{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 272,
  "global_step": 2719,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000367816091954023,
      "grad_norm": 1.6896029710769653,
      "learning_rate": 0.0,
      "loss": 5.7428,
      "num_input_tokens_seen": 3536,
      "step": 1,
      "train_runtime": 4.3917,
      "train_tokens_per_second": 805.149
    },
    {
      "epoch": 0.000735632183908046,
      "grad_norm": 1.2463716268539429,
      "learning_rate": 4e-05,
      "loss": 4.9214,
      "num_input_tokens_seen": 10672,
      "step": 2,
      "train_runtime": 10.5288,
      "train_tokens_per_second": 1013.597
    },
    {
      "epoch": 0.001103448275862069,
      "grad_norm": 1.4572627544403076,
      "learning_rate": 8e-05,
      "loss": 5.5503,
      "num_input_tokens_seen": 14976,
      "step": 3,
      "train_runtime": 13.4789,
      "train_tokens_per_second": 1111.07
    },
    {
      "epoch": 0.001471264367816092,
      "grad_norm": 2.21268367767334,
      "learning_rate": 0.00012,
      "loss": 5.2587,
      "num_input_tokens_seen": 18768,
      "step": 4,
      "train_runtime": 16.172,
      "train_tokens_per_second": 1160.521
    },
    {
      "epoch": 0.0018390804597701149,
      "grad_norm": 1.9235576391220093,
      "learning_rate": 0.00016,
      "loss": 3.9436,
      "num_input_tokens_seen": 25744,
      "step": 5,
      "train_runtime": 20.5776,
      "train_tokens_per_second": 1251.07
    },
    {
      "epoch": 0.002206896551724138,
      "grad_norm": 2.0342299938201904,
      "learning_rate": 0.0002,
      "loss": 3.1697,
      "num_input_tokens_seen": 30800,
      "step": 6,
      "train_runtime": 23.8891,
      "train_tokens_per_second": 1289.288
    },
    {
      "epoch": 0.002574712643678161,
      "grad_norm": 1.2459954023361206,
      "learning_rate": 0.00019992630803242447,
      "loss": 2.8914,
      "num_input_tokens_seen": 36976,
      "step": 7,
      "train_runtime": 27.7393,
      "train_tokens_per_second": 1332.981
    },
    {
      "epoch": 0.002942528735632184,
      "grad_norm": 1.0456387996673584,
      "learning_rate": 0.00019985261606484894,
      "loss": 2.4046,
      "num_input_tokens_seen": 42256,
      "step": 8,
      "train_runtime": 31.2325,
      "train_tokens_per_second": 1352.949
    },
    {
      "epoch": 0.003310344827586207,
      "grad_norm": 0.9048787355422974,
      "learning_rate": 0.0001997789240972734,
      "loss": 2.3319,
      "num_input_tokens_seen": 46512,
      "step": 9,
      "train_runtime": 34.1051,
      "train_tokens_per_second": 1363.784
    },
    {
      "epoch": 0.0036781609195402297,
      "grad_norm": 0.9737868905067444,
      "learning_rate": 0.00019970523212969787,
      "loss": 2.0549,
      "num_input_tokens_seen": 49696,
      "step": 10,
      "train_runtime": 36.44,
      "train_tokens_per_second": 1363.775
    },
    {
      "epoch": 0.004045977011494253,
      "grad_norm": 0.6913642883300781,
      "learning_rate": 0.00019963154016212233,
      "loss": 1.8302,
      "num_input_tokens_seen": 55584,
      "step": 11,
      "train_runtime": 40.1665,
      "train_tokens_per_second": 1383.839
    },
    {
      "epoch": 0.004413793103448276,
      "grad_norm": 1.6613901853561401,
      "learning_rate": 0.00019955784819454683,
      "loss": 1.8091,
      "num_input_tokens_seen": 62592,
      "step": 12,
      "train_runtime": 44.641,
      "train_tokens_per_second": 1402.12
    },
    {
      "epoch": 0.0047816091954022985,
      "grad_norm": 0.4619045853614807,
      "learning_rate": 0.00019948415622697126,
      "loss": 1.6159,
      "num_input_tokens_seen": 66032,
      "step": 13,
      "train_runtime": 47.0748,
      "train_tokens_per_second": 1402.704
    },
    {
      "epoch": 0.005149425287356322,
      "grad_norm": 0.47745728492736816,
      "learning_rate": 0.00019941046425939573,
      "loss": 1.6412,
      "num_input_tokens_seen": 73744,
      "step": 14,
      "train_runtime": 51.9028,
      "train_tokens_per_second": 1420.81
    },
    {
      "epoch": 0.005517241379310344,
      "grad_norm": 0.4540255069732666,
      "learning_rate": 0.0001993367722918202,
      "loss": 1.6662,
      "num_input_tokens_seen": 77504,
      "step": 15,
      "train_runtime": 54.4595,
      "train_tokens_per_second": 1423.149
    },
    {
      "epoch": 0.005885057471264368,
      "grad_norm": 0.3938988745212555,
      "learning_rate": 0.00019926308032424468,
      "loss": 1.4674,
      "num_input_tokens_seen": 82928,
      "step": 16,
      "train_runtime": 58.0412,
      "train_tokens_per_second": 1428.777
    },
    {
      "epoch": 0.006252873563218391,
      "grad_norm": 0.39807504415512085,
      "learning_rate": 0.00019918938835666915,
      "loss": 1.3737,
      "num_input_tokens_seen": 86832,
      "step": 17,
      "train_runtime": 60.8146,
      "train_tokens_per_second": 1427.814
    },
    {
      "epoch": 0.006620689655172414,
      "grad_norm": 0.3505794107913971,
      "learning_rate": 0.0001991156963890936,
      "loss": 1.5536,
      "num_input_tokens_seen": 91056,
      "step": 18,
      "train_runtime": 63.7535,
      "train_tokens_per_second": 1428.252
    },
    {
      "epoch": 0.006988505747126437,
      "grad_norm": 0.3215710520744324,
      "learning_rate": 0.00019904200442151805,
      "loss": 1.4985,
      "num_input_tokens_seen": 97776,
      "step": 19,
      "train_runtime": 68.0162,
      "train_tokens_per_second": 1437.54
    },
    {
      "epoch": 0.0073563218390804595,
      "grad_norm": 0.29926809668540955,
      "learning_rate": 0.00019896831245394254,
      "loss": 1.546,
      "num_input_tokens_seen": 103888,
      "step": 20,
      "train_runtime": 71.8153,
      "train_tokens_per_second": 1446.599
    },
    {
      "epoch": 0.007724137931034483,
      "grad_norm": 0.28521886467933655,
      "learning_rate": 0.000198894620486367,
      "loss": 1.4762,
      "num_input_tokens_seen": 113280,
      "step": 21,
      "train_runtime": 77.4735,
      "train_tokens_per_second": 1462.177
    },
    {
      "epoch": 0.008091954022988505,
      "grad_norm": 0.3224157691001892,
      "learning_rate": 0.00019882092851879147,
      "loss": 1.5076,
      "num_input_tokens_seen": 117072,
      "step": 22,
      "train_runtime": 80.1039,
      "train_tokens_per_second": 1461.502
    },
    {
      "epoch": 0.00845977011494253,
      "grad_norm": 0.31227555871009827,
      "learning_rate": 0.0001987472365512159,
      "loss": 1.5553,
      "num_input_tokens_seen": 123536,
      "step": 23,
      "train_runtime": 84.2535,
      "train_tokens_per_second": 1466.241
    },
    {
      "epoch": 0.008827586206896552,
      "grad_norm": 0.39832526445388794,
      "learning_rate": 0.0001986735445836404,
      "loss": 1.4392,
      "num_input_tokens_seen": 127312,
      "step": 24,
      "train_runtime": 86.9827,
      "train_tokens_per_second": 1463.648
    },
    {
      "epoch": 0.009195402298850575,
      "grad_norm": 0.34754133224487305,
      "learning_rate": 0.00019859985261606487,
      "loss": 1.3843,
      "num_input_tokens_seen": 130896,
      "step": 25,
      "train_runtime": 89.569,
      "train_tokens_per_second": 1461.398
    },
    {
      "epoch": 0.009563218390804597,
      "grad_norm": 0.3295925259590149,
      "learning_rate": 0.00019852616064848933,
      "loss": 1.5071,
      "num_input_tokens_seen": 134416,
      "step": 26,
      "train_runtime": 92.054,
      "train_tokens_per_second": 1460.186
    },
    {
      "epoch": 0.009931034482758621,
      "grad_norm": 0.4134153127670288,
      "learning_rate": 0.00019845246868091377,
      "loss": 1.5796,
      "num_input_tokens_seen": 138304,
      "step": 27,
      "train_runtime": 94.8348,
      "train_tokens_per_second": 1458.368
    },
    {
      "epoch": 0.010298850574712644,
      "grad_norm": 0.3494221270084381,
      "learning_rate": 0.00019837877671333826,
      "loss": 1.6521,
      "num_input_tokens_seen": 146096,
      "step": 28,
      "train_runtime": 99.7402,
      "train_tokens_per_second": 1464.766
    },
    {
      "epoch": 0.010666666666666666,
      "grad_norm": 0.3704027235507965,
      "learning_rate": 0.00019830508474576273,
      "loss": 1.5089,
      "num_input_tokens_seen": 150480,
      "step": 29,
      "train_runtime": 102.7398,
      "train_tokens_per_second": 1464.671
    },
    {
      "epoch": 0.011034482758620689,
      "grad_norm": 0.4166102707386017,
      "learning_rate": 0.0001982313927781872,
      "loss": 1.3976,
      "num_input_tokens_seen": 154704,
      "step": 30,
      "train_runtime": 105.7012,
      "train_tokens_per_second": 1463.597
    },
    {
      "epoch": 0.011402298850574713,
      "grad_norm": 0.3029329180717468,
      "learning_rate": 0.00019815770081061166,
      "loss": 1.451,
      "num_input_tokens_seen": 162656,
      "step": 31,
      "train_runtime": 113.0576,
      "train_tokens_per_second": 1438.7
    },
    {
      "epoch": 0.011770114942528736,
      "grad_norm": 0.3309914171695709,
      "learning_rate": 0.00019808400884303612,
      "loss": 1.464,
      "num_input_tokens_seen": 169392,
      "step": 32,
      "train_runtime": 117.2968,
      "train_tokens_per_second": 1444.132
    },
    {
      "epoch": 0.012137931034482758,
      "grad_norm": 0.3266368806362152,
      "learning_rate": 0.0001980103168754606,
      "loss": 1.3301,
      "num_input_tokens_seen": 173984,
      "step": 33,
      "train_runtime": 120.3639,
      "train_tokens_per_second": 1445.483
    },
    {
      "epoch": 0.012505747126436782,
      "grad_norm": 0.3363005816936493,
      "learning_rate": 0.00019793662490788505,
      "loss": 1.3506,
      "num_input_tokens_seen": 179200,
      "step": 34,
      "train_runtime": 123.8413,
      "train_tokens_per_second": 1447.013
    },
    {
      "epoch": 0.012873563218390805,
      "grad_norm": 0.3295206129550934,
      "learning_rate": 0.00019786293294030952,
      "loss": 1.4824,
      "num_input_tokens_seen": 184384,
      "step": 35,
      "train_runtime": 127.2886,
      "train_tokens_per_second": 1448.551
    },
    {
      "epoch": 0.013241379310344827,
      "grad_norm": 0.31550225615501404,
      "learning_rate": 0.00019778924097273398,
      "loss": 1.3962,
      "num_input_tokens_seen": 190720,
      "step": 36,
      "train_runtime": 131.3805,
      "train_tokens_per_second": 1451.662
    },
    {
      "epoch": 0.01360919540229885,
      "grad_norm": 0.3353164792060852,
      "learning_rate": 0.00019771554900515845,
      "loss": 1.4282,
      "num_input_tokens_seen": 195200,
      "step": 37,
      "train_runtime": 134.4224,
      "train_tokens_per_second": 1452.139
    },
    {
      "epoch": 0.013977011494252874,
      "grad_norm": 0.3152875304222107,
      "learning_rate": 0.0001976418570375829,
      "loss": 1.4053,
      "num_input_tokens_seen": 199600,
      "step": 38,
      "train_runtime": 137.4833,
      "train_tokens_per_second": 1451.813
    },
    {
      "epoch": 0.014344827586206896,
      "grad_norm": 0.3619515597820282,
      "learning_rate": 0.00019756816507000738,
      "loss": 1.4137,
      "num_input_tokens_seen": 203696,
      "step": 39,
      "train_runtime": 140.2899,
      "train_tokens_per_second": 1451.965
    },
    {
      "epoch": 0.014712643678160919,
      "grad_norm": 0.3786216080188751,
      "learning_rate": 0.00019749447310243184,
      "loss": 1.4399,
      "num_input_tokens_seen": 209632,
      "step": 40,
      "train_runtime": 144.1946,
      "train_tokens_per_second": 1453.813
    },
    {
      "epoch": 0.015080459770114943,
      "grad_norm": 0.3696210980415344,
      "learning_rate": 0.0001974207811348563,
      "loss": 1.4629,
      "num_input_tokens_seen": 213136,
      "step": 41,
      "train_runtime": 146.7748,
      "train_tokens_per_second": 1452.129
    },
    {
      "epoch": 0.015448275862068966,
      "grad_norm": 0.3269001841545105,
      "learning_rate": 0.00019734708916728077,
      "loss": 1.465,
      "num_input_tokens_seen": 220544,
      "step": 42,
      "train_runtime": 151.4699,
      "train_tokens_per_second": 1456.025
    },
    {
      "epoch": 0.01581609195402299,
      "grad_norm": 0.33651697635650635,
      "learning_rate": 0.00019727339719970524,
      "loss": 1.3787,
      "num_input_tokens_seen": 224128,
      "step": 43,
      "train_runtime": 154.0748,
      "train_tokens_per_second": 1454.67
    },
    {
      "epoch": 0.01618390804597701,
      "grad_norm": 0.49768590927124023,
      "learning_rate": 0.00019719970523212973,
      "loss": 1.3881,
      "num_input_tokens_seen": 227808,
      "step": 44,
      "train_runtime": 156.7743,
      "train_tokens_per_second": 1453.095
    },
    {
      "epoch": 0.016551724137931035,
      "grad_norm": 0.3080534040927887,
      "learning_rate": 0.00019712601326455416,
      "loss": 1.3805,
      "num_input_tokens_seen": 232656,
      "step": 45,
      "train_runtime": 160.0058,
      "train_tokens_per_second": 1454.048
    },
    {
      "epoch": 0.01691954022988506,
      "grad_norm": 0.3345535397529602,
      "learning_rate": 0.00019705232129697863,
      "loss": 1.4768,
      "num_input_tokens_seen": 237120,
      "step": 46,
      "train_runtime": 163.0919,
      "train_tokens_per_second": 1453.904
    },
    {
      "epoch": 0.01728735632183908,
      "grad_norm": 0.34982284903526306,
      "learning_rate": 0.0001969786293294031,
      "loss": 1.4573,
      "num_input_tokens_seen": 241696,
      "step": 47,
      "train_runtime": 166.2746,
      "train_tokens_per_second": 1453.595
    },
    {
      "epoch": 0.017655172413793104,
      "grad_norm": 0.33389902114868164,
      "learning_rate": 0.00019690493736182759,
      "loss": 1.2849,
      "num_input_tokens_seen": 248656,
      "step": 48,
      "train_runtime": 170.7243,
      "train_tokens_per_second": 1456.477
    },
    {
      "epoch": 0.018022988505747125,
      "grad_norm": 0.3556322157382965,
      "learning_rate": 0.00019683124539425205,
      "loss": 1.4287,
      "num_input_tokens_seen": 252240,
      "step": 49,
      "train_runtime": 173.2828,
      "train_tokens_per_second": 1455.655
    },
    {
      "epoch": 0.01839080459770115,
      "grad_norm": 0.37753573060035706,
      "learning_rate": 0.0001967575534266765,
      "loss": 1.315,
      "num_input_tokens_seen": 256400,
      "step": 50,
      "train_runtime": 176.2014,
      "train_tokens_per_second": 1455.153
    },
    {
      "epoch": 0.018758620689655173,
      "grad_norm": 0.36437854170799255,
      "learning_rate": 0.00019668386145910095,
      "loss": 1.4924,
      "num_input_tokens_seen": 261280,
      "step": 51,
      "train_runtime": 179.5139,
      "train_tokens_per_second": 1455.486
    },
    {
      "epoch": 0.019126436781609194,
      "grad_norm": 0.30621838569641113,
      "learning_rate": 0.00019661016949152545,
      "loss": 1.3666,
      "num_input_tokens_seen": 267856,
      "step": 52,
      "train_runtime": 183.8031,
      "train_tokens_per_second": 1457.299
    },
    {
      "epoch": 0.01949425287356322,
      "grad_norm": 0.38290640711784363,
      "learning_rate": 0.0001965364775239499,
      "loss": 1.458,
      "num_input_tokens_seen": 272320,
      "step": 53,
      "train_runtime": 186.9012,
      "train_tokens_per_second": 1457.027
    },
    {
      "epoch": 0.019862068965517243,
      "grad_norm": 0.32105013728141785,
      "learning_rate": 0.00019646278555637438,
      "loss": 1.4186,
      "num_input_tokens_seen": 277392,
      "step": 54,
      "train_runtime": 190.3458,
      "train_tokens_per_second": 1457.306
    },
    {
      "epoch": 0.020229885057471263,
      "grad_norm": 0.3524461090564728,
      "learning_rate": 0.0001963890935887988,
      "loss": 1.4047,
      "num_input_tokens_seen": 280832,
      "step": 55,
      "train_runtime": 192.9167,
      "train_tokens_per_second": 1455.716
    },
    {
      "epoch": 0.020597701149425288,
      "grad_norm": 0.3453088402748108,
      "learning_rate": 0.0001963154016212233,
      "loss": 1.3904,
      "num_input_tokens_seen": 285504,
      "step": 56,
      "train_runtime": 196.0411,
      "train_tokens_per_second": 1456.347
    },
    {
      "epoch": 0.020965517241379312,
      "grad_norm": 0.3295742869377136,
      "learning_rate": 0.00019624170965364777,
      "loss": 1.278,
      "num_input_tokens_seen": 289984,
      "step": 57,
      "train_runtime": 199.0408,
      "train_tokens_per_second": 1456.907
    },
    {
      "epoch": 0.021333333333333333,
      "grad_norm": 0.3913592994213104,
      "learning_rate": 0.00019616801768607223,
      "loss": 1.405,
      "num_input_tokens_seen": 296512,
      "step": 58,
      "train_runtime": 203.198,
      "train_tokens_per_second": 1459.227
    },
    {
      "epoch": 0.021701149425287357,
      "grad_norm": 0.33533012866973877,
      "learning_rate": 0.00019609432571849667,
      "loss": 1.427,
      "num_input_tokens_seen": 300992,
      "step": 59,
      "train_runtime": 206.2951,
      "train_tokens_per_second": 1459.036
    },
    {
      "epoch": 0.022068965517241378,
      "grad_norm": 0.37227025628089905,
      "learning_rate": 0.00019602063375092116,
      "loss": 1.4043,
      "num_input_tokens_seen": 305952,
      "step": 60,
      "train_runtime": 209.6428,
      "train_tokens_per_second": 1459.396
    },
    {
      "epoch": 0.022436781609195402,
      "grad_norm": 0.35552677512168884,
      "learning_rate": 0.00019594694178334563,
      "loss": 1.4877,
      "num_input_tokens_seen": 314752,
      "step": 61,
      "train_runtime": 217.2613,
      "train_tokens_per_second": 1448.725
    },
    {
      "epoch": 0.022804597701149426,
      "grad_norm": 0.3486213684082031,
      "learning_rate": 0.0001958732498157701,
      "loss": 1.5117,
      "num_input_tokens_seen": 319488,
      "step": 62,
      "train_runtime": 220.4752,
      "train_tokens_per_second": 1449.088
    },
    {
      "epoch": 0.023172413793103447,
      "grad_norm": 0.3574698567390442,
      "learning_rate": 0.00019579955784819456,
      "loss": 1.4744,
      "num_input_tokens_seen": 324656,
      "step": 63,
      "train_runtime": 223.9049,
      "train_tokens_per_second": 1449.973
    },
    {
      "epoch": 0.02354022988505747,
      "grad_norm": 0.35368311405181885,
      "learning_rate": 0.00019572586588061902,
      "loss": 1.4704,
      "num_input_tokens_seen": 330208,
      "step": 64,
      "train_runtime": 227.6251,
      "train_tokens_per_second": 1450.666
    },
    {
      "epoch": 0.023908045977011495,
      "grad_norm": 0.3501671254634857,
      "learning_rate": 0.0001956521739130435,
      "loss": 1.5263,
      "num_input_tokens_seen": 334320,
      "step": 65,
      "train_runtime": 230.4913,
      "train_tokens_per_second": 1450.467
    },
    {
      "epoch": 0.024275862068965516,
      "grad_norm": 0.3009653389453888,
      "learning_rate": 0.00019557848194546795,
      "loss": 1.4436,
      "num_input_tokens_seen": 340592,
      "step": 66,
      "train_runtime": 234.5821,
      "train_tokens_per_second": 1451.91
    },
    {
      "epoch": 0.02464367816091954,
      "grad_norm": 0.3149993419647217,
      "learning_rate": 0.00019550478997789242,
      "loss": 1.4211,
      "num_input_tokens_seen": 345856,
      "step": 67,
      "train_runtime": 238.027,
      "train_tokens_per_second": 1453.012
    },
    {
      "epoch": 0.025011494252873565,
      "grad_norm": 0.38887569308280945,
      "learning_rate": 0.00019543109801031688,
      "loss": 1.5181,
      "num_input_tokens_seen": 349728,
      "step": 68,
      "train_runtime": 240.699,
      "train_tokens_per_second": 1452.968
    },
    {
      "epoch": 0.025379310344827585,
      "grad_norm": 0.3457549810409546,
      "learning_rate": 0.00019535740604274135,
      "loss": 1.3213,
      "num_input_tokens_seen": 354640,
      "step": 69,
      "train_runtime": 244.0063,
      "train_tokens_per_second": 1453.405
    },
    {
      "epoch": 0.02574712643678161,
      "grad_norm": 0.3572216033935547,
      "learning_rate": 0.0001952837140751658,
      "loss": 1.4212,
      "num_input_tokens_seen": 359808,
      "step": 70,
      "train_runtime": 247.5256,
      "train_tokens_per_second": 1453.619
    },
    {
      "epoch": 0.026114942528735634,
      "grad_norm": 0.3528576195240021,
      "learning_rate": 0.00019521002210759028,
      "loss": 1.5576,
      "num_input_tokens_seen": 363856,
      "step": 71,
      "train_runtime": 250.3053,
      "train_tokens_per_second": 1453.649
    },
    {
      "epoch": 0.026482758620689655,
      "grad_norm": 0.30931103229522705,
      "learning_rate": 0.00019513633014001477,
      "loss": 1.3123,
      "num_input_tokens_seen": 368320,
      "step": 72,
      "train_runtime": 253.4049,
      "train_tokens_per_second": 1453.484
    },
    {
      "epoch": 0.02685057471264368,
      "grad_norm": 0.31904783844947815,
      "learning_rate": 0.0001950626381724392,
      "loss": 1.3431,
      "num_input_tokens_seen": 372272,
      "step": 73,
      "train_runtime": 256.1682,
      "train_tokens_per_second": 1453.233
    },
    {
      "epoch": 0.0272183908045977,
      "grad_norm": 0.35796838998794556,
      "learning_rate": 0.00019498894620486367,
      "loss": 1.2837,
      "num_input_tokens_seen": 375664,
      "step": 74,
      "train_runtime": 258.6393,
      "train_tokens_per_second": 1452.463
    },
    {
      "epoch": 0.027586206896551724,
      "grad_norm": 0.3376178741455078,
      "learning_rate": 0.00019491525423728814,
      "loss": 1.4641,
      "num_input_tokens_seen": 379536,
      "step": 75,
      "train_runtime": 261.4376,
      "train_tokens_per_second": 1451.727
    },
    {
      "epoch": 0.027954022988505748,
      "grad_norm": 0.3116322457790375,
      "learning_rate": 0.00019484156226971263,
      "loss": 1.3307,
      "num_input_tokens_seen": 382944,
      "step": 76,
      "train_runtime": 263.8597,
      "train_tokens_per_second": 1451.317
    },
    {
      "epoch": 0.02832183908045977,
      "grad_norm": 0.290591299533844,
      "learning_rate": 0.00019476787030213707,
      "loss": 1.411,
      "num_input_tokens_seen": 387696,
      "step": 77,
      "train_runtime": 267.0011,
      "train_tokens_per_second": 1452.039
    },
    {
      "epoch": 0.028689655172413793,
      "grad_norm": 0.34067365527153015,
      "learning_rate": 0.00019469417833456153,
      "loss": 1.3646,
      "num_input_tokens_seen": 392016,
      "step": 78,
      "train_runtime": 269.9263,
      "train_tokens_per_second": 1452.307
    },
    {
      "epoch": 0.029057471264367817,
      "grad_norm": 0.31595438718795776,
      "learning_rate": 0.000194620486366986,
      "loss": 1.4867,
      "num_input_tokens_seen": 400864,
      "step": 79,
      "train_runtime": 275.3932,
      "train_tokens_per_second": 1455.606
    },
    {
      "epoch": 0.029425287356321838,
      "grad_norm": 0.3195711672306061,
      "learning_rate": 0.0001945467943994105,
      "loss": 1.4826,
      "num_input_tokens_seen": 406096,
      "step": 80,
      "train_runtime": 278.8267,
      "train_tokens_per_second": 1456.446
    },
    {
      "epoch": 0.029793103448275862,
      "grad_norm": 0.3230014443397522,
      "learning_rate": 0.00019447310243183495,
      "loss": 1.4192,
      "num_input_tokens_seen": 409888,
      "step": 81,
      "train_runtime": 281.5895,
      "train_tokens_per_second": 1455.623
    },
    {
      "epoch": 0.030160919540229886,
      "grad_norm": 0.3409586548805237,
      "learning_rate": 0.0001943994104642594,
      "loss": 1.3791,
      "num_input_tokens_seen": 415520,
      "step": 82,
      "train_runtime": 285.3375,
      "train_tokens_per_second": 1456.24
    },
    {
      "epoch": 0.030528735632183907,
      "grad_norm": 0.3579333424568176,
      "learning_rate": 0.00019432571849668386,
      "loss": 1.6083,
      "num_input_tokens_seen": 419472,
      "step": 83,
      "train_runtime": 288.0587,
      "train_tokens_per_second": 1456.203
    },
    {
      "epoch": 0.03089655172413793,
      "grad_norm": 0.3503994643688202,
      "learning_rate": 0.00019425202652910835,
      "loss": 1.279,
      "num_input_tokens_seen": 424192,
      "step": 84,
      "train_runtime": 291.3277,
      "train_tokens_per_second": 1456.065
    },
    {
      "epoch": 0.03126436781609195,
      "grad_norm": 0.3124304711818695,
      "learning_rate": 0.0001941783345615328,
      "loss": 1.2399,
      "num_input_tokens_seen": 428928,
      "step": 85,
      "train_runtime": 294.4718,
      "train_tokens_per_second": 1456.601
    },
    {
      "epoch": 0.03163218390804598,
      "grad_norm": 0.3414918780326843,
      "learning_rate": 0.00019410464259395728,
      "loss": 1.277,
      "num_input_tokens_seen": 433936,
      "step": 86,
      "train_runtime": 297.8887,
      "train_tokens_per_second": 1456.705
    },
    {
      "epoch": 0.032,
      "grad_norm": 0.3402044177055359,
      "learning_rate": 0.00019403095062638171,
      "loss": 1.3645,
      "num_input_tokens_seen": 439344,
      "step": 87,
      "train_runtime": 301.4631,
      "train_tokens_per_second": 1457.372
    },
    {
      "epoch": 0.03236781609195402,
      "grad_norm": 0.3228797912597656,
      "learning_rate": 0.0001939572586588062,
      "loss": 1.3342,
      "num_input_tokens_seen": 443456,
      "step": 88,
      "train_runtime": 304.3388,
      "train_tokens_per_second": 1457.113
    },
    {
      "epoch": 0.03273563218390805,
      "grad_norm": 0.362979531288147,
      "learning_rate": 0.00019388356669123067,
      "loss": 1.5969,
      "num_input_tokens_seen": 448336,
      "step": 89,
      "train_runtime": 307.7029,
      "train_tokens_per_second": 1457.042
    },
    {
      "epoch": 0.03310344827586207,
      "grad_norm": 0.31592848896980286,
      "learning_rate": 0.00019380987472365514,
      "loss": 1.3154,
      "num_input_tokens_seen": 452912,
      "step": 90,
      "train_runtime": 310.8176,
      "train_tokens_per_second": 1457.163
    },
    {
      "epoch": 0.03347126436781609,
      "grad_norm": 0.37438368797302246,
      "learning_rate": 0.0001937361827560796,
      "loss": 1.279,
      "num_input_tokens_seen": 458864,
      "step": 91,
      "train_runtime": 317.0511,
      "train_tokens_per_second": 1447.287
    },
    {
      "epoch": 0.03383908045977012,
      "grad_norm": 0.3996255695819855,
      "learning_rate": 0.00019366249078850407,
      "loss": 1.4509,
      "num_input_tokens_seen": 462208,
      "step": 92,
      "train_runtime": 319.4927,
      "train_tokens_per_second": 1446.693
    },
    {
      "epoch": 0.03420689655172414,
      "grad_norm": 0.35339540243148804,
      "learning_rate": 0.00019358879882092853,
      "loss": 1.315,
      "num_input_tokens_seen": 466336,
      "step": 93,
      "train_runtime": 322.2881,
      "train_tokens_per_second": 1446.954
    },
    {
      "epoch": 0.03457471264367816,
      "grad_norm": 0.33935806155204773,
      "learning_rate": 0.000193515106853353,
      "loss": 1.4,
      "num_input_tokens_seen": 473200,
      "step": 94,
      "train_runtime": 326.6885,
      "train_tokens_per_second": 1448.475
    },
    {
      "epoch": 0.03494252873563218,
      "grad_norm": 0.37198519706726074,
      "learning_rate": 0.00019344141488577746,
      "loss": 1.2492,
      "num_input_tokens_seen": 479120,
      "step": 95,
      "train_runtime": 330.6315,
      "train_tokens_per_second": 1449.106
    },
    {
      "epoch": 0.03531034482758621,
      "grad_norm": 0.3687741756439209,
      "learning_rate": 0.00019336772291820192,
      "loss": 1.377,
      "num_input_tokens_seen": 483088,
      "step": 96,
      "train_runtime": 333.422,
      "train_tokens_per_second": 1448.878
    },
    {
      "epoch": 0.03567816091954023,
      "grad_norm": 0.34657415747642517,
      "learning_rate": 0.0001932940309506264,
      "loss": 1.4442,
      "num_input_tokens_seen": 487088,
      "step": 97,
      "train_runtime": 336.2316,
      "train_tokens_per_second": 1448.668
    },
    {
      "epoch": 0.03604597701149425,
      "grad_norm": 0.3943544030189514,
      "learning_rate": 0.00019322033898305085,
      "loss": 1.3582,
      "num_input_tokens_seen": 490912,
      "step": 98,
      "train_runtime": 338.9424,
      "train_tokens_per_second": 1448.364
    },
    {
      "epoch": 0.03641379310344828,
      "grad_norm": 0.35165315866470337,
      "learning_rate": 0.00019314664701547532,
      "loss": 1.3541,
      "num_input_tokens_seen": 494032,
      "step": 99,
      "train_runtime": 341.2972,
      "train_tokens_per_second": 1447.513
    },
    {
      "epoch": 0.0367816091954023,
      "grad_norm": 0.35324153304100037,
      "learning_rate": 0.00019307295504789978,
      "loss": 1.4833,
      "num_input_tokens_seen": 502432,
      "step": 100,
      "train_runtime": 346.6342,
      "train_tokens_per_second": 1449.459
    },
    {
      "epoch": 0.03714942528735632,
      "grad_norm": 0.3213910460472107,
      "learning_rate": 0.00019299926308032425,
      "loss": 1.3556,
      "num_input_tokens_seen": 507248,
      "step": 101,
      "train_runtime": 349.8347,
      "train_tokens_per_second": 1449.965
    },
    {
      "epoch": 0.03751724137931035,
      "grad_norm": 0.3446355164051056,
      "learning_rate": 0.00019292557111274871,
      "loss": 1.3074,
      "num_input_tokens_seen": 510336,
      "step": 102,
      "train_runtime": 352.2047,
      "train_tokens_per_second": 1448.976
    },
    {
      "epoch": 0.03788505747126437,
      "grad_norm": 0.3802237808704376,
      "learning_rate": 0.00019285187914517318,
      "loss": 1.4042,
      "num_input_tokens_seen": 513392,
      "step": 103,
      "train_runtime": 354.5584,
      "train_tokens_per_second": 1447.976
    },
    {
      "epoch": 0.03825287356321839,
      "grad_norm": 0.3681686222553253,
      "learning_rate": 0.00019277818717759767,
      "loss": 1.3513,
      "num_input_tokens_seen": 517072,
      "step": 104,
      "train_runtime": 357.169,
      "train_tokens_per_second": 1447.696
    },
    {
      "epoch": 0.038620689655172416,
      "grad_norm": 0.3716645836830139,
      "learning_rate": 0.0001927044952100221,
      "loss": 1.3036,
      "num_input_tokens_seen": 521168,
      "step": 105,
      "train_runtime": 360.0461,
      "train_tokens_per_second": 1447.504
    },
    {
      "epoch": 0.03898850574712644,
      "grad_norm": 0.34930476546287537,
      "learning_rate": 0.00019263080324244657,
      "loss": 1.3586,
      "num_input_tokens_seen": 524528,
      "step": 106,
      "train_runtime": 362.559,
      "train_tokens_per_second": 1446.738
    },
    {
      "epoch": 0.03935632183908046,
      "grad_norm": 0.3603226840496063,
      "learning_rate": 0.00019255711127487104,
      "loss": 1.4267,
      "num_input_tokens_seen": 528400,
      "step": 107,
      "train_runtime": 365.3047,
      "train_tokens_per_second": 1446.464
    },
    {
      "epoch": 0.039724137931034485,
      "grad_norm": 0.3882533609867096,
      "learning_rate": 0.00019248341930729553,
      "loss": 1.364,
      "num_input_tokens_seen": 531488,
      "step": 108,
      "train_runtime": 367.6695,
      "train_tokens_per_second": 1445.559
    },
    {
      "epoch": 0.040091954022988506,
      "grad_norm": 0.3208283483982086,
      "learning_rate": 0.00019240972733971997,
      "loss": 1.4234,
      "num_input_tokens_seen": 537232,
      "step": 109,
      "train_runtime": 371.4983,
      "train_tokens_per_second": 1446.123
    },
    {
      "epoch": 0.04045977011494253,
      "grad_norm": 0.37671953439712524,
      "learning_rate": 0.00019233603537214443,
      "loss": 1.3872,
      "num_input_tokens_seen": 541552,
      "step": 110,
      "train_runtime": 374.5105,
      "train_tokens_per_second": 1446.026
    },
    {
      "epoch": 0.040827586206896554,
      "grad_norm": 0.36036062240600586,
      "learning_rate": 0.0001922623434045689,
      "loss": 1.2919,
      "num_input_tokens_seen": 547280,
      "step": 111,
      "train_runtime": 378.2819,
      "train_tokens_per_second": 1446.752
    },
    {
      "epoch": 0.041195402298850575,
      "grad_norm": 0.4014540910720825,
      "learning_rate": 0.0001921886514369934,
      "loss": 1.4405,
      "num_input_tokens_seen": 550816,
      "step": 112,
      "train_runtime": 380.8525,
      "train_tokens_per_second": 1446.271
    },
    {
      "epoch": 0.041563218390804596,
      "grad_norm": 0.3496030271053314,
      "learning_rate": 0.00019211495946941785,
      "loss": 1.3179,
      "num_input_tokens_seen": 555056,
      "step": 113,
      "train_runtime": 383.6774,
      "train_tokens_per_second": 1446.674
    },
    {
      "epoch": 0.041931034482758624,
      "grad_norm": 0.318844199180603,
      "learning_rate": 0.0001920412675018423,
      "loss": 1.2955,
      "num_input_tokens_seen": 559472,
      "step": 114,
      "train_runtime": 386.6985,
      "train_tokens_per_second": 1446.791
    },
    {
      "epoch": 0.042298850574712644,
      "grad_norm": 0.35854312777519226,
      "learning_rate": 0.00019196757553426676,
      "loss": 1.284,
      "num_input_tokens_seen": 562672,
      "step": 115,
      "train_runtime": 389.1303,
      "train_tokens_per_second": 1445.973
    },
    {
      "epoch": 0.042666666666666665,
      "grad_norm": 0.41531145572662354,
      "learning_rate": 0.00019189388356669125,
      "loss": 1.4152,
      "num_input_tokens_seen": 568096,
      "step": 116,
      "train_runtime": 392.6683,
      "train_tokens_per_second": 1446.758
    },
    {
      "epoch": 0.04303448275862069,
      "grad_norm": 0.3431352972984314,
      "learning_rate": 0.0001918201915991157,
      "loss": 1.4023,
      "num_input_tokens_seen": 571680,
      "step": 117,
      "train_runtime": 395.3275,
      "train_tokens_per_second": 1446.092
    },
    {
      "epoch": 0.043402298850574714,
      "grad_norm": 0.35485905408859253,
      "learning_rate": 0.00019174649963154018,
      "loss": 1.4465,
      "num_input_tokens_seen": 576048,
      "step": 118,
      "train_runtime": 398.2846,
      "train_tokens_per_second": 1446.322
    },
    {
      "epoch": 0.043770114942528734,
      "grad_norm": 0.3515053391456604,
      "learning_rate": 0.00019167280766396462,
      "loss": 1.3382,
      "num_input_tokens_seen": 579824,
      "step": 119,
      "train_runtime": 400.9737,
      "train_tokens_per_second": 1446.04
    },
    {
      "epoch": 0.044137931034482755,
      "grad_norm": 0.35786688327789307,
      "learning_rate": 0.0001915991156963891,
      "loss": 1.3336,
      "num_input_tokens_seen": 585760,
      "step": 120,
      "train_runtime": 404.9769,
      "train_tokens_per_second": 1446.403
    },
    {
      "epoch": 0.04450574712643678,
      "grad_norm": 0.33091384172439575,
      "learning_rate": 0.00019152542372881357,
      "loss": 1.3462,
      "num_input_tokens_seen": 589200,
      "step": 121,
      "train_runtime": 409.5632,
      "train_tokens_per_second": 1438.606
    },
    {
      "epoch": 0.044873563218390804,
      "grad_norm": 0.3029051423072815,
      "learning_rate": 0.00019145173176123804,
      "loss": 1.4302,
      "num_input_tokens_seen": 595152,
      "step": 122,
      "train_runtime": 413.4911,
      "train_tokens_per_second": 1439.334
    },
    {
      "epoch": 0.045241379310344824,
      "grad_norm": 0.3376113176345825,
      "learning_rate": 0.0001913780397936625,
      "loss": 1.3544,
      "num_input_tokens_seen": 599360,
      "step": 123,
      "train_runtime": 416.3831,
      "train_tokens_per_second": 1439.444
    },
    {
      "epoch": 0.04560919540229885,
      "grad_norm": 0.33625632524490356,
      "learning_rate": 0.00019130434782608697,
      "loss": 1.4003,
      "num_input_tokens_seen": 603616,
      "step": 124,
      "train_runtime": 419.2971,
      "train_tokens_per_second": 1439.59
    },
    {
      "epoch": 0.04597701149425287,
      "grad_norm": 0.40393340587615967,
      "learning_rate": 0.00019123065585851143,
      "loss": 1.4769,
      "num_input_tokens_seen": 609008,
      "step": 125,
      "train_runtime": 422.8715,
      "train_tokens_per_second": 1440.173
    },
    {
      "epoch": 0.046344827586206894,
      "grad_norm": 0.3525630831718445,
      "learning_rate": 0.0001911569638909359,
      "loss": 1.2941,
      "num_input_tokens_seen": 613712,
      "step": 126,
      "train_runtime": 426.0505,
      "train_tokens_per_second": 1440.468
    },
    {
      "epoch": 0.04671264367816092,
      "grad_norm": 0.3761114478111267,
      "learning_rate": 0.00019108327192336036,
      "loss": 1.2634,
      "num_input_tokens_seen": 618240,
      "step": 127,
      "train_runtime": 429.1152,
      "train_tokens_per_second": 1440.732
    },
    {
      "epoch": 0.04708045977011494,
      "grad_norm": 0.35774338245391846,
      "learning_rate": 0.00019100957995578483,
      "loss": 1.2951,
      "num_input_tokens_seen": 627248,
      "step": 128,
      "train_runtime": 434.7159,
      "train_tokens_per_second": 1442.892
    },
    {
      "epoch": 0.04744827586206896,
      "grad_norm": 0.3444937765598297,
      "learning_rate": 0.0001909358879882093,
      "loss": 1.1266,
      "num_input_tokens_seen": 631424,
      "step": 129,
      "train_runtime": 437.6847,
      "train_tokens_per_second": 1442.646
    },
    {
      "epoch": 0.04781609195402299,
      "grad_norm": 0.3678358793258667,
      "learning_rate": 0.00019086219602063376,
      "loss": 1.4521,
      "num_input_tokens_seen": 636256,
      "step": 130,
      "train_runtime": 440.9446,
      "train_tokens_per_second": 1442.939
    },
    {
      "epoch": 0.04818390804597701,
      "grad_norm": 0.32934266328811646,
      "learning_rate": 0.00019078850405305825,
      "loss": 1.1365,
      "num_input_tokens_seen": 640400,
      "step": 131,
      "train_runtime": 443.8612,
      "train_tokens_per_second": 1442.793
    },
    {
      "epoch": 0.04855172413793103,
      "grad_norm": 0.3722214996814728,
      "learning_rate": 0.00019071481208548269,
      "loss": 1.4527,
      "num_input_tokens_seen": 644416,
      "step": 132,
      "train_runtime": 446.6875,
      "train_tokens_per_second": 1442.655
    },
    {
      "epoch": 0.04891954022988506,
      "grad_norm": 0.4163220226764679,
      "learning_rate": 0.00019064112011790715,
      "loss": 1.4116,
      "num_input_tokens_seen": 648288,
      "step": 133,
      "train_runtime": 449.4526,
      "train_tokens_per_second": 1442.395
    },
    {
      "epoch": 0.04928735632183908,
      "grad_norm": 0.3490603566169739,
      "learning_rate": 0.00019056742815033162,
      "loss": 1.3319,
      "num_input_tokens_seen": 652416,
      "step": 134,
      "train_runtime": 452.3438,
      "train_tokens_per_second": 1442.301
    },
    {
      "epoch": 0.0496551724137931,
      "grad_norm": 0.34180787205696106,
      "learning_rate": 0.0001904937361827561,
      "loss": 1.2957,
      "num_input_tokens_seen": 657360,
      "step": 135,
      "train_runtime": 455.6559,
      "train_tokens_per_second": 1442.667
    },
    {
      "epoch": 0.05002298850574713,
      "grad_norm": 0.36564236879348755,
      "learning_rate": 0.00019042004421518057,
      "loss": 1.2748,
      "num_input_tokens_seen": 661840,
      "step": 136,
      "train_runtime": 458.7545,
      "train_tokens_per_second": 1442.689
    },
    {
      "epoch": 0.05039080459770115,
      "grad_norm": 0.3804440200328827,
      "learning_rate": 0.000190346352247605,
      "loss": 1.3977,
      "num_input_tokens_seen": 665712,
      "step": 137,
      "train_runtime": 461.5453,
      "train_tokens_per_second": 1442.355
    },
    {
      "epoch": 0.05075862068965517,
      "grad_norm": 0.3501209318637848,
      "learning_rate": 0.00019027266028002947,
      "loss": 1.4359,
      "num_input_tokens_seen": 668976,
      "step": 138,
      "train_runtime": 463.9435,
      "train_tokens_per_second": 1441.934
    },
    {
      "epoch": 0.0511264367816092,
      "grad_norm": 0.42087605595588684,
      "learning_rate": 0.00019019896831245397,
      "loss": 1.2101,
      "num_input_tokens_seen": 672672,
      "step": 139,
      "train_runtime": 466.5359,
      "train_tokens_per_second": 1441.844
    },
    {
      "epoch": 0.05149425287356322,
      "grad_norm": 0.3437168300151825,
      "learning_rate": 0.00019012527634487843,
      "loss": 1.4146,
      "num_input_tokens_seen": 676944,
      "step": 140,
      "train_runtime": 469.4428,
      "train_tokens_per_second": 1442.016
    },
    {
      "epoch": 0.05186206896551724,
      "grad_norm": 0.37178879976272583,
      "learning_rate": 0.0001900515843773029,
      "loss": 1.3036,
      "num_input_tokens_seen": 683776,
      "step": 141,
      "train_runtime": 473.8884,
      "train_tokens_per_second": 1442.905
    },
    {
      "epoch": 0.05222988505747127,
      "grad_norm": 0.41747716069221497,
      "learning_rate": 0.00018997789240972733,
      "loss": 1.2937,
      "num_input_tokens_seen": 689152,
      "step": 142,
      "train_runtime": 477.4701,
      "train_tokens_per_second": 1443.341
    },
    {
      "epoch": 0.05259770114942529,
      "grad_norm": 0.3657858669757843,
      "learning_rate": 0.00018990420044215183,
      "loss": 1.2967,
      "num_input_tokens_seen": 692880,
      "step": 143,
      "train_runtime": 480.1332,
      "train_tokens_per_second": 1443.1
    },
    {
      "epoch": 0.05296551724137931,
      "grad_norm": 0.353496253490448,
      "learning_rate": 0.0001898305084745763,
      "loss": 1.3631,
      "num_input_tokens_seen": 697040,
      "step": 144,
      "train_runtime": 483.115,
      "train_tokens_per_second": 1442.803
    },
    {
      "epoch": 0.05333333333333334,
      "grad_norm": 0.3528427183628082,
      "learning_rate": 0.00018975681650700076,
      "loss": 1.4055,
      "num_input_tokens_seen": 704880,
      "step": 145,
      "train_runtime": 488.0491,
      "train_tokens_per_second": 1444.281
    },
    {
      "epoch": 0.05370114942528736,
      "grad_norm": 0.362886518239975,
      "learning_rate": 0.0001896831245394252,
      "loss": 1.3992,
      "num_input_tokens_seen": 709392,
      "step": 146,
      "train_runtime": 491.1384,
      "train_tokens_per_second": 1444.383
    },
    {
      "epoch": 0.05406896551724138,
      "grad_norm": 0.34845200181007385,
      "learning_rate": 0.00018960943257184968,
      "loss": 1.3727,
      "num_input_tokens_seen": 713856,
      "step": 147,
      "train_runtime": 494.1905,
      "train_tokens_per_second": 1444.496
    },
    {
      "epoch": 0.0544367816091954,
      "grad_norm": 0.3569094240665436,
      "learning_rate": 0.00018953574060427415,
      "loss": 1.3373,
      "num_input_tokens_seen": 717872,
      "step": 148,
      "train_runtime": 497.0183,
      "train_tokens_per_second": 1444.357
    },
    {
      "epoch": 0.05480459770114943,
      "grad_norm": 0.36434394121170044,
      "learning_rate": 0.00018946204863669861,
      "loss": 1.2812,
      "num_input_tokens_seen": 722768,
      "step": 149,
      "train_runtime": 500.211,
      "train_tokens_per_second": 1444.926
    },
    {
      "epoch": 0.05517241379310345,
      "grad_norm": 0.3712742030620575,
      "learning_rate": 0.00018938835666912308,
      "loss": 1.3474,
      "num_input_tokens_seen": 729504,
      "step": 150,
      "train_runtime": 504.5186,
      "train_tokens_per_second": 1445.941
    },
    {
      "epoch": 0.05554022988505747,
      "grad_norm": 0.3964684009552002,
      "learning_rate": 0.00018931466470154754,
      "loss": 1.4236,
      "num_input_tokens_seen": 733488,
      "step": 151,
      "train_runtime": 509.5647,
      "train_tokens_per_second": 1439.44
    },
    {
      "epoch": 0.055908045977011496,
      "grad_norm": 0.3355660140514374,
      "learning_rate": 0.000189240972733972,
      "loss": 1.3053,
      "num_input_tokens_seen": 737424,
      "step": 152,
      "train_runtime": 512.2659,
      "train_tokens_per_second": 1439.534
    },
    {
      "epoch": 0.05627586206896552,
      "grad_norm": 0.40488651394844055,
      "learning_rate": 0.00018916728076639647,
      "loss": 1.2128,
      "num_input_tokens_seen": 741600,
      "step": 153,
      "train_runtime": 515.1577,
      "train_tokens_per_second": 1439.559
    },
    {
      "epoch": 0.05664367816091954,
      "grad_norm": 0.3622351884841919,
      "learning_rate": 0.00018909358879882094,
      "loss": 1.2979,
      "num_input_tokens_seen": 746752,
      "step": 154,
      "train_runtime": 518.5181,
      "train_tokens_per_second": 1440.166
    },
    {
      "epoch": 0.057011494252873565,
      "grad_norm": 0.3706391751766205,
      "learning_rate": 0.0001890198968312454,
      "loss": 1.2089,
      "num_input_tokens_seen": 750496,
      "step": 155,
      "train_runtime": 521.1917,
      "train_tokens_per_second": 1439.962
    },
    {
      "epoch": 0.057379310344827586,
      "grad_norm": 0.4029405117034912,
      "learning_rate": 0.00018894620486366987,
      "loss": 1.2611,
      "num_input_tokens_seen": 754064,
      "step": 156,
      "train_runtime": 523.798,
      "train_tokens_per_second": 1439.608
    },
    {
      "epoch": 0.05774712643678161,
      "grad_norm": 0.38390249013900757,
      "learning_rate": 0.00018887251289609433,
      "loss": 1.3915,
      "num_input_tokens_seen": 757776,
      "step": 157,
      "train_runtime": 526.5468,
      "train_tokens_per_second": 1439.143
    },
    {
      "epoch": 0.058114942528735634,
      "grad_norm": 0.36949944496154785,
      "learning_rate": 0.0001887988209285188,
      "loss": 1.1603,
      "num_input_tokens_seen": 761584,
      "step": 158,
      "train_runtime": 529.2233,
      "train_tokens_per_second": 1439.06
    },
    {
      "epoch": 0.058482758620689655,
      "grad_norm": 0.4374903738498688,
      "learning_rate": 0.00018872512896094326,
      "loss": 1.3452,
      "num_input_tokens_seen": 765376,
      "step": 159,
      "train_runtime": 531.9152,
      "train_tokens_per_second": 1438.906
    },
    {
      "epoch": 0.058850574712643676,
      "grad_norm": 0.3176044821739197,
      "learning_rate": 0.00018865143699336773,
      "loss": 1.275,
      "num_input_tokens_seen": 770336,
      "step": 160,
      "train_runtime": 535.2815,
      "train_tokens_per_second": 1439.123
    },
    {
      "epoch": 0.059218390804597704,
      "grad_norm": 0.34038686752319336,
      "learning_rate": 0.0001885777450257922,
      "loss": 1.2403,
      "num_input_tokens_seen": 775664,
      "step": 161,
      "train_runtime": 538.7396,
      "train_tokens_per_second": 1439.775
    },
    {
      "epoch": 0.059586206896551724,
      "grad_norm": 0.32963263988494873,
      "learning_rate": 0.00018850405305821666,
      "loss": 1.331,
      "num_input_tokens_seen": 781216,
      "step": 162,
      "train_runtime": 542.4392,
      "train_tokens_per_second": 1440.191
    },
    {
      "epoch": 0.059954022988505745,
      "grad_norm": 0.3501686751842499,
      "learning_rate": 0.00018843036109064115,
      "loss": 1.3088,
      "num_input_tokens_seen": 785248,
      "step": 163,
      "train_runtime": 545.2676,
      "train_tokens_per_second": 1440.115
    },
    {
      "epoch": 0.06032183908045977,
      "grad_norm": 0.35881733894348145,
      "learning_rate": 0.0001883566691230656,
      "loss": 1.5079,
      "num_input_tokens_seen": 788960,
      "step": 164,
      "train_runtime": 547.9221,
      "train_tokens_per_second": 1439.913
    },
    {
      "epoch": 0.060689655172413794,
      "grad_norm": 0.33264872431755066,
      "learning_rate": 0.00018828297715549005,
      "loss": 1.1775,
      "num_input_tokens_seen": 798256,
      "step": 165,
      "train_runtime": 553.7521,
      "train_tokens_per_second": 1441.54
    },
    {
      "epoch": 0.061057471264367814,
      "grad_norm": 0.3732989728450775,
      "learning_rate": 0.00018820928518791452,
      "loss": 1.4259,
      "num_input_tokens_seen": 801872,
      "step": 166,
      "train_runtime": 556.2786,
      "train_tokens_per_second": 1441.493
    },
    {
      "epoch": 0.06142528735632184,
      "grad_norm": 0.3586452007293701,
      "learning_rate": 0.000188135593220339,
      "loss": 1.3993,
      "num_input_tokens_seen": 804960,
      "step": 167,
      "train_runtime": 558.6359,
      "train_tokens_per_second": 1440.939
    },
    {
      "epoch": 0.06179310344827586,
      "grad_norm": 0.3551160991191864,
      "learning_rate": 0.00018806190125276347,
      "loss": 1.3561,
      "num_input_tokens_seen": 809408,
      "step": 168,
      "train_runtime": 561.6759,
      "train_tokens_per_second": 1441.059
    },
    {
      "epoch": 0.062160919540229884,
      "grad_norm": 0.42596322298049927,
      "learning_rate": 0.0001879882092851879,
      "loss": 1.4839,
      "num_input_tokens_seen": 814688,
      "step": 169,
      "train_runtime": 565.2366,
      "train_tokens_per_second": 1441.322
    },
    {
      "epoch": 0.0625287356321839,
      "grad_norm": 0.33639150857925415,
      "learning_rate": 0.00018791451731761238,
      "loss": 1.2385,
      "num_input_tokens_seen": 823520,
      "step": 170,
      "train_runtime": 570.7503,
      "train_tokens_per_second": 1442.873
    },
    {
      "epoch": 0.06289655172413793,
      "grad_norm": 0.3430693447589874,
      "learning_rate": 0.00018784082535003687,
      "loss": 1.3859,
      "num_input_tokens_seen": 827872,
      "step": 171,
      "train_runtime": 573.7823,
      "train_tokens_per_second": 1442.833
    },
    {
      "epoch": 0.06326436781609196,
      "grad_norm": 0.3821863532066345,
      "learning_rate": 0.00018776713338246133,
      "loss": 1.4072,
      "num_input_tokens_seen": 832000,
      "step": 172,
      "train_runtime": 576.7272,
      "train_tokens_per_second": 1442.623
    },
    {
      "epoch": 0.06363218390804598,
      "grad_norm": 0.3700842261314392,
      "learning_rate": 0.0001876934414148858,
      "loss": 1.401,
      "num_input_tokens_seen": 838384,
      "step": 173,
      "train_runtime": 580.8296,
      "train_tokens_per_second": 1443.425
    },
    {
      "epoch": 0.064,
      "grad_norm": 0.3754938840866089,
      "learning_rate": 0.00018761974944731024,
      "loss": 1.3558,
      "num_input_tokens_seen": 842000,
      "step": 174,
      "train_runtime": 583.4802,
      "train_tokens_per_second": 1443.065
    },
    {
      "epoch": 0.06436781609195402,
      "grad_norm": 0.32714033126831055,
      "learning_rate": 0.00018754605747973473,
      "loss": 1.3782,
      "num_input_tokens_seen": 848800,
      "step": 175,
      "train_runtime": 587.9049,
      "train_tokens_per_second": 1443.771
    },
    {
      "epoch": 0.06473563218390804,
      "grad_norm": 0.3436740040779114,
      "learning_rate": 0.0001874723655121592,
      "loss": 1.4766,
      "num_input_tokens_seen": 855120,
      "step": 176,
      "train_runtime": 591.9989,
      "train_tokens_per_second": 1444.462
    },
    {
      "epoch": 0.06510344827586206,
      "grad_norm": 0.37590697407722473,
      "learning_rate": 0.00018739867354458366,
      "loss": 1.4055,
      "num_input_tokens_seen": 858960,
      "step": 177,
      "train_runtime": 594.7385,
      "train_tokens_per_second": 1444.265
    },
    {
      "epoch": 0.0654712643678161,
      "grad_norm": 0.36898568272590637,
      "learning_rate": 0.0001873249815770081,
      "loss": 1.3238,
      "num_input_tokens_seen": 864256,
      "step": 178,
      "train_runtime": 598.3052,
      "train_tokens_per_second": 1444.507
    },
    {
      "epoch": 0.06583908045977012,
      "grad_norm": 0.32526057958602905,
      "learning_rate": 0.00018725128960943259,
      "loss": 1.2297,
      "num_input_tokens_seen": 870912,
      "step": 179,
      "train_runtime": 602.5428,
      "train_tokens_per_second": 1445.394
    },
    {
      "epoch": 0.06620689655172414,
      "grad_norm": 0.34319373965263367,
      "learning_rate": 0.00018717759764185705,
      "loss": 1.3742,
      "num_input_tokens_seen": 874720,
      "step": 180,
      "train_runtime": 605.2113,
      "train_tokens_per_second": 1445.313
    },
    {
      "epoch": 0.06657471264367816,
      "grad_norm": 0.30405378341674805,
      "learning_rate": 0.00018710390567428152,
      "loss": 1.1638,
      "num_input_tokens_seen": 882656,
      "step": 181,
      "train_runtime": 612.2259,
      "train_tokens_per_second": 1441.716
    },
    {
      "epoch": 0.06694252873563218,
      "grad_norm": 0.3960118293762207,
      "learning_rate": 0.00018703021370670598,
      "loss": 1.2142,
      "num_input_tokens_seen": 887728,
      "step": 182,
      "train_runtime": 615.6804,
      "train_tokens_per_second": 1441.865
    },
    {
      "epoch": 0.0673103448275862,
      "grad_norm": 0.36837369203567505,
      "learning_rate": 0.00018695652173913045,
      "loss": 1.4049,
      "num_input_tokens_seen": 893824,
      "step": 183,
      "train_runtime": 619.6219,
      "train_tokens_per_second": 1442.531
    },
    {
      "epoch": 0.06767816091954024,
      "grad_norm": 0.3938405513763428,
      "learning_rate": 0.0001868828297715549,
      "loss": 1.3685,
      "num_input_tokens_seen": 898352,
      "step": 184,
      "train_runtime": 622.6986,
      "train_tokens_per_second": 1442.675
    },
    {
      "epoch": 0.06804597701149426,
      "grad_norm": 0.35122594237327576,
      "learning_rate": 0.00018680913780397938,
      "loss": 1.1811,
      "num_input_tokens_seen": 901744,
      "step": 185,
      "train_runtime": 625.1633,
      "train_tokens_per_second": 1442.413
    },
    {
      "epoch": 0.06841379310344828,
      "grad_norm": 0.38768187165260315,
      "learning_rate": 0.00018673544583640384,
      "loss": 1.4322,
      "num_input_tokens_seen": 906928,
      "step": 186,
      "train_runtime": 628.5985,
      "train_tokens_per_second": 1442.778
    },
    {
      "epoch": 0.0687816091954023,
      "grad_norm": 0.3452443778514862,
      "learning_rate": 0.0001866617538688283,
      "loss": 1.3319,
      "num_input_tokens_seen": 912016,
      "step": 187,
      "train_runtime": 632.0175,
      "train_tokens_per_second": 1443.023
    },
    {
      "epoch": 0.06914942528735632,
      "grad_norm": 0.33809736371040344,
      "learning_rate": 0.00018658806190125277,
      "loss": 1.3433,
      "num_input_tokens_seen": 918256,
      "step": 188,
      "train_runtime": 636.0156,
      "train_tokens_per_second": 1443.763
    },
    {
      "epoch": 0.06951724137931034,
      "grad_norm": 0.35644927620887756,
      "learning_rate": 0.00018651436993367723,
      "loss": 1.3148,
      "num_input_tokens_seen": 922656,
      "step": 189,
      "train_runtime": 639.0061,
      "train_tokens_per_second": 1443.892
    },
    {
      "epoch": 0.06988505747126436,
      "grad_norm": 0.34792211651802063,
      "learning_rate": 0.0001864406779661017,
      "loss": 1.323,
      "num_input_tokens_seen": 926560,
      "step": 190,
      "train_runtime": 641.8039,
      "train_tokens_per_second": 1443.681
    },
    {
      "epoch": 0.0702528735632184,
      "grad_norm": 0.3884986340999603,
      "learning_rate": 0.0001863669859985262,
      "loss": 1.2509,
      "num_input_tokens_seen": 931408,
      "step": 191,
      "train_runtime": 645.0317,
      "train_tokens_per_second": 1443.972
    },
    {
      "epoch": 0.07062068965517242,
      "grad_norm": 0.37056735157966614,
      "learning_rate": 0.00018629329403095063,
      "loss": 1.3281,
      "num_input_tokens_seen": 936560,
      "step": 192,
      "train_runtime": 648.4631,
      "train_tokens_per_second": 1444.277
    },
    {
      "epoch": 0.07098850574712644,
      "grad_norm": 0.32564520835876465,
      "learning_rate": 0.0001862196020633751,
      "loss": 1.3638,
      "num_input_tokens_seen": 944752,
      "step": 193,
      "train_runtime": 653.5674,
      "train_tokens_per_second": 1445.531
    },
    {
      "epoch": 0.07135632183908046,
      "grad_norm": 0.37514346837997437,
      "learning_rate": 0.00018614591009579956,
      "loss": 1.319,
      "num_input_tokens_seen": 948752,
      "step": 194,
      "train_runtime": 656.4037,
      "train_tokens_per_second": 1445.379
    },
    {
      "epoch": 0.07172413793103448,
      "grad_norm": 0.4033873975276947,
      "learning_rate": 0.00018607221812822405,
      "loss": 1.2798,
      "num_input_tokens_seen": 953440,
      "step": 195,
      "train_runtime": 659.5663,
      "train_tokens_per_second": 1445.556
    },
    {
      "epoch": 0.0720919540229885,
      "grad_norm": 0.3493629992008209,
      "learning_rate": 0.0001859985261606485,
      "loss": 1.2205,
      "num_input_tokens_seen": 959104,
      "step": 196,
      "train_runtime": 663.291,
      "train_tokens_per_second": 1445.978
    },
    {
      "epoch": 0.07245977011494253,
      "grad_norm": 0.3406820297241211,
      "learning_rate": 0.00018592483419307295,
      "loss": 1.3426,
      "num_input_tokens_seen": 964304,
      "step": 197,
      "train_runtime": 666.7029,
      "train_tokens_per_second": 1446.377
    },
    {
      "epoch": 0.07282758620689656,
      "grad_norm": 0.3422794044017792,
      "learning_rate": 0.00018585114222549742,
      "loss": 1.2411,
      "num_input_tokens_seen": 967392,
      "step": 198,
      "train_runtime": 669.0131,
      "train_tokens_per_second": 1445.999
    },
    {
      "epoch": 0.07319540229885058,
      "grad_norm": 0.4126717448234558,
      "learning_rate": 0.0001857774502579219,
      "loss": 1.5401,
      "num_input_tokens_seen": 971648,
      "step": 199,
      "train_runtime": 671.9361,
      "train_tokens_per_second": 1446.042
    },
    {
      "epoch": 0.0735632183908046,
      "grad_norm": 0.45348045229911804,
      "learning_rate": 0.00018570375829034637,
      "loss": 1.2581,
      "num_input_tokens_seen": 975040,
      "step": 200,
      "train_runtime": 674.4174,
      "train_tokens_per_second": 1445.752
    },
    {
      "epoch": 0.07393103448275862,
      "grad_norm": 0.3561207056045532,
      "learning_rate": 0.0001856300663227708,
      "loss": 1.3483,
      "num_input_tokens_seen": 980496,
      "step": 201,
      "train_runtime": 678.0653,
      "train_tokens_per_second": 1446.02
    },
    {
      "epoch": 0.07429885057471264,
      "grad_norm": 0.3428356945514679,
      "learning_rate": 0.00018555637435519528,
      "loss": 1.4107,
      "num_input_tokens_seen": 984352,
      "step": 202,
      "train_runtime": 680.8003,
      "train_tokens_per_second": 1445.875
    },
    {
      "epoch": 0.07466666666666667,
      "grad_norm": 0.3632093667984009,
      "learning_rate": 0.00018548268238761977,
      "loss": 1.4813,
      "num_input_tokens_seen": 989776,
      "step": 203,
      "train_runtime": 684.3976,
      "train_tokens_per_second": 1446.2
    },
    {
      "epoch": 0.0750344827586207,
      "grad_norm": 0.3488505780696869,
      "learning_rate": 0.00018540899042004423,
      "loss": 1.1855,
      "num_input_tokens_seen": 993376,
      "step": 204,
      "train_runtime": 686.9903,
      "train_tokens_per_second": 1445.983
    },
    {
      "epoch": 0.07540229885057471,
      "grad_norm": 0.3393392264842987,
      "learning_rate": 0.0001853352984524687,
      "loss": 1.4086,
      "num_input_tokens_seen": 997376,
      "step": 205,
      "train_runtime": 689.8343,
      "train_tokens_per_second": 1445.82
    },
    {
      "epoch": 0.07577011494252874,
      "grad_norm": 0.34985506534576416,
      "learning_rate": 0.00018526160648489314,
      "loss": 1.2578,
      "num_input_tokens_seen": 1001456,
      "step": 206,
      "train_runtime": 692.7212,
      "train_tokens_per_second": 1445.684
    },
    {
      "epoch": 0.07613793103448276,
      "grad_norm": 0.3500637114048004,
      "learning_rate": 0.00018518791451731763,
      "loss": 1.2264,
      "num_input_tokens_seen": 1009216,
      "step": 207,
      "train_runtime": 697.551,
      "train_tokens_per_second": 1446.799
    },
    {
      "epoch": 0.07650574712643678,
      "grad_norm": 0.38763341307640076,
      "learning_rate": 0.0001851142225497421,
      "loss": 1.515,
      "num_input_tokens_seen": 1014288,
      "step": 208,
      "train_runtime": 701.0069,
      "train_tokens_per_second": 1446.902
    },
    {
      "epoch": 0.07687356321839081,
      "grad_norm": 0.34014877676963806,
      "learning_rate": 0.00018504053058216656,
      "loss": 1.2957,
      "num_input_tokens_seen": 1018640,
      "step": 209,
      "train_runtime": 704.0858,
      "train_tokens_per_second": 1446.756
    },
    {
      "epoch": 0.07724137931034483,
      "grad_norm": 0.41478997468948364,
      "learning_rate": 0.00018496683861459102,
      "loss": 1.3014,
      "num_input_tokens_seen": 1022784,
      "step": 210,
      "train_runtime": 706.9888,
      "train_tokens_per_second": 1446.676
    },
    {
      "epoch": 0.07760919540229885,
      "grad_norm": 0.3496479392051697,
      "learning_rate": 0.0001848931466470155,
      "loss": 1.3461,
      "num_input_tokens_seen": 1027360,
      "step": 211,
      "train_runtime": 711.8041,
      "train_tokens_per_second": 1443.318
    },
    {
      "epoch": 0.07797701149425287,
      "grad_norm": 0.3588869571685791,
      "learning_rate": 0.00018481945467943995,
      "loss": 1.4353,
      "num_input_tokens_seen": 1033456,
      "step": 212,
      "train_runtime": 715.6723,
      "train_tokens_per_second": 1444.035
    },
    {
      "epoch": 0.0783448275862069,
      "grad_norm": 0.35271963477134705,
      "learning_rate": 0.00018474576271186442,
      "loss": 1.0946,
      "num_input_tokens_seen": 1036304,
      "step": 213,
      "train_runtime": 717.8905,
      "train_tokens_per_second": 1443.54
    },
    {
      "epoch": 0.07871264367816092,
      "grad_norm": 0.39200204610824585,
      "learning_rate": 0.00018467207074428888,
      "loss": 1.15,
      "num_input_tokens_seen": 1039936,
      "step": 214,
      "train_runtime": 720.4172,
      "train_tokens_per_second": 1443.519
    },
    {
      "epoch": 0.07908045977011494,
      "grad_norm": 0.4076039791107178,
      "learning_rate": 0.00018459837877671335,
      "loss": 1.3758,
      "num_input_tokens_seen": 1045536,
      "step": 215,
      "train_runtime": 724.1097,
      "train_tokens_per_second": 1443.892
    },
    {
      "epoch": 0.07944827586206897,
      "grad_norm": 0.37207478284835815,
      "learning_rate": 0.0001845246868091378,
      "loss": 1.3664,
      "num_input_tokens_seen": 1049056,
      "step": 216,
      "train_runtime": 726.6894,
      "train_tokens_per_second": 1443.61
    },
    {
      "epoch": 0.07981609195402299,
      "grad_norm": 0.3728545904159546,
      "learning_rate": 0.00018445099484156228,
      "loss": 1.3072,
      "num_input_tokens_seen": 1052304,
      "step": 217,
      "train_runtime": 729.0769,
      "train_tokens_per_second": 1443.338
    },
    {
      "epoch": 0.08018390804597701,
      "grad_norm": 0.3534899652004242,
      "learning_rate": 0.00018437730287398674,
      "loss": 1.2736,
      "num_input_tokens_seen": 1057936,
      "step": 218,
      "train_runtime": 732.8309,
      "train_tokens_per_second": 1443.629
    },
    {
      "epoch": 0.08055172413793103,
      "grad_norm": 0.3420320749282837,
      "learning_rate": 0.0001843036109064112,
      "loss": 1.2494,
      "num_input_tokens_seen": 1061312,
      "step": 219,
      "train_runtime": 735.2926,
      "train_tokens_per_second": 1443.387
    },
    {
      "epoch": 0.08091954022988505,
      "grad_norm": 0.34126317501068115,
      "learning_rate": 0.00018422991893883567,
      "loss": 1.257,
      "num_input_tokens_seen": 1066144,
      "step": 220,
      "train_runtime": 738.6091,
      "train_tokens_per_second": 1443.448
    },
    {
      "epoch": 0.08128735632183907,
      "grad_norm": 0.34283459186553955,
      "learning_rate": 0.00018415622697126014,
      "loss": 1.1256,
      "num_input_tokens_seen": 1071584,
      "step": 221,
      "train_runtime": 742.1626,
      "train_tokens_per_second": 1443.867
    },
    {
      "epoch": 0.08165517241379311,
      "grad_norm": 0.33624303340911865,
      "learning_rate": 0.0001840825350036846,
      "loss": 1.1587,
      "num_input_tokens_seen": 1076064,
      "step": 222,
      "train_runtime": 745.1788,
      "train_tokens_per_second": 1444.035
    },
    {
      "epoch": 0.08202298850574713,
      "grad_norm": 0.3778267800807953,
      "learning_rate": 0.0001840088430361091,
      "loss": 1.2877,
      "num_input_tokens_seen": 1079504,
      "step": 223,
      "train_runtime": 747.7495,
      "train_tokens_per_second": 1443.671
    },
    {
      "epoch": 0.08239080459770115,
      "grad_norm": 0.37978050112724304,
      "learning_rate": 0.00018393515106853353,
      "loss": 1.425,
      "num_input_tokens_seen": 1084768,
      "step": 224,
      "train_runtime": 751.1746,
      "train_tokens_per_second": 1444.096
    },
    {
      "epoch": 0.08275862068965517,
      "grad_norm": 0.36541327834129333,
      "learning_rate": 0.000183861459100958,
      "loss": 1.2708,
      "num_input_tokens_seen": 1088720,
      "step": 225,
      "train_runtime": 753.9348,
      "train_tokens_per_second": 1444.051
    },
    {
      "epoch": 0.08312643678160919,
      "grad_norm": 0.31823718547821045,
      "learning_rate": 0.00018378776713338246,
      "loss": 1.3592,
      "num_input_tokens_seen": 1093152,
      "step": 226,
      "train_runtime": 757.0196,
      "train_tokens_per_second": 1444.021
    },
    {
      "epoch": 0.08349425287356321,
      "grad_norm": 0.35182780027389526,
      "learning_rate": 0.00018371407516580695,
      "loss": 1.3989,
      "num_input_tokens_seen": 1097904,
      "step": 227,
      "train_runtime": 760.2168,
      "train_tokens_per_second": 1444.199
    },
    {
      "epoch": 0.08386206896551725,
      "grad_norm": 0.3908786475658417,
      "learning_rate": 0.0001836403831982314,
      "loss": 1.1939,
      "num_input_tokens_seen": 1101776,
      "step": 228,
      "train_runtime": 762.9056,
      "train_tokens_per_second": 1444.184
    },
    {
      "epoch": 0.08422988505747127,
      "grad_norm": 0.3624025285243988,
      "learning_rate": 0.00018356669123065585,
      "loss": 1.3764,
      "num_input_tokens_seen": 1108368,
      "step": 229,
      "train_runtime": 767.1154,
      "train_tokens_per_second": 1444.852
    },
    {
      "epoch": 0.08459770114942529,
      "grad_norm": 0.41629868745803833,
      "learning_rate": 0.00018349299926308032,
      "loss": 1.2741,
      "num_input_tokens_seen": 1111984,
      "step": 230,
      "train_runtime": 769.7166,
      "train_tokens_per_second": 1444.667
    },
    {
      "epoch": 0.08496551724137931,
      "grad_norm": 0.39580047130584717,
      "learning_rate": 0.0001834193072955048,
      "loss": 1.4093,
      "num_input_tokens_seen": 1117008,
      "step": 231,
      "train_runtime": 773.035,
      "train_tokens_per_second": 1444.964
    },
    {
      "epoch": 0.08533333333333333,
      "grad_norm": 0.3515931963920593,
      "learning_rate": 0.00018334561532792928,
      "loss": 1.2854,
      "num_input_tokens_seen": 1120448,
      "step": 232,
      "train_runtime": 775.5121,
      "train_tokens_per_second": 1444.785
    },
    {
      "epoch": 0.08570114942528735,
      "grad_norm": 0.34929874539375305,
      "learning_rate": 0.00018327192336035371,
      "loss": 1.2306,
      "num_input_tokens_seen": 1124304,
      "step": 233,
      "train_runtime": 778.2103,
      "train_tokens_per_second": 1444.73
    },
    {
      "epoch": 0.08606896551724139,
      "grad_norm": 0.3431086540222168,
      "learning_rate": 0.00018319823139277818,
      "loss": 1.3612,
      "num_input_tokens_seen": 1130192,
      "step": 234,
      "train_runtime": 782.0628,
      "train_tokens_per_second": 1445.142
    },
    {
      "epoch": 0.0864367816091954,
      "grad_norm": 0.3813183605670929,
      "learning_rate": 0.00018312453942520267,
      "loss": 1.2279,
      "num_input_tokens_seen": 1135392,
      "step": 235,
      "train_runtime": 785.5012,
      "train_tokens_per_second": 1445.436
    },
    {
      "epoch": 0.08680459770114943,
      "grad_norm": 0.44352656602859497,
      "learning_rate": 0.00018305084745762714,
      "loss": 1.3339,
      "num_input_tokens_seen": 1139280,
      "step": 236,
      "train_runtime": 788.2441,
      "train_tokens_per_second": 1445.339
    },
    {
      "epoch": 0.08717241379310345,
      "grad_norm": 0.38422566652297974,
      "learning_rate": 0.0001829771554900516,
      "loss": 1.2985,
      "num_input_tokens_seen": 1143744,
      "step": 237,
      "train_runtime": 791.3359,
      "train_tokens_per_second": 1445.333
    },
    {
      "epoch": 0.08754022988505747,
      "grad_norm": 0.3571647107601166,
      "learning_rate": 0.00018290346352247604,
      "loss": 1.4091,
      "num_input_tokens_seen": 1148784,
      "step": 238,
      "train_runtime": 794.709,
      "train_tokens_per_second": 1445.54
    },
    {
      "epoch": 0.08790804597701149,
      "grad_norm": 0.4143889844417572,
      "learning_rate": 0.00018282977155490053,
      "loss": 1.3371,
      "num_input_tokens_seen": 1153568,
      "step": 239,
      "train_runtime": 797.9099,
      "train_tokens_per_second": 1445.737
    },
    {
      "epoch": 0.08827586206896551,
      "grad_norm": 0.399534672498703,
      "learning_rate": 0.000182756079587325,
      "loss": 1.2132,
      "num_input_tokens_seen": 1158576,
      "step": 240,
      "train_runtime": 801.2734,
      "train_tokens_per_second": 1445.919
    },
    {
      "epoch": 0.08864367816091955,
      "grad_norm": 0.36167678236961365,
      "learning_rate": 0.00018268238761974946,
      "loss": 1.2518,
      "num_input_tokens_seen": 1163104,
      "step": 241,
      "train_runtime": 806.239,
      "train_tokens_per_second": 1442.629
    },
    {
      "epoch": 0.08901149425287357,
      "grad_norm": 0.32895228266716003,
      "learning_rate": 0.00018260869565217392,
      "loss": 1.3456,
      "num_input_tokens_seen": 1171792,
      "step": 242,
      "train_runtime": 811.5482,
      "train_tokens_per_second": 1443.897
    },
    {
      "epoch": 0.08937931034482759,
      "grad_norm": 0.3777807354927063,
      "learning_rate": 0.0001825350036845984,
      "loss": 1.3755,
      "num_input_tokens_seen": 1175696,
      "step": 243,
      "train_runtime": 814.3467,
      "train_tokens_per_second": 1443.729
    },
    {
      "epoch": 0.08974712643678161,
      "grad_norm": 0.39932459592819214,
      "learning_rate": 0.00018246131171702285,
      "loss": 1.3449,
      "num_input_tokens_seen": 1179888,
      "step": 244,
      "train_runtime": 817.3093,
      "train_tokens_per_second": 1443.625
    },
    {
      "epoch": 0.09011494252873563,
      "grad_norm": 0.4072055518627167,
      "learning_rate": 0.00018238761974944732,
      "loss": 1.5742,
      "num_input_tokens_seen": 1185408,
      "step": 245,
      "train_runtime": 820.9659,
      "train_tokens_per_second": 1443.919
    },
    {
      "epoch": 0.09048275862068965,
      "grad_norm": 0.31552496552467346,
      "learning_rate": 0.00018231392778187178,
      "loss": 1.3173,
      "num_input_tokens_seen": 1188960,
      "step": 246,
      "train_runtime": 823.5105,
      "train_tokens_per_second": 1443.77
    },
    {
      "epoch": 0.09085057471264368,
      "grad_norm": 0.3510066866874695,
      "learning_rate": 0.00018224023581429625,
      "loss": 1.46,
      "num_input_tokens_seen": 1192240,
      "step": 247,
      "train_runtime": 825.9115,
      "train_tokens_per_second": 1443.544
    },
    {
      "epoch": 0.0912183908045977,
      "grad_norm": 0.3247656226158142,
      "learning_rate": 0.0001821665438467207,
      "loss": 1.3657,
      "num_input_tokens_seen": 1196592,
      "step": 248,
      "train_runtime": 828.9675,
      "train_tokens_per_second": 1443.473
    },
    {
      "epoch": 0.09158620689655173,
      "grad_norm": 0.4057351350784302,
      "learning_rate": 0.00018209285187914518,
      "loss": 1.451,
      "num_input_tokens_seen": 1200816,
      "step": 249,
      "train_runtime": 831.8998,
      "train_tokens_per_second": 1443.462
    },
    {
      "epoch": 0.09195402298850575,
      "grad_norm": 0.31586751341819763,
      "learning_rate": 0.00018201915991156964,
      "loss": 1.2866,
      "num_input_tokens_seen": 1205264,
      "step": 250,
      "train_runtime": 834.9428,
      "train_tokens_per_second": 1443.529
    },
    {
      "epoch": 0.09232183908045977,
      "grad_norm": 0.35121411085128784,
      "learning_rate": 0.0001819454679439941,
      "loss": 1.4238,
      "num_input_tokens_seen": 1208912,
      "step": 251,
      "train_runtime": 837.6257,
      "train_tokens_per_second": 1443.26
    },
    {
      "epoch": 0.09268965517241379,
      "grad_norm": 0.39062920212745667,
      "learning_rate": 0.00018187177597641857,
      "loss": 1.3669,
      "num_input_tokens_seen": 1212624,
      "step": 252,
      "train_runtime": 840.3025,
      "train_tokens_per_second": 1443.08
    },
    {
      "epoch": 0.09305747126436782,
      "grad_norm": 0.32011306285858154,
      "learning_rate": 0.00018179808400884304,
      "loss": 1.2237,
      "num_input_tokens_seen": 1217712,
      "step": 253,
      "train_runtime": 843.6832,
      "train_tokens_per_second": 1443.329
    },
    {
      "epoch": 0.09342528735632184,
      "grad_norm": 0.3875138461589813,
      "learning_rate": 0.00018172439204126753,
      "loss": 1.4564,
      "num_input_tokens_seen": 1223008,
      "step": 254,
      "train_runtime": 847.2923,
      "train_tokens_per_second": 1443.431
    },
    {
      "epoch": 0.09379310344827586,
      "grad_norm": 0.3403080999851227,
      "learning_rate": 0.000181650700073692,
      "loss": 1.3242,
      "num_input_tokens_seen": 1226864,
      "step": 255,
      "train_runtime": 849.9705,
      "train_tokens_per_second": 1443.419
    },
    {
      "epoch": 0.09416091954022988,
      "grad_norm": 0.36226019263267517,
      "learning_rate": 0.00018157700810611643,
      "loss": 1.2468,
      "num_input_tokens_seen": 1235056,
      "step": 256,
      "train_runtime": 855.0662,
      "train_tokens_per_second": 1444.398
    },
    {
      "epoch": 0.0945287356321839,
      "grad_norm": 0.377728670835495,
      "learning_rate": 0.0001815033161385409,
      "loss": 1.3328,
      "num_input_tokens_seen": 1240816,
      "step": 257,
      "train_runtime": 858.8108,
      "train_tokens_per_second": 1444.807
    },
    {
      "epoch": 0.09489655172413793,
      "grad_norm": 0.380744069814682,
      "learning_rate": 0.0001814296241709654,
      "loss": 1.2312,
      "num_input_tokens_seen": 1244576,
      "step": 258,
      "train_runtime": 861.483,
      "train_tokens_per_second": 1444.69
    },
    {
      "epoch": 0.09526436781609196,
      "grad_norm": 0.3609914481639862,
      "learning_rate": 0.00018135593220338985,
      "loss": 1.3088,
      "num_input_tokens_seen": 1248608,
      "step": 259,
      "train_runtime": 864.353,
      "train_tokens_per_second": 1444.558
    },
    {
      "epoch": 0.09563218390804598,
      "grad_norm": 0.35212814807891846,
      "learning_rate": 0.00018128224023581432,
      "loss": 1.3645,
      "num_input_tokens_seen": 1252336,
      "step": 260,
      "train_runtime": 867.1025,
      "train_tokens_per_second": 1444.277
    },
    {
      "epoch": 0.096,
      "grad_norm": 0.30638137459754944,
      "learning_rate": 0.00018120854826823876,
      "loss": 1.4044,
      "num_input_tokens_seen": 1261744,
      "step": 261,
      "train_runtime": 872.8491,
      "train_tokens_per_second": 1445.547
    },
    {
      "epoch": 0.09636781609195402,
      "grad_norm": 0.3573840856552124,
      "learning_rate": 0.00018113485630066325,
      "loss": 1.2195,
      "num_input_tokens_seen": 1265440,
      "step": 262,
      "train_runtime": 875.5308,
      "train_tokens_per_second": 1445.34
    },
    {
      "epoch": 0.09673563218390804,
      "grad_norm": 0.3774884343147278,
      "learning_rate": 0.0001810611643330877,
      "loss": 1.3894,
      "num_input_tokens_seen": 1269168,
      "step": 263,
      "train_runtime": 878.2173,
      "train_tokens_per_second": 1445.164
    },
    {
      "epoch": 0.09710344827586206,
      "grad_norm": 0.410137802362442,
      "learning_rate": 0.00018098747236551218,
      "loss": 1.2193,
      "num_input_tokens_seen": 1274224,
      "step": 264,
      "train_runtime": 881.5883,
      "train_tokens_per_second": 1445.373
    },
    {
      "epoch": 0.09747126436781609,
      "grad_norm": 0.3597395420074463,
      "learning_rate": 0.00018091378039793662,
      "loss": 1.3967,
      "num_input_tokens_seen": 1278160,
      "step": 265,
      "train_runtime": 884.3763,
      "train_tokens_per_second": 1445.267
    },
    {
      "epoch": 0.09783908045977012,
      "grad_norm": 0.3798868656158447,
      "learning_rate": 0.0001808400884303611,
      "loss": 1.3412,
      "num_input_tokens_seen": 1282352,
      "step": 266,
      "train_runtime": 887.2775,
      "train_tokens_per_second": 1445.266
    },
    {
      "epoch": 0.09820689655172414,
      "grad_norm": 0.3923279345035553,
      "learning_rate": 0.00018076639646278557,
      "loss": 1.346,
      "num_input_tokens_seen": 1287392,
      "step": 267,
      "train_runtime": 890.7253,
      "train_tokens_per_second": 1445.33
    },
    {
      "epoch": 0.09857471264367816,
      "grad_norm": 0.41670191287994385,
      "learning_rate": 0.00018069270449521004,
      "loss": 1.2068,
      "num_input_tokens_seen": 1290848,
      "step": 268,
      "train_runtime": 893.2523,
      "train_tokens_per_second": 1445.11
    },
    {
      "epoch": 0.09894252873563218,
      "grad_norm": 0.3499112129211426,
      "learning_rate": 0.0001806190125276345,
      "loss": 1.1829,
      "num_input_tokens_seen": 1295056,
      "step": 269,
      "train_runtime": 896.1665,
      "train_tokens_per_second": 1445.106
    },
    {
      "epoch": 0.0993103448275862,
      "grad_norm": 0.3606136441230774,
      "learning_rate": 0.00018054532056005897,
      "loss": 1.3168,
      "num_input_tokens_seen": 1300800,
      "step": 270,
      "train_runtime": 899.9738,
      "train_tokens_per_second": 1445.375
    },
    {
      "epoch": 0.09967816091954022,
      "grad_norm": 0.3754916191101074,
      "learning_rate": 0.00018047162859248343,
      "loss": 1.4176,
      "num_input_tokens_seen": 1304848,
      "step": 271,
      "train_runtime": 904.7959,
      "train_tokens_per_second": 1442.146
    },
    {
      "epoch": 0.10004597701149426,
      "grad_norm": 0.40887537598609924,
      "learning_rate": 0.0001803979366249079,
      "loss": 1.5235,
      "num_input_tokens_seen": 1309296,
      "step": 272,
      "train_runtime": 907.8267,
      "train_tokens_per_second": 1442.231
    },
    {
      "epoch": 0.10004597701149426,
      "eval_loss": 1.3052144050598145,
      "eval_runtime": 15.5033,
      "eval_samples_per_second": 64.503,
      "eval_steps_per_second": 4.064,
      "num_input_tokens_seen": 1309296,
      "step": 272
    },
    {
      "epoch": 0.10041379310344828,
      "grad_norm": 0.4006941318511963,
      "learning_rate": 0.00018032424465733236,
      "loss": 1.2035,
      "num_input_tokens_seen": 1313280,
      "step": 273,
      "train_runtime": 926.1272,
      "train_tokens_per_second": 1418.034
    },
    {
      "epoch": 0.1007816091954023,
      "grad_norm": 0.3529790937900543,
      "learning_rate": 0.00018025055268975683,
      "loss": 1.2604,
      "num_input_tokens_seen": 1318528,
      "step": 274,
      "train_runtime": 929.6168,
      "train_tokens_per_second": 1418.356
    },
    {
      "epoch": 0.10114942528735632,
      "grad_norm": 0.403448224067688,
      "learning_rate": 0.0001801768607221813,
      "loss": 1.34,
      "num_input_tokens_seen": 1323008,
      "step": 275,
      "train_runtime": 932.6673,
      "train_tokens_per_second": 1418.521
    },
    {
      "epoch": 0.10151724137931034,
      "grad_norm": 0.34158387780189514,
      "learning_rate": 0.00018010316875460576,
      "loss": 1.3079,
      "num_input_tokens_seen": 1327296,
      "step": 276,
      "train_runtime": 935.6092,
      "train_tokens_per_second": 1418.644
    },
    {
      "epoch": 0.10188505747126436,
      "grad_norm": 0.3838740885257721,
      "learning_rate": 0.00018002947678703022,
      "loss": 1.4125,
      "num_input_tokens_seen": 1332400,
      "step": 277,
      "train_runtime": 939.0933,
      "train_tokens_per_second": 1418.815
    },
    {
      "epoch": 0.1022528735632184,
      "grad_norm": 0.3493935763835907,
      "learning_rate": 0.00017995578481945468,
      "loss": 1.2636,
      "num_input_tokens_seen": 1340480,
      "step": 278,
      "train_runtime": 944.1631,
      "train_tokens_per_second": 1419.755
    },
    {
      "epoch": 0.10262068965517242,
      "grad_norm": 0.3681679666042328,
      "learning_rate": 0.00017988209285187915,
      "loss": 1.3285,
      "num_input_tokens_seen": 1344400,
      "step": 279,
      "train_runtime": 946.8945,
      "train_tokens_per_second": 1419.799
    },
    {
      "epoch": 0.10298850574712644,
      "grad_norm": 0.3373294472694397,
      "learning_rate": 0.00017980840088430361,
      "loss": 1.3849,
      "num_input_tokens_seen": 1348496,
      "step": 280,
      "train_runtime": 949.8071,
      "train_tokens_per_second": 1419.758
    },
    {
      "epoch": 0.10335632183908046,
      "grad_norm": 0.32788580656051636,
      "learning_rate": 0.00017973470891672808,
      "loss": 1.2813,
      "num_input_tokens_seen": 1357056,
      "step": 281,
      "train_runtime": 955.1543,
      "train_tokens_per_second": 1420.771
    },
    {
      "epoch": 0.10372413793103448,
      "grad_norm": 0.3648260831832886,
      "learning_rate": 0.00017966101694915257,
      "loss": 1.17,
      "num_input_tokens_seen": 1362240,
      "step": 282,
      "train_runtime": 958.7496,
      "train_tokens_per_second": 1420.851
    },
    {
      "epoch": 0.1040919540229885,
      "grad_norm": 0.3272045850753784,
      "learning_rate": 0.000179587324981577,
      "loss": 1.2861,
      "num_input_tokens_seen": 1369136,
      "step": 283,
      "train_runtime": 963.1842,
      "train_tokens_per_second": 1421.469
    },
    {
      "epoch": 0.10445977011494254,
      "grad_norm": 0.4226033091545105,
      "learning_rate": 0.00017951363301400147,
      "loss": 1.4487,
      "num_input_tokens_seen": 1374848,
      "step": 284,
      "train_runtime": 966.9991,
      "train_tokens_per_second": 1421.768
    },
    {
      "epoch": 0.10482758620689656,
      "grad_norm": 0.3636164665222168,
      "learning_rate": 0.00017943994104642594,
      "loss": 1.3257,
      "num_input_tokens_seen": 1379760,
      "step": 285,
      "train_runtime": 970.2296,
      "train_tokens_per_second": 1422.096
    },
    {
      "epoch": 0.10519540229885058,
      "grad_norm": 0.3344646394252777,
      "learning_rate": 0.00017936624907885043,
      "loss": 1.4043,
      "num_input_tokens_seen": 1385920,
      "step": 286,
      "train_runtime": 974.1866,
      "train_tokens_per_second": 1422.643
    },
    {
      "epoch": 0.1055632183908046,
      "grad_norm": 0.35804253816604614,
      "learning_rate": 0.0001792925571112749,
      "loss": 1.3529,
      "num_input_tokens_seen": 1389824,
      "step": 287,
      "train_runtime": 976.8889,
      "train_tokens_per_second": 1422.704
    },
    {
      "epoch": 0.10593103448275862,
      "grad_norm": 0.36252444982528687,
      "learning_rate": 0.00017921886514369933,
      "loss": 1.3426,
      "num_input_tokens_seen": 1395008,
      "step": 288,
      "train_runtime": 980.2779,
      "train_tokens_per_second": 1423.074
    },
    {
      "epoch": 0.10629885057471264,
      "grad_norm": 0.41811102628707886,
      "learning_rate": 0.0001791451731761238,
      "loss": 1.4098,
      "num_input_tokens_seen": 1403952,
      "step": 289,
      "train_runtime": 985.7571,
      "train_tokens_per_second": 1424.237
    },
    {
      "epoch": 0.10666666666666667,
      "grad_norm": 0.41715559363365173,
      "learning_rate": 0.0001790714812085483,
      "loss": 1.3552,
      "num_input_tokens_seen": 1408304,
      "step": 290,
      "train_runtime": 988.6836,
      "train_tokens_per_second": 1424.423
    },
    {
      "epoch": 0.1070344827586207,
      "grad_norm": 0.4006524682044983,
      "learning_rate": 0.00017899778924097275,
      "loss": 1.3274,
      "num_input_tokens_seen": 1413904,
      "step": 291,
      "train_runtime": 992.4063,
      "train_tokens_per_second": 1424.723
    },
    {
      "epoch": 0.10740229885057472,
      "grad_norm": 0.37396982312202454,
      "learning_rate": 0.00017892409727339722,
      "loss": 1.2964,
      "num_input_tokens_seen": 1422160,
      "step": 292,
      "train_runtime": 997.5305,
      "train_tokens_per_second": 1425.681
    },
    {
      "epoch": 0.10777011494252874,
      "grad_norm": 0.3324640691280365,
      "learning_rate": 0.00017885040530582166,
      "loss": 1.2427,
      "num_input_tokens_seen": 1425712,
      "step": 293,
      "train_runtime": 1000.0263,
      "train_tokens_per_second": 1425.674
    },
    {
      "epoch": 0.10813793103448276,
      "grad_norm": 0.33719897270202637,
      "learning_rate": 0.00017877671333824615,
      "loss": 1.3184,
      "num_input_tokens_seen": 1430064,
      "step": 294,
      "train_runtime": 1002.9771,
      "train_tokens_per_second": 1425.819
    },
    {
      "epoch": 0.10850574712643678,
      "grad_norm": 0.3992919623851776,
      "learning_rate": 0.00017870302137067061,
      "loss": 1.2449,
      "num_input_tokens_seen": 1433520,
      "step": 295,
      "train_runtime": 1005.4712,
      "train_tokens_per_second": 1425.72
    },
    {
      "epoch": 0.1088735632183908,
      "grad_norm": 0.4201892614364624,
      "learning_rate": 0.00017862932940309508,
      "loss": 1.2702,
      "num_input_tokens_seen": 1438624,
      "step": 296,
      "train_runtime": 1008.8583,
      "train_tokens_per_second": 1425.992
    },
    {
      "epoch": 0.10924137931034483,
      "grad_norm": 0.4159906506538391,
      "learning_rate": 0.00017855563743551952,
      "loss": 1.4088,
      "num_input_tokens_seen": 1443616,
      "step": 297,
      "train_runtime": 1012.2013,
      "train_tokens_per_second": 1426.214
    },
    {
      "epoch": 0.10960919540229885,
      "grad_norm": 0.35260722041130066,
      "learning_rate": 0.000178481945467944,
      "loss": 1.3253,
      "num_input_tokens_seen": 1449504,
      "step": 298,
      "train_runtime": 1015.9359,
      "train_tokens_per_second": 1426.767
    },
    {
      "epoch": 0.10997701149425287,
      "grad_norm": 0.3388993740081787,
      "learning_rate": 0.00017840825350036847,
      "loss": 1.4653,
      "num_input_tokens_seen": 1454432,
      "step": 299,
      "train_runtime": 1019.1331,
      "train_tokens_per_second": 1427.127
    },
    {
      "epoch": 0.1103448275862069,
      "grad_norm": 0.3940805494785309,
      "learning_rate": 0.00017833456153279294,
      "loss": 1.2955,
      "num_input_tokens_seen": 1459664,
      "step": 300,
      "train_runtime": 1022.638,
      "train_tokens_per_second": 1427.352
    },
    {
      "epoch": 0.11071264367816092,
      "grad_norm": 0.3170969486236572,
      "learning_rate": 0.0001782608695652174,
      "loss": 1.1511,
      "num_input_tokens_seen": 1462800,
      "step": 301,
      "train_runtime": 1026.9756,
      "train_tokens_per_second": 1424.377
    },
    {
      "epoch": 0.11108045977011494,
      "grad_norm": 0.3472290337085724,
      "learning_rate": 0.00017818717759764187,
      "loss": 1.2986,
      "num_input_tokens_seen": 1468608,
      "step": 302,
      "train_runtime": 1030.7486,
      "train_tokens_per_second": 1424.798
    },
    {
      "epoch": 0.11144827586206897,
      "grad_norm": 0.33495432138442993,
      "learning_rate": 0.00017811348563006633,
      "loss": 1.2575,
      "num_input_tokens_seen": 1472784,
      "step": 303,
      "train_runtime": 1033.6794,
      "train_tokens_per_second": 1424.798
    },
    {
      "epoch": 0.11181609195402299,
      "grad_norm": 0.36378833651542664,
      "learning_rate": 0.0001780397936624908,
      "loss": 1.1909,
      "num_input_tokens_seen": 1477312,
      "step": 304,
      "train_runtime": 1036.7518,
      "train_tokens_per_second": 1424.943
    },
    {
      "epoch": 0.11218390804597701,
      "grad_norm": 0.3660285770893097,
      "learning_rate": 0.00017796610169491526,
      "loss": 1.2604,
      "num_input_tokens_seen": 1484480,
      "step": 305,
      "train_runtime": 1041.3434,
      "train_tokens_per_second": 1425.543
    },
    {
      "epoch": 0.11255172413793103,
      "grad_norm": 0.3172094523906708,
      "learning_rate": 0.00017789240972733973,
      "loss": 1.2402,
      "num_input_tokens_seen": 1494912,
      "step": 306,
      "train_runtime": 1047.642,
      "train_tokens_per_second": 1426.93
    },
    {
      "epoch": 0.11291954022988505,
      "grad_norm": 0.3556981384754181,
      "learning_rate": 0.0001778187177597642,
      "loss": 1.2463,
      "num_input_tokens_seen": 1503152,
      "step": 307,
      "train_runtime": 1052.8105,
      "train_tokens_per_second": 1427.752
    },
    {
      "epoch": 0.11328735632183908,
      "grad_norm": 0.37304648756980896,
      "learning_rate": 0.00017774502579218866,
      "loss": 1.2684,
      "num_input_tokens_seen": 1511104,
      "step": 308,
      "train_runtime": 1057.7744,
      "train_tokens_per_second": 1428.569
    },
    {
      "epoch": 0.11365517241379311,
      "grad_norm": 0.36995697021484375,
      "learning_rate": 0.00017767133382461312,
      "loss": 1.227,
      "num_input_tokens_seen": 1514752,
      "step": 309,
      "train_runtime": 1060.3849,
      "train_tokens_per_second": 1428.493
    },
    {
      "epoch": 0.11402298850574713,
      "grad_norm": 0.3357091546058655,
      "learning_rate": 0.0001775976418570376,
      "loss": 1.2474,
      "num_input_tokens_seen": 1520704,
      "step": 310,
      "train_runtime": 1064.2988,
      "train_tokens_per_second": 1428.832
    },
    {
      "epoch": 0.11439080459770115,
      "grad_norm": 0.32784029841423035,
      "learning_rate": 0.00017752394988946205,
      "loss": 1.1841,
      "num_input_tokens_seen": 1524512,
      "step": 311,
      "train_runtime": 1067.0563,
      "train_tokens_per_second": 1428.708
    },
    {
      "epoch": 0.11475862068965517,
      "grad_norm": 0.3780510723590851,
      "learning_rate": 0.00017745025792188652,
      "loss": 1.0959,
      "num_input_tokens_seen": 1529264,
      "step": 312,
      "train_runtime": 1070.2708,
      "train_tokens_per_second": 1428.857
    },
    {
      "epoch": 0.11512643678160919,
      "grad_norm": 0.37813234329223633,
      "learning_rate": 0.00017737656595431098,
      "loss": 1.2165,
      "num_input_tokens_seen": 1537888,
      "step": 313,
      "train_runtime": 1075.614,
      "train_tokens_per_second": 1429.777
    },
    {
      "epoch": 0.11549425287356321,
      "grad_norm": 0.3664109706878662,
      "learning_rate": 0.00017730287398673547,
      "loss": 1.3155,
      "num_input_tokens_seen": 1544080,
      "step": 314,
      "train_runtime": 1079.6735,
      "train_tokens_per_second": 1430.136
    },
    {
      "epoch": 0.11586206896551725,
      "grad_norm": 0.32953914999961853,
      "learning_rate": 0.0001772291820191599,
      "loss": 1.3003,
      "num_input_tokens_seen": 1548144,
      "step": 315,
      "train_runtime": 1082.5469,
      "train_tokens_per_second": 1430.094
    },
    {
      "epoch": 0.11622988505747127,
      "grad_norm": 0.32690170407295227,
      "learning_rate": 0.00017715549005158438,
      "loss": 1.3154,
      "num_input_tokens_seen": 1553200,
      "step": 316,
      "train_runtime": 1086.0107,
      "train_tokens_per_second": 1430.188
    },
    {
      "epoch": 0.11659770114942529,
      "grad_norm": 0.3233182728290558,
      "learning_rate": 0.00017708179808400884,
      "loss": 1.2467,
      "num_input_tokens_seen": 1558304,
      "step": 317,
      "train_runtime": 1089.4037,
      "train_tokens_per_second": 1430.419
    },
    {
      "epoch": 0.11696551724137931,
      "grad_norm": 0.3660307824611664,
      "learning_rate": 0.00017700810611643333,
      "loss": 1.3191,
      "num_input_tokens_seen": 1562912,
      "step": 318,
      "train_runtime": 1092.6112,
      "train_tokens_per_second": 1430.437
    },
    {
      "epoch": 0.11733333333333333,
      "grad_norm": 0.40045297145843506,
      "learning_rate": 0.0001769344141488578,
      "loss": 1.1062,
      "num_input_tokens_seen": 1567984,
      "step": 319,
      "train_runtime": 1096.0431,
      "train_tokens_per_second": 1430.586
    },
    {
      "epoch": 0.11770114942528735,
      "grad_norm": 0.4017600417137146,
      "learning_rate": 0.00017686072218128223,
      "loss": 1.0655,
      "num_input_tokens_seen": 1572240,
      "step": 320,
      "train_runtime": 1098.9569,
      "train_tokens_per_second": 1430.666
    },
    {
      "epoch": 0.11806896551724137,
      "grad_norm": 0.4012167155742645,
      "learning_rate": 0.0001767870302137067,
      "loss": 1.3542,
      "num_input_tokens_seen": 1576912,
      "step": 321,
      "train_runtime": 1102.1809,
      "train_tokens_per_second": 1430.72
    },
    {
      "epoch": 0.11843678160919541,
      "grad_norm": 0.3356454074382782,
      "learning_rate": 0.0001767133382461312,
      "loss": 1.3085,
      "num_input_tokens_seen": 1581520,
      "step": 322,
      "train_runtime": 1105.2999,
      "train_tokens_per_second": 1430.852
    },
    {
      "epoch": 0.11880459770114943,
      "grad_norm": 0.3676871955394745,
      "learning_rate": 0.00017663964627855566,
      "loss": 1.1754,
      "num_input_tokens_seen": 1585024,
      "step": 323,
      "train_runtime": 1107.8524,
      "train_tokens_per_second": 1430.718
    },
    {
      "epoch": 0.11917241379310345,
      "grad_norm": 0.3801829516887665,
      "learning_rate": 0.00017656595431098012,
      "loss": 1.3644,
      "num_input_tokens_seen": 1589952,
      "step": 324,
      "train_runtime": 1111.1965,
      "train_tokens_per_second": 1430.847
    },
    {
      "epoch": 0.11954022988505747,
      "grad_norm": 0.334855854511261,
      "learning_rate": 0.00017649226234340456,
      "loss": 1.3466,
      "num_input_tokens_seen": 1594864,
      "step": 325,
      "train_runtime": 1114.4727,
      "train_tokens_per_second": 1431.048
    },
    {
      "epoch": 0.11990804597701149,
      "grad_norm": 0.3540022671222687,
      "learning_rate": 0.00017641857037582905,
      "loss": 1.3566,
      "num_input_tokens_seen": 1600256,
      "step": 326,
      "train_runtime": 1118.006,
      "train_tokens_per_second": 1431.348
    },
    {
      "epoch": 0.12027586206896551,
      "grad_norm": 0.3691379725933075,
      "learning_rate": 0.00017634487840825352,
      "loss": 1.4399,
      "num_input_tokens_seen": 1607072,
      "step": 327,
      "train_runtime": 1122.4744,
      "train_tokens_per_second": 1431.723
    },
    {
      "epoch": 0.12064367816091955,
      "grad_norm": 0.3574798107147217,
      "learning_rate": 0.00017627118644067798,
      "loss": 1.2646,
      "num_input_tokens_seen": 1611088,
      "step": 328,
      "train_runtime": 1125.3089,
      "train_tokens_per_second": 1431.685
    },
    {
      "epoch": 0.12101149425287357,
      "grad_norm": 0.33182525634765625,
      "learning_rate": 0.00017619749447310242,
      "loss": 1.2291,
      "num_input_tokens_seen": 1614976,
      "step": 329,
      "train_runtime": 1128.0716,
      "train_tokens_per_second": 1431.625
    },
    {
      "epoch": 0.12137931034482759,
      "grad_norm": 0.3294955790042877,
      "learning_rate": 0.0001761238025055269,
      "loss": 1.29,
      "num_input_tokens_seen": 1619008,
      "step": 330,
      "train_runtime": 1130.9122,
      "train_tokens_per_second": 1431.595
    },
    {
      "epoch": 0.12174712643678161,
      "grad_norm": 0.32767629623413086,
      "learning_rate": 0.00017605011053795137,
      "loss": 1.2442,
      "num_input_tokens_seen": 1622832,
      "step": 331,
      "train_runtime": 1135.2508,
      "train_tokens_per_second": 1429.492
    },
    {
      "epoch": 0.12211494252873563,
      "grad_norm": 0.34242504835128784,
      "learning_rate": 0.00017597641857037584,
      "loss": 1.214,
      "num_input_tokens_seen": 1626144,
      "step": 332,
      "train_runtime": 1137.649,
      "train_tokens_per_second": 1429.39
    },
    {
      "epoch": 0.12248275862068965,
      "grad_norm": 0.29884108901023865,
      "learning_rate": 0.0001759027266028003,
      "loss": 1.0849,
      "num_input_tokens_seen": 1637072,
      "step": 333,
      "train_runtime": 1148.1181,
      "train_tokens_per_second": 1425.874
    },
    {
      "epoch": 0.12285057471264368,
      "grad_norm": 0.36340370774269104,
      "learning_rate": 0.00017582903463522477,
      "loss": 1.3963,
      "num_input_tokens_seen": 1640976,
      "step": 334,
      "train_runtime": 1150.9221,
      "train_tokens_per_second": 1425.792
    },
    {
      "epoch": 0.1232183908045977,
      "grad_norm": 0.3351694345474243,
      "learning_rate": 0.00017575534266764923,
      "loss": 1.2571,
      "num_input_tokens_seen": 1644432,
      "step": 335,
      "train_runtime": 1153.4056,
      "train_tokens_per_second": 1425.719
    },
    {
      "epoch": 0.12358620689655173,
      "grad_norm": 0.39297762513160706,
      "learning_rate": 0.0001756816507000737,
      "loss": 1.3074,
      "num_input_tokens_seen": 1649184,
      "step": 336,
      "train_runtime": 1156.678,
      "train_tokens_per_second": 1425.794
    },
    {
      "epoch": 0.12395402298850575,
      "grad_norm": 0.33720138669013977,
      "learning_rate": 0.00017560795873249816,
      "loss": 1.2003,
      "num_input_tokens_seen": 1654160,
      "step": 337,
      "train_runtime": 1160.0199,
      "train_tokens_per_second": 1425.976
    },
    {
      "epoch": 0.12432183908045977,
      "grad_norm": 0.3647576868534088,
      "learning_rate": 0.00017553426676492263,
      "loss": 1.3519,
      "num_input_tokens_seen": 1659792,
      "step": 338,
      "train_runtime": 1163.6894,
      "train_tokens_per_second": 1426.319
    },
    {
      "epoch": 0.12468965517241379,
      "grad_norm": 0.3617439866065979,
      "learning_rate": 0.0001754605747973471,
      "loss": 1.2487,
      "num_input_tokens_seen": 1665264,
      "step": 339,
      "train_runtime": 1167.2748,
      "train_tokens_per_second": 1426.625
    },
    {
      "epoch": 0.1250574712643678,
      "grad_norm": 0.3797023892402649,
      "learning_rate": 0.00017538688282977156,
      "loss": 1.2513,
      "num_input_tokens_seen": 1668864,
      "step": 340,
      "train_runtime": 1169.8841,
      "train_tokens_per_second": 1426.521
    },
    {
      "epoch": 0.12542528735632183,
      "grad_norm": 0.3398475646972656,
      "learning_rate": 0.00017531319086219602,
      "loss": 1.1765,
      "num_input_tokens_seen": 1672240,
      "step": 341,
      "train_runtime": 1172.3457,
      "train_tokens_per_second": 1426.405
    },
    {
      "epoch": 0.12579310344827585,
      "grad_norm": 0.36390501260757446,
      "learning_rate": 0.00017523949889462051,
      "loss": 1.3972,
      "num_input_tokens_seen": 1679920,
      "step": 342,
      "train_runtime": 1177.0879,
      "train_tokens_per_second": 1427.183
    },
    {
      "epoch": 0.1261609195402299,
      "grad_norm": 0.3486977815628052,
      "learning_rate": 0.00017516580692704495,
      "loss": 1.3217,
      "num_input_tokens_seen": 1684208,
      "step": 343,
      "train_runtime": 1180.0993,
      "train_tokens_per_second": 1427.175
    },
    {
      "epoch": 0.12652873563218392,
      "grad_norm": 0.38147181272506714,
      "learning_rate": 0.00017509211495946942,
      "loss": 1.3411,
      "num_input_tokens_seen": 1687824,
      "step": 344,
      "train_runtime": 1182.656,
      "train_tokens_per_second": 1427.147
    },
    {
      "epoch": 0.12689655172413794,
      "grad_norm": 0.3574577569961548,
      "learning_rate": 0.00017501842299189388,
      "loss": 1.3632,
      "num_input_tokens_seen": 1692064,
      "step": 345,
      "train_runtime": 1185.5686,
      "train_tokens_per_second": 1427.217
    },
    {
      "epoch": 0.12726436781609196,
      "grad_norm": 0.3400028347969055,
      "learning_rate": 0.00017494473102431837,
      "loss": 1.1166,
      "num_input_tokens_seen": 1696112,
      "step": 346,
      "train_runtime": 1188.356,
      "train_tokens_per_second": 1427.276
    },
    {
      "epoch": 0.12763218390804598,
      "grad_norm": 0.3180460035800934,
      "learning_rate": 0.0001748710390567428,
      "loss": 1.2668,
      "num_input_tokens_seen": 1699824,
      "step": 347,
      "train_runtime": 1191.0307,
      "train_tokens_per_second": 1427.187
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.3954669237136841,
      "learning_rate": 0.00017479734708916728,
      "loss": 1.3874,
      "num_input_tokens_seen": 1703840,
      "step": 348,
      "train_runtime": 1193.7618,
      "train_tokens_per_second": 1427.286
    },
    {
      "epoch": 0.12836781609195402,
      "grad_norm": 0.33969107270240784,
      "learning_rate": 0.00017472365512159174,
      "loss": 1.4295,
      "num_input_tokens_seen": 1707936,
      "step": 349,
      "train_runtime": 1196.5025,
      "train_tokens_per_second": 1427.44
    },
    {
      "epoch": 0.12873563218390804,
      "grad_norm": 0.4104045033454895,
      "learning_rate": 0.00017464996315401623,
      "loss": 1.482,
      "num_input_tokens_seen": 1712528,
      "step": 350,
      "train_runtime": 1199.712,
      "train_tokens_per_second": 1427.449
    },
    {
      "epoch": 0.12910344827586206,
      "grad_norm": 0.3302443325519562,
      "learning_rate": 0.0001745762711864407,
      "loss": 1.1095,
      "num_input_tokens_seen": 1716928,
      "step": 351,
      "train_runtime": 1202.69,
      "train_tokens_per_second": 1427.573
    },
    {
      "epoch": 0.12947126436781609,
      "grad_norm": 0.3219512403011322,
      "learning_rate": 0.00017450257921886514,
      "loss": 1.165,
      "num_input_tokens_seen": 1723392,
      "step": 352,
      "train_runtime": 1206.8425,
      "train_tokens_per_second": 1428.017
    },
    {
      "epoch": 0.1298390804597701,
      "grad_norm": 0.33770471811294556,
      "learning_rate": 0.0001744288872512896,
      "loss": 1.4201,
      "num_input_tokens_seen": 1727408,
      "step": 353,
      "train_runtime": 1209.6783,
      "train_tokens_per_second": 1427.99
    },
    {
      "epoch": 0.13020689655172413,
      "grad_norm": 0.35589832067489624,
      "learning_rate": 0.0001743551952837141,
      "loss": 1.393,
      "num_input_tokens_seen": 1732096,
      "step": 354,
      "train_runtime": 1212.9339,
      "train_tokens_per_second": 1428.022
    },
    {
      "epoch": 0.13057471264367815,
      "grad_norm": 0.36610502004623413,
      "learning_rate": 0.00017428150331613856,
      "loss": 1.1047,
      "num_input_tokens_seen": 1738816,
      "step": 355,
      "train_runtime": 1217.3457,
      "train_tokens_per_second": 1428.367
    },
    {
      "epoch": 0.1309425287356322,
      "grad_norm": 0.37327292561531067,
      "learning_rate": 0.00017420781134856302,
      "loss": 1.3669,
      "num_input_tokens_seen": 1746192,
      "step": 356,
      "train_runtime": 1222.019,
      "train_tokens_per_second": 1428.94
    },
    {
      "epoch": 0.13131034482758622,
      "grad_norm": 0.33279553055763245,
      "learning_rate": 0.00017413411938098746,
      "loss": 1.1447,
      "num_input_tokens_seen": 1752176,
      "step": 357,
      "train_runtime": 1225.8912,
      "train_tokens_per_second": 1429.308
    },
    {
      "epoch": 0.13167816091954024,
      "grad_norm": 0.34545910358428955,
      "learning_rate": 0.00017406042741341195,
      "loss": 1.251,
      "num_input_tokens_seen": 1755136,
      "step": 358,
      "train_runtime": 1228.1842,
      "train_tokens_per_second": 1429.049
    },
    {
      "epoch": 0.13204597701149426,
      "grad_norm": 0.3195616900920868,
      "learning_rate": 0.00017398673544583642,
      "loss": 1.2323,
      "num_input_tokens_seen": 1760656,
      "step": 359,
      "train_runtime": 1231.8626,
      "train_tokens_per_second": 1429.263
    },
    {
      "epoch": 0.13241379310344828,
      "grad_norm": 0.3158402144908905,
      "learning_rate": 0.00017391304347826088,
      "loss": 1.1428,
      "num_input_tokens_seen": 1766080,
      "step": 360,
      "train_runtime": 1235.5088,
      "train_tokens_per_second": 1429.435
    },
    {
      "epoch": 0.1327816091954023,
      "grad_norm": 0.3613432049751282,
      "learning_rate": 0.00017383935151068535,
      "loss": 1.3147,
      "num_input_tokens_seen": 1771264,
      "step": 361,
      "train_runtime": 1240.8473,
      "train_tokens_per_second": 1427.463
    },
    {
      "epoch": 0.13314942528735632,
      "grad_norm": 0.31645074486732483,
      "learning_rate": 0.0001737656595431098,
      "loss": 1.2493,
      "num_input_tokens_seen": 1774912,
      "step": 362,
      "train_runtime": 1243.4147,
      "train_tokens_per_second": 1427.45
    },
    {
      "epoch": 0.13351724137931034,
      "grad_norm": 0.38054177165031433,
      "learning_rate": 0.00017369196757553428,
      "loss": 1.4005,
      "num_input_tokens_seen": 1780800,
      "step": 363,
      "train_runtime": 1247.353,
      "train_tokens_per_second": 1427.663
    },
    {
      "epoch": 0.13388505747126436,
      "grad_norm": 0.3522668778896332,
      "learning_rate": 0.00017361827560795874,
      "loss": 1.2367,
      "num_input_tokens_seen": 1785168,
      "step": 364,
      "train_runtime": 1250.295,
      "train_tokens_per_second": 1427.797
    },
    {
      "epoch": 0.13425287356321838,
      "grad_norm": 0.36981090903282166,
      "learning_rate": 0.0001735445836403832,
      "loss": 1.31,
      "num_input_tokens_seen": 1790560,
      "step": 365,
      "train_runtime": 1253.8682,
      "train_tokens_per_second": 1428.029
    },
    {
      "epoch": 0.1346206896551724,
      "grad_norm": 0.38494235277175903,
      "learning_rate": 0.00017347089167280767,
      "loss": 1.2425,
      "num_input_tokens_seen": 1794720,
      "step": 366,
      "train_runtime": 1256.7756,
      "train_tokens_per_second": 1428.035
    },
    {
      "epoch": 0.13498850574712642,
      "grad_norm": 0.3471364378929138,
      "learning_rate": 0.00017339719970523214,
      "loss": 1.2496,
      "num_input_tokens_seen": 1801584,
      "step": 367,
      "train_runtime": 1261.0931,
      "train_tokens_per_second": 1428.589
    },
    {
      "epoch": 0.13535632183908047,
      "grad_norm": 0.36003199219703674,
      "learning_rate": 0.0001733235077376566,
      "loss": 1.2732,
      "num_input_tokens_seen": 1806048,
      "step": 368,
      "train_runtime": 1264.1584,
      "train_tokens_per_second": 1428.656
    },
    {
      "epoch": 0.1357241379310345,
      "grad_norm": 0.36972132325172424,
      "learning_rate": 0.00017324981577008106,
      "loss": 1.2372,
      "num_input_tokens_seen": 1814080,
      "step": 369,
      "train_runtime": 1269.3247,
      "train_tokens_per_second": 1429.169
    },
    {
      "epoch": 0.13609195402298851,
      "grad_norm": 0.3443167209625244,
      "learning_rate": 0.00017317612380250553,
      "loss": 1.2402,
      "num_input_tokens_seen": 1818784,
      "step": 370,
      "train_runtime": 1272.5184,
      "train_tokens_per_second": 1429.279
    },
    {
      "epoch": 0.13645977011494254,
      "grad_norm": 0.38251325488090515,
      "learning_rate": 0.00017310243183493,
      "loss": 1.2565,
      "num_input_tokens_seen": 1824608,
      "step": 371,
      "train_runtime": 1276.2877,
      "train_tokens_per_second": 1429.621
    },
    {
      "epoch": 0.13682758620689656,
      "grad_norm": 0.4466417729854584,
      "learning_rate": 0.00017302873986735446,
      "loss": 1.3363,
      "num_input_tokens_seen": 1828992,
      "step": 372,
      "train_runtime": 1279.2636,
      "train_tokens_per_second": 1429.723
    },
    {
      "epoch": 0.13719540229885058,
      "grad_norm": 0.35756662487983704,
      "learning_rate": 0.00017295504789977892,
      "loss": 1.4238,
      "num_input_tokens_seen": 1833424,
      "step": 373,
      "train_runtime": 1282.3214,
      "train_tokens_per_second": 1429.769
    },
    {
      "epoch": 0.1375632183908046,
      "grad_norm": 0.36335867643356323,
      "learning_rate": 0.00017288135593220342,
      "loss": 1.3744,
      "num_input_tokens_seen": 1837504,
      "step": 374,
      "train_runtime": 1285.1716,
      "train_tokens_per_second": 1429.773
    },
    {
      "epoch": 0.13793103448275862,
      "grad_norm": 0.3521118760108948,
      "learning_rate": 0.00017280766396462785,
      "loss": 1.1201,
      "num_input_tokens_seen": 1841888,
      "step": 375,
      "train_runtime": 1288.1896,
      "train_tokens_per_second": 1429.827
    },
    {
      "epoch": 0.13829885057471264,
      "grad_norm": 0.33666011691093445,
      "learning_rate": 0.00017273397199705232,
      "loss": 1.2316,
      "num_input_tokens_seen": 1849456,
      "step": 376,
      "train_runtime": 1292.8825,
      "train_tokens_per_second": 1430.49
    },
    {
      "epoch": 0.13866666666666666,
      "grad_norm": 0.40880200266838074,
      "learning_rate": 0.0001726602800294768,
      "loss": 1.3432,
      "num_input_tokens_seen": 1854640,
      "step": 377,
      "train_runtime": 1296.3377,
      "train_tokens_per_second": 1430.677
    },
    {
      "epoch": 0.13903448275862068,
      "grad_norm": 0.34930717945098877,
      "learning_rate": 0.00017258658806190128,
      "loss": 1.3542,
      "num_input_tokens_seen": 1864832,
      "step": 378,
      "train_runtime": 1302.513,
      "train_tokens_per_second": 1431.719
    },
    {
      "epoch": 0.1394022988505747,
      "grad_norm": 0.38795962929725647,
      "learning_rate": 0.00017251289609432574,
      "loss": 1.3718,
      "num_input_tokens_seen": 1869536,
      "step": 379,
      "train_runtime": 1305.6807,
      "train_tokens_per_second": 1431.848
    },
    {
      "epoch": 0.13977011494252872,
      "grad_norm": 0.34071630239486694,
      "learning_rate": 0.00017243920412675018,
      "loss": 1.3193,
      "num_input_tokens_seen": 1875632,
      "step": 380,
      "train_runtime": 1309.7488,
      "train_tokens_per_second": 1432.055
    },
    {
      "epoch": 0.14013793103448277,
      "grad_norm": 0.3350682854652405,
      "learning_rate": 0.00017236551215917467,
      "loss": 1.2953,
      "num_input_tokens_seen": 1880576,
      "step": 381,
      "train_runtime": 1313.0213,
      "train_tokens_per_second": 1432.251
    },
    {
      "epoch": 0.1405057471264368,
      "grad_norm": 0.35839366912841797,
      "learning_rate": 0.00017229182019159913,
      "loss": 1.2984,
      "num_input_tokens_seen": 1884256,
      "step": 382,
      "train_runtime": 1315.6269,
      "train_tokens_per_second": 1432.212
    },
    {
      "epoch": 0.1408735632183908,
      "grad_norm": 0.3593604862689972,
      "learning_rate": 0.0001722181282240236,
      "loss": 1.3049,
      "num_input_tokens_seen": 1888016,
      "step": 383,
      "train_runtime": 1318.3432,
      "train_tokens_per_second": 1432.113
    },
    {
      "epoch": 0.14124137931034483,
      "grad_norm": 0.35749468207359314,
      "learning_rate": 0.00017214443625644804,
      "loss": 1.15,
      "num_input_tokens_seen": 1891888,
      "step": 384,
      "train_runtime": 1321.1286,
      "train_tokens_per_second": 1432.024
    },
    {
      "epoch": 0.14160919540229885,
      "grad_norm": 0.4240542948246002,
      "learning_rate": 0.00017207074428887253,
      "loss": 1.2574,
      "num_input_tokens_seen": 1896720,
      "step": 385,
      "train_runtime": 1324.4688,
      "train_tokens_per_second": 1432.061
    },
    {
      "epoch": 0.14197701149425287,
      "grad_norm": 0.3442443609237671,
      "learning_rate": 0.000171997052321297,
      "loss": 1.3985,
      "num_input_tokens_seen": 1901360,
      "step": 386,
      "train_runtime": 1327.6283,
      "train_tokens_per_second": 1432.148
    },
    {
      "epoch": 0.1423448275862069,
      "grad_norm": 0.43367230892181396,
      "learning_rate": 0.00017192336035372146,
      "loss": 1.4627,
      "num_input_tokens_seen": 1905408,
      "step": 387,
      "train_runtime": 1330.4959,
      "train_tokens_per_second": 1432.104
    },
    {
      "epoch": 0.14271264367816092,
      "grad_norm": 0.4000196158885956,
      "learning_rate": 0.00017184966838614592,
      "loss": 1.3171,
      "num_input_tokens_seen": 1910544,
      "step": 388,
      "train_runtime": 1333.9981,
      "train_tokens_per_second": 1432.194
    },
    {
      "epoch": 0.14308045977011494,
      "grad_norm": 0.4140348732471466,
      "learning_rate": 0.0001717759764185704,
      "loss": 1.2402,
      "num_input_tokens_seen": 1915472,
      "step": 389,
      "train_runtime": 1337.3269,
      "train_tokens_per_second": 1432.314
    },
    {
      "epoch": 0.14344827586206896,
      "grad_norm": 0.3835417628288269,
      "learning_rate": 0.00017170228445099485,
      "loss": 1.2875,
      "num_input_tokens_seen": 1920368,
      "step": 390,
      "train_runtime": 1340.6689,
      "train_tokens_per_second": 1432.395
    },
    {
      "epoch": 0.14381609195402298,
      "grad_norm": 0.3389739990234375,
      "learning_rate": 0.00017162859248341932,
      "loss": 1.3775,
      "num_input_tokens_seen": 1923568,
      "step": 391,
      "train_runtime": 1344.9367,
      "train_tokens_per_second": 1430.229
    },
    {
      "epoch": 0.144183908045977,
      "grad_norm": 0.34423133730888367,
      "learning_rate": 0.00017155490051584378,
      "loss": 1.2596,
      "num_input_tokens_seen": 1929104,
      "step": 392,
      "train_runtime": 1348.4832,
      "train_tokens_per_second": 1430.573
    },
    {
      "epoch": 0.14455172413793105,
      "grad_norm": 0.31878727674484253,
      "learning_rate": 0.00017148120854826825,
      "loss": 1.1904,
      "num_input_tokens_seen": 1933120,
      "step": 393,
      "train_runtime": 1351.3408,
      "train_tokens_per_second": 1430.52
    },
    {
      "epoch": 0.14491954022988507,
      "grad_norm": 0.39297088980674744,
      "learning_rate": 0.0001714075165806927,
      "loss": 1.2576,
      "num_input_tokens_seen": 1937232,
      "step": 394,
      "train_runtime": 1354.3271,
      "train_tokens_per_second": 1430.402
    },
    {
      "epoch": 0.1452873563218391,
      "grad_norm": 0.3405814468860626,
      "learning_rate": 0.00017133382461311718,
      "loss": 1.3358,
      "num_input_tokens_seen": 1941232,
      "step": 395,
      "train_runtime": 1357.1096,
      "train_tokens_per_second": 1430.417
    },
    {
      "epoch": 0.1456551724137931,
      "grad_norm": 0.31894049048423767,
      "learning_rate": 0.00017126013264554164,
      "loss": 1.2919,
      "num_input_tokens_seen": 1946096,
      "step": 396,
      "train_runtime": 1360.4301,
      "train_tokens_per_second": 1430.501
    },
    {
      "epoch": 0.14602298850574713,
      "grad_norm": 0.30518004298210144,
      "learning_rate": 0.0001711864406779661,
      "loss": 1.2456,
      "num_input_tokens_seen": 1957120,
      "step": 397,
      "train_runtime": 1367.1031,
      "train_tokens_per_second": 1431.582
    },
    {
      "epoch": 0.14639080459770115,
      "grad_norm": 0.3538399934768677,
      "learning_rate": 0.00017111274871039057,
      "loss": 1.4283,
      "num_input_tokens_seen": 1961840,
      "step": 398,
      "train_runtime": 1370.2305,
      "train_tokens_per_second": 1431.759
    },
    {
      "epoch": 0.14675862068965517,
      "grad_norm": 0.4415174126625061,
      "learning_rate": 0.00017103905674281504,
      "loss": 1.2578,
      "num_input_tokens_seen": 1966112,
      "step": 399,
      "train_runtime": 1373.2288,
      "train_tokens_per_second": 1431.744
    },
    {
      "epoch": 0.1471264367816092,
      "grad_norm": 0.4373137950897217,
      "learning_rate": 0.0001709653647752395,
      "loss": 1.4214,
      "num_input_tokens_seen": 1969536,
      "step": 400,
      "train_runtime": 1375.7143,
      "train_tokens_per_second": 1431.646
    },
    {
      "epoch": 0.14749425287356321,
      "grad_norm": 0.3483783006668091,
      "learning_rate": 0.000170891672807664,
      "loss": 1.453,
      "num_input_tokens_seen": 1976128,
      "step": 401,
      "train_runtime": 1379.9346,
      "train_tokens_per_second": 1432.045
    },
    {
      "epoch": 0.14786206896551723,
      "grad_norm": 0.3300085663795471,
      "learning_rate": 0.00017081798084008843,
      "loss": 1.2722,
      "num_input_tokens_seen": 1979696,
      "step": 402,
      "train_runtime": 1382.5772,
      "train_tokens_per_second": 1431.888
    },
    {
      "epoch": 0.14822988505747126,
      "grad_norm": 0.3039983808994293,
      "learning_rate": 0.0001707442888725129,
      "loss": 1.1046,
      "num_input_tokens_seen": 1989376,
      "step": 403,
      "train_runtime": 1388.5576,
      "train_tokens_per_second": 1432.692
    },
    {
      "epoch": 0.14859770114942528,
      "grad_norm": 0.3359440863132477,
      "learning_rate": 0.00017067059690493736,
      "loss": 1.4073,
      "num_input_tokens_seen": 1994928,
      "step": 404,
      "train_runtime": 1392.2734,
      "train_tokens_per_second": 1432.856
    },
    {
      "epoch": 0.1489655172413793,
      "grad_norm": 0.381528377532959,
      "learning_rate": 0.00017059690493736185,
      "loss": 1.3089,
      "num_input_tokens_seen": 1999056,
      "step": 405,
      "train_runtime": 1395.1902,
      "train_tokens_per_second": 1432.82
    },
    {
      "epoch": 0.14933333333333335,
      "grad_norm": 0.32278943061828613,
      "learning_rate": 0.00017052321296978632,
      "loss": 1.3792,
      "num_input_tokens_seen": 2005632,
      "step": 406,
      "train_runtime": 1399.3481,
      "train_tokens_per_second": 1433.262
    },
    {
      "epoch": 0.14970114942528737,
      "grad_norm": 0.4173831045627594,
      "learning_rate": 0.00017044952100221076,
      "loss": 1.4602,
      "num_input_tokens_seen": 2009440,
      "step": 407,
      "train_runtime": 1402.0397,
      "train_tokens_per_second": 1433.226
    },
    {
      "epoch": 0.1500689655172414,
      "grad_norm": 0.335906058549881,
      "learning_rate": 0.00017037582903463522,
      "loss": 1.2627,
      "num_input_tokens_seen": 2013456,
      "step": 408,
      "train_runtime": 1404.8643,
      "train_tokens_per_second": 1433.203
    },
    {
      "epoch": 0.1504367816091954,
      "grad_norm": 0.3314789831638336,
      "learning_rate": 0.0001703021370670597,
      "loss": 1.2863,
      "num_input_tokens_seen": 2018272,
      "step": 409,
      "train_runtime": 1408.0975,
      "train_tokens_per_second": 1433.333
    },
    {
      "epoch": 0.15080459770114943,
      "grad_norm": 0.3679824769496918,
      "learning_rate": 0.00017022844509948418,
      "loss": 1.28,
      "num_input_tokens_seen": 2022832,
      "step": 410,
      "train_runtime": 1411.2397,
      "train_tokens_per_second": 1433.372
    },
    {
      "epoch": 0.15117241379310345,
      "grad_norm": 0.3312888741493225,
      "learning_rate": 0.00017015475313190864,
      "loss": 1.2087,
      "num_input_tokens_seen": 2027792,
      "step": 411,
      "train_runtime": 1414.5455,
      "train_tokens_per_second": 1433.529
    },
    {
      "epoch": 0.15154022988505747,
      "grad_norm": 0.3557579219341278,
      "learning_rate": 0.00017008106116433308,
      "loss": 1.3052,
      "num_input_tokens_seen": 2032048,
      "step": 412,
      "train_runtime": 1417.5023,
      "train_tokens_per_second": 1433.541
    },
    {
      "epoch": 0.1519080459770115,
      "grad_norm": 0.38084933161735535,
      "learning_rate": 0.00017000736919675757,
      "loss": 1.2378,
      "num_input_tokens_seen": 2036224,
      "step": 413,
      "train_runtime": 1420.4101,
      "train_tokens_per_second": 1433.547
    },
    {
      "epoch": 0.1522758620689655,
      "grad_norm": 0.40804681181907654,
      "learning_rate": 0.00016993367722918204,
      "loss": 1.3993,
      "num_input_tokens_seen": 2042672,
      "step": 414,
      "train_runtime": 1424.6429,
      "train_tokens_per_second": 1433.813
    },
    {
      "epoch": 0.15264367816091953,
      "grad_norm": 0.356699675321579,
      "learning_rate": 0.0001698599852616065,
      "loss": 1.3581,
      "num_input_tokens_seen": 2047392,
      "step": 415,
      "train_runtime": 1427.7212,
      "train_tokens_per_second": 1434.028
    },
    {
      "epoch": 0.15301149425287355,
      "grad_norm": 0.36977219581604004,
      "learning_rate": 0.00016978629329403094,
      "loss": 1.3457,
      "num_input_tokens_seen": 2054368,
      "step": 416,
      "train_runtime": 1432.1738,
      "train_tokens_per_second": 1434.44
    },
    {
      "epoch": 0.15337931034482757,
      "grad_norm": 0.3046276569366455,
      "learning_rate": 0.00016971260132645543,
      "loss": 1.1488,
      "num_input_tokens_seen": 2060224,
      "step": 417,
      "train_runtime": 1436.0747,
      "train_tokens_per_second": 1434.622
    },
    {
      "epoch": 0.15374712643678162,
      "grad_norm": 0.35704097151756287,
      "learning_rate": 0.0001696389093588799,
      "loss": 1.2044,
      "num_input_tokens_seen": 2064544,
      "step": 418,
      "train_runtime": 1439.0453,
      "train_tokens_per_second": 1434.662
    },
    {
      "epoch": 0.15411494252873564,
      "grad_norm": 0.33867743611335754,
      "learning_rate": 0.00016956521739130436,
      "loss": 1.2563,
      "num_input_tokens_seen": 2068560,
      "step": 419,
      "train_runtime": 1441.933,
      "train_tokens_per_second": 1434.574
    },
    {
      "epoch": 0.15448275862068966,
      "grad_norm": 0.336556077003479,
      "learning_rate": 0.00016949152542372882,
      "loss": 1.356,
      "num_input_tokens_seen": 2080528,
      "step": 420,
      "train_runtime": 1449.0782,
      "train_tokens_per_second": 1435.76
    },
    {
      "epoch": 0.15485057471264368,
      "grad_norm": 0.3474124073982239,
      "learning_rate": 0.0001694178334561533,
      "loss": 1.2698,
      "num_input_tokens_seen": 2084000,
      "step": 421,
      "train_runtime": 1453.5086,
      "train_tokens_per_second": 1433.772
    },
    {
      "epoch": 0.1552183908045977,
      "grad_norm": 0.35970601439476013,
      "learning_rate": 0.00016934414148857775,
      "loss": 1.287,
      "num_input_tokens_seen": 2089888,
      "step": 422,
      "train_runtime": 1457.3837,
      "train_tokens_per_second": 1434.0
    },
    {
      "epoch": 0.15558620689655173,
      "grad_norm": 0.31702131032943726,
      "learning_rate": 0.00016927044952100222,
      "loss": 1.1189,
      "num_input_tokens_seen": 2093472,
      "step": 423,
      "train_runtime": 1459.9758,
      "train_tokens_per_second": 1433.909
    },
    {
      "epoch": 0.15595402298850575,
      "grad_norm": 0.32167336344718933,
      "learning_rate": 0.00016919675755342668,
      "loss": 1.2787,
      "num_input_tokens_seen": 2098096,
      "step": 424,
      "train_runtime": 1463.1316,
      "train_tokens_per_second": 1433.976
    },
    {
      "epoch": 0.15632183908045977,
      "grad_norm": 0.3513907194137573,
      "learning_rate": 0.00016912306558585115,
      "loss": 1.2525,
      "num_input_tokens_seen": 2103440,
      "step": 425,
      "train_runtime": 1466.7478,
      "train_tokens_per_second": 1434.084
    },
    {
      "epoch": 0.1566896551724138,
      "grad_norm": 0.3481293022632599,
      "learning_rate": 0.00016904937361827561,
      "loss": 1.3526,
      "num_input_tokens_seen": 2108112,
      "step": 426,
      "train_runtime": 1469.8695,
      "train_tokens_per_second": 1434.217
    },
    {
      "epoch": 0.1570574712643678,
      "grad_norm": 0.3324897885322571,
      "learning_rate": 0.00016897568165070008,
      "loss": 1.3382,
      "num_input_tokens_seen": 2112240,
      "step": 427,
      "train_runtime": 1472.7851,
      "train_tokens_per_second": 1434.181
    },
    {
      "epoch": 0.15742528735632183,
      "grad_norm": 0.3185710310935974,
      "learning_rate": 0.00016890198968312454,
      "loss": 1.2524,
      "num_input_tokens_seen": 2120640,
      "step": 428,
      "train_runtime": 1478.0089,
      "train_tokens_per_second": 1434.795
    },
    {
      "epoch": 0.15779310344827585,
      "grad_norm": 0.3544529974460602,
      "learning_rate": 0.00016882829771554904,
      "loss": 1.2214,
      "num_input_tokens_seen": 2125760,
      "step": 429,
      "train_runtime": 1481.4673,
      "train_tokens_per_second": 1434.902
    },
    {
      "epoch": 0.15816091954022987,
      "grad_norm": 0.33514347672462463,
      "learning_rate": 0.00016875460574797347,
      "loss": 1.3911,
      "num_input_tokens_seen": 2130016,
      "step": 430,
      "train_runtime": 1484.3044,
      "train_tokens_per_second": 1435.026
    },
    {
      "epoch": 0.15852873563218392,
      "grad_norm": 0.33542317152023315,
      "learning_rate": 0.00016868091378039794,
      "loss": 1.3581,
      "num_input_tokens_seen": 2135024,
      "step": 431,
      "train_runtime": 1487.6554,
      "train_tokens_per_second": 1435.16
    },
    {
      "epoch": 0.15889655172413794,
      "grad_norm": 0.3252404034137726,
      "learning_rate": 0.0001686072218128224,
      "loss": 1.3684,
      "num_input_tokens_seen": 2140656,
      "step": 432,
      "train_runtime": 1491.4062,
      "train_tokens_per_second": 1435.327
    },
    {
      "epoch": 0.15926436781609196,
      "grad_norm": 0.35924550890922546,
      "learning_rate": 0.0001685335298452469,
      "loss": 1.4129,
      "num_input_tokens_seen": 2144416,
      "step": 433,
      "train_runtime": 1494.0865,
      "train_tokens_per_second": 1435.269
    },
    {
      "epoch": 0.15963218390804598,
      "grad_norm": 0.32163017988204956,
      "learning_rate": 0.00016845983787767133,
      "loss": 1.3567,
      "num_input_tokens_seen": 2150144,
      "step": 434,
      "train_runtime": 1497.9466,
      "train_tokens_per_second": 1435.394
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.38988184928894043,
      "learning_rate": 0.0001683861459100958,
      "loss": 1.2821,
      "num_input_tokens_seen": 2153296,
      "step": 435,
      "train_runtime": 1500.2712,
      "train_tokens_per_second": 1435.271
    },
    {
      "epoch": 0.16036781609195402,
      "grad_norm": 0.3610256612300873,
      "learning_rate": 0.00016831245394252026,
      "loss": 1.2969,
      "num_input_tokens_seen": 2156688,
      "step": 436,
      "train_runtime": 1502.7894,
      "train_tokens_per_second": 1435.123
    },
    {
      "epoch": 0.16073563218390804,
      "grad_norm": 0.35343337059020996,
      "learning_rate": 0.00016823876197494475,
      "loss": 1.3713,
      "num_input_tokens_seen": 2160752,
      "step": 437,
      "train_runtime": 1505.6297,
      "train_tokens_per_second": 1435.115
    },
    {
      "epoch": 0.16110344827586207,
      "grad_norm": 0.3229235112667084,
      "learning_rate": 0.00016816507000736922,
      "loss": 1.2714,
      "num_input_tokens_seen": 2164832,
      "step": 438,
      "train_runtime": 1508.4733,
      "train_tokens_per_second": 1435.115
    },
    {
      "epoch": 0.1614712643678161,
      "grad_norm": 0.34250307083129883,
      "learning_rate": 0.00016809137803979366,
      "loss": 1.1873,
      "num_input_tokens_seen": 2168880,
      "step": 439,
      "train_runtime": 1511.3412,
      "train_tokens_per_second": 1435.07
    },
    {
      "epoch": 0.1618390804597701,
      "grad_norm": 0.36971795558929443,
      "learning_rate": 0.00016801768607221812,
      "loss": 1.256,
      "num_input_tokens_seen": 2174864,
      "step": 440,
      "train_runtime": 1515.3279,
      "train_tokens_per_second": 1435.243
    },
    {
      "epoch": 0.16220689655172413,
      "grad_norm": 0.3265184760093689,
      "learning_rate": 0.0001679439941046426,
      "loss": 1.2169,
      "num_input_tokens_seen": 2180304,
      "step": 441,
      "train_runtime": 1519.0008,
      "train_tokens_per_second": 1435.354
    },
    {
      "epoch": 0.16257471264367815,
      "grad_norm": 0.37642019987106323,
      "learning_rate": 0.00016787030213706708,
      "loss": 1.261,
      "num_input_tokens_seen": 2185216,
      "step": 442,
      "train_runtime": 1522.3337,
      "train_tokens_per_second": 1435.438
    },
    {
      "epoch": 0.1629425287356322,
      "grad_norm": 0.35478144884109497,
      "learning_rate": 0.00016779661016949154,
      "loss": 1.1235,
      "num_input_tokens_seen": 2189168,
      "step": 443,
      "train_runtime": 1525.0951,
      "train_tokens_per_second": 1435.431
    },
    {
      "epoch": 0.16331034482758622,
      "grad_norm": 0.3368073105812073,
      "learning_rate": 0.00016772291820191598,
      "loss": 1.2992,
      "num_input_tokens_seen": 2193648,
      "step": 444,
      "train_runtime": 1528.1172,
      "train_tokens_per_second": 1435.523
    },
    {
      "epoch": 0.16367816091954024,
      "grad_norm": 0.43296244740486145,
      "learning_rate": 0.00016764922623434047,
      "loss": 1.4187,
      "num_input_tokens_seen": 2197392,
      "step": 445,
      "train_runtime": 1530.7159,
      "train_tokens_per_second": 1435.532
    },
    {
      "epoch": 0.16404597701149426,
      "grad_norm": 0.4027699828147888,
      "learning_rate": 0.00016757553426676494,
      "loss": 1.3037,
      "num_input_tokens_seen": 2200880,
      "step": 446,
      "train_runtime": 1533.2024,
      "train_tokens_per_second": 1435.479
    },
    {
      "epoch": 0.16441379310344828,
      "grad_norm": 0.3729379177093506,
      "learning_rate": 0.0001675018422991894,
      "loss": 1.2391,
      "num_input_tokens_seen": 2203952,
      "step": 447,
      "train_runtime": 1535.5547,
      "train_tokens_per_second": 1435.281
    },
    {
      "epoch": 0.1647816091954023,
      "grad_norm": 0.3396291732788086,
      "learning_rate": 0.00016742815033161384,
      "loss": 1.2727,
      "num_input_tokens_seen": 2207584,
      "step": 448,
      "train_runtime": 1538.2027,
      "train_tokens_per_second": 1435.171
    },
    {
      "epoch": 0.16514942528735632,
      "grad_norm": 0.33090853691101074,
      "learning_rate": 0.00016735445836403833,
      "loss": 1.3714,
      "num_input_tokens_seen": 2214384,
      "step": 449,
      "train_runtime": 1542.5859,
      "train_tokens_per_second": 1435.501
    },
    {
      "epoch": 0.16551724137931034,
      "grad_norm": 0.3716526925563812,
      "learning_rate": 0.0001672807663964628,
      "loss": 1.3241,
      "num_input_tokens_seen": 2218624,
      "step": 450,
      "train_runtime": 1545.5852,
      "train_tokens_per_second": 1435.459
    },
    {
      "epoch": 0.16588505747126436,
      "grad_norm": 0.32018139958381653,
      "learning_rate": 0.00016720707442888726,
      "loss": 1.1954,
      "num_input_tokens_seen": 2226688,
      "step": 451,
      "train_runtime": 1552.659,
      "train_tokens_per_second": 1434.113
    },
    {
      "epoch": 0.16625287356321838,
      "grad_norm": 0.3496841788291931,
      "learning_rate": 0.00016713338246131173,
      "loss": 1.3889,
      "num_input_tokens_seen": 2230656,
      "step": 452,
      "train_runtime": 1555.5039,
      "train_tokens_per_second": 1434.041
    },
    {
      "epoch": 0.1666206896551724,
      "grad_norm": 0.32573357224464417,
      "learning_rate": 0.0001670596904937362,
      "loss": 1.3581,
      "num_input_tokens_seen": 2235152,
      "step": 453,
      "train_runtime": 1558.5137,
      "train_tokens_per_second": 1434.156
    },
    {
      "epoch": 0.16698850574712643,
      "grad_norm": 0.3681710958480835,
      "learning_rate": 0.00016698599852616066,
      "loss": 1.3239,
      "num_input_tokens_seen": 2240256,
      "step": 454,
      "train_runtime": 1561.8838,
      "train_tokens_per_second": 1434.329
    },
    {
      "epoch": 0.16735632183908045,
      "grad_norm": 0.31902021169662476,
      "learning_rate": 0.00016691230655858512,
      "loss": 1.3712,
      "num_input_tokens_seen": 2247984,
      "step": 455,
      "train_runtime": 1566.7686,
      "train_tokens_per_second": 1434.79
    },
    {
      "epoch": 0.1677241379310345,
      "grad_norm": 0.37209877371788025,
      "learning_rate": 0.00016683861459100959,
      "loss": 1.2074,
      "num_input_tokens_seen": 2251200,
      "step": 456,
      "train_runtime": 1569.1997,
      "train_tokens_per_second": 1434.617
    },
    {
      "epoch": 0.16809195402298852,
      "grad_norm": 0.3461419343948364,
      "learning_rate": 0.00016676492262343405,
      "loss": 1.2965,
      "num_input_tokens_seen": 2256976,
      "step": 457,
      "train_runtime": 1572.9015,
      "train_tokens_per_second": 1434.912
    },
    {
      "epoch": 0.16845977011494254,
      "grad_norm": 0.331448495388031,
      "learning_rate": 0.00016669123065585852,
      "loss": 1.2546,
      "num_input_tokens_seen": 2260768,
      "step": 458,
      "train_runtime": 1575.6662,
      "train_tokens_per_second": 1434.801
    },
    {
      "epoch": 0.16882758620689656,
      "grad_norm": 0.388141393661499,
      "learning_rate": 0.00016661753868828298,
      "loss": 1.3422,
      "num_input_tokens_seen": 2266432,
      "step": 459,
      "train_runtime": 1579.4389,
      "train_tokens_per_second": 1434.96
    },
    {
      "epoch": 0.16919540229885058,
      "grad_norm": 0.34401068091392517,
      "learning_rate": 0.00016654384672070744,
      "loss": 1.2571,
      "num_input_tokens_seen": 2271392,
      "step": 460,
      "train_runtime": 1582.8521,
      "train_tokens_per_second": 1435.0
    },
    {
      "epoch": 0.1695632183908046,
      "grad_norm": 0.353071928024292,
      "learning_rate": 0.00016647015475313194,
      "loss": 1.1926,
      "num_input_tokens_seen": 2276016,
      "step": 461,
      "train_runtime": 1586.0107,
      "train_tokens_per_second": 1435.057
    },
    {
      "epoch": 0.16993103448275862,
      "grad_norm": 0.3545515239238739,
      "learning_rate": 0.00016639646278555637,
      "loss": 1.3086,
      "num_input_tokens_seen": 2281568,
      "step": 462,
      "train_runtime": 1589.6648,
      "train_tokens_per_second": 1435.251
    },
    {
      "epoch": 0.17029885057471264,
      "grad_norm": 0.33702024817466736,
      "learning_rate": 0.00016632277081798084,
      "loss": 1.3195,
      "num_input_tokens_seen": 2285024,
      "step": 463,
      "train_runtime": 1592.153,
      "train_tokens_per_second": 1435.179
    },
    {
      "epoch": 0.17066666666666666,
      "grad_norm": 0.396621972322464,
      "learning_rate": 0.0001662490788504053,
      "loss": 1.2395,
      "num_input_tokens_seen": 2291584,
      "step": 464,
      "train_runtime": 1596.4086,
      "train_tokens_per_second": 1435.462
    },
    {
      "epoch": 0.17103448275862068,
      "grad_norm": 0.34218865633010864,
      "learning_rate": 0.0001661753868828298,
      "loss": 1.2186,
      "num_input_tokens_seen": 2295456,
      "step": 465,
      "train_runtime": 1599.1769,
      "train_tokens_per_second": 1435.398
    },
    {
      "epoch": 0.1714022988505747,
      "grad_norm": 0.36157292127609253,
      "learning_rate": 0.00016610169491525423,
      "loss": 1.3077,
      "num_input_tokens_seen": 2301776,
      "step": 466,
      "train_runtime": 1603.2734,
      "train_tokens_per_second": 1435.673
    },
    {
      "epoch": 0.17177011494252872,
      "grad_norm": 0.3664705157279968,
      "learning_rate": 0.0001660280029476787,
      "loss": 1.3709,
      "num_input_tokens_seen": 2305968,
      "step": 467,
      "train_runtime": 1606.0913,
      "train_tokens_per_second": 1435.764
    },
    {
      "epoch": 0.17213793103448277,
      "grad_norm": 0.3933860957622528,
      "learning_rate": 0.00016595431098010316,
      "loss": 1.3227,
      "num_input_tokens_seen": 2310768,
      "step": 468,
      "train_runtime": 1609.3503,
      "train_tokens_per_second": 1435.839
    },
    {
      "epoch": 0.1725057471264368,
      "grad_norm": 0.35732167959213257,
      "learning_rate": 0.00016588061901252766,
      "loss": 1.2522,
      "num_input_tokens_seen": 2314752,
      "step": 469,
      "train_runtime": 1612.1725,
      "train_tokens_per_second": 1435.797
    },
    {
      "epoch": 0.1728735632183908,
      "grad_norm": 0.3423457145690918,
      "learning_rate": 0.00016580692704495212,
      "loss": 1.1958,
      "num_input_tokens_seen": 2319984,
      "step": 470,
      "train_runtime": 1615.6381,
      "train_tokens_per_second": 1435.955
    },
    {
      "epoch": 0.17324137931034483,
      "grad_norm": 0.37787607312202454,
      "learning_rate": 0.00016573323507737656,
      "loss": 1.3952,
      "num_input_tokens_seen": 2323184,
      "step": 471,
      "train_runtime": 1618.0736,
      "train_tokens_per_second": 1435.772
    },
    {
      "epoch": 0.17360919540229885,
      "grad_norm": 0.3515986204147339,
      "learning_rate": 0.00016565954310980102,
      "loss": 1.3096,
      "num_input_tokens_seen": 2329216,
      "step": 472,
      "train_runtime": 1621.9949,
      "train_tokens_per_second": 1436.019
    },
    {
      "epoch": 0.17397701149425288,
      "grad_norm": 0.37123268842697144,
      "learning_rate": 0.00016558585114222551,
      "loss": 1.404,
      "num_input_tokens_seen": 2334000,
      "step": 473,
      "train_runtime": 1625.2386,
      "train_tokens_per_second": 1436.097
    },
    {
      "epoch": 0.1743448275862069,
      "grad_norm": 0.3338044583797455,
      "learning_rate": 0.00016551215917464998,
      "loss": 1.4175,
      "num_input_tokens_seen": 2341040,
      "step": 474,
      "train_runtime": 1629.7076,
      "train_tokens_per_second": 1436.479
    },
    {
      "epoch": 0.17471264367816092,
      "grad_norm": 0.3517896234989166,
      "learning_rate": 0.00016543846720707444,
      "loss": 1.433,
      "num_input_tokens_seen": 2347440,
      "step": 475,
      "train_runtime": 1633.9029,
      "train_tokens_per_second": 1436.707
    },
    {
      "epoch": 0.17508045977011494,
      "grad_norm": 0.32335546612739563,
      "learning_rate": 0.00016536477523949888,
      "loss": 1.3124,
      "num_input_tokens_seen": 2354816,
      "step": 476,
      "train_runtime": 1638.6389,
      "train_tokens_per_second": 1437.056
    },
    {
      "epoch": 0.17544827586206896,
      "grad_norm": 0.3357498049736023,
      "learning_rate": 0.00016529108327192337,
      "loss": 1.3454,
      "num_input_tokens_seen": 2358736,
      "step": 477,
      "train_runtime": 1641.4382,
      "train_tokens_per_second": 1436.993
    },
    {
      "epoch": 0.17581609195402298,
      "grad_norm": 0.3742138147354126,
      "learning_rate": 0.00016521739130434784,
      "loss": 1.3781,
      "num_input_tokens_seen": 2362496,
      "step": 478,
      "train_runtime": 1644.0364,
      "train_tokens_per_second": 1437.01
    },
    {
      "epoch": 0.176183908045977,
      "grad_norm": 0.3452815115451813,
      "learning_rate": 0.0001651436993367723,
      "loss": 1.3116,
      "num_input_tokens_seen": 2366656,
      "step": 479,
      "train_runtime": 1646.9372,
      "train_tokens_per_second": 1437.004
    },
    {
      "epoch": 0.17655172413793102,
      "grad_norm": 0.31542301177978516,
      "learning_rate": 0.00016507000736919677,
      "loss": 1.4107,
      "num_input_tokens_seen": 2371408,
      "step": 480,
      "train_runtime": 1650.2199,
      "train_tokens_per_second": 1437.025
    },
    {
      "epoch": 0.17691954022988507,
      "grad_norm": 0.3817698359489441,
      "learning_rate": 0.00016499631540162123,
      "loss": 1.2445,
      "num_input_tokens_seen": 2375584,
      "step": 481,
      "train_runtime": 1655.7192,
      "train_tokens_per_second": 1434.775
    },
    {
      "epoch": 0.1772873563218391,
      "grad_norm": 0.37903985381126404,
      "learning_rate": 0.0001649226234340457,
      "loss": 1.2607,
      "num_input_tokens_seen": 2381008,
      "step": 482,
      "train_runtime": 1659.2591,
      "train_tokens_per_second": 1434.983
    },
    {
      "epoch": 0.1776551724137931,
      "grad_norm": 0.37075647711753845,
      "learning_rate": 0.00016484893146647016,
      "loss": 1.2386,
      "num_input_tokens_seen": 2385280,
      "step": 483,
      "train_runtime": 1662.2592,
      "train_tokens_per_second": 1434.963
    },
    {
      "epoch": 0.17802298850574713,
      "grad_norm": 0.3169528841972351,
      "learning_rate": 0.00016477523949889463,
      "loss": 1.3309,
      "num_input_tokens_seen": 2391648,
      "step": 484,
      "train_runtime": 1666.3337,
      "train_tokens_per_second": 1435.276
    },
    {
      "epoch": 0.17839080459770115,
      "grad_norm": 0.3421877920627594,
      "learning_rate": 0.0001647015475313191,
      "loss": 1.4214,
      "num_input_tokens_seen": 2395536,
      "step": 485,
      "train_runtime": 1669.0601,
      "train_tokens_per_second": 1435.26
    },
    {
      "epoch": 0.17875862068965517,
      "grad_norm": 0.35942426323890686,
      "learning_rate": 0.00016462785556374356,
      "loss": 1.3059,
      "num_input_tokens_seen": 2399312,
      "step": 486,
      "train_runtime": 1671.8617,
      "train_tokens_per_second": 1435.114
    },
    {
      "epoch": 0.1791264367816092,
      "grad_norm": 0.3300853967666626,
      "learning_rate": 0.00016455416359616802,
      "loss": 1.3302,
      "num_input_tokens_seen": 2404816,
      "step": 487,
      "train_runtime": 1675.5811,
      "train_tokens_per_second": 1435.213
    },
    {
      "epoch": 0.17949425287356321,
      "grad_norm": 0.32718613743782043,
      "learning_rate": 0.0001644804716285925,
      "loss": 1.2576,
      "num_input_tokens_seen": 2408816,
      "step": 488,
      "train_runtime": 1678.3337,
      "train_tokens_per_second": 1435.243
    },
    {
      "epoch": 0.17986206896551724,
      "grad_norm": 0.3190445601940155,
      "learning_rate": 0.00016440677966101695,
      "loss": 1.2407,
      "num_input_tokens_seen": 2414576,
      "step": 489,
      "train_runtime": 1682.1122,
      "train_tokens_per_second": 1435.443
    },
    {
      "epoch": 0.18022988505747126,
      "grad_norm": 0.38515108823776245,
      "learning_rate": 0.00016433308769344142,
      "loss": 1.3565,
      "num_input_tokens_seen": 2418080,
      "step": 490,
      "train_runtime": 1684.6028,
      "train_tokens_per_second": 1435.401
    },
    {
      "epoch": 0.18059770114942528,
      "grad_norm": 0.34255078434944153,
      "learning_rate": 0.00016425939572586588,
      "loss": 1.2903,
      "num_input_tokens_seen": 2424032,
      "step": 491,
      "train_runtime": 1688.5065,
      "train_tokens_per_second": 1435.607
    },
    {
      "epoch": 0.1809655172413793,
      "grad_norm": 0.37366923689842224,
      "learning_rate": 0.00016418570375829035,
      "loss": 1.3299,
      "num_input_tokens_seen": 2428160,
      "step": 492,
      "train_runtime": 1691.4537,
      "train_tokens_per_second": 1435.546
    },
    {
      "epoch": 0.18133333333333335,
      "grad_norm": 0.33095481991767883,
      "learning_rate": 0.00016411201179071484,
      "loss": 1.14,
      "num_input_tokens_seen": 2433200,
      "step": 493,
      "train_runtime": 1694.816,
      "train_tokens_per_second": 1435.672
    },
    {
      "epoch": 0.18170114942528737,
      "grad_norm": 0.31714820861816406,
      "learning_rate": 0.00016403831982313928,
      "loss": 1.3351,
      "num_input_tokens_seen": 2439088,
      "step": 494,
      "train_runtime": 1698.6442,
      "train_tokens_per_second": 1435.903
    },
    {
      "epoch": 0.1820689655172414,
      "grad_norm": 0.34027791023254395,
      "learning_rate": 0.00016396462785556374,
      "loss": 1.3259,
      "num_input_tokens_seen": 2444304,
      "step": 495,
      "train_runtime": 1702.1216,
      "train_tokens_per_second": 1436.034
    },
    {
      "epoch": 0.1824367816091954,
      "grad_norm": 0.34417229890823364,
      "learning_rate": 0.0001638909358879882,
      "loss": 1.3235,
      "num_input_tokens_seen": 2448736,
      "step": 496,
      "train_runtime": 1705.1225,
      "train_tokens_per_second": 1436.106
    },
    {
      "epoch": 0.18280459770114943,
      "grad_norm": 0.3466030955314636,
      "learning_rate": 0.0001638172439204127,
      "loss": 1.3202,
      "num_input_tokens_seen": 2454352,
      "step": 497,
      "train_runtime": 1708.8569,
      "train_tokens_per_second": 1436.254
    },
    {
      "epoch": 0.18317241379310345,
      "grad_norm": 0.3568847179412842,
      "learning_rate": 0.00016374355195283714,
      "loss": 1.3818,
      "num_input_tokens_seen": 2458944,
      "step": 498,
      "train_runtime": 1712.0227,
      "train_tokens_per_second": 1436.28
    },
    {
      "epoch": 0.18354022988505747,
      "grad_norm": 0.3216519355773926,
      "learning_rate": 0.0001636698599852616,
      "loss": 1.3315,
      "num_input_tokens_seen": 2462848,
      "step": 499,
      "train_runtime": 1714.7409,
      "train_tokens_per_second": 1436.28
    },
    {
      "epoch": 0.1839080459770115,
      "grad_norm": 0.32724398374557495,
      "learning_rate": 0.0001635961680176861,
      "loss": 1.2368,
      "num_input_tokens_seen": 2466752,
      "step": 500,
      "train_runtime": 1717.5292,
      "train_tokens_per_second": 1436.221
    },
    {
      "epoch": 0.1842758620689655,
      "grad_norm": 0.37495991587638855,
      "learning_rate": 0.00016352247605011056,
      "loss": 1.319,
      "num_input_tokens_seen": 2470656,
      "step": 501,
      "train_runtime": 1720.2957,
      "train_tokens_per_second": 1436.181
    },
    {
      "epoch": 0.18464367816091953,
      "grad_norm": 0.30906328558921814,
      "learning_rate": 0.00016344878408253502,
      "loss": 1.1545,
      "num_input_tokens_seen": 2475792,
      "step": 502,
      "train_runtime": 1723.7368,
      "train_tokens_per_second": 1436.294
    },
    {
      "epoch": 0.18501149425287355,
      "grad_norm": 0.34888771176338196,
      "learning_rate": 0.00016337509211495946,
      "loss": 1.2204,
      "num_input_tokens_seen": 2480400,
      "step": 503,
      "train_runtime": 1726.9576,
      "train_tokens_per_second": 1436.283
    },
    {
      "epoch": 0.18537931034482757,
      "grad_norm": 0.3419767916202545,
      "learning_rate": 0.00016330140014738395,
      "loss": 1.2839,
      "num_input_tokens_seen": 2485456,
      "step": 504,
      "train_runtime": 1730.2338,
      "train_tokens_per_second": 1436.486
    },
    {
      "epoch": 0.1857471264367816,
      "grad_norm": 0.3836863338947296,
      "learning_rate": 0.00016322770817980842,
      "loss": 1.3313,
      "num_input_tokens_seen": 2488864,
      "step": 505,
      "train_runtime": 1732.7458,
      "train_tokens_per_second": 1436.37
    },
    {
      "epoch": 0.18611494252873564,
      "grad_norm": 0.3446231186389923,
      "learning_rate": 0.00016315401621223288,
      "loss": 1.1375,
      "num_input_tokens_seen": 2492512,
      "step": 506,
      "train_runtime": 1735.3503,
      "train_tokens_per_second": 1436.316
    },
    {
      "epoch": 0.18648275862068966,
      "grad_norm": 0.33684930205345154,
      "learning_rate": 0.00016308032424465735,
      "loss": 1.1534,
      "num_input_tokens_seen": 2496864,
      "step": 507,
      "train_runtime": 1738.3742,
      "train_tokens_per_second": 1436.321
    },
    {
      "epoch": 0.18685057471264369,
      "grad_norm": 0.36668455600738525,
      "learning_rate": 0.0001630066322770818,
      "loss": 1.2738,
      "num_input_tokens_seen": 2500288,
      "step": 508,
      "train_runtime": 1740.9184,
      "train_tokens_per_second": 1436.189
    },
    {
      "epoch": 0.1872183908045977,
      "grad_norm": 0.43545395135879517,
      "learning_rate": 0.00016293294030950628,
      "loss": 1.2834,
      "num_input_tokens_seen": 2506624,
      "step": 509,
      "train_runtime": 1745.0035,
      "train_tokens_per_second": 1436.458
    },
    {
      "epoch": 0.18758620689655173,
      "grad_norm": 0.3866064250469208,
      "learning_rate": 0.00016285924834193074,
      "loss": 1.2799,
      "num_input_tokens_seen": 2511216,
      "step": 510,
      "train_runtime": 1748.1168,
      "train_tokens_per_second": 1436.526
    },
    {
      "epoch": 0.18795402298850575,
      "grad_norm": 0.43094202876091003,
      "learning_rate": 0.0001627855563743552,
      "loss": 1.3075,
      "num_input_tokens_seen": 2515040,
      "step": 511,
      "train_runtime": 1752.7918,
      "train_tokens_per_second": 1434.877
    },
    {
      "epoch": 0.18832183908045977,
      "grad_norm": 0.3287125527858734,
      "learning_rate": 0.00016271186440677967,
      "loss": 1.2294,
      "num_input_tokens_seen": 2518608,
      "step": 512,
      "train_runtime": 1755.292,
      "train_tokens_per_second": 1434.866
    },
    {
      "epoch": 0.1886896551724138,
      "grad_norm": 0.38001322746276855,
      "learning_rate": 0.00016263817243920413,
      "loss": 1.2967,
      "num_input_tokens_seen": 2524528,
      "step": 513,
      "train_runtime": 1759.0632,
      "train_tokens_per_second": 1435.155
    },
    {
      "epoch": 0.1890574712643678,
      "grad_norm": 0.36320173740386963,
      "learning_rate": 0.0001625644804716286,
      "loss": 1.2356,
      "num_input_tokens_seen": 2528528,
      "step": 514,
      "train_runtime": 1761.8651,
      "train_tokens_per_second": 1435.143
    },
    {
      "epoch": 0.18942528735632183,
      "grad_norm": 0.32575723528862,
      "learning_rate": 0.00016249078850405306,
      "loss": 1.2527,
      "num_input_tokens_seen": 2533584,
      "step": 515,
      "train_runtime": 1765.2838,
      "train_tokens_per_second": 1435.228
    },
    {
      "epoch": 0.18979310344827585,
      "grad_norm": 0.37583526968955994,
      "learning_rate": 0.00016241709653647753,
      "loss": 1.2864,
      "num_input_tokens_seen": 2537456,
      "step": 516,
      "train_runtime": 1767.9687,
      "train_tokens_per_second": 1435.238
    },
    {
      "epoch": 0.19016091954022987,
      "grad_norm": 0.3317106068134308,
      "learning_rate": 0.000162343404568902,
      "loss": 1.1857,
      "num_input_tokens_seen": 2540880,
      "step": 517,
      "train_runtime": 1770.4916,
      "train_tokens_per_second": 1435.127
    },
    {
      "epoch": 0.19052873563218392,
      "grad_norm": 0.29974696040153503,
      "learning_rate": 0.00016226971260132646,
      "loss": 1.2705,
      "num_input_tokens_seen": 2548032,
      "step": 518,
      "train_runtime": 1775.034,
      "train_tokens_per_second": 1435.484
    },
    {
      "epoch": 0.19089655172413794,
      "grad_norm": 0.3364499807357788,
      "learning_rate": 0.00016219602063375092,
      "loss": 1.2728,
      "num_input_tokens_seen": 2552400,
      "step": 519,
      "train_runtime": 1778.0662,
      "train_tokens_per_second": 1435.492
    },
    {
      "epoch": 0.19126436781609196,
      "grad_norm": 0.3244878649711609,
      "learning_rate": 0.00016212232866617542,
      "loss": 1.2699,
      "num_input_tokens_seen": 2559888,
      "step": 520,
      "train_runtime": 1782.8718,
      "train_tokens_per_second": 1435.823
    },
    {
      "epoch": 0.19163218390804598,
      "grad_norm": 0.3346767723560333,
      "learning_rate": 0.00016204863669859985,
      "loss": 1.1798,
      "num_input_tokens_seen": 2563360,
      "step": 521,
      "train_runtime": 1785.3748,
      "train_tokens_per_second": 1435.755
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.3296874761581421,
      "learning_rate": 0.00016197494473102432,
      "loss": 1.2463,
      "num_input_tokens_seen": 2566000,
      "step": 522,
      "train_runtime": 1787.4619,
      "train_tokens_per_second": 1435.555
    },
    {
      "epoch": 0.19236781609195402,
      "grad_norm": 0.3505805730819702,
      "learning_rate": 0.00016190125276344878,
      "loss": 1.2946,
      "num_input_tokens_seen": 2570928,
      "step": 523,
      "train_runtime": 1790.8093,
      "train_tokens_per_second": 1435.624
    },
    {
      "epoch": 0.19273563218390805,
      "grad_norm": 0.3358633816242218,
      "learning_rate": 0.00016182756079587327,
      "loss": 1.3498,
      "num_input_tokens_seen": 2576352,
      "step": 524,
      "train_runtime": 1794.2966,
      "train_tokens_per_second": 1435.856
    },
    {
      "epoch": 0.19310344827586207,
      "grad_norm": 0.3377569615840912,
      "learning_rate": 0.00016175386882829774,
      "loss": 1.452,
      "num_input_tokens_seen": 2579856,
      "step": 525,
      "train_runtime": 1796.8332,
      "train_tokens_per_second": 1435.779
    },
    {
      "epoch": 0.1934712643678161,
      "grad_norm": 0.39165809750556946,
      "learning_rate": 0.00016168017686072218,
      "loss": 1.2713,
      "num_input_tokens_seen": 2583376,
      "step": 526,
      "train_runtime": 1799.3331,
      "train_tokens_per_second": 1435.741
    },
    {
      "epoch": 0.1938390804597701,
      "grad_norm": 0.33097031712532043,
      "learning_rate": 0.00016160648489314664,
      "loss": 1.1863,
      "num_input_tokens_seen": 2588144,
      "step": 527,
      "train_runtime": 1802.5206,
      "train_tokens_per_second": 1435.847
    },
    {
      "epoch": 0.19420689655172413,
      "grad_norm": 0.32664138078689575,
      "learning_rate": 0.00016153279292557113,
      "loss": 1.2256,
      "num_input_tokens_seen": 2592272,
      "step": 528,
      "train_runtime": 1805.4258,
      "train_tokens_per_second": 1435.823
    },
    {
      "epoch": 0.19457471264367815,
      "grad_norm": 0.35744956135749817,
      "learning_rate": 0.0001614591009579956,
      "loss": 1.2275,
      "num_input_tokens_seen": 2598816,
      "step": 529,
      "train_runtime": 1809.5927,
      "train_tokens_per_second": 1436.133
    },
    {
      "epoch": 0.19494252873563217,
      "grad_norm": 0.36398911476135254,
      "learning_rate": 0.00016138540899042006,
      "loss": 1.3013,
      "num_input_tokens_seen": 2603792,
      "step": 530,
      "train_runtime": 1812.8989,
      "train_tokens_per_second": 1436.259
    },
    {
      "epoch": 0.19531034482758622,
      "grad_norm": 0.3198075592517853,
      "learning_rate": 0.0001613117170228445,
      "loss": 1.3522,
      "num_input_tokens_seen": 2611088,
      "step": 531,
      "train_runtime": 1817.6382,
      "train_tokens_per_second": 1436.528
    },
    {
      "epoch": 0.19567816091954024,
      "grad_norm": 0.3598136305809021,
      "learning_rate": 0.000161238025055269,
      "loss": 1.3588,
      "num_input_tokens_seen": 2614496,
      "step": 532,
      "train_runtime": 1820.2013,
      "train_tokens_per_second": 1436.377
    },
    {
      "epoch": 0.19604597701149426,
      "grad_norm": 0.35406559705734253,
      "learning_rate": 0.00016116433308769346,
      "loss": 1.27,
      "num_input_tokens_seen": 2620320,
      "step": 533,
      "train_runtime": 1823.9909,
      "train_tokens_per_second": 1436.586
    },
    {
      "epoch": 0.19641379310344828,
      "grad_norm": 0.37063267827033997,
      "learning_rate": 0.00016109064112011792,
      "loss": 1.311,
      "num_input_tokens_seen": 2627840,
      "step": 534,
      "train_runtime": 1828.7531,
      "train_tokens_per_second": 1436.957
    },
    {
      "epoch": 0.1967816091954023,
      "grad_norm": 0.3614117205142975,
      "learning_rate": 0.00016101694915254236,
      "loss": 1.1944,
      "num_input_tokens_seen": 2634128,
      "step": 535,
      "train_runtime": 1832.8949,
      "train_tokens_per_second": 1437.141
    },
    {
      "epoch": 0.19714942528735632,
      "grad_norm": 0.3317558765411377,
      "learning_rate": 0.00016094325718496685,
      "loss": 1.3098,
      "num_input_tokens_seen": 2643760,
      "step": 536,
      "train_runtime": 1838.7941,
      "train_tokens_per_second": 1437.768
    },
    {
      "epoch": 0.19751724137931034,
      "grad_norm": 0.34399816393852234,
      "learning_rate": 0.00016086956521739132,
      "loss": 1.1071,
      "num_input_tokens_seen": 2647184,
      "step": 537,
      "train_runtime": 1841.2836,
      "train_tokens_per_second": 1437.684
    },
    {
      "epoch": 0.19788505747126436,
      "grad_norm": 0.3315516412258148,
      "learning_rate": 0.00016079587324981578,
      "loss": 1.3203,
      "num_input_tokens_seen": 2653200,
      "step": 538,
      "train_runtime": 1845.2351,
      "train_tokens_per_second": 1437.866
    },
    {
      "epoch": 0.19825287356321838,
      "grad_norm": 0.3801349401473999,
      "learning_rate": 0.00016072218128224025,
      "loss": 1.2821,
      "num_input_tokens_seen": 2657136,
      "step": 539,
      "train_runtime": 1848.0085,
      "train_tokens_per_second": 1437.838
    },
    {
      "epoch": 0.1986206896551724,
      "grad_norm": 0.34804943203926086,
      "learning_rate": 0.0001606484893146647,
      "loss": 1.0843,
      "num_input_tokens_seen": 2660224,
      "step": 540,
      "train_runtime": 1850.3064,
      "train_tokens_per_second": 1437.721
    },
    {
      "epoch": 0.19898850574712643,
      "grad_norm": 0.42285385727882385,
      "learning_rate": 0.00016057479734708918,
      "loss": 1.3023,
      "num_input_tokens_seen": 2663152,
      "step": 541,
      "train_runtime": 1854.7352,
      "train_tokens_per_second": 1435.866
    },
    {
      "epoch": 0.19935632183908045,
      "grad_norm": 0.33356553316116333,
      "learning_rate": 0.00016050110537951364,
      "loss": 1.3245,
      "num_input_tokens_seen": 2667472,
      "step": 542,
      "train_runtime": 1857.6688,
      "train_tokens_per_second": 1435.924
    },
    {
      "epoch": 0.1997241379310345,
      "grad_norm": 0.322724848985672,
      "learning_rate": 0.0001604274134119381,
      "loss": 1.1768,
      "num_input_tokens_seen": 2672304,
      "step": 543,
      "train_runtime": 1860.8516,
      "train_tokens_per_second": 1436.065
    },
    {
      "epoch": 0.20009195402298852,
      "grad_norm": 0.36868584156036377,
      "learning_rate": 0.00016035372144436257,
      "loss": 1.3712,
      "num_input_tokens_seen": 2676272,
      "step": 544,
      "train_runtime": 1863.6677,
      "train_tokens_per_second": 1436.024
    },
    {
      "epoch": 0.20009195402298852,
      "eval_loss": 1.2722872495651245,
      "eval_runtime": 15.4745,
      "eval_samples_per_second": 64.623,
      "eval_steps_per_second": 4.071,
      "num_input_tokens_seen": 2676272,
      "step": 544
    },
    {
      "epoch": 0.20045977011494254,
      "grad_norm": 0.36055392026901245,
      "learning_rate": 0.00016028002947678704,
      "loss": 1.3256,
      "num_input_tokens_seen": 2682256,
      "step": 545,
      "train_runtime": 1882.9622,
      "train_tokens_per_second": 1424.487
    },
    {
      "epoch": 0.20082758620689656,
      "grad_norm": 0.3377590775489807,
      "learning_rate": 0.0001602063375092115,
      "loss": 1.1309,
      "num_input_tokens_seen": 2687840,
      "step": 546,
      "train_runtime": 1886.6051,
      "train_tokens_per_second": 1424.697
    },
    {
      "epoch": 0.20119540229885058,
      "grad_norm": 0.3443763256072998,
      "learning_rate": 0.00016013264554163597,
      "loss": 1.2654,
      "num_input_tokens_seen": 2691312,
      "step": 547,
      "train_runtime": 1889.0955,
      "train_tokens_per_second": 1424.656
    },
    {
      "epoch": 0.2015632183908046,
      "grad_norm": 0.39379337430000305,
      "learning_rate": 0.00016005895357406046,
      "loss": 1.42,
      "num_input_tokens_seen": 2695120,
      "step": 548,
      "train_runtime": 1891.816,
      "train_tokens_per_second": 1424.621
    },
    {
      "epoch": 0.20193103448275862,
      "grad_norm": 0.29482370615005493,
      "learning_rate": 0.0001599852616064849,
      "loss": 1.2013,
      "num_input_tokens_seen": 2699568,
      "step": 549,
      "train_runtime": 1894.7995,
      "train_tokens_per_second": 1424.725
    },
    {
      "epoch": 0.20229885057471264,
      "grad_norm": 0.3238631784915924,
      "learning_rate": 0.00015991156963890936,
      "loss": 1.3776,
      "num_input_tokens_seen": 2704608,
      "step": 550,
      "train_runtime": 1898.1532,
      "train_tokens_per_second": 1424.863
    },
    {
      "epoch": 0.20266666666666666,
      "grad_norm": 0.31853145360946655,
      "learning_rate": 0.00015983787767133382,
      "loss": 1.2264,
      "num_input_tokens_seen": 2709088,
      "step": 551,
      "train_runtime": 1901.2593,
      "train_tokens_per_second": 1424.891
    },
    {
      "epoch": 0.20303448275862068,
      "grad_norm": 0.34621456265449524,
      "learning_rate": 0.00015976418570375832,
      "loss": 1.1675,
      "num_input_tokens_seen": 2713008,
      "step": 552,
      "train_runtime": 1904.0395,
      "train_tokens_per_second": 1424.87
    },
    {
      "epoch": 0.2034022988505747,
      "grad_norm": 0.32944974303245544,
      "learning_rate": 0.00015969049373618275,
      "loss": 1.4432,
      "num_input_tokens_seen": 2720256,
      "step": 553,
      "train_runtime": 1908.6301,
      "train_tokens_per_second": 1425.24
    },
    {
      "epoch": 0.20377011494252872,
      "grad_norm": 0.38964101672172546,
      "learning_rate": 0.00015961680176860722,
      "loss": 1.4091,
      "num_input_tokens_seen": 2723888,
      "step": 554,
      "train_runtime": 1911.2318,
      "train_tokens_per_second": 1425.2
    },
    {
      "epoch": 0.20413793103448277,
      "grad_norm": 0.3687337338924408,
      "learning_rate": 0.00015954310980103168,
      "loss": 1.3001,
      "num_input_tokens_seen": 2728176,
      "step": 555,
      "train_runtime": 1914.1401,
      "train_tokens_per_second": 1425.275
    },
    {
      "epoch": 0.2045057471264368,
      "grad_norm": 0.33243492245674133,
      "learning_rate": 0.00015946941783345618,
      "loss": 1.3448,
      "num_input_tokens_seen": 2732048,
      "step": 556,
      "train_runtime": 1916.8689,
      "train_tokens_per_second": 1425.266
    },
    {
      "epoch": 0.20487356321839081,
      "grad_norm": 0.3270266652107239,
      "learning_rate": 0.00015939572586588064,
      "loss": 1.1297,
      "num_input_tokens_seen": 2738000,
      "step": 557,
      "train_runtime": 1920.8298,
      "train_tokens_per_second": 1425.426
    },
    {
      "epoch": 0.20524137931034483,
      "grad_norm": 0.35209283232688904,
      "learning_rate": 0.00015932203389830508,
      "loss": 1.2505,
      "num_input_tokens_seen": 2741328,
      "step": 558,
      "train_runtime": 1923.3322,
      "train_tokens_per_second": 1425.301
    },
    {
      "epoch": 0.20560919540229886,
      "grad_norm": 0.31420478224754333,
      "learning_rate": 0.00015924834193072954,
      "loss": 1.1736,
      "num_input_tokens_seen": 2748224,
      "step": 559,
      "train_runtime": 1927.8068,
      "train_tokens_per_second": 1425.57
    },
    {
      "epoch": 0.20597701149425288,
      "grad_norm": 0.3479740619659424,
      "learning_rate": 0.00015917464996315404,
      "loss": 1.1734,
      "num_input_tokens_seen": 2751024,
      "step": 560,
      "train_runtime": 1930.0153,
      "train_tokens_per_second": 1425.39
    },
    {
      "epoch": 0.2063448275862069,
      "grad_norm": 0.33000895380973816,
      "learning_rate": 0.0001591009579955785,
      "loss": 1.411,
      "num_input_tokens_seen": 2758640,
      "step": 561,
      "train_runtime": 1934.8688,
      "train_tokens_per_second": 1425.75
    },
    {
      "epoch": 0.20671264367816092,
      "grad_norm": 0.3301312327384949,
      "learning_rate": 0.00015902726602800296,
      "loss": 1.3243,
      "num_input_tokens_seen": 2770816,
      "step": 562,
      "train_runtime": 1942.2134,
      "train_tokens_per_second": 1426.628
    },
    {
      "epoch": 0.20708045977011494,
      "grad_norm": 0.358114629983902,
      "learning_rate": 0.0001589535740604274,
      "loss": 1.1099,
      "num_input_tokens_seen": 2775056,
      "step": 563,
      "train_runtime": 1945.1652,
      "train_tokens_per_second": 1426.643
    },
    {
      "epoch": 0.20744827586206896,
      "grad_norm": 0.39302560687065125,
      "learning_rate": 0.0001588798820928519,
      "loss": 1.1804,
      "num_input_tokens_seen": 2780848,
      "step": 564,
      "train_runtime": 1949.0215,
      "train_tokens_per_second": 1426.792
    },
    {
      "epoch": 0.20781609195402298,
      "grad_norm": 0.33622652292251587,
      "learning_rate": 0.00015880619012527636,
      "loss": 1.2061,
      "num_input_tokens_seen": 2785424,
      "step": 565,
      "train_runtime": 1952.1362,
      "train_tokens_per_second": 1426.859
    },
    {
      "epoch": 0.208183908045977,
      "grad_norm": 0.35917094349861145,
      "learning_rate": 0.00015873249815770082,
      "loss": 1.2738,
      "num_input_tokens_seen": 2790352,
      "step": 566,
      "train_runtime": 1955.4066,
      "train_tokens_per_second": 1426.993
    },
    {
      "epoch": 0.20855172413793102,
      "grad_norm": 0.41483211517333984,
      "learning_rate": 0.00015865880619012526,
      "loss": 1.3201,
      "num_input_tokens_seen": 2793760,
      "step": 567,
      "train_runtime": 1957.8828,
      "train_tokens_per_second": 1426.929
    },
    {
      "epoch": 0.20891954022988507,
      "grad_norm": 0.367186039686203,
      "learning_rate": 0.00015858511422254975,
      "loss": 1.2594,
      "num_input_tokens_seen": 2797104,
      "step": 568,
      "train_runtime": 1960.3465,
      "train_tokens_per_second": 1426.842
    },
    {
      "epoch": 0.2092873563218391,
      "grad_norm": 0.35813453793525696,
      "learning_rate": 0.00015851142225497422,
      "loss": 1.3491,
      "num_input_tokens_seen": 2805328,
      "step": 569,
      "train_runtime": 1965.4896,
      "train_tokens_per_second": 1427.292
    },
    {
      "epoch": 0.2096551724137931,
      "grad_norm": 0.34246182441711426,
      "learning_rate": 0.00015843773028739868,
      "loss": 1.2319,
      "num_input_tokens_seen": 2811904,
      "step": 570,
      "train_runtime": 1969.6662,
      "train_tokens_per_second": 1427.604
    },
    {
      "epoch": 0.21002298850574713,
      "grad_norm": 0.364341139793396,
      "learning_rate": 0.00015836403831982315,
      "loss": 1.4294,
      "num_input_tokens_seen": 2819184,
      "step": 571,
      "train_runtime": 1976.3213,
      "train_tokens_per_second": 1426.481
    },
    {
      "epoch": 0.21039080459770115,
      "grad_norm": 0.3529500961303711,
      "learning_rate": 0.0001582903463522476,
      "loss": 1.3549,
      "num_input_tokens_seen": 2823456,
      "step": 572,
      "train_runtime": 1979.2725,
      "train_tokens_per_second": 1426.512
    },
    {
      "epoch": 0.21075862068965517,
      "grad_norm": 0.3735297918319702,
      "learning_rate": 0.00015821665438467208,
      "loss": 1.3024,
      "num_input_tokens_seen": 2827664,
      "step": 573,
      "train_runtime": 1982.2422,
      "train_tokens_per_second": 1426.498
    },
    {
      "epoch": 0.2111264367816092,
      "grad_norm": 0.3497689664363861,
      "learning_rate": 0.00015814296241709654,
      "loss": 1.1662,
      "num_input_tokens_seen": 2831728,
      "step": 574,
      "train_runtime": 1985.0397,
      "train_tokens_per_second": 1426.535
    },
    {
      "epoch": 0.21149425287356322,
      "grad_norm": 0.4035835266113281,
      "learning_rate": 0.000158069270449521,
      "loss": 1.3886,
      "num_input_tokens_seen": 2835472,
      "step": 575,
      "train_runtime": 1987.6621,
      "train_tokens_per_second": 1426.536
    },
    {
      "epoch": 0.21186206896551724,
      "grad_norm": 0.33833611011505127,
      "learning_rate": 0.00015799557848194547,
      "loss": 1.3622,
      "num_input_tokens_seen": 2839664,
      "step": 576,
      "train_runtime": 1990.5618,
      "train_tokens_per_second": 1426.564
    },
    {
      "epoch": 0.21222988505747126,
      "grad_norm": 0.33022499084472656,
      "learning_rate": 0.00015792188651436994,
      "loss": 1.2793,
      "num_input_tokens_seen": 2845008,
      "step": 577,
      "train_runtime": 1994.1098,
      "train_tokens_per_second": 1426.706
    },
    {
      "epoch": 0.21259770114942528,
      "grad_norm": 0.3520048260688782,
      "learning_rate": 0.0001578481945467944,
      "loss": 1.353,
      "num_input_tokens_seen": 2848944,
      "step": 578,
      "train_runtime": 1996.8617,
      "train_tokens_per_second": 1426.711
    },
    {
      "epoch": 0.2129655172413793,
      "grad_norm": 0.33845770359039307,
      "learning_rate": 0.00015777450257921887,
      "loss": 1.2187,
      "num_input_tokens_seen": 2853120,
      "step": 579,
      "train_runtime": 1999.7525,
      "train_tokens_per_second": 1426.737
    },
    {
      "epoch": 0.21333333333333335,
      "grad_norm": 0.37855276465415955,
      "learning_rate": 0.00015770081061164336,
      "loss": 1.2547,
      "num_input_tokens_seen": 2856768,
      "step": 580,
      "train_runtime": 2002.3631,
      "train_tokens_per_second": 1426.698
    },
    {
      "epoch": 0.21370114942528737,
      "grad_norm": 0.32887569069862366,
      "learning_rate": 0.0001576271186440678,
      "loss": 1.2656,
      "num_input_tokens_seen": 2861136,
      "step": 581,
      "train_runtime": 2005.3736,
      "train_tokens_per_second": 1426.735
    },
    {
      "epoch": 0.2140689655172414,
      "grad_norm": 0.3663339912891388,
      "learning_rate": 0.00015755342667649226,
      "loss": 1.3917,
      "num_input_tokens_seen": 2866848,
      "step": 582,
      "train_runtime": 2009.0935,
      "train_tokens_per_second": 1426.936
    },
    {
      "epoch": 0.2144367816091954,
      "grad_norm": 0.37052851915359497,
      "learning_rate": 0.00015747973470891673,
      "loss": 1.3527,
      "num_input_tokens_seen": 2870224,
      "step": 583,
      "train_runtime": 2011.4821,
      "train_tokens_per_second": 1426.92
    },
    {
      "epoch": 0.21480459770114943,
      "grad_norm": 0.33706924319267273,
      "learning_rate": 0.00015740604274134122,
      "loss": 1.2399,
      "num_input_tokens_seen": 2874144,
      "step": 584,
      "train_runtime": 2014.1927,
      "train_tokens_per_second": 1426.946
    },
    {
      "epoch": 0.21517241379310345,
      "grad_norm": 0.3673574924468994,
      "learning_rate": 0.00015733235077376566,
      "loss": 1.2418,
      "num_input_tokens_seen": 2880528,
      "step": 585,
      "train_runtime": 2018.2948,
      "train_tokens_per_second": 1427.209
    },
    {
      "epoch": 0.21554022988505747,
      "grad_norm": 0.3524503707885742,
      "learning_rate": 0.00015725865880619012,
      "loss": 1.3842,
      "num_input_tokens_seen": 2887648,
      "step": 586,
      "train_runtime": 2022.8254,
      "train_tokens_per_second": 1427.532
    },
    {
      "epoch": 0.2159080459770115,
      "grad_norm": 0.3508240282535553,
      "learning_rate": 0.00015718496683861459,
      "loss": 1.4001,
      "num_input_tokens_seen": 2892816,
      "step": 587,
      "train_runtime": 2026.3138,
      "train_tokens_per_second": 1427.625
    },
    {
      "epoch": 0.2162758620689655,
      "grad_norm": 0.34460121393203735,
      "learning_rate": 0.00015711127487103908,
      "loss": 1.1965,
      "num_input_tokens_seen": 2896704,
      "step": 588,
      "train_runtime": 2029.0787,
      "train_tokens_per_second": 1427.596
    },
    {
      "epoch": 0.21664367816091953,
      "grad_norm": 0.3157898187637329,
      "learning_rate": 0.00015703758290346354,
      "loss": 1.2097,
      "num_input_tokens_seen": 2900912,
      "step": 589,
      "train_runtime": 2032.0618,
      "train_tokens_per_second": 1427.571
    },
    {
      "epoch": 0.21701149425287355,
      "grad_norm": 0.37705808877944946,
      "learning_rate": 0.00015696389093588798,
      "loss": 1.297,
      "num_input_tokens_seen": 2905200,
      "step": 590,
      "train_runtime": 2035.0686,
      "train_tokens_per_second": 1427.569
    },
    {
      "epoch": 0.21737931034482758,
      "grad_norm": 0.33461806178092957,
      "learning_rate": 0.00015689019896831244,
      "loss": 1.1567,
      "num_input_tokens_seen": 2910864,
      "step": 591,
      "train_runtime": 2038.7705,
      "train_tokens_per_second": 1427.755
    },
    {
      "epoch": 0.2177471264367816,
      "grad_norm": 0.3506442606449127,
      "learning_rate": 0.00015681650700073694,
      "loss": 1.2028,
      "num_input_tokens_seen": 2917024,
      "step": 592,
      "train_runtime": 2042.699,
      "train_tokens_per_second": 1428.024
    },
    {
      "epoch": 0.21811494252873564,
      "grad_norm": 0.3706601560115814,
      "learning_rate": 0.0001567428150331614,
      "loss": 1.3113,
      "num_input_tokens_seen": 2920928,
      "step": 593,
      "train_runtime": 2045.5071,
      "train_tokens_per_second": 1427.973
    },
    {
      "epoch": 0.21848275862068967,
      "grad_norm": 0.3287620544433594,
      "learning_rate": 0.00015666912306558587,
      "loss": 1.3289,
      "num_input_tokens_seen": 2928656,
      "step": 594,
      "train_runtime": 2050.3697,
      "train_tokens_per_second": 1428.355
    },
    {
      "epoch": 0.2188505747126437,
      "grad_norm": 0.30894625186920166,
      "learning_rate": 0.0001565954310980103,
      "loss": 1.1406,
      "num_input_tokens_seen": 2931920,
      "step": 595,
      "train_runtime": 2052.7269,
      "train_tokens_per_second": 1428.305
    },
    {
      "epoch": 0.2192183908045977,
      "grad_norm": 0.3249402344226837,
      "learning_rate": 0.0001565217391304348,
      "loss": 1.1654,
      "num_input_tokens_seen": 2936672,
      "step": 596,
      "train_runtime": 2055.9805,
      "train_tokens_per_second": 1428.356
    },
    {
      "epoch": 0.21958620689655173,
      "grad_norm": 0.3295873999595642,
      "learning_rate": 0.00015644804716285926,
      "loss": 1.2612,
      "num_input_tokens_seen": 2942768,
      "step": 597,
      "train_runtime": 2059.9486,
      "train_tokens_per_second": 1428.564
    },
    {
      "epoch": 0.21995402298850575,
      "grad_norm": 0.42991822957992554,
      "learning_rate": 0.00015637435519528373,
      "loss": 1.3704,
      "num_input_tokens_seen": 2947712,
      "step": 598,
      "train_runtime": 2063.2948,
      "train_tokens_per_second": 1428.643
    },
    {
      "epoch": 0.22032183908045977,
      "grad_norm": 0.3534640967845917,
      "learning_rate": 0.0001563006632277082,
      "loss": 1.2164,
      "num_input_tokens_seen": 2951280,
      "step": 599,
      "train_runtime": 2065.8086,
      "train_tokens_per_second": 1428.632
    },
    {
      "epoch": 0.2206896551724138,
      "grad_norm": 0.348354697227478,
      "learning_rate": 0.00015622697126013266,
      "loss": 1.2605,
      "num_input_tokens_seen": 2955248,
      "step": 600,
      "train_runtime": 2068.6458,
      "train_tokens_per_second": 1428.591
    },
    {
      "epoch": 0.2210574712643678,
      "grad_norm": 0.3614390790462494,
      "learning_rate": 0.00015615327929255712,
      "loss": 1.3987,
      "num_input_tokens_seen": 2958992,
      "step": 601,
      "train_runtime": 2073.2854,
      "train_tokens_per_second": 1427.2
    },
    {
      "epoch": 0.22142528735632183,
      "grad_norm": 0.3639795482158661,
      "learning_rate": 0.00015607958732498158,
      "loss": 1.3568,
      "num_input_tokens_seen": 2963344,
      "step": 602,
      "train_runtime": 2076.2633,
      "train_tokens_per_second": 1427.249
    },
    {
      "epoch": 0.22179310344827585,
      "grad_norm": 0.4041186571121216,
      "learning_rate": 0.00015600589535740605,
      "loss": 1.1738,
      "num_input_tokens_seen": 2967424,
      "step": 603,
      "train_runtime": 2079.0606,
      "train_tokens_per_second": 1427.291
    },
    {
      "epoch": 0.22216091954022987,
      "grad_norm": 0.3762064576148987,
      "learning_rate": 0.00015593220338983051,
      "loss": 1.2329,
      "num_input_tokens_seen": 2971952,
      "step": 604,
      "train_runtime": 2082.1878,
      "train_tokens_per_second": 1427.322
    },
    {
      "epoch": 0.22252873563218392,
      "grad_norm": 0.3315386474132538,
      "learning_rate": 0.00015585851142225498,
      "loss": 1.4211,
      "num_input_tokens_seen": 2975424,
      "step": 605,
      "train_runtime": 2084.7305,
      "train_tokens_per_second": 1427.246
    },
    {
      "epoch": 0.22289655172413794,
      "grad_norm": 0.35764938592910767,
      "learning_rate": 0.00015578481945467944,
      "loss": 1.374,
      "num_input_tokens_seen": 2978880,
      "step": 606,
      "train_runtime": 2087.2191,
      "train_tokens_per_second": 1427.2
    },
    {
      "epoch": 0.22326436781609196,
      "grad_norm": 0.3803774118423462,
      "learning_rate": 0.0001557111274871039,
      "loss": 1.2384,
      "num_input_tokens_seen": 2982160,
      "step": 607,
      "train_runtime": 2089.6994,
      "train_tokens_per_second": 1427.076
    },
    {
      "epoch": 0.22363218390804598,
      "grad_norm": 0.37342971563339233,
      "learning_rate": 0.00015563743551952837,
      "loss": 1.2087,
      "num_input_tokens_seen": 2985520,
      "step": 608,
      "train_runtime": 2092.2483,
      "train_tokens_per_second": 1426.943
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.38321128487586975,
      "learning_rate": 0.00015556374355195284,
      "loss": 1.2303,
      "num_input_tokens_seen": 2989152,
      "step": 609,
      "train_runtime": 2094.8329,
      "train_tokens_per_second": 1426.917
    },
    {
      "epoch": 0.22436781609195403,
      "grad_norm": 0.34238046407699585,
      "learning_rate": 0.0001554900515843773,
      "loss": 1.2595,
      "num_input_tokens_seen": 2995040,
      "step": 610,
      "train_runtime": 2098.636,
      "train_tokens_per_second": 1427.136
    },
    {
      "epoch": 0.22473563218390805,
      "grad_norm": 0.3561033308506012,
      "learning_rate": 0.00015541635961680177,
      "loss": 1.3446,
      "num_input_tokens_seen": 2998640,
      "step": 611,
      "train_runtime": 2101.2837,
      "train_tokens_per_second": 1427.052
    },
    {
      "epoch": 0.22510344827586207,
      "grad_norm": 0.35131025314331055,
      "learning_rate": 0.00015534266764922626,
      "loss": 1.1862,
      "num_input_tokens_seen": 3003376,
      "step": 612,
      "train_runtime": 2104.4455,
      "train_tokens_per_second": 1427.158
    },
    {
      "epoch": 0.2254712643678161,
      "grad_norm": 0.37832045555114746,
      "learning_rate": 0.0001552689756816507,
      "loss": 1.3236,
      "num_input_tokens_seen": 3008112,
      "step": 613,
      "train_runtime": 2107.5896,
      "train_tokens_per_second": 1427.276
    },
    {
      "epoch": 0.2258390804597701,
      "grad_norm": 0.35437673330307007,
      "learning_rate": 0.00015519528371407516,
      "loss": 1.1955,
      "num_input_tokens_seen": 3012304,
      "step": 614,
      "train_runtime": 2110.5478,
      "train_tokens_per_second": 1427.262
    },
    {
      "epoch": 0.22620689655172413,
      "grad_norm": 0.33689314126968384,
      "learning_rate": 0.00015512159174649963,
      "loss": 1.1469,
      "num_input_tokens_seen": 3016448,
      "step": 615,
      "train_runtime": 2113.422,
      "train_tokens_per_second": 1427.281
    },
    {
      "epoch": 0.22657471264367815,
      "grad_norm": 0.37024691700935364,
      "learning_rate": 0.00015504789977892412,
      "loss": 1.216,
      "num_input_tokens_seen": 3021536,
      "step": 616,
      "train_runtime": 2116.7711,
      "train_tokens_per_second": 1427.427
    },
    {
      "epoch": 0.22694252873563217,
      "grad_norm": 0.34888485074043274,
      "learning_rate": 0.00015497420781134856,
      "loss": 1.2034,
      "num_input_tokens_seen": 3031472,
      "step": 617,
      "train_runtime": 2122.8118,
      "train_tokens_per_second": 1428.046
    },
    {
      "epoch": 0.22731034482758622,
      "grad_norm": 0.3368101119995117,
      "learning_rate": 0.00015490051584377302,
      "loss": 1.1672,
      "num_input_tokens_seen": 3036832,
      "step": 618,
      "train_runtime": 2126.3711,
      "train_tokens_per_second": 1428.176
    },
    {
      "epoch": 0.22767816091954024,
      "grad_norm": 0.3308377265930176,
      "learning_rate": 0.0001548268238761975,
      "loss": 1.2279,
      "num_input_tokens_seen": 3040640,
      "step": 619,
      "train_runtime": 2129.0783,
      "train_tokens_per_second": 1428.148
    },
    {
      "epoch": 0.22804597701149426,
      "grad_norm": 0.34152933955192566,
      "learning_rate": 0.00015475313190862198,
      "loss": 1.259,
      "num_input_tokens_seen": 3048544,
      "step": 620,
      "train_runtime": 2133.8726,
      "train_tokens_per_second": 1428.644
    },
    {
      "epoch": 0.22841379310344828,
      "grad_norm": 0.33772867918014526,
      "learning_rate": 0.00015467943994104644,
      "loss": 1.1684,
      "num_input_tokens_seen": 3054352,
      "step": 621,
      "train_runtime": 2137.7081,
      "train_tokens_per_second": 1428.798
    },
    {
      "epoch": 0.2287816091954023,
      "grad_norm": 0.370492547750473,
      "learning_rate": 0.00015460574797347088,
      "loss": 1.4057,
      "num_input_tokens_seen": 3060176,
      "step": 622,
      "train_runtime": 2141.4916,
      "train_tokens_per_second": 1428.993
    },
    {
      "epoch": 0.22914942528735632,
      "grad_norm": 0.41318953037261963,
      "learning_rate": 0.00015453205600589537,
      "loss": 1.1883,
      "num_input_tokens_seen": 3064016,
      "step": 623,
      "train_runtime": 2144.2031,
      "train_tokens_per_second": 1428.977
    },
    {
      "epoch": 0.22951724137931034,
      "grad_norm": 0.3290882706642151,
      "learning_rate": 0.00015445836403831984,
      "loss": 1.2908,
      "num_input_tokens_seen": 3068288,
      "step": 624,
      "train_runtime": 2147.1118,
      "train_tokens_per_second": 1429.03
    },
    {
      "epoch": 0.22988505747126436,
      "grad_norm": 0.331515371799469,
      "learning_rate": 0.0001543846720707443,
      "loss": 1.3962,
      "num_input_tokens_seen": 3072016,
      "step": 625,
      "train_runtime": 2149.8027,
      "train_tokens_per_second": 1428.976
    },
    {
      "epoch": 0.23025287356321839,
      "grad_norm": 0.3490446209907532,
      "learning_rate": 0.00015431098010316877,
      "loss": 1.2267,
      "num_input_tokens_seen": 3076016,
      "step": 626,
      "train_runtime": 2152.5587,
      "train_tokens_per_second": 1429.004
    },
    {
      "epoch": 0.2306206896551724,
      "grad_norm": 0.34286853671073914,
      "learning_rate": 0.00015423728813559323,
      "loss": 1.3973,
      "num_input_tokens_seen": 3083056,
      "step": 627,
      "train_runtime": 2157.0813,
      "train_tokens_per_second": 1429.272
    },
    {
      "epoch": 0.23098850574712643,
      "grad_norm": 0.3316624164581299,
      "learning_rate": 0.0001541635961680177,
      "loss": 1.2641,
      "num_input_tokens_seen": 3087984,
      "step": 628,
      "train_runtime": 2160.3462,
      "train_tokens_per_second": 1429.393
    },
    {
      "epoch": 0.23135632183908045,
      "grad_norm": 0.32378822565078735,
      "learning_rate": 0.00015408990420044216,
      "loss": 1.1981,
      "num_input_tokens_seen": 3094496,
      "step": 629,
      "train_runtime": 2164.5291,
      "train_tokens_per_second": 1429.639
    },
    {
      "epoch": 0.2317241379310345,
      "grad_norm": 0.3374817669391632,
      "learning_rate": 0.00015401621223286663,
      "loss": 1.2519,
      "num_input_tokens_seen": 3097744,
      "step": 630,
      "train_runtime": 2166.9657,
      "train_tokens_per_second": 1429.531
    },
    {
      "epoch": 0.23209195402298852,
      "grad_norm": 0.3515070378780365,
      "learning_rate": 0.0001539425202652911,
      "loss": 1.4411,
      "num_input_tokens_seen": 3101824,
      "step": 631,
      "train_runtime": 2171.7969,
      "train_tokens_per_second": 1428.229
    },
    {
      "epoch": 0.23245977011494254,
      "grad_norm": 0.3360339403152466,
      "learning_rate": 0.00015386882829771556,
      "loss": 1.229,
      "num_input_tokens_seen": 3106960,
      "step": 632,
      "train_runtime": 2175.1845,
      "train_tokens_per_second": 1428.366
    },
    {
      "epoch": 0.23282758620689656,
      "grad_norm": 0.3126661479473114,
      "learning_rate": 0.00015379513633014002,
      "loss": 1.2524,
      "num_input_tokens_seen": 3111024,
      "step": 633,
      "train_runtime": 2177.9921,
      "train_tokens_per_second": 1428.391
    },
    {
      "epoch": 0.23319540229885058,
      "grad_norm": 0.32186412811279297,
      "learning_rate": 0.00015372144436256449,
      "loss": 1.2399,
      "num_input_tokens_seen": 3115136,
      "step": 634,
      "train_runtime": 2180.8308,
      "train_tokens_per_second": 1428.417
    },
    {
      "epoch": 0.2335632183908046,
      "grad_norm": 0.3627889156341553,
      "learning_rate": 0.00015364775239498895,
      "loss": 1.1929,
      "num_input_tokens_seen": 3120464,
      "step": 635,
      "train_runtime": 2184.3213,
      "train_tokens_per_second": 1428.574
    },
    {
      "epoch": 0.23393103448275862,
      "grad_norm": 0.3592092990875244,
      "learning_rate": 0.00015357406042741342,
      "loss": 1.4087,
      "num_input_tokens_seen": 3126080,
      "step": 636,
      "train_runtime": 2187.9955,
      "train_tokens_per_second": 1428.742
    },
    {
      "epoch": 0.23429885057471264,
      "grad_norm": 0.3188234865665436,
      "learning_rate": 0.00015350036845983788,
      "loss": 1.248,
      "num_input_tokens_seen": 3131056,
      "step": 637,
      "train_runtime": 2191.3305,
      "train_tokens_per_second": 1428.838
    },
    {
      "epoch": 0.23466666666666666,
      "grad_norm": 0.3150631785392761,
      "learning_rate": 0.00015342667649226235,
      "loss": 1.3449,
      "num_input_tokens_seen": 3138384,
      "step": 638,
      "train_runtime": 2196.0512,
      "train_tokens_per_second": 1429.103
    },
    {
      "epoch": 0.23503448275862068,
      "grad_norm": 0.36316341161727905,
      "learning_rate": 0.00015335298452468684,
      "loss": 1.3747,
      "num_input_tokens_seen": 3142352,
      "step": 639,
      "train_runtime": 2198.8694,
      "train_tokens_per_second": 1429.076
    },
    {
      "epoch": 0.2354022988505747,
      "grad_norm": 0.3514518141746521,
      "learning_rate": 0.00015327929255711128,
      "loss": 1.2717,
      "num_input_tokens_seen": 3146864,
      "step": 640,
      "train_runtime": 2201.9838,
      "train_tokens_per_second": 1429.104
    },
    {
      "epoch": 0.23577011494252872,
      "grad_norm": 0.34906163811683655,
      "learning_rate": 0.00015320560058953574,
      "loss": 1.1923,
      "num_input_tokens_seen": 3152784,
      "step": 641,
      "train_runtime": 2205.8944,
      "train_tokens_per_second": 1429.254
    },
    {
      "epoch": 0.23613793103448275,
      "grad_norm": 0.3031584620475769,
      "learning_rate": 0.0001531319086219602,
      "loss": 1.2025,
      "num_input_tokens_seen": 3160096,
      "step": 642,
      "train_runtime": 2210.4306,
      "train_tokens_per_second": 1429.629
    },
    {
      "epoch": 0.2365057471264368,
      "grad_norm": 0.3557109534740448,
      "learning_rate": 0.0001530582166543847,
      "loss": 1.2924,
      "num_input_tokens_seen": 3166704,
      "step": 643,
      "train_runtime": 2214.5628,
      "train_tokens_per_second": 1429.945
    },
    {
      "epoch": 0.23687356321839081,
      "grad_norm": 0.33845528960227966,
      "learning_rate": 0.00015298452468680916,
      "loss": 1.2173,
      "num_input_tokens_seen": 3172432,
      "step": 644,
      "train_runtime": 2218.4093,
      "train_tokens_per_second": 1430.048
    },
    {
      "epoch": 0.23724137931034484,
      "grad_norm": 0.3227993845939636,
      "learning_rate": 0.0001529108327192336,
      "loss": 1.1944,
      "num_input_tokens_seen": 3176624,
      "step": 645,
      "train_runtime": 2221.3908,
      "train_tokens_per_second": 1430.016
    },
    {
      "epoch": 0.23760919540229886,
      "grad_norm": 0.33973607420921326,
      "learning_rate": 0.00015283714075165806,
      "loss": 1.2882,
      "num_input_tokens_seen": 3181744,
      "step": 646,
      "train_runtime": 2224.8343,
      "train_tokens_per_second": 1430.104
    },
    {
      "epoch": 0.23797701149425288,
      "grad_norm": 0.35896897315979004,
      "learning_rate": 0.00015276344878408256,
      "loss": 1.2382,
      "num_input_tokens_seen": 3185936,
      "step": 647,
      "train_runtime": 2227.8318,
      "train_tokens_per_second": 1430.061
    },
    {
      "epoch": 0.2383448275862069,
      "grad_norm": 0.37523940205574036,
      "learning_rate": 0.00015268975681650702,
      "loss": 1.2286,
      "num_input_tokens_seen": 3189488,
      "step": 648,
      "train_runtime": 2230.3394,
      "train_tokens_per_second": 1430.046
    },
    {
      "epoch": 0.23871264367816092,
      "grad_norm": 0.3502814471721649,
      "learning_rate": 0.00015261606484893149,
      "loss": 1.2112,
      "num_input_tokens_seen": 3194720,
      "step": 649,
      "train_runtime": 2233.9256,
      "train_tokens_per_second": 1430.092
    },
    {
      "epoch": 0.23908045977011494,
      "grad_norm": 0.35990142822265625,
      "learning_rate": 0.00015254237288135592,
      "loss": 1.3038,
      "num_input_tokens_seen": 3202496,
      "step": 650,
      "train_runtime": 2238.8049,
      "train_tokens_per_second": 1430.449
    },
    {
      "epoch": 0.23944827586206896,
      "grad_norm": 0.3870561718940735,
      "learning_rate": 0.00015246868091378042,
      "loss": 1.3588,
      "num_input_tokens_seen": 3206320,
      "step": 651,
      "train_runtime": 2241.5207,
      "train_tokens_per_second": 1430.422
    },
    {
      "epoch": 0.23981609195402298,
      "grad_norm": 0.36193016171455383,
      "learning_rate": 0.00015239498894620488,
      "loss": 1.2423,
      "num_input_tokens_seen": 3214016,
      "step": 652,
      "train_runtime": 2246.3831,
      "train_tokens_per_second": 1430.751
    },
    {
      "epoch": 0.240183908045977,
      "grad_norm": 0.4229816794395447,
      "learning_rate": 0.00015232129697862935,
      "loss": 1.2857,
      "num_input_tokens_seen": 3218112,
      "step": 653,
      "train_runtime": 2249.2306,
      "train_tokens_per_second": 1430.761
    },
    {
      "epoch": 0.24055172413793102,
      "grad_norm": 0.4027053415775299,
      "learning_rate": 0.00015224760501105378,
      "loss": 1.2014,
      "num_input_tokens_seen": 3222432,
      "step": 654,
      "train_runtime": 2252.2611,
      "train_tokens_per_second": 1430.754
    },
    {
      "epoch": 0.24091954022988507,
      "grad_norm": 0.3395320177078247,
      "learning_rate": 0.00015217391304347827,
      "loss": 1.2439,
      "num_input_tokens_seen": 3229632,
      "step": 655,
      "train_runtime": 2256.8678,
      "train_tokens_per_second": 1431.024
    },
    {
      "epoch": 0.2412873563218391,
      "grad_norm": 0.3750961124897003,
      "learning_rate": 0.00015210022107590274,
      "loss": 1.2973,
      "num_input_tokens_seen": 3233120,
      "step": 656,
      "train_runtime": 2259.4178,
      "train_tokens_per_second": 1430.953
    },
    {
      "epoch": 0.2416551724137931,
      "grad_norm": 0.3356698751449585,
      "learning_rate": 0.0001520265291083272,
      "loss": 1.2474,
      "num_input_tokens_seen": 3236640,
      "step": 657,
      "train_runtime": 2262.0109,
      "train_tokens_per_second": 1430.868
    },
    {
      "epoch": 0.24202298850574713,
      "grad_norm": 0.31609296798706055,
      "learning_rate": 0.00015195283714075167,
      "loss": 1.1402,
      "num_input_tokens_seen": 3242816,
      "step": 658,
      "train_runtime": 2265.9665,
      "train_tokens_per_second": 1431.096
    },
    {
      "epoch": 0.24239080459770115,
      "grad_norm": 0.3269003629684448,
      "learning_rate": 0.00015187914517317613,
      "loss": 1.3452,
      "num_input_tokens_seen": 3251440,
      "step": 659,
      "train_runtime": 2271.333,
      "train_tokens_per_second": 1431.512
    },
    {
      "epoch": 0.24275862068965517,
      "grad_norm": 0.3321452736854553,
      "learning_rate": 0.0001518054532056006,
      "loss": 1.2074,
      "num_input_tokens_seen": 3255072,
      "step": 660,
      "train_runtime": 2273.952,
      "train_tokens_per_second": 1431.46
    },
    {
      "epoch": 0.2431264367816092,
      "grad_norm": 0.359015554189682,
      "learning_rate": 0.00015173176123802506,
      "loss": 1.3853,
      "num_input_tokens_seen": 3258992,
      "step": 661,
      "train_runtime": 2278.7706,
      "train_tokens_per_second": 1430.154
    },
    {
      "epoch": 0.24349425287356322,
      "grad_norm": 0.33611130714416504,
      "learning_rate": 0.00015165806927044953,
      "loss": 1.136,
      "num_input_tokens_seen": 3263104,
      "step": 662,
      "train_runtime": 2281.5981,
      "train_tokens_per_second": 1430.183
    },
    {
      "epoch": 0.24386206896551724,
      "grad_norm": 0.3454429805278778,
      "learning_rate": 0.000151584377302874,
      "loss": 1.2523,
      "num_input_tokens_seen": 3267648,
      "step": 663,
      "train_runtime": 2284.6702,
      "train_tokens_per_second": 1430.249
    },
    {
      "epoch": 0.24422988505747126,
      "grad_norm": 0.3487210273742676,
      "learning_rate": 0.00015151068533529846,
      "loss": 1.1537,
      "num_input_tokens_seen": 3271728,
      "step": 664,
      "train_runtime": 2287.547,
      "train_tokens_per_second": 1430.234
    },
    {
      "epoch": 0.24459770114942528,
      "grad_norm": 0.3361564576625824,
      "learning_rate": 0.00015143699336772292,
      "loss": 1.3944,
      "num_input_tokens_seen": 3275936,
      "step": 665,
      "train_runtime": 2290.4519,
      "train_tokens_per_second": 1430.257
    },
    {
      "epoch": 0.2449655172413793,
      "grad_norm": 0.34907296299934387,
      "learning_rate": 0.0001513633014001474,
      "loss": 1.2817,
      "num_input_tokens_seen": 3280256,
      "step": 666,
      "train_runtime": 2293.4253,
      "train_tokens_per_second": 1430.287
    },
    {
      "epoch": 0.24533333333333332,
      "grad_norm": 0.34399649500846863,
      "learning_rate": 0.00015128960943257185,
      "loss": 1.1997,
      "num_input_tokens_seen": 3284928,
      "step": 667,
      "train_runtime": 2296.6832,
      "train_tokens_per_second": 1430.292
    },
    {
      "epoch": 0.24570114942528737,
      "grad_norm": 0.328960120677948,
      "learning_rate": 0.00015121591746499632,
      "loss": 1.3485,
      "num_input_tokens_seen": 3289328,
      "step": 668,
      "train_runtime": 2299.6873,
      "train_tokens_per_second": 1430.337
    },
    {
      "epoch": 0.2460689655172414,
      "grad_norm": 0.3279968500137329,
      "learning_rate": 0.00015114222549742078,
      "loss": 1.2129,
      "num_input_tokens_seen": 3294288,
      "step": 669,
      "train_runtime": 2302.9694,
      "train_tokens_per_second": 1430.452
    },
    {
      "epoch": 0.2464367816091954,
      "grad_norm": 0.3576563000679016,
      "learning_rate": 0.00015106853352984525,
      "loss": 1.1662,
      "num_input_tokens_seen": 3302624,
      "step": 670,
      "train_runtime": 2308.1902,
      "train_tokens_per_second": 1430.828
    },
    {
      "epoch": 0.24680459770114943,
      "grad_norm": 0.33015739917755127,
      "learning_rate": 0.00015099484156226974,
      "loss": 1.4062,
      "num_input_tokens_seen": 3306256,
      "step": 671,
      "train_runtime": 2310.8257,
      "train_tokens_per_second": 1430.768
    },
    {
      "epoch": 0.24717241379310345,
      "grad_norm": 0.3386419415473938,
      "learning_rate": 0.00015092114959469418,
      "loss": 1.1945,
      "num_input_tokens_seen": 3310096,
      "step": 672,
      "train_runtime": 2313.522,
      "train_tokens_per_second": 1430.761
    },
    {
      "epoch": 0.24754022988505747,
      "grad_norm": 0.3697667717933655,
      "learning_rate": 0.00015084745762711864,
      "loss": 1.1111,
      "num_input_tokens_seen": 3314272,
      "step": 673,
      "train_runtime": 2316.4016,
      "train_tokens_per_second": 1430.785
    },
    {
      "epoch": 0.2479080459770115,
      "grad_norm": 0.38225889205932617,
      "learning_rate": 0.0001507737656595431,
      "loss": 1.1731,
      "num_input_tokens_seen": 3317856,
      "step": 674,
      "train_runtime": 2318.9524,
      "train_tokens_per_second": 1430.756
    },
    {
      "epoch": 0.2482758620689655,
      "grad_norm": 0.4129439890384674,
      "learning_rate": 0.0001507000736919676,
      "loss": 1.2218,
      "num_input_tokens_seen": 3321136,
      "step": 675,
      "train_runtime": 2321.362,
      "train_tokens_per_second": 1430.684
    },
    {
      "epoch": 0.24864367816091953,
      "grad_norm": 0.33089080452919006,
      "learning_rate": 0.00015062638172439206,
      "loss": 1.332,
      "num_input_tokens_seen": 3325712,
      "step": 676,
      "train_runtime": 2324.4806,
      "train_tokens_per_second": 1430.733
    },
    {
      "epoch": 0.24901149425287356,
      "grad_norm": 0.31337490677833557,
      "learning_rate": 0.0001505526897568165,
      "loss": 1.1848,
      "num_input_tokens_seen": 3330656,
      "step": 677,
      "train_runtime": 2327.7262,
      "train_tokens_per_second": 1430.862
    },
    {
      "epoch": 0.24937931034482758,
      "grad_norm": 0.3710203468799591,
      "learning_rate": 0.00015047899778924097,
      "loss": 1.3389,
      "num_input_tokens_seen": 3335504,
      "step": 678,
      "train_runtime": 2330.9035,
      "train_tokens_per_second": 1430.992
    },
    {
      "epoch": 0.2497471264367816,
      "grad_norm": 0.3893492519855499,
      "learning_rate": 0.00015040530582166546,
      "loss": 1.2191,
      "num_input_tokens_seen": 3340224,
      "step": 679,
      "train_runtime": 2334.0724,
      "train_tokens_per_second": 1431.071
    },
    {
      "epoch": 0.2501149425287356,
      "grad_norm": 0.3619728684425354,
      "learning_rate": 0.00015033161385408992,
      "loss": 1.2595,
      "num_input_tokens_seen": 3344176,
      "step": 680,
      "train_runtime": 2336.8979,
      "train_tokens_per_second": 1431.032
    },
    {
      "epoch": 0.25048275862068964,
      "grad_norm": 0.3799172043800354,
      "learning_rate": 0.0001502579218865144,
      "loss": 1.2605,
      "num_input_tokens_seen": 3347440,
      "step": 681,
      "train_runtime": 2339.2882,
      "train_tokens_per_second": 1430.965
    },
    {
      "epoch": 0.25085057471264366,
      "grad_norm": 0.3770306408405304,
      "learning_rate": 0.00015018422991893882,
      "loss": 1.2391,
      "num_input_tokens_seen": 3351424,
      "step": 682,
      "train_runtime": 2342.0126,
      "train_tokens_per_second": 1431.002
    },
    {
      "epoch": 0.2512183908045977,
      "grad_norm": 0.33930251002311707,
      "learning_rate": 0.00015011053795136332,
      "loss": 1.3171,
      "num_input_tokens_seen": 3355168,
      "step": 683,
      "train_runtime": 2344.7316,
      "train_tokens_per_second": 1430.939
    },
    {
      "epoch": 0.2515862068965517,
      "grad_norm": 0.3217930197715759,
      "learning_rate": 0.00015003684598378778,
      "loss": 1.2046,
      "num_input_tokens_seen": 3361392,
      "step": 684,
      "train_runtime": 2348.6808,
      "train_tokens_per_second": 1431.183
    },
    {
      "epoch": 0.2519540229885057,
      "grad_norm": 0.3601853847503662,
      "learning_rate": 0.00014996315401621225,
      "loss": 1.2486,
      "num_input_tokens_seen": 3365360,
      "step": 685,
      "train_runtime": 2351.4679,
      "train_tokens_per_second": 1431.174
    },
    {
      "epoch": 0.2523218390804598,
      "grad_norm": 0.33223655819892883,
      "learning_rate": 0.00014988946204863668,
      "loss": 1.3917,
      "num_input_tokens_seen": 3370176,
      "step": 686,
      "train_runtime": 2354.6809,
      "train_tokens_per_second": 1431.267
    },
    {
      "epoch": 0.2526896551724138,
      "grad_norm": 0.3874072730541229,
      "learning_rate": 0.00014981577008106118,
      "loss": 1.3981,
      "num_input_tokens_seen": 3375168,
      "step": 687,
      "train_runtime": 2358.0085,
      "train_tokens_per_second": 1431.364
    },
    {
      "epoch": 0.25305747126436784,
      "grad_norm": 0.3030243217945099,
      "learning_rate": 0.00014974207811348564,
      "loss": 1.2081,
      "num_input_tokens_seen": 3382064,
      "step": 688,
      "train_runtime": 2362.475,
      "train_tokens_per_second": 1431.577
    },
    {
      "epoch": 0.25342528735632186,
      "grad_norm": 0.309259295463562,
      "learning_rate": 0.0001496683861459101,
      "loss": 1.2918,
      "num_input_tokens_seen": 3389872,
      "step": 689,
      "train_runtime": 2367.395,
      "train_tokens_per_second": 1431.9
    },
    {
      "epoch": 0.2537931034482759,
      "grad_norm": 0.3466988205909729,
      "learning_rate": 0.00014959469417833457,
      "loss": 1.171,
      "num_input_tokens_seen": 3393456,
      "step": 690,
      "train_runtime": 2369.9487,
      "train_tokens_per_second": 1431.869
    },
    {
      "epoch": 0.2541609195402299,
      "grad_norm": 0.3572579026222229,
      "learning_rate": 0.00014952100221075904,
      "loss": 1.3573,
      "num_input_tokens_seen": 3399648,
      "step": 691,
      "train_runtime": 2376.0364,
      "train_tokens_per_second": 1430.806
    },
    {
      "epoch": 0.2545287356321839,
      "grad_norm": 0.35369816422462463,
      "learning_rate": 0.0001494473102431835,
      "loss": 1.1169,
      "num_input_tokens_seen": 3403392,
      "step": 692,
      "train_runtime": 2378.7839,
      "train_tokens_per_second": 1430.728
    },
    {
      "epoch": 0.25489655172413794,
      "grad_norm": 0.339080810546875,
      "learning_rate": 0.00014937361827560796,
      "loss": 1.1781,
      "num_input_tokens_seen": 3411264,
      "step": 693,
      "train_runtime": 2383.7404,
      "train_tokens_per_second": 1431.055
    },
    {
      "epoch": 0.25526436781609196,
      "grad_norm": 0.3506729006767273,
      "learning_rate": 0.00014929992630803243,
      "loss": 1.1749,
      "num_input_tokens_seen": 3415440,
      "step": 694,
      "train_runtime": 2386.6959,
      "train_tokens_per_second": 1431.033
    },
    {
      "epoch": 0.255632183908046,
      "grad_norm": 0.3485318720340729,
      "learning_rate": 0.0001492262343404569,
      "loss": 1.2567,
      "num_input_tokens_seen": 3419168,
      "step": 695,
      "train_runtime": 2389.363,
      "train_tokens_per_second": 1430.996
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.38310953974723816,
      "learning_rate": 0.00014915254237288136,
      "loss": 1.2558,
      "num_input_tokens_seen": 3423680,
      "step": 696,
      "train_runtime": 2392.46,
      "train_tokens_per_second": 1431.029
    },
    {
      "epoch": 0.256367816091954,
      "grad_norm": 0.34740230441093445,
      "learning_rate": 0.00014907885040530582,
      "loss": 1.2216,
      "num_input_tokens_seen": 3427920,
      "step": 697,
      "train_runtime": 2395.4158,
      "train_tokens_per_second": 1431.033
    },
    {
      "epoch": 0.25673563218390805,
      "grad_norm": 0.31901097297668457,
      "learning_rate": 0.0001490051584377303,
      "loss": 1.206,
      "num_input_tokens_seen": 3432432,
      "step": 698,
      "train_runtime": 2398.4438,
      "train_tokens_per_second": 1431.108
    },
    {
      "epoch": 0.25710344827586207,
      "grad_norm": 0.3171537220478058,
      "learning_rate": 0.00014893146647015478,
      "loss": 1.1877,
      "num_input_tokens_seen": 3437312,
      "step": 699,
      "train_runtime": 2401.7849,
      "train_tokens_per_second": 1431.149
    },
    {
      "epoch": 0.2574712643678161,
      "grad_norm": 0.38633808493614197,
      "learning_rate": 0.00014885777450257922,
      "loss": 1.3395,
      "num_input_tokens_seen": 3440704,
      "step": 700,
      "train_runtime": 2404.2488,
      "train_tokens_per_second": 1431.093
    },
    {
      "epoch": 0.2578390804597701,
      "grad_norm": 0.36124300956726074,
      "learning_rate": 0.00014878408253500368,
      "loss": 1.376,
      "num_input_tokens_seen": 3446160,
      "step": 701,
      "train_runtime": 2407.9137,
      "train_tokens_per_second": 1431.181
    },
    {
      "epoch": 0.25820689655172413,
      "grad_norm": 0.3738381862640381,
      "learning_rate": 0.00014871039056742815,
      "loss": 1.2993,
      "num_input_tokens_seen": 3450000,
      "step": 702,
      "train_runtime": 2410.6088,
      "train_tokens_per_second": 1431.174
    },
    {
      "epoch": 0.25857471264367815,
      "grad_norm": 0.41211527585983276,
      "learning_rate": 0.00014863669859985264,
      "loss": 1.3984,
      "num_input_tokens_seen": 3454160,
      "step": 703,
      "train_runtime": 2413.5428,
      "train_tokens_per_second": 1431.158
    },
    {
      "epoch": 0.25894252873563217,
      "grad_norm": 0.3279429078102112,
      "learning_rate": 0.00014856300663227708,
      "loss": 1.2641,
      "num_input_tokens_seen": 3457440,
      "step": 704,
      "train_runtime": 2415.9856,
      "train_tokens_per_second": 1431.068
    },
    {
      "epoch": 0.2593103448275862,
      "grad_norm": 0.305835098028183,
      "learning_rate": 0.00014848931466470154,
      "loss": 0.9934,
      "num_input_tokens_seen": 3461680,
      "step": 705,
      "train_runtime": 2418.9059,
      "train_tokens_per_second": 1431.093
    },
    {
      "epoch": 0.2596781609195402,
      "grad_norm": 0.25167855620384216,
      "learning_rate": 0.000148415622697126,
      "loss": 1.0183,
      "num_input_tokens_seen": 3475792,
      "step": 706,
      "train_runtime": 2427.4112,
      "train_tokens_per_second": 1431.893
    },
    {
      "epoch": 0.26004597701149423,
      "grad_norm": 0.34948840737342834,
      "learning_rate": 0.0001483419307295505,
      "loss": 1.1742,
      "num_input_tokens_seen": 3480096,
      "step": 707,
      "train_runtime": 2430.4296,
      "train_tokens_per_second": 1431.885
    },
    {
      "epoch": 0.26041379310344825,
      "grad_norm": 0.33695265650749207,
      "learning_rate": 0.00014826823876197496,
      "loss": 1.3288,
      "num_input_tokens_seen": 3483664,
      "step": 708,
      "train_runtime": 2432.9934,
      "train_tokens_per_second": 1431.843
    },
    {
      "epoch": 0.2607816091954023,
      "grad_norm": 0.3297808766365051,
      "learning_rate": 0.0001481945467943994,
      "loss": 1.1475,
      "num_input_tokens_seen": 3489824,
      "step": 709,
      "train_runtime": 2437.0575,
      "train_tokens_per_second": 1431.983
    },
    {
      "epoch": 0.2611494252873563,
      "grad_norm": 0.340537428855896,
      "learning_rate": 0.00014812085482682387,
      "loss": 1.3406,
      "num_input_tokens_seen": 3493296,
      "step": 710,
      "train_runtime": 2439.5964,
      "train_tokens_per_second": 1431.916
    },
    {
      "epoch": 0.2615172413793104,
      "grad_norm": 0.3442951738834381,
      "learning_rate": 0.00014804716285924836,
      "loss": 1.3046,
      "num_input_tokens_seen": 3499600,
      "step": 711,
      "train_runtime": 2443.7034,
      "train_tokens_per_second": 1432.089
    },
    {
      "epoch": 0.2618850574712644,
      "grad_norm": 0.33428308367729187,
      "learning_rate": 0.00014797347089167282,
      "loss": 1.1656,
      "num_input_tokens_seen": 3506608,
      "step": 712,
      "train_runtime": 2448.2463,
      "train_tokens_per_second": 1432.294
    },
    {
      "epoch": 0.2622528735632184,
      "grad_norm": 0.3170371353626251,
      "learning_rate": 0.0001478997789240973,
      "loss": 1.2762,
      "num_input_tokens_seen": 3511072,
      "step": 713,
      "train_runtime": 2451.3211,
      "train_tokens_per_second": 1432.318
    },
    {
      "epoch": 0.26262068965517243,
      "grad_norm": 0.35202986001968384,
      "learning_rate": 0.00014782608695652173,
      "loss": 1.1412,
      "num_input_tokens_seen": 3514560,
      "step": 714,
      "train_runtime": 2453.8613,
      "train_tokens_per_second": 1432.257
    },
    {
      "epoch": 0.26298850574712646,
      "grad_norm": 0.3663095235824585,
      "learning_rate": 0.00014775239498894622,
      "loss": 1.308,
      "num_input_tokens_seen": 3519328,
      "step": 715,
      "train_runtime": 2457.0199,
      "train_tokens_per_second": 1432.356
    },
    {
      "epoch": 0.2633563218390805,
      "grad_norm": 0.3512764871120453,
      "learning_rate": 0.00014767870302137068,
      "loss": 1.0741,
      "num_input_tokens_seen": 3522624,
      "step": 716,
      "train_runtime": 2459.5229,
      "train_tokens_per_second": 1432.239
    },
    {
      "epoch": 0.2637241379310345,
      "grad_norm": 0.323486864566803,
      "learning_rate": 0.00014760501105379515,
      "loss": 1.2378,
      "num_input_tokens_seen": 3528512,
      "step": 717,
      "train_runtime": 2463.3762,
      "train_tokens_per_second": 1432.389
    },
    {
      "epoch": 0.2640919540229885,
      "grad_norm": 0.34799066185951233,
      "learning_rate": 0.0001475313190862196,
      "loss": 1.405,
      "num_input_tokens_seen": 3534688,
      "step": 718,
      "train_runtime": 2467.4415,
      "train_tokens_per_second": 1432.532
    },
    {
      "epoch": 0.26445977011494254,
      "grad_norm": 0.41243162751197815,
      "learning_rate": 0.00014745762711864408,
      "loss": 1.3616,
      "num_input_tokens_seen": 3539584,
      "step": 719,
      "train_runtime": 2470.7963,
      "train_tokens_per_second": 1432.568
    },
    {
      "epoch": 0.26482758620689656,
      "grad_norm": 0.32138335704803467,
      "learning_rate": 0.00014738393515106854,
      "loss": 1.1256,
      "num_input_tokens_seen": 3547360,
      "step": 720,
      "train_runtime": 2475.6586,
      "train_tokens_per_second": 1432.895
    },
    {
      "epoch": 0.2651954022988506,
      "grad_norm": 0.3641934096813202,
      "learning_rate": 0.000147310243183493,
      "loss": 1.1522,
      "num_input_tokens_seen": 3551760,
      "step": 721,
      "train_runtime": 2480.4266,
      "train_tokens_per_second": 1431.915
    },
    {
      "epoch": 0.2655632183908046,
      "grad_norm": 0.35957276821136475,
      "learning_rate": 0.00014723655121591747,
      "loss": 1.3313,
      "num_input_tokens_seen": 3555520,
      "step": 722,
      "train_runtime": 2483.1092,
      "train_tokens_per_second": 1431.882
    },
    {
      "epoch": 0.2659310344827586,
      "grad_norm": 0.40129801630973816,
      "learning_rate": 0.00014716285924834194,
      "loss": 1.2624,
      "num_input_tokens_seen": 3559632,
      "step": 723,
      "train_runtime": 2485.9943,
      "train_tokens_per_second": 1431.875
    },
    {
      "epoch": 0.26629885057471264,
      "grad_norm": 0.3500661849975586,
      "learning_rate": 0.0001470891672807664,
      "loss": 1.3401,
      "num_input_tokens_seen": 3563920,
      "step": 724,
      "train_runtime": 2488.9607,
      "train_tokens_per_second": 1431.891
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 0.35723426938056946,
      "learning_rate": 0.00014701547531319087,
      "loss": 1.2805,
      "num_input_tokens_seen": 3567712,
      "step": 725,
      "train_runtime": 2491.6605,
      "train_tokens_per_second": 1431.861
    },
    {
      "epoch": 0.2670344827586207,
      "grad_norm": 0.39261963963508606,
      "learning_rate": 0.00014694178334561533,
      "loss": 1.3561,
      "num_input_tokens_seen": 3572800,
      "step": 726,
      "train_runtime": 2495.1222,
      "train_tokens_per_second": 1431.914
    },
    {
      "epoch": 0.2674022988505747,
      "grad_norm": 0.3508819043636322,
      "learning_rate": 0.0001468680913780398,
      "loss": 1.2249,
      "num_input_tokens_seen": 3577008,
      "step": 727,
      "train_runtime": 2498.0399,
      "train_tokens_per_second": 1431.926
    },
    {
      "epoch": 0.2677701149425287,
      "grad_norm": 0.3440887928009033,
      "learning_rate": 0.00014679439941046426,
      "loss": 1.1955,
      "num_input_tokens_seen": 3582048,
      "step": 728,
      "train_runtime": 2501.4453,
      "train_tokens_per_second": 1431.991
    },
    {
      "epoch": 0.26813793103448275,
      "grad_norm": 0.35490262508392334,
      "learning_rate": 0.00014672070744288873,
      "loss": 1.2922,
      "num_input_tokens_seen": 3585728,
      "step": 729,
      "train_runtime": 2504.0363,
      "train_tokens_per_second": 1431.979
    },
    {
      "epoch": 0.26850574712643677,
      "grad_norm": 0.3532816469669342,
      "learning_rate": 0.0001466470154753132,
      "loss": 1.2326,
      "num_input_tokens_seen": 3590192,
      "step": 730,
      "train_runtime": 2507.1604,
      "train_tokens_per_second": 1431.975
    },
    {
      "epoch": 0.2688735632183908,
      "grad_norm": 0.3759276568889618,
      "learning_rate": 0.00014657332350773768,
      "loss": 1.3548,
      "num_input_tokens_seen": 3593472,
      "step": 731,
      "train_runtime": 2509.6458,
      "train_tokens_per_second": 1431.864
    },
    {
      "epoch": 0.2692413793103448,
      "grad_norm": 0.3194345533847809,
      "learning_rate": 0.00014649963154016212,
      "loss": 1.2122,
      "num_input_tokens_seen": 3597552,
      "step": 732,
      "train_runtime": 2512.4704,
      "train_tokens_per_second": 1431.878
    },
    {
      "epoch": 0.26960919540229883,
      "grad_norm": 0.36756789684295654,
      "learning_rate": 0.00014642593957258658,
      "loss": 1.3636,
      "num_input_tokens_seen": 3602544,
      "step": 733,
      "train_runtime": 2515.9041,
      "train_tokens_per_second": 1431.908
    },
    {
      "epoch": 0.26997701149425285,
      "grad_norm": 0.35430723428726196,
      "learning_rate": 0.00014635224760501105,
      "loss": 1.228,
      "num_input_tokens_seen": 3606368,
      "step": 734,
      "train_runtime": 2518.7316,
      "train_tokens_per_second": 1431.819
    },
    {
      "epoch": 0.27034482758620687,
      "grad_norm": 0.3631165325641632,
      "learning_rate": 0.00014627855563743554,
      "loss": 1.3099,
      "num_input_tokens_seen": 3610240,
      "step": 735,
      "train_runtime": 2521.4438,
      "train_tokens_per_second": 1431.815
    },
    {
      "epoch": 0.27071264367816095,
      "grad_norm": 0.3637069761753082,
      "learning_rate": 0.00014620486366985998,
      "loss": 1.1917,
      "num_input_tokens_seen": 3613984,
      "step": 736,
      "train_runtime": 2524.0764,
      "train_tokens_per_second": 1431.804
    },
    {
      "epoch": 0.27108045977011497,
      "grad_norm": 0.3546808063983917,
      "learning_rate": 0.00014613117170228444,
      "loss": 1.2892,
      "num_input_tokens_seen": 3618560,
      "step": 737,
      "train_runtime": 2527.1452,
      "train_tokens_per_second": 1431.877
    },
    {
      "epoch": 0.271448275862069,
      "grad_norm": 0.3709617853164673,
      "learning_rate": 0.0001460574797347089,
      "loss": 1.1953,
      "num_input_tokens_seen": 3623408,
      "step": 738,
      "train_runtime": 2530.4085,
      "train_tokens_per_second": 1431.946
    },
    {
      "epoch": 0.271816091954023,
      "grad_norm": 0.35431450605392456,
      "learning_rate": 0.0001459837877671334,
      "loss": 1.3198,
      "num_input_tokens_seen": 3627232,
      "step": 739,
      "train_runtime": 2533.1074,
      "train_tokens_per_second": 1431.93
    },
    {
      "epoch": 0.27218390804597703,
      "grad_norm": 0.37465333938598633,
      "learning_rate": 0.00014591009579955787,
      "loss": 1.3503,
      "num_input_tokens_seen": 3633280,
      "step": 740,
      "train_runtime": 2537.0819,
      "train_tokens_per_second": 1432.07
    },
    {
      "epoch": 0.27255172413793105,
      "grad_norm": 0.43079355359077454,
      "learning_rate": 0.0001458364038319823,
      "loss": 1.1911,
      "num_input_tokens_seen": 3637856,
      "step": 741,
      "train_runtime": 2540.1823,
      "train_tokens_per_second": 1432.124
    },
    {
      "epoch": 0.27291954022988507,
      "grad_norm": 0.3299473226070404,
      "learning_rate": 0.00014576271186440677,
      "loss": 1.1145,
      "num_input_tokens_seen": 3641328,
      "step": 742,
      "train_runtime": 2542.7578,
      "train_tokens_per_second": 1432.039
    },
    {
      "epoch": 0.2732873563218391,
      "grad_norm": 0.3805535137653351,
      "learning_rate": 0.00014568901989683126,
      "loss": 1.0764,
      "num_input_tokens_seen": 3645056,
      "step": 743,
      "train_runtime": 2545.4204,
      "train_tokens_per_second": 1432.006
    },
    {
      "epoch": 0.2736551724137931,
      "grad_norm": 0.37547579407691956,
      "learning_rate": 0.00014561532792925573,
      "loss": 1.3328,
      "num_input_tokens_seen": 3648336,
      "step": 744,
      "train_runtime": 2547.8761,
      "train_tokens_per_second": 1431.913
    },
    {
      "epoch": 0.27402298850574713,
      "grad_norm": 0.3682103455066681,
      "learning_rate": 0.0001455416359616802,
      "loss": 1.1308,
      "num_input_tokens_seen": 3652496,
      "step": 745,
      "train_runtime": 2550.7942,
      "train_tokens_per_second": 1431.905
    },
    {
      "epoch": 0.27439080459770115,
      "grad_norm": 0.3358018398284912,
      "learning_rate": 0.00014546794399410463,
      "loss": 1.2951,
      "num_input_tokens_seen": 3658128,
      "step": 746,
      "train_runtime": 2554.5524,
      "train_tokens_per_second": 1432.003
    },
    {
      "epoch": 0.2747586206896552,
      "grad_norm": 0.3322310745716095,
      "learning_rate": 0.00014539425202652912,
      "loss": 1.3045,
      "num_input_tokens_seen": 3664848,
      "step": 747,
      "train_runtime": 2558.8606,
      "train_tokens_per_second": 1432.219
    },
    {
      "epoch": 0.2751264367816092,
      "grad_norm": 0.4266640245914459,
      "learning_rate": 0.00014532056005895358,
      "loss": 1.3532,
      "num_input_tokens_seen": 3671216,
      "step": 748,
      "train_runtime": 2562.9936,
      "train_tokens_per_second": 1432.394
    },
    {
      "epoch": 0.2754942528735632,
      "grad_norm": 0.3696272075176239,
      "learning_rate": 0.00014524686809137805,
      "loss": 1.3086,
      "num_input_tokens_seen": 3676736,
      "step": 749,
      "train_runtime": 2566.6764,
      "train_tokens_per_second": 1432.489
    },
    {
      "epoch": 0.27586206896551724,
      "grad_norm": 0.3480525016784668,
      "learning_rate": 0.00014517317612380251,
      "loss": 1.309,
      "num_input_tokens_seen": 3679520,
      "step": 750,
      "train_runtime": 2568.8727,
      "train_tokens_per_second": 1432.348
    },
    {
      "epoch": 0.27622988505747126,
      "grad_norm": 0.38454583287239075,
      "learning_rate": 0.00014509948415622698,
      "loss": 1.1951,
      "num_input_tokens_seen": 3686448,
      "step": 751,
      "train_runtime": 2574.8935,
      "train_tokens_per_second": 1431.689
    },
    {
      "epoch": 0.2765977011494253,
      "grad_norm": 0.3347504436969757,
      "learning_rate": 0.00014502579218865144,
      "loss": 1.1769,
      "num_input_tokens_seen": 3690368,
      "step": 752,
      "train_runtime": 2577.6272,
      "train_tokens_per_second": 1431.692
    },
    {
      "epoch": 0.2769655172413793,
      "grad_norm": 0.3606038987636566,
      "learning_rate": 0.0001449521002210759,
      "loss": 1.2629,
      "num_input_tokens_seen": 3694816,
      "step": 753,
      "train_runtime": 2580.6924,
      "train_tokens_per_second": 1431.715
    },
    {
      "epoch": 0.2773333333333333,
      "grad_norm": 0.34804779291152954,
      "learning_rate": 0.00014487840825350037,
      "loss": 1.27,
      "num_input_tokens_seen": 3698912,
      "step": 754,
      "train_runtime": 2583.5852,
      "train_tokens_per_second": 1431.697
    },
    {
      "epoch": 0.27770114942528734,
      "grad_norm": 0.3687753975391388,
      "learning_rate": 0.00014480471628592484,
      "loss": 1.0952,
      "num_input_tokens_seen": 3702352,
      "step": 755,
      "train_runtime": 2586.0673,
      "train_tokens_per_second": 1431.653
    },
    {
      "epoch": 0.27806896551724136,
      "grad_norm": 0.39494216442108154,
      "learning_rate": 0.0001447310243183493,
      "loss": 1.1941,
      "num_input_tokens_seen": 3707136,
      "step": 756,
      "train_runtime": 2589.3509,
      "train_tokens_per_second": 1431.685
    },
    {
      "epoch": 0.2784367816091954,
      "grad_norm": 0.3711467981338501,
      "learning_rate": 0.00014465733235077377,
      "loss": 1.2653,
      "num_input_tokens_seen": 3711856,
      "step": 757,
      "train_runtime": 2592.5878,
      "train_tokens_per_second": 1431.719
    },
    {
      "epoch": 0.2788045977011494,
      "grad_norm": 0.374125599861145,
      "learning_rate": 0.00014458364038319826,
      "loss": 1.1164,
      "num_input_tokens_seen": 3715200,
      "step": 758,
      "train_runtime": 2595.0367,
      "train_tokens_per_second": 1431.656
    },
    {
      "epoch": 0.2791724137931034,
      "grad_norm": 0.37214595079421997,
      "learning_rate": 0.0001445099484156227,
      "loss": 1.2924,
      "num_input_tokens_seen": 3719248,
      "step": 759,
      "train_runtime": 2597.8939,
      "train_tokens_per_second": 1431.64
    },
    {
      "epoch": 0.27954022988505745,
      "grad_norm": 0.3690154552459717,
      "learning_rate": 0.00014443625644804716,
      "loss": 1.1959,
      "num_input_tokens_seen": 3723152,
      "step": 760,
      "train_runtime": 2600.6256,
      "train_tokens_per_second": 1431.637
    },
    {
      "epoch": 0.2799080459770115,
      "grad_norm": 0.3575315773487091,
      "learning_rate": 0.00014436256448047163,
      "loss": 1.1708,
      "num_input_tokens_seen": 3726816,
      "step": 761,
      "train_runtime": 2603.2924,
      "train_tokens_per_second": 1431.578
    },
    {
      "epoch": 0.28027586206896554,
      "grad_norm": 0.4088453948497772,
      "learning_rate": 0.00014428887251289612,
      "loss": 1.2665,
      "num_input_tokens_seen": 3731472,
      "step": 762,
      "train_runtime": 2606.4992,
      "train_tokens_per_second": 1431.603
    },
    {
      "epoch": 0.28064367816091956,
      "grad_norm": 0.35154807567596436,
      "learning_rate": 0.00014421518054532058,
      "loss": 1.2909,
      "num_input_tokens_seen": 3735184,
      "step": 763,
      "train_runtime": 2609.1233,
      "train_tokens_per_second": 1431.586
    },
    {
      "epoch": 0.2810114942528736,
      "grad_norm": 0.36001941561698914,
      "learning_rate": 0.00014414148857774502,
      "loss": 1.1534,
      "num_input_tokens_seen": 3740592,
      "step": 764,
      "train_runtime": 2612.6182,
      "train_tokens_per_second": 1431.741
    },
    {
      "epoch": 0.2813793103448276,
      "grad_norm": 0.38879626989364624,
      "learning_rate": 0.00014406779661016949,
      "loss": 1.4079,
      "num_input_tokens_seen": 3744448,
      "step": 765,
      "train_runtime": 2615.3836,
      "train_tokens_per_second": 1431.701
    },
    {
      "epoch": 0.2817471264367816,
      "grad_norm": 0.38309088349342346,
      "learning_rate": 0.00014399410464259398,
      "loss": 1.3834,
      "num_input_tokens_seen": 3751168,
      "step": 766,
      "train_runtime": 2619.6309,
      "train_tokens_per_second": 1431.945
    },
    {
      "epoch": 0.28211494252873565,
      "grad_norm": 0.37338167428970337,
      "learning_rate": 0.00014392041267501844,
      "loss": 1.2634,
      "num_input_tokens_seen": 3755472,
      "step": 767,
      "train_runtime": 2622.5824,
      "train_tokens_per_second": 1431.975
    },
    {
      "epoch": 0.28248275862068967,
      "grad_norm": 0.35141757130622864,
      "learning_rate": 0.0001438467207074429,
      "loss": 1.187,
      "num_input_tokens_seen": 3760624,
      "step": 768,
      "train_runtime": 2626.0891,
      "train_tokens_per_second": 1432.025
    },
    {
      "epoch": 0.2828505747126437,
      "grad_norm": 0.3733404278755188,
      "learning_rate": 0.00014377302873986735,
      "loss": 1.2634,
      "num_input_tokens_seen": 3764880,
      "step": 769,
      "train_runtime": 2629.1064,
      "train_tokens_per_second": 1432.0
    },
    {
      "epoch": 0.2832183908045977,
      "grad_norm": 0.35038959980010986,
      "learning_rate": 0.00014369933677229184,
      "loss": 1.2856,
      "num_input_tokens_seen": 3768560,
      "step": 770,
      "train_runtime": 2631.7664,
      "train_tokens_per_second": 1431.951
    },
    {
      "epoch": 0.28358620689655173,
      "grad_norm": 0.4066001772880554,
      "learning_rate": 0.0001436256448047163,
      "loss": 1.2666,
      "num_input_tokens_seen": 3771728,
      "step": 771,
      "train_runtime": 2634.1337,
      "train_tokens_per_second": 1431.867
    },
    {
      "epoch": 0.28395402298850575,
      "grad_norm": 0.36203107237815857,
      "learning_rate": 0.00014355195283714077,
      "loss": 1.2935,
      "num_input_tokens_seen": 3779408,
      "step": 772,
      "train_runtime": 2638.9535,
      "train_tokens_per_second": 1432.162
    },
    {
      "epoch": 0.28432183908045977,
      "grad_norm": 0.3711211681365967,
      "learning_rate": 0.0001434782608695652,
      "loss": 1.2945,
      "num_input_tokens_seen": 3783328,
      "step": 773,
      "train_runtime": 2641.7507,
      "train_tokens_per_second": 1432.129
    },
    {
      "epoch": 0.2846896551724138,
      "grad_norm": 0.3628079295158386,
      "learning_rate": 0.0001434045689019897,
      "loss": 1.285,
      "num_input_tokens_seen": 3792112,
      "step": 774,
      "train_runtime": 2647.1875,
      "train_tokens_per_second": 1432.506
    },
    {
      "epoch": 0.2850574712643678,
      "grad_norm": 0.3806452453136444,
      "learning_rate": 0.00014333087693441416,
      "loss": 1.2512,
      "num_input_tokens_seen": 3795872,
      "step": 775,
      "train_runtime": 2649.8756,
      "train_tokens_per_second": 1432.472
    },
    {
      "epoch": 0.28542528735632183,
      "grad_norm": 0.37157365679740906,
      "learning_rate": 0.00014325718496683863,
      "loss": 1.3842,
      "num_input_tokens_seen": 3802144,
      "step": 776,
      "train_runtime": 2653.9266,
      "train_tokens_per_second": 1432.649
    },
    {
      "epoch": 0.28579310344827585,
      "grad_norm": 0.35492148995399475,
      "learning_rate": 0.0001431834929992631,
      "loss": 1.2058,
      "num_input_tokens_seen": 3804944,
      "step": 777,
      "train_runtime": 2656.098,
      "train_tokens_per_second": 1432.532
    },
    {
      "epoch": 0.2861609195402299,
      "grad_norm": 0.3418777585029602,
      "learning_rate": 0.00014310980103168756,
      "loss": 1.2545,
      "num_input_tokens_seen": 3810304,
      "step": 778,
      "train_runtime": 2659.6487,
      "train_tokens_per_second": 1432.634
    },
    {
      "epoch": 0.2865287356321839,
      "grad_norm": 0.34898510575294495,
      "learning_rate": 0.00014303610906411202,
      "loss": 1.2304,
      "num_input_tokens_seen": 3814720,
      "step": 779,
      "train_runtime": 2662.6579,
      "train_tokens_per_second": 1432.674
    },
    {
      "epoch": 0.2868965517241379,
      "grad_norm": 0.351762980222702,
      "learning_rate": 0.00014296241709653649,
      "loss": 1.1135,
      "num_input_tokens_seen": 3817904,
      "step": 780,
      "train_runtime": 2665.0414,
      "train_tokens_per_second": 1432.587
    },
    {
      "epoch": 0.28726436781609194,
      "grad_norm": 0.33062925934791565,
      "learning_rate": 0.00014288872512896095,
      "loss": 1.1541,
      "num_input_tokens_seen": 3822288,
      "step": 781,
      "train_runtime": 2670.1056,
      "train_tokens_per_second": 1431.512
    },
    {
      "epoch": 0.28763218390804596,
      "grad_norm": 0.37835007905960083,
      "learning_rate": 0.00014281503316138542,
      "loss": 1.3553,
      "num_input_tokens_seen": 3826272,
      "step": 782,
      "train_runtime": 2672.8532,
      "train_tokens_per_second": 1431.531
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.3742969334125519,
      "learning_rate": 0.00014274134119380988,
      "loss": 1.1948,
      "num_input_tokens_seen": 3830544,
      "step": 783,
      "train_runtime": 2675.7975,
      "train_tokens_per_second": 1431.552
    },
    {
      "epoch": 0.288367816091954,
      "grad_norm": 0.3816350996494293,
      "learning_rate": 0.00014266764922623434,
      "loss": 1.373,
      "num_input_tokens_seen": 3834896,
      "step": 784,
      "train_runtime": 2678.7294,
      "train_tokens_per_second": 1431.61
    },
    {
      "epoch": 0.288735632183908,
      "grad_norm": 0.31429529190063477,
      "learning_rate": 0.0001425939572586588,
      "loss": 1.1678,
      "num_input_tokens_seen": 3838720,
      "step": 785,
      "train_runtime": 2681.4032,
      "train_tokens_per_second": 1431.609
    },
    {
      "epoch": 0.2891034482758621,
      "grad_norm": 0.34054291248321533,
      "learning_rate": 0.00014252026529108327,
      "loss": 1.1999,
      "num_input_tokens_seen": 3842800,
      "step": 786,
      "train_runtime": 2684.2349,
      "train_tokens_per_second": 1431.618
    },
    {
      "epoch": 0.2894712643678161,
      "grad_norm": 0.3909076452255249,
      "learning_rate": 0.00014244657332350774,
      "loss": 1.1922,
      "num_input_tokens_seen": 3848688,
      "step": 787,
      "train_runtime": 2688.1308,
      "train_tokens_per_second": 1431.734
    },
    {
      "epoch": 0.28983908045977014,
      "grad_norm": 0.3771749436855316,
      "learning_rate": 0.0001423728813559322,
      "loss": 1.2433,
      "num_input_tokens_seen": 3853808,
      "step": 788,
      "train_runtime": 2691.583,
      "train_tokens_per_second": 1431.8
    },
    {
      "epoch": 0.29020689655172416,
      "grad_norm": 0.39928269386291504,
      "learning_rate": 0.00014229918938835667,
      "loss": 1.4231,
      "num_input_tokens_seen": 3858016,
      "step": 789,
      "train_runtime": 2694.4798,
      "train_tokens_per_second": 1431.822
    },
    {
      "epoch": 0.2905747126436782,
      "grad_norm": 0.345500111579895,
      "learning_rate": 0.00014222549742078116,
      "loss": 1.3395,
      "num_input_tokens_seen": 3864576,
      "step": 790,
      "train_runtime": 2698.6146,
      "train_tokens_per_second": 1432.059
    },
    {
      "epoch": 0.2909425287356322,
      "grad_norm": 0.33229881525039673,
      "learning_rate": 0.0001421518054532056,
      "loss": 1.3211,
      "num_input_tokens_seen": 3869104,
      "step": 791,
      "train_runtime": 2701.67,
      "train_tokens_per_second": 1432.116
    },
    {
      "epoch": 0.2913103448275862,
      "grad_norm": 0.3693390488624573,
      "learning_rate": 0.00014207811348563006,
      "loss": 1.4466,
      "num_input_tokens_seen": 3872880,
      "step": 792,
      "train_runtime": 2704.2742,
      "train_tokens_per_second": 1432.133
    },
    {
      "epoch": 0.29167816091954024,
      "grad_norm": 0.42008399963378906,
      "learning_rate": 0.00014200442151805453,
      "loss": 1.2877,
      "num_input_tokens_seen": 3878096,
      "step": 793,
      "train_runtime": 2707.7595,
      "train_tokens_per_second": 1432.216
    },
    {
      "epoch": 0.29204597701149426,
      "grad_norm": 0.3797346353530884,
      "learning_rate": 0.00014193072955047902,
      "loss": 1.2511,
      "num_input_tokens_seen": 3882688,
      "step": 794,
      "train_runtime": 2710.8803,
      "train_tokens_per_second": 1432.261
    },
    {
      "epoch": 0.2924137931034483,
      "grad_norm": 0.3746705949306488,
      "learning_rate": 0.00014185703758290349,
      "loss": 1.2108,
      "num_input_tokens_seen": 3889456,
      "step": 795,
      "train_runtime": 2715.1749,
      "train_tokens_per_second": 1432.488
    },
    {
      "epoch": 0.2927816091954023,
      "grad_norm": 0.38660576939582825,
      "learning_rate": 0.00014178334561532792,
      "loss": 1.1945,
      "num_input_tokens_seen": 3893216,
      "step": 796,
      "train_runtime": 2717.8414,
      "train_tokens_per_second": 1432.466
    },
    {
      "epoch": 0.2931494252873563,
      "grad_norm": 0.35182976722717285,
      "learning_rate": 0.0001417096536477524,
      "loss": 1.2695,
      "num_input_tokens_seen": 3898144,
      "step": 797,
      "train_runtime": 2721.2337,
      "train_tokens_per_second": 1432.491
    },
    {
      "epoch": 0.29351724137931035,
      "grad_norm": 0.374563604593277,
      "learning_rate": 0.00014163596168017688,
      "loss": 1.3673,
      "num_input_tokens_seen": 3902384,
      "step": 798,
      "train_runtime": 2724.1908,
      "train_tokens_per_second": 1432.493
    },
    {
      "epoch": 0.29388505747126437,
      "grad_norm": 0.33761313557624817,
      "learning_rate": 0.00014156226971260134,
      "loss": 1.1371,
      "num_input_tokens_seen": 3908016,
      "step": 799,
      "train_runtime": 2727.9158,
      "train_tokens_per_second": 1432.601
    },
    {
      "epoch": 0.2942528735632184,
      "grad_norm": 0.36212578415870667,
      "learning_rate": 0.0001414885777450258,
      "loss": 1.3324,
      "num_input_tokens_seen": 3912384,
      "step": 800,
      "train_runtime": 2730.9785,
      "train_tokens_per_second": 1432.594
    },
    {
      "epoch": 0.2946206896551724,
      "grad_norm": 0.3794098198413849,
      "learning_rate": 0.00014141488577745025,
      "loss": 1.2653,
      "num_input_tokens_seen": 3918592,
      "step": 801,
      "train_runtime": 2734.9526,
      "train_tokens_per_second": 1432.782
    },
    {
      "epoch": 0.29498850574712643,
      "grad_norm": 0.33707788586616516,
      "learning_rate": 0.00014134119380987474,
      "loss": 1.1963,
      "num_input_tokens_seen": 3924432,
      "step": 802,
      "train_runtime": 2738.8145,
      "train_tokens_per_second": 1432.894
    },
    {
      "epoch": 0.29535632183908045,
      "grad_norm": 0.3408147096633911,
      "learning_rate": 0.0001412675018422992,
      "loss": 1.2344,
      "num_input_tokens_seen": 3928400,
      "step": 803,
      "train_runtime": 2741.6374,
      "train_tokens_per_second": 1432.866
    },
    {
      "epoch": 0.29572413793103447,
      "grad_norm": 0.4036368131637573,
      "learning_rate": 0.00014119380987472367,
      "loss": 1.3421,
      "num_input_tokens_seen": 3931920,
      "step": 804,
      "train_runtime": 2744.1924,
      "train_tokens_per_second": 1432.815
    },
    {
      "epoch": 0.2960919540229885,
      "grad_norm": 0.3561180531978607,
      "learning_rate": 0.0001411201179071481,
      "loss": 1.3033,
      "num_input_tokens_seen": 3937200,
      "step": 805,
      "train_runtime": 2747.6812,
      "train_tokens_per_second": 1432.917
    },
    {
      "epoch": 0.2964597701149425,
      "grad_norm": 0.3317915201187134,
      "learning_rate": 0.0001410464259395726,
      "loss": 1.2202,
      "num_input_tokens_seen": 3941488,
      "step": 806,
      "train_runtime": 2750.6343,
      "train_tokens_per_second": 1432.938
    },
    {
      "epoch": 0.29682758620689653,
      "grad_norm": 0.3520815670490265,
      "learning_rate": 0.00014097273397199706,
      "loss": 1.2568,
      "num_input_tokens_seen": 3946640,
      "step": 807,
      "train_runtime": 2754.0538,
      "train_tokens_per_second": 1433.029
    },
    {
      "epoch": 0.29719540229885055,
      "grad_norm": 0.3856048285961151,
      "learning_rate": 0.00014089904200442153,
      "loss": 1.3211,
      "num_input_tokens_seen": 3949952,
      "step": 808,
      "train_runtime": 2756.4612,
      "train_tokens_per_second": 1432.979
    },
    {
      "epoch": 0.2975632183908046,
      "grad_norm": 0.3464462161064148,
      "learning_rate": 0.000140825350036846,
      "loss": 1.2971,
      "num_input_tokens_seen": 3953760,
      "step": 809,
      "train_runtime": 2759.174,
      "train_tokens_per_second": 1432.951
    },
    {
      "epoch": 0.2979310344827586,
      "grad_norm": 0.3052591383457184,
      "learning_rate": 0.00014075165806927046,
      "loss": 1.0736,
      "num_input_tokens_seen": 3973888,
      "step": 810,
      "train_runtime": 2777.3701,
      "train_tokens_per_second": 1430.81
    },
    {
      "epoch": 0.29829885057471267,
      "grad_norm": 0.37174367904663086,
      "learning_rate": 0.00014067796610169492,
      "loss": 1.1998,
      "num_input_tokens_seen": 3978432,
      "step": 811,
      "train_runtime": 2782.5261,
      "train_tokens_per_second": 1429.791
    },
    {
      "epoch": 0.2986666666666667,
      "grad_norm": 0.33270713686943054,
      "learning_rate": 0.0001406042741341194,
      "loss": 1.3258,
      "num_input_tokens_seen": 3982688,
      "step": 812,
      "train_runtime": 2785.4371,
      "train_tokens_per_second": 1429.825
    },
    {
      "epoch": 0.2990344827586207,
      "grad_norm": 0.3413871228694916,
      "learning_rate": 0.00014053058216654385,
      "loss": 1.1881,
      "num_input_tokens_seen": 3986352,
      "step": 813,
      "train_runtime": 2787.996,
      "train_tokens_per_second": 1429.827
    },
    {
      "epoch": 0.29940229885057473,
      "grad_norm": 0.3527846932411194,
      "learning_rate": 0.00014045689019896832,
      "loss": 1.1748,
      "num_input_tokens_seen": 3993104,
      "step": 814,
      "train_runtime": 2792.2466,
      "train_tokens_per_second": 1430.069
    },
    {
      "epoch": 0.29977011494252875,
      "grad_norm": 0.35125231742858887,
      "learning_rate": 0.00014038319823139278,
      "loss": 1.2081,
      "num_input_tokens_seen": 3997856,
      "step": 815,
      "train_runtime": 2795.5103,
      "train_tokens_per_second": 1430.099
    },
    {
      "epoch": 0.3001379310344828,
      "grad_norm": 0.3626596927642822,
      "learning_rate": 0.00014030950626381725,
      "loss": 1.2606,
      "num_input_tokens_seen": 4002688,
      "step": 816,
      "train_runtime": 2798.856,
      "train_tokens_per_second": 1430.116
    },
    {
      "epoch": 0.3001379310344828,
      "eval_loss": 1.2582504749298096,
      "eval_runtime": 15.4607,
      "eval_samples_per_second": 64.68,
      "eval_steps_per_second": 4.075,
      "num_input_tokens_seen": 4002688,
      "step": 816
    },
    {
      "epoch": 0.3005057471264368,
      "grad_norm": 0.34670722484588623,
      "learning_rate": 0.0001402358142962417,
      "loss": 1.222,
      "num_input_tokens_seen": 4013536,
      "step": 817,
      "train_runtime": 2821.0771,
      "train_tokens_per_second": 1422.696
    },
    {
      "epoch": 0.3008735632183908,
      "grad_norm": 0.3271850645542145,
      "learning_rate": 0.0001401621223286662,
      "loss": 1.1984,
      "num_input_tokens_seen": 4019024,
      "step": 818,
      "train_runtime": 2824.7163,
      "train_tokens_per_second": 1422.806
    },
    {
      "epoch": 0.30124137931034484,
      "grad_norm": 0.36956825852394104,
      "learning_rate": 0.00014008843036109064,
      "loss": 1.2282,
      "num_input_tokens_seen": 4027216,
      "step": 819,
      "train_runtime": 2829.7549,
      "train_tokens_per_second": 1423.168
    },
    {
      "epoch": 0.30160919540229886,
      "grad_norm": 0.37119704484939575,
      "learning_rate": 0.0001400147383935151,
      "loss": 1.2695,
      "num_input_tokens_seen": 4031088,
      "step": 820,
      "train_runtime": 2832.6029,
      "train_tokens_per_second": 1423.104
    },
    {
      "epoch": 0.3019770114942529,
      "grad_norm": 0.3455086648464203,
      "learning_rate": 0.00013994104642593957,
      "loss": 1.1874,
      "num_input_tokens_seen": 4036592,
      "step": 821,
      "train_runtime": 2836.2378,
      "train_tokens_per_second": 1423.221
    },
    {
      "epoch": 0.3023448275862069,
      "grad_norm": 0.32548725605010986,
      "learning_rate": 0.00013986735445836406,
      "loss": 1.259,
      "num_input_tokens_seen": 4042752,
      "step": 822,
      "train_runtime": 2840.2091,
      "train_tokens_per_second": 1423.399
    },
    {
      "epoch": 0.3027126436781609,
      "grad_norm": 0.38643863797187805,
      "learning_rate": 0.0001397936624907885,
      "loss": 1.3641,
      "num_input_tokens_seen": 4049392,
      "step": 823,
      "train_runtime": 2844.4764,
      "train_tokens_per_second": 1423.598
    },
    {
      "epoch": 0.30308045977011494,
      "grad_norm": 0.33685386180877686,
      "learning_rate": 0.00013971997052321296,
      "loss": 1.161,
      "num_input_tokens_seen": 4052960,
      "step": 824,
      "train_runtime": 2847.1058,
      "train_tokens_per_second": 1423.537
    },
    {
      "epoch": 0.30344827586206896,
      "grad_norm": 0.34240755438804626,
      "learning_rate": 0.00013964627855563743,
      "loss": 1.3907,
      "num_input_tokens_seen": 4056784,
      "step": 825,
      "train_runtime": 2849.7671,
      "train_tokens_per_second": 1423.549
    },
    {
      "epoch": 0.303816091954023,
      "grad_norm": 0.36610493063926697,
      "learning_rate": 0.00013957258658806192,
      "loss": 1.1741,
      "num_input_tokens_seen": 4061424,
      "step": 826,
      "train_runtime": 2852.9142,
      "train_tokens_per_second": 1423.605
    },
    {
      "epoch": 0.304183908045977,
      "grad_norm": 0.3073885440826416,
      "learning_rate": 0.0001394988946204864,
      "loss": 1.224,
      "num_input_tokens_seen": 4070176,
      "step": 827,
      "train_runtime": 2858.4139,
      "train_tokens_per_second": 1423.928
    },
    {
      "epoch": 0.304551724137931,
      "grad_norm": 0.3592829406261444,
      "learning_rate": 0.00013942520265291082,
      "loss": 1.2882,
      "num_input_tokens_seen": 4073824,
      "step": 828,
      "train_runtime": 2861.0739,
      "train_tokens_per_second": 1423.879
    },
    {
      "epoch": 0.30491954022988504,
      "grad_norm": 0.34713849425315857,
      "learning_rate": 0.0001393515106853353,
      "loss": 1.2273,
      "num_input_tokens_seen": 4077792,
      "step": 829,
      "train_runtime": 2863.9315,
      "train_tokens_per_second": 1423.844
    },
    {
      "epoch": 0.30528735632183907,
      "grad_norm": 0.3161631226539612,
      "learning_rate": 0.00013927781871775978,
      "loss": 1.2369,
      "num_input_tokens_seen": 4082384,
      "step": 830,
      "train_runtime": 2867.1532,
      "train_tokens_per_second": 1423.846
    },
    {
      "epoch": 0.3056551724137931,
      "grad_norm": 0.3770533502101898,
      "learning_rate": 0.00013920412675018425,
      "loss": 1.2978,
      "num_input_tokens_seen": 4087136,
      "step": 831,
      "train_runtime": 2870.4385,
      "train_tokens_per_second": 1423.872
    },
    {
      "epoch": 0.3060229885057471,
      "grad_norm": 0.34934914112091064,
      "learning_rate": 0.0001391304347826087,
      "loss": 1.2095,
      "num_input_tokens_seen": 4091360,
      "step": 832,
      "train_runtime": 2873.4129,
      "train_tokens_per_second": 1423.868
    },
    {
      "epoch": 0.3063908045977011,
      "grad_norm": 0.3711075484752655,
      "learning_rate": 0.00013905674281503315,
      "loss": 1.3721,
      "num_input_tokens_seen": 4095984,
      "step": 833,
      "train_runtime": 2876.5479,
      "train_tokens_per_second": 1423.923
    },
    {
      "epoch": 0.30675862068965515,
      "grad_norm": 0.40674078464508057,
      "learning_rate": 0.00013898305084745764,
      "loss": 1.195,
      "num_input_tokens_seen": 4099296,
      "step": 834,
      "train_runtime": 2879.0,
      "train_tokens_per_second": 1423.861
    },
    {
      "epoch": 0.30712643678160917,
      "grad_norm": 0.3847304880619049,
      "learning_rate": 0.0001389093588798821,
      "loss": 1.2805,
      "num_input_tokens_seen": 4103968,
      "step": 835,
      "train_runtime": 2882.2299,
      "train_tokens_per_second": 1423.886
    },
    {
      "epoch": 0.30749425287356325,
      "grad_norm": 0.3264797031879425,
      "learning_rate": 0.00013883566691230657,
      "loss": 1.2059,
      "num_input_tokens_seen": 4107600,
      "step": 836,
      "train_runtime": 2884.8201,
      "train_tokens_per_second": 1423.867
    },
    {
      "epoch": 0.30786206896551727,
      "grad_norm": 0.31736651062965393,
      "learning_rate": 0.00013876197494473103,
      "loss": 1.2878,
      "num_input_tokens_seen": 4119152,
      "step": 837,
      "train_runtime": 2891.7585,
      "train_tokens_per_second": 1424.445
    },
    {
      "epoch": 0.3082298850574713,
      "grad_norm": 0.37409865856170654,
      "learning_rate": 0.0001386882829771555,
      "loss": 1.1886,
      "num_input_tokens_seen": 4123792,
      "step": 838,
      "train_runtime": 2894.9126,
      "train_tokens_per_second": 1424.496
    },
    {
      "epoch": 0.3085977011494253,
      "grad_norm": 0.31045854091644287,
      "learning_rate": 0.00013861459100957996,
      "loss": 1.2056,
      "num_input_tokens_seen": 4129296,
      "step": 839,
      "train_runtime": 2898.6389,
      "train_tokens_per_second": 1424.564
    },
    {
      "epoch": 0.30896551724137933,
      "grad_norm": 0.38559892773628235,
      "learning_rate": 0.00013854089904200443,
      "loss": 1.3225,
      "num_input_tokens_seen": 4133088,
      "step": 840,
      "train_runtime": 2901.3552,
      "train_tokens_per_second": 1424.537
    },
    {
      "epoch": 0.30933333333333335,
      "grad_norm": 0.3517187535762787,
      "learning_rate": 0.0001384672070744289,
      "loss": 1.1359,
      "num_input_tokens_seen": 4136672,
      "step": 841,
      "train_runtime": 2906.2971,
      "train_tokens_per_second": 1423.348
    },
    {
      "epoch": 0.30970114942528737,
      "grad_norm": 0.32055985927581787,
      "learning_rate": 0.00013839351510685336,
      "loss": 1.1527,
      "num_input_tokens_seen": 4141024,
      "step": 842,
      "train_runtime": 2909.2928,
      "train_tokens_per_second": 1423.378
    },
    {
      "epoch": 0.3100689655172414,
      "grad_norm": 0.34698328375816345,
      "learning_rate": 0.00013831982313927782,
      "loss": 1.0366,
      "num_input_tokens_seen": 4146720,
      "step": 843,
      "train_runtime": 2913.0825,
      "train_tokens_per_second": 1423.482
    },
    {
      "epoch": 0.3104367816091954,
      "grad_norm": 0.3303372859954834,
      "learning_rate": 0.0001382461311717023,
      "loss": 1.1924,
      "num_input_tokens_seen": 4152224,
      "step": 844,
      "train_runtime": 2916.847,
      "train_tokens_per_second": 1423.532
    },
    {
      "epoch": 0.31080459770114943,
      "grad_norm": 0.34514808654785156,
      "learning_rate": 0.00013817243920412675,
      "loss": 1.1748,
      "num_input_tokens_seen": 4155280,
      "step": 845,
      "train_runtime": 2919.2146,
      "train_tokens_per_second": 1423.424
    },
    {
      "epoch": 0.31117241379310345,
      "grad_norm": 0.3158557415008545,
      "learning_rate": 0.00013809874723655122,
      "loss": 1.1718,
      "num_input_tokens_seen": 4162064,
      "step": 846,
      "train_runtime": 2923.603,
      "train_tokens_per_second": 1423.608
    },
    {
      "epoch": 0.3115402298850575,
      "grad_norm": 0.41774293780326843,
      "learning_rate": 0.00013802505526897568,
      "loss": 1.2337,
      "num_input_tokens_seen": 4166496,
      "step": 847,
      "train_runtime": 2926.6241,
      "train_tokens_per_second": 1423.653
    },
    {
      "epoch": 0.3119080459770115,
      "grad_norm": 0.3888136148452759,
      "learning_rate": 0.00013795136330140015,
      "loss": 1.1873,
      "num_input_tokens_seen": 4170656,
      "step": 848,
      "train_runtime": 2929.5387,
      "train_tokens_per_second": 1423.656
    },
    {
      "epoch": 0.3122758620689655,
      "grad_norm": 0.36558911204338074,
      "learning_rate": 0.0001378776713338246,
      "loss": 1.2043,
      "num_input_tokens_seen": 4174304,
      "step": 849,
      "train_runtime": 2932.1898,
      "train_tokens_per_second": 1423.613
    },
    {
      "epoch": 0.31264367816091954,
      "grad_norm": 0.34464168548583984,
      "learning_rate": 0.0001378039793662491,
      "loss": 1.3803,
      "num_input_tokens_seen": 4182480,
      "step": 850,
      "train_runtime": 2937.1931,
      "train_tokens_per_second": 1423.972
    },
    {
      "epoch": 0.31301149425287356,
      "grad_norm": 0.38367730379104614,
      "learning_rate": 0.00013773028739867354,
      "loss": 1.2086,
      "num_input_tokens_seen": 4187792,
      "step": 851,
      "train_runtime": 2940.7539,
      "train_tokens_per_second": 1424.054
    },
    {
      "epoch": 0.3133793103448276,
      "grad_norm": 0.36858445405960083,
      "learning_rate": 0.000137656595431098,
      "loss": 1.2054,
      "num_input_tokens_seen": 4197328,
      "step": 852,
      "train_runtime": 2946.6626,
      "train_tokens_per_second": 1424.435
    },
    {
      "epoch": 0.3137471264367816,
      "grad_norm": 0.36488014459609985,
      "learning_rate": 0.00013758290346352247,
      "loss": 1.2279,
      "num_input_tokens_seen": 4200848,
      "step": 853,
      "train_runtime": 2949.2275,
      "train_tokens_per_second": 1424.389
    },
    {
      "epoch": 0.3141149425287356,
      "grad_norm": 0.3878115713596344,
      "learning_rate": 0.00013750921149594696,
      "loss": 1.1837,
      "num_input_tokens_seen": 4206064,
      "step": 854,
      "train_runtime": 2952.7468,
      "train_tokens_per_second": 1424.458
    },
    {
      "epoch": 0.31448275862068964,
      "grad_norm": 0.374013751745224,
      "learning_rate": 0.0001374355195283714,
      "loss": 1.1583,
      "num_input_tokens_seen": 4212608,
      "step": 855,
      "train_runtime": 2956.9923,
      "train_tokens_per_second": 1424.626
    },
    {
      "epoch": 0.31485057471264366,
      "grad_norm": 0.32198452949523926,
      "learning_rate": 0.00013736182756079587,
      "loss": 1.0892,
      "num_input_tokens_seen": 4218368,
      "step": 856,
      "train_runtime": 2960.6824,
      "train_tokens_per_second": 1424.796
    },
    {
      "epoch": 0.3152183908045977,
      "grad_norm": 0.3488846719264984,
      "learning_rate": 0.00013728813559322033,
      "loss": 1.22,
      "num_input_tokens_seen": 4224032,
      "step": 857,
      "train_runtime": 2964.4227,
      "train_tokens_per_second": 1424.909
    },
    {
      "epoch": 0.3155862068965517,
      "grad_norm": 0.35620829463005066,
      "learning_rate": 0.00013721444362564482,
      "loss": 1.2182,
      "num_input_tokens_seen": 4232656,
      "step": 858,
      "train_runtime": 2969.731,
      "train_tokens_per_second": 1425.266
    },
    {
      "epoch": 0.3159540229885057,
      "grad_norm": 0.3545152544975281,
      "learning_rate": 0.0001371407516580693,
      "loss": 1.2895,
      "num_input_tokens_seen": 4236976,
      "step": 859,
      "train_runtime": 2972.7789,
      "train_tokens_per_second": 1425.258
    },
    {
      "epoch": 0.31632183908045974,
      "grad_norm": 0.3103776276111603,
      "learning_rate": 0.00013706705969049373,
      "loss": 1.2293,
      "num_input_tokens_seen": 4244128,
      "step": 860,
      "train_runtime": 2977.2706,
      "train_tokens_per_second": 1425.51
    },
    {
      "epoch": 0.3166896551724138,
      "grad_norm": 0.3932916820049286,
      "learning_rate": 0.0001369933677229182,
      "loss": 1.2819,
      "num_input_tokens_seen": 4247472,
      "step": 861,
      "train_runtime": 2979.6948,
      "train_tokens_per_second": 1425.472
    },
    {
      "epoch": 0.31705747126436784,
      "grad_norm": 0.3986624777317047,
      "learning_rate": 0.00013691967575534268,
      "loss": 1.2419,
      "num_input_tokens_seen": 4251840,
      "step": 862,
      "train_runtime": 2982.7573,
      "train_tokens_per_second": 1425.473
    },
    {
      "epoch": 0.31742528735632186,
      "grad_norm": 0.3446524739265442,
      "learning_rate": 0.00013684598378776715,
      "loss": 1.2602,
      "num_input_tokens_seen": 4257024,
      "step": 863,
      "train_runtime": 2986.2977,
      "train_tokens_per_second": 1425.519
    },
    {
      "epoch": 0.3177931034482759,
      "grad_norm": 0.35753557085990906,
      "learning_rate": 0.0001367722918201916,
      "loss": 1.2159,
      "num_input_tokens_seen": 4260304,
      "step": 864,
      "train_runtime": 2988.7913,
      "train_tokens_per_second": 1425.427
    },
    {
      "epoch": 0.3181609195402299,
      "grad_norm": 0.36021220684051514,
      "learning_rate": 0.00013669859985261605,
      "loss": 1.1423,
      "num_input_tokens_seen": 4265776,
      "step": 865,
      "train_runtime": 2992.402,
      "train_tokens_per_second": 1425.536
    },
    {
      "epoch": 0.3185287356321839,
      "grad_norm": 0.36315014958381653,
      "learning_rate": 0.00013662490788504054,
      "loss": 1.2235,
      "num_input_tokens_seen": 4269584,
      "step": 866,
      "train_runtime": 2995.0121,
      "train_tokens_per_second": 1425.565
    },
    {
      "epoch": 0.31889655172413794,
      "grad_norm": 0.37905824184417725,
      "learning_rate": 0.000136551215917465,
      "loss": 1.2489,
      "num_input_tokens_seen": 4273360,
      "step": 867,
      "train_runtime": 2997.7276,
      "train_tokens_per_second": 1425.533
    },
    {
      "epoch": 0.31926436781609197,
      "grad_norm": 0.3535110056400299,
      "learning_rate": 0.00013647752394988947,
      "loss": 1.3607,
      "num_input_tokens_seen": 4278000,
      "step": 868,
      "train_runtime": 3000.9476,
      "train_tokens_per_second": 1425.55
    },
    {
      "epoch": 0.319632183908046,
      "grad_norm": 0.41026008129119873,
      "learning_rate": 0.00013640383198231394,
      "loss": 1.3032,
      "num_input_tokens_seen": 4281264,
      "step": 869,
      "train_runtime": 3003.3614,
      "train_tokens_per_second": 1425.491
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.36517083644866943,
      "learning_rate": 0.0001363301400147384,
      "loss": 1.1699,
      "num_input_tokens_seen": 4285504,
      "step": 870,
      "train_runtime": 3006.6246,
      "train_tokens_per_second": 1425.354
    },
    {
      "epoch": 0.320367816091954,
      "grad_norm": 0.38390645384788513,
      "learning_rate": 0.00013625644804716287,
      "loss": 1.232,
      "num_input_tokens_seen": 4291184,
      "step": 871,
      "train_runtime": 3012.417,
      "train_tokens_per_second": 1424.499
    },
    {
      "epoch": 0.32073563218390805,
      "grad_norm": 0.34316956996917725,
      "learning_rate": 0.00013618275607958733,
      "loss": 1.2474,
      "num_input_tokens_seen": 4295120,
      "step": 872,
      "train_runtime": 3015.2289,
      "train_tokens_per_second": 1424.476
    },
    {
      "epoch": 0.32110344827586207,
      "grad_norm": 0.3473661541938782,
      "learning_rate": 0.0001361090641120118,
      "loss": 1.2379,
      "num_input_tokens_seen": 4299536,
      "step": 873,
      "train_runtime": 3018.3052,
      "train_tokens_per_second": 1424.487
    },
    {
      "epoch": 0.3214712643678161,
      "grad_norm": 0.3860956132411957,
      "learning_rate": 0.00013603537214443626,
      "loss": 1.1806,
      "num_input_tokens_seen": 4303776,
      "step": 874,
      "train_runtime": 3021.1917,
      "train_tokens_per_second": 1424.529
    },
    {
      "epoch": 0.3218390804597701,
      "grad_norm": 0.3867058753967285,
      "learning_rate": 0.00013596168017686072,
      "loss": 1.2291,
      "num_input_tokens_seen": 4308144,
      "step": 875,
      "train_runtime": 3024.2323,
      "train_tokens_per_second": 1424.541
    },
    {
      "epoch": 0.32220689655172413,
      "grad_norm": 0.3715849220752716,
      "learning_rate": 0.0001358879882092852,
      "loss": 1.3374,
      "num_input_tokens_seen": 4313248,
      "step": 876,
      "train_runtime": 3027.5392,
      "train_tokens_per_second": 1424.671
    },
    {
      "epoch": 0.32257471264367815,
      "grad_norm": 0.39966294169425964,
      "learning_rate": 0.00013581429624170968,
      "loss": 1.1806,
      "num_input_tokens_seen": 4318528,
      "step": 877,
      "train_runtime": 3031.1347,
      "train_tokens_per_second": 1424.723
    },
    {
      "epoch": 0.3229425287356322,
      "grad_norm": 0.34188470244407654,
      "learning_rate": 0.00013574060427413412,
      "loss": 1.2269,
      "num_input_tokens_seen": 4322608,
      "step": 878,
      "train_runtime": 3033.9163,
      "train_tokens_per_second": 1424.762
    },
    {
      "epoch": 0.3233103448275862,
      "grad_norm": 0.3809266984462738,
      "learning_rate": 0.00013566691230655858,
      "loss": 1.264,
      "num_input_tokens_seen": 4327824,
      "step": 879,
      "train_runtime": 3037.38,
      "train_tokens_per_second": 1424.854
    },
    {
      "epoch": 0.3236781609195402,
      "grad_norm": 0.33910226821899414,
      "learning_rate": 0.00013559322033898305,
      "loss": 1.202,
      "num_input_tokens_seen": 4332736,
      "step": 880,
      "train_runtime": 3040.7361,
      "train_tokens_per_second": 1424.897
    },
    {
      "epoch": 0.32404597701149424,
      "grad_norm": 0.38752177357673645,
      "learning_rate": 0.00013551952837140754,
      "loss": 1.2626,
      "num_input_tokens_seen": 4336864,
      "step": 881,
      "train_runtime": 3043.6928,
      "train_tokens_per_second": 1424.869
    },
    {
      "epoch": 0.32441379310344826,
      "grad_norm": 0.3363444209098816,
      "learning_rate": 0.000135445836403832,
      "loss": 1.2612,
      "num_input_tokens_seen": 4340288,
      "step": 882,
      "train_runtime": 3046.2645,
      "train_tokens_per_second": 1424.79
    },
    {
      "epoch": 0.3247816091954023,
      "grad_norm": 0.34718260169029236,
      "learning_rate": 0.00013537214443625644,
      "loss": 1.155,
      "num_input_tokens_seen": 4343376,
      "step": 883,
      "train_runtime": 3048.5796,
      "train_tokens_per_second": 1424.721
    },
    {
      "epoch": 0.3251494252873563,
      "grad_norm": 0.4002636671066284,
      "learning_rate": 0.0001352984524686809,
      "loss": 1.3875,
      "num_input_tokens_seen": 4349152,
      "step": 884,
      "train_runtime": 3052.385,
      "train_tokens_per_second": 1424.837
    },
    {
      "epoch": 0.3255172413793103,
      "grad_norm": 0.35120221972465515,
      "learning_rate": 0.0001352247605011054,
      "loss": 1.1387,
      "num_input_tokens_seen": 4353232,
      "step": 885,
      "train_runtime": 3055.2488,
      "train_tokens_per_second": 1424.837
    },
    {
      "epoch": 0.3258850574712644,
      "grad_norm": 0.3691864311695099,
      "learning_rate": 0.00013515106853352987,
      "loss": 1.2017,
      "num_input_tokens_seen": 4358288,
      "step": 886,
      "train_runtime": 3058.7541,
      "train_tokens_per_second": 1424.857
    },
    {
      "epoch": 0.3262528735632184,
      "grad_norm": 0.3733488619327545,
      "learning_rate": 0.00013507737656595433,
      "loss": 1.2489,
      "num_input_tokens_seen": 4362208,
      "step": 887,
      "train_runtime": 3061.5269,
      "train_tokens_per_second": 1424.847
    },
    {
      "epoch": 0.32662068965517244,
      "grad_norm": 0.3438883125782013,
      "learning_rate": 0.00013500368459837877,
      "loss": 1.2437,
      "num_input_tokens_seen": 4366880,
      "step": 888,
      "train_runtime": 3064.7051,
      "train_tokens_per_second": 1424.894
    },
    {
      "epoch": 0.32698850574712646,
      "grad_norm": 0.3556538224220276,
      "learning_rate": 0.00013492999263080326,
      "loss": 1.1462,
      "num_input_tokens_seen": 4370704,
      "step": 889,
      "train_runtime": 3067.3671,
      "train_tokens_per_second": 1424.904
    },
    {
      "epoch": 0.3273563218390805,
      "grad_norm": 0.35471776127815247,
      "learning_rate": 0.00013485630066322772,
      "loss": 1.1785,
      "num_input_tokens_seen": 4376976,
      "step": 890,
      "train_runtime": 3071.4714,
      "train_tokens_per_second": 1425.042
    },
    {
      "epoch": 0.3277241379310345,
      "grad_norm": 0.37259984016418457,
      "learning_rate": 0.0001347826086956522,
      "loss": 1.2901,
      "num_input_tokens_seen": 4381040,
      "step": 891,
      "train_runtime": 3074.2758,
      "train_tokens_per_second": 1425.064
    },
    {
      "epoch": 0.3280919540229885,
      "grad_norm": 0.3603115379810333,
      "learning_rate": 0.00013470891672807663,
      "loss": 1.3182,
      "num_input_tokens_seen": 4384912,
      "step": 892,
      "train_runtime": 3077.0913,
      "train_tokens_per_second": 1425.019
    },
    {
      "epoch": 0.32845977011494254,
      "grad_norm": 0.3494568169116974,
      "learning_rate": 0.00013463522476050112,
      "loss": 1.2556,
      "num_input_tokens_seen": 4390144,
      "step": 893,
      "train_runtime": 3080.5308,
      "train_tokens_per_second": 1425.126
    },
    {
      "epoch": 0.32882758620689656,
      "grad_norm": 0.3400580585002899,
      "learning_rate": 0.00013456153279292558,
      "loss": 1.2658,
      "num_input_tokens_seen": 4394352,
      "step": 894,
      "train_runtime": 3083.4253,
      "train_tokens_per_second": 1425.153
    },
    {
      "epoch": 0.3291954022988506,
      "grad_norm": 0.3196541666984558,
      "learning_rate": 0.00013448784082535005,
      "loss": 1.2209,
      "num_input_tokens_seen": 4400384,
      "step": 895,
      "train_runtime": 3087.3267,
      "train_tokens_per_second": 1425.306
    },
    {
      "epoch": 0.3295632183908046,
      "grad_norm": 0.36949363350868225,
      "learning_rate": 0.0001344141488577745,
      "loss": 1.1314,
      "num_input_tokens_seen": 4406416,
      "step": 896,
      "train_runtime": 3091.2935,
      "train_tokens_per_second": 1425.428
    },
    {
      "epoch": 0.3299310344827586,
      "grad_norm": 0.33267271518707275,
      "learning_rate": 0.00013434045689019898,
      "loss": 1.2986,
      "num_input_tokens_seen": 4410320,
      "step": 897,
      "train_runtime": 3094.0737,
      "train_tokens_per_second": 1425.409
    },
    {
      "epoch": 0.33029885057471264,
      "grad_norm": 0.32091811299324036,
      "learning_rate": 0.00013426676492262344,
      "loss": 1.221,
      "num_input_tokens_seen": 4418288,
      "step": 898,
      "train_runtime": 3099.0573,
      "train_tokens_per_second": 1425.688
    },
    {
      "epoch": 0.33066666666666666,
      "grad_norm": 0.3471040427684784,
      "learning_rate": 0.0001341930729550479,
      "loss": 1.2233,
      "num_input_tokens_seen": 4421936,
      "step": 899,
      "train_runtime": 3101.5775,
      "train_tokens_per_second": 1425.705
    },
    {
      "epoch": 0.3310344827586207,
      "grad_norm": 0.32608795166015625,
      "learning_rate": 0.00013411938098747237,
      "loss": 1.1769,
      "num_input_tokens_seen": 4425792,
      "step": 900,
      "train_runtime": 3104.3238,
      "train_tokens_per_second": 1425.686
    },
    {
      "epoch": 0.3314022988505747,
      "grad_norm": 0.3313031792640686,
      "learning_rate": 0.00013404568901989684,
      "loss": 1.0861,
      "num_input_tokens_seen": 4430528,
      "step": 901,
      "train_runtime": 3109.7062,
      "train_tokens_per_second": 1424.742
    },
    {
      "epoch": 0.3317701149425287,
      "grad_norm": 0.3731558322906494,
      "learning_rate": 0.0001339719970523213,
      "loss": 1.252,
      "num_input_tokens_seen": 4436208,
      "step": 902,
      "train_runtime": 3113.4198,
      "train_tokens_per_second": 1424.867
    },
    {
      "epoch": 0.33213793103448275,
      "grad_norm": 0.3392798900604248,
      "learning_rate": 0.00013389830508474577,
      "loss": 1.1572,
      "num_input_tokens_seen": 4439680,
      "step": 903,
      "train_runtime": 3115.987,
      "train_tokens_per_second": 1424.807
    },
    {
      "epoch": 0.33250574712643677,
      "grad_norm": 0.3852826654911041,
      "learning_rate": 0.00013382461311717023,
      "loss": 1.2067,
      "num_input_tokens_seen": 4444576,
      "step": 904,
      "train_runtime": 3119.2539,
      "train_tokens_per_second": 1424.884
    },
    {
      "epoch": 0.3328735632183908,
      "grad_norm": 0.3363071084022522,
      "learning_rate": 0.0001337509211495947,
      "loss": 1.0688,
      "num_input_tokens_seen": 4448080,
      "step": 905,
      "train_runtime": 3121.8841,
      "train_tokens_per_second": 1424.806
    },
    {
      "epoch": 0.3332413793103448,
      "grad_norm": 0.3272119462490082,
      "learning_rate": 0.00013367722918201916,
      "loss": 1.2854,
      "num_input_tokens_seen": 4453152,
      "step": 906,
      "train_runtime": 3125.2852,
      "train_tokens_per_second": 1424.879
    },
    {
      "epoch": 0.33360919540229883,
      "grad_norm": 0.37685421109199524,
      "learning_rate": 0.00013360353721444363,
      "loss": 1.2419,
      "num_input_tokens_seen": 4458672,
      "step": 907,
      "train_runtime": 3128.9176,
      "train_tokens_per_second": 1424.989
    },
    {
      "epoch": 0.33397701149425285,
      "grad_norm": 0.36570993065834045,
      "learning_rate": 0.0001335298452468681,
      "loss": 1.2802,
      "num_input_tokens_seen": 4462336,
      "step": 908,
      "train_runtime": 3131.5372,
      "train_tokens_per_second": 1424.967
    },
    {
      "epoch": 0.33434482758620687,
      "grad_norm": 0.406149685382843,
      "learning_rate": 0.00013345615327929258,
      "loss": 1.2721,
      "num_input_tokens_seen": 4467568,
      "step": 909,
      "train_runtime": 3135.0733,
      "train_tokens_per_second": 1425.028
    },
    {
      "epoch": 0.3347126436781609,
      "grad_norm": 0.3404579758644104,
      "learning_rate": 0.00013338246131171702,
      "loss": 1.1655,
      "num_input_tokens_seen": 4473040,
      "step": 910,
      "train_runtime": 3138.7322,
      "train_tokens_per_second": 1425.11
    },
    {
      "epoch": 0.33508045977011497,
      "grad_norm": 0.39098358154296875,
      "learning_rate": 0.00013330876934414149,
      "loss": 1.3671,
      "num_input_tokens_seen": 4478624,
      "step": 911,
      "train_runtime": 3142.6039,
      "train_tokens_per_second": 1425.132
    },
    {
      "epoch": 0.335448275862069,
      "grad_norm": 0.37660473585128784,
      "learning_rate": 0.00013323507737656595,
      "loss": 1.2616,
      "num_input_tokens_seen": 4483696,
      "step": 912,
      "train_runtime": 3146.0172,
      "train_tokens_per_second": 1425.198
    },
    {
      "epoch": 0.335816091954023,
      "grad_norm": 0.353767067193985,
      "learning_rate": 0.00013316138540899044,
      "loss": 1.2041,
      "num_input_tokens_seen": 4490336,
      "step": 913,
      "train_runtime": 3150.3059,
      "train_tokens_per_second": 1425.365
    },
    {
      "epoch": 0.33618390804597703,
      "grad_norm": 0.3491032123565674,
      "learning_rate": 0.0001330876934414149,
      "loss": 1.2778,
      "num_input_tokens_seen": 4496720,
      "step": 914,
      "train_runtime": 3154.4769,
      "train_tokens_per_second": 1425.504
    },
    {
      "epoch": 0.33655172413793105,
      "grad_norm": 0.3640507459640503,
      "learning_rate": 0.00013301400147383934,
      "loss": 1.2892,
      "num_input_tokens_seen": 4501040,
      "step": 915,
      "train_runtime": 3157.4112,
      "train_tokens_per_second": 1425.548
    },
    {
      "epoch": 0.3369195402298851,
      "grad_norm": 0.40220198035240173,
      "learning_rate": 0.0001329403095062638,
      "loss": 1.3624,
      "num_input_tokens_seen": 4504624,
      "step": 916,
      "train_runtime": 3160.0123,
      "train_tokens_per_second": 1425.508
    },
    {
      "epoch": 0.3372873563218391,
      "grad_norm": 0.3427709639072418,
      "learning_rate": 0.0001328666175386883,
      "loss": 1.2901,
      "num_input_tokens_seen": 4511392,
      "step": 917,
      "train_runtime": 3164.3259,
      "train_tokens_per_second": 1425.704
    },
    {
      "epoch": 0.3376551724137931,
      "grad_norm": 0.3462558388710022,
      "learning_rate": 0.00013279292557111277,
      "loss": 1.2467,
      "num_input_tokens_seen": 4516704,
      "step": 918,
      "train_runtime": 3167.8003,
      "train_tokens_per_second": 1425.817
    },
    {
      "epoch": 0.33802298850574714,
      "grad_norm": 0.371421754360199,
      "learning_rate": 0.00013271923360353723,
      "loss": 1.3749,
      "num_input_tokens_seen": 4524016,
      "step": 919,
      "train_runtime": 3172.4881,
      "train_tokens_per_second": 1426.015
    },
    {
      "epoch": 0.33839080459770116,
      "grad_norm": 0.41616541147232056,
      "learning_rate": 0.00013264554163596167,
      "loss": 1.3602,
      "num_input_tokens_seen": 4530240,
      "step": 920,
      "train_runtime": 3176.6814,
      "train_tokens_per_second": 1426.092
    },
    {
      "epoch": 0.3387586206896552,
      "grad_norm": 0.33587488532066345,
      "learning_rate": 0.00013257184966838616,
      "loss": 1.1841,
      "num_input_tokens_seen": 4535056,
      "step": 921,
      "train_runtime": 3179.9795,
      "train_tokens_per_second": 1426.127
    },
    {
      "epoch": 0.3391264367816092,
      "grad_norm": 0.31753021478652954,
      "learning_rate": 0.00013249815770081063,
      "loss": 1.2611,
      "num_input_tokens_seen": 4540384,
      "step": 922,
      "train_runtime": 3183.5169,
      "train_tokens_per_second": 1426.216
    },
    {
      "epoch": 0.3394942528735632,
      "grad_norm": 0.3612062931060791,
      "learning_rate": 0.0001324244657332351,
      "loss": 1.3056,
      "num_input_tokens_seen": 4543744,
      "step": 923,
      "train_runtime": 3186.0013,
      "train_tokens_per_second": 1426.159
    },
    {
      "epoch": 0.33986206896551724,
      "grad_norm": 0.3625712990760803,
      "learning_rate": 0.00013235077376565953,
      "loss": 1.2293,
      "num_input_tokens_seen": 4547408,
      "step": 924,
      "train_runtime": 3188.6542,
      "train_tokens_per_second": 1426.121
    },
    {
      "epoch": 0.34022988505747126,
      "grad_norm": 0.34556490182876587,
      "learning_rate": 0.00013227708179808402,
      "loss": 1.2502,
      "num_input_tokens_seen": 4551360,
      "step": 925,
      "train_runtime": 3191.4758,
      "train_tokens_per_second": 1426.099
    },
    {
      "epoch": 0.3405977011494253,
      "grad_norm": 0.3499390482902527,
      "learning_rate": 0.00013220338983050849,
      "loss": 1.232,
      "num_input_tokens_seen": 4554560,
      "step": 926,
      "train_runtime": 3193.8558,
      "train_tokens_per_second": 1426.038
    },
    {
      "epoch": 0.3409655172413793,
      "grad_norm": 0.3291085362434387,
      "learning_rate": 0.00013212969786293295,
      "loss": 1.2285,
      "num_input_tokens_seen": 4560256,
      "step": 927,
      "train_runtime": 3197.5843,
      "train_tokens_per_second": 1426.157
    },
    {
      "epoch": 0.3413333333333333,
      "grad_norm": 0.34726017713546753,
      "learning_rate": 0.00013205600589535741,
      "loss": 1.2759,
      "num_input_tokens_seen": 4564288,
      "step": 928,
      "train_runtime": 3200.3212,
      "train_tokens_per_second": 1426.197
    },
    {
      "epoch": 0.34170114942528734,
      "grad_norm": 0.3745157718658447,
      "learning_rate": 0.00013198231392778188,
      "loss": 1.3754,
      "num_input_tokens_seen": 4569280,
      "step": 929,
      "train_runtime": 3203.6142,
      "train_tokens_per_second": 1426.289
    },
    {
      "epoch": 0.34206896551724136,
      "grad_norm": 0.3378608822822571,
      "learning_rate": 0.00013190862196020634,
      "loss": 1.2081,
      "num_input_tokens_seen": 4572592,
      "step": 930,
      "train_runtime": 3206.0812,
      "train_tokens_per_second": 1426.225
    },
    {
      "epoch": 0.3424367816091954,
      "grad_norm": 0.3454687297344208,
      "learning_rate": 0.0001318349299926308,
      "loss": 1.3007,
      "num_input_tokens_seen": 4579328,
      "step": 931,
      "train_runtime": 3212.5794,
      "train_tokens_per_second": 1425.437
    },
    {
      "epoch": 0.3428045977011494,
      "grad_norm": 0.4085744619369507,
      "learning_rate": 0.00013176123802505527,
      "loss": 1.4149,
      "num_input_tokens_seen": 4588288,
      "step": 932,
      "train_runtime": 3218.0075,
      "train_tokens_per_second": 1425.816
    },
    {
      "epoch": 0.3431724137931034,
      "grad_norm": 0.3527067005634308,
      "learning_rate": 0.00013168754605747974,
      "loss": 1.074,
      "num_input_tokens_seen": 4591760,
      "step": 933,
      "train_runtime": 3220.5356,
      "train_tokens_per_second": 1425.775
    },
    {
      "epoch": 0.34354022988505745,
      "grad_norm": 0.40089431405067444,
      "learning_rate": 0.0001316138540899042,
      "loss": 1.2684,
      "num_input_tokens_seen": 4595872,
      "step": 934,
      "train_runtime": 3223.4665,
      "train_tokens_per_second": 1425.755
    },
    {
      "epoch": 0.34390804597701147,
      "grad_norm": 0.39275211095809937,
      "learning_rate": 0.00013154016212232867,
      "loss": 1.2534,
      "num_input_tokens_seen": 4599536,
      "step": 935,
      "train_runtime": 3226.1158,
      "train_tokens_per_second": 1425.719
    },
    {
      "epoch": 0.34427586206896554,
      "grad_norm": 0.38221997022628784,
      "learning_rate": 0.00013146647015475313,
      "loss": 1.1681,
      "num_input_tokens_seen": 4604496,
      "step": 936,
      "train_runtime": 3229.4178,
      "train_tokens_per_second": 1425.798
    },
    {
      "epoch": 0.34464367816091956,
      "grad_norm": 0.37064045667648315,
      "learning_rate": 0.00013139277818717763,
      "loss": 1.4688,
      "num_input_tokens_seen": 4608336,
      "step": 937,
      "train_runtime": 3232.0935,
      "train_tokens_per_second": 1425.805
    },
    {
      "epoch": 0.3450114942528736,
      "grad_norm": 0.38732385635375977,
      "learning_rate": 0.00013131908621960206,
      "loss": 1.3264,
      "num_input_tokens_seen": 4614224,
      "step": 938,
      "train_runtime": 3235.8867,
      "train_tokens_per_second": 1425.954
    },
    {
      "epoch": 0.3453793103448276,
      "grad_norm": 0.3807511627674103,
      "learning_rate": 0.00013124539425202653,
      "loss": 1.1603,
      "num_input_tokens_seen": 4619024,
      "step": 939,
      "train_runtime": 3239.1126,
      "train_tokens_per_second": 1426.015
    },
    {
      "epoch": 0.3457471264367816,
      "grad_norm": 0.3849567472934723,
      "learning_rate": 0.000131171702284451,
      "loss": 1.2313,
      "num_input_tokens_seen": 4624240,
      "step": 940,
      "train_runtime": 3242.5871,
      "train_tokens_per_second": 1426.096
    },
    {
      "epoch": 0.34611494252873565,
      "grad_norm": 0.4086439609527588,
      "learning_rate": 0.00013109801031687548,
      "loss": 1.0856,
      "num_input_tokens_seen": 4628928,
      "step": 941,
      "train_runtime": 3245.7458,
      "train_tokens_per_second": 1426.152
    },
    {
      "epoch": 0.34648275862068967,
      "grad_norm": 0.34102317690849304,
      "learning_rate": 0.00013102431834929992,
      "loss": 1.2654,
      "num_input_tokens_seen": 4632928,
      "step": 942,
      "train_runtime": 3248.543,
      "train_tokens_per_second": 1426.156
    },
    {
      "epoch": 0.3468505747126437,
      "grad_norm": 0.34738168120384216,
      "learning_rate": 0.0001309506263817244,
      "loss": 1.2962,
      "num_input_tokens_seen": 4637584,
      "step": 943,
      "train_runtime": 3251.6173,
      "train_tokens_per_second": 1426.239
    },
    {
      "epoch": 0.3472183908045977,
      "grad_norm": 0.36523938179016113,
      "learning_rate": 0.00013087693441414885,
      "loss": 1.1524,
      "num_input_tokens_seen": 4642128,
      "step": 944,
      "train_runtime": 3254.7734,
      "train_tokens_per_second": 1426.252
    },
    {
      "epoch": 0.34758620689655173,
      "grad_norm": 0.3553354740142822,
      "learning_rate": 0.00013080324244657334,
      "loss": 1.173,
      "num_input_tokens_seen": 4647184,
      "step": 945,
      "train_runtime": 3258.1284,
      "train_tokens_per_second": 1426.335
    },
    {
      "epoch": 0.34795402298850575,
      "grad_norm": 0.35105282068252563,
      "learning_rate": 0.0001307295504789978,
      "loss": 1.2189,
      "num_input_tokens_seen": 4651216,
      "step": 946,
      "train_runtime": 3260.969,
      "train_tokens_per_second": 1426.329
    },
    {
      "epoch": 0.34832183908045977,
      "grad_norm": 0.3944171369075775,
      "learning_rate": 0.00013065585851142225,
      "loss": 1.2914,
      "num_input_tokens_seen": 4656112,
      "step": 947,
      "train_runtime": 3264.2687,
      "train_tokens_per_second": 1426.387
    },
    {
      "epoch": 0.3486896551724138,
      "grad_norm": 0.3476920425891876,
      "learning_rate": 0.0001305821665438467,
      "loss": 1.1136,
      "num_input_tokens_seen": 4663328,
      "step": 948,
      "train_runtime": 3268.8119,
      "train_tokens_per_second": 1426.613
    },
    {
      "epoch": 0.3490574712643678,
      "grad_norm": 0.34117260575294495,
      "learning_rate": 0.0001305084745762712,
      "loss": 1.1779,
      "num_input_tokens_seen": 4672080,
      "step": 949,
      "train_runtime": 3274.2392,
      "train_tokens_per_second": 1426.921
    },
    {
      "epoch": 0.34942528735632183,
      "grad_norm": 0.40067145228385925,
      "learning_rate": 0.00013043478260869567,
      "loss": 1.2826,
      "num_input_tokens_seen": 4679568,
      "step": 950,
      "train_runtime": 3278.9844,
      "train_tokens_per_second": 1427.139
    },
    {
      "epoch": 0.34979310344827585,
      "grad_norm": 0.3745213449001312,
      "learning_rate": 0.00013036109064112013,
      "loss": 1.1688,
      "num_input_tokens_seen": 4683456,
      "step": 951,
      "train_runtime": 3281.7133,
      "train_tokens_per_second": 1427.137
    },
    {
      "epoch": 0.3501609195402299,
      "grad_norm": 0.3064570724964142,
      "learning_rate": 0.00013028739867354457,
      "loss": 1.1255,
      "num_input_tokens_seen": 4689072,
      "step": 952,
      "train_runtime": 3285.3409,
      "train_tokens_per_second": 1427.271
    },
    {
      "epoch": 0.3505287356321839,
      "grad_norm": 0.349273681640625,
      "learning_rate": 0.00013021370670596906,
      "loss": 1.1948,
      "num_input_tokens_seen": 4692464,
      "step": 953,
      "train_runtime": 3287.8948,
      "train_tokens_per_second": 1427.194
    },
    {
      "epoch": 0.3508965517241379,
      "grad_norm": 0.32461249828338623,
      "learning_rate": 0.00013014001473839353,
      "loss": 1.2896,
      "num_input_tokens_seen": 4699344,
      "step": 954,
      "train_runtime": 3292.3643,
      "train_tokens_per_second": 1427.346
    },
    {
      "epoch": 0.35126436781609194,
      "grad_norm": 0.4004507064819336,
      "learning_rate": 0.000130066322770818,
      "loss": 1.3518,
      "num_input_tokens_seen": 4705856,
      "step": 955,
      "train_runtime": 3296.5393,
      "train_tokens_per_second": 1427.514
    },
    {
      "epoch": 0.35163218390804596,
      "grad_norm": 0.32188403606414795,
      "learning_rate": 0.00012999263080324246,
      "loss": 1.1949,
      "num_input_tokens_seen": 4709616,
      "step": 956,
      "train_runtime": 3299.2295,
      "train_tokens_per_second": 1427.49
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.35095903277397156,
      "learning_rate": 0.00012991893883566692,
      "loss": 1.174,
      "num_input_tokens_seen": 4713936,
      "step": 957,
      "train_runtime": 3302.1052,
      "train_tokens_per_second": 1427.555
    },
    {
      "epoch": 0.352367816091954,
      "grad_norm": 0.3721597492694855,
      "learning_rate": 0.00012984524686809139,
      "loss": 1.2676,
      "num_input_tokens_seen": 4718448,
      "step": 958,
      "train_runtime": 3305.1539,
      "train_tokens_per_second": 1427.603
    },
    {
      "epoch": 0.352735632183908,
      "grad_norm": 0.35347044467926025,
      "learning_rate": 0.00012977155490051585,
      "loss": 1.3514,
      "num_input_tokens_seen": 4727920,
      "step": 959,
      "train_runtime": 3310.8936,
      "train_tokens_per_second": 1427.989
    },
    {
      "epoch": 0.35310344827586204,
      "grad_norm": 0.3443051874637604,
      "learning_rate": 0.00012969786293294032,
      "loss": 1.2247,
      "num_input_tokens_seen": 4731696,
      "step": 960,
      "train_runtime": 3313.537,
      "train_tokens_per_second": 1427.989
    },
    {
      "epoch": 0.3534712643678161,
      "grad_norm": 0.34611454606056213,
      "learning_rate": 0.00012962417096536478,
      "loss": 1.2259,
      "num_input_tokens_seen": 4736896,
      "step": 961,
      "train_runtime": 3318.983,
      "train_tokens_per_second": 1427.213
    },
    {
      "epoch": 0.35383908045977014,
      "grad_norm": 0.33163970708847046,
      "learning_rate": 0.00012955047899778925,
      "loss": 1.2611,
      "num_input_tokens_seen": 4740432,
      "step": 962,
      "train_runtime": 3321.5338,
      "train_tokens_per_second": 1427.182
    },
    {
      "epoch": 0.35420689655172416,
      "grad_norm": 0.3706001937389374,
      "learning_rate": 0.0001294767870302137,
      "loss": 1.1212,
      "num_input_tokens_seen": 4744208,
      "step": 963,
      "train_runtime": 3324.1479,
      "train_tokens_per_second": 1427.195
    },
    {
      "epoch": 0.3545747126436782,
      "grad_norm": 0.3629588484764099,
      "learning_rate": 0.00012940309506263818,
      "loss": 1.1623,
      "num_input_tokens_seen": 4751520,
      "step": 964,
      "train_runtime": 3328.7503,
      "train_tokens_per_second": 1427.419
    },
    {
      "epoch": 0.3549425287356322,
      "grad_norm": 0.3300287425518036,
      "learning_rate": 0.00012932940309506264,
      "loss": 1.1344,
      "num_input_tokens_seen": 4756784,
      "step": 965,
      "train_runtime": 3332.1707,
      "train_tokens_per_second": 1427.533
    },
    {
      "epoch": 0.3553103448275862,
      "grad_norm": 0.36114373803138733,
      "learning_rate": 0.0001292557111274871,
      "loss": 1.2865,
      "num_input_tokens_seen": 4761280,
      "step": 966,
      "train_runtime": 3335.2529,
      "train_tokens_per_second": 1427.562
    },
    {
      "epoch": 0.35567816091954024,
      "grad_norm": 0.36843106150627136,
      "learning_rate": 0.00012918201915991157,
      "loss": 1.3296,
      "num_input_tokens_seen": 4765984,
      "step": 967,
      "train_runtime": 3338.3947,
      "train_tokens_per_second": 1427.627
    },
    {
      "epoch": 0.35604597701149426,
      "grad_norm": 0.3429054915904999,
      "learning_rate": 0.00012910832719233603,
      "loss": 1.195,
      "num_input_tokens_seen": 4770784,
      "step": 968,
      "train_runtime": 3341.6176,
      "train_tokens_per_second": 1427.687
    },
    {
      "epoch": 0.3564137931034483,
      "grad_norm": 0.3531000316143036,
      "learning_rate": 0.00012903463522476053,
      "loss": 1.3216,
      "num_input_tokens_seen": 4776048,
      "step": 969,
      "train_runtime": 3345.1242,
      "train_tokens_per_second": 1427.764
    },
    {
      "epoch": 0.3567816091954023,
      "grad_norm": 0.35412395000457764,
      "learning_rate": 0.00012896094325718496,
      "loss": 1.2612,
      "num_input_tokens_seen": 4782544,
      "step": 970,
      "train_runtime": 3349.256,
      "train_tokens_per_second": 1427.942
    },
    {
      "epoch": 0.3571494252873563,
      "grad_norm": 0.35063204169273376,
      "learning_rate": 0.00012888725128960943,
      "loss": 1.2589,
      "num_input_tokens_seen": 4788160,
      "step": 971,
      "train_runtime": 3352.908,
      "train_tokens_per_second": 1428.062
    },
    {
      "epoch": 0.35751724137931035,
      "grad_norm": 0.4041181802749634,
      "learning_rate": 0.0001288135593220339,
      "loss": 1.4218,
      "num_input_tokens_seen": 4798224,
      "step": 972,
      "train_runtime": 3358.9967,
      "train_tokens_per_second": 1428.469
    },
    {
      "epoch": 0.35788505747126437,
      "grad_norm": 0.37234970927238464,
      "learning_rate": 0.00012873986735445839,
      "loss": 1.1347,
      "num_input_tokens_seen": 4803760,
      "step": 973,
      "train_runtime": 3362.6615,
      "train_tokens_per_second": 1428.559
    },
    {
      "epoch": 0.3582528735632184,
      "grad_norm": 0.3577387034893036,
      "learning_rate": 0.00012866617538688282,
      "loss": 1.2643,
      "num_input_tokens_seen": 4807904,
      "step": 974,
      "train_runtime": 3365.5693,
      "train_tokens_per_second": 1428.556
    },
    {
      "epoch": 0.3586206896551724,
      "grad_norm": 0.320300817489624,
      "learning_rate": 0.0001285924834193073,
      "loss": 1.2285,
      "num_input_tokens_seen": 4812032,
      "step": 975,
      "train_runtime": 3368.4502,
      "train_tokens_per_second": 1428.56
    },
    {
      "epoch": 0.35898850574712643,
      "grad_norm": 0.34907546639442444,
      "learning_rate": 0.00012851879145173175,
      "loss": 1.287,
      "num_input_tokens_seen": 4817520,
      "step": 976,
      "train_runtime": 3371.9347,
      "train_tokens_per_second": 1428.711
    },
    {
      "epoch": 0.35935632183908045,
      "grad_norm": 0.36896204948425293,
      "learning_rate": 0.00012844509948415625,
      "loss": 1.2219,
      "num_input_tokens_seen": 4822112,
      "step": 977,
      "train_runtime": 3375.0871,
      "train_tokens_per_second": 1428.737
    },
    {
      "epoch": 0.35972413793103447,
      "grad_norm": 0.3503137528896332,
      "learning_rate": 0.0001283714075165807,
      "loss": 1.2138,
      "num_input_tokens_seen": 4826384,
      "step": 978,
      "train_runtime": 3378.0733,
      "train_tokens_per_second": 1428.739
    },
    {
      "epoch": 0.3600919540229885,
      "grad_norm": 0.32998669147491455,
      "learning_rate": 0.00012829771554900515,
      "loss": 1.1317,
      "num_input_tokens_seen": 4829760,
      "step": 979,
      "train_runtime": 3380.5225,
      "train_tokens_per_second": 1428.702
    },
    {
      "epoch": 0.3604597701149425,
      "grad_norm": 0.3305336833000183,
      "learning_rate": 0.0001282240235814296,
      "loss": 1.1226,
      "num_input_tokens_seen": 4835984,
      "step": 980,
      "train_runtime": 3384.7151,
      "train_tokens_per_second": 1428.771
    },
    {
      "epoch": 0.36082758620689653,
      "grad_norm": 0.3566102087497711,
      "learning_rate": 0.0001281503316138541,
      "loss": 1.0873,
      "num_input_tokens_seen": 4840800,
      "step": 981,
      "train_runtime": 3387.9279,
      "train_tokens_per_second": 1428.838
    },
    {
      "epoch": 0.36119540229885055,
      "grad_norm": 0.3306235074996948,
      "learning_rate": 0.00012807663964627857,
      "loss": 1.1995,
      "num_input_tokens_seen": 4846384,
      "step": 982,
      "train_runtime": 3391.6798,
      "train_tokens_per_second": 1428.904
    },
    {
      "epoch": 0.3615632183908046,
      "grad_norm": 0.3327372074127197,
      "learning_rate": 0.00012800294767870303,
      "loss": 1.2495,
      "num_input_tokens_seen": 4850208,
      "step": 983,
      "train_runtime": 3394.3494,
      "train_tokens_per_second": 1428.907
    },
    {
      "epoch": 0.3619310344827586,
      "grad_norm": 0.32935208082199097,
      "learning_rate": 0.00012792925571112747,
      "loss": 1.2944,
      "num_input_tokens_seen": 4854432,
      "step": 984,
      "train_runtime": 3397.2232,
      "train_tokens_per_second": 1428.941
    },
    {
      "epoch": 0.3622988505747126,
      "grad_norm": 0.3445834219455719,
      "learning_rate": 0.00012785556374355196,
      "loss": 1.3027,
      "num_input_tokens_seen": 4862320,
      "step": 985,
      "train_runtime": 3402.1135,
      "train_tokens_per_second": 1429.206
    },
    {
      "epoch": 0.3626666666666667,
      "grad_norm": 0.36261871457099915,
      "learning_rate": 0.00012778187177597643,
      "loss": 1.3979,
      "num_input_tokens_seen": 4866848,
      "step": 986,
      "train_runtime": 3405.0901,
      "train_tokens_per_second": 1429.286
    },
    {
      "epoch": 0.3630344827586207,
      "grad_norm": 0.37101832032203674,
      "learning_rate": 0.0001277081798084009,
      "loss": 1.2875,
      "num_input_tokens_seen": 4871568,
      "step": 987,
      "train_runtime": 3408.25,
      "train_tokens_per_second": 1429.346
    },
    {
      "epoch": 0.36340229885057473,
      "grad_norm": 0.31869280338287354,
      "learning_rate": 0.00012763448784082536,
      "loss": 1.1483,
      "num_input_tokens_seen": 4880112,
      "step": 988,
      "train_runtime": 3413.4524,
      "train_tokens_per_second": 1429.67
    },
    {
      "epoch": 0.36377011494252876,
      "grad_norm": 0.37963733077049255,
      "learning_rate": 0.00012756079587324982,
      "loss": 1.2981,
      "num_input_tokens_seen": 4884320,
      "step": 989,
      "train_runtime": 3416.3965,
      "train_tokens_per_second": 1429.67
    },
    {
      "epoch": 0.3641379310344828,
      "grad_norm": 0.3426075279712677,
      "learning_rate": 0.0001274871039056743,
      "loss": 1.3062,
      "num_input_tokens_seen": 4892016,
      "step": 990,
      "train_runtime": 3421.2239,
      "train_tokens_per_second": 1429.902
    },
    {
      "epoch": 0.3645057471264368,
      "grad_norm": 0.36241981387138367,
      "learning_rate": 0.00012741341193809875,
      "loss": 1.1224,
      "num_input_tokens_seen": 4896128,
      "step": 991,
      "train_runtime": 3426.1408,
      "train_tokens_per_second": 1429.05
    },
    {
      "epoch": 0.3648735632183908,
      "grad_norm": 0.3554629385471344,
      "learning_rate": 0.00012733971997052322,
      "loss": 1.3198,
      "num_input_tokens_seen": 4900880,
      "step": 992,
      "train_runtime": 3429.3353,
      "train_tokens_per_second": 1429.105
    },
    {
      "epoch": 0.36524137931034484,
      "grad_norm": 0.3605075180530548,
      "learning_rate": 0.00012726602800294768,
      "loss": 1.2784,
      "num_input_tokens_seen": 4906304,
      "step": 993,
      "train_runtime": 3432.8568,
      "train_tokens_per_second": 1429.219
    },
    {
      "epoch": 0.36560919540229886,
      "grad_norm": 0.3608716130256653,
      "learning_rate": 0.00012719233603537215,
      "loss": 1.2771,
      "num_input_tokens_seen": 4911216,
      "step": 994,
      "train_runtime": 3436.0811,
      "train_tokens_per_second": 1429.307
    },
    {
      "epoch": 0.3659770114942529,
      "grad_norm": 0.3807471990585327,
      "learning_rate": 0.0001271186440677966,
      "loss": 1.3637,
      "num_input_tokens_seen": 4916960,
      "step": 995,
      "train_runtime": 3439.7987,
      "train_tokens_per_second": 1429.433
    },
    {
      "epoch": 0.3663448275862069,
      "grad_norm": 0.3879953622817993,
      "learning_rate": 0.0001270449521002211,
      "loss": 1.3138,
      "num_input_tokens_seen": 4921472,
      "step": 996,
      "train_runtime": 3442.7866,
      "train_tokens_per_second": 1429.502
    },
    {
      "epoch": 0.3667126436781609,
      "grad_norm": 0.38234058022499084,
      "learning_rate": 0.00012697126013264554,
      "loss": 1.1955,
      "num_input_tokens_seen": 4925840,
      "step": 997,
      "train_runtime": 3445.7929,
      "train_tokens_per_second": 1429.523
    },
    {
      "epoch": 0.36708045977011494,
      "grad_norm": 0.36829519271850586,
      "learning_rate": 0.00012689756816507,
      "loss": 1.2126,
      "num_input_tokens_seen": 4935904,
      "step": 998,
      "train_runtime": 3451.8197,
      "train_tokens_per_second": 1429.943
    },
    {
      "epoch": 0.36744827586206896,
      "grad_norm": 0.378055214881897,
      "learning_rate": 0.00012682387619749447,
      "loss": 1.1742,
      "num_input_tokens_seen": 4939904,
      "step": 999,
      "train_runtime": 3454.5562,
      "train_tokens_per_second": 1429.968
    },
    {
      "epoch": 0.367816091954023,
      "grad_norm": 0.33647602796554565,
      "learning_rate": 0.00012675018422991896,
      "loss": 1.1596,
      "num_input_tokens_seen": 4945104,
      "step": 1000,
      "train_runtime": 3458.0131,
      "train_tokens_per_second": 1430.042
    },
    {
      "epoch": 0.368183908045977,
      "grad_norm": 0.3450307250022888,
      "learning_rate": 0.00012667649226234343,
      "loss": 1.3509,
      "num_input_tokens_seen": 4949616,
      "step": 1001,
      "train_runtime": 3461.0538,
      "train_tokens_per_second": 1430.089
    },
    {
      "epoch": 0.368551724137931,
      "grad_norm": 0.3409826457500458,
      "learning_rate": 0.00012660280029476787,
      "loss": 1.2557,
      "num_input_tokens_seen": 4955568,
      "step": 1002,
      "train_runtime": 3464.8782,
      "train_tokens_per_second": 1430.229
    },
    {
      "epoch": 0.36891954022988505,
      "grad_norm": 0.344571590423584,
      "learning_rate": 0.00012652910832719233,
      "loss": 1.1829,
      "num_input_tokens_seen": 4964048,
      "step": 1003,
      "train_runtime": 3470.0093,
      "train_tokens_per_second": 1430.558
    },
    {
      "epoch": 0.36928735632183907,
      "grad_norm": 0.33147361874580383,
      "learning_rate": 0.00012645541635961682,
      "loss": 1.2453,
      "num_input_tokens_seen": 4971616,
      "step": 1004,
      "train_runtime": 3474.7641,
      "train_tokens_per_second": 1430.778
    },
    {
      "epoch": 0.3696551724137931,
      "grad_norm": 0.33022987842559814,
      "learning_rate": 0.0001263817243920413,
      "loss": 1.2299,
      "num_input_tokens_seen": 4977968,
      "step": 1005,
      "train_runtime": 3478.7439,
      "train_tokens_per_second": 1430.967
    },
    {
      "epoch": 0.3700229885057471,
      "grad_norm": 0.33206284046173096,
      "learning_rate": 0.00012630803242446575,
      "loss": 1.1737,
      "num_input_tokens_seen": 4981744,
      "step": 1006,
      "train_runtime": 3481.3857,
      "train_tokens_per_second": 1430.966
    },
    {
      "epoch": 0.37039080459770113,
      "grad_norm": 0.3408454656600952,
      "learning_rate": 0.0001262343404568902,
      "loss": 1.3833,
      "num_input_tokens_seen": 4987616,
      "step": 1007,
      "train_runtime": 3485.0999,
      "train_tokens_per_second": 1431.126
    },
    {
      "epoch": 0.37075862068965515,
      "grad_norm": 0.3427160382270813,
      "learning_rate": 0.00012616064848931468,
      "loss": 1.2127,
      "num_input_tokens_seen": 4992992,
      "step": 1008,
      "train_runtime": 3488.5941,
      "train_tokens_per_second": 1431.233
    },
    {
      "epoch": 0.37112643678160917,
      "grad_norm": 0.36299392580986023,
      "learning_rate": 0.00012608695652173915,
      "loss": 1.2585,
      "num_input_tokens_seen": 4996336,
      "step": 1009,
      "train_runtime": 3491.0463,
      "train_tokens_per_second": 1431.186
    },
    {
      "epoch": 0.3714942528735632,
      "grad_norm": 0.3574354946613312,
      "learning_rate": 0.0001260132645541636,
      "loss": 1.1855,
      "num_input_tokens_seen": 5000064,
      "step": 1010,
      "train_runtime": 3493.7222,
      "train_tokens_per_second": 1431.157
    },
    {
      "epoch": 0.37186206896551727,
      "grad_norm": 0.35885098576545715,
      "learning_rate": 0.00012593957258658805,
      "loss": 1.1812,
      "num_input_tokens_seen": 5003440,
      "step": 1011,
      "train_runtime": 3496.1463,
      "train_tokens_per_second": 1431.13
    },
    {
      "epoch": 0.3722298850574713,
      "grad_norm": 0.34946689009666443,
      "learning_rate": 0.00012586588061901254,
      "loss": 1.2643,
      "num_input_tokens_seen": 5008176,
      "step": 1012,
      "train_runtime": 3499.2998,
      "train_tokens_per_second": 1431.194
    },
    {
      "epoch": 0.3725977011494253,
      "grad_norm": 0.3375435173511505,
      "learning_rate": 0.000125792188651437,
      "loss": 1.2861,
      "num_input_tokens_seen": 5012208,
      "step": 1013,
      "train_runtime": 3502.1225,
      "train_tokens_per_second": 1431.191
    },
    {
      "epoch": 0.37296551724137933,
      "grad_norm": 0.3539506196975708,
      "learning_rate": 0.00012571849668386147,
      "loss": 1.2749,
      "num_input_tokens_seen": 5016544,
      "step": 1014,
      "train_runtime": 3505.0395,
      "train_tokens_per_second": 1431.238
    },
    {
      "epoch": 0.37333333333333335,
      "grad_norm": 0.372263103723526,
      "learning_rate": 0.00012564480471628594,
      "loss": 1.4011,
      "num_input_tokens_seen": 5021040,
      "step": 1015,
      "train_runtime": 3508.0542,
      "train_tokens_per_second": 1431.289
    },
    {
      "epoch": 0.37370114942528737,
      "grad_norm": 0.34841859340667725,
      "learning_rate": 0.0001255711127487104,
      "loss": 1.2685,
      "num_input_tokens_seen": 5025776,
      "step": 1016,
      "train_runtime": 3511.2199,
      "train_tokens_per_second": 1431.348
    },
    {
      "epoch": 0.3740689655172414,
      "grad_norm": 0.3994417190551758,
      "learning_rate": 0.00012549742078113487,
      "loss": 1.1669,
      "num_input_tokens_seen": 5029760,
      "step": 1017,
      "train_runtime": 3513.9952,
      "train_tokens_per_second": 1431.351
    },
    {
      "epoch": 0.3744367816091954,
      "grad_norm": 0.39053210616111755,
      "learning_rate": 0.00012542372881355933,
      "loss": 1.2908,
      "num_input_tokens_seen": 5035312,
      "step": 1018,
      "train_runtime": 3517.6172,
      "train_tokens_per_second": 1431.455
    },
    {
      "epoch": 0.37480459770114943,
      "grad_norm": 0.33367887139320374,
      "learning_rate": 0.0001253500368459838,
      "loss": 1.1579,
      "num_input_tokens_seen": 5040048,
      "step": 1019,
      "train_runtime": 3520.7724,
      "train_tokens_per_second": 1431.518
    },
    {
      "epoch": 0.37517241379310345,
      "grad_norm": 0.40413036942481995,
      "learning_rate": 0.00012527634487840826,
      "loss": 1.4021,
      "num_input_tokens_seen": 5044320,
      "step": 1020,
      "train_runtime": 3523.6925,
      "train_tokens_per_second": 1431.544
    },
    {
      "epoch": 0.3755402298850575,
      "grad_norm": 0.35045018792152405,
      "learning_rate": 0.00012520265291083272,
      "loss": 1.1458,
      "num_input_tokens_seen": 5048512,
      "step": 1021,
      "train_runtime": 3528.7259,
      "train_tokens_per_second": 1430.69
    },
    {
      "epoch": 0.3759080459770115,
      "grad_norm": 0.34468555450439453,
      "learning_rate": 0.0001251289609432572,
      "loss": 1.2014,
      "num_input_tokens_seen": 5052768,
      "step": 1022,
      "train_runtime": 3531.5804,
      "train_tokens_per_second": 1430.738
    },
    {
      "epoch": 0.3762758620689655,
      "grad_norm": 0.32207825779914856,
      "learning_rate": 0.00012505526897568165,
      "loss": 1.2258,
      "num_input_tokens_seen": 5057568,
      "step": 1023,
      "train_runtime": 3534.7308,
      "train_tokens_per_second": 1430.821
    },
    {
      "epoch": 0.37664367816091954,
      "grad_norm": 0.3663703203201294,
      "learning_rate": 0.00012498157700810612,
      "loss": 1.3653,
      "num_input_tokens_seen": 5062048,
      "step": 1024,
      "train_runtime": 3537.7337,
      "train_tokens_per_second": 1430.873
    },
    {
      "epoch": 0.37701149425287356,
      "grad_norm": 0.34460243582725525,
      "learning_rate": 0.00012490788504053058,
      "loss": 1.2623,
      "num_input_tokens_seen": 5066080,
      "step": 1025,
      "train_runtime": 3540.4588,
      "train_tokens_per_second": 1430.911
    },
    {
      "epoch": 0.3773793103448276,
      "grad_norm": 0.3607901334762573,
      "learning_rate": 0.00012483419307295505,
      "loss": 1.0976,
      "num_input_tokens_seen": 5071136,
      "step": 1026,
      "train_runtime": 3543.8172,
      "train_tokens_per_second": 1430.981
    },
    {
      "epoch": 0.3777471264367816,
      "grad_norm": 0.36720073223114014,
      "learning_rate": 0.0001247605011053795,
      "loss": 1.2161,
      "num_input_tokens_seen": 5076224,
      "step": 1027,
      "train_runtime": 3547.181,
      "train_tokens_per_second": 1431.059
    },
    {
      "epoch": 0.3781149425287356,
      "grad_norm": 0.33790916204452515,
      "learning_rate": 0.000124686809137804,
      "loss": 1.1806,
      "num_input_tokens_seen": 5081168,
      "step": 1028,
      "train_runtime": 3550.3945,
      "train_tokens_per_second": 1431.156
    },
    {
      "epoch": 0.37848275862068964,
      "grad_norm": 0.35793933272361755,
      "learning_rate": 0.00012461311717022844,
      "loss": 1.0849,
      "num_input_tokens_seen": 5085968,
      "step": 1029,
      "train_runtime": 3553.5919,
      "train_tokens_per_second": 1431.219
    },
    {
      "epoch": 0.37885057471264366,
      "grad_norm": 0.31205907464027405,
      "learning_rate": 0.0001245394252026529,
      "loss": 1.2986,
      "num_input_tokens_seen": 5091168,
      "step": 1030,
      "train_runtime": 3557.1312,
      "train_tokens_per_second": 1431.257
    },
    {
      "epoch": 0.3792183908045977,
      "grad_norm": 0.3675948679447174,
      "learning_rate": 0.00012446573323507737,
      "loss": 1.298,
      "num_input_tokens_seen": 5096304,
      "step": 1031,
      "train_runtime": 3560.774,
      "train_tokens_per_second": 1431.235
    },
    {
      "epoch": 0.3795862068965517,
      "grad_norm": 0.3270435631275177,
      "learning_rate": 0.00012439204126750186,
      "loss": 1.1731,
      "num_input_tokens_seen": 5105872,
      "step": 1032,
      "train_runtime": 3566.6098,
      "train_tokens_per_second": 1431.576
    },
    {
      "epoch": 0.3799540229885057,
      "grad_norm": 0.3730880916118622,
      "learning_rate": 0.00012431834929992633,
      "loss": 1.2093,
      "num_input_tokens_seen": 5110688,
      "step": 1033,
      "train_runtime": 3569.8467,
      "train_tokens_per_second": 1431.627
    },
    {
      "epoch": 0.38032183908045974,
      "grad_norm": 0.3573417663574219,
      "learning_rate": 0.00012424465733235077,
      "loss": 1.1703,
      "num_input_tokens_seen": 5114576,
      "step": 1034,
      "train_runtime": 3572.6081,
      "train_tokens_per_second": 1431.608
    },
    {
      "epoch": 0.38068965517241377,
      "grad_norm": 0.3891686201095581,
      "learning_rate": 0.00012417096536477523,
      "loss": 1.3439,
      "num_input_tokens_seen": 5119088,
      "step": 1035,
      "train_runtime": 3575.6517,
      "train_tokens_per_second": 1431.652
    },
    {
      "epoch": 0.38105747126436784,
      "grad_norm": 0.3356412649154663,
      "learning_rate": 0.00012409727339719972,
      "loss": 1.1172,
      "num_input_tokens_seen": 5122720,
      "step": 1036,
      "train_runtime": 3578.2597,
      "train_tokens_per_second": 1431.623
    },
    {
      "epoch": 0.38142528735632186,
      "grad_norm": 0.3492901027202606,
      "learning_rate": 0.0001240235814296242,
      "loss": 1.1965,
      "num_input_tokens_seen": 5126288,
      "step": 1037,
      "train_runtime": 3580.7981,
      "train_tokens_per_second": 1431.605
    },
    {
      "epoch": 0.3817931034482759,
      "grad_norm": 0.3627444803714752,
      "learning_rate": 0.00012394988946204865,
      "loss": 1.2814,
      "num_input_tokens_seen": 5130912,
      "step": 1038,
      "train_runtime": 3583.9062,
      "train_tokens_per_second": 1431.654
    },
    {
      "epoch": 0.3821609195402299,
      "grad_norm": 0.4016261398792267,
      "learning_rate": 0.0001238761974944731,
      "loss": 1.3845,
      "num_input_tokens_seen": 5134592,
      "step": 1039,
      "train_runtime": 3586.5609,
      "train_tokens_per_second": 1431.62
    },
    {
      "epoch": 0.3825287356321839,
      "grad_norm": 0.3393212556838989,
      "learning_rate": 0.00012380250552689758,
      "loss": 1.2972,
      "num_input_tokens_seen": 5138016,
      "step": 1040,
      "train_runtime": 3588.9888,
      "train_tokens_per_second": 1431.605
    },
    {
      "epoch": 0.38289655172413795,
      "grad_norm": 0.36610278487205505,
      "learning_rate": 0.00012372881355932205,
      "loss": 1.1849,
      "num_input_tokens_seen": 5140912,
      "step": 1041,
      "train_runtime": 3591.2023,
      "train_tokens_per_second": 1431.53
    },
    {
      "epoch": 0.38326436781609197,
      "grad_norm": 0.3711596131324768,
      "learning_rate": 0.0001236551215917465,
      "loss": 1.2966,
      "num_input_tokens_seen": 5145600,
      "step": 1042,
      "train_runtime": 3594.5427,
      "train_tokens_per_second": 1431.503
    },
    {
      "epoch": 0.383632183908046,
      "grad_norm": 0.33255723118782043,
      "learning_rate": 0.00012358142962417095,
      "loss": 1.1432,
      "num_input_tokens_seen": 5149904,
      "step": 1043,
      "train_runtime": 3597.4764,
      "train_tokens_per_second": 1431.532
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.35760384798049927,
      "learning_rate": 0.00012350773765659544,
      "loss": 1.2683,
      "num_input_tokens_seen": 5154608,
      "step": 1044,
      "train_runtime": 3600.5796,
      "train_tokens_per_second": 1431.605
    },
    {
      "epoch": 0.38436781609195403,
      "grad_norm": 0.37174364924430847,
      "learning_rate": 0.0001234340456890199,
      "loss": 1.2198,
      "num_input_tokens_seen": 5158784,
      "step": 1045,
      "train_runtime": 3603.4169,
      "train_tokens_per_second": 1431.637
    },
    {
      "epoch": 0.38473563218390805,
      "grad_norm": 0.3620304465293884,
      "learning_rate": 0.00012336035372144437,
      "loss": 1.2691,
      "num_input_tokens_seen": 5162032,
      "step": 1046,
      "train_runtime": 3605.8137,
      "train_tokens_per_second": 1431.586
    },
    {
      "epoch": 0.38510344827586207,
      "grad_norm": 0.32890260219573975,
      "learning_rate": 0.00012328666175386884,
      "loss": 1.1396,
      "num_input_tokens_seen": 5166944,
      "step": 1047,
      "train_runtime": 3609.0055,
      "train_tokens_per_second": 1431.681
    },
    {
      "epoch": 0.3854712643678161,
      "grad_norm": 0.35326966643333435,
      "learning_rate": 0.0001232129697862933,
      "loss": 1.1628,
      "num_input_tokens_seen": 5169952,
      "step": 1048,
      "train_runtime": 3611.3092,
      "train_tokens_per_second": 1431.6
    },
    {
      "epoch": 0.3858390804597701,
      "grad_norm": 0.32441720366477966,
      "learning_rate": 0.00012313927781871777,
      "loss": 1.1884,
      "num_input_tokens_seen": 5173600,
      "step": 1049,
      "train_runtime": 3613.8808,
      "train_tokens_per_second": 1431.591
    },
    {
      "epoch": 0.38620689655172413,
      "grad_norm": 0.36308610439300537,
      "learning_rate": 0.00012306558585114223,
      "loss": 1.1562,
      "num_input_tokens_seen": 5177712,
      "step": 1050,
      "train_runtime": 3616.6444,
      "train_tokens_per_second": 1431.634
    },
    {
      "epoch": 0.38657471264367815,
      "grad_norm": 0.3232983648777008,
      "learning_rate": 0.0001229918938835667,
      "loss": 1.1819,
      "num_input_tokens_seen": 5185008,
      "step": 1051,
      "train_runtime": 3623.2532,
      "train_tokens_per_second": 1431.037
    },
    {
      "epoch": 0.3869425287356322,
      "grad_norm": 0.3701878786087036,
      "learning_rate": 0.00012291820191599116,
      "loss": 1.1784,
      "num_input_tokens_seen": 5190944,
      "step": 1052,
      "train_runtime": 3627.0151,
      "train_tokens_per_second": 1431.189
    },
    {
      "epoch": 0.3873103448275862,
      "grad_norm": 0.34583258628845215,
      "learning_rate": 0.00012284450994841563,
      "loss": 1.2214,
      "num_input_tokens_seen": 5198704,
      "step": 1053,
      "train_runtime": 3631.7461,
      "train_tokens_per_second": 1431.461
    },
    {
      "epoch": 0.3876781609195402,
      "grad_norm": 0.3805807828903198,
      "learning_rate": 0.0001227708179808401,
      "loss": 1.303,
      "num_input_tokens_seen": 5203696,
      "step": 1054,
      "train_runtime": 3635.0008,
      "train_tokens_per_second": 1431.553
    },
    {
      "epoch": 0.38804597701149424,
      "grad_norm": 0.39817777276039124,
      "learning_rate": 0.00012269712601326456,
      "loss": 1.3051,
      "num_input_tokens_seen": 5207680,
      "step": 1055,
      "train_runtime": 3637.682,
      "train_tokens_per_second": 1431.593
    },
    {
      "epoch": 0.38841379310344826,
      "grad_norm": 0.3692329227924347,
      "learning_rate": 0.00012262343404568905,
      "loss": 1.2677,
      "num_input_tokens_seen": 5212496,
      "step": 1056,
      "train_runtime": 3640.9108,
      "train_tokens_per_second": 1431.646
    },
    {
      "epoch": 0.3887816091954023,
      "grad_norm": 0.3591819703578949,
      "learning_rate": 0.00012254974207811349,
      "loss": 1.2449,
      "num_input_tokens_seen": 5217264,
      "step": 1057,
      "train_runtime": 3644.1407,
      "train_tokens_per_second": 1431.686
    },
    {
      "epoch": 0.3891494252873563,
      "grad_norm": 0.3706558346748352,
      "learning_rate": 0.00012247605011053795,
      "loss": 1.1845,
      "num_input_tokens_seen": 5220848,
      "step": 1058,
      "train_runtime": 3646.6948,
      "train_tokens_per_second": 1431.666
    },
    {
      "epoch": 0.3895172413793103,
      "grad_norm": 0.35494890809059143,
      "learning_rate": 0.00012240235814296241,
      "loss": 1.2491,
      "num_input_tokens_seen": 5224128,
      "step": 1059,
      "train_runtime": 3649.0922,
      "train_tokens_per_second": 1431.624
    },
    {
      "epoch": 0.38988505747126434,
      "grad_norm": 0.3704403340816498,
      "learning_rate": 0.0001223286661753869,
      "loss": 1.3511,
      "num_input_tokens_seen": 5227776,
      "step": 1060,
      "train_runtime": 3651.7018,
      "train_tokens_per_second": 1431.6
    },
    {
      "epoch": 0.3902528735632184,
      "grad_norm": 0.37094739079475403,
      "learning_rate": 0.00012225497420781134,
      "loss": 1.1495,
      "num_input_tokens_seen": 5231776,
      "step": 1061,
      "train_runtime": 3654.5698,
      "train_tokens_per_second": 1431.571
    },
    {
      "epoch": 0.39062068965517244,
      "grad_norm": 0.3463454246520996,
      "learning_rate": 0.0001221812822402358,
      "loss": 1.2725,
      "num_input_tokens_seen": 5236256,
      "step": 1062,
      "train_runtime": 3657.5643,
      "train_tokens_per_second": 1431.624
    },
    {
      "epoch": 0.39098850574712646,
      "grad_norm": 0.421294629573822,
      "learning_rate": 0.00012210759027266027,
      "loss": 1.4387,
      "num_input_tokens_seen": 5240720,
      "step": 1063,
      "train_runtime": 3660.5191,
      "train_tokens_per_second": 1431.688
    },
    {
      "epoch": 0.3913563218390805,
      "grad_norm": 0.3991212844848633,
      "learning_rate": 0.00012203389830508477,
      "loss": 1.3859,
      "num_input_tokens_seen": 5246624,
      "step": 1064,
      "train_runtime": 3664.3405,
      "train_tokens_per_second": 1431.806
    },
    {
      "epoch": 0.3917241379310345,
      "grad_norm": 0.3889915347099304,
      "learning_rate": 0.00012196020633750922,
      "loss": 1.2784,
      "num_input_tokens_seen": 5252688,
      "step": 1065,
      "train_runtime": 3668.2583,
      "train_tokens_per_second": 1431.93
    },
    {
      "epoch": 0.3920919540229885,
      "grad_norm": 0.3894662857055664,
      "learning_rate": 0.00012188651436993368,
      "loss": 1.3165,
      "num_input_tokens_seen": 5257264,
      "step": 1066,
      "train_runtime": 3671.291,
      "train_tokens_per_second": 1431.993
    },
    {
      "epoch": 0.39245977011494254,
      "grad_norm": 0.34579962491989136,
      "learning_rate": 0.00012181282240235813,
      "loss": 1.2072,
      "num_input_tokens_seen": 5261792,
      "step": 1067,
      "train_runtime": 3674.3245,
      "train_tokens_per_second": 1432.043
    },
    {
      "epoch": 0.39282758620689656,
      "grad_norm": 0.32758286595344543,
      "learning_rate": 0.00012173913043478263,
      "loss": 1.2067,
      "num_input_tokens_seen": 5265856,
      "step": 1068,
      "train_runtime": 3677.0774,
      "train_tokens_per_second": 1432.076
    },
    {
      "epoch": 0.3931954022988506,
      "grad_norm": 0.3422364294528961,
      "learning_rate": 0.00012166543846720708,
      "loss": 1.2137,
      "num_input_tokens_seen": 5270336,
      "step": 1069,
      "train_runtime": 3680.1711,
      "train_tokens_per_second": 1432.09
    },
    {
      "epoch": 0.3935632183908046,
      "grad_norm": 0.3267379403114319,
      "learning_rate": 0.00012159174649963154,
      "loss": 1.2792,
      "num_input_tokens_seen": 5274528,
      "step": 1070,
      "train_runtime": 3683.1396,
      "train_tokens_per_second": 1432.074
    },
    {
      "epoch": 0.3939310344827586,
      "grad_norm": 0.3592357635498047,
      "learning_rate": 0.000121518054532056,
      "loss": 1.2102,
      "num_input_tokens_seen": 5279264,
      "step": 1071,
      "train_runtime": 3686.2913,
      "train_tokens_per_second": 1432.134
    },
    {
      "epoch": 0.39429885057471264,
      "grad_norm": 0.344771146774292,
      "learning_rate": 0.00012144436256448048,
      "loss": 1.212,
      "num_input_tokens_seen": 5285984,
      "step": 1072,
      "train_runtime": 3690.6102,
      "train_tokens_per_second": 1432.279
    },
    {
      "epoch": 0.39466666666666667,
      "grad_norm": 0.34492745995521545,
      "learning_rate": 0.00012137067059690495,
      "loss": 1.1521,
      "num_input_tokens_seen": 5290384,
      "step": 1073,
      "train_runtime": 3693.5284,
      "train_tokens_per_second": 1432.339
    },
    {
      "epoch": 0.3950344827586207,
      "grad_norm": 0.30357855558395386,
      "learning_rate": 0.0001212969786293294,
      "loss": 1.152,
      "num_input_tokens_seen": 5298320,
      "step": 1074,
      "train_runtime": 3698.5493,
      "train_tokens_per_second": 1432.54
    },
    {
      "epoch": 0.3954022988505747,
      "grad_norm": 0.3603195548057556,
      "learning_rate": 0.00012122328666175387,
      "loss": 1.1997,
      "num_input_tokens_seen": 5303712,
      "step": 1075,
      "train_runtime": 3702.175,
      "train_tokens_per_second": 1432.594
    },
    {
      "epoch": 0.39577011494252873,
      "grad_norm": 0.3354949355125427,
      "learning_rate": 0.00012114959469417834,
      "loss": 1.1641,
      "num_input_tokens_seen": 5309264,
      "step": 1076,
      "train_runtime": 3705.7836,
      "train_tokens_per_second": 1432.697
    },
    {
      "epoch": 0.39613793103448275,
      "grad_norm": 0.374262273311615,
      "learning_rate": 0.00012107590272660281,
      "loss": 1.316,
      "num_input_tokens_seen": 5313056,
      "step": 1077,
      "train_runtime": 3708.3959,
      "train_tokens_per_second": 1432.71
    },
    {
      "epoch": 0.39650574712643677,
      "grad_norm": 0.34046006202697754,
      "learning_rate": 0.00012100221075902727,
      "loss": 1.2667,
      "num_input_tokens_seen": 5317088,
      "step": 1078,
      "train_runtime": 3711.2048,
      "train_tokens_per_second": 1432.712
    },
    {
      "epoch": 0.3968735632183908,
      "grad_norm": 0.4357326328754425,
      "learning_rate": 0.00012092851879145172,
      "loss": 1.3433,
      "num_input_tokens_seen": 5321136,
      "step": 1079,
      "train_runtime": 3714.0049,
      "train_tokens_per_second": 1432.722
    },
    {
      "epoch": 0.3972413793103448,
      "grad_norm": 0.3166910707950592,
      "learning_rate": 0.00012085482682387622,
      "loss": 1.1968,
      "num_input_tokens_seen": 5327184,
      "step": 1080,
      "train_runtime": 3717.9476,
      "train_tokens_per_second": 1432.829
    },
    {
      "epoch": 0.39760919540229883,
      "grad_norm": 0.3849300146102905,
      "learning_rate": 0.00012078113485630067,
      "loss": 1.173,
      "num_input_tokens_seen": 5333392,
      "step": 1081,
      "train_runtime": 3723.9439,
      "train_tokens_per_second": 1432.189
    },
    {
      "epoch": 0.39797701149425285,
      "grad_norm": 0.3317451775074005,
      "learning_rate": 0.00012070744288872513,
      "loss": 1.1804,
      "num_input_tokens_seen": 5338208,
      "step": 1082,
      "train_runtime": 3727.1008,
      "train_tokens_per_second": 1432.268
    },
    {
      "epoch": 0.3983448275862069,
      "grad_norm": 0.31019216775894165,
      "learning_rate": 0.0001206337509211496,
      "loss": 1.2013,
      "num_input_tokens_seen": 5342896,
      "step": 1083,
      "train_runtime": 3730.256,
      "train_tokens_per_second": 1432.313
    },
    {
      "epoch": 0.3987126436781609,
      "grad_norm": 0.3194737136363983,
      "learning_rate": 0.00012056005895357408,
      "loss": 1.2644,
      "num_input_tokens_seen": 5347552,
      "step": 1084,
      "train_runtime": 3733.2594,
      "train_tokens_per_second": 1432.408
    },
    {
      "epoch": 0.3990804597701149,
      "grad_norm": 0.35167330503463745,
      "learning_rate": 0.00012048636698599853,
      "loss": 1.2333,
      "num_input_tokens_seen": 5352048,
      "step": 1085,
      "train_runtime": 3736.1995,
      "train_tokens_per_second": 1432.485
    },
    {
      "epoch": 0.399448275862069,
      "grad_norm": 0.3975164294242859,
      "learning_rate": 0.00012041267501842299,
      "loss": 1.1807,
      "num_input_tokens_seen": 5356624,
      "step": 1086,
      "train_runtime": 3739.2608,
      "train_tokens_per_second": 1432.536
    },
    {
      "epoch": 0.399816091954023,
      "grad_norm": 0.33640673756599426,
      "learning_rate": 0.00012033898305084746,
      "loss": 1.0823,
      "num_input_tokens_seen": 5360752,
      "step": 1087,
      "train_runtime": 3742.1242,
      "train_tokens_per_second": 1432.543
    },
    {
      "epoch": 0.40018390804597703,
      "grad_norm": 0.3588925004005432,
      "learning_rate": 0.00012026529108327194,
      "loss": 1.1697,
      "num_input_tokens_seen": 5364016,
      "step": 1088,
      "train_runtime": 3744.4739,
      "train_tokens_per_second": 1432.515
    },
    {
      "epoch": 0.40018390804597703,
      "eval_loss": 1.2428745031356812,
      "eval_runtime": 15.1969,
      "eval_samples_per_second": 65.803,
      "eval_steps_per_second": 4.146,
      "num_input_tokens_seen": 5364016,
      "step": 1088
    },
    {
      "epoch": 0.40055172413793105,
      "grad_norm": 0.33775821328163147,
      "learning_rate": 0.0001201915991156964,
      "loss": 1.2168,
      "num_input_tokens_seen": 5367936,
      "step": 1089,
      "train_runtime": 3762.4401,
      "train_tokens_per_second": 1426.717
    },
    {
      "epoch": 0.4009195402298851,
      "grad_norm": 0.3454897999763489,
      "learning_rate": 0.00012011790714812085,
      "loss": 1.3041,
      "num_input_tokens_seen": 5378176,
      "step": 1090,
      "train_runtime": 3768.5169,
      "train_tokens_per_second": 1427.133
    },
    {
      "epoch": 0.4012873563218391,
      "grad_norm": 0.34462302923202515,
      "learning_rate": 0.00012004421518054532,
      "loss": 1.2234,
      "num_input_tokens_seen": 5381872,
      "step": 1091,
      "train_runtime": 3771.0969,
      "train_tokens_per_second": 1427.137
    },
    {
      "epoch": 0.4016551724137931,
      "grad_norm": 0.3986961841583252,
      "learning_rate": 0.0001199705232129698,
      "loss": 1.3212,
      "num_input_tokens_seen": 5386576,
      "step": 1092,
      "train_runtime": 3774.1991,
      "train_tokens_per_second": 1427.21
    },
    {
      "epoch": 0.40202298850574714,
      "grad_norm": 0.3640199601650238,
      "learning_rate": 0.00011989683124539426,
      "loss": 1.2773,
      "num_input_tokens_seen": 5391744,
      "step": 1093,
      "train_runtime": 3777.5051,
      "train_tokens_per_second": 1427.329
    },
    {
      "epoch": 0.40239080459770116,
      "grad_norm": 0.3769136369228363,
      "learning_rate": 0.00011982313927781872,
      "loss": 1.241,
      "num_input_tokens_seen": 5396416,
      "step": 1094,
      "train_runtime": 3780.5673,
      "train_tokens_per_second": 1427.409
    },
    {
      "epoch": 0.4027586206896552,
      "grad_norm": 0.3686259388923645,
      "learning_rate": 0.00011974944731024318,
      "loss": 1.313,
      "num_input_tokens_seen": 5400320,
      "step": 1095,
      "train_runtime": 3783.3186,
      "train_tokens_per_second": 1427.403
    },
    {
      "epoch": 0.4031264367816092,
      "grad_norm": 0.3390229046344757,
      "learning_rate": 0.00011967575534266767,
      "loss": 1.1134,
      "num_input_tokens_seen": 5405328,
      "step": 1096,
      "train_runtime": 3786.6879,
      "train_tokens_per_second": 1427.455
    },
    {
      "epoch": 0.4034942528735632,
      "grad_norm": 0.3467003107070923,
      "learning_rate": 0.00011960206337509212,
      "loss": 1.1132,
      "num_input_tokens_seen": 5410832,
      "step": 1097,
      "train_runtime": 3790.2789,
      "train_tokens_per_second": 1427.555
    },
    {
      "epoch": 0.40386206896551724,
      "grad_norm": 0.3521224558353424,
      "learning_rate": 0.00011952837140751658,
      "loss": 1.2108,
      "num_input_tokens_seen": 5417024,
      "step": 1098,
      "train_runtime": 3794.3081,
      "train_tokens_per_second": 1427.671
    },
    {
      "epoch": 0.40422988505747126,
      "grad_norm": 0.3469099700450897,
      "learning_rate": 0.00011945467943994105,
      "loss": 1.3315,
      "num_input_tokens_seen": 5420608,
      "step": 1099,
      "train_runtime": 3796.8821,
      "train_tokens_per_second": 1427.647
    },
    {
      "epoch": 0.4045977011494253,
      "grad_norm": 0.35697469115257263,
      "learning_rate": 0.00011938098747236553,
      "loss": 1.2473,
      "num_input_tokens_seen": 5425936,
      "step": 1100,
      "train_runtime": 3800.4101,
      "train_tokens_per_second": 1427.724
    },
    {
      "epoch": 0.4049655172413793,
      "grad_norm": 0.32476937770843506,
      "learning_rate": 0.00011930729550478998,
      "loss": 1.2057,
      "num_input_tokens_seen": 5430400,
      "step": 1101,
      "train_runtime": 3803.349,
      "train_tokens_per_second": 1427.794
    },
    {
      "epoch": 0.4053333333333333,
      "grad_norm": 0.34049010276794434,
      "learning_rate": 0.00011923360353721444,
      "loss": 1.3402,
      "num_input_tokens_seen": 5434176,
      "step": 1102,
      "train_runtime": 3805.9782,
      "train_tokens_per_second": 1427.8
    },
    {
      "epoch": 0.40570114942528734,
      "grad_norm": 0.3352499008178711,
      "learning_rate": 0.00011915991156963891,
      "loss": 0.9929,
      "num_input_tokens_seen": 5438304,
      "step": 1103,
      "train_runtime": 3808.7535,
      "train_tokens_per_second": 1427.844
    },
    {
      "epoch": 0.40606896551724136,
      "grad_norm": 0.3618566691875458,
      "learning_rate": 0.00011908621960206339,
      "loss": 1.2383,
      "num_input_tokens_seen": 5442208,
      "step": 1104,
      "train_runtime": 3811.4145,
      "train_tokens_per_second": 1427.871
    },
    {
      "epoch": 0.4064367816091954,
      "grad_norm": 0.36079367995262146,
      "learning_rate": 0.00011901252763448785,
      "loss": 1.367,
      "num_input_tokens_seen": 5447200,
      "step": 1105,
      "train_runtime": 3814.6324,
      "train_tokens_per_second": 1427.975
    },
    {
      "epoch": 0.4068045977011494,
      "grad_norm": 0.3755130171775818,
      "learning_rate": 0.0001189388356669123,
      "loss": 1.1368,
      "num_input_tokens_seen": 5451536,
      "step": 1106,
      "train_runtime": 3817.6161,
      "train_tokens_per_second": 1427.995
    },
    {
      "epoch": 0.4071724137931034,
      "grad_norm": 0.3543834388256073,
      "learning_rate": 0.00011886514369933677,
      "loss": 1.2331,
      "num_input_tokens_seen": 5455488,
      "step": 1107,
      "train_runtime": 3820.3957,
      "train_tokens_per_second": 1427.99
    },
    {
      "epoch": 0.40754022988505745,
      "grad_norm": 0.3687877357006073,
      "learning_rate": 0.00011879145173176125,
      "loss": 1.3265,
      "num_input_tokens_seen": 5459648,
      "step": 1108,
      "train_runtime": 3823.2891,
      "train_tokens_per_second": 1427.998
    },
    {
      "epoch": 0.40790804597701147,
      "grad_norm": 0.3760135769844055,
      "learning_rate": 0.00011871775976418571,
      "loss": 1.3001,
      "num_input_tokens_seen": 5465184,
      "step": 1109,
      "train_runtime": 3826.8726,
      "train_tokens_per_second": 1428.107
    },
    {
      "epoch": 0.40827586206896554,
      "grad_norm": 0.3332124352455139,
      "learning_rate": 0.00011864406779661017,
      "loss": 1.1981,
      "num_input_tokens_seen": 5470768,
      "step": 1110,
      "train_runtime": 3830.573,
      "train_tokens_per_second": 1428.185
    },
    {
      "epoch": 0.40864367816091957,
      "grad_norm": 0.3272831439971924,
      "learning_rate": 0.00011857037582903463,
      "loss": 1.1764,
      "num_input_tokens_seen": 5474064,
      "step": 1111,
      "train_runtime": 3835.1153,
      "train_tokens_per_second": 1427.353
    },
    {
      "epoch": 0.4090114942528736,
      "grad_norm": 0.3936535716056824,
      "learning_rate": 0.00011849668386145912,
      "loss": 1.2311,
      "num_input_tokens_seen": 5481728,
      "step": 1112,
      "train_runtime": 3839.9344,
      "train_tokens_per_second": 1427.558
    },
    {
      "epoch": 0.4093793103448276,
      "grad_norm": 0.37798210978507996,
      "learning_rate": 0.00011842299189388357,
      "loss": 1.3299,
      "num_input_tokens_seen": 5485536,
      "step": 1113,
      "train_runtime": 3842.6111,
      "train_tokens_per_second": 1427.554
    },
    {
      "epoch": 0.40974712643678163,
      "grad_norm": 0.3663480579853058,
      "learning_rate": 0.00011834929992630803,
      "loss": 1.1728,
      "num_input_tokens_seen": 5491392,
      "step": 1114,
      "train_runtime": 3846.436,
      "train_tokens_per_second": 1427.657
    },
    {
      "epoch": 0.41011494252873565,
      "grad_norm": 0.36282312870025635,
      "learning_rate": 0.0001182756079587325,
      "loss": 1.2436,
      "num_input_tokens_seen": 5496432,
      "step": 1115,
      "train_runtime": 3849.8243,
      "train_tokens_per_second": 1427.71
    },
    {
      "epoch": 0.41048275862068967,
      "grad_norm": 0.3399622142314911,
      "learning_rate": 0.00011820191599115698,
      "loss": 1.1582,
      "num_input_tokens_seen": 5500816,
      "step": 1116,
      "train_runtime": 3852.8612,
      "train_tokens_per_second": 1427.722
    },
    {
      "epoch": 0.4108505747126437,
      "grad_norm": 0.3496538996696472,
      "learning_rate": 0.00011812822402358144,
      "loss": 1.2399,
      "num_input_tokens_seen": 5507408,
      "step": 1117,
      "train_runtime": 3857.091,
      "train_tokens_per_second": 1427.866
    },
    {
      "epoch": 0.4112183908045977,
      "grad_norm": 0.33703044056892395,
      "learning_rate": 0.0001180545320560059,
      "loss": 1.2687,
      "num_input_tokens_seen": 5511920,
      "step": 1118,
      "train_runtime": 3860.1916,
      "train_tokens_per_second": 1427.888
    },
    {
      "epoch": 0.41158620689655173,
      "grad_norm": 0.3628682494163513,
      "learning_rate": 0.00011798084008843037,
      "loss": 1.3816,
      "num_input_tokens_seen": 5515376,
      "step": 1119,
      "train_runtime": 3862.7681,
      "train_tokens_per_second": 1427.83
    },
    {
      "epoch": 0.41195402298850575,
      "grad_norm": 0.3717058300971985,
      "learning_rate": 0.00011790714812085484,
      "loss": 1.2138,
      "num_input_tokens_seen": 5520752,
      "step": 1120,
      "train_runtime": 3866.3199,
      "train_tokens_per_second": 1427.909
    },
    {
      "epoch": 0.4123218390804598,
      "grad_norm": 0.3626756966114044,
      "learning_rate": 0.0001178334561532793,
      "loss": 1.2262,
      "num_input_tokens_seen": 5524640,
      "step": 1121,
      "train_runtime": 3869.1016,
      "train_tokens_per_second": 1427.887
    },
    {
      "epoch": 0.4126896551724138,
      "grad_norm": 0.31174394488334656,
      "learning_rate": 0.00011775976418570375,
      "loss": 1.1207,
      "num_input_tokens_seen": 5529072,
      "step": 1122,
      "train_runtime": 3872.1255,
      "train_tokens_per_second": 1427.917
    },
    {
      "epoch": 0.4130574712643678,
      "grad_norm": 0.37022605538368225,
      "learning_rate": 0.00011768607221812824,
      "loss": 1.3078,
      "num_input_tokens_seen": 5537360,
      "step": 1123,
      "train_runtime": 3877.2642,
      "train_tokens_per_second": 1428.162
    },
    {
      "epoch": 0.41342528735632184,
      "grad_norm": 0.34345361590385437,
      "learning_rate": 0.0001176123802505527,
      "loss": 1.3267,
      "num_input_tokens_seen": 5543360,
      "step": 1124,
      "train_runtime": 3881.2358,
      "train_tokens_per_second": 1428.246
    },
    {
      "epoch": 0.41379310344827586,
      "grad_norm": 0.34921231865882874,
      "learning_rate": 0.00011753868828297716,
      "loss": 1.2609,
      "num_input_tokens_seen": 5551456,
      "step": 1125,
      "train_runtime": 3886.3287,
      "train_tokens_per_second": 1428.458
    },
    {
      "epoch": 0.4141609195402299,
      "grad_norm": 0.3794352114200592,
      "learning_rate": 0.00011746499631540163,
      "loss": 1.2211,
      "num_input_tokens_seen": 5556896,
      "step": 1126,
      "train_runtime": 3889.9685,
      "train_tokens_per_second": 1428.52
    },
    {
      "epoch": 0.4145287356321839,
      "grad_norm": 0.35784193873405457,
      "learning_rate": 0.0001173913043478261,
      "loss": 1.1596,
      "num_input_tokens_seen": 5561696,
      "step": 1127,
      "train_runtime": 3893.1934,
      "train_tokens_per_second": 1428.569
    },
    {
      "epoch": 0.4148965517241379,
      "grad_norm": 0.32974863052368164,
      "learning_rate": 0.00011731761238025057,
      "loss": 1.2636,
      "num_input_tokens_seen": 5566560,
      "step": 1128,
      "train_runtime": 3896.4539,
      "train_tokens_per_second": 1428.622
    },
    {
      "epoch": 0.41526436781609194,
      "grad_norm": 0.34522899985313416,
      "learning_rate": 0.00011724392041267502,
      "loss": 1.1703,
      "num_input_tokens_seen": 5571744,
      "step": 1129,
      "train_runtime": 3899.8517,
      "train_tokens_per_second": 1428.707
    },
    {
      "epoch": 0.41563218390804596,
      "grad_norm": 0.3814712166786194,
      "learning_rate": 0.00011717022844509948,
      "loss": 1.2741,
      "num_input_tokens_seen": 5575744,
      "step": 1130,
      "train_runtime": 3902.6921,
      "train_tokens_per_second": 1428.692
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.37492799758911133,
      "learning_rate": 0.00011709653647752396,
      "loss": 1.2739,
      "num_input_tokens_seen": 5581072,
      "step": 1131,
      "train_runtime": 3906.2097,
      "train_tokens_per_second": 1428.769
    },
    {
      "epoch": 0.416367816091954,
      "grad_norm": 0.35911741852760315,
      "learning_rate": 0.00011702284450994843,
      "loss": 1.1149,
      "num_input_tokens_seen": 5585680,
      "step": 1132,
      "train_runtime": 3909.3336,
      "train_tokens_per_second": 1428.806
    },
    {
      "epoch": 0.416735632183908,
      "grad_norm": 0.3211018741130829,
      "learning_rate": 0.00011694915254237289,
      "loss": 1.2267,
      "num_input_tokens_seen": 5590096,
      "step": 1133,
      "train_runtime": 3912.3798,
      "train_tokens_per_second": 1428.822
    },
    {
      "epoch": 0.41710344827586204,
      "grad_norm": 0.3716791570186615,
      "learning_rate": 0.00011687546057479734,
      "loss": 1.3457,
      "num_input_tokens_seen": 5593680,
      "step": 1134,
      "train_runtime": 3914.9356,
      "train_tokens_per_second": 1428.805
    },
    {
      "epoch": 0.4174712643678161,
      "grad_norm": 0.3171387314796448,
      "learning_rate": 0.00011680176860722182,
      "loss": 1.0486,
      "num_input_tokens_seen": 5599104,
      "step": 1135,
      "train_runtime": 3918.5086,
      "train_tokens_per_second": 1428.886
    },
    {
      "epoch": 0.41783908045977014,
      "grad_norm": 0.34421542286872864,
      "learning_rate": 0.00011672807663964629,
      "loss": 1.1311,
      "num_input_tokens_seen": 5606176,
      "step": 1136,
      "train_runtime": 3922.9796,
      "train_tokens_per_second": 1429.061
    },
    {
      "epoch": 0.41820689655172416,
      "grad_norm": 0.37732580304145813,
      "learning_rate": 0.00011665438467207075,
      "loss": 1.2176,
      "num_input_tokens_seen": 5610048,
      "step": 1137,
      "train_runtime": 3925.7245,
      "train_tokens_per_second": 1429.048
    },
    {
      "epoch": 0.4185747126436782,
      "grad_norm": 0.3768428862094879,
      "learning_rate": 0.0001165806927044952,
      "loss": 1.2225,
      "num_input_tokens_seen": 5614240,
      "step": 1138,
      "train_runtime": 3928.6497,
      "train_tokens_per_second": 1429.051
    },
    {
      "epoch": 0.4189425287356322,
      "grad_norm": 0.377395898103714,
      "learning_rate": 0.0001165070007369197,
      "loss": 1.2007,
      "num_input_tokens_seen": 5617600,
      "step": 1139,
      "train_runtime": 3931.1105,
      "train_tokens_per_second": 1429.011
    },
    {
      "epoch": 0.4193103448275862,
      "grad_norm": 0.34873583912849426,
      "learning_rate": 0.00011643330876934415,
      "loss": 1.2882,
      "num_input_tokens_seen": 5623024,
      "step": 1140,
      "train_runtime": 3934.786,
      "train_tokens_per_second": 1429.055
    },
    {
      "epoch": 0.41967816091954024,
      "grad_norm": 0.32570594549179077,
      "learning_rate": 0.00011635961680176861,
      "loss": 1.1753,
      "num_input_tokens_seen": 5626896,
      "step": 1141,
      "train_runtime": 3939.4902,
      "train_tokens_per_second": 1428.331
    },
    {
      "epoch": 0.42004597701149426,
      "grad_norm": 0.32342666387557983,
      "learning_rate": 0.00011628592483419308,
      "loss": 1.1558,
      "num_input_tokens_seen": 5633504,
      "step": 1142,
      "train_runtime": 3943.6904,
      "train_tokens_per_second": 1428.485
    },
    {
      "epoch": 0.4204137931034483,
      "grad_norm": 0.3442724347114563,
      "learning_rate": 0.00011621223286661755,
      "loss": 1.2874,
      "num_input_tokens_seen": 5637360,
      "step": 1143,
      "train_runtime": 3946.3761,
      "train_tokens_per_second": 1428.49
    },
    {
      "epoch": 0.4207816091954023,
      "grad_norm": 0.3737703859806061,
      "learning_rate": 0.00011613854089904202,
      "loss": 1.3107,
      "num_input_tokens_seen": 5641600,
      "step": 1144,
      "train_runtime": 3949.2657,
      "train_tokens_per_second": 1428.519
    },
    {
      "epoch": 0.4211494252873563,
      "grad_norm": 0.3702608644962311,
      "learning_rate": 0.00011606484893146647,
      "loss": 1.2314,
      "num_input_tokens_seen": 5647200,
      "step": 1145,
      "train_runtime": 3952.8818,
      "train_tokens_per_second": 1428.629
    },
    {
      "epoch": 0.42151724137931035,
      "grad_norm": 0.3603191375732422,
      "learning_rate": 0.00011599115696389094,
      "loss": 1.284,
      "num_input_tokens_seen": 5650688,
      "step": 1146,
      "train_runtime": 3955.4686,
      "train_tokens_per_second": 1428.576
    },
    {
      "epoch": 0.42188505747126437,
      "grad_norm": 0.3344424366950989,
      "learning_rate": 0.00011591746499631541,
      "loss": 1.4159,
      "num_input_tokens_seen": 5656464,
      "step": 1147,
      "train_runtime": 3959.227,
      "train_tokens_per_second": 1428.679
    },
    {
      "epoch": 0.4222528735632184,
      "grad_norm": 0.37944796681404114,
      "learning_rate": 0.00011584377302873988,
      "loss": 1.2642,
      "num_input_tokens_seen": 5660176,
      "step": 1148,
      "train_runtime": 3961.8482,
      "train_tokens_per_second": 1428.671
    },
    {
      "epoch": 0.4226206896551724,
      "grad_norm": 0.36027365922927856,
      "learning_rate": 0.00011577008106116434,
      "loss": 1.1157,
      "num_input_tokens_seen": 5667648,
      "step": 1149,
      "train_runtime": 3966.6452,
      "train_tokens_per_second": 1428.827
    },
    {
      "epoch": 0.42298850574712643,
      "grad_norm": 0.34349489212036133,
      "learning_rate": 0.0001156963890935888,
      "loss": 1.2262,
      "num_input_tokens_seen": 5671328,
      "step": 1150,
      "train_runtime": 3969.2535,
      "train_tokens_per_second": 1428.815
    },
    {
      "epoch": 0.42335632183908045,
      "grad_norm": 0.36051565408706665,
      "learning_rate": 0.00011562269712601327,
      "loss": 1.2615,
      "num_input_tokens_seen": 5676528,
      "step": 1151,
      "train_runtime": 3972.7327,
      "train_tokens_per_second": 1428.872
    },
    {
      "epoch": 0.4237241379310345,
      "grad_norm": 0.39821335673332214,
      "learning_rate": 0.00011554900515843774,
      "loss": 1.2289,
      "num_input_tokens_seen": 5681392,
      "step": 1152,
      "train_runtime": 3975.933,
      "train_tokens_per_second": 1428.946
    },
    {
      "epoch": 0.4240919540229885,
      "grad_norm": 0.3495023846626282,
      "learning_rate": 0.0001154753131908622,
      "loss": 1.301,
      "num_input_tokens_seen": 5686832,
      "step": 1153,
      "train_runtime": 3979.5207,
      "train_tokens_per_second": 1429.024
    },
    {
      "epoch": 0.4244597701149425,
      "grad_norm": 0.3514029383659363,
      "learning_rate": 0.00011540162122328665,
      "loss": 1.3733,
      "num_input_tokens_seen": 5690480,
      "step": 1154,
      "train_runtime": 3982.1354,
      "train_tokens_per_second": 1429.002
    },
    {
      "epoch": 0.42482758620689653,
      "grad_norm": 0.31979987025260925,
      "learning_rate": 0.00011532792925571115,
      "loss": 1.1896,
      "num_input_tokens_seen": 5694896,
      "step": 1155,
      "train_runtime": 3985.1922,
      "train_tokens_per_second": 1429.014
    },
    {
      "epoch": 0.42519540229885056,
      "grad_norm": 0.3527876138687134,
      "learning_rate": 0.0001152542372881356,
      "loss": 1.344,
      "num_input_tokens_seen": 5698896,
      "step": 1156,
      "train_runtime": 3987.9732,
      "train_tokens_per_second": 1429.021
    },
    {
      "epoch": 0.4255632183908046,
      "grad_norm": 0.4048612713813782,
      "learning_rate": 0.00011518054532056006,
      "loss": 1.278,
      "num_input_tokens_seen": 5702384,
      "step": 1157,
      "train_runtime": 3990.4991,
      "train_tokens_per_second": 1428.99
    },
    {
      "epoch": 0.4259310344827586,
      "grad_norm": 0.3892061114311218,
      "learning_rate": 0.00011510685335298453,
      "loss": 1.3818,
      "num_input_tokens_seen": 5710000,
      "step": 1158,
      "train_runtime": 3995.3618,
      "train_tokens_per_second": 1429.157
    },
    {
      "epoch": 0.4262988505747126,
      "grad_norm": 0.37503063678741455,
      "learning_rate": 0.000115033161385409,
      "loss": 1.1966,
      "num_input_tokens_seen": 5717456,
      "step": 1159,
      "train_runtime": 4000.1757,
      "train_tokens_per_second": 1429.301
    },
    {
      "epoch": 0.4266666666666667,
      "grad_norm": 0.3696037828922272,
      "learning_rate": 0.00011495946941783347,
      "loss": 1.2336,
      "num_input_tokens_seen": 5721744,
      "step": 1160,
      "train_runtime": 4003.2383,
      "train_tokens_per_second": 1429.279
    },
    {
      "epoch": 0.4270344827586207,
      "grad_norm": 0.36025142669677734,
      "learning_rate": 0.00011488577745025792,
      "loss": 1.2687,
      "num_input_tokens_seen": 5725504,
      "step": 1161,
      "train_runtime": 4005.9685,
      "train_tokens_per_second": 1429.243
    },
    {
      "epoch": 0.42740229885057474,
      "grad_norm": 0.37280407547950745,
      "learning_rate": 0.00011481208548268239,
      "loss": 1.3401,
      "num_input_tokens_seen": 5729856,
      "step": 1162,
      "train_runtime": 4009.0174,
      "train_tokens_per_second": 1429.242
    },
    {
      "epoch": 0.42777011494252876,
      "grad_norm": 0.5364356637001038,
      "learning_rate": 0.00011473839351510686,
      "loss": 1.2517,
      "num_input_tokens_seen": 5734672,
      "step": 1163,
      "train_runtime": 4012.2114,
      "train_tokens_per_second": 1429.305
    },
    {
      "epoch": 0.4281379310344828,
      "grad_norm": 0.33989518880844116,
      "learning_rate": 0.00011466470154753133,
      "loss": 1.077,
      "num_input_tokens_seen": 5738016,
      "step": 1164,
      "train_runtime": 4014.721,
      "train_tokens_per_second": 1429.244
    },
    {
      "epoch": 0.4285057471264368,
      "grad_norm": 0.3356742262840271,
      "learning_rate": 0.0001145910095799558,
      "loss": 1.2565,
      "num_input_tokens_seen": 5742352,
      "step": 1165,
      "train_runtime": 4017.754,
      "train_tokens_per_second": 1429.244
    },
    {
      "epoch": 0.4288735632183908,
      "grad_norm": 0.33251529932022095,
      "learning_rate": 0.00011451731761238025,
      "loss": 1.1648,
      "num_input_tokens_seen": 5747456,
      "step": 1166,
      "train_runtime": 4021.137,
      "train_tokens_per_second": 1429.311
    },
    {
      "epoch": 0.42924137931034484,
      "grad_norm": 0.3364160358905792,
      "learning_rate": 0.00011444362564480474,
      "loss": 1.2334,
      "num_input_tokens_seen": 5752896,
      "step": 1167,
      "train_runtime": 4024.7353,
      "train_tokens_per_second": 1429.385
    },
    {
      "epoch": 0.42960919540229886,
      "grad_norm": 0.36181768774986267,
      "learning_rate": 0.00011436993367722919,
      "loss": 1.2421,
      "num_input_tokens_seen": 5756608,
      "step": 1168,
      "train_runtime": 4027.3286,
      "train_tokens_per_second": 1429.386
    },
    {
      "epoch": 0.4299770114942529,
      "grad_norm": 0.3630189299583435,
      "learning_rate": 0.00011429624170965365,
      "loss": 1.0806,
      "num_input_tokens_seen": 5761264,
      "step": 1169,
      "train_runtime": 4030.5433,
      "train_tokens_per_second": 1429.401
    },
    {
      "epoch": 0.4303448275862069,
      "grad_norm": 0.3320649266242981,
      "learning_rate": 0.0001142225497420781,
      "loss": 1.2026,
      "num_input_tokens_seen": 5767712,
      "step": 1170,
      "train_runtime": 4034.7117,
      "train_tokens_per_second": 1429.523
    },
    {
      "epoch": 0.4307126436781609,
      "grad_norm": 0.34025683999061584,
      "learning_rate": 0.0001141488577745026,
      "loss": 1.1531,
      "num_input_tokens_seen": 5776464,
      "step": 1171,
      "train_runtime": 4042.2467,
      "train_tokens_per_second": 1429.023
    },
    {
      "epoch": 0.43108045977011494,
      "grad_norm": 0.362405389547348,
      "learning_rate": 0.00011407516580692705,
      "loss": 1.2082,
      "num_input_tokens_seen": 5782448,
      "step": 1172,
      "train_runtime": 4046.1714,
      "train_tokens_per_second": 1429.116
    },
    {
      "epoch": 0.43144827586206896,
      "grad_norm": 0.3381401002407074,
      "learning_rate": 0.00011400147383935151,
      "loss": 1.3184,
      "num_input_tokens_seen": 5786352,
      "step": 1173,
      "train_runtime": 4048.9633,
      "train_tokens_per_second": 1429.095
    },
    {
      "epoch": 0.431816091954023,
      "grad_norm": 0.41051506996154785,
      "learning_rate": 0.00011392778187177598,
      "loss": 1.3395,
      "num_input_tokens_seen": 5790864,
      "step": 1174,
      "train_runtime": 4052.0158,
      "train_tokens_per_second": 1429.132
    },
    {
      "epoch": 0.432183908045977,
      "grad_norm": 0.3421885371208191,
      "learning_rate": 0.00011385408990420046,
      "loss": 1.1728,
      "num_input_tokens_seen": 5798192,
      "step": 1175,
      "train_runtime": 4056.7068,
      "train_tokens_per_second": 1429.285
    },
    {
      "epoch": 0.432551724137931,
      "grad_norm": 0.3680936396121979,
      "learning_rate": 0.00011378039793662492,
      "loss": 1.3333,
      "num_input_tokens_seen": 5802368,
      "step": 1176,
      "train_runtime": 4059.6054,
      "train_tokens_per_second": 1429.294
    },
    {
      "epoch": 0.43291954022988505,
      "grad_norm": 0.397718608379364,
      "learning_rate": 0.00011370670596904937,
      "loss": 1.3145,
      "num_input_tokens_seen": 5806176,
      "step": 1177,
      "train_runtime": 4062.2614,
      "train_tokens_per_second": 1429.296
    },
    {
      "epoch": 0.43328735632183907,
      "grad_norm": 0.33635425567626953,
      "learning_rate": 0.00011363301400147384,
      "loss": 1.2446,
      "num_input_tokens_seen": 5811376,
      "step": 1178,
      "train_runtime": 4065.7383,
      "train_tokens_per_second": 1429.353
    },
    {
      "epoch": 0.4336551724137931,
      "grad_norm": 0.32526007294654846,
      "learning_rate": 0.00011355932203389832,
      "loss": 1.2973,
      "num_input_tokens_seen": 5818240,
      "step": 1179,
      "train_runtime": 4070.131,
      "train_tokens_per_second": 1429.497
    },
    {
      "epoch": 0.4340229885057471,
      "grad_norm": 0.3307048976421356,
      "learning_rate": 0.00011348563006632278,
      "loss": 1.2138,
      "num_input_tokens_seen": 5821696,
      "step": 1180,
      "train_runtime": 4072.6649,
      "train_tokens_per_second": 1429.456
    },
    {
      "epoch": 0.43439080459770113,
      "grad_norm": 0.4009729325771332,
      "learning_rate": 0.00011341193809874724,
      "loss": 1.1518,
      "num_input_tokens_seen": 5827008,
      "step": 1181,
      "train_runtime": 4076.1265,
      "train_tokens_per_second": 1429.545
    },
    {
      "epoch": 0.43475862068965515,
      "grad_norm": 0.39996057748794556,
      "learning_rate": 0.0001133382461311717,
      "loss": 1.4002,
      "num_input_tokens_seen": 5834432,
      "step": 1182,
      "train_runtime": 4080.8989,
      "train_tokens_per_second": 1429.693
    },
    {
      "epoch": 0.43512643678160917,
      "grad_norm": 0.3483218550682068,
      "learning_rate": 0.00011326455416359619,
      "loss": 1.0175,
      "num_input_tokens_seen": 5837360,
      "step": 1183,
      "train_runtime": 4083.1852,
      "train_tokens_per_second": 1429.609
    },
    {
      "epoch": 0.4354942528735632,
      "grad_norm": 0.3229388892650604,
      "learning_rate": 0.00011319086219602064,
      "loss": 1.2757,
      "num_input_tokens_seen": 5842272,
      "step": 1184,
      "train_runtime": 4086.4878,
      "train_tokens_per_second": 1429.656
    },
    {
      "epoch": 0.43586206896551727,
      "grad_norm": 0.3437669575214386,
      "learning_rate": 0.0001131171702284451,
      "loss": 1.2059,
      "num_input_tokens_seen": 5848128,
      "step": 1185,
      "train_runtime": 4090.3537,
      "train_tokens_per_second": 1429.737
    },
    {
      "epoch": 0.4362298850574713,
      "grad_norm": 0.3559442460536957,
      "learning_rate": 0.00011304347826086956,
      "loss": 1.2055,
      "num_input_tokens_seen": 5852800,
      "step": 1186,
      "train_runtime": 4093.4785,
      "train_tokens_per_second": 1429.786
    },
    {
      "epoch": 0.4365977011494253,
      "grad_norm": 0.3661964237689972,
      "learning_rate": 0.00011296978629329405,
      "loss": 1.3441,
      "num_input_tokens_seen": 5857936,
      "step": 1187,
      "train_runtime": 4096.9417,
      "train_tokens_per_second": 1429.831
    },
    {
      "epoch": 0.43696551724137933,
      "grad_norm": 0.36302024126052856,
      "learning_rate": 0.0001128960943257185,
      "loss": 1.1984,
      "num_input_tokens_seen": 5861712,
      "step": 1188,
      "train_runtime": 4099.6586,
      "train_tokens_per_second": 1429.805
    },
    {
      "epoch": 0.43733333333333335,
      "grad_norm": 0.3588753342628479,
      "learning_rate": 0.00011282240235814296,
      "loss": 1.2675,
      "num_input_tokens_seen": 5865856,
      "step": 1189,
      "train_runtime": 4102.608,
      "train_tokens_per_second": 1429.787
    },
    {
      "epoch": 0.4377011494252874,
      "grad_norm": 0.38231196999549866,
      "learning_rate": 0.00011274871039056743,
      "loss": 1.2625,
      "num_input_tokens_seen": 5870288,
      "step": 1190,
      "train_runtime": 4105.6332,
      "train_tokens_per_second": 1429.813
    },
    {
      "epoch": 0.4380689655172414,
      "grad_norm": 0.3439554274082184,
      "learning_rate": 0.0001126750184229919,
      "loss": 1.1443,
      "num_input_tokens_seen": 5875040,
      "step": 1191,
      "train_runtime": 4108.9182,
      "train_tokens_per_second": 1429.826
    },
    {
      "epoch": 0.4384367816091954,
      "grad_norm": 0.3980753719806671,
      "learning_rate": 0.00011260132645541637,
      "loss": 1.1575,
      "num_input_tokens_seen": 5879184,
      "step": 1192,
      "train_runtime": 4111.8306,
      "train_tokens_per_second": 1429.822
    },
    {
      "epoch": 0.43880459770114943,
      "grad_norm": 0.34824803471565247,
      "learning_rate": 0.00011252763448784082,
      "loss": 1.2916,
      "num_input_tokens_seen": 5883888,
      "step": 1193,
      "train_runtime": 4115.1211,
      "train_tokens_per_second": 1429.821
    },
    {
      "epoch": 0.43917241379310346,
      "grad_norm": 0.34816622734069824,
      "learning_rate": 0.00011245394252026529,
      "loss": 1.2445,
      "num_input_tokens_seen": 5888640,
      "step": 1194,
      "train_runtime": 4118.3069,
      "train_tokens_per_second": 1429.869
    },
    {
      "epoch": 0.4395402298850575,
      "grad_norm": 0.38447898626327515,
      "learning_rate": 0.00011238025055268977,
      "loss": 1.1334,
      "num_input_tokens_seen": 5893984,
      "step": 1195,
      "train_runtime": 4121.8758,
      "train_tokens_per_second": 1429.928
    },
    {
      "epoch": 0.4399080459770115,
      "grad_norm": 0.39275237917900085,
      "learning_rate": 0.00011230655858511423,
      "loss": 1.4061,
      "num_input_tokens_seen": 5897616,
      "step": 1196,
      "train_runtime": 4124.4836,
      "train_tokens_per_second": 1429.904
    },
    {
      "epoch": 0.4402758620689655,
      "grad_norm": 0.35407283902168274,
      "learning_rate": 0.0001122328666175387,
      "loss": 1.2315,
      "num_input_tokens_seen": 5902736,
      "step": 1197,
      "train_runtime": 4127.9273,
      "train_tokens_per_second": 1429.952
    },
    {
      "epoch": 0.44064367816091954,
      "grad_norm": 0.35835179686546326,
      "learning_rate": 0.00011215917464996315,
      "loss": 1.2255,
      "num_input_tokens_seen": 5906816,
      "step": 1198,
      "train_runtime": 4130.7407,
      "train_tokens_per_second": 1429.965
    },
    {
      "epoch": 0.44101149425287356,
      "grad_norm": 0.3749407231807709,
      "learning_rate": 0.00011208548268238764,
      "loss": 1.2637,
      "num_input_tokens_seen": 5910784,
      "step": 1199,
      "train_runtime": 4133.505,
      "train_tokens_per_second": 1429.969
    },
    {
      "epoch": 0.4413793103448276,
      "grad_norm": 0.3695550560951233,
      "learning_rate": 0.00011201179071481209,
      "loss": 1.2706,
      "num_input_tokens_seen": 5919504,
      "step": 1200,
      "train_runtime": 4138.8237,
      "train_tokens_per_second": 1430.238
    },
    {
      "epoch": 0.4417471264367816,
      "grad_norm": 0.33353039622306824,
      "learning_rate": 0.00011193809874723655,
      "loss": 1.2446,
      "num_input_tokens_seen": 5925264,
      "step": 1201,
      "train_runtime": 4144.715,
      "train_tokens_per_second": 1429.595
    },
    {
      "epoch": 0.4421149425287356,
      "grad_norm": 0.3443011939525604,
      "learning_rate": 0.00011186440677966102,
      "loss": 1.2483,
      "num_input_tokens_seen": 5929904,
      "step": 1202,
      "train_runtime": 4147.8556,
      "train_tokens_per_second": 1429.631
    },
    {
      "epoch": 0.44248275862068964,
      "grad_norm": 0.3490920662879944,
      "learning_rate": 0.0001117907148120855,
      "loss": 1.2847,
      "num_input_tokens_seen": 5934512,
      "step": 1203,
      "train_runtime": 4151.0616,
      "train_tokens_per_second": 1429.637
    },
    {
      "epoch": 0.44285057471264366,
      "grad_norm": 0.311960369348526,
      "learning_rate": 0.00011171702284450995,
      "loss": 1.1286,
      "num_input_tokens_seen": 5939200,
      "step": 1204,
      "train_runtime": 4154.2897,
      "train_tokens_per_second": 1429.655
    },
    {
      "epoch": 0.4432183908045977,
      "grad_norm": 0.3815508484840393,
      "learning_rate": 0.00011164333087693441,
      "loss": 1.2408,
      "num_input_tokens_seen": 5944768,
      "step": 1205,
      "train_runtime": 4157.9165,
      "train_tokens_per_second": 1429.747
    },
    {
      "epoch": 0.4435862068965517,
      "grad_norm": 0.3875328600406647,
      "learning_rate": 0.00011156963890935888,
      "loss": 1.258,
      "num_input_tokens_seen": 5948304,
      "step": 1206,
      "train_runtime": 4160.4839,
      "train_tokens_per_second": 1429.714
    },
    {
      "epoch": 0.4439540229885057,
      "grad_norm": 0.36638692021369934,
      "learning_rate": 0.00011149594694178336,
      "loss": 1.3522,
      "num_input_tokens_seen": 5951808,
      "step": 1207,
      "train_runtime": 4163.032,
      "train_tokens_per_second": 1429.681
    },
    {
      "epoch": 0.44432183908045975,
      "grad_norm": 0.3697892129421234,
      "learning_rate": 0.00011142225497420782,
      "loss": 1.1856,
      "num_input_tokens_seen": 5956128,
      "step": 1208,
      "train_runtime": 4166.0194,
      "train_tokens_per_second": 1429.693
    },
    {
      "epoch": 0.44468965517241377,
      "grad_norm": 0.3350822925567627,
      "learning_rate": 0.00011134856300663227,
      "loss": 1.2526,
      "num_input_tokens_seen": 5959904,
      "step": 1209,
      "train_runtime": 4168.6596,
      "train_tokens_per_second": 1429.693
    },
    {
      "epoch": 0.44505747126436784,
      "grad_norm": 0.3635978400707245,
      "learning_rate": 0.00011127487103905674,
      "loss": 1.1705,
      "num_input_tokens_seen": 5964608,
      "step": 1210,
      "train_runtime": 4171.7609,
      "train_tokens_per_second": 1429.758
    },
    {
      "epoch": 0.44542528735632186,
      "grad_norm": 0.3655666708946228,
      "learning_rate": 0.00011120117907148122,
      "loss": 1.2799,
      "num_input_tokens_seen": 5975136,
      "step": 1211,
      "train_runtime": 4178.305,
      "train_tokens_per_second": 1430.038
    },
    {
      "epoch": 0.4457931034482759,
      "grad_norm": 0.3381154537200928,
      "learning_rate": 0.00011112748710390568,
      "loss": 1.2709,
      "num_input_tokens_seen": 5978880,
      "step": 1212,
      "train_runtime": 4181.0461,
      "train_tokens_per_second": 1429.996
    },
    {
      "epoch": 0.4461609195402299,
      "grad_norm": 0.34990912675857544,
      "learning_rate": 0.00011105379513633015,
      "loss": 1.1869,
      "num_input_tokens_seen": 5984496,
      "step": 1213,
      "train_runtime": 4184.7765,
      "train_tokens_per_second": 1430.063
    },
    {
      "epoch": 0.4465287356321839,
      "grad_norm": 0.3995387554168701,
      "learning_rate": 0.0001109801031687546,
      "loss": 1.2954,
      "num_input_tokens_seen": 5989872,
      "step": 1214,
      "train_runtime": 4188.3403,
      "train_tokens_per_second": 1430.13
    },
    {
      "epoch": 0.44689655172413795,
      "grad_norm": 0.4230879247188568,
      "learning_rate": 0.00011090641120117909,
      "loss": 1.2431,
      "num_input_tokens_seen": 5993008,
      "step": 1215,
      "train_runtime": 4190.7061,
      "train_tokens_per_second": 1430.071
    },
    {
      "epoch": 0.44726436781609197,
      "grad_norm": 0.33954480290412903,
      "learning_rate": 0.00011083271923360354,
      "loss": 1.1937,
      "num_input_tokens_seen": 5996592,
      "step": 1216,
      "train_runtime": 4193.2649,
      "train_tokens_per_second": 1430.053
    },
    {
      "epoch": 0.447632183908046,
      "grad_norm": 0.417182981967926,
      "learning_rate": 0.000110759027266028,
      "loss": 1.2867,
      "num_input_tokens_seen": 6003120,
      "step": 1217,
      "train_runtime": 4197.5178,
      "train_tokens_per_second": 1430.159
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.382184773683548,
      "learning_rate": 0.00011068533529845247,
      "loss": 1.1929,
      "num_input_tokens_seen": 6008240,
      "step": 1218,
      "train_runtime": 4200.9443,
      "train_tokens_per_second": 1430.212
    },
    {
      "epoch": 0.44836781609195403,
      "grad_norm": 0.3704416751861572,
      "learning_rate": 0.00011061164333087695,
      "loss": 1.1453,
      "num_input_tokens_seen": 6012944,
      "step": 1219,
      "train_runtime": 4204.0888,
      "train_tokens_per_second": 1430.261
    },
    {
      "epoch": 0.44873563218390805,
      "grad_norm": 0.3858819305896759,
      "learning_rate": 0.0001105379513633014,
      "loss": 1.3237,
      "num_input_tokens_seen": 6020160,
      "step": 1220,
      "train_runtime": 4208.7156,
      "train_tokens_per_second": 1430.403
    },
    {
      "epoch": 0.44910344827586207,
      "grad_norm": 0.3485792875289917,
      "learning_rate": 0.00011046425939572586,
      "loss": 1.3126,
      "num_input_tokens_seen": 6023616,
      "step": 1221,
      "train_runtime": 4211.3003,
      "train_tokens_per_second": 1430.346
    },
    {
      "epoch": 0.4494712643678161,
      "grad_norm": 0.36032938957214355,
      "learning_rate": 0.00011039056742815033,
      "loss": 1.2516,
      "num_input_tokens_seen": 6026944,
      "step": 1222,
      "train_runtime": 4213.7157,
      "train_tokens_per_second": 1430.316
    },
    {
      "epoch": 0.4498390804597701,
      "grad_norm": 0.33639198541641235,
      "learning_rate": 0.00011031687546057481,
      "loss": 1.1945,
      "num_input_tokens_seen": 6030512,
      "step": 1223,
      "train_runtime": 4216.2695,
      "train_tokens_per_second": 1430.296
    },
    {
      "epoch": 0.45020689655172413,
      "grad_norm": 0.3770942986011505,
      "learning_rate": 0.00011024318349299927,
      "loss": 1.1821,
      "num_input_tokens_seen": 6033808,
      "step": 1224,
      "train_runtime": 4218.6686,
      "train_tokens_per_second": 1430.264
    },
    {
      "epoch": 0.45057471264367815,
      "grad_norm": 0.35923275351524353,
      "learning_rate": 0.00011016949152542372,
      "loss": 1.2543,
      "num_input_tokens_seen": 6042512,
      "step": 1225,
      "train_runtime": 4224.0798,
      "train_tokens_per_second": 1430.492
    },
    {
      "epoch": 0.4509425287356322,
      "grad_norm": 0.3461097478866577,
      "learning_rate": 0.00011009579955784819,
      "loss": 1.3266,
      "num_input_tokens_seen": 6046368,
      "step": 1226,
      "train_runtime": 4226.8599,
      "train_tokens_per_second": 1430.463
    },
    {
      "epoch": 0.4513103448275862,
      "grad_norm": 0.3811730742454529,
      "learning_rate": 0.00011002210759027267,
      "loss": 1.1791,
      "num_input_tokens_seen": 6050656,
      "step": 1227,
      "train_runtime": 4229.7985,
      "train_tokens_per_second": 1430.483
    },
    {
      "epoch": 0.4516781609195402,
      "grad_norm": 0.3463786542415619,
      "learning_rate": 0.00010994841562269713,
      "loss": 1.2673,
      "num_input_tokens_seen": 6054304,
      "step": 1228,
      "train_runtime": 4232.4393,
      "train_tokens_per_second": 1430.453
    },
    {
      "epoch": 0.45204597701149424,
      "grad_norm": 0.3345196545124054,
      "learning_rate": 0.0001098747236551216,
      "loss": 1.3401,
      "num_input_tokens_seen": 6058720,
      "step": 1229,
      "train_runtime": 4235.5187,
      "train_tokens_per_second": 1430.455
    },
    {
      "epoch": 0.45241379310344826,
      "grad_norm": 0.3485439419746399,
      "learning_rate": 0.00010980103168754605,
      "loss": 1.1461,
      "num_input_tokens_seen": 6065264,
      "step": 1230,
      "train_runtime": 4239.719,
      "train_tokens_per_second": 1430.582
    },
    {
      "epoch": 0.4527816091954023,
      "grad_norm": 0.3488841652870178,
      "learning_rate": 0.00010972733971997054,
      "loss": 1.0778,
      "num_input_tokens_seen": 6070448,
      "step": 1231,
      "train_runtime": 4245.283,
      "train_tokens_per_second": 1429.928
    },
    {
      "epoch": 0.4531494252873563,
      "grad_norm": 0.37106987833976746,
      "learning_rate": 0.00010965364775239499,
      "loss": 1.4014,
      "num_input_tokens_seen": 6074320,
      "step": 1232,
      "train_runtime": 4248.0271,
      "train_tokens_per_second": 1429.916
    },
    {
      "epoch": 0.4535172413793103,
      "grad_norm": 0.3421705365180969,
      "learning_rate": 0.00010957995578481946,
      "loss": 1.3484,
      "num_input_tokens_seen": 6082816,
      "step": 1233,
      "train_runtime": 4253.3233,
      "train_tokens_per_second": 1430.133
    },
    {
      "epoch": 0.45388505747126434,
      "grad_norm": 0.3919673264026642,
      "learning_rate": 0.00010950626381724392,
      "loss": 1.2733,
      "num_input_tokens_seen": 6086736,
      "step": 1234,
      "train_runtime": 4256.1252,
      "train_tokens_per_second": 1430.112
    },
    {
      "epoch": 0.4542528735632184,
      "grad_norm": 0.42163845896720886,
      "learning_rate": 0.0001094325718496684,
      "loss": 1.3779,
      "num_input_tokens_seen": 6090256,
      "step": 1235,
      "train_runtime": 4258.6724,
      "train_tokens_per_second": 1430.083
    },
    {
      "epoch": 0.45462068965517244,
      "grad_norm": 0.31236323714256287,
      "learning_rate": 0.00010935887988209285,
      "loss": 1.2582,
      "num_input_tokens_seen": 6096592,
      "step": 1236,
      "train_runtime": 4262.7771,
      "train_tokens_per_second": 1430.193
    },
    {
      "epoch": 0.45498850574712646,
      "grad_norm": 0.3515360951423645,
      "learning_rate": 0.00010928518791451732,
      "loss": 1.3312,
      "num_input_tokens_seen": 6100256,
      "step": 1237,
      "train_runtime": 4265.4034,
      "train_tokens_per_second": 1430.171
    },
    {
      "epoch": 0.4553563218390805,
      "grad_norm": 0.35068124532699585,
      "learning_rate": 0.00010921149594694178,
      "loss": 1.2454,
      "num_input_tokens_seen": 6105776,
      "step": 1238,
      "train_runtime": 4269.0479,
      "train_tokens_per_second": 1430.243
    },
    {
      "epoch": 0.4557241379310345,
      "grad_norm": 0.36568590998649597,
      "learning_rate": 0.00010913780397936626,
      "loss": 1.1489,
      "num_input_tokens_seen": 6112448,
      "step": 1239,
      "train_runtime": 4273.3726,
      "train_tokens_per_second": 1430.357
    },
    {
      "epoch": 0.4560919540229885,
      "grad_norm": 0.4195767939090729,
      "learning_rate": 0.00010906411201179072,
      "loss": 1.2759,
      "num_input_tokens_seen": 6116112,
      "step": 1240,
      "train_runtime": 4276.0327,
      "train_tokens_per_second": 1430.324
    },
    {
      "epoch": 0.45645977011494254,
      "grad_norm": 0.3691119849681854,
      "learning_rate": 0.00010899042004421517,
      "loss": 1.3591,
      "num_input_tokens_seen": 6122784,
      "step": 1241,
      "train_runtime": 4280.2975,
      "train_tokens_per_second": 1430.458
    },
    {
      "epoch": 0.45682758620689656,
      "grad_norm": 0.37664496898651123,
      "learning_rate": 0.00010891672807663967,
      "loss": 1.19,
      "num_input_tokens_seen": 6126368,
      "step": 1242,
      "train_runtime": 4282.9,
      "train_tokens_per_second": 1430.425
    },
    {
      "epoch": 0.4571954022988506,
      "grad_norm": 0.3803861141204834,
      "learning_rate": 0.00010884303610906412,
      "loss": 1.4098,
      "num_input_tokens_seen": 6130176,
      "step": 1243,
      "train_runtime": 4285.6065,
      "train_tokens_per_second": 1430.41
    },
    {
      "epoch": 0.4575632183908046,
      "grad_norm": 0.3393901586532593,
      "learning_rate": 0.00010876934414148858,
      "loss": 1.1721,
      "num_input_tokens_seen": 6135680,
      "step": 1244,
      "train_runtime": 4289.2725,
      "train_tokens_per_second": 1430.471
    },
    {
      "epoch": 0.4579310344827586,
      "grad_norm": 0.3811853229999542,
      "learning_rate": 0.00010869565217391305,
      "loss": 1.3572,
      "num_input_tokens_seen": 6142368,
      "step": 1245,
      "train_runtime": 4293.5758,
      "train_tokens_per_second": 1430.595
    },
    {
      "epoch": 0.45829885057471265,
      "grad_norm": 0.3287411630153656,
      "learning_rate": 0.00010862196020633753,
      "loss": 1.1254,
      "num_input_tokens_seen": 6148128,
      "step": 1246,
      "train_runtime": 4297.4141,
      "train_tokens_per_second": 1430.658
    },
    {
      "epoch": 0.45866666666666667,
      "grad_norm": 0.3776017427444458,
      "learning_rate": 0.00010854826823876199,
      "loss": 1.1667,
      "num_input_tokens_seen": 6151232,
      "step": 1247,
      "train_runtime": 4299.7626,
      "train_tokens_per_second": 1430.598
    },
    {
      "epoch": 0.4590344827586207,
      "grad_norm": 0.3635268807411194,
      "learning_rate": 0.00010847457627118644,
      "loss": 1.2381,
      "num_input_tokens_seen": 6155104,
      "step": 1248,
      "train_runtime": 4302.4617,
      "train_tokens_per_second": 1430.601
    },
    {
      "epoch": 0.4594022988505747,
      "grad_norm": 0.34583720564842224,
      "learning_rate": 0.00010840088430361091,
      "loss": 1.2012,
      "num_input_tokens_seen": 6161088,
      "step": 1249,
      "train_runtime": 4306.3849,
      "train_tokens_per_second": 1430.687
    },
    {
      "epoch": 0.45977011494252873,
      "grad_norm": 0.37768661975860596,
      "learning_rate": 0.00010832719233603539,
      "loss": 1.3583,
      "num_input_tokens_seen": 6166640,
      "step": 1250,
      "train_runtime": 4310.0186,
      "train_tokens_per_second": 1430.769
    },
    {
      "epoch": 0.46013793103448275,
      "grad_norm": 0.41959473490715027,
      "learning_rate": 0.00010825350036845985,
      "loss": 1.3245,
      "num_input_tokens_seen": 6172032,
      "step": 1251,
      "train_runtime": 4313.61,
      "train_tokens_per_second": 1430.828
    },
    {
      "epoch": 0.46050574712643677,
      "grad_norm": 0.3565807640552521,
      "learning_rate": 0.00010817980840088431,
      "loss": 1.2997,
      "num_input_tokens_seen": 6179120,
      "step": 1252,
      "train_runtime": 4318.1718,
      "train_tokens_per_second": 1430.957
    },
    {
      "epoch": 0.4608735632183908,
      "grad_norm": 0.35290762782096863,
      "learning_rate": 0.00010810611643330877,
      "loss": 1.2975,
      "num_input_tokens_seen": 6182800,
      "step": 1253,
      "train_runtime": 4320.7099,
      "train_tokens_per_second": 1430.969
    },
    {
      "epoch": 0.4612413793103448,
      "grad_norm": 0.3748103678226471,
      "learning_rate": 0.00010803242446573324,
      "loss": 1.3029,
      "num_input_tokens_seen": 6187296,
      "step": 1254,
      "train_runtime": 4323.8035,
      "train_tokens_per_second": 1430.985
    },
    {
      "epoch": 0.46160919540229883,
      "grad_norm": 0.36973637342453003,
      "learning_rate": 0.00010795873249815771,
      "loss": 1.2859,
      "num_input_tokens_seen": 6193424,
      "step": 1255,
      "train_runtime": 4327.8263,
      "train_tokens_per_second": 1431.07
    },
    {
      "epoch": 0.46197701149425285,
      "grad_norm": 0.3441025912761688,
      "learning_rate": 0.00010788504053058217,
      "loss": 1.278,
      "num_input_tokens_seen": 6197440,
      "step": 1256,
      "train_runtime": 4330.6671,
      "train_tokens_per_second": 1431.059
    },
    {
      "epoch": 0.4623448275862069,
      "grad_norm": 0.3740573823451996,
      "learning_rate": 0.00010781134856300663,
      "loss": 1.1992,
      "num_input_tokens_seen": 6201168,
      "step": 1257,
      "train_runtime": 4333.3016,
      "train_tokens_per_second": 1431.049
    },
    {
      "epoch": 0.4627126436781609,
      "grad_norm": 0.36677950620651245,
      "learning_rate": 0.00010773765659543112,
      "loss": 1.2677,
      "num_input_tokens_seen": 6204752,
      "step": 1258,
      "train_runtime": 4335.8207,
      "train_tokens_per_second": 1431.044
    },
    {
      "epoch": 0.4630804597701149,
      "grad_norm": 0.33686837553977966,
      "learning_rate": 0.00010766396462785557,
      "loss": 1.1555,
      "num_input_tokens_seen": 6208272,
      "step": 1259,
      "train_runtime": 4338.3877,
      "train_tokens_per_second": 1431.009
    },
    {
      "epoch": 0.463448275862069,
      "grad_norm": 0.3118646740913391,
      "learning_rate": 0.00010759027266028003,
      "loss": 1.2825,
      "num_input_tokens_seen": 6212320,
      "step": 1260,
      "train_runtime": 4341.2198,
      "train_tokens_per_second": 1431.008
    },
    {
      "epoch": 0.463816091954023,
      "grad_norm": 0.3848208487033844,
      "learning_rate": 0.0001075165806927045,
      "loss": 1.3584,
      "num_input_tokens_seen": 6221824,
      "step": 1261,
      "train_runtime": 4349.5918,
      "train_tokens_per_second": 1430.439
    },
    {
      "epoch": 0.46418390804597703,
      "grad_norm": 0.35913828015327454,
      "learning_rate": 0.00010744288872512898,
      "loss": 1.2013,
      "num_input_tokens_seen": 6227248,
      "step": 1262,
      "train_runtime": 4353.2599,
      "train_tokens_per_second": 1430.479
    },
    {
      "epoch": 0.46455172413793105,
      "grad_norm": 0.41634610295295715,
      "learning_rate": 0.00010736919675755344,
      "loss": 1.1727,
      "num_input_tokens_seen": 6233472,
      "step": 1263,
      "train_runtime": 4357.3071,
      "train_tokens_per_second": 1430.579
    },
    {
      "epoch": 0.4649195402298851,
      "grad_norm": 0.38134899735450745,
      "learning_rate": 0.00010729550478997789,
      "loss": 1.2966,
      "num_input_tokens_seen": 6237136,
      "step": 1264,
      "train_runtime": 4359.9793,
      "train_tokens_per_second": 1430.543
    },
    {
      "epoch": 0.4652873563218391,
      "grad_norm": 0.38777005672454834,
      "learning_rate": 0.00010722181282240236,
      "loss": 1.2314,
      "num_input_tokens_seen": 6241104,
      "step": 1265,
      "train_runtime": 4362.7074,
      "train_tokens_per_second": 1430.558
    },
    {
      "epoch": 0.4656551724137931,
      "grad_norm": 0.38832637667655945,
      "learning_rate": 0.00010714812085482684,
      "loss": 1.2465,
      "num_input_tokens_seen": 6247824,
      "step": 1266,
      "train_runtime": 4367.0115,
      "train_tokens_per_second": 1430.686
    },
    {
      "epoch": 0.46602298850574714,
      "grad_norm": 0.36237862706184387,
      "learning_rate": 0.0001070744288872513,
      "loss": 1.0657,
      "num_input_tokens_seen": 6252944,
      "step": 1267,
      "train_runtime": 4370.469,
      "train_tokens_per_second": 1430.726
    },
    {
      "epoch": 0.46639080459770116,
      "grad_norm": 0.390701562166214,
      "learning_rate": 0.00010700073691967577,
      "loss": 1.4827,
      "num_input_tokens_seen": 6256736,
      "step": 1268,
      "train_runtime": 4373.2168,
      "train_tokens_per_second": 1430.694
    },
    {
      "epoch": 0.4667586206896552,
      "grad_norm": 0.36114194989204407,
      "learning_rate": 0.00010692704495210022,
      "loss": 1.2605,
      "num_input_tokens_seen": 6260432,
      "step": 1269,
      "train_runtime": 4375.8548,
      "train_tokens_per_second": 1430.676
    },
    {
      "epoch": 0.4671264367816092,
      "grad_norm": 0.40937262773513794,
      "learning_rate": 0.0001068533529845247,
      "loss": 1.1864,
      "num_input_tokens_seen": 6264000,
      "step": 1270,
      "train_runtime": 4378.4245,
      "train_tokens_per_second": 1430.652
    },
    {
      "epoch": 0.4674942528735632,
      "grad_norm": 0.33223870396614075,
      "learning_rate": 0.00010677966101694916,
      "loss": 1.1763,
      "num_input_tokens_seen": 6267760,
      "step": 1271,
      "train_runtime": 4381.1289,
      "train_tokens_per_second": 1430.627
    },
    {
      "epoch": 0.46786206896551724,
      "grad_norm": 0.3518359363079071,
      "learning_rate": 0.00010670596904937362,
      "loss": 1.2233,
      "num_input_tokens_seen": 6273456,
      "step": 1272,
      "train_runtime": 4384.906,
      "train_tokens_per_second": 1430.693
    },
    {
      "epoch": 0.46822988505747126,
      "grad_norm": 0.3486008048057556,
      "learning_rate": 0.00010663227708179808,
      "loss": 1.1921,
      "num_input_tokens_seen": 6277344,
      "step": 1273,
      "train_runtime": 4387.6846,
      "train_tokens_per_second": 1430.673
    },
    {
      "epoch": 0.4685977011494253,
      "grad_norm": 0.3802790939807892,
      "learning_rate": 0.00010655858511422257,
      "loss": 1.2854,
      "num_input_tokens_seen": 6281616,
      "step": 1274,
      "train_runtime": 4390.615,
      "train_tokens_per_second": 1430.692
    },
    {
      "epoch": 0.4689655172413793,
      "grad_norm": 0.3698502779006958,
      "learning_rate": 0.00010648489314664702,
      "loss": 1.263,
      "num_input_tokens_seen": 6285152,
      "step": 1275,
      "train_runtime": 4393.2049,
      "train_tokens_per_second": 1430.653
    },
    {
      "epoch": 0.4693333333333333,
      "grad_norm": 0.397744745016098,
      "learning_rate": 0.00010641120117907148,
      "loss": 1.2312,
      "num_input_tokens_seen": 6288624,
      "step": 1276,
      "train_runtime": 4395.7017,
      "train_tokens_per_second": 1430.63
    },
    {
      "epoch": 0.46970114942528735,
      "grad_norm": 0.3773235082626343,
      "learning_rate": 0.00010633750921149595,
      "loss": 1.0731,
      "num_input_tokens_seen": 6292848,
      "step": 1277,
      "train_runtime": 4398.6347,
      "train_tokens_per_second": 1430.637
    },
    {
      "epoch": 0.47006896551724137,
      "grad_norm": 0.33362874388694763,
      "learning_rate": 0.00010626381724392043,
      "loss": 1.2441,
      "num_input_tokens_seen": 6298832,
      "step": 1278,
      "train_runtime": 4402.6273,
      "train_tokens_per_second": 1430.698
    },
    {
      "epoch": 0.4704367816091954,
      "grad_norm": 0.36299988627433777,
      "learning_rate": 0.00010619012527634489,
      "loss": 1.1648,
      "num_input_tokens_seen": 6302304,
      "step": 1279,
      "train_runtime": 4405.1823,
      "train_tokens_per_second": 1430.657
    },
    {
      "epoch": 0.4708045977011494,
      "grad_norm": 0.4033413827419281,
      "learning_rate": 0.00010611643330876934,
      "loss": 1.334,
      "num_input_tokens_seen": 6308736,
      "step": 1280,
      "train_runtime": 4409.3679,
      "train_tokens_per_second": 1430.757
    },
    {
      "epoch": 0.47117241379310343,
      "grad_norm": 0.3534686267375946,
      "learning_rate": 0.00010604274134119381,
      "loss": 1.2313,
      "num_input_tokens_seen": 6313872,
      "step": 1281,
      "train_runtime": 4412.8017,
      "train_tokens_per_second": 1430.808
    },
    {
      "epoch": 0.47154022988505745,
      "grad_norm": 0.3610369861125946,
      "learning_rate": 0.00010596904937361829,
      "loss": 1.4434,
      "num_input_tokens_seen": 6318544,
      "step": 1282,
      "train_runtime": 4416.0242,
      "train_tokens_per_second": 1430.822
    },
    {
      "epoch": 0.47190804597701147,
      "grad_norm": 0.3407459855079651,
      "learning_rate": 0.00010589535740604275,
      "loss": 1.2618,
      "num_input_tokens_seen": 6323072,
      "step": 1283,
      "train_runtime": 4419.1162,
      "train_tokens_per_second": 1430.845
    },
    {
      "epoch": 0.4722758620689655,
      "grad_norm": 0.37021225690841675,
      "learning_rate": 0.00010582166543846722,
      "loss": 1.2868,
      "num_input_tokens_seen": 6328976,
      "step": 1284,
      "train_runtime": 4422.9672,
      "train_tokens_per_second": 1430.934
    },
    {
      "epoch": 0.47264367816091957,
      "grad_norm": 0.406423956155777,
      "learning_rate": 0.00010574797347089167,
      "loss": 1.1297,
      "num_input_tokens_seen": 6333776,
      "step": 1285,
      "train_runtime": 4426.2085,
      "train_tokens_per_second": 1430.971
    },
    {
      "epoch": 0.4730114942528736,
      "grad_norm": 0.3668529987335205,
      "learning_rate": 0.00010567428150331616,
      "loss": 1.2369,
      "num_input_tokens_seen": 6336976,
      "step": 1286,
      "train_runtime": 4428.6273,
      "train_tokens_per_second": 1430.912
    },
    {
      "epoch": 0.4733793103448276,
      "grad_norm": 0.4330931603908539,
      "learning_rate": 0.00010560058953574061,
      "loss": 1.4164,
      "num_input_tokens_seen": 6340608,
      "step": 1287,
      "train_runtime": 4431.2583,
      "train_tokens_per_second": 1430.882
    },
    {
      "epoch": 0.47374712643678163,
      "grad_norm": 0.36734557151794434,
      "learning_rate": 0.00010552689756816508,
      "loss": 1.2767,
      "num_input_tokens_seen": 6344992,
      "step": 1288,
      "train_runtime": 4434.2433,
      "train_tokens_per_second": 1430.907
    },
    {
      "epoch": 0.47411494252873565,
      "grad_norm": 0.34771713614463806,
      "learning_rate": 0.00010545320560058953,
      "loss": 1.1343,
      "num_input_tokens_seen": 6349456,
      "step": 1289,
      "train_runtime": 4437.326,
      "train_tokens_per_second": 1430.919
    },
    {
      "epoch": 0.47448275862068967,
      "grad_norm": 0.33599287271499634,
      "learning_rate": 0.00010537951363301402,
      "loss": 1.0908,
      "num_input_tokens_seen": 6353600,
      "step": 1290,
      "train_runtime": 4440.1962,
      "train_tokens_per_second": 1430.928
    },
    {
      "epoch": 0.4748505747126437,
      "grad_norm": 0.3521447479724884,
      "learning_rate": 0.00010530582166543847,
      "loss": 1.22,
      "num_input_tokens_seen": 6359744,
      "step": 1291,
      "train_runtime": 4446.2197,
      "train_tokens_per_second": 1430.371
    },
    {
      "epoch": 0.4752183908045977,
      "grad_norm": 0.3271966278553009,
      "learning_rate": 0.00010523212969786293,
      "loss": 1.1256,
      "num_input_tokens_seen": 6363504,
      "step": 1292,
      "train_runtime": 4448.8778,
      "train_tokens_per_second": 1430.362
    },
    {
      "epoch": 0.47558620689655173,
      "grad_norm": 0.38395509123802185,
      "learning_rate": 0.0001051584377302874,
      "loss": 1.2161,
      "num_input_tokens_seen": 6366288,
      "step": 1293,
      "train_runtime": 4451.0679,
      "train_tokens_per_second": 1430.283
    },
    {
      "epoch": 0.47595402298850575,
      "grad_norm": 0.3537251353263855,
      "learning_rate": 0.00010508474576271188,
      "loss": 1.341,
      "num_input_tokens_seen": 6371728,
      "step": 1294,
      "train_runtime": 4454.7019,
      "train_tokens_per_second": 1430.338
    },
    {
      "epoch": 0.4763218390804598,
      "grad_norm": 0.3379925489425659,
      "learning_rate": 0.00010501105379513634,
      "loss": 1.2562,
      "num_input_tokens_seen": 6375328,
      "step": 1295,
      "train_runtime": 4457.2202,
      "train_tokens_per_second": 1430.337
    },
    {
      "epoch": 0.4766896551724138,
      "grad_norm": 0.3717350363731384,
      "learning_rate": 0.0001049373618275608,
      "loss": 1.1476,
      "num_input_tokens_seen": 6380464,
      "step": 1296,
      "train_runtime": 4460.6865,
      "train_tokens_per_second": 1430.377
    },
    {
      "epoch": 0.4770574712643678,
      "grad_norm": 0.3562183976173401,
      "learning_rate": 0.00010486366985998526,
      "loss": 1.2088,
      "num_input_tokens_seen": 6385040,
      "step": 1297,
      "train_runtime": 4463.8364,
      "train_tokens_per_second": 1430.393
    },
    {
      "epoch": 0.47742528735632184,
      "grad_norm": 0.3495655059814453,
      "learning_rate": 0.00010478997789240974,
      "loss": 1.0605,
      "num_input_tokens_seen": 6388704,
      "step": 1298,
      "train_runtime": 4466.3586,
      "train_tokens_per_second": 1430.406
    },
    {
      "epoch": 0.47779310344827586,
      "grad_norm": 0.3826037347316742,
      "learning_rate": 0.0001047162859248342,
      "loss": 1.3307,
      "num_input_tokens_seen": 6394224,
      "step": 1299,
      "train_runtime": 4470.0406,
      "train_tokens_per_second": 1430.462
    },
    {
      "epoch": 0.4781609195402299,
      "grad_norm": 0.360375314950943,
      "learning_rate": 0.00010464259395725867,
      "loss": 1.1802,
      "num_input_tokens_seen": 6398496,
      "step": 1300,
      "train_runtime": 4472.9385,
      "train_tokens_per_second": 1430.491
    },
    {
      "epoch": 0.4785287356321839,
      "grad_norm": 0.39885199069976807,
      "learning_rate": 0.00010456890198968312,
      "loss": 1.2222,
      "num_input_tokens_seen": 6403088,
      "step": 1301,
      "train_runtime": 4476.0208,
      "train_tokens_per_second": 1430.531
    },
    {
      "epoch": 0.4788965517241379,
      "grad_norm": 0.4102533757686615,
      "learning_rate": 0.00010449521002210761,
      "loss": 1.1182,
      "num_input_tokens_seen": 6406432,
      "step": 1302,
      "train_runtime": 4478.4776,
      "train_tokens_per_second": 1430.493
    },
    {
      "epoch": 0.47926436781609194,
      "grad_norm": 0.3570133149623871,
      "learning_rate": 0.00010442151805453206,
      "loss": 1.301,
      "num_input_tokens_seen": 6410080,
      "step": 1303,
      "train_runtime": 4481.0902,
      "train_tokens_per_second": 1430.473
    },
    {
      "epoch": 0.47963218390804596,
      "grad_norm": 0.38149404525756836,
      "learning_rate": 0.00010434782608695653,
      "loss": 1.2595,
      "num_input_tokens_seen": 6415984,
      "step": 1304,
      "train_runtime": 4484.8975,
      "train_tokens_per_second": 1430.575
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.33905965089797974,
      "learning_rate": 0.00010427413411938098,
      "loss": 1.0739,
      "num_input_tokens_seen": 6422784,
      "step": 1305,
      "train_runtime": 4489.2181,
      "train_tokens_per_second": 1430.713
    },
    {
      "epoch": 0.480367816091954,
      "grad_norm": 0.39441245794296265,
      "learning_rate": 0.00010420044215180547,
      "loss": 1.1301,
      "num_input_tokens_seen": 6426384,
      "step": 1306,
      "train_runtime": 4491.8253,
      "train_tokens_per_second": 1430.684
    },
    {
      "epoch": 0.480735632183908,
      "grad_norm": 0.34629571437835693,
      "learning_rate": 0.00010412675018422992,
      "loss": 1.2152,
      "num_input_tokens_seen": 6432912,
      "step": 1307,
      "train_runtime": 4496.0448,
      "train_tokens_per_second": 1430.794
    },
    {
      "epoch": 0.48110344827586204,
      "grad_norm": 0.35656386613845825,
      "learning_rate": 0.00010405305821665439,
      "loss": 1.3009,
      "num_input_tokens_seen": 6441328,
      "step": 1308,
      "train_runtime": 4501.2577,
      "train_tokens_per_second": 1431.006
    },
    {
      "epoch": 0.48147126436781607,
      "grad_norm": 0.33565548062324524,
      "learning_rate": 0.00010397936624907885,
      "loss": 1.1521,
      "num_input_tokens_seen": 6446576,
      "step": 1309,
      "train_runtime": 4504.6836,
      "train_tokens_per_second": 1431.083
    },
    {
      "epoch": 0.48183908045977014,
      "grad_norm": 0.3866482377052307,
      "learning_rate": 0.00010390567428150333,
      "loss": 1.3206,
      "num_input_tokens_seen": 6453680,
      "step": 1310,
      "train_runtime": 4509.2201,
      "train_tokens_per_second": 1431.219
    },
    {
      "epoch": 0.48220689655172416,
      "grad_norm": 0.3796980082988739,
      "learning_rate": 0.0001038319823139278,
      "loss": 1.2841,
      "num_input_tokens_seen": 6457776,
      "step": 1311,
      "train_runtime": 4512.0656,
      "train_tokens_per_second": 1431.224
    },
    {
      "epoch": 0.4825747126436782,
      "grad_norm": 0.3984972834587097,
      "learning_rate": 0.00010375829034635224,
      "loss": 1.3831,
      "num_input_tokens_seen": 6461776,
      "step": 1312,
      "train_runtime": 4514.8796,
      "train_tokens_per_second": 1431.218
    },
    {
      "epoch": 0.4829425287356322,
      "grad_norm": 0.34100502729415894,
      "learning_rate": 0.00010368459837877671,
      "loss": 1.0627,
      "num_input_tokens_seen": 6465712,
      "step": 1313,
      "train_runtime": 4517.6207,
      "train_tokens_per_second": 1431.221
    },
    {
      "epoch": 0.4833103448275862,
      "grad_norm": 0.3587813079357147,
      "learning_rate": 0.00010361090641120119,
      "loss": 1.2466,
      "num_input_tokens_seen": 6469312,
      "step": 1314,
      "train_runtime": 4520.2359,
      "train_tokens_per_second": 1431.189
    },
    {
      "epoch": 0.48367816091954025,
      "grad_norm": 0.34215959906578064,
      "learning_rate": 0.00010353721444362565,
      "loss": 1.1964,
      "num_input_tokens_seen": 6473920,
      "step": 1315,
      "train_runtime": 4523.385,
      "train_tokens_per_second": 1431.211
    },
    {
      "epoch": 0.48404597701149427,
      "grad_norm": 0.3500029444694519,
      "learning_rate": 0.00010346352247605012,
      "loss": 1.1479,
      "num_input_tokens_seen": 6480384,
      "step": 1316,
      "train_runtime": 4527.4588,
      "train_tokens_per_second": 1431.351
    },
    {
      "epoch": 0.4844137931034483,
      "grad_norm": 0.3848787248134613,
      "learning_rate": 0.00010338983050847457,
      "loss": 1.2403,
      "num_input_tokens_seen": 6484512,
      "step": 1317,
      "train_runtime": 4530.4302,
      "train_tokens_per_second": 1431.324
    },
    {
      "epoch": 0.4847816091954023,
      "grad_norm": 0.3544035851955414,
      "learning_rate": 0.00010331613854089906,
      "loss": 1.2762,
      "num_input_tokens_seen": 6488672,
      "step": 1318,
      "train_runtime": 4533.3076,
      "train_tokens_per_second": 1431.333
    },
    {
      "epoch": 0.48514942528735633,
      "grad_norm": 0.35067224502563477,
      "learning_rate": 0.00010324244657332351,
      "loss": 1.354,
      "num_input_tokens_seen": 6492832,
      "step": 1319,
      "train_runtime": 4536.2511,
      "train_tokens_per_second": 1431.321
    },
    {
      "epoch": 0.48551724137931035,
      "grad_norm": 0.3909823000431061,
      "learning_rate": 0.00010316875460574798,
      "loss": 1.2899,
      "num_input_tokens_seen": 6497024,
      "step": 1320,
      "train_runtime": 4539.1968,
      "train_tokens_per_second": 1431.316
    },
    {
      "epoch": 0.48588505747126437,
      "grad_norm": 0.3614167273044586,
      "learning_rate": 0.00010309506263817244,
      "loss": 1.2836,
      "num_input_tokens_seen": 6500864,
      "step": 1321,
      "train_runtime": 4543.9491,
      "train_tokens_per_second": 1430.664
    },
    {
      "epoch": 0.4862528735632184,
      "grad_norm": 0.3644118905067444,
      "learning_rate": 0.00010302137067059692,
      "loss": 1.447,
      "num_input_tokens_seen": 6504784,
      "step": 1322,
      "train_runtime": 4546.6647,
      "train_tokens_per_second": 1430.672
    },
    {
      "epoch": 0.4866206896551724,
      "grad_norm": 0.3544730544090271,
      "learning_rate": 0.00010294767870302137,
      "loss": 1.252,
      "num_input_tokens_seen": 6508144,
      "step": 1323,
      "train_runtime": 4549.0708,
      "train_tokens_per_second": 1430.653
    },
    {
      "epoch": 0.48698850574712643,
      "grad_norm": 0.3122788071632385,
      "learning_rate": 0.00010287398673544584,
      "loss": 1.1436,
      "num_input_tokens_seen": 6513664,
      "step": 1324,
      "train_runtime": 4552.7466,
      "train_tokens_per_second": 1430.711
    },
    {
      "epoch": 0.48735632183908045,
      "grad_norm": 0.3575507700443268,
      "learning_rate": 0.0001028002947678703,
      "loss": 1.1869,
      "num_input_tokens_seen": 6518896,
      "step": 1325,
      "train_runtime": 4556.1955,
      "train_tokens_per_second": 1430.776
    },
    {
      "epoch": 0.4877241379310345,
      "grad_norm": 0.37507253885269165,
      "learning_rate": 0.00010272660280029478,
      "loss": 1.3157,
      "num_input_tokens_seen": 6522880,
      "step": 1326,
      "train_runtime": 4558.9646,
      "train_tokens_per_second": 1430.781
    },
    {
      "epoch": 0.4880919540229885,
      "grad_norm": 0.3586728274822235,
      "learning_rate": 0.00010265291083271924,
      "loss": 1.4229,
      "num_input_tokens_seen": 6526928,
      "step": 1327,
      "train_runtime": 4561.8625,
      "train_tokens_per_second": 1430.759
    },
    {
      "epoch": 0.4884597701149425,
      "grad_norm": 0.3385562002658844,
      "learning_rate": 0.0001025792188651437,
      "loss": 1.0994,
      "num_input_tokens_seen": 6531136,
      "step": 1328,
      "train_runtime": 4564.7751,
      "train_tokens_per_second": 1430.768
    },
    {
      "epoch": 0.48882758620689654,
      "grad_norm": 0.35088372230529785,
      "learning_rate": 0.00010250552689756816,
      "loss": 1.3271,
      "num_input_tokens_seen": 6535568,
      "step": 1329,
      "train_runtime": 4567.8035,
      "train_tokens_per_second": 1430.79
    },
    {
      "epoch": 0.48919540229885056,
      "grad_norm": 0.38748985528945923,
      "learning_rate": 0.00010243183492999264,
      "loss": 1.2878,
      "num_input_tokens_seen": 6540736,
      "step": 1330,
      "train_runtime": 4571.2934,
      "train_tokens_per_second": 1430.828
    },
    {
      "epoch": 0.4895632183908046,
      "grad_norm": 0.3386608362197876,
      "learning_rate": 0.0001023581429624171,
      "loss": 1.2251,
      "num_input_tokens_seen": 6546160,
      "step": 1331,
      "train_runtime": 4574.9183,
      "train_tokens_per_second": 1430.88
    },
    {
      "epoch": 0.4899310344827586,
      "grad_norm": 0.3482246994972229,
      "learning_rate": 0.00010228445099484157,
      "loss": 1.1065,
      "num_input_tokens_seen": 6550768,
      "step": 1332,
      "train_runtime": 4578.1044,
      "train_tokens_per_second": 1430.891
    },
    {
      "epoch": 0.4902988505747126,
      "grad_norm": 0.3264620006084442,
      "learning_rate": 0.00010221075902726602,
      "loss": 1.2931,
      "num_input_tokens_seen": 6555584,
      "step": 1333,
      "train_runtime": 4581.3368,
      "train_tokens_per_second": 1430.933
    },
    {
      "epoch": 0.49066666666666664,
      "grad_norm": 0.33076241612434387,
      "learning_rate": 0.00010213706705969051,
      "loss": 1.3031,
      "num_input_tokens_seen": 6559728,
      "step": 1334,
      "train_runtime": 4584.1738,
      "train_tokens_per_second": 1430.951
    },
    {
      "epoch": 0.4910344827586207,
      "grad_norm": 0.3627939224243164,
      "learning_rate": 0.00010206337509211496,
      "loss": 1.4815,
      "num_input_tokens_seen": 6563360,
      "step": 1335,
      "train_runtime": 4586.7038,
      "train_tokens_per_second": 1430.954
    },
    {
      "epoch": 0.49140229885057474,
      "grad_norm": 0.33887478709220886,
      "learning_rate": 0.00010198968312453943,
      "loss": 1.1529,
      "num_input_tokens_seen": 6567840,
      "step": 1336,
      "train_runtime": 4589.7288,
      "train_tokens_per_second": 1430.987
    },
    {
      "epoch": 0.49177011494252876,
      "grad_norm": 0.3291918933391571,
      "learning_rate": 0.00010191599115696389,
      "loss": 1.2626,
      "num_input_tokens_seen": 6573024,
      "step": 1337,
      "train_runtime": 4593.1026,
      "train_tokens_per_second": 1431.064
    },
    {
      "epoch": 0.4921379310344828,
      "grad_norm": 0.3514147400856018,
      "learning_rate": 0.00010184229918938837,
      "loss": 1.2106,
      "num_input_tokens_seen": 6576208,
      "step": 1338,
      "train_runtime": 4595.536,
      "train_tokens_per_second": 1430.999
    },
    {
      "epoch": 0.4925057471264368,
      "grad_norm": 0.34737372398376465,
      "learning_rate": 0.00010176860722181282,
      "loss": 1.3024,
      "num_input_tokens_seen": 6581408,
      "step": 1339,
      "train_runtime": 4598.9773,
      "train_tokens_per_second": 1431.059
    },
    {
      "epoch": 0.4928735632183908,
      "grad_norm": 0.35335344076156616,
      "learning_rate": 0.00010169491525423729,
      "loss": 1.2746,
      "num_input_tokens_seen": 6585552,
      "step": 1340,
      "train_runtime": 4601.8569,
      "train_tokens_per_second": 1431.064
    },
    {
      "epoch": 0.49324137931034484,
      "grad_norm": 0.36211875081062317,
      "learning_rate": 0.00010162122328666175,
      "loss": 1.1778,
      "num_input_tokens_seen": 6593152,
      "step": 1341,
      "train_runtime": 4606.5882,
      "train_tokens_per_second": 1431.244
    },
    {
      "epoch": 0.49360919540229886,
      "grad_norm": 0.3507351875305176,
      "learning_rate": 0.00010154753131908623,
      "loss": 1.2887,
      "num_input_tokens_seen": 6597936,
      "step": 1342,
      "train_runtime": 4609.8619,
      "train_tokens_per_second": 1431.265
    },
    {
      "epoch": 0.4939770114942529,
      "grad_norm": 0.3622756004333496,
      "learning_rate": 0.0001014738393515107,
      "loss": 1.3343,
      "num_input_tokens_seen": 6602608,
      "step": 1343,
      "train_runtime": 4613.0846,
      "train_tokens_per_second": 1431.278
    },
    {
      "epoch": 0.4943448275862069,
      "grad_norm": 0.3753416836261749,
      "learning_rate": 0.00010140014738393515,
      "loss": 1.1923,
      "num_input_tokens_seen": 6610240,
      "step": 1344,
      "train_runtime": 4617.8029,
      "train_tokens_per_second": 1431.469
    },
    {
      "epoch": 0.4947126436781609,
      "grad_norm": 0.3428008258342743,
      "learning_rate": 0.00010132645541635961,
      "loss": 1.2438,
      "num_input_tokens_seen": 6615056,
      "step": 1345,
      "train_runtime": 4621.0526,
      "train_tokens_per_second": 1431.504
    },
    {
      "epoch": 0.49508045977011494,
      "grad_norm": 0.3488641381263733,
      "learning_rate": 0.00010125276344878409,
      "loss": 1.2486,
      "num_input_tokens_seen": 6619712,
      "step": 1346,
      "train_runtime": 4624.2059,
      "train_tokens_per_second": 1431.535
    },
    {
      "epoch": 0.49544827586206897,
      "grad_norm": 0.38282790780067444,
      "learning_rate": 0.00010117907148120855,
      "loss": 1.3044,
      "num_input_tokens_seen": 6624128,
      "step": 1347,
      "train_runtime": 4627.2046,
      "train_tokens_per_second": 1431.562
    },
    {
      "epoch": 0.495816091954023,
      "grad_norm": 0.3703167736530304,
      "learning_rate": 0.00010110537951363302,
      "loss": 1.1798,
      "num_input_tokens_seen": 6627712,
      "step": 1348,
      "train_runtime": 4629.8505,
      "train_tokens_per_second": 1431.517
    },
    {
      "epoch": 0.496183908045977,
      "grad_norm": 0.3638324737548828,
      "learning_rate": 0.00010103168754605747,
      "loss": 1.3771,
      "num_input_tokens_seen": 6631808,
      "step": 1349,
      "train_runtime": 4632.7253,
      "train_tokens_per_second": 1431.513
    },
    {
      "epoch": 0.496551724137931,
      "grad_norm": 0.36770474910736084,
      "learning_rate": 0.00010095799557848196,
      "loss": 1.2583,
      "num_input_tokens_seen": 6635152,
      "step": 1350,
      "train_runtime": 4635.2262,
      "train_tokens_per_second": 1431.462
    },
    {
      "epoch": 0.49691954022988505,
      "grad_norm": 0.33029529452323914,
      "learning_rate": 0.00010088430361090641,
      "loss": 1.1961,
      "num_input_tokens_seen": 6639088,
      "step": 1351,
      "train_runtime": 4640.0733,
      "train_tokens_per_second": 1430.815
    },
    {
      "epoch": 0.49728735632183907,
      "grad_norm": 0.38102835416793823,
      "learning_rate": 0.00010081061164333088,
      "loss": 1.1888,
      "num_input_tokens_seen": 6643072,
      "step": 1352,
      "train_runtime": 4642.8299,
      "train_tokens_per_second": 1430.824
    },
    {
      "epoch": 0.4976551724137931,
      "grad_norm": 0.4221692681312561,
      "learning_rate": 0.00010073691967575534,
      "loss": 1.2982,
      "num_input_tokens_seen": 6646832,
      "step": 1353,
      "train_runtime": 4645.5409,
      "train_tokens_per_second": 1430.798
    },
    {
      "epoch": 0.4980229885057471,
      "grad_norm": 0.3427620530128479,
      "learning_rate": 0.00010066322770817982,
      "loss": 1.2295,
      "num_input_tokens_seen": 6653344,
      "step": 1354,
      "train_runtime": 4649.7806,
      "train_tokens_per_second": 1430.894
    },
    {
      "epoch": 0.49839080459770113,
      "grad_norm": 0.3641730546951294,
      "learning_rate": 0.00010058953574060427,
      "loss": 1.2287,
      "num_input_tokens_seen": 6658560,
      "step": 1355,
      "train_runtime": 4653.2374,
      "train_tokens_per_second": 1430.952
    },
    {
      "epoch": 0.49875862068965515,
      "grad_norm": 0.27332767844200134,
      "learning_rate": 0.00010051584377302874,
      "loss": 0.9757,
      "num_input_tokens_seen": 6671968,
      "step": 1356,
      "train_runtime": 4661.2082,
      "train_tokens_per_second": 1431.382
    },
    {
      "epoch": 0.4991264367816092,
      "grad_norm": 0.326118141412735,
      "learning_rate": 0.0001004421518054532,
      "loss": 1.1437,
      "num_input_tokens_seen": 6675136,
      "step": 1357,
      "train_runtime": 4663.5794,
      "train_tokens_per_second": 1431.333
    },
    {
      "epoch": 0.4994942528735632,
      "grad_norm": 0.33779895305633545,
      "learning_rate": 0.00010036845983787768,
      "loss": 1.155,
      "num_input_tokens_seen": 6680528,
      "step": 1358,
      "train_runtime": 4667.2464,
      "train_tokens_per_second": 1431.364
    },
    {
      "epoch": 0.4998620689655172,
      "grad_norm": 0.3250412940979004,
      "learning_rate": 0.00010029476787030215,
      "loss": 1.0596,
      "num_input_tokens_seen": 6683824,
      "step": 1359,
      "train_runtime": 4669.7042,
      "train_tokens_per_second": 1431.316
    },
    {
      "epoch": 0.5002298850574712,
      "grad_norm": 0.3165808916091919,
      "learning_rate": 0.0001002210759027266,
      "loss": 1.186,
      "num_input_tokens_seen": 6688752,
      "step": 1360,
      "train_runtime": 4673.0562,
      "train_tokens_per_second": 1431.344
    },
    {
      "epoch": 0.5002298850574712,
      "eval_loss": 1.234187126159668,
      "eval_runtime": 15.5047,
      "eval_samples_per_second": 64.497,
      "eval_steps_per_second": 4.063,
      "num_input_tokens_seen": 6688752,
      "step": 1360
    },
    {
      "epoch": 0.5005977011494253,
      "grad_norm": 0.3671439588069916,
      "learning_rate": 0.00010014738393515106,
      "loss": 1.1942,
      "num_input_tokens_seen": 6692464,
      "step": 1361,
      "train_runtime": 4691.2291,
      "train_tokens_per_second": 1426.591
    },
    {
      "epoch": 0.5009655172413793,
      "grad_norm": 0.38166704773902893,
      "learning_rate": 0.00010007369196757554,
      "loss": 1.3655,
      "num_input_tokens_seen": 6697280,
      "step": 1362,
      "train_runtime": 4694.5481,
      "train_tokens_per_second": 1426.608
    },
    {
      "epoch": 0.5013333333333333,
      "grad_norm": 0.43827009201049805,
      "learning_rate": 0.0001,
      "loss": 1.2117,
      "num_input_tokens_seen": 6700880,
      "step": 1363,
      "train_runtime": 4697.1177,
      "train_tokens_per_second": 1426.594
    },
    {
      "epoch": 0.5017011494252873,
      "grad_norm": 0.3412015438079834,
      "learning_rate": 9.992630803242447e-05,
      "loss": 1.1908,
      "num_input_tokens_seen": 6706288,
      "step": 1364,
      "train_runtime": 4700.7699,
      "train_tokens_per_second": 1426.636
    },
    {
      "epoch": 0.5020689655172413,
      "grad_norm": 0.35053691267967224,
      "learning_rate": 9.985261606484893e-05,
      "loss": 1.2373,
      "num_input_tokens_seen": 6710736,
      "step": 1365,
      "train_runtime": 4703.7755,
      "train_tokens_per_second": 1426.67
    },
    {
      "epoch": 0.5024367816091954,
      "grad_norm": 0.33263084292411804,
      "learning_rate": 9.977892409727341e-05,
      "loss": 1.3199,
      "num_input_tokens_seen": 6715456,
      "step": 1366,
      "train_runtime": 4707.0267,
      "train_tokens_per_second": 1426.687
    },
    {
      "epoch": 0.5028045977011494,
      "grad_norm": 0.3621940612792969,
      "learning_rate": 9.970523212969786e-05,
      "loss": 1.2034,
      "num_input_tokens_seen": 6719072,
      "step": 1367,
      "train_runtime": 4709.6324,
      "train_tokens_per_second": 1426.666
    },
    {
      "epoch": 0.5031724137931034,
      "grad_norm": 0.3386554718017578,
      "learning_rate": 9.963154016212234e-05,
      "loss": 1.1709,
      "num_input_tokens_seen": 6722064,
      "step": 1368,
      "train_runtime": 4711.9308,
      "train_tokens_per_second": 1426.605
    },
    {
      "epoch": 0.5035402298850574,
      "grad_norm": 0.3810170292854309,
      "learning_rate": 9.95578481945468e-05,
      "loss": 1.3903,
      "num_input_tokens_seen": 6726720,
      "step": 1369,
      "train_runtime": 4715.0819,
      "train_tokens_per_second": 1426.639
    },
    {
      "epoch": 0.5039080459770114,
      "grad_norm": 0.33149290084838867,
      "learning_rate": 9.948415622697127e-05,
      "loss": 1.2115,
      "num_input_tokens_seen": 6732768,
      "step": 1370,
      "train_runtime": 4719.0674,
      "train_tokens_per_second": 1426.716
    },
    {
      "epoch": 0.5042758620689655,
      "grad_norm": 0.3657072186470032,
      "learning_rate": 9.941046425939574e-05,
      "loss": 1.4278,
      "num_input_tokens_seen": 6736592,
      "step": 1371,
      "train_runtime": 4721.7078,
      "train_tokens_per_second": 1426.728
    },
    {
      "epoch": 0.5046436781609196,
      "grad_norm": 0.3391696512699127,
      "learning_rate": 9.93367722918202e-05,
      "loss": 1.1667,
      "num_input_tokens_seen": 6742800,
      "step": 1372,
      "train_runtime": 4725.6766,
      "train_tokens_per_second": 1426.843
    },
    {
      "epoch": 0.5050114942528736,
      "grad_norm": 0.37643301486968994,
      "learning_rate": 9.926308032424467e-05,
      "loss": 1.2547,
      "num_input_tokens_seen": 6746240,
      "step": 1373,
      "train_runtime": 4728.2489,
      "train_tokens_per_second": 1426.795
    },
    {
      "epoch": 0.5053793103448276,
      "grad_norm": 0.39713141322135925,
      "learning_rate": 9.918938835666913e-05,
      "loss": 1.3092,
      "num_input_tokens_seen": 6750016,
      "step": 1374,
      "train_runtime": 4731.0326,
      "train_tokens_per_second": 1426.753
    },
    {
      "epoch": 0.5057471264367817,
      "grad_norm": 0.3885817527770996,
      "learning_rate": 9.91156963890936e-05,
      "loss": 1.035,
      "num_input_tokens_seen": 6754576,
      "step": 1375,
      "train_runtime": 4734.1275,
      "train_tokens_per_second": 1426.784
    },
    {
      "epoch": 0.5061149425287357,
      "grad_norm": 0.37782612442970276,
      "learning_rate": 9.904200442151806e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 6761088,
      "step": 1376,
      "train_runtime": 4738.3231,
      "train_tokens_per_second": 1426.895
    },
    {
      "epoch": 0.5064827586206897,
      "grad_norm": 0.37984251976013184,
      "learning_rate": 9.896831245394253e-05,
      "loss": 1.2703,
      "num_input_tokens_seen": 6766368,
      "step": 1377,
      "train_runtime": 4741.8911,
      "train_tokens_per_second": 1426.934
    },
    {
      "epoch": 0.5068505747126437,
      "grad_norm": 0.34817638993263245,
      "learning_rate": 9.889462048636699e-05,
      "loss": 1.23,
      "num_input_tokens_seen": 6773664,
      "step": 1378,
      "train_runtime": 4746.5275,
      "train_tokens_per_second": 1427.078
    },
    {
      "epoch": 0.5072183908045977,
      "grad_norm": 0.3688991963863373,
      "learning_rate": 9.882092851879146e-05,
      "loss": 1.3542,
      "num_input_tokens_seen": 6779040,
      "step": 1379,
      "train_runtime": 4750.1649,
      "train_tokens_per_second": 1427.117
    },
    {
      "epoch": 0.5075862068965518,
      "grad_norm": 0.4449474513530731,
      "learning_rate": 9.874723655121592e-05,
      "loss": 1.2917,
      "num_input_tokens_seen": 6784096,
      "step": 1380,
      "train_runtime": 4753.5825,
      "train_tokens_per_second": 1427.154
    },
    {
      "epoch": 0.5079540229885058,
      "grad_norm": 0.3971673548221588,
      "learning_rate": 9.867354458364039e-05,
      "loss": 1.2483,
      "num_input_tokens_seen": 6788176,
      "step": 1381,
      "train_runtime": 4758.3867,
      "train_tokens_per_second": 1426.571
    },
    {
      "epoch": 0.5083218390804598,
      "grad_norm": 0.32641008496284485,
      "learning_rate": 9.859985261606486e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 6793824,
      "step": 1382,
      "train_runtime": 4762.1058,
      "train_tokens_per_second": 1426.643
    },
    {
      "epoch": 0.5086896551724138,
      "grad_norm": 0.36454838514328003,
      "learning_rate": 9.852616064848931e-05,
      "loss": 1.0799,
      "num_input_tokens_seen": 6801056,
      "step": 1383,
      "train_runtime": 4766.6858,
      "train_tokens_per_second": 1426.789
    },
    {
      "epoch": 0.5090574712643678,
      "grad_norm": 0.3873193860054016,
      "learning_rate": 9.845246868091379e-05,
      "loss": 1.3625,
      "num_input_tokens_seen": 6805920,
      "step": 1384,
      "train_runtime": 4770.0645,
      "train_tokens_per_second": 1426.798
    },
    {
      "epoch": 0.5094252873563219,
      "grad_norm": 0.3755558431148529,
      "learning_rate": 9.837877671333824e-05,
      "loss": 1.2728,
      "num_input_tokens_seen": 6810080,
      "step": 1385,
      "train_runtime": 4773.0058,
      "train_tokens_per_second": 1426.791
    },
    {
      "epoch": 0.5097931034482759,
      "grad_norm": 0.35195860266685486,
      "learning_rate": 9.830508474576272e-05,
      "loss": 1.295,
      "num_input_tokens_seen": 6816080,
      "step": 1386,
      "train_runtime": 4776.9224,
      "train_tokens_per_second": 1426.877
    },
    {
      "epoch": 0.5101609195402299,
      "grad_norm": 0.3220309913158417,
      "learning_rate": 9.823139277818719e-05,
      "loss": 1.2098,
      "num_input_tokens_seen": 6821344,
      "step": 1387,
      "train_runtime": 4780.425,
      "train_tokens_per_second": 1426.933
    },
    {
      "epoch": 0.5105287356321839,
      "grad_norm": 0.3427046835422516,
      "learning_rate": 9.815770081061165e-05,
      "loss": 1.2658,
      "num_input_tokens_seen": 6828784,
      "step": 1388,
      "train_runtime": 4785.1989,
      "train_tokens_per_second": 1427.064
    },
    {
      "epoch": 0.510896551724138,
      "grad_norm": 0.4271998703479767,
      "learning_rate": 9.808400884303612e-05,
      "loss": 1.2027,
      "num_input_tokens_seen": 6833344,
      "step": 1389,
      "train_runtime": 4788.2983,
      "train_tokens_per_second": 1427.092
    },
    {
      "epoch": 0.511264367816092,
      "grad_norm": 0.3685409724712372,
      "learning_rate": 9.801031687546058e-05,
      "loss": 1.1689,
      "num_input_tokens_seen": 6838928,
      "step": 1390,
      "train_runtime": 4792.0227,
      "train_tokens_per_second": 1427.149
    },
    {
      "epoch": 0.511632183908046,
      "grad_norm": 0.3594321608543396,
      "learning_rate": 9.793662490788505e-05,
      "loss": 1.2483,
      "num_input_tokens_seen": 6843216,
      "step": 1391,
      "train_runtime": 4794.9385,
      "train_tokens_per_second": 1427.175
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.34025707840919495,
      "learning_rate": 9.786293294030951e-05,
      "loss": 1.1582,
      "num_input_tokens_seen": 6849104,
      "step": 1392,
      "train_runtime": 4798.818,
      "train_tokens_per_second": 1427.248
    },
    {
      "epoch": 0.512367816091954,
      "grad_norm": 0.3693334758281708,
      "learning_rate": 9.778924097273398e-05,
      "loss": 1.3104,
      "num_input_tokens_seen": 6853184,
      "step": 1393,
      "train_runtime": 4801.7021,
      "train_tokens_per_second": 1427.241
    },
    {
      "epoch": 0.512735632183908,
      "grad_norm": 0.31527236104011536,
      "learning_rate": 9.771554900515844e-05,
      "loss": 1.3171,
      "num_input_tokens_seen": 6860464,
      "step": 1394,
      "train_runtime": 4806.2938,
      "train_tokens_per_second": 1427.392
    },
    {
      "epoch": 0.5131034482758621,
      "grad_norm": 0.32845792174339294,
      "learning_rate": 9.76418570375829e-05,
      "loss": 1.1775,
      "num_input_tokens_seen": 6864464,
      "step": 1395,
      "train_runtime": 4809.1549,
      "train_tokens_per_second": 1427.374
    },
    {
      "epoch": 0.5134712643678161,
      "grad_norm": 0.33912116289138794,
      "learning_rate": 9.756816507000738e-05,
      "loss": 1.1522,
      "num_input_tokens_seen": 6867952,
      "step": 1396,
      "train_runtime": 4811.7325,
      "train_tokens_per_second": 1427.335
    },
    {
      "epoch": 0.5138390804597701,
      "grad_norm": 0.3654550015926361,
      "learning_rate": 9.749447310243184e-05,
      "loss": 1.1388,
      "num_input_tokens_seen": 6872032,
      "step": 1397,
      "train_runtime": 4814.6619,
      "train_tokens_per_second": 1427.314
    },
    {
      "epoch": 0.5142068965517241,
      "grad_norm": 0.34746071696281433,
      "learning_rate": 9.742078113485631e-05,
      "loss": 1.2127,
      "num_input_tokens_seen": 6877360,
      "step": 1398,
      "train_runtime": 4818.1965,
      "train_tokens_per_second": 1427.372
    },
    {
      "epoch": 0.5145747126436782,
      "grad_norm": 0.34591183066368103,
      "learning_rate": 9.734708916728077e-05,
      "loss": 1.2691,
      "num_input_tokens_seen": 6881008,
      "step": 1399,
      "train_runtime": 4820.8572,
      "train_tokens_per_second": 1427.341
    },
    {
      "epoch": 0.5149425287356322,
      "grad_norm": 0.36108094453811646,
      "learning_rate": 9.727339719970524e-05,
      "loss": 1.2256,
      "num_input_tokens_seen": 6885152,
      "step": 1400,
      "train_runtime": 4823.6595,
      "train_tokens_per_second": 1427.371
    },
    {
      "epoch": 0.5153103448275862,
      "grad_norm": 0.3229909837245941,
      "learning_rate": 9.71997052321297e-05,
      "loss": 1.0822,
      "num_input_tokens_seen": 6888768,
      "step": 1401,
      "train_runtime": 4826.2824,
      "train_tokens_per_second": 1427.345
    },
    {
      "epoch": 0.5156781609195402,
      "grad_norm": 0.3228570222854614,
      "learning_rate": 9.712601326455417e-05,
      "loss": 1.2148,
      "num_input_tokens_seen": 6898720,
      "step": 1402,
      "train_runtime": 4832.3362,
      "train_tokens_per_second": 1427.616
    },
    {
      "epoch": 0.5160459770114942,
      "grad_norm": 0.32842516899108887,
      "learning_rate": 9.705232129697864e-05,
      "loss": 1.1983,
      "num_input_tokens_seen": 6906928,
      "step": 1403,
      "train_runtime": 4837.4113,
      "train_tokens_per_second": 1427.815
    },
    {
      "epoch": 0.5164137931034483,
      "grad_norm": 0.32673364877700806,
      "learning_rate": 9.69786293294031e-05,
      "loss": 1.1338,
      "num_input_tokens_seen": 6913296,
      "step": 1404,
      "train_runtime": 4841.5113,
      "train_tokens_per_second": 1427.921
    },
    {
      "epoch": 0.5167816091954023,
      "grad_norm": 0.33890852332115173,
      "learning_rate": 9.690493736182757e-05,
      "loss": 1.1692,
      "num_input_tokens_seen": 6917824,
      "step": 1405,
      "train_runtime": 4844.6083,
      "train_tokens_per_second": 1427.943
    },
    {
      "epoch": 0.5171494252873563,
      "grad_norm": 0.34418174624443054,
      "learning_rate": 9.683124539425203e-05,
      "loss": 1.2507,
      "num_input_tokens_seen": 6921520,
      "step": 1406,
      "train_runtime": 4847.2199,
      "train_tokens_per_second": 1427.936
    },
    {
      "epoch": 0.5175172413793103,
      "grad_norm": 0.33793869614601135,
      "learning_rate": 9.67575534266765e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 6926432,
      "step": 1407,
      "train_runtime": 4850.5251,
      "train_tokens_per_second": 1427.976
    },
    {
      "epoch": 0.5178850574712643,
      "grad_norm": 0.36337676644325256,
      "learning_rate": 9.668386145910096e-05,
      "loss": 1.3168,
      "num_input_tokens_seen": 6931856,
      "step": 1408,
      "train_runtime": 4854.0802,
      "train_tokens_per_second": 1428.047
    },
    {
      "epoch": 0.5182528735632184,
      "grad_norm": 0.34077227115631104,
      "learning_rate": 9.661016949152543e-05,
      "loss": 1.2219,
      "num_input_tokens_seen": 6939136,
      "step": 1409,
      "train_runtime": 4858.7154,
      "train_tokens_per_second": 1428.183
    },
    {
      "epoch": 0.5186206896551724,
      "grad_norm": 0.35585498809814453,
      "learning_rate": 9.653647752394989e-05,
      "loss": 1.2912,
      "num_input_tokens_seen": 6942992,
      "step": 1410,
      "train_runtime": 4861.4351,
      "train_tokens_per_second": 1428.177
    },
    {
      "epoch": 0.5189885057471264,
      "grad_norm": 0.3456760048866272,
      "learning_rate": 9.646278555637436e-05,
      "loss": 1.1418,
      "num_input_tokens_seen": 6947824,
      "step": 1411,
      "train_runtime": 4866.8124,
      "train_tokens_per_second": 1427.592
    },
    {
      "epoch": 0.5193563218390804,
      "grad_norm": 0.3090837001800537,
      "learning_rate": 9.638909358879884e-05,
      "loss": 1.1177,
      "num_input_tokens_seen": 6954096,
      "step": 1412,
      "train_runtime": 4870.843,
      "train_tokens_per_second": 1427.699
    },
    {
      "epoch": 0.5197241379310344,
      "grad_norm": 0.3940297067165375,
      "learning_rate": 9.631540162122329e-05,
      "loss": 1.167,
      "num_input_tokens_seen": 6959008,
      "step": 1413,
      "train_runtime": 4874.0793,
      "train_tokens_per_second": 1427.758
    },
    {
      "epoch": 0.5200919540229885,
      "grad_norm": 0.35202744603157043,
      "learning_rate": 9.624170965364776e-05,
      "loss": 1.2153,
      "num_input_tokens_seen": 6965280,
      "step": 1414,
      "train_runtime": 4878.043,
      "train_tokens_per_second": 1427.884
    },
    {
      "epoch": 0.5204597701149425,
      "grad_norm": 0.3534502387046814,
      "learning_rate": 9.616801768607222e-05,
      "loss": 1.2572,
      "num_input_tokens_seen": 6968352,
      "step": 1415,
      "train_runtime": 4880.3989,
      "train_tokens_per_second": 1427.824
    },
    {
      "epoch": 0.5208275862068965,
      "grad_norm": 0.36782732605934143,
      "learning_rate": 9.60943257184967e-05,
      "loss": 1.1387,
      "num_input_tokens_seen": 6972256,
      "step": 1416,
      "train_runtime": 4883.1046,
      "train_tokens_per_second": 1427.833
    },
    {
      "epoch": 0.5211954022988505,
      "grad_norm": 0.38094982504844666,
      "learning_rate": 9.602063375092115e-05,
      "loss": 1.1373,
      "num_input_tokens_seen": 6979472,
      "step": 1417,
      "train_runtime": 4887.6965,
      "train_tokens_per_second": 1427.968
    },
    {
      "epoch": 0.5215632183908046,
      "grad_norm": 0.3493528962135315,
      "learning_rate": 9.594694178334562e-05,
      "loss": 1.153,
      "num_input_tokens_seen": 6985504,
      "step": 1418,
      "train_runtime": 4891.6238,
      "train_tokens_per_second": 1428.054
    },
    {
      "epoch": 0.5219310344827586,
      "grad_norm": 0.3825155794620514,
      "learning_rate": 9.587324981577009e-05,
      "loss": 1.1144,
      "num_input_tokens_seen": 6990112,
      "step": 1419,
      "train_runtime": 4894.7302,
      "train_tokens_per_second": 1428.089
    },
    {
      "epoch": 0.5222988505747126,
      "grad_norm": 0.3287540376186371,
      "learning_rate": 9.579955784819455e-05,
      "loss": 1.2909,
      "num_input_tokens_seen": 6997648,
      "step": 1420,
      "train_runtime": 4899.508,
      "train_tokens_per_second": 1428.235
    },
    {
      "epoch": 0.5226666666666666,
      "grad_norm": 0.3626883327960968,
      "learning_rate": 9.572586588061902e-05,
      "loss": 1.2793,
      "num_input_tokens_seen": 7001648,
      "step": 1421,
      "train_runtime": 4902.3196,
      "train_tokens_per_second": 1428.232
    },
    {
      "epoch": 0.5230344827586207,
      "grad_norm": 0.4033142626285553,
      "learning_rate": 9.565217391304348e-05,
      "loss": 1.1882,
      "num_input_tokens_seen": 7005488,
      "step": 1422,
      "train_runtime": 4904.9947,
      "train_tokens_per_second": 1428.236
    },
    {
      "epoch": 0.5234022988505748,
      "grad_norm": 0.3712122142314911,
      "learning_rate": 9.557848194546795e-05,
      "loss": 1.2233,
      "num_input_tokens_seen": 7008720,
      "step": 1423,
      "train_runtime": 4907.3812,
      "train_tokens_per_second": 1428.2
    },
    {
      "epoch": 0.5237701149425288,
      "grad_norm": 0.3646475672721863,
      "learning_rate": 9.550478997789241e-05,
      "loss": 1.2466,
      "num_input_tokens_seen": 7013504,
      "step": 1424,
      "train_runtime": 4910.6601,
      "train_tokens_per_second": 1428.22
    },
    {
      "epoch": 0.5241379310344828,
      "grad_norm": 0.385348379611969,
      "learning_rate": 9.543109801031688e-05,
      "loss": 1.2902,
      "num_input_tokens_seen": 7017872,
      "step": 1425,
      "train_runtime": 4913.6948,
      "train_tokens_per_second": 1428.227
    },
    {
      "epoch": 0.5245057471264368,
      "grad_norm": 0.3624279201030731,
      "learning_rate": 9.535740604274134e-05,
      "loss": 1.2451,
      "num_input_tokens_seen": 7024192,
      "step": 1426,
      "train_runtime": 4917.7704,
      "train_tokens_per_second": 1428.329
    },
    {
      "epoch": 0.5248735632183908,
      "grad_norm": 0.3650817573070526,
      "learning_rate": 9.528371407516581e-05,
      "loss": 1.2346,
      "num_input_tokens_seen": 7027712,
      "step": 1427,
      "train_runtime": 4920.22,
      "train_tokens_per_second": 1428.333
    },
    {
      "epoch": 0.5252413793103449,
      "grad_norm": 0.38149765133857727,
      "learning_rate": 9.521002210759029e-05,
      "loss": 1.2122,
      "num_input_tokens_seen": 7031200,
      "step": 1428,
      "train_runtime": 4922.7569,
      "train_tokens_per_second": 1428.305
    },
    {
      "epoch": 0.5256091954022989,
      "grad_norm": 0.3471353352069855,
      "learning_rate": 9.513633014001474e-05,
      "loss": 1.1466,
      "num_input_tokens_seen": 7034560,
      "step": 1429,
      "train_runtime": 4925.2256,
      "train_tokens_per_second": 1428.272
    },
    {
      "epoch": 0.5259770114942529,
      "grad_norm": 0.3440505564212799,
      "learning_rate": 9.506263817243922e-05,
      "loss": 1.1731,
      "num_input_tokens_seen": 7039472,
      "step": 1430,
      "train_runtime": 4928.5523,
      "train_tokens_per_second": 1428.304
    },
    {
      "epoch": 0.5263448275862069,
      "grad_norm": 0.3367503881454468,
      "learning_rate": 9.498894620486367e-05,
      "loss": 1.1563,
      "num_input_tokens_seen": 7043056,
      "step": 1431,
      "train_runtime": 4931.1903,
      "train_tokens_per_second": 1428.267
    },
    {
      "epoch": 0.526712643678161,
      "grad_norm": 0.37291184067726135,
      "learning_rate": 9.491525423728815e-05,
      "loss": 1.2209,
      "num_input_tokens_seen": 7047216,
      "step": 1432,
      "train_runtime": 4934.1741,
      "train_tokens_per_second": 1428.246
    },
    {
      "epoch": 0.527080459770115,
      "grad_norm": 0.33535143733024597,
      "learning_rate": 9.48415622697126e-05,
      "loss": 1.2184,
      "num_input_tokens_seen": 7053808,
      "step": 1433,
      "train_runtime": 4938.4373,
      "train_tokens_per_second": 1428.348
    },
    {
      "epoch": 0.527448275862069,
      "grad_norm": 0.3484176993370056,
      "learning_rate": 9.476787030213707e-05,
      "loss": 1.2478,
      "num_input_tokens_seen": 7058144,
      "step": 1434,
      "train_runtime": 4941.4719,
      "train_tokens_per_second": 1428.348
    },
    {
      "epoch": 0.527816091954023,
      "grad_norm": 0.36846980452537537,
      "learning_rate": 9.469417833456154e-05,
      "loss": 1.1772,
      "num_input_tokens_seen": 7063456,
      "step": 1435,
      "train_runtime": 4944.96,
      "train_tokens_per_second": 1428.415
    },
    {
      "epoch": 0.528183908045977,
      "grad_norm": 0.3910638689994812,
      "learning_rate": 9.4620486366986e-05,
      "loss": 1.2322,
      "num_input_tokens_seen": 7069552,
      "step": 1436,
      "train_runtime": 4948.8915,
      "train_tokens_per_second": 1428.512
    },
    {
      "epoch": 0.5285517241379311,
      "grad_norm": 0.41100719571113586,
      "learning_rate": 9.454679439941047e-05,
      "loss": 1.4005,
      "num_input_tokens_seen": 7074240,
      "step": 1437,
      "train_runtime": 4952.0394,
      "train_tokens_per_second": 1428.551
    },
    {
      "epoch": 0.5289195402298851,
      "grad_norm": 0.3543812334537506,
      "learning_rate": 9.447310243183493e-05,
      "loss": 1.174,
      "num_input_tokens_seen": 7078064,
      "step": 1438,
      "train_runtime": 4954.7094,
      "train_tokens_per_second": 1428.553
    },
    {
      "epoch": 0.5292873563218391,
      "grad_norm": 0.3710232675075531,
      "learning_rate": 9.43994104642594e-05,
      "loss": 1.3199,
      "num_input_tokens_seen": 7083088,
      "step": 1439,
      "train_runtime": 4958.0439,
      "train_tokens_per_second": 1428.605
    },
    {
      "epoch": 0.5296551724137931,
      "grad_norm": 0.34313949942588806,
      "learning_rate": 9.432571849668386e-05,
      "loss": 1.153,
      "num_input_tokens_seen": 7086640,
      "step": 1440,
      "train_runtime": 4960.6411,
      "train_tokens_per_second": 1428.573
    },
    {
      "epoch": 0.5300229885057471,
      "grad_norm": 0.3154895007610321,
      "learning_rate": 9.425202652910833e-05,
      "loss": 1.1942,
      "num_input_tokens_seen": 7093472,
      "step": 1441,
      "train_runtime": 4967.1111,
      "train_tokens_per_second": 1428.088
    },
    {
      "epoch": 0.5303908045977012,
      "grad_norm": 0.33226457238197327,
      "learning_rate": 9.41783345615328e-05,
      "loss": 1.4049,
      "num_input_tokens_seen": 7098416,
      "step": 1442,
      "train_runtime": 4970.4813,
      "train_tokens_per_second": 1428.114
    },
    {
      "epoch": 0.5307586206896552,
      "grad_norm": 0.35245200991630554,
      "learning_rate": 9.410464259395726e-05,
      "loss": 1.4884,
      "num_input_tokens_seen": 7106176,
      "step": 1443,
      "train_runtime": 4975.4591,
      "train_tokens_per_second": 1428.245
    },
    {
      "epoch": 0.5311264367816092,
      "grad_norm": 0.3660103380680084,
      "learning_rate": 9.403095062638174e-05,
      "loss": 1.0861,
      "num_input_tokens_seen": 7109744,
      "step": 1444,
      "train_runtime": 4978.0762,
      "train_tokens_per_second": 1428.211
    },
    {
      "epoch": 0.5314942528735632,
      "grad_norm": 0.3780525326728821,
      "learning_rate": 9.395725865880619e-05,
      "loss": 1.2801,
      "num_input_tokens_seen": 7114240,
      "step": 1445,
      "train_runtime": 4981.273,
      "train_tokens_per_second": 1428.197
    },
    {
      "epoch": 0.5318620689655172,
      "grad_norm": 0.3413727879524231,
      "learning_rate": 9.388356669123067e-05,
      "loss": 1.1914,
      "num_input_tokens_seen": 7118448,
      "step": 1446,
      "train_runtime": 4984.2092,
      "train_tokens_per_second": 1428.2
    },
    {
      "epoch": 0.5322298850574713,
      "grad_norm": 0.34549179673194885,
      "learning_rate": 9.380987472365512e-05,
      "loss": 1.3134,
      "num_input_tokens_seen": 7124640,
      "step": 1447,
      "train_runtime": 4988.2442,
      "train_tokens_per_second": 1428.286
    },
    {
      "epoch": 0.5325977011494253,
      "grad_norm": 0.35517168045043945,
      "learning_rate": 9.37361827560796e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 7130592,
      "step": 1448,
      "train_runtime": 4992.1535,
      "train_tokens_per_second": 1428.36
    },
    {
      "epoch": 0.5329655172413793,
      "grad_norm": 0.33144551515579224,
      "learning_rate": 9.366249078850405e-05,
      "loss": 1.1364,
      "num_input_tokens_seen": 7135360,
      "step": 1449,
      "train_runtime": 4995.3349,
      "train_tokens_per_second": 1428.405
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.332100510597229,
      "learning_rate": 9.358879882092853e-05,
      "loss": 1.2122,
      "num_input_tokens_seen": 7141248,
      "step": 1450,
      "train_runtime": 4999.1342,
      "train_tokens_per_second": 1428.497
    },
    {
      "epoch": 0.5337011494252873,
      "grad_norm": 0.3594553768634796,
      "learning_rate": 9.351510685335299e-05,
      "loss": 1.2517,
      "num_input_tokens_seen": 7146752,
      "step": 1451,
      "train_runtime": 5002.7979,
      "train_tokens_per_second": 1428.551
    },
    {
      "epoch": 0.5340689655172414,
      "grad_norm": 0.3523523509502411,
      "learning_rate": 9.344141488577746e-05,
      "loss": 1.2063,
      "num_input_tokens_seen": 7150448,
      "step": 1452,
      "train_runtime": 5005.4119,
      "train_tokens_per_second": 1428.543
    },
    {
      "epoch": 0.5344367816091954,
      "grad_norm": 0.33943164348602295,
      "learning_rate": 9.336772291820192e-05,
      "loss": 1.1565,
      "num_input_tokens_seen": 7154672,
      "step": 1453,
      "train_runtime": 5008.3272,
      "train_tokens_per_second": 1428.555
    },
    {
      "epoch": 0.5348045977011494,
      "grad_norm": 0.35627835988998413,
      "learning_rate": 9.329403095062638e-05,
      "loss": 1.2225,
      "num_input_tokens_seen": 7158416,
      "step": 1454,
      "train_runtime": 5011.0472,
      "train_tokens_per_second": 1428.527
    },
    {
      "epoch": 0.5351724137931034,
      "grad_norm": 0.35724586248397827,
      "learning_rate": 9.322033898305085e-05,
      "loss": 1.3647,
      "num_input_tokens_seen": 7162480,
      "step": 1455,
      "train_runtime": 5013.913,
      "train_tokens_per_second": 1428.521
    },
    {
      "epoch": 0.5355402298850575,
      "grad_norm": 0.33584490418434143,
      "learning_rate": 9.314664701547531e-05,
      "loss": 1.3424,
      "num_input_tokens_seen": 7166864,
      "step": 1456,
      "train_runtime": 5016.8585,
      "train_tokens_per_second": 1428.556
    },
    {
      "epoch": 0.5359080459770115,
      "grad_norm": 0.3489488959312439,
      "learning_rate": 9.307295504789978e-05,
      "loss": 1.3379,
      "num_input_tokens_seen": 7171152,
      "step": 1457,
      "train_runtime": 5019.7897,
      "train_tokens_per_second": 1428.576
    },
    {
      "epoch": 0.5362758620689655,
      "grad_norm": 0.3750684857368469,
      "learning_rate": 9.299926308032424e-05,
      "loss": 1.3421,
      "num_input_tokens_seen": 7175072,
      "step": 1458,
      "train_runtime": 5022.5173,
      "train_tokens_per_second": 1428.581
    },
    {
      "epoch": 0.5366436781609195,
      "grad_norm": 0.3445059061050415,
      "learning_rate": 9.292557111274871e-05,
      "loss": 1.143,
      "num_input_tokens_seen": 7178800,
      "step": 1459,
      "train_runtime": 5025.1948,
      "train_tokens_per_second": 1428.562
    },
    {
      "epoch": 0.5370114942528735,
      "grad_norm": 0.339100182056427,
      "learning_rate": 9.285187914517319e-05,
      "loss": 1.2948,
      "num_input_tokens_seen": 7184144,
      "step": 1460,
      "train_runtime": 5028.7712,
      "train_tokens_per_second": 1428.608
    },
    {
      "epoch": 0.5373793103448276,
      "grad_norm": 0.33574753999710083,
      "learning_rate": 9.277818717759764e-05,
      "loss": 1.1967,
      "num_input_tokens_seen": 7188400,
      "step": 1461,
      "train_runtime": 5031.7253,
      "train_tokens_per_second": 1428.615
    },
    {
      "epoch": 0.5377471264367816,
      "grad_norm": 0.3918282389640808,
      "learning_rate": 9.270449521002212e-05,
      "loss": 1.2405,
      "num_input_tokens_seen": 7192928,
      "step": 1462,
      "train_runtime": 5034.8113,
      "train_tokens_per_second": 1428.639
    },
    {
      "epoch": 0.5381149425287356,
      "grad_norm": 0.3533143103122711,
      "learning_rate": 9.263080324244657e-05,
      "loss": 1.3731,
      "num_input_tokens_seen": 7197952,
      "step": 1463,
      "train_runtime": 5038.203,
      "train_tokens_per_second": 1428.674
    },
    {
      "epoch": 0.5384827586206896,
      "grad_norm": 0.3659995496273041,
      "learning_rate": 9.255711127487105e-05,
      "loss": 1.2669,
      "num_input_tokens_seen": 7203536,
      "step": 1464,
      "train_runtime": 5041.9099,
      "train_tokens_per_second": 1428.732
    },
    {
      "epoch": 0.5388505747126436,
      "grad_norm": 0.3859991133213043,
      "learning_rate": 9.248341930729551e-05,
      "loss": 1.2924,
      "num_input_tokens_seen": 7210736,
      "step": 1465,
      "train_runtime": 5046.4629,
      "train_tokens_per_second": 1428.869
    },
    {
      "epoch": 0.5392183908045977,
      "grad_norm": 0.33815860748291016,
      "learning_rate": 9.240972733971998e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 7214736,
      "step": 1466,
      "train_runtime": 5049.3082,
      "train_tokens_per_second": 1428.856
    },
    {
      "epoch": 0.5395862068965517,
      "grad_norm": 0.37804386019706726,
      "learning_rate": 9.233603537214444e-05,
      "loss": 1.2564,
      "num_input_tokens_seen": 7221104,
      "step": 1467,
      "train_runtime": 5053.4827,
      "train_tokens_per_second": 1428.936
    },
    {
      "epoch": 0.5399540229885057,
      "grad_norm": 0.3668687641620636,
      "learning_rate": 9.22623434045689e-05,
      "loss": 1.1892,
      "num_input_tokens_seen": 7225408,
      "step": 1468,
      "train_runtime": 5056.4195,
      "train_tokens_per_second": 1428.957
    },
    {
      "epoch": 0.5403218390804597,
      "grad_norm": 0.3522321581840515,
      "learning_rate": 9.218865143699337e-05,
      "loss": 1.116,
      "num_input_tokens_seen": 7231248,
      "step": 1469,
      "train_runtime": 5060.2624,
      "train_tokens_per_second": 1429.026
    },
    {
      "epoch": 0.5406896551724137,
      "grad_norm": 0.34292274713516235,
      "learning_rate": 9.211495946941784e-05,
      "loss": 1.1913,
      "num_input_tokens_seen": 7235232,
      "step": 1470,
      "train_runtime": 5063.0051,
      "train_tokens_per_second": 1429.039
    },
    {
      "epoch": 0.5410574712643678,
      "grad_norm": 0.39232829213142395,
      "learning_rate": 9.20412675018423e-05,
      "loss": 1.1431,
      "num_input_tokens_seen": 7239360,
      "step": 1471,
      "train_runtime": 5067.8903,
      "train_tokens_per_second": 1428.476
    },
    {
      "epoch": 0.5414252873563219,
      "grad_norm": 0.3699335753917694,
      "learning_rate": 9.196757553426677e-05,
      "loss": 1.2261,
      "num_input_tokens_seen": 7244096,
      "step": 1472,
      "train_runtime": 5071.1896,
      "train_tokens_per_second": 1428.481
    },
    {
      "epoch": 0.5417931034482759,
      "grad_norm": 0.34099844098091125,
      "learning_rate": 9.189388356669123e-05,
      "loss": 1.2193,
      "num_input_tokens_seen": 7249120,
      "step": 1473,
      "train_runtime": 5074.5549,
      "train_tokens_per_second": 1428.523
    },
    {
      "epoch": 0.5421609195402299,
      "grad_norm": 0.39217302203178406,
      "learning_rate": 9.18201915991157e-05,
      "loss": 1.3514,
      "num_input_tokens_seen": 7252608,
      "step": 1474,
      "train_runtime": 5077.0821,
      "train_tokens_per_second": 1428.499
    },
    {
      "epoch": 0.542528735632184,
      "grad_norm": 0.3623465895652771,
      "learning_rate": 9.174649963154016e-05,
      "loss": 1.1799,
      "num_input_tokens_seen": 7256240,
      "step": 1475,
      "train_runtime": 5079.669,
      "train_tokens_per_second": 1428.487
    },
    {
      "epoch": 0.542896551724138,
      "grad_norm": 0.36307892203330994,
      "learning_rate": 9.167280766396464e-05,
      "loss": 1.131,
      "num_input_tokens_seen": 7260016,
      "step": 1476,
      "train_runtime": 5082.324,
      "train_tokens_per_second": 1428.483
    },
    {
      "epoch": 0.543264367816092,
      "grad_norm": 0.3880380094051361,
      "learning_rate": 9.159911569638909e-05,
      "loss": 1.2634,
      "num_input_tokens_seen": 7263824,
      "step": 1477,
      "train_runtime": 5085.0841,
      "train_tokens_per_second": 1428.457
    },
    {
      "epoch": 0.543632183908046,
      "grad_norm": 0.38017377257347107,
      "learning_rate": 9.152542372881357e-05,
      "loss": 1.2838,
      "num_input_tokens_seen": 7271696,
      "step": 1478,
      "train_runtime": 5090.0658,
      "train_tokens_per_second": 1428.605
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.37988346815109253,
      "learning_rate": 9.145173176123802e-05,
      "loss": 1.3246,
      "num_input_tokens_seen": 7277808,
      "step": 1479,
      "train_runtime": 5094.0697,
      "train_tokens_per_second": 1428.682
    },
    {
      "epoch": 0.5443678160919541,
      "grad_norm": 0.40572696924209595,
      "learning_rate": 9.13780397936625e-05,
      "loss": 1.1419,
      "num_input_tokens_seen": 7284496,
      "step": 1480,
      "train_runtime": 5098.4092,
      "train_tokens_per_second": 1428.778
    },
    {
      "epoch": 0.5447356321839081,
      "grad_norm": 0.3993561863899231,
      "learning_rate": 9.130434782608696e-05,
      "loss": 1.3171,
      "num_input_tokens_seen": 7288880,
      "step": 1481,
      "train_runtime": 5101.3945,
      "train_tokens_per_second": 1428.801
    },
    {
      "epoch": 0.5451034482758621,
      "grad_norm": 0.3429904878139496,
      "learning_rate": 9.123065585851143e-05,
      "loss": 1.1614,
      "num_input_tokens_seen": 7294032,
      "step": 1482,
      "train_runtime": 5104.8663,
      "train_tokens_per_second": 1428.839
    },
    {
      "epoch": 0.5454712643678161,
      "grad_norm": 0.3459399938583374,
      "learning_rate": 9.115696389093589e-05,
      "loss": 1.1332,
      "num_input_tokens_seen": 7298528,
      "step": 1483,
      "train_runtime": 5107.9103,
      "train_tokens_per_second": 1428.868
    },
    {
      "epoch": 0.5458390804597701,
      "grad_norm": 0.33769845962524414,
      "learning_rate": 9.108327192336036e-05,
      "loss": 1.0937,
      "num_input_tokens_seen": 7302544,
      "step": 1484,
      "train_runtime": 5110.6401,
      "train_tokens_per_second": 1428.89
    },
    {
      "epoch": 0.5462068965517242,
      "grad_norm": 0.37438470125198364,
      "learning_rate": 9.100957995578482e-05,
      "loss": 1.3188,
      "num_input_tokens_seen": 7309600,
      "step": 1485,
      "train_runtime": 5115.2012,
      "train_tokens_per_second": 1428.996
    },
    {
      "epoch": 0.5465747126436782,
      "grad_norm": 0.3531028628349304,
      "learning_rate": 9.093588798820929e-05,
      "loss": 1.3727,
      "num_input_tokens_seen": 7312896,
      "step": 1486,
      "train_runtime": 5117.6116,
      "train_tokens_per_second": 1428.967
    },
    {
      "epoch": 0.5469425287356322,
      "grad_norm": 0.35417547821998596,
      "learning_rate": 9.086219602063376e-05,
      "loss": 1.2232,
      "num_input_tokens_seen": 7316544,
      "step": 1487,
      "train_runtime": 5120.2516,
      "train_tokens_per_second": 1428.942
    },
    {
      "epoch": 0.5473103448275862,
      "grad_norm": 0.3925384283065796,
      "learning_rate": 9.078850405305822e-05,
      "loss": 1.3288,
      "num_input_tokens_seen": 7320544,
      "step": 1488,
      "train_runtime": 5122.9592,
      "train_tokens_per_second": 1428.968
    },
    {
      "epoch": 0.5476781609195402,
      "grad_norm": 0.3428603410720825,
      "learning_rate": 9.07148120854827e-05,
      "loss": 1.1466,
      "num_input_tokens_seen": 7325136,
      "step": 1489,
      "train_runtime": 5126.0314,
      "train_tokens_per_second": 1429.007
    },
    {
      "epoch": 0.5480459770114943,
      "grad_norm": 0.34437793493270874,
      "learning_rate": 9.064112011790716e-05,
      "loss": 1.1909,
      "num_input_tokens_seen": 7330576,
      "step": 1490,
      "train_runtime": 5129.7072,
      "train_tokens_per_second": 1429.044
    },
    {
      "epoch": 0.5484137931034483,
      "grad_norm": 0.35966598987579346,
      "learning_rate": 9.056742815033162e-05,
      "loss": 1.2855,
      "num_input_tokens_seen": 7334560,
      "step": 1491,
      "train_runtime": 5132.51,
      "train_tokens_per_second": 1429.04
    },
    {
      "epoch": 0.5487816091954023,
      "grad_norm": 0.3296428620815277,
      "learning_rate": 9.049373618275609e-05,
      "loss": 1.1789,
      "num_input_tokens_seen": 7338880,
      "step": 1492,
      "train_runtime": 5135.5379,
      "train_tokens_per_second": 1429.038
    },
    {
      "epoch": 0.5491494252873563,
      "grad_norm": 0.33403250575065613,
      "learning_rate": 9.042004421518055e-05,
      "loss": 1.0913,
      "num_input_tokens_seen": 7342592,
      "step": 1493,
      "train_runtime": 5138.23,
      "train_tokens_per_second": 1429.012
    },
    {
      "epoch": 0.5495172413793104,
      "grad_norm": 0.3356791138648987,
      "learning_rate": 9.034635224760502e-05,
      "loss": 1.1621,
      "num_input_tokens_seen": 7346608,
      "step": 1494,
      "train_runtime": 5141.0444,
      "train_tokens_per_second": 1429.011
    },
    {
      "epoch": 0.5498850574712644,
      "grad_norm": 0.36152005195617676,
      "learning_rate": 9.027266028002948e-05,
      "loss": 1.1491,
      "num_input_tokens_seen": 7351232,
      "step": 1495,
      "train_runtime": 5144.1615,
      "train_tokens_per_second": 1429.044
    },
    {
      "epoch": 0.5502528735632184,
      "grad_norm": 0.3728543221950531,
      "learning_rate": 9.019896831245395e-05,
      "loss": 1.3075,
      "num_input_tokens_seen": 7355008,
      "step": 1496,
      "train_runtime": 5146.8104,
      "train_tokens_per_second": 1429.042
    },
    {
      "epoch": 0.5506206896551724,
      "grad_norm": 0.3417280614376068,
      "learning_rate": 9.012527634487841e-05,
      "loss": 1.317,
      "num_input_tokens_seen": 7359856,
      "step": 1497,
      "train_runtime": 5150.1016,
      "train_tokens_per_second": 1429.07
    },
    {
      "epoch": 0.5509885057471264,
      "grad_norm": 0.38008061051368713,
      "learning_rate": 9.005158437730288e-05,
      "loss": 1.2398,
      "num_input_tokens_seen": 7363424,
      "step": 1498,
      "train_runtime": 5152.6932,
      "train_tokens_per_second": 1429.044
    },
    {
      "epoch": 0.5513563218390805,
      "grad_norm": 0.3422344923019409,
      "learning_rate": 8.997789240972734e-05,
      "loss": 1.2273,
      "num_input_tokens_seen": 7368832,
      "step": 1499,
      "train_runtime": 5156.2332,
      "train_tokens_per_second": 1429.111
    },
    {
      "epoch": 0.5517241379310345,
      "grad_norm": 0.3224169909954071,
      "learning_rate": 8.990420044215181e-05,
      "loss": 1.2853,
      "num_input_tokens_seen": 7372432,
      "step": 1500,
      "train_runtime": 5158.8867,
      "train_tokens_per_second": 1429.074
    },
    {
      "epoch": 0.5520919540229885,
      "grad_norm": 0.35514792799949646,
      "learning_rate": 8.983050847457629e-05,
      "loss": 1.0594,
      "num_input_tokens_seen": 7376400,
      "step": 1501,
      "train_runtime": 5163.7448,
      "train_tokens_per_second": 1428.498
    },
    {
      "epoch": 0.5524597701149425,
      "grad_norm": 0.3709370493888855,
      "learning_rate": 8.975681650700074e-05,
      "loss": 1.2589,
      "num_input_tokens_seen": 7380432,
      "step": 1502,
      "train_runtime": 5166.5773,
      "train_tokens_per_second": 1428.495
    },
    {
      "epoch": 0.5528275862068965,
      "grad_norm": 0.3260691165924072,
      "learning_rate": 8.968312453942522e-05,
      "loss": 1.1489,
      "num_input_tokens_seen": 7384992,
      "step": 1503,
      "train_runtime": 5169.6657,
      "train_tokens_per_second": 1428.524
    },
    {
      "epoch": 0.5531954022988506,
      "grad_norm": 0.3615417778491974,
      "learning_rate": 8.960943257184967e-05,
      "loss": 1.1495,
      "num_input_tokens_seen": 7389872,
      "step": 1504,
      "train_runtime": 5172.9174,
      "train_tokens_per_second": 1428.569
    },
    {
      "epoch": 0.5535632183908046,
      "grad_norm": 0.3600541353225708,
      "learning_rate": 8.953574060427414e-05,
      "loss": 1.2627,
      "num_input_tokens_seen": 7394800,
      "step": 1505,
      "train_runtime": 5176.2298,
      "train_tokens_per_second": 1428.607
    },
    {
      "epoch": 0.5539310344827586,
      "grad_norm": 0.32656240463256836,
      "learning_rate": 8.946204863669861e-05,
      "loss": 1.2052,
      "num_input_tokens_seen": 7398848,
      "step": 1506,
      "train_runtime": 5179.0722,
      "train_tokens_per_second": 1428.605
    },
    {
      "epoch": 0.5542988505747126,
      "grad_norm": 0.35126733779907227,
      "learning_rate": 8.938835666912307e-05,
      "loss": 1.2179,
      "num_input_tokens_seen": 7406288,
      "step": 1507,
      "train_runtime": 5183.7464,
      "train_tokens_per_second": 1428.752
    },
    {
      "epoch": 0.5546666666666666,
      "grad_norm": 0.3632078170776367,
      "learning_rate": 8.931466470154754e-05,
      "loss": 1.2913,
      "num_input_tokens_seen": 7411152,
      "step": 1508,
      "train_runtime": 5187.0165,
      "train_tokens_per_second": 1428.789
    },
    {
      "epoch": 0.5550344827586207,
      "grad_norm": 0.3276430666446686,
      "learning_rate": 8.9240972733972e-05,
      "loss": 1.1918,
      "num_input_tokens_seen": 7414992,
      "step": 1509,
      "train_runtime": 5189.763,
      "train_tokens_per_second": 1428.773
    },
    {
      "epoch": 0.5554022988505747,
      "grad_norm": 0.40329283475875854,
      "learning_rate": 8.916728076639647e-05,
      "loss": 1.2579,
      "num_input_tokens_seen": 7419104,
      "step": 1510,
      "train_runtime": 5192.6238,
      "train_tokens_per_second": 1428.777
    },
    {
      "epoch": 0.5557701149425287,
      "grad_norm": 0.3633747696876526,
      "learning_rate": 8.909358879882093e-05,
      "loss": 1.2176,
      "num_input_tokens_seen": 7423280,
      "step": 1511,
      "train_runtime": 5195.4816,
      "train_tokens_per_second": 1428.795
    },
    {
      "epoch": 0.5561379310344827,
      "grad_norm": 0.37626203894615173,
      "learning_rate": 8.90198968312454e-05,
      "loss": 1.1892,
      "num_input_tokens_seen": 7429328,
      "step": 1512,
      "train_runtime": 5199.4065,
      "train_tokens_per_second": 1428.88
    },
    {
      "epoch": 0.5565057471264367,
      "grad_norm": 0.3440423309803009,
      "learning_rate": 8.894620486366986e-05,
      "loss": 1.2773,
      "num_input_tokens_seen": 7435152,
      "step": 1513,
      "train_runtime": 5203.1178,
      "train_tokens_per_second": 1428.98
    },
    {
      "epoch": 0.5568735632183908,
      "grad_norm": 0.3534166216850281,
      "learning_rate": 8.887251289609433e-05,
      "loss": 1.2264,
      "num_input_tokens_seen": 7439568,
      "step": 1514,
      "train_runtime": 5206.2377,
      "train_tokens_per_second": 1428.972
    },
    {
      "epoch": 0.5572413793103448,
      "grad_norm": 0.34926795959472656,
      "learning_rate": 8.87988209285188e-05,
      "loss": 1.2183,
      "num_input_tokens_seen": 7443664,
      "step": 1515,
      "train_runtime": 5209.0466,
      "train_tokens_per_second": 1428.988
    },
    {
      "epoch": 0.5576091954022988,
      "grad_norm": 0.35664013028144836,
      "learning_rate": 8.872512896094326e-05,
      "loss": 1.2729,
      "num_input_tokens_seen": 7447312,
      "step": 1516,
      "train_runtime": 5211.6163,
      "train_tokens_per_second": 1428.983
    },
    {
      "epoch": 0.5579770114942528,
      "grad_norm": 0.31531840562820435,
      "learning_rate": 8.865143699336774e-05,
      "loss": 1.122,
      "num_input_tokens_seen": 7453424,
      "step": 1517,
      "train_runtime": 5215.5911,
      "train_tokens_per_second": 1429.066
    },
    {
      "epoch": 0.5583448275862068,
      "grad_norm": 0.35419803857803345,
      "learning_rate": 8.857774502579219e-05,
      "loss": 1.2165,
      "num_input_tokens_seen": 7459232,
      "step": 1518,
      "train_runtime": 5219.4465,
      "train_tokens_per_second": 1429.123
    },
    {
      "epoch": 0.5587126436781609,
      "grad_norm": 0.3428138196468353,
      "learning_rate": 8.850405305821667e-05,
      "loss": 1.0832,
      "num_input_tokens_seen": 7464944,
      "step": 1519,
      "train_runtime": 5223.1947,
      "train_tokens_per_second": 1429.191
    },
    {
      "epoch": 0.5590804597701149,
      "grad_norm": 0.389974981546402,
      "learning_rate": 8.843036109064112e-05,
      "loss": 1.2682,
      "num_input_tokens_seen": 7468640,
      "step": 1520,
      "train_runtime": 5225.8292,
      "train_tokens_per_second": 1429.178
    },
    {
      "epoch": 0.5594482758620689,
      "grad_norm": 0.3580048382282257,
      "learning_rate": 8.83566691230656e-05,
      "loss": 1.2654,
      "num_input_tokens_seen": 7473520,
      "step": 1521,
      "train_runtime": 5229.0927,
      "train_tokens_per_second": 1429.219
    },
    {
      "epoch": 0.559816091954023,
      "grad_norm": 0.34142670035362244,
      "learning_rate": 8.828297715549006e-05,
      "loss": 1.1834,
      "num_input_tokens_seen": 7478080,
      "step": 1522,
      "train_runtime": 5232.2702,
      "train_tokens_per_second": 1429.223
    },
    {
      "epoch": 0.5601839080459771,
      "grad_norm": 0.35978788137435913,
      "learning_rate": 8.820928518791453e-05,
      "loss": 1.2212,
      "num_input_tokens_seen": 7482064,
      "step": 1523,
      "train_runtime": 5235.0182,
      "train_tokens_per_second": 1429.234
    },
    {
      "epoch": 0.5605517241379311,
      "grad_norm": 0.3566664457321167,
      "learning_rate": 8.813559322033899e-05,
      "loss": 1.2121,
      "num_input_tokens_seen": 7487008,
      "step": 1524,
      "train_runtime": 5238.4107,
      "train_tokens_per_second": 1429.252
    },
    {
      "epoch": 0.5609195402298851,
      "grad_norm": 0.37518709897994995,
      "learning_rate": 8.806190125276345e-05,
      "loss": 1.322,
      "num_input_tokens_seen": 7494880,
      "step": 1525,
      "train_runtime": 5243.3558,
      "train_tokens_per_second": 1429.405
    },
    {
      "epoch": 0.5612873563218391,
      "grad_norm": 0.3378138542175293,
      "learning_rate": 8.798820928518792e-05,
      "loss": 1.3046,
      "num_input_tokens_seen": 7498992,
      "step": 1526,
      "train_runtime": 5246.1702,
      "train_tokens_per_second": 1429.422
    },
    {
      "epoch": 0.5616551724137931,
      "grad_norm": 0.33979520201683044,
      "learning_rate": 8.791451731761238e-05,
      "loss": 1.2412,
      "num_input_tokens_seen": 7504512,
      "step": 1527,
      "train_runtime": 5249.7796,
      "train_tokens_per_second": 1429.491
    },
    {
      "epoch": 0.5620229885057472,
      "grad_norm": 0.3645789325237274,
      "learning_rate": 8.784082535003685e-05,
      "loss": 1.1553,
      "num_input_tokens_seen": 7510080,
      "step": 1528,
      "train_runtime": 5253.3499,
      "train_tokens_per_second": 1429.579
    },
    {
      "epoch": 0.5623908045977012,
      "grad_norm": 0.3475429117679596,
      "learning_rate": 8.776713338246131e-05,
      "loss": 1.2072,
      "num_input_tokens_seen": 7515136,
      "step": 1529,
      "train_runtime": 5256.7019,
      "train_tokens_per_second": 1429.629
    },
    {
      "epoch": 0.5627586206896552,
      "grad_norm": 0.39109325408935547,
      "learning_rate": 8.769344141488578e-05,
      "loss": 1.2434,
      "num_input_tokens_seen": 7519008,
      "step": 1530,
      "train_runtime": 5259.5112,
      "train_tokens_per_second": 1429.602
    },
    {
      "epoch": 0.5631264367816092,
      "grad_norm": 0.3514973223209381,
      "learning_rate": 8.761974944731026e-05,
      "loss": 1.0944,
      "num_input_tokens_seen": 7525248,
      "step": 1531,
      "train_runtime": 5265.5413,
      "train_tokens_per_second": 1429.15
    },
    {
      "epoch": 0.5634942528735633,
      "grad_norm": 0.36799776554107666,
      "learning_rate": 8.754605747973471e-05,
      "loss": 1.3268,
      "num_input_tokens_seen": 7529232,
      "step": 1532,
      "train_runtime": 5268.3203,
      "train_tokens_per_second": 1429.152
    },
    {
      "epoch": 0.5638620689655173,
      "grad_norm": 0.38167804479599,
      "learning_rate": 8.747236551215919e-05,
      "loss": 1.1402,
      "num_input_tokens_seen": 7533152,
      "step": 1533,
      "train_runtime": 5271.0938,
      "train_tokens_per_second": 1429.144
    },
    {
      "epoch": 0.5642298850574713,
      "grad_norm": 0.36129435896873474,
      "learning_rate": 8.739867354458364e-05,
      "loss": 1.157,
      "num_input_tokens_seen": 7537488,
      "step": 1534,
      "train_runtime": 5274.1097,
      "train_tokens_per_second": 1429.149
    },
    {
      "epoch": 0.5645977011494253,
      "grad_norm": 0.31032782793045044,
      "learning_rate": 8.732498157700812e-05,
      "loss": 1.2479,
      "num_input_tokens_seen": 7542896,
      "step": 1535,
      "train_runtime": 5277.6584,
      "train_tokens_per_second": 1429.213
    },
    {
      "epoch": 0.5649655172413793,
      "grad_norm": 0.33601248264312744,
      "learning_rate": 8.725128960943257e-05,
      "loss": 1.1178,
      "num_input_tokens_seen": 7548576,
      "step": 1536,
      "train_runtime": 5281.4201,
      "train_tokens_per_second": 1429.27
    },
    {
      "epoch": 0.5653333333333334,
      "grad_norm": 0.32082489132881165,
      "learning_rate": 8.717759764185705e-05,
      "loss": 1.2136,
      "num_input_tokens_seen": 7553744,
      "step": 1537,
      "train_runtime": 5284.7894,
      "train_tokens_per_second": 1429.337
    },
    {
      "epoch": 0.5657011494252874,
      "grad_norm": 0.3529646694660187,
      "learning_rate": 8.710390567428151e-05,
      "loss": 1.1445,
      "num_input_tokens_seen": 7559152,
      "step": 1538,
      "train_runtime": 5288.4881,
      "train_tokens_per_second": 1429.36
    },
    {
      "epoch": 0.5660689655172414,
      "grad_norm": 0.3776971995830536,
      "learning_rate": 8.703021370670598e-05,
      "loss": 1.1139,
      "num_input_tokens_seen": 7563968,
      "step": 1539,
      "train_runtime": 5291.7811,
      "train_tokens_per_second": 1429.38
    },
    {
      "epoch": 0.5664367816091954,
      "grad_norm": 0.3826366662979126,
      "learning_rate": 8.695652173913044e-05,
      "loss": 1.2649,
      "num_input_tokens_seen": 7568208,
      "step": 1540,
      "train_runtime": 5294.7419,
      "train_tokens_per_second": 1429.382
    },
    {
      "epoch": 0.5668045977011494,
      "grad_norm": 0.3375912308692932,
      "learning_rate": 8.68828297715549e-05,
      "loss": 1.1855,
      "num_input_tokens_seen": 7573648,
      "step": 1541,
      "train_runtime": 5298.334,
      "train_tokens_per_second": 1429.44
    },
    {
      "epoch": 0.5671724137931035,
      "grad_norm": 0.3581699728965759,
      "learning_rate": 8.680913780397937e-05,
      "loss": 1.2801,
      "num_input_tokens_seen": 7578288,
      "step": 1542,
      "train_runtime": 5301.5126,
      "train_tokens_per_second": 1429.458
    },
    {
      "epoch": 0.5675402298850575,
      "grad_norm": 0.3506247103214264,
      "learning_rate": 8.673544583640384e-05,
      "loss": 1.0998,
      "num_input_tokens_seen": 7581680,
      "step": 1543,
      "train_runtime": 5303.9404,
      "train_tokens_per_second": 1429.443
    },
    {
      "epoch": 0.5679080459770115,
      "grad_norm": 0.3562047779560089,
      "learning_rate": 8.66617538688283e-05,
      "loss": 1.249,
      "num_input_tokens_seen": 7586400,
      "step": 1544,
      "train_runtime": 5307.0955,
      "train_tokens_per_second": 1429.482
    },
    {
      "epoch": 0.5682758620689655,
      "grad_norm": 0.4296574294567108,
      "learning_rate": 8.658806190125276e-05,
      "loss": 1.3815,
      "num_input_tokens_seen": 7592752,
      "step": 1545,
      "train_runtime": 5311.2583,
      "train_tokens_per_second": 1429.558
    },
    {
      "epoch": 0.5686436781609195,
      "grad_norm": 0.3460198640823364,
      "learning_rate": 8.651436993367723e-05,
      "loss": 1.3356,
      "num_input_tokens_seen": 7597296,
      "step": 1546,
      "train_runtime": 5314.312,
      "train_tokens_per_second": 1429.592
    },
    {
      "epoch": 0.5690114942528736,
      "grad_norm": 0.37243911623954773,
      "learning_rate": 8.644067796610171e-05,
      "loss": 1.2701,
      "num_input_tokens_seen": 7600352,
      "step": 1547,
      "train_runtime": 5316.6655,
      "train_tokens_per_second": 1429.534
    },
    {
      "epoch": 0.5693793103448276,
      "grad_norm": 0.37210336327552795,
      "learning_rate": 8.636698599852616e-05,
      "loss": 1.2796,
      "num_input_tokens_seen": 7604144,
      "step": 1548,
      "train_runtime": 5319.2615,
      "train_tokens_per_second": 1429.549
    },
    {
      "epoch": 0.5697471264367816,
      "grad_norm": 0.35311219096183777,
      "learning_rate": 8.629329403095064e-05,
      "loss": 1.2223,
      "num_input_tokens_seen": 7607872,
      "step": 1549,
      "train_runtime": 5321.9327,
      "train_tokens_per_second": 1429.532
    },
    {
      "epoch": 0.5701149425287356,
      "grad_norm": 0.33246049284935,
      "learning_rate": 8.621960206337509e-05,
      "loss": 1.1023,
      "num_input_tokens_seen": 7611424,
      "step": 1550,
      "train_runtime": 5324.5287,
      "train_tokens_per_second": 1429.502
    },
    {
      "epoch": 0.5704827586206896,
      "grad_norm": 0.3722507357597351,
      "learning_rate": 8.614591009579957e-05,
      "loss": 1.2594,
      "num_input_tokens_seen": 7617152,
      "step": 1551,
      "train_runtime": 5328.3123,
      "train_tokens_per_second": 1429.562
    },
    {
      "epoch": 0.5708505747126437,
      "grad_norm": 0.3631453216075897,
      "learning_rate": 8.607221812822402e-05,
      "loss": 1.212,
      "num_input_tokens_seen": 7620800,
      "step": 1552,
      "train_runtime": 5330.9807,
      "train_tokens_per_second": 1429.531
    },
    {
      "epoch": 0.5712183908045977,
      "grad_norm": 0.35323989391326904,
      "learning_rate": 8.59985261606485e-05,
      "loss": 1.2393,
      "num_input_tokens_seen": 7625632,
      "step": 1553,
      "train_runtime": 5334.2467,
      "train_tokens_per_second": 1429.561
    },
    {
      "epoch": 0.5715862068965517,
      "grad_norm": 0.36066100001335144,
      "learning_rate": 8.592483419307296e-05,
      "loss": 1.2232,
      "num_input_tokens_seen": 7632032,
      "step": 1554,
      "train_runtime": 5338.3866,
      "train_tokens_per_second": 1429.651
    },
    {
      "epoch": 0.5719540229885057,
      "grad_norm": 0.3540048599243164,
      "learning_rate": 8.585114222549743e-05,
      "loss": 1.2513,
      "num_input_tokens_seen": 7637328,
      "step": 1555,
      "train_runtime": 5341.8947,
      "train_tokens_per_second": 1429.704
    },
    {
      "epoch": 0.5723218390804597,
      "grad_norm": 0.36024242639541626,
      "learning_rate": 8.577745025792189e-05,
      "loss": 1.239,
      "num_input_tokens_seen": 7642720,
      "step": 1556,
      "train_runtime": 5345.4446,
      "train_tokens_per_second": 1429.763
    },
    {
      "epoch": 0.5726896551724138,
      "grad_norm": 0.34701937437057495,
      "learning_rate": 8.570375829034636e-05,
      "loss": 1.2198,
      "num_input_tokens_seen": 7648496,
      "step": 1557,
      "train_runtime": 5349.2536,
      "train_tokens_per_second": 1429.825
    },
    {
      "epoch": 0.5730574712643678,
      "grad_norm": 0.33805257081985474,
      "learning_rate": 8.563006632277082e-05,
      "loss": 1.3458,
      "num_input_tokens_seen": 7653152,
      "step": 1558,
      "train_runtime": 5352.4503,
      "train_tokens_per_second": 1429.841
    },
    {
      "epoch": 0.5734252873563218,
      "grad_norm": 0.3685722053050995,
      "learning_rate": 8.555637435519529e-05,
      "loss": 1.2973,
      "num_input_tokens_seen": 7664784,
      "step": 1559,
      "train_runtime": 5359.5333,
      "train_tokens_per_second": 1430.122
    },
    {
      "epoch": 0.5737931034482758,
      "grad_norm": 0.34516969323158264,
      "learning_rate": 8.548268238761975e-05,
      "loss": 1.3319,
      "num_input_tokens_seen": 7673408,
      "step": 1560,
      "train_runtime": 5364.7883,
      "train_tokens_per_second": 1430.328
    },
    {
      "epoch": 0.5741609195402299,
      "grad_norm": 0.3682008385658264,
      "learning_rate": 8.540899042004422e-05,
      "loss": 1.3695,
      "num_input_tokens_seen": 7678816,
      "step": 1561,
      "train_runtime": 5370.4033,
      "train_tokens_per_second": 1429.84
    },
    {
      "epoch": 0.5745287356321839,
      "grad_norm": 0.351102739572525,
      "learning_rate": 8.533529845246868e-05,
      "loss": 1.0719,
      "num_input_tokens_seen": 7683680,
      "step": 1562,
      "train_runtime": 5373.6442,
      "train_tokens_per_second": 1429.883
    },
    {
      "epoch": 0.5748965517241379,
      "grad_norm": 0.3792226016521454,
      "learning_rate": 8.526160648489316e-05,
      "loss": 1.1213,
      "num_input_tokens_seen": 7688160,
      "step": 1563,
      "train_runtime": 5376.6922,
      "train_tokens_per_second": 1429.905
    },
    {
      "epoch": 0.5752643678160919,
      "grad_norm": 0.37403708696365356,
      "learning_rate": 8.518791451731761e-05,
      "loss": 1.3069,
      "num_input_tokens_seen": 7693200,
      "step": 1564,
      "train_runtime": 5380.154,
      "train_tokens_per_second": 1429.922
    },
    {
      "epoch": 0.5756321839080459,
      "grad_norm": 0.3389625549316406,
      "learning_rate": 8.511422254974209e-05,
      "loss": 1.1747,
      "num_input_tokens_seen": 7697888,
      "step": 1565,
      "train_runtime": 5383.286,
      "train_tokens_per_second": 1429.961
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.38302186131477356,
      "learning_rate": 8.504053058216654e-05,
      "loss": 1.2867,
      "num_input_tokens_seen": 7702640,
      "step": 1566,
      "train_runtime": 5386.4768,
      "train_tokens_per_second": 1429.996
    },
    {
      "epoch": 0.576367816091954,
      "grad_norm": 0.3931235671043396,
      "learning_rate": 8.496683861459102e-05,
      "loss": 1.3135,
      "num_input_tokens_seen": 7706704,
      "step": 1567,
      "train_runtime": 5389.3049,
      "train_tokens_per_second": 1430.0
    },
    {
      "epoch": 0.576735632183908,
      "grad_norm": 0.3440674841403961,
      "learning_rate": 8.489314664701547e-05,
      "loss": 1.2389,
      "num_input_tokens_seen": 7711328,
      "step": 1568,
      "train_runtime": 5392.4639,
      "train_tokens_per_second": 1430.019
    },
    {
      "epoch": 0.577103448275862,
      "grad_norm": 0.36354872584342957,
      "learning_rate": 8.481945467943995e-05,
      "loss": 1.1611,
      "num_input_tokens_seen": 7715776,
      "step": 1569,
      "train_runtime": 5395.5077,
      "train_tokens_per_second": 1430.037
    },
    {
      "epoch": 0.577471264367816,
      "grad_norm": 0.34597697854042053,
      "learning_rate": 8.474576271186441e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 7720128,
      "step": 1570,
      "train_runtime": 5398.5457,
      "train_tokens_per_second": 1430.038
    },
    {
      "epoch": 0.5778390804597701,
      "grad_norm": 0.34855154156684875,
      "learning_rate": 8.467207074428888e-05,
      "loss": 1.117,
      "num_input_tokens_seen": 7724576,
      "step": 1571,
      "train_runtime": 5401.702,
      "train_tokens_per_second": 1430.026
    },
    {
      "epoch": 0.5782068965517242,
      "grad_norm": 0.37674474716186523,
      "learning_rate": 8.459837877671334e-05,
      "loss": 1.2255,
      "num_input_tokens_seen": 7731168,
      "step": 1572,
      "train_runtime": 5406.021,
      "train_tokens_per_second": 1430.103
    },
    {
      "epoch": 0.5785747126436782,
      "grad_norm": 0.3757065534591675,
      "learning_rate": 8.452468680913781e-05,
      "loss": 1.1987,
      "num_input_tokens_seen": 7738704,
      "step": 1573,
      "train_runtime": 5410.7509,
      "train_tokens_per_second": 1430.246
    },
    {
      "epoch": 0.5789425287356322,
      "grad_norm": 0.35863828659057617,
      "learning_rate": 8.445099484156227e-05,
      "loss": 1.1842,
      "num_input_tokens_seen": 7744400,
      "step": 1574,
      "train_runtime": 5414.5252,
      "train_tokens_per_second": 1430.301
    },
    {
      "epoch": 0.5793103448275863,
      "grad_norm": 0.3289389908313751,
      "learning_rate": 8.437730287398674e-05,
      "loss": 1.3038,
      "num_input_tokens_seen": 7753232,
      "step": 1575,
      "train_runtime": 5420.0126,
      "train_tokens_per_second": 1430.482
    },
    {
      "epoch": 0.5796781609195403,
      "grad_norm": 0.31837430596351624,
      "learning_rate": 8.43036109064112e-05,
      "loss": 1.4199,
      "num_input_tokens_seen": 7764304,
      "step": 1576,
      "train_runtime": 5426.7195,
      "train_tokens_per_second": 1430.755
    },
    {
      "epoch": 0.5800459770114943,
      "grad_norm": 0.34056252241134644,
      "learning_rate": 8.422991893883567e-05,
      "loss": 0.9954,
      "num_input_tokens_seen": 7767440,
      "step": 1577,
      "train_runtime": 5429.1391,
      "train_tokens_per_second": 1430.695
    },
    {
      "epoch": 0.5804137931034483,
      "grad_norm": 0.3496263921260834,
      "learning_rate": 8.415622697126013e-05,
      "loss": 1.1382,
      "num_input_tokens_seen": 7772704,
      "step": 1578,
      "train_runtime": 5432.6302,
      "train_tokens_per_second": 1430.744
    },
    {
      "epoch": 0.5807816091954023,
      "grad_norm": 0.36682674288749695,
      "learning_rate": 8.408253500368461e-05,
      "loss": 1.3116,
      "num_input_tokens_seen": 7778304,
      "step": 1579,
      "train_runtime": 5436.3262,
      "train_tokens_per_second": 1430.802
    },
    {
      "epoch": 0.5811494252873564,
      "grad_norm": 0.3761027455329895,
      "learning_rate": 8.400884303610906e-05,
      "loss": 1.0612,
      "num_input_tokens_seen": 7782928,
      "step": 1580,
      "train_runtime": 5439.4989,
      "train_tokens_per_second": 1430.817
    },
    {
      "epoch": 0.5815172413793104,
      "grad_norm": 0.3057478964328766,
      "learning_rate": 8.393515106853354e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 7792784,
      "step": 1581,
      "train_runtime": 5445.7031,
      "train_tokens_per_second": 1430.997
    },
    {
      "epoch": 0.5818850574712644,
      "grad_norm": 0.3722552955150604,
      "learning_rate": 8.386145910095799e-05,
      "loss": 1.264,
      "num_input_tokens_seen": 7798480,
      "step": 1582,
      "train_runtime": 5449.4447,
      "train_tokens_per_second": 1431.06
    },
    {
      "epoch": 0.5822528735632184,
      "grad_norm": 0.38350504636764526,
      "learning_rate": 8.378776713338247e-05,
      "loss": 1.2435,
      "num_input_tokens_seen": 7806656,
      "step": 1583,
      "train_runtime": 5454.449,
      "train_tokens_per_second": 1431.246
    },
    {
      "epoch": 0.5826206896551724,
      "grad_norm": 0.3597047030925751,
      "learning_rate": 8.371407516580692e-05,
      "loss": 1.2914,
      "num_input_tokens_seen": 7811504,
      "step": 1584,
      "train_runtime": 5457.7825,
      "train_tokens_per_second": 1431.26
    },
    {
      "epoch": 0.5829885057471265,
      "grad_norm": 0.35457006096839905,
      "learning_rate": 8.36403831982314e-05,
      "loss": 1.2324,
      "num_input_tokens_seen": 7818128,
      "step": 1585,
      "train_runtime": 5462.1282,
      "train_tokens_per_second": 1431.334
    },
    {
      "epoch": 0.5833563218390805,
      "grad_norm": 0.4037728011608124,
      "learning_rate": 8.356669123065586e-05,
      "loss": 1.2973,
      "num_input_tokens_seen": 7821376,
      "step": 1586,
      "train_runtime": 5464.5292,
      "train_tokens_per_second": 1431.299
    },
    {
      "epoch": 0.5837241379310345,
      "grad_norm": 0.3242194950580597,
      "learning_rate": 8.349299926308033e-05,
      "loss": 1.1664,
      "num_input_tokens_seen": 7825424,
      "step": 1587,
      "train_runtime": 5467.3301,
      "train_tokens_per_second": 1431.306
    },
    {
      "epoch": 0.5840919540229885,
      "grad_norm": 0.3614206314086914,
      "learning_rate": 8.341930729550479e-05,
      "loss": 1.0722,
      "num_input_tokens_seen": 7833552,
      "step": 1588,
      "train_runtime": 5472.4124,
      "train_tokens_per_second": 1431.462
    },
    {
      "epoch": 0.5844597701149425,
      "grad_norm": 0.36593663692474365,
      "learning_rate": 8.334561532792926e-05,
      "loss": 1.3346,
      "num_input_tokens_seen": 7837776,
      "step": 1589,
      "train_runtime": 5475.3155,
      "train_tokens_per_second": 1431.475
    },
    {
      "epoch": 0.5848275862068966,
      "grad_norm": 0.32114365696907043,
      "learning_rate": 8.327192336035372e-05,
      "loss": 1.2286,
      "num_input_tokens_seen": 7843584,
      "step": 1590,
      "train_runtime": 5479.1522,
      "train_tokens_per_second": 1431.532
    },
    {
      "epoch": 0.5851954022988506,
      "grad_norm": 0.35319972038269043,
      "learning_rate": 8.319823139277819e-05,
      "loss": 1.2805,
      "num_input_tokens_seen": 7847504,
      "step": 1591,
      "train_runtime": 5483.9903,
      "train_tokens_per_second": 1430.984
    },
    {
      "epoch": 0.5855632183908046,
      "grad_norm": 0.37851041555404663,
      "learning_rate": 8.312453942520265e-05,
      "loss": 1.3903,
      "num_input_tokens_seen": 7852048,
      "step": 1592,
      "train_runtime": 5487.0549,
      "train_tokens_per_second": 1431.013
    },
    {
      "epoch": 0.5859310344827586,
      "grad_norm": 0.361674964427948,
      "learning_rate": 8.305084745762712e-05,
      "loss": 1.2077,
      "num_input_tokens_seen": 7856528,
      "step": 1593,
      "train_runtime": 5490.1351,
      "train_tokens_per_second": 1431.026
    },
    {
      "epoch": 0.5862988505747126,
      "grad_norm": 0.3516424298286438,
      "learning_rate": 8.297715549005158e-05,
      "loss": 1.2541,
      "num_input_tokens_seen": 7859712,
      "step": 1594,
      "train_runtime": 5492.6153,
      "train_tokens_per_second": 1430.96
    },
    {
      "epoch": 0.5866666666666667,
      "grad_norm": 0.36594077944755554,
      "learning_rate": 8.290346352247606e-05,
      "loss": 1.2502,
      "num_input_tokens_seen": 7864448,
      "step": 1595,
      "train_runtime": 5495.8083,
      "train_tokens_per_second": 1430.99
    },
    {
      "epoch": 0.5870344827586207,
      "grad_norm": 0.3563876152038574,
      "learning_rate": 8.282977155490051e-05,
      "loss": 1.3529,
      "num_input_tokens_seen": 7869936,
      "step": 1596,
      "train_runtime": 5499.3828,
      "train_tokens_per_second": 1431.058
    },
    {
      "epoch": 0.5874022988505747,
      "grad_norm": 0.3402424156665802,
      "learning_rate": 8.275607958732499e-05,
      "loss": 1.1151,
      "num_input_tokens_seen": 7876272,
      "step": 1597,
      "train_runtime": 5503.5174,
      "train_tokens_per_second": 1431.134
    },
    {
      "epoch": 0.5877701149425287,
      "grad_norm": 0.37097758054733276,
      "learning_rate": 8.268238761974944e-05,
      "loss": 1.1777,
      "num_input_tokens_seen": 7882080,
      "step": 1598,
      "train_runtime": 5507.3417,
      "train_tokens_per_second": 1431.195
    },
    {
      "epoch": 0.5881379310344828,
      "grad_norm": 0.35348713397979736,
      "learning_rate": 8.260869565217392e-05,
      "loss": 1.2938,
      "num_input_tokens_seen": 7886048,
      "step": 1599,
      "train_runtime": 5510.1555,
      "train_tokens_per_second": 1431.184
    },
    {
      "epoch": 0.5885057471264368,
      "grad_norm": 0.36142581701278687,
      "learning_rate": 8.253500368459838e-05,
      "loss": 1.2518,
      "num_input_tokens_seen": 7892912,
      "step": 1600,
      "train_runtime": 5514.5341,
      "train_tokens_per_second": 1431.293
    },
    {
      "epoch": 0.5888735632183908,
      "grad_norm": 0.34320446848869324,
      "learning_rate": 8.246131171702285e-05,
      "loss": 1.3004,
      "num_input_tokens_seen": 7896464,
      "step": 1601,
      "train_runtime": 5517.0436,
      "train_tokens_per_second": 1431.285
    },
    {
      "epoch": 0.5892413793103448,
      "grad_norm": 0.35260260105133057,
      "learning_rate": 8.238761974944731e-05,
      "loss": 1.1938,
      "num_input_tokens_seen": 7900624,
      "step": 1602,
      "train_runtime": 5519.912,
      "train_tokens_per_second": 1431.295
    },
    {
      "epoch": 0.5896091954022988,
      "grad_norm": 0.37798693776130676,
      "learning_rate": 8.231392778187178e-05,
      "loss": 1.4233,
      "num_input_tokens_seen": 7906464,
      "step": 1603,
      "train_runtime": 5523.7664,
      "train_tokens_per_second": 1431.354
    },
    {
      "epoch": 0.5899770114942529,
      "grad_norm": 0.3189416825771332,
      "learning_rate": 8.224023581429624e-05,
      "loss": 1.1915,
      "num_input_tokens_seen": 7909920,
      "step": 1604,
      "train_runtime": 5526.3058,
      "train_tokens_per_second": 1431.321
    },
    {
      "epoch": 0.5903448275862069,
      "grad_norm": 0.3486345112323761,
      "learning_rate": 8.216654384672071e-05,
      "loss": 1.1839,
      "num_input_tokens_seen": 7914416,
      "step": 1605,
      "train_runtime": 5529.4389,
      "train_tokens_per_second": 1431.324
    },
    {
      "epoch": 0.5907126436781609,
      "grad_norm": 0.36112046241760254,
      "learning_rate": 8.209285187914517e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 7920560,
      "step": 1606,
      "train_runtime": 5533.4603,
      "train_tokens_per_second": 1431.394
    },
    {
      "epoch": 0.5910804597701149,
      "grad_norm": 0.3525996208190918,
      "learning_rate": 8.201915991156964e-05,
      "loss": 1.3363,
      "num_input_tokens_seen": 7924608,
      "step": 1607,
      "train_runtime": 5536.2677,
      "train_tokens_per_second": 1431.399
    },
    {
      "epoch": 0.5914482758620689,
      "grad_norm": 0.35944250226020813,
      "learning_rate": 8.19454679439941e-05,
      "loss": 1.3444,
      "num_input_tokens_seen": 7929232,
      "step": 1608,
      "train_runtime": 5539.4608,
      "train_tokens_per_second": 1431.409
    },
    {
      "epoch": 0.591816091954023,
      "grad_norm": 0.3544812798500061,
      "learning_rate": 8.187177597641857e-05,
      "loss": 1.1643,
      "num_input_tokens_seen": 7933472,
      "step": 1609,
      "train_runtime": 5542.5033,
      "train_tokens_per_second": 1431.388
    },
    {
      "epoch": 0.592183908045977,
      "grad_norm": 0.3513464033603668,
      "learning_rate": 8.179808400884305e-05,
      "loss": 1.0798,
      "num_input_tokens_seen": 7938256,
      "step": 1610,
      "train_runtime": 5545.6623,
      "train_tokens_per_second": 1431.435
    },
    {
      "epoch": 0.592551724137931,
      "grad_norm": 0.3400366008281708,
      "learning_rate": 8.172439204126751e-05,
      "loss": 1.0744,
      "num_input_tokens_seen": 7944656,
      "step": 1611,
      "train_runtime": 5549.8162,
      "train_tokens_per_second": 1431.517
    },
    {
      "epoch": 0.592919540229885,
      "grad_norm": 0.3588066101074219,
      "learning_rate": 8.165070007369198e-05,
      "loss": 1.1398,
      "num_input_tokens_seen": 7950960,
      "step": 1612,
      "train_runtime": 5553.9969,
      "train_tokens_per_second": 1431.574
    },
    {
      "epoch": 0.593287356321839,
      "grad_norm": 0.3882015347480774,
      "learning_rate": 8.157700810611644e-05,
      "loss": 1.3625,
      "num_input_tokens_seen": 7956848,
      "step": 1613,
      "train_runtime": 5557.8548,
      "train_tokens_per_second": 1431.64
    },
    {
      "epoch": 0.5936551724137931,
      "grad_norm": 0.3832549750804901,
      "learning_rate": 8.15033161385409e-05,
      "loss": 1.2174,
      "num_input_tokens_seen": 7962336,
      "step": 1614,
      "train_runtime": 5561.5085,
      "train_tokens_per_second": 1431.686
    },
    {
      "epoch": 0.5940229885057471,
      "grad_norm": 0.34523752331733704,
      "learning_rate": 8.142962417096537e-05,
      "loss": 1.1365,
      "num_input_tokens_seen": 7966576,
      "step": 1615,
      "train_runtime": 5564.5522,
      "train_tokens_per_second": 1431.665
    },
    {
      "epoch": 0.5943908045977011,
      "grad_norm": 0.3599925637245178,
      "learning_rate": 8.135593220338983e-05,
      "loss": 1.2142,
      "num_input_tokens_seen": 7971392,
      "step": 1616,
      "train_runtime": 5567.8557,
      "train_tokens_per_second": 1431.681
    },
    {
      "epoch": 0.5947586206896551,
      "grad_norm": 0.3558957874774933,
      "learning_rate": 8.12822402358143e-05,
      "loss": 1.3562,
      "num_input_tokens_seen": 7976240,
      "step": 1617,
      "train_runtime": 5571.2462,
      "train_tokens_per_second": 1431.68
    },
    {
      "epoch": 0.5951264367816091,
      "grad_norm": 0.39159995317459106,
      "learning_rate": 8.120854826823876e-05,
      "loss": 1.2635,
      "num_input_tokens_seen": 7981296,
      "step": 1618,
      "train_runtime": 5574.6939,
      "train_tokens_per_second": 1431.701
    },
    {
      "epoch": 0.5954942528735632,
      "grad_norm": 0.38039717078208923,
      "learning_rate": 8.113485630066323e-05,
      "loss": 1.1417,
      "num_input_tokens_seen": 7984496,
      "step": 1619,
      "train_runtime": 5577.1275,
      "train_tokens_per_second": 1431.65
    },
    {
      "epoch": 0.5958620689655172,
      "grad_norm": 0.35429835319519043,
      "learning_rate": 8.106116433308771e-05,
      "loss": 1.27,
      "num_input_tokens_seen": 7988720,
      "step": 1620,
      "train_runtime": 5580.0287,
      "train_tokens_per_second": 1431.663
    },
    {
      "epoch": 0.5962298850574712,
      "grad_norm": 0.3473983108997345,
      "learning_rate": 8.098747236551216e-05,
      "loss": 1.1802,
      "num_input_tokens_seen": 7994096,
      "step": 1621,
      "train_runtime": 5585.7775,
      "train_tokens_per_second": 1431.152
    },
    {
      "epoch": 0.5965977011494253,
      "grad_norm": 0.35925814509391785,
      "learning_rate": 8.091378039793664e-05,
      "loss": 1.164,
      "num_input_tokens_seen": 7997472,
      "step": 1622,
      "train_runtime": 5588.2055,
      "train_tokens_per_second": 1431.134
    },
    {
      "epoch": 0.5969655172413794,
      "grad_norm": 0.3355098068714142,
      "learning_rate": 8.084008843036109e-05,
      "loss": 1.2693,
      "num_input_tokens_seen": 8004656,
      "step": 1623,
      "train_runtime": 5592.8504,
      "train_tokens_per_second": 1431.23
    },
    {
      "epoch": 0.5973333333333334,
      "grad_norm": 0.3596530854701996,
      "learning_rate": 8.076639646278557e-05,
      "loss": 1.4136,
      "num_input_tokens_seen": 8010496,
      "step": 1624,
      "train_runtime": 5596.6575,
      "train_tokens_per_second": 1431.3
    },
    {
      "epoch": 0.5977011494252874,
      "grad_norm": 0.35510095953941345,
      "learning_rate": 8.069270449521003e-05,
      "loss": 1.2209,
      "num_input_tokens_seen": 8015312,
      "step": 1625,
      "train_runtime": 5599.9028,
      "train_tokens_per_second": 1431.331
    },
    {
      "epoch": 0.5980689655172414,
      "grad_norm": 0.33866769075393677,
      "learning_rate": 8.06190125276345e-05,
      "loss": 1.1828,
      "num_input_tokens_seen": 8019280,
      "step": 1626,
      "train_runtime": 5602.6376,
      "train_tokens_per_second": 1431.34
    },
    {
      "epoch": 0.5984367816091954,
      "grad_norm": 0.40151500701904297,
      "learning_rate": 8.054532056005896e-05,
      "loss": 1.1759,
      "num_input_tokens_seen": 8024464,
      "step": 1627,
      "train_runtime": 5606.0198,
      "train_tokens_per_second": 1431.401
    },
    {
      "epoch": 0.5988045977011495,
      "grad_norm": 0.3653234541416168,
      "learning_rate": 8.047162859248343e-05,
      "loss": 1.1834,
      "num_input_tokens_seen": 8028960,
      "step": 1628,
      "train_runtime": 5609.0426,
      "train_tokens_per_second": 1431.431
    },
    {
      "epoch": 0.5991724137931035,
      "grad_norm": 0.34185248613357544,
      "learning_rate": 8.039793662490789e-05,
      "loss": 1.1726,
      "num_input_tokens_seen": 8033584,
      "step": 1629,
      "train_runtime": 5612.2245,
      "train_tokens_per_second": 1431.444
    },
    {
      "epoch": 0.5995402298850575,
      "grad_norm": 0.4381316304206848,
      "learning_rate": 8.032424465733236e-05,
      "loss": 1.3913,
      "num_input_tokens_seen": 8038368,
      "step": 1630,
      "train_runtime": 5615.5364,
      "train_tokens_per_second": 1431.452
    },
    {
      "epoch": 0.5999080459770115,
      "grad_norm": 0.33758002519607544,
      "learning_rate": 8.025055268975682e-05,
      "loss": 1.292,
      "num_input_tokens_seen": 8042352,
      "step": 1631,
      "train_runtime": 5618.3713,
      "train_tokens_per_second": 1431.438
    },
    {
      "epoch": 0.6002758620689655,
      "grad_norm": 1.249489426612854,
      "learning_rate": 8.017686072218129e-05,
      "loss": 1.2454,
      "num_input_tokens_seen": 8049712,
      "step": 1632,
      "train_runtime": 5623.0957,
      "train_tokens_per_second": 1431.545
    },
    {
      "epoch": 0.6002758620689655,
      "eval_loss": 1.2253098487854004,
      "eval_runtime": 15.5364,
      "eval_samples_per_second": 64.365,
      "eval_steps_per_second": 4.055,
      "num_input_tokens_seen": 8049712,
      "step": 1632
    },
    {
      "epoch": 0.6006436781609196,
      "grad_norm": 0.34538760781288147,
      "learning_rate": 8.010316875460575e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 8055312,
      "step": 1633,
      "train_runtime": 5642.3456,
      "train_tokens_per_second": 1427.653
    },
    {
      "epoch": 0.6010114942528736,
      "grad_norm": 0.3712161183357239,
      "learning_rate": 8.002947678703023e-05,
      "loss": 1.2901,
      "num_input_tokens_seen": 8060576,
      "step": 1634,
      "train_runtime": 5645.8862,
      "train_tokens_per_second": 1427.69
    },
    {
      "epoch": 0.6013793103448276,
      "grad_norm": 0.38065770268440247,
      "learning_rate": 7.995578481945468e-05,
      "loss": 1.339,
      "num_input_tokens_seen": 8069712,
      "step": 1635,
      "train_runtime": 5651.4722,
      "train_tokens_per_second": 1427.896
    },
    {
      "epoch": 0.6017471264367816,
      "grad_norm": 0.3502139151096344,
      "learning_rate": 7.988209285187916e-05,
      "loss": 1.3599,
      "num_input_tokens_seen": 8076080,
      "step": 1636,
      "train_runtime": 5655.5522,
      "train_tokens_per_second": 1427.991
    },
    {
      "epoch": 0.6021149425287357,
      "grad_norm": 0.37293684482574463,
      "learning_rate": 7.980840088430361e-05,
      "loss": 1.2456,
      "num_input_tokens_seen": 8081248,
      "step": 1637,
      "train_runtime": 5658.9642,
      "train_tokens_per_second": 1428.044
    },
    {
      "epoch": 0.6024827586206897,
      "grad_norm": 0.3323860764503479,
      "learning_rate": 7.973470891672809e-05,
      "loss": 1.2696,
      "num_input_tokens_seen": 8100240,
      "step": 1638,
      "train_runtime": 5670.0685,
      "train_tokens_per_second": 1428.597
    },
    {
      "epoch": 0.6028505747126437,
      "grad_norm": 0.37038612365722656,
      "learning_rate": 7.966101694915254e-05,
      "loss": 1.34,
      "num_input_tokens_seen": 8104528,
      "step": 1639,
      "train_runtime": 5673.0309,
      "train_tokens_per_second": 1428.606
    },
    {
      "epoch": 0.6032183908045977,
      "grad_norm": 0.41014528274536133,
      "learning_rate": 7.958732498157702e-05,
      "loss": 1.1197,
      "num_input_tokens_seen": 8109264,
      "step": 1640,
      "train_runtime": 5676.2848,
      "train_tokens_per_second": 1428.622
    },
    {
      "epoch": 0.6035862068965517,
      "grad_norm": 0.36411958932876587,
      "learning_rate": 7.951363301400148e-05,
      "loss": 1.1882,
      "num_input_tokens_seen": 8113520,
      "step": 1641,
      "train_runtime": 5679.2976,
      "train_tokens_per_second": 1428.613
    },
    {
      "epoch": 0.6039540229885058,
      "grad_norm": 0.3903307616710663,
      "learning_rate": 7.943994104642595e-05,
      "loss": 1.4,
      "num_input_tokens_seen": 8117360,
      "step": 1642,
      "train_runtime": 5682.0151,
      "train_tokens_per_second": 1428.606
    },
    {
      "epoch": 0.6043218390804598,
      "grad_norm": 0.34962713718414307,
      "learning_rate": 7.936624907885041e-05,
      "loss": 1.1911,
      "num_input_tokens_seen": 8121984,
      "step": 1643,
      "train_runtime": 5685.0339,
      "train_tokens_per_second": 1428.661
    },
    {
      "epoch": 0.6046896551724138,
      "grad_norm": 0.3581300675868988,
      "learning_rate": 7.929255711127488e-05,
      "loss": 1.343,
      "num_input_tokens_seen": 8127648,
      "step": 1644,
      "train_runtime": 5688.7355,
      "train_tokens_per_second": 1428.727
    },
    {
      "epoch": 0.6050574712643678,
      "grad_norm": 0.3512268662452698,
      "learning_rate": 7.921886514369934e-05,
      "loss": 1.211,
      "num_input_tokens_seen": 8132432,
      "step": 1645,
      "train_runtime": 5692.0152,
      "train_tokens_per_second": 1428.744
    },
    {
      "epoch": 0.6054252873563218,
      "grad_norm": 0.3820294439792633,
      "learning_rate": 7.91451731761238e-05,
      "loss": 1.2591,
      "num_input_tokens_seen": 8145056,
      "step": 1646,
      "train_runtime": 5699.6026,
      "train_tokens_per_second": 1429.057
    },
    {
      "epoch": 0.6057931034482759,
      "grad_norm": 0.366929292678833,
      "learning_rate": 7.907148120854827e-05,
      "loss": 1.3052,
      "num_input_tokens_seen": 8149600,
      "step": 1647,
      "train_runtime": 5702.7167,
      "train_tokens_per_second": 1429.073
    },
    {
      "epoch": 0.6061609195402299,
      "grad_norm": 0.4067031741142273,
      "learning_rate": 7.899778924097274e-05,
      "loss": 1.2315,
      "num_input_tokens_seen": 8153280,
      "step": 1648,
      "train_runtime": 5705.4079,
      "train_tokens_per_second": 1429.044
    },
    {
      "epoch": 0.6065287356321839,
      "grad_norm": 0.39430102705955505,
      "learning_rate": 7.89240972733972e-05,
      "loss": 1.1819,
      "num_input_tokens_seen": 8157264,
      "step": 1649,
      "train_runtime": 5708.2292,
      "train_tokens_per_second": 1429.036
    },
    {
      "epoch": 0.6068965517241379,
      "grad_norm": 0.3571094274520874,
      "learning_rate": 7.885040530582168e-05,
      "loss": 1.319,
      "num_input_tokens_seen": 8162848,
      "step": 1650,
      "train_runtime": 5711.9407,
      "train_tokens_per_second": 1429.085
    },
    {
      "epoch": 0.6072643678160919,
      "grad_norm": 0.40118053555488586,
      "learning_rate": 7.877671333824613e-05,
      "loss": 1.2625,
      "num_input_tokens_seen": 8166320,
      "step": 1651,
      "train_runtime": 5716.5158,
      "train_tokens_per_second": 1428.548
    },
    {
      "epoch": 0.607632183908046,
      "grad_norm": 0.3696447014808655,
      "learning_rate": 7.870302137067061e-05,
      "loss": 1.2195,
      "num_input_tokens_seen": 8172272,
      "step": 1652,
      "train_runtime": 5720.3147,
      "train_tokens_per_second": 1428.64
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.3668355345726013,
      "learning_rate": 7.862932940309506e-05,
      "loss": 1.3403,
      "num_input_tokens_seen": 8176416,
      "step": 1653,
      "train_runtime": 5723.1675,
      "train_tokens_per_second": 1428.652
    },
    {
      "epoch": 0.608367816091954,
      "grad_norm": 0.3291745185852051,
      "learning_rate": 7.855563743551954e-05,
      "loss": 1.1815,
      "num_input_tokens_seen": 8180896,
      "step": 1654,
      "train_runtime": 5726.2525,
      "train_tokens_per_second": 1428.665
    },
    {
      "epoch": 0.608735632183908,
      "grad_norm": 0.3443072736263275,
      "learning_rate": 7.848194546794399e-05,
      "loss": 1.1011,
      "num_input_tokens_seen": 8184224,
      "step": 1655,
      "train_runtime": 5728.7426,
      "train_tokens_per_second": 1428.625
    },
    {
      "epoch": 0.609103448275862,
      "grad_norm": 0.355884313583374,
      "learning_rate": 7.840825350036847e-05,
      "loss": 1.2486,
      "num_input_tokens_seen": 8188320,
      "step": 1656,
      "train_runtime": 5731.6319,
      "train_tokens_per_second": 1428.619
    },
    {
      "epoch": 0.6094712643678161,
      "grad_norm": 0.346994549036026,
      "learning_rate": 7.833456153279293e-05,
      "loss": 1.238,
      "num_input_tokens_seen": 8198144,
      "step": 1657,
      "train_runtime": 5737.627,
      "train_tokens_per_second": 1428.839
    },
    {
      "epoch": 0.6098390804597701,
      "grad_norm": 0.3324006199836731,
      "learning_rate": 7.82608695652174e-05,
      "loss": 1.245,
      "num_input_tokens_seen": 8202336,
      "step": 1658,
      "train_runtime": 5740.5104,
      "train_tokens_per_second": 1428.851
    },
    {
      "epoch": 0.6102068965517241,
      "grad_norm": 0.3342224955558777,
      "learning_rate": 7.818717759764186e-05,
      "loss": 1.1237,
      "num_input_tokens_seen": 8207344,
      "step": 1659,
      "train_runtime": 5743.8458,
      "train_tokens_per_second": 1428.894
    },
    {
      "epoch": 0.6105747126436781,
      "grad_norm": 0.3513737916946411,
      "learning_rate": 7.811348563006633e-05,
      "loss": 1.2363,
      "num_input_tokens_seen": 8211168,
      "step": 1660,
      "train_runtime": 5746.5795,
      "train_tokens_per_second": 1428.879
    },
    {
      "epoch": 0.6109425287356322,
      "grad_norm": 0.33271244168281555,
      "learning_rate": 7.803979366249079e-05,
      "loss": 1.2682,
      "num_input_tokens_seen": 8216992,
      "step": 1661,
      "train_runtime": 5750.3954,
      "train_tokens_per_second": 1428.944
    },
    {
      "epoch": 0.6113103448275862,
      "grad_norm": 0.330715537071228,
      "learning_rate": 7.796610169491526e-05,
      "loss": 1.1741,
      "num_input_tokens_seen": 8220848,
      "step": 1662,
      "train_runtime": 5753.1671,
      "train_tokens_per_second": 1428.926
    },
    {
      "epoch": 0.6116781609195402,
      "grad_norm": 0.3354067802429199,
      "learning_rate": 7.789240972733972e-05,
      "loss": 1.2572,
      "num_input_tokens_seen": 8227328,
      "step": 1663,
      "train_runtime": 5757.3748,
      "train_tokens_per_second": 1429.007
    },
    {
      "epoch": 0.6120459770114942,
      "grad_norm": 0.38026654720306396,
      "learning_rate": 7.781871775976419e-05,
      "loss": 1.2293,
      "num_input_tokens_seen": 8231152,
      "step": 1664,
      "train_runtime": 5760.0483,
      "train_tokens_per_second": 1429.007
    },
    {
      "epoch": 0.6124137931034482,
      "grad_norm": 0.38024336099624634,
      "learning_rate": 7.774502579218865e-05,
      "loss": 1.3161,
      "num_input_tokens_seen": 8237152,
      "step": 1665,
      "train_runtime": 5763.9649,
      "train_tokens_per_second": 1429.077
    },
    {
      "epoch": 0.6127816091954023,
      "grad_norm": 0.36663687229156494,
      "learning_rate": 7.767133382461313e-05,
      "loss": 1.2095,
      "num_input_tokens_seen": 8241568,
      "step": 1666,
      "train_runtime": 5767.0356,
      "train_tokens_per_second": 1429.082
    },
    {
      "epoch": 0.6131494252873563,
      "grad_norm": 0.35605117678642273,
      "learning_rate": 7.759764185703758e-05,
      "loss": 1.1832,
      "num_input_tokens_seen": 8248112,
      "step": 1667,
      "train_runtime": 5771.2867,
      "train_tokens_per_second": 1429.163
    },
    {
      "epoch": 0.6135172413793103,
      "grad_norm": 0.34538206458091736,
      "learning_rate": 7.752394988946206e-05,
      "loss": 1.2124,
      "num_input_tokens_seen": 8252032,
      "step": 1668,
      "train_runtime": 5773.9686,
      "train_tokens_per_second": 1429.179
    },
    {
      "epoch": 0.6138850574712643,
      "grad_norm": 0.37975063920021057,
      "learning_rate": 7.745025792188651e-05,
      "loss": 1.21,
      "num_input_tokens_seen": 8256160,
      "step": 1669,
      "train_runtime": 5776.8044,
      "train_tokens_per_second": 1429.192
    },
    {
      "epoch": 0.6142528735632183,
      "grad_norm": 0.385015070438385,
      "learning_rate": 7.737656595431099e-05,
      "loss": 1.2157,
      "num_input_tokens_seen": 8260176,
      "step": 1670,
      "train_runtime": 5779.6329,
      "train_tokens_per_second": 1429.187
    },
    {
      "epoch": 0.6146206896551724,
      "grad_norm": 0.3640705645084381,
      "learning_rate": 7.730287398673544e-05,
      "loss": 1.2995,
      "num_input_tokens_seen": 8266448,
      "step": 1671,
      "train_runtime": 5783.7064,
      "train_tokens_per_second": 1429.265
    },
    {
      "epoch": 0.6149885057471265,
      "grad_norm": 0.3843008279800415,
      "learning_rate": 7.722918201915992e-05,
      "loss": 1.1236,
      "num_input_tokens_seen": 8270528,
      "step": 1672,
      "train_runtime": 5786.5618,
      "train_tokens_per_second": 1429.265
    },
    {
      "epoch": 0.6153563218390805,
      "grad_norm": 0.3321741223335266,
      "learning_rate": 7.715549005158438e-05,
      "loss": 1.1477,
      "num_input_tokens_seen": 8274448,
      "step": 1673,
      "train_runtime": 5789.318,
      "train_tokens_per_second": 1429.261
    },
    {
      "epoch": 0.6157241379310345,
      "grad_norm": 0.363680899143219,
      "learning_rate": 7.708179808400885e-05,
      "loss": 1.4119,
      "num_input_tokens_seen": 8278752,
      "step": 1674,
      "train_runtime": 5792.3767,
      "train_tokens_per_second": 1429.25
    },
    {
      "epoch": 0.6160919540229886,
      "grad_norm": 0.3656877875328064,
      "learning_rate": 7.700810611643331e-05,
      "loss": 1.3734,
      "num_input_tokens_seen": 8282112,
      "step": 1675,
      "train_runtime": 5794.8791,
      "train_tokens_per_second": 1429.212
    },
    {
      "epoch": 0.6164597701149426,
      "grad_norm": 0.3490371108055115,
      "learning_rate": 7.693441414885778e-05,
      "loss": 1.0974,
      "num_input_tokens_seen": 8291488,
      "step": 1676,
      "train_runtime": 5800.5803,
      "train_tokens_per_second": 1429.424
    },
    {
      "epoch": 0.6168275862068966,
      "grad_norm": 0.35831621289253235,
      "learning_rate": 7.686072218128224e-05,
      "loss": 1.2945,
      "num_input_tokens_seen": 8294896,
      "step": 1677,
      "train_runtime": 5803.2077,
      "train_tokens_per_second": 1429.364
    },
    {
      "epoch": 0.6171954022988506,
      "grad_norm": 0.3728373050689697,
      "learning_rate": 7.678703021370671e-05,
      "loss": 1.2804,
      "num_input_tokens_seen": 8299376,
      "step": 1678,
      "train_runtime": 5806.2739,
      "train_tokens_per_second": 1429.381
    },
    {
      "epoch": 0.6175632183908046,
      "grad_norm": 0.35205045342445374,
      "learning_rate": 7.671333824613117e-05,
      "loss": 1.1876,
      "num_input_tokens_seen": 8306080,
      "step": 1679,
      "train_runtime": 5810.491,
      "train_tokens_per_second": 1429.497
    },
    {
      "epoch": 0.6179310344827587,
      "grad_norm": 0.34353363513946533,
      "learning_rate": 7.663964627855564e-05,
      "loss": 1.2488,
      "num_input_tokens_seen": 8311088,
      "step": 1680,
      "train_runtime": 5813.894,
      "train_tokens_per_second": 1429.522
    },
    {
      "epoch": 0.6182988505747127,
      "grad_norm": 0.3545430600643158,
      "learning_rate": 7.65659543109801e-05,
      "loss": 1.2714,
      "num_input_tokens_seen": 8316720,
      "step": 1681,
      "train_runtime": 5819.7315,
      "train_tokens_per_second": 1429.056
    },
    {
      "epoch": 0.6186666666666667,
      "grad_norm": 0.3599246144294739,
      "learning_rate": 7.649226234340458e-05,
      "loss": 1.0997,
      "num_input_tokens_seen": 8320144,
      "step": 1682,
      "train_runtime": 5822.2177,
      "train_tokens_per_second": 1429.033
    },
    {
      "epoch": 0.6190344827586207,
      "grad_norm": 0.33515429496765137,
      "learning_rate": 7.641857037582903e-05,
      "loss": 1.254,
      "num_input_tokens_seen": 8324192,
      "step": 1683,
      "train_runtime": 5825.0793,
      "train_tokens_per_second": 1429.026
    },
    {
      "epoch": 0.6194022988505747,
      "grad_norm": 0.35460397601127625,
      "learning_rate": 7.634487840825351e-05,
      "loss": 1.1662,
      "num_input_tokens_seen": 8328320,
      "step": 1684,
      "train_runtime": 5827.9777,
      "train_tokens_per_second": 1429.024
    },
    {
      "epoch": 0.6197701149425288,
      "grad_norm": 0.33717334270477295,
      "learning_rate": 7.627118644067796e-05,
      "loss": 1.1742,
      "num_input_tokens_seen": 8332064,
      "step": 1685,
      "train_runtime": 5830.6363,
      "train_tokens_per_second": 1429.015
    },
    {
      "epoch": 0.6201379310344828,
      "grad_norm": 0.39270511269569397,
      "learning_rate": 7.619749447310244e-05,
      "loss": 1.2929,
      "num_input_tokens_seen": 8336096,
      "step": 1686,
      "train_runtime": 5833.4447,
      "train_tokens_per_second": 1429.018
    },
    {
      "epoch": 0.6205057471264368,
      "grad_norm": 0.354690283536911,
      "learning_rate": 7.612380250552689e-05,
      "loss": 1.1895,
      "num_input_tokens_seen": 8339424,
      "step": 1687,
      "train_runtime": 5835.9405,
      "train_tokens_per_second": 1428.977
    },
    {
      "epoch": 0.6208735632183908,
      "grad_norm": 0.34419217705726624,
      "learning_rate": 7.605011053795137e-05,
      "loss": 1.2396,
      "num_input_tokens_seen": 8344176,
      "step": 1688,
      "train_runtime": 5839.2142,
      "train_tokens_per_second": 1428.99
    },
    {
      "epoch": 0.6212413793103448,
      "grad_norm": 0.38825803995132446,
      "learning_rate": 7.597641857037583e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 8352576,
      "step": 1689,
      "train_runtime": 5844.4953,
      "train_tokens_per_second": 1429.136
    },
    {
      "epoch": 0.6216091954022989,
      "grad_norm": 0.32436880469322205,
      "learning_rate": 7.59027266028003e-05,
      "loss": 1.1041,
      "num_input_tokens_seen": 8357584,
      "step": 1690,
      "train_runtime": 5847.8726,
      "train_tokens_per_second": 1429.167
    },
    {
      "epoch": 0.6219770114942529,
      "grad_norm": 0.35893768072128296,
      "learning_rate": 7.582903463522476e-05,
      "loss": 1.1916,
      "num_input_tokens_seen": 8361472,
      "step": 1691,
      "train_runtime": 5850.5938,
      "train_tokens_per_second": 1429.166
    },
    {
      "epoch": 0.6223448275862069,
      "grad_norm": 0.3316190540790558,
      "learning_rate": 7.575534266764923e-05,
      "loss": 1.1044,
      "num_input_tokens_seen": 8365648,
      "step": 1692,
      "train_runtime": 5853.4557,
      "train_tokens_per_second": 1429.181
    },
    {
      "epoch": 0.6227126436781609,
      "grad_norm": 0.34904029965400696,
      "learning_rate": 7.56816507000737e-05,
      "loss": 1.1768,
      "num_input_tokens_seen": 8369600,
      "step": 1693,
      "train_runtime": 5856.2494,
      "train_tokens_per_second": 1429.174
    },
    {
      "epoch": 0.623080459770115,
      "grad_norm": 0.34622156620025635,
      "learning_rate": 7.560795873249816e-05,
      "loss": 1.1642,
      "num_input_tokens_seen": 8372896,
      "step": 1694,
      "train_runtime": 5858.7018,
      "train_tokens_per_second": 1429.138
    },
    {
      "epoch": 0.623448275862069,
      "grad_norm": 0.39177048206329346,
      "learning_rate": 7.553426676492262e-05,
      "loss": 1.4457,
      "num_input_tokens_seen": 8377296,
      "step": 1695,
      "train_runtime": 5861.7435,
      "train_tokens_per_second": 1429.147
    },
    {
      "epoch": 0.623816091954023,
      "grad_norm": 0.36550185084342957,
      "learning_rate": 7.546057479734709e-05,
      "loss": 1.2474,
      "num_input_tokens_seen": 8384384,
      "step": 1696,
      "train_runtime": 5866.1669,
      "train_tokens_per_second": 1429.278
    },
    {
      "epoch": 0.624183908045977,
      "grad_norm": 0.3592674136161804,
      "learning_rate": 7.538688282977155e-05,
      "loss": 1.1728,
      "num_input_tokens_seen": 8389152,
      "step": 1697,
      "train_runtime": 5869.4197,
      "train_tokens_per_second": 1429.298
    },
    {
      "epoch": 0.624551724137931,
      "grad_norm": 0.36256006360054016,
      "learning_rate": 7.531319086219603e-05,
      "loss": 1.1662,
      "num_input_tokens_seen": 8395920,
      "step": 1698,
      "train_runtime": 5873.7059,
      "train_tokens_per_second": 1429.408
    },
    {
      "epoch": 0.624919540229885,
      "grad_norm": 0.37653040885925293,
      "learning_rate": 7.523949889462048e-05,
      "loss": 1.3271,
      "num_input_tokens_seen": 8399632,
      "step": 1699,
      "train_runtime": 5876.417,
      "train_tokens_per_second": 1429.38
    },
    {
      "epoch": 0.6252873563218391,
      "grad_norm": 0.366256982088089,
      "learning_rate": 7.516580692704496e-05,
      "loss": 1.1338,
      "num_input_tokens_seen": 8403280,
      "step": 1700,
      "train_runtime": 5879.1221,
      "train_tokens_per_second": 1429.343
    },
    {
      "epoch": 0.6256551724137931,
      "grad_norm": 0.372652143239975,
      "learning_rate": 7.509211495946941e-05,
      "loss": 1.2001,
      "num_input_tokens_seen": 8406480,
      "step": 1701,
      "train_runtime": 5881.503,
      "train_tokens_per_second": 1429.308
    },
    {
      "epoch": 0.6260229885057471,
      "grad_norm": 0.3536698520183563,
      "learning_rate": 7.501842299189389e-05,
      "loss": 1.1611,
      "num_input_tokens_seen": 8414080,
      "step": 1702,
      "train_runtime": 5886.3073,
      "train_tokens_per_second": 1429.433
    },
    {
      "epoch": 0.6263908045977011,
      "grad_norm": 0.38645151257514954,
      "learning_rate": 7.494473102431834e-05,
      "loss": 1.2416,
      "num_input_tokens_seen": 8418352,
      "step": 1703,
      "train_runtime": 5889.2821,
      "train_tokens_per_second": 1429.436
    },
    {
      "epoch": 0.6267586206896552,
      "grad_norm": 0.44208791851997375,
      "learning_rate": 7.487103905674282e-05,
      "loss": 1.2463,
      "num_input_tokens_seen": 8423376,
      "step": 1704,
      "train_runtime": 5892.592,
      "train_tokens_per_second": 1429.486
    },
    {
      "epoch": 0.6271264367816092,
      "grad_norm": 0.36852338910102844,
      "learning_rate": 7.479734708916729e-05,
      "loss": 1.2882,
      "num_input_tokens_seen": 8429136,
      "step": 1705,
      "train_runtime": 5896.3849,
      "train_tokens_per_second": 1429.543
    },
    {
      "epoch": 0.6274942528735632,
      "grad_norm": 0.44526517391204834,
      "learning_rate": 7.472365512159175e-05,
      "loss": 1.1301,
      "num_input_tokens_seen": 8438560,
      "step": 1706,
      "train_runtime": 5902.2457,
      "train_tokens_per_second": 1429.72
    },
    {
      "epoch": 0.6278620689655172,
      "grad_norm": 0.3883841931819916,
      "learning_rate": 7.464996315401621e-05,
      "loss": 1.2752,
      "num_input_tokens_seen": 8442128,
      "step": 1707,
      "train_runtime": 5904.8067,
      "train_tokens_per_second": 1429.704
    },
    {
      "epoch": 0.6282298850574712,
      "grad_norm": 0.37330782413482666,
      "learning_rate": 7.457627118644068e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 8447296,
      "step": 1708,
      "train_runtime": 5908.2233,
      "train_tokens_per_second": 1429.752
    },
    {
      "epoch": 0.6285977011494253,
      "grad_norm": 0.3426826000213623,
      "learning_rate": 7.450257921886514e-05,
      "loss": 1.224,
      "num_input_tokens_seen": 8451120,
      "step": 1709,
      "train_runtime": 5910.8956,
      "train_tokens_per_second": 1429.753
    },
    {
      "epoch": 0.6289655172413793,
      "grad_norm": 0.33743831515312195,
      "learning_rate": 7.442888725128961e-05,
      "loss": 1.094,
      "num_input_tokens_seen": 8455648,
      "step": 1710,
      "train_runtime": 5914.0251,
      "train_tokens_per_second": 1429.762
    },
    {
      "epoch": 0.6293333333333333,
      "grad_norm": 0.33115074038505554,
      "learning_rate": 7.435519528371407e-05,
      "loss": 1.1234,
      "num_input_tokens_seen": 8460160,
      "step": 1711,
      "train_runtime": 5919.2328,
      "train_tokens_per_second": 1429.266
    },
    {
      "epoch": 0.6297011494252873,
      "grad_norm": 0.3846089243888855,
      "learning_rate": 7.428150331613854e-05,
      "loss": 1.2594,
      "num_input_tokens_seen": 8466016,
      "step": 1712,
      "train_runtime": 5923.0893,
      "train_tokens_per_second": 1429.324
    },
    {
      "epoch": 0.6300689655172413,
      "grad_norm": 0.37581756711006165,
      "learning_rate": 7.4207811348563e-05,
      "loss": 1.2047,
      "num_input_tokens_seen": 8470976,
      "step": 1713,
      "train_runtime": 5926.4594,
      "train_tokens_per_second": 1429.349
    },
    {
      "epoch": 0.6304367816091954,
      "grad_norm": 0.3285261392593384,
      "learning_rate": 7.413411938098748e-05,
      "loss": 1.1702,
      "num_input_tokens_seen": 8475072,
      "step": 1714,
      "train_runtime": 5929.2624,
      "train_tokens_per_second": 1429.364
    },
    {
      "epoch": 0.6308045977011494,
      "grad_norm": 0.3255341947078705,
      "learning_rate": 7.406042741341193e-05,
      "loss": 1.1993,
      "num_input_tokens_seen": 8480336,
      "step": 1715,
      "train_runtime": 5932.8379,
      "train_tokens_per_second": 1429.389
    },
    {
      "epoch": 0.6311724137931034,
      "grad_norm": 0.3650467097759247,
      "learning_rate": 7.398673544583641e-05,
      "loss": 1.1929,
      "num_input_tokens_seen": 8485184,
      "step": 1716,
      "train_runtime": 5936.0374,
      "train_tokens_per_second": 1429.436
    },
    {
      "epoch": 0.6315402298850574,
      "grad_norm": 0.32892972230911255,
      "learning_rate": 7.391304347826086e-05,
      "loss": 1.1589,
      "num_input_tokens_seen": 8488448,
      "step": 1717,
      "train_runtime": 5938.468,
      "train_tokens_per_second": 1429.4
    },
    {
      "epoch": 0.6319080459770114,
      "grad_norm": 0.37333914637565613,
      "learning_rate": 7.383935151068534e-05,
      "loss": 1.275,
      "num_input_tokens_seen": 8493840,
      "step": 1718,
      "train_runtime": 5942.016,
      "train_tokens_per_second": 1429.454
    },
    {
      "epoch": 0.6322758620689655,
      "grad_norm": 0.3702186644077301,
      "learning_rate": 7.37656595431098e-05,
      "loss": 1.2836,
      "num_input_tokens_seen": 8498704,
      "step": 1719,
      "train_runtime": 5945.371,
      "train_tokens_per_second": 1429.466
    },
    {
      "epoch": 0.6326436781609195,
      "grad_norm": 0.39659491181373596,
      "learning_rate": 7.369196757553427e-05,
      "loss": 1.2489,
      "num_input_tokens_seen": 8502144,
      "step": 1720,
      "train_runtime": 5947.8739,
      "train_tokens_per_second": 1429.443
    },
    {
      "epoch": 0.6330114942528735,
      "grad_norm": 0.34572821855545044,
      "learning_rate": 7.361827560795874e-05,
      "loss": 1.2434,
      "num_input_tokens_seen": 8505824,
      "step": 1721,
      "train_runtime": 5950.5455,
      "train_tokens_per_second": 1429.419
    },
    {
      "epoch": 0.6333793103448276,
      "grad_norm": 0.3726195991039276,
      "learning_rate": 7.35445836403832e-05,
      "loss": 1.2399,
      "num_input_tokens_seen": 8510784,
      "step": 1722,
      "train_runtime": 5953.8061,
      "train_tokens_per_second": 1429.469
    },
    {
      "epoch": 0.6337471264367817,
      "grad_norm": 0.35379770398139954,
      "learning_rate": 7.347089167280767e-05,
      "loss": 1.3305,
      "num_input_tokens_seen": 8514208,
      "step": 1723,
      "train_runtime": 5956.3298,
      "train_tokens_per_second": 1429.439
    },
    {
      "epoch": 0.6341149425287357,
      "grad_norm": 0.36143144965171814,
      "learning_rate": 7.339719970523213e-05,
      "loss": 1.1167,
      "num_input_tokens_seen": 8518400,
      "step": 1724,
      "train_runtime": 5959.2424,
      "train_tokens_per_second": 1429.443
    },
    {
      "epoch": 0.6344827586206897,
      "grad_norm": 0.3915243148803711,
      "learning_rate": 7.33235077376566e-05,
      "loss": 1.3042,
      "num_input_tokens_seen": 8524928,
      "step": 1725,
      "train_runtime": 5963.3754,
      "train_tokens_per_second": 1429.547
    },
    {
      "epoch": 0.6348505747126437,
      "grad_norm": 0.3500584363937378,
      "learning_rate": 7.324981577008106e-05,
      "loss": 1.1139,
      "num_input_tokens_seen": 8534320,
      "step": 1726,
      "train_runtime": 5969.1137,
      "train_tokens_per_second": 1429.747
    },
    {
      "epoch": 0.6352183908045977,
      "grad_norm": 0.41126441955566406,
      "learning_rate": 7.317612380250552e-05,
      "loss": 1.2846,
      "num_input_tokens_seen": 8540400,
      "step": 1727,
      "train_runtime": 5973.053,
      "train_tokens_per_second": 1429.822
    },
    {
      "epoch": 0.6355862068965518,
      "grad_norm": 0.323826402425766,
      "learning_rate": 7.310243183492999e-05,
      "loss": 1.2211,
      "num_input_tokens_seen": 8547040,
      "step": 1728,
      "train_runtime": 5977.3366,
      "train_tokens_per_second": 1429.908
    },
    {
      "epoch": 0.6359540229885058,
      "grad_norm": 0.3922103941440582,
      "learning_rate": 7.302873986735445e-05,
      "loss": 1.2249,
      "num_input_tokens_seen": 8553584,
      "step": 1729,
      "train_runtime": 5981.6217,
      "train_tokens_per_second": 1429.977
    },
    {
      "epoch": 0.6363218390804598,
      "grad_norm": 0.3563559651374817,
      "learning_rate": 7.295504789977893e-05,
      "loss": 1.1275,
      "num_input_tokens_seen": 8562672,
      "step": 1730,
      "train_runtime": 5987.2531,
      "train_tokens_per_second": 1430.15
    },
    {
      "epoch": 0.6366896551724138,
      "grad_norm": 0.34606972336769104,
      "learning_rate": 7.288135593220338e-05,
      "loss": 1.1647,
      "num_input_tokens_seen": 8567040,
      "step": 1731,
      "train_runtime": 5990.2886,
      "train_tokens_per_second": 1430.155
    },
    {
      "epoch": 0.6370574712643678,
      "grad_norm": 0.36875972151756287,
      "learning_rate": 7.280766396462786e-05,
      "loss": 1.2265,
      "num_input_tokens_seen": 8571376,
      "step": 1732,
      "train_runtime": 5993.2536,
      "train_tokens_per_second": 1430.171
    },
    {
      "epoch": 0.6374252873563219,
      "grad_norm": 0.3302783668041229,
      "learning_rate": 7.273397199705231e-05,
      "loss": 1.2149,
      "num_input_tokens_seen": 8579904,
      "step": 1733,
      "train_runtime": 5998.5997,
      "train_tokens_per_second": 1430.318
    },
    {
      "epoch": 0.6377931034482759,
      "grad_norm": 0.3487246036529541,
      "learning_rate": 7.266028002947679e-05,
      "loss": 1.2711,
      "num_input_tokens_seen": 8585664,
      "step": 1734,
      "train_runtime": 6002.3843,
      "train_tokens_per_second": 1430.376
    },
    {
      "epoch": 0.6381609195402299,
      "grad_norm": 0.34984487295150757,
      "learning_rate": 7.258658806190126e-05,
      "loss": 1.0415,
      "num_input_tokens_seen": 8589168,
      "step": 1735,
      "train_runtime": 6004.8669,
      "train_tokens_per_second": 1430.368
    },
    {
      "epoch": 0.6385287356321839,
      "grad_norm": 0.33944281935691833,
      "learning_rate": 7.251289609432572e-05,
      "loss": 1.2221,
      "num_input_tokens_seen": 8593280,
      "step": 1736,
      "train_runtime": 6007.7107,
      "train_tokens_per_second": 1430.375
    },
    {
      "epoch": 0.638896551724138,
      "grad_norm": 0.3744630515575409,
      "learning_rate": 7.243920412675019e-05,
      "loss": 1.3095,
      "num_input_tokens_seen": 8597760,
      "step": 1737,
      "train_runtime": 6010.8044,
      "train_tokens_per_second": 1430.384
    },
    {
      "epoch": 0.639264367816092,
      "grad_norm": 0.38167956471443176,
      "learning_rate": 7.236551215917465e-05,
      "loss": 1.2901,
      "num_input_tokens_seen": 8608656,
      "step": 1738,
      "train_runtime": 6017.3674,
      "train_tokens_per_second": 1430.635
    },
    {
      "epoch": 0.639632183908046,
      "grad_norm": 0.35250505805015564,
      "learning_rate": 7.229182019159913e-05,
      "loss": 1.2224,
      "num_input_tokens_seen": 8611776,
      "step": 1739,
      "train_runtime": 6019.6915,
      "train_tokens_per_second": 1430.601
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3379589319229126,
      "learning_rate": 7.221812822402358e-05,
      "loss": 1.0406,
      "num_input_tokens_seen": 8616384,
      "step": 1740,
      "train_runtime": 6022.7709,
      "train_tokens_per_second": 1430.635
    },
    {
      "epoch": 0.640367816091954,
      "grad_norm": 0.4123132526874542,
      "learning_rate": 7.214443625644806e-05,
      "loss": 1.3314,
      "num_input_tokens_seen": 8620400,
      "step": 1741,
      "train_runtime": 6027.706,
      "train_tokens_per_second": 1430.129
    },
    {
      "epoch": 0.640735632183908,
      "grad_norm": 0.38162675499916077,
      "learning_rate": 7.207074428887251e-05,
      "loss": 1.2229,
      "num_input_tokens_seen": 8624976,
      "step": 1742,
      "train_runtime": 6030.8467,
      "train_tokens_per_second": 1430.143
    },
    {
      "epoch": 0.6411034482758621,
      "grad_norm": 0.3775383234024048,
      "learning_rate": 7.199705232129699e-05,
      "loss": 1.3291,
      "num_input_tokens_seen": 8629472,
      "step": 1743,
      "train_runtime": 6033.9782,
      "train_tokens_per_second": 1430.146
    },
    {
      "epoch": 0.6414712643678161,
      "grad_norm": 0.36281177401542664,
      "learning_rate": 7.192336035372145e-05,
      "loss": 1.2229,
      "num_input_tokens_seen": 8634112,
      "step": 1744,
      "train_runtime": 6037.1953,
      "train_tokens_per_second": 1430.153
    },
    {
      "epoch": 0.6418390804597701,
      "grad_norm": 0.38396233320236206,
      "learning_rate": 7.184966838614592e-05,
      "loss": 1.2697,
      "num_input_tokens_seen": 8641824,
      "step": 1745,
      "train_runtime": 6042.0258,
      "train_tokens_per_second": 1430.286
    },
    {
      "epoch": 0.6422068965517241,
      "grad_norm": 0.39428532123565674,
      "learning_rate": 7.177597641857038e-05,
      "loss": 1.2465,
      "num_input_tokens_seen": 8645904,
      "step": 1746,
      "train_runtime": 6044.8586,
      "train_tokens_per_second": 1430.291
    },
    {
      "epoch": 0.6425747126436782,
      "grad_norm": 0.3777850568294525,
      "learning_rate": 7.170228445099485e-05,
      "loss": 1.2897,
      "num_input_tokens_seen": 8649856,
      "step": 1747,
      "train_runtime": 6047.5995,
      "train_tokens_per_second": 1430.296
    },
    {
      "epoch": 0.6429425287356322,
      "grad_norm": 0.37085989117622375,
      "learning_rate": 7.162859248341931e-05,
      "loss": 1.3786,
      "num_input_tokens_seen": 8654768,
      "step": 1748,
      "train_runtime": 6050.8791,
      "train_tokens_per_second": 1430.332
    },
    {
      "epoch": 0.6433103448275862,
      "grad_norm": 0.35489970445632935,
      "learning_rate": 7.155490051584378e-05,
      "loss": 1.3616,
      "num_input_tokens_seen": 8658880,
      "step": 1749,
      "train_runtime": 6053.7682,
      "train_tokens_per_second": 1430.329
    },
    {
      "epoch": 0.6436781609195402,
      "grad_norm": 0.37047791481018066,
      "learning_rate": 7.148120854826824e-05,
      "loss": 1.3575,
      "num_input_tokens_seen": 8662144,
      "step": 1750,
      "train_runtime": 6056.1594,
      "train_tokens_per_second": 1430.303
    },
    {
      "epoch": 0.6440459770114942,
      "grad_norm": 0.3515414297580719,
      "learning_rate": 7.140751658069271e-05,
      "loss": 1.354,
      "num_input_tokens_seen": 8666096,
      "step": 1751,
      "train_runtime": 6058.9181,
      "train_tokens_per_second": 1430.304
    },
    {
      "epoch": 0.6444137931034483,
      "grad_norm": 0.398267924785614,
      "learning_rate": 7.133382461311717e-05,
      "loss": 1.1688,
      "num_input_tokens_seen": 8670672,
      "step": 1752,
      "train_runtime": 6062.0909,
      "train_tokens_per_second": 1430.31
    },
    {
      "epoch": 0.6447816091954023,
      "grad_norm": 0.35337409377098083,
      "learning_rate": 7.126013264554164e-05,
      "loss": 1.2443,
      "num_input_tokens_seen": 8674656,
      "step": 1753,
      "train_runtime": 6064.9413,
      "train_tokens_per_second": 1430.295
    },
    {
      "epoch": 0.6451494252873563,
      "grad_norm": 0.3913950026035309,
      "learning_rate": 7.11864406779661e-05,
      "loss": 1.2636,
      "num_input_tokens_seen": 8678352,
      "step": 1754,
      "train_runtime": 6067.5314,
      "train_tokens_per_second": 1430.294
    },
    {
      "epoch": 0.6455172413793103,
      "grad_norm": 0.3143499791622162,
      "learning_rate": 7.111274871039058e-05,
      "loss": 1.1459,
      "num_input_tokens_seen": 8684656,
      "step": 1755,
      "train_runtime": 6071.6067,
      "train_tokens_per_second": 1430.372
    },
    {
      "epoch": 0.6458850574712643,
      "grad_norm": 0.3837677538394928,
      "learning_rate": 7.103905674281503e-05,
      "loss": 1.1757,
      "num_input_tokens_seen": 8687936,
      "step": 1756,
      "train_runtime": 6074.0305,
      "train_tokens_per_second": 1430.341
    },
    {
      "epoch": 0.6462528735632184,
      "grad_norm": 0.34883350133895874,
      "learning_rate": 7.096536477523951e-05,
      "loss": 1.1783,
      "num_input_tokens_seen": 8692688,
      "step": 1757,
      "train_runtime": 6077.2516,
      "train_tokens_per_second": 1430.365
    },
    {
      "epoch": 0.6466206896551724,
      "grad_norm": 0.37379708886146545,
      "learning_rate": 7.089167280766396e-05,
      "loss": 1.1724,
      "num_input_tokens_seen": 8697776,
      "step": 1758,
      "train_runtime": 6080.6046,
      "train_tokens_per_second": 1430.413
    },
    {
      "epoch": 0.6469885057471264,
      "grad_norm": 0.3478778600692749,
      "learning_rate": 7.081798084008844e-05,
      "loss": 1.1709,
      "num_input_tokens_seen": 8702768,
      "step": 1759,
      "train_runtime": 6084.0177,
      "train_tokens_per_second": 1430.431
    },
    {
      "epoch": 0.6473563218390804,
      "grad_norm": 0.3637995719909668,
      "learning_rate": 7.07442888725129e-05,
      "loss": 1.2735,
      "num_input_tokens_seen": 8708448,
      "step": 1760,
      "train_runtime": 6087.7386,
      "train_tokens_per_second": 1430.49
    },
    {
      "epoch": 0.6477241379310344,
      "grad_norm": 0.3367466926574707,
      "learning_rate": 7.067059690493737e-05,
      "loss": 1.1286,
      "num_input_tokens_seen": 8712896,
      "step": 1761,
      "train_runtime": 6090.7652,
      "train_tokens_per_second": 1430.509
    },
    {
      "epoch": 0.6480919540229885,
      "grad_norm": 0.3800182342529297,
      "learning_rate": 7.059690493736183e-05,
      "loss": 1.1854,
      "num_input_tokens_seen": 8719328,
      "step": 1762,
      "train_runtime": 6094.9137,
      "train_tokens_per_second": 1430.591
    },
    {
      "epoch": 0.6484597701149425,
      "grad_norm": 0.40403521060943604,
      "learning_rate": 7.05232129697863e-05,
      "loss": 1.3117,
      "num_input_tokens_seen": 8724560,
      "step": 1763,
      "train_runtime": 6098.4609,
      "train_tokens_per_second": 1430.617
    },
    {
      "epoch": 0.6488275862068965,
      "grad_norm": 0.3457203507423401,
      "learning_rate": 7.044952100221076e-05,
      "loss": 1.1243,
      "num_input_tokens_seen": 8728928,
      "step": 1764,
      "train_runtime": 6101.4944,
      "train_tokens_per_second": 1430.621
    },
    {
      "epoch": 0.6491954022988505,
      "grad_norm": 0.3725492060184479,
      "learning_rate": 7.037582903463523e-05,
      "loss": 1.2636,
      "num_input_tokens_seen": 8733952,
      "step": 1765,
      "train_runtime": 6104.8527,
      "train_tokens_per_second": 1430.657
    },
    {
      "epoch": 0.6495632183908046,
      "grad_norm": 0.34399518370628357,
      "learning_rate": 7.03021370670597e-05,
      "loss": 1.2447,
      "num_input_tokens_seen": 8739456,
      "step": 1766,
      "train_runtime": 6108.5181,
      "train_tokens_per_second": 1430.7
    },
    {
      "epoch": 0.6499310344827586,
      "grad_norm": 0.4023468494415283,
      "learning_rate": 7.022844509948416e-05,
      "loss": 1.2799,
      "num_input_tokens_seen": 8743008,
      "step": 1767,
      "train_runtime": 6111.0663,
      "train_tokens_per_second": 1430.685
    },
    {
      "epoch": 0.6502988505747126,
      "grad_norm": 0.3690298795700073,
      "learning_rate": 7.015475313190862e-05,
      "loss": 1.1633,
      "num_input_tokens_seen": 8746288,
      "step": 1768,
      "train_runtime": 6113.4582,
      "train_tokens_per_second": 1430.661
    },
    {
      "epoch": 0.6506666666666666,
      "grad_norm": 0.360746830701828,
      "learning_rate": 7.00810611643331e-05,
      "loss": 1.2309,
      "num_input_tokens_seen": 8749696,
      "step": 1769,
      "train_runtime": 6115.9301,
      "train_tokens_per_second": 1430.64
    },
    {
      "epoch": 0.6510344827586206,
      "grad_norm": 0.35331016778945923,
      "learning_rate": 7.000736919675755e-05,
      "loss": 1.2066,
      "num_input_tokens_seen": 8753840,
      "step": 1770,
      "train_runtime": 6118.8282,
      "train_tokens_per_second": 1430.64
    },
    {
      "epoch": 0.6514022988505748,
      "grad_norm": 0.31782907247543335,
      "learning_rate": 6.993367722918203e-05,
      "loss": 1.2213,
      "num_input_tokens_seen": 8760928,
      "step": 1771,
      "train_runtime": 6125.3261,
      "train_tokens_per_second": 1430.279
    },
    {
      "epoch": 0.6517701149425288,
      "grad_norm": 0.35930871963500977,
      "learning_rate": 6.985998526160648e-05,
      "loss": 1.1641,
      "num_input_tokens_seen": 8766544,
      "step": 1772,
      "train_runtime": 6129.1413,
      "train_tokens_per_second": 1430.305
    },
    {
      "epoch": 0.6521379310344828,
      "grad_norm": 0.3347841799259186,
      "learning_rate": 6.978629329403096e-05,
      "loss": 1.2085,
      "num_input_tokens_seen": 8771408,
      "step": 1773,
      "train_runtime": 6132.4403,
      "train_tokens_per_second": 1430.329
    },
    {
      "epoch": 0.6525057471264368,
      "grad_norm": 0.3424447178840637,
      "learning_rate": 6.971260132645541e-05,
      "loss": 1.2922,
      "num_input_tokens_seen": 8775616,
      "step": 1774,
      "train_runtime": 6135.3754,
      "train_tokens_per_second": 1430.331
    },
    {
      "epoch": 0.6528735632183909,
      "grad_norm": 0.36627665162086487,
      "learning_rate": 6.963890935887989e-05,
      "loss": 1.1149,
      "num_input_tokens_seen": 8780352,
      "step": 1775,
      "train_runtime": 6138.555,
      "train_tokens_per_second": 1430.361
    },
    {
      "epoch": 0.6532413793103449,
      "grad_norm": 0.37700387835502625,
      "learning_rate": 6.956521739130436e-05,
      "loss": 1.2168,
      "num_input_tokens_seen": 8788112,
      "step": 1776,
      "train_runtime": 6143.4436,
      "train_tokens_per_second": 1430.486
    },
    {
      "epoch": 0.6536091954022989,
      "grad_norm": 0.33836108446121216,
      "learning_rate": 6.949152542372882e-05,
      "loss": 1.1739,
      "num_input_tokens_seen": 8792128,
      "step": 1777,
      "train_runtime": 6146.168,
      "train_tokens_per_second": 1430.506
    },
    {
      "epoch": 0.6539770114942529,
      "grad_norm": 0.3440338671207428,
      "learning_rate": 6.941783345615328e-05,
      "loss": 1.3012,
      "num_input_tokens_seen": 8797120,
      "step": 1778,
      "train_runtime": 6149.5407,
      "train_tokens_per_second": 1430.533
    },
    {
      "epoch": 0.6543448275862069,
      "grad_norm": 0.35122546553611755,
      "learning_rate": 6.934414148857775e-05,
      "loss": 1.3452,
      "num_input_tokens_seen": 8801632,
      "step": 1779,
      "train_runtime": 6152.5495,
      "train_tokens_per_second": 1430.567
    },
    {
      "epoch": 0.654712643678161,
      "grad_norm": 0.326999306678772,
      "learning_rate": 6.927044952100221e-05,
      "loss": 1.2079,
      "num_input_tokens_seen": 8807632,
      "step": 1780,
      "train_runtime": 6156.4956,
      "train_tokens_per_second": 1430.624
    },
    {
      "epoch": 0.655080459770115,
      "grad_norm": 0.35183876752853394,
      "learning_rate": 6.919675755342668e-05,
      "loss": 1.1984,
      "num_input_tokens_seen": 8811376,
      "step": 1781,
      "train_runtime": 6159.2182,
      "train_tokens_per_second": 1430.6
    },
    {
      "epoch": 0.655448275862069,
      "grad_norm": 0.360561341047287,
      "learning_rate": 6.912306558585114e-05,
      "loss": 1.1207,
      "num_input_tokens_seen": 8816576,
      "step": 1782,
      "train_runtime": 6162.6637,
      "train_tokens_per_second": 1430.644
    },
    {
      "epoch": 0.655816091954023,
      "grad_norm": 0.3827435076236725,
      "learning_rate": 6.904937361827561e-05,
      "loss": 1.0998,
      "num_input_tokens_seen": 8822512,
      "step": 1783,
      "train_runtime": 6166.6277,
      "train_tokens_per_second": 1430.687
    },
    {
      "epoch": 0.656183908045977,
      "grad_norm": 0.3594070374965668,
      "learning_rate": 6.897568165070007e-05,
      "loss": 1.244,
      "num_input_tokens_seen": 8825856,
      "step": 1784,
      "train_runtime": 6169.0415,
      "train_tokens_per_second": 1430.669
    },
    {
      "epoch": 0.6565517241379311,
      "grad_norm": 0.36453524231910706,
      "learning_rate": 6.890198968312455e-05,
      "loss": 1.2792,
      "num_input_tokens_seen": 8829952,
      "step": 1785,
      "train_runtime": 6171.9656,
      "train_tokens_per_second": 1430.655
    },
    {
      "epoch": 0.6569195402298851,
      "grad_norm": 0.383881539106369,
      "learning_rate": 6.8828297715549e-05,
      "loss": 1.3045,
      "num_input_tokens_seen": 8834192,
      "step": 1786,
      "train_runtime": 6174.9483,
      "train_tokens_per_second": 1430.65
    },
    {
      "epoch": 0.6572873563218391,
      "grad_norm": 0.3531823754310608,
      "learning_rate": 6.875460574797348e-05,
      "loss": 1.0776,
      "num_input_tokens_seen": 8842064,
      "step": 1787,
      "train_runtime": 6179.8857,
      "train_tokens_per_second": 1430.781
    },
    {
      "epoch": 0.6576551724137931,
      "grad_norm": 0.3610166311264038,
      "learning_rate": 6.868091378039793e-05,
      "loss": 1.2063,
      "num_input_tokens_seen": 8846544,
      "step": 1788,
      "train_runtime": 6182.9988,
      "train_tokens_per_second": 1430.785
    },
    {
      "epoch": 0.6580229885057471,
      "grad_norm": 0.41903409361839294,
      "learning_rate": 6.860722181282241e-05,
      "loss": 1.1257,
      "num_input_tokens_seen": 8854112,
      "step": 1789,
      "train_runtime": 6187.8122,
      "train_tokens_per_second": 1430.895
    },
    {
      "epoch": 0.6583908045977012,
      "grad_norm": 0.35946381092071533,
      "learning_rate": 6.853352984524686e-05,
      "loss": 1.1565,
      "num_input_tokens_seen": 8859184,
      "step": 1790,
      "train_runtime": 6191.2282,
      "train_tokens_per_second": 1430.925
    },
    {
      "epoch": 0.6587586206896552,
      "grad_norm": 0.3825937509536743,
      "learning_rate": 6.845983787767134e-05,
      "loss": 1.3219,
      "num_input_tokens_seen": 8863216,
      "step": 1791,
      "train_runtime": 6193.9898,
      "train_tokens_per_second": 1430.938
    },
    {
      "epoch": 0.6591264367816092,
      "grad_norm": 0.31378173828125,
      "learning_rate": 6.83861459100958e-05,
      "loss": 1.0779,
      "num_input_tokens_seen": 8869888,
      "step": 1792,
      "train_runtime": 6198.1329,
      "train_tokens_per_second": 1431.058
    },
    {
      "epoch": 0.6594942528735632,
      "grad_norm": 0.3681214451789856,
      "learning_rate": 6.831245394252027e-05,
      "loss": 1.2146,
      "num_input_tokens_seen": 8875712,
      "step": 1793,
      "train_runtime": 6201.9276,
      "train_tokens_per_second": 1431.121
    },
    {
      "epoch": 0.6598620689655172,
      "grad_norm": 0.3513132333755493,
      "learning_rate": 6.823876197494474e-05,
      "loss": 1.1929,
      "num_input_tokens_seen": 8883408,
      "step": 1794,
      "train_runtime": 6206.8064,
      "train_tokens_per_second": 1431.237
    },
    {
      "epoch": 0.6602298850574713,
      "grad_norm": 0.35128530859947205,
      "learning_rate": 6.81650700073692e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 8892832,
      "step": 1795,
      "train_runtime": 6212.4787,
      "train_tokens_per_second": 1431.447
    },
    {
      "epoch": 0.6605977011494253,
      "grad_norm": 0.3785873353481293,
      "learning_rate": 6.809137803979367e-05,
      "loss": 1.2206,
      "num_input_tokens_seen": 8899152,
      "step": 1796,
      "train_runtime": 6216.5977,
      "train_tokens_per_second": 1431.515
    },
    {
      "epoch": 0.6609655172413793,
      "grad_norm": 0.3689374029636383,
      "learning_rate": 6.801768607221813e-05,
      "loss": 1.2212,
      "num_input_tokens_seen": 8903200,
      "step": 1797,
      "train_runtime": 6219.5073,
      "train_tokens_per_second": 1431.496
    },
    {
      "epoch": 0.6613333333333333,
      "grad_norm": 0.35654157400131226,
      "learning_rate": 6.79439941046426e-05,
      "loss": 1.1282,
      "num_input_tokens_seen": 8910192,
      "step": 1798,
      "train_runtime": 6223.9936,
      "train_tokens_per_second": 1431.588
    },
    {
      "epoch": 0.6617011494252873,
      "grad_norm": 0.3701092302799225,
      "learning_rate": 6.787030213706706e-05,
      "loss": 1.252,
      "num_input_tokens_seen": 8913952,
      "step": 1799,
      "train_runtime": 6226.6117,
      "train_tokens_per_second": 1431.59
    },
    {
      "epoch": 0.6620689655172414,
      "grad_norm": 0.36222463846206665,
      "learning_rate": 6.779661016949152e-05,
      "loss": 1.2092,
      "num_input_tokens_seen": 8916896,
      "step": 1800,
      "train_runtime": 6228.8995,
      "train_tokens_per_second": 1431.536
    },
    {
      "epoch": 0.6624367816091954,
      "grad_norm": 0.3272439241409302,
      "learning_rate": 6.7722918201916e-05,
      "loss": 1.1512,
      "num_input_tokens_seen": 8920624,
      "step": 1801,
      "train_runtime": 6233.5561,
      "train_tokens_per_second": 1431.065
    },
    {
      "epoch": 0.6628045977011494,
      "grad_norm": 0.3526923656463623,
      "learning_rate": 6.764922623434045e-05,
      "loss": 1.1256,
      "num_input_tokens_seen": 8925584,
      "step": 1802,
      "train_runtime": 6236.922,
      "train_tokens_per_second": 1431.088
    },
    {
      "epoch": 0.6631724137931034,
      "grad_norm": 0.3336963653564453,
      "learning_rate": 6.757553426676493e-05,
      "loss": 1.1617,
      "num_input_tokens_seen": 8930064,
      "step": 1803,
      "train_runtime": 6239.9543,
      "train_tokens_per_second": 1431.11
    },
    {
      "epoch": 0.6635402298850575,
      "grad_norm": 0.3636209964752197,
      "learning_rate": 6.750184229918938e-05,
      "loss": 1.1992,
      "num_input_tokens_seen": 8935552,
      "step": 1804,
      "train_runtime": 6243.6211,
      "train_tokens_per_second": 1431.149
    },
    {
      "epoch": 0.6639080459770115,
      "grad_norm": 0.3536466956138611,
      "learning_rate": 6.742815033161386e-05,
      "loss": 1.2405,
      "num_input_tokens_seen": 8938960,
      "step": 1805,
      "train_runtime": 6246.1387,
      "train_tokens_per_second": 1431.118
    },
    {
      "epoch": 0.6642758620689655,
      "grad_norm": 0.37251850962638855,
      "learning_rate": 6.735445836403831e-05,
      "loss": 1.2534,
      "num_input_tokens_seen": 8944688,
      "step": 1806,
      "train_runtime": 6249.8934,
      "train_tokens_per_second": 1431.174
    },
    {
      "epoch": 0.6646436781609195,
      "grad_norm": 0.3681562840938568,
      "learning_rate": 6.728076639646279e-05,
      "loss": 1.1789,
      "num_input_tokens_seen": 8950016,
      "step": 1807,
      "train_runtime": 6253.4018,
      "train_tokens_per_second": 1431.224
    },
    {
      "epoch": 0.6650114942528735,
      "grad_norm": 0.35245054960250854,
      "learning_rate": 6.720707442888726e-05,
      "loss": 1.3615,
      "num_input_tokens_seen": 8954848,
      "step": 1808,
      "train_runtime": 6256.6823,
      "train_tokens_per_second": 1431.245
    },
    {
      "epoch": 0.6653793103448276,
      "grad_norm": 0.3475450277328491,
      "learning_rate": 6.713338246131172e-05,
      "loss": 1.2396,
      "num_input_tokens_seen": 8959280,
      "step": 1809,
      "train_runtime": 6259.8175,
      "train_tokens_per_second": 1431.237
    },
    {
      "epoch": 0.6657471264367816,
      "grad_norm": 0.3447854518890381,
      "learning_rate": 6.705969049373619e-05,
      "loss": 1.2954,
      "num_input_tokens_seen": 8963472,
      "step": 1810,
      "train_runtime": 6262.7232,
      "train_tokens_per_second": 1431.242
    },
    {
      "epoch": 0.6661149425287356,
      "grad_norm": 0.3926207721233368,
      "learning_rate": 6.698599852616065e-05,
      "loss": 1.182,
      "num_input_tokens_seen": 8971232,
      "step": 1811,
      "train_runtime": 6267.6476,
      "train_tokens_per_second": 1431.356
    },
    {
      "epoch": 0.6664827586206896,
      "grad_norm": 0.3552563488483429,
      "learning_rate": 6.691230655858512e-05,
      "loss": 1.153,
      "num_input_tokens_seen": 8975536,
      "step": 1812,
      "train_runtime": 6270.6742,
      "train_tokens_per_second": 1431.351
    },
    {
      "epoch": 0.6668505747126436,
      "grad_norm": 0.3268927335739136,
      "learning_rate": 6.683861459100958e-05,
      "loss": 1.1059,
      "num_input_tokens_seen": 8981040,
      "step": 1813,
      "train_runtime": 6274.3463,
      "train_tokens_per_second": 1431.391
    },
    {
      "epoch": 0.6672183908045977,
      "grad_norm": 0.36756736040115356,
      "learning_rate": 6.676492262343405e-05,
      "loss": 1.1799,
      "num_input_tokens_seen": 8985056,
      "step": 1814,
      "train_runtime": 6277.0844,
      "train_tokens_per_second": 1431.406
    },
    {
      "epoch": 0.6675862068965517,
      "grad_norm": 0.37198054790496826,
      "learning_rate": 6.669123065585851e-05,
      "loss": 1.1645,
      "num_input_tokens_seen": 8989008,
      "step": 1815,
      "train_runtime": 6279.8848,
      "train_tokens_per_second": 1431.397
    },
    {
      "epoch": 0.6679540229885057,
      "grad_norm": 0.36678871512413025,
      "learning_rate": 6.661753868828298e-05,
      "loss": 1.3249,
      "num_input_tokens_seen": 8996080,
      "step": 1816,
      "train_runtime": 6284.3199,
      "train_tokens_per_second": 1431.512
    },
    {
      "epoch": 0.6683218390804597,
      "grad_norm": 0.33646324276924133,
      "learning_rate": 6.654384672070745e-05,
      "loss": 1.1876,
      "num_input_tokens_seen": 9000176,
      "step": 1817,
      "train_runtime": 6287.2273,
      "train_tokens_per_second": 1431.502
    },
    {
      "epoch": 0.6686896551724137,
      "grad_norm": 0.3421087861061096,
      "learning_rate": 6.64701547531319e-05,
      "loss": 1.3747,
      "num_input_tokens_seen": 9004032,
      "step": 1818,
      "train_runtime": 6290.0007,
      "train_tokens_per_second": 1431.483
    },
    {
      "epoch": 0.6690574712643678,
      "grad_norm": 0.4438786208629608,
      "learning_rate": 6.639646278555638e-05,
      "loss": 1.312,
      "num_input_tokens_seen": 9008288,
      "step": 1819,
      "train_runtime": 6292.9153,
      "train_tokens_per_second": 1431.497
    },
    {
      "epoch": 0.6694252873563218,
      "grad_norm": 0.339505672454834,
      "learning_rate": 6.632277081798083e-05,
      "loss": 1.1924,
      "num_input_tokens_seen": 9011968,
      "step": 1820,
      "train_runtime": 6295.5831,
      "train_tokens_per_second": 1431.475
    },
    {
      "epoch": 0.6697931034482759,
      "grad_norm": 0.34662675857543945,
      "learning_rate": 6.624907885040531e-05,
      "loss": 1.2922,
      "num_input_tokens_seen": 9015872,
      "step": 1821,
      "train_runtime": 6298.4191,
      "train_tokens_per_second": 1431.45
    },
    {
      "epoch": 0.6701609195402299,
      "grad_norm": 0.3761582374572754,
      "learning_rate": 6.617538688282976e-05,
      "loss": 1.1373,
      "num_input_tokens_seen": 9019824,
      "step": 1822,
      "train_runtime": 6301.2307,
      "train_tokens_per_second": 1431.438
    },
    {
      "epoch": 0.670528735632184,
      "grad_norm": 0.3703620731830597,
      "learning_rate": 6.610169491525424e-05,
      "loss": 1.1427,
      "num_input_tokens_seen": 9026384,
      "step": 1823,
      "train_runtime": 6305.5133,
      "train_tokens_per_second": 1431.507
    },
    {
      "epoch": 0.670896551724138,
      "grad_norm": 0.3850911259651184,
      "learning_rate": 6.602800294767871e-05,
      "loss": 1.28,
      "num_input_tokens_seen": 9030576,
      "step": 1824,
      "train_runtime": 6308.3937,
      "train_tokens_per_second": 1431.518
    },
    {
      "epoch": 0.671264367816092,
      "grad_norm": 0.3600991368293762,
      "learning_rate": 6.595431098010317e-05,
      "loss": 1.2281,
      "num_input_tokens_seen": 9036560,
      "step": 1825,
      "train_runtime": 6312.2085,
      "train_tokens_per_second": 1431.6
    },
    {
      "epoch": 0.671632183908046,
      "grad_norm": 0.3798394203186035,
      "learning_rate": 6.588061901252764e-05,
      "loss": 1.3111,
      "num_input_tokens_seen": 9040352,
      "step": 1826,
      "train_runtime": 6314.8561,
      "train_tokens_per_second": 1431.601
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.4036741852760315,
      "learning_rate": 6.58069270449521e-05,
      "loss": 1.2882,
      "num_input_tokens_seen": 9046352,
      "step": 1827,
      "train_runtime": 6318.8109,
      "train_tokens_per_second": 1431.654
    },
    {
      "epoch": 0.6723678160919541,
      "grad_norm": 0.3849335014820099,
      "learning_rate": 6.573323507737657e-05,
      "loss": 1.1751,
      "num_input_tokens_seen": 9049760,
      "step": 1828,
      "train_runtime": 6321.3295,
      "train_tokens_per_second": 1431.623
    },
    {
      "epoch": 0.6727356321839081,
      "grad_norm": 0.3657708466053009,
      "learning_rate": 6.565954310980103e-05,
      "loss": 1.2536,
      "num_input_tokens_seen": 9053680,
      "step": 1829,
      "train_runtime": 6324.0259,
      "train_tokens_per_second": 1431.632
    },
    {
      "epoch": 0.6731034482758621,
      "grad_norm": 0.4086407721042633,
      "learning_rate": 6.55858511422255e-05,
      "loss": 1.4151,
      "num_input_tokens_seen": 9057600,
      "step": 1830,
      "train_runtime": 6326.8135,
      "train_tokens_per_second": 1431.621
    },
    {
      "epoch": 0.6734712643678161,
      "grad_norm": 0.3436093032360077,
      "learning_rate": 6.551215917464996e-05,
      "loss": 1.1478,
      "num_input_tokens_seen": 9062992,
      "step": 1831,
      "train_runtime": 6332.6858,
      "train_tokens_per_second": 1431.145
    },
    {
      "epoch": 0.6738390804597701,
      "grad_norm": 0.3341924846172333,
      "learning_rate": 6.543846720707443e-05,
      "loss": 1.1927,
      "num_input_tokens_seen": 9066352,
      "step": 1832,
      "train_runtime": 6335.1493,
      "train_tokens_per_second": 1431.119
    },
    {
      "epoch": 0.6742068965517242,
      "grad_norm": 0.36981868743896484,
      "learning_rate": 6.53647752394989e-05,
      "loss": 1.2557,
      "num_input_tokens_seen": 9070304,
      "step": 1833,
      "train_runtime": 6337.9877,
      "train_tokens_per_second": 1431.102
    },
    {
      "epoch": 0.6745747126436782,
      "grad_norm": 0.3636479675769806,
      "learning_rate": 6.529108327192336e-05,
      "loss": 1.3673,
      "num_input_tokens_seen": 9074496,
      "step": 1834,
      "train_runtime": 6340.9341,
      "train_tokens_per_second": 1431.098
    },
    {
      "epoch": 0.6749425287356322,
      "grad_norm": 0.38223516941070557,
      "learning_rate": 6.521739130434783e-05,
      "loss": 1.1859,
      "num_input_tokens_seen": 9078144,
      "step": 1835,
      "train_runtime": 6343.5448,
      "train_tokens_per_second": 1431.084
    },
    {
      "epoch": 0.6753103448275862,
      "grad_norm": 0.37790757417678833,
      "learning_rate": 6.514369933677229e-05,
      "loss": 1.304,
      "num_input_tokens_seen": 9081648,
      "step": 1836,
      "train_runtime": 6346.127,
      "train_tokens_per_second": 1431.054
    },
    {
      "epoch": 0.6756781609195402,
      "grad_norm": 0.4128868877887726,
      "learning_rate": 6.507000736919676e-05,
      "loss": 1.2607,
      "num_input_tokens_seen": 9085184,
      "step": 1837,
      "train_runtime": 6348.6613,
      "train_tokens_per_second": 1431.039
    },
    {
      "epoch": 0.6760459770114943,
      "grad_norm": 0.3606361746788025,
      "learning_rate": 6.499631540162123e-05,
      "loss": 1.2152,
      "num_input_tokens_seen": 9089296,
      "step": 1838,
      "train_runtime": 6351.5085,
      "train_tokens_per_second": 1431.045
    },
    {
      "epoch": 0.6764137931034483,
      "grad_norm": 0.369467556476593,
      "learning_rate": 6.492262343404569e-05,
      "loss": 1.2048,
      "num_input_tokens_seen": 9098016,
      "step": 1839,
      "train_runtime": 6356.8465,
      "train_tokens_per_second": 1431.215
    },
    {
      "epoch": 0.6767816091954023,
      "grad_norm": 0.3669975996017456,
      "learning_rate": 6.484893146647016e-05,
      "loss": 1.1768,
      "num_input_tokens_seen": 9103168,
      "step": 1840,
      "train_runtime": 6360.2995,
      "train_tokens_per_second": 1431.248
    },
    {
      "epoch": 0.6771494252873563,
      "grad_norm": 0.3667323887348175,
      "learning_rate": 6.477523949889462e-05,
      "loss": 1.1826,
      "num_input_tokens_seen": 9107904,
      "step": 1841,
      "train_runtime": 6363.5143,
      "train_tokens_per_second": 1431.27
    },
    {
      "epoch": 0.6775172413793104,
      "grad_norm": 0.372950941324234,
      "learning_rate": 6.470154753131909e-05,
      "loss": 1.2482,
      "num_input_tokens_seen": 9111808,
      "step": 1842,
      "train_runtime": 6366.4978,
      "train_tokens_per_second": 1431.212
    },
    {
      "epoch": 0.6778850574712644,
      "grad_norm": 0.34295400977134705,
      "learning_rate": 6.462785556374355e-05,
      "loss": 1.1972,
      "num_input_tokens_seen": 9115616,
      "step": 1843,
      "train_runtime": 6369.2423,
      "train_tokens_per_second": 1431.193
    },
    {
      "epoch": 0.6782528735632184,
      "grad_norm": 0.35535502433776855,
      "learning_rate": 6.455416359616802e-05,
      "loss": 1.155,
      "num_input_tokens_seen": 9120000,
      "step": 1844,
      "train_runtime": 6372.2647,
      "train_tokens_per_second": 1431.202
    },
    {
      "epoch": 0.6786206896551724,
      "grad_norm": 0.3609297275543213,
      "learning_rate": 6.448047162859248e-05,
      "loss": 1.1983,
      "num_input_tokens_seen": 9125168,
      "step": 1845,
      "train_runtime": 6375.7512,
      "train_tokens_per_second": 1431.23
    },
    {
      "epoch": 0.6789885057471264,
      "grad_norm": 0.3910728693008423,
      "learning_rate": 6.440677966101695e-05,
      "loss": 1.376,
      "num_input_tokens_seen": 9130928,
      "step": 1846,
      "train_runtime": 6379.5748,
      "train_tokens_per_second": 1431.275
    },
    {
      "epoch": 0.6793563218390805,
      "grad_norm": 0.34996485710144043,
      "learning_rate": 6.433308769344141e-05,
      "loss": 1.2125,
      "num_input_tokens_seen": 9134880,
      "step": 1847,
      "train_runtime": 6382.3282,
      "train_tokens_per_second": 1431.277
    },
    {
      "epoch": 0.6797241379310345,
      "grad_norm": 0.36965256929397583,
      "learning_rate": 6.425939572586588e-05,
      "loss": 1.1865,
      "num_input_tokens_seen": 9139920,
      "step": 1848,
      "train_runtime": 6385.626,
      "train_tokens_per_second": 1431.327
    },
    {
      "epoch": 0.6800919540229885,
      "grad_norm": 0.36250296235084534,
      "learning_rate": 6.418570375829035e-05,
      "loss": 1.2429,
      "num_input_tokens_seen": 9144064,
      "step": 1849,
      "train_runtime": 6388.5808,
      "train_tokens_per_second": 1431.314
    },
    {
      "epoch": 0.6804597701149425,
      "grad_norm": 0.40359920263290405,
      "learning_rate": 6.41120117907148e-05,
      "loss": 1.2092,
      "num_input_tokens_seen": 9148256,
      "step": 1850,
      "train_runtime": 6391.5033,
      "train_tokens_per_second": 1431.315
    },
    {
      "epoch": 0.6808275862068965,
      "grad_norm": 0.4024823307991028,
      "learning_rate": 6.403831982313928e-05,
      "loss": 1.1363,
      "num_input_tokens_seen": 9154480,
      "step": 1851,
      "train_runtime": 6395.5385,
      "train_tokens_per_second": 1431.385
    },
    {
      "epoch": 0.6811954022988506,
      "grad_norm": 0.37563154101371765,
      "learning_rate": 6.396462785556374e-05,
      "loss": 1.0199,
      "num_input_tokens_seen": 9161280,
      "step": 1852,
      "train_runtime": 6400.1541,
      "train_tokens_per_second": 1431.416
    },
    {
      "epoch": 0.6815632183908046,
      "grad_norm": 0.38258638978004456,
      "learning_rate": 6.389093588798821e-05,
      "loss": 1.2741,
      "num_input_tokens_seen": 9166544,
      "step": 1853,
      "train_runtime": 6403.696,
      "train_tokens_per_second": 1431.446
    },
    {
      "epoch": 0.6819310344827586,
      "grad_norm": 0.37171056866645813,
      "learning_rate": 6.381724392041268e-05,
      "loss": 1.2824,
      "num_input_tokens_seen": 9170240,
      "step": 1854,
      "train_runtime": 6406.3479,
      "train_tokens_per_second": 1431.43
    },
    {
      "epoch": 0.6822988505747126,
      "grad_norm": 0.36011645197868347,
      "learning_rate": 6.374355195283714e-05,
      "loss": 1.1489,
      "num_input_tokens_seen": 9175040,
      "step": 1855,
      "train_runtime": 6409.6583,
      "train_tokens_per_second": 1431.44
    },
    {
      "epoch": 0.6826666666666666,
      "grad_norm": 0.37741726636886597,
      "learning_rate": 6.366985998526161e-05,
      "loss": 1.1609,
      "num_input_tokens_seen": 9179296,
      "step": 1856,
      "train_runtime": 6412.5882,
      "train_tokens_per_second": 1431.449
    },
    {
      "epoch": 0.6830344827586207,
      "grad_norm": 0.35376110672950745,
      "learning_rate": 6.359616801768607e-05,
      "loss": 1.2731,
      "num_input_tokens_seen": 9183008,
      "step": 1857,
      "train_runtime": 6415.1709,
      "train_tokens_per_second": 1431.452
    },
    {
      "epoch": 0.6834022988505747,
      "grad_norm": 0.35565951466560364,
      "learning_rate": 6.352247605011055e-05,
      "loss": 1.2774,
      "num_input_tokens_seen": 9186608,
      "step": 1858,
      "train_runtime": 6417.6818,
      "train_tokens_per_second": 1431.453
    },
    {
      "epoch": 0.6837701149425287,
      "grad_norm": 0.34637174010276794,
      "learning_rate": 6.3448784082535e-05,
      "loss": 1.0784,
      "num_input_tokens_seen": 9190896,
      "step": 1859,
      "train_runtime": 6420.6898,
      "train_tokens_per_second": 1431.45
    },
    {
      "epoch": 0.6841379310344827,
      "grad_norm": 0.3382808268070221,
      "learning_rate": 6.337509211495948e-05,
      "loss": 1.1368,
      "num_input_tokens_seen": 9196160,
      "step": 1860,
      "train_runtime": 6424.2139,
      "train_tokens_per_second": 1431.484
    },
    {
      "epoch": 0.6845057471264367,
      "grad_norm": 0.3636821210384369,
      "learning_rate": 6.330140014738393e-05,
      "loss": 1.0651,
      "num_input_tokens_seen": 9200048,
      "step": 1861,
      "train_runtime": 6428.9818,
      "train_tokens_per_second": 1431.027
    },
    {
      "epoch": 0.6848735632183908,
      "grad_norm": 0.3772403597831726,
      "learning_rate": 6.322770817980841e-05,
      "loss": 1.1643,
      "num_input_tokens_seen": 9206688,
      "step": 1862,
      "train_runtime": 6433.2295,
      "train_tokens_per_second": 1431.114
    },
    {
      "epoch": 0.6852413793103448,
      "grad_norm": 0.3455715775489807,
      "learning_rate": 6.315401621223288e-05,
      "loss": 1.1697,
      "num_input_tokens_seen": 9211296,
      "step": 1863,
      "train_runtime": 6436.3186,
      "train_tokens_per_second": 1431.144
    },
    {
      "epoch": 0.6856091954022988,
      "grad_norm": 0.32315951585769653,
      "learning_rate": 6.308032424465734e-05,
      "loss": 1.2082,
      "num_input_tokens_seen": 9219232,
      "step": 1864,
      "train_runtime": 6441.1762,
      "train_tokens_per_second": 1431.296
    },
    {
      "epoch": 0.6859770114942528,
      "grad_norm": 0.35601308941841125,
      "learning_rate": 6.30066322770818e-05,
      "loss": 1.1205,
      "num_input_tokens_seen": 9223504,
      "step": 1865,
      "train_runtime": 6444.1893,
      "train_tokens_per_second": 1431.29
    },
    {
      "epoch": 0.6863448275862069,
      "grad_norm": 0.3578878343105316,
      "learning_rate": 6.293294030950627e-05,
      "loss": 1.3482,
      "num_input_tokens_seen": 9228288,
      "step": 1866,
      "train_runtime": 6447.4622,
      "train_tokens_per_second": 1431.305
    },
    {
      "epoch": 0.6867126436781609,
      "grad_norm": 0.38044896721839905,
      "learning_rate": 6.285924834193074e-05,
      "loss": 1.2671,
      "num_input_tokens_seen": 9232192,
      "step": 1867,
      "train_runtime": 6450.2302,
      "train_tokens_per_second": 1431.297
    },
    {
      "epoch": 0.6870804597701149,
      "grad_norm": 0.35539358854293823,
      "learning_rate": 6.27855563743552e-05,
      "loss": 1.0913,
      "num_input_tokens_seen": 9237776,
      "step": 1868,
      "train_runtime": 6453.8758,
      "train_tokens_per_second": 1431.353
    },
    {
      "epoch": 0.6874482758620689,
      "grad_norm": 0.4087512195110321,
      "learning_rate": 6.271186440677966e-05,
      "loss": 1.4946,
      "num_input_tokens_seen": 9244736,
      "step": 1869,
      "train_runtime": 6458.3793,
      "train_tokens_per_second": 1431.433
    },
    {
      "epoch": 0.6878160919540229,
      "grad_norm": 0.38253363966941833,
      "learning_rate": 6.263817243920413e-05,
      "loss": 1.1972,
      "num_input_tokens_seen": 9250560,
      "step": 1870,
      "train_runtime": 6462.1429,
      "train_tokens_per_second": 1431.5
    },
    {
      "epoch": 0.6881839080459771,
      "grad_norm": 0.33830374479293823,
      "learning_rate": 6.25644804716286e-05,
      "loss": 1.0928,
      "num_input_tokens_seen": 9255024,
      "step": 1871,
      "train_runtime": 6465.1691,
      "train_tokens_per_second": 1431.521
    },
    {
      "epoch": 0.6885517241379311,
      "grad_norm": 0.3524548411369324,
      "learning_rate": 6.249078850405306e-05,
      "loss": 1.274,
      "num_input_tokens_seen": 9258224,
      "step": 1872,
      "train_runtime": 6467.5569,
      "train_tokens_per_second": 1431.487
    },
    {
      "epoch": 0.6889195402298851,
      "grad_norm": 0.38763436675071716,
      "learning_rate": 6.241709653647752e-05,
      "loss": 1.0899,
      "num_input_tokens_seen": 9262272,
      "step": 1873,
      "train_runtime": 6470.3487,
      "train_tokens_per_second": 1431.495
    },
    {
      "epoch": 0.6892873563218391,
      "grad_norm": 0.3447798490524292,
      "learning_rate": 6.2343404568902e-05,
      "loss": 1.1273,
      "num_input_tokens_seen": 9267568,
      "step": 1874,
      "train_runtime": 6473.9055,
      "train_tokens_per_second": 1431.527
    },
    {
      "epoch": 0.6896551724137931,
      "grad_norm": 0.34311696887016296,
      "learning_rate": 6.226971260132645e-05,
      "loss": 1.1664,
      "num_input_tokens_seen": 9272192,
      "step": 1875,
      "train_runtime": 6477.063,
      "train_tokens_per_second": 1431.543
    },
    {
      "epoch": 0.6900229885057472,
      "grad_norm": 0.3738171458244324,
      "learning_rate": 6.219602063375093e-05,
      "loss": 1.3383,
      "num_input_tokens_seen": 9276832,
      "step": 1876,
      "train_runtime": 6480.2122,
      "train_tokens_per_second": 1431.563
    },
    {
      "epoch": 0.6903908045977012,
      "grad_norm": 0.3437025845050812,
      "learning_rate": 6.212232866617538e-05,
      "loss": 1.2243,
      "num_input_tokens_seen": 9282960,
      "step": 1877,
      "train_runtime": 6484.2288,
      "train_tokens_per_second": 1431.621
    },
    {
      "epoch": 0.6907586206896552,
      "grad_norm": 0.3527766466140747,
      "learning_rate": 6.204863669859986e-05,
      "loss": 1.2312,
      "num_input_tokens_seen": 9286672,
      "step": 1878,
      "train_runtime": 6486.8531,
      "train_tokens_per_second": 1431.614
    },
    {
      "epoch": 0.6911264367816092,
      "grad_norm": 0.34566250443458557,
      "learning_rate": 6.197494473102433e-05,
      "loss": 1.1619,
      "num_input_tokens_seen": 9291568,
      "step": 1879,
      "train_runtime": 6490.0455,
      "train_tokens_per_second": 1431.665
    },
    {
      "epoch": 0.6914942528735633,
      "grad_norm": 0.3459474742412567,
      "learning_rate": 6.190125276344879e-05,
      "loss": 1.281,
      "num_input_tokens_seen": 9294944,
      "step": 1880,
      "train_runtime": 6492.5178,
      "train_tokens_per_second": 1431.639
    },
    {
      "epoch": 0.6918620689655173,
      "grad_norm": 0.3621971905231476,
      "learning_rate": 6.182756079587326e-05,
      "loss": 1.2377,
      "num_input_tokens_seen": 9298144,
      "step": 1881,
      "train_runtime": 6494.946,
      "train_tokens_per_second": 1431.597
    },
    {
      "epoch": 0.6922298850574713,
      "grad_norm": 0.3718065023422241,
      "learning_rate": 6.175386882829772e-05,
      "loss": 1.1842,
      "num_input_tokens_seen": 9302624,
      "step": 1882,
      "train_runtime": 6498.0336,
      "train_tokens_per_second": 1431.606
    },
    {
      "epoch": 0.6925977011494253,
      "grad_norm": 0.3204992115497589,
      "learning_rate": 6.168017686072219e-05,
      "loss": 1.0915,
      "num_input_tokens_seen": 9308000,
      "step": 1883,
      "train_runtime": 6501.6177,
      "train_tokens_per_second": 1431.644
    },
    {
      "epoch": 0.6929655172413793,
      "grad_norm": 0.3641936480998993,
      "learning_rate": 6.160648489314665e-05,
      "loss": 1.2156,
      "num_input_tokens_seen": 9314032,
      "step": 1884,
      "train_runtime": 6505.5074,
      "train_tokens_per_second": 1431.715
    },
    {
      "epoch": 0.6933333333333334,
      "grad_norm": 0.33905354142189026,
      "learning_rate": 6.153279292557112e-05,
      "loss": 1.255,
      "num_input_tokens_seen": 9317776,
      "step": 1885,
      "train_runtime": 6508.1929,
      "train_tokens_per_second": 1431.699
    },
    {
      "epoch": 0.6937011494252874,
      "grad_norm": 0.3809105455875397,
      "learning_rate": 6.145910095799558e-05,
      "loss": 1.1799,
      "num_input_tokens_seen": 9324080,
      "step": 1886,
      "train_runtime": 6512.2401,
      "train_tokens_per_second": 1431.778
    },
    {
      "epoch": 0.6940689655172414,
      "grad_norm": 0.4171127676963806,
      "learning_rate": 6.138540899042005e-05,
      "loss": 1.2886,
      "num_input_tokens_seen": 9327216,
      "step": 1887,
      "train_runtime": 6514.6094,
      "train_tokens_per_second": 1431.738
    },
    {
      "epoch": 0.6944367816091954,
      "grad_norm": 0.3351636528968811,
      "learning_rate": 6.131171702284452e-05,
      "loss": 1.1496,
      "num_input_tokens_seen": 9331952,
      "step": 1888,
      "train_runtime": 6517.8359,
      "train_tokens_per_second": 1431.756
    },
    {
      "epoch": 0.6948045977011494,
      "grad_norm": 0.34816426038742065,
      "learning_rate": 6.123802505526897e-05,
      "loss": 1.1456,
      "num_input_tokens_seen": 9337376,
      "step": 1889,
      "train_runtime": 6521.3646,
      "train_tokens_per_second": 1431.813
    },
    {
      "epoch": 0.6951724137931035,
      "grad_norm": 0.33218374848365784,
      "learning_rate": 6.116433308769345e-05,
      "loss": 1.053,
      "num_input_tokens_seen": 9342416,
      "step": 1890,
      "train_runtime": 6524.7475,
      "train_tokens_per_second": 1431.843
    },
    {
      "epoch": 0.6955402298850575,
      "grad_norm": 0.37147361040115356,
      "learning_rate": 6.10906411201179e-05,
      "loss": 1.1641,
      "num_input_tokens_seen": 9350512,
      "step": 1891,
      "train_runtime": 6531.6901,
      "train_tokens_per_second": 1431.561
    },
    {
      "epoch": 0.6959080459770115,
      "grad_norm": 0.377197265625,
      "learning_rate": 6.101694915254238e-05,
      "loss": 1.2976,
      "num_input_tokens_seen": 9357072,
      "step": 1892,
      "train_runtime": 6535.9712,
      "train_tokens_per_second": 1431.627
    },
    {
      "epoch": 0.6962758620689655,
      "grad_norm": 0.3650781810283661,
      "learning_rate": 6.094325718496684e-05,
      "loss": 1.1569,
      "num_input_tokens_seen": 9361984,
      "step": 1893,
      "train_runtime": 6539.3028,
      "train_tokens_per_second": 1431.649
    },
    {
      "epoch": 0.6966436781609195,
      "grad_norm": 0.34893959760665894,
      "learning_rate": 6.086956521739131e-05,
      "loss": 1.1522,
      "num_input_tokens_seen": 9367536,
      "step": 1894,
      "train_runtime": 6542.9072,
      "train_tokens_per_second": 1431.709
    },
    {
      "epoch": 0.6970114942528736,
      "grad_norm": 0.35472455620765686,
      "learning_rate": 6.079587324981577e-05,
      "loss": 1.134,
      "num_input_tokens_seen": 9372240,
      "step": 1895,
      "train_runtime": 6546.0875,
      "train_tokens_per_second": 1431.732
    },
    {
      "epoch": 0.6973793103448276,
      "grad_norm": 0.3683747947216034,
      "learning_rate": 6.072218128224024e-05,
      "loss": 1.1356,
      "num_input_tokens_seen": 9376096,
      "step": 1896,
      "train_runtime": 6548.8341,
      "train_tokens_per_second": 1431.72
    },
    {
      "epoch": 0.6977471264367816,
      "grad_norm": 0.34663519263267517,
      "learning_rate": 6.06484893146647e-05,
      "loss": 1.0954,
      "num_input_tokens_seen": 9381408,
      "step": 1897,
      "train_runtime": 6552.3427,
      "train_tokens_per_second": 1431.764
    },
    {
      "epoch": 0.6981149425287356,
      "grad_norm": 0.38044968247413635,
      "learning_rate": 6.057479734708917e-05,
      "loss": 1.1224,
      "num_input_tokens_seen": 9386816,
      "step": 1898,
      "train_runtime": 6555.9786,
      "train_tokens_per_second": 1431.795
    },
    {
      "epoch": 0.6984827586206896,
      "grad_norm": 0.3662697970867157,
      "learning_rate": 6.050110537951364e-05,
      "loss": 1.2003,
      "num_input_tokens_seen": 9391200,
      "step": 1899,
      "train_runtime": 6559.042,
      "train_tokens_per_second": 1431.794
    },
    {
      "epoch": 0.6988505747126437,
      "grad_norm": 0.33618733286857605,
      "learning_rate": 6.042741341193811e-05,
      "loss": 1.1694,
      "num_input_tokens_seen": 9395264,
      "step": 1900,
      "train_runtime": 6561.8635,
      "train_tokens_per_second": 1431.798
    },
    {
      "epoch": 0.6992183908045977,
      "grad_norm": 0.40789663791656494,
      "learning_rate": 6.0353721444362566e-05,
      "loss": 1.1237,
      "num_input_tokens_seen": 9398848,
      "step": 1901,
      "train_runtime": 6564.4486,
      "train_tokens_per_second": 1431.78
    },
    {
      "epoch": 0.6995862068965517,
      "grad_norm": 0.36087027192115784,
      "learning_rate": 6.028002947678704e-05,
      "loss": 1.082,
      "num_input_tokens_seen": 9402144,
      "step": 1902,
      "train_runtime": 6566.94,
      "train_tokens_per_second": 1431.739
    },
    {
      "epoch": 0.6999540229885057,
      "grad_norm": 0.36859339475631714,
      "learning_rate": 6.0206337509211496e-05,
      "loss": 1.1294,
      "num_input_tokens_seen": 9407472,
      "step": 1903,
      "train_runtime": 6570.4776,
      "train_tokens_per_second": 1431.779
    },
    {
      "epoch": 0.7003218390804598,
      "grad_norm": 0.3652470111846924,
      "learning_rate": 6.013264554163597e-05,
      "loss": 1.1833,
      "num_input_tokens_seen": 9414960,
      "step": 1904,
      "train_runtime": 6575.2716,
      "train_tokens_per_second": 1431.874
    },
    {
      "epoch": 0.7003218390804598,
      "eval_loss": 1.2160385847091675,
      "eval_runtime": 15.4936,
      "eval_samples_per_second": 64.543,
      "eval_steps_per_second": 4.066,
      "num_input_tokens_seen": 9414960,
      "step": 1904
    },
    {
      "epoch": 0.7006896551724138,
      "grad_norm": 0.3549385666847229,
      "learning_rate": 6.0058953574060426e-05,
      "loss": 1.0774,
      "num_input_tokens_seen": 9419824,
      "step": 1905,
      "train_runtime": 6594.1029,
      "train_tokens_per_second": 1428.522
    },
    {
      "epoch": 0.7010574712643678,
      "grad_norm": 0.40111044049263,
      "learning_rate": 5.99852616064849e-05,
      "loss": 1.3336,
      "num_input_tokens_seen": 9426320,
      "step": 1906,
      "train_runtime": 6598.2909,
      "train_tokens_per_second": 1428.6
    },
    {
      "epoch": 0.7014252873563218,
      "grad_norm": 0.3566766679286957,
      "learning_rate": 5.991156963890936e-05,
      "loss": 1.2749,
      "num_input_tokens_seen": 9433712,
      "step": 1907,
      "train_runtime": 6602.9706,
      "train_tokens_per_second": 1428.707
    },
    {
      "epoch": 0.7017931034482758,
      "grad_norm": 0.34691503643989563,
      "learning_rate": 5.9837877671333834e-05,
      "loss": 1.2687,
      "num_input_tokens_seen": 9437584,
      "step": 1908,
      "train_runtime": 6605.7028,
      "train_tokens_per_second": 1428.702
    },
    {
      "epoch": 0.7021609195402299,
      "grad_norm": 0.3621295690536499,
      "learning_rate": 5.976418570375829e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 9442336,
      "step": 1909,
      "train_runtime": 6608.9436,
      "train_tokens_per_second": 1428.721
    },
    {
      "epoch": 0.7025287356321839,
      "grad_norm": 0.3740859925746918,
      "learning_rate": 5.969049373618276e-05,
      "loss": 1.2191,
      "num_input_tokens_seen": 9447008,
      "step": 1910,
      "train_runtime": 6612.1062,
      "train_tokens_per_second": 1428.744
    },
    {
      "epoch": 0.7028965517241379,
      "grad_norm": 0.36735615134239197,
      "learning_rate": 5.961680176860722e-05,
      "loss": 1.408,
      "num_input_tokens_seen": 9453824,
      "step": 1911,
      "train_runtime": 6616.5543,
      "train_tokens_per_second": 1428.814
    },
    {
      "epoch": 0.7032643678160919,
      "grad_norm": 0.33453720808029175,
      "learning_rate": 5.954310980103169e-05,
      "loss": 1.018,
      "num_input_tokens_seen": 9458240,
      "step": 1912,
      "train_runtime": 6619.5646,
      "train_tokens_per_second": 1428.831
    },
    {
      "epoch": 0.7036321839080459,
      "grad_norm": 0.3643985986709595,
      "learning_rate": 5.946941783345615e-05,
      "loss": 1.2118,
      "num_input_tokens_seen": 9463200,
      "step": 1913,
      "train_runtime": 6622.9093,
      "train_tokens_per_second": 1428.858
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.3474371135234833,
      "learning_rate": 5.939572586588062e-05,
      "loss": 1.0819,
      "num_input_tokens_seen": 9467536,
      "step": 1914,
      "train_runtime": 6625.8575,
      "train_tokens_per_second": 1428.877
    },
    {
      "epoch": 0.704367816091954,
      "grad_norm": 0.36465346813201904,
      "learning_rate": 5.932203389830509e-05,
      "loss": 1.2481,
      "num_input_tokens_seen": 9471808,
      "step": 1915,
      "train_runtime": 6628.8327,
      "train_tokens_per_second": 1428.88
    },
    {
      "epoch": 0.704735632183908,
      "grad_norm": 0.35101088881492615,
      "learning_rate": 5.924834193072956e-05,
      "loss": 1.2469,
      "num_input_tokens_seen": 9478512,
      "step": 1916,
      "train_runtime": 6633.1398,
      "train_tokens_per_second": 1428.963
    },
    {
      "epoch": 0.705103448275862,
      "grad_norm": 0.3579235076904297,
      "learning_rate": 5.917464996315402e-05,
      "loss": 1.0934,
      "num_input_tokens_seen": 9483392,
      "step": 1917,
      "train_runtime": 6636.3826,
      "train_tokens_per_second": 1429.0
    },
    {
      "epoch": 0.705471264367816,
      "grad_norm": 0.38996049761772156,
      "learning_rate": 5.910095799557849e-05,
      "loss": 1.29,
      "num_input_tokens_seen": 9487472,
      "step": 1918,
      "train_runtime": 6639.2459,
      "train_tokens_per_second": 1428.998
    },
    {
      "epoch": 0.7058390804597701,
      "grad_norm": 0.35924428701400757,
      "learning_rate": 5.902726602800295e-05,
      "loss": 1.2439,
      "num_input_tokens_seen": 9494416,
      "step": 1919,
      "train_runtime": 6643.6823,
      "train_tokens_per_second": 1429.089
    },
    {
      "epoch": 0.7062068965517241,
      "grad_norm": 0.3391769230365753,
      "learning_rate": 5.895357406042742e-05,
      "loss": 1.12,
      "num_input_tokens_seen": 9501152,
      "step": 1920,
      "train_runtime": 6647.8737,
      "train_tokens_per_second": 1429.202
    },
    {
      "epoch": 0.7065747126436782,
      "grad_norm": 0.34716150164604187,
      "learning_rate": 5.8879882092851876e-05,
      "loss": 1.1751,
      "num_input_tokens_seen": 9506160,
      "step": 1921,
      "train_runtime": 6653.1794,
      "train_tokens_per_second": 1428.815
    },
    {
      "epoch": 0.7069425287356322,
      "grad_norm": 0.36143097281455994,
      "learning_rate": 5.880619012527635e-05,
      "loss": 1.3488,
      "num_input_tokens_seen": 9510160,
      "step": 1922,
      "train_runtime": 6655.9287,
      "train_tokens_per_second": 1428.825
    },
    {
      "epoch": 0.7073103448275863,
      "grad_norm": 0.3692696690559387,
      "learning_rate": 5.873249815770081e-05,
      "loss": 1.2048,
      "num_input_tokens_seen": 9515392,
      "step": 1923,
      "train_runtime": 6659.3707,
      "train_tokens_per_second": 1428.873
    },
    {
      "epoch": 0.7076781609195403,
      "grad_norm": 0.35383716225624084,
      "learning_rate": 5.8658806190125284e-05,
      "loss": 1.2238,
      "num_input_tokens_seen": 9519216,
      "step": 1924,
      "train_runtime": 6662.0237,
      "train_tokens_per_second": 1428.878
    },
    {
      "epoch": 0.7080459770114943,
      "grad_norm": 0.37925904989242554,
      "learning_rate": 5.858511422254974e-05,
      "loss": 1.2483,
      "num_input_tokens_seen": 9522752,
      "step": 1925,
      "train_runtime": 6664.6275,
      "train_tokens_per_second": 1428.85
    },
    {
      "epoch": 0.7084137931034483,
      "grad_norm": 0.33492371439933777,
      "learning_rate": 5.8511422254974214e-05,
      "loss": 1.2185,
      "num_input_tokens_seen": 9530688,
      "step": 1926,
      "train_runtime": 6669.6051,
      "train_tokens_per_second": 1428.973
    },
    {
      "epoch": 0.7087816091954023,
      "grad_norm": 0.36645299196243286,
      "learning_rate": 5.843773028739867e-05,
      "loss": 1.115,
      "num_input_tokens_seen": 9534176,
      "step": 1927,
      "train_runtime": 6672.1486,
      "train_tokens_per_second": 1428.951
    },
    {
      "epoch": 0.7091494252873564,
      "grad_norm": 0.33114898204803467,
      "learning_rate": 5.8364038319823144e-05,
      "loss": 1.0731,
      "num_input_tokens_seen": 9537760,
      "step": 1928,
      "train_runtime": 6674.7271,
      "train_tokens_per_second": 1428.936
    },
    {
      "epoch": 0.7095172413793104,
      "grad_norm": 0.36399704217910767,
      "learning_rate": 5.82903463522476e-05,
      "loss": 1.1155,
      "num_input_tokens_seen": 9546368,
      "step": 1929,
      "train_runtime": 6680.0861,
      "train_tokens_per_second": 1429.079
    },
    {
      "epoch": 0.7098850574712644,
      "grad_norm": 0.3734672963619232,
      "learning_rate": 5.821665438467207e-05,
      "loss": 1.1804,
      "num_input_tokens_seen": 9550528,
      "step": 1930,
      "train_runtime": 6682.9829,
      "train_tokens_per_second": 1429.082
    },
    {
      "epoch": 0.7102528735632184,
      "grad_norm": 0.3941843509674072,
      "learning_rate": 5.814296241709654e-05,
      "loss": 1.1189,
      "num_input_tokens_seen": 9554112,
      "step": 1931,
      "train_runtime": 6685.616,
      "train_tokens_per_second": 1429.055
    },
    {
      "epoch": 0.7106206896551724,
      "grad_norm": 0.3478284478187561,
      "learning_rate": 5.806927044952101e-05,
      "loss": 1.139,
      "num_input_tokens_seen": 9560208,
      "step": 1932,
      "train_runtime": 6689.5866,
      "train_tokens_per_second": 1429.118
    },
    {
      "epoch": 0.7109885057471265,
      "grad_norm": 0.37273889780044556,
      "learning_rate": 5.799557848194547e-05,
      "loss": 1.2675,
      "num_input_tokens_seen": 9565504,
      "step": 1933,
      "train_runtime": 6693.0761,
      "train_tokens_per_second": 1429.164
    },
    {
      "epoch": 0.7113563218390805,
      "grad_norm": 0.33759135007858276,
      "learning_rate": 5.792188651436994e-05,
      "loss": 1.0943,
      "num_input_tokens_seen": 9569184,
      "step": 1934,
      "train_runtime": 6695.7085,
      "train_tokens_per_second": 1429.152
    },
    {
      "epoch": 0.7117241379310345,
      "grad_norm": 0.3741821348667145,
      "learning_rate": 5.78481945467944e-05,
      "loss": 1.3709,
      "num_input_tokens_seen": 9572320,
      "step": 1935,
      "train_runtime": 6698.0788,
      "train_tokens_per_second": 1429.114
    },
    {
      "epoch": 0.7120919540229885,
      "grad_norm": 0.36867663264274597,
      "learning_rate": 5.777450257921887e-05,
      "loss": 1.0813,
      "num_input_tokens_seen": 9577968,
      "step": 1936,
      "train_runtime": 6701.775,
      "train_tokens_per_second": 1429.169
    },
    {
      "epoch": 0.7124597701149425,
      "grad_norm": 0.33169442415237427,
      "learning_rate": 5.770081061164333e-05,
      "loss": 1.0825,
      "num_input_tokens_seen": 9581776,
      "step": 1937,
      "train_runtime": 6704.4381,
      "train_tokens_per_second": 1429.169
    },
    {
      "epoch": 0.7128275862068966,
      "grad_norm": 0.35956621170043945,
      "learning_rate": 5.76271186440678e-05,
      "loss": 1.2338,
      "num_input_tokens_seen": 9585360,
      "step": 1938,
      "train_runtime": 6707.0328,
      "train_tokens_per_second": 1429.151
    },
    {
      "epoch": 0.7131954022988506,
      "grad_norm": 0.36173951625823975,
      "learning_rate": 5.7553426676492263e-05,
      "loss": 1.1735,
      "num_input_tokens_seen": 9591344,
      "step": 1939,
      "train_runtime": 6710.918,
      "train_tokens_per_second": 1429.215
    },
    {
      "epoch": 0.7135632183908046,
      "grad_norm": 0.4012773931026459,
      "learning_rate": 5.7479734708916735e-05,
      "loss": 1.2553,
      "num_input_tokens_seen": 9595280,
      "step": 1940,
      "train_runtime": 6713.7203,
      "train_tokens_per_second": 1429.205
    },
    {
      "epoch": 0.7139310344827586,
      "grad_norm": 0.36968138813972473,
      "learning_rate": 5.740604274134119e-05,
      "loss": 1.0821,
      "num_input_tokens_seen": 9599344,
      "step": 1941,
      "train_runtime": 6716.5218,
      "train_tokens_per_second": 1429.214
    },
    {
      "epoch": 0.7142988505747127,
      "grad_norm": 0.3253781795501709,
      "learning_rate": 5.7332350773765665e-05,
      "loss": 1.2087,
      "num_input_tokens_seen": 9604096,
      "step": 1942,
      "train_runtime": 6719.7493,
      "train_tokens_per_second": 1429.234
    },
    {
      "epoch": 0.7146666666666667,
      "grad_norm": 0.34529078006744385,
      "learning_rate": 5.725865880619012e-05,
      "loss": 1.2329,
      "num_input_tokens_seen": 9611168,
      "step": 1943,
      "train_runtime": 6724.193,
      "train_tokens_per_second": 1429.341
    },
    {
      "epoch": 0.7150344827586207,
      "grad_norm": 0.31104111671447754,
      "learning_rate": 5.7184966838614594e-05,
      "loss": 1.1661,
      "num_input_tokens_seen": 9619568,
      "step": 1944,
      "train_runtime": 6729.3699,
      "train_tokens_per_second": 1429.49
    },
    {
      "epoch": 0.7154022988505747,
      "grad_norm": 0.36981260776519775,
      "learning_rate": 5.711127487103905e-05,
      "loss": 1.2791,
      "num_input_tokens_seen": 9622624,
      "step": 1945,
      "train_runtime": 6731.669,
      "train_tokens_per_second": 1429.456
    },
    {
      "epoch": 0.7157701149425287,
      "grad_norm": 0.3541151285171509,
      "learning_rate": 5.7037582903463524e-05,
      "loss": 1.2053,
      "num_input_tokens_seen": 9626432,
      "step": 1946,
      "train_runtime": 6734.3535,
      "train_tokens_per_second": 1429.452
    },
    {
      "epoch": 0.7161379310344828,
      "grad_norm": 0.38977038860321045,
      "learning_rate": 5.696389093588799e-05,
      "loss": 1.2271,
      "num_input_tokens_seen": 9631600,
      "step": 1947,
      "train_runtime": 6737.8178,
      "train_tokens_per_second": 1429.484
    },
    {
      "epoch": 0.7165057471264368,
      "grad_norm": 0.3207605481147766,
      "learning_rate": 5.689019896831246e-05,
      "loss": 1.1311,
      "num_input_tokens_seen": 9634656,
      "step": 1948,
      "train_runtime": 6740.1266,
      "train_tokens_per_second": 1429.447
    },
    {
      "epoch": 0.7168735632183908,
      "grad_norm": 0.36478957533836365,
      "learning_rate": 5.681650700073692e-05,
      "loss": 1.3074,
      "num_input_tokens_seen": 9639008,
      "step": 1949,
      "train_runtime": 6743.1415,
      "train_tokens_per_second": 1429.454
    },
    {
      "epoch": 0.7172413793103448,
      "grad_norm": 0.3832957446575165,
      "learning_rate": 5.674281503316139e-05,
      "loss": 0.978,
      "num_input_tokens_seen": 9644464,
      "step": 1950,
      "train_runtime": 6746.6232,
      "train_tokens_per_second": 1429.525
    },
    {
      "epoch": 0.7176091954022988,
      "grad_norm": 0.3567690849304199,
      "learning_rate": 5.666912306558585e-05,
      "loss": 1.2101,
      "num_input_tokens_seen": 9650656,
      "step": 1951,
      "train_runtime": 6752.649,
      "train_tokens_per_second": 1429.166
    },
    {
      "epoch": 0.7179770114942529,
      "grad_norm": 0.30054405331611633,
      "learning_rate": 5.659543109801032e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 9656368,
      "step": 1952,
      "train_runtime": 6756.4971,
      "train_tokens_per_second": 1429.197
    },
    {
      "epoch": 0.7183448275862069,
      "grad_norm": 0.3316233456134796,
      "learning_rate": 5.652173913043478e-05,
      "loss": 0.9968,
      "num_input_tokens_seen": 9662848,
      "step": 1953,
      "train_runtime": 6760.6282,
      "train_tokens_per_second": 1429.283
    },
    {
      "epoch": 0.7187126436781609,
      "grad_norm": 0.38092342019081116,
      "learning_rate": 5.644804716285925e-05,
      "loss": 1.2992,
      "num_input_tokens_seen": 9666080,
      "step": 1954,
      "train_runtime": 6763.012,
      "train_tokens_per_second": 1429.257
    },
    {
      "epoch": 0.7190804597701149,
      "grad_norm": 0.3728266656398773,
      "learning_rate": 5.6374355195283714e-05,
      "loss": 1.3001,
      "num_input_tokens_seen": 9670416,
      "step": 1955,
      "train_runtime": 6766.0335,
      "train_tokens_per_second": 1429.259
    },
    {
      "epoch": 0.7194482758620689,
      "grad_norm": 0.3573281764984131,
      "learning_rate": 5.6300663227708186e-05,
      "loss": 1.0498,
      "num_input_tokens_seen": 9675520,
      "step": 1956,
      "train_runtime": 6769.5092,
      "train_tokens_per_second": 1429.279
    },
    {
      "epoch": 0.719816091954023,
      "grad_norm": 0.3668102025985718,
      "learning_rate": 5.6226971260132644e-05,
      "loss": 1.3063,
      "num_input_tokens_seen": 9679248,
      "step": 1957,
      "train_runtime": 6772.1412,
      "train_tokens_per_second": 1429.274
    },
    {
      "epoch": 0.720183908045977,
      "grad_norm": 0.3725837171077728,
      "learning_rate": 5.6153279292557115e-05,
      "loss": 1.2212,
      "num_input_tokens_seen": 9683536,
      "step": 1958,
      "train_runtime": 6775.0333,
      "train_tokens_per_second": 1429.297
    },
    {
      "epoch": 0.720551724137931,
      "grad_norm": 0.37918996810913086,
      "learning_rate": 5.607958732498157e-05,
      "loss": 1.1328,
      "num_input_tokens_seen": 9688480,
      "step": 1959,
      "train_runtime": 6778.4599,
      "train_tokens_per_second": 1429.304
    },
    {
      "epoch": 0.720919540229885,
      "grad_norm": 0.3398841917514801,
      "learning_rate": 5.6005895357406045e-05,
      "loss": 1.0351,
      "num_input_tokens_seen": 9693968,
      "step": 1960,
      "train_runtime": 6781.9684,
      "train_tokens_per_second": 1429.374
    },
    {
      "epoch": 0.721287356321839,
      "grad_norm": 0.35706812143325806,
      "learning_rate": 5.593220338983051e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 9699376,
      "step": 1961,
      "train_runtime": 6785.5186,
      "train_tokens_per_second": 1429.423
    },
    {
      "epoch": 0.7216551724137931,
      "grad_norm": 0.3878159821033478,
      "learning_rate": 5.5858511422254975e-05,
      "loss": 1.3867,
      "num_input_tokens_seen": 9703024,
      "step": 1962,
      "train_runtime": 6788.0374,
      "train_tokens_per_second": 1429.43
    },
    {
      "epoch": 0.7220229885057471,
      "grad_norm": 0.35455986857414246,
      "learning_rate": 5.578481945467944e-05,
      "loss": 1.072,
      "num_input_tokens_seen": 9707088,
      "step": 1963,
      "train_runtime": 6790.8753,
      "train_tokens_per_second": 1429.431
    },
    {
      "epoch": 0.7223908045977011,
      "grad_norm": 0.3683864176273346,
      "learning_rate": 5.571112748710391e-05,
      "loss": 1.2966,
      "num_input_tokens_seen": 9710800,
      "step": 1964,
      "train_runtime": 6793.453,
      "train_tokens_per_second": 1429.435
    },
    {
      "epoch": 0.7227586206896551,
      "grad_norm": 0.3618786633014679,
      "learning_rate": 5.563743551952837e-05,
      "loss": 1.1108,
      "num_input_tokens_seen": 9714576,
      "step": 1965,
      "train_runtime": 6796.1236,
      "train_tokens_per_second": 1429.429
    },
    {
      "epoch": 0.7231264367816091,
      "grad_norm": 0.44552484154701233,
      "learning_rate": 5.556374355195284e-05,
      "loss": 1.1554,
      "num_input_tokens_seen": 9718384,
      "step": 1966,
      "train_runtime": 6798.8977,
      "train_tokens_per_second": 1429.406
    },
    {
      "epoch": 0.7234942528735632,
      "grad_norm": 0.36063694953918457,
      "learning_rate": 5.54900515843773e-05,
      "loss": 1.2525,
      "num_input_tokens_seen": 9723264,
      "step": 1967,
      "train_runtime": 6802.1861,
      "train_tokens_per_second": 1429.432
    },
    {
      "epoch": 0.7238620689655172,
      "grad_norm": 0.39343586564064026,
      "learning_rate": 5.541635961680177e-05,
      "loss": 1.1781,
      "num_input_tokens_seen": 9727472,
      "step": 1968,
      "train_runtime": 6805.0746,
      "train_tokens_per_second": 1429.444
    },
    {
      "epoch": 0.7242298850574712,
      "grad_norm": 0.3574298918247223,
      "learning_rate": 5.5342667649226235e-05,
      "loss": 1.1394,
      "num_input_tokens_seen": 9735920,
      "step": 1969,
      "train_runtime": 6810.3377,
      "train_tokens_per_second": 1429.58
    },
    {
      "epoch": 0.7245977011494252,
      "grad_norm": 0.3609323501586914,
      "learning_rate": 5.52689756816507e-05,
      "loss": 1.2069,
      "num_input_tokens_seen": 9740112,
      "step": 1970,
      "train_runtime": 6813.3353,
      "train_tokens_per_second": 1429.566
    },
    {
      "epoch": 0.7249655172413794,
      "grad_norm": 0.3511142134666443,
      "learning_rate": 5.5195283714075165e-05,
      "loss": 1.0898,
      "num_input_tokens_seen": 9748624,
      "step": 1971,
      "train_runtime": 6818.7,
      "train_tokens_per_second": 1429.69
    },
    {
      "epoch": 0.7253333333333334,
      "grad_norm": 0.3796879053115845,
      "learning_rate": 5.5121591746499636e-05,
      "loss": 1.2731,
      "num_input_tokens_seen": 9752032,
      "step": 1972,
      "train_runtime": 6821.2686,
      "train_tokens_per_second": 1429.651
    },
    {
      "epoch": 0.7257011494252874,
      "grad_norm": 0.3540513217449188,
      "learning_rate": 5.5047899778924094e-05,
      "loss": 1.1299,
      "num_input_tokens_seen": 9757760,
      "step": 1973,
      "train_runtime": 6825.0448,
      "train_tokens_per_second": 1429.699
    },
    {
      "epoch": 0.7260689655172414,
      "grad_norm": 0.33710286021232605,
      "learning_rate": 5.4974207811348566e-05,
      "loss": 1.0978,
      "num_input_tokens_seen": 9761824,
      "step": 1974,
      "train_runtime": 6827.8741,
      "train_tokens_per_second": 1429.702
    },
    {
      "epoch": 0.7264367816091954,
      "grad_norm": 0.35127952694892883,
      "learning_rate": 5.4900515843773024e-05,
      "loss": 1.1884,
      "num_input_tokens_seen": 9765088,
      "step": 1975,
      "train_runtime": 6830.2605,
      "train_tokens_per_second": 1429.68
    },
    {
      "epoch": 0.7268045977011495,
      "grad_norm": 0.4071696102619171,
      "learning_rate": 5.4826823876197496e-05,
      "loss": 1.2901,
      "num_input_tokens_seen": 9769408,
      "step": 1976,
      "train_runtime": 6833.2158,
      "train_tokens_per_second": 1429.694
    },
    {
      "epoch": 0.7271724137931035,
      "grad_norm": 0.38829708099365234,
      "learning_rate": 5.475313190862196e-05,
      "loss": 1.2401,
      "num_input_tokens_seen": 9775056,
      "step": 1977,
      "train_runtime": 6836.9118,
      "train_tokens_per_second": 1429.747
    },
    {
      "epoch": 0.7275402298850575,
      "grad_norm": 0.4717738628387451,
      "learning_rate": 5.4679439941046425e-05,
      "loss": 1.622,
      "num_input_tokens_seen": 9779168,
      "step": 1978,
      "train_runtime": 6839.8034,
      "train_tokens_per_second": 1429.744
    },
    {
      "epoch": 0.7279080459770115,
      "grad_norm": 0.39373430609703064,
      "learning_rate": 5.460574797347089e-05,
      "loss": 1.2383,
      "num_input_tokens_seen": 9782848,
      "step": 1979,
      "train_runtime": 6842.4602,
      "train_tokens_per_second": 1429.727
    },
    {
      "epoch": 0.7282758620689656,
      "grad_norm": 0.38938674330711365,
      "learning_rate": 5.453205600589536e-05,
      "loss": 1.1347,
      "num_input_tokens_seen": 9787456,
      "step": 1980,
      "train_runtime": 6845.5015,
      "train_tokens_per_second": 1429.765
    },
    {
      "epoch": 0.7286436781609196,
      "grad_norm": 0.37794309854507446,
      "learning_rate": 5.445836403831983e-05,
      "loss": 1.3076,
      "num_input_tokens_seen": 9793664,
      "step": 1981,
      "train_runtime": 6851.719,
      "train_tokens_per_second": 1429.373
    },
    {
      "epoch": 0.7290114942528736,
      "grad_norm": 0.4039067327976227,
      "learning_rate": 5.438467207074429e-05,
      "loss": 1.1806,
      "num_input_tokens_seen": 9797184,
      "step": 1982,
      "train_runtime": 6854.2879,
      "train_tokens_per_second": 1429.351
    },
    {
      "epoch": 0.7293793103448276,
      "grad_norm": 0.37168362736701965,
      "learning_rate": 5.431098010316876e-05,
      "loss": 1.1072,
      "num_input_tokens_seen": 9803904,
      "step": 1983,
      "train_runtime": 6858.5662,
      "train_tokens_per_second": 1429.439
    },
    {
      "epoch": 0.7297471264367816,
      "grad_norm": 0.37610507011413574,
      "learning_rate": 5.423728813559322e-05,
      "loss": 1.2056,
      "num_input_tokens_seen": 9808304,
      "step": 1984,
      "train_runtime": 6861.5665,
      "train_tokens_per_second": 1429.456
    },
    {
      "epoch": 0.7301149425287357,
      "grad_norm": 0.36628156900405884,
      "learning_rate": 5.416359616801769e-05,
      "loss": 1.1143,
      "num_input_tokens_seen": 9812736,
      "step": 1985,
      "train_runtime": 6864.6168,
      "train_tokens_per_second": 1429.466
    },
    {
      "epoch": 0.7304827586206897,
      "grad_norm": 0.36420294642448425,
      "learning_rate": 5.408990420044216e-05,
      "loss": 1.2567,
      "num_input_tokens_seen": 9816304,
      "step": 1986,
      "train_runtime": 6867.2194,
      "train_tokens_per_second": 1429.444
    },
    {
      "epoch": 0.7308505747126437,
      "grad_norm": 0.3996455669403076,
      "learning_rate": 5.401621223286662e-05,
      "loss": 1.2983,
      "num_input_tokens_seen": 9820896,
      "step": 1987,
      "train_runtime": 6870.3501,
      "train_tokens_per_second": 1429.461
    },
    {
      "epoch": 0.7312183908045977,
      "grad_norm": 0.36641180515289307,
      "learning_rate": 5.394252026529109e-05,
      "loss": 1.2829,
      "num_input_tokens_seen": 9824720,
      "step": 1988,
      "train_runtime": 6873.1052,
      "train_tokens_per_second": 1429.444
    },
    {
      "epoch": 0.7315862068965517,
      "grad_norm": 0.40277162194252014,
      "learning_rate": 5.386882829771556e-05,
      "loss": 1.2458,
      "num_input_tokens_seen": 9828624,
      "step": 1989,
      "train_runtime": 6875.8833,
      "train_tokens_per_second": 1429.434
    },
    {
      "epoch": 0.7319540229885058,
      "grad_norm": 0.3662918508052826,
      "learning_rate": 5.379513633014002e-05,
      "loss": 1.4541,
      "num_input_tokens_seen": 9834880,
      "step": 1990,
      "train_runtime": 6879.9502,
      "train_tokens_per_second": 1429.499
    },
    {
      "epoch": 0.7323218390804598,
      "grad_norm": 0.3359386622905731,
      "learning_rate": 5.372144436256449e-05,
      "loss": 1.1518,
      "num_input_tokens_seen": 9839440,
      "step": 1991,
      "train_runtime": 6883.0285,
      "train_tokens_per_second": 1429.522
    },
    {
      "epoch": 0.7326896551724138,
      "grad_norm": 0.34676942229270935,
      "learning_rate": 5.3647752394988946e-05,
      "loss": 1.2788,
      "num_input_tokens_seen": 9844256,
      "step": 1992,
      "train_runtime": 6886.3564,
      "train_tokens_per_second": 1429.53
    },
    {
      "epoch": 0.7330574712643678,
      "grad_norm": 0.3641933500766754,
      "learning_rate": 5.357406042741342e-05,
      "loss": 1.2664,
      "num_input_tokens_seen": 9848016,
      "step": 1993,
      "train_runtime": 6889.1178,
      "train_tokens_per_second": 1429.503
    },
    {
      "epoch": 0.7334252873563218,
      "grad_norm": 0.3946431875228882,
      "learning_rate": 5.350036845983788e-05,
      "loss": 1.2664,
      "num_input_tokens_seen": 9852464,
      "step": 1994,
      "train_runtime": 6892.2322,
      "train_tokens_per_second": 1429.503
    },
    {
      "epoch": 0.7337931034482759,
      "grad_norm": 0.3413759469985962,
      "learning_rate": 5.342667649226235e-05,
      "loss": 1.3558,
      "num_input_tokens_seen": 9857840,
      "step": 1995,
      "train_runtime": 6895.7472,
      "train_tokens_per_second": 1429.554
    },
    {
      "epoch": 0.7341609195402299,
      "grad_norm": 0.3781610429286957,
      "learning_rate": 5.335298452468681e-05,
      "loss": 1.2013,
      "num_input_tokens_seen": 9864672,
      "step": 1996,
      "train_runtime": 6900.1912,
      "train_tokens_per_second": 1429.623
    },
    {
      "epoch": 0.7345287356321839,
      "grad_norm": 0.375271201133728,
      "learning_rate": 5.3279292557111284e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 9868688,
      "step": 1997,
      "train_runtime": 6903.0303,
      "train_tokens_per_second": 1429.617
    },
    {
      "epoch": 0.7348965517241379,
      "grad_norm": 0.3385009765625,
      "learning_rate": 5.320560058953574e-05,
      "loss": 1.2295,
      "num_input_tokens_seen": 9873968,
      "step": 1998,
      "train_runtime": 6906.6201,
      "train_tokens_per_second": 1429.638
    },
    {
      "epoch": 0.735264367816092,
      "grad_norm": 0.35097870230674744,
      "learning_rate": 5.3131908621960214e-05,
      "loss": 1.2998,
      "num_input_tokens_seen": 9880016,
      "step": 1999,
      "train_runtime": 6910.643,
      "train_tokens_per_second": 1429.681
    },
    {
      "epoch": 0.735632183908046,
      "grad_norm": 0.3186858892440796,
      "learning_rate": 5.305821665438467e-05,
      "loss": 1.31,
      "num_input_tokens_seen": 9885664,
      "step": 2000,
      "train_runtime": 6914.3079,
      "train_tokens_per_second": 1429.74
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.37143802642822266,
      "learning_rate": 5.298452468680914e-05,
      "loss": 1.2539,
      "num_input_tokens_seen": 9889280,
      "step": 2001,
      "train_runtime": 6916.9153,
      "train_tokens_per_second": 1429.724
    },
    {
      "epoch": 0.736367816091954,
      "grad_norm": 0.34992706775665283,
      "learning_rate": 5.291083271923361e-05,
      "loss": 1.2064,
      "num_input_tokens_seen": 9894432,
      "step": 2002,
      "train_runtime": 6920.3477,
      "train_tokens_per_second": 1429.759
    },
    {
      "epoch": 0.736735632183908,
      "grad_norm": 0.3759583830833435,
      "learning_rate": 5.283714075165808e-05,
      "loss": 1.2319,
      "num_input_tokens_seen": 9902912,
      "step": 2003,
      "train_runtime": 6925.7199,
      "train_tokens_per_second": 1429.875
    },
    {
      "epoch": 0.737103448275862,
      "grad_norm": 0.30717042088508606,
      "learning_rate": 5.276344878408254e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 9907472,
      "step": 2004,
      "train_runtime": 6928.7942,
      "train_tokens_per_second": 1429.898
    },
    {
      "epoch": 0.7374712643678161,
      "grad_norm": 0.3640108108520508,
      "learning_rate": 5.268975681650701e-05,
      "loss": 1.0412,
      "num_input_tokens_seen": 9912592,
      "step": 2005,
      "train_runtime": 6932.1369,
      "train_tokens_per_second": 1429.948
    },
    {
      "epoch": 0.7378390804597701,
      "grad_norm": 0.37335336208343506,
      "learning_rate": 5.261606484893147e-05,
      "loss": 1.1765,
      "num_input_tokens_seen": 9916336,
      "step": 2006,
      "train_runtime": 6934.7502,
      "train_tokens_per_second": 1429.949
    },
    {
      "epoch": 0.7382068965517241,
      "grad_norm": 0.3571365475654602,
      "learning_rate": 5.254237288135594e-05,
      "loss": 1.3221,
      "num_input_tokens_seen": 9921552,
      "step": 2007,
      "train_runtime": 6938.2039,
      "train_tokens_per_second": 1429.989
    },
    {
      "epoch": 0.7385747126436781,
      "grad_norm": 0.35716331005096436,
      "learning_rate": 5.24686809137804e-05,
      "loss": 1.0862,
      "num_input_tokens_seen": 9928768,
      "step": 2008,
      "train_runtime": 6942.8361,
      "train_tokens_per_second": 1430.074
    },
    {
      "epoch": 0.7389425287356322,
      "grad_norm": 0.36532944440841675,
      "learning_rate": 5.239498894620487e-05,
      "loss": 1.1792,
      "num_input_tokens_seen": 9933120,
      "step": 2009,
      "train_runtime": 6945.8431,
      "train_tokens_per_second": 1430.081
    },
    {
      "epoch": 0.7393103448275862,
      "grad_norm": 0.3846795856952667,
      "learning_rate": 5.2321296978629333e-05,
      "loss": 1.2027,
      "num_input_tokens_seen": 9937104,
      "step": 2010,
      "train_runtime": 6948.6349,
      "train_tokens_per_second": 1430.08
    },
    {
      "epoch": 0.7396781609195402,
      "grad_norm": 0.3514174818992615,
      "learning_rate": 5.2247605011053805e-05,
      "loss": 1.1343,
      "num_input_tokens_seen": 9940672,
      "step": 2011,
      "train_runtime": 6953.7508,
      "train_tokens_per_second": 1429.541
    },
    {
      "epoch": 0.7400459770114942,
      "grad_norm": 0.35756951570510864,
      "learning_rate": 5.217391304347826e-05,
      "loss": 1.1809,
      "num_input_tokens_seen": 9945344,
      "step": 2012,
      "train_runtime": 6956.8756,
      "train_tokens_per_second": 1429.57
    },
    {
      "epoch": 0.7404137931034482,
      "grad_norm": 0.39096799492836,
      "learning_rate": 5.2100221075902735e-05,
      "loss": 1.3275,
      "num_input_tokens_seen": 9948576,
      "step": 2013,
      "train_runtime": 6959.2651,
      "train_tokens_per_second": 1429.544
    },
    {
      "epoch": 0.7407816091954023,
      "grad_norm": 0.38563162088394165,
      "learning_rate": 5.202652910832719e-05,
      "loss": 1.1817,
      "num_input_tokens_seen": 9951904,
      "step": 2014,
      "train_runtime": 6961.7263,
      "train_tokens_per_second": 1429.517
    },
    {
      "epoch": 0.7411494252873563,
      "grad_norm": 0.3683525323867798,
      "learning_rate": 5.1952837140751664e-05,
      "loss": 1.2449,
      "num_input_tokens_seen": 9955904,
      "step": 2015,
      "train_runtime": 6964.6006,
      "train_tokens_per_second": 1429.501
    },
    {
      "epoch": 0.7415172413793103,
      "grad_norm": 0.4213060736656189,
      "learning_rate": 5.187914517317612e-05,
      "loss": 1.3772,
      "num_input_tokens_seen": 9961088,
      "step": 2016,
      "train_runtime": 6968.1352,
      "train_tokens_per_second": 1429.52
    },
    {
      "epoch": 0.7418850574712643,
      "grad_norm": 0.4007192552089691,
      "learning_rate": 5.1805453205600594e-05,
      "loss": 1.2908,
      "num_input_tokens_seen": 9966800,
      "step": 2017,
      "train_runtime": 6971.9148,
      "train_tokens_per_second": 1429.564
    },
    {
      "epoch": 0.7422528735632183,
      "grad_norm": 0.3853626251220703,
      "learning_rate": 5.173176123802506e-05,
      "loss": 1.3185,
      "num_input_tokens_seen": 9971472,
      "step": 2018,
      "train_runtime": 6975.0795,
      "train_tokens_per_second": 1429.585
    },
    {
      "epoch": 0.7426206896551724,
      "grad_norm": 0.39276832342147827,
      "learning_rate": 5.165806927044953e-05,
      "loss": 1.3106,
      "num_input_tokens_seen": 9976048,
      "step": 2019,
      "train_runtime": 6978.2776,
      "train_tokens_per_second": 1429.586
    },
    {
      "epoch": 0.7429885057471264,
      "grad_norm": 0.3647022247314453,
      "learning_rate": 5.158437730287399e-05,
      "loss": 1.2766,
      "num_input_tokens_seen": 9984800,
      "step": 2020,
      "train_runtime": 6983.6494,
      "train_tokens_per_second": 1429.74
    },
    {
      "epoch": 0.7433563218390805,
      "grad_norm": 0.3334828317165375,
      "learning_rate": 5.151068533529846e-05,
      "loss": 1.0391,
      "num_input_tokens_seen": 9992192,
      "step": 2021,
      "train_runtime": 6988.5539,
      "train_tokens_per_second": 1429.794
    },
    {
      "epoch": 0.7437241379310345,
      "grad_norm": 0.3582146465778351,
      "learning_rate": 5.143699336772292e-05,
      "loss": 1.1515,
      "num_input_tokens_seen": 9997376,
      "step": 2022,
      "train_runtime": 6992.075,
      "train_tokens_per_second": 1429.815
    },
    {
      "epoch": 0.7440919540229886,
      "grad_norm": 0.3577844798564911,
      "learning_rate": 5.136330140014739e-05,
      "loss": 1.2602,
      "num_input_tokens_seen": 10001856,
      "step": 2023,
      "train_runtime": 6995.0471,
      "train_tokens_per_second": 1429.848
    },
    {
      "epoch": 0.7444597701149426,
      "grad_norm": 0.36794063448905945,
      "learning_rate": 5.128960943257185e-05,
      "loss": 1.1986,
      "num_input_tokens_seen": 10005824,
      "step": 2024,
      "train_runtime": 6997.8952,
      "train_tokens_per_second": 1429.833
    },
    {
      "epoch": 0.7448275862068966,
      "grad_norm": 0.33305928111076355,
      "learning_rate": 5.121591746499632e-05,
      "loss": 1.1856,
      "num_input_tokens_seen": 10011776,
      "step": 2025,
      "train_runtime": 7001.6977,
      "train_tokens_per_second": 1429.907
    },
    {
      "epoch": 0.7451954022988506,
      "grad_norm": 0.3727646470069885,
      "learning_rate": 5.1142225497420784e-05,
      "loss": 1.1093,
      "num_input_tokens_seen": 10016912,
      "step": 2026,
      "train_runtime": 7005.0587,
      "train_tokens_per_second": 1429.954
    },
    {
      "epoch": 0.7455632183908046,
      "grad_norm": 0.42127686738967896,
      "learning_rate": 5.1068533529845256e-05,
      "loss": 1.3303,
      "num_input_tokens_seen": 10022400,
      "step": 2027,
      "train_runtime": 7008.6693,
      "train_tokens_per_second": 1430.0
    },
    {
      "epoch": 0.7459310344827587,
      "grad_norm": 0.36830392479896545,
      "learning_rate": 5.0994841562269714e-05,
      "loss": 1.3581,
      "num_input_tokens_seen": 10026128,
      "step": 2028,
      "train_runtime": 7011.3494,
      "train_tokens_per_second": 1429.986
    },
    {
      "epoch": 0.7462988505747127,
      "grad_norm": 0.3702619969844818,
      "learning_rate": 5.0921149594694185e-05,
      "loss": 1.1841,
      "num_input_tokens_seen": 10031888,
      "step": 2029,
      "train_runtime": 7015.1022,
      "train_tokens_per_second": 1430.042
    },
    {
      "epoch": 0.7466666666666667,
      "grad_norm": 0.3536003828048706,
      "learning_rate": 5.0847457627118643e-05,
      "loss": 1.2594,
      "num_input_tokens_seen": 10035856,
      "step": 2030,
      "train_runtime": 7017.9464,
      "train_tokens_per_second": 1430.027
    },
    {
      "epoch": 0.7470344827586207,
      "grad_norm": 0.38960978388786316,
      "learning_rate": 5.0773765659543115e-05,
      "loss": 1.2737,
      "num_input_tokens_seen": 10039216,
      "step": 2031,
      "train_runtime": 7020.6099,
      "train_tokens_per_second": 1429.964
    },
    {
      "epoch": 0.7474022988505747,
      "grad_norm": 0.4064590632915497,
      "learning_rate": 5.070007369196757e-05,
      "loss": 1.2461,
      "num_input_tokens_seen": 10045440,
      "step": 2032,
      "train_runtime": 7024.6007,
      "train_tokens_per_second": 1430.037
    },
    {
      "epoch": 0.7477701149425288,
      "grad_norm": 0.3747829794883728,
      "learning_rate": 5.0626381724392045e-05,
      "loss": 1.2312,
      "num_input_tokens_seen": 10048752,
      "step": 2033,
      "train_runtime": 7026.9578,
      "train_tokens_per_second": 1430.029
    },
    {
      "epoch": 0.7481379310344828,
      "grad_norm": 0.3794178068637848,
      "learning_rate": 5.055268975681651e-05,
      "loss": 1.337,
      "num_input_tokens_seen": 10053440,
      "step": 2034,
      "train_runtime": 7030.1801,
      "train_tokens_per_second": 1430.04
    },
    {
      "epoch": 0.7485057471264368,
      "grad_norm": 0.3810683488845825,
      "learning_rate": 5.047899778924098e-05,
      "loss": 1.2665,
      "num_input_tokens_seen": 10058368,
      "step": 2035,
      "train_runtime": 7033.4954,
      "train_tokens_per_second": 1430.067
    },
    {
      "epoch": 0.7488735632183908,
      "grad_norm": 0.3513549566268921,
      "learning_rate": 5.040530582166544e-05,
      "loss": 1.2018,
      "num_input_tokens_seen": 10062400,
      "step": 2036,
      "train_runtime": 7036.3181,
      "train_tokens_per_second": 1430.066
    },
    {
      "epoch": 0.7492413793103448,
      "grad_norm": 0.3836992383003235,
      "learning_rate": 5.033161385408991e-05,
      "loss": 1.1851,
      "num_input_tokens_seen": 10065888,
      "step": 2037,
      "train_runtime": 7038.7933,
      "train_tokens_per_second": 1430.059
    },
    {
      "epoch": 0.7496091954022989,
      "grad_norm": 0.34747523069381714,
      "learning_rate": 5.025792188651437e-05,
      "loss": 1.1054,
      "num_input_tokens_seen": 10069808,
      "step": 2038,
      "train_runtime": 7041.575,
      "train_tokens_per_second": 1430.051
    },
    {
      "epoch": 0.7499770114942529,
      "grad_norm": 0.37715470790863037,
      "learning_rate": 5.018422991893884e-05,
      "loss": 1.1988,
      "num_input_tokens_seen": 10074704,
      "step": 2039,
      "train_runtime": 7044.7484,
      "train_tokens_per_second": 1430.101
    },
    {
      "epoch": 0.7503448275862069,
      "grad_norm": 0.33090466260910034,
      "learning_rate": 5.01105379513633e-05,
      "loss": 1.1553,
      "num_input_tokens_seen": 10082032,
      "step": 2040,
      "train_runtime": 7049.3306,
      "train_tokens_per_second": 1430.211
    },
    {
      "epoch": 0.7507126436781609,
      "grad_norm": 0.39747190475463867,
      "learning_rate": 5.003684598378777e-05,
      "loss": 1.0709,
      "num_input_tokens_seen": 10086928,
      "step": 2041,
      "train_runtime": 7054.9362,
      "train_tokens_per_second": 1429.769
    },
    {
      "epoch": 0.751080459770115,
      "grad_norm": 0.3875967562198639,
      "learning_rate": 4.9963154016212235e-05,
      "loss": 1.2019,
      "num_input_tokens_seen": 10091072,
      "step": 2042,
      "train_runtime": 7057.7623,
      "train_tokens_per_second": 1429.783
    },
    {
      "epoch": 0.751448275862069,
      "grad_norm": 0.3462505638599396,
      "learning_rate": 4.9889462048636706e-05,
      "loss": 1.0891,
      "num_input_tokens_seen": 10096384,
      "step": 2043,
      "train_runtime": 7061.3305,
      "train_tokens_per_second": 1429.813
    },
    {
      "epoch": 0.751816091954023,
      "grad_norm": 0.36672234535217285,
      "learning_rate": 4.981577008106117e-05,
      "loss": 1.2469,
      "num_input_tokens_seen": 10099776,
      "step": 2044,
      "train_runtime": 7063.8023,
      "train_tokens_per_second": 1429.793
    },
    {
      "epoch": 0.752183908045977,
      "grad_norm": 0.32974258065223694,
      "learning_rate": 4.9742078113485636e-05,
      "loss": 1.2321,
      "num_input_tokens_seen": 10106224,
      "step": 2045,
      "train_runtime": 7068.0646,
      "train_tokens_per_second": 1429.843
    },
    {
      "epoch": 0.752551724137931,
      "grad_norm": 0.3559560179710388,
      "learning_rate": 4.96683861459101e-05,
      "loss": 1.103,
      "num_input_tokens_seen": 10112880,
      "step": 2046,
      "train_runtime": 7072.4201,
      "train_tokens_per_second": 1429.904
    },
    {
      "epoch": 0.752919540229885,
      "grad_norm": 0.3330234885215759,
      "learning_rate": 4.9594694178334566e-05,
      "loss": 1.187,
      "num_input_tokens_seen": 10116688,
      "step": 2047,
      "train_runtime": 7075.1164,
      "train_tokens_per_second": 1429.897
    },
    {
      "epoch": 0.7532873563218391,
      "grad_norm": 0.36663639545440674,
      "learning_rate": 4.952100221075903e-05,
      "loss": 1.1006,
      "num_input_tokens_seen": 10120224,
      "step": 2048,
      "train_runtime": 7077.6891,
      "train_tokens_per_second": 1429.877
    },
    {
      "epoch": 0.7536551724137931,
      "grad_norm": 0.3912426233291626,
      "learning_rate": 4.9447310243183495e-05,
      "loss": 1.3455,
      "num_input_tokens_seen": 10125088,
      "step": 2049,
      "train_runtime": 7080.9942,
      "train_tokens_per_second": 1429.896
    },
    {
      "epoch": 0.7540229885057471,
      "grad_norm": 0.34763529896736145,
      "learning_rate": 4.937361827560796e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 10130208,
      "step": 2050,
      "train_runtime": 7084.3911,
      "train_tokens_per_second": 1429.933
    },
    {
      "epoch": 0.7543908045977011,
      "grad_norm": 0.3467884957790375,
      "learning_rate": 4.929992630803243e-05,
      "loss": 1.2697,
      "num_input_tokens_seen": 10136432,
      "step": 2051,
      "train_runtime": 7088.47,
      "train_tokens_per_second": 1429.989
    },
    {
      "epoch": 0.7547586206896552,
      "grad_norm": 0.3817930817604065,
      "learning_rate": 4.9226234340456897e-05,
      "loss": 1.188,
      "num_input_tokens_seen": 10141984,
      "step": 2052,
      "train_runtime": 7092.1059,
      "train_tokens_per_second": 1430.038
    },
    {
      "epoch": 0.7551264367816092,
      "grad_norm": 0.3571681082248688,
      "learning_rate": 4.915254237288136e-05,
      "loss": 1.1959,
      "num_input_tokens_seen": 10151216,
      "step": 2053,
      "train_runtime": 7097.7786,
      "train_tokens_per_second": 1430.196
    },
    {
      "epoch": 0.7554942528735632,
      "grad_norm": 0.397423654794693,
      "learning_rate": 4.9078850405305826e-05,
      "loss": 1.2399,
      "num_input_tokens_seen": 10157984,
      "step": 2054,
      "train_runtime": 7102.1167,
      "train_tokens_per_second": 1430.276
    },
    {
      "epoch": 0.7558620689655172,
      "grad_norm": 0.38847312331199646,
      "learning_rate": 4.900515843773029e-05,
      "loss": 1.2781,
      "num_input_tokens_seen": 10166256,
      "step": 2055,
      "train_runtime": 7107.2779,
      "train_tokens_per_second": 1430.401
    },
    {
      "epoch": 0.7562298850574712,
      "grad_norm": 0.35216590762138367,
      "learning_rate": 4.8931466470154756e-05,
      "loss": 1.1085,
      "num_input_tokens_seen": 10170496,
      "step": 2056,
      "train_runtime": 7110.1578,
      "train_tokens_per_second": 1430.418
    },
    {
      "epoch": 0.7565977011494253,
      "grad_norm": 0.3902183175086975,
      "learning_rate": 4.885777450257922e-05,
      "loss": 1.262,
      "num_input_tokens_seen": 10176368,
      "step": 2057,
      "train_runtime": 7113.9998,
      "train_tokens_per_second": 1430.471
    },
    {
      "epoch": 0.7569655172413793,
      "grad_norm": 0.34320852160453796,
      "learning_rate": 4.878408253500369e-05,
      "loss": 1.0795,
      "num_input_tokens_seen": 10184000,
      "step": 2058,
      "train_runtime": 7118.7297,
      "train_tokens_per_second": 1430.592
    },
    {
      "epoch": 0.7573333333333333,
      "grad_norm": 0.35162466764450073,
      "learning_rate": 4.871039056742816e-05,
      "loss": 1.2492,
      "num_input_tokens_seen": 10187632,
      "step": 2059,
      "train_runtime": 7121.4246,
      "train_tokens_per_second": 1430.561
    },
    {
      "epoch": 0.7577011494252873,
      "grad_norm": 0.3756692111492157,
      "learning_rate": 4.863669859985262e-05,
      "loss": 1.214,
      "num_input_tokens_seen": 10191984,
      "step": 2060,
      "train_runtime": 7124.4596,
      "train_tokens_per_second": 1430.562
    },
    {
      "epoch": 0.7580689655172413,
      "grad_norm": 0.3153895437717438,
      "learning_rate": 4.856300663227709e-05,
      "loss": 1.0061,
      "num_input_tokens_seen": 10199488,
      "step": 2061,
      "train_runtime": 7129.2065,
      "train_tokens_per_second": 1430.662
    },
    {
      "epoch": 0.7584367816091954,
      "grad_norm": 0.3751019537448883,
      "learning_rate": 4.848931466470155e-05,
      "loss": 1.2012,
      "num_input_tokens_seen": 10202912,
      "step": 2062,
      "train_runtime": 7131.7605,
      "train_tokens_per_second": 1430.63
    },
    {
      "epoch": 0.7588045977011494,
      "grad_norm": 0.3560206890106201,
      "learning_rate": 4.8415622697126016e-05,
      "loss": 1.0746,
      "num_input_tokens_seen": 10207184,
      "step": 2063,
      "train_runtime": 7134.7109,
      "train_tokens_per_second": 1430.637
    },
    {
      "epoch": 0.7591724137931034,
      "grad_norm": 0.3726266920566559,
      "learning_rate": 4.834193072955048e-05,
      "loss": 1.2619,
      "num_input_tokens_seen": 10212992,
      "step": 2064,
      "train_runtime": 7138.4889,
      "train_tokens_per_second": 1430.694
    },
    {
      "epoch": 0.7595402298850574,
      "grad_norm": 0.39077332615852356,
      "learning_rate": 4.8268238761974946e-05,
      "loss": 1.2568,
      "num_input_tokens_seen": 10219312,
      "step": 2065,
      "train_runtime": 7142.6261,
      "train_tokens_per_second": 1430.75
    },
    {
      "epoch": 0.7599080459770114,
      "grad_norm": 0.3722257912158966,
      "learning_rate": 4.819454679439942e-05,
      "loss": 1.113,
      "num_input_tokens_seen": 10224768,
      "step": 2066,
      "train_runtime": 7146.332,
      "train_tokens_per_second": 1430.771
    },
    {
      "epoch": 0.7602758620689655,
      "grad_norm": 0.3605303466320038,
      "learning_rate": 4.812085482682388e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 10230400,
      "step": 2067,
      "train_runtime": 7150.0216,
      "train_tokens_per_second": 1430.821
    },
    {
      "epoch": 0.7606436781609195,
      "grad_norm": 0.36080196499824524,
      "learning_rate": 4.804716285924835e-05,
      "loss": 1.1532,
      "num_input_tokens_seen": 10238480,
      "step": 2068,
      "train_runtime": 7155.0315,
      "train_tokens_per_second": 1430.948
    },
    {
      "epoch": 0.7610114942528735,
      "grad_norm": 0.35726282000541687,
      "learning_rate": 4.797347089167281e-05,
      "loss": 1.1386,
      "num_input_tokens_seen": 10242720,
      "step": 2069,
      "train_runtime": 7158.0099,
      "train_tokens_per_second": 1430.945
    },
    {
      "epoch": 0.7613793103448275,
      "grad_norm": 0.38103166222572327,
      "learning_rate": 4.789977892409728e-05,
      "loss": 1.1797,
      "num_input_tokens_seen": 10246576,
      "step": 2070,
      "train_runtime": 7160.7166,
      "train_tokens_per_second": 1430.943
    },
    {
      "epoch": 0.7617471264367817,
      "grad_norm": 0.3718661367893219,
      "learning_rate": 4.782608695652174e-05,
      "loss": 1.1831,
      "num_input_tokens_seen": 10250816,
      "step": 2071,
      "train_runtime": 7165.7955,
      "train_tokens_per_second": 1430.52
    },
    {
      "epoch": 0.7621149425287357,
      "grad_norm": 0.32570597529411316,
      "learning_rate": 4.7752394988946207e-05,
      "loss": 1.1222,
      "num_input_tokens_seen": 10256656,
      "step": 2072,
      "train_runtime": 7169.5766,
      "train_tokens_per_second": 1430.58
    },
    {
      "epoch": 0.7624827586206897,
      "grad_norm": 0.37853527069091797,
      "learning_rate": 4.767870302137067e-05,
      "loss": 1.297,
      "num_input_tokens_seen": 10261056,
      "step": 2073,
      "train_runtime": 7172.5789,
      "train_tokens_per_second": 1430.595
    },
    {
      "epoch": 0.7628505747126437,
      "grad_norm": 0.3393378257751465,
      "learning_rate": 4.760501105379514e-05,
      "loss": 1.0361,
      "num_input_tokens_seen": 10264256,
      "step": 2074,
      "train_runtime": 7174.9787,
      "train_tokens_per_second": 1430.563
    },
    {
      "epoch": 0.7632183908045977,
      "grad_norm": 0.3383787274360657,
      "learning_rate": 4.753131908621961e-05,
      "loss": 1.1802,
      "num_input_tokens_seen": 10268752,
      "step": 2075,
      "train_runtime": 7178.038,
      "train_tokens_per_second": 1430.579
    },
    {
      "epoch": 0.7635862068965518,
      "grad_norm": 0.3437236249446869,
      "learning_rate": 4.745762711864407e-05,
      "loss": 1.0817,
      "num_input_tokens_seen": 10272240,
      "step": 2076,
      "train_runtime": 7180.5496,
      "train_tokens_per_second": 1430.565
    },
    {
      "epoch": 0.7639540229885058,
      "grad_norm": 0.39617833495140076,
      "learning_rate": 4.738393515106854e-05,
      "loss": 1.3022,
      "num_input_tokens_seen": 10277056,
      "step": 2077,
      "train_runtime": 7183.7675,
      "train_tokens_per_second": 1430.594
    },
    {
      "epoch": 0.7643218390804598,
      "grad_norm": 0.34657150506973267,
      "learning_rate": 4.7310243183493e-05,
      "loss": 1.2007,
      "num_input_tokens_seen": 10281104,
      "step": 2078,
      "train_runtime": 7186.5477,
      "train_tokens_per_second": 1430.604
    },
    {
      "epoch": 0.7646896551724138,
      "grad_norm": 0.37454310059547424,
      "learning_rate": 4.723655121591747e-05,
      "loss": 1.2335,
      "num_input_tokens_seen": 10285968,
      "step": 2079,
      "train_runtime": 7189.7481,
      "train_tokens_per_second": 1430.644
    },
    {
      "epoch": 0.7650574712643678,
      "grad_norm": 0.3676322400569916,
      "learning_rate": 4.716285924834193e-05,
      "loss": 1.0868,
      "num_input_tokens_seen": 10291136,
      "step": 2080,
      "train_runtime": 7193.1111,
      "train_tokens_per_second": 1430.693
    },
    {
      "epoch": 0.7654252873563219,
      "grad_norm": 0.39924076199531555,
      "learning_rate": 4.70891672807664e-05,
      "loss": 1.2072,
      "num_input_tokens_seen": 10296560,
      "step": 2081,
      "train_runtime": 7196.8275,
      "train_tokens_per_second": 1430.708
    },
    {
      "epoch": 0.7657931034482759,
      "grad_norm": 0.3459792137145996,
      "learning_rate": 4.701547531319087e-05,
      "loss": 1.1102,
      "num_input_tokens_seen": 10302400,
      "step": 2082,
      "train_runtime": 7200.6383,
      "train_tokens_per_second": 1430.762
    },
    {
      "epoch": 0.7661609195402299,
      "grad_norm": 0.37597450613975525,
      "learning_rate": 4.694178334561533e-05,
      "loss": 1.1711,
      "num_input_tokens_seen": 10308656,
      "step": 2083,
      "train_runtime": 7204.6843,
      "train_tokens_per_second": 1430.827
    },
    {
      "epoch": 0.7665287356321839,
      "grad_norm": 0.3689568042755127,
      "learning_rate": 4.68680913780398e-05,
      "loss": 1.0629,
      "num_input_tokens_seen": 10317856,
      "step": 2084,
      "train_runtime": 7210.2449,
      "train_tokens_per_second": 1430.999
    },
    {
      "epoch": 0.766896551724138,
      "grad_norm": 0.4066593050956726,
      "learning_rate": 4.679439941046426e-05,
      "loss": 1.2215,
      "num_input_tokens_seen": 10322560,
      "step": 2085,
      "train_runtime": 7213.4809,
      "train_tokens_per_second": 1431.01
    },
    {
      "epoch": 0.767264367816092,
      "grad_norm": 0.3608400523662567,
      "learning_rate": 4.672070744288873e-05,
      "loss": 1.2937,
      "num_input_tokens_seen": 10329680,
      "step": 2086,
      "train_runtime": 7218.0218,
      "train_tokens_per_second": 1431.096
    },
    {
      "epoch": 0.767632183908046,
      "grad_norm": 0.3691215217113495,
      "learning_rate": 4.664701547531319e-05,
      "loss": 1.1806,
      "num_input_tokens_seen": 10336144,
      "step": 2087,
      "train_runtime": 7222.1782,
      "train_tokens_per_second": 1431.167
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.3937698304653168,
      "learning_rate": 4.657332350773766e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 10339504,
      "step": 2088,
      "train_runtime": 7224.5675,
      "train_tokens_per_second": 1431.159
    },
    {
      "epoch": 0.768367816091954,
      "grad_norm": 0.3723428249359131,
      "learning_rate": 4.649963154016212e-05,
      "loss": 1.1998,
      "num_input_tokens_seen": 10343408,
      "step": 2089,
      "train_runtime": 7227.3515,
      "train_tokens_per_second": 1431.148
    },
    {
      "epoch": 0.7687356321839081,
      "grad_norm": 0.3708900213241577,
      "learning_rate": 4.6425939572586594e-05,
      "loss": 1.1635,
      "num_input_tokens_seen": 10348672,
      "step": 2090,
      "train_runtime": 7230.8118,
      "train_tokens_per_second": 1431.191
    },
    {
      "epoch": 0.7691034482758621,
      "grad_norm": 0.36263126134872437,
      "learning_rate": 4.635224760501106e-05,
      "loss": 1.3147,
      "num_input_tokens_seen": 10354880,
      "step": 2091,
      "train_runtime": 7234.7381,
      "train_tokens_per_second": 1431.272
    },
    {
      "epoch": 0.7694712643678161,
      "grad_norm": 0.3652060329914093,
      "learning_rate": 4.627855563743552e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 10358752,
      "step": 2092,
      "train_runtime": 7237.4355,
      "train_tokens_per_second": 1431.274
    },
    {
      "epoch": 0.7698390804597701,
      "grad_norm": 0.3801341652870178,
      "learning_rate": 4.620486366985999e-05,
      "loss": 1.2374,
      "num_input_tokens_seen": 10362608,
      "step": 2093,
      "train_runtime": 7240.1998,
      "train_tokens_per_second": 1431.26
    },
    {
      "epoch": 0.7702068965517241,
      "grad_norm": 0.35437583923339844,
      "learning_rate": 4.613117170228445e-05,
      "loss": 1.2502,
      "num_input_tokens_seen": 10366560,
      "step": 2094,
      "train_runtime": 7243.007,
      "train_tokens_per_second": 1431.251
    },
    {
      "epoch": 0.7705747126436782,
      "grad_norm": 0.3846490681171417,
      "learning_rate": 4.605747973470892e-05,
      "loss": 1.2677,
      "num_input_tokens_seen": 10372224,
      "step": 2095,
      "train_runtime": 7246.6955,
      "train_tokens_per_second": 1431.304
    },
    {
      "epoch": 0.7709425287356322,
      "grad_norm": 0.3832799792289734,
      "learning_rate": 4.598378776713338e-05,
      "loss": 1.2278,
      "num_input_tokens_seen": 10379280,
      "step": 2096,
      "train_runtime": 7251.1839,
      "train_tokens_per_second": 1431.391
    },
    {
      "epoch": 0.7713103448275862,
      "grad_norm": 0.3387010395526886,
      "learning_rate": 4.591009579955785e-05,
      "loss": 1.1699,
      "num_input_tokens_seen": 10385904,
      "step": 2097,
      "train_runtime": 7255.3935,
      "train_tokens_per_second": 1431.474
    },
    {
      "epoch": 0.7716781609195402,
      "grad_norm": 0.41282057762145996,
      "learning_rate": 4.583640383198232e-05,
      "loss": 1.1454,
      "num_input_tokens_seen": 10390848,
      "step": 2098,
      "train_runtime": 7258.7237,
      "train_tokens_per_second": 1431.498
    },
    {
      "epoch": 0.7720459770114942,
      "grad_norm": 0.36254802346229553,
      "learning_rate": 4.5762711864406784e-05,
      "loss": 1.0989,
      "num_input_tokens_seen": 10396160,
      "step": 2099,
      "train_runtime": 7262.1879,
      "train_tokens_per_second": 1431.547
    },
    {
      "epoch": 0.7724137931034483,
      "grad_norm": 0.35990336537361145,
      "learning_rate": 4.568901989683125e-05,
      "loss": 1.2364,
      "num_input_tokens_seen": 10401408,
      "step": 2100,
      "train_runtime": 7265.6634,
      "train_tokens_per_second": 1431.584
    },
    {
      "epoch": 0.7727816091954023,
      "grad_norm": 0.3373374044895172,
      "learning_rate": 4.5615327929255713e-05,
      "loss": 1.0395,
      "num_input_tokens_seen": 10405232,
      "step": 2101,
      "train_runtime": 7270.6098,
      "train_tokens_per_second": 1431.136
    },
    {
      "epoch": 0.7731494252873563,
      "grad_norm": 0.4272858202457428,
      "learning_rate": 4.554163596168018e-05,
      "loss": 1.1808,
      "num_input_tokens_seen": 10409104,
      "step": 2102,
      "train_runtime": 7273.3011,
      "train_tokens_per_second": 1431.139
    },
    {
      "epoch": 0.7735172413793103,
      "grad_norm": 0.35010093450546265,
      "learning_rate": 4.546794399410464e-05,
      "loss": 1.0808,
      "num_input_tokens_seen": 10412544,
      "step": 2103,
      "train_runtime": 7275.7651,
      "train_tokens_per_second": 1431.127
    },
    {
      "epoch": 0.7738850574712643,
      "grad_norm": 0.3812684714794159,
      "learning_rate": 4.539425202652911e-05,
      "loss": 1.1303,
      "num_input_tokens_seen": 10416704,
      "step": 2104,
      "train_runtime": 7278.5331,
      "train_tokens_per_second": 1431.154
    },
    {
      "epoch": 0.7742528735632184,
      "grad_norm": 0.3618868291378021,
      "learning_rate": 4.532056005895358e-05,
      "loss": 1.224,
      "num_input_tokens_seen": 10421824,
      "step": 2105,
      "train_runtime": 7281.899,
      "train_tokens_per_second": 1431.196
    },
    {
      "epoch": 0.7746206896551724,
      "grad_norm": 0.32397451996803284,
      "learning_rate": 4.5246868091378044e-05,
      "loss": 1.2166,
      "num_input_tokens_seen": 10427488,
      "step": 2106,
      "train_runtime": 7285.6429,
      "train_tokens_per_second": 1431.238
    },
    {
      "epoch": 0.7749885057471264,
      "grad_norm": 0.3255363404750824,
      "learning_rate": 4.517317612380251e-05,
      "loss": 1.117,
      "num_input_tokens_seen": 10434816,
      "step": 2107,
      "train_runtime": 7290.2,
      "train_tokens_per_second": 1431.348
    },
    {
      "epoch": 0.7753563218390804,
      "grad_norm": 0.39174115657806396,
      "learning_rate": 4.5099484156226974e-05,
      "loss": 1.1382,
      "num_input_tokens_seen": 10439248,
      "step": 2108,
      "train_runtime": 7293.539,
      "train_tokens_per_second": 1431.301
    },
    {
      "epoch": 0.7757241379310345,
      "grad_norm": 0.3679260015487671,
      "learning_rate": 4.502579218865144e-05,
      "loss": 1.1387,
      "num_input_tokens_seen": 10442592,
      "step": 2109,
      "train_runtime": 7296.008,
      "train_tokens_per_second": 1431.275
    },
    {
      "epoch": 0.7760919540229885,
      "grad_norm": 0.3300345242023468,
      "learning_rate": 4.4952100221075904e-05,
      "loss": 1.0843,
      "num_input_tokens_seen": 10446832,
      "step": 2110,
      "train_runtime": 7298.9582,
      "train_tokens_per_second": 1431.277
    },
    {
      "epoch": 0.7764597701149425,
      "grad_norm": 0.36731183528900146,
      "learning_rate": 4.487840825350037e-05,
      "loss": 1.1997,
      "num_input_tokens_seen": 10452608,
      "step": 2111,
      "train_runtime": 7302.6544,
      "train_tokens_per_second": 1431.344
    },
    {
      "epoch": 0.7768275862068965,
      "grad_norm": 0.3751862645149231,
      "learning_rate": 4.480471628592483e-05,
      "loss": 1.234,
      "num_input_tokens_seen": 10458272,
      "step": 2112,
      "train_runtime": 7306.3281,
      "train_tokens_per_second": 1431.399
    },
    {
      "epoch": 0.7771954022988505,
      "grad_norm": 0.39682427048683167,
      "learning_rate": 4.4731024318349305e-05,
      "loss": 1.1767,
      "num_input_tokens_seen": 10464096,
      "step": 2113,
      "train_runtime": 7310.1416,
      "train_tokens_per_second": 1431.449
    },
    {
      "epoch": 0.7775632183908046,
      "grad_norm": 0.3733166754245758,
      "learning_rate": 4.465733235077377e-05,
      "loss": 1.1189,
      "num_input_tokens_seen": 10468480,
      "step": 2114,
      "train_runtime": 7313.1152,
      "train_tokens_per_second": 1431.467
    },
    {
      "epoch": 0.7779310344827586,
      "grad_norm": 0.3759706914424896,
      "learning_rate": 4.4583640383198234e-05,
      "loss": 1.2198,
      "num_input_tokens_seen": 10473504,
      "step": 2115,
      "train_runtime": 7316.4262,
      "train_tokens_per_second": 1431.505
    },
    {
      "epoch": 0.7782988505747126,
      "grad_norm": 0.3869493901729584,
      "learning_rate": 4.45099484156227e-05,
      "loss": 1.2462,
      "num_input_tokens_seen": 10478944,
      "step": 2116,
      "train_runtime": 7320.021,
      "train_tokens_per_second": 1431.546
    },
    {
      "epoch": 0.7786666666666666,
      "grad_norm": 0.392558217048645,
      "learning_rate": 4.4436256448047164e-05,
      "loss": 1.112,
      "num_input_tokens_seen": 10485296,
      "step": 2117,
      "train_runtime": 7324.0797,
      "train_tokens_per_second": 1431.62
    },
    {
      "epoch": 0.7790344827586206,
      "grad_norm": 0.41828736662864685,
      "learning_rate": 4.436256448047163e-05,
      "loss": 1.2317,
      "num_input_tokens_seen": 10488912,
      "step": 2118,
      "train_runtime": 7326.8515,
      "train_tokens_per_second": 1431.572
    },
    {
      "epoch": 0.7794022988505747,
      "grad_norm": 0.3480420410633087,
      "learning_rate": 4.4288872512896094e-05,
      "loss": 1.1233,
      "num_input_tokens_seen": 10492272,
      "step": 2119,
      "train_runtime": 7329.3035,
      "train_tokens_per_second": 1431.551
    },
    {
      "epoch": 0.7797701149425287,
      "grad_norm": 0.3556925356388092,
      "learning_rate": 4.421518054532056e-05,
      "loss": 1.2759,
      "num_input_tokens_seen": 10498576,
      "step": 2120,
      "train_runtime": 7333.33,
      "train_tokens_per_second": 1431.625
    },
    {
      "epoch": 0.7801379310344828,
      "grad_norm": 0.391644686460495,
      "learning_rate": 4.414148857774503e-05,
      "loss": 1.2512,
      "num_input_tokens_seen": 10504592,
      "step": 2121,
      "train_runtime": 7337.2486,
      "train_tokens_per_second": 1431.68
    },
    {
      "epoch": 0.7805057471264368,
      "grad_norm": 0.3730500042438507,
      "learning_rate": 4.4067796610169495e-05,
      "loss": 1.2504,
      "num_input_tokens_seen": 10514624,
      "step": 2122,
      "train_runtime": 7343.2973,
      "train_tokens_per_second": 1431.867
    },
    {
      "epoch": 0.7808735632183909,
      "grad_norm": 0.35692721605300903,
      "learning_rate": 4.399410464259396e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 10518320,
      "step": 2123,
      "train_runtime": 7345.9169,
      "train_tokens_per_second": 1431.859
    },
    {
      "epoch": 0.7812413793103449,
      "grad_norm": 0.3872193992137909,
      "learning_rate": 4.3920412675018425e-05,
      "loss": 1.1556,
      "num_input_tokens_seen": 10524128,
      "step": 2124,
      "train_runtime": 7349.7361,
      "train_tokens_per_second": 1431.906
    },
    {
      "epoch": 0.7816091954022989,
      "grad_norm": 0.34768128395080566,
      "learning_rate": 4.384672070744289e-05,
      "loss": 1.1912,
      "num_input_tokens_seen": 10527616,
      "step": 2125,
      "train_runtime": 7352.201,
      "train_tokens_per_second": 1431.9
    },
    {
      "epoch": 0.7819770114942529,
      "grad_norm": 0.35486099123954773,
      "learning_rate": 4.3773028739867354e-05,
      "loss": 1.1583,
      "num_input_tokens_seen": 10533632,
      "step": 2126,
      "train_runtime": 7356.021,
      "train_tokens_per_second": 1431.974
    },
    {
      "epoch": 0.7823448275862069,
      "grad_norm": 0.3728826940059662,
      "learning_rate": 4.369933677229182e-05,
      "loss": 1.3217,
      "num_input_tokens_seen": 10537456,
      "step": 2127,
      "train_runtime": 7358.6015,
      "train_tokens_per_second": 1431.992
    },
    {
      "epoch": 0.782712643678161,
      "grad_norm": 0.3817051947116852,
      "learning_rate": 4.3625644804716284e-05,
      "loss": 1.1696,
      "num_input_tokens_seen": 10544192,
      "step": 2128,
      "train_runtime": 7362.7906,
      "train_tokens_per_second": 1432.092
    },
    {
      "epoch": 0.783080459770115,
      "grad_norm": 0.39544183015823364,
      "learning_rate": 4.3551952837140756e-05,
      "loss": 1.2827,
      "num_input_tokens_seen": 10550016,
      "step": 2129,
      "train_runtime": 7366.56,
      "train_tokens_per_second": 1432.15
    },
    {
      "epoch": 0.783448275862069,
      "grad_norm": 0.42103514075279236,
      "learning_rate": 4.347826086956522e-05,
      "loss": 1.1435,
      "num_input_tokens_seen": 10554192,
      "step": 2130,
      "train_runtime": 7369.4241,
      "train_tokens_per_second": 1432.16
    },
    {
      "epoch": 0.783816091954023,
      "grad_norm": 0.3514348268508911,
      "learning_rate": 4.3404568901989685e-05,
      "loss": 1.3526,
      "num_input_tokens_seen": 10559488,
      "step": 2131,
      "train_runtime": 7375.0906,
      "train_tokens_per_second": 1431.777
    },
    {
      "epoch": 0.784183908045977,
      "grad_norm": 0.4109044075012207,
      "learning_rate": 4.333087693441415e-05,
      "loss": 1.1822,
      "num_input_tokens_seen": 10564192,
      "step": 2132,
      "train_runtime": 7378.1956,
      "train_tokens_per_second": 1431.812
    },
    {
      "epoch": 0.7845517241379311,
      "grad_norm": 0.33501356840133667,
      "learning_rate": 4.3257184966838615e-05,
      "loss": 1.2004,
      "num_input_tokens_seen": 10570400,
      "step": 2133,
      "train_runtime": 7382.2336,
      "train_tokens_per_second": 1431.87
    },
    {
      "epoch": 0.7849195402298851,
      "grad_norm": 0.3741925060749054,
      "learning_rate": 4.318349299926308e-05,
      "loss": 1.0734,
      "num_input_tokens_seen": 10575824,
      "step": 2134,
      "train_runtime": 7385.8048,
      "train_tokens_per_second": 1431.912
    },
    {
      "epoch": 0.7852873563218391,
      "grad_norm": 0.3550627827644348,
      "learning_rate": 4.3109801031687544e-05,
      "loss": 1.2629,
      "num_input_tokens_seen": 10582336,
      "step": 2135,
      "train_runtime": 7389.8617,
      "train_tokens_per_second": 1432.007
    },
    {
      "epoch": 0.7856551724137931,
      "grad_norm": 0.3435611426830292,
      "learning_rate": 4.303610906411201e-05,
      "loss": 1.0417,
      "num_input_tokens_seen": 10585904,
      "step": 2136,
      "train_runtime": 7392.3224,
      "train_tokens_per_second": 1432.013
    },
    {
      "epoch": 0.7860229885057471,
      "grad_norm": 0.37348830699920654,
      "learning_rate": 4.296241709653648e-05,
      "loss": 1.258,
      "num_input_tokens_seen": 10590464,
      "step": 2137,
      "train_runtime": 7395.412,
      "train_tokens_per_second": 1432.032
    },
    {
      "epoch": 0.7863908045977012,
      "grad_norm": 0.341295063495636,
      "learning_rate": 4.2888725128960946e-05,
      "loss": 1.1567,
      "num_input_tokens_seen": 10594848,
      "step": 2138,
      "train_runtime": 7398.3796,
      "train_tokens_per_second": 1432.05
    },
    {
      "epoch": 0.7867586206896552,
      "grad_norm": 0.34383878111839294,
      "learning_rate": 4.281503316138541e-05,
      "loss": 1.1565,
      "num_input_tokens_seen": 10598880,
      "step": 2139,
      "train_runtime": 7401.1578,
      "train_tokens_per_second": 1432.057
    },
    {
      "epoch": 0.7871264367816092,
      "grad_norm": 0.3610958755016327,
      "learning_rate": 4.2741341193809875e-05,
      "loss": 1.2002,
      "num_input_tokens_seen": 10603104,
      "step": 2140,
      "train_runtime": 7404.0138,
      "train_tokens_per_second": 1432.075
    },
    {
      "epoch": 0.7874942528735632,
      "grad_norm": 0.3590060770511627,
      "learning_rate": 4.266764922623434e-05,
      "loss": 1.3145,
      "num_input_tokens_seen": 10610400,
      "step": 2141,
      "train_runtime": 7408.4911,
      "train_tokens_per_second": 1432.194
    },
    {
      "epoch": 0.7878620689655172,
      "grad_norm": 0.36746180057525635,
      "learning_rate": 4.2593957258658805e-05,
      "loss": 1.1656,
      "num_input_tokens_seen": 10614160,
      "step": 2142,
      "train_runtime": 7411.183,
      "train_tokens_per_second": 1432.182
    },
    {
      "epoch": 0.7882298850574713,
      "grad_norm": 0.4026757776737213,
      "learning_rate": 4.252026529108327e-05,
      "loss": 1.3014,
      "num_input_tokens_seen": 10618192,
      "step": 2143,
      "train_runtime": 7413.9877,
      "train_tokens_per_second": 1432.184
    },
    {
      "epoch": 0.7885977011494253,
      "grad_norm": 0.37475791573524475,
      "learning_rate": 4.2446573323507735e-05,
      "loss": 1.4171,
      "num_input_tokens_seen": 10622368,
      "step": 2144,
      "train_runtime": 7416.9176,
      "train_tokens_per_second": 1432.181
    },
    {
      "epoch": 0.7889655172413793,
      "grad_norm": 0.35653188824653625,
      "learning_rate": 4.2372881355932206e-05,
      "loss": 1.1512,
      "num_input_tokens_seen": 10629968,
      "step": 2145,
      "train_runtime": 7421.668,
      "train_tokens_per_second": 1432.288
    },
    {
      "epoch": 0.7893333333333333,
      "grad_norm": 0.3764266073703766,
      "learning_rate": 4.229918938835667e-05,
      "loss": 1.1602,
      "num_input_tokens_seen": 10636928,
      "step": 2146,
      "train_runtime": 7426.0222,
      "train_tokens_per_second": 1432.386
    },
    {
      "epoch": 0.7897011494252874,
      "grad_norm": 0.3702530860900879,
      "learning_rate": 4.2225497420781136e-05,
      "loss": 1.212,
      "num_input_tokens_seen": 10644272,
      "step": 2147,
      "train_runtime": 7430.5594,
      "train_tokens_per_second": 1432.499
    },
    {
      "epoch": 0.7900689655172414,
      "grad_norm": 0.3757065236568451,
      "learning_rate": 4.21518054532056e-05,
      "loss": 1.2507,
      "num_input_tokens_seen": 10649552,
      "step": 2148,
      "train_runtime": 7434.0824,
      "train_tokens_per_second": 1432.531
    },
    {
      "epoch": 0.7904367816091954,
      "grad_norm": 0.3695482611656189,
      "learning_rate": 4.2078113485630066e-05,
      "loss": 1.0818,
      "num_input_tokens_seen": 10653616,
      "step": 2149,
      "train_runtime": 7436.8915,
      "train_tokens_per_second": 1432.536
    },
    {
      "epoch": 0.7908045977011494,
      "grad_norm": 0.39494508504867554,
      "learning_rate": 4.200442151805453e-05,
      "loss": 1.3179,
      "num_input_tokens_seen": 10660784,
      "step": 2150,
      "train_runtime": 7441.3266,
      "train_tokens_per_second": 1432.646
    },
    {
      "epoch": 0.7911724137931034,
      "grad_norm": 0.3682766556739807,
      "learning_rate": 4.1930729550478995e-05,
      "loss": 1.1862,
      "num_input_tokens_seen": 10665456,
      "step": 2151,
      "train_runtime": 7444.4,
      "train_tokens_per_second": 1432.682
    },
    {
      "epoch": 0.7915402298850575,
      "grad_norm": 0.38218170404434204,
      "learning_rate": 4.185703758290346e-05,
      "loss": 1.3696,
      "num_input_tokens_seen": 10670624,
      "step": 2152,
      "train_runtime": 7447.7892,
      "train_tokens_per_second": 1432.724
    },
    {
      "epoch": 0.7919080459770115,
      "grad_norm": 0.34641164541244507,
      "learning_rate": 4.178334561532793e-05,
      "loss": 1.2038,
      "num_input_tokens_seen": 10677104,
      "step": 2153,
      "train_runtime": 7451.8744,
      "train_tokens_per_second": 1432.808
    },
    {
      "epoch": 0.7922758620689655,
      "grad_norm": 0.382000207901001,
      "learning_rate": 4.1709653647752396e-05,
      "loss": 1.1334,
      "num_input_tokens_seen": 10687360,
      "step": 2154,
      "train_runtime": 7458.0642,
      "train_tokens_per_second": 1432.994
    },
    {
      "epoch": 0.7926436781609195,
      "grad_norm": 0.35297074913978577,
      "learning_rate": 4.163596168017686e-05,
      "loss": 1.0983,
      "num_input_tokens_seen": 10691584,
      "step": 2155,
      "train_runtime": 7460.9209,
      "train_tokens_per_second": 1433.011
    },
    {
      "epoch": 0.7930114942528735,
      "grad_norm": 0.33983173966407776,
      "learning_rate": 4.1562269712601326e-05,
      "loss": 1.2581,
      "num_input_tokens_seen": 10697632,
      "step": 2156,
      "train_runtime": 7464.7815,
      "train_tokens_per_second": 1433.08
    },
    {
      "epoch": 0.7933793103448276,
      "grad_norm": 0.38186582922935486,
      "learning_rate": 4.148857774502579e-05,
      "loss": 1.2129,
      "num_input_tokens_seen": 10701152,
      "step": 2157,
      "train_runtime": 7467.3031,
      "train_tokens_per_second": 1433.068
    },
    {
      "epoch": 0.7937471264367816,
      "grad_norm": 0.36720943450927734,
      "learning_rate": 4.1414885777450256e-05,
      "loss": 1.3464,
      "num_input_tokens_seen": 10704976,
      "step": 2158,
      "train_runtime": 7469.9774,
      "train_tokens_per_second": 1433.067
    },
    {
      "epoch": 0.7941149425287356,
      "grad_norm": 0.35176119208335876,
      "learning_rate": 4.134119380987472e-05,
      "loss": 1.1796,
      "num_input_tokens_seen": 10711024,
      "step": 2159,
      "train_runtime": 7473.8773,
      "train_tokens_per_second": 1433.128
    },
    {
      "epoch": 0.7944827586206896,
      "grad_norm": 0.36766719818115234,
      "learning_rate": 4.126750184229919e-05,
      "loss": 1.2194,
      "num_input_tokens_seen": 10715472,
      "step": 2160,
      "train_runtime": 7476.8314,
      "train_tokens_per_second": 1433.157
    },
    {
      "epoch": 0.7948505747126436,
      "grad_norm": 0.3604617714881897,
      "learning_rate": 4.119380987472366e-05,
      "loss": 1.297,
      "num_input_tokens_seen": 10719584,
      "step": 2161,
      "train_runtime": 7482.2381,
      "train_tokens_per_second": 1432.671
    },
    {
      "epoch": 0.7952183908045977,
      "grad_norm": 0.35564473271369934,
      "learning_rate": 4.112011790714812e-05,
      "loss": 1.2508,
      "num_input_tokens_seen": 10723424,
      "step": 2162,
      "train_runtime": 7484.9512,
      "train_tokens_per_second": 1432.665
    },
    {
      "epoch": 0.7955862068965517,
      "grad_norm": 0.358351469039917,
      "learning_rate": 4.1046425939572587e-05,
      "loss": 1.1678,
      "num_input_tokens_seen": 10727888,
      "step": 2163,
      "train_runtime": 7488.0283,
      "train_tokens_per_second": 1432.672
    },
    {
      "epoch": 0.7959540229885057,
      "grad_norm": 0.3639545440673828,
      "learning_rate": 4.097273397199705e-05,
      "loss": 1.1054,
      "num_input_tokens_seen": 10732912,
      "step": 2164,
      "train_runtime": 7491.3757,
      "train_tokens_per_second": 1432.702
    },
    {
      "epoch": 0.7963218390804597,
      "grad_norm": 0.38622066378593445,
      "learning_rate": 4.089904200442152e-05,
      "loss": 1.2808,
      "num_input_tokens_seen": 10738496,
      "step": 2165,
      "train_runtime": 7495.0496,
      "train_tokens_per_second": 1432.745
    },
    {
      "epoch": 0.7966896551724137,
      "grad_norm": 0.35459399223327637,
      "learning_rate": 4.082535003684599e-05,
      "loss": 1.1949,
      "num_input_tokens_seen": 10742624,
      "step": 2166,
      "train_runtime": 7497.8397,
      "train_tokens_per_second": 1432.763
    },
    {
      "epoch": 0.7970574712643678,
      "grad_norm": 0.3667961657047272,
      "learning_rate": 4.075165806927045e-05,
      "loss": 1.2046,
      "num_input_tokens_seen": 10746592,
      "step": 2167,
      "train_runtime": 7500.6226,
      "train_tokens_per_second": 1432.76
    },
    {
      "epoch": 0.7974252873563218,
      "grad_norm": 0.36509400606155396,
      "learning_rate": 4.067796610169492e-05,
      "loss": 1.2754,
      "num_input_tokens_seen": 10754144,
      "step": 2168,
      "train_runtime": 7505.255,
      "train_tokens_per_second": 1432.882
    },
    {
      "epoch": 0.7977931034482758,
      "grad_norm": 0.35090118646621704,
      "learning_rate": 4.060427413411938e-05,
      "loss": 1.2592,
      "num_input_tokens_seen": 10765440,
      "step": 2169,
      "train_runtime": 7512.0198,
      "train_tokens_per_second": 1433.095
    },
    {
      "epoch": 0.7981609195402298,
      "grad_norm": 0.31430619955062866,
      "learning_rate": 4.0530582166543854e-05,
      "loss": 1.1928,
      "num_input_tokens_seen": 10776496,
      "step": 2170,
      "train_runtime": 7518.6907,
      "train_tokens_per_second": 1433.294
    },
    {
      "epoch": 0.798528735632184,
      "grad_norm": 0.3578290343284607,
      "learning_rate": 4.045689019896832e-05,
      "loss": 1.1974,
      "num_input_tokens_seen": 10780768,
      "step": 2171,
      "train_runtime": 7521.6037,
      "train_tokens_per_second": 1433.307
    },
    {
      "epoch": 0.798896551724138,
      "grad_norm": 0.35479453206062317,
      "learning_rate": 4.0383198231392783e-05,
      "loss": 1.1515,
      "num_input_tokens_seen": 10786080,
      "step": 2172,
      "train_runtime": 7525.1122,
      "train_tokens_per_second": 1433.345
    },
    {
      "epoch": 0.799264367816092,
      "grad_norm": 0.3450087010860443,
      "learning_rate": 4.030950626381725e-05,
      "loss": 1.246,
      "num_input_tokens_seen": 10789728,
      "step": 2173,
      "train_runtime": 7527.6927,
      "train_tokens_per_second": 1433.338
    },
    {
      "epoch": 0.799632183908046,
      "grad_norm": 0.36223742365837097,
      "learning_rate": 4.023581429624171e-05,
      "loss": 1.3354,
      "num_input_tokens_seen": 10794064,
      "step": 2174,
      "train_runtime": 7530.6743,
      "train_tokens_per_second": 1433.346
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3862453103065491,
      "learning_rate": 4.016212232866618e-05,
      "loss": 1.2895,
      "num_input_tokens_seen": 10804720,
      "step": 2175,
      "train_runtime": 7537.0892,
      "train_tokens_per_second": 1433.54
    },
    {
      "epoch": 0.8003678160919541,
      "grad_norm": 0.3645070493221283,
      "learning_rate": 4.008843036109064e-05,
      "loss": 1.1983,
      "num_input_tokens_seen": 10811552,
      "step": 2176,
      "train_runtime": 7541.3356,
      "train_tokens_per_second": 1433.639
    },
    {
      "epoch": 0.8003678160919541,
      "eval_loss": 1.2086011171340942,
      "eval_runtime": 15.5238,
      "eval_samples_per_second": 64.417,
      "eval_steps_per_second": 4.058,
      "num_input_tokens_seen": 10811552,
      "step": 2176
    },
    {
      "epoch": 0.8007356321839081,
      "grad_norm": 0.39066165685653687,
      "learning_rate": 4.0014738393515114e-05,
      "loss": 1.2683,
      "num_input_tokens_seen": 10815472,
      "step": 2177,
      "train_runtime": 7559.6274,
      "train_tokens_per_second": 1430.689
    },
    {
      "epoch": 0.8011034482758621,
      "grad_norm": 0.3674662709236145,
      "learning_rate": 3.994104642593958e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 10820000,
      "step": 2178,
      "train_runtime": 7562.7024,
      "train_tokens_per_second": 1430.706
    },
    {
      "epoch": 0.8014712643678161,
      "grad_norm": 0.39019522070884705,
      "learning_rate": 3.9867354458364044e-05,
      "loss": 1.2636,
      "num_input_tokens_seen": 10824704,
      "step": 2179,
      "train_runtime": 7565.9112,
      "train_tokens_per_second": 1430.72
    },
    {
      "epoch": 0.8018390804597701,
      "grad_norm": 0.36135849356651306,
      "learning_rate": 3.979366249078851e-05,
      "loss": 1.1094,
      "num_input_tokens_seen": 10829824,
      "step": 2180,
      "train_runtime": 7569.2761,
      "train_tokens_per_second": 1430.761
    },
    {
      "epoch": 0.8022068965517242,
      "grad_norm": 0.33432087302207947,
      "learning_rate": 3.9719970523212974e-05,
      "loss": 1.2246,
      "num_input_tokens_seen": 10833600,
      "step": 2181,
      "train_runtime": 7571.856,
      "train_tokens_per_second": 1430.772
    },
    {
      "epoch": 0.8025747126436782,
      "grad_norm": 0.34472835063934326,
      "learning_rate": 3.964627855563744e-05,
      "loss": 1.1524,
      "num_input_tokens_seen": 10840256,
      "step": 2182,
      "train_runtime": 7576.1201,
      "train_tokens_per_second": 1430.845
    },
    {
      "epoch": 0.8029425287356322,
      "grad_norm": 0.36489275097846985,
      "learning_rate": 3.95725865880619e-05,
      "loss": 1.2153,
      "num_input_tokens_seen": 10843888,
      "step": 2183,
      "train_runtime": 7578.7159,
      "train_tokens_per_second": 1430.834
    },
    {
      "epoch": 0.8033103448275862,
      "grad_norm": 0.3496350944042206,
      "learning_rate": 3.949889462048637e-05,
      "loss": 1.1331,
      "num_input_tokens_seen": 10847504,
      "step": 2184,
      "train_runtime": 7581.4713,
      "train_tokens_per_second": 1430.791
    },
    {
      "epoch": 0.8036781609195403,
      "grad_norm": 0.34375157952308655,
      "learning_rate": 3.942520265291084e-05,
      "loss": 1.0461,
      "num_input_tokens_seen": 10851472,
      "step": 2185,
      "train_runtime": 7584.2024,
      "train_tokens_per_second": 1430.799
    },
    {
      "epoch": 0.8040459770114943,
      "grad_norm": 0.3992886245250702,
      "learning_rate": 3.9351510685335305e-05,
      "loss": 1.3437,
      "num_input_tokens_seen": 10856912,
      "step": 2186,
      "train_runtime": 7587.7458,
      "train_tokens_per_second": 1430.848
    },
    {
      "epoch": 0.8044137931034483,
      "grad_norm": 0.3416062295436859,
      "learning_rate": 3.927781871775977e-05,
      "loss": 0.9705,
      "num_input_tokens_seen": 10860896,
      "step": 2187,
      "train_runtime": 7590.5901,
      "train_tokens_per_second": 1430.837
    },
    {
      "epoch": 0.8047816091954023,
      "grad_norm": 0.4134158492088318,
      "learning_rate": 3.9204126750184234e-05,
      "loss": 1.3652,
      "num_input_tokens_seen": 10866416,
      "step": 2188,
      "train_runtime": 7594.2799,
      "train_tokens_per_second": 1430.869
    },
    {
      "epoch": 0.8051494252873563,
      "grad_norm": 0.3477571904659271,
      "learning_rate": 3.91304347826087e-05,
      "loss": 1.1635,
      "num_input_tokens_seen": 10872944,
      "step": 2189,
      "train_runtime": 7598.469,
      "train_tokens_per_second": 1430.939
    },
    {
      "epoch": 0.8055172413793104,
      "grad_norm": 0.3680364787578583,
      "learning_rate": 3.9056742815033164e-05,
      "loss": 1.1893,
      "num_input_tokens_seen": 10880304,
      "step": 2190,
      "train_runtime": 7603.0756,
      "train_tokens_per_second": 1431.04
    },
    {
      "epoch": 0.8058850574712644,
      "grad_norm": 0.33584916591644287,
      "learning_rate": 3.898305084745763e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 10883856,
      "step": 2191,
      "train_runtime": 7607.8385,
      "train_tokens_per_second": 1430.611
    },
    {
      "epoch": 0.8062528735632184,
      "grad_norm": 0.3721490800380707,
      "learning_rate": 3.8909358879882093e-05,
      "loss": 1.1083,
      "num_input_tokens_seen": 10888176,
      "step": 2192,
      "train_runtime": 7610.7579,
      "train_tokens_per_second": 1430.63
    },
    {
      "epoch": 0.8066206896551724,
      "grad_norm": 0.3533916771411896,
      "learning_rate": 3.8835666912306565e-05,
      "loss": 1.278,
      "num_input_tokens_seen": 10892560,
      "step": 2193,
      "train_runtime": 7613.6766,
      "train_tokens_per_second": 1430.657
    },
    {
      "epoch": 0.8069885057471264,
      "grad_norm": 0.38169029355049133,
      "learning_rate": 3.876197494473103e-05,
      "loss": 1.2225,
      "num_input_tokens_seen": 10896208,
      "step": 2194,
      "train_runtime": 7616.339,
      "train_tokens_per_second": 1430.636
    },
    {
      "epoch": 0.8073563218390805,
      "grad_norm": 0.3399798572063446,
      "learning_rate": 3.8688282977155495e-05,
      "loss": 1.1528,
      "num_input_tokens_seen": 10900240,
      "step": 2195,
      "train_runtime": 7619.1576,
      "train_tokens_per_second": 1430.636
    },
    {
      "epoch": 0.8077241379310345,
      "grad_norm": 0.374094694852829,
      "learning_rate": 3.861459100957996e-05,
      "loss": 1.2981,
      "num_input_tokens_seen": 10905696,
      "step": 2196,
      "train_runtime": 7622.7302,
      "train_tokens_per_second": 1430.681
    },
    {
      "epoch": 0.8080919540229885,
      "grad_norm": 0.3852567672729492,
      "learning_rate": 3.8540899042004424e-05,
      "loss": 1.3074,
      "num_input_tokens_seen": 10909904,
      "step": 2197,
      "train_runtime": 7625.612,
      "train_tokens_per_second": 1430.692
    },
    {
      "epoch": 0.8084597701149425,
      "grad_norm": 0.3825359642505646,
      "learning_rate": 3.846720707442889e-05,
      "loss": 1.2137,
      "num_input_tokens_seen": 10913824,
      "step": 2198,
      "train_runtime": 7628.3177,
      "train_tokens_per_second": 1430.699
    },
    {
      "epoch": 0.8088275862068965,
      "grad_norm": 0.37729591131210327,
      "learning_rate": 3.8393515106853354e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 10919216,
      "step": 2199,
      "train_runtime": 7631.8241,
      "train_tokens_per_second": 1430.748
    },
    {
      "epoch": 0.8091954022988506,
      "grad_norm": 0.35168570280075073,
      "learning_rate": 3.831982313927782e-05,
      "loss": 1.1989,
      "num_input_tokens_seen": 10923136,
      "step": 2200,
      "train_runtime": 7634.5589,
      "train_tokens_per_second": 1430.749
    },
    {
      "epoch": 0.8095632183908046,
      "grad_norm": 0.373208224773407,
      "learning_rate": 3.824613117170229e-05,
      "loss": 1.1844,
      "num_input_tokens_seen": 10928224,
      "step": 2201,
      "train_runtime": 7637.8986,
      "train_tokens_per_second": 1430.789
    },
    {
      "epoch": 0.8099310344827586,
      "grad_norm": 0.3567676842212677,
      "learning_rate": 3.8172439204126755e-05,
      "loss": 1.1947,
      "num_input_tokens_seen": 10932864,
      "step": 2202,
      "train_runtime": 7641.0302,
      "train_tokens_per_second": 1430.81
    },
    {
      "epoch": 0.8102988505747126,
      "grad_norm": 0.4092943072319031,
      "learning_rate": 3.809874723655122e-05,
      "loss": 1.1706,
      "num_input_tokens_seen": 10939552,
      "step": 2203,
      "train_runtime": 7645.3388,
      "train_tokens_per_second": 1430.879
    },
    {
      "epoch": 0.8106666666666666,
      "grad_norm": 0.35322707891464233,
      "learning_rate": 3.8025055268975685e-05,
      "loss": 1.2156,
      "num_input_tokens_seen": 10944144,
      "step": 2204,
      "train_runtime": 7648.3843,
      "train_tokens_per_second": 1430.909
    },
    {
      "epoch": 0.8110344827586207,
      "grad_norm": 0.39123642444610596,
      "learning_rate": 3.795136330140015e-05,
      "loss": 1.136,
      "num_input_tokens_seen": 10949328,
      "step": 2205,
      "train_runtime": 7651.8119,
      "train_tokens_per_second": 1430.946
    },
    {
      "epoch": 0.8114022988505747,
      "grad_norm": 0.3603310286998749,
      "learning_rate": 3.7877671333824615e-05,
      "loss": 1.1049,
      "num_input_tokens_seen": 10955296,
      "step": 2206,
      "train_runtime": 7655.525,
      "train_tokens_per_second": 1431.031
    },
    {
      "epoch": 0.8117701149425287,
      "grad_norm": 0.39141908288002014,
      "learning_rate": 3.780397936624908e-05,
      "loss": 1.176,
      "num_input_tokens_seen": 10958592,
      "step": 2207,
      "train_runtime": 7657.9013,
      "train_tokens_per_second": 1431.018
    },
    {
      "epoch": 0.8121379310344827,
      "grad_norm": 0.3875255882740021,
      "learning_rate": 3.7730287398673544e-05,
      "loss": 1.2343,
      "num_input_tokens_seen": 10964544,
      "step": 2208,
      "train_runtime": 7661.8026,
      "train_tokens_per_second": 1431.066
    },
    {
      "epoch": 0.8125057471264368,
      "grad_norm": 0.3603978455066681,
      "learning_rate": 3.7656595431098016e-05,
      "loss": 1.1432,
      "num_input_tokens_seen": 10968384,
      "step": 2209,
      "train_runtime": 7664.5354,
      "train_tokens_per_second": 1431.057
    },
    {
      "epoch": 0.8128735632183908,
      "grad_norm": 0.3609590530395508,
      "learning_rate": 3.758290346352248e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 10973136,
      "step": 2210,
      "train_runtime": 7667.6589,
      "train_tokens_per_second": 1431.093
    },
    {
      "epoch": 0.8132413793103448,
      "grad_norm": 0.4301469326019287,
      "learning_rate": 3.7509211495946945e-05,
      "loss": 1.1845,
      "num_input_tokens_seen": 10978128,
      "step": 2211,
      "train_runtime": 7670.981,
      "train_tokens_per_second": 1431.124
    },
    {
      "epoch": 0.8136091954022988,
      "grad_norm": 0.36393284797668457,
      "learning_rate": 3.743551952837141e-05,
      "loss": 1.2357,
      "num_input_tokens_seen": 10982400,
      "step": 2212,
      "train_runtime": 7674.0159,
      "train_tokens_per_second": 1431.115
    },
    {
      "epoch": 0.8139770114942528,
      "grad_norm": 0.3666031062602997,
      "learning_rate": 3.7361827560795875e-05,
      "loss": 1.0967,
      "num_input_tokens_seen": 10986512,
      "step": 2213,
      "train_runtime": 7676.8909,
      "train_tokens_per_second": 1431.115
    },
    {
      "epoch": 0.8143448275862069,
      "grad_norm": 0.340594083070755,
      "learning_rate": 3.728813559322034e-05,
      "loss": 1.195,
      "num_input_tokens_seen": 10991888,
      "step": 2214,
      "train_runtime": 7680.4017,
      "train_tokens_per_second": 1431.161
    },
    {
      "epoch": 0.8147126436781609,
      "grad_norm": 0.38579240441322327,
      "learning_rate": 3.7214443625644805e-05,
      "loss": 1.2717,
      "num_input_tokens_seen": 10995872,
      "step": 2215,
      "train_runtime": 7683.1653,
      "train_tokens_per_second": 1431.164
    },
    {
      "epoch": 0.8150804597701149,
      "grad_norm": 0.374298095703125,
      "learning_rate": 3.714075165806927e-05,
      "loss": 1.072,
      "num_input_tokens_seen": 11002304,
      "step": 2216,
      "train_runtime": 7687.325,
      "train_tokens_per_second": 1431.227
    },
    {
      "epoch": 0.8154482758620689,
      "grad_norm": 0.36412107944488525,
      "learning_rate": 3.706705969049374e-05,
      "loss": 1.1551,
      "num_input_tokens_seen": 11006096,
      "step": 2217,
      "train_runtime": 7690.0185,
      "train_tokens_per_second": 1431.218
    },
    {
      "epoch": 0.8158160919540229,
      "grad_norm": 0.3489569425582886,
      "learning_rate": 3.6993367722918206e-05,
      "loss": 1.2231,
      "num_input_tokens_seen": 11013152,
      "step": 2218,
      "train_runtime": 7694.5239,
      "train_tokens_per_second": 1431.297
    },
    {
      "epoch": 0.816183908045977,
      "grad_norm": 0.3949083387851715,
      "learning_rate": 3.691967575534267e-05,
      "loss": 1.2647,
      "num_input_tokens_seen": 11017792,
      "step": 2219,
      "train_runtime": 7697.6911,
      "train_tokens_per_second": 1431.311
    },
    {
      "epoch": 0.8165517241379311,
      "grad_norm": 0.3573471009731293,
      "learning_rate": 3.6845983787767136e-05,
      "loss": 1.2449,
      "num_input_tokens_seen": 11021088,
      "step": 2220,
      "train_runtime": 7700.0701,
      "train_tokens_per_second": 1431.297
    },
    {
      "epoch": 0.8169195402298851,
      "grad_norm": 0.367899090051651,
      "learning_rate": 3.67722918201916e-05,
      "loss": 1.2773,
      "num_input_tokens_seen": 11025216,
      "step": 2221,
      "train_runtime": 7705.1169,
      "train_tokens_per_second": 1430.895
    },
    {
      "epoch": 0.8172873563218391,
      "grad_norm": 0.43307703733444214,
      "learning_rate": 3.6698599852616065e-05,
      "loss": 1.3283,
      "num_input_tokens_seen": 11029488,
      "step": 2222,
      "train_runtime": 7708.0306,
      "train_tokens_per_second": 1430.909
    },
    {
      "epoch": 0.8176551724137932,
      "grad_norm": 0.381436824798584,
      "learning_rate": 3.662490788504053e-05,
      "loss": 1.1985,
      "num_input_tokens_seen": 11032928,
      "step": 2223,
      "train_runtime": 7710.4838,
      "train_tokens_per_second": 1430.9
    },
    {
      "epoch": 0.8180229885057472,
      "grad_norm": 0.360688179731369,
      "learning_rate": 3.6551215917464995e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 11037488,
      "step": 2224,
      "train_runtime": 7713.5048,
      "train_tokens_per_second": 1430.93
    },
    {
      "epoch": 0.8183908045977012,
      "grad_norm": 0.3755568563938141,
      "learning_rate": 3.6477523949889466e-05,
      "loss": 1.106,
      "num_input_tokens_seen": 11042016,
      "step": 2225,
      "train_runtime": 7716.6039,
      "train_tokens_per_second": 1430.942
    },
    {
      "epoch": 0.8187586206896552,
      "grad_norm": 0.34972530603408813,
      "learning_rate": 3.640383198231393e-05,
      "loss": 0.9777,
      "num_input_tokens_seen": 11046384,
      "step": 2226,
      "train_runtime": 7719.5414,
      "train_tokens_per_second": 1430.964
    },
    {
      "epoch": 0.8191264367816092,
      "grad_norm": 0.4256415069103241,
      "learning_rate": 3.6330140014738396e-05,
      "loss": 1.3821,
      "num_input_tokens_seen": 11049760,
      "step": 2227,
      "train_runtime": 7721.9841,
      "train_tokens_per_second": 1430.948
    },
    {
      "epoch": 0.8194942528735633,
      "grad_norm": 0.3539048135280609,
      "learning_rate": 3.625644804716286e-05,
      "loss": 1.219,
      "num_input_tokens_seen": 11053232,
      "step": 2228,
      "train_runtime": 7724.4618,
      "train_tokens_per_second": 1430.939
    },
    {
      "epoch": 0.8198620689655173,
      "grad_norm": 0.34744372963905334,
      "learning_rate": 3.6182756079587326e-05,
      "loss": 1.186,
      "num_input_tokens_seen": 11057280,
      "step": 2229,
      "train_runtime": 7727.1989,
      "train_tokens_per_second": 1430.956
    },
    {
      "epoch": 0.8202298850574713,
      "grad_norm": 0.3572700023651123,
      "learning_rate": 3.610906411201179e-05,
      "loss": 1.131,
      "num_input_tokens_seen": 11062912,
      "step": 2230,
      "train_runtime": 7730.8321,
      "train_tokens_per_second": 1431.012
    },
    {
      "epoch": 0.8205977011494253,
      "grad_norm": 0.34894219040870667,
      "learning_rate": 3.6035372144436255e-05,
      "loss": 1.3993,
      "num_input_tokens_seen": 11071264,
      "step": 2231,
      "train_runtime": 7736.0791,
      "train_tokens_per_second": 1431.121
    },
    {
      "epoch": 0.8209655172413793,
      "grad_norm": 0.36651894450187683,
      "learning_rate": 3.596168017686073e-05,
      "loss": 1.0528,
      "num_input_tokens_seen": 11074352,
      "step": 2232,
      "train_runtime": 7738.383,
      "train_tokens_per_second": 1431.094
    },
    {
      "epoch": 0.8213333333333334,
      "grad_norm": 0.37314262986183167,
      "learning_rate": 3.588798820928519e-05,
      "loss": 1.1481,
      "num_input_tokens_seen": 11080928,
      "step": 2233,
      "train_runtime": 7742.5022,
      "train_tokens_per_second": 1431.182
    },
    {
      "epoch": 0.8217011494252874,
      "grad_norm": 0.3408365845680237,
      "learning_rate": 3.5814296241709657e-05,
      "loss": 1.1784,
      "num_input_tokens_seen": 11085616,
      "step": 2234,
      "train_runtime": 7745.757,
      "train_tokens_per_second": 1431.186
    },
    {
      "epoch": 0.8220689655172414,
      "grad_norm": 0.409574031829834,
      "learning_rate": 3.574060427413412e-05,
      "loss": 1.1684,
      "num_input_tokens_seen": 11089136,
      "step": 2235,
      "train_runtime": 7748.2761,
      "train_tokens_per_second": 1431.175
    },
    {
      "epoch": 0.8224367816091954,
      "grad_norm": 0.382395476102829,
      "learning_rate": 3.5666912306558586e-05,
      "loss": 1.1465,
      "num_input_tokens_seen": 11097488,
      "step": 2236,
      "train_runtime": 7753.5567,
      "train_tokens_per_second": 1431.277
    },
    {
      "epoch": 0.8228045977011494,
      "grad_norm": 0.35066017508506775,
      "learning_rate": 3.559322033898305e-05,
      "loss": 1.2212,
      "num_input_tokens_seen": 11101600,
      "step": 2237,
      "train_runtime": 7756.4119,
      "train_tokens_per_second": 1431.28
    },
    {
      "epoch": 0.8231724137931035,
      "grad_norm": 0.3712022006511688,
      "learning_rate": 3.5519528371407516e-05,
      "loss": 1.2685,
      "num_input_tokens_seen": 11105312,
      "step": 2238,
      "train_runtime": 7759.0206,
      "train_tokens_per_second": 1431.278
    },
    {
      "epoch": 0.8235402298850575,
      "grad_norm": 0.3650802969932556,
      "learning_rate": 3.544583640383198e-05,
      "loss": 1.0869,
      "num_input_tokens_seen": 11108944,
      "step": 2239,
      "train_runtime": 7761.6182,
      "train_tokens_per_second": 1431.266
    },
    {
      "epoch": 0.8239080459770115,
      "grad_norm": 0.34153664112091064,
      "learning_rate": 3.537214443625645e-05,
      "loss": 1.2786,
      "num_input_tokens_seen": 11115968,
      "step": 2240,
      "train_runtime": 7766.0717,
      "train_tokens_per_second": 1431.35
    },
    {
      "epoch": 0.8242758620689655,
      "grad_norm": 0.3589034676551819,
      "learning_rate": 3.529845246868092e-05,
      "loss": 1.1396,
      "num_input_tokens_seen": 11120800,
      "step": 2241,
      "train_runtime": 7769.3388,
      "train_tokens_per_second": 1431.37
    },
    {
      "epoch": 0.8246436781609195,
      "grad_norm": 0.3497578501701355,
      "learning_rate": 3.522476050110538e-05,
      "loss": 1.2812,
      "num_input_tokens_seen": 11124496,
      "step": 2242,
      "train_runtime": 7771.8913,
      "train_tokens_per_second": 1431.376
    },
    {
      "epoch": 0.8250114942528736,
      "grad_norm": 0.3567488193511963,
      "learning_rate": 3.515106853352985e-05,
      "loss": 1.1238,
      "num_input_tokens_seen": 11128224,
      "step": 2243,
      "train_runtime": 7774.5255,
      "train_tokens_per_second": 1431.37
    },
    {
      "epoch": 0.8253793103448276,
      "grad_norm": 0.37901973724365234,
      "learning_rate": 3.507737656595431e-05,
      "loss": 1.2452,
      "num_input_tokens_seen": 11131872,
      "step": 2244,
      "train_runtime": 7777.1698,
      "train_tokens_per_second": 1431.353
    },
    {
      "epoch": 0.8257471264367816,
      "grad_norm": 0.3546604514122009,
      "learning_rate": 3.5003684598378776e-05,
      "loss": 1.2136,
      "num_input_tokens_seen": 11135600,
      "step": 2245,
      "train_runtime": 7779.832,
      "train_tokens_per_second": 1431.342
    },
    {
      "epoch": 0.8261149425287356,
      "grad_norm": 0.39348286390304565,
      "learning_rate": 3.492999263080324e-05,
      "loss": 1.1029,
      "num_input_tokens_seen": 11139184,
      "step": 2246,
      "train_runtime": 7782.354,
      "train_tokens_per_second": 1431.339
    },
    {
      "epoch": 0.8264827586206897,
      "grad_norm": 0.34764671325683594,
      "learning_rate": 3.4856300663227706e-05,
      "loss": 1.0123,
      "num_input_tokens_seen": 11143712,
      "step": 2247,
      "train_runtime": 7785.4539,
      "train_tokens_per_second": 1431.35
    },
    {
      "epoch": 0.8268505747126437,
      "grad_norm": 0.35627540946006775,
      "learning_rate": 3.478260869565218e-05,
      "loss": 1.157,
      "num_input_tokens_seen": 11149728,
      "step": 2248,
      "train_runtime": 7789.3734,
      "train_tokens_per_second": 1431.402
    },
    {
      "epoch": 0.8272183908045977,
      "grad_norm": 0.35804876685142517,
      "learning_rate": 3.470891672807664e-05,
      "loss": 1.1953,
      "num_input_tokens_seen": 11153520,
      "step": 2249,
      "train_runtime": 7792.0955,
      "train_tokens_per_second": 1431.389
    },
    {
      "epoch": 0.8275862068965517,
      "grad_norm": 0.3648607134819031,
      "learning_rate": 3.463522476050111e-05,
      "loss": 1.2814,
      "num_input_tokens_seen": 11160592,
      "step": 2250,
      "train_runtime": 7796.6243,
      "train_tokens_per_second": 1431.465
    },
    {
      "epoch": 0.8279540229885057,
      "grad_norm": 0.4124414920806885,
      "learning_rate": 3.456153279292557e-05,
      "loss": 1.2525,
      "num_input_tokens_seen": 11164992,
      "step": 2251,
      "train_runtime": 7801.6859,
      "train_tokens_per_second": 1431.1
    },
    {
      "epoch": 0.8283218390804598,
      "grad_norm": 0.4313027560710907,
      "learning_rate": 3.448784082535004e-05,
      "loss": 1.2966,
      "num_input_tokens_seen": 11169216,
      "step": 2252,
      "train_runtime": 7804.5487,
      "train_tokens_per_second": 1431.116
    },
    {
      "epoch": 0.8286896551724138,
      "grad_norm": 0.34469956159591675,
      "learning_rate": 3.44141488577745e-05,
      "loss": 1.1824,
      "num_input_tokens_seen": 11174400,
      "step": 2253,
      "train_runtime": 7808.0642,
      "train_tokens_per_second": 1431.136
    },
    {
      "epoch": 0.8290574712643678,
      "grad_norm": 0.38931646943092346,
      "learning_rate": 3.4340456890198967e-05,
      "loss": 1.2728,
      "num_input_tokens_seen": 11178592,
      "step": 2254,
      "train_runtime": 7810.959,
      "train_tokens_per_second": 1431.142
    },
    {
      "epoch": 0.8294252873563218,
      "grad_norm": 0.3734469711780548,
      "learning_rate": 3.426676492262343e-05,
      "loss": 1.2284,
      "num_input_tokens_seen": 11183664,
      "step": 2255,
      "train_runtime": 7814.3296,
      "train_tokens_per_second": 1431.174
    },
    {
      "epoch": 0.8297931034482758,
      "grad_norm": 0.34597718715667725,
      "learning_rate": 3.41930729550479e-05,
      "loss": 1.2758,
      "num_input_tokens_seen": 11188928,
      "step": 2256,
      "train_runtime": 7817.7848,
      "train_tokens_per_second": 1431.215
    },
    {
      "epoch": 0.8301609195402299,
      "grad_norm": 0.35527363419532776,
      "learning_rate": 3.411938098747237e-05,
      "loss": 1.1719,
      "num_input_tokens_seen": 11194368,
      "step": 2257,
      "train_runtime": 7821.3777,
      "train_tokens_per_second": 1431.253
    },
    {
      "epoch": 0.8305287356321839,
      "grad_norm": 0.3635798990726471,
      "learning_rate": 3.404568901989683e-05,
      "loss": 1.0338,
      "num_input_tokens_seen": 11201376,
      "step": 2258,
      "train_runtime": 7825.8569,
      "train_tokens_per_second": 1431.329
    },
    {
      "epoch": 0.8308965517241379,
      "grad_norm": 0.36341825127601624,
      "learning_rate": 3.39719970523213e-05,
      "loss": 1.1644,
      "num_input_tokens_seen": 11206656,
      "step": 2259,
      "train_runtime": 7829.3427,
      "train_tokens_per_second": 1431.366
    },
    {
      "epoch": 0.8312643678160919,
      "grad_norm": 0.4638970196247101,
      "learning_rate": 3.389830508474576e-05,
      "loss": 1.3294,
      "num_input_tokens_seen": 11210160,
      "step": 2260,
      "train_runtime": 7831.8261,
      "train_tokens_per_second": 1431.36
    },
    {
      "epoch": 0.8316321839080459,
      "grad_norm": 0.38175779581069946,
      "learning_rate": 3.382461311717023e-05,
      "loss": 1.2148,
      "num_input_tokens_seen": 11213232,
      "step": 2261,
      "train_runtime": 7834.1561,
      "train_tokens_per_second": 1431.326
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.3692260682582855,
      "learning_rate": 3.375092114959469e-05,
      "loss": 1.215,
      "num_input_tokens_seen": 11218016,
      "step": 2262,
      "train_runtime": 7837.3362,
      "train_tokens_per_second": 1431.356
    },
    {
      "epoch": 0.832367816091954,
      "grad_norm": 0.36855754256248474,
      "learning_rate": 3.367722918201916e-05,
      "loss": 1.1863,
      "num_input_tokens_seen": 11222256,
      "step": 2263,
      "train_runtime": 7840.3362,
      "train_tokens_per_second": 1431.349
    },
    {
      "epoch": 0.832735632183908,
      "grad_norm": 0.3529607653617859,
      "learning_rate": 3.360353721444363e-05,
      "loss": 1.2546,
      "num_input_tokens_seen": 11228112,
      "step": 2264,
      "train_runtime": 7844.23,
      "train_tokens_per_second": 1431.385
    },
    {
      "epoch": 0.833103448275862,
      "grad_norm": 0.33194971084594727,
      "learning_rate": 3.352984524686809e-05,
      "loss": 1.1762,
      "num_input_tokens_seen": 11235600,
      "step": 2265,
      "train_runtime": 7848.9221,
      "train_tokens_per_second": 1431.483
    },
    {
      "epoch": 0.833471264367816,
      "grad_norm": 0.3698999285697937,
      "learning_rate": 3.345615327929256e-05,
      "loss": 1.324,
      "num_input_tokens_seen": 11241920,
      "step": 2266,
      "train_runtime": 7853.0252,
      "train_tokens_per_second": 1431.54
    },
    {
      "epoch": 0.8338390804597701,
      "grad_norm": 0.3352484703063965,
      "learning_rate": 3.338246131171702e-05,
      "loss": 1.1774,
      "num_input_tokens_seen": 11248640,
      "step": 2267,
      "train_runtime": 7857.293,
      "train_tokens_per_second": 1431.618
    },
    {
      "epoch": 0.8342068965517241,
      "grad_norm": 0.3569459617137909,
      "learning_rate": 3.330876934414149e-05,
      "loss": 1.1983,
      "num_input_tokens_seen": 11253344,
      "step": 2268,
      "train_runtime": 7860.4294,
      "train_tokens_per_second": 1431.645
    },
    {
      "epoch": 0.8345747126436781,
      "grad_norm": 0.3377792239189148,
      "learning_rate": 3.323507737656595e-05,
      "loss": 1.1995,
      "num_input_tokens_seen": 11257616,
      "step": 2269,
      "train_runtime": 7863.4488,
      "train_tokens_per_second": 1431.638
    },
    {
      "epoch": 0.8349425287356322,
      "grad_norm": 0.3700810372829437,
      "learning_rate": 3.316138540899042e-05,
      "loss": 1.1721,
      "num_input_tokens_seen": 11260928,
      "step": 2270,
      "train_runtime": 7865.9019,
      "train_tokens_per_second": 1431.613
    },
    {
      "epoch": 0.8353103448275863,
      "grad_norm": 0.37969207763671875,
      "learning_rate": 3.308769344141488e-05,
      "loss": 1.2362,
      "num_input_tokens_seen": 11265936,
      "step": 2271,
      "train_runtime": 7869.2105,
      "train_tokens_per_second": 1431.648
    },
    {
      "epoch": 0.8356781609195403,
      "grad_norm": 0.3042827546596527,
      "learning_rate": 3.3014001473839354e-05,
      "loss": 1.0416,
      "num_input_tokens_seen": 11271264,
      "step": 2272,
      "train_runtime": 7872.7361,
      "train_tokens_per_second": 1431.683
    },
    {
      "epoch": 0.8360459770114943,
      "grad_norm": 0.3895162343978882,
      "learning_rate": 3.294030950626382e-05,
      "loss": 1.2003,
      "num_input_tokens_seen": 11277920,
      "step": 2273,
      "train_runtime": 7876.9019,
      "train_tokens_per_second": 1431.771
    },
    {
      "epoch": 0.8364137931034483,
      "grad_norm": 0.34355947375297546,
      "learning_rate": 3.286661753868828e-05,
      "loss": 1.1156,
      "num_input_tokens_seen": 11281072,
      "step": 2274,
      "train_runtime": 7879.2639,
      "train_tokens_per_second": 1431.742
    },
    {
      "epoch": 0.8367816091954023,
      "grad_norm": 0.3605348765850067,
      "learning_rate": 3.279292557111275e-05,
      "loss": 1.2835,
      "num_input_tokens_seen": 11287072,
      "step": 2275,
      "train_runtime": 7883.2458,
      "train_tokens_per_second": 1431.78
    },
    {
      "epoch": 0.8371494252873564,
      "grad_norm": 0.37450411915779114,
      "learning_rate": 3.271923360353721e-05,
      "loss": 1.1573,
      "num_input_tokens_seen": 11290656,
      "step": 2276,
      "train_runtime": 7885.8722,
      "train_tokens_per_second": 1431.757
    },
    {
      "epoch": 0.8375172413793104,
      "grad_norm": 0.39006364345550537,
      "learning_rate": 3.264554163596168e-05,
      "loss": 1.1191,
      "num_input_tokens_seen": 11295024,
      "step": 2277,
      "train_runtime": 7888.8491,
      "train_tokens_per_second": 1431.771
    },
    {
      "epoch": 0.8378850574712644,
      "grad_norm": 0.39977237582206726,
      "learning_rate": 3.257184966838614e-05,
      "loss": 1.1304,
      "num_input_tokens_seen": 11298704,
      "step": 2278,
      "train_runtime": 7891.4723,
      "train_tokens_per_second": 1431.761
    },
    {
      "epoch": 0.8382528735632184,
      "grad_norm": 0.3386874794960022,
      "learning_rate": 3.2498157700810614e-05,
      "loss": 1.0305,
      "num_input_tokens_seen": 11303984,
      "step": 2279,
      "train_runtime": 7894.975,
      "train_tokens_per_second": 1431.795
    },
    {
      "epoch": 0.8386206896551724,
      "grad_norm": 0.37369704246520996,
      "learning_rate": 3.242446573323508e-05,
      "loss": 1.2185,
      "num_input_tokens_seen": 11308736,
      "step": 2280,
      "train_runtime": 7898.1684,
      "train_tokens_per_second": 1431.818
    },
    {
      "epoch": 0.8389885057471265,
      "grad_norm": 0.3337636888027191,
      "learning_rate": 3.2350773765659544e-05,
      "loss": 1.1213,
      "num_input_tokens_seen": 11314416,
      "step": 2281,
      "train_runtime": 7903.9686,
      "train_tokens_per_second": 1431.485
    },
    {
      "epoch": 0.8393563218390805,
      "grad_norm": 0.40354880690574646,
      "learning_rate": 3.227708179808401e-05,
      "loss": 1.3279,
      "num_input_tokens_seen": 11318368,
      "step": 2282,
      "train_runtime": 7906.7065,
      "train_tokens_per_second": 1431.49
    },
    {
      "epoch": 0.8397241379310345,
      "grad_norm": 0.37759336829185486,
      "learning_rate": 3.2203389830508473e-05,
      "loss": 1.2324,
      "num_input_tokens_seen": 11324336,
      "step": 2283,
      "train_runtime": 7910.5392,
      "train_tokens_per_second": 1431.55
    },
    {
      "epoch": 0.8400919540229885,
      "grad_norm": 0.3506891131401062,
      "learning_rate": 3.212969786293294e-05,
      "loss": 1.2108,
      "num_input_tokens_seen": 11328160,
      "step": 2284,
      "train_runtime": 7913.2181,
      "train_tokens_per_second": 1431.549
    },
    {
      "epoch": 0.8404597701149426,
      "grad_norm": 0.38749098777770996,
      "learning_rate": 3.20560058953574e-05,
      "loss": 1.2173,
      "num_input_tokens_seen": 11333504,
      "step": 2285,
      "train_runtime": 7916.7276,
      "train_tokens_per_second": 1431.589
    },
    {
      "epoch": 0.8408275862068966,
      "grad_norm": 0.3610440492630005,
      "learning_rate": 3.198231392778187e-05,
      "loss": 1.194,
      "num_input_tokens_seen": 11337648,
      "step": 2286,
      "train_runtime": 7919.6302,
      "train_tokens_per_second": 1431.588
    },
    {
      "epoch": 0.8411954022988506,
      "grad_norm": 0.34255412220954895,
      "learning_rate": 3.190862196020634e-05,
      "loss": 1.0885,
      "num_input_tokens_seen": 11343856,
      "step": 2287,
      "train_runtime": 7923.6604,
      "train_tokens_per_second": 1431.643
    },
    {
      "epoch": 0.8415632183908046,
      "grad_norm": 0.3525124192237854,
      "learning_rate": 3.1834929992630804e-05,
      "loss": 1.1454,
      "num_input_tokens_seen": 11348512,
      "step": 2288,
      "train_runtime": 7926.8055,
      "train_tokens_per_second": 1431.663
    },
    {
      "epoch": 0.8419310344827586,
      "grad_norm": 0.3775132894515991,
      "learning_rate": 3.1761238025055276e-05,
      "loss": 1.3019,
      "num_input_tokens_seen": 11353040,
      "step": 2289,
      "train_runtime": 7929.8157,
      "train_tokens_per_second": 1431.69
    },
    {
      "epoch": 0.8422988505747127,
      "grad_norm": 0.3541698157787323,
      "learning_rate": 3.168754605747974e-05,
      "loss": 1.0868,
      "num_input_tokens_seen": 11357936,
      "step": 2290,
      "train_runtime": 7933.1536,
      "train_tokens_per_second": 1431.705
    },
    {
      "epoch": 0.8426666666666667,
      "grad_norm": 0.3667600154876709,
      "learning_rate": 3.1613854089904206e-05,
      "loss": 1.1032,
      "num_input_tokens_seen": 11361568,
      "step": 2291,
      "train_runtime": 7935.7209,
      "train_tokens_per_second": 1431.7
    },
    {
      "epoch": 0.8430344827586207,
      "grad_norm": 0.39227744936943054,
      "learning_rate": 3.154016212232867e-05,
      "loss": 1.1824,
      "num_input_tokens_seen": 11365344,
      "step": 2292,
      "train_runtime": 7938.3863,
      "train_tokens_per_second": 1431.694
    },
    {
      "epoch": 0.8434022988505747,
      "grad_norm": 0.34459492564201355,
      "learning_rate": 3.1466470154753135e-05,
      "loss": 1.2114,
      "num_input_tokens_seen": 11371376,
      "step": 2293,
      "train_runtime": 7942.2651,
      "train_tokens_per_second": 1431.755
    },
    {
      "epoch": 0.8437701149425287,
      "grad_norm": 0.3658595383167267,
      "learning_rate": 3.13927781871776e-05,
      "loss": 1.2453,
      "num_input_tokens_seen": 11375232,
      "step": 2294,
      "train_runtime": 7944.9421,
      "train_tokens_per_second": 1431.758
    },
    {
      "epoch": 0.8441379310344828,
      "grad_norm": 0.3694017827510834,
      "learning_rate": 3.1319086219602065e-05,
      "loss": 1.2455,
      "num_input_tokens_seen": 11381216,
      "step": 2295,
      "train_runtime": 7948.7883,
      "train_tokens_per_second": 1431.818
    },
    {
      "epoch": 0.8445057471264368,
      "grad_norm": 0.37624162435531616,
      "learning_rate": 3.124539425202653e-05,
      "loss": 1.1637,
      "num_input_tokens_seen": 11384464,
      "step": 2296,
      "train_runtime": 7951.1329,
      "train_tokens_per_second": 1431.804
    },
    {
      "epoch": 0.8448735632183908,
      "grad_norm": 0.36825355887413025,
      "learning_rate": 3.1171702284451e-05,
      "loss": 1.2203,
      "num_input_tokens_seen": 11389600,
      "step": 2297,
      "train_runtime": 7954.5341,
      "train_tokens_per_second": 1431.837
    },
    {
      "epoch": 0.8452413793103448,
      "grad_norm": 0.496214896440506,
      "learning_rate": 3.1098010316875466e-05,
      "loss": 1.228,
      "num_input_tokens_seen": 11393712,
      "step": 2298,
      "train_runtime": 7957.4135,
      "train_tokens_per_second": 1431.836
    },
    {
      "epoch": 0.8456091954022988,
      "grad_norm": 0.3443017899990082,
      "learning_rate": 3.102431834929993e-05,
      "loss": 1.1756,
      "num_input_tokens_seen": 11397776,
      "step": 2299,
      "train_runtime": 7960.2278,
      "train_tokens_per_second": 1431.84
    },
    {
      "epoch": 0.8459770114942529,
      "grad_norm": 0.35658949613571167,
      "learning_rate": 3.0950626381724396e-05,
      "loss": 1.1155,
      "num_input_tokens_seen": 11402944,
      "step": 2300,
      "train_runtime": 7963.6341,
      "train_tokens_per_second": 1431.877
    },
    {
      "epoch": 0.8463448275862069,
      "grad_norm": 0.3757800757884979,
      "learning_rate": 3.087693441414886e-05,
      "loss": 1.2837,
      "num_input_tokens_seen": 11407648,
      "step": 2301,
      "train_runtime": 7966.7998,
      "train_tokens_per_second": 1431.898
    },
    {
      "epoch": 0.8467126436781609,
      "grad_norm": 0.41095036268234253,
      "learning_rate": 3.0803242446573325e-05,
      "loss": 1.2237,
      "num_input_tokens_seen": 11414336,
      "step": 2302,
      "train_runtime": 7971.0219,
      "train_tokens_per_second": 1431.979
    },
    {
      "epoch": 0.8470804597701149,
      "grad_norm": 0.40145936608314514,
      "learning_rate": 3.072955047899779e-05,
      "loss": 1.2939,
      "num_input_tokens_seen": 11417760,
      "step": 2303,
      "train_runtime": 7973.4476,
      "train_tokens_per_second": 1431.973
    },
    {
      "epoch": 0.847448275862069,
      "grad_norm": 0.3589552342891693,
      "learning_rate": 3.065585851142226e-05,
      "loss": 1.2338,
      "num_input_tokens_seen": 11425040,
      "step": 2304,
      "train_runtime": 7977.9205,
      "train_tokens_per_second": 1432.082
    },
    {
      "epoch": 0.847816091954023,
      "grad_norm": 0.36281466484069824,
      "learning_rate": 3.058216654384673e-05,
      "loss": 1.197,
      "num_input_tokens_seen": 11429312,
      "step": 2305,
      "train_runtime": 7980.7975,
      "train_tokens_per_second": 1432.101
    },
    {
      "epoch": 0.848183908045977,
      "grad_norm": 0.34846553206443787,
      "learning_rate": 3.050847457627119e-05,
      "loss": 1.2261,
      "num_input_tokens_seen": 11434960,
      "step": 2306,
      "train_runtime": 7984.5259,
      "train_tokens_per_second": 1432.14
    },
    {
      "epoch": 0.848551724137931,
      "grad_norm": 0.35105523467063904,
      "learning_rate": 3.0434782608695656e-05,
      "loss": 1.0301,
      "num_input_tokens_seen": 11442864,
      "step": 2307,
      "train_runtime": 7989.3839,
      "train_tokens_per_second": 1432.259
    },
    {
      "epoch": 0.848919540229885,
      "grad_norm": 0.43849313259124756,
      "learning_rate": 3.036109064112012e-05,
      "loss": 1.3086,
      "num_input_tokens_seen": 11449632,
      "step": 2308,
      "train_runtime": 7993.6769,
      "train_tokens_per_second": 1432.336
    },
    {
      "epoch": 0.849287356321839,
      "grad_norm": 0.3420996367931366,
      "learning_rate": 3.0287398673544586e-05,
      "loss": 1.2129,
      "num_input_tokens_seen": 11453632,
      "step": 2309,
      "train_runtime": 7996.3677,
      "train_tokens_per_second": 1432.354
    },
    {
      "epoch": 0.8496551724137931,
      "grad_norm": 0.3746173083782196,
      "learning_rate": 3.0213706705969054e-05,
      "loss": 1.252,
      "num_input_tokens_seen": 11458880,
      "step": 2310,
      "train_runtime": 7999.7986,
      "train_tokens_per_second": 1432.396
    },
    {
      "epoch": 0.8500229885057471,
      "grad_norm": 0.33470419049263,
      "learning_rate": 3.014001473839352e-05,
      "loss": 1.0921,
      "num_input_tokens_seen": 11462624,
      "step": 2311,
      "train_runtime": 8004.5234,
      "train_tokens_per_second": 1432.018
    },
    {
      "epoch": 0.8503908045977011,
      "grad_norm": 0.3931296467781067,
      "learning_rate": 3.0066322770817984e-05,
      "loss": 1.2004,
      "num_input_tokens_seen": 11469040,
      "step": 2312,
      "train_runtime": 8008.5529,
      "train_tokens_per_second": 1432.099
    },
    {
      "epoch": 0.8507586206896551,
      "grad_norm": 0.3638696074485779,
      "learning_rate": 2.999263080324245e-05,
      "loss": 1.1885,
      "num_input_tokens_seen": 11473184,
      "step": 2313,
      "train_runtime": 8011.4053,
      "train_tokens_per_second": 1432.106
    },
    {
      "epoch": 0.8511264367816092,
      "grad_norm": 0.37879887223243713,
      "learning_rate": 2.9918938835666917e-05,
      "loss": 1.3213,
      "num_input_tokens_seen": 11476880,
      "step": 2314,
      "train_runtime": 8014.0154,
      "train_tokens_per_second": 1432.101
    },
    {
      "epoch": 0.8514942528735632,
      "grad_norm": 0.3558363616466522,
      "learning_rate": 2.984524686809138e-05,
      "loss": 1.1794,
      "num_input_tokens_seen": 11482480,
      "step": 2315,
      "train_runtime": 8017.6018,
      "train_tokens_per_second": 1432.159
    },
    {
      "epoch": 0.8518620689655172,
      "grad_norm": 0.3671194612979889,
      "learning_rate": 2.9771554900515846e-05,
      "loss": 1.2351,
      "num_input_tokens_seen": 11487008,
      "step": 2316,
      "train_runtime": 8020.6021,
      "train_tokens_per_second": 1432.188
    },
    {
      "epoch": 0.8522298850574712,
      "grad_norm": 0.41130730509757996,
      "learning_rate": 2.969786293294031e-05,
      "loss": 1.3448,
      "num_input_tokens_seen": 11491648,
      "step": 2317,
      "train_runtime": 8023.6487,
      "train_tokens_per_second": 1432.222
    },
    {
      "epoch": 0.8525977011494252,
      "grad_norm": 0.3718991279602051,
      "learning_rate": 2.962417096536478e-05,
      "loss": 1.1617,
      "num_input_tokens_seen": 11496720,
      "step": 2318,
      "train_runtime": 8027.0472,
      "train_tokens_per_second": 1432.248
    },
    {
      "epoch": 0.8529655172413793,
      "grad_norm": 0.38872313499450684,
      "learning_rate": 2.9550478997789244e-05,
      "loss": 1.247,
      "num_input_tokens_seen": 11500704,
      "step": 2319,
      "train_runtime": 8029.7984,
      "train_tokens_per_second": 1432.253
    },
    {
      "epoch": 0.8533333333333334,
      "grad_norm": 0.3886869549751282,
      "learning_rate": 2.947678703021371e-05,
      "loss": 1.3846,
      "num_input_tokens_seen": 11504224,
      "step": 2320,
      "train_runtime": 8032.321,
      "train_tokens_per_second": 1432.242
    },
    {
      "epoch": 0.8537011494252874,
      "grad_norm": 0.37152817845344543,
      "learning_rate": 2.9403095062638174e-05,
      "loss": 1.3175,
      "num_input_tokens_seen": 11507456,
      "step": 2321,
      "train_runtime": 8034.6608,
      "train_tokens_per_second": 1432.227
    },
    {
      "epoch": 0.8540689655172414,
      "grad_norm": 0.3631071448326111,
      "learning_rate": 2.9329403095062642e-05,
      "loss": 1.2335,
      "num_input_tokens_seen": 11511184,
      "step": 2322,
      "train_runtime": 8037.2787,
      "train_tokens_per_second": 1432.224
    },
    {
      "epoch": 0.8544367816091955,
      "grad_norm": 0.37639376521110535,
      "learning_rate": 2.9255711127487107e-05,
      "loss": 1.1944,
      "num_input_tokens_seen": 11516992,
      "step": 2323,
      "train_runtime": 8041.1031,
      "train_tokens_per_second": 1432.265
    },
    {
      "epoch": 0.8548045977011495,
      "grad_norm": 0.37705475091934204,
      "learning_rate": 2.9182019159911572e-05,
      "loss": 1.2569,
      "num_input_tokens_seen": 11522464,
      "step": 2324,
      "train_runtime": 8044.6137,
      "train_tokens_per_second": 1432.32
    },
    {
      "epoch": 0.8551724137931035,
      "grad_norm": 0.392872154712677,
      "learning_rate": 2.9108327192336037e-05,
      "loss": 1.2066,
      "num_input_tokens_seen": 11527312,
      "step": 2325,
      "train_runtime": 8047.8517,
      "train_tokens_per_second": 1432.346
    },
    {
      "epoch": 0.8555402298850575,
      "grad_norm": 0.3412322700023651,
      "learning_rate": 2.9034635224760505e-05,
      "loss": 1.1425,
      "num_input_tokens_seen": 11532544,
      "step": 2326,
      "train_runtime": 8051.2496,
      "train_tokens_per_second": 1432.392
    },
    {
      "epoch": 0.8559080459770115,
      "grad_norm": 0.3914344310760498,
      "learning_rate": 2.896094325718497e-05,
      "loss": 1.18,
      "num_input_tokens_seen": 11538032,
      "step": 2327,
      "train_runtime": 8054.7802,
      "train_tokens_per_second": 1432.445
    },
    {
      "epoch": 0.8562758620689656,
      "grad_norm": 0.3546917140483856,
      "learning_rate": 2.8887251289609434e-05,
      "loss": 1.2681,
      "num_input_tokens_seen": 11543168,
      "step": 2328,
      "train_runtime": 8058.2268,
      "train_tokens_per_second": 1432.47
    },
    {
      "epoch": 0.8566436781609196,
      "grad_norm": 0.41026362776756287,
      "learning_rate": 2.88135593220339e-05,
      "loss": 1.4283,
      "num_input_tokens_seen": 11547056,
      "step": 2329,
      "train_runtime": 8060.8879,
      "train_tokens_per_second": 1432.479
    },
    {
      "epoch": 0.8570114942528736,
      "grad_norm": 0.3765309154987335,
      "learning_rate": 2.8739867354458367e-05,
      "loss": 1.0912,
      "num_input_tokens_seen": 11551456,
      "step": 2330,
      "train_runtime": 8063.8757,
      "train_tokens_per_second": 1432.494
    },
    {
      "epoch": 0.8573793103448276,
      "grad_norm": 0.4597894549369812,
      "learning_rate": 2.8666175386882832e-05,
      "loss": 1.3273,
      "num_input_tokens_seen": 11559200,
      "step": 2331,
      "train_runtime": 8068.5559,
      "train_tokens_per_second": 1432.623
    },
    {
      "epoch": 0.8577471264367816,
      "grad_norm": 0.3727620542049408,
      "learning_rate": 2.8592483419307297e-05,
      "loss": 1.1774,
      "num_input_tokens_seen": 11563056,
      "step": 2332,
      "train_runtime": 8071.2183,
      "train_tokens_per_second": 1432.628
    },
    {
      "epoch": 0.8581149425287357,
      "grad_norm": 0.3988867402076721,
      "learning_rate": 2.8518791451731762e-05,
      "loss": 1.1822,
      "num_input_tokens_seen": 11566320,
      "step": 2333,
      "train_runtime": 8073.5278,
      "train_tokens_per_second": 1432.623
    },
    {
      "epoch": 0.8584827586206897,
      "grad_norm": 0.3431839942932129,
      "learning_rate": 2.844509948415623e-05,
      "loss": 1.2776,
      "num_input_tokens_seen": 11570448,
      "step": 2334,
      "train_runtime": 8076.3766,
      "train_tokens_per_second": 1432.629
    },
    {
      "epoch": 0.8588505747126437,
      "grad_norm": 0.3561246693134308,
      "learning_rate": 2.8371407516580695e-05,
      "loss": 1.1009,
      "num_input_tokens_seen": 11574304,
      "step": 2335,
      "train_runtime": 8079.0243,
      "train_tokens_per_second": 1432.636
    },
    {
      "epoch": 0.8592183908045977,
      "grad_norm": 0.36138033866882324,
      "learning_rate": 2.829771554900516e-05,
      "loss": 1.2186,
      "num_input_tokens_seen": 11577936,
      "step": 2336,
      "train_runtime": 8081.5805,
      "train_tokens_per_second": 1432.633
    },
    {
      "epoch": 0.8595862068965517,
      "grad_norm": 0.382767915725708,
      "learning_rate": 2.8224023581429625e-05,
      "loss": 1.2227,
      "num_input_tokens_seen": 11581488,
      "step": 2337,
      "train_runtime": 8084.0905,
      "train_tokens_per_second": 1432.627
    },
    {
      "epoch": 0.8599540229885058,
      "grad_norm": 0.37188485264778137,
      "learning_rate": 2.8150331613854093e-05,
      "loss": 1.2467,
      "num_input_tokens_seen": 11590336,
      "step": 2338,
      "train_runtime": 8089.3785,
      "train_tokens_per_second": 1432.784
    },
    {
      "epoch": 0.8603218390804598,
      "grad_norm": 0.36899954080581665,
      "learning_rate": 2.8076639646278558e-05,
      "loss": 1.1596,
      "num_input_tokens_seen": 11593856,
      "step": 2339,
      "train_runtime": 8091.9096,
      "train_tokens_per_second": 1432.771
    },
    {
      "epoch": 0.8606896551724138,
      "grad_norm": 0.3485456705093384,
      "learning_rate": 2.8002947678703022e-05,
      "loss": 1.1614,
      "num_input_tokens_seen": 11597680,
      "step": 2340,
      "train_runtime": 8094.5559,
      "train_tokens_per_second": 1432.775
    },
    {
      "epoch": 0.8610574712643678,
      "grad_norm": 0.3754168152809143,
      "learning_rate": 2.7929255711127487e-05,
      "loss": 1.1861,
      "num_input_tokens_seen": 11604432,
      "step": 2341,
      "train_runtime": 8100.8765,
      "train_tokens_per_second": 1432.491
    },
    {
      "epoch": 0.8614252873563218,
      "grad_norm": 0.3656279146671295,
      "learning_rate": 2.7855563743551956e-05,
      "loss": 1.1541,
      "num_input_tokens_seen": 11609104,
      "step": 2342,
      "train_runtime": 8104.0379,
      "train_tokens_per_second": 1432.509
    },
    {
      "epoch": 0.8617931034482759,
      "grad_norm": 0.36324989795684814,
      "learning_rate": 2.778187177597642e-05,
      "loss": 1.2718,
      "num_input_tokens_seen": 11615248,
      "step": 2343,
      "train_runtime": 8107.9474,
      "train_tokens_per_second": 1432.576
    },
    {
      "epoch": 0.8621609195402299,
      "grad_norm": 0.39646971225738525,
      "learning_rate": 2.7708179808400885e-05,
      "loss": 1.1867,
      "num_input_tokens_seen": 11621872,
      "step": 2344,
      "train_runtime": 8112.1238,
      "train_tokens_per_second": 1432.655
    },
    {
      "epoch": 0.8625287356321839,
      "grad_norm": 0.3751087486743927,
      "learning_rate": 2.763448784082535e-05,
      "loss": 1.2859,
      "num_input_tokens_seen": 11626000,
      "step": 2345,
      "train_runtime": 8114.8918,
      "train_tokens_per_second": 1432.675
    },
    {
      "epoch": 0.8628965517241379,
      "grad_norm": 0.3649898171424866,
      "learning_rate": 2.7560795873249818e-05,
      "loss": 1.1061,
      "num_input_tokens_seen": 11631792,
      "step": 2346,
      "train_runtime": 8118.5796,
      "train_tokens_per_second": 1432.737
    },
    {
      "epoch": 0.863264367816092,
      "grad_norm": 0.3596176505088806,
      "learning_rate": 2.7487103905674283e-05,
      "loss": 1.1859,
      "num_input_tokens_seen": 11637808,
      "step": 2347,
      "train_runtime": 8122.4444,
      "train_tokens_per_second": 1432.796
    },
    {
      "epoch": 0.863632183908046,
      "grad_norm": 0.4087933301925659,
      "learning_rate": 2.7413411938098748e-05,
      "loss": 1.2644,
      "num_input_tokens_seen": 11641392,
      "step": 2348,
      "train_runtime": 8124.9819,
      "train_tokens_per_second": 1432.79
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.3594960570335388,
      "learning_rate": 2.7339719970523213e-05,
      "loss": 1.2586,
      "num_input_tokens_seen": 11647424,
      "step": 2349,
      "train_runtime": 8128.9176,
      "train_tokens_per_second": 1432.838
    },
    {
      "epoch": 0.864367816091954,
      "grad_norm": 0.37164241075515747,
      "learning_rate": 2.726602800294768e-05,
      "loss": 1.245,
      "num_input_tokens_seen": 11653200,
      "step": 2350,
      "train_runtime": 8132.6316,
      "train_tokens_per_second": 1432.894
    },
    {
      "epoch": 0.864735632183908,
      "grad_norm": 0.3846262991428375,
      "learning_rate": 2.7192336035372146e-05,
      "loss": 1.0308,
      "num_input_tokens_seen": 11657504,
      "step": 2351,
      "train_runtime": 8135.5037,
      "train_tokens_per_second": 1432.917
    },
    {
      "epoch": 0.865103448275862,
      "grad_norm": 0.4109100103378296,
      "learning_rate": 2.711864406779661e-05,
      "loss": 1.1317,
      "num_input_tokens_seen": 11661808,
      "step": 2352,
      "train_runtime": 8138.4569,
      "train_tokens_per_second": 1432.926
    },
    {
      "epoch": 0.8654712643678161,
      "grad_norm": 0.38728392124176025,
      "learning_rate": 2.704495210022108e-05,
      "loss": 1.304,
      "num_input_tokens_seen": 11666240,
      "step": 2353,
      "train_runtime": 8141.482,
      "train_tokens_per_second": 1432.938
    },
    {
      "epoch": 0.8658390804597701,
      "grad_norm": 0.38073766231536865,
      "learning_rate": 2.6971260132645544e-05,
      "loss": 1.0767,
      "num_input_tokens_seen": 11669392,
      "step": 2354,
      "train_runtime": 8143.8651,
      "train_tokens_per_second": 1432.906
    },
    {
      "epoch": 0.8662068965517241,
      "grad_norm": 0.3477320671081543,
      "learning_rate": 2.689756816507001e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 11673856,
      "step": 2355,
      "train_runtime": 8146.8508,
      "train_tokens_per_second": 1432.929
    },
    {
      "epoch": 0.8665747126436781,
      "grad_norm": 0.3762071430683136,
      "learning_rate": 2.6823876197494473e-05,
      "loss": 1.1214,
      "num_input_tokens_seen": 11679568,
      "step": 2356,
      "train_runtime": 8150.4795,
      "train_tokens_per_second": 1432.992
    },
    {
      "epoch": 0.8669425287356322,
      "grad_norm": 0.36011943221092224,
      "learning_rate": 2.675018422991894e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 11683952,
      "step": 2357,
      "train_runtime": 8153.3689,
      "train_tokens_per_second": 1433.021
    },
    {
      "epoch": 0.8673103448275862,
      "grad_norm": 0.3860645294189453,
      "learning_rate": 2.6676492262343406e-05,
      "loss": 1.2445,
      "num_input_tokens_seen": 11689376,
      "step": 2358,
      "train_runtime": 8156.904,
      "train_tokens_per_second": 1433.065
    },
    {
      "epoch": 0.8676781609195402,
      "grad_norm": 0.3797599971294403,
      "learning_rate": 2.660280029476787e-05,
      "loss": 1.1711,
      "num_input_tokens_seen": 11695104,
      "step": 2359,
      "train_runtime": 8160.684,
      "train_tokens_per_second": 1433.103
    },
    {
      "epoch": 0.8680459770114942,
      "grad_norm": 0.39891761541366577,
      "learning_rate": 2.6529108327192336e-05,
      "loss": 1.2976,
      "num_input_tokens_seen": 11699504,
      "step": 2360,
      "train_runtime": 8163.6629,
      "train_tokens_per_second": 1433.119
    },
    {
      "epoch": 0.8684137931034482,
      "grad_norm": 0.3743274211883545,
      "learning_rate": 2.6455416359616804e-05,
      "loss": 1.0993,
      "num_input_tokens_seen": 11705600,
      "step": 2361,
      "train_runtime": 8167.5662,
      "train_tokens_per_second": 1433.181
    },
    {
      "epoch": 0.8687816091954023,
      "grad_norm": 0.3728826940059662,
      "learning_rate": 2.638172439204127e-05,
      "loss": 1.1218,
      "num_input_tokens_seen": 11710704,
      "step": 2362,
      "train_runtime": 8170.9151,
      "train_tokens_per_second": 1433.218
    },
    {
      "epoch": 0.8691494252873563,
      "grad_norm": 0.36612221598625183,
      "learning_rate": 2.6308032424465734e-05,
      "loss": 1.124,
      "num_input_tokens_seen": 11716400,
      "step": 2363,
      "train_runtime": 8174.5546,
      "train_tokens_per_second": 1433.277
    },
    {
      "epoch": 0.8695172413793103,
      "grad_norm": 0.35945627093315125,
      "learning_rate": 2.62343404568902e-05,
      "loss": 1.2693,
      "num_input_tokens_seen": 11719888,
      "step": 2364,
      "train_runtime": 8176.994,
      "train_tokens_per_second": 1433.276
    },
    {
      "epoch": 0.8698850574712643,
      "grad_norm": 0.3450169861316681,
      "learning_rate": 2.6160648489314667e-05,
      "loss": 1.0919,
      "num_input_tokens_seen": 11723200,
      "step": 2365,
      "train_runtime": 8179.3493,
      "train_tokens_per_second": 1433.268
    },
    {
      "epoch": 0.8702528735632183,
      "grad_norm": 0.36018115282058716,
      "learning_rate": 2.608695652173913e-05,
      "loss": 1.0416,
      "num_input_tokens_seen": 11727456,
      "step": 2366,
      "train_runtime": 8182.2025,
      "train_tokens_per_second": 1433.288
    },
    {
      "epoch": 0.8706206896551724,
      "grad_norm": 0.3790985941886902,
      "learning_rate": 2.6013264554163596e-05,
      "loss": 1.2022,
      "num_input_tokens_seen": 11733024,
      "step": 2367,
      "train_runtime": 8185.7586,
      "train_tokens_per_second": 1433.346
    },
    {
      "epoch": 0.8709885057471264,
      "grad_norm": 0.37325188517570496,
      "learning_rate": 2.593957258658806e-05,
      "loss": 1.1948,
      "num_input_tokens_seen": 11737232,
      "step": 2368,
      "train_runtime": 8188.6892,
      "train_tokens_per_second": 1433.347
    },
    {
      "epoch": 0.8713563218390804,
      "grad_norm": 0.35205021500587463,
      "learning_rate": 2.586588061901253e-05,
      "loss": 1.2889,
      "num_input_tokens_seen": 11740784,
      "step": 2369,
      "train_runtime": 8191.1948,
      "train_tokens_per_second": 1433.342
    },
    {
      "epoch": 0.8717241379310345,
      "grad_norm": 0.3960356116294861,
      "learning_rate": 2.5792188651436994e-05,
      "loss": 1.303,
      "num_input_tokens_seen": 11746192,
      "step": 2370,
      "train_runtime": 8194.7635,
      "train_tokens_per_second": 1433.378
    },
    {
      "epoch": 0.8720919540229886,
      "grad_norm": 0.39275914430618286,
      "learning_rate": 2.571849668386146e-05,
      "loss": 1.2338,
      "num_input_tokens_seen": 11750960,
      "step": 2371,
      "train_runtime": 8200.1646,
      "train_tokens_per_second": 1433.015
    },
    {
      "epoch": 0.8724597701149426,
      "grad_norm": 0.37443453073501587,
      "learning_rate": 2.5644804716285924e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 11753840,
      "step": 2372,
      "train_runtime": 8202.3889,
      "train_tokens_per_second": 1432.978
    },
    {
      "epoch": 0.8728275862068966,
      "grad_norm": 0.3933907151222229,
      "learning_rate": 2.5571112748710392e-05,
      "loss": 1.24,
      "num_input_tokens_seen": 11758256,
      "step": 2373,
      "train_runtime": 8205.3341,
      "train_tokens_per_second": 1433.001
    },
    {
      "epoch": 0.8731954022988506,
      "grad_norm": 0.3632223606109619,
      "learning_rate": 2.5497420781134857e-05,
      "loss": 1.2591,
      "num_input_tokens_seen": 11763040,
      "step": 2374,
      "train_runtime": 8208.4342,
      "train_tokens_per_second": 1433.043
    },
    {
      "epoch": 0.8735632183908046,
      "grad_norm": 0.3614504635334015,
      "learning_rate": 2.5423728813559322e-05,
      "loss": 1.2917,
      "num_input_tokens_seen": 11767712,
      "step": 2375,
      "train_runtime": 8211.5212,
      "train_tokens_per_second": 1433.073
    },
    {
      "epoch": 0.8739310344827587,
      "grad_norm": 0.3508528470993042,
      "learning_rate": 2.5350036845983787e-05,
      "loss": 1.2213,
      "num_input_tokens_seen": 11771280,
      "step": 2376,
      "train_runtime": 8214.0226,
      "train_tokens_per_second": 1433.071
    },
    {
      "epoch": 0.8742988505747127,
      "grad_norm": 0.3659219741821289,
      "learning_rate": 2.5276344878408255e-05,
      "loss": 1.2203,
      "num_input_tokens_seen": 11778752,
      "step": 2377,
      "train_runtime": 8218.5787,
      "train_tokens_per_second": 1433.186
    },
    {
      "epoch": 0.8746666666666667,
      "grad_norm": 0.4087873697280884,
      "learning_rate": 2.520265291083272e-05,
      "loss": 1.2277,
      "num_input_tokens_seen": 11782480,
      "step": 2378,
      "train_runtime": 8221.2016,
      "train_tokens_per_second": 1433.182
    },
    {
      "epoch": 0.8750344827586207,
      "grad_norm": 0.3752245306968689,
      "learning_rate": 2.5128960943257184e-05,
      "loss": 1.1933,
      "num_input_tokens_seen": 11786624,
      "step": 2379,
      "train_runtime": 8224.0211,
      "train_tokens_per_second": 1433.195
    },
    {
      "epoch": 0.8754022988505747,
      "grad_norm": 0.36517125368118286,
      "learning_rate": 2.505526897568165e-05,
      "loss": 1.2637,
      "num_input_tokens_seen": 11789904,
      "step": 2380,
      "train_runtime": 8226.4211,
      "train_tokens_per_second": 1433.175
    },
    {
      "epoch": 0.8757701149425288,
      "grad_norm": 0.3739364445209503,
      "learning_rate": 2.4981577008106117e-05,
      "loss": 1.1952,
      "num_input_tokens_seen": 11795040,
      "step": 2381,
      "train_runtime": 8229.7892,
      "train_tokens_per_second": 1433.213
    },
    {
      "epoch": 0.8761379310344828,
      "grad_norm": 0.39218613505363464,
      "learning_rate": 2.4907885040530586e-05,
      "loss": 1.3,
      "num_input_tokens_seen": 11799552,
      "step": 2382,
      "train_runtime": 8232.786,
      "train_tokens_per_second": 1433.239
    },
    {
      "epoch": 0.8765057471264368,
      "grad_norm": 0.364801287651062,
      "learning_rate": 2.483419307295505e-05,
      "loss": 1.1685,
      "num_input_tokens_seen": 11804800,
      "step": 2383,
      "train_runtime": 8236.2268,
      "train_tokens_per_second": 1433.278
    },
    {
      "epoch": 0.8768735632183908,
      "grad_norm": 0.4224899411201477,
      "learning_rate": 2.4760501105379515e-05,
      "loss": 1.0667,
      "num_input_tokens_seen": 11810272,
      "step": 2384,
      "train_runtime": 8239.785,
      "train_tokens_per_second": 1433.323
    },
    {
      "epoch": 0.8772413793103448,
      "grad_norm": 0.3898802101612091,
      "learning_rate": 2.468680913780398e-05,
      "loss": 1.1808,
      "num_input_tokens_seen": 11815056,
      "step": 2385,
      "train_runtime": 8242.9541,
      "train_tokens_per_second": 1433.352
    },
    {
      "epoch": 0.8776091954022989,
      "grad_norm": 0.36671003699302673,
      "learning_rate": 2.4613117170228448e-05,
      "loss": 1.1094,
      "num_input_tokens_seen": 11819008,
      "step": 2386,
      "train_runtime": 8245.6328,
      "train_tokens_per_second": 1433.366
    },
    {
      "epoch": 0.8779770114942529,
      "grad_norm": 0.3437158167362213,
      "learning_rate": 2.4539425202652913e-05,
      "loss": 1.1374,
      "num_input_tokens_seen": 11823488,
      "step": 2387,
      "train_runtime": 8248.6682,
      "train_tokens_per_second": 1433.381
    },
    {
      "epoch": 0.8783448275862069,
      "grad_norm": 0.3442572057247162,
      "learning_rate": 2.4465733235077378e-05,
      "loss": 1.0513,
      "num_input_tokens_seen": 11831648,
      "step": 2388,
      "train_runtime": 8253.7413,
      "train_tokens_per_second": 1433.489
    },
    {
      "epoch": 0.8787126436781609,
      "grad_norm": 0.42760980129241943,
      "learning_rate": 2.4392041267501846e-05,
      "loss": 1.2651,
      "num_input_tokens_seen": 11836720,
      "step": 2389,
      "train_runtime": 8257.1131,
      "train_tokens_per_second": 1433.518
    },
    {
      "epoch": 0.879080459770115,
      "grad_norm": 0.36846694350242615,
      "learning_rate": 2.431834929992631e-05,
      "loss": 1.1469,
      "num_input_tokens_seen": 11842512,
      "step": 2390,
      "train_runtime": 8260.9375,
      "train_tokens_per_second": 1433.555
    },
    {
      "epoch": 0.879448275862069,
      "grad_norm": 0.36522969603538513,
      "learning_rate": 2.4244657332350776e-05,
      "loss": 1.1699,
      "num_input_tokens_seen": 11847168,
      "step": 2391,
      "train_runtime": 8264.0625,
      "train_tokens_per_second": 1433.577
    },
    {
      "epoch": 0.879816091954023,
      "grad_norm": 0.3250475525856018,
      "learning_rate": 2.417096536477524e-05,
      "loss": 1.1779,
      "num_input_tokens_seen": 11852832,
      "step": 2392,
      "train_runtime": 8267.7978,
      "train_tokens_per_second": 1433.614
    },
    {
      "epoch": 0.880183908045977,
      "grad_norm": 0.3469904959201813,
      "learning_rate": 2.409727339719971e-05,
      "loss": 1.0961,
      "num_input_tokens_seen": 11856096,
      "step": 2393,
      "train_runtime": 8270.2039,
      "train_tokens_per_second": 1433.592
    },
    {
      "epoch": 0.880551724137931,
      "grad_norm": 0.36842042207717896,
      "learning_rate": 2.4023581429624174e-05,
      "loss": 1.136,
      "num_input_tokens_seen": 11861136,
      "step": 2394,
      "train_runtime": 8273.4665,
      "train_tokens_per_second": 1433.636
    },
    {
      "epoch": 0.8809195402298851,
      "grad_norm": 0.3704577386379242,
      "learning_rate": 2.394988946204864e-05,
      "loss": 1.2025,
      "num_input_tokens_seen": 11864768,
      "step": 2395,
      "train_runtime": 8275.9489,
      "train_tokens_per_second": 1433.644
    },
    {
      "epoch": 0.8812873563218391,
      "grad_norm": 0.3620859682559967,
      "learning_rate": 2.3876197494473103e-05,
      "loss": 1.2302,
      "num_input_tokens_seen": 11868896,
      "step": 2396,
      "train_runtime": 8278.8079,
      "train_tokens_per_second": 1433.648
    },
    {
      "epoch": 0.8816551724137931,
      "grad_norm": 0.3965453505516052,
      "learning_rate": 2.380250552689757e-05,
      "loss": 1.1548,
      "num_input_tokens_seen": 11872864,
      "step": 2397,
      "train_runtime": 8281.5575,
      "train_tokens_per_second": 1433.651
    },
    {
      "epoch": 0.8820229885057471,
      "grad_norm": 0.36142468452453613,
      "learning_rate": 2.3728813559322036e-05,
      "loss": 1.0315,
      "num_input_tokens_seen": 11877536,
      "step": 2398,
      "train_runtime": 8284.7576,
      "train_tokens_per_second": 1433.661
    },
    {
      "epoch": 0.8823908045977011,
      "grad_norm": 0.350455641746521,
      "learning_rate": 2.36551215917465e-05,
      "loss": 1.1378,
      "num_input_tokens_seen": 11882064,
      "step": 2399,
      "train_runtime": 8287.7531,
      "train_tokens_per_second": 1433.689
    },
    {
      "epoch": 0.8827586206896552,
      "grad_norm": 0.38785919547080994,
      "learning_rate": 2.3581429624170966e-05,
      "loss": 1.2479,
      "num_input_tokens_seen": 11886336,
      "step": 2400,
      "train_runtime": 8290.6085,
      "train_tokens_per_second": 1433.711
    },
    {
      "epoch": 0.8831264367816092,
      "grad_norm": 0.31994694471359253,
      "learning_rate": 2.3507737656595434e-05,
      "loss": 1.1543,
      "num_input_tokens_seen": 11893536,
      "step": 2401,
      "train_runtime": 8297.0721,
      "train_tokens_per_second": 1433.462
    },
    {
      "epoch": 0.8834942528735632,
      "grad_norm": 0.36156028509140015,
      "learning_rate": 2.34340456890199e-05,
      "loss": 1.0872,
      "num_input_tokens_seen": 11897856,
      "step": 2402,
      "train_runtime": 8299.9955,
      "train_tokens_per_second": 1433.477
    },
    {
      "epoch": 0.8838620689655172,
      "grad_norm": 0.3720942735671997,
      "learning_rate": 2.3360353721444364e-05,
      "loss": 1.348,
      "num_input_tokens_seen": 11902208,
      "step": 2403,
      "train_runtime": 8303.002,
      "train_tokens_per_second": 1433.482
    },
    {
      "epoch": 0.8842298850574712,
      "grad_norm": 0.41489970684051514,
      "learning_rate": 2.328666175386883e-05,
      "loss": 1.2107,
      "num_input_tokens_seen": 11906096,
      "step": 2404,
      "train_runtime": 8305.6874,
      "train_tokens_per_second": 1433.487
    },
    {
      "epoch": 0.8845977011494253,
      "grad_norm": 0.3562310039997101,
      "learning_rate": 2.3212969786293297e-05,
      "loss": 1.2216,
      "num_input_tokens_seen": 11911024,
      "step": 2405,
      "train_runtime": 8308.9931,
      "train_tokens_per_second": 1433.51
    },
    {
      "epoch": 0.8849655172413793,
      "grad_norm": 0.4116514027118683,
      "learning_rate": 2.313927781871776e-05,
      "loss": 1.3029,
      "num_input_tokens_seen": 11914736,
      "step": 2406,
      "train_runtime": 8311.6595,
      "train_tokens_per_second": 1433.497
    },
    {
      "epoch": 0.8853333333333333,
      "grad_norm": 0.35498255491256714,
      "learning_rate": 2.3065585851142226e-05,
      "loss": 1.1718,
      "num_input_tokens_seen": 11918320,
      "step": 2407,
      "train_runtime": 8314.2888,
      "train_tokens_per_second": 1433.474
    },
    {
      "epoch": 0.8857011494252873,
      "grad_norm": 0.36428338289260864,
      "learning_rate": 2.299189388356669e-05,
      "loss": 1.2047,
      "num_input_tokens_seen": 11921616,
      "step": 2408,
      "train_runtime": 8316.7034,
      "train_tokens_per_second": 1433.455
    },
    {
      "epoch": 0.8860689655172413,
      "grad_norm": 0.3640618920326233,
      "learning_rate": 2.291820191599116e-05,
      "loss": 1.2264,
      "num_input_tokens_seen": 11926944,
      "step": 2409,
      "train_runtime": 8320.1945,
      "train_tokens_per_second": 1433.493
    },
    {
      "epoch": 0.8864367816091954,
      "grad_norm": 0.4203958511352539,
      "learning_rate": 2.2844509948415624e-05,
      "loss": 1.2425,
      "num_input_tokens_seen": 11930480,
      "step": 2410,
      "train_runtime": 8322.717,
      "train_tokens_per_second": 1433.484
    },
    {
      "epoch": 0.8868045977011494,
      "grad_norm": 0.37219858169555664,
      "learning_rate": 2.277081798084009e-05,
      "loss": 1.229,
      "num_input_tokens_seen": 11934480,
      "step": 2411,
      "train_runtime": 8325.487,
      "train_tokens_per_second": 1433.487
    },
    {
      "epoch": 0.8871724137931034,
      "grad_norm": 0.37115055322647095,
      "learning_rate": 2.2697126013264554e-05,
      "loss": 1.1379,
      "num_input_tokens_seen": 11938832,
      "step": 2412,
      "train_runtime": 8328.4305,
      "train_tokens_per_second": 1433.503
    },
    {
      "epoch": 0.8875402298850574,
      "grad_norm": 0.3520651161670685,
      "learning_rate": 2.2623434045689022e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 11944752,
      "step": 2413,
      "train_runtime": 8332.2515,
      "train_tokens_per_second": 1433.556
    },
    {
      "epoch": 0.8879080459770115,
      "grad_norm": 0.3684951066970825,
      "learning_rate": 2.2549742078113487e-05,
      "loss": 1.0369,
      "num_input_tokens_seen": 11951408,
      "step": 2414,
      "train_runtime": 8336.5168,
      "train_tokens_per_second": 1433.621
    },
    {
      "epoch": 0.8882758620689655,
      "grad_norm": 0.3673064410686493,
      "learning_rate": 2.2476050110537952e-05,
      "loss": 1.2615,
      "num_input_tokens_seen": 11959376,
      "step": 2415,
      "train_runtime": 8341.5768,
      "train_tokens_per_second": 1433.707
    },
    {
      "epoch": 0.8886436781609195,
      "grad_norm": 0.360674113035202,
      "learning_rate": 2.2402358142962417e-05,
      "loss": 1.0447,
      "num_input_tokens_seen": 11963120,
      "step": 2416,
      "train_runtime": 8344.2185,
      "train_tokens_per_second": 1433.702
    },
    {
      "epoch": 0.8890114942528735,
      "grad_norm": 0.37171250581741333,
      "learning_rate": 2.2328666175386885e-05,
      "loss": 1.1654,
      "num_input_tokens_seen": 11966944,
      "step": 2417,
      "train_runtime": 8346.8483,
      "train_tokens_per_second": 1433.708
    },
    {
      "epoch": 0.8893793103448275,
      "grad_norm": 0.3843598961830139,
      "learning_rate": 2.225497420781135e-05,
      "loss": 1.19,
      "num_input_tokens_seen": 11973888,
      "step": 2418,
      "train_runtime": 8351.1857,
      "train_tokens_per_second": 1433.795
    },
    {
      "epoch": 0.8897471264367816,
      "grad_norm": 0.41382521390914917,
      "learning_rate": 2.2181282240235814e-05,
      "loss": 1.3103,
      "num_input_tokens_seen": 11981024,
      "step": 2419,
      "train_runtime": 8355.7133,
      "train_tokens_per_second": 1433.872
    },
    {
      "epoch": 0.8901149425287357,
      "grad_norm": 0.32924750447273254,
      "learning_rate": 2.210759027266028e-05,
      "loss": 1.2093,
      "num_input_tokens_seen": 11985840,
      "step": 2420,
      "train_runtime": 8358.9326,
      "train_tokens_per_second": 1433.896
    },
    {
      "epoch": 0.8904827586206897,
      "grad_norm": 0.391390860080719,
      "learning_rate": 2.2033898305084748e-05,
      "loss": 1.0946,
      "num_input_tokens_seen": 11990752,
      "step": 2421,
      "train_runtime": 8362.207,
      "train_tokens_per_second": 1433.922
    },
    {
      "epoch": 0.8908505747126437,
      "grad_norm": 0.36776503920555115,
      "learning_rate": 2.1960206337509212e-05,
      "loss": 1.0493,
      "num_input_tokens_seen": 11995136,
      "step": 2422,
      "train_runtime": 8365.1772,
      "train_tokens_per_second": 1433.937
    },
    {
      "epoch": 0.8912183908045977,
      "grad_norm": 0.3330918848514557,
      "learning_rate": 2.1886514369933677e-05,
      "loss": 1.1392,
      "num_input_tokens_seen": 12000784,
      "step": 2423,
      "train_runtime": 8368.8469,
      "train_tokens_per_second": 1433.983
    },
    {
      "epoch": 0.8915862068965518,
      "grad_norm": 0.38090014457702637,
      "learning_rate": 2.1812822402358142e-05,
      "loss": 1.1236,
      "num_input_tokens_seen": 12004464,
      "step": 2424,
      "train_runtime": 8371.3916,
      "train_tokens_per_second": 1433.987
    },
    {
      "epoch": 0.8919540229885058,
      "grad_norm": 0.3741970956325531,
      "learning_rate": 2.173913043478261e-05,
      "loss": 1.2068,
      "num_input_tokens_seen": 12009664,
      "step": 2425,
      "train_runtime": 8374.7362,
      "train_tokens_per_second": 1434.035
    },
    {
      "epoch": 0.8923218390804598,
      "grad_norm": 0.3657437264919281,
      "learning_rate": 2.1665438467207075e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 12014784,
      "step": 2426,
      "train_runtime": 8378.0849,
      "train_tokens_per_second": 1434.073
    },
    {
      "epoch": 0.8926896551724138,
      "grad_norm": 0.38468652963638306,
      "learning_rate": 2.159174649963154e-05,
      "loss": 1.2108,
      "num_input_tokens_seen": 12019056,
      "step": 2427,
      "train_runtime": 8380.9722,
      "train_tokens_per_second": 1434.089
    },
    {
      "epoch": 0.8930574712643679,
      "grad_norm": 0.3464387059211731,
      "learning_rate": 2.1518054532056005e-05,
      "loss": 1.2607,
      "num_input_tokens_seen": 12023072,
      "step": 2428,
      "train_runtime": 8383.7336,
      "train_tokens_per_second": 1434.095
    },
    {
      "epoch": 0.8934252873563219,
      "grad_norm": 0.3254941999912262,
      "learning_rate": 2.1444362564480473e-05,
      "loss": 1.1208,
      "num_input_tokens_seen": 12028672,
      "step": 2429,
      "train_runtime": 8387.2997,
      "train_tokens_per_second": 1434.153
    },
    {
      "epoch": 0.8937931034482759,
      "grad_norm": 0.383190780878067,
      "learning_rate": 2.1370670596904938e-05,
      "loss": 1.2589,
      "num_input_tokens_seen": 12032576,
      "step": 2430,
      "train_runtime": 8389.9919,
      "train_tokens_per_second": 1434.158
    },
    {
      "epoch": 0.8941609195402299,
      "grad_norm": 0.3317315876483917,
      "learning_rate": 2.1296978629329402e-05,
      "loss": 1.0787,
      "num_input_tokens_seen": 12036272,
      "step": 2431,
      "train_runtime": 8394.624,
      "train_tokens_per_second": 1433.807
    },
    {
      "epoch": 0.8945287356321839,
      "grad_norm": 0.34554919600486755,
      "learning_rate": 2.1223286661753867e-05,
      "loss": 1.0744,
      "num_input_tokens_seen": 12039504,
      "step": 2432,
      "train_runtime": 8396.98,
      "train_tokens_per_second": 1433.79
    },
    {
      "epoch": 0.894896551724138,
      "grad_norm": 0.34781625866889954,
      "learning_rate": 2.1149594694178336e-05,
      "loss": 1.2288,
      "num_input_tokens_seen": 12046848,
      "step": 2433,
      "train_runtime": 8401.5451,
      "train_tokens_per_second": 1433.885
    },
    {
      "epoch": 0.895264367816092,
      "grad_norm": 0.3810427486896515,
      "learning_rate": 2.10759027266028e-05,
      "loss": 1.2652,
      "num_input_tokens_seen": 12051744,
      "step": 2434,
      "train_runtime": 8404.8086,
      "train_tokens_per_second": 1433.911
    },
    {
      "epoch": 0.895632183908046,
      "grad_norm": 0.35943153500556946,
      "learning_rate": 2.1002210759027265e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 12055360,
      "step": 2435,
      "train_runtime": 8407.4064,
      "train_tokens_per_second": 1433.898
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.3595828413963318,
      "learning_rate": 2.092851879145173e-05,
      "loss": 1.3073,
      "num_input_tokens_seen": 12059888,
      "step": 2436,
      "train_runtime": 8410.4383,
      "train_tokens_per_second": 1433.919
    },
    {
      "epoch": 0.896367816091954,
      "grad_norm": 0.3479820787906647,
      "learning_rate": 2.0854826823876198e-05,
      "loss": 1.0591,
      "num_input_tokens_seen": 12063344,
      "step": 2437,
      "train_runtime": 8412.9306,
      "train_tokens_per_second": 1433.905
    },
    {
      "epoch": 0.8967356321839081,
      "grad_norm": 0.4258720576763153,
      "learning_rate": 2.0781134856300663e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 12069552,
      "step": 2438,
      "train_runtime": 8416.9393,
      "train_tokens_per_second": 1433.96
    },
    {
      "epoch": 0.8971034482758621,
      "grad_norm": 0.38927775621414185,
      "learning_rate": 2.0707442888725128e-05,
      "loss": 1.3002,
      "num_input_tokens_seen": 12072912,
      "step": 2439,
      "train_runtime": 8419.4031,
      "train_tokens_per_second": 1433.939
    },
    {
      "epoch": 0.8974712643678161,
      "grad_norm": 0.3939809203147888,
      "learning_rate": 2.0633750921149596e-05,
      "loss": 1.2204,
      "num_input_tokens_seen": 12076928,
      "step": 2440,
      "train_runtime": 8422.1889,
      "train_tokens_per_second": 1433.942
    },
    {
      "epoch": 0.8978390804597701,
      "grad_norm": 0.37524181604385376,
      "learning_rate": 2.056005895357406e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 12082016,
      "step": 2441,
      "train_runtime": 8425.5927,
      "train_tokens_per_second": 1433.966
    },
    {
      "epoch": 0.8982068965517241,
      "grad_norm": 0.3529812693595886,
      "learning_rate": 2.0486366985998526e-05,
      "loss": 1.1422,
      "num_input_tokens_seen": 12085712,
      "step": 2442,
      "train_runtime": 8428.1437,
      "train_tokens_per_second": 1433.971
    },
    {
      "epoch": 0.8985747126436782,
      "grad_norm": 0.36355653405189514,
      "learning_rate": 2.0412675018422994e-05,
      "loss": 1.1259,
      "num_input_tokens_seen": 12089248,
      "step": 2443,
      "train_runtime": 8430.7314,
      "train_tokens_per_second": 1433.95
    },
    {
      "epoch": 0.8989425287356322,
      "grad_norm": 0.38057613372802734,
      "learning_rate": 2.033898305084746e-05,
      "loss": 1.0715,
      "num_input_tokens_seen": 12093120,
      "step": 2444,
      "train_runtime": 8433.4656,
      "train_tokens_per_second": 1433.944
    },
    {
      "epoch": 0.8993103448275862,
      "grad_norm": 0.3674173653125763,
      "learning_rate": 2.0265291083271927e-05,
      "loss": 1.2794,
      "num_input_tokens_seen": 12103136,
      "step": 2445,
      "train_runtime": 8439.5163,
      "train_tokens_per_second": 1434.103
    },
    {
      "epoch": 0.8996781609195402,
      "grad_norm": 0.3886840343475342,
      "learning_rate": 2.0191599115696392e-05,
      "loss": 1.2219,
      "num_input_tokens_seen": 12107664,
      "step": 2446,
      "train_runtime": 8442.5025,
      "train_tokens_per_second": 1434.132
    },
    {
      "epoch": 0.9000459770114942,
      "grad_norm": 0.36319077014923096,
      "learning_rate": 2.0117907148120857e-05,
      "loss": 1.1283,
      "num_input_tokens_seen": 12112640,
      "step": 2447,
      "train_runtime": 8445.7575,
      "train_tokens_per_second": 1434.169
    },
    {
      "epoch": 0.9004137931034483,
      "grad_norm": 0.365339994430542,
      "learning_rate": 2.004421518054532e-05,
      "loss": 1.0311,
      "num_input_tokens_seen": 12116192,
      "step": 2448,
      "train_runtime": 8448.3127,
      "train_tokens_per_second": 1434.155
    },
    {
      "epoch": 0.9004137931034483,
      "eval_loss": 1.2033369541168213,
      "eval_runtime": 15.3222,
      "eval_samples_per_second": 65.265,
      "eval_steps_per_second": 4.112,
      "num_input_tokens_seen": 12116192,
      "step": 2448
    },
    {
      "epoch": 0.9007816091954023,
      "grad_norm": 0.37838661670684814,
      "learning_rate": 1.997052321296979e-05,
      "loss": 1.2124,
      "num_input_tokens_seen": 12123168,
      "step": 2449,
      "train_runtime": 8468.1029,
      "train_tokens_per_second": 1431.627
    },
    {
      "epoch": 0.9011494252873563,
      "grad_norm": 0.379877507686615,
      "learning_rate": 1.9896831245394254e-05,
      "loss": 1.1355,
      "num_input_tokens_seen": 12128208,
      "step": 2450,
      "train_runtime": 8471.4127,
      "train_tokens_per_second": 1431.663
    },
    {
      "epoch": 0.9015172413793103,
      "grad_norm": 0.38910603523254395,
      "learning_rate": 1.982313927781872e-05,
      "loss": 1.1078,
      "num_input_tokens_seen": 12131680,
      "step": 2451,
      "train_runtime": 8473.9126,
      "train_tokens_per_second": 1431.65
    },
    {
      "epoch": 0.9018850574712644,
      "grad_norm": 0.3610929846763611,
      "learning_rate": 1.9749447310243184e-05,
      "loss": 1.1257,
      "num_input_tokens_seen": 12134944,
      "step": 2452,
      "train_runtime": 8476.3088,
      "train_tokens_per_second": 1431.631
    },
    {
      "epoch": 0.9022528735632184,
      "grad_norm": 0.37164750695228577,
      "learning_rate": 1.9675755342667652e-05,
      "loss": 1.1522,
      "num_input_tokens_seen": 12138240,
      "step": 2453,
      "train_runtime": 8478.6699,
      "train_tokens_per_second": 1431.621
    },
    {
      "epoch": 0.9026206896551724,
      "grad_norm": 0.3993352949619293,
      "learning_rate": 1.9602063375092117e-05,
      "loss": 1.2239,
      "num_input_tokens_seen": 12143008,
      "step": 2454,
      "train_runtime": 8481.843,
      "train_tokens_per_second": 1431.647
    },
    {
      "epoch": 0.9029885057471264,
      "grad_norm": 0.37187516689300537,
      "learning_rate": 1.9528371407516582e-05,
      "loss": 1.1776,
      "num_input_tokens_seen": 12148320,
      "step": 2455,
      "train_runtime": 8485.3594,
      "train_tokens_per_second": 1431.68
    },
    {
      "epoch": 0.9033563218390804,
      "grad_norm": 0.3434256315231323,
      "learning_rate": 1.9454679439941047e-05,
      "loss": 1.1513,
      "num_input_tokens_seen": 12151616,
      "step": 2456,
      "train_runtime": 8487.7751,
      "train_tokens_per_second": 1431.661
    },
    {
      "epoch": 0.9037241379310345,
      "grad_norm": 0.3511615991592407,
      "learning_rate": 1.9380987472365515e-05,
      "loss": 1.3006,
      "num_input_tokens_seen": 12156032,
      "step": 2457,
      "train_runtime": 8490.8279,
      "train_tokens_per_second": 1431.666
    },
    {
      "epoch": 0.9040919540229885,
      "grad_norm": 0.3882412016391754,
      "learning_rate": 1.930729550478998e-05,
      "loss": 1.1631,
      "num_input_tokens_seen": 12160480,
      "step": 2458,
      "train_runtime": 8493.8436,
      "train_tokens_per_second": 1431.682
    },
    {
      "epoch": 0.9044597701149425,
      "grad_norm": 0.36855489015579224,
      "learning_rate": 1.9233603537214445e-05,
      "loss": 1.2051,
      "num_input_tokens_seen": 12167392,
      "step": 2459,
      "train_runtime": 8498.0703,
      "train_tokens_per_second": 1431.783
    },
    {
      "epoch": 0.9048275862068965,
      "grad_norm": 0.3787379264831543,
      "learning_rate": 1.915991156963891e-05,
      "loss": 1.1731,
      "num_input_tokens_seen": 12172544,
      "step": 2460,
      "train_runtime": 8501.5298,
      "train_tokens_per_second": 1431.806
    },
    {
      "epoch": 0.9051954022988505,
      "grad_norm": 0.357272207736969,
      "learning_rate": 1.9086219602063378e-05,
      "loss": 1.1767,
      "num_input_tokens_seen": 12176064,
      "step": 2461,
      "train_runtime": 8506.5689,
      "train_tokens_per_second": 1431.372
    },
    {
      "epoch": 0.9055632183908046,
      "grad_norm": 0.3510552942752838,
      "learning_rate": 1.9012527634487842e-05,
      "loss": 1.1362,
      "num_input_tokens_seen": 12179440,
      "step": 2462,
      "train_runtime": 8508.9599,
      "train_tokens_per_second": 1431.366
    },
    {
      "epoch": 0.9059310344827586,
      "grad_norm": 0.36177706718444824,
      "learning_rate": 1.8938835666912307e-05,
      "loss": 1.2651,
      "num_input_tokens_seen": 12184320,
      "step": 2463,
      "train_runtime": 8512.245,
      "train_tokens_per_second": 1431.387
    },
    {
      "epoch": 0.9062988505747126,
      "grad_norm": 0.35650742053985596,
      "learning_rate": 1.8865143699336772e-05,
      "loss": 1.1847,
      "num_input_tokens_seen": 12188928,
      "step": 2464,
      "train_runtime": 8515.338,
      "train_tokens_per_second": 1431.409
    },
    {
      "epoch": 0.9066666666666666,
      "grad_norm": 0.35153520107269287,
      "learning_rate": 1.879145173176124e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 12193216,
      "step": 2465,
      "train_runtime": 8518.2521,
      "train_tokens_per_second": 1431.422
    },
    {
      "epoch": 0.9070344827586206,
      "grad_norm": 0.3745214343070984,
      "learning_rate": 1.8717759764185705e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 12198560,
      "step": 2466,
      "train_runtime": 8521.6546,
      "train_tokens_per_second": 1431.478
    },
    {
      "epoch": 0.9074022988505747,
      "grad_norm": 0.6005309820175171,
      "learning_rate": 1.864406779661017e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 12202640,
      "step": 2467,
      "train_runtime": 8524.5246,
      "train_tokens_per_second": 1431.475
    },
    {
      "epoch": 0.9077701149425287,
      "grad_norm": 0.3786957859992981,
      "learning_rate": 1.8570375829034635e-05,
      "loss": 1.1231,
      "num_input_tokens_seen": 12209984,
      "step": 2468,
      "train_runtime": 8529.0861,
      "train_tokens_per_second": 1431.57
    },
    {
      "epoch": 0.9081379310344827,
      "grad_norm": 0.37910813093185425,
      "learning_rate": 1.8496683861459103e-05,
      "loss": 1.0957,
      "num_input_tokens_seen": 12214944,
      "step": 2469,
      "train_runtime": 8532.3487,
      "train_tokens_per_second": 1431.604
    },
    {
      "epoch": 0.9085057471264368,
      "grad_norm": 0.39273855090141296,
      "learning_rate": 1.8422991893883568e-05,
      "loss": 1.1858,
      "num_input_tokens_seen": 12219808,
      "step": 2470,
      "train_runtime": 8535.5674,
      "train_tokens_per_second": 1431.634
    },
    {
      "epoch": 0.9088735632183909,
      "grad_norm": 0.39350464940071106,
      "learning_rate": 1.8349299926308033e-05,
      "loss": 1.1622,
      "num_input_tokens_seen": 12224800,
      "step": 2471,
      "train_runtime": 8538.9557,
      "train_tokens_per_second": 1431.65
    },
    {
      "epoch": 0.9092413793103449,
      "grad_norm": 0.3747163414955139,
      "learning_rate": 1.8275607958732497e-05,
      "loss": 1.1086,
      "num_input_tokens_seen": 12233248,
      "step": 2472,
      "train_runtime": 8544.1552,
      "train_tokens_per_second": 1431.768
    },
    {
      "epoch": 0.9096091954022989,
      "grad_norm": 0.36554378271102905,
      "learning_rate": 1.8201915991156966e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 12236976,
      "step": 2473,
      "train_runtime": 8546.7917,
      "train_tokens_per_second": 1431.763
    },
    {
      "epoch": 0.9099770114942529,
      "grad_norm": 0.3639166057109833,
      "learning_rate": 1.812822402358143e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 12242512,
      "step": 2474,
      "train_runtime": 8550.4639,
      "train_tokens_per_second": 1431.795
    },
    {
      "epoch": 0.9103448275862069,
      "grad_norm": 0.4087698757648468,
      "learning_rate": 1.8054532056005895e-05,
      "loss": 1.1927,
      "num_input_tokens_seen": 12251312,
      "step": 2475,
      "train_runtime": 8555.745,
      "train_tokens_per_second": 1431.94
    },
    {
      "epoch": 0.910712643678161,
      "grad_norm": 0.35056373476982117,
      "learning_rate": 1.7980840088430363e-05,
      "loss": 1.0884,
      "num_input_tokens_seen": 12255104,
      "step": 2476,
      "train_runtime": 8558.4163,
      "train_tokens_per_second": 1431.936
    },
    {
      "epoch": 0.911080459770115,
      "grad_norm": 0.39144542813301086,
      "learning_rate": 1.7907148120854828e-05,
      "loss": 1.249,
      "num_input_tokens_seen": 12260080,
      "step": 2477,
      "train_runtime": 8561.6894,
      "train_tokens_per_second": 1431.97
    },
    {
      "epoch": 0.911448275862069,
      "grad_norm": 0.3819919228553772,
      "learning_rate": 1.7833456153279293e-05,
      "loss": 1.2446,
      "num_input_tokens_seen": 12264432,
      "step": 2478,
      "train_runtime": 8564.7175,
      "train_tokens_per_second": 1431.972
    },
    {
      "epoch": 0.911816091954023,
      "grad_norm": 0.37486758828163147,
      "learning_rate": 1.7759764185703758e-05,
      "loss": 1.2748,
      "num_input_tokens_seen": 12268384,
      "step": 2479,
      "train_runtime": 8567.4436,
      "train_tokens_per_second": 1431.977
    },
    {
      "epoch": 0.912183908045977,
      "grad_norm": 0.3709360659122467,
      "learning_rate": 1.7686072218128226e-05,
      "loss": 1.2766,
      "num_input_tokens_seen": 12273120,
      "step": 2480,
      "train_runtime": 8570.5952,
      "train_tokens_per_second": 1432.003
    },
    {
      "epoch": 0.9125517241379311,
      "grad_norm": 0.3885132372379303,
      "learning_rate": 1.761238025055269e-05,
      "loss": 1.2032,
      "num_input_tokens_seen": 12277584,
      "step": 2481,
      "train_runtime": 8573.7554,
      "train_tokens_per_second": 1431.996
    },
    {
      "epoch": 0.9129195402298851,
      "grad_norm": 0.3441966772079468,
      "learning_rate": 1.7538688282977156e-05,
      "loss": 1.1289,
      "num_input_tokens_seen": 12281088,
      "step": 2482,
      "train_runtime": 8576.3067,
      "train_tokens_per_second": 1431.979
    },
    {
      "epoch": 0.9132873563218391,
      "grad_norm": 0.332922101020813,
      "learning_rate": 1.746499631540162e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 12286544,
      "step": 2483,
      "train_runtime": 8579.9477,
      "train_tokens_per_second": 1432.007
    },
    {
      "epoch": 0.9136551724137931,
      "grad_norm": 0.3894675374031067,
      "learning_rate": 1.739130434782609e-05,
      "loss": 1.205,
      "num_input_tokens_seen": 12292000,
      "step": 2484,
      "train_runtime": 8583.5373,
      "train_tokens_per_second": 1432.044
    },
    {
      "epoch": 0.9140229885057471,
      "grad_norm": 0.3474193215370178,
      "learning_rate": 1.7317612380250554e-05,
      "loss": 1.2556,
      "num_input_tokens_seen": 12297152,
      "step": 2485,
      "train_runtime": 8587.0062,
      "train_tokens_per_second": 1432.065
    },
    {
      "epoch": 0.9143908045977012,
      "grad_norm": 0.39839980006217957,
      "learning_rate": 1.724392041267502e-05,
      "loss": 1.064,
      "num_input_tokens_seen": 12300752,
      "step": 2486,
      "train_runtime": 8589.5539,
      "train_tokens_per_second": 1432.059
    },
    {
      "epoch": 0.9147586206896552,
      "grad_norm": 0.33552855253219604,
      "learning_rate": 1.7170228445099483e-05,
      "loss": 1.1282,
      "num_input_tokens_seen": 12304608,
      "step": 2487,
      "train_runtime": 8592.2305,
      "train_tokens_per_second": 1432.062
    },
    {
      "epoch": 0.9151264367816092,
      "grad_norm": 0.3747282326221466,
      "learning_rate": 1.709653647752395e-05,
      "loss": 1.0453,
      "num_input_tokens_seen": 12311968,
      "step": 2488,
      "train_runtime": 8596.8005,
      "train_tokens_per_second": 1432.157
    },
    {
      "epoch": 0.9154942528735632,
      "grad_norm": 0.3756624162197113,
      "learning_rate": 1.7022844509948416e-05,
      "loss": 1.2382,
      "num_input_tokens_seen": 12316528,
      "step": 2489,
      "train_runtime": 8599.8703,
      "train_tokens_per_second": 1432.176
    },
    {
      "epoch": 0.9158620689655173,
      "grad_norm": 0.37141913175582886,
      "learning_rate": 1.694915254237288e-05,
      "loss": 1.0413,
      "num_input_tokens_seen": 12320016,
      "step": 2490,
      "train_runtime": 8602.3705,
      "train_tokens_per_second": 1432.165
    },
    {
      "epoch": 0.9162298850574713,
      "grad_norm": 0.3895915150642395,
      "learning_rate": 1.6875460574797346e-05,
      "loss": 1.2209,
      "num_input_tokens_seen": 12323440,
      "step": 2491,
      "train_runtime": 8606.8155,
      "train_tokens_per_second": 1431.823
    },
    {
      "epoch": 0.9165977011494253,
      "grad_norm": 0.4433688223361969,
      "learning_rate": 1.6801768607221814e-05,
      "loss": 1.2876,
      "num_input_tokens_seen": 12327728,
      "step": 2492,
      "train_runtime": 8609.7421,
      "train_tokens_per_second": 1431.835
    },
    {
      "epoch": 0.9169655172413793,
      "grad_norm": 0.3654894232749939,
      "learning_rate": 1.672807663964628e-05,
      "loss": 1.1085,
      "num_input_tokens_seen": 12331840,
      "step": 2493,
      "train_runtime": 8612.6297,
      "train_tokens_per_second": 1431.832
    },
    {
      "epoch": 0.9173333333333333,
      "grad_norm": 0.3452780842781067,
      "learning_rate": 1.6654384672070744e-05,
      "loss": 1.0985,
      "num_input_tokens_seen": 12336688,
      "step": 2494,
      "train_runtime": 8615.8882,
      "train_tokens_per_second": 1431.853
    },
    {
      "epoch": 0.9177011494252874,
      "grad_norm": 0.3547157943248749,
      "learning_rate": 1.658069270449521e-05,
      "loss": 1.1784,
      "num_input_tokens_seen": 12342240,
      "step": 2495,
      "train_runtime": 8619.5322,
      "train_tokens_per_second": 1431.892
    },
    {
      "epoch": 0.9180689655172414,
      "grad_norm": 0.37024760246276855,
      "learning_rate": 1.6507000736919677e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 12352240,
      "step": 2496,
      "train_runtime": 8625.5367,
      "train_tokens_per_second": 1432.055
    },
    {
      "epoch": 0.9184367816091954,
      "grad_norm": 0.4070953130722046,
      "learning_rate": 1.643330876934414e-05,
      "loss": 1.2314,
      "num_input_tokens_seen": 12358288,
      "step": 2497,
      "train_runtime": 8629.3706,
      "train_tokens_per_second": 1432.119
    },
    {
      "epoch": 0.9188045977011494,
      "grad_norm": 0.42430737614631653,
      "learning_rate": 1.6359616801768606e-05,
      "loss": 1.1096,
      "num_input_tokens_seen": 12363984,
      "step": 2498,
      "train_runtime": 8633.0736,
      "train_tokens_per_second": 1432.165
    },
    {
      "epoch": 0.9191724137931034,
      "grad_norm": 0.4190928637981415,
      "learning_rate": 1.628592483419307e-05,
      "loss": 1.1957,
      "num_input_tokens_seen": 12367520,
      "step": 2499,
      "train_runtime": 8635.6013,
      "train_tokens_per_second": 1432.155
    },
    {
      "epoch": 0.9195402298850575,
      "grad_norm": 0.36764538288116455,
      "learning_rate": 1.621223286661754e-05,
      "loss": 1.2321,
      "num_input_tokens_seen": 12372640,
      "step": 2500,
      "train_runtime": 8639.0174,
      "train_tokens_per_second": 1432.181
    },
    {
      "epoch": 0.9199080459770115,
      "grad_norm": 0.3932344317436218,
      "learning_rate": 1.6138540899042004e-05,
      "loss": 1.0861,
      "num_input_tokens_seen": 12376704,
      "step": 2501,
      "train_runtime": 8641.8016,
      "train_tokens_per_second": 1432.19
    },
    {
      "epoch": 0.9202758620689655,
      "grad_norm": 0.390630841255188,
      "learning_rate": 1.606484893146647e-05,
      "loss": 1.2483,
      "num_input_tokens_seen": 12383472,
      "step": 2502,
      "train_runtime": 8646.1442,
      "train_tokens_per_second": 1432.254
    },
    {
      "epoch": 0.9206436781609195,
      "grad_norm": 0.3736308217048645,
      "learning_rate": 1.5991156963890934e-05,
      "loss": 1.2097,
      "num_input_tokens_seen": 12388912,
      "step": 2503,
      "train_runtime": 8649.7678,
      "train_tokens_per_second": 1432.283
    },
    {
      "epoch": 0.9210114942528735,
      "grad_norm": 0.3684477210044861,
      "learning_rate": 1.5917464996315402e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 12391984,
      "step": 2504,
      "train_runtime": 8652.0962,
      "train_tokens_per_second": 1432.252
    },
    {
      "epoch": 0.9213793103448276,
      "grad_norm": 0.3766310214996338,
      "learning_rate": 1.584377302873987e-05,
      "loss": 1.3325,
      "num_input_tokens_seen": 12397216,
      "step": 2505,
      "train_runtime": 8655.5247,
      "train_tokens_per_second": 1432.289
    },
    {
      "epoch": 0.9217471264367816,
      "grad_norm": 0.40472573041915894,
      "learning_rate": 1.5770081061164335e-05,
      "loss": 1.3397,
      "num_input_tokens_seen": 12401088,
      "step": 2506,
      "train_runtime": 8658.1997,
      "train_tokens_per_second": 1432.294
    },
    {
      "epoch": 0.9221149425287356,
      "grad_norm": 0.36540356278419495,
      "learning_rate": 1.56963890935888e-05,
      "loss": 1.2042,
      "num_input_tokens_seen": 12407328,
      "step": 2507,
      "train_runtime": 8662.1559,
      "train_tokens_per_second": 1432.36
    },
    {
      "epoch": 0.9224827586206896,
      "grad_norm": 0.3810074031352997,
      "learning_rate": 1.5622697126013265e-05,
      "loss": 1.1581,
      "num_input_tokens_seen": 12412256,
      "step": 2508,
      "train_runtime": 8665.4316,
      "train_tokens_per_second": 1432.388
    },
    {
      "epoch": 0.9228505747126436,
      "grad_norm": 0.373441606760025,
      "learning_rate": 1.5549005158437733e-05,
      "loss": 1.2202,
      "num_input_tokens_seen": 12416256,
      "step": 2509,
      "train_runtime": 8668.2917,
      "train_tokens_per_second": 1432.376
    },
    {
      "epoch": 0.9232183908045977,
      "grad_norm": 0.39730438590049744,
      "learning_rate": 1.5475313190862198e-05,
      "loss": 1.2659,
      "num_input_tokens_seen": 12422336,
      "step": 2510,
      "train_runtime": 8672.2362,
      "train_tokens_per_second": 1432.426
    },
    {
      "epoch": 0.9235862068965517,
      "grad_norm": 0.3227142095565796,
      "learning_rate": 1.5401621223286663e-05,
      "loss": 1.1228,
      "num_input_tokens_seen": 12428512,
      "step": 2511,
      "train_runtime": 8676.2907,
      "train_tokens_per_second": 1432.468
    },
    {
      "epoch": 0.9239540229885057,
      "grad_norm": 0.35089659690856934,
      "learning_rate": 1.532792925571113e-05,
      "loss": 1.1555,
      "num_input_tokens_seen": 12434928,
      "step": 2512,
      "train_runtime": 8680.4554,
      "train_tokens_per_second": 1432.52
    },
    {
      "epoch": 0.9243218390804597,
      "grad_norm": 0.38634735345840454,
      "learning_rate": 1.5254237288135596e-05,
      "loss": 1.2185,
      "num_input_tokens_seen": 12438384,
      "step": 2513,
      "train_runtime": 8682.9606,
      "train_tokens_per_second": 1432.505
    },
    {
      "epoch": 0.9246896551724137,
      "grad_norm": 0.36329758167266846,
      "learning_rate": 1.518054532056006e-05,
      "loss": 1.1645,
      "num_input_tokens_seen": 12441984,
      "step": 2514,
      "train_runtime": 8685.4576,
      "train_tokens_per_second": 1432.508
    },
    {
      "epoch": 0.9250574712643678,
      "grad_norm": 0.3851502239704132,
      "learning_rate": 1.5106853352984527e-05,
      "loss": 1.1352,
      "num_input_tokens_seen": 12447440,
      "step": 2515,
      "train_runtime": 8689.0797,
      "train_tokens_per_second": 1432.538
    },
    {
      "epoch": 0.9254252873563218,
      "grad_norm": 0.37889960408210754,
      "learning_rate": 1.5033161385408992e-05,
      "loss": 1.2014,
      "num_input_tokens_seen": 12454224,
      "step": 2516,
      "train_runtime": 8693.439,
      "train_tokens_per_second": 1432.6
    },
    {
      "epoch": 0.9257931034482758,
      "grad_norm": 0.39130479097366333,
      "learning_rate": 1.4959469417833458e-05,
      "loss": 1.2034,
      "num_input_tokens_seen": 12458032,
      "step": 2517,
      "train_runtime": 8696.1442,
      "train_tokens_per_second": 1432.593
    },
    {
      "epoch": 0.9261609195402298,
      "grad_norm": 0.3867555856704712,
      "learning_rate": 1.4885777450257923e-05,
      "loss": 1.1046,
      "num_input_tokens_seen": 12463840,
      "step": 2518,
      "train_runtime": 8699.9164,
      "train_tokens_per_second": 1432.639
    },
    {
      "epoch": 0.9265287356321839,
      "grad_norm": 0.3491637408733368,
      "learning_rate": 1.481208548268239e-05,
      "loss": 1.1492,
      "num_input_tokens_seen": 12467376,
      "step": 2519,
      "train_runtime": 8702.4386,
      "train_tokens_per_second": 1432.63
    },
    {
      "epoch": 0.926896551724138,
      "grad_norm": 0.3786649703979492,
      "learning_rate": 1.4738393515106855e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 12472976,
      "step": 2520,
      "train_runtime": 8706.0471,
      "train_tokens_per_second": 1432.68
    },
    {
      "epoch": 0.927264367816092,
      "grad_norm": 0.33817291259765625,
      "learning_rate": 1.4664701547531321e-05,
      "loss": 1.0623,
      "num_input_tokens_seen": 12476560,
      "step": 2521,
      "train_runtime": 8710.5778,
      "train_tokens_per_second": 1432.346
    },
    {
      "epoch": 0.927632183908046,
      "grad_norm": 0.4019729197025299,
      "learning_rate": 1.4591009579955786e-05,
      "loss": 1.3792,
      "num_input_tokens_seen": 12480688,
      "step": 2522,
      "train_runtime": 8713.3667,
      "train_tokens_per_second": 1432.361
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.36243730783462524,
      "learning_rate": 1.4517317612380252e-05,
      "loss": 1.1777,
      "num_input_tokens_seen": 12486608,
      "step": 2523,
      "train_runtime": 8717.2246,
      "train_tokens_per_second": 1432.406
    },
    {
      "epoch": 0.9283678160919541,
      "grad_norm": 0.40555229783058167,
      "learning_rate": 1.4443625644804717e-05,
      "loss": 1.2722,
      "num_input_tokens_seen": 12490032,
      "step": 2524,
      "train_runtime": 8719.6435,
      "train_tokens_per_second": 1432.402
    },
    {
      "epoch": 0.9287356321839081,
      "grad_norm": 0.40432339906692505,
      "learning_rate": 1.4369933677229184e-05,
      "loss": 1.1675,
      "num_input_tokens_seen": 12494720,
      "step": 2525,
      "train_runtime": 8722.8133,
      "train_tokens_per_second": 1432.419
    },
    {
      "epoch": 0.9291034482758621,
      "grad_norm": 0.4040227234363556,
      "learning_rate": 1.4296241709653649e-05,
      "loss": 1.2355,
      "num_input_tokens_seen": 12500048,
      "step": 2526,
      "train_runtime": 8726.2638,
      "train_tokens_per_second": 1432.463
    },
    {
      "epoch": 0.9294712643678161,
      "grad_norm": 0.36101672053337097,
      "learning_rate": 1.4222549742078115e-05,
      "loss": 1.1843,
      "num_input_tokens_seen": 12503296,
      "step": 2527,
      "train_runtime": 8728.6895,
      "train_tokens_per_second": 1432.437
    },
    {
      "epoch": 0.9298390804597702,
      "grad_norm": 0.3777821660041809,
      "learning_rate": 1.414885777450258e-05,
      "loss": 1.1627,
      "num_input_tokens_seen": 12507456,
      "step": 2528,
      "train_runtime": 8731.5779,
      "train_tokens_per_second": 1432.439
    },
    {
      "epoch": 0.9302068965517242,
      "grad_norm": 0.38835227489471436,
      "learning_rate": 1.4075165806927046e-05,
      "loss": 1.2043,
      "num_input_tokens_seen": 12512368,
      "step": 2529,
      "train_runtime": 8734.8654,
      "train_tokens_per_second": 1432.463
    },
    {
      "epoch": 0.9305747126436782,
      "grad_norm": 0.3832857608795166,
      "learning_rate": 1.4001473839351511e-05,
      "loss": 1.1951,
      "num_input_tokens_seen": 12516848,
      "step": 2530,
      "train_runtime": 8737.891,
      "train_tokens_per_second": 1432.479
    },
    {
      "epoch": 0.9309425287356322,
      "grad_norm": 0.35440483689308167,
      "learning_rate": 1.3927781871775978e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 12527152,
      "step": 2531,
      "train_runtime": 8744.115,
      "train_tokens_per_second": 1432.638
    },
    {
      "epoch": 0.9313103448275862,
      "grad_norm": 0.3388694226741791,
      "learning_rate": 1.3854089904200443e-05,
      "loss": 1.1632,
      "num_input_tokens_seen": 12531728,
      "step": 2532,
      "train_runtime": 8747.1868,
      "train_tokens_per_second": 1432.658
    },
    {
      "epoch": 0.9316781609195403,
      "grad_norm": 0.37156593799591064,
      "learning_rate": 1.3780397936624909e-05,
      "loss": 1.0575,
      "num_input_tokens_seen": 12537328,
      "step": 2533,
      "train_runtime": 8750.7462,
      "train_tokens_per_second": 1432.715
    },
    {
      "epoch": 0.9320459770114943,
      "grad_norm": 0.36473792791366577,
      "learning_rate": 1.3706705969049374e-05,
      "loss": 1.2026,
      "num_input_tokens_seen": 12541328,
      "step": 2534,
      "train_runtime": 8753.4487,
      "train_tokens_per_second": 1432.73
    },
    {
      "epoch": 0.9324137931034483,
      "grad_norm": 0.3509726822376251,
      "learning_rate": 1.363301400147384e-05,
      "loss": 1.1565,
      "num_input_tokens_seen": 12546368,
      "step": 2535,
      "train_runtime": 8756.8869,
      "train_tokens_per_second": 1432.743
    },
    {
      "epoch": 0.9327816091954023,
      "grad_norm": 0.3323367238044739,
      "learning_rate": 1.3559322033898305e-05,
      "loss": 1.1789,
      "num_input_tokens_seen": 12554608,
      "step": 2536,
      "train_runtime": 8761.9476,
      "train_tokens_per_second": 1432.856
    },
    {
      "epoch": 0.9331494252873563,
      "grad_norm": 0.39954280853271484,
      "learning_rate": 1.3485630066322772e-05,
      "loss": 1.2408,
      "num_input_tokens_seen": 12558608,
      "step": 2537,
      "train_runtime": 8764.8355,
      "train_tokens_per_second": 1432.84
    },
    {
      "epoch": 0.9335172413793104,
      "grad_norm": 0.378231018781662,
      "learning_rate": 1.3411938098747237e-05,
      "loss": 1.3176,
      "num_input_tokens_seen": 12561872,
      "step": 2538,
      "train_runtime": 8767.2727,
      "train_tokens_per_second": 1432.814
    },
    {
      "epoch": 0.9338850574712644,
      "grad_norm": 0.38415512442588806,
      "learning_rate": 1.3338246131171703e-05,
      "loss": 1.2404,
      "num_input_tokens_seen": 12567104,
      "step": 2539,
      "train_runtime": 8770.8248,
      "train_tokens_per_second": 1432.83
    },
    {
      "epoch": 0.9342528735632184,
      "grad_norm": 0.3389279246330261,
      "learning_rate": 1.3264554163596168e-05,
      "loss": 1.1042,
      "num_input_tokens_seen": 12570432,
      "step": 2540,
      "train_runtime": 8773.2796,
      "train_tokens_per_second": 1432.809
    },
    {
      "epoch": 0.9346206896551724,
      "grad_norm": 0.33893299102783203,
      "learning_rate": 1.3190862196020634e-05,
      "loss": 1.0801,
      "num_input_tokens_seen": 12574432,
      "step": 2541,
      "train_runtime": 8776.0549,
      "train_tokens_per_second": 1432.811
    },
    {
      "epoch": 0.9349885057471264,
      "grad_norm": 0.403434693813324,
      "learning_rate": 1.31171702284451e-05,
      "loss": 1.1139,
      "num_input_tokens_seen": 12579056,
      "step": 2542,
      "train_runtime": 8779.074,
      "train_tokens_per_second": 1432.845
    },
    {
      "epoch": 0.9353563218390805,
      "grad_norm": 0.39922621846199036,
      "learning_rate": 1.3043478260869566e-05,
      "loss": 1.2206,
      "num_input_tokens_seen": 12583968,
      "step": 2543,
      "train_runtime": 8782.3478,
      "train_tokens_per_second": 1432.871
    },
    {
      "epoch": 0.9357241379310345,
      "grad_norm": 0.3386867940425873,
      "learning_rate": 1.296978629329403e-05,
      "loss": 1.1265,
      "num_input_tokens_seen": 12588064,
      "step": 2544,
      "train_runtime": 8785.1712,
      "train_tokens_per_second": 1432.876
    },
    {
      "epoch": 0.9360919540229885,
      "grad_norm": 0.3719545006752014,
      "learning_rate": 1.2896094325718497e-05,
      "loss": 1.1342,
      "num_input_tokens_seen": 12592336,
      "step": 2545,
      "train_runtime": 8788.1685,
      "train_tokens_per_second": 1432.874
    },
    {
      "epoch": 0.9364597701149425,
      "grad_norm": 0.4040217399597168,
      "learning_rate": 1.2822402358142962e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 12596976,
      "step": 2546,
      "train_runtime": 8791.3135,
      "train_tokens_per_second": 1432.889
    },
    {
      "epoch": 0.9368275862068965,
      "grad_norm": 0.3798476755619049,
      "learning_rate": 1.2748710390567428e-05,
      "loss": 1.1829,
      "num_input_tokens_seen": 12604400,
      "step": 2547,
      "train_runtime": 8796.0338,
      "train_tokens_per_second": 1432.964
    },
    {
      "epoch": 0.9371954022988506,
      "grad_norm": 0.36366572976112366,
      "learning_rate": 1.2675018422991893e-05,
      "loss": 1.1791,
      "num_input_tokens_seen": 12609392,
      "step": 2548,
      "train_runtime": 8799.3262,
      "train_tokens_per_second": 1432.995
    },
    {
      "epoch": 0.9375632183908046,
      "grad_norm": 0.3968220055103302,
      "learning_rate": 1.260132645541636e-05,
      "loss": 1.159,
      "num_input_tokens_seen": 12615264,
      "step": 2549,
      "train_runtime": 8803.1812,
      "train_tokens_per_second": 1433.035
    },
    {
      "epoch": 0.9379310344827586,
      "grad_norm": 0.34876206517219543,
      "learning_rate": 1.2527634487840825e-05,
      "loss": 1.1937,
      "num_input_tokens_seen": 12621904,
      "step": 2550,
      "train_runtime": 8807.4091,
      "train_tokens_per_second": 1433.101
    },
    {
      "epoch": 0.9382988505747126,
      "grad_norm": 0.36600789427757263,
      "learning_rate": 1.2453942520265293e-05,
      "loss": 1.1573,
      "num_input_tokens_seen": 12625760,
      "step": 2551,
      "train_runtime": 8812.2221,
      "train_tokens_per_second": 1432.756
    },
    {
      "epoch": 0.9386666666666666,
      "grad_norm": 0.4065242111682892,
      "learning_rate": 1.2380250552689758e-05,
      "loss": 1.2907,
      "num_input_tokens_seen": 12630928,
      "step": 2552,
      "train_runtime": 8815.6538,
      "train_tokens_per_second": 1432.784
    },
    {
      "epoch": 0.9390344827586207,
      "grad_norm": 0.3677788972854614,
      "learning_rate": 1.2306558585114224e-05,
      "loss": 1.1828,
      "num_input_tokens_seen": 12635392,
      "step": 2553,
      "train_runtime": 8818.6534,
      "train_tokens_per_second": 1432.803
    },
    {
      "epoch": 0.9394022988505747,
      "grad_norm": 0.38127607107162476,
      "learning_rate": 1.2232866617538689e-05,
      "loss": 1.2158,
      "num_input_tokens_seen": 12639776,
      "step": 2554,
      "train_runtime": 8821.6439,
      "train_tokens_per_second": 1432.814
    },
    {
      "epoch": 0.9397701149425287,
      "grad_norm": 0.3584839105606079,
      "learning_rate": 1.2159174649963155e-05,
      "loss": 1.129,
      "num_input_tokens_seen": 12647728,
      "step": 2555,
      "train_runtime": 8826.6192,
      "train_tokens_per_second": 1432.907
    },
    {
      "epoch": 0.9401379310344827,
      "grad_norm": 0.35018038749694824,
      "learning_rate": 1.208548268238762e-05,
      "loss": 1.1501,
      "num_input_tokens_seen": 12653712,
      "step": 2556,
      "train_runtime": 8830.5315,
      "train_tokens_per_second": 1432.95
    },
    {
      "epoch": 0.9405057471264368,
      "grad_norm": 0.3529202342033386,
      "learning_rate": 1.2011790714812087e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 12658416,
      "step": 2557,
      "train_runtime": 8833.6765,
      "train_tokens_per_second": 1432.973
    },
    {
      "epoch": 0.9408735632183908,
      "grad_norm": 0.36576128005981445,
      "learning_rate": 1.1938098747236552e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 12665728,
      "step": 2558,
      "train_runtime": 8838.2206,
      "train_tokens_per_second": 1433.063
    },
    {
      "epoch": 0.9412413793103448,
      "grad_norm": 0.38364243507385254,
      "learning_rate": 1.1864406779661018e-05,
      "loss": 1.2179,
      "num_input_tokens_seen": 12673776,
      "step": 2559,
      "train_runtime": 8843.2391,
      "train_tokens_per_second": 1433.16
    },
    {
      "epoch": 0.9416091954022988,
      "grad_norm": 0.38225045800209045,
      "learning_rate": 1.1790714812085483e-05,
      "loss": 1.0608,
      "num_input_tokens_seen": 12679968,
      "step": 2560,
      "train_runtime": 8847.2828,
      "train_tokens_per_second": 1433.205
    },
    {
      "epoch": 0.9419770114942528,
      "grad_norm": 0.3517354428768158,
      "learning_rate": 1.171702284450995e-05,
      "loss": 1.0932,
      "num_input_tokens_seen": 12686352,
      "step": 2561,
      "train_runtime": 8851.4069,
      "train_tokens_per_second": 1433.258
    },
    {
      "epoch": 0.9423448275862069,
      "grad_norm": 0.3529907464981079,
      "learning_rate": 1.1643330876934414e-05,
      "loss": 1.1662,
      "num_input_tokens_seen": 12691264,
      "step": 2562,
      "train_runtime": 8854.6544,
      "train_tokens_per_second": 1433.287
    },
    {
      "epoch": 0.9427126436781609,
      "grad_norm": 0.44911494851112366,
      "learning_rate": 1.156963890935888e-05,
      "loss": 1.1467,
      "num_input_tokens_seen": 12698480,
      "step": 2563,
      "train_runtime": 8859.2483,
      "train_tokens_per_second": 1433.359
    },
    {
      "epoch": 0.9430804597701149,
      "grad_norm": 0.3625054955482483,
      "learning_rate": 1.1495946941783346e-05,
      "loss": 1.123,
      "num_input_tokens_seen": 12702864,
      "step": 2564,
      "train_runtime": 8862.2128,
      "train_tokens_per_second": 1433.374
    },
    {
      "epoch": 0.9434482758620689,
      "grad_norm": 0.3507109582424164,
      "learning_rate": 1.1422254974207812e-05,
      "loss": 1.2098,
      "num_input_tokens_seen": 12706896,
      "step": 2565,
      "train_runtime": 8865.0445,
      "train_tokens_per_second": 1433.371
    },
    {
      "epoch": 0.9438160919540229,
      "grad_norm": 0.37830185890197754,
      "learning_rate": 1.1348563006632277e-05,
      "loss": 1.2039,
      "num_input_tokens_seen": 12710448,
      "step": 2566,
      "train_runtime": 8867.6649,
      "train_tokens_per_second": 1433.348
    },
    {
      "epoch": 0.944183908045977,
      "grad_norm": 0.37593886256217957,
      "learning_rate": 1.1274871039056743e-05,
      "loss": 1.0765,
      "num_input_tokens_seen": 12715456,
      "step": 2567,
      "train_runtime": 8870.9802,
      "train_tokens_per_second": 1433.377
    },
    {
      "epoch": 0.944551724137931,
      "grad_norm": 0.36541810631752014,
      "learning_rate": 1.1201179071481208e-05,
      "loss": 1.1554,
      "num_input_tokens_seen": 12718928,
      "step": 2568,
      "train_runtime": 8873.5351,
      "train_tokens_per_second": 1433.355
    },
    {
      "epoch": 0.944919540229885,
      "grad_norm": 0.34366193413734436,
      "learning_rate": 1.1127487103905675e-05,
      "loss": 1.0039,
      "num_input_tokens_seen": 12724192,
      "step": 2569,
      "train_runtime": 8877.0485,
      "train_tokens_per_second": 1433.381
    },
    {
      "epoch": 0.9452873563218391,
      "grad_norm": 0.3616051971912384,
      "learning_rate": 1.105379513633014e-05,
      "loss": 1.2761,
      "num_input_tokens_seen": 12728064,
      "step": 2570,
      "train_runtime": 8879.7794,
      "train_tokens_per_second": 1433.376
    },
    {
      "epoch": 0.9456551724137932,
      "grad_norm": 0.3893207907676697,
      "learning_rate": 1.0980103168754606e-05,
      "loss": 1.2434,
      "num_input_tokens_seen": 12733440,
      "step": 2571,
      "train_runtime": 8883.3757,
      "train_tokens_per_second": 1433.401
    },
    {
      "epoch": 0.9460229885057472,
      "grad_norm": 0.38037747144699097,
      "learning_rate": 1.0906411201179071e-05,
      "loss": 1.3139,
      "num_input_tokens_seen": 12737264,
      "step": 2572,
      "train_runtime": 8886.0055,
      "train_tokens_per_second": 1433.407
    },
    {
      "epoch": 0.9463908045977012,
      "grad_norm": 0.41482865810394287,
      "learning_rate": 1.0832719233603538e-05,
      "loss": 1.1259,
      "num_input_tokens_seen": 12743360,
      "step": 2573,
      "train_runtime": 8889.95,
      "train_tokens_per_second": 1433.457
    },
    {
      "epoch": 0.9467586206896552,
      "grad_norm": 0.4118165969848633,
      "learning_rate": 1.0759027266028002e-05,
      "loss": 1.2237,
      "num_input_tokens_seen": 12747872,
      "step": 2574,
      "train_runtime": 8892.9988,
      "train_tokens_per_second": 1433.473
    },
    {
      "epoch": 0.9471264367816092,
      "grad_norm": 0.4312781095504761,
      "learning_rate": 1.0685335298452469e-05,
      "loss": 1.2888,
      "num_input_tokens_seen": 12752000,
      "step": 2575,
      "train_runtime": 8895.8468,
      "train_tokens_per_second": 1433.478
    },
    {
      "epoch": 0.9474942528735633,
      "grad_norm": 0.31879571080207825,
      "learning_rate": 1.0611643330876934e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 12756016,
      "step": 2576,
      "train_runtime": 8898.5845,
      "train_tokens_per_second": 1433.488
    },
    {
      "epoch": 0.9478620689655173,
      "grad_norm": 0.37517669796943665,
      "learning_rate": 1.05379513633014e-05,
      "loss": 1.1017,
      "num_input_tokens_seen": 12761696,
      "step": 2577,
      "train_runtime": 8902.2535,
      "train_tokens_per_second": 1433.535
    },
    {
      "epoch": 0.9482298850574713,
      "grad_norm": 0.38680213689804077,
      "learning_rate": 1.0464259395725865e-05,
      "loss": 1.2187,
      "num_input_tokens_seen": 12767776,
      "step": 2578,
      "train_runtime": 8906.1848,
      "train_tokens_per_second": 1433.585
    },
    {
      "epoch": 0.9485977011494253,
      "grad_norm": 0.37336546182632446,
      "learning_rate": 1.0390567428150332e-05,
      "loss": 1.2181,
      "num_input_tokens_seen": 12772320,
      "step": 2579,
      "train_runtime": 8909.2796,
      "train_tokens_per_second": 1433.597
    },
    {
      "epoch": 0.9489655172413793,
      "grad_norm": 0.3453748822212219,
      "learning_rate": 1.0316875460574798e-05,
      "loss": 1.2581,
      "num_input_tokens_seen": 12777440,
      "step": 2580,
      "train_runtime": 8912.7079,
      "train_tokens_per_second": 1433.62
    },
    {
      "epoch": 0.9493333333333334,
      "grad_norm": 0.349387526512146,
      "learning_rate": 1.0243183492999263e-05,
      "loss": 1.1171,
      "num_input_tokens_seen": 12781008,
      "step": 2581,
      "train_runtime": 8917.4782,
      "train_tokens_per_second": 1433.254
    },
    {
      "epoch": 0.9497011494252874,
      "grad_norm": 0.35860323905944824,
      "learning_rate": 1.016949152542373e-05,
      "loss": 1.1587,
      "num_input_tokens_seen": 12787392,
      "step": 2582,
      "train_runtime": 8921.6211,
      "train_tokens_per_second": 1433.304
    },
    {
      "epoch": 0.9500689655172414,
      "grad_norm": 0.3526651859283447,
      "learning_rate": 1.0095799557848196e-05,
      "loss": 1.1402,
      "num_input_tokens_seen": 12791872,
      "step": 2583,
      "train_runtime": 8924.68,
      "train_tokens_per_second": 1433.314
    },
    {
      "epoch": 0.9504367816091954,
      "grad_norm": 0.3840789198875427,
      "learning_rate": 1.002210759027266e-05,
      "loss": 1.3412,
      "num_input_tokens_seen": 12799696,
      "step": 2584,
      "train_runtime": 8929.6252,
      "train_tokens_per_second": 1433.397
    },
    {
      "epoch": 0.9508045977011494,
      "grad_norm": 0.3874008357524872,
      "learning_rate": 9.948415622697127e-06,
      "loss": 1.1957,
      "num_input_tokens_seen": 12805120,
      "step": 2585,
      "train_runtime": 8933.1839,
      "train_tokens_per_second": 1433.433
    },
    {
      "epoch": 0.9511724137931035,
      "grad_norm": 0.3895336985588074,
      "learning_rate": 9.874723655121592e-06,
      "loss": 1.2269,
      "num_input_tokens_seen": 12810320,
      "step": 2586,
      "train_runtime": 8936.6843,
      "train_tokens_per_second": 1433.453
    },
    {
      "epoch": 0.9515402298850575,
      "grad_norm": 0.38132545351982117,
      "learning_rate": 9.801031687546059e-06,
      "loss": 1.2226,
      "num_input_tokens_seen": 12814624,
      "step": 2587,
      "train_runtime": 8939.6801,
      "train_tokens_per_second": 1433.454
    },
    {
      "epoch": 0.9519080459770115,
      "grad_norm": 0.38565635681152344,
      "learning_rate": 9.727339719970523e-06,
      "loss": 1.3191,
      "num_input_tokens_seen": 12819920,
      "step": 2588,
      "train_runtime": 8943.1142,
      "train_tokens_per_second": 1433.496
    },
    {
      "epoch": 0.9522758620689655,
      "grad_norm": 0.3734370172023773,
      "learning_rate": 9.65364775239499e-06,
      "loss": 1.2672,
      "num_input_tokens_seen": 12824256,
      "step": 2589,
      "train_runtime": 8946.0142,
      "train_tokens_per_second": 1433.516
    },
    {
      "epoch": 0.9526436781609195,
      "grad_norm": 0.38229501247406006,
      "learning_rate": 9.579955784819455e-06,
      "loss": 1.067,
      "num_input_tokens_seen": 12827936,
      "step": 2590,
      "train_runtime": 8948.607,
      "train_tokens_per_second": 1433.512
    },
    {
      "epoch": 0.9530114942528736,
      "grad_norm": 0.35377880930900574,
      "learning_rate": 9.506263817243921e-06,
      "loss": 1.1216,
      "num_input_tokens_seen": 12831744,
      "step": 2591,
      "train_runtime": 8951.3088,
      "train_tokens_per_second": 1433.505
    },
    {
      "epoch": 0.9533793103448276,
      "grad_norm": 0.3394309878349304,
      "learning_rate": 9.432571849668386e-06,
      "loss": 1.2458,
      "num_input_tokens_seen": 12837200,
      "step": 2592,
      "train_runtime": 8954.9699,
      "train_tokens_per_second": 1433.528
    },
    {
      "epoch": 0.9537471264367816,
      "grad_norm": 0.3668551743030548,
      "learning_rate": 9.358879882092853e-06,
      "loss": 1.2165,
      "num_input_tokens_seen": 12840992,
      "step": 2593,
      "train_runtime": 8957.6589,
      "train_tokens_per_second": 1433.521
    },
    {
      "epoch": 0.9541149425287356,
      "grad_norm": 0.32961776852607727,
      "learning_rate": 9.285187914517317e-06,
      "loss": 1.2615,
      "num_input_tokens_seen": 12845296,
      "step": 2594,
      "train_runtime": 8960.653,
      "train_tokens_per_second": 1433.522
    },
    {
      "epoch": 0.9544827586206897,
      "grad_norm": 0.40517279505729675,
      "learning_rate": 9.211495946941784e-06,
      "loss": 1.1891,
      "num_input_tokens_seen": 12850880,
      "step": 2595,
      "train_runtime": 8964.3592,
      "train_tokens_per_second": 1433.553
    },
    {
      "epoch": 0.9548505747126437,
      "grad_norm": 0.37425488233566284,
      "learning_rate": 9.137803979366249e-06,
      "loss": 1.1559,
      "num_input_tokens_seen": 12855984,
      "step": 2596,
      "train_runtime": 8967.7607,
      "train_tokens_per_second": 1433.578
    },
    {
      "epoch": 0.9552183908045977,
      "grad_norm": 0.37122002243995667,
      "learning_rate": 9.064112011790715e-06,
      "loss": 1.1885,
      "num_input_tokens_seen": 12860224,
      "step": 2597,
      "train_runtime": 8970.7211,
      "train_tokens_per_second": 1433.578
    },
    {
      "epoch": 0.9555862068965517,
      "grad_norm": 0.3670824468135834,
      "learning_rate": 8.990420044215182e-06,
      "loss": 1.149,
      "num_input_tokens_seen": 12863360,
      "step": 2598,
      "train_runtime": 8973.0784,
      "train_tokens_per_second": 1433.55
    },
    {
      "epoch": 0.9559540229885057,
      "grad_norm": 0.41599032282829285,
      "learning_rate": 8.916728076639647e-06,
      "loss": 1.299,
      "num_input_tokens_seen": 12867632,
      "step": 2599,
      "train_runtime": 8975.9753,
      "train_tokens_per_second": 1433.564
    },
    {
      "epoch": 0.9563218390804598,
      "grad_norm": 0.3932607173919678,
      "learning_rate": 8.843036109064113e-06,
      "loss": 1.2306,
      "num_input_tokens_seen": 12872960,
      "step": 2600,
      "train_runtime": 8979.5026,
      "train_tokens_per_second": 1433.594
    },
    {
      "epoch": 0.9566896551724138,
      "grad_norm": 0.3910403847694397,
      "learning_rate": 8.769344141488578e-06,
      "loss": 1.3124,
      "num_input_tokens_seen": 12877552,
      "step": 2601,
      "train_runtime": 8982.6007,
      "train_tokens_per_second": 1433.611
    },
    {
      "epoch": 0.9570574712643678,
      "grad_norm": 0.38028043508529663,
      "learning_rate": 8.695652173913044e-06,
      "loss": 1.1597,
      "num_input_tokens_seen": 12883952,
      "step": 2602,
      "train_runtime": 8986.69,
      "train_tokens_per_second": 1433.67
    },
    {
      "epoch": 0.9574252873563218,
      "grad_norm": 0.39262303709983826,
      "learning_rate": 8.62196020633751e-06,
      "loss": 1.2596,
      "num_input_tokens_seen": 12887696,
      "step": 2603,
      "train_runtime": 8989.271,
      "train_tokens_per_second": 1433.675
    },
    {
      "epoch": 0.9577931034482758,
      "grad_norm": 0.32645902037620544,
      "learning_rate": 8.548268238761976e-06,
      "loss": 1.1844,
      "num_input_tokens_seen": 12898432,
      "step": 2604,
      "train_runtime": 8995.6472,
      "train_tokens_per_second": 1433.853
    },
    {
      "epoch": 0.9581609195402299,
      "grad_norm": 0.34202808141708374,
      "learning_rate": 8.47457627118644e-06,
      "loss": 1.1345,
      "num_input_tokens_seen": 12904416,
      "step": 2605,
      "train_runtime": 8999.5357,
      "train_tokens_per_second": 1433.898
    },
    {
      "epoch": 0.9585287356321839,
      "grad_norm": 0.3665388524532318,
      "learning_rate": 8.400884303610907e-06,
      "loss": 1.2556,
      "num_input_tokens_seen": 12909888,
      "step": 2606,
      "train_runtime": 9003.0978,
      "train_tokens_per_second": 1433.938
    },
    {
      "epoch": 0.9588965517241379,
      "grad_norm": 0.35638895630836487,
      "learning_rate": 8.327192336035372e-06,
      "loss": 1.1493,
      "num_input_tokens_seen": 12915232,
      "step": 2607,
      "train_runtime": 9006.5265,
      "train_tokens_per_second": 1433.986
    },
    {
      "epoch": 0.9592643678160919,
      "grad_norm": 0.3463881313800812,
      "learning_rate": 8.253500368459838e-06,
      "loss": 1.099,
      "num_input_tokens_seen": 12919280,
      "step": 2608,
      "train_runtime": 9009.3218,
      "train_tokens_per_second": 1433.99
    },
    {
      "epoch": 0.9596321839080459,
      "grad_norm": 0.37970536947250366,
      "learning_rate": 8.179808400884303e-06,
      "loss": 1.1595,
      "num_input_tokens_seen": 12925504,
      "step": 2609,
      "train_runtime": 9013.3023,
      "train_tokens_per_second": 1434.048
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.33949676156044006,
      "learning_rate": 8.10611643330877e-06,
      "loss": 1.1059,
      "num_input_tokens_seen": 12932624,
      "step": 2610,
      "train_runtime": 9017.7466,
      "train_tokens_per_second": 1434.13
    },
    {
      "epoch": 0.960367816091954,
      "grad_norm": 0.38554108142852783,
      "learning_rate": 8.032424465733235e-06,
      "loss": 1.1818,
      "num_input_tokens_seen": 12936144,
      "step": 2611,
      "train_runtime": 9022.348,
      "train_tokens_per_second": 1433.789
    },
    {
      "epoch": 0.960735632183908,
      "grad_norm": 0.40498021245002747,
      "learning_rate": 7.958732498157701e-06,
      "loss": 1.1977,
      "num_input_tokens_seen": 12944112,
      "step": 2612,
      "train_runtime": 9027.3734,
      "train_tokens_per_second": 1433.874
    },
    {
      "epoch": 0.961103448275862,
      "grad_norm": 0.35562485456466675,
      "learning_rate": 7.885040530582168e-06,
      "loss": 1.2872,
      "num_input_tokens_seen": 12947984,
      "step": 2613,
      "train_runtime": 9030.127,
      "train_tokens_per_second": 1433.865
    },
    {
      "epoch": 0.961471264367816,
      "grad_norm": 0.3915903866291046,
      "learning_rate": 7.811348563006632e-06,
      "loss": 1.2613,
      "num_input_tokens_seen": 12952080,
      "step": 2614,
      "train_runtime": 9033.0331,
      "train_tokens_per_second": 1433.857
    },
    {
      "epoch": 0.9618390804597701,
      "grad_norm": 0.38376888632774353,
      "learning_rate": 7.737656595431099e-06,
      "loss": 1.1156,
      "num_input_tokens_seen": 12955440,
      "step": 2615,
      "train_runtime": 9035.5185,
      "train_tokens_per_second": 1433.835
    },
    {
      "epoch": 0.9622068965517241,
      "grad_norm": 0.3606862723827362,
      "learning_rate": 7.663964627855565e-06,
      "loss": 1.2131,
      "num_input_tokens_seen": 12958896,
      "step": 2616,
      "train_runtime": 9038.0468,
      "train_tokens_per_second": 1433.816
    },
    {
      "epoch": 0.9625747126436781,
      "grad_norm": 0.35248076915740967,
      "learning_rate": 7.59027266028003e-06,
      "loss": 1.2635,
      "num_input_tokens_seen": 12964512,
      "step": 2617,
      "train_runtime": 9041.7146,
      "train_tokens_per_second": 1433.855
    },
    {
      "epoch": 0.9629425287356321,
      "grad_norm": 0.4109298586845398,
      "learning_rate": 7.516580692704496e-06,
      "loss": 1.1903,
      "num_input_tokens_seen": 12970128,
      "step": 2618,
      "train_runtime": 9045.401,
      "train_tokens_per_second": 1433.892
    },
    {
      "epoch": 0.9633103448275863,
      "grad_norm": 0.38341623544692993,
      "learning_rate": 7.442888725128962e-06,
      "loss": 1.3705,
      "num_input_tokens_seen": 12973488,
      "step": 2619,
      "train_runtime": 9047.8819,
      "train_tokens_per_second": 1433.87
    },
    {
      "epoch": 0.9636781609195403,
      "grad_norm": 0.37579721212387085,
      "learning_rate": 7.369196757553427e-06,
      "loss": 1.1324,
      "num_input_tokens_seen": 12981168,
      "step": 2620,
      "train_runtime": 9052.7103,
      "train_tokens_per_second": 1433.954
    },
    {
      "epoch": 0.9640459770114943,
      "grad_norm": 0.388622522354126,
      "learning_rate": 7.295504789977893e-06,
      "loss": 1.2855,
      "num_input_tokens_seen": 12985168,
      "step": 2621,
      "train_runtime": 9055.4927,
      "train_tokens_per_second": 1433.955
    },
    {
      "epoch": 0.9644137931034483,
      "grad_norm": 0.36083441972732544,
      "learning_rate": 7.221812822402359e-06,
      "loss": 1.147,
      "num_input_tokens_seen": 12990240,
      "step": 2622,
      "train_runtime": 9058.9246,
      "train_tokens_per_second": 1433.972
    },
    {
      "epoch": 0.9647816091954023,
      "grad_norm": 0.3841739892959595,
      "learning_rate": 7.148120854826824e-06,
      "loss": 1.1187,
      "num_input_tokens_seen": 12996256,
      "step": 2623,
      "train_runtime": 9062.7806,
      "train_tokens_per_second": 1434.025
    },
    {
      "epoch": 0.9651494252873564,
      "grad_norm": 0.35147175192832947,
      "learning_rate": 7.07442888725129e-06,
      "loss": 1.1573,
      "num_input_tokens_seen": 13001728,
      "step": 2624,
      "train_runtime": 9066.4402,
      "train_tokens_per_second": 1434.05
    },
    {
      "epoch": 0.9655172413793104,
      "grad_norm": 0.38508978486061096,
      "learning_rate": 7.000736919675756e-06,
      "loss": 1.2295,
      "num_input_tokens_seen": 13006512,
      "step": 2625,
      "train_runtime": 9069.6659,
      "train_tokens_per_second": 1434.067
    },
    {
      "epoch": 0.9658850574712644,
      "grad_norm": 0.3841729164123535,
      "learning_rate": 6.927044952100221e-06,
      "loss": 1.2373,
      "num_input_tokens_seen": 13013792,
      "step": 2626,
      "train_runtime": 9074.1357,
      "train_tokens_per_second": 1434.163
    },
    {
      "epoch": 0.9662528735632184,
      "grad_norm": 0.3278234601020813,
      "learning_rate": 6.853352984524687e-06,
      "loss": 1.2258,
      "num_input_tokens_seen": 13022400,
      "step": 2627,
      "train_runtime": 9079.4286,
      "train_tokens_per_second": 1434.275
    },
    {
      "epoch": 0.9666206896551724,
      "grad_norm": 0.3791029453277588,
      "learning_rate": 6.779661016949153e-06,
      "loss": 1.0661,
      "num_input_tokens_seen": 13026704,
      "step": 2628,
      "train_runtime": 9082.4326,
      "train_tokens_per_second": 1434.275
    },
    {
      "epoch": 0.9669885057471265,
      "grad_norm": 0.45164456963539124,
      "learning_rate": 6.705969049373618e-06,
      "loss": 1.1975,
      "num_input_tokens_seen": 13030656,
      "step": 2629,
      "train_runtime": 9085.2075,
      "train_tokens_per_second": 1434.272
    },
    {
      "epoch": 0.9673563218390805,
      "grad_norm": 0.3532894551753998,
      "learning_rate": 6.632277081798084e-06,
      "loss": 1.2566,
      "num_input_tokens_seen": 13035440,
      "step": 2630,
      "train_runtime": 9088.4308,
      "train_tokens_per_second": 1434.289
    },
    {
      "epoch": 0.9677241379310345,
      "grad_norm": 0.35183894634246826,
      "learning_rate": 6.55858511422255e-06,
      "loss": 1.2102,
      "num_input_tokens_seen": 13038544,
      "step": 2631,
      "train_runtime": 9090.7743,
      "train_tokens_per_second": 1434.261
    },
    {
      "epoch": 0.9680919540229885,
      "grad_norm": 0.3607577085494995,
      "learning_rate": 6.484893146647015e-06,
      "loss": 1.1183,
      "num_input_tokens_seen": 13043536,
      "step": 2632,
      "train_runtime": 9094.0634,
      "train_tokens_per_second": 1434.291
    },
    {
      "epoch": 0.9684597701149426,
      "grad_norm": 0.3834487795829773,
      "learning_rate": 6.411201179071481e-06,
      "loss": 1.3596,
      "num_input_tokens_seen": 13050496,
      "step": 2633,
      "train_runtime": 9098.3932,
      "train_tokens_per_second": 1434.374
    },
    {
      "epoch": 0.9688275862068966,
      "grad_norm": 0.3820025324821472,
      "learning_rate": 6.337509211495947e-06,
      "loss": 1.145,
      "num_input_tokens_seen": 13056144,
      "step": 2634,
      "train_runtime": 9102.0635,
      "train_tokens_per_second": 1434.416
    },
    {
      "epoch": 0.9691954022988506,
      "grad_norm": 0.361589252948761,
      "learning_rate": 6.263817243920412e-06,
      "loss": 1.1775,
      "num_input_tokens_seen": 13061920,
      "step": 2635,
      "train_runtime": 9105.7989,
      "train_tokens_per_second": 1434.462
    },
    {
      "epoch": 0.9695632183908046,
      "grad_norm": 0.3981812596321106,
      "learning_rate": 6.190125276344879e-06,
      "loss": 1.1227,
      "num_input_tokens_seen": 13069408,
      "step": 2636,
      "train_runtime": 9110.5384,
      "train_tokens_per_second": 1434.537
    },
    {
      "epoch": 0.9699310344827586,
      "grad_norm": 0.3458286225795746,
      "learning_rate": 6.1164333087693445e-06,
      "loss": 1.1424,
      "num_input_tokens_seen": 13073968,
      "step": 2637,
      "train_runtime": 9113.6476,
      "train_tokens_per_second": 1434.548
    },
    {
      "epoch": 0.9702988505747127,
      "grad_norm": 0.3819259703159332,
      "learning_rate": 6.04274134119381e-06,
      "loss": 1.1816,
      "num_input_tokens_seen": 13083936,
      "step": 2638,
      "train_runtime": 9119.7206,
      "train_tokens_per_second": 1434.686
    },
    {
      "epoch": 0.9706666666666667,
      "grad_norm": 0.3737925589084625,
      "learning_rate": 5.969049373618276e-06,
      "loss": 1.2839,
      "num_input_tokens_seen": 13087856,
      "step": 2639,
      "train_runtime": 9122.459,
      "train_tokens_per_second": 1434.685
    },
    {
      "epoch": 0.9710344827586207,
      "grad_norm": 0.3687480390071869,
      "learning_rate": 5.8953574060427415e-06,
      "loss": 1.1696,
      "num_input_tokens_seen": 13091808,
      "step": 2640,
      "train_runtime": 9125.2682,
      "train_tokens_per_second": 1434.677
    },
    {
      "epoch": 0.9714022988505747,
      "grad_norm": 0.35901394486427307,
      "learning_rate": 5.821665438467207e-06,
      "loss": 1.0964,
      "num_input_tokens_seen": 13096064,
      "step": 2641,
      "train_runtime": 9130.1924,
      "train_tokens_per_second": 1434.369
    },
    {
      "epoch": 0.9717701149425287,
      "grad_norm": 0.36465173959732056,
      "learning_rate": 5.747973470891673e-06,
      "loss": 1.1083,
      "num_input_tokens_seen": 13100192,
      "step": 2642,
      "train_runtime": 9132.9918,
      "train_tokens_per_second": 1434.381
    },
    {
      "epoch": 0.9721379310344828,
      "grad_norm": 0.37539973855018616,
      "learning_rate": 5.6742815033161385e-06,
      "loss": 1.3061,
      "num_input_tokens_seen": 13103952,
      "step": 2643,
      "train_runtime": 9135.65,
      "train_tokens_per_second": 1434.375
    },
    {
      "epoch": 0.9725057471264368,
      "grad_norm": 0.3692229986190796,
      "learning_rate": 5.600589535740604e-06,
      "loss": 1.1398,
      "num_input_tokens_seen": 13108592,
      "step": 2644,
      "train_runtime": 9138.7596,
      "train_tokens_per_second": 1434.395
    },
    {
      "epoch": 0.9728735632183908,
      "grad_norm": 0.3598920702934265,
      "learning_rate": 5.52689756816507e-06,
      "loss": 1.0703,
      "num_input_tokens_seen": 13112192,
      "step": 2645,
      "train_runtime": 9141.3593,
      "train_tokens_per_second": 1434.381
    },
    {
      "epoch": 0.9732413793103448,
      "grad_norm": 0.39512544870376587,
      "learning_rate": 5.4532056005895355e-06,
      "loss": 1.267,
      "num_input_tokens_seen": 13115952,
      "step": 2646,
      "train_runtime": 9144.1131,
      "train_tokens_per_second": 1434.36
    },
    {
      "epoch": 0.9736091954022988,
      "grad_norm": 0.3510258197784424,
      "learning_rate": 5.379513633014001e-06,
      "loss": 1.1143,
      "num_input_tokens_seen": 13119696,
      "step": 2647,
      "train_runtime": 9146.7728,
      "train_tokens_per_second": 1434.352
    },
    {
      "epoch": 0.9739770114942529,
      "grad_norm": 0.378836989402771,
      "learning_rate": 5.305821665438467e-06,
      "loss": 1.1973,
      "num_input_tokens_seen": 13125200,
      "step": 2648,
      "train_runtime": 9150.422,
      "train_tokens_per_second": 1434.382
    },
    {
      "epoch": 0.9743448275862069,
      "grad_norm": 0.3684450387954712,
      "learning_rate": 5.2321296978629325e-06,
      "loss": 1.1651,
      "num_input_tokens_seen": 13129376,
      "step": 2649,
      "train_runtime": 9153.419,
      "train_tokens_per_second": 1434.369
    },
    {
      "epoch": 0.9747126436781609,
      "grad_norm": 0.3807239234447479,
      "learning_rate": 5.158437730287399e-06,
      "loss": 1.0743,
      "num_input_tokens_seen": 13135184,
      "step": 2650,
      "train_runtime": 9157.2302,
      "train_tokens_per_second": 1434.406
    },
    {
      "epoch": 0.9750804597701149,
      "grad_norm": 0.3260813355445862,
      "learning_rate": 5.084745762711865e-06,
      "loss": 1.0861,
      "num_input_tokens_seen": 13139328,
      "step": 2651,
      "train_runtime": 9160.0983,
      "train_tokens_per_second": 1434.409
    },
    {
      "epoch": 0.975448275862069,
      "grad_norm": 0.35605546832084656,
      "learning_rate": 5.01105379513633e-06,
      "loss": 1.1935,
      "num_input_tokens_seen": 13142864,
      "step": 2652,
      "train_runtime": 9162.7131,
      "train_tokens_per_second": 1434.386
    },
    {
      "epoch": 0.975816091954023,
      "grad_norm": 0.4067990481853485,
      "learning_rate": 4.937361827560796e-06,
      "loss": 1.2467,
      "num_input_tokens_seen": 13149600,
      "step": 2653,
      "train_runtime": 9167.0399,
      "train_tokens_per_second": 1434.443
    },
    {
      "epoch": 0.976183908045977,
      "grad_norm": 0.3621925413608551,
      "learning_rate": 4.863669859985262e-06,
      "loss": 1.1553,
      "num_input_tokens_seen": 13152624,
      "step": 2654,
      "train_runtime": 9169.3626,
      "train_tokens_per_second": 1434.41
    },
    {
      "epoch": 0.976551724137931,
      "grad_norm": 0.39945676922798157,
      "learning_rate": 4.789977892409727e-06,
      "loss": 1.256,
      "num_input_tokens_seen": 13156864,
      "step": 2655,
      "train_runtime": 9172.2776,
      "train_tokens_per_second": 1434.416
    },
    {
      "epoch": 0.976919540229885,
      "grad_norm": 0.34752747416496277,
      "learning_rate": 4.716285924834193e-06,
      "loss": 1.2401,
      "num_input_tokens_seen": 13161104,
      "step": 2656,
      "train_runtime": 9175.1876,
      "train_tokens_per_second": 1434.423
    },
    {
      "epoch": 0.977287356321839,
      "grad_norm": 0.36894023418426514,
      "learning_rate": 4.642593957258659e-06,
      "loss": 1.1463,
      "num_input_tokens_seen": 13167216,
      "step": 2657,
      "train_runtime": 9179.0885,
      "train_tokens_per_second": 1434.48
    },
    {
      "epoch": 0.9776551724137931,
      "grad_norm": 0.3994079828262329,
      "learning_rate": 4.568901989683124e-06,
      "loss": 1.1976,
      "num_input_tokens_seen": 13172848,
      "step": 2658,
      "train_runtime": 9182.85,
      "train_tokens_per_second": 1434.505
    },
    {
      "epoch": 0.9780229885057471,
      "grad_norm": 0.3812773525714874,
      "learning_rate": 4.495210022107591e-06,
      "loss": 1.2223,
      "num_input_tokens_seen": 13177632,
      "step": 2659,
      "train_runtime": 9186.0748,
      "train_tokens_per_second": 1434.523
    },
    {
      "epoch": 0.9783908045977011,
      "grad_norm": 0.3517211973667145,
      "learning_rate": 4.4215180545320565e-06,
      "loss": 1.0889,
      "num_input_tokens_seen": 13185376,
      "step": 2660,
      "train_runtime": 9190.9199,
      "train_tokens_per_second": 1434.609
    },
    {
      "epoch": 0.9787586206896551,
      "grad_norm": 0.37851017713546753,
      "learning_rate": 4.347826086956522e-06,
      "loss": 1.0837,
      "num_input_tokens_seen": 13192432,
      "step": 2661,
      "train_runtime": 9195.2901,
      "train_tokens_per_second": 1434.694
    },
    {
      "epoch": 0.9791264367816092,
      "grad_norm": 0.3822534680366516,
      "learning_rate": 4.274134119380988e-06,
      "loss": 1.277,
      "num_input_tokens_seen": 13196880,
      "step": 2662,
      "train_runtime": 9198.3425,
      "train_tokens_per_second": 1434.702
    },
    {
      "epoch": 0.9794942528735632,
      "grad_norm": 0.3740699589252472,
      "learning_rate": 4.2004421518054535e-06,
      "loss": 1.2112,
      "num_input_tokens_seen": 13201440,
      "step": 2663,
      "train_runtime": 9201.3801,
      "train_tokens_per_second": 1434.724
    },
    {
      "epoch": 0.9798620689655172,
      "grad_norm": 0.40683114528656006,
      "learning_rate": 4.126750184229919e-06,
      "loss": 1.2945,
      "num_input_tokens_seen": 13206576,
      "step": 2664,
      "train_runtime": 9204.7476,
      "train_tokens_per_second": 1434.757
    },
    {
      "epoch": 0.9802298850574712,
      "grad_norm": 0.3855752944946289,
      "learning_rate": 4.053058216654385e-06,
      "loss": 1.1944,
      "num_input_tokens_seen": 13212512,
      "step": 2665,
      "train_runtime": 9208.5481,
      "train_tokens_per_second": 1434.809
    },
    {
      "epoch": 0.9805977011494252,
      "grad_norm": 0.3845732510089874,
      "learning_rate": 3.9793662490788505e-06,
      "loss": 1.1988,
      "num_input_tokens_seen": 13221056,
      "step": 2666,
      "train_runtime": 9213.7912,
      "train_tokens_per_second": 1434.92
    },
    {
      "epoch": 0.9809655172413793,
      "grad_norm": 0.38011282682418823,
      "learning_rate": 3.905674281503316e-06,
      "loss": 1.1812,
      "num_input_tokens_seen": 13224976,
      "step": 2667,
      "train_runtime": 9216.5361,
      "train_tokens_per_second": 1434.918
    },
    {
      "epoch": 0.9813333333333333,
      "grad_norm": 0.3793037235736847,
      "learning_rate": 3.831982313927783e-06,
      "loss": 1.2654,
      "num_input_tokens_seen": 13229072,
      "step": 2668,
      "train_runtime": 9219.3938,
      "train_tokens_per_second": 1434.918
    },
    {
      "epoch": 0.9817011494252874,
      "grad_norm": 0.3945206105709076,
      "learning_rate": 3.758290346352248e-06,
      "loss": 1.1835,
      "num_input_tokens_seen": 13233872,
      "step": 2669,
      "train_runtime": 9222.6265,
      "train_tokens_per_second": 1434.935
    },
    {
      "epoch": 0.9820689655172414,
      "grad_norm": 0.3916941285133362,
      "learning_rate": 3.6845983787767136e-06,
      "loss": 1.2922,
      "num_input_tokens_seen": 13237392,
      "step": 2670,
      "train_runtime": 9225.2451,
      "train_tokens_per_second": 1434.91
    },
    {
      "epoch": 0.9824367816091955,
      "grad_norm": 0.35781216621398926,
      "learning_rate": 3.6109064112011793e-06,
      "loss": 1.1003,
      "num_input_tokens_seen": 13241072,
      "step": 2671,
      "train_runtime": 9229.8816,
      "train_tokens_per_second": 1434.587
    },
    {
      "epoch": 0.9828045977011495,
      "grad_norm": 0.39518725872039795,
      "learning_rate": 3.537214443625645e-06,
      "loss": 1.1756,
      "num_input_tokens_seen": 13245280,
      "step": 2672,
      "train_runtime": 9232.7503,
      "train_tokens_per_second": 1434.597
    },
    {
      "epoch": 0.9831724137931035,
      "grad_norm": 0.3739321529865265,
      "learning_rate": 3.4635224760501106e-06,
      "loss": 1.3331,
      "num_input_tokens_seen": 13248752,
      "step": 2673,
      "train_runtime": 9235.2515,
      "train_tokens_per_second": 1434.585
    },
    {
      "epoch": 0.9835402298850575,
      "grad_norm": 0.3432806432247162,
      "learning_rate": 3.3898305084745763e-06,
      "loss": 1.0935,
      "num_input_tokens_seen": 13253216,
      "step": 2674,
      "train_runtime": 9238.3072,
      "train_tokens_per_second": 1434.594
    },
    {
      "epoch": 0.9839080459770115,
      "grad_norm": 0.3220776915550232,
      "learning_rate": 3.316138540899042e-06,
      "loss": 1.0416,
      "num_input_tokens_seen": 13263680,
      "step": 2675,
      "train_runtime": 9244.5786,
      "train_tokens_per_second": 1434.752
    },
    {
      "epoch": 0.9842758620689656,
      "grad_norm": 0.3478752374649048,
      "learning_rate": 3.2424465733235076e-06,
      "loss": 1.139,
      "num_input_tokens_seen": 13268816,
      "step": 2676,
      "train_runtime": 9247.9822,
      "train_tokens_per_second": 1434.78
    },
    {
      "epoch": 0.9846436781609196,
      "grad_norm": 0.37755367159843445,
      "learning_rate": 3.1687546057479733e-06,
      "loss": 1.0579,
      "num_input_tokens_seen": 13274720,
      "step": 2677,
      "train_runtime": 9251.7823,
      "train_tokens_per_second": 1434.828
    },
    {
      "epoch": 0.9850114942528736,
      "grad_norm": 0.35921499133110046,
      "learning_rate": 3.0950626381724394e-06,
      "loss": 1.2193,
      "num_input_tokens_seen": 13280816,
      "step": 2678,
      "train_runtime": 9255.7344,
      "train_tokens_per_second": 1434.874
    },
    {
      "epoch": 0.9853793103448276,
      "grad_norm": 0.36000990867614746,
      "learning_rate": 3.021370670596905e-06,
      "loss": 1.2805,
      "num_input_tokens_seen": 13284896,
      "step": 2679,
      "train_runtime": 9258.4904,
      "train_tokens_per_second": 1434.888
    },
    {
      "epoch": 0.9857471264367816,
      "grad_norm": 0.34659695625305176,
      "learning_rate": 2.9476787030213707e-06,
      "loss": 1.065,
      "num_input_tokens_seen": 13289152,
      "step": 2680,
      "train_runtime": 9261.3648,
      "train_tokens_per_second": 1434.902
    },
    {
      "epoch": 0.9861149425287357,
      "grad_norm": 0.36805298924446106,
      "learning_rate": 2.8739867354458364e-06,
      "loss": 1.2633,
      "num_input_tokens_seen": 13294432,
      "step": 2681,
      "train_runtime": 9264.8288,
      "train_tokens_per_second": 1434.936
    },
    {
      "epoch": 0.9864827586206897,
      "grad_norm": 0.3604376018047333,
      "learning_rate": 2.800294767870302e-06,
      "loss": 1.1101,
      "num_input_tokens_seen": 13297696,
      "step": 2682,
      "train_runtime": 9267.2003,
      "train_tokens_per_second": 1434.921
    },
    {
      "epoch": 0.9868505747126437,
      "grad_norm": 0.4133100211620331,
      "learning_rate": 2.7266028002947677e-06,
      "loss": 1.268,
      "num_input_tokens_seen": 13301488,
      "step": 2683,
      "train_runtime": 9269.9022,
      "train_tokens_per_second": 1434.911
    },
    {
      "epoch": 0.9872183908045977,
      "grad_norm": 0.36944735050201416,
      "learning_rate": 2.6529108327192334e-06,
      "loss": 1.1549,
      "num_input_tokens_seen": 13308800,
      "step": 2684,
      "train_runtime": 9274.5558,
      "train_tokens_per_second": 1434.98
    },
    {
      "epoch": 0.9875862068965517,
      "grad_norm": 0.36020931601524353,
      "learning_rate": 2.5792188651436995e-06,
      "loss": 0.9927,
      "num_input_tokens_seen": 13311904,
      "step": 2685,
      "train_runtime": 9276.8974,
      "train_tokens_per_second": 1434.952
    },
    {
      "epoch": 0.9879540229885058,
      "grad_norm": 0.3761769235134125,
      "learning_rate": 2.505526897568165e-06,
      "loss": 1.1888,
      "num_input_tokens_seen": 13315360,
      "step": 2686,
      "train_runtime": 9279.4447,
      "train_tokens_per_second": 1434.931
    },
    {
      "epoch": 0.9883218390804598,
      "grad_norm": 0.3514864146709442,
      "learning_rate": 2.431834929992631e-06,
      "loss": 1.1336,
      "num_input_tokens_seen": 13319872,
      "step": 2687,
      "train_runtime": 9282.5809,
      "train_tokens_per_second": 1434.932
    },
    {
      "epoch": 0.9886896551724138,
      "grad_norm": 0.4007459282875061,
      "learning_rate": 2.3581429624170965e-06,
      "loss": 1.1083,
      "num_input_tokens_seen": 13325088,
      "step": 2688,
      "train_runtime": 9286.0272,
      "train_tokens_per_second": 1434.961
    },
    {
      "epoch": 0.9890574712643678,
      "grad_norm": 0.4084776043891907,
      "learning_rate": 2.284450994841562e-06,
      "loss": 1.1634,
      "num_input_tokens_seen": 13329920,
      "step": 2689,
      "train_runtime": 9289.264,
      "train_tokens_per_second": 1434.981
    },
    {
      "epoch": 0.9894252873563218,
      "grad_norm": 0.39426979422569275,
      "learning_rate": 2.2107590272660283e-06,
      "loss": 1.1331,
      "num_input_tokens_seen": 13333968,
      "step": 2690,
      "train_runtime": 9291.9757,
      "train_tokens_per_second": 1434.998
    },
    {
      "epoch": 0.9897931034482759,
      "grad_norm": 0.3781372010707855,
      "learning_rate": 2.137067059690494e-06,
      "loss": 1.161,
      "num_input_tokens_seen": 13337968,
      "step": 2691,
      "train_runtime": 9294.7259,
      "train_tokens_per_second": 1435.004
    },
    {
      "epoch": 0.9901609195402299,
      "grad_norm": 0.3930552303791046,
      "learning_rate": 2.0633750921149596e-06,
      "loss": 1.2381,
      "num_input_tokens_seen": 13341200,
      "step": 2692,
      "train_runtime": 9297.1622,
      "train_tokens_per_second": 1434.976
    },
    {
      "epoch": 0.9905287356321839,
      "grad_norm": 0.3918300271034241,
      "learning_rate": 1.9896831245394253e-06,
      "loss": 1.1624,
      "num_input_tokens_seen": 13346416,
      "step": 2693,
      "train_runtime": 9300.6979,
      "train_tokens_per_second": 1434.991
    },
    {
      "epoch": 0.9908965517241379,
      "grad_norm": 0.4089981019496918,
      "learning_rate": 1.9159911569638914e-06,
      "loss": 1.3183,
      "num_input_tokens_seen": 13351536,
      "step": 2694,
      "train_runtime": 9304.1306,
      "train_tokens_per_second": 1435.012
    },
    {
      "epoch": 0.991264367816092,
      "grad_norm": 0.3748939633369446,
      "learning_rate": 1.8422991893883568e-06,
      "loss": 1.2059,
      "num_input_tokens_seen": 13356304,
      "step": 2695,
      "train_runtime": 9307.3,
      "train_tokens_per_second": 1435.035
    },
    {
      "epoch": 0.991632183908046,
      "grad_norm": 0.3489076495170593,
      "learning_rate": 1.7686072218128225e-06,
      "loss": 1.1273,
      "num_input_tokens_seen": 13360720,
      "step": 2696,
      "train_runtime": 9310.3048,
      "train_tokens_per_second": 1435.046
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.3759075999259949,
      "learning_rate": 1.6949152542372882e-06,
      "loss": 1.2313,
      "num_input_tokens_seen": 13364704,
      "step": 2697,
      "train_runtime": 9312.994,
      "train_tokens_per_second": 1435.06
    },
    {
      "epoch": 0.992367816091954,
      "grad_norm": 0.3778208792209625,
      "learning_rate": 1.6212232866617538e-06,
      "loss": 1.2682,
      "num_input_tokens_seen": 13368560,
      "step": 2698,
      "train_runtime": 9315.6845,
      "train_tokens_per_second": 1435.059
    },
    {
      "epoch": 0.992735632183908,
      "grad_norm": 0.39586445689201355,
      "learning_rate": 1.5475313190862197e-06,
      "loss": 1.1028,
      "num_input_tokens_seen": 13373104,
      "step": 2699,
      "train_runtime": 9318.7639,
      "train_tokens_per_second": 1435.073
    },
    {
      "epoch": 0.993103448275862,
      "grad_norm": 0.34999415278434753,
      "learning_rate": 1.4738393515106854e-06,
      "loss": 1.1526,
      "num_input_tokens_seen": 13377872,
      "step": 2700,
      "train_runtime": 9322.0195,
      "train_tokens_per_second": 1435.083
    },
    {
      "epoch": 0.9934712643678161,
      "grad_norm": 0.39181894063949585,
      "learning_rate": 1.400147383935151e-06,
      "loss": 1.1644,
      "num_input_tokens_seen": 13382752,
      "step": 2701,
      "train_runtime": 9327.35,
      "train_tokens_per_second": 1434.786
    },
    {
      "epoch": 0.9938390804597701,
      "grad_norm": 0.35605180263519287,
      "learning_rate": 1.3264554163596167e-06,
      "loss": 1.123,
      "num_input_tokens_seen": 13387296,
      "step": 2702,
      "train_runtime": 9330.5136,
      "train_tokens_per_second": 1434.787
    },
    {
      "epoch": 0.9942068965517241,
      "grad_norm": 0.389503538608551,
      "learning_rate": 1.2527634487840826e-06,
      "loss": 1.2023,
      "num_input_tokens_seen": 13391424,
      "step": 2703,
      "train_runtime": 9333.3876,
      "train_tokens_per_second": 1434.787
    },
    {
      "epoch": 0.9945747126436781,
      "grad_norm": 0.356380432844162,
      "learning_rate": 1.1790714812085483e-06,
      "loss": 1.2694,
      "num_input_tokens_seen": 13395712,
      "step": 2704,
      "train_runtime": 9336.2019,
      "train_tokens_per_second": 1434.814
    },
    {
      "epoch": 0.9949425287356322,
      "grad_norm": 0.38404738903045654,
      "learning_rate": 1.1053795136330141e-06,
      "loss": 1.1005,
      "num_input_tokens_seen": 13404560,
      "step": 2705,
      "train_runtime": 9341.5956,
      "train_tokens_per_second": 1434.933
    },
    {
      "epoch": 0.9953103448275862,
      "grad_norm": 0.39288413524627686,
      "learning_rate": 1.0316875460574798e-06,
      "loss": 1.235,
      "num_input_tokens_seen": 13409056,
      "step": 2706,
      "train_runtime": 9344.6909,
      "train_tokens_per_second": 1434.938
    },
    {
      "epoch": 0.9956781609195402,
      "grad_norm": 0.35604438185691833,
      "learning_rate": 9.579955784819457e-07,
      "loss": 1.2303,
      "num_input_tokens_seen": 13415040,
      "step": 2707,
      "train_runtime": 9348.606,
      "train_tokens_per_second": 1434.978
    },
    {
      "epoch": 0.9960459770114942,
      "grad_norm": 0.3633836507797241,
      "learning_rate": 8.843036109064112e-07,
      "loss": 1.1844,
      "num_input_tokens_seen": 13420800,
      "step": 2708,
      "train_runtime": 9352.438,
      "train_tokens_per_second": 1435.006
    },
    {
      "epoch": 0.9964137931034482,
      "grad_norm": 0.335910826921463,
      "learning_rate": 8.106116433308769e-07,
      "loss": 1.0622,
      "num_input_tokens_seen": 13425072,
      "step": 2709,
      "train_runtime": 9355.3056,
      "train_tokens_per_second": 1435.022
    },
    {
      "epoch": 0.9967816091954023,
      "grad_norm": 0.3533064126968384,
      "learning_rate": 7.369196757553427e-07,
      "loss": 1.2354,
      "num_input_tokens_seen": 13428880,
      "step": 2710,
      "train_runtime": 9358.031,
      "train_tokens_per_second": 1435.011
    },
    {
      "epoch": 0.9971494252873563,
      "grad_norm": 0.3506322503089905,
      "learning_rate": 6.632277081798084e-07,
      "loss": 1.2809,
      "num_input_tokens_seen": 13434128,
      "step": 2711,
      "train_runtime": 9361.4711,
      "train_tokens_per_second": 1435.045
    },
    {
      "epoch": 0.9975172413793103,
      "grad_norm": 0.3724497854709625,
      "learning_rate": 5.895357406042741e-07,
      "loss": 1.2338,
      "num_input_tokens_seen": 13439616,
      "step": 2712,
      "train_runtime": 9365.1353,
      "train_tokens_per_second": 1435.069
    },
    {
      "epoch": 0.9978850574712643,
      "grad_norm": 0.35516542196273804,
      "learning_rate": 5.158437730287399e-07,
      "loss": 1.1765,
      "num_input_tokens_seen": 13445984,
      "step": 2713,
      "train_runtime": 9369.1385,
      "train_tokens_per_second": 1435.136
    },
    {
      "epoch": 0.9982528735632183,
      "grad_norm": 0.3694939613342285,
      "learning_rate": 4.421518054532056e-07,
      "loss": 1.2878,
      "num_input_tokens_seen": 13450576,
      "step": 2714,
      "train_runtime": 9372.1452,
      "train_tokens_per_second": 1435.165
    },
    {
      "epoch": 0.9986206896551724,
      "grad_norm": 0.3756771385669708,
      "learning_rate": 3.6845983787767134e-07,
      "loss": 1.1811,
      "num_input_tokens_seen": 13454272,
      "step": 2715,
      "train_runtime": 9374.8052,
      "train_tokens_per_second": 1435.152
    },
    {
      "epoch": 0.9989885057471264,
      "grad_norm": 0.3870641887187958,
      "learning_rate": 2.9476787030213706e-07,
      "loss": 1.0927,
      "num_input_tokens_seen": 13460928,
      "step": 2716,
      "train_runtime": 9379.0286,
      "train_tokens_per_second": 1435.216
    },
    {
      "epoch": 0.9993563218390804,
      "grad_norm": 0.3609115183353424,
      "learning_rate": 2.210759027266028e-07,
      "loss": 1.2228,
      "num_input_tokens_seen": 13465152,
      "step": 2717,
      "train_runtime": 9381.9636,
      "train_tokens_per_second": 1435.217
    },
    {
      "epoch": 0.9997241379310344,
      "grad_norm": 0.36089998483657837,
      "learning_rate": 1.4738393515106853e-07,
      "loss": 1.1128,
      "num_input_tokens_seen": 13468816,
      "step": 2718,
      "train_runtime": 9384.5989,
      "train_tokens_per_second": 1435.204
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.4428732395172119,
      "learning_rate": 7.369196757553427e-08,
      "loss": 1.1919,
      "num_input_tokens_seen": 13472656,
      "step": 2719,
      "train_runtime": 9387.199,
      "train_tokens_per_second": 1435.216
    }
  ],
  "logging_steps": 1,
  "max_steps": 2719,
  "num_input_tokens_seen": 13472656,
  "num_train_epochs": 1,
  "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": 2.6071728602210304e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}