{
  "best_global_step": 40000,
  "best_metric": 0.198301300406456,
  "best_model_checkpoint": "/media/user/Expansion1/multilingual-e5-small-aligned-v2-fineweb2hq-vs-c4-classifier-run2/checkpoint-40000",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 120000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0125,
      "grad_norm": 5.874658584594727,
      "learning_rate": 4.979208333333333e-05,
      "loss": 0.3436,
      "num_input_tokens_seen": 512000,
      "step": 500
    },
    {
      "epoch": 0.025,
      "grad_norm": 0.6375065445899963,
      "learning_rate": 4.958375e-05,
      "loss": 0.281,
      "num_input_tokens_seen": 1024000,
      "step": 1000
    },
    {
      "epoch": 0.0375,
      "grad_norm": 0.48116546869277954,
      "learning_rate": 4.937541666666667e-05,
      "loss": 0.2595,
      "num_input_tokens_seen": 1536000,
      "step": 1500
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.6305193305015564,
      "learning_rate": 4.9167083333333336e-05,
      "loss": 0.2466,
      "num_input_tokens_seen": 2048000,
      "step": 2000
    },
    {
      "epoch": 0.0625,
      "grad_norm": 3.560100555419922,
      "learning_rate": 4.8958750000000006e-05,
      "loss": 0.2283,
      "num_input_tokens_seen": 2560000,
      "step": 2500
    },
    {
      "epoch": 0.075,
      "grad_norm": 43.007347106933594,
      "learning_rate": 4.875041666666667e-05,
      "loss": 0.2106,
      "num_input_tokens_seen": 3072000,
      "step": 3000
    },
    {
      "epoch": 0.0875,
      "grad_norm": 27.75348472595215,
      "learning_rate": 4.854208333333334e-05,
      "loss": 0.2198,
      "num_input_tokens_seen": 3584000,
      "step": 3500
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.11089582741260529,
      "learning_rate": 4.833375e-05,
      "loss": 0.2145,
      "num_input_tokens_seen": 4096000,
      "step": 4000
    },
    {
      "epoch": 0.1125,
      "grad_norm": 0.08869020640850067,
      "learning_rate": 4.812541666666667e-05,
      "loss": 0.2033,
      "num_input_tokens_seen": 4608000,
      "step": 4500
    },
    {
      "epoch": 0.125,
      "grad_norm": 22.97330665588379,
      "learning_rate": 4.791708333333333e-05,
      "loss": 0.2007,
      "num_input_tokens_seen": 5120000,
      "step": 5000
    },
    {
      "epoch": 0.1375,
      "grad_norm": 0.12722039222717285,
      "learning_rate": 4.770875e-05,
      "loss": 0.1893,
      "num_input_tokens_seen": 5632000,
      "step": 5500
    },
    {
      "epoch": 0.15,
      "grad_norm": 24.923276901245117,
      "learning_rate": 4.750041666666667e-05,
      "loss": 0.2073,
      "num_input_tokens_seen": 6144000,
      "step": 6000
    },
    {
      "epoch": 0.1625,
      "grad_norm": 0.17447435855865479,
      "learning_rate": 4.729208333333334e-05,
      "loss": 0.2024,
      "num_input_tokens_seen": 6656000,
      "step": 6500
    },
    {
      "epoch": 0.175,
      "grad_norm": 5.2057671546936035,
      "learning_rate": 4.708375e-05,
      "loss": 0.2168,
      "num_input_tokens_seen": 7168000,
      "step": 7000
    },
    {
      "epoch": 0.1875,
      "grad_norm": 0.35154542326927185,
      "learning_rate": 4.687541666666667e-05,
      "loss": 0.2002,
      "num_input_tokens_seen": 7680000,
      "step": 7500
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.17855411767959595,
      "learning_rate": 4.6667083333333336e-05,
      "loss": 0.1863,
      "num_input_tokens_seen": 8192000,
      "step": 8000
    },
    {
      "epoch": 0.2125,
      "grad_norm": 0.26302626729011536,
      "learning_rate": 4.645875e-05,
      "loss": 0.2023,
      "num_input_tokens_seen": 8704000,
      "step": 8500
    },
    {
      "epoch": 0.225,
      "grad_norm": 1.0355998277664185,
      "learning_rate": 4.625041666666667e-05,
      "loss": 0.1996,
      "num_input_tokens_seen": 9216000,
      "step": 9000
    },
    {
      "epoch": 0.2375,
      "grad_norm": 3.5437514781951904,
      "learning_rate": 4.6042083333333334e-05,
      "loss": 0.1928,
      "num_input_tokens_seen": 9728000,
      "step": 9500
    },
    {
      "epoch": 0.25,
      "grad_norm": 18.394441604614258,
      "learning_rate": 4.5833750000000005e-05,
      "loss": 0.1968,
      "num_input_tokens_seen": 10240000,
      "step": 10000
    },
    {
      "epoch": 0.2625,
      "grad_norm": 2.9275922775268555,
      "learning_rate": 4.562541666666667e-05,
      "loss": 0.2081,
      "num_input_tokens_seen": 10752000,
      "step": 10500
    },
    {
      "epoch": 0.275,
      "grad_norm": 3.922402858734131,
      "learning_rate": 4.541708333333334e-05,
      "loss": 0.1769,
      "num_input_tokens_seen": 11264000,
      "step": 11000
    },
    {
      "epoch": 0.2875,
      "grad_norm": 0.25219565629959106,
      "learning_rate": 4.5208749999999996e-05,
      "loss": 0.1722,
      "num_input_tokens_seen": 11776000,
      "step": 11500
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.10917416214942932,
      "learning_rate": 4.500041666666667e-05,
      "loss": 0.1638,
      "num_input_tokens_seen": 12288000,
      "step": 12000
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.1928466558456421,
      "learning_rate": 4.479208333333334e-05,
      "loss": 0.1826,
      "num_input_tokens_seen": 12800000,
      "step": 12500
    },
    {
      "epoch": 0.325,
      "grad_norm": 1.1688095331192017,
      "learning_rate": 4.458375e-05,
      "loss": 0.186,
      "num_input_tokens_seen": 13312000,
      "step": 13000
    },
    {
      "epoch": 0.3375,
      "grad_norm": 0.07879551500082016,
      "learning_rate": 4.437541666666667e-05,
      "loss": 0.1638,
      "num_input_tokens_seen": 13824000,
      "step": 13500
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.294481635093689,
      "learning_rate": 4.4167083333333336e-05,
      "loss": 0.1695,
      "num_input_tokens_seen": 14336000,
      "step": 14000
    },
    {
      "epoch": 0.3625,
      "grad_norm": 21.836345672607422,
      "learning_rate": 4.395875000000001e-05,
      "loss": 0.1713,
      "num_input_tokens_seen": 14848000,
      "step": 14500
    },
    {
      "epoch": 0.375,
      "grad_norm": 0.1329026073217392,
      "learning_rate": 4.375041666666667e-05,
      "loss": 0.178,
      "num_input_tokens_seen": 15360000,
      "step": 15000
    },
    {
      "epoch": 0.3875,
      "grad_norm": 2.196962594985962,
      "learning_rate": 4.3542083333333334e-05,
      "loss": 0.1616,
      "num_input_tokens_seen": 15872000,
      "step": 15500
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.06707575172185898,
      "learning_rate": 4.333375e-05,
      "loss": 0.1563,
      "num_input_tokens_seen": 16384000,
      "step": 16000
    },
    {
      "epoch": 0.4125,
      "grad_norm": 4.03335428237915,
      "learning_rate": 4.312541666666667e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 16896000,
      "step": 16500
    },
    {
      "epoch": 0.425,
      "grad_norm": 0.7723323702812195,
      "learning_rate": 4.291708333333333e-05,
      "loss": 0.1695,
      "num_input_tokens_seen": 17408000,
      "step": 17000
    },
    {
      "epoch": 0.4375,
      "grad_norm": 0.14005227386951447,
      "learning_rate": 4.270875e-05,
      "loss": 0.1771,
      "num_input_tokens_seen": 17920000,
      "step": 17500
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.22879253327846527,
      "learning_rate": 4.250041666666667e-05,
      "loss": 0.1652,
      "num_input_tokens_seen": 18432000,
      "step": 18000
    },
    {
      "epoch": 0.4625,
      "grad_norm": 41.407352447509766,
      "learning_rate": 4.229208333333334e-05,
      "loss": 0.1697,
      "num_input_tokens_seen": 18944000,
      "step": 18500
    },
    {
      "epoch": 0.475,
      "grad_norm": 0.2552178204059601,
      "learning_rate": 4.208375e-05,
      "loss": 0.169,
      "num_input_tokens_seen": 19456000,
      "step": 19000
    },
    {
      "epoch": 0.4875,
      "grad_norm": 4.208723545074463,
      "learning_rate": 4.1875416666666666e-05,
      "loss": 0.1887,
      "num_input_tokens_seen": 19968000,
      "step": 19500
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.6019179821014404,
      "learning_rate": 4.1667083333333336e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 20480000,
      "step": 20000
    },
    {
      "epoch": 0.5125,
      "grad_norm": 11.652756690979004,
      "learning_rate": 4.145875e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 20992000,
      "step": 20500
    },
    {
      "epoch": 0.525,
      "grad_norm": 3.790767192840576,
      "learning_rate": 4.125041666666667e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 21504000,
      "step": 21000
    },
    {
      "epoch": 0.5375,
      "grad_norm": 0.0643579512834549,
      "learning_rate": 4.1042083333333335e-05,
      "loss": 0.156,
      "num_input_tokens_seen": 22016000,
      "step": 21500
    },
    {
      "epoch": 0.55,
      "grad_norm": 6.347803115844727,
      "learning_rate": 4.0833750000000005e-05,
      "loss": 0.1743,
      "num_input_tokens_seen": 22528000,
      "step": 22000
    },
    {
      "epoch": 0.5625,
      "grad_norm": 0.45395129919052124,
      "learning_rate": 4.062541666666667e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 23040000,
      "step": 22500
    },
    {
      "epoch": 0.575,
      "grad_norm": 4.8587470054626465,
      "learning_rate": 4.041708333333333e-05,
      "loss": 0.156,
      "num_input_tokens_seen": 23552000,
      "step": 23000
    },
    {
      "epoch": 0.5875,
      "grad_norm": 0.8771836757659912,
      "learning_rate": 4.0208750000000004e-05,
      "loss": 0.1473,
      "num_input_tokens_seen": 24064000,
      "step": 23500
    },
    {
      "epoch": 0.6,
      "grad_norm": 7.632213592529297,
      "learning_rate": 4.000041666666667e-05,
      "loss": 0.1523,
      "num_input_tokens_seen": 24576000,
      "step": 24000
    },
    {
      "epoch": 0.6125,
      "grad_norm": 4.013466835021973,
      "learning_rate": 3.979208333333334e-05,
      "loss": 0.1646,
      "num_input_tokens_seen": 25088000,
      "step": 24500
    },
    {
      "epoch": 0.625,
      "grad_norm": 0.08471076935529709,
      "learning_rate": 3.958375e-05,
      "loss": 0.1653,
      "num_input_tokens_seen": 25600000,
      "step": 25000
    },
    {
      "epoch": 0.6375,
      "grad_norm": 4.263392448425293,
      "learning_rate": 3.937541666666667e-05,
      "loss": 0.1513,
      "num_input_tokens_seen": 26112000,
      "step": 25500
    },
    {
      "epoch": 0.65,
      "grad_norm": 17.002439498901367,
      "learning_rate": 3.9167083333333336e-05,
      "loss": 0.144,
      "num_input_tokens_seen": 26624000,
      "step": 26000
    },
    {
      "epoch": 0.6625,
      "grad_norm": 0.056964244693517685,
      "learning_rate": 3.895875e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 27136000,
      "step": 26500
    },
    {
      "epoch": 0.675,
      "grad_norm": 0.4403606653213501,
      "learning_rate": 3.8750416666666664e-05,
      "loss": 0.1501,
      "num_input_tokens_seen": 27648000,
      "step": 27000
    },
    {
      "epoch": 0.6875,
      "grad_norm": 0.11431742459535599,
      "learning_rate": 3.8542083333333335e-05,
      "loss": 0.1597,
      "num_input_tokens_seen": 28160000,
      "step": 27500
    },
    {
      "epoch": 0.7,
      "grad_norm": 7.810299396514893,
      "learning_rate": 3.833375e-05,
      "loss": 0.1437,
      "num_input_tokens_seen": 28672000,
      "step": 28000
    },
    {
      "epoch": 0.7125,
      "grad_norm": 5.501016616821289,
      "learning_rate": 3.812541666666667e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 29184000,
      "step": 28500
    },
    {
      "epoch": 0.725,
      "grad_norm": 0.20471607148647308,
      "learning_rate": 3.791708333333333e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 29696000,
      "step": 29000
    },
    {
      "epoch": 0.7375,
      "grad_norm": 4.6759748458862305,
      "learning_rate": 3.7708750000000004e-05,
      "loss": 0.1431,
      "num_input_tokens_seen": 30208000,
      "step": 29500
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.12430200725793839,
      "learning_rate": 3.7500416666666674e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 30720000,
      "step": 30000
    },
    {
      "epoch": 0.7625,
      "grad_norm": 0.19309231638908386,
      "learning_rate": 3.729208333333333e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 31232000,
      "step": 30500
    },
    {
      "epoch": 0.775,
      "grad_norm": 0.18272100389003754,
      "learning_rate": 3.708375e-05,
      "loss": 0.1504,
      "num_input_tokens_seen": 31744000,
      "step": 31000
    },
    {
      "epoch": 0.7875,
      "grad_norm": 0.08077079057693481,
      "learning_rate": 3.6875416666666666e-05,
      "loss": 0.13,
      "num_input_tokens_seen": 32256000,
      "step": 31500
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3213728368282318,
      "learning_rate": 3.666708333333334e-05,
      "loss": 0.1406,
      "num_input_tokens_seen": 32768000,
      "step": 32000
    },
    {
      "epoch": 0.8125,
      "grad_norm": 2.6233737468719482,
      "learning_rate": 3.645875e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 33280000,
      "step": 32500
    },
    {
      "epoch": 0.825,
      "grad_norm": 0.1083766371011734,
      "learning_rate": 3.625041666666667e-05,
      "loss": 0.1361,
      "num_input_tokens_seen": 33792000,
      "step": 33000
    },
    {
      "epoch": 0.8375,
      "grad_norm": 0.08282948285341263,
      "learning_rate": 3.6042083333333335e-05,
      "loss": 0.1483,
      "num_input_tokens_seen": 34304000,
      "step": 33500
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.10612857341766357,
      "learning_rate": 3.583375e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 34816000,
      "step": 34000
    },
    {
      "epoch": 0.8625,
      "grad_norm": 1.134366750717163,
      "learning_rate": 3.562541666666667e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 35328000,
      "step": 34500
    },
    {
      "epoch": 0.875,
      "grad_norm": 32.71923065185547,
      "learning_rate": 3.5417083333333333e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 35840000,
      "step": 35000
    },
    {
      "epoch": 0.8875,
      "grad_norm": 0.14211738109588623,
      "learning_rate": 3.5208750000000004e-05,
      "loss": 0.1543,
      "num_input_tokens_seen": 36352000,
      "step": 35500
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.04908756539225578,
      "learning_rate": 3.500041666666667e-05,
      "loss": 0.1229,
      "num_input_tokens_seen": 36864000,
      "step": 36000
    },
    {
      "epoch": 0.9125,
      "grad_norm": 0.47926560044288635,
      "learning_rate": 3.479208333333334e-05,
      "loss": 0.1251,
      "num_input_tokens_seen": 37376000,
      "step": 36500
    },
    {
      "epoch": 0.925,
      "grad_norm": 1.389073371887207,
      "learning_rate": 3.458375e-05,
      "loss": 0.1718,
      "num_input_tokens_seen": 37888000,
      "step": 37000
    },
    {
      "epoch": 0.9375,
      "grad_norm": 0.14962467551231384,
      "learning_rate": 3.437541666666667e-05,
      "loss": 0.1579,
      "num_input_tokens_seen": 38400000,
      "step": 37500
    },
    {
      "epoch": 0.95,
      "grad_norm": 3.8970487117767334,
      "learning_rate": 3.416708333333333e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 38912000,
      "step": 38000
    },
    {
      "epoch": 0.9625,
      "grad_norm": 0.1765402853488922,
      "learning_rate": 3.395875e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 39424000,
      "step": 38500
    },
    {
      "epoch": 0.975,
      "grad_norm": 0.1407538652420044,
      "learning_rate": 3.3750416666666665e-05,
      "loss": 0.1336,
      "num_input_tokens_seen": 39936000,
      "step": 39000
    },
    {
      "epoch": 0.9875,
      "grad_norm": 8.721837997436523,
      "learning_rate": 3.3542083333333335e-05,
      "loss": 0.1268,
      "num_input_tokens_seen": 40448000,
      "step": 39500
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.15711814165115356,
      "learning_rate": 3.333375e-05,
      "loss": 0.1387,
      "num_input_tokens_seen": 40960000,
      "step": 40000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.951525,
      "eval_combined_score": 1.3493705490187269,
      "eval_loss": 0.198301300406456,
      "eval_runtime": 39.36,
      "eval_samples_per_second": 2032.519,
      "eval_steps_per_second": 254.065,
      "num_input_tokens_seen": 40960000,
      "step": 40000
    },
    {
      "epoch": 1.0125,
      "grad_norm": 0.02203565090894699,
      "learning_rate": 3.312541666666667e-05,
      "loss": 0.1092,
      "num_input_tokens_seen": 41472000,
      "step": 40500
    },
    {
      "epoch": 1.025,
      "grad_norm": 328.5529479980469,
      "learning_rate": 3.291708333333334e-05,
      "loss": 0.0844,
      "num_input_tokens_seen": 41984000,
      "step": 41000
    },
    {
      "epoch": 1.0375,
      "grad_norm": 0.11391662806272507,
      "learning_rate": 3.270875e-05,
      "loss": 0.1115,
      "num_input_tokens_seen": 42496000,
      "step": 41500
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.07919144630432129,
      "learning_rate": 3.250041666666667e-05,
      "loss": 0.0878,
      "num_input_tokens_seen": 43008000,
      "step": 42000
    },
    {
      "epoch": 1.0625,
      "grad_norm": 0.029733452945947647,
      "learning_rate": 3.229208333333333e-05,
      "loss": 0.0914,
      "num_input_tokens_seen": 43520000,
      "step": 42500
    },
    {
      "epoch": 1.075,
      "grad_norm": 0.277174711227417,
      "learning_rate": 3.208375e-05,
      "loss": 0.1028,
      "num_input_tokens_seen": 44032000,
      "step": 43000
    },
    {
      "epoch": 1.0875,
      "grad_norm": 0.11592718958854675,
      "learning_rate": 3.1875416666666666e-05,
      "loss": 0.1032,
      "num_input_tokens_seen": 44544000,
      "step": 43500
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.2565874755382538,
      "learning_rate": 3.166708333333334e-05,
      "loss": 0.0912,
      "num_input_tokens_seen": 45056000,
      "step": 44000
    },
    {
      "epoch": 1.1125,
      "grad_norm": 0.03557795658707619,
      "learning_rate": 3.145875e-05,
      "loss": 0.0882,
      "num_input_tokens_seen": 45568000,
      "step": 44500
    },
    {
      "epoch": 1.125,
      "grad_norm": 4.72824764251709,
      "learning_rate": 3.125041666666667e-05,
      "loss": 0.0918,
      "num_input_tokens_seen": 46080000,
      "step": 45000
    },
    {
      "epoch": 1.1375,
      "grad_norm": 0.10075237601995468,
      "learning_rate": 3.1042083333333335e-05,
      "loss": 0.1117,
      "num_input_tokens_seen": 46592000,
      "step": 45500
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.07608671486377716,
      "learning_rate": 3.083375e-05,
      "loss": 0.1383,
      "num_input_tokens_seen": 47104000,
      "step": 46000
    },
    {
      "epoch": 1.1625,
      "grad_norm": 14.187911033630371,
      "learning_rate": 3.062541666666667e-05,
      "loss": 0.1062,
      "num_input_tokens_seen": 47616000,
      "step": 46500
    },
    {
      "epoch": 1.175,
      "grad_norm": 0.0889461487531662,
      "learning_rate": 3.0417083333333334e-05,
      "loss": 0.1255,
      "num_input_tokens_seen": 48128000,
      "step": 47000
    },
    {
      "epoch": 1.1875,
      "grad_norm": 0.10722101479768753,
      "learning_rate": 3.020875e-05,
      "loss": 0.0987,
      "num_input_tokens_seen": 48640000,
      "step": 47500
    },
    {
      "epoch": 1.2,
      "grad_norm": 42.01722717285156,
      "learning_rate": 3.000041666666667e-05,
      "loss": 0.1047,
      "num_input_tokens_seen": 49152000,
      "step": 48000
    },
    {
      "epoch": 1.2125,
      "grad_norm": 0.08817047625780106,
      "learning_rate": 2.9792083333333336e-05,
      "loss": 0.0931,
      "num_input_tokens_seen": 49664000,
      "step": 48500
    },
    {
      "epoch": 1.225,
      "grad_norm": 0.058988332748413086,
      "learning_rate": 2.958375e-05,
      "loss": 0.1048,
      "num_input_tokens_seen": 50176000,
      "step": 49000
    },
    {
      "epoch": 1.2375,
      "grad_norm": 0.025500474497675896,
      "learning_rate": 2.9375416666666667e-05,
      "loss": 0.0845,
      "num_input_tokens_seen": 50688000,
      "step": 49500
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.07621040940284729,
      "learning_rate": 2.9167083333333334e-05,
      "loss": 0.0893,
      "num_input_tokens_seen": 51200000,
      "step": 50000
    },
    {
      "epoch": 1.2625,
      "grad_norm": 0.14059029519557953,
      "learning_rate": 2.895875e-05,
      "loss": 0.0911,
      "num_input_tokens_seen": 51712000,
      "step": 50500
    },
    {
      "epoch": 1.275,
      "grad_norm": 0.04900716617703438,
      "learning_rate": 2.875041666666667e-05,
      "loss": 0.0984,
      "num_input_tokens_seen": 52224000,
      "step": 51000
    },
    {
      "epoch": 1.2875,
      "grad_norm": 0.0568259134888649,
      "learning_rate": 2.8542083333333336e-05,
      "loss": 0.0995,
      "num_input_tokens_seen": 52736000,
      "step": 51500
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.052453652024269104,
      "learning_rate": 2.8333750000000003e-05,
      "loss": 0.0875,
      "num_input_tokens_seen": 53248000,
      "step": 52000
    },
    {
      "epoch": 1.3125,
      "grad_norm": 0.05766982212662697,
      "learning_rate": 2.812541666666667e-05,
      "loss": 0.0772,
      "num_input_tokens_seen": 53760000,
      "step": 52500
    },
    {
      "epoch": 1.325,
      "grad_norm": 0.06809753179550171,
      "learning_rate": 2.7917083333333334e-05,
      "loss": 0.0849,
      "num_input_tokens_seen": 54272000,
      "step": 53000
    },
    {
      "epoch": 1.3375,
      "grad_norm": 0.5180730223655701,
      "learning_rate": 2.770875e-05,
      "loss": 0.089,
      "num_input_tokens_seen": 54784000,
      "step": 53500
    },
    {
      "epoch": 1.35,
      "grad_norm": 12.439111709594727,
      "learning_rate": 2.750041666666667e-05,
      "loss": 0.0927,
      "num_input_tokens_seen": 55296000,
      "step": 54000
    },
    {
      "epoch": 1.3625,
      "grad_norm": 0.12473966181278229,
      "learning_rate": 2.7292083333333336e-05,
      "loss": 0.1109,
      "num_input_tokens_seen": 55808000,
      "step": 54500
    },
    {
      "epoch": 1.375,
      "grad_norm": 1.5988309383392334,
      "learning_rate": 2.7083750000000003e-05,
      "loss": 0.0962,
      "num_input_tokens_seen": 56320000,
      "step": 55000
    },
    {
      "epoch": 1.3875,
      "grad_norm": 0.09568974375724792,
      "learning_rate": 2.687541666666667e-05,
      "loss": 0.1041,
      "num_input_tokens_seen": 56832000,
      "step": 55500
    },
    {
      "epoch": 1.4,
      "grad_norm": 105.98346710205078,
      "learning_rate": 2.6667083333333338e-05,
      "loss": 0.0895,
      "num_input_tokens_seen": 57344000,
      "step": 56000
    },
    {
      "epoch": 1.4125,
      "grad_norm": 0.04284173250198364,
      "learning_rate": 2.6458749999999998e-05,
      "loss": 0.09,
      "num_input_tokens_seen": 57856000,
      "step": 56500
    },
    {
      "epoch": 1.425,
      "grad_norm": 0.06827585399150848,
      "learning_rate": 2.6250416666666665e-05,
      "loss": 0.1064,
      "num_input_tokens_seen": 58368000,
      "step": 57000
    },
    {
      "epoch": 1.4375,
      "grad_norm": 0.10431079566478729,
      "learning_rate": 2.6042083333333333e-05,
      "loss": 0.1033,
      "num_input_tokens_seen": 58880000,
      "step": 57500
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.08796288073062897,
      "learning_rate": 2.583375e-05,
      "loss": 0.0781,
      "num_input_tokens_seen": 59392000,
      "step": 58000
    },
    {
      "epoch": 1.4625,
      "grad_norm": 1.883025884628296,
      "learning_rate": 2.5625416666666667e-05,
      "loss": 0.1001,
      "num_input_tokens_seen": 59904000,
      "step": 58500
    },
    {
      "epoch": 1.475,
      "grad_norm": 0.2901429533958435,
      "learning_rate": 2.5417083333333334e-05,
      "loss": 0.0965,
      "num_input_tokens_seen": 60416000,
      "step": 59000
    },
    {
      "epoch": 1.4875,
      "grad_norm": 0.05163797363638878,
      "learning_rate": 2.520875e-05,
      "loss": 0.1064,
      "num_input_tokens_seen": 60928000,
      "step": 59500
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.03811231628060341,
      "learning_rate": 2.5000416666666672e-05,
      "loss": 0.0865,
      "num_input_tokens_seen": 61440000,
      "step": 60000
    },
    {
      "epoch": 1.5125,
      "grad_norm": 8.308381080627441,
      "learning_rate": 2.4792083333333336e-05,
      "loss": 0.093,
      "num_input_tokens_seen": 61952000,
      "step": 60500
    },
    {
      "epoch": 1.525,
      "grad_norm": 0.03729177638888359,
      "learning_rate": 2.458375e-05,
      "loss": 0.077,
      "num_input_tokens_seen": 62464000,
      "step": 61000
    },
    {
      "epoch": 1.5375,
      "grad_norm": 5.803088188171387,
      "learning_rate": 2.4375416666666667e-05,
      "loss": 0.101,
      "num_input_tokens_seen": 62976000,
      "step": 61500
    },
    {
      "epoch": 1.55,
      "grad_norm": 2.017042636871338,
      "learning_rate": 2.4167083333333334e-05,
      "loss": 0.0896,
      "num_input_tokens_seen": 63488000,
      "step": 62000
    },
    {
      "epoch": 1.5625,
      "grad_norm": 32.310630798339844,
      "learning_rate": 2.395875e-05,
      "loss": 0.1145,
      "num_input_tokens_seen": 64000000,
      "step": 62500
    },
    {
      "epoch": 1.575,
      "grad_norm": 0.37863266468048096,
      "learning_rate": 2.3750416666666665e-05,
      "loss": 0.1017,
      "num_input_tokens_seen": 64512000,
      "step": 63000
    },
    {
      "epoch": 1.5875,
      "grad_norm": 0.05939273163676262,
      "learning_rate": 2.3542083333333333e-05,
      "loss": 0.0962,
      "num_input_tokens_seen": 65024000,
      "step": 63500
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.045398563146591187,
      "learning_rate": 2.333375e-05,
      "loss": 0.0895,
      "num_input_tokens_seen": 65536000,
      "step": 64000
    },
    {
      "epoch": 1.6125,
      "grad_norm": 1.5717942714691162,
      "learning_rate": 2.3125416666666667e-05,
      "loss": 0.0788,
      "num_input_tokens_seen": 66048000,
      "step": 64500
    },
    {
      "epoch": 1.625,
      "grad_norm": 6.278552532196045,
      "learning_rate": 2.2917083333333334e-05,
      "loss": 0.0876,
      "num_input_tokens_seen": 66560000,
      "step": 65000
    },
    {
      "epoch": 1.6375,
      "grad_norm": 0.048090457916259766,
      "learning_rate": 2.2708750000000002e-05,
      "loss": 0.0709,
      "num_input_tokens_seen": 67072000,
      "step": 65500
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.5121225714683533,
      "learning_rate": 2.250041666666667e-05,
      "loss": 0.0906,
      "num_input_tokens_seen": 67584000,
      "step": 66000
    },
    {
      "epoch": 1.6625,
      "grad_norm": 0.04399065673351288,
      "learning_rate": 2.2292083333333336e-05,
      "loss": 0.097,
      "num_input_tokens_seen": 68096000,
      "step": 66500
    },
    {
      "epoch": 1.675,
      "grad_norm": 0.051211412996053696,
      "learning_rate": 2.208375e-05,
      "loss": 0.0931,
      "num_input_tokens_seen": 68608000,
      "step": 67000
    },
    {
      "epoch": 1.6875,
      "grad_norm": 58.19650650024414,
      "learning_rate": 2.1875416666666667e-05,
      "loss": 0.0794,
      "num_input_tokens_seen": 69120000,
      "step": 67500
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.07303386926651001,
      "learning_rate": 2.1667083333333335e-05,
      "loss": 0.1015,
      "num_input_tokens_seen": 69632000,
      "step": 68000
    },
    {
      "epoch": 1.7125,
      "grad_norm": 0.02853875607252121,
      "learning_rate": 2.145875e-05,
      "loss": 0.0684,
      "num_input_tokens_seen": 70144000,
      "step": 68500
    },
    {
      "epoch": 1.725,
      "grad_norm": 0.06810135394334793,
      "learning_rate": 2.1250416666666666e-05,
      "loss": 0.101,
      "num_input_tokens_seen": 70656000,
      "step": 69000
    },
    {
      "epoch": 1.7375,
      "grad_norm": 0.019835174083709717,
      "learning_rate": 2.1042083333333333e-05,
      "loss": 0.0719,
      "num_input_tokens_seen": 71168000,
      "step": 69500
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.07850214093923569,
      "learning_rate": 2.083375e-05,
      "loss": 0.0808,
      "num_input_tokens_seen": 71680000,
      "step": 70000
    },
    {
      "epoch": 1.7625,
      "grad_norm": 0.050091035664081573,
      "learning_rate": 2.0625416666666667e-05,
      "loss": 0.0835,
      "num_input_tokens_seen": 72192000,
      "step": 70500
    },
    {
      "epoch": 1.775,
      "grad_norm": 0.01498348731547594,
      "learning_rate": 2.0417083333333335e-05,
      "loss": 0.0918,
      "num_input_tokens_seen": 72704000,
      "step": 71000
    },
    {
      "epoch": 1.7875,
      "grad_norm": 0.034435465931892395,
      "learning_rate": 2.0208750000000002e-05,
      "loss": 0.0777,
      "num_input_tokens_seen": 73216000,
      "step": 71500
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.045340586453676224,
      "learning_rate": 2.000041666666667e-05,
      "loss": 0.0894,
      "num_input_tokens_seen": 73728000,
      "step": 72000
    },
    {
      "epoch": 1.8125,
      "grad_norm": 0.28705134987831116,
      "learning_rate": 1.9792083333333333e-05,
      "loss": 0.0741,
      "num_input_tokens_seen": 74240000,
      "step": 72500
    },
    {
      "epoch": 1.825,
      "grad_norm": 155.12445068359375,
      "learning_rate": 1.958375e-05,
      "loss": 0.0813,
      "num_input_tokens_seen": 74752000,
      "step": 73000
    },
    {
      "epoch": 1.8375,
      "grad_norm": 0.08065774291753769,
      "learning_rate": 1.9375416666666668e-05,
      "loss": 0.0819,
      "num_input_tokens_seen": 75264000,
      "step": 73500
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.40842482447624207,
      "learning_rate": 1.9167083333333335e-05,
      "loss": 0.0669,
      "num_input_tokens_seen": 75776000,
      "step": 74000
    },
    {
      "epoch": 1.8625,
      "grad_norm": 0.09376771003007889,
      "learning_rate": 1.895875e-05,
      "loss": 0.0749,
      "num_input_tokens_seen": 76288000,
      "step": 74500
    },
    {
      "epoch": 1.875,
      "grad_norm": 0.035654786974191666,
      "learning_rate": 1.8750416666666666e-05,
      "loss": 0.0757,
      "num_input_tokens_seen": 76800000,
      "step": 75000
    },
    {
      "epoch": 1.8875,
      "grad_norm": 0.04763418436050415,
      "learning_rate": 1.8542083333333337e-05,
      "loss": 0.0652,
      "num_input_tokens_seen": 77312000,
      "step": 75500
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.04799911379814148,
      "learning_rate": 1.833375e-05,
      "loss": 0.0688,
      "num_input_tokens_seen": 77824000,
      "step": 76000
    },
    {
      "epoch": 1.9125,
      "grad_norm": 0.22011104226112366,
      "learning_rate": 1.8125416666666668e-05,
      "loss": 0.0797,
      "num_input_tokens_seen": 78336000,
      "step": 76500
    },
    {
      "epoch": 1.925,
      "grad_norm": 0.05009845644235611,
      "learning_rate": 1.7917083333333335e-05,
      "loss": 0.0542,
      "num_input_tokens_seen": 78848000,
      "step": 77000
    },
    {
      "epoch": 1.9375,
      "grad_norm": 1.8440918922424316,
      "learning_rate": 1.7708750000000002e-05,
      "loss": 0.0773,
      "num_input_tokens_seen": 79360000,
      "step": 77500
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.025103362277150154,
      "learning_rate": 1.7500416666666666e-05,
      "loss": 0.0692,
      "num_input_tokens_seen": 79872000,
      "step": 78000
    },
    {
      "epoch": 1.9625,
      "grad_norm": 14.473067283630371,
      "learning_rate": 1.7292083333333333e-05,
      "loss": 0.0658,
      "num_input_tokens_seen": 80384000,
      "step": 78500
    },
    {
      "epoch": 1.975,
      "grad_norm": 0.051201172173023224,
      "learning_rate": 1.708375e-05,
      "loss": 0.0648,
      "num_input_tokens_seen": 80896000,
      "step": 79000
    },
    {
      "epoch": 1.9875,
      "grad_norm": 0.05888315662741661,
      "learning_rate": 1.6875416666666668e-05,
      "loss": 0.0792,
      "num_input_tokens_seen": 81408000,
      "step": 79500
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.08364333212375641,
      "learning_rate": 1.666708333333333e-05,
      "loss": 0.0682,
      "num_input_tokens_seen": 81920000,
      "step": 80000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9528,
      "eval_combined_score": 1.327030219983165,
      "eval_loss": 0.22639968991279602,
      "eval_runtime": 39.5018,
      "eval_samples_per_second": 2025.225,
      "eval_steps_per_second": 253.153,
      "num_input_tokens_seen": 81920000,
      "step": 80000
    },
    {
      "epoch": 2.0125,
      "grad_norm": 1.8050274848937988,
      "learning_rate": 1.645875e-05,
      "loss": 0.0508,
      "num_input_tokens_seen": 82432000,
      "step": 80500
    },
    {
      "epoch": 2.025,
      "grad_norm": 0.02845793031156063,
      "learning_rate": 1.625041666666667e-05,
      "loss": 0.0525,
      "num_input_tokens_seen": 82944000,
      "step": 81000
    },
    {
      "epoch": 2.0375,
      "grad_norm": 0.025743141770362854,
      "learning_rate": 1.6042083333333337e-05,
      "loss": 0.0483,
      "num_input_tokens_seen": 83456000,
      "step": 81500
    },
    {
      "epoch": 2.05,
      "grad_norm": 8.399374008178711,
      "learning_rate": 1.583375e-05,
      "loss": 0.0652,
      "num_input_tokens_seen": 83968000,
      "step": 82000
    },
    {
      "epoch": 2.0625,
      "grad_norm": 0.017835861071944237,
      "learning_rate": 1.5625416666666668e-05,
      "loss": 0.0482,
      "num_input_tokens_seen": 84480000,
      "step": 82500
    },
    {
      "epoch": 2.075,
      "grad_norm": 0.06692575663328171,
      "learning_rate": 1.5417083333333335e-05,
      "loss": 0.0521,
      "num_input_tokens_seen": 84992000,
      "step": 83000
    },
    {
      "epoch": 2.0875,
      "grad_norm": 0.035685233771800995,
      "learning_rate": 1.5208749999999999e-05,
      "loss": 0.0582,
      "num_input_tokens_seen": 85504000,
      "step": 83500
    },
    {
      "epoch": 2.1,
      "grad_norm": 0.13481773436069489,
      "learning_rate": 1.5000416666666666e-05,
      "loss": 0.0468,
      "num_input_tokens_seen": 86016000,
      "step": 84000
    },
    {
      "epoch": 2.1125,
      "grad_norm": 17.486289978027344,
      "learning_rate": 1.4792083333333334e-05,
      "loss": 0.0559,
      "num_input_tokens_seen": 86528000,
      "step": 84500
    },
    {
      "epoch": 2.125,
      "grad_norm": 0.03201691806316376,
      "learning_rate": 1.458375e-05,
      "loss": 0.0425,
      "num_input_tokens_seen": 87040000,
      "step": 85000
    },
    {
      "epoch": 2.1375,
      "grad_norm": 0.021227147430181503,
      "learning_rate": 1.4375416666666666e-05,
      "loss": 0.05,
      "num_input_tokens_seen": 87552000,
      "step": 85500
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.040053412318229675,
      "learning_rate": 1.4167083333333334e-05,
      "loss": 0.048,
      "num_input_tokens_seen": 88064000,
      "step": 86000
    },
    {
      "epoch": 2.1625,
      "grad_norm": 0.01835712045431137,
      "learning_rate": 1.395875e-05,
      "loss": 0.038,
      "num_input_tokens_seen": 88576000,
      "step": 86500
    },
    {
      "epoch": 2.175,
      "grad_norm": 0.03529110550880432,
      "learning_rate": 1.3750416666666668e-05,
      "loss": 0.051,
      "num_input_tokens_seen": 89088000,
      "step": 87000
    },
    {
      "epoch": 2.1875,
      "grad_norm": 97.47898864746094,
      "learning_rate": 1.3542083333333334e-05,
      "loss": 0.0373,
      "num_input_tokens_seen": 89600000,
      "step": 87500
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.014346601441502571,
      "learning_rate": 1.3333750000000001e-05,
      "loss": 0.0557,
      "num_input_tokens_seen": 90112000,
      "step": 88000
    },
    {
      "epoch": 2.2125,
      "grad_norm": 0.01878521591424942,
      "learning_rate": 1.3125416666666668e-05,
      "loss": 0.0526,
      "num_input_tokens_seen": 90624000,
      "step": 88500
    },
    {
      "epoch": 2.225,
      "grad_norm": 0.04132438451051712,
      "learning_rate": 1.2917083333333335e-05,
      "loss": 0.0464,
      "num_input_tokens_seen": 91136000,
      "step": 89000
    },
    {
      "epoch": 2.2375,
      "grad_norm": 0.02642699144780636,
      "learning_rate": 1.270875e-05,
      "loss": 0.0464,
      "num_input_tokens_seen": 91648000,
      "step": 89500
    },
    {
      "epoch": 2.25,
      "grad_norm": 877.631591796875,
      "learning_rate": 1.2500416666666666e-05,
      "loss": 0.0374,
      "num_input_tokens_seen": 92160000,
      "step": 90000
    },
    {
      "epoch": 2.2625,
      "grad_norm": 0.4528743326663971,
      "learning_rate": 1.2292083333333334e-05,
      "loss": 0.059,
      "num_input_tokens_seen": 92672000,
      "step": 90500
    },
    {
      "epoch": 2.275,
      "grad_norm": 0.1183973178267479,
      "learning_rate": 1.2083750000000001e-05,
      "loss": 0.0505,
      "num_input_tokens_seen": 93184000,
      "step": 91000
    },
    {
      "epoch": 2.2875,
      "grad_norm": 0.04196188971400261,
      "learning_rate": 1.1875416666666667e-05,
      "loss": 0.0517,
      "num_input_tokens_seen": 93696000,
      "step": 91500
    },
    {
      "epoch": 2.3,
      "grad_norm": 0.03194092586636543,
      "learning_rate": 1.1667083333333334e-05,
      "loss": 0.0511,
      "num_input_tokens_seen": 94208000,
      "step": 92000
    },
    {
      "epoch": 2.3125,
      "grad_norm": 0.0184203889220953,
      "learning_rate": 1.145875e-05,
      "loss": 0.0501,
      "num_input_tokens_seen": 94720000,
      "step": 92500
    },
    {
      "epoch": 2.325,
      "grad_norm": 0.053758785128593445,
      "learning_rate": 1.1250416666666667e-05,
      "loss": 0.0412,
      "num_input_tokens_seen": 95232000,
      "step": 93000
    },
    {
      "epoch": 2.3375,
      "grad_norm": 0.036872394382953644,
      "learning_rate": 1.1042083333333334e-05,
      "loss": 0.0418,
      "num_input_tokens_seen": 95744000,
      "step": 93500
    },
    {
      "epoch": 2.35,
      "grad_norm": 0.014922083355486393,
      "learning_rate": 1.0833750000000001e-05,
      "loss": 0.0507,
      "num_input_tokens_seen": 96256000,
      "step": 94000
    },
    {
      "epoch": 2.3625,
      "grad_norm": 0.020171664655208588,
      "learning_rate": 1.0625416666666667e-05,
      "loss": 0.0444,
      "num_input_tokens_seen": 96768000,
      "step": 94500
    },
    {
      "epoch": 2.375,
      "grad_norm": 1.7357385158538818,
      "learning_rate": 1.0417083333333334e-05,
      "loss": 0.0545,
      "num_input_tokens_seen": 97280000,
      "step": 95000
    },
    {
      "epoch": 2.3875,
      "grad_norm": 0.023304857313632965,
      "learning_rate": 1.020875e-05,
      "loss": 0.0515,
      "num_input_tokens_seen": 97792000,
      "step": 95500
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.01939631998538971,
      "learning_rate": 1.0000416666666668e-05,
      "loss": 0.0495,
      "num_input_tokens_seen": 98304000,
      "step": 96000
    },
    {
      "epoch": 2.4125,
      "grad_norm": 0.019845569506287575,
      "learning_rate": 9.792083333333334e-06,
      "loss": 0.0411,
      "num_input_tokens_seen": 98816000,
      "step": 96500
    },
    {
      "epoch": 2.425,
      "grad_norm": 0.014959324151277542,
      "learning_rate": 9.583750000000001e-06,
      "loss": 0.0342,
      "num_input_tokens_seen": 99328000,
      "step": 97000
    },
    {
      "epoch": 2.4375,
      "grad_norm": 0.01649474911391735,
      "learning_rate": 9.375416666666667e-06,
      "loss": 0.0472,
      "num_input_tokens_seen": 99840000,
      "step": 97500
    },
    {
      "epoch": 2.45,
      "grad_norm": 0.031929146498441696,
      "learning_rate": 9.167083333333332e-06,
      "loss": 0.0384,
      "num_input_tokens_seen": 100352000,
      "step": 98000
    },
    {
      "epoch": 2.4625,
      "grad_norm": 0.10980285704135895,
      "learning_rate": 8.958750000000001e-06,
      "loss": 0.0513,
      "num_input_tokens_seen": 100864000,
      "step": 98500
    },
    {
      "epoch": 2.475,
      "grad_norm": 0.020933426916599274,
      "learning_rate": 8.750416666666667e-06,
      "loss": 0.0478,
      "num_input_tokens_seen": 101376000,
      "step": 99000
    },
    {
      "epoch": 2.4875,
      "grad_norm": 0.017404716461896896,
      "learning_rate": 8.542083333333334e-06,
      "loss": 0.039,
      "num_input_tokens_seen": 101888000,
      "step": 99500
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.11211936920881271,
      "learning_rate": 8.33375e-06,
      "loss": 0.0473,
      "num_input_tokens_seen": 102400000,
      "step": 100000
    },
    {
      "epoch": 2.5125,
      "grad_norm": 0.023433908820152283,
      "learning_rate": 8.125416666666667e-06,
      "loss": 0.0417,
      "num_input_tokens_seen": 102912000,
      "step": 100500
    },
    {
      "epoch": 2.525,
      "grad_norm": 0.011504637077450752,
      "learning_rate": 7.917083333333334e-06,
      "loss": 0.0396,
      "num_input_tokens_seen": 103424000,
      "step": 101000
    },
    {
      "epoch": 2.5375,
      "grad_norm": 0.012843768112361431,
      "learning_rate": 7.708750000000001e-06,
      "loss": 0.0475,
      "num_input_tokens_seen": 103936000,
      "step": 101500
    },
    {
      "epoch": 2.55,
      "grad_norm": 0.020524220541119576,
      "learning_rate": 7.500416666666667e-06,
      "loss": 0.0472,
      "num_input_tokens_seen": 104448000,
      "step": 102000
    },
    {
      "epoch": 2.5625,
      "grad_norm": 0.01861303672194481,
      "learning_rate": 7.292083333333334e-06,
      "loss": 0.0294,
      "num_input_tokens_seen": 104960000,
      "step": 102500
    },
    {
      "epoch": 2.575,
      "grad_norm": 0.021343663334846497,
      "learning_rate": 7.08375e-06,
      "loss": 0.0376,
      "num_input_tokens_seen": 105472000,
      "step": 103000
    },
    {
      "epoch": 2.5875,
      "grad_norm": 0.8902124166488647,
      "learning_rate": 6.875416666666668e-06,
      "loss": 0.0456,
      "num_input_tokens_seen": 105984000,
      "step": 103500
    },
    {
      "epoch": 2.6,
      "grad_norm": 0.06912536174058914,
      "learning_rate": 6.667083333333333e-06,
      "loss": 0.0365,
      "num_input_tokens_seen": 106496000,
      "step": 104000
    },
    {
      "epoch": 2.6125,
      "grad_norm": 0.17968295514583588,
      "learning_rate": 6.458750000000001e-06,
      "loss": 0.0372,
      "num_input_tokens_seen": 107008000,
      "step": 104500
    },
    {
      "epoch": 2.625,
      "grad_norm": 0.01591988280415535,
      "learning_rate": 6.250416666666667e-06,
      "loss": 0.0334,
      "num_input_tokens_seen": 107520000,
      "step": 105000
    },
    {
      "epoch": 2.6375,
      "grad_norm": 0.015928415581583977,
      "learning_rate": 6.0420833333333334e-06,
      "loss": 0.0425,
      "num_input_tokens_seen": 108032000,
      "step": 105500
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.04025963693857193,
      "learning_rate": 5.833750000000001e-06,
      "loss": 0.0371,
      "num_input_tokens_seen": 108544000,
      "step": 106000
    },
    {
      "epoch": 2.6625,
      "grad_norm": 0.07384547591209412,
      "learning_rate": 5.625416666666667e-06,
      "loss": 0.043,
      "num_input_tokens_seen": 109056000,
      "step": 106500
    },
    {
      "epoch": 2.675,
      "grad_norm": 0.0424518883228302,
      "learning_rate": 5.4170833333333335e-06,
      "loss": 0.0349,
      "num_input_tokens_seen": 109568000,
      "step": 107000
    },
    {
      "epoch": 2.6875,
      "grad_norm": 0.01949002780020237,
      "learning_rate": 5.208750000000001e-06,
      "loss": 0.04,
      "num_input_tokens_seen": 110080000,
      "step": 107500
    },
    {
      "epoch": 2.7,
      "grad_norm": 0.011860487051308155,
      "learning_rate": 5.000416666666667e-06,
      "loss": 0.0274,
      "num_input_tokens_seen": 110592000,
      "step": 108000
    },
    {
      "epoch": 2.7125,
      "grad_norm": 0.010354108177125454,
      "learning_rate": 4.7920833333333335e-06,
      "loss": 0.0305,
      "num_input_tokens_seen": 111104000,
      "step": 108500
    },
    {
      "epoch": 2.725,
      "grad_norm": 13.753798484802246,
      "learning_rate": 4.583750000000001e-06,
      "loss": 0.0446,
      "num_input_tokens_seen": 111616000,
      "step": 109000
    },
    {
      "epoch": 2.7375,
      "grad_norm": 0.017097918316721916,
      "learning_rate": 4.375416666666666e-06,
      "loss": 0.0398,
      "num_input_tokens_seen": 112128000,
      "step": 109500
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.01976764015853405,
      "learning_rate": 4.167083333333334e-06,
      "loss": 0.0318,
      "num_input_tokens_seen": 112640000,
      "step": 110000
    },
    {
      "epoch": 2.7625,
      "grad_norm": 6.2214531898498535,
      "learning_rate": 3.95875e-06,
      "loss": 0.0424,
      "num_input_tokens_seen": 113152000,
      "step": 110500
    },
    {
      "epoch": 2.775,
      "grad_norm": 0.007995002903044224,
      "learning_rate": 3.750416666666667e-06,
      "loss": 0.0442,
      "num_input_tokens_seen": 113664000,
      "step": 111000
    },
    {
      "epoch": 2.7875,
      "grad_norm": 0.05575885996222496,
      "learning_rate": 3.5420833333333332e-06,
      "loss": 0.037,
      "num_input_tokens_seen": 114176000,
      "step": 111500
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.1336035579442978,
      "learning_rate": 3.33375e-06,
      "loss": 0.0385,
      "num_input_tokens_seen": 114688000,
      "step": 112000
    },
    {
      "epoch": 2.8125,
      "grad_norm": 0.019037237390875816,
      "learning_rate": 3.125416666666667e-06,
      "loss": 0.0326,
      "num_input_tokens_seen": 115200000,
      "step": 112500
    },
    {
      "epoch": 2.825,
      "grad_norm": 0.00875825248658657,
      "learning_rate": 2.9170833333333333e-06,
      "loss": 0.0277,
      "num_input_tokens_seen": 115712000,
      "step": 113000
    },
    {
      "epoch": 2.8375,
      "grad_norm": 0.052571795880794525,
      "learning_rate": 2.70875e-06,
      "loss": 0.037,
      "num_input_tokens_seen": 116224000,
      "step": 113500
    },
    {
      "epoch": 2.85,
      "grad_norm": 0.019845254719257355,
      "learning_rate": 2.500416666666667e-06,
      "loss": 0.0445,
      "num_input_tokens_seen": 116736000,
      "step": 114000
    },
    {
      "epoch": 2.8625,
      "grad_norm": 0.01046211551874876,
      "learning_rate": 2.2920833333333338e-06,
      "loss": 0.0294,
      "num_input_tokens_seen": 117248000,
      "step": 114500
    },
    {
      "epoch": 2.875,
      "grad_norm": 0.012272953987121582,
      "learning_rate": 2.0837499999999997e-06,
      "loss": 0.0396,
      "num_input_tokens_seen": 117760000,
      "step": 115000
    },
    {
      "epoch": 2.8875,
      "grad_norm": 0.022472262382507324,
      "learning_rate": 1.8754166666666666e-06,
      "loss": 0.047,
      "num_input_tokens_seen": 118272000,
      "step": 115500
    },
    {
      "epoch": 2.9,
      "grad_norm": 0.07120255380868912,
      "learning_rate": 1.6670833333333334e-06,
      "loss": 0.0437,
      "num_input_tokens_seen": 118784000,
      "step": 116000
    },
    {
      "epoch": 2.9125,
      "grad_norm": 0.060723673552274704,
      "learning_rate": 1.45875e-06,
      "loss": 0.035,
      "num_input_tokens_seen": 119296000,
      "step": 116500
    },
    {
      "epoch": 2.925,
      "grad_norm": 0.012350406497716904,
      "learning_rate": 1.2504166666666668e-06,
      "loss": 0.03,
      "num_input_tokens_seen": 119808000,
      "step": 117000
    },
    {
      "epoch": 2.9375,
      "grad_norm": 0.18025244772434235,
      "learning_rate": 1.0420833333333334e-06,
      "loss": 0.0306,
      "num_input_tokens_seen": 120320000,
      "step": 117500
    },
    {
      "epoch": 2.95,
      "grad_norm": 0.007316856179386377,
      "learning_rate": 8.3375e-07,
      "loss": 0.0328,
      "num_input_tokens_seen": 120832000,
      "step": 118000
    },
    {
      "epoch": 2.9625,
      "grad_norm": 0.06496240198612213,
      "learning_rate": 6.254166666666667e-07,
      "loss": 0.039,
      "num_input_tokens_seen": 121344000,
      "step": 118500
    },
    {
      "epoch": 2.975,
      "grad_norm": 0.0057182470336556435,
      "learning_rate": 4.170833333333334e-07,
      "loss": 0.0367,
      "num_input_tokens_seen": 121856000,
      "step": 119000
    },
    {
      "epoch": 2.9875,
      "grad_norm": 0.017466630786657333,
      "learning_rate": 2.0875e-07,
      "loss": 0.0381,
      "num_input_tokens_seen": 122368000,
      "step": 119500
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.013719202019274235,
      "learning_rate": 4.1666666666666673e-10,
      "loss": 0.0424,
      "num_input_tokens_seen": 122880000,
      "step": 120000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.955225,
      "eval_combined_score": 1.2845397902488602,
      "eval_loss": 0.2597787082195282,
      "eval_runtime": 39.5285,
      "eval_samples_per_second": 2023.856,
      "eval_steps_per_second": 252.982,
      "num_input_tokens_seen": 122880000,
      "step": 120000
    },
    {
      "epoch": 3.0,
      "num_input_tokens_seen": 122880000,
      "step": 120000,
      "total_flos": 1.580945522688e+16,
      "train_loss": 0.10143799341519674,
      "train_runtime": 4258.2582,
      "train_samples_per_second": 225.444,
      "train_steps_per_second": 28.181,
      "train_tokens_per_second": 28856.869
    }
  ],
  "logging_steps": 500,
  "max_steps": 120000,
  "num_input_tokens_seen": 122880000,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.580945522688e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}