens_seen": 22412464,
      "step": 3702,
      "train_runtime": 3758.0576,
      "train_tokens_per_second": 5963.843
    },
    {
      "epoch": 1.3619310344827587,
      "grad_norm": 0.8829864263534546,
      "learning_rate": 6.39057610896374e-05,
      "loss": 1.3871,
      "num_input_tokens_seen": 22418432,
      "step": 3703,
      "train_runtime": 3759.024,
      "train_tokens_per_second": 5963.897
    },
    {
      "epoch": 1.3622988505747127,
      "grad_norm": 0.9515782594680786,
      "learning_rate": 6.386894901527701e-05,
      "loss": 1.3696,
      "num_input_tokens_seen": 22427488,
      "step": 3704,
      "train_runtime": 3760.3897,
      "train_tokens_per_second": 5964.139
    },
    {
      "epoch": 1.3626666666666667,
      "grad_norm": 0.9590629935264587,
      "learning_rate": 6.383213694091662e-05,
      "loss": 1.4148,
      "num_input_tokens_seen": 22432688,
      "step": 3705,
      "train_runtime": 3761.2764,
      "train_tokens_per_second": 5964.116
    },
    {
      "epoch": 1.3630344827586207,
      "grad_norm": 0.9900871515274048,
      "learning_rate": 6.379532486655624e-05,
      "loss": 1.7678,
      "num_input_tokens_seen": 22438384,
      "step": 3706,
      "train_runtime": 3762.2145,
      "train_tokens_per_second": 5964.143
    },
    {
      "epoch": 1.3634022988505747,
      "grad_norm": 0.9610579013824463,
      "learning_rate": 6.375851279219584e-05,
      "loss": 1.595,
      "num_input_tokens_seen": 22443296,
      "step": 3707,
      "train_runtime": 3763.0421,
      "train_tokens_per_second": 5964.136
    },
    {
      "epoch": 1.3637701149425288,
      "grad_norm": 1.0513726472854614,
      "learning_rate": 6.372170071783545e-05,
      "loss": 1.475,
      "num_input_tokens_seen": 22448768,
      "step": 3708,
      "train_runtime": 3763.971,
      "train_tokens_per_second": 5964.118
    },
    {
      "epoch": 1.3641379310344828,
      "grad_norm": 0.9578030109405518,
      "learning_rate": 6.368488864347506e-05,
      "loss": 1.4103,
      "num_input_tokens_seen": 22454512,
      "step": 3709,
      "train_runtime": 3764.9307,
      "train_tokens_per_second": 5964.124
    },
    {
      "epoch": 1.3645057471264368,
      "grad_norm": 1.000126600265503,
      "learning_rate": 6.364807656911467e-05,
      "loss": 1.5431,
      "num_input_tokens_seen": 22460944,
      "step": 3710,
      "train_runtime": 3765.9901,
      "train_tokens_per_second": 5964.154
    },
    {
      "epoch": 1.3648735632183908,
      "grad_norm": 0.9518430829048157,
      "learning_rate": 6.361126449475429e-05,
      "loss": 1.7073,
      "num_input_tokens_seen": 22465600,
      "step": 3711,
      "train_runtime": 3766.7729,
      "train_tokens_per_second": 5964.15
    },
    {
      "epoch": 1.3652413793103448,
      "grad_norm": 0.9710787534713745,
      "learning_rate": 6.35744524203939e-05,
      "loss": 1.4007,
      "num_input_tokens_seen": 22472816,
      "step": 3712,
      "train_runtime": 3767.905,
      "train_tokens_per_second": 5964.273
    },
    {
      "epoch": 1.3656091954022989,
      "grad_norm": 1.0492024421691895,
      "learning_rate": 6.35376403460335e-05,
      "loss": 1.6356,
      "num_input_tokens_seen": 22479712,
      "step": 3713,
      "train_runtime": 3768.991,
      "train_tokens_per_second": 5964.385
    },
    {
      "epoch": 1.3659770114942529,
      "grad_norm": 0.9096162915229797,
      "learning_rate": 6.35008282716731e-05,
      "loss": 1.3954,
      "num_input_tokens_seen": 22484544,
      "step": 3714,
      "train_runtime": 3769.8286,
      "train_tokens_per_second": 5964.341
    },
    {
      "epoch": 1.366344827586207,
      "grad_norm": 0.9547939300537109,
      "learning_rate": 6.346401619731272e-05,
      "loss": 1.5207,
      "num_input_tokens_seen": 22489792,
      "step": 3715,
      "train_runtime": 3770.7095,
      "train_tokens_per_second": 5964.34
    },
    {
      "epoch": 1.366712643678161,
      "grad_norm": 0.9628258347511292,
      "learning_rate": 6.342720412295233e-05,
      "loss": 1.5337,
      "num_input_tokens_seen": 22495216,
      "step": 3716,
      "train_runtime": 3771.6426,
      "train_tokens_per_second": 5964.302
    },
    {
      "epoch": 1.367080459770115,
      "grad_norm": 0.9746856093406677,
      "learning_rate": 6.339039204859194e-05,
      "loss": 1.6134,
      "num_input_tokens_seen": 22499936,
      "step": 3717,
      "train_runtime": 3772.4733,
      "train_tokens_per_second": 5964.24
    },
    {
      "epoch": 1.367448275862069,
      "grad_norm": 0.9883073568344116,
      "learning_rate": 6.335357997423155e-05,
      "loss": 1.4514,
      "num_input_tokens_seen": 22506736,
      "step": 3718,
      "train_runtime": 3773.5581,
      "train_tokens_per_second": 5964.327
    },
    {
      "epoch": 1.367816091954023,
      "grad_norm": 1.0037277936935425,
      "learning_rate": 6.331676789987116e-05,
      "loss": 1.6926,
      "num_input_tokens_seen": 22513664,
      "step": 3719,
      "train_runtime": 3774.6735,
      "train_tokens_per_second": 5964.4
    },
    {
      "epoch": 1.368183908045977,
      "grad_norm": 0.8952387571334839,
      "learning_rate": 6.327995582551077e-05,
      "loss": 1.3581,
      "num_input_tokens_seen": 22519008,
      "step": 3720,
      "train_runtime": 3775.5473,
      "train_tokens_per_second": 5964.435
    },
    {
      "epoch": 1.368551724137931,
      "grad_norm": 0.9532164335250854,
      "learning_rate": 6.324314375115038e-05,
      "loss": 1.5126,
      "num_input_tokens_seen": 22523840,
      "step": 3721,
      "train_runtime": 3776.924,
      "train_tokens_per_second": 5963.541
    },
    {
      "epoch": 1.368919540229885,
      "grad_norm": 0.9461865425109863,
      "learning_rate": 6.320633167678999e-05,
      "loss": 1.7382,
      "num_input_tokens_seen": 22529616,
      "step": 3722,
      "train_runtime": 3777.9044,
      "train_tokens_per_second": 5963.522
    },
    {
      "epoch": 1.369287356321839,
      "grad_norm": 0.9506685733795166,
      "learning_rate": 6.316951960242961e-05,
      "loss": 1.6842,
      "num_input_tokens_seen": 22534864,
      "step": 3723,
      "train_runtime": 3778.8041,
      "train_tokens_per_second": 5963.491
    },
    {
      "epoch": 1.369655172413793,
      "grad_norm": 0.8528960347175598,
      "learning_rate": 6.313270752806922e-05,
      "loss": 1.5325,
      "num_input_tokens_seen": 22541376,
      "step": 3724,
      "train_runtime": 3779.8939,
      "train_tokens_per_second": 5963.494
    },
    {
      "epoch": 1.370022988505747,
      "grad_norm": 0.8038332462310791,
      "learning_rate": 6.309589545370881e-05,
      "loss": 1.5505,
      "num_input_tokens_seen": 22549648,
      "step": 3725,
      "train_runtime": 3781.1884,
      "train_tokens_per_second": 5963.64
    },
    {
      "epoch": 1.3703908045977011,
      "grad_norm": 0.9963042140007019,
      "learning_rate": 6.305908337934842e-05,
      "loss": 1.6605,
      "num_input_tokens_seen": 22553616,
      "step": 3726,
      "train_runtime": 3781.8811,
      "train_tokens_per_second": 5963.597
    },
    {
      "epoch": 1.3707586206896551,
      "grad_norm": 0.8822771310806274,
      "learning_rate": 6.302227130498803e-05,
      "loss": 1.3603,
      "num_input_tokens_seen": 22558448,
      "step": 3727,
      "train_runtime": 3782.7202,
      "train_tokens_per_second": 5963.552
    },
    {
      "epoch": 1.3711264367816092,
      "grad_norm": 0.8757033348083496,
      "learning_rate": 6.298545923062765e-05,
      "loss": 1.2826,
      "num_input_tokens_seen": 22564800,
      "step": 3728,
      "train_runtime": 3783.745,
      "train_tokens_per_second": 5963.615
    },
    {
      "epoch": 1.3714942528735632,
      "grad_norm": 0.9956175684928894,
      "learning_rate": 6.294864715626726e-05,
      "loss": 1.548,
      "num_input_tokens_seen": 22569344,
      "step": 3729,
      "train_runtime": 3784.5373,
      "train_tokens_per_second": 5963.568
    },
    {
      "epoch": 1.3718620689655172,
      "grad_norm": 1.0362850427627563,
      "learning_rate": 6.291183508190687e-05,
      "loss": 1.5132,
      "num_input_tokens_seen": 22576592,
      "step": 3730,
      "train_runtime": 3785.6673,
      "train_tokens_per_second": 5963.702
    },
    {
      "epoch": 1.3722298850574712,
      "grad_norm": 0.9091210961341858,
      "learning_rate": 6.287502300754648e-05,
      "loss": 1.4306,
      "num_input_tokens_seen": 22584384,
      "step": 3731,
      "train_runtime": 3786.8804,
      "train_tokens_per_second": 5963.849
    },
    {
      "epoch": 1.3725977011494253,
      "grad_norm": 1.0187913179397583,
      "learning_rate": 6.283821093318609e-05,
      "loss": 1.6564,
      "num_input_tokens_seen": 22591088,
      "step": 3732,
      "train_runtime": 3787.9616,
      "train_tokens_per_second": 5963.917
    },
    {
      "epoch": 1.3729655172413793,
      "grad_norm": 0.9769201278686523,
      "learning_rate": 6.28013988588257e-05,
      "loss": 1.4914,
      "num_input_tokens_seen": 22596080,
      "step": 3733,
      "train_runtime": 3788.8281,
      "train_tokens_per_second": 5963.87
    },
    {
      "epoch": 1.3733333333333333,
      "grad_norm": 1.039028525352478,
      "learning_rate": 6.276458678446531e-05,
      "loss": 1.6656,
      "num_input_tokens_seen": 22600944,
      "step": 3734,
      "train_runtime": 3789.6701,
      "train_tokens_per_second": 5963.829
    },
    {
      "epoch": 1.3737011494252873,
      "grad_norm": 0.8745564222335815,
      "learning_rate": 6.272777471010492e-05,
      "loss": 1.4106,
      "num_input_tokens_seen": 22610016,
      "step": 3735,
      "train_runtime": 3791.0464,
      "train_tokens_per_second": 5964.057
    },
    {
      "epoch": 1.3740689655172413,
      "grad_norm": 0.9492440223693848,
      "learning_rate": 6.269096263574454e-05,
      "loss": 1.4987,
      "num_input_tokens_seen": 22616496,
      "step": 3736,
      "train_runtime": 3792.1119,
      "train_tokens_per_second": 5964.09
    },
    {
      "epoch": 1.3744367816091954,
      "grad_norm": 1.0139230489730835,
      "learning_rate": 6.265415056138413e-05,
      "loss": 1.5855,
      "num_input_tokens_seen": 22621088,
      "step": 3737,
      "train_runtime": 3792.9052,
      "train_tokens_per_second": 5964.053
    },
    {
      "epoch": 1.3748045977011494,
      "grad_norm": 1.0495636463165283,
      "learning_rate": 6.261733848702374e-05,
      "loss": 1.5869,
      "num_input_tokens_seen": 22625648,
      "step": 3738,
      "train_runtime": 3793.6646,
      "train_tokens_per_second": 5964.061
    },
    {
      "epoch": 1.3751724137931034,
      "grad_norm": 0.938824474811554,
      "learning_rate": 6.258052641266335e-05,
      "loss": 1.5636,
      "num_input_tokens_seen": 22630816,
      "step": 3739,
      "train_runtime": 3794.4998,
      "train_tokens_per_second": 5964.111
    },
    {
      "epoch": 1.3755402298850574,
      "grad_norm": 0.9487571716308594,
      "learning_rate": 6.254371433830296e-05,
      "loss": 1.3919,
      "num_input_tokens_seen": 22638208,
      "step": 3740,
      "train_runtime": 3795.6587,
      "train_tokens_per_second": 5964.237
    },
    {
      "epoch": 1.3759080459770114,
      "grad_norm": 1.2077070474624634,
      "learning_rate": 6.250690226394258e-05,
      "loss": 1.3998,
      "num_input_tokens_seen": 22643920,
      "step": 3741,
      "train_runtime": 3796.6168,
      "train_tokens_per_second": 5964.236
    },
    {
      "epoch": 1.3762758620689655,
      "grad_norm": 0.8746975064277649,
      "learning_rate": 6.247009018958219e-05,
      "loss": 1.4574,
      "num_input_tokens_seen": 22648192,
      "step": 3742,
      "train_runtime": 3797.3543,
      "train_tokens_per_second": 5964.203
    },
    {
      "epoch": 1.3766436781609195,
      "grad_norm": 0.910503625869751,
      "learning_rate": 6.243327811522179e-05,
      "loss": 1.2189,
      "num_input_tokens_seen": 22654592,
      "step": 3743,
      "train_runtime": 3798.3986,
      "train_tokens_per_second": 5964.248
    },
    {
      "epoch": 1.3770114942528735,
      "grad_norm": 1.0025415420532227,
      "learning_rate": 6.23964660408614e-05,
      "loss": 1.4245,
      "num_input_tokens_seen": 22659616,
      "step": 3744,
      "train_runtime": 3799.2432,
      "train_tokens_per_second": 5964.245
    },
    {
      "epoch": 1.3773793103448275,
      "grad_norm": 0.8953913450241089,
      "learning_rate": 6.235965396650102e-05,
      "loss": 1.432,
      "num_input_tokens_seen": 22665488,
      "step": 3745,
      "train_runtime": 3800.1936,
      "train_tokens_per_second": 5964.298
    },
    {
      "epoch": 1.3777471264367815,
      "grad_norm": 0.893984854221344,
      "learning_rate": 6.232284189214063e-05,
      "loss": 1.4849,
      "num_input_tokens_seen": 22671952,
      "step": 3746,
      "train_runtime": 3801.2172,
      "train_tokens_per_second": 5964.393
    },
    {
      "epoch": 1.3781149425287356,
      "grad_norm": 0.8688514232635498,
      "learning_rate": 6.228602981778024e-05,
      "loss": 1.3543,
      "num_input_tokens_seen": 22680576,
      "step": 3747,
      "train_runtime": 3802.5607,
      "train_tokens_per_second": 5964.553
    },
    {
      "epoch": 1.3784827586206896,
      "grad_norm": 0.9506460428237915,
      "learning_rate": 6.224921774341985e-05,
      "loss": 1.5756,
      "num_input_tokens_seen": 22685200,
      "step": 3748,
      "train_runtime": 3803.3845,
      "train_tokens_per_second": 5964.477
    },
    {
      "epoch": 1.3788505747126436,
      "grad_norm": 1.0392670631408691,
      "learning_rate": 6.221240566905945e-05,
      "loss": 1.4427,
      "num_input_tokens_seen": 22694224,
      "step": 3749,
      "train_runtime": 3804.787,
      "train_tokens_per_second": 5964.65
    },
    {
      "epoch": 1.3792183908045976,
      "grad_norm": 0.888479471206665,
      "learning_rate": 6.217559359469906e-05,
      "loss": 1.4515,
      "num_input_tokens_seen": 22700784,
      "step": 3750,
      "train_runtime": 3805.8518,
      "train_tokens_per_second": 5964.705
    },
    {
      "epoch": 1.3795862068965516,
      "grad_norm": 0.9241269826889038,
      "learning_rate": 6.213878152033867e-05,
      "loss": 1.5077,
      "num_input_tokens_seen": 22706096,
      "step": 3751,
      "train_runtime": 3807.3533,
      "train_tokens_per_second": 5963.748
    },
    {
      "epoch": 1.3799540229885057,
      "grad_norm": 0.9859208464622498,
      "learning_rate": 6.210196944597828e-05,
      "loss": 1.5239,
      "num_input_tokens_seen": 22710528,
      "step": 3752,
      "train_runtime": 3808.1141,
      "train_tokens_per_second": 5963.721
    },
    {
      "epoch": 1.3803218390804597,
      "grad_norm": 0.9163882732391357,
      "learning_rate": 6.20651573716179e-05,
      "loss": 1.5234,
      "num_input_tokens_seen": 22715776,
      "step": 3753,
      "train_runtime": 3808.9773,
      "train_tokens_per_second": 5963.747
    },
    {
      "epoch": 1.3806896551724137,
      "grad_norm": 0.9750805497169495,
      "learning_rate": 6.202834529725751e-05,
      "loss": 1.5073,
      "num_input_tokens_seen": 22722672,
      "step": 3754,
      "train_runtime": 3810.0805,
      "train_tokens_per_second": 5963.83
    },
    {
      "epoch": 1.3810574712643677,
      "grad_norm": 0.9713814854621887,
      "learning_rate": 6.199153322289711e-05,
      "loss": 1.4555,
      "num_input_tokens_seen": 22729888,
      "step": 3755,
      "train_runtime": 3811.2371,
      "train_tokens_per_second": 5963.913
    },
    {
      "epoch": 1.381425287356322,
      "grad_norm": 0.9078335165977478,
      "learning_rate": 6.195472114853672e-05,
      "loss": 1.4817,
      "num_input_tokens_seen": 22736288,
      "step": 3756,
      "train_runtime": 3812.2592,
      "train_tokens_per_second": 5963.993
    },
    {
      "epoch": 1.3817931034482758,
      "grad_norm": 0.9787589311599731,
      "learning_rate": 6.191790907417633e-05,
      "loss": 1.5476,
      "num_input_tokens_seen": 22740816,
      "step": 3757,
      "train_runtime": 3813.0684,
      "train_tokens_per_second": 5963.915
    },
    {
      "epoch": 1.38216091954023,
      "grad_norm": 1.0089010000228882,
      "learning_rate": 6.188109699981595e-05,
      "loss": 1.5927,
      "num_input_tokens_seen": 22746560,
      "step": 3758,
      "train_runtime": 3814.0428,
      "train_tokens_per_second": 5963.897
    },
    {
      "epoch": 1.3825287356321838,
      "grad_norm": 0.9962058067321777,
      "learning_rate": 6.184428492545556e-05,
      "loss": 1.3135,
      "num_input_tokens_seen": 22751584,
      "step": 3759,
      "train_runtime": 3814.8893,
      "train_tokens_per_second": 5963.891
    },
    {
      "epoch": 1.382896551724138,
      "grad_norm": 1.0328810214996338,
      "learning_rate": 6.180747285109517e-05,
      "loss": 1.6177,
      "num_input_tokens_seen": 22757520,
      "step": 3760,
      "train_runtime": 3815.8668,
      "train_tokens_per_second": 5963.919
    },
    {
      "epoch": 1.3832643678160919,
      "grad_norm": 1.1473071575164795,
      "learning_rate": 6.177066077673477e-05,
      "loss": 1.6739,
      "num_input_tokens_seen": 22761824,
      "step": 3761,
      "train_runtime": 3816.6343,
      "train_tokens_per_second": 5963.847
    },
    {
      "epoch": 1.383632183908046,
      "grad_norm": 1.0062990188598633,
      "learning_rate": 6.173384870237438e-05,
      "loss": 1.7144,
      "num_input_tokens_seen": 22768272,
      "step": 3762,
      "train_runtime": 3817.6882,
      "train_tokens_per_second": 5963.89
    },
    {
      "epoch": 1.384,
      "grad_norm": 0.9547836184501648,
      "learning_rate": 6.169703662801399e-05,
      "loss": 1.439,
      "num_input_tokens_seen": 22775728,
      "step": 3763,
      "train_runtime": 3818.8529,
      "train_tokens_per_second": 5964.023
    },
    {
      "epoch": 1.3843678160919541,
      "grad_norm": 0.9581781029701233,
      "learning_rate": 6.16602245536536e-05,
      "loss": 1.5565,
      "num_input_tokens_seen": 22781584,
      "step": 3764,
      "train_runtime": 3819.8387,
      "train_tokens_per_second": 5964.017
    },
    {
      "epoch": 1.384735632183908,
      "grad_norm": 0.9812362194061279,
      "learning_rate": 6.162341247929321e-05,
      "loss": 1.4536,
      "num_input_tokens_seen": 22787184,
      "step": 3765,
      "train_runtime": 3820.7584,
      "train_tokens_per_second": 5964.047
    },
    {
      "epoch": 1.3851034482758622,
      "grad_norm": 0.9955962896347046,
      "learning_rate": 6.158660040493283e-05,
      "loss": 1.5392,
      "num_input_tokens_seen": 22792768,
      "step": 3766,
      "train_runtime": 3821.6751,
      "train_tokens_per_second": 5964.078
    },
    {
      "epoch": 1.385471264367816,
      "grad_norm": 0.9818433523178101,
      "learning_rate": 6.154978833057243e-05,
      "loss": 1.5408,
      "num_input_tokens_seen": 22797616,
      "step": 3767,
      "train_runtime": 3822.5145,
      "train_tokens_per_second": 5964.036
    },
    {
      "epoch": 1.3858390804597702,
      "grad_norm": 0.9573441743850708,
      "learning_rate": 6.151297625621204e-05,
      "loss": 1.5261,
      "num_input_tokens_seen": 22802896,
      "step": 3768,
      "train_runtime": 3823.4131,
      "train_tokens_per_second": 5964.016
    },
    {
      "epoch": 1.386206896551724,
      "grad_norm": 1.0647966861724854,
      "learning_rate": 6.147616418185165e-05,
      "loss": 1.5025,
      "num_input_tokens_seen": 22809168,
      "step": 3769,
      "train_runtime": 3824.4262,
      "train_tokens_per_second": 5964.076
    },
    {
      "epoch": 1.3865747126436783,
      "grad_norm": 0.9207072854042053,
      "learning_rate": 6.143935210749125e-05,
      "loss": 1.3226,
      "num_input_tokens_seen": 22816768,
      "step": 3770,
      "train_runtime": 3825.6089,
      "train_tokens_per_second": 5964.219
    },
    {
      "epoch": 1.386942528735632,
      "grad_norm": 0.9285719990730286,
      "learning_rate": 6.140254003313088e-05,
      "loss": 1.5345,
      "num_input_tokens_seen": 22821360,
      "step": 3771,
      "train_runtime": 3826.4205,
      "train_tokens_per_second": 5964.154
    },
    {
      "epoch": 1.3873103448275863,
      "grad_norm": 0.9359105229377747,
      "learning_rate": 6.136572795877049e-05,
      "loss": 1.5052,
      "num_input_tokens_seen": 22826032,
      "step": 3772,
      "train_runtime": 3827.2038,
      "train_tokens_per_second": 5964.154
    },
    {
      "epoch": 1.38767816091954,
      "grad_norm": 0.9463181495666504,
      "learning_rate": 6.132891588441008e-05,
      "loss": 1.5824,
      "num_input_tokens_seen": 22831456,
      "step": 3773,
      "train_runtime": 3828.1096,
      "train_tokens_per_second": 5964.159
    },
    {
      "epoch": 1.3880459770114943,
      "grad_norm": 1.0038729906082153,
      "learning_rate": 6.129210381004969e-05,
      "loss": 1.7986,
      "num_input_tokens_seen": 22837248,
      "step": 3774,
      "train_runtime": 3829.0618,
      "train_tokens_per_second": 5964.189
    },
    {
      "epoch": 1.3884137931034481,
      "grad_norm": 1.0014309883117676,
      "learning_rate": 6.125529173568931e-05,
      "loss": 1.3315,
      "num_input_tokens_seen": 22843824,
      "step": 3775,
      "train_runtime": 3830.1297,
      "train_tokens_per_second": 5964.243
    },
    {
      "epoch": 1.3887816091954024,
      "grad_norm": 1.010796308517456,
      "learning_rate": 6.121847966132892e-05,
      "loss": 1.6359,
      "num_input_tokens_seen": 22848416,
      "step": 3776,
      "train_runtime": 3830.9322,
      "train_tokens_per_second": 5964.192
    },
    {
      "epoch": 1.3891494252873562,
      "grad_norm": 0.8719692826271057,
      "learning_rate": 6.118166758696853e-05,
      "loss": 1.2227,
      "num_input_tokens_seen": 22855392,
      "step": 3777,
      "train_runtime": 3832.0296,
      "train_tokens_per_second": 5964.305
    },
    {
      "epoch": 1.3895172413793104,
      "grad_norm": 0.9220796823501587,
      "learning_rate": 6.114485551260814e-05,
      "loss": 1.6286,
      "num_input_tokens_seen": 22860992,
      "step": 3778,
      "train_runtime": 3832.9503,
      "train_tokens_per_second": 5964.333
    },
    {
      "epoch": 1.3898850574712642,
      "grad_norm": 0.8386502861976624,
      "learning_rate": 6.110804343824775e-05,
      "loss": 1.5701,
      "num_input_tokens_seen": 22868720,
      "step": 3779,
      "train_runtime": 3834.156,
      "train_tokens_per_second": 5964.473
    },
    {
      "epoch": 1.3902528735632185,
      "grad_norm": 1.0410337448120117,
      "learning_rate": 6.107123136388736e-05,
      "loss": 1.4751,
      "num_input_tokens_seen": 22874128,
      "step": 3780,
      "train_runtime": 3835.0712,
      "train_tokens_per_second": 5964.46
    },
    {
      "epoch": 1.3906206896551725,
      "grad_norm": 0.9953624606132507,
      "learning_rate": 6.1034419289526966e-05,
      "loss": 1.531,
      "num_input_tokens_seen": 22879664,
      "step": 3781,
      "train_runtime": 3836.4798,
      "train_tokens_per_second": 5963.713
    },
    {
      "epoch": 1.3909885057471265,
      "grad_norm": 0.9081664085388184,
      "learning_rate": 6.0997607215166575e-05,
      "loss": 1.5514,
      "num_input_tokens_seen": 22886384,
      "step": 3782,
      "train_runtime": 3837.5811,
      "train_tokens_per_second": 5963.753
    },
    {
      "epoch": 1.3913563218390805,
      "grad_norm": 0.9312737584114075,
      "learning_rate": 6.096079514080619e-05,
      "loss": 1.5961,
      "num_input_tokens_seen": 22891680,
      "step": 3783,
      "train_runtime": 3838.4725,
      "train_tokens_per_second": 5963.747
    },
    {
      "epoch": 1.3917241379310346,
      "grad_norm": 0.9960929751396179,
      "learning_rate": 6.09239830664458e-05,
      "loss": 1.4714,
      "num_input_tokens_seen": 22896640,
      "step": 3784,
      "train_runtime": 3839.3212,
      "train_tokens_per_second": 5963.721
    },
    {
      "epoch": 1.3920919540229886,
      "grad_norm": 0.9982829093933105,
      "learning_rate": 6.088717099208541e-05,
      "loss": 1.5787,
      "num_input_tokens_seen": 22902176,
      "step": 3785,
      "train_runtime": 3840.227,
      "train_tokens_per_second": 5963.756
    },
    {
      "epoch": 1.3924597701149426,
      "grad_norm": 0.9263778924942017,
      "learning_rate": 6.085035891772502e-05,
      "loss": 1.271,
      "num_input_tokens_seen": 22909008,
      "step": 3786,
      "train_runtime": 3841.3466,
      "train_tokens_per_second": 5963.796
    },
    {
      "epoch": 1.3928275862068966,
      "grad_norm": 1.0715011358261108,
      "learning_rate": 6.081354684336462e-05,
      "loss": 1.7286,
      "num_input_tokens_seen": 22914048,
      "step": 3787,
      "train_runtime": 3842.2085,
      "train_tokens_per_second": 5963.77
    },
    {
      "epoch": 1.3931954022988506,
      "grad_norm": 0.9937467575073242,
      "learning_rate": 6.077673476900424e-05,
      "loss": 1.5261,
      "num_input_tokens_seen": 22918096,
      "step": 3788,
      "train_runtime": 3842.9174,
      "train_tokens_per_second": 5963.723
    },
    {
      "epoch": 1.3935632183908047,
      "grad_norm": 0.957568347454071,
      "learning_rate": 6.0739922694643844e-05,
      "loss": 1.3541,
      "num_input_tokens_seen": 22925840,
      "step": 3789,
      "train_runtime": 3844.1151,
      "train_tokens_per_second": 5963.88
    },
    {
      "epoch": 1.3939310344827587,
      "grad_norm": 1.0398995876312256,
      "learning_rate": 6.070311062028345e-05,
      "loss": 1.4543,
      "num_input_tokens_seen": 22930800,
      "step": 3790,
      "train_runtime": 3844.9689,
      "train_tokens_per_second": 5963.845
    },
    {
      "epoch": 1.3942988505747127,
      "grad_norm": 0.8865105509757996,
      "learning_rate": 6.066629854592306e-05,
      "loss": 1.4975,
      "num_input_tokens_seen": 22937376,
      "step": 3791,
      "train_runtime": 3846.0368,
      "train_tokens_per_second": 5963.899
    },
    {
      "epoch": 1.3946666666666667,
      "grad_norm": 0.9229094386100769,
      "learning_rate": 6.062948647156268e-05,
      "loss": 1.395,
      "num_input_tokens_seen": 22946304,
      "step": 3792,
      "train_runtime": 3847.4191,
      "train_tokens_per_second": 5964.077
    },
    {
      "epoch": 1.3950344827586207,
      "grad_norm": 0.9891628623008728,
      "learning_rate": 6.0592674397202286e-05,
      "loss": 1.4104,
      "num_input_tokens_seen": 22951600,
      "step": 3793,
      "train_runtime": 3848.2947,
      "train_tokens_per_second": 5964.096
    },
    {
      "epoch": 1.3954022988505748,
      "grad_norm": 0.8762285709381104,
      "learning_rate": 6.0555862322841895e-05,
      "loss": 1.5228,
      "num_input_tokens_seen": 22957424,
      "step": 3794,
      "train_runtime": 3849.2556,
      "train_tokens_per_second": 5964.12
    },
    {
      "epoch": 1.3957701149425288,
      "grad_norm": 0.8926065564155579,
      "learning_rate": 6.0519050248481504e-05,
      "loss": 1.3875,
      "num_input_tokens_seen": 22965360,
      "step": 3795,
      "train_runtime": 3850.4955,
      "train_tokens_per_second": 5964.261
    },
    {
      "epoch": 1.3961379310344828,
      "grad_norm": 1.046146035194397,
      "learning_rate": 6.048223817412112e-05,
      "loss": 1.6837,
      "num_input_tokens_seen": 22971984,
      "step": 3796,
      "train_runtime": 3851.5765,
      "train_tokens_per_second": 5964.307
    },
    {
      "epoch": 1.3965057471264368,
      "grad_norm": 0.9228570461273193,
      "learning_rate": 6.044542609976073e-05,
      "loss": 1.543,
      "num_input_tokens_seen": 22978080,
      "step": 3797,
      "train_runtime": 3852.5518,
      "train_tokens_per_second": 5964.379
    },
    {
      "epoch": 1.3968735632183908,
      "grad_norm": 0.9805880784988403,
      "learning_rate": 6.040861402540034e-05,
      "loss": 1.6692,
      "num_input_tokens_seen": 22983472,
      "step": 3798,
      "train_runtime": 3853.4308,
      "train_tokens_per_second": 5964.418
    },
    {
      "epoch": 1.3972413793103449,
      "grad_norm": 1.104706048965454,
      "learning_rate": 6.037180195103994e-05,
      "loss": 1.6094,
      "num_input_tokens_seen": 22990816,
      "step": 3799,
      "train_runtime": 3854.6019,
      "train_tokens_per_second": 5964.511
    },
    {
      "epoch": 1.3976091954022989,
      "grad_norm": 0.9283113479614258,
      "learning_rate": 6.033498987667955e-05,
      "loss": 1.3129,
      "num_input_tokens_seen": 22996432,
      "step": 3800,
      "train_runtime": 3855.5216,
      "train_tokens_per_second": 5964.545
    },
    {
      "epoch": 1.397977011494253,
      "grad_norm": 0.8897649049758911,
      "learning_rate": 6.0298177802319164e-05,
      "loss": 1.225,
      "num_input_tokens_seen": 23007296,
      "step": 3801,
      "train_runtime": 3857.1737,
      "train_tokens_per_second": 5964.807
    },
    {
      "epoch": 1.398344827586207,
      "grad_norm": 0.9260315299034119,
      "learning_rate": 6.026136572795877e-05,
      "loss": 1.3904,
      "num_input_tokens_seen": 23013584,
      "step": 3802,
      "train_runtime": 3858.2205,
      "train_tokens_per_second": 5964.818
    },
    {
      "epoch": 1.398712643678161,
      "grad_norm": 0.9527813196182251,
      "learning_rate": 6.022455365359838e-05,
      "loss": 1.552,
      "num_input_tokens_seen": 23017776,
      "step": 3803,
      "train_runtime": 3858.9784,
      "train_tokens_per_second": 5964.733
    },
    {
      "epoch": 1.399080459770115,
      "grad_norm": 0.9974001049995422,
      "learning_rate": 6.018774157923799e-05,
      "loss": 1.5802,
      "num_input_tokens_seen": 23023056,
      "step": 3804,
      "train_runtime": 3859.8617,
      "train_tokens_per_second": 5964.736
    },
    {
      "epoch": 1.399448275862069,
      "grad_norm": 0.9293568134307861,
      "learning_rate": 6.0150929504877606e-05,
      "loss": 1.4446,
      "num_input_tokens_seen": 23028912,
      "step": 3805,
      "train_runtime": 3860.8209,
      "train_tokens_per_second": 5964.771
    },
    {
      "epoch": 1.399816091954023,
      "grad_norm": 0.9839887022972107,
      "learning_rate": 6.0114117430517215e-05,
      "loss": 1.525,
      "num_input_tokens_seen": 23033072,
      "step": 3806,
      "train_runtime": 3861.5858,
      "train_tokens_per_second": 5964.667
    },
    {
      "epoch": 1.400183908045977,
      "grad_norm": 0.9744802117347717,
      "learning_rate": 6.0077305356156824e-05,
      "loss": 1.5001,
      "num_input_tokens_seen": 23038432,
      "step": 3807,
      "train_runtime": 3862.4497,
      "train_tokens_per_second": 5964.72
    },
    {
      "epoch": 1.400551724137931,
      "grad_norm": 0.9339486360549927,
      "learning_rate": 6.0040493281796426e-05,
      "loss": 1.5087,
      "num_input_tokens_seen": 23043248,
      "step": 3808,
      "train_runtime": 3863.2824,
      "train_tokens_per_second": 5964.681
    },
    {
      "epoch": 1.400551724137931,
      "eval_loss": 1.762308955192566,
      "eval_runtime": 4.783,
      "eval_samples_per_second": 209.072,
      "eval_steps_per_second": 13.172,
      "num_input_tokens_seen": 23043248,
      "step": 3808
    },
    {
      "epoch": 1.400919540229885,
      "grad_norm": 0.9982841610908508,
      "learning_rate": 6.000368120743605e-05,
      "loss": 1.5034,
      "num_input_tokens_seen": 23053232,
      "step": 3809,
      "train_runtime": 3869.574,
      "train_tokens_per_second": 5957.563
    },
    {
      "epoch": 1.401287356321839,
      "grad_norm": 1.0063250064849854,
      "learning_rate": 5.996686913307565e-05,
      "loss": 1.6184,
      "num_input_tokens_seen": 23057872,
      "step": 3810,
      "train_runtime": 3870.3639,
      "train_tokens_per_second": 5957.546
    },
    {
      "epoch": 1.4016551724137931,
      "grad_norm": 0.9848638772964478,
      "learning_rate": 5.993005705871526e-05,
      "loss": 1.4884,
      "num_input_tokens_seen": 23062704,
      "step": 3811,
      "train_runtime": 3871.6846,
      "train_tokens_per_second": 5956.762
    },
    {
      "epoch": 1.4020229885057471,
      "grad_norm": 0.9566230773925781,
      "learning_rate": 5.989324498435487e-05,
      "loss": 1.665,
      "num_input_tokens_seen": 23067440,
      "step": 3812,
      "train_runtime": 3872.4944,
      "train_tokens_per_second": 5956.739
    },
    {
      "epoch": 1.4023908045977012,
      "grad_norm": 1.0285876989364624,
      "learning_rate": 5.9856432909994484e-05,
      "loss": 1.7322,
      "num_input_tokens_seen": 23072864,
      "step": 3813,
      "train_runtime": 3873.3823,
      "train_tokens_per_second": 5956.774
    },
    {
      "epoch": 1.4027586206896552,
      "grad_norm": 1.0430469512939453,
      "learning_rate": 5.981962083563409e-05,
      "loss": 1.7483,
      "num_input_tokens_seen": 23077632,
      "step": 3814,
      "train_runtime": 3874.1922,
      "train_tokens_per_second": 5956.76
    },
    {
      "epoch": 1.4031264367816092,
      "grad_norm": 1.0014046430587769,
      "learning_rate": 5.97828087612737e-05,
      "loss": 1.6424,
      "num_input_tokens_seen": 23082896,
      "step": 3815,
      "train_runtime": 3875.0743,
      "train_tokens_per_second": 5956.762
    },
    {
      "epoch": 1.4034942528735632,
      "grad_norm": 1.055223822593689,
      "learning_rate": 5.974599668691331e-05,
      "loss": 1.6634,
      "num_input_tokens_seen": 23088544,
      "step": 3816,
      "train_runtime": 3876.0102,
      "train_tokens_per_second": 5956.781
    },
    {
      "epoch": 1.4038620689655172,
      "grad_norm": 1.0828168392181396,
      "learning_rate": 5.970918461255291e-05,
      "loss": 1.3683,
      "num_input_tokens_seen": 23094048,
      "step": 3817,
      "train_runtime": 3876.9499,
      "train_tokens_per_second": 5956.757
    },
    {
      "epoch": 1.4042298850574713,
      "grad_norm": 1.054440975189209,
      "learning_rate": 5.9672372538192536e-05,
      "loss": 1.4349,
      "num_input_tokens_seen": 23099184,
      "step": 3818,
      "train_runtime": 3877.8099,
      "train_tokens_per_second": 5956.76
    },
    {
      "epoch": 1.4045977011494253,
      "grad_norm": 1.0225321054458618,
      "learning_rate": 5.963556046383214e-05,
      "loss": 1.5644,
      "num_input_tokens_seen": 23103584,
      "step": 3819,
      "train_runtime": 3878.583,
      "train_tokens_per_second": 5956.707
    },
    {
      "epoch": 1.4049655172413793,
      "grad_norm": 1.0918307304382324,
      "learning_rate": 5.9598748389471747e-05,
      "loss": 1.5909,
      "num_input_tokens_seen": 23108672,
      "step": 3820,
      "train_runtime": 3879.4368,
      "train_tokens_per_second": 5956.708
    },
    {
      "epoch": 1.4053333333333333,
      "grad_norm": 0.8967226147651672,
      "learning_rate": 5.9561936315111355e-05,
      "loss": 1.3601,
      "num_input_tokens_seen": 23118160,
      "step": 3821,
      "train_runtime": 3880.9012,
      "train_tokens_per_second": 5956.905
    },
    {
      "epoch": 1.4057011494252873,
      "grad_norm": 0.9362155795097351,
      "learning_rate": 5.952512424075097e-05,
      "loss": 1.4752,
      "num_input_tokens_seen": 23127648,
      "step": 3822,
      "train_runtime": 3882.3632,
      "train_tokens_per_second": 5957.106
    },
    {
      "epoch": 1.4060689655172414,
      "grad_norm": 1.0105961561203003,
      "learning_rate": 5.948831216639058e-05,
      "loss": 1.4078,
      "num_input_tokens_seen": 23132624,
      "step": 3823,
      "train_runtime": 3883.208,
      "train_tokens_per_second": 5957.091
    },
    {
      "epoch": 1.4064367816091954,
      "grad_norm": 0.9605310559272766,
      "learning_rate": 5.945150009203019e-05,
      "loss": 1.5139,
      "num_input_tokens_seen": 23137968,
      "step": 3824,
      "train_runtime": 3884.1071,
      "train_tokens_per_second": 5957.088
    },
    {
      "epoch": 1.4068045977011494,
      "grad_norm": 0.9292888641357422,
      "learning_rate": 5.94146880176698e-05,
      "loss": 1.7353,
      "num_input_tokens_seen": 23142672,
      "step": 3825,
      "train_runtime": 3884.926,
      "train_tokens_per_second": 5957.043
    },
    {
      "epoch": 1.4071724137931034,
      "grad_norm": 1.0281490087509155,
      "learning_rate": 5.9377875943309413e-05,
      "loss": 1.6638,
      "num_input_tokens_seen": 23147952,
      "step": 3826,
      "train_runtime": 3885.8256,
      "train_tokens_per_second": 5957.023
    },
    {
      "epoch": 1.4075402298850574,
      "grad_norm": 1.0020662546157837,
      "learning_rate": 5.934106386894902e-05,
      "loss": 1.7889,
      "num_input_tokens_seen": 23152432,
      "step": 3827,
      "train_runtime": 3886.5937,
      "train_tokens_per_second": 5956.998
    },
    {
      "epoch": 1.4079080459770115,
      "grad_norm": 0.9764525890350342,
      "learning_rate": 5.9304251794588624e-05,
      "loss": 1.3644,
      "num_input_tokens_seen": 23160048,
      "step": 3828,
      "train_runtime": 3887.8057,
      "train_tokens_per_second": 5957.1
    },
    {
      "epoch": 1.4082758620689655,
      "grad_norm": 0.9640796184539795,
      "learning_rate": 5.926743972022823e-05,
      "loss": 1.6267,
      "num_input_tokens_seen": 23168208,
      "step": 3829,
      "train_runtime": 3889.0765,
      "train_tokens_per_second": 5957.252
    },
    {
      "epoch": 1.4086436781609195,
      "grad_norm": 1.030874490737915,
      "learning_rate": 5.923062764586784e-05,
      "loss": 1.6074,
      "num_input_tokens_seen": 23174080,
      "step": 3830,
      "train_runtime": 3890.0517,
      "train_tokens_per_second": 5957.268
    },
    {
      "epoch": 1.4090114942528735,
      "grad_norm": 0.8311935067176819,
      "learning_rate": 5.919381557150746e-05,
      "loss": 1.4108,
      "num_input_tokens_seen": 23184736,
      "step": 3831,
      "train_runtime": 3891.6405,
      "train_tokens_per_second": 5957.574
    },
    {
      "epoch": 1.4093793103448276,
      "grad_norm": 0.9829455018043518,
      "learning_rate": 5.915700349714707e-05,
      "loss": 1.6012,
      "num_input_tokens_seen": 23190080,
      "step": 3832,
      "train_runtime": 3892.5102,
      "train_tokens_per_second": 5957.616
    },
    {
      "epoch": 1.4097471264367816,
      "grad_norm": 0.9618329405784607,
      "learning_rate": 5.9120191422786676e-05,
      "loss": 1.5514,
      "num_input_tokens_seen": 23197360,
      "step": 3833,
      "train_runtime": 3893.6569,
      "train_tokens_per_second": 5957.731
    },
    {
      "epoch": 1.4101149425287356,
      "grad_norm": 0.9474486112594604,
      "learning_rate": 5.9083379348426285e-05,
      "loss": 1.5553,
      "num_input_tokens_seen": 23202336,
      "step": 3834,
      "train_runtime": 3894.5055,
      "train_tokens_per_second": 5957.71
    },
    {
      "epoch": 1.4104827586206896,
      "grad_norm": 0.9888517260551453,
      "learning_rate": 5.90465672740659e-05,
      "loss": 1.4854,
      "num_input_tokens_seen": 23207424,
      "step": 3835,
      "train_runtime": 3895.346,
      "train_tokens_per_second": 5957.731
    },
    {
      "epoch": 1.4108505747126436,
      "grad_norm": 0.9446732401847839,
      "learning_rate": 5.900975519970551e-05,
      "loss": 1.5645,
      "num_input_tokens_seen": 23213904,
      "step": 3836,
      "train_runtime": 3896.4324,
      "train_tokens_per_second": 5957.733
    },
    {
      "epoch": 1.4112183908045977,
      "grad_norm": 0.9339814186096191,
      "learning_rate": 5.897294312534512e-05,
      "loss": 1.4718,
      "num_input_tokens_seen": 23219184,
      "step": 3837,
      "train_runtime": 3897.3091,
      "train_tokens_per_second": 5957.748
    },
    {
      "epoch": 1.4115862068965517,
      "grad_norm": 1.0465707778930664,
      "learning_rate": 5.893613105098472e-05,
      "loss": 1.5653,
      "num_input_tokens_seen": 23224288,
      "step": 3838,
      "train_runtime": 3898.161,
      "train_tokens_per_second": 5957.755
    },
    {
      "epoch": 1.4119540229885057,
      "grad_norm": 0.9296630024909973,
      "learning_rate": 5.889931897662434e-05,
      "loss": 1.6597,
      "num_input_tokens_seen": 23228992,
      "step": 3839,
      "train_runtime": 3898.99,
      "train_tokens_per_second": 5957.695
    },
    {
      "epoch": 1.4123218390804597,
      "grad_norm": 0.9607882499694824,
      "learning_rate": 5.8862506902263945e-05,
      "loss": 1.5334,
      "num_input_tokens_seen": 23236160,
      "step": 3840,
      "train_runtime": 3900.1417,
      "train_tokens_per_second": 5957.773
    },
    {
      "epoch": 1.4126896551724137,
      "grad_norm": 0.9619476795196533,
      "learning_rate": 5.8825694827903554e-05,
      "loss": 1.4458,
      "num_input_tokens_seen": 23241232,
      "step": 3841,
      "train_runtime": 3901.4996,
      "train_tokens_per_second": 5957.0
    },
    {
      "epoch": 1.4130574712643678,
      "grad_norm": 0.9586007595062256,
      "learning_rate": 5.878888275354316e-05,
      "loss": 1.5467,
      "num_input_tokens_seen": 23247328,
      "step": 3842,
      "train_runtime": 3902.4774,
      "train_tokens_per_second": 5957.069
    },
    {
      "epoch": 1.4134252873563218,
      "grad_norm": 0.9935484528541565,
      "learning_rate": 5.875207067918278e-05,
      "loss": 1.5199,
      "num_input_tokens_seen": 23253312,
      "step": 3843,
      "train_runtime": 3903.4533,
      "train_tokens_per_second": 5957.113
    },
    {
      "epoch": 1.4137931034482758,
      "grad_norm": 0.9330466389656067,
      "learning_rate": 5.871525860482239e-05,
      "loss": 1.455,
      "num_input_tokens_seen": 23260080,
      "step": 3844,
      "train_runtime": 3904.5295,
      "train_tokens_per_second": 5957.204
    },
    {
      "epoch": 1.4141609195402298,
      "grad_norm": 0.9788523316383362,
      "learning_rate": 5.8678446530461996e-05,
      "loss": 1.5322,
      "num_input_tokens_seen": 23264768,
      "step": 3845,
      "train_runtime": 3905.3446,
      "train_tokens_per_second": 5957.161
    },
    {
      "epoch": 1.4145287356321838,
      "grad_norm": 0.9855858683586121,
      "learning_rate": 5.8641634456101605e-05,
      "loss": 1.6398,
      "num_input_tokens_seen": 23270192,
      "step": 3846,
      "train_runtime": 3906.2588,
      "train_tokens_per_second": 5957.156
    },
    {
      "epoch": 1.4148965517241379,
      "grad_norm": 0.9445512294769287,
      "learning_rate": 5.860482238174121e-05,
      "loss": 1.4612,
      "num_input_tokens_seen": 23274672,
      "step": 3847,
      "train_runtime": 3907.0612,
      "train_tokens_per_second": 5957.079
    },
    {
      "epoch": 1.4152643678160919,
      "grad_norm": 0.9998615980148315,
      "learning_rate": 5.856801030738083e-05,
      "loss": 1.4926,
      "num_input_tokens_seen": 23280144,
      "step": 3848,
      "train_runtime": 3907.9501,
      "train_tokens_per_second": 5957.124
    },
    {
      "epoch": 1.415632183908046,
      "grad_norm": 1.034893274307251,
      "learning_rate": 5.853119823302043e-05,
      "loss": 1.4794,
      "num_input_tokens_seen": 23287024,
      "step": 3849,
      "train_runtime": 3909.03,
      "train_tokens_per_second": 5957.238
    },
    {
      "epoch": 1.416,
      "grad_norm": 0.9571827054023743,
      "learning_rate": 5.849438615866004e-05,
      "loss": 1.662,
      "num_input_tokens_seen": 23293760,
      "step": 3850,
      "train_runtime": 3910.1194,
      "train_tokens_per_second": 5957.301
    },
    {
      "epoch": 1.416367816091954,
      "grad_norm": 0.951391875743866,
      "learning_rate": 5.845757408429965e-05,
      "loss": 1.4558,
      "num_input_tokens_seen": 23299232,
      "step": 3851,
      "train_runtime": 3911.0287,
      "train_tokens_per_second": 5957.316
    },
    {
      "epoch": 1.416735632183908,
      "grad_norm": 1.0006654262542725,
      "learning_rate": 5.8420762009939265e-05,
      "loss": 1.4558,
      "num_input_tokens_seen": 23303216,
      "step": 3852,
      "train_runtime": 3911.7532,
      "train_tokens_per_second": 5957.231
    },
    {
      "epoch": 1.417103448275862,
      "grad_norm": 1.0957709550857544,
      "learning_rate": 5.8383949935578874e-05,
      "loss": 1.6442,
      "num_input_tokens_seen": 23307824,
      "step": 3853,
      "train_runtime": 3912.5467,
      "train_tokens_per_second": 5957.2
    },
    {
      "epoch": 1.417471264367816,
      "grad_norm": 0.957355797290802,
      "learning_rate": 5.834713786121848e-05,
      "loss": 1.5419,
      "num_input_tokens_seen": 23312144,
      "step": 3854,
      "train_runtime": 3913.3081,
      "train_tokens_per_second": 5957.145
    },
    {
      "epoch": 1.4178390804597703,
      "grad_norm": 0.8509191274642944,
      "learning_rate": 5.831032578685809e-05,
      "loss": 1.4699,
      "num_input_tokens_seen": 23318800,
      "step": 3855,
      "train_runtime": 3914.351,
      "train_tokens_per_second": 5957.258
    },
    {
      "epoch": 1.418206896551724,
      "grad_norm": 1.0836807489395142,
      "learning_rate": 5.827351371249771e-05,
      "loss": 1.5796,
      "num_input_tokens_seen": 23323488,
      "step": 3856,
      "train_runtime": 3915.1682,
      "train_tokens_per_second": 5957.212
    },
    {
      "epoch": 1.4185747126436783,
      "grad_norm": 1.02164626121521,
      "learning_rate": 5.8236701638137316e-05,
      "loss": 1.7552,
      "num_input_tokens_seen": 23328480,
      "step": 3857,
      "train_runtime": 3916.0216,
      "train_tokens_per_second": 5957.189
    },
    {
      "epoch": 1.418942528735632,
      "grad_norm": 1.0840938091278076,
      "learning_rate": 5.819988956377692e-05,
      "loss": 1.9078,
      "num_input_tokens_seen": 23334000,
      "step": 3858,
      "train_runtime": 3916.9267,
      "train_tokens_per_second": 5957.222
    },
    {
      "epoch": 1.4193103448275863,
      "grad_norm": 0.9423296451568604,
      "learning_rate": 5.816307748941653e-05,
      "loss": 1.441,
      "num_input_tokens_seen": 23338752,
      "step": 3859,
      "train_runtime": 3917.7444,
      "train_tokens_per_second": 5957.191
    },
    {
      "epoch": 1.4196781609195401,
      "grad_norm": 0.9281581044197083,
      "learning_rate": 5.8126265415056136e-05,
      "loss": 1.3866,
      "num_input_tokens_seen": 23344160,
      "step": 3860,
      "train_runtime": 3918.6296,
      "train_tokens_per_second": 5957.226
    },
    {
      "epoch": 1.4200459770114944,
      "grad_norm": 1.0541623830795288,
      "learning_rate": 5.808945334069575e-05,
      "loss": 1.5146,
      "num_input_tokens_seen": 23349264,
      "step": 3861,
      "train_runtime": 3919.4916,
      "train_tokens_per_second": 5957.218
    },
    {
      "epoch": 1.4204137931034482,
      "grad_norm": 0.9313670992851257,
      "learning_rate": 5.805264126633536e-05,
      "loss": 1.4677,
      "num_input_tokens_seen": 23354464,
      "step": 3862,
      "train_runtime": 3920.3386,
      "train_tokens_per_second": 5957.257
    },
    {
      "epoch": 1.4207816091954024,
      "grad_norm": 1.0147207975387573,
      "learning_rate": 5.801582919197497e-05,
      "loss": 1.6991,
      "num_input_tokens_seen": 23359408,
      "step": 3863,
      "train_runtime": 3921.1478,
      "train_tokens_per_second": 5957.288
    },
    {
      "epoch": 1.4211494252873562,
      "grad_norm": 0.9353586435317993,
      "learning_rate": 5.797901711761458e-05,
      "loss": 1.685,
      "num_input_tokens_seen": 23364432,
      "step": 3864,
      "train_runtime": 3921.9993,
      "train_tokens_per_second": 5957.276
    },
    {
      "epoch": 1.4215172413793105,
      "grad_norm": 1.0175602436065674,
      "learning_rate": 5.7942205043254194e-05,
      "loss": 1.4763,
      "num_input_tokens_seen": 23368880,
      "step": 3865,
      "train_runtime": 3922.8016,
      "train_tokens_per_second": 5957.191
    },
    {
      "epoch": 1.4218850574712643,
      "grad_norm": 0.7078250050544739,
      "learning_rate": 5.79053929688938e-05,
      "loss": 1.3493,
      "num_input_tokens_seen": 23381696,
      "step": 3866,
      "train_runtime": 3924.6875,
      "train_tokens_per_second": 5957.594
    },
    {
      "epoch": 1.4222528735632185,
      "grad_norm": 0.9604660868644714,
      "learning_rate": 5.7868580894533405e-05,
      "loss": 1.5835,
      "num_input_tokens_seen": 23388480,
      "step": 3867,
      "train_runtime": 3925.7763,
      "train_tokens_per_second": 5957.67
    },
    {
      "epoch": 1.4226206896551723,
      "grad_norm": 1.0410925149917603,
      "learning_rate": 5.7831768820173014e-05,
      "loss": 1.4953,
      "num_input_tokens_seen": 23394224,
      "step": 3868,
      "train_runtime": 3926.7134,
      "train_tokens_per_second": 5957.711
    },
    {
      "epoch": 1.4229885057471265,
      "grad_norm": 1.0154757499694824,
      "learning_rate": 5.7794956745812637e-05,
      "loss": 1.4721,
      "num_input_tokens_seen": 23399056,
      "step": 3869,
      "train_runtime": 3927.538,
      "train_tokens_per_second": 5957.69
    },
    {
      "epoch": 1.4233563218390803,
      "grad_norm": 1.0183285474777222,
      "learning_rate": 5.775814467145224e-05,
      "loss": 1.6538,
      "num_input_tokens_seen": 23403488,
      "step": 3870,
      "train_runtime": 3928.3073,
      "train_tokens_per_second": 5957.652
    },
    {
      "epoch": 1.4237241379310346,
      "grad_norm": 0.8870113492012024,
      "learning_rate": 5.772133259709185e-05,
      "loss": 1.3219,
      "num_input_tokens_seen": 23412752,
      "step": 3871,
      "train_runtime": 3930.1993,
      "train_tokens_per_second": 5957.141
    },
    {
      "epoch": 1.4240919540229884,
      "grad_norm": 0.9855868220329285,
      "learning_rate": 5.7684520522731456e-05,
      "loss": 1.4956,
      "num_input_tokens_seen": 23417376,
      "step": 3872,
      "train_runtime": 3931.0125,
      "train_tokens_per_second": 5957.085
    },
    {
      "epoch": 1.4244597701149426,
      "grad_norm": 1.0047303438186646,
      "learning_rate": 5.764770844837107e-05,
      "loss": 1.4862,
      "num_input_tokens_seen": 23424688,
      "step": 3873,
      "train_runtime": 3932.1756,
      "train_tokens_per_second": 5957.183
    },
    {
      "epoch": 1.4248275862068964,
      "grad_norm": 0.9415448307991028,
      "learning_rate": 5.761089637401068e-05,
      "loss": 1.5932,
      "num_input_tokens_seen": 23429312,
      "step": 3874,
      "train_runtime": 3932.9722,
      "train_tokens_per_second": 5957.152
    },
    {
      "epoch": 1.4251954022988507,
      "grad_norm": 0.948373556137085,
      "learning_rate": 5.757408429965029e-05,
      "loss": 1.565,
      "num_input_tokens_seen": 23436512,
      "step": 3875,
      "train_runtime": 3934.1334,
      "train_tokens_per_second": 5957.223
    },
    {
      "epoch": 1.4255632183908045,
      "grad_norm": 0.9539574980735779,
      "learning_rate": 5.75372722252899e-05,
      "loss": 1.6299,
      "num_input_tokens_seen": 23441472,
      "step": 3876,
      "train_runtime": 3934.963,
      "train_tokens_per_second": 5957.228
    },
    {
      "epoch": 1.4259310344827587,
      "grad_norm": 1.014194130897522,
      "learning_rate": 5.75004601509295e-05,
      "loss": 1.574,
      "num_input_tokens_seen": 23447888,
      "step": 3877,
      "train_runtime": 3935.9843,
      "train_tokens_per_second": 5957.312
    },
    {
      "epoch": 1.4262988505747125,
      "grad_norm": 0.9765027165412903,
      "learning_rate": 5.746364807656912e-05,
      "loss": 1.5121,
      "num_input_tokens_seen": 23454160,
      "step": 3878,
      "train_runtime": 3937.0199,
      "train_tokens_per_second": 5957.338
    },
    {
      "epoch": 1.4266666666666667,
      "grad_norm": 0.9368051290512085,
      "learning_rate": 5.7426836002208725e-05,
      "loss": 1.4899,
      "num_input_tokens_seen": 23459936,
      "step": 3879,
      "train_runtime": 3937.9919,
      "train_tokens_per_second": 5957.335
    },
    {
      "epoch": 1.4270344827586208,
      "grad_norm": 0.9212564826011658,
      "learning_rate": 5.7390023927848334e-05,
      "loss": 1.3824,
      "num_input_tokens_seen": 23466496,
      "step": 3880,
      "train_runtime": 3939.064,
      "train_tokens_per_second": 5957.379
    },
    {
      "epoch": 1.4274022988505748,
      "grad_norm": 1.0353937149047852,
      "learning_rate": 5.735321185348794e-05,
      "loss": 1.6984,
      "num_input_tokens_seen": 23471824,
      "step": 3881,
      "train_runtime": 3939.9433,
      "train_tokens_per_second": 5957.401
    },
    {
      "epoch": 1.4277701149425288,
      "grad_norm": 0.9657158851623535,
      "learning_rate": 5.731639977912756e-05,
      "loss": 1.3684,
      "num_input_tokens_seen": 23476624,
      "step": 3882,
      "train_runtime": 3940.7634,
      "train_tokens_per_second": 5957.38
    },
    {
      "epoch": 1.4281379310344828,
      "grad_norm": 0.8945689797401428,
      "learning_rate": 5.727958770476717e-05,
      "loss": 1.4206,
      "num_input_tokens_seen": 23481728,
      "step": 3883,
      "train_runtime": 3941.5979,
      "train_tokens_per_second": 5957.413
    },
    {
      "epoch": 1.4285057471264369,
      "grad_norm": 0.9939314126968384,
      "learning_rate": 5.724277563040678e-05,
      "loss": 1.4417,
      "num_input_tokens_seen": 23490016,
      "step": 3884,
      "train_runtime": 3942.9102,
      "train_tokens_per_second": 5957.533
    },
    {
      "epoch": 1.4288735632183909,
      "grad_norm": 0.9665780663490295,
      "learning_rate": 5.7205963556046386e-05,
      "loss": 1.6332,
      "num_input_tokens_seen": 23495936,
      "step": 3885,
      "train_runtime": 3943.8607,
      "train_tokens_per_second": 5957.598
    },
    {
      "epoch": 1.429241379310345,
      "grad_norm": 0.913955569267273,
      "learning_rate": 5.7169151481686e-05,
      "loss": 1.5729,
      "num_input_tokens_seen": 23504432,
      "step": 3886,
      "train_runtime": 3945.1835,
      "train_tokens_per_second": 5957.754
    },
    {
      "epoch": 1.429609195402299,
      "grad_norm": 1.0110636949539185,
      "learning_rate": 5.713233940732561e-05,
      "loss": 1.7369,
      "num_input_tokens_seen": 23509408,
      "step": 3887,
      "train_runtime": 3946.0357,
      "train_tokens_per_second": 5957.728
    },
    {
      "epoch": 1.429977011494253,
      "grad_norm": 0.9137610793113708,
      "learning_rate": 5.709552733296521e-05,
      "loss": 1.5215,
      "num_input_tokens_seen": 23515664,
      "step": 3888,
      "train_runtime": 3947.0532,
      "train_tokens_per_second": 5957.777
    },
    {
      "epoch": 1.430344827586207,
      "grad_norm": 0.9536697864532471,
      "learning_rate": 5.705871525860482e-05,
      "loss": 1.4227,
      "num_input_tokens_seen": 23520624,
      "step": 3889,
      "train_runtime": 3947.8823,
      "train_tokens_per_second": 5957.783
    },
    {
      "epoch": 1.430712643678161,
      "grad_norm": 0.8703693151473999,
      "learning_rate": 5.702190318424443e-05,
      "loss": 1.4911,
      "num_input_tokens_seen": 23527328,
      "step": 3890,
      "train_runtime": 3948.9175,
      "train_tokens_per_second": 5957.918
    },
    {
      "epoch": 1.431080459770115,
      "grad_norm": 0.878696084022522,
      "learning_rate": 5.6985091109884046e-05,
      "loss": 1.5087,
      "num_input_tokens_seen": 23532896,
      "step": 3891,
      "train_runtime": 3949.8306,
      "train_tokens_per_second": 5957.951
    },
    {
      "epoch": 1.431448275862069,
      "grad_norm": 1.0287691354751587,
      "learning_rate": 5.6948279035523655e-05,
      "loss": 1.4682,
      "num_input_tokens_seen": 23542352,
      "step": 3892,
      "train_runtime": 3951.2423,
      "train_tokens_per_second": 5958.215
    },
    {
      "epoch": 1.431816091954023,
      "grad_norm": 0.9357215762138367,
      "learning_rate": 5.6911466961163263e-05,
      "loss": 1.628,
      "num_input_tokens_seen": 23547104,
      "step": 3893,
      "train_runtime": 3952.0493,
      "train_tokens_per_second": 5958.201
    },
    {
      "epoch": 1.432183908045977,
      "grad_norm": 0.9912281632423401,
      "learning_rate": 5.687465488680287e-05,
      "loss": 1.6233,
      "num_input_tokens_seen": 23553072,
      "step": 3894,
      "train_runtime": 3953.003,
      "train_tokens_per_second": 5958.273
    },
    {
      "epoch": 1.432551724137931,
      "grad_norm": 0.9902219772338867,
      "learning_rate": 5.683784281244249e-05,
      "loss": 1.587,
      "num_input_tokens_seen": 23557856,
      "step": 3895,
      "train_runtime": 3953.8029,
      "train_tokens_per_second": 5958.278
    },
    {
      "epoch": 1.432919540229885,
      "grad_norm": 1.0217221975326538,
      "learning_rate": 5.68010307380821e-05,
      "loss": 1.8398,
      "num_input_tokens_seen": 23562656,
      "step": 3896,
      "train_runtime": 3954.6292,
      "train_tokens_per_second": 5958.247
    },
    {
      "epoch": 1.4332873563218391,
      "grad_norm": 0.9312261343002319,
      "learning_rate": 5.67642186637217e-05,
      "loss": 1.7105,
      "num_input_tokens_seen": 23569664,
      "step": 3897,
      "train_runtime": 3955.761,
      "train_tokens_per_second": 5958.313
    },
    {
      "epoch": 1.4336551724137931,
      "grad_norm": 0.9893278479576111,
      "learning_rate": 5.672740658936131e-05,
      "loss": 1.5993,
      "num_input_tokens_seen": 23574384,
      "step": 3898,
      "train_runtime": 3956.5777,
      "train_tokens_per_second": 5958.277
    },
    {
      "epoch": 1.4340229885057472,
      "grad_norm": 0.9132021069526672,
      "learning_rate": 5.669059451500093e-05,
      "loss": 1.4003,
      "num_input_tokens_seen": 23579392,
      "step": 3899,
      "train_runtime": 3957.3917,
      "train_tokens_per_second": 5958.316
    },
    {
      "epoch": 1.4343908045977012,
      "grad_norm": 0.996238648891449,
      "learning_rate": 5.665378244064053e-05,
      "loss": 1.5099,
      "num_input_tokens_seen": 23584816,
      "step": 3900,
      "train_runtime": 3958.2583,
      "train_tokens_per_second": 5958.382
    },
    {
      "epoch": 1.4347586206896552,
      "grad_norm": 0.9297142624855042,
      "learning_rate": 5.661697036628014e-05,
      "loss": 1.4056,
      "num_input_tokens_seen": 23590544,
      "step": 3901,
      "train_runtime": 3959.6809,
      "train_tokens_per_second": 5957.688
    },
    {
      "epoch": 1.4351264367816092,
      "grad_norm": 0.9852582812309265,
      "learning_rate": 5.658015829191975e-05,
      "loss": 1.6708,
      "num_input_tokens_seen": 23596016,
      "step": 3902,
      "train_runtime": 3960.5906,
      "train_tokens_per_second": 5957.701
    },
    {
      "epoch": 1.4354942528735632,
      "grad_norm": 0.9678986668586731,
      "learning_rate": 5.6543346217559366e-05,
      "loss": 1.3661,
      "num_input_tokens_seen": 23602144,
      "step": 3903,
      "train_runtime": 3961.595,
      "train_tokens_per_second": 5957.738
    },
    {
      "epoch": 1.4358620689655173,
      "grad_norm": 0.9360491633415222,
      "learning_rate": 5.6506534143198975e-05,
      "loss": 1.6038,
      "num_input_tokens_seen": 23607216,
      "step": 3904,
      "train_runtime": 3962.4478,
      "train_tokens_per_second": 5957.736
    },
    {
      "epoch": 1.4362298850574713,
      "grad_norm": 0.921673059463501,
      "learning_rate": 5.6469722068838584e-05,
      "loss": 1.5799,
      "num_input_tokens_seen": 23614592,
      "step": 3905,
      "train_runtime": 3963.5808,
      "train_tokens_per_second": 5957.893
    },
    {
      "epoch": 1.4365977011494253,
      "grad_norm": 1.0383217334747314,
      "learning_rate": 5.6432909994478186e-05,
      "loss": 1.6317,
      "num_input_tokens_seen": 23621648,
      "step": 3906,
      "train_runtime": 3964.7094,
      "train_tokens_per_second": 5957.977
    },
    {
      "epoch": 1.4369655172413793,
      "grad_norm": 0.9556329250335693,
      "learning_rate": 5.6396097920117795e-05,
      "loss": 1.4928,
      "num_input_tokens_seen": 23626624,
      "step": 3907,
      "train_runtime": 3965.5508,
      "train_tokens_per_second": 5957.968
    },
    {
      "epoch": 1.4373333333333334,
      "grad_norm": 0.9423363208770752,
      "learning_rate": 5.635928584575742e-05,
      "loss": 1.4834,
      "num_input_tokens_seen": 23631792,
      "step": 3908,
      "train_runtime": 3966.422,
      "train_tokens_per_second": 5957.962
    },
    {
      "epoch": 1.4377011494252874,
      "grad_norm": 0.9611340165138245,
      "learning_rate": 5.632247377139702e-05,
      "loss": 1.5116,
      "num_input_tokens_seen": 23636320,
      "step": 3909,
      "train_runtime": 3967.2077,
      "train_tokens_per_second": 5957.923
    },
    {
      "epoch": 1.4380689655172414,
      "grad_norm": 0.9286181330680847,
      "learning_rate": 5.628566169703663e-05,
      "loss": 1.4953,
      "num_input_tokens_seen": 23642144,
      "step": 3910,
      "train_runtime": 3968.1666,
      "train_tokens_per_second": 5957.951
    },
    {
      "epoch": 1.4384367816091954,
      "grad_norm": 1.0016546249389648,
      "learning_rate": 5.624884962267624e-05,
      "loss": 1.4092,
      "num_input_tokens_seen": 23646368,
      "step": 3911,
      "train_runtime": 3968.8987,
      "train_tokens_per_second": 5957.917
    },
    {
      "epoch": 1.4388045977011494,
      "grad_norm": 0.929370105266571,
      "learning_rate": 5.621203754831585e-05,
      "loss": 1.4827,
      "num_input_tokens_seen": 23652128,
      "step": 3912,
      "train_runtime": 3969.8477,
      "train_tokens_per_second": 5957.943
    },
    {
      "epoch": 1.4391724137931035,
      "grad_norm": 0.9945060610771179,
      "learning_rate": 5.617522547395546e-05,
      "loss": 1.2793,
      "num_input_tokens_seen": 23657120,
      "step": 3913,
      "train_runtime": 3970.6861,
      "train_tokens_per_second": 5957.943
    },
    {
      "epoch": 1.4395402298850575,
      "grad_norm": 1.0037338733673096,
      "learning_rate": 5.613841339959507e-05,
      "loss": 1.5655,
      "num_input_tokens_seen": 23662656,
      "step": 3914,
      "train_runtime": 3971.5749,
      "train_tokens_per_second": 5958.003
    },
    {
      "epoch": 1.4399080459770115,
      "grad_norm": 0.9895558953285217,
      "learning_rate": 5.610160132523468e-05,
      "loss": 1.4959,
      "num_input_tokens_seen": 23667504,
      "step": 3915,
      "train_runtime": 3972.3952,
      "train_tokens_per_second": 5957.993
    },
    {
      "epoch": 1.4402758620689655,
      "grad_norm": 0.9794352054595947,
      "learning_rate": 5.6064789250874295e-05,
      "loss": 1.4233,
      "num_input_tokens_seen": 23672416,
      "step": 3916,
      "train_runtime": 3973.2274,
      "train_tokens_per_second": 5957.982
    },
    {
      "epoch": 1.4406436781609195,
      "grad_norm": 1.0201325416564941,
      "learning_rate": 5.6027977176513904e-05,
      "loss": 1.4731,
      "num_input_tokens_seen": 23676880,
      "step": 3917,
      "train_runtime": 3974.0054,
      "train_tokens_per_second": 5957.939
    },
    {
      "epoch": 1.4410114942528736,
      "grad_norm": 1.0298972129821777,
      "learning_rate": 5.5991165102153506e-05,
      "loss": 1.6392,
      "num_input_tokens_seen": 23681776,
      "step": 3918,
      "train_runtime": 3974.8313,
      "train_tokens_per_second": 5957.932
    },
    {
      "epoch": 1.4413793103448276,
      "grad_norm": 0.9182288646697998,
      "learning_rate": 5.5954353027793115e-05,
      "loss": 1.298,
      "num_input_tokens_seen": 23688976,
      "step": 3919,
      "train_runtime": 3975.9796,
      "train_tokens_per_second": 5958.023
    },
    {
      "epoch": 1.4417471264367816,
      "grad_norm": 0.9700836539268494,
      "learning_rate": 5.5917540953432724e-05,
      "loss": 1.4472,
      "num_input_tokens_seen": 23693808,
      "step": 3920,
      "train_runtime": 3976.807,
      "train_tokens_per_second": 5957.998
    },
    {
      "epoch": 1.4421149425287356,
      "grad_norm": 0.9916850924491882,
      "learning_rate": 5.588072887907234e-05,
      "loss": 1.4742,
      "num_input_tokens_seen": 23702576,
      "step": 3921,
      "train_runtime": 3978.1476,
      "train_tokens_per_second": 5958.194
    },
    {
      "epoch": 1.4424827586206896,
      "grad_norm": 0.9078749418258667,
      "learning_rate": 5.584391680471195e-05,
      "loss": 1.3597,
      "num_input_tokens_seen": 23707856,
      "step": 3922,
      "train_runtime": 3979.0346,
      "train_tokens_per_second": 5958.193
    },
    {
      "epoch": 1.4428505747126437,
      "grad_norm": 1.0607084035873413,
      "learning_rate": 5.580710473035156e-05,
      "loss": 1.6505,
      "num_input_tokens_seen": 23712576,
      "step": 3923,
      "train_runtime": 3979.8481,
      "train_tokens_per_second": 5958.161
    },
    {
      "epoch": 1.4432183908045977,
      "grad_norm": 0.9192100167274475,
      "learning_rate": 5.5770292655991166e-05,
      "loss": 1.3215,
      "num_input_tokens_seen": 23718112,
      "step": 3924,
      "train_runtime": 3980.7565,
      "train_tokens_per_second": 5958.192
    },
    {
      "epoch": 1.4435862068965517,
      "grad_norm": 1.0487370491027832,
      "learning_rate": 5.573348058163078e-05,
      "loss": 1.5791,
      "num_input_tokens_seen": 23722816,
      "step": 3925,
      "train_runtime": 3981.5574,
      "train_tokens_per_second": 5958.175
    },
    {
      "epoch": 1.4439540229885057,
      "grad_norm": 1.0452100038528442,
      "learning_rate": 5.569666850727039e-05,
      "loss": 1.7352,
      "num_input_tokens_seen": 23729184,
      "step": 3926,
      "train_runtime": 3982.5871,
      "train_tokens_per_second": 5958.234
    },
    {
      "epoch": 1.4443218390804597,
      "grad_norm": 1.0129988193511963,
      "learning_rate": 5.565985643290999e-05,
      "loss": 1.6463,
      "num_input_tokens_seen": 23734000,
      "step": 3927,
      "train_runtime": 3983.3857,
      "train_tokens_per_second": 5958.248
    },
    {
      "epoch": 1.4446896551724138,
      "grad_norm": 0.9765925407409668,
      "learning_rate": 5.56230443585496e-05,
      "loss": 1.4746,
      "num_input_tokens_seen": 23739488,
      "step": 3928,
      "train_runtime": 3984.2667,
      "train_tokens_per_second": 5958.308
    },
    {
      "epoch": 1.4450574712643678,
      "grad_norm": 0.902500331401825,
      "learning_rate": 5.5586232284189224e-05,
      "loss": 1.6207,
      "num_input_tokens_seen": 23744640,
      "step": 3929,
      "train_runtime": 3985.1347,
      "train_tokens_per_second": 5958.303
    },
    {
      "epoch": 1.4454252873563218,
      "grad_norm": 0.9580575823783875,
      "learning_rate": 5.5549420209828826e-05,
      "loss": 1.3617,
      "num_input_tokens_seen": 23753776,
      "step": 3930,
      "train_runtime": 3986.4911,
      "train_tokens_per_second": 5958.567
    },
    {
      "epoch": 1.4457931034482758,
      "grad_norm": 0.9147399663925171,
      "learning_rate": 5.5512608135468435e-05,
      "loss": 1.4165,
      "num_input_tokens_seen": 23760080,
      "step": 3931,
      "train_runtime": 3987.9724,
      "train_tokens_per_second": 5957.935
    },
    {
      "epoch": 1.4461609195402298,
      "grad_norm": 0.8609374165534973,
      "learning_rate": 5.5475796061108044e-05,
      "loss": 1.4154,
      "num_input_tokens_seen": 23765296,
      "step": 3932,
      "train_runtime": 3988.8481,
      "train_tokens_per_second": 5957.935
    },
    {
      "epoch": 1.4465287356321839,
      "grad_norm": 0.9341060519218445,
      "learning_rate": 5.543898398674766e-05,
      "loss": 1.4064,
      "num_input_tokens_seen": 23771616,
      "step": 3933,
      "train_runtime": 3989.8857,
      "train_tokens_per_second": 5957.969
    },
    {
      "epoch": 1.446896551724138,
      "grad_norm": 0.9795817732810974,
      "learning_rate": 5.540217191238727e-05,
      "loss": 1.7441,
      "num_input_tokens_seen": 23776336,
      "step": 3934,
      "train_runtime": 3990.6981,
      "train_tokens_per_second": 5957.939
    },
    {
      "epoch": 1.447264367816092,
      "grad_norm": 1.0131844282150269,
      "learning_rate": 5.536535983802688e-05,
      "loss": 1.5081,
      "num_input_tokens_seen": 23782736,
      "step": 3935,
      "train_runtime": 3991.7219,
      "train_tokens_per_second": 5958.014
    },
    {
      "epoch": 1.447632183908046,
      "grad_norm": 0.8310518860816956,
      "learning_rate": 5.532854776366648e-05,
      "loss": 1.4058,
      "num_input_tokens_seen": 23788448,
      "step": 3936,
      "train_runtime": 3992.6285,
      "train_tokens_per_second": 5958.092
    },
    {
      "epoch": 1.448,
      "grad_norm": 1.0397616624832153,
      "learning_rate": 5.529173568930609e-05,
      "loss": 1.5659,
      "num_input_tokens_seen": 23794400,
      "step": 3937,
      "train_runtime": 3993.5751,
      "train_tokens_per_second": 5958.17
    },
    {
      "epoch": 1.448367816091954,
      "grad_norm": 1.0304861068725586,
      "learning_rate": 5.525492361494571e-05,
      "loss": 1.5069,
      "num_input_tokens_seen": 23799424,
      "step": 3938,
      "train_runtime": 3994.4241,
      "train_tokens_per_second": 5958.161
    },
    {
      "epoch": 1.448735632183908,
      "grad_norm": 0.9898491501808167,
      "learning_rate": 5.521811154058531e-05,
      "loss": 1.5935,
      "num_input_tokens_seen": 23804160,
      "step": 3939,
      "train_runtime": 3995.2412,
      "train_tokens_per_second": 5958.128
    },
    {
      "epoch": 1.449103448275862,
      "grad_norm": 0.9128115773200989,
      "learning_rate": 5.518129946622492e-05,
      "loss": 1.4474,
      "num_input_tokens_seen": 23810656,
      "step": 3940,
      "train_runtime": 3996.2863,
      "train_tokens_per_second": 5958.196
    },
    {
      "epoch": 1.449471264367816,
      "grad_norm": 1.0561201572418213,
      "learning_rate": 5.514448739186453e-05,
      "loss": 1.3838,
      "num_input_tokens_seen": 23816976,
      "step": 3941,
      "train_runtime": 3997.3115,
      "train_tokens_per_second": 5958.249
    },
    {
      "epoch": 1.44983908045977,
      "grad_norm": 1.0236408710479736,
      "learning_rate": 5.510767531750415e-05,
      "loss": 1.5291,
      "num_input_tokens_seen": 23822160,
      "step": 3942,
      "train_runtime": 3998.1788,
      "train_tokens_per_second": 5958.253
    },
    {
      "epoch": 1.450206896551724,
      "grad_norm": 0.9636337161064148,
      "learning_rate": 5.5070863243143756e-05,
      "loss": 1.5549,
      "num_input_tokens_seen": 23829696,
      "step": 3943,
      "train_runtime": 3999.3635,
      "train_tokens_per_second": 5958.372
    },
    {
      "epoch": 1.450574712643678,
      "grad_norm": 0.8566339015960693,
      "learning_rate": 5.5034051168783364e-05,
      "loss": 1.382,
      "num_input_tokens_seen": 23835232,
      "step": 3944,
      "train_runtime": 4000.2786,
      "train_tokens_per_second": 5958.393
    },
    {
      "epoch": 1.4509425287356321,
      "grad_norm": 0.9837653040885925,
      "learning_rate": 5.4997239094422967e-05,
      "loss": 1.5552,
      "num_input_tokens_seen": 23840704,
      "step": 3945,
      "train_runtime": 4001.186,
      "train_tokens_per_second": 5958.409
    },
    {
      "epoch": 1.4513103448275861,
      "grad_norm": 0.9445203542709351,
      "learning_rate": 5.496042702006259e-05,
      "loss": 1.5177,
      "num_input_tokens_seen": 23845520,
      "step": 3946,
      "train_runtime": 4002.022,
      "train_tokens_per_second": 5958.368
    },
    {
      "epoch": 1.4516781609195402,
      "grad_norm": 0.9818313121795654,
      "learning_rate": 5.49236149457022e-05,
      "loss": 1.5207,
      "num_input_tokens_seen": 23850720,
      "step": 3947,
      "train_runtime": 4002.8686,
      "train_tokens_per_second": 5958.407
    },
    {
      "epoch": 1.4520459770114942,
      "grad_norm": 1.0295852422714233,
      "learning_rate": 5.48868028713418e-05,
      "loss": 1.5124,
      "num_input_tokens_seen": 23856064,
      "step": 3948,
      "train_runtime": 4003.78,
      "train_tokens_per_second": 5958.385
    },
    {
      "epoch": 1.4524137931034482,
      "grad_norm": 1.0007249116897583,
      "learning_rate": 5.484999079698141e-05,
      "loss": 1.5903,
      "num_input_tokens_seen": 23861344,
      "step": 3949,
      "train_runtime": 4004.6662,
      "train_tokens_per_second": 5958.385
    },
    {
      "epoch": 1.4527816091954022,
      "grad_norm": 0.9333604574203491,
      "learning_rate": 5.481317872262102e-05,
      "loss": 1.4851,
      "num_input_tokens_seen": 23868768,
      "step": 3950,
      "train_runtime": 4005.8037,
      "train_tokens_per_second": 5958.547
    },
    {
      "epoch": 1.4531494252873562,
      "grad_norm": 0.9393358826637268,
      "learning_rate": 5.4776366648260633e-05,
      "loss": 1.5632,
      "num_input_tokens_seen": 23874336,
      "step": 3951,
      "train_runtime": 4006.6983,
      "train_tokens_per_second": 5958.606
    },
    {
      "epoch": 1.4535172413793103,
      "grad_norm": 0.9209961295127869,
      "learning_rate": 5.473955457390024e-05,
      "loss": 1.6251,
      "num_input_tokens_seen": 23879888,
      "step": 3952,
      "train_runtime": 4007.5863,
      "train_tokens_per_second": 5958.671
    },
    {
      "epoch": 1.4538850574712643,
      "grad_norm": 1.0529483556747437,
      "learning_rate": 5.470274249953985e-05,
      "loss": 1.5663,
      "num_input_tokens_seen": 23885008,
      "step": 3953,
      "train_runtime": 4008.458,
      "train_tokens_per_second": 5958.652
    },
    {
      "epoch": 1.4542528735632185,
      "grad_norm": 0.9830681681632996,
      "learning_rate": 5.466593042517946e-05,
      "loss": 1.5933,
      "num_input_tokens_seen": 23890208,
      "step": 3954,
      "train_runtime": 4009.3302,
      "train_tokens_per_second": 5958.653
    },
    {
      "epoch": 1.4546206896551723,
      "grad_norm": 0.9561806917190552,
      "learning_rate": 5.4629118350819076e-05,
      "loss": 1.6661,
      "num_input_tokens_seen": 23895904,
      "step": 3955,
      "train_runtime": 4010.2585,
      "train_tokens_per_second": 5958.694
    },
    {
      "epoch": 1.4549885057471266,
      "grad_norm": 1.0198067426681519,
      "learning_rate": 5.4592306276458685e-05,
      "loss": 1.1809,
      "num_input_tokens_seen": 23901312,
      "step": 3956,
      "train_runtime": 4011.1577,
      "train_tokens_per_second": 5958.707
    },
    {
      "epoch": 1.4553563218390804,
      "grad_norm": 0.9848325848579407,
      "learning_rate": 5.455549420209829e-05,
      "loss": 1.7004,
      "num_input_tokens_seen": 23906176,
      "step": 3957,
      "train_runtime": 4011.9873,
      "train_tokens_per_second": 5958.687
    },
    {
      "epoch": 1.4557241379310346,
      "grad_norm": 0.9442578554153442,
      "learning_rate": 5.4518682127737896e-05,
      "loss": 1.5273,
      "num_input_tokens_seen": 23911280,
      "step": 3958,
      "train_runtime": 4012.8615,
      "train_tokens_per_second": 5958.661
    },
    {
      "epoch": 1.4560919540229884,
      "grad_norm": 0.969097912311554,
      "learning_rate": 5.448187005337752e-05,
      "loss": 1.6106,
      "num_input_tokens_seen": 23916688,
      "step": 3959,
      "train_runtime": 4013.7588,
      "train_tokens_per_second": 5958.676
    },
    {
      "epoch": 1.4564597701149427,
      "grad_norm": 0.9606767296791077,
      "learning_rate": 5.444505797901712e-05,
      "loss": 1.3795,
      "num_input_tokens_seen": 23925536,
      "step": 3960,
      "train_runtime": 4015.1427,
      "train_tokens_per_second": 5958.826
    },
    {
      "epoch": 1.4568275862068965,
      "grad_norm": 0.9321314096450806,
      "learning_rate": 5.440824590465673e-05,
      "loss": 1.4685,
      "num_input_tokens_seen": 23932160,
      "step": 3961,
      "train_runtime": 4016.6968,
      "train_tokens_per_second": 5958.169
    },
    {
      "epoch": 1.4571954022988507,
      "grad_norm": 0.9932836890220642,
      "learning_rate": 5.437143383029634e-05,
      "loss": 1.5359,
      "num_input_tokens_seen": 23937536,
      "step": 3962,
      "train_runtime": 4017.5924,
      "train_tokens_per_second": 5958.179
    },
    {
      "epoch": 1.4575632183908045,
      "grad_norm": 1.011725664138794,
      "learning_rate": 5.4334621755935954e-05,
      "loss": 1.5599,
      "num_input_tokens_seen": 23942208,
      "step": 3963,
      "train_runtime": 4018.3864,
      "train_tokens_per_second": 5958.165
    },
    {
      "epoch": 1.4579310344827587,
      "grad_norm": 0.9977604746818542,
      "learning_rate": 5.429780968157556e-05,
      "loss": 1.4931,
      "num_input_tokens_seen": 23946976,
      "step": 3964,
      "train_runtime": 4019.2157,
      "train_tokens_per_second": 5958.122
    },
    {
      "epoch": 1.4582988505747125,
      "grad_norm": 0.9668065309524536,
      "learning_rate": 5.426099760721517e-05,
      "loss": 1.7853,
      "num_input_tokens_seen": 23953120,
      "step": 3965,
      "train_runtime": 4020.2037,
      "train_tokens_per_second": 5958.186
    },
    {
      "epoch": 1.4586666666666668,
      "grad_norm": 1.0396864414215088,
      "learning_rate": 5.4224185532854774e-05,
      "loss": 1.4023,
      "num_input_tokens_seen": 23960480,
      "step": 3966,
      "train_runtime": 4021.3462,
      "train_tokens_per_second": 5958.323
    },
    {
      "epoch": 1.4590344827586206,
      "grad_norm": 0.8447195291519165,
      "learning_rate": 5.418737345849438e-05,
      "loss": 1.3982,
      "num_input_tokens_seen": 23967216,
      "step": 3967,
      "train_runtime": 4022.42,
      "train_tokens_per_second": 5958.407
    },
    {
      "epoch": 1.4594022988505748,
      "grad_norm": 1.0791133642196655,
      "learning_rate": 5.4150561384134005e-05,
      "loss": 1.502,
      "num_input_tokens_seen": 23974928,
      "step": 3968,
      "train_runtime": 4023.6332,
      "train_tokens_per_second": 5958.527
    },
    {
      "epoch": 1.4597701149425286,
      "grad_norm": 1.0344927310943604,
      "learning_rate": 5.411374930977361e-05,
      "loss": 1.4992,
      "num_input_tokens_seen": 23982032,
      "step": 3969,
      "train_runtime": 4024.7424,
      "train_tokens_per_second": 5958.65
    },
    {
      "epoch": 1.4601379310344829,
      "grad_norm": 1.0100144147872925,
      "learning_rate": 5.4076937235413216e-05,
      "loss": 1.562,
      "num_input_tokens_seen": 23989600,
      "step": 3970,
      "train_runtime": 4025.922,
      "train_tokens_per_second": 5958.784
    },
    {
      "epoch": 1.4605057471264367,
      "grad_norm": 1.0503922700881958,
      "learning_rate": 5.4040125161052825e-05,
      "loss": 1.5329,
      "num_input_tokens_seen": 23996992,
      "step": 3971,
      "train_runtime": 4027.0618,
      "train_tokens_per_second": 5958.933
    },
    {
      "epoch": 1.460873563218391,
      "grad_norm": 0.9958587288856506,
      "learning_rate": 5.400331308669244e-05,
      "loss": 1.498,
      "num_input_tokens_seen": 24002832,
      "step": 3972,
      "train_runtime": 4028.0027,
      "train_tokens_per_second": 5958.991
    },
    {
      "epoch": 1.4612413793103447,
      "grad_norm": 1.0499320030212402,
      "learning_rate": 5.396650101233205e-05,
      "loss": 1.4402,
      "num_input_tokens_seen": 24008864,
      "step": 3973,
      "train_runtime": 4028.9593,
      "train_tokens_per_second": 5959.073
    },
    {
      "epoch": 1.461609195402299,
      "grad_norm": 0.980239748954773,
      "learning_rate": 5.392968893797166e-05,
      "loss": 1.3266,
      "num_input_tokens_seen": 24014368,
      "step": 3974,
      "train_runtime": 4029.8671,
      "train_tokens_per_second": 5959.097
    },
    {
      "epoch": 1.4619770114942527,
      "grad_norm": 1.0136046409606934,
      "learning_rate": 5.389287686361126e-05,
      "loss": 1.5789,
      "num_input_tokens_seen": 24019376,
      "step": 3975,
      "train_runtime": 4030.7019,
      "train_tokens_per_second": 5959.105
    },
    {
      "epoch": 1.462344827586207,
      "grad_norm": 1.0189021825790405,
      "learning_rate": 5.385606478925088e-05,
      "loss": 1.5808,
      "num_input_tokens_seen": 24024016,
      "step": 3976,
      "train_runtime": 4031.4968,
      "train_tokens_per_second": 5959.081
    },
    {
      "epoch": 1.4627126436781608,
      "grad_norm": 0.9636145830154419,
      "learning_rate": 5.381925271489049e-05,
      "loss": 1.5115,
      "num_input_tokens_seen": 24028480,
      "step": 3977,
      "train_runtime": 4032.2736,
      "train_tokens_per_second": 5959.04
    },
    {
      "epoch": 1.463080459770115,
      "grad_norm": 1.0247414112091064,
      "learning_rate": 5.3782440640530094e-05,
      "loss": 1.5167,
      "num_input_tokens_seen": 24033184,
      "step": 3978,
      "train_runtime": 4033.0716,
      "train_tokens_per_second": 5959.027
    },
    {
      "epoch": 1.463448275862069,
      "grad_norm": 1.0392184257507324,
      "learning_rate": 5.37456285661697e-05,
      "loss": 1.4035,
      "num_input_tokens_seen": 24037664,
      "step": 3979,
      "train_runtime": 4033.842,
      "train_tokens_per_second": 5959.0
    },
    {
      "epoch": 1.463816091954023,
      "grad_norm": 1.0267279148101807,
      "learning_rate": 5.370881649180931e-05,
      "loss": 1.6334,
      "num_input_tokens_seen": 24043504,
      "step": 3980,
      "train_runtime": 4034.7993,
      "train_tokens_per_second": 5959.034
    },
    {
      "epoch": 1.464183908045977,
      "grad_norm": 0.9114202260971069,
      "learning_rate": 5.367200441744893e-05,
      "loss": 1.4836,
      "num_input_tokens_seen": 24049296,
      "step": 3981,
      "train_runtime": 4035.7378,
      "train_tokens_per_second": 5959.083
    },
    {
      "epoch": 1.464551724137931,
      "grad_norm": 0.9234261512756348,
      "learning_rate": 5.3635192343088536e-05,
      "loss": 1.2929,
      "num_input_tokens_seen": 24057264,
      "step": 3982,
      "train_runtime": 4036.9513,
      "train_tokens_per_second": 5959.265
    },
    {
      "epoch": 1.4649195402298851,
      "grad_norm": 0.9922747015953064,
      "learning_rate": 5.3598380268728145e-05,
      "loss": 1.4076,
      "num_input_tokens_seen": 24061936,
      "step": 3983,
      "train_runtime": 4037.7658,
      "train_tokens_per_second": 5959.22
    },
    {
      "epoch": 1.4652873563218392,
      "grad_norm": 1.0076498985290527,
      "learning_rate": 5.356156819436775e-05,
      "loss": 1.5843,
      "num_input_tokens_seen": 24067104,
      "step": 3984,
      "train_runtime": 4038.6409,
      "train_tokens_per_second": 5959.209
    },
    {
      "epoch": 1.4656551724137932,
      "grad_norm": 1.0241613388061523,
      "learning_rate": 5.352475612000737e-05,
      "loss": 1.5223,
      "num_input_tokens_seen": 24072576,
      "step": 3985,
      "train_runtime": 4039.5324,
      "train_tokens_per_second": 5959.248
    },
    {
      "epoch": 1.4660229885057472,
      "grad_norm": 0.9356365203857422,
      "learning_rate": 5.348794404564698e-05,
      "loss": 1.4281,
      "num_input_tokens_seen": 24078368,
      "step": 3986,
      "train_runtime": 4040.4765,
      "train_tokens_per_second": 5959.289
    },
    {
      "epoch": 1.4663908045977012,
      "grad_norm": 0.9243559241294861,
      "learning_rate": 5.345113197128658e-05,
      "loss": 1.4318,
      "num_input_tokens_seen": 24084576,
      "step": 3987,
      "train_runtime": 4041.4565,
      "train_tokens_per_second": 5959.38
    },
    {
      "epoch": 1.4667586206896552,
      "grad_norm": 0.8942695260047913,
      "learning_rate": 5.341431989692619e-05,
      "loss": 1.3723,
      "num_input_tokens_seen": 24091280,
      "step": 3988,
      "train_runtime": 4042.5297,
      "train_tokens_per_second": 5959.457
    },
    {
      "epoch": 1.4671264367816093,
      "grad_norm": 1.0101655721664429,
      "learning_rate": 5.337750782256581e-05,
      "loss": 1.6068,
      "num_input_tokens_seen": 24096192,
      "step": 3989,
      "train_runtime": 4043.3485,
      "train_tokens_per_second": 5959.465
    },
    {
      "epoch": 1.4674942528735633,
      "grad_norm": 0.9235145449638367,
      "learning_rate": 5.3340695748205414e-05,
      "loss": 1.5572,
      "num_input_tokens_seen": 24100816,
      "step": 3990,
      "train_runtime": 4044.1371,
      "train_tokens_per_second": 5959.446
    },
    {
      "epoch": 1.4678620689655173,
      "grad_norm": 1.0393980741500854,
      "learning_rate": 5.330388367384502e-05,
      "loss": 1.4833,
      "num_input_tokens_seen": 24108800,
      "step": 3991,
      "train_runtime": 4045.8696,
      "train_tokens_per_second": 5958.867
    },
    {
      "epoch": 1.4682298850574713,
      "grad_norm": 0.9173269867897034,
      "learning_rate": 5.326707159948463e-05,
      "loss": 1.4363,
      "num_input_tokens_seen": 24114816,
      "step": 3992,
      "train_runtime": 4046.8569,
      "train_tokens_per_second": 5958.9
    },
    {
      "epoch": 1.4685977011494253,
      "grad_norm": 0.9402369856834412,
      "learning_rate": 5.323025952512425e-05,
      "loss": 1.5039,
      "num_input_tokens_seen": 24120784,
      "step": 3993,
      "train_runtime": 4047.8128,
      "train_tokens_per_second": 5958.967
    },
    {
      "epoch": 1.4689655172413794,
      "grad_norm": 1.1363598108291626,
      "learning_rate": 5.3193447450763856e-05,
      "loss": 1.7887,
      "num_input_tokens_seen": 24125072,
      "step": 3994,
      "train_runtime": 4048.5535,
      "train_tokens_per_second": 5958.936
    },
    {
      "epoch": 1.4693333333333334,
      "grad_norm": 0.9854596257209778,
      "learning_rate": 5.3156635376403465e-05,
      "loss": 1.6261,
      "num_input_tokens_seen": 24130384,
      "step": 3995,
      "train_runtime": 4049.4271,
      "train_tokens_per_second": 5958.962
    },
    {
      "epoch": 1.4697011494252874,
      "grad_norm": 1.0667961835861206,
      "learning_rate": 5.311982330204307e-05,
      "loss": 1.5302,
      "num_input_tokens_seen": 24135776,
      "step": 3996,
      "train_runtime": 4050.3218,
      "train_tokens_per_second": 5958.977
    },
    {
      "epoch": 1.4700689655172414,
      "grad_norm": 1.052289605140686,
      "learning_rate": 5.3083011227682676e-05,
      "loss": 1.682,
      "num_input_tokens_seen": 24140496,
      "step": 3997,
      "train_runtime": 4051.1089,
      "train_tokens_per_second": 5958.985
    },
    {
      "epoch": 1.4704367816091954,
      "grad_norm": 1.0311145782470703,
      "learning_rate": 5.30461991533223e-05,
      "loss": 1.6228,
      "num_input_tokens_seen": 24144832,
      "step": 3998,
      "train_runtime": 4051.8618,
      "train_tokens_per_second": 5958.947
    },
    {
      "epoch": 1.4708045977011495,
      "grad_norm": 1.0431197881698608,
      "learning_rate": 5.30093870789619e-05,
      "loss": 1.4621,
      "num_input_tokens_seen": 24152464,
      "step": 3999,
      "train_runtime": 4053.0928,
      "train_tokens_per_second": 5959.021
    },
    {
      "epoch": 1.4711724137931035,
      "grad_norm": 0.9911890625953674,
      "learning_rate": 5.297257500460151e-05,
      "loss": 1.5715,
      "num_input_tokens_seen": 24160880,
      "step": 4000,
      "train_runtime": 4054.3684,
      "train_tokens_per_second": 5959.222
    },
    {
      "epoch": 1.4715402298850575,
      "grad_norm": 0.91972815990448,
      "learning_rate": 5.293576293024112e-05,
      "loss": 1.3507,
      "num_input_tokens_seen": 24169168,
      "step": 4001,
      "train_runtime": 4055.6576,
      "train_tokens_per_second": 5959.371
    },
    {
      "epoch": 1.4719080459770115,
      "grad_norm": 0.998271644115448,
      "learning_rate": 5.2898950855880734e-05,
      "loss": 1.5106,
      "num_input_tokens_seen": 24174256,
      "step": 4002,
      "train_runtime": 4056.51,
      "train_tokens_per_second": 5959.373
    },
    {
      "epoch": 1.4722758620689655,
      "grad_norm": 0.8969280123710632,
      "learning_rate": 5.286213878152034e-05,
      "loss": 1.4304,
      "num_input_tokens_seen": 24182336,
      "step": 4003,
      "train_runtime": 4057.7493,
      "train_tokens_per_second": 5959.544
    },
    {
      "epoch": 1.4726436781609196,
      "grad_norm": 1.0865023136138916,
      "learning_rate": 5.282532670715995e-05,
      "loss": 1.5719,
      "num_input_tokens_seen": 24186992,
      "step": 4004,
      "train_runtime": 4058.5463,
      "train_tokens_per_second": 5959.521
    },
    {
      "epoch": 1.4730114942528736,
      "grad_norm": 0.9584881067276001,
      "learning_rate": 5.2788514632799554e-05,
      "loss": 1.5604,
      "num_input_tokens_seen": 24193616,
      "step": 4005,
      "train_runtime": 4059.6164,
      "train_tokens_per_second": 5959.582
    },
    {
      "epoch": 1.4733793103448276,
      "grad_norm": 0.956598699092865,
      "learning_rate": 5.275170255843918e-05,
      "loss": 1.6063,
      "num_input_tokens_seen": 24199376,
      "step": 4006,
      "train_runtime": 4060.5674,
      "train_tokens_per_second": 5959.605
    },
    {
      "epoch": 1.4737471264367816,
      "grad_norm": 0.9755147695541382,
      "learning_rate": 5.2714890484078786e-05,
      "loss": 1.3879,
      "num_input_tokens_seen": 24207920,
      "step": 4007,
      "train_runtime": 4061.8704,
      "train_tokens_per_second": 5959.796
    },
    {
      "epoch": 1.4741149425287357,
      "grad_norm": 0.900531530380249,
      "learning_rate": 5.267807840971839e-05,
      "loss": 1.3409,
      "num_input_tokens_seen": 24216720,
      "step": 4008,
      "train_runtime": 4063.2073,
      "train_tokens_per_second": 5960.001
    },
    {
      "epoch": 1.4744827586206897,
      "grad_norm": 1.003485083580017,
      "learning_rate": 5.2641266335358e-05,
      "loss": 1.4971,
      "num_input_tokens_seen": 24229248,
      "step": 4009,
      "train_runtime": 4065.0502,
      "train_tokens_per_second": 5960.381
    },
    {
      "epoch": 1.4748505747126437,
      "grad_norm": 1.0460050106048584,
      "learning_rate": 5.2604454260997606e-05,
      "loss": 1.4891,
      "num_input_tokens_seen": 24235264,
      "step": 4010,
      "train_runtime": 4066.005,
      "train_tokens_per_second": 5960.461
    },
    {
      "epoch": 1.4752183908045977,
      "grad_norm": 0.9604606032371521,
      "learning_rate": 5.256764218663722e-05,
      "loss": 1.4463,
      "num_input_tokens_seen": 24239808,
      "step": 4011,
      "train_runtime": 4066.7945,
      "train_tokens_per_second": 5960.421
    },
    {
      "epoch": 1.4755862068965517,
      "grad_norm": 0.9596603512763977,
      "learning_rate": 5.253083011227683e-05,
      "loss": 1.4025,
      "num_input_tokens_seen": 24246096,
      "step": 4012,
      "train_runtime": 4067.7933,
      "train_tokens_per_second": 5960.504
    },
    {
      "epoch": 1.4759540229885058,
      "grad_norm": 0.9584473967552185,
      "learning_rate": 5.249401803791644e-05,
      "loss": 1.3761,
      "num_input_tokens_seen": 24250160,
      "step": 4013,
      "train_runtime": 4068.4905,
      "train_tokens_per_second": 5960.481
    },
    {
      "epoch": 1.4763218390804598,
      "grad_norm": 1.0671095848083496,
      "learning_rate": 5.245720596355604e-05,
      "loss": 1.5959,
      "num_input_tokens_seen": 24256480,
      "step": 4014,
      "train_runtime": 4069.528,
      "train_tokens_per_second": 5960.514
    },
    {
      "epoch": 1.4766896551724138,
      "grad_norm": 0.9837397933006287,
      "learning_rate": 5.2420393889195664e-05,
      "loss": 1.5743,
      "num_input_tokens_seen": 24262560,
      "step": 4015,
      "train_runtime": 4070.4795,
      "train_tokens_per_second": 5960.615
    },
    {
      "epoch": 1.4770574712643678,
      "grad_norm": 0.9995848536491394,
      "learning_rate": 5.238358181483527e-05,
      "loss": 1.4061,
      "num_input_tokens_seen": 24267104,
      "step": 4016,
      "train_runtime": 4071.2706,
      "train_tokens_per_second": 5960.573
    },
    {
      "epoch": 1.4774252873563218,
      "grad_norm": 0.9571737051010132,
      "learning_rate": 5.2346769740474875e-05,
      "loss": 1.4699,
      "num_input_tokens_seen": 24274896,
      "step": 4017,
      "train_runtime": 4072.5066,
      "train_tokens_per_second": 5960.677
    },
    {
      "epoch": 1.4777931034482759,
      "grad_norm": 0.9793049693107605,
      "learning_rate": 5.2309957666114483e-05,
      "loss": 1.4763,
      "num_input_tokens_seen": 24280256,
      "step": 4018,
      "train_runtime": 4073.408,
      "train_tokens_per_second": 5960.674
    },
    {
      "epoch": 1.4781609195402299,
      "grad_norm": 0.876334011554718,
      "learning_rate": 5.22731455917541e-05,
      "loss": 1.4638,
      "num_input_tokens_seen": 24287520,
      "step": 4019,
      "train_runtime": 4074.54,
      "train_tokens_per_second": 5960.8
    },
    {
      "epoch": 1.478528735632184,
      "grad_norm": 0.9345941543579102,
      "learning_rate": 5.223633351739371e-05,
      "loss": 1.3881,
      "num_input_tokens_seen": 24291872,
      "step": 4020,
      "train_runtime": 4075.3064,
      "train_tokens_per_second": 5960.747
    },
    {
      "epoch": 1.478896551724138,
      "grad_norm": 0.9565058350563049,
      "learning_rate": 5.219952144303332e-05,
      "loss": 1.4838,
      "num_input_tokens_seen": 24297856,
      "step": 4021,
      "train_runtime": 4076.7539,
      "train_tokens_per_second": 5960.099
    },
    {
      "epoch": 1.479264367816092,
      "grad_norm": 0.6849691867828369,
      "learning_rate": 5.2162709368672926e-05,
      "loss": 1.1094,
      "num_input_tokens_seen": 24319344,
      "step": 4022,
      "train_runtime": 4079.7715,
      "train_tokens_per_second": 5960.957
    },
    {
      "epoch": 1.479632183908046,
      "grad_norm": 0.9383494853973389,
      "learning_rate": 5.212589729431254e-05,
      "loss": 1.3657,
      "num_input_tokens_seen": 24326480,
      "step": 4023,
      "train_runtime": 4080.8851,
      "train_tokens_per_second": 5961.079
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.9587903022766113,
      "learning_rate": 5.208908521995215e-05,
      "loss": 1.5376,
      "num_input_tokens_seen": 24333040,
      "step": 4024,
      "train_runtime": 4081.9149,
      "train_tokens_per_second": 5961.183
    },
    {
      "epoch": 1.480367816091954,
      "grad_norm": 0.9399529099464417,
      "learning_rate": 5.205227314559176e-05,
      "loss": 1.3645,
      "num_input_tokens_seen": 24339872,
      "step": 4025,
      "train_runtime": 4082.9763,
      "train_tokens_per_second": 5961.306
    },
    {
      "epoch": 1.480735632183908,
      "grad_norm": 0.9669511318206787,
      "learning_rate": 5.201546107123136e-05,
      "loss": 1.6092,
      "num_input_tokens_seen": 24345856,
      "step": 4026,
      "train_runtime": 4083.9317,
      "train_tokens_per_second": 5961.377
    },
    {
      "epoch": 1.481103448275862,
      "grad_norm": 1.1014961004257202,
      "learning_rate": 5.197864899687097e-05,
      "loss": 1.557,
      "num_input_tokens_seen": 24353664,
      "step": 4027,
      "train_runtime": 4085.172,
      "train_tokens_per_second": 5961.478
    },
    {
      "epoch": 1.481471264367816,
      "grad_norm": 1.007810115814209,
      "learning_rate": 5.194183692251059e-05,
      "loss": 1.4677,
      "num_input_tokens_seen": 24359168,
      "step": 4028,
      "train_runtime": 4086.0586,
      "train_tokens_per_second": 5961.532
    },
    {
      "epoch": 1.48183908045977,
      "grad_norm": 1.085648775100708,
      "learning_rate": 5.1905024848150195e-05,
      "loss": 1.6783,
      "num_input_tokens_seen": 24365088,
      "step": 4029,
      "train_runtime": 4086.9885,
      "train_tokens_per_second": 5961.624
    },
    {
      "epoch": 1.482206896551724,
      "grad_norm": 1.0785069465637207,
      "learning_rate": 5.1868212773789804e-05,
      "loss": 1.6227,
      "num_input_tokens_seen": 24369328,
      "step": 4030,
      "train_runtime": 4087.7473,
      "train_tokens_per_second": 5961.554
    },
    {
      "epoch": 1.4825747126436781,
      "grad_norm": 1.0458329916000366,
      "learning_rate": 5.183140069942941e-05,
      "loss": 1.4058,
      "num_input_tokens_seen": 24374352,
      "step": 4031,
      "train_runtime": 4088.5752,
      "train_tokens_per_second": 5961.576
    },
    {
      "epoch": 1.4829425287356321,
      "grad_norm": 1.00020170211792,
      "learning_rate": 5.179458862506903e-05,
      "loss": 1.5499,
      "num_input_tokens_seen": 24379344,
      "step": 4032,
      "train_runtime": 4089.3924,
      "train_tokens_per_second": 5961.605
    },
    {
      "epoch": 1.4833103448275862,
      "grad_norm": 0.91293865442276,
      "learning_rate": 5.175777655070864e-05,
      "loss": 1.4224,
      "num_input_tokens_seen": 24385840,
      "step": 4033,
      "train_runtime": 4090.4516,
      "train_tokens_per_second": 5961.65
    },
    {
      "epoch": 1.4836781609195402,
      "grad_norm": 1.145096778869629,
      "learning_rate": 5.1720964476348246e-05,
      "loss": 1.7098,
      "num_input_tokens_seen": 24391520,
      "step": 4034,
      "train_runtime": 4091.4016,
      "train_tokens_per_second": 5961.654
    },
    {
      "epoch": 1.4840459770114942,
      "grad_norm": 0.781602680683136,
      "learning_rate": 5.168415240198785e-05,
      "loss": 1.2461,
      "num_input_tokens_seen": 24399008,
      "step": 4035,
      "train_runtime": 4092.5895,
      "train_tokens_per_second": 5961.753
    },
    {
      "epoch": 1.4844137931034482,
      "grad_norm": 1.0057408809661865,
      "learning_rate": 5.164734032762747e-05,
      "loss": 1.5055,
      "num_input_tokens_seen": 24404448,
      "step": 4036,
      "train_runtime": 4093.507,
      "train_tokens_per_second": 5961.746
    },
    {
      "epoch": 1.4847816091954023,
      "grad_norm": 0.9656437039375305,
      "learning_rate": 5.161052825326708e-05,
      "loss": 1.5091,
      "num_input_tokens_seen": 24411280,
      "step": 4037,
      "train_runtime": 4094.5885,
      "train_tokens_per_second": 5961.84
    },
    {
      "epoch": 1.4851494252873563,
      "grad_norm": 0.9602141976356506,
      "learning_rate": 5.157371617890668e-05,
      "loss": 1.3938,
      "num_input_tokens_seen": 24418800,
      "step": 4038,
      "train_runtime": 4095.7714,
      "train_tokens_per_second": 5961.954
    },
    {
      "epoch": 1.4855172413793103,
      "grad_norm": 0.9323406219482422,
      "learning_rate": 5.153690410454629e-05,
      "loss": 1.6365,
      "num_input_tokens_seen": 24424000,
      "step": 4039,
      "train_runtime": 4096.6273,
      "train_tokens_per_second": 5961.978
    },
    {
      "epoch": 1.4858850574712643,
      "grad_norm": 1.044651746749878,
      "learning_rate": 5.15000920301859e-05,
      "loss": 1.6499,
      "num_input_tokens_seen": 24428048,
      "step": 4040,
      "train_runtime": 4097.3468,
      "train_tokens_per_second": 5961.919
    },
    {
      "epoch": 1.4862528735632183,
      "grad_norm": 0.9557268619537354,
      "learning_rate": 5.1463279955825515e-05,
      "loss": 1.4491,
      "num_input_tokens_seen": 24432688,
      "step": 4041,
      "train_runtime": 4098.1433,
      "train_tokens_per_second": 5961.892
    },
    {
      "epoch": 1.4866206896551724,
      "grad_norm": 0.930274486541748,
      "learning_rate": 5.1426467881465124e-05,
      "loss": 1.4152,
      "num_input_tokens_seen": 24437936,
      "step": 4042,
      "train_runtime": 4099.0031,
      "train_tokens_per_second": 5961.922
    },
    {
      "epoch": 1.4869885057471264,
      "grad_norm": 0.9440920352935791,
      "learning_rate": 5.138965580710473e-05,
      "loss": 1.2971,
      "num_input_tokens_seen": 24443600,
      "step": 4043,
      "train_runtime": 4099.9281,
      "train_tokens_per_second": 5961.958
    },
    {
      "epoch": 1.4873563218390804,
      "grad_norm": 1.0481032133102417,
      "learning_rate": 5.1352843732744335e-05,
      "loss": 1.4669,
      "num_input_tokens_seen": 24448512,
      "step": 4044,
      "train_runtime": 4100.7674,
      "train_tokens_per_second": 5961.936
    },
    {
      "epoch": 1.4877241379310344,
      "grad_norm": 1.0776386260986328,
      "learning_rate": 5.131603165838396e-05,
      "loss": 1.7256,
      "num_input_tokens_seen": 24452720,
      "step": 4045,
      "train_runtime": 4101.5358,
      "train_tokens_per_second": 5961.845
    },
    {
      "epoch": 1.4880919540229884,
      "grad_norm": 0.9875791668891907,
      "learning_rate": 5.1279219584023566e-05,
      "loss": 1.4522,
      "num_input_tokens_seen": 24458512,
      "step": 4046,
      "train_runtime": 4102.4779,
      "train_tokens_per_second": 5961.888
    },
    {
      "epoch": 1.4884597701149425,
      "grad_norm": 0.948467493057251,
      "learning_rate": 5.124240750966317e-05,
      "loss": 1.3109,
      "num_input_tokens_seen": 24464672,
      "step": 4047,
      "train_runtime": 4103.4725,
      "train_tokens_per_second": 5961.944
    },
    {
      "epoch": 1.4888275862068965,
      "grad_norm": 1.076884388923645,
      "learning_rate": 5.120559543530278e-05,
      "loss": 1.6678,
      "num_input_tokens_seen": 24470768,
      "step": 4048,
      "train_runtime": 4104.4708,
      "train_tokens_per_second": 5961.979
    },
    {
      "epoch": 1.4891954022988505,
      "grad_norm": 0.9881492853164673,
      "learning_rate": 5.116878336094239e-05,
      "loss": 1.4313,
      "num_input_tokens_seen": 24478768,
      "step": 4049,
      "train_runtime": 4105.7322,
      "train_tokens_per_second": 5962.096
    },
    {
      "epoch": 1.4895632183908045,
      "grad_norm": 0.906985878944397,
      "learning_rate": 5.1131971286582e-05,
      "loss": 1.481,
      "num_input_tokens_seen": 24484000,
      "step": 4050,
      "train_runtime": 4106.5983,
      "train_tokens_per_second": 5962.112
    },
    {
      "epoch": 1.4899310344827585,
      "grad_norm": 0.8949291110038757,
      "learning_rate": 5.109515921222161e-05,
      "loss": 1.31,
      "num_input_tokens_seen": 24492896,
      "step": 4051,
      "train_runtime": 4108.5125,
      "train_tokens_per_second": 5961.5
    },
    {
      "epoch": 1.4902988505747126,
      "grad_norm": 1.043642282485962,
      "learning_rate": 5.105834713786122e-05,
      "loss": 1.4841,
      "num_input_tokens_seen": 24497888,
      "step": 4052,
      "train_runtime": 4109.3588,
      "train_tokens_per_second": 5961.487
    },
    {
      "epoch": 1.4906666666666666,
      "grad_norm": 1.0845096111297607,
      "learning_rate": 5.1021535063500835e-05,
      "loss": 1.608,
      "num_input_tokens_seen": 24502560,
      "step": 4053,
      "train_runtime": 4110.1501,
      "train_tokens_per_second": 5961.476
    },
    {
      "epoch": 1.4910344827586206,
      "grad_norm": 1.0524910688400269,
      "learning_rate": 5.0984722989140444e-05,
      "loss": 1.6301,
      "num_input_tokens_seen": 24508064,
      "step": 4054,
      "train_runtime": 4111.0458,
      "train_tokens_per_second": 5961.516
    },
    {
      "epoch": 1.4914022988505748,
      "grad_norm": 1.0239331722259521,
      "learning_rate": 5.094791091478005e-05,
      "loss": 1.5205,
      "num_input_tokens_seen": 24512144,
      "step": 4055,
      "train_runtime": 4111.7427,
      "train_tokens_per_second": 5961.498
    },
    {
      "epoch": 1.4917701149425286,
      "grad_norm": 0.9673982858657837,
      "learning_rate": 5.0911098840419655e-05,
      "loss": 1.5423,
      "num_input_tokens_seen": 24519920,
      "step": 4056,
      "train_runtime": 4112.9593,
      "train_tokens_per_second": 5961.625
    },
    {
      "epoch": 1.492137931034483,
      "grad_norm": 1.0554982423782349,
      "learning_rate": 5.0874286766059264e-05,
      "loss": 1.5934,
      "num_input_tokens_seen": 24525360,
      "step": 4057,
      "train_runtime": 4113.8594,
      "train_tokens_per_second": 5961.643
    },
    {
      "epoch": 1.4925057471264367,
      "grad_norm": 0.95525723695755,
      "learning_rate": 5.083747469169888e-05,
      "loss": 1.4316,
      "num_input_tokens_seen": 24530736,
      "step": 4058,
      "train_runtime": 4114.7572,
      "train_tokens_per_second": 5961.648
    },
    {
      "epoch": 1.492873563218391,
      "grad_norm": 0.934636116027832,
      "learning_rate": 5.080066261733849e-05,
      "loss": 1.6114,
      "num_input_tokens_seen": 24537056,
      "step": 4059,
      "train_runtime": 4115.7718,
      "train_tokens_per_second": 5961.714
    },
    {
      "epoch": 1.4932413793103447,
      "grad_norm": 1.023818850517273,
      "learning_rate": 5.07638505429781e-05,
      "loss": 1.5031,
      "num_input_tokens_seen": 24541808,
      "step": 4060,
      "train_runtime": 4116.5724,
      "train_tokens_per_second": 5961.709
    },
    {
      "epoch": 1.493609195402299,
      "grad_norm": 0.9134024381637573,
      "learning_rate": 5.0727038468617706e-05,
      "loss": 1.721,
      "num_input_tokens_seen": 24549264,
      "step": 4061,
      "train_runtime": 4117.7379,
      "train_tokens_per_second": 5961.833
    },
    {
      "epoch": 1.4939770114942528,
      "grad_norm": 0.988197386264801,
      "learning_rate": 5.069022639425732e-05,
      "loss": 1.475,
      "num_input_tokens_seen": 24554176,
      "step": 4062,
      "train_runtime": 4118.5675,
      "train_tokens_per_second": 5961.824
    },
    {
      "epoch": 1.494344827586207,
      "grad_norm": 1.0888458490371704,
      "learning_rate": 5.065341431989693e-05,
      "loss": 1.5108,
      "num_input_tokens_seen": 24558320,
      "step": 4063,
      "train_runtime": 4119.2937,
      "train_tokens_per_second": 5961.779
    },
    {
      "epoch": 1.4947126436781608,
      "grad_norm": 0.9004330635070801,
      "learning_rate": 5.061660224553654e-05,
      "loss": 1.5406,
      "num_input_tokens_seen": 24564784,
      "step": 4064,
      "train_runtime": 4120.3084,
      "train_tokens_per_second": 5961.88
    },
    {
      "epoch": 1.495080459770115,
      "grad_norm": 0.9852762222290039,
      "learning_rate": 5.057979017117614e-05,
      "loss": 1.577,
      "num_input_tokens_seen": 24569744,
      "step": 4065,
      "train_runtime": 4121.1456,
      "train_tokens_per_second": 5961.872
    },
    {
      "epoch": 1.4954482758620689,
      "grad_norm": 1.106040120124817,
      "learning_rate": 5.0542978096815764e-05,
      "loss": 1.6727,
      "num_input_tokens_seen": 24574416,
      "step": 4066,
      "train_runtime": 4121.9589,
      "train_tokens_per_second": 5961.829
    },
    {
      "epoch": 1.495816091954023,
      "grad_norm": 0.972218930721283,
      "learning_rate": 5.050616602245537e-05,
      "loss": 1.4289,
      "num_input_tokens_seen": 24581744,
      "step": 4067,
      "train_runtime": 4123.1106,
      "train_tokens_per_second": 5961.941
    },
    {
      "epoch": 1.496183908045977,
      "grad_norm": 0.9878373146057129,
      "learning_rate": 5.0469353948094975e-05,
      "loss": 1.6709,
      "num_input_tokens_seen": 24587936,
      "step": 4068,
      "train_runtime": 4124.0833,
      "train_tokens_per_second": 5962.037
    },
    {
      "epoch": 1.4965517241379311,
      "grad_norm": 0.9404173493385315,
      "learning_rate": 5.0432541873734584e-05,
      "loss": 1.522,
      "num_input_tokens_seen": 24593008,
      "step": 4069,
      "train_runtime": 4124.9337,
      "train_tokens_per_second": 5962.037
    },
    {
      "epoch": 1.496919540229885,
      "grad_norm": 1.0829967260360718,
      "learning_rate": 5.039572979937419e-05,
      "loss": 1.5438,
      "num_input_tokens_seen": 24597488,
      "step": 4070,
      "train_runtime": 4125.6955,
      "train_tokens_per_second": 5962.022
    },
    {
      "epoch": 1.4972873563218392,
      "grad_norm": 0.8731368184089661,
      "learning_rate": 5.035891772501381e-05,
      "loss": 1.3352,
      "num_input_tokens_seen": 24608784,
      "step": 4071,
      "train_runtime": 4127.3506,
      "train_tokens_per_second": 5962.368
    },
    {
      "epoch": 1.497655172413793,
      "grad_norm": 1.0104687213897705,
      "learning_rate": 5.032210565065342e-05,
      "loss": 1.4738,
      "num_input_tokens_seen": 24616208,
      "step": 4072,
      "train_runtime": 4128.5454,
      "train_tokens_per_second": 5962.441
    },
    {
      "epoch": 1.4980229885057472,
      "grad_norm": 1.0959763526916504,
      "learning_rate": 5.028529357629303e-05,
      "loss": 1.8198,
      "num_input_tokens_seen": 24620720,
      "step": 4073,
      "train_runtime": 4129.3171,
      "train_tokens_per_second": 5962.419
    },
    {
      "epoch": 1.498390804597701,
      "grad_norm": 0.8429267406463623,
      "learning_rate": 5.024848150193263e-05,
      "loss": 1.4432,
      "num_input_tokens_seen": 24628336,
      "step": 4074,
      "train_runtime": 4130.4818,
      "train_tokens_per_second": 5962.582
    },
    {
      "epoch": 1.4987586206896553,
      "grad_norm": 1.0586172342300415,
      "learning_rate": 5.021166942757225e-05,
      "loss": 1.5656,
      "num_input_tokens_seen": 24632240,
      "step": 4075,
      "train_runtime": 4131.1707,
      "train_tokens_per_second": 5962.533
    },
    {
      "epoch": 1.499126436781609,
      "grad_norm": 0.8929089307785034,
      "learning_rate": 5.017485735321186e-05,
      "loss": 1.4776,
      "num_input_tokens_seen": 24641392,
      "step": 4076,
      "train_runtime": 4132.5523,
      "train_tokens_per_second": 5962.754
    },
    {
      "epoch": 1.4994942528735633,
      "grad_norm": 1.0645390748977661,
      "learning_rate": 5.013804527885146e-05,
      "loss": 1.3586,
      "num_input_tokens_seen": 24649328,
      "step": 4077,
      "train_runtime": 4133.7655,
      "train_tokens_per_second": 5962.924
    },
    {
      "epoch": 1.499862068965517,
      "grad_norm": 1.0506035089492798,
      "learning_rate": 5.010123320449107e-05,
      "loss": 1.7051,
      "num_input_tokens_seen": 24655008,
      "step": 4078,
      "train_runtime": 4134.7006,
      "train_tokens_per_second": 5962.949
    },
    {
      "epoch": 1.5002298850574713,
      "grad_norm": 0.9689409732818604,
      "learning_rate": 5.006442113013069e-05,
      "loss": 1.4383,
      "num_input_tokens_seen": 24661296,
      "step": 4079,
      "train_runtime": 4135.6891,
      "train_tokens_per_second": 5963.044
    },
    {
      "epoch": 1.5005977011494251,
      "grad_norm": 0.9953100085258484,
      "learning_rate": 5.0027609055770296e-05,
      "loss": 1.619,
      "num_input_tokens_seen": 24666128,
      "step": 4080,
      "train_runtime": 4136.5049,
      "train_tokens_per_second": 5963.036
    },
    {
      "epoch": 1.5009655172413794,
      "grad_norm": 0.9611397981643677,
      "learning_rate": 4.9990796981409905e-05,
      "loss": 1.4576,
      "num_input_tokens_seen": 24671568,
      "step": 4081,
      "train_runtime": 4137.916,
      "train_tokens_per_second": 5962.317
    },
    {
      "epoch": 1.5013333333333332,
      "grad_norm": 0.9430239200592041,
      "learning_rate": 4.995398490704952e-05,
      "loss": 1.3471,
      "num_input_tokens_seen": 24680624,
      "step": 4082,
      "train_runtime": 4139.2854,
      "train_tokens_per_second": 5962.533
    },
    {
      "epoch": 1.5017011494252874,
      "grad_norm": 1.0238491296768188,
      "learning_rate": 4.991717283268912e-05,
      "loss": 1.7808,
      "num_input_tokens_seen": 24686096,
      "step": 4083,
      "train_runtime": 4140.1816,
      "train_tokens_per_second": 5962.564
    },
    {
      "epoch": 1.5020689655172412,
      "grad_norm": 0.9407117366790771,
      "learning_rate": 4.988036075832874e-05,
      "loss": 1.3638,
      "num_input_tokens_seen": 24691264,
      "step": 4084,
      "train_runtime": 4141.0517,
      "train_tokens_per_second": 5962.559
    },
    {
      "epoch": 1.5024367816091955,
      "grad_norm": 0.9622142910957336,
      "learning_rate": 4.984354868396835e-05,
      "loss": 1.4942,
      "num_input_tokens_seen": 24695904,
      "step": 4085,
      "train_runtime": 4141.8608,
      "train_tokens_per_second": 5962.514
    },
    {
      "epoch": 1.5028045977011493,
      "grad_norm": 1.085375189781189,
      "learning_rate": 4.980673660960795e-05,
      "loss": 1.7551,
      "num_input_tokens_seen": 24701328,
      "step": 4086,
      "train_runtime": 4142.7468,
      "train_tokens_per_second": 5962.548
    },
    {
      "epoch": 1.5031724137931035,
      "grad_norm": 0.9670528173446655,
      "learning_rate": 4.9769924535247565e-05,
      "loss": 1.3634,
      "num_input_tokens_seen": 24709632,
      "step": 4087,
      "train_runtime": 4144.0106,
      "train_tokens_per_second": 5962.734
    },
    {
      "epoch": 1.5035402298850573,
      "grad_norm": 1.0086263418197632,
      "learning_rate": 4.9733112460887174e-05,
      "loss": 1.5774,
      "num_input_tokens_seen": 24715024,
      "step": 4088,
      "train_runtime": 4144.8817,
      "train_tokens_per_second": 5962.782
    },
    {
      "epoch": 1.5039080459770116,
      "grad_norm": 0.9485494494438171,
      "learning_rate": 4.969630038652678e-05,
      "loss": 1.5046,
      "num_input_tokens_seen": 24721136,
      "step": 4089,
      "train_runtime": 4145.8708,
      "train_tokens_per_second": 5962.833
    },
    {
      "epoch": 1.5042758620689654,
      "grad_norm": 0.8939463496208191,
      "learning_rate": 4.965948831216639e-05,
      "loss": 1.5576,
      "num_input_tokens_seen": 24726976,
      "step": 4090,
      "train_runtime": 4146.8093,
      "train_tokens_per_second": 5962.892
    },
    {
      "epoch": 1.5046436781609196,
      "grad_norm": 1.0472095012664795,
      "learning_rate": 4.962267623780601e-05,
      "loss": 1.4858,
      "num_input_tokens_seen": 24732192,
      "step": 4091,
      "train_runtime": 4147.6751,
      "train_tokens_per_second": 5962.905
    },
    {
      "epoch": 1.5050114942528736,
      "grad_norm": 0.9629560708999634,
      "learning_rate": 4.958586416344561e-05,
      "loss": 1.5479,
      "num_input_tokens_seen": 24737472,
      "step": 4092,
      "train_runtime": 4148.5413,
      "train_tokens_per_second": 5962.933
    },
    {
      "epoch": 1.5053793103448276,
      "grad_norm": 1.1243983507156372,
      "learning_rate": 4.9549052089085225e-05,
      "loss": 1.7423,
      "num_input_tokens_seen": 24741792,
      "step": 4093,
      "train_runtime": 4149.2838,
      "train_tokens_per_second": 5962.907
    },
    {
      "epoch": 1.5057471264367817,
      "grad_norm": 0.9103943705558777,
      "learning_rate": 4.9512240014724834e-05,
      "loss": 1.5661,
      "num_input_tokens_seen": 24746432,
      "step": 4094,
      "train_runtime": 4150.055,
      "train_tokens_per_second": 5962.917
    },
    {
      "epoch": 1.5061149425287357,
      "grad_norm": 1.1104499101638794,
      "learning_rate": 4.947542794036444e-05,
      "loss": 1.4933,
      "num_input_tokens_seen": 24751216,
      "step": 4095,
      "train_runtime": 4150.8667,
      "train_tokens_per_second": 5962.903
    },
    {
      "epoch": 1.5064827586206897,
      "grad_norm": 1.0149651765823364,
      "learning_rate": 4.943861586600405e-05,
      "loss": 1.5268,
      "num_input_tokens_seen": 24755840,
      "step": 4096,
      "train_runtime": 4151.6817,
      "train_tokens_per_second": 5962.846
    },
    {
      "epoch": 1.5068505747126437,
      "grad_norm": 0.89632648229599,
      "learning_rate": 4.940180379164366e-05,
      "loss": 1.5609,
      "num_input_tokens_seen": 24763888,
      "step": 4097,
      "train_runtime": 4152.9465,
      "train_tokens_per_second": 5962.968
    },
    {
      "epoch": 1.5072183908045977,
      "grad_norm": 0.9414806365966797,
      "learning_rate": 4.936499171728327e-05,
      "loss": 1.7155,
      "num_input_tokens_seen": 24769216,
      "step": 4098,
      "train_runtime": 4153.8057,
      "train_tokens_per_second": 5963.018
    },
    {
      "epoch": 1.5075862068965518,
      "grad_norm": 0.9200591444969177,
      "learning_rate": 4.9328179642922885e-05,
      "loss": 1.3421,
      "num_input_tokens_seen": 24775600,
      "step": 4099,
      "train_runtime": 4154.8403,
      "train_tokens_per_second": 5963.069
    },
    {
      "epoch": 1.5079540229885058,
      "grad_norm": 1.033595323562622,
      "learning_rate": 4.9291367568562494e-05,
      "loss": 1.4535,
      "num_input_tokens_seen": 24782208,
      "step": 4100,
      "train_runtime": 4155.9004,
      "train_tokens_per_second": 5963.138
    },
    {
      "epoch": 1.5083218390804598,
      "grad_norm": 0.9664585590362549,
      "learning_rate": 4.9254555494202096e-05,
      "loss": 1.4396,
      "num_input_tokens_seen": 24789632,
      "step": 4101,
      "train_runtime": 4157.0715,
      "train_tokens_per_second": 5963.244
    },
    {
      "epoch": 1.5086896551724138,
      "grad_norm": 0.9104776978492737,
      "learning_rate": 4.921774341984171e-05,
      "loss": 1.4124,
      "num_input_tokens_seen": 24797296,
      "step": 4102,
      "train_runtime": 4158.2672,
      "train_tokens_per_second": 5963.372
    },
    {
      "epoch": 1.5090574712643678,
      "grad_norm": 0.8871594071388245,
      "learning_rate": 4.918093134548132e-05,
      "loss": 1.4965,
      "num_input_tokens_seen": 24804656,
      "step": 4103,
      "train_runtime": 4159.4018,
      "train_tokens_per_second": 5963.515
    },
    {
      "epoch": 1.5094252873563219,
      "grad_norm": 0.889460027217865,
      "learning_rate": 4.914411927112093e-05,
      "loss": 1.4599,
      "num_input_tokens_seen": 24811328,
      "step": 4104,
      "train_runtime": 4160.4737,
      "train_tokens_per_second": 5963.582
    },
    {
      "epoch": 1.5097931034482759,
      "grad_norm": 1.0685147047042847,
      "learning_rate": 4.910730719676054e-05,
      "loss": 1.6426,
      "num_input_tokens_seen": 24818272,
      "step": 4105,
      "train_runtime": 4161.5735,
      "train_tokens_per_second": 5963.675
    },
    {
      "epoch": 1.51016091954023,
      "grad_norm": 1.033657431602478,
      "learning_rate": 4.9070495122400154e-05,
      "loss": 1.6988,
      "num_input_tokens_seen": 24823200,
      "step": 4106,
      "train_runtime": 4162.4215,
      "train_tokens_per_second": 5963.644
    },
    {
      "epoch": 1.510528735632184,
      "grad_norm": 1.0012242794036865,
      "learning_rate": 4.9033683048039756e-05,
      "loss": 1.6408,
      "num_input_tokens_seen": 24829312,
      "step": 4107,
      "train_runtime": 4163.4148,
      "train_tokens_per_second": 5963.689
    },
    {
      "epoch": 1.510896551724138,
      "grad_norm": 1.0517055988311768,
      "learning_rate": 4.899687097367937e-05,
      "loss": 1.7441,
      "num_input_tokens_seen": 24834848,
      "step": 4108,
      "train_runtime": 4164.3306,
      "train_tokens_per_second": 5963.707
    },
    {
      "epoch": 1.511264367816092,
      "grad_norm": 1.0701045989990234,
      "learning_rate": 4.896005889931898e-05,
      "loss": 1.5684,
      "num_input_tokens_seen": 24840112,
      "step": 4109,
      "train_runtime": 4165.1928,
      "train_tokens_per_second": 5963.736
    },
    {
      "epoch": 1.511632183908046,
      "grad_norm": 0.8956212401390076,
      "learning_rate": 4.892324682495859e-05,
      "loss": 1.3929,
      "num_input_tokens_seen": 24846560,
      "step": 4110,
      "train_runtime": 4166.2165,
      "train_tokens_per_second": 5963.819
    },
    {
      "epoch": 1.512,
      "grad_norm": 0.8714888095855713,
      "learning_rate": 4.88864347505982e-05,
      "loss": 1.1294,
      "num_input_tokens_seen": 24856336,
      "step": 4111,
      "train_runtime": 4168.1964,
      "train_tokens_per_second": 5963.331
    },
    {
      "epoch": 1.512367816091954,
      "grad_norm": 0.9745454788208008,
      "learning_rate": 4.884962267623781e-05,
      "loss": 1.5583,
      "num_input_tokens_seen": 24860560,
      "step": 4112,
      "train_runtime": 4168.9591,
      "train_tokens_per_second": 5963.254
    },
    {
      "epoch": 1.512735632183908,
      "grad_norm": 0.9520863890647888,
      "learning_rate": 4.8812810601877416e-05,
      "loss": 1.3723,
      "num_input_tokens_seen": 24868544,
      "step": 4113,
      "train_runtime": 4170.1785,
      "train_tokens_per_second": 5963.424
    },
    {
      "epoch": 1.513103448275862,
      "grad_norm": 0.9671286940574646,
      "learning_rate": 4.877599852751703e-05,
      "loss": 1.6032,
      "num_input_tokens_seen": 24873488,
      "step": 4114,
      "train_runtime": 4171.0224,
      "train_tokens_per_second": 5963.403
    },
    {
      "epoch": 1.513471264367816,
      "grad_norm": 0.8513003587722778,
      "learning_rate": 4.873918645315664e-05,
      "loss": 1.5025,
      "num_input_tokens_seen": 24882528,
      "step": 4115,
      "train_runtime": 4172.3907,
      "train_tokens_per_second": 5963.614
    },
    {
      "epoch": 1.5138390804597701,
      "grad_norm": 1.0138033628463745,
      "learning_rate": 4.870237437879624e-05,
      "loss": 1.4907,
      "num_input_tokens_seen": 24887248,
      "step": 4116,
      "train_runtime": 4173.2136,
      "train_tokens_per_second": 5963.569
    },
    {
      "epoch": 1.5142068965517241,
      "grad_norm": 0.950100302696228,
      "learning_rate": 4.866556230443586e-05,
      "loss": 1.6034,
      "num_input_tokens_seen": 24892720,
      "step": 4117,
      "train_runtime": 4174.104,
      "train_tokens_per_second": 5963.608
    },
    {
      "epoch": 1.5145747126436782,
      "grad_norm": 1.117335557937622,
      "learning_rate": 4.862875023007547e-05,
      "loss": 1.688,
      "num_input_tokens_seen": 24897776,
      "step": 4118,
      "train_runtime": 4174.9498,
      "train_tokens_per_second": 5963.611
    },
    {
      "epoch": 1.5149425287356322,
      "grad_norm": 1.0012331008911133,
      "learning_rate": 4.8591938155715076e-05,
      "loss": 1.5859,
      "num_input_tokens_seen": 24902304,
      "step": 4119,
      "train_runtime": 4175.7303,
      "train_tokens_per_second": 5963.581
    },
    {
      "epoch": 1.5153103448275862,
      "grad_norm": 0.9911288022994995,
      "learning_rate": 4.8555126081354685e-05,
      "loss": 1.7102,
      "num_input_tokens_seen": 24907216,
      "step": 4120,
      "train_runtime": 4176.5505,
      "train_tokens_per_second": 5963.586
    },
    {
      "epoch": 1.5156781609195402,
      "grad_norm": 1.0760222673416138,
      "learning_rate": 4.85183140069943e-05,
      "loss": 1.6123,
      "num_input_tokens_seen": 24912816,
      "step": 4121,
      "train_runtime": 4177.4426,
      "train_tokens_per_second": 5963.653
    },
    {
      "epoch": 1.5160459770114942,
      "grad_norm": 0.9216646552085876,
      "learning_rate": 4.84815019326339e-05,
      "loss": 1.6326,
      "num_input_tokens_seen": 24920096,
      "step": 4122,
      "train_runtime": 4178.5865,
      "train_tokens_per_second": 5963.762
    },
    {
      "epoch": 1.5164137931034483,
      "grad_norm": 0.9923339486122131,
      "learning_rate": 4.844468985827352e-05,
      "loss": 1.4925,
      "num_input_tokens_seen": 24924960,
      "step": 4123,
      "train_runtime": 4179.4155,
      "train_tokens_per_second": 5963.743
    },
    {
      "epoch": 1.5167816091954023,
      "grad_norm": 1.0101779699325562,
      "learning_rate": 4.840787778391313e-05,
      "loss": 1.5584,
      "num_input_tokens_seen": 24929744,
      "step": 4124,
      "train_runtime": 4180.2393,
      "train_tokens_per_second": 5963.712
    },
    {
      "epoch": 1.5171494252873563,
      "grad_norm": 1.010554552078247,
      "learning_rate": 4.8371065709552737e-05,
      "loss": 1.5622,
      "num_input_tokens_seen": 24935744,
      "step": 4125,
      "train_runtime": 4181.2202,
      "train_tokens_per_second": 5963.748
    },
    {
      "epoch": 1.5175172413793103,
      "grad_norm": 0.9562215209007263,
      "learning_rate": 4.8334253635192345e-05,
      "loss": 1.5257,
      "num_input_tokens_seen": 24941504,
      "step": 4126,
      "train_runtime": 4182.1862,
      "train_tokens_per_second": 5963.748
    },
    {
      "epoch": 1.5178850574712643,
      "grad_norm": 0.8532366156578064,
      "learning_rate": 4.8297441560831954e-05,
      "loss": 1.3883,
      "num_input_tokens_seen": 24948640,
      "step": 4127,
      "train_runtime": 4183.3082,
      "train_tokens_per_second": 5963.854
    },
    {
      "epoch": 1.5182528735632184,
      "grad_norm": 0.9355764389038086,
      "learning_rate": 4.826062948647156e-05,
      "loss": 1.4522,
      "num_input_tokens_seen": 24955776,
      "step": 4128,
      "train_runtime": 4184.4569,
      "train_tokens_per_second": 5963.922
    },
    {
      "epoch": 1.5186206896551724,
      "grad_norm": 0.9794521331787109,
      "learning_rate": 4.822381741211118e-05,
      "loss": 1.5783,
      "num_input_tokens_seen": 24963456,
      "step": 4129,
      "train_runtime": 4185.6409,
      "train_tokens_per_second": 5964.07
    },
    {
      "epoch": 1.5189885057471264,
      "grad_norm": 0.9304761290550232,
      "learning_rate": 4.818700533775079e-05,
      "loss": 1.3663,
      "num_input_tokens_seen": 24969200,
      "step": 4130,
      "train_runtime": 4186.5793,
      "train_tokens_per_second": 5964.105
    },
    {
      "epoch": 1.5193563218390804,
      "grad_norm": 1.066969394683838,
      "learning_rate": 4.815019326339039e-05,
      "loss": 1.4422,
      "num_input_tokens_seen": 24973392,
      "step": 4131,
      "train_runtime": 4187.3061,
      "train_tokens_per_second": 5964.071
    },
    {
      "epoch": 1.5197241379310344,
      "grad_norm": 1.0843766927719116,
      "learning_rate": 4.8113381189030006e-05,
      "loss": 1.6431,
      "num_input_tokens_seen": 24977712,
      "step": 4132,
      "train_runtime": 4188.0839,
      "train_tokens_per_second": 5963.995
    },
    {
      "epoch": 1.5200919540229885,
      "grad_norm": 1.0205549001693726,
      "learning_rate": 4.8076569114669614e-05,
      "loss": 1.6441,
      "num_input_tokens_seen": 24984640,
      "step": 4133,
      "train_runtime": 4189.1809,
      "train_tokens_per_second": 5964.087
    },
    {
      "epoch": 1.5204597701149425,
      "grad_norm": 0.9489389657974243,
      "learning_rate": 4.803975704030922e-05,
      "loss": 1.5125,
      "num_input_tokens_seen": 24991376,
      "step": 4134,
      "train_runtime": 4190.2574,
      "train_tokens_per_second": 5964.163
    },
    {
      "epoch": 1.5208275862068965,
      "grad_norm": 1.0028316974639893,
      "learning_rate": 4.800294496594883e-05,
      "loss": 1.4758,
      "num_input_tokens_seen": 24998704,
      "step": 4135,
      "train_runtime": 4191.4033,
      "train_tokens_per_second": 5964.28
    },
    {
      "epoch": 1.5211954022988505,
      "grad_norm": 1.1025384664535522,
      "learning_rate": 4.796613289158844e-05,
      "loss": 1.4589,
      "num_input_tokens_seen": 25005824,
      "step": 4136,
      "train_runtime": 4192.5139,
      "train_tokens_per_second": 5964.399
    },
    {
      "epoch": 1.5215632183908046,
      "grad_norm": 1.0178321599960327,
      "learning_rate": 4.792932081722805e-05,
      "loss": 1.6515,
      "num_input_tokens_seen": 25012192,
      "step": 4137,
      "train_runtime": 4193.5443,
      "train_tokens_per_second": 5964.452
    },
    {
      "epoch": 1.5219310344827586,
      "grad_norm": 0.9515804052352905,
      "learning_rate": 4.7892508742867666e-05,
      "loss": 1.4498,
      "num_input_tokens_seen": 25016480,
      "step": 4138,
      "train_runtime": 4194.3014,
      "train_tokens_per_second": 5964.397
    },
    {
      "epoch": 1.5222988505747126,
      "grad_norm": 0.8750317692756653,
      "learning_rate": 4.7855696668507275e-05,
      "loss": 1.4558,
      "num_input_tokens_seen": 25024096,
      "step": 4139,
      "train_runtime": 4195.5012,
      "train_tokens_per_second": 5964.507
    },
    {
      "epoch": 1.5226666666666666,
      "grad_norm": 1.0486810207366943,
      "learning_rate": 4.7818884594146883e-05,
      "loss": 1.5602,
      "num_input_tokens_seen": 25030112,
      "step": 4140,
      "train_runtime": 4196.4868,
      "train_tokens_per_second": 5964.54
    },
    {
      "epoch": 1.5230344827586206,
      "grad_norm": 1.0021171569824219,
      "learning_rate": 4.778207251978649e-05,
      "loss": 1.6507,
      "num_input_tokens_seen": 25034240,
      "step": 4141,
      "train_runtime": 4197.825,
      "train_tokens_per_second": 5963.622
    },
    {
      "epoch": 1.5234022988505749,
      "grad_norm": 0.9792951345443726,
      "learning_rate": 4.77452604454261e-05,
      "loss": 1.6869,
      "num_input_tokens_seen": 25040016,
      "step": 4142,
      "train_runtime": 4198.7976,
      "train_tokens_per_second": 5963.616
    },
    {
      "epoch": 1.5237701149425287,
      "grad_norm": 1.01872980594635,
      "learning_rate": 4.770844837106571e-05,
      "loss": 1.5767,
      "num_input_tokens_seen": 25045520,
      "step": 4143,
      "train_runtime": 4199.7403,
      "train_tokens_per_second": 5963.588
    },
    {
      "epoch": 1.524137931034483,
      "grad_norm": 0.9187114238739014,
      "learning_rate": 4.7671636296705326e-05,
      "loss": 1.4714,
      "num_input_tokens_seen": 25050400,
      "step": 4144,
      "train_runtime": 4200.5766,
      "train_tokens_per_second": 5963.562
    },
    {
      "epoch": 1.5245057471264367,
      "grad_norm": 0.9544577598571777,
      "learning_rate": 4.7634824222344935e-05,
      "loss": 1.4923,
      "num_input_tokens_seen": 25057056,
      "step": 4145,
      "train_runtime": 4201.6359,
      "train_tokens_per_second": 5963.643
    },
    {
      "epoch": 1.524873563218391,
      "grad_norm": 0.9472795724868774,
      "learning_rate": 4.759801214798454e-05,
      "loss": 1.5866,
      "num_input_tokens_seen": 25061824,
      "step": 4146,
      "train_runtime": 4202.4383,
      "train_tokens_per_second": 5963.639
    },
    {
      "epoch": 1.5252413793103448,
      "grad_norm": 1.027949571609497,
      "learning_rate": 4.756120007362415e-05,
      "loss": 1.4579,
      "num_input_tokens_seen": 25068048,
      "step": 4147,
      "train_runtime": 4203.4057,
      "train_tokens_per_second": 5963.747
    },
    {
      "epoch": 1.525609195402299,
      "grad_norm": 0.9217515587806702,
      "learning_rate": 4.752438799926376e-05,
      "loss": 1.473,
      "num_input_tokens_seen": 25073872,
      "step": 4148,
      "train_runtime": 4204.349,
      "train_tokens_per_second": 5963.794
    },
    {
      "epoch": 1.5259770114942528,
      "grad_norm": 1.0180093050003052,
      "learning_rate": 4.748757592490337e-05,
      "loss": 1.6773,
      "num_input_tokens_seen": 25080192,
      "step": 4149,
      "train_runtime": 4205.356,
      "train_tokens_per_second": 5963.869
    },
    {
      "epoch": 1.526344827586207,
      "grad_norm": 0.9596949219703674,
      "learning_rate": 4.745076385054298e-05,
      "loss": 1.456,
      "num_input_tokens_seen": 25086128,
      "step": 4150,
      "train_runtime": 4206.3394,
      "train_tokens_per_second": 5963.886
    },
    {
      "epoch": 1.5267126436781608,
      "grad_norm": 0.9876999855041504,
      "learning_rate": 4.741395177618259e-05,
      "loss": 1.6332,
      "num_input_tokens_seen": 25090896,
      "step": 4151,
      "train_runtime": 4207.1435,
      "train_tokens_per_second": 5963.879
    },
    {
      "epoch": 1.527080459770115,
      "grad_norm": 1.0236774682998657,
      "learning_rate": 4.73771397018222e-05,
      "loss": 1.6292,
      "num_input_tokens_seen": 25095936,
      "step": 4152,
      "train_runtime": 4207.9994,
      "train_tokens_per_second": 5963.864
    },
    {
      "epoch": 1.5274482758620689,
      "grad_norm": 0.8478060960769653,
      "learning_rate": 4.734032762746181e-05,
      "loss": 1.4633,
      "num_input_tokens_seen": 25102512,
      "step": 4153,
      "train_runtime": 4209.0723,
      "train_tokens_per_second": 5963.906
    },
    {
      "epoch": 1.5278160919540231,
      "grad_norm": 0.9411160945892334,
      "learning_rate": 4.730351555310142e-05,
      "loss": 1.4515,
      "num_input_tokens_seen": 25109568,
      "step": 4154,
      "train_runtime": 4210.1801,
      "train_tokens_per_second": 5964.013
    },
    {
      "epoch": 1.528183908045977,
      "grad_norm": 1.0756803750991821,
      "learning_rate": 4.726670347874103e-05,
      "loss": 1.7328,
      "num_input_tokens_seen": 25115968,
      "step": 4155,
      "train_runtime": 4211.1894,
      "train_tokens_per_second": 5964.103
    },
    {
      "epoch": 1.5285517241379312,
      "grad_norm": 1.0040136575698853,
      "learning_rate": 4.722989140438064e-05,
      "loss": 1.6318,
      "num_input_tokens_seen": 25123168,
      "step": 4156,
      "train_runtime": 4212.3299,
      "train_tokens_per_second": 5964.198
    },
    {
      "epoch": 1.528919540229885,
      "grad_norm": 0.9725233912467957,
      "learning_rate": 4.719307933002025e-05,
      "loss": 1.5209,
      "num_input_tokens_seen": 25127792,
      "step": 4157,
      "train_runtime": 4213.1147,
      "train_tokens_per_second": 5964.184
    },
    {
      "epoch": 1.5292873563218392,
      "grad_norm": 0.8511664271354675,
      "learning_rate": 4.715626725565986e-05,
      "loss": 1.3947,
      "num_input_tokens_seen": 25135232,
      "step": 4158,
      "train_runtime": 4214.2871,
      "train_tokens_per_second": 5964.29
    },
    {
      "epoch": 1.529655172413793,
      "grad_norm": 1.1127713918685913,
      "learning_rate": 4.711945518129947e-05,
      "loss": 1.6247,
      "num_input_tokens_seen": 25141536,
      "step": 4159,
      "train_runtime": 4215.2878,
      "train_tokens_per_second": 5964.37
    },
    {
      "epoch": 1.5300229885057473,
      "grad_norm": 0.9434221982955933,
      "learning_rate": 4.708264310693908e-05,
      "loss": 1.5217,
      "num_input_tokens_seen": 25147536,
      "step": 4160,
      "train_runtime": 4216.2439,
      "train_tokens_per_second": 5964.441
    },
    {
      "epoch": 1.530390804597701,
      "grad_norm": 1.0339033603668213,
      "learning_rate": 4.7045831032578684e-05,
      "loss": 1.4969,
      "num_input_tokens_seen": 25152688,
      "step": 4161,
      "train_runtime": 4217.0872,
      "train_tokens_per_second": 5964.469
    },
    {
      "epoch": 1.5307586206896553,
      "grad_norm": 1.085526704788208,
      "learning_rate": 4.70090189582183e-05,
      "loss": 1.4692,
      "num_input_tokens_seen": 25157856,
      "step": 4162,
      "train_runtime": 4217.9258,
      "train_tokens_per_second": 5964.509
    },
    {
      "epoch": 1.531126436781609,
      "grad_norm": 1.0283817052841187,
      "learning_rate": 4.697220688385791e-05,
      "loss": 1.4176,
      "num_input_tokens_seen": 25163376,
      "step": 4163,
      "train_runtime": 4218.8127,
      "train_tokens_per_second": 5964.563
    },
    {
      "epoch": 1.5314942528735633,
      "grad_norm": 1.0186675786972046,
      "learning_rate": 4.693539480949752e-05,
      "loss": 1.5648,
      "num_input_tokens_seen": 25168752,
      "step": 4164,
      "train_runtime": 4219.7008,
      "train_tokens_per_second": 5964.582
    },
    {
      "epoch": 1.5318620689655171,
      "grad_norm": 0.9804777503013611,
      "learning_rate": 4.6898582735137126e-05,
      "loss": 1.5878,
      "num_input_tokens_seen": 25174736,
      "step": 4165,
      "train_runtime": 4220.6685,
      "train_tokens_per_second": 5964.632
    },
    {
      "epoch": 1.5322298850574714,
      "grad_norm": 0.9473086595535278,
      "learning_rate": 4.6861770660776735e-05,
      "loss": 1.6777,
      "num_input_tokens_seen": 25179776,
      "step": 4166,
      "train_runtime": 4221.5182,
      "train_tokens_per_second": 5964.626
    },
    {
      "epoch": 1.5325977011494252,
      "grad_norm": 0.9539825320243835,
      "learning_rate": 4.6824958586416344e-05,
      "loss": 1.2358,
      "num_input_tokens_seen": 25185168,
      "step": 4167,
      "train_runtime": 4222.4028,
      "train_tokens_per_second": 5964.653
    },
    {
      "epoch": 1.5329655172413794,
      "grad_norm": 0.9325450658798218,
      "learning_rate": 4.678814651205596e-05,
      "loss": 1.4247,
      "num_input_tokens_seen": 25189824,
      "step": 4168,
      "train_runtime": 4223.1948,
      "train_tokens_per_second": 5964.637
    },
    {
      "epoch": 1.5333333333333332,
      "grad_norm": 1.0673909187316895,
      "learning_rate": 4.675133443769557e-05,
      "loss": 1.3801,
      "num_input_tokens_seen": 25197040,
      "step": 4169,
      "train_runtime": 4224.3212,
      "train_tokens_per_second": 5964.755
    },
    {
      "epoch": 1.5337011494252875,
      "grad_norm": 1.0286864042282104,
      "learning_rate": 4.671452236333518e-05,
      "loss": 1.4338,
      "num_input_tokens_seen": 25202352,
      "step": 4170,
      "train_runtime": 4225.2072,
      "train_tokens_per_second": 5964.761
    },
    {
      "epoch": 1.5340689655172413,
      "grad_norm": 0.949355959892273,
      "learning_rate": 4.6677710288974786e-05,
      "loss": 1.4864,
      "num_input_tokens_seen": 25207040,
      "step": 4171,
      "train_runtime": 4226.5611,
      "train_tokens_per_second": 5963.96
    },
    {
      "epoch": 1.5344367816091955,
      "grad_norm": 1.077665090560913,
      "learning_rate": 4.6640898214614395e-05,
      "loss": 1.5227,
      "num_input_tokens_seen": 25213952,
      "step": 4172,
      "train_runtime": 4227.6524,
      "train_tokens_per_second": 5964.055
    },
    {
      "epoch": 1.5348045977011493,
      "grad_norm": 0.9799932241439819,
      "learning_rate": 4.6604086140254004e-05,
      "loss": 1.5721,
      "num_input_tokens_seen": 25219760,
      "step": 4173,
      "train_runtime": 4228.6,
      "train_tokens_per_second": 5964.092
    },
    {
      "epoch": 1.5351724137931035,
      "grad_norm": 1.0068010091781616,
      "learning_rate": 4.656727406589362e-05,
      "loss": 1.4747,
      "num_input_tokens_seen": 25226000,
      "step": 4174,
      "train_runtime": 4229.6066,
      "train_tokens_per_second": 5964.148
    },
    {
      "epoch": 1.5355402298850573,
      "grad_norm": 0.9431117177009583,
      "learning_rate": 4.653046199153322e-05,
      "loss": 1.3369,
      "num_input_tokens_seen": 25235328,
      "step": 4175,
      "train_runtime": 4231.0229,
      "train_tokens_per_second": 5964.356
    },
    {
      "epoch": 1.5359080459770116,
      "grad_norm": 0.9573741555213928,
      "learning_rate": 4.649364991717283e-05,
      "loss": 1.3994,
      "num_input_tokens_seen": 25241616,
      "step": 4176,
      "train_runtime": 4232.0425,
      "train_tokens_per_second": 5964.405
    },
    {
      "epoch": 1.5362758620689654,
      "grad_norm": 0.9425127506256104,
      "learning_rate": 4.6456837842812446e-05,
      "loss": 1.601,
      "num_input_tokens_seen": 25246608,
      "step": 4177,
      "train_runtime": 4232.8678,
      "train_tokens_per_second": 5964.422
    },
    {
      "epoch": 1.5366436781609196,
      "grad_norm": 0.9227352738380432,
      "learning_rate": 4.6420025768452055e-05,
      "loss": 1.5689,
      "num_input_tokens_seen": 25252000,
      "step": 4178,
      "train_runtime": 4233.7557,
      "train_tokens_per_second": 5964.444
    },
    {
      "epoch": 1.5370114942528734,
      "grad_norm": 1.0466983318328857,
      "learning_rate": 4.6383213694091664e-05,
      "loss": 1.6838,
      "num_input_tokens_seen": 25257392,
      "step": 4179,
      "train_runtime": 4234.633,
      "train_tokens_per_second": 5964.482
    },
    {
      "epoch": 1.5373793103448277,
      "grad_norm": 1.0222901105880737,
      "learning_rate": 4.634640161973127e-05,
      "loss": 1.5077,
      "num_input_tokens_seen": 25263824,
      "step": 4180,
      "train_runtime": 4235.6714,
      "train_tokens_per_second": 5964.538
    },
    {
      "epoch": 1.5377471264367815,
      "grad_norm": 0.9365169405937195,
      "learning_rate": 4.630958954537088e-05,
      "loss": 1.3421,
      "num_input_tokens_seen": 25268720,
      "step": 4181,
      "train_runtime": 4236.5145,
      "train_tokens_per_second": 5964.507
    },
    {
      "epoch": 1.5381149425287357,
      "grad_norm": 1.005853295326233,
      "learning_rate": 4.627277747101049e-05,
      "loss": 1.4659,
      "num_input_tokens_seen": 25274592,
      "step": 4182,
      "train_runtime": 4237.4889,
      "train_tokens_per_second": 5964.521
    },
    {
      "epoch": 1.5384827586206895,
      "grad_norm": 0.9372466206550598,
      "learning_rate": 4.6235965396650106e-05,
      "loss": 1.3908,
      "num_input_tokens_seen": 25282816,
      "step": 4183,
      "train_runtime": 4238.7576,
      "train_tokens_per_second": 5964.676
    },
    {
      "epoch": 1.5388505747126437,
      "grad_norm": 0.9906845092773438,
      "learning_rate": 4.6199153322289715e-05,
      "loss": 1.6273,
      "num_input_tokens_seen": 25287360,
      "step": 4184,
      "train_runtime": 4239.5462,
      "train_tokens_per_second": 5964.638
    },
    {
      "epoch": 1.5392183908045975,
      "grad_norm": 0.9468985795974731,
      "learning_rate": 4.6162341247929324e-05,
      "loss": 1.603,
      "num_input_tokens_seen": 25292432,
      "step": 4185,
      "train_runtime": 4240.4134,
      "train_tokens_per_second": 5964.615
    },
    {
      "epoch": 1.5395862068965518,
      "grad_norm": 0.9000393152236938,
      "learning_rate": 4.612552917356893e-05,
      "loss": 1.3251,
      "num_input_tokens_seen": 25299840,
      "step": 4186,
      "train_runtime": 4241.5485,
      "train_tokens_per_second": 5964.765
    },
    {
      "epoch": 1.5399540229885056,
      "grad_norm": 1.0468236207962036,
      "learning_rate": 4.608871709920854e-05,
      "loss": 1.5165,
      "num_input_tokens_seen": 25306208,
      "step": 4187,
      "train_runtime": 4242.5796,
      "train_tokens_per_second": 5964.816
    },
    {
      "epoch": 1.5403218390804598,
      "grad_norm": 0.9615391492843628,
      "learning_rate": 4.605190502484815e-05,
      "loss": 1.4022,
      "num_input_tokens_seen": 25313952,
      "step": 4188,
      "train_runtime": 4243.7995,
      "train_tokens_per_second": 5964.926
    },
    {
      "epoch": 1.5406896551724136,
      "grad_norm": 0.9920722842216492,
      "learning_rate": 4.601509295048777e-05,
      "loss": 1.3898,
      "num_input_tokens_seen": 25321264,
      "step": 4189,
      "train_runtime": 4244.9596,
      "train_tokens_per_second": 5965.019
    },
    {
      "epoch": 1.5410574712643679,
      "grad_norm": 1.0014163255691528,
      "learning_rate": 4.597828087612737e-05,
      "loss": 1.6046,
      "num_input_tokens_seen": 25326144,
      "step": 4190,
      "train_runtime": 4245.7742,
      "train_tokens_per_second": 5965.024
    },
    {
      "epoch": 1.541425287356322,
      "grad_norm": 1.0710721015930176,
      "learning_rate": 4.594146880176698e-05,
      "loss": 1.6376,
      "num_input_tokens_seen": 25331040,
      "step": 4191,
      "train_runtime": 4246.5845,
      "train_tokens_per_second": 5965.038
    },
    {
      "epoch": 1.541793103448276,
      "grad_norm": 0.9985161423683167,
      "learning_rate": 4.590465672740659e-05,
      "loss": 1.4707,
      "num_input_tokens_seen": 25336752,
      "step": 4192,
      "train_runtime": 4247.544,
      "train_tokens_per_second": 5965.036
    },
    {
      "epoch": 1.54216091954023,
      "grad_norm": 0.871497631072998,
      "learning_rate": 4.58678446530462e-05,
      "loss": 1.327,
      "num_input_tokens_seen": 25344096,
      "step": 4193,
      "train_runtime": 4248.7113,
      "train_tokens_per_second": 5965.125
    },
    {
      "epoch": 1.542528735632184,
      "grad_norm": 1.0417569875717163,
      "learning_rate": 4.583103257868581e-05,
      "loss": 1.8051,
      "num_input_tokens_seen": 25349008,
      "step": 4194,
      "train_runtime": 4249.5687,
      "train_tokens_per_second": 5965.078
    },
    {
      "epoch": 1.542896551724138,
      "grad_norm": 0.8819166421890259,
      "learning_rate": 4.579422050432542e-05,
      "loss": 1.3929,
      "num_input_tokens_seen": 25354656,
      "step": 4195,
      "train_runtime": 4250.4892,
      "train_tokens_per_second": 5965.115
    },
    {
      "epoch": 1.543264367816092,
      "grad_norm": 0.9865065217018127,
      "learning_rate": 4.575740842996503e-05,
      "loss": 1.5959,
      "num_input_tokens_seen": 25359984,
      "step": 4196,
      "train_runtime": 4251.367,
      "train_tokens_per_second": 5965.136
    },
    {
      "epoch": 1.543632183908046,
      "grad_norm": 1.061691403388977,
      "learning_rate": 4.572059635560464e-05,
      "loss": 1.4419,
      "num_input_tokens_seen": 25365984,
      "step": 4197,
      "train_runtime": 4252.3376,
      "train_tokens_per_second": 5965.186
    },
    {
      "epoch": 1.544,
      "grad_norm": 0.995140016078949,
      "learning_rate": 4.5683784281244253e-05,
      "loss": 1.692,
      "num_input_tokens_seen": 25371152,
      "step": 4198,
      "train_runtime": 4253.1952,
      "train_tokens_per_second": 5965.198
    },
    {
      "epoch": 1.544367816091954,
      "grad_norm": 1.0438783168792725,
      "learning_rate": 4.564697220688386e-05,
      "loss": 1.4333,
      "num_input_tokens_seen": 25376464,
      "step": 4199,
      "train_runtime": 4254.0914,
      "train_tokens_per_second": 5965.19
    },
    {
      "epoch": 1.544735632183908,
      "grad_norm": 1.0793952941894531,
      "learning_rate": 4.561016013252347e-05,
      "loss": 1.6111,
      "num_input_tokens_seen": 25381264,
      "step": 4200,
      "train_runtime": 4254.9234,
      "train_tokens_per_second": 5965.152
    },
    {
      "epoch": 1.545103448275862,
      "grad_norm": 0.8983263969421387,
      "learning_rate": 4.557334805816308e-05,
      "loss": 1.3303,
      "num_input_tokens_seen": 25388560,
      "step": 4201,
      "train_runtime": 4256.6089,
      "train_tokens_per_second": 5964.504
    },
    {
      "epoch": 1.5454712643678161,
      "grad_norm": 1.0643035173416138,
      "learning_rate": 4.553653598380269e-05,
      "loss": 1.8699,
      "num_input_tokens_seen": 25393488,
      "step": 4202,
      "train_runtime": 4257.4344,
      "train_tokens_per_second": 5964.505
    },
    {
      "epoch": 1.5458390804597701,
      "grad_norm": 1.009087324142456,
      "learning_rate": 4.54997239094423e-05,
      "loss": 1.4647,
      "num_input_tokens_seen": 25398352,
      "step": 4203,
      "train_runtime": 4258.266,
      "train_tokens_per_second": 5964.482
    },
    {
      "epoch": 1.5462068965517242,
      "grad_norm": 1.0162410736083984,
      "learning_rate": 4.5462911835081914e-05,
      "loss": 1.5835,
      "num_input_tokens_seen": 25402864,
      "step": 4204,
      "train_runtime": 4259.0376,
      "train_tokens_per_second": 5964.461
    },
    {
      "epoch": 1.5465747126436782,
      "grad_norm": 0.8939027786254883,
      "learning_rate": 4.5426099760721516e-05,
      "loss": 1.4868,
      "num_input_tokens_seen": 25409664,
      "step": 4205,
      "train_runtime": 4260.1181,
      "train_tokens_per_second": 5964.544
    },
    {
      "epoch": 1.5469425287356322,
      "grad_norm": 0.9433472752571106,
      "learning_rate": 4.5389287686361125e-05,
      "loss": 1.4043,
      "num_input_tokens_seen": 25415472,
      "step": 4206,
      "train_runtime": 4261.0676,
      "train_tokens_per_second": 5964.578
    },
    {
      "epoch": 1.5473103448275862,
      "grad_norm": 1.017948865890503,
      "learning_rate": 4.535247561200074e-05,
      "loss": 1.6007,
      "num_input_tokens_seen": 25422096,
      "step": 4207,
      "train_runtime": 4262.1359,
      "train_tokens_per_second": 5964.638
    },
    {
      "epoch": 1.5476781609195402,
      "grad_norm": 1.0156141519546509,
      "learning_rate": 4.531566353764035e-05,
      "loss": 1.4328,
      "num_input_tokens_seen": 25429008,
      "step": 4208,
      "train_runtime": 4263.2514,
      "train_tokens_per_second": 5964.698
    },
    {
      "epoch": 1.5480459770114943,
      "grad_norm": 1.0091875791549683,
      "learning_rate": 4.527885146327996e-05,
      "loss": 1.5701,
      "num_input_tokens_seen": 25433552,
      "step": 4209,
      "train_runtime": 4264.0407,
      "train_tokens_per_second": 5964.66
    },
    {
      "epoch": 1.5484137931034483,
      "grad_norm": 1.064543604850769,
      "learning_rate": 4.524203938891957e-05,
      "loss": 1.5772,
      "num_input_tokens_seen": 25438416,
      "step": 4210,
      "train_runtime": 4264.8918,
      "train_tokens_per_second": 5964.61
    },
    {
      "epoch": 1.5487816091954023,
      "grad_norm": 0.87115079164505,
      "learning_rate": 4.5205227314559176e-05,
      "loss": 1.4506,
      "num_input_tokens_seen": 25445584,
      "step": 4211,
      "train_runtime": 4266.028,
      "train_tokens_per_second": 5964.702
    },
    {
      "epoch": 1.5491494252873563,
      "grad_norm": 1.1344107389450073,
      "learning_rate": 4.5168415240198785e-05,
      "loss": 1.6297,
      "num_input_tokens_seen": 25453632,
      "step": 4212,
      "train_runtime": 4267.2692,
      "train_tokens_per_second": 5964.853
    },
    {
      "epoch": 1.5495172413793104,
      "grad_norm": 1.004704475402832,
      "learning_rate": 4.51316031658384e-05,
      "loss": 1.598,
      "num_input_tokens_seen": 25459328,
      "step": 4213,
      "train_runtime": 4268.207,
      "train_tokens_per_second": 5964.877
    },
    {
      "epoch": 1.5498850574712644,
      "grad_norm": 0.9495462775230408,
      "learning_rate": 4.5094791091478e-05,
      "loss": 1.4677,
      "num_input_tokens_seen": 25466512,
      "step": 4214,
      "train_runtime": 4269.3573,
      "train_tokens_per_second": 5964.952
    },
    {
      "epoch": 1.5502528735632184,
      "grad_norm": 1.073548674583435,
      "learning_rate": 4.505797901711762e-05,
      "loss": 1.6857,
      "num_input_tokens_seen": 25471200,
      "step": 4215,
      "train_runtime": 4270.1647,
      "train_tokens_per_second": 5964.922
    },
    {
      "epoch": 1.5506206896551724,
      "grad_norm": 0.9707559943199158,
      "learning_rate": 4.502116694275723e-05,
      "loss": 1.3454,
      "num_input_tokens_seen": 25478144,
      "step": 4216,
      "train_runtime": 4271.2747,
      "train_tokens_per_second": 5964.998
    },
    {
      "epoch": 1.5509885057471264,
      "grad_norm": 0.9022209644317627,
      "learning_rate": 4.4984354868396836e-05,
      "loss": 1.3278,
      "num_input_tokens_seen": 25482480,
      "step": 4217,
      "train_runtime": 4272.03,
      "train_tokens_per_second": 5964.958
    },
    {
      "epoch": 1.5513563218390805,
      "grad_norm": 0.9404050707817078,
      "learning_rate": 4.4947542794036445e-05,
      "loss": 1.5802,
      "num_input_tokens_seen": 25487120,
      "step": 4218,
      "train_runtime": 4272.8469,
      "train_tokens_per_second": 5964.904
    },
    {
      "epoch": 1.5517241379310345,
      "grad_norm": 1.0079388618469238,
      "learning_rate": 4.491073071967606e-05,
      "loss": 1.6163,
      "num_input_tokens_seen": 25491840,
      "step": 4219,
      "train_runtime": 4273.669,
      "train_tokens_per_second": 5964.861
    },
    {
      "epoch": 1.5520919540229885,
      "grad_norm": 0.9895066022872925,
      "learning_rate": 4.487391864531566e-05,
      "loss": 1.4185,
      "num_input_tokens_seen": 25500432,
      "step": 4220,
      "train_runtime": 4274.984,
      "train_tokens_per_second": 5965.036
    },
    {
      "epoch": 1.5524597701149425,
      "grad_norm": 0.8870865702629089,
      "learning_rate": 4.483710657095527e-05,
      "loss": 1.4676,
      "num_input_tokens_seen": 25505568,
      "step": 4221,
      "train_runtime": 4275.8602,
      "train_tokens_per_second": 5965.015
    },
    {
      "epoch": 1.5528275862068965,
      "grad_norm": 0.9963148832321167,
      "learning_rate": 4.480029449659489e-05,
      "loss": 1.5035,
      "num_input_tokens_seen": 25512480,
      "step": 4222,
      "train_runtime": 4276.9646,
      "train_tokens_per_second": 5965.09
    },
    {
      "epoch": 1.5531954022988506,
      "grad_norm": 1.0009320974349976,
      "learning_rate": 4.4763482422234496e-05,
      "loss": 1.4106,
      "num_input_tokens_seen": 25517120,
      "step": 4223,
      "train_runtime": 4277.7604,
      "train_tokens_per_second": 5965.065
    },
    {
      "epoch": 1.5535632183908046,
      "grad_norm": 0.9632704257965088,
      "learning_rate": 4.4726670347874105e-05,
      "loss": 1.513,
      "num_input_tokens_seen": 25522368,
      "step": 4224,
      "train_runtime": 4278.6413,
      "train_tokens_per_second": 5965.064
    },
    {
      "epoch": 1.5539310344827586,
      "grad_norm": 0.9824567437171936,
      "learning_rate": 4.4689858273513714e-05,
      "loss": 1.609,
      "num_input_tokens_seen": 25529312,
      "step": 4225,
      "train_runtime": 4279.6985,
      "train_tokens_per_second": 5965.213
    },
    {
      "epoch": 1.5542988505747126,
      "grad_norm": 0.9744192361831665,
      "learning_rate": 4.465304619915332e-05,
      "loss": 1.4286,
      "num_input_tokens_seen": 25534416,
      "step": 4226,
      "train_runtime": 4280.5279,
      "train_tokens_per_second": 5965.249
    },
    {
      "epoch": 1.5546666666666666,
      "grad_norm": 0.9338300824165344,
      "learning_rate": 4.461623412479293e-05,
      "loss": 1.246,
      "num_input_tokens_seen": 25541056,
      "step": 4227,
      "train_runtime": 4281.5844,
      "train_tokens_per_second": 5965.328
    },
    {
      "epoch": 1.5550344827586207,
      "grad_norm": 0.9846093058586121,
      "learning_rate": 4.457942205043255e-05,
      "loss": 1.5499,
      "num_input_tokens_seen": 25546512,
      "step": 4228,
      "train_runtime": 4282.4988,
      "train_tokens_per_second": 5965.329
    },
    {
      "epoch": 1.5554022988505747,
      "grad_norm": 1.0264620780944824,
      "learning_rate": 4.454260997607215e-05,
      "loss": 1.5117,
      "num_input_tokens_seen": 25553536,
      "step": 4229,
      "train_runtime": 4283.6258,
      "train_tokens_per_second": 5965.399
    },
    {
      "epoch": 1.5557701149425287,
      "grad_norm": 0.9882364273071289,
      "learning_rate": 4.4505797901711765e-05,
      "loss": 1.4894,
      "num_input_tokens_seen": 25559968,
      "step": 4230,
      "train_runtime": 4284.6592,
      "train_tokens_per_second": 5965.461
    },
    {
      "epoch": 1.5561379310344827,
      "grad_norm": 0.9190003275871277,
      "learning_rate": 4.4468985827351374e-05,
      "loss": 1.5542,
      "num_input_tokens_seen": 25566528,
      "step": 4231,
      "train_runtime": 4286.2509,
      "train_tokens_per_second": 5964.776
    },
    {
      "epoch": 1.5565057471264367,
      "grad_norm": 1.0381605625152588,
      "learning_rate": 4.443217375299098e-05,
      "loss": 1.5309,
      "num_input_tokens_seen": 25570944,
      "step": 4232,
      "train_runtime": 4287.0307,
      "train_tokens_per_second": 5964.721
    },
    {
      "epoch": 1.5568735632183908,
      "grad_norm": 0.8768330812454224,
      "learning_rate": 4.439536167863059e-05,
      "loss": 1.4222,
      "num_input_tokens_seen": 25578896,
      "step": 4233,
      "train_runtime": 4288.2464,
      "train_tokens_per_second": 5964.885
    },
    {
      "epoch": 1.5572413793103448,
      "grad_norm": 0.9919173121452332,
      "learning_rate": 4.435854960427021e-05,
      "loss": 1.7583,
      "num_input_tokens_seen": 25584544,
      "step": 4234,
      "train_runtime": 4289.1658,
      "train_tokens_per_second": 5964.923
    },
    {
      "epoch": 1.5576091954022988,
      "grad_norm": 1.0327153205871582,
      "learning_rate": 4.432173752990981e-05,
      "loss": 1.8772,
      "num_input_tokens_seen": 25590336,
      "step": 4235,
      "train_runtime": 4290.1098,
      "train_tokens_per_second": 5964.961
    },
    {
      "epoch": 1.5579770114942528,
      "grad_norm": 0.9097049832344055,
      "learning_rate": 4.428492545554942e-05,
      "loss": 1.496,
      "num_input_tokens_seen": 25595232,
      "step": 4236,
      "train_runtime": 4290.95,
      "train_tokens_per_second": 5964.934
    },
    {
      "epoch": 1.5583448275862068,
      "grad_norm": 0.9581717252731323,
      "learning_rate": 4.4248113381189034e-05,
      "loss": 1.6014,
      "num_input_tokens_seen": 25600032,
      "step": 4237,
      "train_runtime": 4291.7589,
      "train_tokens_per_second": 5964.928
    },
    {
      "epoch": 1.5587126436781609,
      "grad_norm": 1.0112465620040894,
      "learning_rate": 4.421130130682864e-05,
      "loss": 1.64,
      "num_input_tokens_seen": 25607088,
      "step": 4238,
      "train_runtime": 4292.9047,
      "train_tokens_per_second": 5964.979
    },
    {
      "epoch": 1.559080459770115,
      "grad_norm": 1.033936619758606,
      "learning_rate": 4.417448923246825e-05,
      "loss": 1.6785,
      "num_input_tokens_seen": 25612768,
      "step": 4239,
      "train_runtime": 4293.8363,
      "train_tokens_per_second": 5965.008
    },
    {
      "epoch": 1.559448275862069,
      "grad_norm": 1.0251233577728271,
      "learning_rate": 4.413767715810786e-05,
      "loss": 1.6126,
      "num_input_tokens_seen": 25617856,
      "step": 4240,
      "train_runtime": 4294.6895,
      "train_tokens_per_second": 5965.008
    },
    {
      "epoch": 1.5598160919540232,
      "grad_norm": 0.9828919768333435,
      "learning_rate": 4.410086508374747e-05,
      "loss": 1.5709,
      "num_input_tokens_seen": 25626496,
      "step": 4241,
      "train_runtime": 4296.0255,
      "train_tokens_per_second": 5965.164
    },
    {
      "epoch": 1.560183908045977,
      "grad_norm": 0.9944127202033997,
      "learning_rate": 4.406405300938708e-05,
      "loss": 1.6536,
      "num_input_tokens_seen": 25633008,
      "step": 4242,
      "train_runtime": 4297.0788,
      "train_tokens_per_second": 5965.217
    },
    {
      "epoch": 1.5605517241379312,
      "grad_norm": 0.9649595618247986,
      "learning_rate": 4.4027240935026694e-05,
      "loss": 1.5619,
      "num_input_tokens_seen": 25639280,
      "step": 4243,
      "train_runtime": 4298.0853,
      "train_tokens_per_second": 5965.279
    },
    {
      "epoch": 1.560919540229885,
      "grad_norm": 0.9769201278686523,
      "learning_rate": 4.3990428860666296e-05,
      "loss": 1.6019,
      "num_input_tokens_seen": 25644464,
      "step": 4244,
      "train_runtime": 4298.985,
      "train_tokens_per_second": 5965.237
    },
    {
      "epoch": 1.5612873563218392,
      "grad_norm": 0.988795280456543,
      "learning_rate": 4.395361678630591e-05,
      "loss": 1.5341,
      "num_input_tokens_seen": 25653104,
      "step": 4245,
      "train_runtime": 4300.3246,
      "train_tokens_per_second": 5965.388
    },
    {
      "epoch": 1.561655172413793,
      "grad_norm": 0.9538114070892334,
      "learning_rate": 4.391680471194552e-05,
      "loss": 1.4543,
      "num_input_tokens_seen": 25658112,
      "step": 4246,
      "train_runtime": 4301.1773,
      "train_tokens_per_second": 5965.37
    },
    {
      "epoch": 1.5620229885057473,
      "grad_norm": 0.9469668865203857,
      "learning_rate": 4.387999263758513e-05,
      "loss": 1.5033,
      "num_input_tokens_seen": 25664080,
      "step": 4247,
      "train_runtime": 4302.1476,
      "train_tokens_per_second": 5965.411
    },
    {
      "epoch": 1.562390804597701,
      "grad_norm": 1.0212516784667969,
      "learning_rate": 4.384318056322474e-05,
      "loss": 1.5974,
      "num_input_tokens_seen": 25670224,
      "step": 4248,
      "train_runtime": 4303.1535,
      "train_tokens_per_second": 5965.445
    },
    {
      "epoch": 1.5627586206896553,
      "grad_norm": 0.9424988627433777,
      "learning_rate": 4.3806368488864354e-05,
      "loss": 1.4809,
      "num_input_tokens_seen": 25674640,
      "step": 4249,
      "train_runtime": 4303.9045,
      "train_tokens_per_second": 5965.43
    },
    {
      "epoch": 1.5631264367816091,
      "grad_norm": 1.0168170928955078,
      "learning_rate": 4.3769556414503956e-05,
      "loss": 1.3401,
      "num_input_tokens_seen": 25680512,
      "step": 4250,
      "train_runtime": 4304.8691,
      "train_tokens_per_second": 5965.457
    },
    {
      "epoch": 1.5634942528735634,
      "grad_norm": 1.0160890817642212,
      "learning_rate": 4.3732744340143565e-05,
      "loss": 1.437,
      "num_input_tokens_seen": 25686464,
      "step": 4251,
      "train_runtime": 4305.8784,
      "train_tokens_per_second": 5965.441
    },
    {
      "epoch": 1.5638620689655172,
      "grad_norm": 0.8786030411720276,
      "learning_rate": 4.369593226578318e-05,
      "loss": 1.4234,
      "num_input_tokens_seen": 25691472,
      "step": 4252,
      "train_runtime": 4306.7413,
      "train_tokens_per_second": 5965.409
    },
    {
      "epoch": 1.5642298850574714,
      "grad_norm": 0.9332633018493652,
      "learning_rate": 4.365912019142278e-05,
      "loss": 1.4067,
      "num_input_tokens_seen": 25696992,
      "step": 4253,
      "train_runtime": 4307.6637,
      "train_tokens_per_second": 5965.413
    },
    {
      "epoch": 1.5645977011494252,
      "grad_norm": 0.9453607797622681,
      "learning_rate": 4.36223081170624e-05,
      "loss": 1.4662,
      "num_input_tokens_seen": 25702944,
      "step": 4254,
      "train_runtime": 4308.6278,
      "train_tokens_per_second": 5965.459
    },
    {
      "epoch": 1.5649655172413794,
      "grad_norm": 0.9211291670799255,
      "learning_rate": 4.358549604270201e-05,
      "loss": 1.5416,
      "num_input_tokens_seen": 25707792,
      "step": 4255,
      "train_runtime": 4309.4616,
      "train_tokens_per_second": 5965.43
    },
    {
      "epoch": 1.5653333333333332,
      "grad_norm": 0.9791601896286011,
      "learning_rate": 4.354868396834162e-05,
      "loss": 1.3551,
      "num_input_tokens_seen": 25715904,
      "step": 4256,
      "train_runtime": 4310.7334,
      "train_tokens_per_second": 5965.552
    },
    {
      "epoch": 1.5657011494252875,
      "grad_norm": 0.8566781878471375,
      "learning_rate": 4.3511871893981226e-05,
      "loss": 1.5358,
      "num_input_tokens_seen": 25722720,
      "step": 4257,
      "train_runtime": 4311.8114,
      "train_tokens_per_second": 5965.641
    },
    {
      "epoch": 1.5660689655172413,
      "grad_norm": 1.0267621278762817,
      "learning_rate": 4.347505981962084e-05,
      "loss": 1.4572,
      "num_input_tokens_seen": 25727872,
      "step": 4258,
      "train_runtime": 4312.6577,
      "train_tokens_per_second": 5965.665
    },
    {
      "epoch": 1.5664367816091955,
      "grad_norm": 1.033227562904358,
      "learning_rate": 4.343824774526044e-05,
      "loss": 1.6288,
      "num_input_tokens_seen": 25734224,
      "step": 4259,
      "train_runtime": 4313.693,
      "train_tokens_per_second": 5965.706
    },
    {
      "epoch": 1.5668045977011493,
      "grad_norm": 0.9439014792442322,
      "learning_rate": 4.340143567090006e-05,
      "loss": 1.339,
      "num_input_tokens_seen": 25741552,
      "step": 4260,
      "train_runtime": 4314.8433,
      "train_tokens_per_second": 5965.814
    },
    {
      "epoch": 1.5671724137931036,
      "grad_norm": 1.0122768878936768,
      "learning_rate": 4.336462359653967e-05,
      "loss": 1.4836,
      "num_input_tokens_seen": 25746400,
      "step": 4261,
      "train_runtime": 4316.1658,
      "train_tokens_per_second": 5965.109
    },
    {
      "epoch": 1.5675402298850574,
      "grad_norm": 0.8594652414321899,
      "learning_rate": 4.332781152217928e-05,
      "loss": 1.2932,
      "num_input_tokens_seen": 25758032,
      "step": 4262,
      "train_runtime": 4317.8807,
      "train_tokens_per_second": 5965.434
    },
    {
      "epoch": 1.5679080459770116,
      "grad_norm": 1.067428469657898,
      "learning_rate": 4.3290999447818886e-05,
      "loss": 1.7333,
      "num_input_tokens_seen": 25763504,
      "step": 4263,
      "train_runtime": 4318.7712,
      "train_tokens_per_second": 5965.471
    },
    {
      "epoch": 1.5682758620689654,
      "grad_norm": 0.9325382113456726,
      "learning_rate": 4.32541873734585e-05,
      "loss": 1.463,
      "num_input_tokens_seen": 25769344,
      "step": 4264,
      "train_runtime": 4319.7442,
      "train_tokens_per_second": 5965.479
    },
    {
      "epoch": 1.5686436781609197,
      "grad_norm": 1.0096169710159302,
      "learning_rate": 4.3217375299098103e-05,
      "loss": 1.5472,
      "num_input_tokens_seen": 25774464,
      "step": 4265,
      "train_runtime": 4320.6261,
      "train_tokens_per_second": 5965.447
    },
    {
      "epoch": 1.5690114942528735,
      "grad_norm": 1.001533031463623,
      "learning_rate": 4.318056322473771e-05,
      "loss": 1.453,
      "num_input_tokens_seen": 25780464,
      "step": 4266,
      "train_runtime": 4321.592,
      "train_tokens_per_second": 5965.502
    },
    {
      "epoch": 1.5693793103448277,
      "grad_norm": 1.0608218908309937,
      "learning_rate": 4.314375115037733e-05,
      "loss": 1.6735,
      "num_input_tokens_seen": 25787728,
      "step": 4267,
      "train_runtime": 4322.7169,
      "train_tokens_per_second": 5965.63
    },
    {
      "epoch": 1.5697471264367815,
      "grad_norm": 0.9219775795936584,
      "learning_rate": 4.310693907601693e-05,
      "loss": 1.483,
      "num_input_tokens_seen": 25794048,
      "step": 4268,
      "train_runtime": 4323.7428,
      "train_tokens_per_second": 5965.676
    },
    {
      "epoch": 1.5701149425287357,
      "grad_norm": 1.0273150205612183,
      "learning_rate": 4.3070127001656546e-05,
      "loss": 1.543,
      "num_input_tokens_seen": 25802944,
      "step": 4269,
      "train_runtime": 4325.1424,
      "train_tokens_per_second": 5965.802
    },
    {
      "epoch": 1.5704827586206895,
      "grad_norm": 0.9345740079879761,
      "learning_rate": 4.3033314927296155e-05,
      "loss": 1.4351,
      "num_input_tokens_seen": 25807984,
      "step": 4270,
      "train_runtime": 4326.0009,
      "train_tokens_per_second": 5965.783
    },
    {
      "epoch": 1.5708505747126438,
      "grad_norm": 1.0006455183029175,
      "learning_rate": 4.2996502852935764e-05,
      "loss": 1.7909,
      "num_input_tokens_seen": 25813376,
      "step": 4271,
      "train_runtime": 4326.8726,
      "train_tokens_per_second": 5965.828
    },
    {
      "epoch": 1.5712183908045976,
      "grad_norm": 0.9771626591682434,
      "learning_rate": 4.295969077857537e-05,
      "loss": 1.4778,
      "num_input_tokens_seen": 25818656,
      "step": 4272,
      "train_runtime": 4327.756,
      "train_tokens_per_second": 5965.83
    },
    {
      "epoch": 1.5715862068965518,
      "grad_norm": 1.0043179988861084,
      "learning_rate": 4.292287870421499e-05,
      "loss": 1.7215,
      "num_input_tokens_seen": 25823616,
      "step": 4273,
      "train_runtime": 4328.5939,
      "train_tokens_per_second": 5965.821
    },
    {
      "epoch": 1.5719540229885056,
      "grad_norm": 0.9097915887832642,
      "learning_rate": 4.288606662985459e-05,
      "loss": 1.4548,
      "num_input_tokens_seen": 25828752,
      "step": 4274,
      "train_runtime": 4329.4571,
      "train_tokens_per_second": 5965.818
    },
    {
      "epoch": 1.5723218390804599,
      "grad_norm": 0.9500516653060913,
      "learning_rate": 4.2849254555494206e-05,
      "loss": 1.4368,
      "num_input_tokens_seen": 25837152,
      "step": 4275,
      "train_runtime": 4330.76,
      "train_tokens_per_second": 5965.963
    },
    {
      "epoch": 1.5726896551724137,
      "grad_norm": 1.1244925260543823,
      "learning_rate": 4.2812442481133815e-05,
      "loss": 1.515,
      "num_input_tokens_seen": 25843536,
      "step": 4276,
      "train_runtime": 4331.7862,
      "train_tokens_per_second": 5966.023
    },
    {
      "epoch": 1.573057471264368,
      "grad_norm": 1.040409803390503,
      "learning_rate": 4.2775630406773424e-05,
      "loss": 1.452,
      "num_input_tokens_seen": 25849664,
      "step": 4277,
      "train_runtime": 4332.7847,
      "train_tokens_per_second": 5966.062
    },
    {
      "epoch": 1.5734252873563217,
      "grad_norm": 0.8676648736000061,
      "learning_rate": 4.273881833241303e-05,
      "loss": 1.3851,
      "num_input_tokens_seen": 25856576,
      "step": 4278,
      "train_runtime": 4333.874,
      "train_tokens_per_second": 5966.158
    },
    {
      "epoch": 1.573793103448276,
      "grad_norm": 1.0687546730041504,
      "learning_rate": 4.270200625805265e-05,
      "loss": 1.7076,
      "num_input_tokens_seen": 25861776,
      "step": 4279,
      "train_runtime": 4334.7414,
      "train_tokens_per_second": 5966.163
    },
    {
      "epoch": 1.5741609195402297,
      "grad_norm": 0.9815601110458374,
      "learning_rate": 4.266519418369225e-05,
      "loss": 1.5391,
      "num_input_tokens_seen": 25866592,
      "step": 4280,
      "train_runtime": 4335.5616,
      "train_tokens_per_second": 5966.146
    },
    {
      "epoch": 1.574528735632184,
      "grad_norm": 1.0182512998580933,
      "learning_rate": 4.262838210933186e-05,
      "loss": 1.5775,
      "num_input_tokens_seen": 25871296,
      "step": 4281,
      "train_runtime": 4336.3506,
      "train_tokens_per_second": 5966.145
    },
    {
      "epoch": 1.5748965517241378,
      "grad_norm": 0.8868687152862549,
      "learning_rate": 4.2591570034971475e-05,
      "loss": 1.5942,
      "num_input_tokens_seen": 25876944,
      "step": 4282,
      "train_runtime": 4337.2856,
      "train_tokens_per_second": 5966.161
    },
    {
      "epoch": 1.575264367816092,
      "grad_norm": 0.9742661714553833,
      "learning_rate": 4.255475796061108e-05,
      "loss": 1.3796,
      "num_input_tokens_seen": 25883456,
      "step": 4283,
      "train_runtime": 4338.3362,
      "train_tokens_per_second": 5966.217
    },
    {
      "epoch": 1.5756321839080458,
      "grad_norm": 0.7727472186088562,
      "learning_rate": 4.251794588625069e-05,
      "loss": 1.2235,
      "num_input_tokens_seen": 25896320,
      "step": 4284,
      "train_runtime": 4340.2435,
      "train_tokens_per_second": 5966.559
    },
    {
      "epoch": 1.576,
      "grad_norm": 0.965480625629425,
      "learning_rate": 4.24811338118903e-05,
      "loss": 1.4692,
      "num_input_tokens_seen": 25901920,
      "step": 4285,
      "train_runtime": 4341.1553,
      "train_tokens_per_second": 5966.596
    },
    {
      "epoch": 1.5763678160919539,
      "grad_norm": 0.9831876754760742,
      "learning_rate": 4.244432173752991e-05,
      "loss": 1.5166,
      "num_input_tokens_seen": 25907888,
      "step": 4286,
      "train_runtime": 4342.1182,
      "train_tokens_per_second": 5966.647
    },
    {
      "epoch": 1.576735632183908,
      "grad_norm": 1.0034735202789307,
      "learning_rate": 4.240750966316952e-05,
      "loss": 1.4433,
      "num_input_tokens_seen": 25913952,
      "step": 4287,
      "train_runtime": 4343.0958,
      "train_tokens_per_second": 5966.701
    },
    {
      "epoch": 1.577103448275862,
      "grad_norm": 1.0129915475845337,
      "learning_rate": 4.2370697588809135e-05,
      "loss": 1.8215,
      "num_input_tokens_seen": 25918448,
      "step": 4288,
      "train_runtime": 4343.8725,
      "train_tokens_per_second": 5966.669
    },
    {
      "epoch": 1.5774712643678162,
      "grad_norm": 1.0383774042129517,
      "learning_rate": 4.233388551444874e-05,
      "loss": 1.4937,
      "num_input_tokens_seen": 25926240,
      "step": 4289,
      "train_runtime": 4345.0854,
      "train_tokens_per_second": 5966.796
    },
    {
      "epoch": 1.57783908045977,
      "grad_norm": 0.9382445812225342,
      "learning_rate": 4.229707344008835e-05,
      "loss": 1.4022,
      "num_input_tokens_seen": 25933040,
      "step": 4290,
      "train_runtime": 4346.1712,
      "train_tokens_per_second": 5966.87
    },
    {
      "epoch": 1.5782068965517242,
      "grad_norm": 0.9903818964958191,
      "learning_rate": 4.226026136572796e-05,
      "loss": 1.5088,
      "num_input_tokens_seen": 25939168,
      "step": 4291,
      "train_runtime": 4347.6853,
      "train_tokens_per_second": 5966.202
    },
    {
      "epoch": 1.5785747126436782,
      "grad_norm": 1.0315245389938354,
      "learning_rate": 4.222344929136757e-05,
      "loss": 1.6216,
      "num_input_tokens_seen": 25949840,
      "step": 4292,
      "train_runtime": 4349.293,
      "train_tokens_per_second": 5966.45
    },
    {
      "epoch": 1.5789425287356322,
      "grad_norm": 0.9897589087486267,
      "learning_rate": 4.218663721700718e-05,
      "loss": 1.5035,
      "num_input_tokens_seen": 25954160,
      "step": 4293,
      "train_runtime": 4350.0654,
      "train_tokens_per_second": 5966.384
    },
    {
      "epoch": 1.5793103448275863,
      "grad_norm": 0.9769780039787292,
      "learning_rate": 4.2149825142646795e-05,
      "loss": 1.4296,
      "num_input_tokens_seen": 25960448,
      "step": 4294,
      "train_runtime": 4351.068,
      "train_tokens_per_second": 5966.454
    },
    {
      "epoch": 1.5796781609195403,
      "grad_norm": 0.9172228574752808,
      "learning_rate": 4.21130130682864e-05,
      "loss": 1.6009,
      "num_input_tokens_seen": 25966768,
      "step": 4295,
      "train_runtime": 4352.0773,
      "train_tokens_per_second": 5966.523
    },
    {
      "epoch": 1.5800459770114943,
      "grad_norm": 1.1289092302322388,
      "learning_rate": 4.2076200993926006e-05,
      "loss": 1.8916,
      "num_input_tokens_seen": 25972256,
      "step": 4296,
      "train_runtime": 4352.9901,
      "train_tokens_per_second": 5966.532
    },
    {
      "epoch": 1.5804137931034483,
      "grad_norm": 0.957542359828949,
      "learning_rate": 4.203938891956562e-05,
      "loss": 1.4331,
      "num_input_tokens_seen": 25978112,
      "step": 4297,
      "train_runtime": 4353.9408,
      "train_tokens_per_second": 5966.574
    },
    {
      "epoch": 1.5807816091954023,
      "grad_norm": 0.9918944239616394,
      "learning_rate": 4.2002576845205224e-05,
      "loss": 1.407,
      "num_input_tokens_seen": 25984496,
      "step": 4298,
      "train_runtime": 4354.9649,
      "train_tokens_per_second": 5966.637
    },
    {
      "epoch": 1.5811494252873564,
      "grad_norm": 0.9517878293991089,
      "learning_rate": 4.196576477084484e-05,
      "loss": 1.5335,
      "num_input_tokens_seen": 25989456,
      "step": 4299,
      "train_runtime": 4355.7986,
      "train_tokens_per_second": 5966.634
    },
    {
      "epoch": 1.5815172413793104,
      "grad_norm": 0.9874196648597717,
      "learning_rate": 4.192895269648445e-05,
      "loss": 1.6057,
      "num_input_tokens_seen": 25994224,
      "step": 4300,
      "train_runtime": 4356.6078,
      "train_tokens_per_second": 5966.62
    },
    {
      "epoch": 1.5818850574712644,
      "grad_norm": 0.8222333788871765,
      "learning_rate": 4.189214062212406e-05,
      "loss": 1.3939,
      "num_input_tokens_seen": 26008416,
      "step": 4301,
      "train_runtime": 4358.6572,
      "train_tokens_per_second": 5967.071
    },
    {
      "epoch": 1.5822528735632184,
      "grad_norm": 0.9964632987976074,
      "learning_rate": 4.1855328547763666e-05,
      "loss": 1.3196,
      "num_input_tokens_seen": 26013424,
      "step": 4302,
      "train_runtime": 4359.4984,
      "train_tokens_per_second": 5967.068
    },
    {
      "epoch": 1.5826206896551724,
      "grad_norm": 0.9242708683013916,
      "learning_rate": 4.181851647340328e-05,
      "loss": 1.4558,
      "num_input_tokens_seen": 26018000,
      "step": 4303,
      "train_runtime": 4360.2835,
      "train_tokens_per_second": 5967.043
    },
    {
      "epoch": 1.5829885057471265,
      "grad_norm": 1.0612273216247559,
      "learning_rate": 4.1781704399042884e-05,
      "loss": 1.4975,
      "num_input_tokens_seen": 26023120,
      "step": 4304,
      "train_runtime": 4361.1436,
      "train_tokens_per_second": 5967.04
    },
    {
      "epoch": 1.5833563218390805,
      "grad_norm": 0.9777734279632568,
      "learning_rate": 4.17448923246825e-05,
      "loss": 1.5889,
      "num_input_tokens_seen": 26029648,
      "step": 4305,
      "train_runtime": 4362.1909,
      "train_tokens_per_second": 5967.104
    },
    {
      "epoch": 1.5837241379310345,
      "grad_norm": 0.9828711152076721,
      "learning_rate": 4.170808025032211e-05,
      "loss": 1.5231,
      "num_input_tokens_seen": 26038448,
      "step": 4306,
      "train_runtime": 4363.5325,
      "train_tokens_per_second": 5967.286
    },
    {
      "epoch": 1.5840919540229885,
      "grad_norm": 0.9952308535575867,
      "learning_rate": 4.167126817596172e-05,
      "loss": 1.6852,
      "num_input_tokens_seen": 26044128,
      "step": 4307,
      "train_runtime": 4364.4738,
      "train_tokens_per_second": 5967.301
    },
    {
      "epoch": 1.5844597701149425,
      "grad_norm": 0.88361656665802,
      "learning_rate": 4.1634456101601326e-05,
      "loss": 1.4891,
      "num_input_tokens_seen": 26051328,
      "step": 4308,
      "train_runtime": 4365.6289,
      "train_tokens_per_second": 5967.371
    },
    {
      "epoch": 1.5848275862068966,
      "grad_norm": 0.9838547706604004,
      "learning_rate": 4.159764402724094e-05,
      "loss": 1.5035,
      "num_input_tokens_seen": 26056832,
      "step": 4309,
      "train_runtime": 4366.542,
      "train_tokens_per_second": 5967.384
    },
    {
      "epoch": 1.5851954022988506,
      "grad_norm": 0.992989182472229,
      "learning_rate": 4.1560831952880544e-05,
      "loss": 1.6844,
      "num_input_tokens_seen": 26061696,
      "step": 4310,
      "train_runtime": 4367.3527,
      "train_tokens_per_second": 5967.39
    },
    {
      "epoch": 1.5855632183908046,
      "grad_norm": 0.8714199662208557,
      "learning_rate": 4.152401987852015e-05,
      "loss": 1.454,
      "num_input_tokens_seen": 26067264,
      "step": 4311,
      "train_runtime": 4368.3079,
      "train_tokens_per_second": 5967.36
    },
    {
      "epoch": 1.5859310344827586,
      "grad_norm": 0.8716127276420593,
      "learning_rate": 4.148720780415977e-05,
      "loss": 1.5206,
      "num_input_tokens_seen": 26075472,
      "step": 4312,
      "train_runtime": 4369.5835,
      "train_tokens_per_second": 5967.496
    },
    {
      "epoch": 1.5862988505747126,
      "grad_norm": 0.8958300352096558,
      "learning_rate": 4.145039572979937e-05,
      "loss": 1.2265,
      "num_input_tokens_seen": 26080688,
      "step": 4313,
      "train_runtime": 4370.462,
      "train_tokens_per_second": 5967.49
    },
    {
      "epoch": 1.5866666666666667,
      "grad_norm": 0.9490132331848145,
      "learning_rate": 4.1413583655438987e-05,
      "loss": 1.3134,
      "num_input_tokens_seen": 26085712,
      "step": 4314,
      "train_runtime": 4371.3159,
      "train_tokens_per_second": 5967.474
    },
    {
      "epoch": 1.5870344827586207,
      "grad_norm": 0.9401388168334961,
      "learning_rate": 4.1376771581078595e-05,
      "loss": 1.4178,
      "num_input_tokens_seen": 26092368,
      "step": 4315,
      "train_runtime": 4372.3947,
      "train_tokens_per_second": 5967.523
    },
    {
      "epoch": 1.5874022988505747,
      "grad_norm": 0.8958325386047363,
      "learning_rate": 4.1339959506718204e-05,
      "loss": 1.3569,
      "num_input_tokens_seen": 26099904,
      "step": 4316,
      "train_runtime": 4373.5888,
      "train_tokens_per_second": 5967.617
    },
    {
      "epoch": 1.5877701149425287,
      "grad_norm": 0.9853538274765015,
      "learning_rate": 4.130314743235781e-05,
      "loss": 1.5538,
      "num_input_tokens_seen": 26104656,
      "step": 4317,
      "train_runtime": 4374.4115,
      "train_tokens_per_second": 5967.581
    },
    {
      "epoch": 1.5881379310344828,
      "grad_norm": 0.9488229155540466,
      "learning_rate": 4.126633535799743e-05,
      "loss": 1.4395,
      "num_input_tokens_seen": 26111120,
      "step": 4318,
      "train_runtime": 4375.4668,
      "train_tokens_per_second": 5967.619
    },
    {
      "epoch": 1.5885057471264368,
      "grad_norm": 1.0154321193695068,
      "learning_rate": 4.122952328363703e-05,
      "loss": 1.5689,
      "num_input_tokens_seen": 26116320,
      "step": 4319,
      "train_runtime": 4376.3522,
      "train_tokens_per_second": 5967.6
    },
    {
      "epoch": 1.5888735632183908,
      "grad_norm": 1.0700663328170776,
      "learning_rate": 4.119271120927665e-05,
      "loss": 1.7247,
      "num_input_tokens_seen": 26121520,
      "step": 4320,
      "train_runtime": 4377.2232,
      "train_tokens_per_second": 5967.601
    },
    {
      "epoch": 1.5892413793103448,
      "grad_norm": 0.9283444881439209,
      "learning_rate": 4.1155899134916256e-05,
      "loss": 1.3292,
      "num_input_tokens_seen": 26128608,
      "step": 4321,
      "train_runtime": 4378.8542,
      "train_tokens_per_second": 5966.997
    },
    {
      "epoch": 1.5896091954022988,
      "grad_norm": 1.0338287353515625,
      "learning_rate": 4.1119087060555864e-05,
      "loss": 1.5759,
      "num_input_tokens_seen": 26134784,
      "step": 4322,
      "train_runtime": 4379.8632,
      "train_tokens_per_second": 5967.032
    },
    {
      "epoch": 1.5899770114942529,
      "grad_norm": 0.9672470092773438,
      "learning_rate": 4.108227498619547e-05,
      "loss": 1.5717,
      "num_input_tokens_seen": 26139024,
      "step": 4323,
      "train_runtime": 4380.6215,
      "train_tokens_per_second": 5966.967
    },
    {
      "epoch": 1.5903448275862069,
      "grad_norm": 0.9788838028907776,
      "learning_rate": 4.104546291183509e-05,
      "loss": 1.3699,
      "num_input_tokens_seen": 26144496,
      "step": 4324,
      "train_runtime": 4381.5088,
      "train_tokens_per_second": 5967.008
    },
    {
      "epoch": 1.590712643678161,
      "grad_norm": 0.9931208491325378,
      "learning_rate": 4.100865083747469e-05,
      "loss": 1.5796,
      "num_input_tokens_seen": 26151552,
      "step": 4325,
      "train_runtime": 4382.622,
      "train_tokens_per_second": 5967.102
    },
    {
      "epoch": 1.591080459770115,
      "grad_norm": 1.1408939361572266,
      "learning_rate": 4.09718387631143e-05,
      "loss": 1.6329,
      "num_input_tokens_seen": 26158528,
      "step": 4326,
      "train_runtime": 4383.756,
      "train_tokens_per_second": 5967.15
    },
    {
      "epoch": 1.591448275862069,
      "grad_norm": 1.0218188762664795,
      "learning_rate": 4.0935026688753916e-05,
      "loss": 1.539,
      "num_input_tokens_seen": 26163248,
      "step": 4327,
      "train_runtime": 4384.5794,
      "train_tokens_per_second": 5967.106
    },
    {
      "epoch": 1.591816091954023,
      "grad_norm": 0.9635866284370422,
      "learning_rate": 4.089821461439352e-05,
      "loss": 1.4762,
      "num_input_tokens_seen": 26168000,
      "step": 4328,
      "train_runtime": 4385.3741,
      "train_tokens_per_second": 5967.108
    },
    {
      "epoch": 1.592183908045977,
      "grad_norm": 0.9733908772468567,
      "learning_rate": 4.0861402540033134e-05,
      "loss": 1.4928,
      "num_input_tokens_seen": 26172752,
      "step": 4329,
      "train_runtime": 4386.182,
      "train_tokens_per_second": 5967.092
    },
    {
      "epoch": 1.592551724137931,
      "grad_norm": 0.98436439037323,
      "learning_rate": 4.082459046567274e-05,
      "loss": 1.5519,
      "num_input_tokens_seen": 26177696,
      "step": 4330,
      "train_runtime": 4387.0449,
      "train_tokens_per_second": 5967.045
    },
    {
      "epoch": 1.592919540229885,
      "grad_norm": 1.0157173871994019,
      "learning_rate": 4.078777839131235e-05,
      "loss": 1.638,
      "num_input_tokens_seen": 26184416,
      "step": 4331,
      "train_runtime": 4388.1043,
      "train_tokens_per_second": 5967.136
    },
    {
      "epoch": 1.593287356321839,
      "grad_norm": 1.0790306329727173,
      "learning_rate": 4.075096631695196e-05,
      "loss": 1.4374,
      "num_input_tokens_seen": 26191984,
      "step": 4332,
      "train_runtime": 4389.295,
      "train_tokens_per_second": 5967.242
    },
    {
      "epoch": 1.593655172413793,
      "grad_norm": 0.8191881775856018,
      "learning_rate": 4.0714154242591576e-05,
      "loss": 1.3121,
      "num_input_tokens_seen": 26201536,
      "step": 4333,
      "train_runtime": 4390.744,
      "train_tokens_per_second": 5967.448
    },
    {
      "epoch": 1.594022988505747,
      "grad_norm": 0.9245478510856628,
      "learning_rate": 4.067734216823118e-05,
      "loss": 1.3724,
      "num_input_tokens_seen": 26205680,
      "step": 4334,
      "train_runtime": 4391.4851,
      "train_tokens_per_second": 5967.384
    },
    {
      "epoch": 1.594390804597701,
      "grad_norm": 1.0320152044296265,
      "learning_rate": 4.0640530093870794e-05,
      "loss": 1.6407,
      "num_input_tokens_seen": 26210688,
      "step": 4335,
      "train_runtime": 4392.3413,
      "train_tokens_per_second": 5967.361
    },
    {
      "epoch": 1.5947586206896551,
      "grad_norm": 0.9842714667320251,
      "learning_rate": 4.06037180195104e-05,
      "loss": 1.5017,
      "num_input_tokens_seen": 26216512,
      "step": 4336,
      "train_runtime": 4393.309,
      "train_tokens_per_second": 5967.373
    },
    {
      "epoch": 1.5951264367816091,
      "grad_norm": 0.9979831576347351,
      "learning_rate": 4.056690594515001e-05,
      "loss": 1.4886,
      "num_input_tokens_seen": 26222208,
      "step": 4337,
      "train_runtime": 4394.2404,
      "train_tokens_per_second": 5967.404
    },
    {
      "epoch": 1.5954942528735632,
      "grad_norm": 1.0059894323349,
      "learning_rate": 4.053009387078962e-05,
      "loss": 1.4552,
      "num_input_tokens_seen": 26228032,
      "step": 4338,
      "train_runtime": 4395.1846,
      "train_tokens_per_second": 5967.447
    },
    {
      "epoch": 1.5958620689655172,
      "grad_norm": 1.0417289733886719,
      "learning_rate": 4.0493281796429236e-05,
      "loss": 1.5313,
      "num_input_tokens_seen": 26232816,
      "step": 4339,
      "train_runtime": 4395.9948,
      "train_tokens_per_second": 5967.436
    },
    {
      "epoch": 1.5962298850574712,
      "grad_norm": 1.0718395709991455,
      "learning_rate": 4.045646972206884e-05,
      "loss": 1.5523,
      "num_input_tokens_seen": 26237952,
      "step": 4340,
      "train_runtime": 4396.8543,
      "train_tokens_per_second": 5967.437
    },
    {
      "epoch": 1.5965977011494252,
      "grad_norm": 1.0831660032272339,
      "learning_rate": 4.041965764770845e-05,
      "loss": 1.5731,
      "num_input_tokens_seen": 26243200,
      "step": 4341,
      "train_runtime": 4397.7618,
      "train_tokens_per_second": 5967.399
    },
    {
      "epoch": 1.5969655172413795,
      "grad_norm": 1.0683315992355347,
      "learning_rate": 4.038284557334806e-05,
      "loss": 1.3889,
      "num_input_tokens_seen": 26250416,
      "step": 4342,
      "train_runtime": 4398.9218,
      "train_tokens_per_second": 5967.466
    },
    {
      "epoch": 1.5973333333333333,
      "grad_norm": 0.9711940884590149,
      "learning_rate": 4.0346033498987665e-05,
      "loss": 1.4773,
      "num_input_tokens_seen": 26255104,
      "step": 4343,
      "train_runtime": 4399.7449,
      "train_tokens_per_second": 5967.415
    },
    {
      "epoch": 1.5977011494252875,
      "grad_norm": 1.040043830871582,
      "learning_rate": 4.030922142462728e-05,
      "loss": 1.4573,
      "num_input_tokens_seen": 26262064,
      "step": 4344,
      "train_runtime": 4400.8555,
      "train_tokens_per_second": 5967.491
    },
    {
      "epoch": 1.5980689655172413,
      "grad_norm": 0.9906898140907288,
      "learning_rate": 4.027240935026689e-05,
      "loss": 1.5405,
      "num_input_tokens_seen": 26269952,
      "step": 4345,
      "train_runtime": 4402.0723,
      "train_tokens_per_second": 5967.633
    },
    {
      "epoch": 1.5984367816091956,
      "grad_norm": 0.9680057168006897,
      "learning_rate": 4.02355972759065e-05,
      "loss": 1.5904,
      "num_input_tokens_seen": 26274912,
      "step": 4346,
      "train_runtime": 4402.9058,
      "train_tokens_per_second": 5967.63
    },
    {
      "epoch": 1.5988045977011494,
      "grad_norm": 0.9311021566390991,
      "learning_rate": 4.019878520154611e-05,
      "loss": 1.4334,
      "num_input_tokens_seen": 26280448,
      "step": 4347,
      "train_runtime": 4403.861,
      "train_tokens_per_second": 5967.592
    },
    {
      "epoch": 1.5991724137931036,
      "grad_norm": 1.0014636516571045,
      "learning_rate": 4.016197312718572e-05,
      "loss": 1.5175,
      "num_input_tokens_seen": 26288048,
      "step": 4348,
      "train_runtime": 4405.0279,
      "train_tokens_per_second": 5967.737
    },
    {
      "epoch": 1.5995402298850574,
      "grad_norm": 0.8882377743721008,
      "learning_rate": 4.0125161052825325e-05,
      "loss": 1.4356,
      "num_input_tokens_seen": 26294880,
      "step": 4349,
      "train_runtime": 4406.1179,
      "train_tokens_per_second": 5967.811
    },
    {
      "epoch": 1.5999080459770116,
      "grad_norm": 1.0775160789489746,
      "learning_rate": 4.008834897846494e-05,
      "loss": 1.6759,
      "num_input_tokens_seen": 26300256,
      "step": 4350,
      "train_runtime": 4407.019,
      "train_tokens_per_second": 5967.811
    },
    {
      "epoch": 1.6002758620689654,
      "grad_norm": 1.0069363117218018,
      "learning_rate": 4.005153690410455e-05,
      "loss": 1.486,
      "num_input_tokens_seen": 26305344,
      "step": 4351,
      "train_runtime": 4408.368,
      "train_tokens_per_second": 5967.139
    },
    {
      "epoch": 1.6006436781609197,
      "grad_norm": 1.0245729684829712,
      "learning_rate": 4.001472482974416e-05,
      "loss": 1.4882,
      "num_input_tokens_seen": 26310064,
      "step": 4352,
      "train_runtime": 4409.1846,
      "train_tokens_per_second": 5967.104
    },
    {
      "epoch": 1.6006436781609197,
      "eval_loss": 1.755459189414978,
      "eval_runtime": 4.7855,
      "eval_samples_per_second": 208.966,
      "eval_steps_per_second": 13.165,
      "num_input_tokens_seen": 26310064,
      "step": 4352
    },
    {
      "epoch": 1.6010114942528735,
      "grad_norm": 1.0545986890792847,
      "learning_rate": 3.997791275538377e-05,
      "loss": 1.4745,
      "num_input_tokens_seen": 26315584,
      "step": 4353,
      "train_runtime": 4414.9009,
      "train_tokens_per_second": 5960.628
    },
    {
      "epoch": 1.6013793103448277,
      "grad_norm": 1.0666497945785522,
      "learning_rate": 3.994110068102338e-05,
      "loss": 1.6229,
      "num_input_tokens_seen": 26320656,
      "step": 4354,
      "train_runtime": 4415.7601,
      "train_tokens_per_second": 5960.617
    },
    {
      "epoch": 1.6017471264367815,
      "grad_norm": 1.0167460441589355,
      "learning_rate": 3.9904288606662985e-05,
      "loss": 1.4698,
      "num_input_tokens_seen": 26326384,
      "step": 4355,
      "train_runtime": 4416.6835,
      "train_tokens_per_second": 5960.668
    },
    {
      "epoch": 1.6021149425287358,
      "grad_norm": 1.0352330207824707,
      "learning_rate": 3.9867476532302594e-05,
      "loss": 1.564,
      "num_input_tokens_seen": 26332000,
      "step": 4356,
      "train_runtime": 4417.6168,
      "train_tokens_per_second": 5960.68
    },
    {
      "epoch": 1.6024827586206896,
      "grad_norm": 0.9572430849075317,
      "learning_rate": 3.983066445794221e-05,
      "loss": 1.5439,
      "num_input_tokens_seen": 26339232,
      "step": 4357,
      "train_runtime": 4418.7487,
      "train_tokens_per_second": 5960.79
    },
    {
      "epoch": 1.6028505747126438,
      "grad_norm": 1.1000181436538696,
      "learning_rate": 3.979385238358181e-05,
      "loss": 1.5571,
      "num_input_tokens_seen": 26349952,
      "step": 4358,
      "train_runtime": 4420.3504,
      "train_tokens_per_second": 5961.055
    },
    {
      "epoch": 1.6032183908045976,
      "grad_norm": 1.0398907661437988,
      "learning_rate": 3.975704030922143e-05,
      "loss": 1.5054,
      "num_input_tokens_seen": 26357280,
      "step": 4359,
      "train_runtime": 4421.5014,
      "train_tokens_per_second": 5961.161
    },
    {
      "epoch": 1.6035862068965518,
      "grad_norm": 1.129089593887329,
      "learning_rate": 3.9720228234861036e-05,
      "loss": 1.7235,
      "num_input_tokens_seen": 26361408,
      "step": 4360,
      "train_runtime": 4422.2285,
      "train_tokens_per_second": 5961.114
    },
    {
      "epoch": 1.6039540229885056,
      "grad_norm": 0.9978166818618774,
      "learning_rate": 3.9683416160500645e-05,
      "loss": 1.3892,
      "num_input_tokens_seen": 26367840,
      "step": 4361,
      "train_runtime": 4423.2869,
      "train_tokens_per_second": 5961.142
    },
    {
      "epoch": 1.6043218390804599,
      "grad_norm": 1.0726642608642578,
      "learning_rate": 3.9646604086140254e-05,
      "loss": 1.5379,
      "num_input_tokens_seen": 26373296,
      "step": 4362,
      "train_runtime": 4424.1787,
      "train_tokens_per_second": 5961.173
    },
    {
      "epoch": 1.6046896551724137,
      "grad_norm": 1.0192846059799194,
      "learning_rate": 3.960979201177987e-05,
      "loss": 1.4906,
      "num_input_tokens_seen": 26378176,
      "step": 4363,
      "train_runtime": 4424.9928,
      "train_tokens_per_second": 5961.179
    },
    {
      "epoch": 1.605057471264368,
      "grad_norm": 0.9348970055580139,
      "learning_rate": 3.957297993741947e-05,
      "loss": 1.4609,
      "num_input_tokens_seen": 26384576,
      "step": 4364,
      "train_runtime": 4425.9962,
      "train_tokens_per_second": 5961.274
    },
    {
      "epoch": 1.6054252873563217,
      "grad_norm": 0.99070805311203,
      "learning_rate": 3.953616786305909e-05,
      "loss": 1.4948,
      "num_input_tokens_seen": 26389632,
      "step": 4365,
      "train_runtime": 4426.8532,
      "train_tokens_per_second": 5961.262
    },
    {
      "epoch": 1.605793103448276,
      "grad_norm": 1.0859571695327759,
      "learning_rate": 3.9499355788698696e-05,
      "loss": 1.5446,
      "num_input_tokens_seen": 26394400,
      "step": 4366,
      "train_runtime": 4427.6819,
      "train_tokens_per_second": 5961.223
    },
    {
      "epoch": 1.6061609195402298,
      "grad_norm": 1.0329722166061401,
      "learning_rate": 3.9462543714338305e-05,
      "loss": 1.6001,
      "num_input_tokens_seen": 26398832,
      "step": 4367,
      "train_runtime": 4428.4557,
      "train_tokens_per_second": 5961.182
    },
    {
      "epoch": 1.606528735632184,
      "grad_norm": 1.043910026550293,
      "learning_rate": 3.9425731639977914e-05,
      "loss": 1.4536,
      "num_input_tokens_seen": 26404128,
      "step": 4368,
      "train_runtime": 4429.3586,
      "train_tokens_per_second": 5961.163
    },
    {
      "epoch": 1.6068965517241378,
      "grad_norm": 0.9209439754486084,
      "learning_rate": 3.938891956561753e-05,
      "loss": 1.6301,
      "num_input_tokens_seen": 26410192,
      "step": 4369,
      "train_runtime": 4430.3523,
      "train_tokens_per_second": 5961.195
    },
    {
      "epoch": 1.607264367816092,
      "grad_norm": 1.0413321256637573,
      "learning_rate": 3.935210749125713e-05,
      "loss": 1.564,
      "num_input_tokens_seen": 26417152,
      "step": 4370,
      "train_runtime": 4431.4692,
      "train_tokens_per_second": 5961.263
    },
    {
      "epoch": 1.6076321839080459,
      "grad_norm": 0.9783403277397156,
      "learning_rate": 3.931529541689674e-05,
      "loss": 1.2865,
      "num_input_tokens_seen": 26423168,
      "step": 4371,
      "train_runtime": 4432.4636,
      "train_tokens_per_second": 5961.283
    },
    {
      "epoch": 1.608,
      "grad_norm": 1.0528738498687744,
      "learning_rate": 3.9278483342536357e-05,
      "loss": 1.5607,
      "num_input_tokens_seen": 26430544,
      "step": 4372,
      "train_runtime": 4433.6012,
      "train_tokens_per_second": 5961.417
    },
    {
      "epoch": 1.608367816091954,
      "grad_norm": 0.9902295470237732,
      "learning_rate": 3.924167126817596e-05,
      "loss": 1.4594,
      "num_input_tokens_seen": 26434800,
      "step": 4373,
      "train_runtime": 4434.3693,
      "train_tokens_per_second": 5961.344
    },
    {
      "epoch": 1.6087356321839081,
      "grad_norm": 0.9916370511054993,
      "learning_rate": 3.9204859193815574e-05,
      "loss": 1.4159,
      "num_input_tokens_seen": 26439504,
      "step": 4374,
      "train_runtime": 4435.1964,
      "train_tokens_per_second": 5961.293
    },
    {
      "epoch": 1.609103448275862,
      "grad_norm": 0.9547199606895447,
      "learning_rate": 3.916804711945518e-05,
      "loss": 1.5048,
      "num_input_tokens_seen": 26445376,
      "step": 4375,
      "train_runtime": 4436.1658,
      "train_tokens_per_second": 5961.314
    },
    {
      "epoch": 1.6094712643678162,
      "grad_norm": 1.04433274269104,
      "learning_rate": 3.913123504509479e-05,
      "loss": 1.5698,
      "num_input_tokens_seen": 26451664,
      "step": 4376,
      "train_runtime": 4437.177,
      "train_tokens_per_second": 5961.372
    },
    {
      "epoch": 1.60983908045977,
      "grad_norm": 1.0016043186187744,
      "learning_rate": 3.90944229707344e-05,
      "loss": 1.4724,
      "num_input_tokens_seen": 26457664,
      "step": 4377,
      "train_runtime": 4438.1569,
      "train_tokens_per_second": 5961.408
    },
    {
      "epoch": 1.6102068965517242,
      "grad_norm": 0.9706348180770874,
      "learning_rate": 3.905761089637402e-05,
      "loss": 1.5244,
      "num_input_tokens_seen": 26462336,
      "step": 4378,
      "train_runtime": 4438.9356,
      "train_tokens_per_second": 5961.415
    },
    {
      "epoch": 1.610574712643678,
      "grad_norm": 0.967286229133606,
      "learning_rate": 3.902079882201362e-05,
      "loss": 1.5636,
      "num_input_tokens_seen": 26467328,
      "step": 4379,
      "train_runtime": 4439.7835,
      "train_tokens_per_second": 5961.401
    },
    {
      "epoch": 1.6109425287356323,
      "grad_norm": 1.0006687641143799,
      "learning_rate": 3.8983986747653234e-05,
      "loss": 1.4439,
      "num_input_tokens_seen": 26472336,
      "step": 4380,
      "train_runtime": 4440.6076,
      "train_tokens_per_second": 5961.422
    },
    {
      "epoch": 1.611310344827586,
      "grad_norm": 0.9450172781944275,
      "learning_rate": 3.894717467329284e-05,
      "loss": 1.4469,
      "num_input_tokens_seen": 26477408,
      "step": 4381,
      "train_runtime": 4442.0079,
      "train_tokens_per_second": 5960.685
    },
    {
      "epoch": 1.6116781609195403,
      "grad_norm": 0.9812448024749756,
      "learning_rate": 3.891036259893245e-05,
      "loss": 1.5535,
      "num_input_tokens_seen": 26482336,
      "step": 4382,
      "train_runtime": 4442.853,
      "train_tokens_per_second": 5960.66
    },
    {
      "epoch": 1.612045977011494,
      "grad_norm": 1.051989197731018,
      "learning_rate": 3.887355052457206e-05,
      "loss": 1.3843,
      "num_input_tokens_seen": 26487920,
      "step": 4383,
      "train_runtime": 4443.7582,
      "train_tokens_per_second": 5960.702
    },
    {
      "epoch": 1.6124137931034483,
      "grad_norm": 1.1212756633758545,
      "learning_rate": 3.883673845021168e-05,
      "loss": 1.6682,
      "num_input_tokens_seen": 26493056,
      "step": 4384,
      "train_runtime": 4444.6101,
      "train_tokens_per_second": 5960.715
    },
    {
      "epoch": 1.6127816091954021,
      "grad_norm": 0.9472477436065674,
      "learning_rate": 3.879992637585128e-05,
      "loss": 1.5529,
      "num_input_tokens_seen": 26499776,
      "step": 4385,
      "train_runtime": 4445.6682,
      "train_tokens_per_second": 5960.808
    },
    {
      "epoch": 1.6131494252873564,
      "grad_norm": 0.9102217555046082,
      "learning_rate": 3.876311430149089e-05,
      "loss": 1.4751,
      "num_input_tokens_seen": 26509616,
      "step": 4386,
      "train_runtime": 4447.1771,
      "train_tokens_per_second": 5960.998
    },
    {
      "epoch": 1.6135172413793102,
      "grad_norm": 1.003507375717163,
      "learning_rate": 3.8726302227130503e-05,
      "loss": 1.2815,
      "num_input_tokens_seen": 26516128,
      "step": 4387,
      "train_runtime": 4448.2193,
      "train_tokens_per_second": 5961.066
    },
    {
      "epoch": 1.6138850574712644,
      "grad_norm": 0.974827229976654,
      "learning_rate": 3.8689490152770106e-05,
      "loss": 1.4255,
      "num_input_tokens_seen": 26522032,
      "step": 4388,
      "train_runtime": 4449.1929,
      "train_tokens_per_second": 5961.088
    },
    {
      "epoch": 1.6142528735632182,
      "grad_norm": 1.0163264274597168,
      "learning_rate": 3.865267807840972e-05,
      "loss": 1.5271,
      "num_input_tokens_seen": 26528032,
      "step": 4389,
      "train_runtime": 4450.1648,
      "train_tokens_per_second": 5961.135
    },
    {
      "epoch": 1.6146206896551725,
      "grad_norm": 0.9294977784156799,
      "learning_rate": 3.861586600404933e-05,
      "loss": 1.3502,
      "num_input_tokens_seen": 26533104,
      "step": 4390,
      "train_runtime": 4451.0193,
      "train_tokens_per_second": 5961.13
    },
    {
      "epoch": 1.6149885057471265,
      "grad_norm": 1.1055110692977905,
      "learning_rate": 3.857905392968894e-05,
      "loss": 1.3525,
      "num_input_tokens_seen": 26539488,
      "step": 4391,
      "train_runtime": 4452.0468,
      "train_tokens_per_second": 5961.188
    },
    {
      "epoch": 1.6153563218390805,
      "grad_norm": 1.035902500152588,
      "learning_rate": 3.854224185532855e-05,
      "loss": 1.401,
      "num_input_tokens_seen": 26545024,
      "step": 4392,
      "train_runtime": 4452.9484,
      "train_tokens_per_second": 5961.224
    },
    {
      "epoch": 1.6157241379310345,
      "grad_norm": 0.9403799772262573,
      "learning_rate": 3.8505429780968164e-05,
      "loss": 1.412,
      "num_input_tokens_seen": 26551744,
      "step": 4393,
      "train_runtime": 4454.0344,
      "train_tokens_per_second": 5961.279
    },
    {
      "epoch": 1.6160919540229886,
      "grad_norm": 1.0154902935028076,
      "learning_rate": 3.8468617706607766e-05,
      "loss": 1.5268,
      "num_input_tokens_seen": 26556768,
      "step": 4394,
      "train_runtime": 4454.8986,
      "train_tokens_per_second": 5961.251
    },
    {
      "epoch": 1.6164597701149426,
      "grad_norm": 0.8746429085731506,
      "learning_rate": 3.843180563224738e-05,
      "loss": 1.4239,
      "num_input_tokens_seen": 26563600,
      "step": 4395,
      "train_runtime": 4456.0022,
      "train_tokens_per_second": 5961.308
    },
    {
      "epoch": 1.6168275862068966,
      "grad_norm": 0.9715175032615662,
      "learning_rate": 3.839499355788699e-05,
      "loss": 1.5218,
      "num_input_tokens_seen": 26568880,
      "step": 4396,
      "train_runtime": 4456.8982,
      "train_tokens_per_second": 5961.294
    },
    {
      "epoch": 1.6171954022988506,
      "grad_norm": 0.9170100688934326,
      "learning_rate": 3.83581814835266e-05,
      "loss": 1.4938,
      "num_input_tokens_seen": 26576752,
      "step": 4397,
      "train_runtime": 4458.1607,
      "train_tokens_per_second": 5961.371
    },
    {
      "epoch": 1.6175632183908046,
      "grad_norm": 0.9837867617607117,
      "learning_rate": 3.832136940916621e-05,
      "loss": 1.5362,
      "num_input_tokens_seen": 26583680,
      "step": 4398,
      "train_runtime": 4459.2703,
      "train_tokens_per_second": 5961.442
    },
    {
      "epoch": 1.6179310344827587,
      "grad_norm": 1.02565336227417,
      "learning_rate": 3.8284557334805824e-05,
      "loss": 1.5985,
      "num_input_tokens_seen": 26589088,
      "step": 4399,
      "train_runtime": 4460.1741,
      "train_tokens_per_second": 5961.446
    },
    {
      "epoch": 1.6182988505747127,
      "grad_norm": 0.9038709998130798,
      "learning_rate": 3.8247745260445426e-05,
      "loss": 1.3906,
      "num_input_tokens_seen": 26596448,
      "step": 4400,
      "train_runtime": 4461.3159,
      "train_tokens_per_second": 5961.57
    },
    {
      "epoch": 1.6186666666666667,
      "grad_norm": 0.9837738871574402,
      "learning_rate": 3.8210933186085035e-05,
      "loss": 1.5132,
      "num_input_tokens_seen": 26603360,
      "step": 4401,
      "train_runtime": 4462.4194,
      "train_tokens_per_second": 5961.645
    },
    {
      "epoch": 1.6190344827586207,
      "grad_norm": 1.008815050125122,
      "learning_rate": 3.817412111172465e-05,
      "loss": 1.6265,
      "num_input_tokens_seen": 26608272,
      "step": 4402,
      "train_runtime": 4463.2544,
      "train_tokens_per_second": 5961.63
    },
    {
      "epoch": 1.6194022988505747,
      "grad_norm": 0.9824703335762024,
      "learning_rate": 3.813730903736425e-05,
      "loss": 1.5987,
      "num_input_tokens_seen": 26613584,
      "step": 4403,
      "train_runtime": 4464.121,
      "train_tokens_per_second": 5961.663
    },
    {
      "epoch": 1.6197701149425288,
      "grad_norm": 0.9766114950180054,
      "learning_rate": 3.810049696300387e-05,
      "loss": 1.5593,
      "num_input_tokens_seen": 26621856,
      "step": 4404,
      "train_runtime": 4465.3987,
      "train_tokens_per_second": 5961.809
    },
    {
      "epoch": 1.6201379310344828,
      "grad_norm": 0.9039485454559326,
      "learning_rate": 3.806368488864348e-05,
      "loss": 1.3344,
      "num_input_tokens_seen": 26628944,
      "step": 4405,
      "train_runtime": 4466.5147,
      "train_tokens_per_second": 5961.907
    },
    {
      "epoch": 1.6205057471264368,
      "grad_norm": 0.9576097726821899,
      "learning_rate": 3.8026872814283086e-05,
      "loss": 1.5583,
      "num_input_tokens_seen": 26635552,
      "step": 4406,
      "train_runtime": 4467.5474,
      "train_tokens_per_second": 5962.008
    },
    {
      "epoch": 1.6208735632183908,
      "grad_norm": 0.9713906645774841,
      "learning_rate": 3.7990060739922695e-05,
      "loss": 1.3574,
      "num_input_tokens_seen": 26645104,
      "step": 4407,
      "train_runtime": 4468.9825,
      "train_tokens_per_second": 5962.231
    },
    {
      "epoch": 1.6212413793103448,
      "grad_norm": 0.9580845236778259,
      "learning_rate": 3.795324866556231e-05,
      "loss": 1.69,
      "num_input_tokens_seen": 26650544,
      "step": 4408,
      "train_runtime": 4469.8951,
      "train_tokens_per_second": 5962.23
    },
    {
      "epoch": 1.6216091954022989,
      "grad_norm": 1.071306824684143,
      "learning_rate": 3.791643659120191e-05,
      "loss": 1.7586,
      "num_input_tokens_seen": 26655168,
      "step": 4409,
      "train_runtime": 4470.6849,
      "train_tokens_per_second": 5962.211
    },
    {
      "epoch": 1.6219770114942529,
      "grad_norm": 0.9921271204948425,
      "learning_rate": 3.787962451684153e-05,
      "loss": 1.4606,
      "num_input_tokens_seen": 26661200,
      "step": 4410,
      "train_runtime": 4471.6953,
      "train_tokens_per_second": 5962.213
    },
    {
      "epoch": 1.622344827586207,
      "grad_norm": 0.8845428228378296,
      "learning_rate": 3.784281244248114e-05,
      "loss": 1.405,
      "num_input_tokens_seen": 26665904,
      "step": 4411,
      "train_runtime": 4473.0522,
      "train_tokens_per_second": 5961.456
    },
    {
      "epoch": 1.622712643678161,
      "grad_norm": 1.1096715927124023,
      "learning_rate": 3.7806000368120746e-05,
      "loss": 1.7312,
      "num_input_tokens_seen": 26672224,
      "step": 4412,
      "train_runtime": 4474.0865,
      "train_tokens_per_second": 5961.49
    },
    {
      "epoch": 1.623080459770115,
      "grad_norm": 0.9747064709663391,
      "learning_rate": 3.7769188293760355e-05,
      "loss": 1.4633,
      "num_input_tokens_seen": 26677312,
      "step": 4413,
      "train_runtime": 4474.9548,
      "train_tokens_per_second": 5961.471
    },
    {
      "epoch": 1.623448275862069,
      "grad_norm": 1.0329118967056274,
      "learning_rate": 3.773237621939997e-05,
      "loss": 1.4654,
      "num_input_tokens_seen": 26682768,
      "step": 4414,
      "train_runtime": 4475.8597,
      "train_tokens_per_second": 5961.484
    },
    {
      "epoch": 1.623816091954023,
      "grad_norm": 1.0991086959838867,
      "learning_rate": 3.769556414503957e-05,
      "loss": 1.6239,
      "num_input_tokens_seen": 26688336,
      "step": 4415,
      "train_runtime": 4476.7963,
      "train_tokens_per_second": 5961.481
    },
    {
      "epoch": 1.624183908045977,
      "grad_norm": 0.9764090776443481,
      "learning_rate": 3.765875207067918e-05,
      "loss": 1.4923,
      "num_input_tokens_seen": 26693968,
      "step": 4416,
      "train_runtime": 4477.7572,
      "train_tokens_per_second": 5961.459
    },
    {
      "epoch": 1.624551724137931,
      "grad_norm": 0.9504784941673279,
      "learning_rate": 3.76219399963188e-05,
      "loss": 1.3885,
      "num_input_tokens_seen": 26699808,
      "step": 4417,
      "train_runtime": 4478.7185,
      "train_tokens_per_second": 5961.484
    },
    {
      "epoch": 1.624919540229885,
      "grad_norm": 1.0261403322219849,
      "learning_rate": 3.75851279219584e-05,
      "loss": 1.626,
      "num_input_tokens_seen": 26705376,
      "step": 4418,
      "train_runtime": 4479.6485,
      "train_tokens_per_second": 5961.489
    },
    {
      "epoch": 1.625287356321839,
      "grad_norm": 0.905576765537262,
      "learning_rate": 3.7548315847598015e-05,
      "loss": 1.3979,
      "num_input_tokens_seen": 26713312,
      "step": 4419,
      "train_runtime": 4480.8761,
      "train_tokens_per_second": 5961.627
    },
    {
      "epoch": 1.625655172413793,
      "grad_norm": 0.9249241352081299,
      "learning_rate": 3.7511503773237624e-05,
      "loss": 1.4806,
      "num_input_tokens_seen": 26719728,
      "step": 4420,
      "train_runtime": 4481.8908,
      "train_tokens_per_second": 5961.709
    },
    {
      "epoch": 1.6260229885057471,
      "grad_norm": 0.9380692839622498,
      "learning_rate": 3.747469169887723e-05,
      "loss": 1.3959,
      "num_input_tokens_seen": 26725760,
      "step": 4421,
      "train_runtime": 4482.8828,
      "train_tokens_per_second": 5961.735
    },
    {
      "epoch": 1.6263908045977011,
      "grad_norm": 0.9099374413490295,
      "learning_rate": 3.743787962451684e-05,
      "loss": 1.3663,
      "num_input_tokens_seen": 26732608,
      "step": 4422,
      "train_runtime": 4483.9632,
      "train_tokens_per_second": 5961.826
    },
    {
      "epoch": 1.6267586206896552,
      "grad_norm": 1.0278503894805908,
      "learning_rate": 3.740106755015646e-05,
      "loss": 1.5807,
      "num_input_tokens_seen": 26737728,
      "step": 4423,
      "train_runtime": 4484.8464,
      "train_tokens_per_second": 5961.794
    },
    {
      "epoch": 1.6271264367816092,
      "grad_norm": 0.9729159474372864,
      "learning_rate": 3.736425547579606e-05,
      "loss": 1.3266,
      "num_input_tokens_seen": 26744544,
      "step": 4424,
      "train_runtime": 4485.9447,
      "train_tokens_per_second": 5961.853
    },
    {
      "epoch": 1.6274942528735632,
      "grad_norm": 1.0604205131530762,
      "learning_rate": 3.7327443401435675e-05,
      "loss": 1.6733,
      "num_input_tokens_seen": 26750528,
      "step": 4425,
      "train_runtime": 4486.9039,
      "train_tokens_per_second": 5961.912
    },
    {
      "epoch": 1.6278620689655172,
      "grad_norm": 1.0183029174804688,
      "learning_rate": 3.7290631327075284e-05,
      "loss": 1.5815,
      "num_input_tokens_seen": 26756368,
      "step": 4426,
      "train_runtime": 4487.845,
      "train_tokens_per_second": 5961.963
    },
    {
      "epoch": 1.6282298850574712,
      "grad_norm": 0.9771733283996582,
      "learning_rate": 3.725381925271489e-05,
      "loss": 1.5414,
      "num_input_tokens_seen": 26761904,
      "step": 4427,
      "train_runtime": 4488.7939,
      "train_tokens_per_second": 5961.936
    },
    {
      "epoch": 1.6285977011494253,
      "grad_norm": 1.031917929649353,
      "learning_rate": 3.72170071783545e-05,
      "loss": 1.5621,
      "num_input_tokens_seen": 26768304,
      "step": 4428,
      "train_runtime": 4489.8412,
      "train_tokens_per_second": 5961.971
    },
    {
      "epoch": 1.6289655172413793,
      "grad_norm": 0.8578279614448547,
      "learning_rate": 3.718019510399412e-05,
      "loss": 1.3516,
      "num_input_tokens_seen": 26778048,
      "step": 4429,
      "train_runtime": 4491.3178,
      "train_tokens_per_second": 5962.181
    },
    {
      "epoch": 1.6293333333333333,
      "grad_norm": 0.9770480990409851,
      "learning_rate": 3.714338302963372e-05,
      "loss": 1.669,
      "num_input_tokens_seen": 26782928,
      "step": 4430,
      "train_runtime": 4492.1658,
      "train_tokens_per_second": 5962.141
    },
    {
      "epoch": 1.6297011494252873,
      "grad_norm": 0.9898492097854614,
      "learning_rate": 3.710657095527333e-05,
      "loss": 1.4968,
      "num_input_tokens_seen": 26788032,
      "step": 4431,
      "train_runtime": 4493.0261,
      "train_tokens_per_second": 5962.136
    },
    {
      "epoch": 1.6300689655172413,
      "grad_norm": 0.9755603671073914,
      "learning_rate": 3.7069758880912944e-05,
      "loss": 1.4318,
      "num_input_tokens_seen": 26794080,
      "step": 4432,
      "train_runtime": 4494.015,
      "train_tokens_per_second": 5962.17
    },
    {
      "epoch": 1.6304367816091954,
      "grad_norm": 0.9433183073997498,
      "learning_rate": 3.7032946806552546e-05,
      "loss": 1.5087,
      "num_input_tokens_seen": 26801088,
      "step": 4433,
      "train_runtime": 4495.1198,
      "train_tokens_per_second": 5962.263
    },
    {
      "epoch": 1.6308045977011494,
      "grad_norm": 1.0346920490264893,
      "learning_rate": 3.699613473219216e-05,
      "loss": 1.4415,
      "num_input_tokens_seen": 26806352,
      "step": 4434,
      "train_runtime": 4496.0123,
      "train_tokens_per_second": 5962.251
    },
    {
      "epoch": 1.6311724137931034,
      "grad_norm": 1.0566402673721313,
      "learning_rate": 3.695932265783177e-05,
      "loss": 1.506,
      "num_input_tokens_seen": 26811504,
      "step": 4435,
      "train_runtime": 4496.8711,
      "train_tokens_per_second": 5962.258
    },
    {
      "epoch": 1.6315402298850574,
      "grad_norm": 1.0768990516662598,
      "learning_rate": 3.692251058347138e-05,
      "loss": 1.796,
      "num_input_tokens_seen": 26815888,
      "step": 4436,
      "train_runtime": 4497.6445,
      "train_tokens_per_second": 5962.207
    },
    {
      "epoch": 1.6319080459770114,
      "grad_norm": 0.9810062050819397,
      "learning_rate": 3.688569850911099e-05,
      "loss": 1.4282,
      "num_input_tokens_seen": 26821120,
      "step": 4437,
      "train_runtime": 4498.4858,
      "train_tokens_per_second": 5962.255
    },
    {
      "epoch": 1.6322758620689655,
      "grad_norm": 1.186820387840271,
      "learning_rate": 3.6848886434750604e-05,
      "loss": 1.6454,
      "num_input_tokens_seen": 26827920,
      "step": 4438,
      "train_runtime": 4499.5946,
      "train_tokens_per_second": 5962.297
    },
    {
      "epoch": 1.6326436781609195,
      "grad_norm": 1.0648199319839478,
      "learning_rate": 3.6812074360390207e-05,
      "loss": 1.5711,
      "num_input_tokens_seen": 26832544,
      "step": 4439,
      "train_runtime": 4500.3915,
      "train_tokens_per_second": 5962.269
    },
    {
      "epoch": 1.6330114942528735,
      "grad_norm": 1.0071207284927368,
      "learning_rate": 3.677526228602982e-05,
      "loss": 1.5084,
      "num_input_tokens_seen": 26838240,
      "step": 4440,
      "train_runtime": 4501.3225,
      "train_tokens_per_second": 5962.301
    },
    {
      "epoch": 1.6333793103448278,
      "grad_norm": 1.106308102607727,
      "learning_rate": 3.673845021166943e-05,
      "loss": 1.5581,
      "num_input_tokens_seen": 26843200,
      "step": 4441,
      "train_runtime": 4502.6366,
      "train_tokens_per_second": 5961.663
    },
    {
      "epoch": 1.6337471264367815,
      "grad_norm": 1.0576567649841309,
      "learning_rate": 3.670163813730904e-05,
      "loss": 1.5525,
      "num_input_tokens_seen": 26849088,
      "step": 4442,
      "train_runtime": 4503.5946,
      "train_tokens_per_second": 5961.702
    },
    {
      "epoch": 1.6341149425287358,
      "grad_norm": 1.1532714366912842,
      "learning_rate": 3.666482606294865e-05,
      "loss": 1.6447,
      "num_input_tokens_seen": 26853984,
      "step": 4443,
      "train_runtime": 4504.4195,
      "train_tokens_per_second": 5961.697
    },
    {
      "epoch": 1.6344827586206896,
      "grad_norm": 1.0651657581329346,
      "learning_rate": 3.6628013988588265e-05,
      "loss": 1.6349,
      "num_input_tokens_seen": 26859424,
      "step": 4444,
      "train_runtime": 4505.332,
      "train_tokens_per_second": 5961.697
    },
    {
      "epoch": 1.6348505747126438,
      "grad_norm": 1.020686388015747,
      "learning_rate": 3.659120191422787e-05,
      "loss": 1.4747,
      "num_input_tokens_seen": 26864080,
      "step": 4445,
      "train_runtime": 4506.1518,
      "train_tokens_per_second": 5961.646
    },
    {
      "epoch": 1.6352183908045976,
      "grad_norm": 0.9999488592147827,
      "learning_rate": 3.6554389839867476e-05,
      "loss": 1.5482,
      "num_input_tokens_seen": 26869824,
      "step": 4446,
      "train_runtime": 4507.0848,
      "train_tokens_per_second": 5961.686
    },
    {
      "epoch": 1.6355862068965519,
      "grad_norm": 0.9598243832588196,
      "learning_rate": 3.651757776550709e-05,
      "loss": 1.4257,
      "num_input_tokens_seen": 26875136,
      "step": 4447,
      "train_runtime": 4507.9572,
      "train_tokens_per_second": 5961.711
    },
    {
      "epoch": 1.6359540229885057,
      "grad_norm": 0.9431926012039185,
      "learning_rate": 3.648076569114669e-05,
      "loss": 1.6416,
      "num_input_tokens_seen": 26879744,
      "step": 4448,
      "train_runtime": 4508.7289,
      "train_tokens_per_second": 5961.712
    },
    {
      "epoch": 1.63632183908046,
      "grad_norm": 1.0127543210983276,
      "learning_rate": 3.644395361678631e-05,
      "loss": 1.5172,
      "num_input_tokens_seen": 26885392,
      "step": 4449,
      "train_runtime": 4509.6398,
      "train_tokens_per_second": 5961.76
    },
    {
      "epoch": 1.6366896551724137,
      "grad_norm": 0.9811985492706299,
      "learning_rate": 3.640714154242592e-05,
      "loss": 1.4665,
      "num_input_tokens_seen": 26891120,
      "step": 4450,
      "train_runtime": 4510.5923,
      "train_tokens_per_second": 5961.771
    },
    {
      "epoch": 1.637057471264368,
      "grad_norm": 0.9507855772972107,
      "learning_rate": 3.637032946806553e-05,
      "loss": 1.5199,
      "num_input_tokens_seen": 26895392,
      "step": 4451,
      "train_runtime": 4511.3329,
      "train_tokens_per_second": 5961.74
    },
    {
      "epoch": 1.6374252873563218,
      "grad_norm": 0.9510373473167419,
      "learning_rate": 3.6333517393705136e-05,
      "loss": 1.5625,
      "num_input_tokens_seen": 26900480,
      "step": 4452,
      "train_runtime": 4512.205,
      "train_tokens_per_second": 5961.715
    },
    {
      "epoch": 1.637793103448276,
      "grad_norm": 1.1824833154678345,
      "learning_rate": 3.629670531934475e-05,
      "loss": 1.5046,
      "num_input_tokens_seen": 26904640,
      "step": 4453,
      "train_runtime": 4512.9334,
      "train_tokens_per_second": 5961.674
    },
    {
      "epoch": 1.6381609195402298,
      "grad_norm": 1.1188271045684814,
      "learning_rate": 3.6259893244984353e-05,
      "loss": 1.5253,
      "num_input_tokens_seen": 26910416,
      "step": 4454,
      "train_runtime": 4513.8995,
      "train_tokens_per_second": 5961.678
    },
    {
      "epoch": 1.638528735632184,
      "grad_norm": 0.9652191400527954,
      "learning_rate": 3.622308117062397e-05,
      "loss": 1.5941,
      "num_input_tokens_seen": 26915776,
      "step": 4455,
      "train_runtime": 4514.803,
      "train_tokens_per_second": 5961.672
    },
    {
      "epoch": 1.6388965517241378,
      "grad_norm": 0.9670843482017517,
      "learning_rate": 3.618626909626358e-05,
      "loss": 1.4658,
      "num_input_tokens_seen": 26920672,
      "step": 4456,
      "train_runtime": 4515.6342,
      "train_tokens_per_second": 5961.659
    },
    {
      "epoch": 1.639264367816092,
      "grad_norm": 0.9543077945709229,
      "learning_rate": 3.614945702190319e-05,
      "loss": 1.5206,
      "num_input_tokens_seen": 26927616,
      "step": 4457,
      "train_runtime": 4516.7216,
      "train_tokens_per_second": 5961.761
    },
    {
      "epoch": 1.6396321839080459,
      "grad_norm": 0.9574422836303711,
      "learning_rate": 3.6112644947542796e-05,
      "loss": 1.4938,
      "num_input_tokens_seen": 26933504,
      "step": 4458,
      "train_runtime": 4517.6965,
      "train_tokens_per_second": 5961.778
    },
    {
      "epoch": 1.6400000000000001,
      "grad_norm": 1.0249989032745361,
      "learning_rate": 3.6075832873182405e-05,
      "loss": 1.3922,
      "num_input_tokens_seen": 26938912,
      "step": 4459,
      "train_runtime": 4518.5855,
      "train_tokens_per_second": 5961.802
    },
    {
      "epoch": 1.640367816091954,
      "grad_norm": 1.000575304031372,
      "learning_rate": 3.6039020798822014e-05,
      "loss": 1.6231,
      "num_input_tokens_seen": 26944336,
      "step": 4460,
      "train_runtime": 4519.4899,
      "train_tokens_per_second": 5961.809
    },
    {
      "epoch": 1.6407356321839082,
      "grad_norm": 0.9081759452819824,
      "learning_rate": 3.600220872446162e-05,
      "loss": 1.4537,
      "num_input_tokens_seen": 26949728,
      "step": 4461,
      "train_runtime": 4520.3946,
      "train_tokens_per_second": 5961.809
    },
    {
      "epoch": 1.641103448275862,
      "grad_norm": 1.0230835676193237,
      "learning_rate": 3.596539665010124e-05,
      "loss": 1.4631,
      "num_input_tokens_seen": 26956144,
      "step": 4462,
      "train_runtime": 4521.432,
      "train_tokens_per_second": 5961.86
    },
    {
      "epoch": 1.6414712643678162,
      "grad_norm": 1.0817731618881226,
      "learning_rate": 3.592858457574084e-05,
      "loss": 1.5918,
      "num_input_tokens_seen": 26960384,
      "step": 4463,
      "train_runtime": 4522.1782,
      "train_tokens_per_second": 5961.814
    },
    {
      "epoch": 1.64183908045977,
      "grad_norm": 1.0232770442962646,
      "learning_rate": 3.5891772501380456e-05,
      "loss": 1.562,
      "num_input_tokens_seen": 26967840,
      "step": 4464,
      "train_runtime": 4523.3499,
      "train_tokens_per_second": 5961.918
    },
    {
      "epoch": 1.6422068965517242,
      "grad_norm": 0.9815859198570251,
      "learning_rate": 3.5854960427020065e-05,
      "loss": 1.3634,
      "num_input_tokens_seen": 26974320,
      "step": 4465,
      "train_runtime": 4524.3783,
      "train_tokens_per_second": 5961.995
    },
    {
      "epoch": 1.642574712643678,
      "grad_norm": 1.0082767009735107,
      "learning_rate": 3.5818148352659674e-05,
      "loss": 1.4737,
      "num_input_tokens_seen": 26980528,
      "step": 4466,
      "train_runtime": 4525.3892,
      "train_tokens_per_second": 5962.035
    },
    {
      "epoch": 1.6429425287356323,
      "grad_norm": 1.064754605293274,
      "learning_rate": 3.578133627829928e-05,
      "loss": 1.6101,
      "num_input_tokens_seen": 26985024,
      "step": 4467,
      "train_runtime": 4526.1731,
      "train_tokens_per_second": 5961.996
    },
    {
      "epoch": 1.643310344827586,
      "grad_norm": 0.9956942200660706,
      "learning_rate": 3.57445242039389e-05,
      "loss": 1.6064,
      "num_input_tokens_seen": 26989680,
      "step": 4468,
      "train_runtime": 4526.9933,
      "train_tokens_per_second": 5961.944
    },
    {
      "epoch": 1.6436781609195403,
      "grad_norm": 1.0067864656448364,
      "learning_rate": 3.57077121295785e-05,
      "loss": 1.4774,
      "num_input_tokens_seen": 26994688,
      "step": 4469,
      "train_runtime": 4527.8574,
      "train_tokens_per_second": 5961.912
    },
    {
      "epoch": 1.6440459770114941,
      "grad_norm": 1.0546534061431885,
      "learning_rate": 3.5670900055218116e-05,
      "loss": 1.6702,
      "num_input_tokens_seen": 26999616,
      "step": 4470,
      "train_runtime": 4528.7021,
      "train_tokens_per_second": 5961.888
    },
    {
      "epoch": 1.6444137931034484,
      "grad_norm": 0.8732433319091797,
      "learning_rate": 3.5634087980857725e-05,
      "loss": 1.5471,
      "num_input_tokens_seen": 27006048,
      "step": 4471,
      "train_runtime": 4530.2771,
      "train_tokens_per_second": 5961.235
    },
    {
      "epoch": 1.6447816091954022,
      "grad_norm": 1.0325344800949097,
      "learning_rate": 3.5597275906497334e-05,
      "loss": 1.5864,
      "num_input_tokens_seen": 27010912,
      "step": 4472,
      "train_runtime": 4531.0977,
      "train_tokens_per_second": 5961.229
    },
    {
      "epoch": 1.6451494252873564,
      "grad_norm": 1.0923030376434326,
      "learning_rate": 3.556046383213694e-05,
      "loss": 1.7019,
      "num_input_tokens_seen": 27016160,
      "step": 4473,
      "train_runtime": 4531.967,
      "train_tokens_per_second": 5961.244
    },
    {
      "epoch": 1.6455172413793102,
      "grad_norm": 1.070400357246399,
      "learning_rate": 3.552365175777655e-05,
      "loss": 1.5514,
      "num_input_tokens_seen": 27020352,
      "step": 4474,
      "train_runtime": 4532.7047,
      "train_tokens_per_second": 5961.198
    },
    {
      "epoch": 1.6458850574712645,
      "grad_norm": 0.8958239555358887,
      "learning_rate": 3.548683968341616e-05,
      "loss": 1.7059,
      "num_input_tokens_seen": 27029008,
      "step": 4475,
      "train_runtime": 4534.0454,
      "train_tokens_per_second": 5961.345
    },
    {
      "epoch": 1.6462528735632183,
      "grad_norm": 0.9670495390892029,
      "learning_rate": 3.545002760905577e-05,
      "loss": 1.355,
      "num_input_tokens_seen": 27037936,
      "step": 4476,
      "train_runtime": 4535.3878,
      "train_tokens_per_second": 5961.549
    },
    {
      "epoch": 1.6466206896551725,
      "grad_norm": 0.9604555368423462,
      "learning_rate": 3.5413215534695385e-05,
      "loss": 1.5953,
      "num_input_tokens_seen": 27043152,
      "step": 4477,
      "train_runtime": 4536.2709,
      "train_tokens_per_second": 5961.538
    },
    {
      "epoch": 1.6469885057471263,
      "grad_norm": 1.004197597503662,
      "learning_rate": 3.537640346033499e-05,
      "loss": 1.6387,
      "num_input_tokens_seen": 27050720,
      "step": 4478,
      "train_runtime": 4537.4482,
      "train_tokens_per_second": 5961.659
    },
    {
      "epoch": 1.6473563218390805,
      "grad_norm": 1.1304956674575806,
      "learning_rate": 3.53395913859746e-05,
      "loss": 1.6723,
      "num_input_tokens_seen": 27056592,
      "step": 4479,
      "train_runtime": 4538.415,
      "train_tokens_per_second": 5961.683
    },
    {
      "epoch": 1.6477241379310343,
      "grad_norm": 1.074472188949585,
      "learning_rate": 3.530277931161421e-05,
      "loss": 1.732,
      "num_input_tokens_seen": 27061808,
      "step": 4480,
      "train_runtime": 4539.2692,
      "train_tokens_per_second": 5961.71
    },
    {
      "epoch": 1.6480919540229886,
      "grad_norm": 1.0209598541259766,
      "learning_rate": 3.526596723725382e-05,
      "loss": 1.6028,
      "num_input_tokens_seen": 27067696,
      "step": 4481,
      "train_runtime": 4540.2666,
      "train_tokens_per_second": 5961.697
    },
    {
      "epoch": 1.6484597701149424,
      "grad_norm": 0.9989323616027832,
      "learning_rate": 3.522915516289343e-05,
      "loss": 1.6453,
      "num_input_tokens_seen": 27072864,
      "step": 4482,
      "train_runtime": 4541.1432,
      "train_tokens_per_second": 5961.685
    },
    {
      "epoch": 1.6488275862068966,
      "grad_norm": 0.9564641118049622,
      "learning_rate": 3.5192343088533045e-05,
      "loss": 1.5845,
      "num_input_tokens_seen": 27077872,
      "step": 4483,
      "train_runtime": 4542.0138,
      "train_tokens_per_second": 5961.645
    },
    {
      "epoch": 1.6491954022988504,
      "grad_norm": 0.9396659135818481,
      "learning_rate": 3.515553101417265e-05,
      "loss": 1.3699,
      "num_input_tokens_seen": 27084944,
      "step": 4484,
      "train_runtime": 4543.1406,
      "train_tokens_per_second": 5961.723
    },
    {
      "epoch": 1.6495632183908047,
      "grad_norm": 1.0293605327606201,
      "learning_rate": 3.511871893981226e-05,
      "loss": 1.3618,
      "num_input_tokens_seen": 27091280,
      "step": 4485,
      "train_runtime": 4544.1504,
      "train_tokens_per_second": 5961.792
    },
    {
      "epoch": 1.6499310344827585,
      "grad_norm": 0.9557163119316101,
      "learning_rate": 3.508190686545187e-05,
      "loss": 1.5235,
      "num_input_tokens_seen": 27097024,
      "step": 4486,
      "train_runtime": 4545.1248,
      "train_tokens_per_second": 5961.778
    },
    {
      "epoch": 1.6502988505747127,
      "grad_norm": 1.0445588827133179,
      "learning_rate": 3.504509479109148e-05,
      "loss": 1.4695,
      "num_input_tokens_seen": 27104592,
      "step": 4487,
      "train_runtime": 4546.2893,
      "train_tokens_per_second": 5961.915
    },
    {
      "epoch": 1.6506666666666665,
      "grad_norm": 0.9958745837211609,
      "learning_rate": 3.500828271673109e-05,
      "loss": 1.4353,
      "num_input_tokens_seen": 27109936,
      "step": 4488,
      "train_runtime": 4547.1857,
      "train_tokens_per_second": 5961.915
    },
    {
      "epoch": 1.6510344827586207,
      "grad_norm": 0.9378944039344788,
      "learning_rate": 3.49714706423707e-05,
      "loss": 1.4088,
      "num_input_tokens_seen": 27118448,
      "step": 4489,
      "train_runtime": 4548.4777,
      "train_tokens_per_second": 5962.093
    },
    {
      "epoch": 1.6514022988505748,
      "grad_norm": 0.9922360777854919,
      "learning_rate": 3.493465856801031e-05,
      "loss": 1.6083,
      "num_input_tokens_seen": 27123120,
      "step": 4490,
      "train_runtime": 4549.3064,
      "train_tokens_per_second": 5962.034
    },
    {
      "epoch": 1.6517701149425288,
      "grad_norm": 1.0193889141082764,
      "learning_rate": 3.4897846493649916e-05,
      "loss": 1.5953,
      "num_input_tokens_seen": 27128992,
      "step": 4491,
      "train_runtime": 4550.2775,
      "train_tokens_per_second": 5962.052
    },
    {
      "epoch": 1.6521379310344828,
      "grad_norm": 0.9122409224510193,
      "learning_rate": 3.486103441928953e-05,
      "loss": 1.4181,
      "num_input_tokens_seen": 27135408,
      "step": 4492,
      "train_runtime": 4551.3094,
      "train_tokens_per_second": 5962.11
    },
    {
      "epoch": 1.6525057471264368,
      "grad_norm": 1.0294233560562134,
      "learning_rate": 3.4824222344929134e-05,
      "loss": 1.491,
      "num_input_tokens_seen": 27143440,
      "step": 4493,
      "train_runtime": 4552.5664,
      "train_tokens_per_second": 5962.228
    },
    {
      "epoch": 1.6528735632183909,
      "grad_norm": 1.0201284885406494,
      "learning_rate": 3.478741027056875e-05,
      "loss": 1.6388,
      "num_input_tokens_seen": 27150048,
      "step": 4494,
      "train_runtime": 4553.6193,
      "train_tokens_per_second": 5962.301
    },
    {
      "epoch": 1.6532413793103449,
      "grad_norm": 0.9764149785041809,
      "learning_rate": 3.475059819620836e-05,
      "loss": 1.7102,
      "num_input_tokens_seen": 27155840,
      "step": 4495,
      "train_runtime": 4554.5587,
      "train_tokens_per_second": 5962.343
    },
    {
      "epoch": 1.653609195402299,
      "grad_norm": 0.9833749532699585,
      "learning_rate": 3.471378612184797e-05,
      "loss": 1.3359,
      "num_input_tokens_seen": 27161280,
      "step": 4496,
      "train_runtime": 4555.4584,
      "train_tokens_per_second": 5962.359
    },
    {
      "epoch": 1.653977011494253,
      "grad_norm": 1.1207014322280884,
      "learning_rate": 3.4676974047487576e-05,
      "loss": 1.737,
      "num_input_tokens_seen": 27165648,
      "step": 4497,
      "train_runtime": 4556.2126,
      "train_tokens_per_second": 5962.331
    },
    {
      "epoch": 1.654344827586207,
      "grad_norm": 0.9498369097709656,
      "learning_rate": 3.4640161973127185e-05,
      "loss": 1.5512,
      "num_input_tokens_seen": 27172608,
      "step": 4498,
      "train_runtime": 4557.3334,
      "train_tokens_per_second": 5962.392
    },
    {
      "epoch": 1.654712643678161,
      "grad_norm": 1.1282031536102295,
      "learning_rate": 3.4603349898766794e-05,
      "loss": 1.5794,
      "num_input_tokens_seen": 27178400,
      "step": 4499,
      "train_runtime": 4558.2715,
      "train_tokens_per_second": 5962.435
    },
    {
      "epoch": 1.655080459770115,
      "grad_norm": 0.8588025569915771,
      "learning_rate": 3.456653782440641e-05,
      "loss": 1.3368,
      "num_input_tokens_seen": 27185744,
      "step": 4500,
      "train_runtime": 4559.3953,
      "train_tokens_per_second": 5962.577
    },
    {
      "epoch": 1.655448275862069,
      "grad_norm": 0.8851767182350159,
      "learning_rate": 3.452972575004602e-05,
      "loss": 1.5944,
      "num_input_tokens_seen": 27193744,
      "step": 4501,
      "train_runtime": 4561.202,
      "train_tokens_per_second": 5961.969
    },
    {
      "epoch": 1.655816091954023,
      "grad_norm": 0.9626141786575317,
      "learning_rate": 3.449291367568563e-05,
      "loss": 1.452,
      "num_input_tokens_seen": 27198736,
      "step": 4502,
      "train_runtime": 4562.0731,
      "train_tokens_per_second": 5961.925
    },
    {
      "epoch": 1.656183908045977,
      "grad_norm": 1.0072157382965088,
      "learning_rate": 3.445610160132524e-05,
      "loss": 1.4744,
      "num_input_tokens_seen": 27205296,
      "step": 4503,
      "train_runtime": 4563.1053,
      "train_tokens_per_second": 5962.014
    },
    {
      "epoch": 1.656551724137931,
      "grad_norm": 0.9441608786582947,
      "learning_rate": 3.4419289526964846e-05,
      "loss": 1.5023,
      "num_input_tokens_seen": 27211696,
      "step": 4504,
      "train_runtime": 4564.164,
      "train_tokens_per_second": 5962.033
    },
    {
      "epoch": 1.656919540229885,
      "grad_norm": 1.0220075845718384,
      "learning_rate": 3.4382477452604454e-05,
      "loss": 1.4444,
      "num_input_tokens_seen": 27217408,
      "step": 4505,
      "train_runtime": 4565.1009,
      "train_tokens_per_second": 5962.061
    },
    {
      "epoch": 1.657287356321839,
      "grad_norm": 1.0067750215530396,
      "learning_rate": 3.434566537824406e-05,
      "loss": 1.6475,
      "num_input_tokens_seen": 27225088,
      "step": 4506,
      "train_runtime": 4566.3162,
      "train_tokens_per_second": 5962.156
    },
    {
      "epoch": 1.6576551724137931,
      "grad_norm": 0.9772936105728149,
      "learning_rate": 3.430885330388368e-05,
      "loss": 1.4659,
      "num_input_tokens_seen": 27230080,
      "step": 4507,
      "train_runtime": 4567.1652,
      "train_tokens_per_second": 5962.14
    },
    {
      "epoch": 1.6580229885057471,
      "grad_norm": 0.877511739730835,
      "learning_rate": 3.427204122952328e-05,
      "loss": 1.3628,
      "num_input_tokens_seen": 27237840,
      "step": 4508,
      "train_runtime": 4568.3647,
      "train_tokens_per_second": 5962.274
    },
    {
      "epoch": 1.6583908045977012,
      "grad_norm": 1.0704823732376099,
      "learning_rate": 3.42352291551629e-05,
      "loss": 1.4692,
      "num_input_tokens_seen": 27242032,
      "step": 4509,
      "train_runtime": 4569.13,
      "train_tokens_per_second": 5962.192
    },
    {
      "epoch": 1.6587586206896552,
      "grad_norm": 0.9791299700737,
      "learning_rate": 3.4198417080802506e-05,
      "loss": 1.459,
      "num_input_tokens_seen": 27248640,
      "step": 4510,
      "train_runtime": 4570.1798,
      "train_tokens_per_second": 5962.269
    },
    {
      "epoch": 1.6591264367816092,
      "grad_norm": 1.028090000152588,
      "learning_rate": 3.4161605006442115e-05,
      "loss": 1.5709,
      "num_input_tokens_seen": 27253840,
      "step": 4511,
      "train_runtime": 4571.0393,
      "train_tokens_per_second": 5962.285
    },
    {
      "epoch": 1.6594942528735632,
      "grad_norm": 1.0077334642410278,
      "learning_rate": 3.4124792932081723e-05,
      "loss": 1.5776,
      "num_input_tokens_seen": 27259248,
      "step": 4512,
      "train_runtime": 4571.9551,
      "train_tokens_per_second": 5962.274
    },
    {
      "epoch": 1.6598620689655172,
      "grad_norm": 1.001970887184143,
      "learning_rate": 3.408798085772133e-05,
      "loss": 1.5236,
      "num_input_tokens_seen": 27264400,
      "step": 4513,
      "train_runtime": 4572.8161,
      "train_tokens_per_second": 5962.278
    },
    {
      "epoch": 1.6602298850574713,
      "grad_norm": 0.9177581667900085,
      "learning_rate": 3.405116878336094e-05,
      "loss": 1.5772,
      "num_input_tokens_seen": 27270816,
      "step": 4514,
      "train_runtime": 4573.86,
      "train_tokens_per_second": 5962.32
    },
    {
      "epoch": 1.6605977011494253,
      "grad_norm": 1.0000431537628174,
      "learning_rate": 3.401435670900056e-05,
      "loss": 1.5973,
      "num_input_tokens_seen": 27276368,
      "step": 4515,
      "train_runtime": 4574.785,
      "train_tokens_per_second": 5962.328
    },
    {
      "epoch": 1.6609655172413793,
      "grad_norm": 0.948546290397644,
      "learning_rate": 3.3977544634640166e-05,
      "loss": 1.5555,
      "num_input_tokens_seen": 27285008,
      "step": 4516,
      "train_runtime": 4576.1336,
      "train_tokens_per_second": 5962.459
    },
    {
      "epoch": 1.6613333333333333,
      "grad_norm": 0.9299002289772034,
      "learning_rate": 3.3940732560279775e-05,
      "loss": 1.1601,
      "num_input_tokens_seen": 27294912,
      "step": 4517,
      "train_runtime": 4577.6202,
      "train_tokens_per_second": 5962.686
    },
    {
      "epoch": 1.6617011494252873,
      "grad_norm": 1.0480867624282837,
      "learning_rate": 3.3903920485919384e-05,
      "loss": 1.6586,
      "num_input_tokens_seen": 27300704,
      "step": 4518,
      "train_runtime": 4578.5851,
      "train_tokens_per_second": 5962.695
    },
    {
      "epoch": 1.6620689655172414,
      "grad_norm": 1.0556780099868774,
      "learning_rate": 3.386710841155899e-05,
      "loss": 1.6269,
      "num_input_tokens_seen": 27306944,
      "step": 4519,
      "train_runtime": 4579.6053,
      "train_tokens_per_second": 5962.729
    },
    {
      "epoch": 1.6624367816091954,
      "grad_norm": 0.9771064519882202,
      "learning_rate": 3.38302963371986e-05,
      "loss": 1.44,
      "num_input_tokens_seen": 27313024,
      "step": 4520,
      "train_runtime": 4580.5616,
      "train_tokens_per_second": 5962.811
    },
    {
      "epoch": 1.6628045977011494,
      "grad_norm": 1.0678390264511108,
      "learning_rate": 3.379348426283821e-05,
      "loss": 1.6643,
      "num_input_tokens_seen": 27317760,
      "step": 4521,
      "train_runtime": 4581.369,
      "train_tokens_per_second": 5962.794
    },
    {
      "epoch": 1.6631724137931034,
      "grad_norm": 1.0714067220687866,
      "learning_rate": 3.3756672188477826e-05,
      "loss": 1.5341,
      "num_input_tokens_seen": 27324032,
      "step": 4522,
      "train_runtime": 4582.3807,
      "train_tokens_per_second": 5962.846
    },
    {
      "epoch": 1.6635402298850575,
      "grad_norm": 1.062416911125183,
      "learning_rate": 3.371986011411743e-05,
      "loss": 1.4978,
      "num_input_tokens_seen": 27330608,
      "step": 4523,
      "train_runtime": 4583.4693,
      "train_tokens_per_second": 5962.865
    },
    {
      "epoch": 1.6639080459770115,
      "grad_norm": 1.0742371082305908,
      "learning_rate": 3.3683048039757044e-05,
      "loss": 1.7435,
      "num_input_tokens_seen": 27336736,
      "step": 4524,
      "train_runtime": 4584.4696,
      "train_tokens_per_second": 5962.9
    },
    {
      "epoch": 1.6642758620689655,
      "grad_norm": 0.9882484078407288,
      "learning_rate": 3.364623596539665e-05,
      "loss": 1.4981,
      "num_input_tokens_seen": 27342064,
      "step": 4525,
      "train_runtime": 4585.3458,
      "train_tokens_per_second": 5962.923
    },
    {
      "epoch": 1.6646436781609195,
      "grad_norm": 1.0915501117706299,
      "learning_rate": 3.360942389103626e-05,
      "loss": 1.5119,
      "num_input_tokens_seen": 27347328,
      "step": 4526,
      "train_runtime": 4586.2256,
      "train_tokens_per_second": 5962.927
    },
    {
      "epoch": 1.6650114942528735,
      "grad_norm": 0.9395991563796997,
      "learning_rate": 3.357261181667587e-05,
      "loss": 1.4777,
      "num_input_tokens_seen": 27354368,
      "step": 4527,
      "train_runtime": 4587.3364,
      "train_tokens_per_second": 5963.018
    },
    {
      "epoch": 1.6653793103448276,
      "grad_norm": 1.0218963623046875,
      "learning_rate": 3.353579974231548e-05,
      "loss": 1.5474,
      "num_input_tokens_seen": 27359200,
      "step": 4528,
      "train_runtime": 4588.1598,
      "train_tokens_per_second": 5963.001
    },
    {
      "epoch": 1.6657471264367816,
      "grad_norm": 1.0213508605957031,
      "learning_rate": 3.349898766795509e-05,
      "loss": 1.5227,
      "num_input_tokens_seen": 27363600,
      "step": 4529,
      "train_runtime": 4588.9332,
      "train_tokens_per_second": 5962.955
    },
    {
      "epoch": 1.6661149425287356,
      "grad_norm": 0.9922882914543152,
      "learning_rate": 3.3462175593594704e-05,
      "loss": 1.2771,
      "num_input_tokens_seen": 27369536,
      "step": 4530,
      "train_runtime": 4589.9048,
      "train_tokens_per_second": 5962.986
    },
    {
      "epoch": 1.6664827586206896,
      "grad_norm": 1.1102005243301392,
      "learning_rate": 3.342536351923431e-05,
      "loss": 1.6434,
      "num_input_tokens_seen": 27376640,
      "step": 4531,
      "train_runtime": 4591.5539,
      "train_tokens_per_second": 5962.391
    },
    {
      "epoch": 1.6668505747126436,
      "grad_norm": 0.9789210557937622,
      "learning_rate": 3.338855144487392e-05,
      "loss": 1.6367,
      "num_input_tokens_seen": 27382944,
      "step": 4532,
      "train_runtime": 4592.5938,
      "train_tokens_per_second": 5962.414
    },
    {
      "epoch": 1.6672183908045977,
      "grad_norm": 0.9995959401130676,
      "learning_rate": 3.335173937051353e-05,
      "loss": 1.5033,
      "num_input_tokens_seen": 27388912,
      "step": 4533,
      "train_runtime": 4593.5537,
      "train_tokens_per_second": 5962.467
    },
    {
      "epoch": 1.6675862068965517,
      "grad_norm": 0.8130632042884827,
      "learning_rate": 3.331492729615314e-05,
      "loss": 1.4627,
      "num_input_tokens_seen": 27398048,
      "step": 4534,
      "train_runtime": 4594.9871,
      "train_tokens_per_second": 5962.595
    },
    {
      "epoch": 1.6679540229885057,
      "grad_norm": 1.0611449480056763,
      "learning_rate": 3.327811522179275e-05,
      "loss": 1.6465,
      "num_input_tokens_seen": 27402848,
      "step": 4535,
      "train_runtime": 4595.8179,
      "train_tokens_per_second": 5962.562
    },
    {
      "epoch": 1.6683218390804597,
      "grad_norm": 0.9352962970733643,
      "learning_rate": 3.3241303147432364e-05,
      "loss": 1.3001,
      "num_input_tokens_seen": 27410640,
      "step": 4536,
      "train_runtime": 4597.0483,
      "train_tokens_per_second": 5962.661
    },
    {
      "epoch": 1.6686896551724137,
      "grad_norm": 1.0341665744781494,
      "learning_rate": 3.3204491073071966e-05,
      "loss": 1.489,
      "num_input_tokens_seen": 27415680,
      "step": 4537,
      "train_runtime": 4597.902,
      "train_tokens_per_second": 5962.65
    },
    {
      "epoch": 1.6690574712643678,
      "grad_norm": 0.9031906723976135,
      "learning_rate": 3.3167678998711575e-05,
      "loss": 1.3862,
      "num_input_tokens_seen": 27427280,
      "step": 4538,
      "train_runtime": 4599.6234,
      "train_tokens_per_second": 5962.94
    },
    {
      "epoch": 1.6694252873563218,
      "grad_norm": 1.0327389240264893,
      "learning_rate": 3.313086692435119e-05,
      "loss": 1.4689,
      "num_input_tokens_seen": 27432752,
      "step": 4539,
      "train_runtime": 4600.5086,
      "train_tokens_per_second": 5962.982
    },
    {
      "epoch": 1.669793103448276,
      "grad_norm": 1.150347113609314,
      "learning_rate": 3.30940548499908e-05,
      "loss": 1.6378,
      "num_input_tokens_seen": 27438048,
      "step": 4540,
      "train_runtime": 4601.3783,
      "train_tokens_per_second": 5963.006
    },
    {
      "epoch": 1.6701609195402298,
      "grad_norm": 1.0720256567001343,
      "learning_rate": 3.305724277563041e-05,
      "loss": 1.7241,
      "num_input_tokens_seen": 27442720,
      "step": 4541,
      "train_runtime": 4602.1848,
      "train_tokens_per_second": 5962.977
    },
    {
      "epoch": 1.670528735632184,
      "grad_norm": 1.027369499206543,
      "learning_rate": 3.302043070127002e-05,
      "loss": 1.364,
      "num_input_tokens_seen": 27450256,
      "step": 4542,
      "train_runtime": 4603.3852,
      "train_tokens_per_second": 5963.059
    },
    {
      "epoch": 1.6708965517241379,
      "grad_norm": 0.9748852252960205,
      "learning_rate": 3.2983618626909626e-05,
      "loss": 1.424,
      "num_input_tokens_seen": 27458032,
      "step": 4543,
      "train_runtime": 4604.6186,
      "train_tokens_per_second": 5963.15
    },
    {
      "epoch": 1.6712643678160921,
      "grad_norm": 0.936853289604187,
      "learning_rate": 3.2946806552549235e-05,
      "loss": 1.477,
      "num_input_tokens_seen": 27465344,
      "step": 4544,
      "train_runtime": 4605.7982,
      "train_tokens_per_second": 5963.211
    },
    {
      "epoch": 1.671632183908046,
      "grad_norm": 0.9770433902740479,
      "learning_rate": 3.290999447818885e-05,
      "loss": 1.4245,
      "num_input_tokens_seen": 27470656,
      "step": 4545,
      "train_runtime": 4606.681,
      "train_tokens_per_second": 5963.221
    },
    {
      "epoch": 1.6720000000000002,
      "grad_norm": 0.9963710308074951,
      "learning_rate": 3.287318240382846e-05,
      "loss": 1.4901,
      "num_input_tokens_seen": 27476336,
      "step": 4546,
      "train_runtime": 4607.6258,
      "train_tokens_per_second": 5963.231
    },
    {
      "epoch": 1.672367816091954,
      "grad_norm": 1.0379221439361572,
      "learning_rate": 3.283637032946807e-05,
      "loss": 1.634,
      "num_input_tokens_seen": 27481456,
      "step": 4547,
      "train_runtime": 4608.5017,
      "train_tokens_per_second": 5963.208
    },
    {
      "epoch": 1.6727356321839082,
      "grad_norm": 1.0403231382369995,
      "learning_rate": 3.279955825510768e-05,
      "loss": 1.5957,
      "num_input_tokens_seen": 27486112,
      "step": 4548,
      "train_runtime": 4609.2827,
      "train_tokens_per_second": 5963.208
    },
    {
      "epoch": 1.673103448275862,
      "grad_norm": 1.0611093044281006,
      "learning_rate": 3.2762746180747286e-05,
      "loss": 1.6517,
      "num_input_tokens_seen": 27491552,
      "step": 4549,
      "train_runtime": 4610.1888,
      "train_tokens_per_second": 5963.216
    },
    {
      "epoch": 1.6734712643678162,
      "grad_norm": 1.013532042503357,
      "learning_rate": 3.2725934106386895e-05,
      "loss": 1.4266,
      "num_input_tokens_seen": 27497776,
      "step": 4550,
      "train_runtime": 4611.1924,
      "train_tokens_per_second": 5963.268
    },
    {
      "epoch": 1.67383908045977,
      "grad_norm": 0.9836187362670898,
      "learning_rate": 3.268912203202651e-05,
      "loss": 1.486,
      "num_input_tokens_seen": 27506032,
      "step": 4551,
      "train_runtime": 4612.4686,
      "train_tokens_per_second": 5963.408
    },
    {
      "epoch": 1.6742068965517243,
      "grad_norm": 0.9497100710868835,
      "learning_rate": 3.265230995766611e-05,
      "loss": 1.6527,
      "num_input_tokens_seen": 27514528,
      "step": 4552,
      "train_runtime": 4613.7966,
      "train_tokens_per_second": 5963.533
    },
    {
      "epoch": 1.674574712643678,
      "grad_norm": 0.9267294406890869,
      "learning_rate": 3.261549788330572e-05,
      "loss": 1.3514,
      "num_input_tokens_seen": 27521008,
      "step": 4553,
      "train_runtime": 4614.8389,
      "train_tokens_per_second": 5963.59
    },
    {
      "epoch": 1.6749425287356323,
      "grad_norm": 0.9495742917060852,
      "learning_rate": 3.257868580894534e-05,
      "loss": 1.4508,
      "num_input_tokens_seen": 27527216,
      "step": 4554,
      "train_runtime": 4615.8481,
      "train_tokens_per_second": 5963.631
    },
    {
      "epoch": 1.6753103448275861,
      "grad_norm": 0.9232713580131531,
      "learning_rate": 3.2541873734584946e-05,
      "loss": 1.4087,
      "num_input_tokens_seen": 27532752,
      "step": 4555,
      "train_runtime": 4616.7654,
      "train_tokens_per_second": 5963.645
    },
    {
      "epoch": 1.6756781609195404,
      "grad_norm": 0.9398287534713745,
      "learning_rate": 3.2505061660224555e-05,
      "loss": 1.34,
      "num_input_tokens_seen": 27537792,
      "step": 4556,
      "train_runtime": 4617.6334,
      "train_tokens_per_second": 5963.616
    },
    {
      "epoch": 1.6760459770114942,
      "grad_norm": 0.9533345103263855,
      "learning_rate": 3.2468249585864164e-05,
      "loss": 1.3681,
      "num_input_tokens_seen": 27545008,
      "step": 4557,
      "train_runtime": 4618.7945,
      "train_tokens_per_second": 5963.679
    },
    {
      "epoch": 1.6764137931034484,
      "grad_norm": 0.9850859045982361,
      "learning_rate": 3.243143751150377e-05,
      "loss": 1.4773,
      "num_input_tokens_seen": 27550256,
      "step": 4558,
      "train_runtime": 4619.6806,
      "train_tokens_per_second": 5963.671
    },
    {
      "epoch": 1.6767816091954022,
      "grad_norm": 1.0518481731414795,
      "learning_rate": 3.239462543714338e-05,
      "loss": 1.5096,
      "num_input_tokens_seen": 27554992,
      "step": 4559,
      "train_runtime": 4620.482,
      "train_tokens_per_second": 5963.662
    },
    {
      "epoch": 1.6771494252873564,
      "grad_norm": 0.9615206718444824,
      "learning_rate": 3.2357813362783e-05,
      "loss": 1.3367,
      "num_input_tokens_seen": 27569408,
      "step": 4560,
      "train_runtime": 4622.5843,
      "train_tokens_per_second": 5964.068
    },
    {
      "epoch": 1.6775172413793102,
      "grad_norm": 1.047471046447754,
      "learning_rate": 3.2321001288422607e-05,
      "loss": 1.6531,
      "num_input_tokens_seen": 27574208,
      "step": 4561,
      "train_runtime": 4623.8667,
      "train_tokens_per_second": 5963.452
    },
    {
      "epoch": 1.6778850574712645,
      "grad_norm": 0.9237154126167297,
      "learning_rate": 3.2284189214062215e-05,
      "loss": 1.3022,
      "num_input_tokens_seen": 27579520,
      "step": 4562,
      "train_runtime": 4624.7637,
      "train_tokens_per_second": 5963.444
    },
    {
      "epoch": 1.6782528735632183,
      "grad_norm": 1.0633589029312134,
      "learning_rate": 3.2247377139701824e-05,
      "loss": 1.6763,
      "num_input_tokens_seen": 27585184,
      "step": 4563,
      "train_runtime": 4625.699,
      "train_tokens_per_second": 5963.463
    },
    {
      "epoch": 1.6786206896551725,
      "grad_norm": 0.9793161749839783,
      "learning_rate": 3.221056506534143e-05,
      "loss": 1.4456,
      "num_input_tokens_seen": 27590432,
      "step": 4564,
      "train_runtime": 4626.5867,
      "train_tokens_per_second": 5963.453
    },
    {
      "epoch": 1.6789885057471263,
      "grad_norm": 0.9934513568878174,
      "learning_rate": 3.217375299098104e-05,
      "loss": 1.3779,
      "num_input_tokens_seen": 27596944,
      "step": 4565,
      "train_runtime": 4627.6444,
      "train_tokens_per_second": 5963.497
    },
    {
      "epoch": 1.6793563218390806,
      "grad_norm": 1.0272324085235596,
      "learning_rate": 3.213694091662066e-05,
      "loss": 1.6104,
      "num_input_tokens_seen": 27601360,
      "step": 4566,
      "train_runtime": 4628.407,
      "train_tokens_per_second": 5963.469
    },
    {
      "epoch": 1.6797241379310344,
      "grad_norm": 1.1488995552062988,
      "learning_rate": 3.210012884226026e-05,
      "loss": 1.7001,
      "num_input_tokens_seen": 27606032,
      "step": 4567,
      "train_runtime": 4629.2018,
      "train_tokens_per_second": 5963.454
    },
    {
      "epoch": 1.6800919540229886,
      "grad_norm": 1.0806998014450073,
      "learning_rate": 3.206331676789987e-05,
      "loss": 1.4357,
      "num_input_tokens_seen": 27610368,
      "step": 4568,
      "train_runtime": 4629.9715,
      "train_tokens_per_second": 5963.399
    },
    {
      "epoch": 1.6804597701149424,
      "grad_norm": 0.9658429622650146,
      "learning_rate": 3.2026504693539484e-05,
      "loss": 1.4371,
      "num_input_tokens_seen": 27618576,
      "step": 4569,
      "train_runtime": 4631.2459,
      "train_tokens_per_second": 5963.53
    },
    {
      "epoch": 1.6808275862068967,
      "grad_norm": 1.0064529180526733,
      "learning_rate": 3.198969261917909e-05,
      "loss": 1.6742,
      "num_input_tokens_seen": 27624656,
      "step": 4570,
      "train_runtime": 4632.2573,
      "train_tokens_per_second": 5963.541
    },
    {
      "epoch": 1.6811954022988504,
      "grad_norm": 0.9576125741004944,
      "learning_rate": 3.19528805448187e-05,
      "loss": 1.4484,
      "num_input_tokens_seen": 27630400,
      "step": 4571,
      "train_runtime": 4633.2039,
      "train_tokens_per_second": 5963.562
    },
    {
      "epoch": 1.6815632183908047,
      "grad_norm": 0.9752952456474304,
      "learning_rate": 3.191606847045831e-05,
      "loss": 1.3325,
      "num_input_tokens_seen": 27637360,
      "step": 4572,
      "train_runtime": 4634.2836,
      "train_tokens_per_second": 5963.675
    },
    {
      "epoch": 1.6819310344827585,
      "grad_norm": 0.9824012517929077,
      "learning_rate": 3.187925639609792e-05,
      "loss": 1.5745,
      "num_input_tokens_seen": 27643104,
      "step": 4573,
      "train_runtime": 4635.23,
      "train_tokens_per_second": 5963.696
    },
    {
      "epoch": 1.6822988505747127,
      "grad_norm": 0.9704250693321228,
      "learning_rate": 3.184244432173753e-05,
      "loss": 1.5435,
      "num_input_tokens_seen": 27648240,
      "step": 4574,
      "train_runtime": 4636.0737,
      "train_tokens_per_second": 5963.719
    },
    {
      "epoch": 1.6826666666666665,
      "grad_norm": 0.8644093871116638,
      "learning_rate": 3.1805632247377145e-05,
      "loss": 1.3439,
      "num_input_tokens_seen": 27658352,
      "step": 4575,
      "train_runtime": 4637.6061,
      "train_tokens_per_second": 5963.929
    },
    {
      "epoch": 1.6830344827586208,
      "grad_norm": 0.9795041084289551,
      "learning_rate": 3.176882017301675e-05,
      "loss": 1.2999,
      "num_input_tokens_seen": 27667440,
      "step": 4576,
      "train_runtime": 4638.9711,
      "train_tokens_per_second": 5964.133
    },
    {
      "epoch": 1.6834022988505746,
      "grad_norm": 1.0225485563278198,
      "learning_rate": 3.173200809865636e-05,
      "loss": 1.5607,
      "num_input_tokens_seen": 27672688,
      "step": 4577,
      "train_runtime": 4639.8534,
      "train_tokens_per_second": 5964.13
    },
    {
      "epoch": 1.6837701149425288,
      "grad_norm": 0.9260286688804626,
      "learning_rate": 3.169519602429597e-05,
      "loss": 1.5856,
      "num_input_tokens_seen": 27677312,
      "step": 4578,
      "train_runtime": 4640.6708,
      "train_tokens_per_second": 5964.076
    },
    {
      "epoch": 1.6841379310344826,
      "grad_norm": 0.9811348915100098,
      "learning_rate": 3.165838394993558e-05,
      "loss": 1.4791,
      "num_input_tokens_seen": 27685008,
      "step": 4579,
      "train_runtime": 4641.8866,
      "train_tokens_per_second": 5964.172
    },
    {
      "epoch": 1.6845057471264369,
      "grad_norm": 0.9541542530059814,
      "learning_rate": 3.162157187557519e-05,
      "loss": 1.5143,
      "num_input_tokens_seen": 27692064,
      "step": 4580,
      "train_runtime": 4643.023,
      "train_tokens_per_second": 5964.231
    },
    {
      "epoch": 1.6848735632183907,
      "grad_norm": 1.0133819580078125,
      "learning_rate": 3.1584759801214805e-05,
      "loss": 1.3716,
      "num_input_tokens_seen": 27699904,
      "step": 4581,
      "train_runtime": 4644.245,
      "train_tokens_per_second": 5964.35
    },
    {
      "epoch": 1.685241379310345,
      "grad_norm": 0.9617471694946289,
      "learning_rate": 3.154794772685441e-05,
      "loss": 1.6089,
      "num_input_tokens_seen": 27705440,
      "step": 4582,
      "train_runtime": 4645.173,
      "train_tokens_per_second": 5964.351
    },
    {
      "epoch": 1.6856091954022987,
      "grad_norm": 0.9849011301994324,
      "learning_rate": 3.1511135652494016e-05,
      "loss": 1.2325,
      "num_input_tokens_seen": 27711296,
      "step": 4583,
      "train_runtime": 4646.139,
      "train_tokens_per_second": 5964.371
    },
    {
      "epoch": 1.685977011494253,
      "grad_norm": 1.1416987180709839,
      "learning_rate": 3.147432357813363e-05,
      "loss": 1.7682,
      "num_input_tokens_seen": 27716736,
      "step": 4584,
      "train_runtime": 4647.0316,
      "train_tokens_per_second": 5964.396
    },
    {
      "epoch": 1.6863448275862067,
      "grad_norm": 1.0593440532684326,
      "learning_rate": 3.143751150377324e-05,
      "loss": 1.3654,
      "num_input_tokens_seen": 27721952,
      "step": 4585,
      "train_runtime": 4647.8933,
      "train_tokens_per_second": 5964.412
    },
    {
      "epoch": 1.686712643678161,
      "grad_norm": 1.015592336654663,
      "learning_rate": 3.140069942941285e-05,
      "loss": 1.3702,
      "num_input_tokens_seen": 27729296,
      "step": 4586,
      "train_runtime": 4649.0441,
      "train_tokens_per_second": 5964.516
    },
    {
      "epoch": 1.6870804597701148,
      "grad_norm": 0.995673418045044,
      "learning_rate": 3.136388735505246e-05,
      "loss": 1.5328,
      "num_input_tokens_seen": 27734256,
      "step": 4587,
      "train_runtime": 4649.887,
      "train_tokens_per_second": 5964.501
    },
    {
      "epoch": 1.687448275862069,
      "grad_norm": 0.9884728193283081,
      "learning_rate": 3.132707528069207e-05,
      "loss": 1.381,
      "num_input_tokens_seen": 27739616,
      "step": 4588,
      "train_runtime": 4650.7772,
      "train_tokens_per_second": 5964.512
    },
    {
      "epoch": 1.6878160919540228,
      "grad_norm": 1.022894024848938,
      "learning_rate": 3.1290263206331676e-05,
      "loss": 1.4292,
      "num_input_tokens_seen": 27745376,
      "step": 4589,
      "train_runtime": 4651.7295,
      "train_tokens_per_second": 5964.529
    },
    {
      "epoch": 1.688183908045977,
      "grad_norm": 0.9556910991668701,
      "learning_rate": 3.125345113197129e-05,
      "loss": 1.4345,
      "num_input_tokens_seen": 27750096,
      "step": 4590,
      "train_runtime": 4652.547,
      "train_tokens_per_second": 5964.496
    },
    {
      "epoch": 1.688551724137931,
      "grad_norm": 1.0383857488632202,
      "learning_rate": 3.1216639057610894e-05,
      "loss": 1.5837,
      "num_input_tokens_seen": 27755584,
      "step": 4591,
      "train_runtime": 4653.9462,
      "train_tokens_per_second": 5963.882
    },
    {
      "epoch": 1.688919540229885,
      "grad_norm": 1.0677129030227661,
      "learning_rate": 3.117982698325051e-05,
      "loss": 1.3866,
      "num_input_tokens_seen": 27760912,
      "step": 4592,
      "train_runtime": 4654.8228,
      "train_tokens_per_second": 5963.903
    },
    {
      "epoch": 1.6892873563218391,
      "grad_norm": 1.0819765329360962,
      "learning_rate": 3.114301490889012e-05,
      "loss": 1.6699,
      "num_input_tokens_seen": 27766544,
      "step": 4593,
      "train_runtime": 4655.741,
      "train_tokens_per_second": 5963.937
    },
    {
      "epoch": 1.6896551724137931,
      "grad_norm": 1.055228352546692,
      "learning_rate": 3.110620283452973e-05,
      "loss": 1.4852,
      "num_input_tokens_seen": 27772272,
      "step": 4594,
      "train_runtime": 4656.7158,
      "train_tokens_per_second": 5963.918
    },
    {
      "epoch": 1.6900229885057472,
      "grad_norm": 1.12703275680542,
      "learning_rate": 3.1069390760169336e-05,
      "loss": 1.3859,
      "num_input_tokens_seen": 27776720,
      "step": 4595,
      "train_runtime": 4657.4677,
      "train_tokens_per_second": 5963.91
    },
    {
      "epoch": 1.6903908045977012,
      "grad_norm": 1.071090579032898,
      "learning_rate": 3.103257868580895e-05,
      "loss": 1.6972,
      "num_input_tokens_seen": 27782896,
      "step": 4596,
      "train_runtime": 4658.474,
      "train_tokens_per_second": 5963.948
    },
    {
      "epoch": 1.6907586206896552,
      "grad_norm": 0.9115914106369019,
      "learning_rate": 3.0995766611448554e-05,
      "loss": 1.3728,
      "num_input_tokens_seen": 27790224,
      "step": 4597,
      "train_runtime": 4659.6279,
      "train_tokens_per_second": 5964.044
    },
    {
      "epoch": 1.6911264367816092,
      "grad_norm": 0.994540810585022,
      "learning_rate": 3.095895453708816e-05,
      "loss": 1.4326,
      "num_input_tokens_seen": 27797280,
      "step": 4598,
      "train_runtime": 4660.7469,
      "train_tokens_per_second": 5964.126
    },
    {
      "epoch": 1.6914942528735633,
      "grad_norm": 0.9558801651000977,
      "learning_rate": 3.092214246272778e-05,
      "loss": 1.3787,
      "num_input_tokens_seen": 27804352,
      "step": 4599,
      "train_runtime": 4661.8694,
      "train_tokens_per_second": 5964.207
    },
    {
      "epoch": 1.6918620689655173,
      "grad_norm": 0.9633522629737854,
      "learning_rate": 3.088533038836739e-05,
      "loss": 1.5573,
      "num_input_tokens_seen": 27810672,
      "step": 4600,
      "train_runtime": 4662.8929,
      "train_tokens_per_second": 5964.253
    },
    {
      "epoch": 1.6922298850574713,
      "grad_norm": 1.0789103507995605,
      "learning_rate": 3.0848518314006996e-05,
      "loss": 1.6742,
      "num_input_tokens_seen": 27817584,
      "step": 4601,
      "train_runtime": 4663.9774,
      "train_tokens_per_second": 5964.348
    },
    {
      "epoch": 1.6925977011494253,
      "grad_norm": 0.9825541973114014,
      "learning_rate": 3.0811706239646605e-05,
      "loss": 1.4961,
      "num_input_tokens_seen": 27822640,
      "step": 4602,
      "train_runtime": 4664.8262,
      "train_tokens_per_second": 5964.347
    },
    {
      "epoch": 1.6929655172413793,
      "grad_norm": 1.253936529159546,
      "learning_rate": 3.0774894165286214e-05,
      "loss": 1.4032,
      "num_input_tokens_seen": 27829040,
      "step": 4603,
      "train_runtime": 4665.8794,
      "train_tokens_per_second": 5964.372
    },
    {
      "epoch": 1.6933333333333334,
      "grad_norm": 0.9411514401435852,
      "learning_rate": 3.073808209092582e-05,
      "loss": 1.5212,
      "num_input_tokens_seen": 27834480,
      "step": 4604,
      "train_runtime": 4666.7749,
      "train_tokens_per_second": 5964.393
    },
    {
      "epoch": 1.6937011494252874,
      "grad_norm": 0.9370007514953613,
      "learning_rate": 3.070127001656544e-05,
      "loss": 1.3163,
      "num_input_tokens_seen": 27841168,
      "step": 4605,
      "train_runtime": 4667.8657,
      "train_tokens_per_second": 5964.432
    },
    {
      "epoch": 1.6940689655172414,
      "grad_norm": 0.9840338230133057,
      "learning_rate": 3.066445794220504e-05,
      "loss": 1.4556,
      "num_input_tokens_seen": 27847216,
      "step": 4606,
      "train_runtime": 4668.8424,
      "train_tokens_per_second": 5964.48
    },
    {
      "epoch": 1.6944367816091954,
      "grad_norm": 1.042328953742981,
      "learning_rate": 3.0627645867844656e-05,
      "loss": 1.7598,
      "num_input_tokens_seen": 27852688,
      "step": 4607,
      "train_runtime": 4669.751,
      "train_tokens_per_second": 5964.491
    },
    {
      "epoch": 1.6948045977011494,
      "grad_norm": 0.9465333819389343,
      "learning_rate": 3.0590833793484265e-05,
      "loss": 1.446,
      "num_input_tokens_seen": 27859136,
      "step": 4608,
      "train_runtime": 4670.7797,
      "train_tokens_per_second": 5964.558
    },
    {
      "epoch": 1.6951724137931035,
      "grad_norm": 0.8988509774208069,
      "learning_rate": 3.0554021719123874e-05,
      "loss": 1.5137,
      "num_input_tokens_seen": 27866720,
      "step": 4609,
      "train_runtime": 4671.9727,
      "train_tokens_per_second": 5964.658
    },
    {
      "epoch": 1.6955402298850575,
      "grad_norm": 1.0273783206939697,
      "learning_rate": 3.0517209644763483e-05,
      "loss": 1.6119,
      "num_input_tokens_seen": 27871344,
      "step": 4610,
      "train_runtime": 4672.7671,
      "train_tokens_per_second": 5964.634
    },
    {
      "epoch": 1.6959080459770115,
      "grad_norm": 1.0789744853973389,
      "learning_rate": 3.0480397570403095e-05,
      "loss": 1.6015,
      "num_input_tokens_seen": 27876304,
      "step": 4611,
      "train_runtime": 4673.6332,
      "train_tokens_per_second": 5964.59
    },
    {
      "epoch": 1.6962758620689655,
      "grad_norm": 1.0871307849884033,
      "learning_rate": 3.0443585496042704e-05,
      "loss": 1.5275,
      "num_input_tokens_seen": 27883824,
      "step": 4612,
      "train_runtime": 4674.8223,
      "train_tokens_per_second": 5964.681
    },
    {
      "epoch": 1.6966436781609195,
      "grad_norm": 1.0001200437545776,
      "learning_rate": 3.040677342168231e-05,
      "loss": 1.502,
      "num_input_tokens_seen": 27888192,
      "step": 4613,
      "train_runtime": 4675.5861,
      "train_tokens_per_second": 5964.641
    },
    {
      "epoch": 1.6970114942528736,
      "grad_norm": 0.9980902671813965,
      "learning_rate": 3.0369961347321922e-05,
      "loss": 1.5381,
      "num_input_tokens_seen": 27895008,
      "step": 4614,
      "train_runtime": 4676.6711,
      "train_tokens_per_second": 5964.715
    },
    {
      "epoch": 1.6973793103448276,
      "grad_norm": 1.0284391641616821,
      "learning_rate": 3.033314927296153e-05,
      "loss": 1.5916,
      "num_input_tokens_seen": 27899840,
      "step": 4615,
      "train_runtime": 4677.4828,
      "train_tokens_per_second": 5964.713
    },
    {
      "epoch": 1.6977471264367816,
      "grad_norm": 0.9810758233070374,
      "learning_rate": 3.0296337198601143e-05,
      "loss": 1.3792,
      "num_input_tokens_seen": 27906496,
      "step": 4616,
      "train_runtime": 4678.5449,
      "train_tokens_per_second": 5964.781
    },
    {
      "epoch": 1.6981149425287356,
      "grad_norm": 1.088984489440918,
      "learning_rate": 3.0259525124240752e-05,
      "loss": 1.7894,
      "num_input_tokens_seen": 27911424,
      "step": 4617,
      "train_runtime": 4679.3707,
      "train_tokens_per_second": 5964.782
    },
    {
      "epoch": 1.6984827586206896,
      "grad_norm": 1.0045228004455566,
      "learning_rate": 3.0222713049880364e-05,
      "loss": 1.5744,
      "num_input_tokens_seen": 27916288,
      "step": 4618,
      "train_runtime": 4680.2149,
      "train_tokens_per_second": 5964.745
    },
    {
      "epoch": 1.6988505747126437,
      "grad_norm": 0.9596930742263794,
      "learning_rate": 3.018590097551997e-05,
      "loss": 1.4039,
      "num_input_tokens_seen": 27922000,
      "step": 4619,
      "train_runtime": 4681.1671,
      "train_tokens_per_second": 5964.752
    },
    {
      "epoch": 1.6992183908045977,
      "grad_norm": 0.8283352255821228,
      "learning_rate": 3.0149088901159582e-05,
      "loss": 1.1706,
      "num_input_tokens_seen": 27933088,
      "step": 4620,
      "train_runtime": 4682.8063,
      "train_tokens_per_second": 5965.032
    },
    {
      "epoch": 1.6995862068965517,
      "grad_norm": 1.0603177547454834,
      "learning_rate": 3.011227682679919e-05,
      "loss": 1.5413,
      "num_input_tokens_seen": 27939200,
      "step": 4621,
      "train_runtime": 4684.2644,
      "train_tokens_per_second": 5964.48
    },
    {
      "epoch": 1.6999540229885057,
      "grad_norm": 0.9649791121482849,
      "learning_rate": 3.0075464752438803e-05,
      "loss": 1.2902,
      "num_input_tokens_seen": 27945216,
      "step": 4622,
      "train_runtime": 4685.2435,
      "train_tokens_per_second": 5964.517
    },
    {
      "epoch": 1.7003218390804598,
      "grad_norm": 1.0330976247787476,
      "learning_rate": 3.0038652678078412e-05,
      "loss": 1.5685,
      "num_input_tokens_seen": 27951184,
      "step": 4623,
      "train_runtime": 4686.2282,
      "train_tokens_per_second": 5964.538
    },
    {
      "epoch": 1.7006896551724138,
      "grad_norm": 1.0130820274353027,
      "learning_rate": 3.0001840603718024e-05,
      "loss": 1.6268,
      "num_input_tokens_seen": 27957472,
      "step": 4624,
      "train_runtime": 4687.2379,
      "train_tokens_per_second": 5964.594
    },
    {
      "epoch": 1.7010574712643678,
      "grad_norm": 1.0403722524642944,
      "learning_rate": 2.996502852935763e-05,
      "loss": 1.3725,
      "num_input_tokens_seen": 27963168,
      "step": 4625,
      "train_runtime": 4688.1637,
      "train_tokens_per_second": 5964.631
    },
    {
      "epoch": 1.7014252873563218,
      "grad_norm": 1.084006667137146,
      "learning_rate": 2.9928216454997242e-05,
      "loss": 1.5613,
      "num_input_tokens_seen": 27967504,
      "step": 4626,
      "train_runtime": 4688.9311,
      "train_tokens_per_second": 5964.58
    },
    {
      "epoch": 1.7017931034482758,
      "grad_norm": 0.9584836959838867,
      "learning_rate": 2.989140438063685e-05,
      "loss": 1.3322,
      "num_input_tokens_seen": 27972864,
      "step": 4627,
      "train_runtime": 4689.8415,
      "train_tokens_per_second": 5964.565
    },
    {
      "epoch": 1.7021609195402299,
      "grad_norm": 1.0244557857513428,
      "learning_rate": 2.9854592306276457e-05,
      "loss": 1.682,
      "num_input_tokens_seen": 27978160,
      "step": 4628,
      "train_runtime": 4690.7345,
      "train_tokens_per_second": 5964.558
    },
    {
      "epoch": 1.7025287356321839,
      "grad_norm": 0.9913014769554138,
      "learning_rate": 2.981778023191607e-05,
      "loss": 1.418,
      "num_input_tokens_seen": 27983312,
      "step": 4629,
      "train_runtime": 4691.5947,
      "train_tokens_per_second": 5964.563
    },
    {
      "epoch": 1.702896551724138,
      "grad_norm": 1.0452675819396973,
      "learning_rate": 2.9780968157555678e-05,
      "loss": 1.6439,
      "num_input_tokens_seen": 27987968,
      "step": 4630,
      "train_runtime": 4692.4013,
      "train_tokens_per_second": 5964.53
    },
    {
      "epoch": 1.703264367816092,
      "grad_norm": 0.9283667802810669,
      "learning_rate": 2.974415608319529e-05,
      "loss": 1.3235,
      "num_input_tokens_seen": 27992736,
      "step": 4631,
      "train_runtime": 4693.2164,
      "train_tokens_per_second": 5964.51
    },
    {
      "epoch": 1.703632183908046,
      "grad_norm": 0.9419856071472168,
      "learning_rate": 2.97073440088349e-05,
      "loss": 1.3919,
      "num_input_tokens_seen": 27999280,
      "step": 4632,
      "train_runtime": 4694.2743,
      "train_tokens_per_second": 5964.56
    },
    {
      "epoch": 1.704,
      "grad_norm": 1.0104327201843262,
      "learning_rate": 2.967053193447451e-05,
      "loss": 1.4808,
      "num_input_tokens_seen": 28006560,
      "step": 4633,
      "train_runtime": 4695.4532,
      "train_tokens_per_second": 5964.613
    },
    {
      "epoch": 1.704367816091954,
      "grad_norm": 0.9505506157875061,
      "learning_rate": 2.9633719860114117e-05,
      "loss": 1.3957,
      "num_input_tokens_seen": 28012672,
      "step": 4634,
      "train_runtime": 4696.4422,
      "train_tokens_per_second": 5964.658
    },
    {
      "epoch": 1.704735632183908,
      "grad_norm": 1.0302808284759521,
      "learning_rate": 2.959690778575373e-05,
      "loss": 1.6753,
      "num_input_tokens_seen": 28018208,
      "step": 4635,
      "train_runtime": 4697.3881,
      "train_tokens_per_second": 5964.636
    },
    {
      "epoch": 1.705103448275862,
      "grad_norm": 1.0058319568634033,
      "learning_rate": 2.9560095711393338e-05,
      "loss": 1.5491,
      "num_input_tokens_seen": 28022896,
      "step": 4636,
      "train_runtime": 4698.1957,
      "train_tokens_per_second": 5964.608
    },
    {
      "epoch": 1.705471264367816,
      "grad_norm": 0.9292551875114441,
      "learning_rate": 2.952328363703295e-05,
      "loss": 1.5156,
      "num_input_tokens_seen": 28028400,
      "step": 4637,
      "train_runtime": 4699.1131,
      "train_tokens_per_second": 5964.615
    },
    {
      "epoch": 1.70583908045977,
      "grad_norm": 1.0408554077148438,
      "learning_rate": 2.948647156267256e-05,
      "loss": 1.5696,
      "num_input_tokens_seen": 28033040,
      "step": 4638,
      "train_runtime": 4699.9117,
      "train_tokens_per_second": 5964.589
    },
    {
      "epoch": 1.706206896551724,
      "grad_norm": 0.992240846157074,
      "learning_rate": 2.944965948831217e-05,
      "loss": 1.6072,
      "num_input_tokens_seen": 28038896,
      "step": 4639,
      "train_runtime": 4700.8625,
      "train_tokens_per_second": 5964.628
    },
    {
      "epoch": 1.706574712643678,
      "grad_norm": 0.9619226455688477,
      "learning_rate": 2.9412847413951777e-05,
      "loss": 1.4376,
      "num_input_tokens_seen": 28045584,
      "step": 4640,
      "train_runtime": 4701.9366,
      "train_tokens_per_second": 5964.688
    },
    {
      "epoch": 1.7069425287356323,
      "grad_norm": 0.9967951774597168,
      "learning_rate": 2.937603533959139e-05,
      "loss": 1.5017,
      "num_input_tokens_seen": 28050160,
      "step": 4641,
      "train_runtime": 4702.7025,
      "train_tokens_per_second": 5964.689
    },
    {
      "epoch": 1.7073103448275861,
      "grad_norm": 0.9876623749732971,
      "learning_rate": 2.9339223265230998e-05,
      "loss": 1.557,
      "num_input_tokens_seen": 28055232,
      "step": 4642,
      "train_runtime": 4703.5717,
      "train_tokens_per_second": 5964.666
    },
    {
      "epoch": 1.7076781609195404,
      "grad_norm": 0.9602704048156738,
      "learning_rate": 2.9302411190870603e-05,
      "loss": 1.3624,
      "num_input_tokens_seen": 28061936,
      "step": 4643,
      "train_runtime": 4704.6463,
      "train_tokens_per_second": 5964.728
    },
    {
      "epoch": 1.7080459770114942,
      "grad_norm": 1.025449275970459,
      "learning_rate": 2.9265599116510216e-05,
      "loss": 1.5564,
      "num_input_tokens_seen": 28068672,
      "step": 4644,
      "train_runtime": 4705.7051,
      "train_tokens_per_second": 5964.817
    },
    {
      "epoch": 1.7084137931034484,
      "grad_norm": 0.983074963092804,
      "learning_rate": 2.9228787042149825e-05,
      "loss": 1.6266,
      "num_input_tokens_seen": 28074144,
      "step": 4645,
      "train_runtime": 4706.614,
      "train_tokens_per_second": 5964.828
    },
    {
      "epoch": 1.7087816091954022,
      "grad_norm": 1.0376065969467163,
      "learning_rate": 2.9191974967789437e-05,
      "loss": 1.7139,
      "num_input_tokens_seen": 28079168,
      "step": 4646,
      "train_runtime": 4707.4639,
      "train_tokens_per_second": 5964.819
    },
    {
      "epoch": 1.7091494252873565,
      "grad_norm": 1.0465658903121948,
      "learning_rate": 2.9155162893429046e-05,
      "loss": 1.5849,
      "num_input_tokens_seen": 28084592,
      "step": 4647,
      "train_runtime": 4708.3744,
      "train_tokens_per_second": 5964.817
    },
    {
      "epoch": 1.7095172413793103,
      "grad_norm": 0.9902276992797852,
      "learning_rate": 2.9118350819068658e-05,
      "loss": 1.4385,
      "num_input_tokens_seen": 28088752,
      "step": 4648,
      "train_runtime": 4709.0813,
      "train_tokens_per_second": 5964.805
    },
    {
      "epoch": 1.7098850574712645,
      "grad_norm": 1.0716465711593628,
      "learning_rate": 2.9081538744708264e-05,
      "loss": 1.6299,
      "num_input_tokens_seen": 28093872,
      "step": 4649,
      "train_runtime": 4709.9453,
      "train_tokens_per_second": 5964.798
    },
    {
      "epoch": 1.7102528735632183,
      "grad_norm": 0.9719383716583252,
      "learning_rate": 2.9044726670347876e-05,
      "loss": 1.5073,
      "num_input_tokens_seen": 28100368,
      "step": 4650,
      "train_runtime": 4711.0004,
      "train_tokens_per_second": 5964.841
    },
    {
      "epoch": 1.7106206896551726,
      "grad_norm": 0.7893573641777039,
      "learning_rate": 2.9007914595987485e-05,
      "loss": 1.3181,
      "num_input_tokens_seen": 28110960,
      "step": 4651,
      "train_runtime": 4713.0737,
      "train_tokens_per_second": 5964.464
    },
    {
      "epoch": 1.7109885057471264,
      "grad_norm": 0.9521739482879639,
      "learning_rate": 2.8971102521627097e-05,
      "loss": 1.3179,
      "num_input_tokens_seen": 28116720,
      "step": 4652,
      "train_runtime": 4714.0191,
      "train_tokens_per_second": 5964.49
    },
    {
      "epoch": 1.7113563218390806,
      "grad_norm": 0.9731817245483398,
      "learning_rate": 2.8934290447266703e-05,
      "loss": 1.4212,
      "num_input_tokens_seen": 28123008,
      "step": 4653,
      "train_runtime": 4715.0562,
      "train_tokens_per_second": 5964.512
    },
    {
      "epoch": 1.7117241379310344,
      "grad_norm": 0.9910678863525391,
      "learning_rate": 2.8897478372906318e-05,
      "loss": 1.4926,
      "num_input_tokens_seen": 28130192,
      "step": 4654,
      "train_runtime": 4716.2194,
      "train_tokens_per_second": 5964.564
    },
    {
      "epoch": 1.7120919540229886,
      "grad_norm": 1.0696122646331787,
      "learning_rate": 2.8860666298545924e-05,
      "loss": 1.4922,
      "num_input_tokens_seen": 28137040,
      "step": 4655,
      "train_runtime": 4717.3046,
      "train_tokens_per_second": 5964.643
    },
    {
      "epoch": 1.7124597701149424,
      "grad_norm": 1.0817041397094727,
      "learning_rate": 2.8823854224185536e-05,
      "loss": 1.4757,
      "num_input_tokens_seen": 28141776,
      "step": 4656,
      "train_runtime": 4718.1018,
      "train_tokens_per_second": 5964.639
    },
    {
      "epoch": 1.7128275862068967,
      "grad_norm": 1.0063755512237549,
      "learning_rate": 2.8787042149825145e-05,
      "loss": 1.4174,
      "num_input_tokens_seen": 28150544,
      "step": 4657,
      "train_runtime": 4719.4416,
      "train_tokens_per_second": 5964.804
    },
    {
      "epoch": 1.7131954022988505,
      "grad_norm": 1.0067154169082642,
      "learning_rate": 2.875023007546475e-05,
      "loss": 1.4901,
      "num_input_tokens_seen": 28155488,
      "step": 4658,
      "train_runtime": 4720.2856,
      "train_tokens_per_second": 5964.785
    },
    {
      "epoch": 1.7135632183908047,
      "grad_norm": 1.0253180265426636,
      "learning_rate": 2.8713418001104363e-05,
      "loss": 1.4496,
      "num_input_tokens_seen": 28163168,
      "step": 4659,
      "train_runtime": 4721.4569,
      "train_tokens_per_second": 5964.932
    },
    {
      "epoch": 1.7139310344827585,
      "grad_norm": 0.9907005429267883,
      "learning_rate": 2.867660592674397e-05,
      "loss": 1.5636,
      "num_input_tokens_seen": 28168448,
      "step": 4660,
      "train_runtime": 4722.3326,
      "train_tokens_per_second": 5964.944
    },
    {
      "epoch": 1.7142988505747128,
      "grad_norm": 0.9704107642173767,
      "learning_rate": 2.8639793852383584e-05,
      "loss": 1.348,
      "num_input_tokens_seen": 28177408,
      "step": 4661,
      "train_runtime": 4723.705,
      "train_tokens_per_second": 5965.107
    },
    {
      "epoch": 1.7146666666666666,
      "grad_norm": 1.0285683870315552,
      "learning_rate": 2.8602981778023193e-05,
      "loss": 1.5931,
      "num_input_tokens_seen": 28184288,
      "step": 4662,
      "train_runtime": 4724.8051,
      "train_tokens_per_second": 5965.175
    },
    {
      "epoch": 1.7150344827586208,
      "grad_norm": 0.9361411929130554,
      "learning_rate": 2.8566169703662805e-05,
      "loss": 1.4663,
      "num_input_tokens_seen": 28191536,
      "step": 4663,
      "train_runtime": 4725.941,
      "train_tokens_per_second": 5965.275
    },
    {
      "epoch": 1.7154022988505746,
      "grad_norm": 1.0544816255569458,
      "learning_rate": 2.852935762930241e-05,
      "loss": 1.6253,
      "num_input_tokens_seen": 28196080,
      "step": 4664,
      "train_runtime": 4726.7401,
      "train_tokens_per_second": 5965.228
    },
    {
      "epoch": 1.7157701149425288,
      "grad_norm": 1.0819185972213745,
      "learning_rate": 2.8492545554942023e-05,
      "loss": 1.6852,
      "num_input_tokens_seen": 28200656,
      "step": 4665,
      "train_runtime": 4727.541,
      "train_tokens_per_second": 5965.185
    },
    {
      "epoch": 1.7161379310344826,
      "grad_norm": 0.9444822072982788,
      "learning_rate": 2.8455733480581632e-05,
      "loss": 1.506,
      "num_input_tokens_seen": 28208128,
      "step": 4666,
      "train_runtime": 4728.731,
      "train_tokens_per_second": 5965.264
    },
    {
      "epoch": 1.7165057471264369,
      "grad_norm": 1.2216205596923828,
      "learning_rate": 2.8418921406221244e-05,
      "loss": 1.6304,
      "num_input_tokens_seen": 28213872,
      "step": 4667,
      "train_runtime": 4729.6887,
      "train_tokens_per_second": 5965.27
    },
    {
      "epoch": 1.7168735632183907,
      "grad_norm": 0.9831274151802063,
      "learning_rate": 2.838210933186085e-05,
      "loss": 1.4768,
      "num_input_tokens_seen": 28220176,
      "step": 4668,
      "train_runtime": 4730.7036,
      "train_tokens_per_second": 5965.323
    },
    {
      "epoch": 1.717241379310345,
      "grad_norm": 1.070762038230896,
      "learning_rate": 2.8345297257500465e-05,
      "loss": 1.3823,
      "num_input_tokens_seen": 28226944,
      "step": 4669,
      "train_runtime": 4731.798,
      "train_tokens_per_second": 5965.374
    },
    {
      "epoch": 1.7176091954022987,
      "grad_norm": 0.9777244329452515,
      "learning_rate": 2.830848518314007e-05,
      "loss": 1.711,
      "num_input_tokens_seen": 28232320,
      "step": 4670,
      "train_runtime": 4732.6719,
      "train_tokens_per_second": 5965.408
    },
    {
      "epoch": 1.717977011494253,
      "grad_norm": 1.0839136838912964,
      "learning_rate": 2.8271673108779683e-05,
      "loss": 1.5984,
      "num_input_tokens_seen": 28237984,
      "step": 4671,
      "train_runtime": 4733.6058,
      "train_tokens_per_second": 5965.428
    },
    {
      "epoch": 1.7183448275862068,
      "grad_norm": 0.9616579413414001,
      "learning_rate": 2.8234861034419292e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 28245312,
      "step": 4672,
      "train_runtime": 4734.782,
      "train_tokens_per_second": 5965.494
    },
    {
      "epoch": 1.718712643678161,
      "grad_norm": 1.0128158330917358,
      "learning_rate": 2.8198048960058897e-05,
      "loss": 1.4396,
      "num_input_tokens_seen": 28250544,
      "step": 4673,
      "train_runtime": 4735.6455,
      "train_tokens_per_second": 5965.511
    },
    {
      "epoch": 1.7190804597701148,
      "grad_norm": 1.0112683773040771,
      "learning_rate": 2.816123688569851e-05,
      "loss": 1.467,
      "num_input_tokens_seen": 28255760,
      "step": 4674,
      "train_runtime": 4736.5468,
      "train_tokens_per_second": 5965.477
    },
    {
      "epoch": 1.719448275862069,
      "grad_norm": 0.9474003314971924,
      "learning_rate": 2.812442481133812e-05,
      "loss": 1.3938,
      "num_input_tokens_seen": 28261984,
      "step": 4675,
      "train_runtime": 4737.5505,
      "train_tokens_per_second": 5965.527
    },
    {
      "epoch": 1.7198160919540229,
      "grad_norm": 1.0330866575241089,
      "learning_rate": 2.808761273697773e-05,
      "loss": 1.5392,
      "num_input_tokens_seen": 28267120,
      "step": 4676,
      "train_runtime": 4738.4271,
      "train_tokens_per_second": 5965.507
    },
    {
      "epoch": 1.720183908045977,
      "grad_norm": 1.1261749267578125,
      "learning_rate": 2.805080066261734e-05,
      "loss": 1.7794,
      "num_input_tokens_seen": 28273200,
      "step": 4677,
      "train_runtime": 4739.4305,
      "train_tokens_per_second": 5965.527
    },
    {
      "epoch": 1.720551724137931,
      "grad_norm": 0.9648178219795227,
      "learning_rate": 2.8013988588256952e-05,
      "loss": 1.3757,
      "num_input_tokens_seen": 28280048,
      "step": 4678,
      "train_runtime": 4740.548,
      "train_tokens_per_second": 5965.565
    },
    {
      "epoch": 1.7209195402298851,
      "grad_norm": 1.0651392936706543,
      "learning_rate": 2.7977176513896557e-05,
      "loss": 1.7287,
      "num_input_tokens_seen": 28284960,
      "step": 4679,
      "train_runtime": 4741.3716,
      "train_tokens_per_second": 5965.565
    },
    {
      "epoch": 1.721287356321839,
      "grad_norm": 0.9519479870796204,
      "learning_rate": 2.794036443953617e-05,
      "loss": 1.4971,
      "num_input_tokens_seen": 28292464,
      "step": 4680,
      "train_runtime": 4742.5646,
      "train_tokens_per_second": 5965.646
    },
    {
      "epoch": 1.7216551724137932,
      "grad_norm": 0.9936322569847107,
      "learning_rate": 2.790355236517578e-05,
      "loss": 1.5905,
      "num_input_tokens_seen": 28297920,
      "step": 4681,
      "train_runtime": 4744.0386,
      "train_tokens_per_second": 5964.943
    },
    {
      "epoch": 1.722022988505747,
      "grad_norm": 0.9680910110473633,
      "learning_rate": 2.786674029081539e-05,
      "loss": 1.4007,
      "num_input_tokens_seen": 28302400,
      "step": 4682,
      "train_runtime": 4744.8219,
      "train_tokens_per_second": 5964.902
    },
    {
      "epoch": 1.7223908045977012,
      "grad_norm": 1.003828525543213,
      "learning_rate": 2.7829928216454996e-05,
      "loss": 1.4056,
      "num_input_tokens_seen": 28307376,
      "step": 4683,
      "train_runtime": 4745.6753,
      "train_tokens_per_second": 5964.878
    },
    {
      "epoch": 1.722758620689655,
      "grad_norm": 1.0208098888397217,
      "learning_rate": 2.7793116142094612e-05,
      "loss": 1.5971,
      "num_input_tokens_seen": 28313248,
      "step": 4684,
      "train_runtime": 4746.6515,
      "train_tokens_per_second": 5964.889
    },
    {
      "epoch": 1.7231264367816093,
      "grad_norm": 1.088664174079895,
      "learning_rate": 2.7756304067734218e-05,
      "loss": 1.5996,
      "num_input_tokens_seen": 28319312,
      "step": 4685,
      "train_runtime": 4747.6485,
      "train_tokens_per_second": 5964.913
    },
    {
      "epoch": 1.723494252873563,
      "grad_norm": 0.9208300113677979,
      "learning_rate": 2.771949199337383e-05,
      "loss": 1.3616,
      "num_input_tokens_seen": 28326848,
      "step": 4686,
      "train_runtime": 4748.859,
      "train_tokens_per_second": 5964.98
    },
    {
      "epoch": 1.7238620689655173,
      "grad_norm": 1.0395221710205078,
      "learning_rate": 2.768267991901344e-05,
      "loss": 1.391,
      "num_input_tokens_seen": 28334336,
      "step": 4687,
      "train_runtime": 4750.0355,
      "train_tokens_per_second": 5965.079
    },
    {
      "epoch": 1.724229885057471,
      "grad_norm": 1.0180349349975586,
      "learning_rate": 2.7645867844653044e-05,
      "loss": 1.5954,
      "num_input_tokens_seen": 28341184,
      "step": 4688,
      "train_runtime": 4751.1313,
      "train_tokens_per_second": 5965.144
    },
    {
      "epoch": 1.7245977011494253,
      "grad_norm": 0.9864228367805481,
      "learning_rate": 2.7609055770292657e-05,
      "loss": 1.4096,
      "num_input_tokens_seen": 28347024,
      "step": 4689,
      "train_runtime": 4752.0875,
      "train_tokens_per_second": 5965.173
    },
    {
      "epoch": 1.7249655172413794,
      "grad_norm": 1.024919867515564,
      "learning_rate": 2.7572243695932265e-05,
      "loss": 1.5929,
      "num_input_tokens_seen": 28351904,
      "step": 4690,
      "train_runtime": 4752.8925,
      "train_tokens_per_second": 5965.189
    },
    {
      "epoch": 1.7253333333333334,
      "grad_norm": 1.0786160230636597,
      "learning_rate": 2.7535431621571878e-05,
      "loss": 1.4965,
      "num_input_tokens_seen": 28359904,
      "step": 4691,
      "train_runtime": 4754.1516,
      "train_tokens_per_second": 5965.292
    },
    {
      "epoch": 1.7257011494252874,
      "grad_norm": 0.9984427094459534,
      "learning_rate": 2.7498619547211483e-05,
      "loss": 1.5075,
      "num_input_tokens_seen": 28364992,
      "step": 4692,
      "train_runtime": 4755.0029,
      "train_tokens_per_second": 5965.294
    },
    {
      "epoch": 1.7260689655172414,
      "grad_norm": 0.9112803339958191,
      "learning_rate": 2.74618074728511e-05,
      "loss": 1.288,
      "num_input_tokens_seen": 28375504,
      "step": 4693,
      "train_runtime": 4756.5792,
      "train_tokens_per_second": 5965.528
    },
    {
      "epoch": 1.7264367816091954,
      "grad_norm": 0.9645473957061768,
      "learning_rate": 2.7424995398490704e-05,
      "loss": 1.4539,
      "num_input_tokens_seen": 28380624,
      "step": 4694,
      "train_runtime": 4757.4552,
      "train_tokens_per_second": 5965.505
    },
    {
      "epoch": 1.7268045977011495,
      "grad_norm": 1.0410685539245605,
      "learning_rate": 2.7388183324130317e-05,
      "loss": 1.7913,
      "num_input_tokens_seen": 28385808,
      "step": 4695,
      "train_runtime": 4758.3265,
      "train_tokens_per_second": 5965.502
    },
    {
      "epoch": 1.7271724137931035,
      "grad_norm": 0.9809327125549316,
      "learning_rate": 2.7351371249769926e-05,
      "loss": 1.4374,
      "num_input_tokens_seen": 28391280,
      "step": 4696,
      "train_runtime": 4759.2339,
      "train_tokens_per_second": 5965.515
    },
    {
      "epoch": 1.7275402298850575,
      "grad_norm": 1.0407499074935913,
      "learning_rate": 2.7314559175409538e-05,
      "loss": 1.6229,
      "num_input_tokens_seen": 28397584,
      "step": 4697,
      "train_runtime": 4760.2809,
      "train_tokens_per_second": 5965.527
    },
    {
      "epoch": 1.7279080459770115,
      "grad_norm": 0.9742929935455322,
      "learning_rate": 2.7277747101049143e-05,
      "loss": 1.4711,
      "num_input_tokens_seen": 28403488,
      "step": 4698,
      "train_runtime": 4761.2597,
      "train_tokens_per_second": 5965.541
    },
    {
      "epoch": 1.7282758620689656,
      "grad_norm": 1.065129280090332,
      "learning_rate": 2.724093502668876e-05,
      "loss": 1.6719,
      "num_input_tokens_seen": 28408624,
      "step": 4699,
      "train_runtime": 4762.123,
      "train_tokens_per_second": 5965.538
    },
    {
      "epoch": 1.7286436781609196,
      "grad_norm": 1.0128817558288574,
      "learning_rate": 2.7204122952328365e-05,
      "loss": 1.5271,
      "num_input_tokens_seen": 28414128,
      "step": 4700,
      "train_runtime": 4763.0537,
      "train_tokens_per_second": 5965.528
    },
    {
      "epoch": 1.7290114942528736,
      "grad_norm": 1.002425193786621,
      "learning_rate": 2.7167310877967977e-05,
      "loss": 1.5294,
      "num_input_tokens_seen": 28419408,
      "step": 4701,
      "train_runtime": 4763.9028,
      "train_tokens_per_second": 5965.573
    },
    {
      "epoch": 1.7293793103448276,
      "grad_norm": 1.0439168214797974,
      "learning_rate": 2.7130498803607586e-05,
      "loss": 1.3572,
      "num_input_tokens_seen": 28426880,
      "step": 4702,
      "train_runtime": 4765.0851,
      "train_tokens_per_second": 5965.66
    },
    {
      "epoch": 1.7297471264367816,
      "grad_norm": 0.9425588846206665,
      "learning_rate": 2.709368672924719e-05,
      "loss": 1.5627,
      "num_input_tokens_seen": 28433504,
      "step": 4703,
      "train_runtime": 4766.1329,
      "train_tokens_per_second": 5965.739
    },
    {
      "epoch": 1.7301149425287357,
      "grad_norm": 1.1186126470565796,
      "learning_rate": 2.7056874654886804e-05,
      "loss": 1.7109,
      "num_input_tokens_seen": 28438720,
      "step": 4704,
      "train_runtime": 4766.9853,
      "train_tokens_per_second": 5965.766
    },
    {
      "epoch": 1.7304827586206897,
      "grad_norm": 0.8971080183982849,
      "learning_rate": 2.7020062580526412e-05,
      "loss": 1.4155,
      "num_input_tokens_seen": 28445984,
      "step": 4705,
      "train_runtime": 4768.1134,
      "train_tokens_per_second": 5965.878
    },
    {
      "epoch": 1.7308505747126437,
      "grad_norm": 1.0290093421936035,
      "learning_rate": 2.6983250506166025e-05,
      "loss": 1.6032,
      "num_input_tokens_seen": 28451472,
      "step": 4706,
      "train_runtime": 4769.0112,
      "train_tokens_per_second": 5965.906
    },
    {
      "epoch": 1.7312183908045977,
      "grad_norm": 1.020194411277771,
      "learning_rate": 2.694643843180563e-05,
      "loss": 1.5013,
      "num_input_tokens_seen": 28456336,
      "step": 4707,
      "train_runtime": 4769.8552,
      "train_tokens_per_second": 5965.87
    },
    {
      "epoch": 1.7315862068965517,
      "grad_norm": 0.9831212759017944,
      "learning_rate": 2.6909626357445246e-05,
      "loss": 1.5739,
      "num_input_tokens_seen": 28461552,
      "step": 4708,
      "train_runtime": 4770.7213,
      "train_tokens_per_second": 5965.88
    },
    {
      "epoch": 1.7319540229885058,
      "grad_norm": 1.0328938961029053,
      "learning_rate": 2.687281428308485e-05,
      "loss": 1.56,
      "num_input_tokens_seen": 28469680,
      "step": 4709,
      "train_runtime": 4771.9758,
      "train_tokens_per_second": 5966.015
    },
    {
      "epoch": 1.7323218390804598,
      "grad_norm": 1.0327404737472534,
      "learning_rate": 2.6836002208724464e-05,
      "loss": 1.479,
      "num_input_tokens_seen": 28476016,
      "step": 4710,
      "train_runtime": 4772.9924,
      "train_tokens_per_second": 5966.072
    },
    {
      "epoch": 1.7326896551724138,
      "grad_norm": 0.9637169241905212,
      "learning_rate": 2.6799190134364073e-05,
      "loss": 1.3993,
      "num_input_tokens_seen": 28480672,
      "step": 4711,
      "train_runtime": 4774.3143,
      "train_tokens_per_second": 5965.395
    },
    {
      "epoch": 1.7330574712643678,
      "grad_norm": 0.97785884141922,
      "learning_rate": 2.6762378060003685e-05,
      "loss": 1.36,
      "num_input_tokens_seen": 28489792,
      "step": 4712,
      "train_runtime": 4775.7287,
      "train_tokens_per_second": 5965.538
    },
    {
      "epoch": 1.7334252873563218,
      "grad_norm": 1.0487269163131714,
      "learning_rate": 2.672556598564329e-05,
      "loss": 1.5209,
      "num_input_tokens_seen": 28495680,
      "step": 4713,
      "train_runtime": 4776.6925,
      "train_tokens_per_second": 5965.567
    },
    {
      "epoch": 1.7337931034482759,
      "grad_norm": 0.9517030119895935,
      "learning_rate": 2.6688753911282906e-05,
      "loss": 1.412,
      "num_input_tokens_seen": 28499952,
      "step": 4714,
      "train_runtime": 4777.4332,
      "train_tokens_per_second": 5965.536
    },
    {
      "epoch": 1.7341609195402299,
      "grad_norm": 0.9899877309799194,
      "learning_rate": 2.665194183692251e-05,
      "loss": 1.3717,
      "num_input_tokens_seen": 28506128,
      "step": 4715,
      "train_runtime": 4778.4215,
      "train_tokens_per_second": 5965.595
    },
    {
      "epoch": 1.734528735632184,
      "grad_norm": 1.0582808256149292,
      "learning_rate": 2.6615129762562124e-05,
      "loss": 1.6263,
      "num_input_tokens_seen": 28513168,
      "step": 4716,
      "train_runtime": 4779.5335,
      "train_tokens_per_second": 5965.68
    },
    {
      "epoch": 1.734896551724138,
      "grad_norm": 0.9215847849845886,
      "learning_rate": 2.6578317688201733e-05,
      "loss": 1.1623,
      "num_input_tokens_seen": 28520000,
      "step": 4717,
      "train_runtime": 4780.6031,
      "train_tokens_per_second": 5965.775
    },
    {
      "epoch": 1.735264367816092,
      "grad_norm": 1.154850959777832,
      "learning_rate": 2.6541505613841338e-05,
      "loss": 1.7994,
      "num_input_tokens_seen": 28526688,
      "step": 4718,
      "train_runtime": 4781.7226,
      "train_tokens_per_second": 5965.776
    },
    {
      "epoch": 1.735632183908046,
      "grad_norm": 0.7254742383956909,
      "learning_rate": 2.650469353948095e-05,
      "loss": 1.1917,
      "num_input_tokens_seen": 28541424,
      "step": 4719,
      "train_runtime": 4783.8707,
      "train_tokens_per_second": 5966.178
    },
    {
      "epoch": 1.736,
      "grad_norm": 0.9757923483848572,
      "learning_rate": 2.646788146512056e-05,
      "loss": 1.4235,
      "num_input_tokens_seen": 28546768,
      "step": 4720,
      "train_runtime": 4784.771,
      "train_tokens_per_second": 5966.172
    },
    {
      "epoch": 1.736367816091954,
      "grad_norm": 1.0379769802093506,
      "learning_rate": 2.643106939076017e-05,
      "loss": 1.4071,
      "num_input_tokens_seen": 28550944,
      "step": 4721,
      "train_runtime": 4785.531,
      "train_tokens_per_second": 5966.097
    },
    {
      "epoch": 1.736735632183908,
      "grad_norm": 1.041725516319275,
      "learning_rate": 2.6394257316399777e-05,
      "loss": 1.4824,
      "num_input_tokens_seen": 28557264,
      "step": 4722,
      "train_runtime": 4786.5502,
      "train_tokens_per_second": 5966.147
    },
    {
      "epoch": 1.737103448275862,
      "grad_norm": 0.9187132120132446,
      "learning_rate": 2.6357445242039393e-05,
      "loss": 1.4692,
      "num_input_tokens_seen": 28562800,
      "step": 4723,
      "train_runtime": 4787.4731,
      "train_tokens_per_second": 5966.154
    },
    {
      "epoch": 1.737471264367816,
      "grad_norm": 1.0748697519302368,
      "learning_rate": 2.6320633167679e-05,
      "loss": 1.54,
      "num_input_tokens_seen": 28567200,
      "step": 4724,
      "train_runtime": 4788.2434,
      "train_tokens_per_second": 5966.113
    },
    {
      "epoch": 1.73783908045977,
      "grad_norm": 1.0062446594238281,
      "learning_rate": 2.628382109331861e-05,
      "loss": 1.6539,
      "num_input_tokens_seen": 28572352,
      "step": 4725,
      "train_runtime": 4789.1026,
      "train_tokens_per_second": 5966.118
    },
    {
      "epoch": 1.7382068965517241,
      "grad_norm": 1.0973478555679321,
      "learning_rate": 2.624700901895822e-05,
      "loss": 1.6675,
      "num_input_tokens_seen": 28577824,
      "step": 4726,
      "train_runtime": 4790.0021,
      "train_tokens_per_second": 5966.14
    },
    {
      "epoch": 1.7385747126436781,
      "grad_norm": 1.0328434705734253,
      "learning_rate": 2.6210196944597832e-05,
      "loss": 1.514,
      "num_input_tokens_seen": 28586240,
      "step": 4727,
      "train_runtime": 4791.3075,
      "train_tokens_per_second": 5966.271
    },
    {
      "epoch": 1.7389425287356322,
      "grad_norm": 1.0493148565292358,
      "learning_rate": 2.6173384870237437e-05,
      "loss": 1.5932,
      "num_input_tokens_seen": 28593248,
      "step": 4728,
      "train_runtime": 4792.4231,
      "train_tokens_per_second": 5966.345
    },
    {
      "epoch": 1.7393103448275862,
      "grad_norm": 0.9841045141220093,
      "learning_rate": 2.613657279587705e-05,
      "loss": 1.397,
      "num_input_tokens_seen": 28599184,
      "step": 4729,
      "train_runtime": 4793.4131,
      "train_tokens_per_second": 5966.351
    },
    {
      "epoch": 1.7396781609195402,
      "grad_norm": 1.0101946592330933,
      "learning_rate": 2.609976072151666e-05,
      "loss": 1.5192,
      "num_input_tokens_seen": 28604144,
      "step": 4730,
      "train_runtime": 4794.2402,
      "train_tokens_per_second": 5966.356
    },
    {
      "epoch": 1.7400459770114942,
      "grad_norm": 1.1683965921401978,
      "learning_rate": 2.606294864715627e-05,
      "loss": 1.7244,
      "num_input_tokens_seen": 28609184,
      "step": 4731,
      "train_runtime": 4795.1084,
      "train_tokens_per_second": 5966.327
    },
    {
      "epoch": 1.7404137931034482,
      "grad_norm": 1.0627989768981934,
      "learning_rate": 2.602613657279588e-05,
      "loss": 1.5207,
      "num_input_tokens_seen": 28614256,
      "step": 4732,
      "train_runtime": 4795.9651,
      "train_tokens_per_second": 5966.319
    },
    {
      "epoch": 1.7407816091954023,
      "grad_norm": 1.1456577777862549,
      "learning_rate": 2.5989324498435485e-05,
      "loss": 1.4358,
      "num_input_tokens_seen": 28619984,
      "step": 4733,
      "train_runtime": 4796.8959,
      "train_tokens_per_second": 5966.355
    },
    {
      "epoch": 1.7411494252873563,
      "grad_norm": 1.0102976560592651,
      "learning_rate": 2.5952512424075097e-05,
      "loss": 1.4043,
      "num_input_tokens_seen": 28626640,
      "step": 4734,
      "train_runtime": 4797.9706,
      "train_tokens_per_second": 5966.406
    },
    {
      "epoch": 1.7415172413793103,
      "grad_norm": 1.0463898181915283,
      "learning_rate": 2.5915700349714706e-05,
      "loss": 1.6174,
      "num_input_tokens_seen": 28631824,
      "step": 4735,
      "train_runtime": 4798.835,
      "train_tokens_per_second": 5966.411
    },
    {
      "epoch": 1.7418850574712643,
      "grad_norm": 1.050277590751648,
      "learning_rate": 2.587888827535432e-05,
      "loss": 1.4079,
      "num_input_tokens_seen": 28637296,
      "step": 4736,
      "train_runtime": 4799.7242,
      "train_tokens_per_second": 5966.446
    },
    {
      "epoch": 1.7422528735632183,
      "grad_norm": 1.0384160280227661,
      "learning_rate": 2.5842076200993924e-05,
      "loss": 1.5397,
      "num_input_tokens_seen": 28645440,
      "step": 4737,
      "train_runtime": 4801.0199,
      "train_tokens_per_second": 5966.532
    },
    {
      "epoch": 1.7426206896551724,
      "grad_norm": 1.1893243789672852,
      "learning_rate": 2.580526412663354e-05,
      "loss": 1.5227,
      "num_input_tokens_seen": 28650368,
      "step": 4738,
      "train_runtime": 4801.8614,
      "train_tokens_per_second": 5966.513
    },
    {
      "epoch": 1.7429885057471264,
      "grad_norm": 1.1832375526428223,
      "learning_rate": 2.5768452052273145e-05,
      "loss": 1.6056,
      "num_input_tokens_seen": 28657424,
      "step": 4739,
      "train_runtime": 4802.9949,
      "train_tokens_per_second": 5966.574
    },
    {
      "epoch": 1.7433563218390806,
      "grad_norm": 1.0431658029556274,
      "learning_rate": 2.5731639977912758e-05,
      "loss": 1.6738,
      "num_input_tokens_seen": 28663664,
      "step": 4740,
      "train_runtime": 4803.9938,
      "train_tokens_per_second": 5966.632
    },
    {
      "epoch": 1.7437241379310344,
      "grad_norm": 1.0423671007156372,
      "learning_rate": 2.5694827903552366e-05,
      "loss": 1.5012,
      "num_input_tokens_seen": 28668144,
      "step": 4741,
      "train_runtime": 4805.2544,
      "train_tokens_per_second": 5965.999
    },
    {
      "epoch": 1.7440919540229887,
      "grad_norm": 1.0289225578308105,
      "learning_rate": 2.565801582919198e-05,
      "loss": 1.4479,
      "num_input_tokens_seen": 28675248,
      "step": 4742,
      "train_runtime": 4806.4085,
      "train_tokens_per_second": 5966.045
    },
    {
      "epoch": 1.7444597701149425,
      "grad_norm": 0.9507474303245544,
      "learning_rate": 2.5621203754831584e-05,
      "loss": 1.4761,
      "num_input_tokens_seen": 28683248,
      "step": 4743,
      "train_runtime": 4807.648,
      "train_tokens_per_second": 5966.171
    },
    {
      "epoch": 1.7448275862068967,
      "grad_norm": 0.9714196920394897,
      "learning_rate": 2.5584391680471196e-05,
      "loss": 1.5067,
      "num_input_tokens_seen": 28687888,
      "step": 4744,
      "train_runtime": 4808.4446,
      "train_tokens_per_second": 5966.147
    },
    {
      "epoch": 1.7451954022988505,
      "grad_norm": 1.0843784809112549,
      "learning_rate": 2.5547579606110805e-05,
      "loss": 1.2733,
      "num_input_tokens_seen": 28696432,
      "step": 4745,
      "train_runtime": 4809.7729,
      "train_tokens_per_second": 5966.276
    },
    {
      "epoch": 1.7455632183908047,
      "grad_norm": 1.058371663093567,
      "learning_rate": 2.5510767531750418e-05,
      "loss": 1.6213,
      "num_input_tokens_seen": 28701408,
      "step": 4746,
      "train_runtime": 4810.6161,
      "train_tokens_per_second": 5966.264
    },
    {
      "epoch": 1.7459310344827585,
      "grad_norm": 0.9698977470397949,
      "learning_rate": 2.5473955457390027e-05,
      "loss": 1.4538,
      "num_input_tokens_seen": 28706176,
      "step": 4747,
      "train_runtime": 4811.4419,
      "train_tokens_per_second": 5966.231
    },
    {
      "epoch": 1.7462988505747128,
      "grad_norm": 0.9774466156959534,
      "learning_rate": 2.5437143383029632e-05,
      "loss": 1.4971,
      "num_input_tokens_seen": 28712144,
      "step": 4748,
      "train_runtime": 4812.3784,
      "train_tokens_per_second": 5966.311
    },
    {
      "epoch": 1.7466666666666666,
      "grad_norm": 0.934420108795166,
      "learning_rate": 2.5400331308669244e-05,
      "loss": 1.3411,
      "num_input_tokens_seen": 28719840,
      "step": 4749,
      "train_runtime": 4813.5892,
      "train_tokens_per_second": 5966.409
    },
    {
      "epoch": 1.7470344827586208,
      "grad_norm": 0.8894535899162292,
      "learning_rate": 2.5363519234308853e-05,
      "loss": 1.3336,
      "num_input_tokens_seen": 28729344,
      "step": 4750,
      "train_runtime": 4815.0344,
      "train_tokens_per_second": 5966.592
    },
    {
      "epoch": 1.7474022988505746,
      "grad_norm": 1.0110523700714111,
      "learning_rate": 2.5326707159948466e-05,
      "loss": 1.3918,
      "num_input_tokens_seen": 28734800,
      "step": 4751,
      "train_runtime": 4815.9794,
      "train_tokens_per_second": 5966.554
    },
    {
      "epoch": 1.7477701149425289,
      "grad_norm": 1.0272303819656372,
      "learning_rate": 2.528989508558807e-05,
      "loss": 1.3874,
      "num_input_tokens_seen": 28741744,
      "step": 4752,
      "train_runtime": 4817.0988,
      "train_tokens_per_second": 5966.609
    },
    {
      "epoch": 1.7481379310344827,
      "grad_norm": 1.0168406963348389,
      "learning_rate": 2.5253083011227687e-05,
      "loss": 1.5772,
      "num_input_tokens_seen": 28746208,
      "step": 4753,
      "train_runtime": 4817.8578,
      "train_tokens_per_second": 5966.595
    },
    {
      "epoch": 1.748505747126437,
      "grad_norm": 1.0032830238342285,
      "learning_rate": 2.5216270936867292e-05,
      "loss": 1.3717,
      "num_input_tokens_seen": 28751616,
      "step": 4754,
      "train_runtime": 4818.7577,
      "train_tokens_per_second": 5966.603
    },
    {
      "epoch": 1.7488735632183907,
      "grad_norm": 0.9939249753952026,
      "learning_rate": 2.5179458862506904e-05,
      "loss": 1.6187,
      "num_input_tokens_seen": 28756896,
      "step": 4755,
      "train_runtime": 4819.6627,
      "train_tokens_per_second": 5966.579
    },
    {
      "epoch": 1.749241379310345,
      "grad_norm": 1.1142992973327637,
      "learning_rate": 2.5142646788146513e-05,
      "loss": 1.4498,
      "num_input_tokens_seen": 28764800,
      "step": 4756,
      "train_runtime": 4820.9015,
      "train_tokens_per_second": 5966.685
    },
    {
      "epoch": 1.7496091954022988,
      "grad_norm": 0.944139838218689,
      "learning_rate": 2.5105834713786126e-05,
      "loss": 1.5494,
      "num_input_tokens_seen": 28772304,
      "step": 4757,
      "train_runtime": 4822.0737,
      "train_tokens_per_second": 5966.791
    },
    {
      "epoch": 1.749977011494253,
      "grad_norm": 1.033401608467102,
      "learning_rate": 2.506902263942573e-05,
      "loss": 1.5243,
      "num_input_tokens_seen": 28777536,
      "step": 4758,
      "train_runtime": 4822.9721,
      "train_tokens_per_second": 5966.764
    },
    {
      "epoch": 1.7503448275862068,
      "grad_norm": 1.0464811325073242,
      "learning_rate": 2.5032210565065343e-05,
      "loss": 1.4168,
      "num_input_tokens_seen": 28783520,
      "step": 4759,
      "train_runtime": 4823.9558,
      "train_tokens_per_second": 5966.788
    },
    {
      "epoch": 1.750712643678161,
      "grad_norm": 1.0507761240005493,
      "learning_rate": 2.4995398490704952e-05,
      "loss": 1.5518,
      "num_input_tokens_seen": 28788192,
      "step": 4760,
      "train_runtime": 4824.781,
      "train_tokens_per_second": 5966.735
    },
    {
      "epoch": 1.7510804597701148,
      "grad_norm": 1.004483699798584,
      "learning_rate": 2.495858641634456e-05,
      "loss": 1.4975,
      "num_input_tokens_seen": 28793664,
      "step": 4761,
      "train_runtime": 4825.6882,
      "train_tokens_per_second": 5966.748
    },
    {
      "epoch": 1.751448275862069,
      "grad_norm": 1.0057058334350586,
      "learning_rate": 2.4921774341984173e-05,
      "loss": 1.6755,
      "num_input_tokens_seen": 28799536,
      "step": 4762,
      "train_runtime": 4826.6641,
      "train_tokens_per_second": 5966.758
    },
    {
      "epoch": 1.7518160919540229,
      "grad_norm": 1.0779151916503906,
      "learning_rate": 2.4884962267623782e-05,
      "loss": 1.5761,
      "num_input_tokens_seen": 28804432,
      "step": 4763,
      "train_runtime": 4827.5009,
      "train_tokens_per_second": 5966.738
    },
    {
      "epoch": 1.7521839080459771,
      "grad_norm": 1.0455926656723022,
      "learning_rate": 2.484815019326339e-05,
      "loss": 1.6414,
      "num_input_tokens_seen": 28812624,
      "step": 4764,
      "train_runtime": 4828.7923,
      "train_tokens_per_second": 5966.839
    },
    {
      "epoch": 1.752551724137931,
      "grad_norm": 0.9676247835159302,
      "learning_rate": 2.4811338118903004e-05,
      "loss": 1.4324,
      "num_input_tokens_seen": 28819024,
      "step": 4765,
      "train_runtime": 4829.8394,
      "train_tokens_per_second": 5966.87
    },
    {
      "epoch": 1.7529195402298852,
      "grad_norm": 1.0028635263442993,
      "learning_rate": 2.4774526044542612e-05,
      "loss": 1.6043,
      "num_input_tokens_seen": 28823984,
      "step": 4766,
      "train_runtime": 4830.6841,
      "train_tokens_per_second": 5966.853
    },
    {
      "epoch": 1.753287356321839,
      "grad_norm": 1.0500530004501343,
      "learning_rate": 2.473771397018222e-05,
      "loss": 1.4025,
      "num_input_tokens_seen": 28832064,
      "step": 4767,
      "train_runtime": 4831.9186,
      "train_tokens_per_second": 5967.001
    },
    {
      "epoch": 1.7536551724137932,
      "grad_norm": 1.0421392917633057,
      "learning_rate": 2.470090189582183e-05,
      "loss": 1.6018,
      "num_input_tokens_seen": 28838304,
      "step": 4768,
      "train_runtime": 4832.9508,
      "train_tokens_per_second": 5967.018
    },
    {
      "epoch": 1.754022988505747,
      "grad_norm": 1.1408933401107788,
      "learning_rate": 2.4664089821461443e-05,
      "loss": 1.72,
      "num_input_tokens_seen": 28842976,
      "step": 4769,
      "train_runtime": 4833.7403,
      "train_tokens_per_second": 5967.01
    },
    {
      "epoch": 1.7543908045977012,
      "grad_norm": 0.9642238020896912,
      "learning_rate": 2.4627277747101048e-05,
      "loss": 1.5639,
      "num_input_tokens_seen": 28851456,
      "step": 4770,
      "train_runtime": 4835.0521,
      "train_tokens_per_second": 5967.145
    },
    {
      "epoch": 1.754758620689655,
      "grad_norm": 1.0812686681747437,
      "learning_rate": 2.459046567274066e-05,
      "loss": 1.4493,
      "num_input_tokens_seen": 28856544,
      "step": 4771,
      "train_runtime": 4836.5241,
      "train_tokens_per_second": 5966.381
    },
    {
      "epoch": 1.7551264367816093,
      "grad_norm": 1.0130112171173096,
      "learning_rate": 2.455365359838027e-05,
      "loss": 1.5043,
      "num_input_tokens_seen": 28861360,
      "step": 4772,
      "train_runtime": 4837.3546,
      "train_tokens_per_second": 5966.352
    },
    {
      "epoch": 1.755494252873563,
      "grad_norm": 0.991884171962738,
      "learning_rate": 2.4516841524019878e-05,
      "loss": 1.3684,
      "num_input_tokens_seen": 28867664,
      "step": 4773,
      "train_runtime": 4838.3821,
      "train_tokens_per_second": 5966.388
    },
    {
      "epoch": 1.7558620689655173,
      "grad_norm": 0.9944646954536438,
      "learning_rate": 2.448002944965949e-05,
      "loss": 1.5666,
      "num_input_tokens_seen": 28874464,
      "step": 4774,
      "train_runtime": 4839.4788,
      "train_tokens_per_second": 5966.441
    },
    {
      "epoch": 1.7562298850574711,
      "grad_norm": 1.0106456279754639,
      "learning_rate": 2.44432173752991e-05,
      "loss": 1.4483,
      "num_input_tokens_seen": 28881728,
      "step": 4775,
      "train_runtime": 4840.6374,
      "train_tokens_per_second": 5966.513
    },
    {
      "epoch": 1.7565977011494254,
      "grad_norm": 1.0139954090118408,
      "learning_rate": 2.4406405300938708e-05,
      "loss": 1.498,
      "num_input_tokens_seen": 28888144,
      "step": 4776,
      "train_runtime": 4841.6522,
      "train_tokens_per_second": 5966.588
    },
    {
      "epoch": 1.7569655172413792,
      "grad_norm": 0.9798570871353149,
      "learning_rate": 2.436959322657832e-05,
      "loss": 1.6253,
      "num_input_tokens_seen": 28894688,
      "step": 4777,
      "train_runtime": 4842.7185,
      "train_tokens_per_second": 5966.626
    },
    {
      "epoch": 1.7573333333333334,
      "grad_norm": 0.9457206726074219,
      "learning_rate": 2.433278115221793e-05,
      "loss": 1.5637,
      "num_input_tokens_seen": 28901152,
      "step": 4778,
      "train_runtime": 4843.7467,
      "train_tokens_per_second": 5966.694
    },
    {
      "epoch": 1.7577011494252872,
      "grad_norm": 1.0914689302444458,
      "learning_rate": 2.4295969077857538e-05,
      "loss": 1.592,
      "num_input_tokens_seen": 28906304,
      "step": 4779,
      "train_runtime": 4844.6162,
      "train_tokens_per_second": 5966.686
    },
    {
      "epoch": 1.7580689655172415,
      "grad_norm": 1.1174763441085815,
      "learning_rate": 2.425915700349715e-05,
      "loss": 1.4404,
      "num_input_tokens_seen": 28913104,
      "step": 4780,
      "train_runtime": 4845.7408,
      "train_tokens_per_second": 5966.705
    },
    {
      "epoch": 1.7584367816091953,
      "grad_norm": 1.1484270095825195,
      "learning_rate": 2.422234492913676e-05,
      "loss": 1.712,
      "num_input_tokens_seen": 28918976,
      "step": 4781,
      "train_runtime": 4846.6932,
      "train_tokens_per_second": 5966.744
    },
    {
      "epoch": 1.7588045977011495,
      "grad_norm": 1.011078953742981,
      "learning_rate": 2.4185532854776368e-05,
      "loss": 1.5001,
      "num_input_tokens_seen": 28927472,
      "step": 4782,
      "train_runtime": 4848.0154,
      "train_tokens_per_second": 5966.869
    },
    {
      "epoch": 1.7591724137931033,
      "grad_norm": 1.0478185415267944,
      "learning_rate": 2.4148720780415977e-05,
      "loss": 1.6283,
      "num_input_tokens_seen": 28932432,
      "step": 4783,
      "train_runtime": 4848.8425,
      "train_tokens_per_second": 5966.874
    },
    {
      "epoch": 1.7595402298850575,
      "grad_norm": 1.0241420269012451,
      "learning_rate": 2.411190870605559e-05,
      "loss": 1.4226,
      "num_input_tokens_seen": 28937888,
      "step": 4784,
      "train_runtime": 4849.7451,
      "train_tokens_per_second": 5966.888
    },
    {
      "epoch": 1.7599080459770113,
      "grad_norm": 1.0115891695022583,
      "learning_rate": 2.4075096631695195e-05,
      "loss": 1.3334,
      "num_input_tokens_seen": 28944144,
      "step": 4785,
      "train_runtime": 4850.758,
      "train_tokens_per_second": 5966.932
    },
    {
      "epoch": 1.7602758620689656,
      "grad_norm": 0.9646759629249573,
      "learning_rate": 2.4038284557334807e-05,
      "loss": 1.321,
      "num_input_tokens_seen": 28948752,
      "step": 4786,
      "train_runtime": 4851.5707,
      "train_tokens_per_second": 5966.882
    },
    {
      "epoch": 1.7606436781609194,
      "grad_norm": 1.011519193649292,
      "learning_rate": 2.4001472482974416e-05,
      "loss": 1.4329,
      "num_input_tokens_seen": 28953104,
      "step": 4787,
      "train_runtime": 4852.3417,
      "train_tokens_per_second": 5966.831
    },
    {
      "epoch": 1.7610114942528736,
      "grad_norm": 0.9330379962921143,
      "learning_rate": 2.3964660408614025e-05,
      "loss": 1.3982,
      "num_input_tokens_seen": 28958624,
      "step": 4788,
      "train_runtime": 4853.2508,
      "train_tokens_per_second": 5966.851
    },
    {
      "epoch": 1.7613793103448274,
      "grad_norm": 0.9441887736320496,
      "learning_rate": 2.3927848334253637e-05,
      "loss": 1.4537,
      "num_input_tokens_seen": 28963872,
      "step": 4789,
      "train_runtime": 4854.1286,
      "train_tokens_per_second": 5966.853
    },
    {
      "epoch": 1.7617471264367817,
      "grad_norm": 0.9717675447463989,
      "learning_rate": 2.3891036259893246e-05,
      "loss": 1.4119,
      "num_input_tokens_seen": 28968928,
      "step": 4790,
      "train_runtime": 4854.9596,
      "train_tokens_per_second": 5966.873
    },
    {
      "epoch": 1.7621149425287357,
      "grad_norm": 1.0108474493026733,
      "learning_rate": 2.3854224185532855e-05,
      "loss": 1.4986,
      "num_input_tokens_seen": 28975440,
      "step": 4791,
      "train_runtime": 4855.9965,
      "train_tokens_per_second": 5966.94
    },
    {
      "epoch": 1.7624827586206897,
      "grad_norm": 0.8929666876792908,
      "learning_rate": 2.3817412111172467e-05,
      "loss": 1.3344,
      "num_input_tokens_seen": 28982496,
      "step": 4792,
      "train_runtime": 4857.1093,
      "train_tokens_per_second": 5967.026
    },
    {
      "epoch": 1.7628505747126437,
      "grad_norm": 1.0070686340332031,
      "learning_rate": 2.3780600036812076e-05,
      "loss": 1.5368,
      "num_input_tokens_seen": 28987520,
      "step": 4793,
      "train_runtime": 4857.9572,
      "train_tokens_per_second": 5967.018
    },
    {
      "epoch": 1.7632183908045977,
      "grad_norm": 1.1040626764297485,
      "learning_rate": 2.3743787962451685e-05,
      "loss": 1.6217,
      "num_input_tokens_seen": 28992832,
      "step": 4794,
      "train_runtime": 4858.8325,
      "train_tokens_per_second": 5967.037
    },
    {
      "epoch": 1.7635862068965518,
      "grad_norm": 1.103492021560669,
      "learning_rate": 2.3706975888091294e-05,
      "loss": 1.4166,
      "num_input_tokens_seen": 28998112,
      "step": 4795,
      "train_runtime": 4859.6795,
      "train_tokens_per_second": 5967.083
    },
    {
      "epoch": 1.7639540229885058,
      "grad_norm": 1.0375038385391235,
      "learning_rate": 2.3670163813730906e-05,
      "loss": 1.6313,
      "num_input_tokens_seen": 29003408,
      "step": 4796,
      "train_runtime": 4860.5591,
      "train_tokens_per_second": 5967.093
    },
    {
      "epoch": 1.7643218390804598,
      "grad_norm": 1.0778006315231323,
      "learning_rate": 2.3633351739370515e-05,
      "loss": 1.7406,
      "num_input_tokens_seen": 29007696,
      "step": 4797,
      "train_runtime": 4861.2996,
      "train_tokens_per_second": 5967.066
    },
    {
      "epoch": 1.7646896551724138,
      "grad_norm": 0.9532221555709839,
      "learning_rate": 2.3596539665010124e-05,
      "loss": 1.5471,
      "num_input_tokens_seen": 29013392,
      "step": 4798,
      "train_runtime": 4862.2653,
      "train_tokens_per_second": 5967.052
    },
    {
      "epoch": 1.7650574712643678,
      "grad_norm": 1.0531294345855713,
      "learning_rate": 2.3559727590649736e-05,
      "loss": 1.4915,
      "num_input_tokens_seen": 29018992,
      "step": 4799,
      "train_runtime": 4863.1879,
      "train_tokens_per_second": 5967.072
    },
    {
      "epoch": 1.7654252873563219,
      "grad_norm": 1.0663352012634277,
      "learning_rate": 2.3522915516289342e-05,
      "loss": 1.4718,
      "num_input_tokens_seen": 29024624,
      "step": 4800,
      "train_runtime": 4864.1242,
      "train_tokens_per_second": 5967.081
    },
    {
      "epoch": 1.765793103448276,
      "grad_norm": 1.0736812353134155,
      "learning_rate": 2.3486103441928954e-05,
      "loss": 1.5117,
      "num_input_tokens_seen": 29030784,
      "step": 4801,
      "train_runtime": 4865.6616,
      "train_tokens_per_second": 5966.462
    },
    {
      "epoch": 1.76616091954023,
      "grad_norm": 1.0155384540557861,
      "learning_rate": 2.3449291367568563e-05,
      "loss": 1.5503,
      "num_input_tokens_seen": 29035952,
      "step": 4802,
      "train_runtime": 4866.5211,
      "train_tokens_per_second": 5966.47
    },
    {
      "epoch": 1.766528735632184,
      "grad_norm": 0.9911293983459473,
      "learning_rate": 2.3412479293208172e-05,
      "loss": 1.7012,
      "num_input_tokens_seen": 29041440,
      "step": 4803,
      "train_runtime": 4867.4131,
      "train_tokens_per_second": 5966.504
    },
    {
      "epoch": 1.766896551724138,
      "grad_norm": 1.0082165002822876,
      "learning_rate": 2.3375667218847784e-05,
      "loss": 1.5463,
      "num_input_tokens_seen": 29047696,
      "step": 4804,
      "train_runtime": 4868.4224,
      "train_tokens_per_second": 5966.552
    },
    {
      "epoch": 1.767264367816092,
      "grad_norm": 1.087516188621521,
      "learning_rate": 2.3338855144487393e-05,
      "loss": 1.4284,
      "num_input_tokens_seen": 29051888,
      "step": 4805,
      "train_runtime": 4869.1614,
      "train_tokens_per_second": 5966.508
    },
    {
      "epoch": 1.767632183908046,
      "grad_norm": 1.0501154661178589,
      "learning_rate": 2.3302043070127002e-05,
      "loss": 1.3063,
      "num_input_tokens_seen": 29057280,
      "step": 4806,
      "train_runtime": 4870.059,
      "train_tokens_per_second": 5966.515
    },
    {
      "epoch": 1.768,
      "grad_norm": 0.9493128061294556,
      "learning_rate": 2.326523099576661e-05,
      "loss": 1.3305,
      "num_input_tokens_seen": 29067952,
      "step": 4807,
      "train_runtime": 4871.6684,
      "train_tokens_per_second": 5966.734
    },
    {
      "epoch": 1.768367816091954,
      "grad_norm": 1.1272931098937988,
      "learning_rate": 2.3228418921406223e-05,
      "loss": 1.7959,
      "num_input_tokens_seen": 29073728,
      "step": 4808,
      "train_runtime": 4872.6124,
      "train_tokens_per_second": 5966.764
    },
    {
      "epoch": 1.768735632183908,
      "grad_norm": 1.021295428276062,
      "learning_rate": 2.3191606847045832e-05,
      "loss": 1.5178,
      "num_input_tokens_seen": 29078656,
      "step": 4809,
      "train_runtime": 4873.4601,
      "train_tokens_per_second": 5966.737
    },
    {
      "epoch": 1.769103448275862,
      "grad_norm": 0.9110109806060791,
      "learning_rate": 2.315479477268544e-05,
      "loss": 1.4342,
      "num_input_tokens_seen": 29084832,
      "step": 4810,
      "train_runtime": 4874.4722,
      "train_tokens_per_second": 5966.765
    },
    {
      "epoch": 1.769471264367816,
      "grad_norm": 1.0035239458084106,
      "learning_rate": 2.3117982698325053e-05,
      "loss": 1.585,
      "num_input_tokens_seen": 29089328,
      "step": 4811,
      "train_runtime": 4875.2478,
      "train_tokens_per_second": 5966.738
    },
    {
      "epoch": 1.7698390804597701,
      "grad_norm": 0.9894155859947205,
      "learning_rate": 2.3081170623964662e-05,
      "loss": 1.4294,
      "num_input_tokens_seen": 29095456,
      "step": 4812,
      "train_runtime": 4876.2421,
      "train_tokens_per_second": 5966.778
    },
    {
      "epoch": 1.7702068965517241,
      "grad_norm": 0.9737396240234375,
      "learning_rate": 2.304435854960427e-05,
      "loss": 1.5051,
      "num_input_tokens_seen": 29101040,
      "step": 4813,
      "train_runtime": 4877.1553,
      "train_tokens_per_second": 5966.806
    },
    {
      "epoch": 1.7705747126436782,
      "grad_norm": 0.9453579783439636,
      "learning_rate": 2.3007546475243883e-05,
      "loss": 1.4653,
      "num_input_tokens_seen": 29109968,
      "step": 4814,
      "train_runtime": 4878.5028,
      "train_tokens_per_second": 5966.988
    },
    {
      "epoch": 1.7709425287356322,
      "grad_norm": 0.9606562256813049,
      "learning_rate": 2.297073440088349e-05,
      "loss": 1.5069,
      "num_input_tokens_seen": 29116928,
      "step": 4815,
      "train_runtime": 4879.6255,
      "train_tokens_per_second": 5967.042
    },
    {
      "epoch": 1.7713103448275862,
      "grad_norm": 1.0943264961242676,
      "learning_rate": 2.29339223265231e-05,
      "loss": 1.5241,
      "num_input_tokens_seen": 29121520,
      "step": 4816,
      "train_runtime": 4880.4226,
      "train_tokens_per_second": 5967.008
    },
    {
      "epoch": 1.7716781609195402,
      "grad_norm": 1.014162302017212,
      "learning_rate": 2.289711025216271e-05,
      "loss": 1.5497,
      "num_input_tokens_seen": 29125952,
      "step": 4817,
      "train_runtime": 4881.2056,
      "train_tokens_per_second": 5966.959
    },
    {
      "epoch": 1.7720459770114942,
      "grad_norm": 0.9894721508026123,
      "learning_rate": 2.286029817780232e-05,
      "loss": 1.4047,
      "num_input_tokens_seen": 29130480,
      "step": 4818,
      "train_runtime": 4881.9937,
      "train_tokens_per_second": 5966.923
    },
    {
      "epoch": 1.7724137931034483,
      "grad_norm": 0.9608563780784607,
      "learning_rate": 2.282348610344193e-05,
      "loss": 1.4861,
      "num_input_tokens_seen": 29135968,
      "step": 4819,
      "train_runtime": 4882.8846,
      "train_tokens_per_second": 5966.958
    },
    {
      "epoch": 1.7727816091954023,
      "grad_norm": 1.1403229236602783,
      "learning_rate": 2.278667402908154e-05,
      "loss": 1.4569,
      "num_input_tokens_seen": 29140704,
      "step": 4820,
      "train_runtime": 4883.6855,
      "train_tokens_per_second": 5966.949
    },
    {
      "epoch": 1.7731494252873563,
      "grad_norm": 1.0871057510375977,
      "learning_rate": 2.274986195472115e-05,
      "loss": 1.6215,
      "num_input_tokens_seen": 29145136,
      "step": 4821,
      "train_runtime": 4884.487,
      "train_tokens_per_second": 5966.878
    },
    {
      "epoch": 1.7735172413793103,
      "grad_norm": 0.9230962991714478,
      "learning_rate": 2.2713049880360758e-05,
      "loss": 1.4355,
      "num_input_tokens_seen": 29152336,
      "step": 4822,
      "train_runtime": 4885.6285,
      "train_tokens_per_second": 5966.957
    },
    {
      "epoch": 1.7738850574712643,
      "grad_norm": 0.9452508091926575,
      "learning_rate": 2.267623780600037e-05,
      "loss": 1.428,
      "num_input_tokens_seen": 29157696,
      "step": 4823,
      "train_runtime": 4886.5157,
      "train_tokens_per_second": 5966.971
    },
    {
      "epoch": 1.7742528735632184,
      "grad_norm": 1.0570164918899536,
      "learning_rate": 2.263942573163998e-05,
      "loss": 1.5498,
      "num_input_tokens_seen": 29163760,
      "step": 4824,
      "train_runtime": 4887.494,
      "train_tokens_per_second": 5967.017
    },
    {
      "epoch": 1.7746206896551724,
      "grad_norm": 0.952113151550293,
      "learning_rate": 2.2602613657279588e-05,
      "loss": 1.3292,
      "num_input_tokens_seen": 29170272,
      "step": 4825,
      "train_runtime": 4888.5315,
      "train_tokens_per_second": 5967.083
    },
    {
      "epoch": 1.7749885057471264,
      "grad_norm": 0.9960702061653137,
      "learning_rate": 2.25658015829192e-05,
      "loss": 1.448,
      "num_input_tokens_seen": 29176720,
      "step": 4826,
      "train_runtime": 4889.5676,
      "train_tokens_per_second": 5967.137
    },
    {
      "epoch": 1.7753563218390804,
      "grad_norm": 1.0668630599975586,
      "learning_rate": 2.252898950855881e-05,
      "loss": 1.4394,
      "num_input_tokens_seen": 29182928,
      "step": 4827,
      "train_runtime": 4890.573,
      "train_tokens_per_second": 5967.18
    },
    {
      "epoch": 1.7757241379310345,
      "grad_norm": 0.9461020231246948,
      "learning_rate": 2.2492177434198418e-05,
      "loss": 1.5087,
      "num_input_tokens_seen": 29188096,
      "step": 4828,
      "train_runtime": 4891.4342,
      "train_tokens_per_second": 5967.186
    },
    {
      "epoch": 1.7760919540229885,
      "grad_norm": 0.9859970211982727,
      "learning_rate": 2.245536535983803e-05,
      "loss": 1.4893,
      "num_input_tokens_seen": 29194112,
      "step": 4829,
      "train_runtime": 4892.4234,
      "train_tokens_per_second": 5967.209
    },
    {
      "epoch": 1.7764597701149425,
      "grad_norm": 1.0894885063171387,
      "learning_rate": 2.2418553285477636e-05,
      "loss": 1.542,
      "num_input_tokens_seen": 29200160,
      "step": 4830,
      "train_runtime": 4893.384,
      "train_tokens_per_second": 5967.273
    },
    {
      "epoch": 1.7768275862068965,
      "grad_norm": 1.0765576362609863,
      "learning_rate": 2.2381741211117248e-05,
      "loss": 1.6309,
      "num_input_tokens_seen": 29205088,
      "step": 4831,
      "train_runtime": 4894.7037,
      "train_tokens_per_second": 5966.671
    },
    {
      "epoch": 1.7771954022988505,
      "grad_norm": 1.0274276733398438,
      "learning_rate": 2.2344929136756857e-05,
      "loss": 1.5739,
      "num_input_tokens_seen": 29210896,
      "step": 4832,
      "train_runtime": 4895.6499,
      "train_tokens_per_second": 5966.704
    },
    {
      "epoch": 1.7775632183908046,
      "grad_norm": 1.1504994630813599,
      "learning_rate": 2.2308117062396466e-05,
      "loss": 1.4678,
      "num_input_tokens_seen": 29216224,
      "step": 4833,
      "train_runtime": 4896.5272,
      "train_tokens_per_second": 5966.724
    },
    {
      "epoch": 1.7779310344827586,
      "grad_norm": 1.0065832138061523,
      "learning_rate": 2.2271304988036075e-05,
      "loss": 1.585,
      "num_input_tokens_seen": 29221440,
      "step": 4834,
      "train_runtime": 4897.4184,
      "train_tokens_per_second": 5966.703
    },
    {
      "epoch": 1.7782988505747126,
      "grad_norm": 1.0450266599655151,
      "learning_rate": 2.2234492913675687e-05,
      "loss": 1.5991,
      "num_input_tokens_seen": 29227648,
      "step": 4835,
      "train_runtime": 4898.4592,
      "train_tokens_per_second": 5966.702
    },
    {
      "epoch": 1.7786666666666666,
      "grad_norm": 0.8858741521835327,
      "learning_rate": 2.2197680839315296e-05,
      "loss": 1.3514,
      "num_input_tokens_seen": 29236688,
      "step": 4836,
      "train_runtime": 4899.8325,
      "train_tokens_per_second": 5966.875
    },
    {
      "epoch": 1.7790344827586206,
      "grad_norm": 1.0070812702178955,
      "learning_rate": 2.2160868764954905e-05,
      "loss": 1.5729,
      "num_input_tokens_seen": 29241776,
      "step": 4837,
      "train_runtime": 4900.6922,
      "train_tokens_per_second": 5966.866
    },
    {
      "epoch": 1.7794022988505747,
      "grad_norm": 0.9339779019355774,
      "learning_rate": 2.2124056690594517e-05,
      "loss": 1.3212,
      "num_input_tokens_seen": 29247008,
      "step": 4838,
      "train_runtime": 4901.5648,
      "train_tokens_per_second": 5966.872
    },
    {
      "epoch": 1.7797701149425287,
      "grad_norm": 1.102351188659668,
      "learning_rate": 2.2087244616234126e-05,
      "loss": 1.9419,
      "num_input_tokens_seen": 29251728,
      "step": 4839,
      "train_runtime": 4902.3962,
      "train_tokens_per_second": 5966.823
    },
    {
      "epoch": 1.7801379310344827,
      "grad_norm": 0.961283802986145,
      "learning_rate": 2.2050432541873735e-05,
      "loss": 1.3867,
      "num_input_tokens_seen": 29256288,
      "step": 4840,
      "train_runtime": 4903.185,
      "train_tokens_per_second": 5966.793
    },
    {
      "epoch": 1.780505747126437,
      "grad_norm": 1.037157416343689,
      "learning_rate": 2.2013620467513347e-05,
      "loss": 1.6612,
      "num_input_tokens_seen": 29261680,
      "step": 4841,
      "train_runtime": 4904.0736,
      "train_tokens_per_second": 5966.811
    },
    {
      "epoch": 1.7808735632183907,
      "grad_norm": 1.0962274074554443,
      "learning_rate": 2.1976808393152956e-05,
      "loss": 1.6382,
      "num_input_tokens_seen": 29266512,
      "step": 4842,
      "train_runtime": 4904.9043,
      "train_tokens_per_second": 5966.786
    },
    {
      "epoch": 1.781241379310345,
      "grad_norm": 1.1156631708145142,
      "learning_rate": 2.1939996318792565e-05,
      "loss": 1.5553,
      "num_input_tokens_seen": 29271504,
      "step": 4843,
      "train_runtime": 4905.7548,
      "train_tokens_per_second": 5966.769
    },
    {
      "epoch": 1.7816091954022988,
      "grad_norm": 0.9504975080490112,
      "learning_rate": 2.1903184244432177e-05,
      "loss": 1.5963,
      "num_input_tokens_seen": 29278896,
      "step": 4844,
      "train_runtime": 4906.9192,
      "train_tokens_per_second": 5966.859
    },
    {
      "epoch": 1.781977011494253,
      "grad_norm": 0.9749406576156616,
      "learning_rate": 2.1866372170071783e-05,
      "loss": 1.5224,
      "num_input_tokens_seen": 29283696,
      "step": 4845,
      "train_runtime": 4907.7565,
      "train_tokens_per_second": 5966.819
    },
    {
      "epoch": 1.7823448275862068,
      "grad_norm": 1.0290385484695435,
      "learning_rate": 2.182956009571139e-05,
      "loss": 1.5321,
      "num_input_tokens_seen": 29288256,
      "step": 4846,
      "train_runtime": 4908.553,
      "train_tokens_per_second": 5966.78
    },
    {
      "epoch": 1.782712643678161,
      "grad_norm": 1.05253267288208,
      "learning_rate": 2.1792748021351004e-05,
      "loss": 1.5027,
      "num_input_tokens_seen": 29295424,
      "step": 4847,
      "train_runtime": 4909.667,
      "train_tokens_per_second": 5966.886
    },
    {
      "epoch": 1.7830804597701149,
      "grad_norm": 0.9805688261985779,
      "learning_rate": 2.1755935946990613e-05,
      "loss": 1.6608,
      "num_input_tokens_seen": 29300176,
      "step": 4848,
      "train_runtime": 4910.4863,
      "train_tokens_per_second": 5966.858
    },
    {
      "epoch": 1.783448275862069,
      "grad_norm": 1.0403445959091187,
      "learning_rate": 2.171912387263022e-05,
      "loss": 1.6362,
      "num_input_tokens_seen": 29305040,
      "step": 4849,
      "train_runtime": 4911.3193,
      "train_tokens_per_second": 5966.837
    },
    {
      "epoch": 1.783816091954023,
      "grad_norm": 0.9725497364997864,
      "learning_rate": 2.1682311798269834e-05,
      "loss": 1.4648,
      "num_input_tokens_seen": 29312720,
      "step": 4850,
      "train_runtime": 4912.5235,
      "train_tokens_per_second": 5966.937
    },
    {
      "epoch": 1.7841839080459772,
      "grad_norm": 1.0356947183609009,
      "learning_rate": 2.1645499723909443e-05,
      "loss": 1.3962,
      "num_input_tokens_seen": 29316992,
      "step": 4851,
      "train_runtime": 4913.2914,
      "train_tokens_per_second": 5966.874
    },
    {
      "epoch": 1.784551724137931,
      "grad_norm": 0.8753128051757812,
      "learning_rate": 2.1608687649549052e-05,
      "loss": 1.438,
      "num_input_tokens_seen": 29325200,
      "step": 4852,
      "train_runtime": 4914.5591,
      "train_tokens_per_second": 5967.005
    },
    {
      "epoch": 1.7849195402298852,
      "grad_norm": 1.0391000509262085,
      "learning_rate": 2.1571875575188664e-05,
      "loss": 1.5341,
      "num_input_tokens_seen": 29329904,
      "step": 4853,
      "train_runtime": 4915.3783,
      "train_tokens_per_second": 5966.968
    },
    {
      "epoch": 1.785287356321839,
      "grad_norm": 0.9996411800384521,
      "learning_rate": 2.1535063500828273e-05,
      "loss": 1.4621,
      "num_input_tokens_seen": 29335248,
      "step": 4854,
      "train_runtime": 4916.2539,
      "train_tokens_per_second": 5966.992
    },
    {
      "epoch": 1.7856551724137932,
      "grad_norm": 1.1347910165786743,
      "learning_rate": 2.1498251426467882e-05,
      "loss": 1.6497,
      "num_input_tokens_seen": 29340016,
      "step": 4855,
      "train_runtime": 4917.076,
      "train_tokens_per_second": 5966.964
    },
    {
      "epoch": 1.786022988505747,
      "grad_norm": 1.0457426309585571,
      "learning_rate": 2.1461439352107494e-05,
      "loss": 1.7235,
      "num_input_tokens_seen": 29345408,
      "step": 4856,
      "train_runtime": 4917.9835,
      "train_tokens_per_second": 5966.959
    },
    {
      "epoch": 1.7863908045977013,
      "grad_norm": 0.764273464679718,
      "learning_rate": 2.1424627277747103e-05,
      "loss": 1.4666,
      "num_input_tokens_seen": 29357040,
      "step": 4857,
      "train_runtime": 4919.7168,
      "train_tokens_per_second": 5967.221
    },
    {
      "epoch": 1.786758620689655,
      "grad_norm": 0.9954184293746948,
      "learning_rate": 2.1387815203386712e-05,
      "loss": 1.42,
      "num_input_tokens_seen": 29363696,
      "step": 4858,
      "train_runtime": 4920.8014,
      "train_tokens_per_second": 5967.259
    },
    {
      "epoch": 1.7871264367816093,
      "grad_norm": 0.96946781873703,
      "learning_rate": 2.1351003129026324e-05,
      "loss": 1.4562,
      "num_input_tokens_seen": 29368816,
      "step": 4859,
      "train_runtime": 4921.6507,
      "train_tokens_per_second": 5967.269
    },
    {
      "epoch": 1.7874942528735631,
      "grad_norm": 1.0975924730300903,
      "learning_rate": 2.131419105466593e-05,
      "loss": 1.6143,
      "num_input_tokens_seen": 29373392,
      "step": 4860,
      "train_runtime": 4922.4649,
      "train_tokens_per_second": 5967.212
    },
    {
      "epoch": 1.7878620689655174,
      "grad_norm": 1.04430091381073,
      "learning_rate": 2.127737898030554e-05,
      "loss": 1.7536,
      "num_input_tokens_seen": 29378240,
      "step": 4861,
      "train_runtime": 4923.7817,
      "train_tokens_per_second": 5966.601
    },
    {
      "epoch": 1.7882298850574712,
      "grad_norm": 0.9940053820610046,
      "learning_rate": 2.124056690594515e-05,
      "loss": 1.3992,
      "num_input_tokens_seen": 29385072,
      "step": 4862,
      "train_runtime": 4924.8707,
      "train_tokens_per_second": 5966.669
    },
    {
      "epoch": 1.7885977011494254,
      "grad_norm": 1.0616343021392822,
      "learning_rate": 2.120375483158476e-05,
      "loss": 1.5123,
      "num_input_tokens_seen": 29390512,
      "step": 4863,
      "train_runtime": 4925.7738,
      "train_tokens_per_second": 5966.679
    },
    {
      "epoch": 1.7889655172413792,
      "grad_norm": 1.0299736261367798,
      "learning_rate": 2.116694275722437e-05,
      "loss": 1.3869,
      "num_input_tokens_seen": 29396112,
      "step": 4864,
      "train_runtime": 4926.6776,
      "train_tokens_per_second": 5966.721
    },
    {
      "epoch": 1.7893333333333334,
      "grad_norm": 1.010871410369873,
      "learning_rate": 2.113013068286398e-05,
      "loss": 1.4701,
      "num_input_tokens_seen": 29401568,
      "step": 4865,
      "train_runtime": 4927.5734,
      "train_tokens_per_second": 5966.744
    },
    {
      "epoch": 1.7897011494252872,
      "grad_norm": 1.0078232288360596,
      "learning_rate": 2.109331860850359e-05,
      "loss": 1.675,
      "num_input_tokens_seen": 29406912,
      "step": 4866,
      "train_runtime": 4928.4472,
      "train_tokens_per_second": 5966.77
    },
    {
      "epoch": 1.7900689655172415,
      "grad_norm": 1.0409224033355713,
      "learning_rate": 2.10565065341432e-05,
      "loss": 1.6477,
      "num_input_tokens_seen": 29414128,
      "step": 4867,
      "train_runtime": 4929.6125,
      "train_tokens_per_second": 5966.824
    },
    {
      "epoch": 1.7904367816091953,
      "grad_norm": 1.0885225534439087,
      "learning_rate": 2.101969445978281e-05,
      "loss": 1.6365,
      "num_input_tokens_seen": 29419104,
      "step": 4868,
      "train_runtime": 4930.4379,
      "train_tokens_per_second": 5966.834
    },
    {
      "epoch": 1.7908045977011495,
      "grad_norm": 1.1059404611587524,
      "learning_rate": 2.098288238542242e-05,
      "loss": 1.4918,
      "num_input_tokens_seen": 29424800,
      "step": 4869,
      "train_runtime": 4931.4029,
      "train_tokens_per_second": 5966.821
    },
    {
      "epoch": 1.7911724137931033,
      "grad_norm": 1.0508280992507935,
      "learning_rate": 2.094607031106203e-05,
      "loss": 1.7301,
      "num_input_tokens_seen": 29430720,
      "step": 4870,
      "train_runtime": 4932.345,
      "train_tokens_per_second": 5966.882
    },
    {
      "epoch": 1.7915402298850576,
      "grad_norm": 1.060753345489502,
      "learning_rate": 2.090925823670164e-05,
      "loss": 1.5117,
      "num_input_tokens_seen": 29436736,
      "step": 4871,
      "train_runtime": 4933.3426,
      "train_tokens_per_second": 5966.895
    },
    {
      "epoch": 1.7919080459770114,
      "grad_norm": 1.097510814666748,
      "learning_rate": 2.087244616234125e-05,
      "loss": 1.4251,
      "num_input_tokens_seen": 29442992,
      "step": 4872,
      "train_runtime": 4934.3538,
      "train_tokens_per_second": 5966.94
    },
    {
      "epoch": 1.7922758620689656,
      "grad_norm": 1.0107671022415161,
      "learning_rate": 2.083563408798086e-05,
      "loss": 1.3961,
      "num_input_tokens_seen": 29449536,
      "step": 4873,
      "train_runtime": 4935.4079,
      "train_tokens_per_second": 5966.991
    },
    {
      "epoch": 1.7926436781609194,
      "grad_norm": 1.034067153930664,
      "learning_rate": 2.079882201362047e-05,
      "loss": 1.6524,
      "num_input_tokens_seen": 29455280,
      "step": 4874,
      "train_runtime": 4936.3636,
      "train_tokens_per_second": 5967.0
    },
    {
      "epoch": 1.7930114942528736,
      "grad_norm": 1.1237860918045044,
      "learning_rate": 2.0762009939260077e-05,
      "loss": 1.5278,
      "num_input_tokens_seen": 29460208,
      "step": 4875,
      "train_runtime": 4937.2136,
      "train_tokens_per_second": 5966.971
    },
    {
      "epoch": 1.7933793103448274,
      "grad_norm": 1.0731533765792847,
      "learning_rate": 2.0725197864899685e-05,
      "loss": 1.4303,
      "num_input_tokens_seen": 29465712,
      "step": 4876,
      "train_runtime": 4938.1468,
      "train_tokens_per_second": 5966.957
    },
    {
      "epoch": 1.7937471264367817,
      "grad_norm": 0.9860634803771973,
      "learning_rate": 2.0688385790539298e-05,
      "loss": 1.6109,
      "num_input_tokens_seen": 29470576,
      "step": 4877,
      "train_runtime": 4938.9784,
      "train_tokens_per_second": 5966.938
    },
    {
      "epoch": 1.7941149425287355,
      "grad_norm": 1.0423297882080078,
      "learning_rate": 2.0651573716178907e-05,
      "loss": 1.3816,
      "num_input_tokens_seen": 29477376,
      "step": 4878,
      "train_runtime": 4940.0849,
      "train_tokens_per_second": 5966.978
    },
    {
      "epoch": 1.7944827586206897,
      "grad_norm": 0.9954624772071838,
      "learning_rate": 2.0614761641818516e-05,
      "loss": 1.5698,
      "num_input_tokens_seen": 29483520,
      "step": 4879,
      "train_runtime": 4941.1123,
      "train_tokens_per_second": 5966.98
    },
    {
      "epoch": 1.7948505747126435,
      "grad_norm": 1.064196228981018,
      "learning_rate": 2.0577949567458128e-05,
      "loss": 1.5358,
      "num_input_tokens_seen": 29488608,
      "step": 4880,
      "train_runtime": 4941.9799,
      "train_tokens_per_second": 5966.962
    },
    {
      "epoch": 1.7952183908045978,
      "grad_norm": 1.0329298973083496,
      "learning_rate": 2.0541137493097737e-05,
      "loss": 1.4956,
      "num_input_tokens_seen": 29494544,
      "step": 4881,
      "train_runtime": 4942.9524,
      "train_tokens_per_second": 5966.989
    },
    {
      "epoch": 1.7955862068965516,
      "grad_norm": 1.177534818649292,
      "learning_rate": 2.0504325418737346e-05,
      "loss": 1.8409,
      "num_input_tokens_seen": 29499264,
      "step": 4882,
      "train_runtime": 4943.761,
      "train_tokens_per_second": 5966.968
    },
    {
      "epoch": 1.7959540229885058,
      "grad_norm": 1.0485416650772095,
      "learning_rate": 2.0467513344376958e-05,
      "loss": 1.4265,
      "num_input_tokens_seen": 29504288,
      "step": 4883,
      "train_runtime": 4944.6175,
      "train_tokens_per_second": 5966.951
    },
    {
      "epoch": 1.7963218390804596,
      "grad_norm": 0.9922838807106018,
      "learning_rate": 2.0430701270016567e-05,
      "loss": 1.5442,
      "num_input_tokens_seen": 29510000,
      "step": 4884,
      "train_runtime": 4945.5612,
      "train_tokens_per_second": 5966.967
    },
    {
      "epoch": 1.7966896551724139,
      "grad_norm": 1.1228270530700684,
      "learning_rate": 2.0393889195656176e-05,
      "loss": 1.5242,
      "num_input_tokens_seen": 29516288,
      "step": 4885,
      "train_runtime": 4946.5663,
      "train_tokens_per_second": 5967.026
    },
    {
      "epoch": 1.7970574712643677,
      "grad_norm": 0.959301769733429,
      "learning_rate": 2.0357077121295788e-05,
      "loss": 1.4251,
      "num_input_tokens_seen": 29520640,
      "step": 4886,
      "train_runtime": 4947.3123,
      "train_tokens_per_second": 5967.006
    },
    {
      "epoch": 1.797425287356322,
      "grad_norm": 1.0731889009475708,
      "learning_rate": 2.0320265046935397e-05,
      "loss": 1.8014,
      "num_input_tokens_seen": 29525312,
      "step": 4887,
      "train_runtime": 4948.1407,
      "train_tokens_per_second": 5966.951
    },
    {
      "epoch": 1.7977931034482757,
      "grad_norm": 1.0586031675338745,
      "learning_rate": 2.0283452972575006e-05,
      "loss": 1.5994,
      "num_input_tokens_seen": 29532176,
      "step": 4888,
      "train_runtime": 4949.2349,
      "train_tokens_per_second": 5967.018
    },
    {
      "epoch": 1.79816091954023,
      "grad_norm": 1.127091407775879,
      "learning_rate": 2.0246640898214618e-05,
      "loss": 1.6764,
      "num_input_tokens_seen": 29538368,
      "step": 4889,
      "train_runtime": 4950.2623,
      "train_tokens_per_second": 5967.031
    },
    {
      "epoch": 1.798528735632184,
      "grad_norm": 0.9802016615867615,
      "learning_rate": 2.0209828823854223e-05,
      "loss": 1.4589,
      "num_input_tokens_seen": 29545776,
      "step": 4890,
      "train_runtime": 4951.4202,
      "train_tokens_per_second": 5967.132
    },
    {
      "epoch": 1.798896551724138,
      "grad_norm": 0.9854152798652649,
      "learning_rate": 2.0173016749493832e-05,
      "loss": 1.5698,
      "num_input_tokens_seen": 29550576,
      "step": 4891,
      "train_runtime": 4952.7294,
      "train_tokens_per_second": 5966.523
    },
    {
      "epoch": 1.799264367816092,
      "grad_norm": 1.0517677068710327,
      "learning_rate": 2.0136204675133445e-05,
      "loss": 1.2653,
      "num_input_tokens_seen": 29556544,
      "step": 4892,
      "train_runtime": 4953.6982,
      "train_tokens_per_second": 5966.561
    },
    {
      "epoch": 1.799632183908046,
      "grad_norm": 1.0668220520019531,
      "learning_rate": 2.0099392600773054e-05,
      "loss": 1.5209,
      "num_input_tokens_seen": 29566608,
      "step": 4893,
      "train_runtime": 4955.1926,
      "train_tokens_per_second": 5966.793
    },
    {
      "epoch": 1.8,
      "grad_norm": 1.0575467348098755,
      "learning_rate": 2.0062580526412662e-05,
      "loss": 1.4832,
      "num_input_tokens_seen": 29572256,
      "step": 4894,
      "train_runtime": 4956.1134,
      "train_tokens_per_second": 5966.824
    },
    {
      "epoch": 1.800367816091954,
      "grad_norm": 1.0190519094467163,
      "learning_rate": 2.0025768452052275e-05,
      "loss": 1.4475,
      "num_input_tokens_seen": 29577856,
      "step": 4895,
      "train_runtime": 4957.0745,
      "train_tokens_per_second": 5966.797
    },
    {
      "epoch": 1.800735632183908,
      "grad_norm": 0.9429070353507996,
      "learning_rate": 1.9988956377691884e-05,
      "loss": 1.3044,
      "num_input_tokens_seen": 29584496,
      "step": 4896,
      "train_runtime": 4958.1786,
      "train_tokens_per_second": 5966.807
    },
    {
      "epoch": 1.800735632183908,
      "eval_loss": 1.7461637258529663,
      "eval_runtime": 4.7811,
      "eval_samples_per_second": 209.158,
      "eval_steps_per_second": 13.177,
      "num_input_tokens_seen": 29584496,
      "step": 4896
    },
    {
      "epoch": 1.801103448275862,
      "grad_norm": 1.1192717552185059,
      "learning_rate": 1.9952144303331493e-05,
      "loss": 1.6999,
      "num_input_tokens_seen": 29588800,
      "step": 4897,
      "train_runtime": 4963.7078,
      "train_tokens_per_second": 5961.028
    },
    {
      "epoch": 1.8014712643678161,
      "grad_norm": 1.0206178426742554,
      "learning_rate": 1.9915332228971105e-05,
      "loss": 1.4832,
      "num_input_tokens_seen": 29594000,
      "step": 4898,
      "train_runtime": 4964.5758,
      "train_tokens_per_second": 5961.033
    },
    {
      "epoch": 1.8018390804597701,
      "grad_norm": 1.058915615081787,
      "learning_rate": 1.9878520154610714e-05,
      "loss": 1.5559,
      "num_input_tokens_seen": 29602016,
      "step": 4899,
      "train_runtime": 4965.8222,
      "train_tokens_per_second": 5961.151
    },
    {
      "epoch": 1.8022068965517242,
      "grad_norm": 1.0015265941619873,
      "learning_rate": 1.9841708080250323e-05,
      "loss": 1.3341,
      "num_input_tokens_seen": 29609232,
      "step": 4900,
      "train_runtime": 4966.9688,
      "train_tokens_per_second": 5961.228
    },
    {
      "epoch": 1.8025747126436782,
      "grad_norm": 1.034872055053711,
      "learning_rate": 1.9804896005889935e-05,
      "loss": 1.4087,
      "num_input_tokens_seen": 29615632,
      "step": 4901,
      "train_runtime": 4968.0165,
      "train_tokens_per_second": 5961.259
    },
    {
      "epoch": 1.8029425287356322,
      "grad_norm": 0.9960013031959534,
      "learning_rate": 1.9768083931529544e-05,
      "loss": 1.5575,
      "num_input_tokens_seen": 29621152,
      "step": 4902,
      "train_runtime": 4968.9369,
      "train_tokens_per_second": 5961.265
    },
    {
      "epoch": 1.8033103448275862,
      "grad_norm": 1.0279793739318848,
      "learning_rate": 1.9731271857169153e-05,
      "loss": 1.4925,
      "num_input_tokens_seen": 29627232,
      "step": 4903,
      "train_runtime": 4969.9196,
      "train_tokens_per_second": 5961.31
    },
    {
      "epoch": 1.8036781609195403,
      "grad_norm": 1.1436517238616943,
      "learning_rate": 1.9694459782808765e-05,
      "loss": 1.6609,
      "num_input_tokens_seen": 29633072,
      "step": 4904,
      "train_runtime": 4970.8903,
      "train_tokens_per_second": 5961.321
    },
    {
      "epoch": 1.8040459770114943,
      "grad_norm": 1.0318900346755981,
      "learning_rate": 1.965764770844837e-05,
      "loss": 1.4023,
      "num_input_tokens_seen": 29640416,
      "step": 4905,
      "train_runtime": 4972.0698,
      "train_tokens_per_second": 5961.384
    },
    {
      "epoch": 1.8044137931034483,
      "grad_norm": 0.9956551194190979,
      "learning_rate": 1.962083563408798e-05,
      "loss": 1.4185,
      "num_input_tokens_seen": 29645680,
      "step": 4906,
      "train_runtime": 4972.9504,
      "train_tokens_per_second": 5961.387
    },
    {
      "epoch": 1.8047816091954023,
      "grad_norm": 1.0204379558563232,
      "learning_rate": 1.958402355972759e-05,
      "loss": 1.4424,
      "num_input_tokens_seen": 29652608,
      "step": 4907,
      "train_runtime": 4974.0547,
      "train_tokens_per_second": 5961.456
    },
    {
      "epoch": 1.8051494252873563,
      "grad_norm": 0.9458982944488525,
      "learning_rate": 1.95472114853672e-05,
      "loss": 1.5268,
      "num_input_tokens_seen": 29659936,
      "step": 4908,
      "train_runtime": 4975.2291,
      "train_tokens_per_second": 5961.522
    },
    {
      "epoch": 1.8055172413793104,
      "grad_norm": 0.9880759119987488,
      "learning_rate": 1.951039941100681e-05,
      "loss": 1.4263,
      "num_input_tokens_seen": 29668608,
      "step": 4909,
      "train_runtime": 4976.5683,
      "train_tokens_per_second": 5961.66
    },
    {
      "epoch": 1.8058850574712644,
      "grad_norm": 1.0818374156951904,
      "learning_rate": 1.947358733664642e-05,
      "loss": 1.5683,
      "num_input_tokens_seen": 29672544,
      "step": 4910,
      "train_runtime": 4977.2607,
      "train_tokens_per_second": 5961.621
    },
    {
      "epoch": 1.8062528735632184,
      "grad_norm": 0.9811917543411255,
      "learning_rate": 1.943677526228603e-05,
      "loss": 1.4595,
      "num_input_tokens_seen": 29678064,
      "step": 4911,
      "train_runtime": 4978.1486,
      "train_tokens_per_second": 5961.667
    },
    {
      "epoch": 1.8066206896551724,
      "grad_norm": 1.09088933467865,
      "learning_rate": 1.939996318792564e-05,
      "loss": 1.3414,
      "num_input_tokens_seen": 29682560,
      "step": 4912,
      "train_runtime": 4978.9376,
      "train_tokens_per_second": 5961.625
    },
    {
      "epoch": 1.8069885057471264,
      "grad_norm": 0.9992160201072693,
      "learning_rate": 1.9363151113565252e-05,
      "loss": 1.4584,
      "num_input_tokens_seen": 29687968,
      "step": 4913,
      "train_runtime": 4979.7978,
      "train_tokens_per_second": 5961.681
    },
    {
      "epoch": 1.8073563218390805,
      "grad_norm": 0.9398074746131897,
      "learning_rate": 1.932633903920486e-05,
      "loss": 1.3795,
      "num_input_tokens_seen": 29697072,
      "step": 4914,
      "train_runtime": 4981.1889,
      "train_tokens_per_second": 5961.844
    },
    {
      "epoch": 1.8077241379310345,
      "grad_norm": 0.9479207992553711,
      "learning_rate": 1.928952696484447e-05,
      "loss": 1.5688,
      "num_input_tokens_seen": 29704032,
      "step": 4915,
      "train_runtime": 4982.315,
      "train_tokens_per_second": 5961.894
    },
    {
      "epoch": 1.8080919540229885,
      "grad_norm": 1.0464063882827759,
      "learning_rate": 1.9252714890484082e-05,
      "loss": 1.3561,
      "num_input_tokens_seen": 29711520,
      "step": 4916,
      "train_runtime": 4983.522,
      "train_tokens_per_second": 5961.952
    },
    {
      "epoch": 1.8084597701149425,
      "grad_norm": 0.9227623343467712,
      "learning_rate": 1.921590281612369e-05,
      "loss": 1.4569,
      "num_input_tokens_seen": 29719200,
      "step": 4917,
      "train_runtime": 4984.7303,
      "train_tokens_per_second": 5962.048
    },
    {
      "epoch": 1.8088275862068965,
      "grad_norm": 0.999740719795227,
      "learning_rate": 1.91790907417633e-05,
      "loss": 1.1996,
      "num_input_tokens_seen": 29726256,
      "step": 4918,
      "train_runtime": 4985.8429,
      "train_tokens_per_second": 5962.132
    },
    {
      "epoch": 1.8091954022988506,
      "grad_norm": 0.9896454811096191,
      "learning_rate": 1.9142278667402912e-05,
      "loss": 1.5592,
      "num_input_tokens_seen": 29730624,
      "step": 4919,
      "train_runtime": 4986.6278,
      "train_tokens_per_second": 5962.07
    },
    {
      "epoch": 1.8095632183908046,
      "grad_norm": 1.013528823852539,
      "learning_rate": 1.9105466593042517e-05,
      "loss": 1.5625,
      "num_input_tokens_seen": 29735712,
      "step": 4920,
      "train_runtime": 4987.4896,
      "train_tokens_per_second": 5962.06
    },
    {
      "epoch": 1.8099310344827586,
      "grad_norm": 1.0473814010620117,
      "learning_rate": 1.9068654518682126e-05,
      "loss": 1.4979,
      "num_input_tokens_seen": 29742608,
      "step": 4921,
      "train_runtime": 4989.1729,
      "train_tokens_per_second": 5961.431
    },
    {
      "epoch": 1.8102988505747126,
      "grad_norm": 1.0424458980560303,
      "learning_rate": 1.903184244432174e-05,
      "loss": 1.504,
      "num_input_tokens_seen": 29747888,
      "step": 4922,
      "train_runtime": 4990.0563,
      "train_tokens_per_second": 5961.433
    },
    {
      "epoch": 1.8106666666666666,
      "grad_norm": 1.1431024074554443,
      "learning_rate": 1.8995030369961347e-05,
      "loss": 1.6549,
      "num_input_tokens_seen": 29753744,
      "step": 4923,
      "train_runtime": 4991.0235,
      "train_tokens_per_second": 5961.451
    },
    {
      "epoch": 1.8110344827586207,
      "grad_norm": 1.0633981227874756,
      "learning_rate": 1.8958218295600956e-05,
      "loss": 1.2825,
      "num_input_tokens_seen": 29760976,
      "step": 4924,
      "train_runtime": 4992.156,
      "train_tokens_per_second": 5961.548
    },
    {
      "epoch": 1.8114022988505747,
      "grad_norm": 0.9903343915939331,
      "learning_rate": 1.892140622124057e-05,
      "loss": 1.456,
      "num_input_tokens_seen": 29768960,
      "step": 4925,
      "train_runtime": 4993.4291,
      "train_tokens_per_second": 5961.627
    },
    {
      "epoch": 1.8117701149425287,
      "grad_norm": 1.053836464881897,
      "learning_rate": 1.8884594146880177e-05,
      "loss": 1.5849,
      "num_input_tokens_seen": 29773296,
      "step": 4926,
      "train_runtime": 4994.2032,
      "train_tokens_per_second": 5961.571
    },
    {
      "epoch": 1.8121379310344827,
      "grad_norm": 1.002320408821106,
      "learning_rate": 1.8847782072519786e-05,
      "loss": 1.3932,
      "num_input_tokens_seen": 29778848,
      "step": 4927,
      "train_runtime": 4995.1133,
      "train_tokens_per_second": 5961.596
    },
    {
      "epoch": 1.8125057471264368,
      "grad_norm": 1.0374735593795776,
      "learning_rate": 1.88109699981594e-05,
      "loss": 1.7077,
      "num_input_tokens_seen": 29783552,
      "step": 4928,
      "train_runtime": 4995.9102,
      "train_tokens_per_second": 5961.587
    },
    {
      "epoch": 1.8128735632183908,
      "grad_norm": 1.0954574346542358,
      "learning_rate": 1.8774157923799008e-05,
      "loss": 1.5615,
      "num_input_tokens_seen": 29789984,
      "step": 4929,
      "train_runtime": 4996.958,
      "train_tokens_per_second": 5961.624
    },
    {
      "epoch": 1.8132413793103448,
      "grad_norm": 1.0004336833953857,
      "learning_rate": 1.8737345849438616e-05,
      "loss": 1.4386,
      "num_input_tokens_seen": 29795024,
      "step": 4930,
      "train_runtime": 4997.791,
      "train_tokens_per_second": 5961.639
    },
    {
      "epoch": 1.8136091954022988,
      "grad_norm": 0.9620081186294556,
      "learning_rate": 1.870053377507823e-05,
      "loss": 1.6004,
      "num_input_tokens_seen": 29800288,
      "step": 4931,
      "train_runtime": 4998.6833,
      "train_tokens_per_second": 5961.628
    },
    {
      "epoch": 1.8139770114942528,
      "grad_norm": 0.9992434978485107,
      "learning_rate": 1.8663721700717838e-05,
      "loss": 1.385,
      "num_input_tokens_seen": 29805904,
      "step": 4932,
      "train_runtime": 4999.6132,
      "train_tokens_per_second": 5961.642
    },
    {
      "epoch": 1.8143448275862069,
      "grad_norm": 0.947857677936554,
      "learning_rate": 1.8626909626357447e-05,
      "loss": 1.3974,
      "num_input_tokens_seen": 29811360,
      "step": 4933,
      "train_runtime": 5000.5313,
      "train_tokens_per_second": 5961.638
    },
    {
      "epoch": 1.8147126436781609,
      "grad_norm": 0.9161631464958191,
      "learning_rate": 1.859009755199706e-05,
      "loss": 1.3631,
      "num_input_tokens_seen": 29816976,
      "step": 4934,
      "train_runtime": 5001.4925,
      "train_tokens_per_second": 5961.616
    },
    {
      "epoch": 1.815080459770115,
      "grad_norm": 1.0192244052886963,
      "learning_rate": 1.8553285477636664e-05,
      "loss": 1.5762,
      "num_input_tokens_seen": 29824928,
      "step": 4935,
      "train_runtime": 5002.7636,
      "train_tokens_per_second": 5961.69
    },
    {
      "epoch": 1.815448275862069,
      "grad_norm": 0.9719940423965454,
      "learning_rate": 1.8516473403276273e-05,
      "loss": 1.551,
      "num_input_tokens_seen": 29829552,
      "step": 4936,
      "train_runtime": 5003.5594,
      "train_tokens_per_second": 5961.666
    },
    {
      "epoch": 1.815816091954023,
      "grad_norm": 1.1043134927749634,
      "learning_rate": 1.8479661328915885e-05,
      "loss": 1.5287,
      "num_input_tokens_seen": 29835040,
      "step": 4937,
      "train_runtime": 5004.4571,
      "train_tokens_per_second": 5961.694
    },
    {
      "epoch": 1.816183908045977,
      "grad_norm": 0.9655507802963257,
      "learning_rate": 1.8442849254555494e-05,
      "loss": 1.5159,
      "num_input_tokens_seen": 29841312,
      "step": 4938,
      "train_runtime": 5005.4723,
      "train_tokens_per_second": 5961.738
    },
    {
      "epoch": 1.816551724137931,
      "grad_norm": 0.9759043455123901,
      "learning_rate": 1.8406037180195103e-05,
      "loss": 1.6169,
      "num_input_tokens_seen": 29847280,
      "step": 4939,
      "train_runtime": 5006.4614,
      "train_tokens_per_second": 5961.752
    },
    {
      "epoch": 1.8169195402298852,
      "grad_norm": 0.897058367729187,
      "learning_rate": 1.8369225105834716e-05,
      "loss": 1.425,
      "num_input_tokens_seen": 29854144,
      "step": 4940,
      "train_runtime": 5007.5836,
      "train_tokens_per_second": 5961.786
    },
    {
      "epoch": 1.817287356321839,
      "grad_norm": 0.9897091388702393,
      "learning_rate": 1.8332413031474324e-05,
      "loss": 1.5767,
      "num_input_tokens_seen": 29860864,
      "step": 4941,
      "train_runtime": 5008.6566,
      "train_tokens_per_second": 5961.851
    },
    {
      "epoch": 1.8176551724137933,
      "grad_norm": 1.0577644109725952,
      "learning_rate": 1.8295600957113933e-05,
      "loss": 1.5974,
      "num_input_tokens_seen": 29865312,
      "step": 4942,
      "train_runtime": 5009.4388,
      "train_tokens_per_second": 5961.808
    },
    {
      "epoch": 1.818022988505747,
      "grad_norm": 1.2485471963882446,
      "learning_rate": 1.8258788882753546e-05,
      "loss": 1.608,
      "num_input_tokens_seen": 29869232,
      "step": 4943,
      "train_runtime": 5010.1373,
      "train_tokens_per_second": 5961.759
    },
    {
      "epoch": 1.8183908045977013,
      "grad_norm": 1.029123067855835,
      "learning_rate": 1.8221976808393154e-05,
      "loss": 1.4635,
      "num_input_tokens_seen": 29875728,
      "step": 4944,
      "train_runtime": 5011.191,
      "train_tokens_per_second": 5961.802
    },
    {
      "epoch": 1.818758620689655,
      "grad_norm": 0.9065240025520325,
      "learning_rate": 1.8185164734032763e-05,
      "loss": 1.3109,
      "num_input_tokens_seen": 29881136,
      "step": 4945,
      "train_runtime": 5012.0734,
      "train_tokens_per_second": 5961.831
    },
    {
      "epoch": 1.8191264367816093,
      "grad_norm": 1.0743343830108643,
      "learning_rate": 1.8148352659672376e-05,
      "loss": 1.5482,
      "num_input_tokens_seen": 29886016,
      "step": 4946,
      "train_runtime": 5012.9138,
      "train_tokens_per_second": 5961.805
    },
    {
      "epoch": 1.8194942528735631,
      "grad_norm": 0.9645093083381653,
      "learning_rate": 1.8111540585311985e-05,
      "loss": 1.548,
      "num_input_tokens_seen": 29893232,
      "step": 4947,
      "train_runtime": 5014.0596,
      "train_tokens_per_second": 5961.882
    },
    {
      "epoch": 1.8198620689655174,
      "grad_norm": 0.9978023767471313,
      "learning_rate": 1.8074728510951593e-05,
      "loss": 1.6379,
      "num_input_tokens_seen": 29898624,
      "step": 4948,
      "train_runtime": 5014.9542,
      "train_tokens_per_second": 5961.894
    },
    {
      "epoch": 1.8202298850574712,
      "grad_norm": 0.9754540920257568,
      "learning_rate": 1.8037916436591202e-05,
      "loss": 1.2603,
      "num_input_tokens_seen": 29906288,
      "step": 4949,
      "train_runtime": 5016.1689,
      "train_tokens_per_second": 5961.978
    },
    {
      "epoch": 1.8205977011494254,
      "grad_norm": 0.9246507287025452,
      "learning_rate": 1.800110436223081e-05,
      "loss": 1.376,
      "num_input_tokens_seen": 29912096,
      "step": 4950,
      "train_runtime": 5017.1188,
      "train_tokens_per_second": 5962.007
    },
    {
      "epoch": 1.8209655172413792,
      "grad_norm": 1.0795091390609741,
      "learning_rate": 1.796429228787042e-05,
      "loss": 1.5992,
      "num_input_tokens_seen": 29917424,
      "step": 4951,
      "train_runtime": 5018.515,
      "train_tokens_per_second": 5961.41
    },
    {
      "epoch": 1.8213333333333335,
      "grad_norm": 1.0177384614944458,
      "learning_rate": 1.7927480213510032e-05,
      "loss": 1.4097,
      "num_input_tokens_seen": 29925680,
      "step": 4952,
      "train_runtime": 5019.7827,
      "train_tokens_per_second": 5961.549
    },
    {
      "epoch": 1.8217011494252873,
      "grad_norm": 1.0944467782974243,
      "learning_rate": 1.789066813914964e-05,
      "loss": 1.5994,
      "num_input_tokens_seen": 29930848,
      "step": 4953,
      "train_runtime": 5020.6447,
      "train_tokens_per_second": 5961.555
    },
    {
      "epoch": 1.8220689655172415,
      "grad_norm": 0.9916122555732727,
      "learning_rate": 1.785385606478925e-05,
      "loss": 1.3572,
      "num_input_tokens_seen": 29937488,
      "step": 4954,
      "train_runtime": 5021.7292,
      "train_tokens_per_second": 5961.59
    },
    {
      "epoch": 1.8224367816091953,
      "grad_norm": 1.0126090049743652,
      "learning_rate": 1.7817043990428862e-05,
      "loss": 1.2591,
      "num_input_tokens_seen": 29943824,
      "step": 4955,
      "train_runtime": 5022.7418,
      "train_tokens_per_second": 5961.649
    },
    {
      "epoch": 1.8228045977011496,
      "grad_norm": 1.036719799041748,
      "learning_rate": 1.778023191606847e-05,
      "loss": 1.3511,
      "num_input_tokens_seen": 29951008,
      "step": 4956,
      "train_runtime": 5023.8881,
      "train_tokens_per_second": 5961.719
    },
    {
      "epoch": 1.8231724137931034,
      "grad_norm": 0.9798049330711365,
      "learning_rate": 1.774341984170808e-05,
      "loss": 1.5379,
      "num_input_tokens_seen": 29957632,
      "step": 4957,
      "train_runtime": 5024.9772,
      "train_tokens_per_second": 5961.745
    },
    {
      "epoch": 1.8235402298850576,
      "grad_norm": 1.064881682395935,
      "learning_rate": 1.7706607767347693e-05,
      "loss": 1.6163,
      "num_input_tokens_seen": 29963792,
      "step": 4958,
      "train_runtime": 5025.9822,
      "train_tokens_per_second": 5961.778
    },
    {
      "epoch": 1.8239080459770114,
      "grad_norm": 1.0115530490875244,
      "learning_rate": 1.76697956929873e-05,
      "loss": 1.4527,
      "num_input_tokens_seen": 29972816,
      "step": 4959,
      "train_runtime": 5027.369,
      "train_tokens_per_second": 5961.929
    },
    {
      "epoch": 1.8242758620689656,
      "grad_norm": 0.9694092273712158,
      "learning_rate": 1.763298361862691e-05,
      "loss": 1.3431,
      "num_input_tokens_seen": 29977584,
      "step": 4960,
      "train_runtime": 5028.1884,
      "train_tokens_per_second": 5961.905
    },
    {
      "epoch": 1.8246436781609194,
      "grad_norm": 1.1495500802993774,
      "learning_rate": 1.7596171544266523e-05,
      "loss": 1.554,
      "num_input_tokens_seen": 29982912,
      "step": 4961,
      "train_runtime": 5029.0776,
      "train_tokens_per_second": 5961.911
    },
    {
      "epoch": 1.8250114942528737,
      "grad_norm": 1.0569508075714111,
      "learning_rate": 1.755935946990613e-05,
      "loss": 1.6111,
      "num_input_tokens_seen": 29989872,
      "step": 4962,
      "train_runtime": 5030.2012,
      "train_tokens_per_second": 5961.963
    },
    {
      "epoch": 1.8253793103448275,
      "grad_norm": 1.0188040733337402,
      "learning_rate": 1.752254739554574e-05,
      "loss": 1.5268,
      "num_input_tokens_seen": 29994704,
      "step": 4963,
      "train_runtime": 5031.0167,
      "train_tokens_per_second": 5961.957
    },
    {
      "epoch": 1.8257471264367817,
      "grad_norm": 1.0134364366531372,
      "learning_rate": 1.748573532118535e-05,
      "loss": 1.61,
      "num_input_tokens_seen": 29999552,
      "step": 4964,
      "train_runtime": 5031.8552,
      "train_tokens_per_second": 5961.927
    },
    {
      "epoch": 1.8261149425287355,
      "grad_norm": 1.0191245079040527,
      "learning_rate": 1.7448923246824958e-05,
      "loss": 1.5768,
      "num_input_tokens_seen": 30005648,
      "step": 4965,
      "train_runtime": 5032.8494,
      "train_tokens_per_second": 5961.96
    },
    {
      "epoch": 1.8264827586206898,
      "grad_norm": 1.0314635038375854,
      "learning_rate": 1.7412111172464567e-05,
      "loss": 1.631,
      "num_input_tokens_seen": 30010848,
      "step": 4966,
      "train_runtime": 5033.7288,
      "train_tokens_per_second": 5961.952
    },
    {
      "epoch": 1.8268505747126436,
      "grad_norm": 0.9666833281517029,
      "learning_rate": 1.737529909810418e-05,
      "loss": 1.5824,
      "num_input_tokens_seen": 30016016,
      "step": 4967,
      "train_runtime": 5034.6143,
      "train_tokens_per_second": 5961.93
    },
    {
      "epoch": 1.8272183908045978,
      "grad_norm": 1.0156227350234985,
      "learning_rate": 1.7338487023743788e-05,
      "loss": 1.7048,
      "num_input_tokens_seen": 30021456,
      "step": 4968,
      "train_runtime": 5035.5245,
      "train_tokens_per_second": 5961.932
    },
    {
      "epoch": 1.8275862068965516,
      "grad_norm": 0.9774816036224365,
      "learning_rate": 1.7301674949383397e-05,
      "loss": 1.5288,
      "num_input_tokens_seen": 30027440,
      "step": 4969,
      "train_runtime": 5036.4996,
      "train_tokens_per_second": 5961.966
    },
    {
      "epoch": 1.8279540229885058,
      "grad_norm": 1.05318284034729,
      "learning_rate": 1.726486287502301e-05,
      "loss": 1.741,
      "num_input_tokens_seen": 30032416,
      "step": 4970,
      "train_runtime": 5037.334,
      "train_tokens_per_second": 5961.966
    },
    {
      "epoch": 1.8283218390804596,
      "grad_norm": 1.029547095298767,
      "learning_rate": 1.722805080066262e-05,
      "loss": 1.4505,
      "num_input_tokens_seen": 30036864,
      "step": 4971,
      "train_runtime": 5038.109,
      "train_tokens_per_second": 5961.932
    },
    {
      "epoch": 1.8286896551724139,
      "grad_norm": 1.095587968826294,
      "learning_rate": 1.7191238726302227e-05,
      "loss": 1.731,
      "num_input_tokens_seen": 30042736,
      "step": 4972,
      "train_runtime": 5039.0994,
      "train_tokens_per_second": 5961.926
    },
    {
      "epoch": 1.8290574712643677,
      "grad_norm": 0.8790012001991272,
      "learning_rate": 1.715442665194184e-05,
      "loss": 1.1759,
      "num_input_tokens_seen": 30049328,
      "step": 4973,
      "train_runtime": 5040.157,
      "train_tokens_per_second": 5961.983
    },
    {
      "epoch": 1.829425287356322,
      "grad_norm": 1.0671660900115967,
      "learning_rate": 1.711761457758145e-05,
      "loss": 1.5747,
      "num_input_tokens_seen": 30054112,
      "step": 4974,
      "train_runtime": 5040.9681,
      "train_tokens_per_second": 5961.972
    },
    {
      "epoch": 1.8297931034482757,
      "grad_norm": 1.0995584726333618,
      "learning_rate": 1.7080802503221057e-05,
      "loss": 1.6804,
      "num_input_tokens_seen": 30059504,
      "step": 4975,
      "train_runtime": 5041.8852,
      "train_tokens_per_second": 5961.957
    },
    {
      "epoch": 1.83016091954023,
      "grad_norm": 1.0045770406723022,
      "learning_rate": 1.7043990428860666e-05,
      "loss": 1.5507,
      "num_input_tokens_seen": 30065088,
      "step": 4976,
      "train_runtime": 5042.8229,
      "train_tokens_per_second": 5961.956
    },
    {
      "epoch": 1.8305287356321838,
      "grad_norm": 1.0075474977493286,
      "learning_rate": 1.700717835450028e-05,
      "loss": 1.5872,
      "num_input_tokens_seen": 30069680,
      "step": 4977,
      "train_runtime": 5043.639,
      "train_tokens_per_second": 5961.902
    },
    {
      "epoch": 1.830896551724138,
      "grad_norm": 0.8998899459838867,
      "learning_rate": 1.6970366280139887e-05,
      "loss": 1.3291,
      "num_input_tokens_seen": 30076896,
      "step": 4978,
      "train_runtime": 5044.7999,
      "train_tokens_per_second": 5961.96
    },
    {
      "epoch": 1.8312643678160918,
      "grad_norm": 0.924291729927063,
      "learning_rate": 1.6933554205779496e-05,
      "loss": 1.4414,
      "num_input_tokens_seen": 30082960,
      "step": 4979,
      "train_runtime": 5045.8099,
      "train_tokens_per_second": 5961.969
    },
    {
      "epoch": 1.831632183908046,
      "grad_norm": 1.105634093284607,
      "learning_rate": 1.6896742131419105e-05,
      "loss": 1.4625,
      "num_input_tokens_seen": 30088720,
      "step": 4980,
      "train_runtime": 5046.7596,
      "train_tokens_per_second": 5961.988
    },
    {
      "epoch": 1.8319999999999999,
      "grad_norm": 1.005671739578247,
      "learning_rate": 1.6859930057058714e-05,
      "loss": 1.4213,
      "num_input_tokens_seen": 30098528,
      "step": 4981,
      "train_runtime": 5048.746,
      "train_tokens_per_second": 5961.585
    },
    {
      "epoch": 1.832367816091954,
      "grad_norm": 1.0299687385559082,
      "learning_rate": 1.6823117982698326e-05,
      "loss": 1.4318,
      "num_input_tokens_seen": 30107072,
      "step": 4982,
      "train_runtime": 5050.0456,
      "train_tokens_per_second": 5961.743
    },
    {
      "epoch": 1.832735632183908,
      "grad_norm": 0.9860517382621765,
      "learning_rate": 1.6786305908337935e-05,
      "loss": 1.6479,
      "num_input_tokens_seen": 30112096,
      "step": 4983,
      "train_runtime": 5050.9068,
      "train_tokens_per_second": 5961.721
    },
    {
      "epoch": 1.8331034482758621,
      "grad_norm": 1.0216484069824219,
      "learning_rate": 1.6749493833977544e-05,
      "loss": 1.5473,
      "num_input_tokens_seen": 30116768,
      "step": 4984,
      "train_runtime": 5051.728,
      "train_tokens_per_second": 5961.676
    },
    {
      "epoch": 1.833471264367816,
      "grad_norm": 0.98443603515625,
      "learning_rate": 1.6712681759617156e-05,
      "loss": 1.5211,
      "num_input_tokens_seen": 30121824,
      "step": 4985,
      "train_runtime": 5052.5704,
      "train_tokens_per_second": 5961.683
    },
    {
      "epoch": 1.8338390804597702,
      "grad_norm": 0.9827259182929993,
      "learning_rate": 1.6675869685256765e-05,
      "loss": 1.4428,
      "num_input_tokens_seen": 30129152,
      "step": 4986,
      "train_runtime": 5053.7553,
      "train_tokens_per_second": 5961.735
    },
    {
      "epoch": 1.834206896551724,
      "grad_norm": 0.9716022610664368,
      "learning_rate": 1.6639057610896374e-05,
      "loss": 1.4764,
      "num_input_tokens_seen": 30135536,
      "step": 4987,
      "train_runtime": 5054.7983,
      "train_tokens_per_second": 5961.768
    },
    {
      "epoch": 1.8345747126436782,
      "grad_norm": 1.008683204650879,
      "learning_rate": 1.6602245536535983e-05,
      "loss": 1.6244,
      "num_input_tokens_seen": 30140416,
      "step": 4988,
      "train_runtime": 5055.622,
      "train_tokens_per_second": 5961.762
    },
    {
      "epoch": 1.8349425287356322,
      "grad_norm": 1.0419566631317139,
      "learning_rate": 1.6565433462175595e-05,
      "loss": 1.577,
      "num_input_tokens_seen": 30145088,
      "step": 4989,
      "train_runtime": 5056.4486,
      "train_tokens_per_second": 5961.712
    },
    {
      "epoch": 1.8353103448275863,
      "grad_norm": 0.9477977752685547,
      "learning_rate": 1.6528621387815204e-05,
      "loss": 1.3136,
      "num_input_tokens_seen": 30154208,
      "step": 4990,
      "train_runtime": 5057.8425,
      "train_tokens_per_second": 5961.872
    },
    {
      "epoch": 1.8356781609195403,
      "grad_norm": 1.0180165767669678,
      "learning_rate": 1.6491809313454813e-05,
      "loss": 1.4604,
      "num_input_tokens_seen": 30161744,
      "step": 4991,
      "train_runtime": 5059.0286,
      "train_tokens_per_second": 5961.964
    },
    {
      "epoch": 1.8360459770114943,
      "grad_norm": 0.9779078960418701,
      "learning_rate": 1.6454997239094425e-05,
      "loss": 1.3226,
      "num_input_tokens_seen": 30167488,
      "step": 4992,
      "train_runtime": 5059.9687,
      "train_tokens_per_second": 5961.991
    },
    {
      "epoch": 1.8364137931034483,
      "grad_norm": 1.0453846454620361,
      "learning_rate": 1.6418185164734034e-05,
      "loss": 1.5109,
      "num_input_tokens_seen": 30171808,
      "step": 4993,
      "train_runtime": 5060.7563,
      "train_tokens_per_second": 5961.917
    },
    {
      "epoch": 1.8367816091954023,
      "grad_norm": 0.980159342288971,
      "learning_rate": 1.6381373090373643e-05,
      "loss": 1.4922,
      "num_input_tokens_seen": 30176816,
      "step": 4994,
      "train_runtime": 5061.5947,
      "train_tokens_per_second": 5961.919
    },
    {
      "epoch": 1.8371494252873564,
      "grad_norm": 0.9506115913391113,
      "learning_rate": 1.6344561016013255e-05,
      "loss": 1.2191,
      "num_input_tokens_seen": 30184688,
      "step": 4995,
      "train_runtime": 5062.8234,
      "train_tokens_per_second": 5962.027
    },
    {
      "epoch": 1.8375172413793104,
      "grad_norm": 1.0737897157669067,
      "learning_rate": 1.630774894165286e-05,
      "loss": 1.4525,
      "num_input_tokens_seen": 30188896,
      "step": 4996,
      "train_runtime": 5063.5629,
      "train_tokens_per_second": 5961.987
    },
    {
      "epoch": 1.8378850574712644,
      "grad_norm": 1.061200499534607,
      "learning_rate": 1.6270936867292473e-05,
      "loss": 1.6259,
      "num_input_tokens_seen": 30199760,
      "step": 4997,
      "train_runtime": 5065.1752,
      "train_tokens_per_second": 5962.234
    },
    {
      "epoch": 1.8382528735632184,
      "grad_norm": 1.0682674646377563,
      "learning_rate": 1.6234124792932082e-05,
      "loss": 1.6091,
      "num_input_tokens_seen": 30204752,
      "step": 4998,
      "train_runtime": 5066.0282,
      "train_tokens_per_second": 5962.215
    },
    {
      "epoch": 1.8386206896551724,
      "grad_norm": 1.0308079719543457,
      "learning_rate": 1.619731271857169e-05,
      "loss": 1.4901,
      "num_input_tokens_seen": 30209232,
      "step": 4999,
      "train_runtime": 5066.8397,
      "train_tokens_per_second": 5962.145
    },
    {
      "epoch": 1.8389885057471265,
      "grad_norm": 1.0160410404205322,
      "learning_rate": 1.6160500644211303e-05,
      "loss": 1.5683,
      "num_input_tokens_seen": 30214560,
      "step": 5000,
      "train_runtime": 5067.7383,
      "train_tokens_per_second": 5962.139
    },
    {
      "epoch": 1.8393563218390805,
      "grad_norm": 1.0536404848098755,
      "learning_rate": 1.6123688569850912e-05,
      "loss": 1.2674,
      "num_input_tokens_seen": 30222432,
      "step": 5001,
      "train_runtime": 5068.9546,
      "train_tokens_per_second": 5962.261
    },
    {
      "epoch": 1.8397241379310345,
      "grad_norm": 1.0723559856414795,
      "learning_rate": 1.608687649549052e-05,
      "loss": 1.6385,
      "num_input_tokens_seen": 30227088,
      "step": 5002,
      "train_runtime": 5069.7547,
      "train_tokens_per_second": 5962.239
    },
    {
      "epoch": 1.8400919540229885,
      "grad_norm": 1.0325651168823242,
      "learning_rate": 1.605006442113013e-05,
      "loss": 1.4342,
      "num_input_tokens_seen": 30232656,
      "step": 5003,
      "train_runtime": 5070.6568,
      "train_tokens_per_second": 5962.276
    },
    {
      "epoch": 1.8404597701149426,
      "grad_norm": 1.0740904808044434,
      "learning_rate": 1.6013252346769742e-05,
      "loss": 1.4162,
      "num_input_tokens_seen": 30237584,
      "step": 5004,
      "train_runtime": 5071.4831,
      "train_tokens_per_second": 5962.276
    },
    {
      "epoch": 1.8408275862068966,
      "grad_norm": 0.9917394518852234,
      "learning_rate": 1.597644027240935e-05,
      "loss": 1.4873,
      "num_input_tokens_seen": 30243040,
      "step": 5005,
      "train_runtime": 5072.3909,
      "train_tokens_per_second": 5962.285
    },
    {
      "epoch": 1.8411954022988506,
      "grad_norm": 1.0606144666671753,
      "learning_rate": 1.593962819804896e-05,
      "loss": 1.486,
      "num_input_tokens_seen": 30249472,
      "step": 5006,
      "train_runtime": 5073.4178,
      "train_tokens_per_second": 5962.346
    },
    {
      "epoch": 1.8415632183908046,
      "grad_norm": 1.0541677474975586,
      "learning_rate": 1.5902816123688572e-05,
      "loss": 1.5326,
      "num_input_tokens_seen": 30255744,
      "step": 5007,
      "train_runtime": 5074.4427,
      "train_tokens_per_second": 5962.378
    },
    {
      "epoch": 1.8419310344827586,
      "grad_norm": 1.1358091831207275,
      "learning_rate": 1.586600404932818e-05,
      "loss": 1.7369,
      "num_input_tokens_seen": 30260496,
      "step": 5008,
      "train_runtime": 5075.2593,
      "train_tokens_per_second": 5962.355
    },
    {
      "epoch": 1.8422988505747127,
      "grad_norm": 1.0278191566467285,
      "learning_rate": 1.582919197496779e-05,
      "loss": 1.4993,
      "num_input_tokens_seen": 30265968,
      "step": 5009,
      "train_runtime": 5076.1591,
      "train_tokens_per_second": 5962.376
    },
    {
      "epoch": 1.8426666666666667,
      "grad_norm": 0.9608697295188904,
      "learning_rate": 1.5792379900607402e-05,
      "loss": 1.4283,
      "num_input_tokens_seen": 30274544,
      "step": 5010,
      "train_runtime": 5077.5133,
      "train_tokens_per_second": 5962.475
    },
    {
      "epoch": 1.8430344827586207,
      "grad_norm": 0.9913397431373596,
      "learning_rate": 1.5755567826247008e-05,
      "loss": 1.753,
      "num_input_tokens_seen": 30279616,
      "step": 5011,
      "train_runtime": 5078.8607,
      "train_tokens_per_second": 5961.891
    },
    {
      "epoch": 1.8434022988505747,
      "grad_norm": 0.9961228966712952,
      "learning_rate": 1.571875575188662e-05,
      "loss": 1.3563,
      "num_input_tokens_seen": 30284304,
      "step": 5012,
      "train_runtime": 5079.6395,
      "train_tokens_per_second": 5961.9
    },
    {
      "epoch": 1.8437701149425287,
      "grad_norm": 1.109960913658142,
      "learning_rate": 1.568194367752623e-05,
      "loss": 1.5663,
      "num_input_tokens_seen": 30292736,
      "step": 5013,
      "train_runtime": 5080.9552,
      "train_tokens_per_second": 5962.016
    },
    {
      "epoch": 1.8441379310344828,
      "grad_norm": 0.9731327891349792,
      "learning_rate": 1.5645131603165838e-05,
      "loss": 1.5068,
      "num_input_tokens_seen": 30298528,
      "step": 5014,
      "train_runtime": 5081.8938,
      "train_tokens_per_second": 5962.055
    },
    {
      "epoch": 1.8445057471264368,
      "grad_norm": 0.9294212460517883,
      "learning_rate": 1.5608319528805447e-05,
      "loss": 1.2195,
      "num_input_tokens_seen": 30307088,
      "step": 5015,
      "train_runtime": 5083.1898,
      "train_tokens_per_second": 5962.218
    },
    {
      "epoch": 1.8448735632183908,
      "grad_norm": 0.99272620677948,
      "learning_rate": 1.557150745444506e-05,
      "loss": 1.5177,
      "num_input_tokens_seen": 30312224,
      "step": 5016,
      "train_runtime": 5084.0713,
      "train_tokens_per_second": 5962.195
    },
    {
      "epoch": 1.8452413793103448,
      "grad_norm": 0.9845655560493469,
      "learning_rate": 1.5534695380084668e-05,
      "loss": 1.4629,
      "num_input_tokens_seen": 30321664,
      "step": 5017,
      "train_runtime": 5085.525,
      "train_tokens_per_second": 5962.347
    },
    {
      "epoch": 1.8456091954022988,
      "grad_norm": 1.1305726766586304,
      "learning_rate": 1.5497883305724277e-05,
      "loss": 1.715,
      "num_input_tokens_seen": 30327088,
      "step": 5018,
      "train_runtime": 5086.4389,
      "train_tokens_per_second": 5962.342
    },
    {
      "epoch": 1.8459770114942529,
      "grad_norm": 1.0919382572174072,
      "learning_rate": 1.546107123136389e-05,
      "loss": 1.8685,
      "num_input_tokens_seen": 30332208,
      "step": 5019,
      "train_runtime": 5087.3037,
      "train_tokens_per_second": 5962.335
    },
    {
      "epoch": 1.8463448275862069,
      "grad_norm": 1.0870243310928345,
      "learning_rate": 1.5424259157003498e-05,
      "loss": 1.6012,
      "num_input_tokens_seen": 30336976,
      "step": 5020,
      "train_runtime": 5088.1072,
      "train_tokens_per_second": 5962.33
    },
    {
      "epoch": 1.846712643678161,
      "grad_norm": 1.0127289295196533,
      "learning_rate": 1.5387447082643107e-05,
      "loss": 1.7094,
      "num_input_tokens_seen": 30342320,
      "step": 5021,
      "train_runtime": 5088.988,
      "train_tokens_per_second": 5962.348
    },
    {
      "epoch": 1.847080459770115,
      "grad_norm": 1.0184253454208374,
      "learning_rate": 1.535063500828272e-05,
      "loss": 1.4599,
      "num_input_tokens_seen": 30347568,
      "step": 5022,
      "train_runtime": 5089.8803,
      "train_tokens_per_second": 5962.334
    },
    {
      "epoch": 1.847448275862069,
      "grad_norm": 0.9991694688796997,
      "learning_rate": 1.5313822933922328e-05,
      "loss": 1.527,
      "num_input_tokens_seen": 30352848,
      "step": 5023,
      "train_runtime": 5090.7557,
      "train_tokens_per_second": 5962.346
    },
    {
      "epoch": 1.847816091954023,
      "grad_norm": 0.9224336743354797,
      "learning_rate": 1.5277010859561937e-05,
      "loss": 1.4338,
      "num_input_tokens_seen": 30359504,
      "step": 5024,
      "train_runtime": 5091.8406,
      "train_tokens_per_second": 5962.383
    },
    {
      "epoch": 1.848183908045977,
      "grad_norm": 1.082759976387024,
      "learning_rate": 1.5240198785201548e-05,
      "loss": 1.7668,
      "num_input_tokens_seen": 30364144,
      "step": 5025,
      "train_runtime": 5092.6448,
      "train_tokens_per_second": 5962.353
    },
    {
      "epoch": 1.848551724137931,
      "grad_norm": 1.0988906621932983,
      "learning_rate": 1.5203386710841155e-05,
      "loss": 1.463,
      "num_input_tokens_seen": 30369920,
      "step": 5026,
      "train_runtime": 5093.6022,
      "train_tokens_per_second": 5962.366
    },
    {
      "epoch": 1.848919540229885,
      "grad_norm": 0.898306131362915,
      "learning_rate": 1.5166574636480765e-05,
      "loss": 1.5303,
      "num_input_tokens_seen": 30377904,
      "step": 5027,
      "train_runtime": 5094.8497,
      "train_tokens_per_second": 5962.473
    },
    {
      "epoch": 1.849287356321839,
      "grad_norm": 0.9081872701644897,
      "learning_rate": 1.5129762562120376e-05,
      "loss": 1.3248,
      "num_input_tokens_seen": 30384624,
      "step": 5028,
      "train_runtime": 5095.9105,
      "train_tokens_per_second": 5962.551
    },
    {
      "epoch": 1.849655172413793,
      "grad_norm": 1.063267707824707,
      "learning_rate": 1.5092950487759985e-05,
      "loss": 1.4219,
      "num_input_tokens_seen": 30392544,
      "step": 5029,
      "train_runtime": 5097.1409,
      "train_tokens_per_second": 5962.665
    },
    {
      "epoch": 1.850022988505747,
      "grad_norm": 0.9911524653434753,
      "learning_rate": 1.5056138413399595e-05,
      "loss": 1.4586,
      "num_input_tokens_seen": 30399056,
      "step": 5030,
      "train_runtime": 5098.1995,
      "train_tokens_per_second": 5962.704
    },
    {
      "epoch": 1.850390804597701,
      "grad_norm": 0.95414799451828,
      "learning_rate": 1.5019326339039206e-05,
      "loss": 1.3232,
      "num_input_tokens_seen": 30405504,
      "step": 5031,
      "train_runtime": 5099.2523,
      "train_tokens_per_second": 5962.738
    },
    {
      "epoch": 1.8507586206896551,
      "grad_norm": 0.9188612103462219,
      "learning_rate": 1.4982514264678815e-05,
      "loss": 1.6327,
      "num_input_tokens_seen": 30411536,
      "step": 5032,
      "train_runtime": 5100.2201,
      "train_tokens_per_second": 5962.789
    },
    {
      "epoch": 1.8511264367816092,
      "grad_norm": 0.9932575821876526,
      "learning_rate": 1.4945702190318426e-05,
      "loss": 1.4435,
      "num_input_tokens_seen": 30417168,
      "step": 5033,
      "train_runtime": 5101.1551,
      "train_tokens_per_second": 5962.8
    },
    {
      "epoch": 1.8514942528735632,
      "grad_norm": 0.9736252427101135,
      "learning_rate": 1.4908890115958034e-05,
      "loss": 1.4467,
      "num_input_tokens_seen": 30424160,
      "step": 5034,
      "train_runtime": 5102.2557,
      "train_tokens_per_second": 5962.884
    },
    {
      "epoch": 1.8518620689655172,
      "grad_norm": 1.0463744401931763,
      "learning_rate": 1.4872078041597645e-05,
      "loss": 1.4632,
      "num_input_tokens_seen": 30429904,
      "step": 5035,
      "train_runtime": 5103.166,
      "train_tokens_per_second": 5962.946
    },
    {
      "epoch": 1.8522298850574712,
      "grad_norm": 1.0465551614761353,
      "learning_rate": 1.4835265967237256e-05,
      "loss": 1.5998,
      "num_input_tokens_seen": 30434656,
      "step": 5036,
      "train_runtime": 5104.0026,
      "train_tokens_per_second": 5962.9
    },
    {
      "epoch": 1.8525977011494252,
      "grad_norm": 1.0173680782318115,
      "learning_rate": 1.4798453892876864e-05,
      "loss": 1.5473,
      "num_input_tokens_seen": 30441760,
      "step": 5037,
      "train_runtime": 5105.1223,
      "train_tokens_per_second": 5962.983
    },
    {
      "epoch": 1.8529655172413793,
      "grad_norm": 0.9877189993858337,
      "learning_rate": 1.4761641818516475e-05,
      "loss": 1.2774,
      "num_input_tokens_seen": 30449424,
      "step": 5038,
      "train_runtime": 5106.3205,
      "train_tokens_per_second": 5963.085
    },
    {
      "epoch": 1.8533333333333335,
      "grad_norm": 1.026616096496582,
      "learning_rate": 1.4724829744156086e-05,
      "loss": 1.6398,
      "num_input_tokens_seen": 30455664,
      "step": 5039,
      "train_runtime": 5107.3267,
      "train_tokens_per_second": 5963.132
    },
    {
      "epoch": 1.8537011494252873,
      "grad_norm": 1.1925349235534668,
      "learning_rate": 1.4688017669795695e-05,
      "loss": 1.7049,
      "num_input_tokens_seen": 30460944,
      "step": 5040,
      "train_runtime": 5108.1895,
      "train_tokens_per_second": 5963.159
    },
    {
      "epoch": 1.8540689655172415,
      "grad_norm": 1.0198228359222412,
      "learning_rate": 1.4651205595435302e-05,
      "loss": 1.656,
      "num_input_tokens_seen": 30466032,
      "step": 5041,
      "train_runtime": 5109.6124,
      "train_tokens_per_second": 5962.494
    },
    {
      "epoch": 1.8544367816091953,
      "grad_norm": 1.084392786026001,
      "learning_rate": 1.4614393521074912e-05,
      "loss": 1.6761,
      "num_input_tokens_seen": 30470192,
      "step": 5042,
      "train_runtime": 5110.3422,
      "train_tokens_per_second": 5962.456
    },
    {
      "epoch": 1.8548045977011496,
      "grad_norm": 1.002570390701294,
      "learning_rate": 1.4577581446714523e-05,
      "loss": 1.5013,
      "num_input_tokens_seen": 30476192,
      "step": 5043,
      "train_runtime": 5111.3153,
      "train_tokens_per_second": 5962.495
    },
    {
      "epoch": 1.8551724137931034,
      "grad_norm": 1.031084418296814,
      "learning_rate": 1.4540769372354132e-05,
      "loss": 1.5598,
      "num_input_tokens_seen": 30483056,
      "step": 5044,
      "train_runtime": 5112.397,
      "train_tokens_per_second": 5962.576
    },
    {
      "epoch": 1.8555402298850576,
      "grad_norm": 0.9973475933074951,
      "learning_rate": 1.4503957297993742e-05,
      "loss": 1.617,
      "num_input_tokens_seen": 30487616,
      "step": 5045,
      "train_runtime": 5113.1713,
      "train_tokens_per_second": 5962.565
    },
    {
      "epoch": 1.8559080459770114,
      "grad_norm": 0.9751969575881958,
      "learning_rate": 1.4467145223633351e-05,
      "loss": 1.5653,
      "num_input_tokens_seen": 30492896,
      "step": 5046,
      "train_runtime": 5114.0419,
      "train_tokens_per_second": 5962.582
    },
    {
      "epoch": 1.8562758620689657,
      "grad_norm": 1.0036165714263916,
      "learning_rate": 1.4430333149272962e-05,
      "loss": 1.7789,
      "num_input_tokens_seen": 30497696,
      "step": 5047,
      "train_runtime": 5114.8702,
      "train_tokens_per_second": 5962.555
    },
    {
      "epoch": 1.8566436781609195,
      "grad_norm": 1.0237960815429688,
      "learning_rate": 1.4393521074912572e-05,
      "loss": 1.5519,
      "num_input_tokens_seen": 30506880,
      "step": 5048,
      "train_runtime": 5116.2788,
      "train_tokens_per_second": 5962.709
    },
    {
      "epoch": 1.8570114942528737,
      "grad_norm": 0.950262188911438,
      "learning_rate": 1.4356709000552181e-05,
      "loss": 1.417,
      "num_input_tokens_seen": 30515168,
      "step": 5049,
      "train_runtime": 5117.5645,
      "train_tokens_per_second": 5962.83
    },
    {
      "epoch": 1.8573793103448275,
      "grad_norm": 1.030007243156433,
      "learning_rate": 1.4319896926191792e-05,
      "loss": 1.4578,
      "num_input_tokens_seen": 30521376,
      "step": 5050,
      "train_runtime": 5118.5796,
      "train_tokens_per_second": 5962.86
    },
    {
      "epoch": 1.8577471264367817,
      "grad_norm": 0.986011266708374,
      "learning_rate": 1.4283084851831403e-05,
      "loss": 1.4901,
      "num_input_tokens_seen": 30526560,
      "step": 5051,
      "train_runtime": 5119.4287,
      "train_tokens_per_second": 5962.884
    },
    {
      "epoch": 1.8581149425287355,
      "grad_norm": 0.9912033081054688,
      "learning_rate": 1.4246272777471011e-05,
      "loss": 1.5228,
      "num_input_tokens_seen": 30531920,
      "step": 5052,
      "train_runtime": 5120.3288,
      "train_tokens_per_second": 5962.883
    },
    {
      "epoch": 1.8584827586206898,
      "grad_norm": 1.0162124633789062,
      "learning_rate": 1.4209460703110622e-05,
      "loss": 1.5007,
      "num_input_tokens_seen": 30540928,
      "step": 5053,
      "train_runtime": 5121.7022,
      "train_tokens_per_second": 5963.042
    },
    {
      "epoch": 1.8588505747126436,
      "grad_norm": 1.0581971406936646,
      "learning_rate": 1.4172648628750233e-05,
      "loss": 1.4919,
      "num_input_tokens_seen": 30546448,
      "step": 5054,
      "train_runtime": 5122.6102,
      "train_tokens_per_second": 5963.063
    },
    {
      "epoch": 1.8592183908045978,
      "grad_norm": 1.0529959201812744,
      "learning_rate": 1.4135836554389841e-05,
      "loss": 1.5747,
      "num_input_tokens_seen": 30551360,
      "step": 5055,
      "train_runtime": 5123.4277,
      "train_tokens_per_second": 5963.07
    },
    {
      "epoch": 1.8595862068965516,
      "grad_norm": 1.065110445022583,
      "learning_rate": 1.4099024480029449e-05,
      "loss": 1.6546,
      "num_input_tokens_seen": 30557280,
      "step": 5056,
      "train_runtime": 5124.3956,
      "train_tokens_per_second": 5963.099
    },
    {
      "epoch": 1.8599540229885059,
      "grad_norm": 0.9212146997451782,
      "learning_rate": 1.406221240566906e-05,
      "loss": 1.2735,
      "num_input_tokens_seen": 30563264,
      "step": 5057,
      "train_runtime": 5125.3742,
      "train_tokens_per_second": 5963.128
    },
    {
      "epoch": 1.8603218390804597,
      "grad_norm": 0.9818912148475647,
      "learning_rate": 1.402540033130867e-05,
      "loss": 1.4226,
      "num_input_tokens_seen": 30568752,
      "step": 5058,
      "train_runtime": 5126.2659,
      "train_tokens_per_second": 5963.162
    },
    {
      "epoch": 1.860689655172414,
      "grad_norm": 1.0723986625671387,
      "learning_rate": 1.3988588256948279e-05,
      "loss": 1.4278,
      "num_input_tokens_seen": 30574272,
      "step": 5059,
      "train_runtime": 5127.1939,
      "train_tokens_per_second": 5963.159
    },
    {
      "epoch": 1.8610574712643677,
      "grad_norm": 0.98280268907547,
      "learning_rate": 1.395177618258789e-05,
      "loss": 1.2585,
      "num_input_tokens_seen": 30580224,
      "step": 5060,
      "train_runtime": 5128.1961,
      "train_tokens_per_second": 5963.154
    },
    {
      "epoch": 1.861425287356322,
      "grad_norm": 0.9946303367614746,
      "learning_rate": 1.3914964108227498e-05,
      "loss": 1.6314,
      "num_input_tokens_seen": 30585248,
      "step": 5061,
      "train_runtime": 5129.0636,
      "train_tokens_per_second": 5963.125
    },
    {
      "epoch": 1.8617931034482758,
      "grad_norm": 1.0140941143035889,
      "learning_rate": 1.3878152033867109e-05,
      "loss": 1.568,
      "num_input_tokens_seen": 30589872,
      "step": 5062,
      "train_runtime": 5129.8829,
      "train_tokens_per_second": 5963.074
    },
    {
      "epoch": 1.86216091954023,
      "grad_norm": 0.9778771996498108,
      "learning_rate": 1.384133995950672e-05,
      "loss": 1.4906,
      "num_input_tokens_seen": 30594432,
      "step": 5063,
      "train_runtime": 5130.6869,
      "train_tokens_per_second": 5963.028
    },
    {
      "epoch": 1.8625287356321838,
      "grad_norm": 0.9767743945121765,
      "learning_rate": 1.3804527885146328e-05,
      "loss": 1.4655,
      "num_input_tokens_seen": 30600496,
      "step": 5064,
      "train_runtime": 5131.6727,
      "train_tokens_per_second": 5963.065
    },
    {
      "epoch": 1.862896551724138,
      "grad_norm": 0.9685781598091125,
      "learning_rate": 1.3767715810785939e-05,
      "loss": 1.4346,
      "num_input_tokens_seen": 30605856,
      "step": 5065,
      "train_runtime": 5132.5409,
      "train_tokens_per_second": 5963.1
    },
    {
      "epoch": 1.8632643678160918,
      "grad_norm": 1.0256255865097046,
      "learning_rate": 1.373090373642555e-05,
      "loss": 1.4224,
      "num_input_tokens_seen": 30611072,
      "step": 5066,
      "train_runtime": 5133.395,
      "train_tokens_per_second": 5963.124
    },
    {
      "epoch": 1.863632183908046,
      "grad_norm": 0.9459775686264038,
      "learning_rate": 1.3694091662065158e-05,
      "loss": 1.3028,
      "num_input_tokens_seen": 30619168,
      "step": 5067,
      "train_runtime": 5134.6519,
      "train_tokens_per_second": 5963.241
    },
    {
      "epoch": 1.8639999999999999,
      "grad_norm": 1.0559552907943726,
      "learning_rate": 1.3657279587704769e-05,
      "loss": 1.6466,
      "num_input_tokens_seen": 30624000,
      "step": 5068,
      "train_runtime": 5135.4798,
      "train_tokens_per_second": 5963.221
    },
    {
      "epoch": 1.8643678160919541,
      "grad_norm": 1.0232208967208862,
      "learning_rate": 1.362046751334438e-05,
      "loss": 1.5455,
      "num_input_tokens_seen": 30628736,
      "step": 5069,
      "train_runtime": 5136.2864,
      "train_tokens_per_second": 5963.206
    },
    {
      "epoch": 1.864735632183908,
      "grad_norm": 1.013243317604065,
      "learning_rate": 1.3583655438983988e-05,
      "loss": 1.6172,
      "num_input_tokens_seen": 30633520,
      "step": 5070,
      "train_runtime": 5137.1024,
      "train_tokens_per_second": 5963.19
    },
    {
      "epoch": 1.8651034482758622,
      "grad_norm": 1.057968020439148,
      "learning_rate": 1.3546843364623596e-05,
      "loss": 1.5438,
      "num_input_tokens_seen": 30639328,
      "step": 5071,
      "train_runtime": 5138.6041,
      "train_tokens_per_second": 5962.578
    },
    {
      "epoch": 1.865471264367816,
      "grad_norm": 0.9944549202919006,
      "learning_rate": 1.3510031290263206e-05,
      "loss": 1.3446,
      "num_input_tokens_seen": 30647536,
      "step": 5072,
      "train_runtime": 5139.8809,
      "train_tokens_per_second": 5962.694
    },
    {
      "epoch": 1.8658390804597702,
      "grad_norm": 1.0087077617645264,
      "learning_rate": 1.3473219215902815e-05,
      "loss": 1.5324,
      "num_input_tokens_seen": 30652400,
      "step": 5073,
      "train_runtime": 5140.7054,
      "train_tokens_per_second": 5962.684
    },
    {
      "epoch": 1.866206896551724,
      "grad_norm": 0.9828615784645081,
      "learning_rate": 1.3436407141542426e-05,
      "loss": 1.5487,
      "num_input_tokens_seen": 30657328,
      "step": 5074,
      "train_runtime": 5141.5239,
      "train_tokens_per_second": 5962.693
    },
    {
      "epoch": 1.8665747126436782,
      "grad_norm": 1.0491549968719482,
      "learning_rate": 1.3399595067182036e-05,
      "loss": 1.6233,
      "num_input_tokens_seen": 30662384,
      "step": 5075,
      "train_runtime": 5142.3783,
      "train_tokens_per_second": 5962.685
    },
    {
      "epoch": 1.866942528735632,
      "grad_norm": 1.046576976776123,
      "learning_rate": 1.3362782992821645e-05,
      "loss": 1.5585,
      "num_input_tokens_seen": 30667424,
      "step": 5076,
      "train_runtime": 5143.2315,
      "train_tokens_per_second": 5962.676
    },
    {
      "epoch": 1.8673103448275863,
      "grad_norm": 0.9875950813293457,
      "learning_rate": 1.3325970918461256e-05,
      "loss": 1.3376,
      "num_input_tokens_seen": 30672544,
      "step": 5077,
      "train_runtime": 5144.1033,
      "train_tokens_per_second": 5962.661
    },
    {
      "epoch": 1.86767816091954,
      "grad_norm": 1.1327226161956787,
      "learning_rate": 1.3289158844100866e-05,
      "loss": 1.5743,
      "num_input_tokens_seen": 30678240,
      "step": 5078,
      "train_runtime": 5145.0381,
      "train_tokens_per_second": 5962.685
    },
    {
      "epoch": 1.8680459770114943,
      "grad_norm": 1.0779470205307007,
      "learning_rate": 1.3252346769740475e-05,
      "loss": 1.6379,
      "num_input_tokens_seen": 30683408,
      "step": 5079,
      "train_runtime": 5145.8942,
      "train_tokens_per_second": 5962.697
    },
    {
      "epoch": 1.8684137931034481,
      "grad_norm": 0.964714765548706,
      "learning_rate": 1.3215534695380086e-05,
      "loss": 1.3875,
      "num_input_tokens_seen": 30688256,
      "step": 5080,
      "train_runtime": 5146.728,
      "train_tokens_per_second": 5962.673
    },
    {
      "epoch": 1.8687816091954024,
      "grad_norm": 1.1236010789871216,
      "learning_rate": 1.3178722621019696e-05,
      "loss": 1.6559,
      "num_input_tokens_seen": 30693008,
      "step": 5081,
      "train_runtime": 5147.5356,
      "train_tokens_per_second": 5962.661
    },
    {
      "epoch": 1.8691494252873562,
      "grad_norm": 1.012908935546875,
      "learning_rate": 1.3141910546659305e-05,
      "loss": 1.6738,
      "num_input_tokens_seen": 30699792,
      "step": 5082,
      "train_runtime": 5148.6036,
      "train_tokens_per_second": 5962.741
    },
    {
      "epoch": 1.8695172413793104,
      "grad_norm": 1.034468650817871,
      "learning_rate": 1.3105098472298916e-05,
      "loss": 1.5693,
      "num_input_tokens_seen": 30704672,
      "step": 5083,
      "train_runtime": 5149.42,
      "train_tokens_per_second": 5962.744
    },
    {
      "epoch": 1.8698850574712642,
      "grad_norm": 1.038994312286377,
      "learning_rate": 1.3068286397938525e-05,
      "loss": 1.7106,
      "num_input_tokens_seen": 30709232,
      "step": 5084,
      "train_runtime": 5150.1949,
      "train_tokens_per_second": 5962.732
    },
    {
      "epoch": 1.8702528735632185,
      "grad_norm": 1.0251950025558472,
      "learning_rate": 1.3031474323578135e-05,
      "loss": 1.5251,
      "num_input_tokens_seen": 30714336,
      "step": 5085,
      "train_runtime": 5151.0501,
      "train_tokens_per_second": 5962.733
    },
    {
      "epoch": 1.8706206896551723,
      "grad_norm": 1.024907112121582,
      "learning_rate": 1.2994662249217743e-05,
      "loss": 1.4149,
      "num_input_tokens_seen": 30721744,
      "step": 5086,
      "train_runtime": 5152.236,
      "train_tokens_per_second": 5962.798
    },
    {
      "epoch": 1.8709885057471265,
      "grad_norm": 1.0257986783981323,
      "learning_rate": 1.2957850174857353e-05,
      "loss": 1.3384,
      "num_input_tokens_seen": 30728912,
      "step": 5087,
      "train_runtime": 5153.402,
      "train_tokens_per_second": 5962.84
    },
    {
      "epoch": 1.8713563218390803,
      "grad_norm": 0.9821317195892334,
      "learning_rate": 1.2921038100496962e-05,
      "loss": 1.3058,
      "num_input_tokens_seen": 30735760,
      "step": 5088,
      "train_runtime": 5154.5118,
      "train_tokens_per_second": 5962.885
    },
    {
      "epoch": 1.8717241379310345,
      "grad_norm": 1.106223225593567,
      "learning_rate": 1.2884226026136573e-05,
      "loss": 1.6508,
      "num_input_tokens_seen": 30742496,
      "step": 5089,
      "train_runtime": 5155.5644,
      "train_tokens_per_second": 5962.974
    },
    {
      "epoch": 1.8720919540229886,
      "grad_norm": 1.1005597114562988,
      "learning_rate": 1.2847413951776183e-05,
      "loss": 1.3574,
      "num_input_tokens_seen": 30749168,
      "step": 5090,
      "train_runtime": 5156.644,
      "train_tokens_per_second": 5963.019
    },
    {
      "epoch": 1.8724597701149426,
      "grad_norm": 1.0167715549468994,
      "learning_rate": 1.2810601877415792e-05,
      "loss": 1.3957,
      "num_input_tokens_seen": 30756192,
      "step": 5091,
      "train_runtime": 5157.7636,
      "train_tokens_per_second": 5963.087
    },
    {
      "epoch": 1.8728275862068966,
      "grad_norm": 1.0534465312957764,
      "learning_rate": 1.2773789803055403e-05,
      "loss": 1.6777,
      "num_input_tokens_seen": 30762048,
      "step": 5092,
      "train_runtime": 5158.7233,
      "train_tokens_per_second": 5963.113
    },
    {
      "epoch": 1.8731954022988506,
      "grad_norm": 1.136102557182312,
      "learning_rate": 1.2736977728695013e-05,
      "loss": 1.724,
      "num_input_tokens_seen": 30766560,
      "step": 5093,
      "train_runtime": 5159.5324,
      "train_tokens_per_second": 5963.052
    },
    {
      "epoch": 1.8735632183908046,
      "grad_norm": 0.9724223017692566,
      "learning_rate": 1.2700165654334622e-05,
      "loss": 1.3868,
      "num_input_tokens_seen": 30771536,
      "step": 5094,
      "train_runtime": 5160.3963,
      "train_tokens_per_second": 5963.018
    },
    {
      "epoch": 1.8739310344827587,
      "grad_norm": 1.020814299583435,
      "learning_rate": 1.2663353579974233e-05,
      "loss": 1.5598,
      "num_input_tokens_seen": 30777840,
      "step": 5095,
      "train_runtime": 5161.3988,
      "train_tokens_per_second": 5963.081
    },
    {
      "epoch": 1.8742988505747127,
      "grad_norm": 1.0722523927688599,
      "learning_rate": 1.2626541505613843e-05,
      "loss": 1.5814,
      "num_input_tokens_seen": 30783184,
      "step": 5096,
      "train_runtime": 5162.2725,
      "train_tokens_per_second": 5963.107
    },
    {
      "epoch": 1.8746666666666667,
      "grad_norm": 1.038047432899475,
      "learning_rate": 1.2589729431253452e-05,
      "loss": 1.4757,
      "num_input_tokens_seen": 30788912,
      "step": 5097,
      "train_runtime": 5163.2417,
      "train_tokens_per_second": 5963.097
    },
    {
      "epoch": 1.8750344827586207,
      "grad_norm": 1.0603339672088623,
      "learning_rate": 1.2552917356893063e-05,
      "loss": 1.5563,
      "num_input_tokens_seen": 30794864,
      "step": 5098,
      "train_runtime": 5164.2116,
      "train_tokens_per_second": 5963.13
    },
    {
      "epoch": 1.8754022988505747,
      "grad_norm": 0.9655351042747498,
      "learning_rate": 1.2516105282532672e-05,
      "loss": 1.4097,
      "num_input_tokens_seen": 30802736,
      "step": 5099,
      "train_runtime": 5165.4285,
      "train_tokens_per_second": 5963.249
    },
    {
      "epoch": 1.8757701149425288,
      "grad_norm": 0.9132452011108398,
      "learning_rate": 1.247929320817228e-05,
      "loss": 1.3708,
      "num_input_tokens_seen": 30811360,
      "step": 5100,
      "train_runtime": 5166.7524,
      "train_tokens_per_second": 5963.39
    },
    {
      "epoch": 1.8761379310344828,
      "grad_norm": 1.0230684280395508,
      "learning_rate": 1.2442481133811891e-05,
      "loss": 1.4089,
      "num_input_tokens_seen": 30818352,
      "step": 5101,
      "train_runtime": 5168.4608,
      "train_tokens_per_second": 5962.772
    },
    {
      "epoch": 1.8765057471264368,
      "grad_norm": 1.0087424516677856,
      "learning_rate": 1.2405669059451502e-05,
      "loss": 1.3587,
      "num_input_tokens_seen": 30824672,
      "step": 5102,
      "train_runtime": 5169.4818,
      "train_tokens_per_second": 5962.817
    },
    {
      "epoch": 1.8768735632183908,
      "grad_norm": 1.0149543285369873,
      "learning_rate": 1.236885698509111e-05,
      "loss": 1.5482,
      "num_input_tokens_seen": 30829328,
      "step": 5103,
      "train_runtime": 5170.3068,
      "train_tokens_per_second": 5962.766
    },
    {
      "epoch": 1.8772413793103448,
      "grad_norm": 0.9287912845611572,
      "learning_rate": 1.2332044910730721e-05,
      "loss": 1.3568,
      "num_input_tokens_seen": 30838224,
      "step": 5104,
      "train_runtime": 5171.6861,
      "train_tokens_per_second": 5962.896
    },
    {
      "epoch": 1.8776091954022989,
      "grad_norm": 1.0355212688446045,
      "learning_rate": 1.229523283637033e-05,
      "loss": 1.3621,
      "num_input_tokens_seen": 30845760,
      "step": 5105,
      "train_runtime": 5172.853,
      "train_tokens_per_second": 5963.007
    },
    {
      "epoch": 1.877977011494253,
      "grad_norm": 1.085738182067871,
      "learning_rate": 1.2258420762009939e-05,
      "loss": 1.7144,
      "num_input_tokens_seen": 30851264,
      "step": 5106,
      "train_runtime": 5173.7704,
      "train_tokens_per_second": 5963.014
    },
    {
      "epoch": 1.878344827586207,
      "grad_norm": 1.158015489578247,
      "learning_rate": 1.222160868764955e-05,
      "loss": 1.4586,
      "num_input_tokens_seen": 30858160,
      "step": 5107,
      "train_runtime": 5174.8826,
      "train_tokens_per_second": 5963.065
    },
    {
      "epoch": 1.878712643678161,
      "grad_norm": 1.06531822681427,
      "learning_rate": 1.218479661328916e-05,
      "loss": 1.5691,
      "num_input_tokens_seen": 30863520,
      "step": 5108,
      "train_runtime": 5175.7685,
      "train_tokens_per_second": 5963.08
    },
    {
      "epoch": 1.879080459770115,
      "grad_norm": 0.9724718332290649,
      "learning_rate": 1.2147984538928769e-05,
      "loss": 1.4844,
      "num_input_tokens_seen": 30870320,
      "step": 5109,
      "train_runtime": 5176.8622,
      "train_tokens_per_second": 5963.133
    },
    {
      "epoch": 1.879448275862069,
      "grad_norm": 1.0280042886734009,
      "learning_rate": 1.211117246456838e-05,
      "loss": 1.5261,
      "num_input_tokens_seen": 30875648,
      "step": 5110,
      "train_runtime": 5177.756,
      "train_tokens_per_second": 5963.133
    },
    {
      "epoch": 1.879816091954023,
      "grad_norm": 1.0828311443328857,
      "learning_rate": 1.2074360390207989e-05,
      "loss": 1.3906,
      "num_input_tokens_seen": 30880576,
      "step": 5111,
      "train_runtime": 5178.5765,
      "train_tokens_per_second": 5963.14
    },
    {
      "epoch": 1.880183908045977,
      "grad_norm": 1.0483033657073975,
      "learning_rate": 1.2037548315847597e-05,
      "loss": 1.7721,
      "num_input_tokens_seen": 30884864,
      "step": 5112,
      "train_runtime": 5179.3431,
      "train_tokens_per_second": 5963.085
    },
    {
      "epoch": 1.880551724137931,
      "grad_norm": 0.9692808389663696,
      "learning_rate": 1.2000736241487208e-05,
      "loss": 1.4897,
      "num_input_tokens_seen": 30890224,
      "step": 5113,
      "train_runtime": 5180.2285,
      "train_tokens_per_second": 5963.101
    },
    {
      "epoch": 1.880919540229885,
      "grad_norm": 1.1231056451797485,
      "learning_rate": 1.1963924167126819e-05,
      "loss": 1.7476,
      "num_input_tokens_seen": 30894416,
      "step": 5114,
      "train_runtime": 5180.9853,
      "train_tokens_per_second": 5963.039
    },
    {
      "epoch": 1.881287356321839,
      "grad_norm": 0.9407712817192078,
      "learning_rate": 1.1927112092766428e-05,
      "loss": 1.4063,
      "num_input_tokens_seen": 30902528,
      "step": 5115,
      "train_runtime": 5182.2622,
      "train_tokens_per_second": 5963.135
    },
    {
      "epoch": 1.881655172413793,
      "grad_norm": 1.0812958478927612,
      "learning_rate": 1.1890300018406038e-05,
      "loss": 1.5708,
      "num_input_tokens_seen": 30909312,
      "step": 5116,
      "train_runtime": 5183.342,
      "train_tokens_per_second": 5963.201
    },
    {
      "epoch": 1.8820229885057471,
      "grad_norm": 0.99626624584198,
      "learning_rate": 1.1853487944045647e-05,
      "loss": 1.4375,
      "num_input_tokens_seen": 30915328,
      "step": 5117,
      "train_runtime": 5184.3239,
      "train_tokens_per_second": 5963.232
    },
    {
      "epoch": 1.8823908045977011,
      "grad_norm": 0.9325799345970154,
      "learning_rate": 1.1816675869685258e-05,
      "loss": 1.5727,
      "num_input_tokens_seen": 30921824,
      "step": 5118,
      "train_runtime": 5185.3516,
      "train_tokens_per_second": 5963.303
    },
    {
      "epoch": 1.8827586206896552,
      "grad_norm": 1.0887229442596436,
      "learning_rate": 1.1779863795324868e-05,
      "loss": 1.6099,
      "num_input_tokens_seen": 30926448,
      "step": 5119,
      "train_runtime": 5186.14,
      "train_tokens_per_second": 5963.288
    },
    {
      "epoch": 1.8831264367816092,
      "grad_norm": 0.9841151237487793,
      "learning_rate": 1.1743051720964477e-05,
      "loss": 1.4023,
      "num_input_tokens_seen": 30934400,
      "step": 5120,
      "train_runtime": 5187.385,
      "train_tokens_per_second": 5963.39
    },
    {
      "epoch": 1.8834942528735632,
      "grad_norm": 1.0270477533340454,
      "learning_rate": 1.1706239646604086e-05,
      "loss": 1.6605,
      "num_input_tokens_seen": 30940272,
      "step": 5121,
      "train_runtime": 5188.3421,
      "train_tokens_per_second": 5963.422
    },
    {
      "epoch": 1.8838620689655172,
      "grad_norm": 0.9291617274284363,
      "learning_rate": 1.1669427572243697e-05,
      "loss": 1.6315,
      "num_input_tokens_seen": 30947440,
      "step": 5122,
      "train_runtime": 5189.443,
      "train_tokens_per_second": 5963.538
    },
    {
      "epoch": 1.8842298850574712,
      "grad_norm": 1.0358939170837402,
      "learning_rate": 1.1632615497883305e-05,
      "loss": 1.5352,
      "num_input_tokens_seen": 30953536,
      "step": 5123,
      "train_runtime": 5190.4202,
      "train_tokens_per_second": 5963.59
    },
    {
      "epoch": 1.8845977011494253,
      "grad_norm": 1.0341215133666992,
      "learning_rate": 1.1595803423522916e-05,
      "loss": 1.4522,
      "num_input_tokens_seen": 30959968,
      "step": 5124,
      "train_runtime": 5191.4722,
      "train_tokens_per_second": 5963.62
    },
    {
      "epoch": 1.8849655172413793,
      "grad_norm": 1.081660509109497,
      "learning_rate": 1.1558991349162527e-05,
      "loss": 1.4329,
      "num_input_tokens_seen": 30965424,
      "step": 5125,
      "train_runtime": 5192.3878,
      "train_tokens_per_second": 5963.619
    },
    {
      "epoch": 1.8853333333333333,
      "grad_norm": 0.9941276907920837,
      "learning_rate": 1.1522179274802136e-05,
      "loss": 1.5152,
      "num_input_tokens_seen": 30969504,
      "step": 5126,
      "train_runtime": 5193.1151,
      "train_tokens_per_second": 5963.57
    },
    {
      "epoch": 1.8857011494252873,
      "grad_norm": 0.9303632378578186,
      "learning_rate": 1.1485367200441744e-05,
      "loss": 1.3556,
      "num_input_tokens_seen": 30977888,
      "step": 5127,
      "train_runtime": 5194.4158,
      "train_tokens_per_second": 5963.691
    },
    {
      "epoch": 1.8860689655172413,
      "grad_norm": 1.0067976713180542,
      "learning_rate": 1.1448555126081355e-05,
      "loss": 1.3019,
      "num_input_tokens_seen": 30983568,
      "step": 5128,
      "train_runtime": 5195.3456,
      "train_tokens_per_second": 5963.716
    },
    {
      "epoch": 1.8864367816091954,
      "grad_norm": 1.1567789316177368,
      "learning_rate": 1.1411743051720966e-05,
      "loss": 1.7082,
      "num_input_tokens_seen": 30989088,
      "step": 5129,
      "train_runtime": 5196.2779,
      "train_tokens_per_second": 5963.709
    },
    {
      "epoch": 1.8868045977011494,
      "grad_norm": 1.0901026725769043,
      "learning_rate": 1.1374930977360574e-05,
      "loss": 1.7373,
      "num_input_tokens_seen": 30993632,
      "step": 5130,
      "train_runtime": 5197.0676,
      "train_tokens_per_second": 5963.677
    },
    {
      "epoch": 1.8871724137931034,
      "grad_norm": 1.068395972251892,
      "learning_rate": 1.1338118903000185e-05,
      "loss": 1.4796,
      "num_input_tokens_seen": 31000272,
      "step": 5131,
      "train_runtime": 5198.7044,
      "train_tokens_per_second": 5963.076
    },
    {
      "epoch": 1.8875402298850574,
      "grad_norm": 0.9585198760032654,
      "learning_rate": 1.1301306828639794e-05,
      "loss": 1.4375,
      "num_input_tokens_seen": 31010048,
      "step": 5132,
      "train_runtime": 5200.1892,
      "train_tokens_per_second": 5963.254
    },
    {
      "epoch": 1.8879080459770115,
      "grad_norm": 0.9547199606895447,
      "learning_rate": 1.1264494754279405e-05,
      "loss": 1.3827,
      "num_input_tokens_seen": 31015504,
      "step": 5133,
      "train_runtime": 5201.1071,
      "train_tokens_per_second": 5963.25
    },
    {
      "epoch": 1.8882758620689655,
      "grad_norm": 1.1191118955612183,
      "learning_rate": 1.1227682679919015e-05,
      "loss": 1.562,
      "num_input_tokens_seen": 31021152,
      "step": 5134,
      "train_runtime": 5202.0602,
      "train_tokens_per_second": 5963.244
    },
    {
      "epoch": 1.8886436781609195,
      "grad_norm": 0.9584413766860962,
      "learning_rate": 1.1190870605558624e-05,
      "loss": 1.324,
      "num_input_tokens_seen": 31025824,
      "step": 5135,
      "train_runtime": 5202.8468,
      "train_tokens_per_second": 5963.24
    },
    {
      "epoch": 1.8890114942528735,
      "grad_norm": 0.9913816452026367,
      "learning_rate": 1.1154058531198233e-05,
      "loss": 1.3563,
      "num_input_tokens_seen": 31030080,
      "step": 5136,
      "train_runtime": 5203.5844,
      "train_tokens_per_second": 5963.213
    },
    {
      "epoch": 1.8893793103448275,
      "grad_norm": 1.0783475637435913,
      "learning_rate": 1.1117246456837843e-05,
      "loss": 1.5492,
      "num_input_tokens_seen": 31035488,
      "step": 5137,
      "train_runtime": 5204.4588,
      "train_tokens_per_second": 5963.25
    },
    {
      "epoch": 1.8897471264367816,
      "grad_norm": 1.1071451902389526,
      "learning_rate": 1.1080434382477452e-05,
      "loss": 1.6794,
      "num_input_tokens_seen": 31040272,
      "step": 5138,
      "train_runtime": 5205.2863,
      "train_tokens_per_second": 5963.221
    },
    {
      "epoch": 1.8901149425287356,
      "grad_norm": 1.0222585201263428,
      "learning_rate": 1.1043622308117063e-05,
      "loss": 1.6259,
      "num_input_tokens_seen": 31046256,
      "step": 5139,
      "train_runtime": 5206.258,
      "train_tokens_per_second": 5963.257
    },
    {
      "epoch": 1.8904827586206898,
      "grad_norm": 1.0303621292114258,
      "learning_rate": 1.1006810233756674e-05,
      "loss": 1.4607,
      "num_input_tokens_seen": 31052672,
      "step": 5140,
      "train_runtime": 5207.3154,
      "train_tokens_per_second": 5963.278
    },
    {
      "epoch": 1.8908505747126436,
      "grad_norm": 1.0179482698440552,
      "learning_rate": 1.0969998159396282e-05,
      "loss": 1.5509,
      "num_input_tokens_seen": 31058288,
      "step": 5141,
      "train_runtime": 5208.2364,
      "train_tokens_per_second": 5963.302
    },
    {
      "epoch": 1.8912183908045979,
      "grad_norm": 1.0750333070755005,
      "learning_rate": 1.0933186085035891e-05,
      "loss": 1.4189,
      "num_input_tokens_seen": 31064480,
      "step": 5142,
      "train_runtime": 5209.2428,
      "train_tokens_per_second": 5963.339
    },
    {
      "epoch": 1.8915862068965517,
      "grad_norm": 1.103312611579895,
      "learning_rate": 1.0896374010675502e-05,
      "loss": 1.5595,
      "num_input_tokens_seen": 31068832,
      "step": 5143,
      "train_runtime": 5210.0232,
      "train_tokens_per_second": 5963.281
    },
    {
      "epoch": 1.891954022988506,
      "grad_norm": 0.8550114631652832,
      "learning_rate": 1.085956193631511e-05,
      "loss": 1.2791,
      "num_input_tokens_seen": 31075232,
      "step": 5144,
      "train_runtime": 5211.0595,
      "train_tokens_per_second": 5963.323
    },
    {
      "epoch": 1.8923218390804597,
      "grad_norm": 1.001771092414856,
      "learning_rate": 1.0822749861954721e-05,
      "loss": 1.4215,
      "num_input_tokens_seen": 31083056,
      "step": 5145,
      "train_runtime": 5212.265,
      "train_tokens_per_second": 5963.445
    },
    {
      "epoch": 1.892689655172414,
      "grad_norm": 1.0242040157318115,
      "learning_rate": 1.0785937787594332e-05,
      "loss": 1.6836,
      "num_input_tokens_seen": 31089024,
      "step": 5146,
      "train_runtime": 5213.2311,
      "train_tokens_per_second": 5963.485
    },
    {
      "epoch": 1.8930574712643677,
      "grad_norm": 1.105438232421875,
      "learning_rate": 1.0749125713233941e-05,
      "loss": 1.5183,
      "num_input_tokens_seen": 31095232,
      "step": 5147,
      "train_runtime": 5214.2046,
      "train_tokens_per_second": 5963.562
    },
    {
      "epoch": 1.893425287356322,
      "grad_norm": 1.0332019329071045,
      "learning_rate": 1.0712313638873551e-05,
      "loss": 1.6151,
      "num_input_tokens_seen": 31100480,
      "step": 5148,
      "train_runtime": 5215.0946,
      "train_tokens_per_second": 5963.551
    },
    {
      "epoch": 1.8937931034482758,
      "grad_norm": 1.1260578632354736,
      "learning_rate": 1.0675501564513162e-05,
      "loss": 1.4997,
      "num_input_tokens_seen": 31106176,
      "step": 5149,
      "train_runtime": 5216.0286,
      "train_tokens_per_second": 5963.575
    },
    {
      "epoch": 1.89416091954023,
      "grad_norm": 1.1261924505233765,
      "learning_rate": 1.063868949015277e-05,
      "loss": 1.594,
      "num_input_tokens_seen": 31110480,
      "step": 5150,
      "train_runtime": 5216.7977,
      "train_tokens_per_second": 5963.521
    },
    {
      "epoch": 1.8945287356321838,
      "grad_norm": 0.9641284942626953,
      "learning_rate": 1.060187741579238e-05,
      "loss": 1.4913,
      "num_input_tokens_seen": 31115440,
      "step": 5151,
      "train_runtime": 5217.6397,
      "train_tokens_per_second": 5963.509
    },
    {
      "epoch": 1.894896551724138,
      "grad_norm": 0.9839797616004944,
      "learning_rate": 1.056506534143199e-05,
      "loss": 1.556,
      "num_input_tokens_seen": 31121200,
      "step": 5152,
      "train_runtime": 5218.5865,
      "train_tokens_per_second": 5963.531
    },
    {
      "epoch": 1.8952643678160919,
      "grad_norm": 0.9305192232131958,
      "learning_rate": 1.05282532670716e-05,
      "loss": 1.2631,
      "num_input_tokens_seen": 31126784,
      "step": 5153,
      "train_runtime": 5219.516,
      "train_tokens_per_second": 5963.538
    },
    {
      "epoch": 1.895632183908046,
      "grad_norm": 0.9300264716148376,
      "learning_rate": 1.049144119271121e-05,
      "loss": 1.5224,
      "num_input_tokens_seen": 31131792,
      "step": 5154,
      "train_runtime": 5220.3817,
      "train_tokens_per_second": 5963.509
    },
    {
      "epoch": 1.896,
      "grad_norm": 0.9646492004394531,
      "learning_rate": 1.045462911835082e-05,
      "loss": 1.546,
      "num_input_tokens_seen": 31137872,
      "step": 5155,
      "train_runtime": 5221.3764,
      "train_tokens_per_second": 5963.537
    },
    {
      "epoch": 1.8963678160919542,
      "grad_norm": 0.958551824092865,
      "learning_rate": 1.041781704399043e-05,
      "loss": 1.3527,
      "num_input_tokens_seen": 31144128,
      "step": 5156,
      "train_runtime": 5222.3764,
      "train_tokens_per_second": 5963.593
    },
    {
      "epoch": 1.896735632183908,
      "grad_norm": 1.0264830589294434,
      "learning_rate": 1.0381004969630038e-05,
      "loss": 1.6428,
      "num_input_tokens_seen": 31149536,
      "step": 5157,
      "train_runtime": 5223.2814,
      "train_tokens_per_second": 5963.595
    },
    {
      "epoch": 1.8971034482758622,
      "grad_norm": 0.9699311852455139,
      "learning_rate": 1.0344192895269649e-05,
      "loss": 1.4691,
      "num_input_tokens_seen": 31156208,
      "step": 5158,
      "train_runtime": 5224.3513,
      "train_tokens_per_second": 5963.651
    },
    {
      "epoch": 1.897471264367816,
      "grad_norm": 0.9945620894432068,
      "learning_rate": 1.0307380820909258e-05,
      "loss": 1.5895,
      "num_input_tokens_seen": 31161248,
      "step": 5159,
      "train_runtime": 5225.1883,
      "train_tokens_per_second": 5963.66
    },
    {
      "epoch": 1.8978390804597702,
      "grad_norm": 0.953173041343689,
      "learning_rate": 1.0270568746548868e-05,
      "loss": 1.5288,
      "num_input_tokens_seen": 31166720,
      "step": 5160,
      "train_runtime": 5226.1059,
      "train_tokens_per_second": 5963.66
    },
    {
      "epoch": 1.898206896551724,
      "grad_norm": 0.8694062829017639,
      "learning_rate": 1.0233756672188479e-05,
      "loss": 1.4781,
      "num_input_tokens_seen": 31173360,
      "step": 5161,
      "train_runtime": 5227.6412,
      "train_tokens_per_second": 5963.179
    },
    {
      "epoch": 1.8985747126436783,
      "grad_norm": 1.0044862031936646,
      "learning_rate": 1.0196944597828088e-05,
      "loss": 1.4316,
      "num_input_tokens_seen": 31179296,
      "step": 5162,
      "train_runtime": 5228.5806,
      "train_tokens_per_second": 5963.243
    },
    {
      "epoch": 1.898942528735632,
      "grad_norm": 1.098825454711914,
      "learning_rate": 1.0160132523467698e-05,
      "loss": 1.4944,
      "num_input_tokens_seen": 31184288,
      "step": 5163,
      "train_runtime": 5229.4351,
      "train_tokens_per_second": 5963.223
    },
    {
      "epoch": 1.8993103448275863,
      "grad_norm": 1.1091935634613037,
      "learning_rate": 1.0123320449107309e-05,
      "loss": 1.4743,
      "num_input_tokens_seen": 31189440,
      "step": 5164,
      "train_runtime": 5230.2784,
      "train_tokens_per_second": 5963.247
    },
    {
      "epoch": 1.8996781609195401,
      "grad_norm": 0.9583799242973328,
      "learning_rate": 1.0086508374746916e-05,
      "loss": 1.355,
      "num_input_tokens_seen": 31194624,
      "step": 5165,
      "train_runtime": 5231.1459,
      "train_tokens_per_second": 5963.249
    },
    {
      "epoch": 1.9000459770114944,
      "grad_norm": 1.2292479276657104,
      "learning_rate": 1.0049696300386527e-05,
      "loss": 1.6079,
      "num_input_tokens_seen": 31202016,
      "step": 5166,
      "train_runtime": 5232.3001,
      "train_tokens_per_second": 5963.346
    },
    {
      "epoch": 1.9004137931034482,
      "grad_norm": 1.0354876518249512,
      "learning_rate": 1.0012884226026137e-05,
      "loss": 1.4838,
      "num_input_tokens_seen": 31207936,
      "step": 5167,
      "train_runtime": 5233.2551,
      "train_tokens_per_second": 5963.389
    },
    {
      "epoch": 1.9007816091954024,
      "grad_norm": 0.9916900396347046,
      "learning_rate": 9.976072151665746e-06,
      "loss": 1.2896,
      "num_input_tokens_seen": 31218368,
      "step": 5168,
      "train_runtime": 5234.8604,
      "train_tokens_per_second": 5963.553
    },
    {
      "epoch": 1.9011494252873562,
      "grad_norm": 1.0767730474472046,
      "learning_rate": 9.939260077305357e-06,
      "loss": 1.2717,
      "num_input_tokens_seen": 31226064,
      "step": 5169,
      "train_runtime": 5236.0753,
      "train_tokens_per_second": 5963.639
    },
    {
      "epoch": 1.9015172413793104,
      "grad_norm": 0.9991474747657776,
      "learning_rate": 9.902448002944967e-06,
      "loss": 1.3001,
      "num_input_tokens_seen": 31231104,
      "step": 5170,
      "train_runtime": 5236.9048,
      "train_tokens_per_second": 5963.657
    },
    {
      "epoch": 1.9018850574712642,
      "grad_norm": 1.0238171815872192,
      "learning_rate": 9.865635928584576e-06,
      "loss": 1.6231,
      "num_input_tokens_seen": 31235952,
      "step": 5171,
      "train_runtime": 5237.7384,
      "train_tokens_per_second": 5963.633
    },
    {
      "epoch": 1.9022528735632185,
      "grad_norm": 1.0991358757019043,
      "learning_rate": 9.828823854224185e-06,
      "loss": 1.7272,
      "num_input_tokens_seen": 31240192,
      "step": 5172,
      "train_runtime": 5238.4881,
      "train_tokens_per_second": 5963.589
    },
    {
      "epoch": 1.9026206896551723,
      "grad_norm": 0.9541977643966675,
      "learning_rate": 9.792011779863796e-06,
      "loss": 1.61,
      "num_input_tokens_seen": 31245392,
      "step": 5173,
      "train_runtime": 5239.3567,
      "train_tokens_per_second": 5963.593
    },
    {
      "epoch": 1.9029885057471265,
      "grad_norm": 1.1026161909103394,
      "learning_rate": 9.755199705503405e-06,
      "loss": 1.3701,
      "num_input_tokens_seen": 31252752,
      "step": 5174,
      "train_runtime": 5240.54,
      "train_tokens_per_second": 5963.651
    },
    {
      "epoch": 1.9033563218390803,
      "grad_norm": 1.078701376914978,
      "learning_rate": 9.718387631143015e-06,
      "loss": 1.4136,
      "num_input_tokens_seen": 31259136,
      "step": 5175,
      "train_runtime": 5241.5806,
      "train_tokens_per_second": 5963.685
    },
    {
      "epoch": 1.9037241379310346,
      "grad_norm": 0.9863090515136719,
      "learning_rate": 9.681575556782626e-06,
      "loss": 1.4909,
      "num_input_tokens_seen": 31264528,
      "step": 5176,
      "train_runtime": 5242.4533,
      "train_tokens_per_second": 5963.721
    },
    {
      "epoch": 1.9040919540229884,
      "grad_norm": 1.059182047843933,
      "learning_rate": 9.644763482422235e-06,
      "loss": 1.6034,
      "num_input_tokens_seen": 31271536,
      "step": 5177,
      "train_runtime": 5243.5402,
      "train_tokens_per_second": 5963.821
    },
    {
      "epoch": 1.9044597701149426,
      "grad_norm": 0.9668793678283691,
      "learning_rate": 9.607951408061845e-06,
      "loss": 1.498,
      "num_input_tokens_seen": 31277888,
      "step": 5178,
      "train_runtime": 5244.5722,
      "train_tokens_per_second": 5963.859
    },
    {
      "epoch": 1.9048275862068964,
      "grad_norm": 0.8954527378082275,
      "learning_rate": 9.571139333701456e-06,
      "loss": 1.2766,
      "num_input_tokens_seen": 31283296,
      "step": 5179,
      "train_runtime": 5245.4754,
      "train_tokens_per_second": 5963.863
    },
    {
      "epoch": 1.9051954022988506,
      "grad_norm": 0.9767663478851318,
      "learning_rate": 9.534327259341063e-06,
      "loss": 1.4348,
      "num_input_tokens_seen": 31290528,
      "step": 5180,
      "train_runtime": 5246.6416,
      "train_tokens_per_second": 5963.916
    },
    {
      "epoch": 1.9055632183908044,
      "grad_norm": 0.9650322794914246,
      "learning_rate": 9.497515184980674e-06,
      "loss": 1.4885,
      "num_input_tokens_seen": 31296688,
      "step": 5181,
      "train_runtime": 5247.6372,
      "train_tokens_per_second": 5963.958
    },
    {
      "epoch": 1.9059310344827587,
      "grad_norm": 1.0090296268463135,
      "learning_rate": 9.460703110620284e-06,
      "loss": 1.3792,
      "num_input_tokens_seen": 31302496,
      "step": 5182,
      "train_runtime": 5248.6318,
      "train_tokens_per_second": 5963.934
    },
    {
      "epoch": 1.9062988505747125,
      "grad_norm": 0.9824272990226746,
      "learning_rate": 9.423891036259893e-06,
      "loss": 1.5411,
      "num_input_tokens_seen": 31306976,
      "step": 5183,
      "train_runtime": 5249.3936,
      "train_tokens_per_second": 5963.922
    },
    {
      "epoch": 1.9066666666666667,
      "grad_norm": 0.999448299407959,
      "learning_rate": 9.387078961899504e-06,
      "loss": 1.6293,
      "num_input_tokens_seen": 31312288,
      "step": 5184,
      "train_runtime": 5250.2777,
      "train_tokens_per_second": 5963.93
    },
    {
      "epoch": 1.9070344827586205,
      "grad_norm": 0.9954407215118408,
      "learning_rate": 9.350266887539114e-06,
      "loss": 1.3755,
      "num_input_tokens_seen": 31317840,
      "step": 5185,
      "train_runtime": 5251.1845,
      "train_tokens_per_second": 5963.957
    },
    {
      "epoch": 1.9074022988505748,
      "grad_norm": 1.073150634765625,
      "learning_rate": 9.313454813178723e-06,
      "loss": 1.6224,
      "num_input_tokens_seen": 31322272,
      "step": 5186,
      "train_runtime": 5251.9603,
      "train_tokens_per_second": 5963.92
    },
    {
      "epoch": 1.9077701149425286,
      "grad_norm": 1.0602670907974243,
      "learning_rate": 9.276642738818332e-06,
      "loss": 1.4079,
      "num_input_tokens_seen": 31327888,
      "step": 5187,
      "train_runtime": 5252.9206,
      "train_tokens_per_second": 5963.899
    },
    {
      "epoch": 1.9081379310344828,
      "grad_norm": 1.0609350204467773,
      "learning_rate": 9.239830664457943e-06,
      "loss": 1.475,
      "num_input_tokens_seen": 31332544,
      "step": 5188,
      "train_runtime": 5253.7163,
      "train_tokens_per_second": 5963.882
    },
    {
      "epoch": 1.9085057471264368,
      "grad_norm": 0.9889388084411621,
      "learning_rate": 9.203018590097552e-06,
      "loss": 1.367,
      "num_input_tokens_seen": 31337504,
      "step": 5189,
      "train_runtime": 5254.5457,
      "train_tokens_per_second": 5963.885
    },
    {
      "epoch": 1.9088735632183909,
      "grad_norm": 0.9926358461380005,
      "learning_rate": 9.166206515737162e-06,
      "loss": 1.3965,
      "num_input_tokens_seen": 31345360,
      "step": 5190,
      "train_runtime": 5255.7574,
      "train_tokens_per_second": 5964.004
    },
    {
      "epoch": 1.9092413793103449,
      "grad_norm": 1.0332140922546387,
      "learning_rate": 9.129394441376773e-06,
      "loss": 1.4543,
      "num_input_tokens_seen": 31350352,
      "step": 5191,
      "train_runtime": 5257.0693,
      "train_tokens_per_second": 5963.466
    },
    {
      "epoch": 1.909609195402299,
      "grad_norm": 0.9564376473426819,
      "learning_rate": 9.092582367016382e-06,
      "loss": 1.6294,
      "num_input_tokens_seen": 31356528,
      "step": 5192,
      "train_runtime": 5258.0999,
      "train_tokens_per_second": 5963.471
    },
    {
      "epoch": 1.909977011494253,
      "grad_norm": 1.043769121170044,
      "learning_rate": 9.055770292655992e-06,
      "loss": 1.4124,
      "num_input_tokens_seen": 31364288,
      "step": 5193,
      "train_runtime": 5259.3107,
      "train_tokens_per_second": 5963.574
    },
    {
      "epoch": 1.910344827586207,
      "grad_norm": 1.0887336730957031,
      "learning_rate": 9.018958218295601e-06,
      "loss": 1.6384,
      "num_input_tokens_seen": 31370832,
      "step": 5194,
      "train_runtime": 5260.3615,
      "train_tokens_per_second": 5963.627
    },
    {
      "epoch": 1.910712643678161,
      "grad_norm": 0.9897884726524353,
      "learning_rate": 8.98214614393521e-06,
      "loss": 1.6019,
      "num_input_tokens_seen": 31375632,
      "step": 5195,
      "train_runtime": 5261.1987,
      "train_tokens_per_second": 5963.59
    },
    {
      "epoch": 1.911080459770115,
      "grad_norm": 1.119964361190796,
      "learning_rate": 8.94533406957482e-06,
      "loss": 1.4412,
      "num_input_tokens_seen": 31380384,
      "step": 5196,
      "train_runtime": 5262.0319,
      "train_tokens_per_second": 5963.549
    },
    {
      "epoch": 1.911448275862069,
      "grad_norm": 1.0215885639190674,
      "learning_rate": 8.908521995214431e-06,
      "loss": 1.3712,
      "num_input_tokens_seen": 31387552,
      "step": 5197,
      "train_runtime": 5263.1631,
      "train_tokens_per_second": 5963.629
    },
    {
      "epoch": 1.911816091954023,
      "grad_norm": 1.0042119026184082,
      "learning_rate": 8.87170992085404e-06,
      "loss": 1.546,
      "num_input_tokens_seen": 31393088,
      "step": 5198,
      "train_runtime": 5264.082,
      "train_tokens_per_second": 5963.64
    },
    {
      "epoch": 1.912183908045977,
      "grad_norm": 0.9443457126617432,
      "learning_rate": 8.83489784649365e-06,
      "loss": 1.6105,
      "num_input_tokens_seen": 31400144,
      "step": 5199,
      "train_runtime": 5265.2117,
      "train_tokens_per_second": 5963.7
    },
    {
      "epoch": 1.912551724137931,
      "grad_norm": 0.7088549733161926,
      "learning_rate": 8.798085772133261e-06,
      "loss": 1.0782,
      "num_input_tokens_seen": 31416256,
      "step": 5200,
      "train_runtime": 5267.5304,
      "train_tokens_per_second": 5964.134
    },
    {
      "epoch": 1.912919540229885,
      "grad_norm": 0.9911025762557983,
      "learning_rate": 8.76127369777287e-06,
      "loss": 1.4901,
      "num_input_tokens_seen": 31422016,
      "step": 5201,
      "train_runtime": 5268.4789,
      "train_tokens_per_second": 5964.153
    },
    {
      "epoch": 1.913287356321839,
      "grad_norm": 0.9640730619430542,
      "learning_rate": 8.724461623412479e-06,
      "loss": 1.4643,
      "num_input_tokens_seen": 31427824,
      "step": 5202,
      "train_runtime": 5269.4289,
      "train_tokens_per_second": 5964.18
    },
    {
      "epoch": 1.9136551724137931,
      "grad_norm": 0.970166802406311,
      "learning_rate": 8.68764954905209e-06,
      "loss": 1.4615,
      "num_input_tokens_seen": 31433904,
      "step": 5203,
      "train_runtime": 5270.4451,
      "train_tokens_per_second": 5964.184
    },
    {
      "epoch": 1.9140229885057471,
      "grad_norm": 1.080848217010498,
      "learning_rate": 8.650837474691699e-06,
      "loss": 1.5406,
      "num_input_tokens_seen": 31439952,
      "step": 5204,
      "train_runtime": 5271.4238,
      "train_tokens_per_second": 5964.224
    },
    {
      "epoch": 1.9143908045977012,
      "grad_norm": 0.9751112461090088,
      "learning_rate": 8.61402540033131e-06,
      "loss": 1.5039,
      "num_input_tokens_seen": 31446912,
      "step": 5205,
      "train_runtime": 5272.5104,
      "train_tokens_per_second": 5964.315
    },
    {
      "epoch": 1.9147586206896552,
      "grad_norm": 1.0692551136016846,
      "learning_rate": 8.57721332597092e-06,
      "loss": 1.6031,
      "num_input_tokens_seen": 31451504,
      "step": 5206,
      "train_runtime": 5273.2933,
      "train_tokens_per_second": 5964.3
    },
    {
      "epoch": 1.9151264367816092,
      "grad_norm": 1.0313695669174194,
      "learning_rate": 8.540401251610529e-06,
      "loss": 1.5916,
      "num_input_tokens_seen": 31456528,
      "step": 5207,
      "train_runtime": 5274.1527,
      "train_tokens_per_second": 5964.281
    },
    {
      "epoch": 1.9154942528735632,
      "grad_norm": 1.0278955698013306,
      "learning_rate": 8.50358917725014e-06,
      "loss": 1.4985,
      "num_input_tokens_seen": 31461456,
      "step": 5208,
      "train_runtime": 5275.007,
      "train_tokens_per_second": 5964.249
    },
    {
      "epoch": 1.9158620689655173,
      "grad_norm": 0.9766169190406799,
      "learning_rate": 8.466777102889748e-06,
      "loss": 1.4472,
      "num_input_tokens_seen": 31466720,
      "step": 5209,
      "train_runtime": 5275.8893,
      "train_tokens_per_second": 5964.249
    },
    {
      "epoch": 1.9162298850574713,
      "grad_norm": 1.1143441200256348,
      "learning_rate": 8.429965028529357e-06,
      "loss": 1.537,
      "num_input_tokens_seen": 31471712,
      "step": 5210,
      "train_runtime": 5276.741,
      "train_tokens_per_second": 5964.233
    },
    {
      "epoch": 1.9165977011494253,
      "grad_norm": 0.9699863791465759,
      "learning_rate": 8.393152954168968e-06,
      "loss": 1.5436,
      "num_input_tokens_seen": 31477680,
      "step": 5211,
      "train_runtime": 5277.7127,
      "train_tokens_per_second": 5964.266
    },
    {
      "epoch": 1.9169655172413793,
      "grad_norm": 1.1384267807006836,
      "learning_rate": 8.356340879808578e-06,
      "loss": 1.438,
      "num_input_tokens_seen": 31487840,
      "step": 5212,
      "train_runtime": 5279.2373,
      "train_tokens_per_second": 5964.468
    },
    {
      "epoch": 1.9173333333333333,
      "grad_norm": 1.039361596107483,
      "learning_rate": 8.319528805448187e-06,
      "loss": 1.5273,
      "num_input_tokens_seen": 31492272,
      "step": 5213,
      "train_runtime": 5280.0139,
      "train_tokens_per_second": 5964.43
    },
    {
      "epoch": 1.9177011494252874,
      "grad_norm": 1.2111543416976929,
      "learning_rate": 8.282716731087798e-06,
      "loss": 1.6289,
      "num_input_tokens_seen": 31497696,
      "step": 5214,
      "train_runtime": 5280.9006,
      "train_tokens_per_second": 5964.455
    },
    {
      "epoch": 1.9180689655172414,
      "grad_norm": 1.0253822803497314,
      "learning_rate": 8.245904656727407e-06,
      "loss": 1.5035,
      "num_input_tokens_seen": 31502496,
      "step": 5215,
      "train_runtime": 5281.7188,
      "train_tokens_per_second": 5964.44
    },
    {
      "epoch": 1.9184367816091954,
      "grad_norm": 1.0145090818405151,
      "learning_rate": 8.209092582367017e-06,
      "loss": 1.4359,
      "num_input_tokens_seen": 31508688,
      "step": 5216,
      "train_runtime": 5282.7161,
      "train_tokens_per_second": 5964.486
    },
    {
      "epoch": 1.9188045977011494,
      "grad_norm": 1.071014404296875,
      "learning_rate": 8.172280508006628e-06,
      "loss": 1.5631,
      "num_input_tokens_seen": 31513632,
      "step": 5217,
      "train_runtime": 5283.5637,
      "train_tokens_per_second": 5964.465
    },
    {
      "epoch": 1.9191724137931034,
      "grad_norm": 1.0823394060134888,
      "learning_rate": 8.135468433646237e-06,
      "loss": 1.334,
      "num_input_tokens_seen": 31519808,
      "step": 5218,
      "train_runtime": 5284.5715,
      "train_tokens_per_second": 5964.496
    },
    {
      "epoch": 1.9195402298850575,
      "grad_norm": 0.9998838901519775,
      "learning_rate": 8.098656359285846e-06,
      "loss": 1.4185,
      "num_input_tokens_seen": 31526208,
      "step": 5219,
      "train_runtime": 5285.6075,
      "train_tokens_per_second": 5964.538
    },
    {
      "epoch": 1.9199080459770115,
      "grad_norm": 1.0125830173492432,
      "learning_rate": 8.061844284925456e-06,
      "loss": 1.2764,
      "num_input_tokens_seen": 31535088,
      "step": 5220,
      "train_runtime": 5286.9402,
      "train_tokens_per_second": 5964.714
    },
    {
      "epoch": 1.9202758620689655,
      "grad_norm": 1.0879446268081665,
      "learning_rate": 8.025032210565065e-06,
      "loss": 1.2968,
      "num_input_tokens_seen": 31542240,
      "step": 5221,
      "train_runtime": 5288.5831,
      "train_tokens_per_second": 5964.214
    },
    {
      "epoch": 1.9206436781609195,
      "grad_norm": 1.0482430458068848,
      "learning_rate": 7.988220136204676e-06,
      "loss": 1.5569,
      "num_input_tokens_seen": 31546688,
      "step": 5222,
      "train_runtime": 5289.3382,
      "train_tokens_per_second": 5964.203
    },
    {
      "epoch": 1.9210114942528735,
      "grad_norm": 1.0618278980255127,
      "learning_rate": 7.951408061844286e-06,
      "loss": 1.649,
      "num_input_tokens_seen": 31551264,
      "step": 5223,
      "train_runtime": 5290.1443,
      "train_tokens_per_second": 5964.159
    },
    {
      "epoch": 1.9213793103448276,
      "grad_norm": 0.9304054379463196,
      "learning_rate": 7.914595987483895e-06,
      "loss": 1.364,
      "num_input_tokens_seen": 31557488,
      "step": 5224,
      "train_runtime": 5291.1563,
      "train_tokens_per_second": 5964.195
    },
    {
      "epoch": 1.9217471264367816,
      "grad_norm": 1.078218936920166,
      "learning_rate": 7.877783913123504e-06,
      "loss": 1.5614,
      "num_input_tokens_seen": 31562144,
      "step": 5225,
      "train_runtime": 5291.9591,
      "train_tokens_per_second": 5964.17
    },
    {
      "epoch": 1.9221149425287356,
      "grad_norm": 1.0731745958328247,
      "learning_rate": 7.840971838763115e-06,
      "loss": 1.4664,
      "num_input_tokens_seen": 31566320,
      "step": 5226,
      "train_runtime": 5292.6876,
      "train_tokens_per_second": 5964.138
    },
    {
      "epoch": 1.9224827586206896,
      "grad_norm": 1.0793852806091309,
      "learning_rate": 7.804159764402723e-06,
      "loss": 1.7021,
      "num_input_tokens_seen": 31571936,
      "step": 5227,
      "train_runtime": 5293.6037,
      "train_tokens_per_second": 5964.167
    },
    {
      "epoch": 1.9228505747126436,
      "grad_norm": 1.0043725967407227,
      "learning_rate": 7.767347690042334e-06,
      "loss": 1.6227,
      "num_input_tokens_seen": 31577456,
      "step": 5228,
      "train_runtime": 5294.53,
      "train_tokens_per_second": 5964.166
    },
    {
      "epoch": 1.9232183908045977,
      "grad_norm": 0.981519341468811,
      "learning_rate": 7.730535615681945e-06,
      "loss": 1.5114,
      "num_input_tokens_seen": 31582368,
      "step": 5229,
      "train_runtime": 5295.3563,
      "train_tokens_per_second": 5964.163
    },
    {
      "epoch": 1.9235862068965517,
      "grad_norm": 1.0342694520950317,
      "learning_rate": 7.693723541321553e-06,
      "loss": 1.3978,
      "num_input_tokens_seen": 31590080,
      "step": 5230,
      "train_runtime": 5296.5787,
      "train_tokens_per_second": 5964.242
    },
    {
      "epoch": 1.9239540229885057,
      "grad_norm": 1.0602362155914307,
      "learning_rate": 7.656911466961164e-06,
      "loss": 1.349,
      "num_input_tokens_seen": 31594880,
      "step": 5231,
      "train_runtime": 5297.4254,
      "train_tokens_per_second": 5964.195
    },
    {
      "epoch": 1.9243218390804597,
      "grad_norm": 1.0674331188201904,
      "learning_rate": 7.620099392600774e-06,
      "loss": 1.5256,
      "num_input_tokens_seen": 31599808,
      "step": 5232,
      "train_runtime": 5298.2491,
      "train_tokens_per_second": 5964.198
    },
    {
      "epoch": 1.9246896551724137,
      "grad_norm": 1.0152877569198608,
      "learning_rate": 7.583287318240383e-06,
      "loss": 1.2964,
      "num_input_tokens_seen": 31604352,
      "step": 5233,
      "train_runtime": 5299.0243,
      "train_tokens_per_second": 5964.183
    },
    {
      "epoch": 1.9250574712643678,
      "grad_norm": 0.9840177893638611,
      "learning_rate": 7.5464752438799924e-06,
      "loss": 1.3441,
      "num_input_tokens_seen": 31610992,
      "step": 5234,
      "train_runtime": 5300.1224,
      "train_tokens_per_second": 5964.2
    },
    {
      "epoch": 1.9254252873563218,
      "grad_norm": 1.013264775276184,
      "learning_rate": 7.509663169519603e-06,
      "loss": 1.4727,
      "num_input_tokens_seen": 31615696,
      "step": 5235,
      "train_runtime": 5300.9426,
      "train_tokens_per_second": 5964.165
    },
    {
      "epoch": 1.9257931034482758,
      "grad_norm": 1.1546998023986816,
      "learning_rate": 7.472851095159213e-06,
      "loss": 1.64,
      "num_input_tokens_seen": 31621616,
      "step": 5236,
      "train_runtime": 5301.9207,
      "train_tokens_per_second": 5964.181
    },
    {
      "epoch": 1.9261609195402298,
      "grad_norm": 0.9697108268737793,
      "learning_rate": 7.4360390207988225e-06,
      "loss": 1.4139,
      "num_input_tokens_seen": 31628096,
      "step": 5237,
      "train_runtime": 5302.9564,
      "train_tokens_per_second": 5964.238
    },
    {
      "epoch": 1.9265287356321839,
      "grad_norm": 1.0260202884674072,
      "learning_rate": 7.399226946438432e-06,
      "loss": 1.6004,
      "num_input_tokens_seen": 31634080,
      "step": 5238,
      "train_runtime": 5303.9133,
      "train_tokens_per_second": 5964.291
    },
    {
      "epoch": 1.926896551724138,
      "grad_norm": 0.9902703166007996,
      "learning_rate": 7.362414872078043e-06,
      "loss": 1.49,
      "num_input_tokens_seen": 31639872,
      "step": 5239,
      "train_runtime": 5304.8456,
      "train_tokens_per_second": 5964.334
    },
    {
      "epoch": 1.927264367816092,
      "grad_norm": 0.937769889831543,
      "learning_rate": 7.325602797717651e-06,
      "loss": 1.4061,
      "num_input_tokens_seen": 31648176,
      "step": 5240,
      "train_runtime": 5306.1174,
      "train_tokens_per_second": 5964.47
    },
    {
      "epoch": 1.9276321839080461,
      "grad_norm": 1.0571717023849487,
      "learning_rate": 7.2887907233572615e-06,
      "loss": 1.7377,
      "num_input_tokens_seen": 31653968,
      "step": 5241,
      "train_runtime": 5307.0924,
      "train_tokens_per_second": 5964.465
    },
    {
      "epoch": 1.928,
      "grad_norm": 1.071352481842041,
      "learning_rate": 7.251978648996871e-06,
      "loss": 1.6941,
      "num_input_tokens_seen": 31660288,
      "step": 5242,
      "train_runtime": 5308.1041,
      "train_tokens_per_second": 5964.519
    },
    {
      "epoch": 1.9283678160919542,
      "grad_norm": 1.121132493019104,
      "learning_rate": 7.215166574636481e-06,
      "loss": 1.5137,
      "num_input_tokens_seen": 31664544,
      "step": 5243,
      "train_runtime": 5308.8404,
      "train_tokens_per_second": 5964.494
    },
    {
      "epoch": 1.928735632183908,
      "grad_norm": 0.9838693141937256,
      "learning_rate": 7.178354500276091e-06,
      "loss": 1.7087,
      "num_input_tokens_seen": 31670320,
      "step": 5244,
      "train_runtime": 5309.7679,
      "train_tokens_per_second": 5964.539
    },
    {
      "epoch": 1.9291034482758622,
      "grad_norm": 0.9813634753227234,
      "learning_rate": 7.141542425915701e-06,
      "loss": 1.397,
      "num_input_tokens_seen": 31675680,
      "step": 5245,
      "train_runtime": 5310.6475,
      "train_tokens_per_second": 5964.561
    },
    {
      "epoch": 1.929471264367816,
      "grad_norm": 1.0224682092666626,
      "learning_rate": 7.104730351555311e-06,
      "loss": 1.5231,
      "num_input_tokens_seen": 31683088,
      "step": 5246,
      "train_runtime": 5311.8091,
      "train_tokens_per_second": 5964.651
    },
    {
      "epoch": 1.9298390804597703,
      "grad_norm": 0.9386256337165833,
      "learning_rate": 7.067918277194921e-06,
      "loss": 1.3935,
      "num_input_tokens_seen": 31688896,
      "step": 5247,
      "train_runtime": 5312.7325,
      "train_tokens_per_second": 5964.708
    },
    {
      "epoch": 1.930206896551724,
      "grad_norm": 0.9824444651603699,
      "learning_rate": 7.03110620283453e-06,
      "loss": 1.3528,
      "num_input_tokens_seen": 31695344,
      "step": 5248,
      "train_runtime": 5313.7785,
      "train_tokens_per_second": 5964.747
    },
    {
      "epoch": 1.9305747126436783,
      "grad_norm": 0.9770721793174744,
      "learning_rate": 6.994294128474139e-06,
      "loss": 1.473,
      "num_input_tokens_seen": 31701120,
      "step": 5249,
      "train_runtime": 5314.7302,
      "train_tokens_per_second": 5964.766
    },
    {
      "epoch": 1.930942528735632,
      "grad_norm": 1.138535976409912,
      "learning_rate": 6.957482054113749e-06,
      "loss": 1.4634,
      "num_input_tokens_seen": 31706480,
      "step": 5250,
      "train_runtime": 5315.6273,
      "train_tokens_per_second": 5964.767
    },
    {
      "epoch": 1.9313103448275863,
      "grad_norm": 1.027065396308899,
      "learning_rate": 6.92066997975336e-06,
      "loss": 1.5138,
      "num_input_tokens_seen": 31712736,
      "step": 5251,
      "train_runtime": 5317.2325,
      "train_tokens_per_second": 5964.143
    },
    {
      "epoch": 1.9316781609195401,
      "grad_norm": 1.0368155241012573,
      "learning_rate": 6.8838579053929694e-06,
      "loss": 1.4688,
      "num_input_tokens_seen": 31719072,
      "step": 5252,
      "train_runtime": 5318.2804,
      "train_tokens_per_second": 5964.159
    },
    {
      "epoch": 1.9320459770114944,
      "grad_norm": 1.1307930946350098,
      "learning_rate": 6.847045831032579e-06,
      "loss": 1.6219,
      "num_input_tokens_seen": 31724384,
      "step": 5253,
      "train_runtime": 5319.1655,
      "train_tokens_per_second": 5964.166
    },
    {
      "epoch": 1.9324137931034482,
      "grad_norm": 0.9835103154182434,
      "learning_rate": 6.81023375667219e-06,
      "loss": 1.5351,
      "num_input_tokens_seen": 31728976,
      "step": 5254,
      "train_runtime": 5319.9565,
      "train_tokens_per_second": 5964.142
    },
    {
      "epoch": 1.9327816091954024,
      "grad_norm": 0.9250811338424683,
      "learning_rate": 6.773421682311798e-06,
      "loss": 1.313,
      "num_input_tokens_seen": 31734976,
      "step": 5255,
      "train_runtime": 5320.9232,
      "train_tokens_per_second": 5964.186
    },
    {
      "epoch": 1.9331494252873562,
      "grad_norm": 1.0666043758392334,
      "learning_rate": 6.7366096079514075e-06,
      "loss": 1.5281,
      "num_input_tokens_seen": 31740400,
      "step": 5256,
      "train_runtime": 5321.8069,
      "train_tokens_per_second": 5964.215
    },
    {
      "epoch": 1.9335172413793105,
      "grad_norm": 1.0790902376174927,
      "learning_rate": 6.699797533591018e-06,
      "loss": 1.6425,
      "num_input_tokens_seen": 31746528,
      "step": 5257,
      "train_runtime": 5322.8096,
      "train_tokens_per_second": 5964.243
    },
    {
      "epoch": 1.9338850574712643,
      "grad_norm": 1.0127862691879272,
      "learning_rate": 6.662985459230628e-06,
      "loss": 1.4918,
      "num_input_tokens_seen": 31750864,
      "step": 5258,
      "train_runtime": 5323.5557,
      "train_tokens_per_second": 5964.221
    },
    {
      "epoch": 1.9342528735632185,
      "grad_norm": 1.0982059240341187,
      "learning_rate": 6.626173384870238e-06,
      "loss": 1.5349,
      "num_input_tokens_seen": 31757632,
      "step": 5259,
      "train_runtime": 5324.6389,
      "train_tokens_per_second": 5964.279
    },
    {
      "epoch": 1.9346206896551723,
      "grad_norm": 1.0924607515335083,
      "learning_rate": 6.589361310509848e-06,
      "loss": 1.4036,
      "num_input_tokens_seen": 31765440,
      "step": 5260,
      "train_runtime": 5325.8407,
      "train_tokens_per_second": 5964.399
    },
    {
      "epoch": 1.9349885057471266,
      "grad_norm": 1.1054469347000122,
      "learning_rate": 6.552549236149458e-06,
      "loss": 1.658,
      "num_input_tokens_seen": 31769776,
      "step": 5261,
      "train_runtime": 5326.5955,
      "train_tokens_per_second": 5964.368
    },
    {
      "epoch": 1.9353563218390804,
      "grad_norm": 0.9382045269012451,
      "learning_rate": 6.515737161789068e-06,
      "loss": 1.4855,
      "num_input_tokens_seen": 31774304,
      "step": 5262,
      "train_runtime": 5327.4098,
      "train_tokens_per_second": 5964.306
    },
    {
      "epoch": 1.9357241379310346,
      "grad_norm": 0.95528644323349,
      "learning_rate": 6.4789250874286766e-06,
      "loss": 1.2951,
      "num_input_tokens_seen": 31780224,
      "step": 5263,
      "train_runtime": 5328.3698,
      "train_tokens_per_second": 5964.343
    },
    {
      "epoch": 1.9360919540229884,
      "grad_norm": 0.9378964304924011,
      "learning_rate": 6.442113013068286e-06,
      "loss": 1.4056,
      "num_input_tokens_seen": 31784480,
      "step": 5264,
      "train_runtime": 5329.1146,
      "train_tokens_per_second": 5964.308
    },
    {
      "epoch": 1.9364597701149426,
      "grad_norm": 1.047817587852478,
      "learning_rate": 6.405300938707896e-06,
      "loss": 1.3752,
      "num_input_tokens_seen": 31789520,
      "step": 5265,
      "train_runtime": 5329.9458,
      "train_tokens_per_second": 5964.323
    },
    {
      "epoch": 1.9368275862068964,
      "grad_norm": 1.0981109142303467,
      "learning_rate": 6.368488864347507e-06,
      "loss": 1.5143,
      "num_input_tokens_seen": 31793520,
      "step": 5266,
      "train_runtime": 5330.6386,
      "train_tokens_per_second": 5964.298
    },
    {
      "epoch": 1.9371954022988507,
      "grad_norm": 0.9980946779251099,
      "learning_rate": 6.331676789987116e-06,
      "loss": 1.2403,
      "num_input_tokens_seen": 31798512,
      "step": 5267,
      "train_runtime": 5331.4758,
      "train_tokens_per_second": 5964.298
    },
    {
      "epoch": 1.9375632183908045,
      "grad_norm": 0.9951990842819214,
      "learning_rate": 6.294864715626726e-06,
      "loss": 1.4614,
      "num_input_tokens_seen": 31804736,
      "step": 5268,
      "train_runtime": 5332.508,
      "train_tokens_per_second": 5964.311
    },
    {
      "epoch": 1.9379310344827587,
      "grad_norm": 1.0398674011230469,
      "learning_rate": 6.258052641266336e-06,
      "loss": 1.2538,
      "num_input_tokens_seen": 31813264,
      "step": 5269,
      "train_runtime": 5333.8399,
      "train_tokens_per_second": 5964.42
    },
    {
      "epoch": 1.9382988505747125,
      "grad_norm": 0.9988831281661987,
      "learning_rate": 6.221240566905946e-06,
      "loss": 1.4653,
      "num_input_tokens_seen": 31817824,
      "step": 5270,
      "train_runtime": 5334.627,
      "train_tokens_per_second": 5964.395
    },
    {
      "epoch": 1.9386666666666668,
      "grad_norm": 1.0581578016281128,
      "learning_rate": 6.184428492545555e-06,
      "loss": 1.7029,
      "num_input_tokens_seen": 31822528,
      "step": 5271,
      "train_runtime": 5335.446,
      "train_tokens_per_second": 5964.361
    },
    {
      "epoch": 1.9390344827586206,
      "grad_norm": 0.9771072864532471,
      "learning_rate": 6.147616418185165e-06,
      "loss": 1.2942,
      "num_input_tokens_seen": 31830192,
      "step": 5272,
      "train_runtime": 5336.665,
      "train_tokens_per_second": 5964.435
    },
    {
      "epoch": 1.9394022988505748,
      "grad_norm": 1.0883715152740479,
      "learning_rate": 6.110804343824775e-06,
      "loss": 1.4369,
      "num_input_tokens_seen": 31839120,
      "step": 5273,
      "train_runtime": 5338.0342,
      "train_tokens_per_second": 5964.578
    },
    {
      "epoch": 1.9397701149425286,
      "grad_norm": 1.0530915260314941,
      "learning_rate": 6.0739922694643846e-06,
      "loss": 1.4179,
      "num_input_tokens_seen": 31845488,
      "step": 5274,
      "train_runtime": 5339.0807,
      "train_tokens_per_second": 5964.601
    },
    {
      "epoch": 1.9401379310344828,
      "grad_norm": 0.9308196306228638,
      "learning_rate": 6.037180195103994e-06,
      "loss": 1.4165,
      "num_input_tokens_seen": 31853248,
      "step": 5275,
      "train_runtime": 5340.2969,
      "train_tokens_per_second": 5964.696
    },
    {
      "epoch": 1.9405057471264366,
      "grad_norm": 1.0790725946426392,
      "learning_rate": 6.000368120743604e-06,
      "loss": 1.4623,
      "num_input_tokens_seen": 31860864,
      "step": 5276,
      "train_runtime": 5341.4885,
      "train_tokens_per_second": 5964.791
    },
    {
      "epoch": 1.9408735632183909,
      "grad_norm": 1.0767091512680054,
      "learning_rate": 5.963556046383214e-06,
      "loss": 1.409,
      "num_input_tokens_seen": 31865472,
      "step": 5277,
      "train_runtime": 5342.2757,
      "train_tokens_per_second": 5964.775
    },
    {
      "epoch": 1.9412413793103447,
      "grad_norm": 0.9972087740898132,
      "learning_rate": 5.9267439720228235e-06,
      "loss": 1.392,
      "num_input_tokens_seen": 31870384,
      "step": 5278,
      "train_runtime": 5343.111,
      "train_tokens_per_second": 5964.762
    },
    {
      "epoch": 1.941609195402299,
      "grad_norm": 1.0063000917434692,
      "learning_rate": 5.889931897662434e-06,
      "loss": 1.3924,
      "num_input_tokens_seen": 31875392,
      "step": 5279,
      "train_runtime": 5343.9673,
      "train_tokens_per_second": 5964.743
    },
    {
      "epoch": 1.9419770114942527,
      "grad_norm": 1.0639219284057617,
      "learning_rate": 5.853119823302043e-06,
      "loss": 1.5586,
      "num_input_tokens_seen": 31880208,
      "step": 5280,
      "train_runtime": 5344.7643,
      "train_tokens_per_second": 5964.755
    },
    {
      "epoch": 1.942344827586207,
      "grad_norm": 1.0319242477416992,
      "learning_rate": 5.816307748941653e-06,
      "loss": 1.3954,
      "num_input_tokens_seen": 31887952,
      "step": 5281,
      "train_runtime": 5346.5403,
      "train_tokens_per_second": 5964.222
    },
    {
      "epoch": 1.9427126436781608,
      "grad_norm": 1.0243401527404785,
      "learning_rate": 5.779495674581263e-06,
      "loss": 1.4722,
      "num_input_tokens_seen": 31895024,
      "step": 5282,
      "train_runtime": 5347.6663,
      "train_tokens_per_second": 5964.288
    },
    {
      "epoch": 1.943080459770115,
      "grad_norm": 0.9136255383491516,
      "learning_rate": 5.742683600220872e-06,
      "loss": 1.4453,
      "num_input_tokens_seen": 31901216,
      "step": 5283,
      "train_runtime": 5348.6577,
      "train_tokens_per_second": 5964.341
    },
    {
      "epoch": 1.9434482758620688,
      "grad_norm": 1.0555051565170288,
      "learning_rate": 5.705871525860483e-06,
      "loss": 1.5629,
      "num_input_tokens_seen": 31907056,
      "step": 5284,
      "train_runtime": 5349.6089,
      "train_tokens_per_second": 5964.372
    },
    {
      "epoch": 1.943816091954023,
      "grad_norm": 0.9248809218406677,
      "learning_rate": 5.6690594515000925e-06,
      "loss": 1.313,
      "num_input_tokens_seen": 31913520,
      "step": 5285,
      "train_runtime": 5350.6509,
      "train_tokens_per_second": 5964.418
    },
    {
      "epoch": 1.9441839080459768,
      "grad_norm": 0.9594218134880066,
      "learning_rate": 5.632247377139702e-06,
      "loss": 1.4998,
      "num_input_tokens_seen": 31920576,
      "step": 5286,
      "train_runtime": 5351.7527,
      "train_tokens_per_second": 5964.509
    },
    {
      "epoch": 1.944551724137931,
      "grad_norm": 0.8581014275550842,
      "learning_rate": 5.595435302779312e-06,
      "loss": 1.3114,
      "num_input_tokens_seen": 31928128,
      "step": 5287,
      "train_runtime": 5352.9601,
      "train_tokens_per_second": 5964.574
    },
    {
      "epoch": 1.944919540229885,
      "grad_norm": 1.2395435571670532,
      "learning_rate": 5.558623228418922e-06,
      "loss": 1.6761,
      "num_input_tokens_seen": 31932400,
      "step": 5288,
      "train_runtime": 5353.7024,
      "train_tokens_per_second": 5964.545
    },
    {
      "epoch": 1.9452873563218391,
      "grad_norm": 1.0082666873931885,
      "learning_rate": 5.5218111540585315e-06,
      "loss": 1.556,
      "num_input_tokens_seen": 31940480,
      "step": 5289,
      "train_runtime": 5354.9502,
      "train_tokens_per_second": 5964.664
    },
    {
      "epoch": 1.9456551724137932,
      "grad_norm": 1.0668951272964478,
      "learning_rate": 5.484999079698141e-06,
      "loss": 1.5841,
      "num_input_tokens_seen": 31950640,
      "step": 5290,
      "train_runtime": 5356.5053,
      "train_tokens_per_second": 5964.829
    },
    {
      "epoch": 1.9460229885057472,
      "grad_norm": 1.0597999095916748,
      "learning_rate": 5.448187005337751e-06,
      "loss": 1.3845,
      "num_input_tokens_seen": 31955376,
      "step": 5291,
      "train_runtime": 5357.3224,
      "train_tokens_per_second": 5964.804
    },
    {
      "epoch": 1.9463908045977012,
      "grad_norm": 1.0465960502624512,
      "learning_rate": 5.411374930977361e-06,
      "loss": 1.4331,
      "num_input_tokens_seen": 31959424,
      "step": 5292,
      "train_runtime": 5358.019,
      "train_tokens_per_second": 5964.784
    },
    {
      "epoch": 1.9467586206896552,
      "grad_norm": 1.035742163658142,
      "learning_rate": 5.3745628566169704e-06,
      "loss": 1.5636,
      "num_input_tokens_seen": 31966240,
      "step": 5293,
      "train_runtime": 5359.1134,
      "train_tokens_per_second": 5964.837
    },
    {
      "epoch": 1.9471264367816092,
      "grad_norm": 1.0233871936798096,
      "learning_rate": 5.337750782256581e-06,
      "loss": 1.5166,
      "num_input_tokens_seen": 31972032,
      "step": 5294,
      "train_runtime": 5360.0589,
      "train_tokens_per_second": 5964.866
    },
    {
      "epoch": 1.9474942528735633,
      "grad_norm": 1.0100815296173096,
      "learning_rate": 5.30093870789619e-06,
      "loss": 1.4645,
      "num_input_tokens_seen": 31978544,
      "step": 5295,
      "train_runtime": 5361.0804,
      "train_tokens_per_second": 5964.944
    },
    {
      "epoch": 1.9478620689655173,
      "grad_norm": 0.9109177589416504,
      "learning_rate": 5.2641266335358e-06,
      "loss": 1.3745,
      "num_input_tokens_seen": 31984224,
      "step": 5296,
      "train_runtime": 5362.008,
      "train_tokens_per_second": 5964.971
    },
    {
      "epoch": 1.9482298850574713,
      "grad_norm": 1.0548319816589355,
      "learning_rate": 5.22731455917541e-06,
      "loss": 1.4834,
      "num_input_tokens_seen": 31988656,
      "step": 5297,
      "train_runtime": 5362.7867,
      "train_tokens_per_second": 5964.932
    },
    {
      "epoch": 1.9485977011494253,
      "grad_norm": 1.040931224822998,
      "learning_rate": 5.190502484815019e-06,
      "loss": 1.5954,
      "num_input_tokens_seen": 31993616,
      "step": 5298,
      "train_runtime": 5363.6256,
      "train_tokens_per_second": 5964.923
    },
    {
      "epoch": 1.9489655172413793,
      "grad_norm": 0.9623754620552063,
      "learning_rate": 5.153690410454629e-06,
      "loss": 1.4927,
      "num_input_tokens_seen": 32002272,
      "step": 5299,
      "train_runtime": 5364.9954,
      "train_tokens_per_second": 5965.014
    },
    {
      "epoch": 1.9493333333333334,
      "grad_norm": 1.096806526184082,
      "learning_rate": 5.1168783360942395e-06,
      "loss": 1.667,
      "num_input_tokens_seen": 32006736,
      "step": 5300,
      "train_runtime": 5365.7505,
      "train_tokens_per_second": 5965.006
    },
    {
      "epoch": 1.9497011494252874,
      "grad_norm": 0.9942008256912231,
      "learning_rate": 5.080066261733849e-06,
      "loss": 1.4233,
      "num_input_tokens_seen": 32011440,
      "step": 5301,
      "train_runtime": 5366.5594,
      "train_tokens_per_second": 5964.984
    },
    {
      "epoch": 1.9500689655172414,
      "grad_norm": 1.0310285091400146,
      "learning_rate": 5.043254187373458e-06,
      "loss": 1.5672,
      "num_input_tokens_seen": 32015680,
      "step": 5302,
      "train_runtime": 5367.3184,
      "train_tokens_per_second": 5964.93
    },
    {
      "epoch": 1.9504367816091954,
      "grad_norm": 1.0258104801177979,
      "learning_rate": 5.006442113013069e-06,
      "loss": 1.5451,
      "num_input_tokens_seen": 32021808,
      "step": 5303,
      "train_runtime": 5368.3281,
      "train_tokens_per_second": 5964.95
    },
    {
      "epoch": 1.9508045977011494,
      "grad_norm": 1.0516815185546875,
      "learning_rate": 4.969630038652678e-06,
      "loss": 1.6127,
      "num_input_tokens_seen": 32026464,
      "step": 5304,
      "train_runtime": 5369.1231,
      "train_tokens_per_second": 5964.934
    },
    {
      "epoch": 1.9511724137931035,
      "grad_norm": 1.090609073638916,
      "learning_rate": 4.932817964292288e-06,
      "loss": 1.7142,
      "num_input_tokens_seen": 32031472,
      "step": 5305,
      "train_runtime": 5369.9618,
      "train_tokens_per_second": 5964.935
    },
    {
      "epoch": 1.9515402298850575,
      "grad_norm": 0.8959497809410095,
      "learning_rate": 4.896005889931898e-06,
      "loss": 1.3462,
      "num_input_tokens_seen": 32036928,
      "step": 5306,
      "train_runtime": 5370.8535,
      "train_tokens_per_second": 5964.96
    },
    {
      "epoch": 1.9519080459770115,
      "grad_norm": 0.9927746653556824,
      "learning_rate": 4.859193815571508e-06,
      "loss": 1.5212,
      "num_input_tokens_seen": 32041824,
      "step": 5307,
      "train_runtime": 5371.6925,
      "train_tokens_per_second": 5964.94
    },
    {
      "epoch": 1.9522758620689655,
      "grad_norm": 0.9697750210762024,
      "learning_rate": 4.822381741211117e-06,
      "loss": 1.4346,
      "num_input_tokens_seen": 32048928,
      "step": 5308,
      "train_runtime": 5372.8131,
      "train_tokens_per_second": 5965.018
    },
    {
      "epoch": 1.9526436781609195,
      "grad_norm": 1.0766029357910156,
      "learning_rate": 4.785569666850728e-06,
      "loss": 1.4846,
      "num_input_tokens_seen": 32054736,
      "step": 5309,
      "train_runtime": 5373.7493,
      "train_tokens_per_second": 5965.06
    },
    {
      "epoch": 1.9530114942528736,
      "grad_norm": 0.9555527567863464,
      "learning_rate": 4.748757592490337e-06,
      "loss": 1.5427,
      "num_input_tokens_seen": 32061536,
      "step": 5310,
      "train_runtime": 5374.8473,
      "train_tokens_per_second": 5965.106
    },
    {
      "epoch": 1.9533793103448276,
      "grad_norm": 1.0309621095657349,
      "learning_rate": 4.711945518129947e-06,
      "loss": 1.5572,
      "num_input_tokens_seen": 32066240,
      "step": 5311,
      "train_runtime": 5376.2125,
      "train_tokens_per_second": 5964.467
    },
    {
      "epoch": 1.9537471264367816,
      "grad_norm": 0.986017107963562,
      "learning_rate": 4.675133443769557e-06,
      "loss": 1.4435,
      "num_input_tokens_seen": 32073840,
      "step": 5312,
      "train_runtime": 5377.4049,
      "train_tokens_per_second": 5964.557
    },
    {
      "epoch": 1.9541149425287356,
      "grad_norm": 1.0569292306900024,
      "learning_rate": 4.638321369409166e-06,
      "loss": 1.7579,
      "num_input_tokens_seen": 32078640,
      "step": 5313,
      "train_runtime": 5378.236,
      "train_tokens_per_second": 5964.528
    },
    {
      "epoch": 1.9544827586206897,
      "grad_norm": 0.9573275446891785,
      "learning_rate": 4.601509295048776e-06,
      "loss": 1.4357,
      "num_input_tokens_seen": 32082832,
      "step": 5314,
      "train_runtime": 5378.9793,
      "train_tokens_per_second": 5964.483
    },
    {
      "epoch": 1.9548505747126437,
      "grad_norm": 1.1017218828201294,
      "learning_rate": 4.564697220688386e-06,
      "loss": 1.7869,
      "num_input_tokens_seen": 32087696,
      "step": 5315,
      "train_runtime": 5379.8218,
      "train_tokens_per_second": 5964.453
    },
    {
      "epoch": 1.9552183908045977,
      "grad_norm": 0.9713738560676575,
      "learning_rate": 4.527885146327996e-06,
      "loss": 1.6279,
      "num_input_tokens_seen": 32095840,
      "step": 5316,
      "train_runtime": 5381.1109,
      "train_tokens_per_second": 5964.538
    },
    {
      "epoch": 1.9555862068965517,
      "grad_norm": 1.0966476202011108,
      "learning_rate": 4.491073071967605e-06,
      "loss": 1.4297,
      "num_input_tokens_seen": 32101472,
      "step": 5317,
      "train_runtime": 5382.0202,
      "train_tokens_per_second": 5964.577
    },
    {
      "epoch": 1.9559540229885057,
      "grad_norm": 0.9543721675872803,
      "learning_rate": 4.454260997607216e-06,
      "loss": 1.3416,
      "num_input_tokens_seen": 32108688,
      "step": 5318,
      "train_runtime": 5383.1491,
      "train_tokens_per_second": 5964.666
    },
    {
      "epoch": 1.9563218390804598,
      "grad_norm": 1.0037769079208374,
      "learning_rate": 4.417448923246825e-06,
      "loss": 1.5395,
      "num_input_tokens_seen": 32115984,
      "step": 5319,
      "train_runtime": 5384.3019,
      "train_tokens_per_second": 5964.744
    },
    {
      "epoch": 1.9566896551724138,
      "grad_norm": 0.9767951369285583,
      "learning_rate": 4.380636848886435e-06,
      "loss": 1.4984,
      "num_input_tokens_seen": 32121456,
      "step": 5320,
      "train_runtime": 5385.2202,
      "train_tokens_per_second": 5964.743
    },
    {
      "epoch": 1.9570574712643678,
      "grad_norm": 1.0521577596664429,
      "learning_rate": 4.343824774526045e-06,
      "loss": 1.6567,
      "num_input_tokens_seen": 32128112,
      "step": 5321,
      "train_runtime": 5386.3056,
      "train_tokens_per_second": 5964.777
    },
    {
      "epoch": 1.9574252873563218,
      "grad_norm": 1.0742543935775757,
      "learning_rate": 4.307012700165655e-06,
      "loss": 1.4182,
      "num_input_tokens_seen": 32133200,
      "step": 5322,
      "train_runtime": 5387.1552,
      "train_tokens_per_second": 5964.781
    },
    {
      "epoch": 1.9577931034482758,
      "grad_norm": 0.9935344457626343,
      "learning_rate": 4.270200625805264e-06,
      "loss": 1.3829,
      "num_input_tokens_seen": 32140416,
      "step": 5323,
      "train_runtime": 5388.2863,
      "train_tokens_per_second": 5964.868
    },
    {
      "epoch": 1.9581609195402299,
      "grad_norm": 1.0693191289901733,
      "learning_rate": 4.233388551444874e-06,
      "loss": 1.4094,
      "num_input_tokens_seen": 32145296,
      "step": 5324,
      "train_runtime": 5389.1319,
      "train_tokens_per_second": 5964.837
    },
    {
      "epoch": 1.9585287356321839,
      "grad_norm": 1.1128088235855103,
      "learning_rate": 4.196576477084484e-06,
      "loss": 1.5803,
      "num_input_tokens_seen": 32149504,
      "step": 5325,
      "train_runtime": 5389.8676,
      "train_tokens_per_second": 5964.804
    },
    {
      "epoch": 1.958896551724138,
      "grad_norm": 1.1361160278320312,
      "learning_rate": 4.1597644027240935e-06,
      "loss": 1.6932,
      "num_input_tokens_seen": 32154224,
      "step": 5326,
      "train_runtime": 5390.6935,
      "train_tokens_per_second": 5964.766
    },
    {
      "epoch": 1.959264367816092,
      "grad_norm": 1.0320876836776733,
      "learning_rate": 4.122952328363703e-06,
      "loss": 1.4487,
      "num_input_tokens_seen": 32166464,
      "step": 5327,
      "train_runtime": 5392.4932,
      "train_tokens_per_second": 5965.045
    },
    {
      "epoch": 1.959632183908046,
      "grad_norm": 1.0516407489776611,
      "learning_rate": 4.086140254003314e-06,
      "loss": 1.693,
      "num_input_tokens_seen": 32172288,
      "step": 5328,
      "train_runtime": 5393.4447,
      "train_tokens_per_second": 5965.072
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.9320648312568665,
      "learning_rate": 4.049328179642923e-06,
      "loss": 1.3479,
      "num_input_tokens_seen": 32181808,
      "step": 5329,
      "train_runtime": 5394.8944,
      "train_tokens_per_second": 5965.234
    },
    {
      "epoch": 1.960367816091954,
      "grad_norm": 0.9066106677055359,
      "learning_rate": 4.0125161052825325e-06,
      "loss": 1.3445,
      "num_input_tokens_seen": 32188576,
      "step": 5330,
      "train_runtime": 5396.0185,
      "train_tokens_per_second": 5965.246
    },
    {
      "epoch": 1.960735632183908,
      "grad_norm": 0.9577798247337341,
      "learning_rate": 3.975704030922143e-06,
      "loss": 1.4834,
      "num_input_tokens_seen": 32203488,
      "step": 5331,
      "train_runtime": 5398.1948,
      "train_tokens_per_second": 5965.603
    },
    {
      "epoch": 1.961103448275862,
      "grad_norm": 0.9157636165618896,
      "learning_rate": 3.938891956561752e-06,
      "loss": 1.2181,
      "num_input_tokens_seen": 32213120,
      "step": 5332,
      "train_runtime": 5399.6511,
      "train_tokens_per_second": 5965.778
    },
    {
      "epoch": 1.961471264367816,
      "grad_norm": 1.0202064514160156,
      "learning_rate": 3.902079882201362e-06,
      "loss": 1.619,
      "num_input_tokens_seen": 32218208,
      "step": 5333,
      "train_runtime": 5400.4967,
      "train_tokens_per_second": 5965.786
    },
    {
      "epoch": 1.96183908045977,
      "grad_norm": 0.9759100675582886,
      "learning_rate": 3.865267807840972e-06,
      "loss": 1.5136,
      "num_input_tokens_seen": 32226928,
      "step": 5334,
      "train_runtime": 5401.8401,
      "train_tokens_per_second": 5965.917
    },
    {
      "epoch": 1.962206896551724,
      "grad_norm": 1.0097498893737793,
      "learning_rate": 3.828455733480582e-06,
      "loss": 1.615,
      "num_input_tokens_seen": 32232112,
      "step": 5335,
      "train_runtime": 5402.7104,
      "train_tokens_per_second": 5965.915
    },
    {
      "epoch": 1.962574712643678,
      "grad_norm": 0.9370710849761963,
      "learning_rate": 3.7916436591201913e-06,
      "loss": 1.2833,
      "num_input_tokens_seen": 32239552,
      "step": 5336,
      "train_runtime": 5403.8691,
      "train_tokens_per_second": 5966.013
    },
    {
      "epoch": 1.9629425287356321,
      "grad_norm": 0.9166112542152405,
      "learning_rate": 3.7548315847598015e-06,
      "loss": 1.4053,
      "num_input_tokens_seen": 32245984,
      "step": 5337,
      "train_runtime": 5404.9143,
      "train_tokens_per_second": 5966.049
    },
    {
      "epoch": 1.9633103448275864,
      "grad_norm": 1.098130226135254,
      "learning_rate": 3.7180195103994113e-06,
      "loss": 1.6172,
      "num_input_tokens_seen": 32252976,
      "step": 5338,
      "train_runtime": 5406.0157,
      "train_tokens_per_second": 5966.127
    },
    {
      "epoch": 1.9636781609195402,
      "grad_norm": 1.1313691139221191,
      "learning_rate": 3.6812074360390214e-06,
      "loss": 1.604,
      "num_input_tokens_seen": 32259328,
      "step": 5339,
      "train_runtime": 5407.0206,
      "train_tokens_per_second": 5966.193
    },
    {
      "epoch": 1.9640459770114944,
      "grad_norm": 0.8854979872703552,
      "learning_rate": 3.6443953616786307e-06,
      "loss": 1.3713,
      "num_input_tokens_seen": 32265920,
      "step": 5340,
      "train_runtime": 5408.0631,
      "train_tokens_per_second": 5966.262
    },
    {
      "epoch": 1.9644137931034482,
      "grad_norm": 1.0324574708938599,
      "learning_rate": 3.6075832873182405e-06,
      "loss": 1.4795,
      "num_input_tokens_seen": 32271088,
      "step": 5341,
      "train_runtime": 5409.4578,
      "train_tokens_per_second": 5965.679
    },
    {
      "epoch": 1.9647816091954025,
      "grad_norm": 0.948697566986084,
      "learning_rate": 3.5707712129578506e-06,
      "loss": 1.4571,
      "num_input_tokens_seen": 32279936,
      "step": 5342,
      "train_runtime": 5410.8085,
      "train_tokens_per_second": 5965.825
    },
    {
      "epoch": 1.9651494252873563,
      "grad_norm": 0.945921778678894,
      "learning_rate": 3.5339591385974604e-06,
      "loss": 1.3578,
      "num_input_tokens_seen": 32286848,
      "step": 5343,
      "train_runtime": 5411.8957,
      "train_tokens_per_second": 5965.904
    },
    {
      "epoch": 1.9655172413793105,
      "grad_norm": 1.049149990081787,
      "learning_rate": 3.4971470642370697e-06,
      "loss": 1.3977,
      "num_input_tokens_seen": 32292528,
      "step": 5344,
      "train_runtime": 5412.8039,
      "train_tokens_per_second": 5965.952
    },
    {
      "epoch": 1.9658850574712643,
      "grad_norm": 1.106776237487793,
      "learning_rate": 3.46033498987668e-06,
      "loss": 1.7396,
      "num_input_tokens_seen": 32298688,
      "step": 5345,
      "train_runtime": 5413.8333,
      "train_tokens_per_second": 5965.955
    },
    {
      "epoch": 1.9662528735632185,
      "grad_norm": 1.051772117614746,
      "learning_rate": 3.4235229155162896e-06,
      "loss": 1.5489,
      "num_input_tokens_seen": 32303648,
      "step": 5346,
      "train_runtime": 5414.6424,
      "train_tokens_per_second": 5965.98
    },
    {
      "epoch": 1.9666206896551723,
      "grad_norm": 0.9481874108314514,
      "learning_rate": 3.386710841155899e-06,
      "loss": 1.5319,
      "num_input_tokens_seen": 32308880,
      "step": 5347,
      "train_runtime": 5415.5453,
      "train_tokens_per_second": 5965.951
    },
    {
      "epoch": 1.9669885057471266,
      "grad_norm": 0.8972981572151184,
      "learning_rate": 3.349898766795509e-06,
      "loss": 1.3542,
      "num_input_tokens_seen": 32318624,
      "step": 5348,
      "train_runtime": 5417.0419,
      "train_tokens_per_second": 5966.102
    },
    {
      "epoch": 1.9673563218390804,
      "grad_norm": 1.0520169734954834,
      "learning_rate": 3.313086692435119e-06,
      "loss": 1.5498,
      "num_input_tokens_seen": 32323728,
      "step": 5349,
      "train_runtime": 5417.8956,
      "train_tokens_per_second": 5966.104
    },
    {
      "epoch": 1.9677241379310346,
      "grad_norm": 0.9874334931373596,
      "learning_rate": 3.276274618074729e-06,
      "loss": 1.4794,
      "num_input_tokens_seen": 32329168,
      "step": 5350,
      "train_runtime": 5418.8042,
      "train_tokens_per_second": 5966.107
    },
    {
      "epoch": 1.9680919540229884,
      "grad_norm": 1.0940510034561157,
      "learning_rate": 3.2394625437143383e-06,
      "loss": 1.4545,
      "num_input_tokens_seen": 32335024,
      "step": 5351,
      "train_runtime": 5419.78,
      "train_tokens_per_second": 5966.114
    },
    {
      "epoch": 1.9684597701149427,
      "grad_norm": 1.0374449491500854,
      "learning_rate": 3.202650469353948e-06,
      "loss": 1.641,
      "num_input_tokens_seen": 32339888,
      "step": 5352,
      "train_runtime": 5420.6193,
      "train_tokens_per_second": 5966.087
    },
    {
      "epoch": 1.9688275862068965,
      "grad_norm": 1.0712896585464478,
      "learning_rate": 3.165838394993558e-06,
      "loss": 1.3483,
      "num_input_tokens_seen": 32345616,
      "step": 5353,
      "train_runtime": 5421.584,
      "train_tokens_per_second": 5966.082
    },
    {
      "epoch": 1.9691954022988507,
      "grad_norm": 0.9663326144218445,
      "learning_rate": 3.129026320633168e-06,
      "loss": 1.3515,
      "num_input_tokens_seen": 32353664,
      "step": 5354,
      "train_runtime": 5422.7933,
      "train_tokens_per_second": 5966.236
    },
    {
      "epoch": 1.9695632183908045,
      "grad_norm": 1.0441182851791382,
      "learning_rate": 3.0922142462727777e-06,
      "loss": 1.6059,
      "num_input_tokens_seen": 32359360,
      "step": 5355,
      "train_runtime": 5423.7113,
      "train_tokens_per_second": 5966.276
    },
    {
      "epoch": 1.9699310344827587,
      "grad_norm": 0.9823611974716187,
      "learning_rate": 3.0554021719123874e-06,
      "loss": 1.3778,
      "num_input_tokens_seen": 32365040,
      "step": 5356,
      "train_runtime": 5424.6475,
      "train_tokens_per_second": 5966.294
    },
    {
      "epoch": 1.9702988505747125,
      "grad_norm": 1.0174914598464966,
      "learning_rate": 3.018590097551997e-06,
      "loss": 1.4918,
      "num_input_tokens_seen": 32370224,
      "step": 5357,
      "train_runtime": 5425.5185,
      "train_tokens_per_second": 5966.291
    },
    {
      "epoch": 1.9706666666666668,
      "grad_norm": 1.0235735177993774,
      "learning_rate": 2.981778023191607e-06,
      "loss": 1.4855,
      "num_input_tokens_seen": 32375056,
      "step": 5358,
      "train_runtime": 5426.3197,
      "train_tokens_per_second": 5966.301
    },
    {
      "epoch": 1.9710344827586206,
      "grad_norm": 1.0242557525634766,
      "learning_rate": 2.944965948831217e-06,
      "loss": 1.4203,
      "num_input_tokens_seen": 32379888,
      "step": 5359,
      "train_runtime": 5427.1579,
      "train_tokens_per_second": 5966.27
    },
    {
      "epoch": 1.9714022988505748,
      "grad_norm": 1.0372973680496216,
      "learning_rate": 2.9081538744708264e-06,
      "loss": 1.397,
      "num_input_tokens_seen": 32384272,
      "step": 5360,
      "train_runtime": 5427.9331,
      "train_tokens_per_second": 5966.225
    },
    {
      "epoch": 1.9717701149425286,
      "grad_norm": 0.9870122075080872,
      "learning_rate": 2.871341800110436e-06,
      "loss": 1.3466,
      "num_input_tokens_seen": 32389568,
      "step": 5361,
      "train_runtime": 5428.8289,
      "train_tokens_per_second": 5966.216
    },
    {
      "epoch": 1.9721379310344829,
      "grad_norm": 0.935351550579071,
      "learning_rate": 2.8345297257500463e-06,
      "loss": 1.4078,
      "num_input_tokens_seen": 32396880,
      "step": 5362,
      "train_runtime": 5429.9827,
      "train_tokens_per_second": 5966.295
    },
    {
      "epoch": 1.9725057471264367,
      "grad_norm": 1.0815507173538208,
      "learning_rate": 2.797717651389656e-06,
      "loss": 1.5521,
      "num_input_tokens_seen": 32402464,
      "step": 5363,
      "train_runtime": 5430.9015,
      "train_tokens_per_second": 5966.314
    },
    {
      "epoch": 1.972873563218391,
      "grad_norm": 0.942524254322052,
      "learning_rate": 2.7609055770292657e-06,
      "loss": 1.3315,
      "num_input_tokens_seen": 32412512,
      "step": 5364,
      "train_runtime": 5432.4128,
      "train_tokens_per_second": 5966.504
    },
    {
      "epoch": 1.9732413793103447,
      "grad_norm": 0.9533636569976807,
      "learning_rate": 2.7240935026688755e-06,
      "loss": 1.2837,
      "num_input_tokens_seen": 32417232,
      "step": 5365,
      "train_runtime": 5433.2341,
      "train_tokens_per_second": 5966.471
    },
    {
      "epoch": 1.973609195402299,
      "grad_norm": 0.9564442038536072,
      "learning_rate": 2.6872814283084852e-06,
      "loss": 1.4282,
      "num_input_tokens_seen": 32425984,
      "step": 5366,
      "train_runtime": 5434.5632,
      "train_tokens_per_second": 5966.622
    },
    {
      "epoch": 1.9739770114942528,
      "grad_norm": 1.1184884309768677,
      "learning_rate": 2.650469353948095e-06,
      "loss": 1.4477,
      "num_input_tokens_seen": 32436000,
      "step": 5367,
      "train_runtime": 5436.0869,
      "train_tokens_per_second": 5966.792
    },
    {
      "epoch": 1.974344827586207,
      "grad_norm": 0.9816742539405823,
      "learning_rate": 2.613657279587705e-06,
      "loss": 1.3507,
      "num_input_tokens_seen": 32443200,
      "step": 5368,
      "train_runtime": 5437.2264,
      "train_tokens_per_second": 5966.866
    },
    {
      "epoch": 1.9747126436781608,
      "grad_norm": 1.0354766845703125,
      "learning_rate": 2.5768452052273144e-06,
      "loss": 1.5121,
      "num_input_tokens_seen": 32448656,
      "step": 5369,
      "train_runtime": 5438.1423,
      "train_tokens_per_second": 5966.864
    },
    {
      "epoch": 1.975080459770115,
      "grad_norm": 0.990043044090271,
      "learning_rate": 2.5400331308669246e-06,
      "loss": 1.5338,
      "num_input_tokens_seen": 32454816,
      "step": 5370,
      "train_runtime": 5439.1369,
      "train_tokens_per_second": 5966.906
    },
    {
      "epoch": 1.9754482758620688,
      "grad_norm": 0.9449595212936401,
      "learning_rate": 2.5032210565065343e-06,
      "loss": 1.469,
      "num_input_tokens_seen": 32460880,
      "step": 5371,
      "train_runtime": 5440.6079,
      "train_tokens_per_second": 5966.407
    },
    {
      "epoch": 1.975816091954023,
      "grad_norm": 1.0524660348892212,
      "learning_rate": 2.466408982146144e-06,
      "loss": 1.4156,
      "num_input_tokens_seen": 32467184,
      "step": 5372,
      "train_runtime": 5441.6133,
      "train_tokens_per_second": 5966.463
    },
    {
      "epoch": 1.9761839080459769,
      "grad_norm": 0.998673677444458,
      "learning_rate": 2.429596907785754e-06,
      "loss": 1.7635,
      "num_input_tokens_seen": 32472544,
      "step": 5373,
      "train_runtime": 5442.5043,
      "train_tokens_per_second": 5966.471
    },
    {
      "epoch": 1.9765517241379311,
      "grad_norm": 0.969534158706665,
      "learning_rate": 2.392784833425364e-06,
      "loss": 1.4915,
      "num_input_tokens_seen": 32479248,
      "step": 5374,
      "train_runtime": 5443.5556,
      "train_tokens_per_second": 5966.55
    },
    {
      "epoch": 1.976919540229885,
      "grad_norm": 0.9665586948394775,
      "learning_rate": 2.3559727590649733e-06,
      "loss": 1.4788,
      "num_input_tokens_seen": 32487584,
      "step": 5375,
      "train_runtime": 5444.8847,
      "train_tokens_per_second": 5966.625
    },
    {
      "epoch": 1.9772873563218392,
      "grad_norm": 0.9553905129432678,
      "learning_rate": 2.319160684704583e-06,
      "loss": 1.3564,
      "num_input_tokens_seen": 32492544,
      "step": 5376,
      "train_runtime": 5445.7375,
      "train_tokens_per_second": 5966.601
    },
    {
      "epoch": 1.977655172413793,
      "grad_norm": 1.0792388916015625,
      "learning_rate": 2.282348610344193e-06,
      "loss": 1.5903,
      "num_input_tokens_seen": 32497712,
      "step": 5377,
      "train_runtime": 5446.6038,
      "train_tokens_per_second": 5966.601
    },
    {
      "epoch": 1.9780229885057472,
      "grad_norm": 1.0385632514953613,
      "learning_rate": 2.2455365359838025e-06,
      "loss": 1.329,
      "num_input_tokens_seen": 32504880,
      "step": 5378,
      "train_runtime": 5447.7461,
      "train_tokens_per_second": 5966.666
    },
    {
      "epoch": 1.978390804597701,
      "grad_norm": 0.908355176448822,
      "learning_rate": 2.2087244616234127e-06,
      "loss": 1.2065,
      "num_input_tokens_seen": 32513696,
      "step": 5379,
      "train_runtime": 5449.1411,
      "train_tokens_per_second": 5966.756
    },
    {
      "epoch": 1.9787586206896552,
      "grad_norm": 0.9959275722503662,
      "learning_rate": 2.1719123872630224e-06,
      "loss": 1.6937,
      "num_input_tokens_seen": 32517760,
      "step": 5380,
      "train_runtime": 5449.837,
      "train_tokens_per_second": 5966.74
    },
    {
      "epoch": 1.979126436781609,
      "grad_norm": 1.0768471956253052,
      "learning_rate": 2.135100312902632e-06,
      "loss": 1.5575,
      "num_input_tokens_seen": 32522544,
      "step": 5381,
      "train_runtime": 5450.6656,
      "train_tokens_per_second": 5966.711
    },
    {
      "epoch": 1.9794942528735633,
      "grad_norm": 0.9211650490760803,
      "learning_rate": 2.098288238542242e-06,
      "loss": 1.3118,
      "num_input_tokens_seen": 32532800,
      "step": 5382,
      "train_runtime": 5452.2013,
      "train_tokens_per_second": 5966.911
    },
    {
      "epoch": 1.979862068965517,
      "grad_norm": 0.986584484577179,
      "learning_rate": 2.0614761641818516e-06,
      "loss": 1.4974,
      "num_input_tokens_seen": 32538448,
      "step": 5383,
      "train_runtime": 5453.122,
      "train_tokens_per_second": 5966.939
    },
    {
      "epoch": 1.9802298850574713,
      "grad_norm": 1.0325545072555542,
      "learning_rate": 2.0246640898214614e-06,
      "loss": 1.3793,
      "num_input_tokens_seen": 32544624,
      "step": 5384,
      "train_runtime": 5454.1293,
      "train_tokens_per_second": 5966.97
    },
    {
      "epoch": 1.9805977011494251,
      "grad_norm": 0.9728584885597229,
      "learning_rate": 1.9878520154610715e-06,
      "loss": 1.4634,
      "num_input_tokens_seen": 32550976,
      "step": 5385,
      "train_runtime": 5455.1562,
      "train_tokens_per_second": 5967.011
    },
    {
      "epoch": 1.9809655172413794,
      "grad_norm": 1.024316430091858,
      "learning_rate": 1.951039941100681e-06,
      "loss": 1.35,
      "num_input_tokens_seen": 32556688,
      "step": 5386,
      "train_runtime": 5456.1093,
      "train_tokens_per_second": 5967.015
    },
    {
      "epoch": 1.9813333333333332,
      "grad_norm": 1.017407774925232,
      "learning_rate": 1.914227866740291e-06,
      "loss": 1.4562,
      "num_input_tokens_seen": 32562032,
      "step": 5387,
      "train_runtime": 5456.993,
      "train_tokens_per_second": 5967.028
    },
    {
      "epoch": 1.9817011494252874,
      "grad_norm": 0.994735836982727,
      "learning_rate": 1.8774157923799008e-06,
      "loss": 1.6563,
      "num_input_tokens_seen": 32566448,
      "step": 5388,
      "train_runtime": 5457.7637,
      "train_tokens_per_second": 5966.995
    },
    {
      "epoch": 1.9820689655172414,
      "grad_norm": 1.0104857683181763,
      "learning_rate": 1.8406037180195107e-06,
      "loss": 1.5782,
      "num_input_tokens_seen": 32573168,
      "step": 5389,
      "train_runtime": 5458.8495,
      "train_tokens_per_second": 5967.039
    },
    {
      "epoch": 1.9824367816091955,
      "grad_norm": 0.9368088245391846,
      "learning_rate": 1.8037916436591202e-06,
      "loss": 1.3141,
      "num_input_tokens_seen": 32583120,
      "step": 5390,
      "train_runtime": 5460.3467,
      "train_tokens_per_second": 5967.226
    },
    {
      "epoch": 1.9828045977011495,
      "grad_norm": 0.9728841185569763,
      "learning_rate": 1.7669795692987302e-06,
      "loss": 1.5194,
      "num_input_tokens_seen": 32588016,
      "step": 5391,
      "train_runtime": 5461.188,
      "train_tokens_per_second": 5967.203
    },
    {
      "epoch": 1.9831724137931035,
      "grad_norm": 1.044521689414978,
      "learning_rate": 1.73016749493834e-06,
      "loss": 1.5423,
      "num_input_tokens_seen": 32592896,
      "step": 5392,
      "train_runtime": 5462.0224,
      "train_tokens_per_second": 5967.185
    },
    {
      "epoch": 1.9835402298850575,
      "grad_norm": 1.0784873962402344,
      "learning_rate": 1.6933554205779495e-06,
      "loss": 1.4701,
      "num_input_tokens_seen": 32598208,
      "step": 5393,
      "train_runtime": 5462.9158,
      "train_tokens_per_second": 5967.181
    },
    {
      "epoch": 1.9839080459770115,
      "grad_norm": 1.1091370582580566,
      "learning_rate": 1.6565433462175594e-06,
      "loss": 1.5639,
      "num_input_tokens_seen": 32604960,
      "step": 5394,
      "train_runtime": 5464.0048,
      "train_tokens_per_second": 5967.228
    },
    {
      "epoch": 1.9842758620689656,
      "grad_norm": 0.967035710811615,
      "learning_rate": 1.6197312718571691e-06,
      "loss": 1.6944,
      "num_input_tokens_seen": 32610848,
      "step": 5395,
      "train_runtime": 5464.9607,
      "train_tokens_per_second": 5967.261
    },
    {
      "epoch": 1.9846436781609196,
      "grad_norm": 1.0483430624008179,
      "learning_rate": 1.582919197496779e-06,
      "loss": 1.6001,
      "num_input_tokens_seen": 32615088,
      "step": 5396,
      "train_runtime": 5465.7188,
      "train_tokens_per_second": 5967.209
    },
    {
      "epoch": 1.9850114942528736,
      "grad_norm": 1.045504093170166,
      "learning_rate": 1.5461071231363888e-06,
      "loss": 1.5257,
      "num_input_tokens_seen": 32620736,
      "step": 5397,
      "train_runtime": 5466.6494,
      "train_tokens_per_second": 5967.227
    },
    {
      "epoch": 1.9853793103448276,
      "grad_norm": 1.0130599737167358,
      "learning_rate": 1.5092950487759986e-06,
      "loss": 1.4876,
      "num_input_tokens_seen": 32627744,
      "step": 5398,
      "train_runtime": 5467.7756,
      "train_tokens_per_second": 5967.279
    },
    {
      "epoch": 1.9857471264367816,
      "grad_norm": 0.8777824640274048,
      "learning_rate": 1.4724829744156085e-06,
      "loss": 1.1176,
      "num_input_tokens_seen": 32635632,
      "step": 5399,
      "train_runtime": 5469.035,
      "train_tokens_per_second": 5967.347
    },
    {
      "epoch": 1.9861149425287357,
      "grad_norm": 0.7812280058860779,
      "learning_rate": 1.435670900055218e-06,
      "loss": 1.2394,
      "num_input_tokens_seen": 32645344,
      "step": 5400,
      "train_runtime": 5470.4902,
      "train_tokens_per_second": 5967.535
    },
    {
      "epoch": 1.9864827586206897,
      "grad_norm": 1.1060978174209595,
      "learning_rate": 1.398858825694828e-06,
      "loss": 1.5318,
      "num_input_tokens_seen": 32649856,
      "step": 5401,
      "train_runtime": 5471.7562,
      "train_tokens_per_second": 5966.979
    },
    {
      "epoch": 1.9868505747126437,
      "grad_norm": 1.0044702291488647,
      "learning_rate": 1.3620467513344377e-06,
      "loss": 1.5481,
      "num_input_tokens_seen": 32659792,
      "step": 5402,
      "train_runtime": 5473.265,
      "train_tokens_per_second": 5967.15
    },
    {
      "epoch": 1.9872183908045977,
      "grad_norm": 1.1206163167953491,
      "learning_rate": 1.3252346769740475e-06,
      "loss": 1.4078,
      "num_input_tokens_seen": 32665104,
      "step": 5403,
      "train_runtime": 5474.1586,
      "train_tokens_per_second": 5967.146
    },
    {
      "epoch": 1.9875862068965517,
      "grad_norm": 1.097223162651062,
      "learning_rate": 1.2884226026136572e-06,
      "loss": 1.602,
      "num_input_tokens_seen": 32669152,
      "step": 5404,
      "train_runtime": 5474.8887,
      "train_tokens_per_second": 5967.09
    },
    {
      "epoch": 1.9879540229885058,
      "grad_norm": 1.0463114976882935,
      "learning_rate": 1.2516105282532672e-06,
      "loss": 1.6595,
      "num_input_tokens_seen": 32673568,
      "step": 5405,
      "train_runtime": 5475.6598,
      "train_tokens_per_second": 5967.056
    },
    {
      "epoch": 1.9883218390804598,
      "grad_norm": 1.0015771389007568,
      "learning_rate": 1.214798453892877e-06,
      "loss": 1.434,
      "num_input_tokens_seen": 32678272,
      "step": 5406,
      "train_runtime": 5476.4832,
      "train_tokens_per_second": 5967.018
    },
    {
      "epoch": 1.9886896551724138,
      "grad_norm": 1.1400842666625977,
      "learning_rate": 1.1779863795324866e-06,
      "loss": 1.4002,
      "num_input_tokens_seen": 32683376,
      "step": 5407,
      "train_runtime": 5477.3249,
      "train_tokens_per_second": 5967.033
    },
    {
      "epoch": 1.9890574712643678,
      "grad_norm": 1.0390219688415527,
      "learning_rate": 1.1411743051720966e-06,
      "loss": 1.5986,
      "num_input_tokens_seen": 32689328,
      "step": 5408,
      "train_runtime": 5478.2892,
      "train_tokens_per_second": 5967.069
    },
    {
      "epoch": 1.9894252873563218,
      "grad_norm": 0.9824442267417908,
      "learning_rate": 1.1043622308117063e-06,
      "loss": 1.6095,
      "num_input_tokens_seen": 32695088,
      "step": 5409,
      "train_runtime": 5479.2359,
      "train_tokens_per_second": 5967.089
    },
    {
      "epoch": 1.9897931034482759,
      "grad_norm": 0.9292029142379761,
      "learning_rate": 1.067550156451316e-06,
      "loss": 1.3702,
      "num_input_tokens_seen": 32705024,
      "step": 5410,
      "train_runtime": 5480.7403,
      "train_tokens_per_second": 5967.264
    },
    {
      "epoch": 1.99016091954023,
      "grad_norm": 1.1019560098648071,
      "learning_rate": 1.0307380820909258e-06,
      "loss": 1.4871,
      "num_input_tokens_seen": 32710576,
      "step": 5411,
      "train_runtime": 5481.6498,
      "train_tokens_per_second": 5967.287
    },
    {
      "epoch": 1.990528735632184,
      "grad_norm": 0.961487352848053,
      "learning_rate": 9.939260077305358e-07,
      "loss": 1.4016,
      "num_input_tokens_seen": 32716944,
      "step": 5412,
      "train_runtime": 5482.6761,
      "train_tokens_per_second": 5967.331
    },
    {
      "epoch": 1.990896551724138,
      "grad_norm": 0.9317049384117126,
      "learning_rate": 9.571139333701455e-07,
      "loss": 1.3659,
      "num_input_tokens_seen": 32724928,
      "step": 5413,
      "train_runtime": 5483.9384,
      "train_tokens_per_second": 5967.414
    },
    {
      "epoch": 1.991264367816092,
      "grad_norm": 0.9792227745056152,
      "learning_rate": 9.203018590097554e-07,
      "loss": 1.6279,
      "num_input_tokens_seen": 32730464,
      "step": 5414,
      "train_runtime": 5484.8508,
      "train_tokens_per_second": 5967.43
    },
    {
      "epoch": 1.991632183908046,
      "grad_norm": 0.9332373142242432,
      "learning_rate": 8.834897846493651e-07,
      "loss": 1.3293,
      "num_input_tokens_seen": 32736400,
      "step": 5415,
      "train_runtime": 5485.8182,
      "train_tokens_per_second": 5967.46
    },
    {
      "epoch": 1.992,
      "grad_norm": 0.9401854276657104,
      "learning_rate": 8.466777102889747e-07,
      "loss": 1.4474,
      "num_input_tokens_seen": 32745760,
      "step": 5416,
      "train_runtime": 5487.2357,
      "train_tokens_per_second": 5967.624
    },
    {
      "epoch": 1.992367816091954,
      "grad_norm": 1.0248979330062866,
      "learning_rate": 8.098656359285846e-07,
      "loss": 1.5212,
      "num_input_tokens_seen": 32751344,
      "step": 5417,
      "train_runtime": 5488.1559,
      "train_tokens_per_second": 5967.641
    },
    {
      "epoch": 1.992735632183908,
      "grad_norm": 0.9462123513221741,
      "learning_rate": 7.730535615681944e-07,
      "loss": 1.5531,
      "num_input_tokens_seen": 32756352,
      "step": 5418,
      "train_runtime": 5488.9709,
      "train_tokens_per_second": 5967.667
    },
    {
      "epoch": 1.993103448275862,
      "grad_norm": 0.9655794501304626,
      "learning_rate": 7.362414872078043e-07,
      "loss": 1.4801,
      "num_input_tokens_seen": 32761440,
      "step": 5419,
      "train_runtime": 5489.8388,
      "train_tokens_per_second": 5967.651
    },
    {
      "epoch": 1.993471264367816,
      "grad_norm": 0.9820994138717651,
      "learning_rate": 6.99429412847414e-07,
      "loss": 1.3995,
      "num_input_tokens_seen": 32766944,
      "step": 5420,
      "train_runtime": 5490.7655,
      "train_tokens_per_second": 5967.646
    },
    {
      "epoch": 1.99383908045977,
      "grad_norm": 1.161966323852539,
      "learning_rate": 6.626173384870237e-07,
      "loss": 1.645,
      "num_input_tokens_seen": 32774704,
      "step": 5421,
      "train_runtime": 5491.9682,
      "train_tokens_per_second": 5967.752
    },
    {
      "epoch": 1.9942068965517241,
      "grad_norm": 1.0541167259216309,
      "learning_rate": 6.258052641266336e-07,
      "loss": 1.4254,
      "num_input_tokens_seen": 32779696,
      "step": 5422,
      "train_runtime": 5492.8143,
      "train_tokens_per_second": 5967.741
    },
    {
      "epoch": 1.9945747126436781,
      "grad_norm": 1.0184603929519653,
      "learning_rate": 5.889931897662433e-07,
      "loss": 1.4754,
      "num_input_tokens_seen": 32785536,
      "step": 5423,
      "train_runtime": 5493.7647,
      "train_tokens_per_second": 5967.772
    },
    {
      "epoch": 1.9949425287356322,
      "grad_norm": 1.1059820652008057,
      "learning_rate": 5.521811154058532e-07,
      "loss": 1.3308,
      "num_input_tokens_seen": 32793536,
      "step": 5424,
      "train_runtime": 5495.0017,
      "train_tokens_per_second": 5967.885
    },
    {
      "epoch": 1.9953103448275862,
      "grad_norm": 0.8938537836074829,
      "learning_rate": 5.153690410454629e-07,
      "loss": 1.3523,
      "num_input_tokens_seen": 32801824,
      "step": 5425,
      "train_runtime": 5496.3196,
      "train_tokens_per_second": 5967.962
    },
    {
      "epoch": 1.9956781609195402,
      "grad_norm": 0.9378447532653809,
      "learning_rate": 4.785569666850728e-07,
      "loss": 1.6491,
      "num_input_tokens_seen": 32808880,
      "step": 5426,
      "train_runtime": 5497.4354,
      "train_tokens_per_second": 5968.034
    },
    {
      "epoch": 1.9960459770114942,
      "grad_norm": 0.9492788314819336,
      "learning_rate": 4.4174489232468255e-07,
      "loss": 1.352,
      "num_input_tokens_seen": 32815856,
      "step": 5427,
      "train_runtime": 5498.5434,
      "train_tokens_per_second": 5968.1
    },
    {
      "epoch": 1.9964137931034482,
      "grad_norm": 1.075098991394043,
      "learning_rate": 4.049328179642923e-07,
      "loss": 1.4486,
      "num_input_tokens_seen": 32821664,
      "step": 5428,
      "train_runtime": 5499.5237,
      "train_tokens_per_second": 5968.092
    },
    {
      "epoch": 1.9967816091954023,
      "grad_norm": 0.9548684358596802,
      "learning_rate": 3.6812074360390213e-07,
      "loss": 1.4292,
      "num_input_tokens_seen": 32827536,
      "step": 5429,
      "train_runtime": 5500.4718,
      "train_tokens_per_second": 5968.131
    },
    {
      "epoch": 1.9971494252873563,
      "grad_norm": 1.0021536350250244,
      "learning_rate": 3.3130866924351187e-07,
      "loss": 1.6485,
      "num_input_tokens_seen": 32832544,
      "step": 5430,
      "train_runtime": 5501.3226,
      "train_tokens_per_second": 5968.118
    },
    {
      "epoch": 1.9975172413793103,
      "grad_norm": 1.0741565227508545,
      "learning_rate": 2.9449659488312166e-07,
      "loss": 1.6941,
      "num_input_tokens_seen": 32837200,
      "step": 5431,
      "train_runtime": 5502.7376,
      "train_tokens_per_second": 5967.43
    },
    {
      "epoch": 1.9978850574712643,
      "grad_norm": 0.9878013730049133,
      "learning_rate": 2.5768452052273145e-07,
      "loss": 1.3869,
      "num_input_tokens_seen": 32842496,
      "step": 5432,
      "train_runtime": 5503.6287,
      "train_tokens_per_second": 5967.426
    },
    {
      "epoch": 1.9982528735632183,
      "grad_norm": 1.0719490051269531,
      "learning_rate": 2.2087244616234127e-07,
      "loss": 1.4274,
      "num_input_tokens_seen": 32847632,
      "step": 5433,
      "train_runtime": 5504.505,
      "train_tokens_per_second": 5967.409
    },
    {
      "epoch": 1.9986206896551724,
      "grad_norm": 1.1153154373168945,
      "learning_rate": 1.8406037180195107e-07,
      "loss": 1.4851,
      "num_input_tokens_seen": 32852816,
      "step": 5434,
      "train_runtime": 5505.3723,
      "train_tokens_per_second": 5967.41
    },
    {
      "epoch": 1.9989885057471264,
      "grad_norm": 1.0122332572937012,
      "learning_rate": 1.4724829744156083e-07,
      "loss": 1.3487,
      "num_input_tokens_seen": 32858576,
      "step": 5435,
      "train_runtime": 5506.3153,
      "train_tokens_per_second": 5967.435
    },
    {
      "epoch": 1.9993563218390804,
      "grad_norm": 0.9718579649925232,
      "learning_rate": 1.1043622308117064e-07,
      "loss": 1.5405,
      "num_input_tokens_seen": 32864720,
      "step": 5436,
      "train_runtime": 5507.3259,
      "train_tokens_per_second": 5967.455
    },
    {
      "epoch": 1.9997241379310344,
      "grad_norm": 0.9767136573791504,
      "learning_rate": 7.362414872078042e-08,
      "loss": 1.4336,
      "num_input_tokens_seen": 32870096,
      "step": 5437,
      "train_runtime": 5508.209,
      "train_tokens_per_second": 5967.474
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.1507837772369385,
      "learning_rate": 3.681207436039021e-08,
      "loss": 1.6856,
      "num_input_tokens_seen": 32874336,
      "step": 5438,
      "train_runtime": 5508.9238,
      "train_tokens_per_second": 5967.47
    }
  ],
  "logging_steps": 1,
  "max_steps": 5438,
  "num_input_tokens_seen": 32874336,
  "num_train_epochs": 2,
  "save_steps": 30,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.402614311610581e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}