{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9978252990213845,
  "eval_steps": 14,
  "global_step": 1378,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0014498006524102935,
      "grad_norm": 1457492047.9411972,
      "learning_rate": 4.7619047619047623e-07,
      "loss": 2.5508,
      "step": 1
    },
    {
      "epoch": 0.002899601304820587,
      "grad_norm": 1708341776.4369097,
      "learning_rate": 9.523809523809525e-07,
      "loss": 2.6569,
      "step": 2
    },
    {
      "epoch": 0.004349401957230881,
      "grad_norm": 14641.48761650254,
      "learning_rate": 1.4285714285714286e-06,
      "loss": 2.7324,
      "step": 3
    },
    {
      "epoch": 0.005799202609641174,
      "grad_norm": 6087.45573803664,
      "learning_rate": 1.904761904761905e-06,
      "loss": 2.7763,
      "step": 4
    },
    {
      "epoch": 0.007249003262051468,
      "grad_norm": 5270.299882297447,
      "learning_rate": 2.380952380952381e-06,
      "loss": 2.5942,
      "step": 5
    },
    {
      "epoch": 0.008698803914461761,
      "grad_norm": 6471.382403807851,
      "learning_rate": 2.8571428571428573e-06,
      "loss": 2.5252,
      "step": 6
    },
    {
      "epoch": 0.010148604566872055,
      "grad_norm": 7128.933563751969,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 1.8433,
      "step": 7
    },
    {
      "epoch": 0.011598405219282348,
      "grad_norm": 1071.358854209208,
      "learning_rate": 3.80952380952381e-06,
      "loss": 1.7892,
      "step": 8
    },
    {
      "epoch": 0.013048205871692642,
      "grad_norm": 628.265924703897,
      "learning_rate": 4.2857142857142855e-06,
      "loss": 1.732,
      "step": 9
    },
    {
      "epoch": 0.014498006524102935,
      "grad_norm": 126.8989979881357,
      "learning_rate": 4.761904761904762e-06,
      "loss": 1.8633,
      "step": 10
    },
    {
      "epoch": 0.01594780717651323,
      "grad_norm": 108.89689603388551,
      "learning_rate": 5.2380952380952384e-06,
      "loss": 1.6739,
      "step": 11
    },
    {
      "epoch": 0.017397607828923523,
      "grad_norm": 2424.9717655274735,
      "learning_rate": 5.7142857142857145e-06,
      "loss": 2.3872,
      "step": 12
    },
    {
      "epoch": 0.018847408481333816,
      "grad_norm": 37607.39427004566,
      "learning_rate": 6.1904761904761914e-06,
      "loss": 2.4291,
      "step": 13
    },
    {
      "epoch": 0.02029720913374411,
      "grad_norm": 1554.860601015685,
      "learning_rate": 6.666666666666667e-06,
      "loss": 1.6035,
      "step": 14
    },
    {
      "epoch": 0.02029720913374411,
      "eval_loss": 2.472167491912842,
      "eval_runtime": 780.1506,
      "eval_samples_per_second": 22.629,
      "eval_steps_per_second": 0.354,
      "step": 14
    },
    {
      "epoch": 0.021747009786154403,
      "grad_norm": 252.30332846491794,
      "learning_rate": 7.1428571428571436e-06,
      "loss": 1.6409,
      "step": 15
    },
    {
      "epoch": 0.023196810438564697,
      "grad_norm": 16.283385648133912,
      "learning_rate": 7.61904761904762e-06,
      "loss": 1.6375,
      "step": 16
    },
    {
      "epoch": 0.02464661109097499,
      "grad_norm": 8.637178118094772,
      "learning_rate": 8.095238095238097e-06,
      "loss": 1.6096,
      "step": 17
    },
    {
      "epoch": 0.026096411743385284,
      "grad_norm": 7.605002815893089,
      "learning_rate": 8.571428571428571e-06,
      "loss": 1.4773,
      "step": 18
    },
    {
      "epoch": 0.027546212395795577,
      "grad_norm": 13.78342592067249,
      "learning_rate": 9.047619047619049e-06,
      "loss": 1.4452,
      "step": 19
    },
    {
      "epoch": 0.02899601304820587,
      "grad_norm": 1428.7948104485945,
      "learning_rate": 9.523809523809525e-06,
      "loss": 1.5265,
      "step": 20
    },
    {
      "epoch": 0.030445813700616164,
      "grad_norm": 213.11582252486195,
      "learning_rate": 1e-05,
      "loss": 1.4612,
      "step": 21
    },
    {
      "epoch": 0.03189561435302646,
      "grad_norm": 117.70329162763524,
      "learning_rate": 1.0476190476190477e-05,
      "loss": 1.5,
      "step": 22
    },
    {
      "epoch": 0.03334541500543675,
      "grad_norm": 146.3398774276084,
      "learning_rate": 1.0952380952380955e-05,
      "loss": 1.4917,
      "step": 23
    },
    {
      "epoch": 0.034795215657847045,
      "grad_norm": 66.75292101354283,
      "learning_rate": 1.1428571428571429e-05,
      "loss": 1.4443,
      "step": 24
    },
    {
      "epoch": 0.03624501631025734,
      "grad_norm": 5.554245454631336,
      "learning_rate": 1.1904761904761905e-05,
      "loss": 1.4653,
      "step": 25
    },
    {
      "epoch": 0.03769481696266763,
      "grad_norm": 5.168402751519938,
      "learning_rate": 1.2380952380952383e-05,
      "loss": 1.405,
      "step": 26
    },
    {
      "epoch": 0.039144617615077926,
      "grad_norm": 3.846778342377277,
      "learning_rate": 1.2857142857142859e-05,
      "loss": 1.3952,
      "step": 27
    },
    {
      "epoch": 0.04059441826748822,
      "grad_norm": 4.984675114916189,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 1.4311,
      "step": 28
    },
    {
      "epoch": 0.04059441826748822,
      "eval_loss": 2.224747896194458,
      "eval_runtime": 713.0359,
      "eval_samples_per_second": 24.759,
      "eval_steps_per_second": 0.387,
      "step": 28
    },
    {
      "epoch": 0.04204421891989851,
      "grad_norm": 6.220059299424948,
      "learning_rate": 1.3809523809523811e-05,
      "loss": 1.3498,
      "step": 29
    },
    {
      "epoch": 0.043494019572308806,
      "grad_norm": 14.91496839626074,
      "learning_rate": 1.4285714285714287e-05,
      "loss": 1.4387,
      "step": 30
    },
    {
      "epoch": 0.0449438202247191,
      "grad_norm": 3.895470815283485,
      "learning_rate": 1.4761904761904763e-05,
      "loss": 1.487,
      "step": 31
    },
    {
      "epoch": 0.04639362087712939,
      "grad_norm": 4.099711030805065,
      "learning_rate": 1.523809523809524e-05,
      "loss": 1.3014,
      "step": 32
    },
    {
      "epoch": 0.04784342152953969,
      "grad_norm": 4.824765658374792,
      "learning_rate": 1.5714285714285715e-05,
      "loss": 1.5107,
      "step": 33
    },
    {
      "epoch": 0.04929322218194998,
      "grad_norm": 3.7969276411012465,
      "learning_rate": 1.6190476190476193e-05,
      "loss": 1.4654,
      "step": 34
    },
    {
      "epoch": 0.050743022834360274,
      "grad_norm": 3.7332706899493067,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.3853,
      "step": 35
    },
    {
      "epoch": 0.05219282348677057,
      "grad_norm": 4.183430178644755,
      "learning_rate": 1.7142857142857142e-05,
      "loss": 1.3083,
      "step": 36
    },
    {
      "epoch": 0.05364262413918086,
      "grad_norm": 3.626864712504921,
      "learning_rate": 1.761904761904762e-05,
      "loss": 1.4702,
      "step": 37
    },
    {
      "epoch": 0.055092424791591155,
      "grad_norm": 4.189490063354064,
      "learning_rate": 1.8095238095238097e-05,
      "loss": 1.3907,
      "step": 38
    },
    {
      "epoch": 0.05654222544400145,
      "grad_norm": 3.7476410127601665,
      "learning_rate": 1.8571428571428575e-05,
      "loss": 1.3952,
      "step": 39
    },
    {
      "epoch": 0.05799202609641174,
      "grad_norm": 4.025765892969993,
      "learning_rate": 1.904761904761905e-05,
      "loss": 1.3885,
      "step": 40
    },
    {
      "epoch": 0.059441826748822035,
      "grad_norm": 4.08490727317016,
      "learning_rate": 1.9523809523809524e-05,
      "loss": 1.3328,
      "step": 41
    },
    {
      "epoch": 0.06089162740123233,
      "grad_norm": 3.5800725703160747,
      "learning_rate": 2e-05,
      "loss": 1.3568,
      "step": 42
    },
    {
      "epoch": 0.06089162740123233,
      "eval_loss": 2.175341844558716,
      "eval_runtime": 714.9381,
      "eval_samples_per_second": 24.693,
      "eval_steps_per_second": 0.386,
      "step": 42
    },
    {
      "epoch": 0.06234142805364262,
      "grad_norm": 4.539326607933263,
      "learning_rate": 1.99999723524512e-05,
      "loss": 1.375,
      "step": 43
    },
    {
      "epoch": 0.06379122870605292,
      "grad_norm": 3.809257947016862,
      "learning_rate": 1.999988940995767e-05,
      "loss": 1.3114,
      "step": 44
    },
    {
      "epoch": 0.06524102935846321,
      "grad_norm": 3.9956395954846444,
      "learning_rate": 1.9999751172978047e-05,
      "loss": 1.3303,
      "step": 45
    },
    {
      "epoch": 0.0666908300108735,
      "grad_norm": 5.728697479488556,
      "learning_rate": 1.9999557642276715e-05,
      "loss": 1.2766,
      "step": 46
    },
    {
      "epoch": 0.0681406306632838,
      "grad_norm": 4.270744285528382,
      "learning_rate": 1.9999308818923795e-05,
      "loss": 1.3802,
      "step": 47
    },
    {
      "epoch": 0.06959043131569409,
      "grad_norm": 3.5855541892656198,
      "learning_rate": 1.9999004704295167e-05,
      "loss": 1.2665,
      "step": 48
    },
    {
      "epoch": 0.07104023196810438,
      "grad_norm": 6.135582053021324,
      "learning_rate": 1.999864530007243e-05,
      "loss": 1.3791,
      "step": 49
    },
    {
      "epoch": 0.07249003262051468,
      "grad_norm": 3.9821723719368634,
      "learning_rate": 1.9998230608242917e-05,
      "loss": 1.3425,
      "step": 50
    },
    {
      "epoch": 0.07393983327292497,
      "grad_norm": 4.096727085742442,
      "learning_rate": 1.999776063109967e-05,
      "loss": 1.3186,
      "step": 51
    },
    {
      "epoch": 0.07538963392533526,
      "grad_norm": 3.488782934449226,
      "learning_rate": 1.999723537124143e-05,
      "loss": 1.4062,
      "step": 52
    },
    {
      "epoch": 0.07683943457774556,
      "grad_norm": 3.265937418740512,
      "learning_rate": 1.999665483157263e-05,
      "loss": 1.349,
      "step": 53
    },
    {
      "epoch": 0.07828923523015585,
      "grad_norm": 3.630315575066132,
      "learning_rate": 1.9996019015303366e-05,
      "loss": 1.2678,
      "step": 54
    },
    {
      "epoch": 0.07973903588256614,
      "grad_norm": 2.9901280139239512,
      "learning_rate": 1.9995327925949396e-05,
      "loss": 1.3699,
      "step": 55
    },
    {
      "epoch": 0.08118883653497644,
      "grad_norm": 2.9145774778227636,
      "learning_rate": 1.9994581567332095e-05,
      "loss": 1.282,
      "step": 56
    },
    {
      "epoch": 0.08118883653497644,
      "eval_loss": 2.0694785118103027,
      "eval_runtime": 714.6246,
      "eval_samples_per_second": 24.704,
      "eval_steps_per_second": 0.386,
      "step": 56
    },
    {
      "epoch": 0.08263863718738673,
      "grad_norm": 2.790277998276927,
      "learning_rate": 1.999377994357847e-05,
      "loss": 1.3817,
      "step": 57
    },
    {
      "epoch": 0.08408843783979703,
      "grad_norm": 3.3681275010839014,
      "learning_rate": 1.9992923059121107e-05,
      "loss": 1.2804,
      "step": 58
    },
    {
      "epoch": 0.08553823849220732,
      "grad_norm": 4.790106518133967,
      "learning_rate": 1.9992010918698147e-05,
      "loss": 1.3975,
      "step": 59
    },
    {
      "epoch": 0.08698803914461761,
      "grad_norm": 3.857238854469178,
      "learning_rate": 1.999104352735329e-05,
      "loss": 1.3224,
      "step": 60
    },
    {
      "epoch": 0.0884378397970279,
      "grad_norm": 4.416924497424182,
      "learning_rate": 1.9990020890435733e-05,
      "loss": 1.3469,
      "step": 61
    },
    {
      "epoch": 0.0898876404494382,
      "grad_norm": 3.999696931116358,
      "learning_rate": 1.9988943013600156e-05,
      "loss": 1.2731,
      "step": 62
    },
    {
      "epoch": 0.0913374411018485,
      "grad_norm": 5.712211446404219,
      "learning_rate": 1.998780990280669e-05,
      "loss": 1.2586,
      "step": 63
    },
    {
      "epoch": 0.09278724175425879,
      "grad_norm": 3.4221352973541603,
      "learning_rate": 1.9986621564320878e-05,
      "loss": 1.3256,
      "step": 64
    },
    {
      "epoch": 0.09423704240666908,
      "grad_norm": 4.340054969184447,
      "learning_rate": 1.9985378004713657e-05,
      "loss": 1.3228,
      "step": 65
    },
    {
      "epoch": 0.09568684305907937,
      "grad_norm": 11.615412008695657,
      "learning_rate": 1.9984079230861296e-05,
      "loss": 1.2933,
      "step": 66
    },
    {
      "epoch": 0.09713664371148967,
      "grad_norm": 4.4759505965730835,
      "learning_rate": 1.9982725249945378e-05,
      "loss": 1.3441,
      "step": 67
    },
    {
      "epoch": 0.09858644436389996,
      "grad_norm": 3.367870279772255,
      "learning_rate": 1.998131606945276e-05,
      "loss": 1.3545,
      "step": 68
    },
    {
      "epoch": 0.10003624501631025,
      "grad_norm": 6.8022440140916745,
      "learning_rate": 1.9979851697175513e-05,
      "loss": 1.3147,
      "step": 69
    },
    {
      "epoch": 0.10148604566872055,
      "grad_norm": 2.823942307466475,
      "learning_rate": 1.9978332141210894e-05,
      "loss": 1.3395,
      "step": 70
    },
    {
      "epoch": 0.10148604566872055,
      "eval_loss": 2.0606071949005127,
      "eval_runtime": 712.0848,
      "eval_samples_per_second": 24.792,
      "eval_steps_per_second": 0.388,
      "step": 70
    },
    {
      "epoch": 0.10293584632113084,
      "grad_norm": 3.2670350338824394,
      "learning_rate": 1.9976757409961315e-05,
      "loss": 1.2326,
      "step": 71
    },
    {
      "epoch": 0.10438564697354114,
      "grad_norm": 3.086581832740092,
      "learning_rate": 1.997512751213426e-05,
      "loss": 1.2191,
      "step": 72
    },
    {
      "epoch": 0.10583544762595143,
      "grad_norm": 3.6862840968303363,
      "learning_rate": 1.997344245674226e-05,
      "loss": 1.2892,
      "step": 73
    },
    {
      "epoch": 0.10728524827836172,
      "grad_norm": 3.5355963749967523,
      "learning_rate": 1.9971702253102856e-05,
      "loss": 1.1924,
      "step": 74
    },
    {
      "epoch": 0.10873504893077202,
      "grad_norm": 2.448279059054822,
      "learning_rate": 1.9969906910838516e-05,
      "loss": 1.297,
      "step": 75
    },
    {
      "epoch": 0.11018484958318231,
      "grad_norm": 3.6585930927615276,
      "learning_rate": 1.99680564398766e-05,
      "loss": 1.347,
      "step": 76
    },
    {
      "epoch": 0.1116346502355926,
      "grad_norm": 3.35838769965714,
      "learning_rate": 1.9966150850449308e-05,
      "loss": 1.3417,
      "step": 77
    },
    {
      "epoch": 0.1130844508880029,
      "grad_norm": 3.009005634725042,
      "learning_rate": 1.9964190153093613e-05,
      "loss": 1.2716,
      "step": 78
    },
    {
      "epoch": 0.11453425154041319,
      "grad_norm": 3.0830123490946177,
      "learning_rate": 1.9962174358651215e-05,
      "loss": 1.3209,
      "step": 79
    },
    {
      "epoch": 0.11598405219282348,
      "grad_norm": 2.537108547149772,
      "learning_rate": 1.996010347826847e-05,
      "loss": 1.2229,
      "step": 80
    },
    {
      "epoch": 0.11743385284523378,
      "grad_norm": 3.5596824299433925,
      "learning_rate": 1.9957977523396323e-05,
      "loss": 1.2556,
      "step": 81
    },
    {
      "epoch": 0.11888365349764407,
      "grad_norm": 2.54307900070356,
      "learning_rate": 1.9955796505790266e-05,
      "loss": 1.262,
      "step": 82
    },
    {
      "epoch": 0.12033345415005436,
      "grad_norm": 2.243591877483027,
      "learning_rate": 1.9953560437510263e-05,
      "loss": 1.1412,
      "step": 83
    },
    {
      "epoch": 0.12178325480246466,
      "grad_norm": 2.444842284673475,
      "learning_rate": 1.9951269330920667e-05,
      "loss": 1.2856,
      "step": 84
    },
    {
      "epoch": 0.12178325480246466,
      "eval_loss": 2.051546096801758,
      "eval_runtime": 716.5341,
      "eval_samples_per_second": 24.638,
      "eval_steps_per_second": 0.385,
      "step": 84
    },
    {
      "epoch": 0.12323305545487495,
      "grad_norm": 2.423011984312384,
      "learning_rate": 1.9948923198690176e-05,
      "loss": 1.3634,
      "step": 85
    },
    {
      "epoch": 0.12468285610728524,
      "grad_norm": 3.097828032758144,
      "learning_rate": 1.9946522053791755e-05,
      "loss": 1.304,
      "step": 86
    },
    {
      "epoch": 0.12613265675969554,
      "grad_norm": 2.626279011649116,
      "learning_rate": 1.9944065909502557e-05,
      "loss": 1.2304,
      "step": 87
    },
    {
      "epoch": 0.12758245741210583,
      "grad_norm": 2.6070710135252226,
      "learning_rate": 1.9941554779403854e-05,
      "loss": 1.4751,
      "step": 88
    },
    {
      "epoch": 0.12903225806451613,
      "grad_norm": 2.471273744932804,
      "learning_rate": 1.9938988677380963e-05,
      "loss": 1.2688,
      "step": 89
    },
    {
      "epoch": 0.13048205871692642,
      "grad_norm": 2.615335911781161,
      "learning_rate": 1.993636761762317e-05,
      "loss": 1.3525,
      "step": 90
    },
    {
      "epoch": 0.1319318593693367,
      "grad_norm": 2.7800477996075914,
      "learning_rate": 1.9933691614623658e-05,
      "loss": 1.3366,
      "step": 91
    },
    {
      "epoch": 0.133381660021747,
      "grad_norm": 2.521205349596145,
      "learning_rate": 1.9930960683179404e-05,
      "loss": 1.381,
      "step": 92
    },
    {
      "epoch": 0.1348314606741573,
      "grad_norm": 2.6918042954480104,
      "learning_rate": 1.992817483839112e-05,
      "loss": 1.2665,
      "step": 93
    },
    {
      "epoch": 0.1362812613265676,
      "grad_norm": 2.6676771081025765,
      "learning_rate": 1.992533409566317e-05,
      "loss": 1.2721,
      "step": 94
    },
    {
      "epoch": 0.1377310619789779,
      "grad_norm": 2.86149323943376,
      "learning_rate": 1.992243847070346e-05,
      "loss": 1.3342,
      "step": 95
    },
    {
      "epoch": 0.13918086263138818,
      "grad_norm": 2.6222650614251672,
      "learning_rate": 1.991948797952338e-05,
      "loss": 1.3169,
      "step": 96
    },
    {
      "epoch": 0.14063066328379847,
      "grad_norm": 2.673993530188596,
      "learning_rate": 1.9916482638437696e-05,
      "loss": 1.3406,
      "step": 97
    },
    {
      "epoch": 0.14208046393620877,
      "grad_norm": 2.657574026832722,
      "learning_rate": 1.991342246406448e-05,
      "loss": 1.2181,
      "step": 98
    },
    {
      "epoch": 0.14208046393620877,
      "eval_loss": 2.0149688720703125,
      "eval_runtime": 716.6893,
      "eval_samples_per_second": 24.633,
      "eval_steps_per_second": 0.385,
      "step": 98
    },
    {
      "epoch": 0.14353026458861906,
      "grad_norm": 3.8593009802152154,
      "learning_rate": 1.9910307473324986e-05,
      "loss": 1.3003,
      "step": 99
    },
    {
      "epoch": 0.14498006524102935,
      "grad_norm": 3.301766032041174,
      "learning_rate": 1.9907137683443593e-05,
      "loss": 1.2866,
      "step": 100
    },
    {
      "epoch": 0.14642986589343965,
      "grad_norm": 2.846017734729031,
      "learning_rate": 1.9903913111947682e-05,
      "loss": 1.2288,
      "step": 101
    },
    {
      "epoch": 0.14787966654584994,
      "grad_norm": 2.8374148083925768,
      "learning_rate": 1.990063377666755e-05,
      "loss": 1.3923,
      "step": 102
    },
    {
      "epoch": 0.14932946719826024,
      "grad_norm": 2.821632739519889,
      "learning_rate": 1.989729969573632e-05,
      "loss": 1.323,
      "step": 103
    },
    {
      "epoch": 0.15077926785067053,
      "grad_norm": 2.649958703145158,
      "learning_rate": 1.9893910887589817e-05,
      "loss": 1.2218,
      "step": 104
    },
    {
      "epoch": 0.15222906850308082,
      "grad_norm": 2.5837522172787097,
      "learning_rate": 1.9890467370966494e-05,
      "loss": 1.1986,
      "step": 105
    },
    {
      "epoch": 0.15367886915549112,
      "grad_norm": 2.314868918928458,
      "learning_rate": 1.9886969164907306e-05,
      "loss": 1.2556,
      "step": 106
    },
    {
      "epoch": 0.1551286698079014,
      "grad_norm": 2.5151330241302845,
      "learning_rate": 1.9883416288755626e-05,
      "loss": 1.3832,
      "step": 107
    },
    {
      "epoch": 0.1565784704603117,
      "grad_norm": 2.740103855505652,
      "learning_rate": 1.9879808762157106e-05,
      "loss": 1.2824,
      "step": 108
    },
    {
      "epoch": 0.158028271112722,
      "grad_norm": 2.973518709310962,
      "learning_rate": 1.987614660505961e-05,
      "loss": 1.2185,
      "step": 109
    },
    {
      "epoch": 0.1594780717651323,
      "grad_norm": 2.530344121278102,
      "learning_rate": 1.987242983771306e-05,
      "loss": 1.3199,
      "step": 110
    },
    {
      "epoch": 0.16092787241754258,
      "grad_norm": 2.3585265275962275,
      "learning_rate": 1.986865848066937e-05,
      "loss": 1.2452,
      "step": 111
    },
    {
      "epoch": 0.16237767306995288,
      "grad_norm": 2.5801305740341496,
      "learning_rate": 1.986483255478229e-05,
      "loss": 1.3249,
      "step": 112
    },
    {
      "epoch": 0.16237767306995288,
      "eval_loss": 2.0279502868652344,
      "eval_runtime": 709.573,
      "eval_samples_per_second": 24.88,
      "eval_steps_per_second": 0.389,
      "step": 112
    },
    {
      "epoch": 0.16382747372236317,
      "grad_norm": 2.7085500745397977,
      "learning_rate": 1.986095208120731e-05,
      "loss": 1.2648,
      "step": 113
    },
    {
      "epoch": 0.16527727437477346,
      "grad_norm": 2.6202580478625146,
      "learning_rate": 1.985701708140155e-05,
      "loss": 1.2166,
      "step": 114
    },
    {
      "epoch": 0.16672707502718376,
      "grad_norm": 2.7121654289364345,
      "learning_rate": 1.985302757712363e-05,
      "loss": 1.2989,
      "step": 115
    },
    {
      "epoch": 0.16817687567959405,
      "grad_norm": 2.269525342742487,
      "learning_rate": 1.9848983590433555e-05,
      "loss": 1.2264,
      "step": 116
    },
    {
      "epoch": 0.16962667633200434,
      "grad_norm": 2.974768805306543,
      "learning_rate": 1.9844885143692586e-05,
      "loss": 1.3338,
      "step": 117
    },
    {
      "epoch": 0.17107647698441464,
      "grad_norm": 2.635516716517576,
      "learning_rate": 1.984073225956313e-05,
      "loss": 1.2819,
      "step": 118
    },
    {
      "epoch": 0.17252627763682493,
      "grad_norm": 2.2451480335193663,
      "learning_rate": 1.983652496100859e-05,
      "loss": 1.2192,
      "step": 119
    },
    {
      "epoch": 0.17397607828923523,
      "grad_norm": 2.5706376196760896,
      "learning_rate": 1.983226327129327e-05,
      "loss": 1.1193,
      "step": 120
    },
    {
      "epoch": 0.17542587894164552,
      "grad_norm": 2.4611724507270742,
      "learning_rate": 1.9827947213982226e-05,
      "loss": 1.3225,
      "step": 121
    },
    {
      "epoch": 0.1768756795940558,
      "grad_norm": 2.799993795739916,
      "learning_rate": 1.982357681294114e-05,
      "loss": 1.217,
      "step": 122
    },
    {
      "epoch": 0.1783254802464661,
      "grad_norm": 2.448573555484895,
      "learning_rate": 1.981915209233618e-05,
      "loss": 1.1912,
      "step": 123
    },
    {
      "epoch": 0.1797752808988764,
      "grad_norm": 2.5519828960871185,
      "learning_rate": 1.981467307663389e-05,
      "loss": 1.2255,
      "step": 124
    },
    {
      "epoch": 0.1812250815512867,
      "grad_norm": 2.8783426390023434,
      "learning_rate": 1.981013979060103e-05,
      "loss": 1.1928,
      "step": 125
    },
    {
      "epoch": 0.182674882203697,
      "grad_norm": 3.2467527125526776,
      "learning_rate": 1.9805552259304447e-05,
      "loss": 1.3635,
      "step": 126
    },
    {
      "epoch": 0.182674882203697,
      "eval_loss": 2.0441839694976807,
      "eval_runtime": 710.3356,
      "eval_samples_per_second": 24.853,
      "eval_steps_per_second": 0.389,
      "step": 126
    },
    {
      "epoch": 0.18412468285610728,
      "grad_norm": 8.100309364402014,
      "learning_rate": 1.9800910508110934e-05,
      "loss": 1.291,
      "step": 127
    },
    {
      "epoch": 0.18557448350851757,
      "grad_norm": 2.7603692709800716,
      "learning_rate": 1.9796214562687114e-05,
      "loss": 1.2521,
      "step": 128
    },
    {
      "epoch": 0.18702428416092787,
      "grad_norm": 2.702282337073925,
      "learning_rate": 1.979146444899925e-05,
      "loss": 1.2542,
      "step": 129
    },
    {
      "epoch": 0.18847408481333816,
      "grad_norm": 3.1089681822193964,
      "learning_rate": 1.978666019331315e-05,
      "loss": 1.2533,
      "step": 130
    },
    {
      "epoch": 0.18992388546574845,
      "grad_norm": 2.740126643867791,
      "learning_rate": 1.9781801822193988e-05,
      "loss": 1.2252,
      "step": 131
    },
    {
      "epoch": 0.19137368611815875,
      "grad_norm": 3.3489712326082115,
      "learning_rate": 1.9776889362506175e-05,
      "loss": 1.2545,
      "step": 132
    },
    {
      "epoch": 0.19282348677056904,
      "grad_norm": 3.149779985790514,
      "learning_rate": 1.9771922841413205e-05,
      "loss": 1.229,
      "step": 133
    },
    {
      "epoch": 0.19427328742297933,
      "grad_norm": 2.8588085234014384,
      "learning_rate": 1.9766902286377507e-05,
      "loss": 1.2519,
      "step": 134
    },
    {
      "epoch": 0.19572308807538963,
      "grad_norm": 2.222557836479702,
      "learning_rate": 1.9761827725160293e-05,
      "loss": 1.1869,
      "step": 135
    },
    {
      "epoch": 0.19717288872779992,
      "grad_norm": 2.640066454996811,
      "learning_rate": 1.9756699185821387e-05,
      "loss": 1.2738,
      "step": 136
    },
    {
      "epoch": 0.19862268938021022,
      "grad_norm": 2.5394740555362945,
      "learning_rate": 1.9751516696719107e-05,
      "loss": 1.2495,
      "step": 137
    },
    {
      "epoch": 0.2000724900326205,
      "grad_norm": 2.325912214845598,
      "learning_rate": 1.974628028651007e-05,
      "loss": 1.3198,
      "step": 138
    },
    {
      "epoch": 0.2015222906850308,
      "grad_norm": 2.4103714238486593,
      "learning_rate": 1.9740989984149065e-05,
      "loss": 1.2133,
      "step": 139
    },
    {
      "epoch": 0.2029720913374411,
      "grad_norm": 2.3096525100612677,
      "learning_rate": 1.973564581888887e-05,
      "loss": 1.2133,
      "step": 140
    },
    {
      "epoch": 0.2029720913374411,
      "eval_loss": 2.0053794384002686,
      "eval_runtime": 714.7568,
      "eval_samples_per_second": 24.699,
      "eval_steps_per_second": 0.386,
      "step": 140
    },
    {
      "epoch": 0.2044218919898514,
      "grad_norm": 2.391167285672832,
      "learning_rate": 1.9730247820280094e-05,
      "loss": 1.1377,
      "step": 141
    },
    {
      "epoch": 0.20587169264226168,
      "grad_norm": 2.404924110025378,
      "learning_rate": 1.9724796018171022e-05,
      "loss": 1.2287,
      "step": 142
    },
    {
      "epoch": 0.20732149329467198,
      "grad_norm": 2.4852133516784134,
      "learning_rate": 1.971929044270745e-05,
      "loss": 1.2628,
      "step": 143
    },
    {
      "epoch": 0.20877129394708227,
      "grad_norm": 3.3890608538717935,
      "learning_rate": 1.9713731124332513e-05,
      "loss": 1.305,
      "step": 144
    },
    {
      "epoch": 0.21022109459949256,
      "grad_norm": 2.931301723786055,
      "learning_rate": 1.9708118093786515e-05,
      "loss": 1.3619,
      "step": 145
    },
    {
      "epoch": 0.21167089525190286,
      "grad_norm": 2.355758826183243,
      "learning_rate": 1.9702451382106763e-05,
      "loss": 1.1684,
      "step": 146
    },
    {
      "epoch": 0.21312069590431315,
      "grad_norm": 3.165962405802141,
      "learning_rate": 1.9696731020627392e-05,
      "loss": 1.3188,
      "step": 147
    },
    {
      "epoch": 0.21457049655672344,
      "grad_norm": 2.2928077569415355,
      "learning_rate": 1.96909570409792e-05,
      "loss": 1.2364,
      "step": 148
    },
    {
      "epoch": 0.21602029720913374,
      "grad_norm": 2.9036100841184806,
      "learning_rate": 1.9685129475089462e-05,
      "loss": 1.2389,
      "step": 149
    },
    {
      "epoch": 0.21747009786154403,
      "grad_norm": 2.372563378504368,
      "learning_rate": 1.967924835518176e-05,
      "loss": 1.197,
      "step": 150
    },
    {
      "epoch": 0.21891989851395433,
      "grad_norm": 3.4574679770772074,
      "learning_rate": 1.9673313713775805e-05,
      "loss": 1.2934,
      "step": 151
    },
    {
      "epoch": 0.22036969916636462,
      "grad_norm": 2.6845468869168836,
      "learning_rate": 1.9667325583687255e-05,
      "loss": 1.1874,
      "step": 152
    },
    {
      "epoch": 0.2218194998187749,
      "grad_norm": 2.7738471768934243,
      "learning_rate": 1.9661283998027535e-05,
      "loss": 1.3529,
      "step": 153
    },
    {
      "epoch": 0.2232693004711852,
      "grad_norm": 2.211056270178631,
      "learning_rate": 1.9655188990203648e-05,
      "loss": 1.2413,
      "step": 154
    },
    {
      "epoch": 0.2232693004711852,
      "eval_loss": 2.032243251800537,
      "eval_runtime": 715.5606,
      "eval_samples_per_second": 24.672,
      "eval_steps_per_second": 0.386,
      "step": 154
    },
    {
      "epoch": 0.2247191011235955,
      "grad_norm": 2.9032461478639244,
      "learning_rate": 1.9649040593918002e-05,
      "loss": 1.3017,
      "step": 155
    },
    {
      "epoch": 0.2261689017760058,
      "grad_norm": 2.6665920811660215,
      "learning_rate": 1.9642838843168215e-05,
      "loss": 1.3031,
      "step": 156
    },
    {
      "epoch": 0.2276187024284161,
      "grad_norm": 2.3388201388065646,
      "learning_rate": 1.9636583772246926e-05,
      "loss": 1.1396,
      "step": 157
    },
    {
      "epoch": 0.22906850308082638,
      "grad_norm": 2.4021618806318203,
      "learning_rate": 1.963027541574161e-05,
      "loss": 1.1989,
      "step": 158
    },
    {
      "epoch": 0.23051830373323667,
      "grad_norm": 2.3705721741670764,
      "learning_rate": 1.962391380853439e-05,
      "loss": 1.1893,
      "step": 159
    },
    {
      "epoch": 0.23196810438564697,
      "grad_norm": 2.465141724805024,
      "learning_rate": 1.9617498985801834e-05,
      "loss": 1.2362,
      "step": 160
    },
    {
      "epoch": 0.23341790503805726,
      "grad_norm": 3.1359950269078025,
      "learning_rate": 1.9611030983014764e-05,
      "loss": 1.2995,
      "step": 161
    },
    {
      "epoch": 0.23486770569046755,
      "grad_norm": 2.3407504913494996,
      "learning_rate": 1.9604509835938066e-05,
      "loss": 1.2376,
      "step": 162
    },
    {
      "epoch": 0.23631750634287785,
      "grad_norm": 2.821142528520419,
      "learning_rate": 1.9597935580630487e-05,
      "loss": 1.2571,
      "step": 163
    },
    {
      "epoch": 0.23776730699528814,
      "grad_norm": 2.556613441591143,
      "learning_rate": 1.9591308253444437e-05,
      "loss": 1.1393,
      "step": 164
    },
    {
      "epoch": 0.23921710764769843,
      "grad_norm": 2.299837208653212,
      "learning_rate": 1.958462789102578e-05,
      "loss": 1.2674,
      "step": 165
    },
    {
      "epoch": 0.24066690830010873,
      "grad_norm": 2.720542315300884,
      "learning_rate": 1.9577894530313655e-05,
      "loss": 1.2214,
      "step": 166
    },
    {
      "epoch": 0.24211670895251902,
      "grad_norm": 2.715137097664862,
      "learning_rate": 1.9571108208540238e-05,
      "loss": 1.2895,
      "step": 167
    },
    {
      "epoch": 0.24356650960492932,
      "grad_norm": 2.2043013978046218,
      "learning_rate": 1.9564268963230564e-05,
      "loss": 1.2205,
      "step": 168
    },
    {
      "epoch": 0.24356650960492932,
      "eval_loss": 2.027660608291626,
      "eval_runtime": 712.1686,
      "eval_samples_per_second": 24.789,
      "eval_steps_per_second": 0.388,
      "step": 168
    },
    {
      "epoch": 0.2450163102573396,
      "grad_norm": 2.8130397382613745,
      "learning_rate": 1.955737683220231e-05,
      "loss": 1.2164,
      "step": 169
    },
    {
      "epoch": 0.2464661109097499,
      "grad_norm": 2.8067100585659,
      "learning_rate": 1.9550431853565577e-05,
      "loss": 1.2449,
      "step": 170
    },
    {
      "epoch": 0.2479159115621602,
      "grad_norm": 2.4444200212729807,
      "learning_rate": 1.9543434065722692e-05,
      "loss": 1.2631,
      "step": 171
    },
    {
      "epoch": 0.2493657122145705,
      "grad_norm": 2.885941379893374,
      "learning_rate": 1.9536383507367994e-05,
      "loss": 1.3282,
      "step": 172
    },
    {
      "epoch": 0.2508155128669808,
      "grad_norm": 2.3801646133226066,
      "learning_rate": 1.9529280217487617e-05,
      "loss": 1.2318,
      "step": 173
    },
    {
      "epoch": 0.2522653135193911,
      "grad_norm": 2.5087449574507334,
      "learning_rate": 1.9522124235359265e-05,
      "loss": 1.2274,
      "step": 174
    },
    {
      "epoch": 0.25371511417180137,
      "grad_norm": 2.370383123620175,
      "learning_rate": 1.9514915600552014e-05,
      "loss": 1.3315,
      "step": 175
    },
    {
      "epoch": 0.25516491482421166,
      "grad_norm": 2.403628026090708,
      "learning_rate": 1.950765435292608e-05,
      "loss": 1.2031,
      "step": 176
    },
    {
      "epoch": 0.25661471547662196,
      "grad_norm": 2.491681375290092,
      "learning_rate": 1.9500340532632606e-05,
      "loss": 1.2091,
      "step": 177
    },
    {
      "epoch": 0.25806451612903225,
      "grad_norm": 2.718348491711138,
      "learning_rate": 1.9492974180113425e-05,
      "loss": 1.1895,
      "step": 178
    },
    {
      "epoch": 0.25951431678144254,
      "grad_norm": 2.3733699519239297,
      "learning_rate": 1.9485555336100866e-05,
      "loss": 1.2879,
      "step": 179
    },
    {
      "epoch": 0.26096411743385284,
      "grad_norm": 2.603135795982347,
      "learning_rate": 1.9478084041617492e-05,
      "loss": 1.3304,
      "step": 180
    },
    {
      "epoch": 0.26241391808626313,
      "grad_norm": 2.261603937445733,
      "learning_rate": 1.94705603379759e-05,
      "loss": 1.1692,
      "step": 181
    },
    {
      "epoch": 0.2638637187386734,
      "grad_norm": 2.5680299230847057,
      "learning_rate": 1.946298426677848e-05,
      "loss": 1.2292,
      "step": 182
    },
    {
      "epoch": 0.2638637187386734,
      "eval_loss": 2.01969313621521,
      "eval_runtime": 716.543,
      "eval_samples_per_second": 24.638,
      "eval_steps_per_second": 0.385,
      "step": 182
    },
    {
      "epoch": 0.2653135193910837,
      "grad_norm": 2.5199657916183336,
      "learning_rate": 1.94553558699172e-05,
      "loss": 1.3177,
      "step": 183
    },
    {
      "epoch": 0.266763320043494,
      "grad_norm": 2.5371260783689857,
      "learning_rate": 1.944767518957334e-05,
      "loss": 1.2573,
      "step": 184
    },
    {
      "epoch": 0.2682131206959043,
      "grad_norm": 2.4498963846756228,
      "learning_rate": 1.9439942268217318e-05,
      "loss": 1.24,
      "step": 185
    },
    {
      "epoch": 0.2696629213483146,
      "grad_norm": 2.3057335604979254,
      "learning_rate": 1.943215714860838e-05,
      "loss": 1.1961,
      "step": 186
    },
    {
      "epoch": 0.2711127220007249,
      "grad_norm": 2.834310299228459,
      "learning_rate": 1.942431987379443e-05,
      "loss": 1.2122,
      "step": 187
    },
    {
      "epoch": 0.2725625226531352,
      "grad_norm": 2.070254567185668,
      "learning_rate": 1.941643048711175e-05,
      "loss": 1.2458,
      "step": 188
    },
    {
      "epoch": 0.2740123233055455,
      "grad_norm": 2.2466884957355546,
      "learning_rate": 1.9408489032184784e-05,
      "loss": 1.1571,
      "step": 189
    },
    {
      "epoch": 0.2754621239579558,
      "grad_norm": 2.5636925869052285,
      "learning_rate": 1.9400495552925887e-05,
      "loss": 1.1661,
      "step": 190
    },
    {
      "epoch": 0.27691192461036607,
      "grad_norm": 2.662319246292031,
      "learning_rate": 1.9392450093535074e-05,
      "loss": 1.3203,
      "step": 191
    },
    {
      "epoch": 0.27836172526277636,
      "grad_norm": 2.313908196313624,
      "learning_rate": 1.93843526984998e-05,
      "loss": 1.2685,
      "step": 192
    },
    {
      "epoch": 0.27981152591518665,
      "grad_norm": 2.341133180212495,
      "learning_rate": 1.9376203412594683e-05,
      "loss": 1.3031,
      "step": 193
    },
    {
      "epoch": 0.28126132656759695,
      "grad_norm": 2.9432243730901004,
      "learning_rate": 1.936800228088128e-05,
      "loss": 1.2407,
      "step": 194
    },
    {
      "epoch": 0.28271112722000724,
      "grad_norm": 2.445194048578033,
      "learning_rate": 1.935974934870783e-05,
      "loss": 1.1704,
      "step": 195
    },
    {
      "epoch": 0.28416092787241753,
      "grad_norm": 2.5515347336473977,
      "learning_rate": 1.9351444661709e-05,
      "loss": 1.2047,
      "step": 196
    },
    {
      "epoch": 0.28416092787241753,
      "eval_loss": 2.016186475753784,
      "eval_runtime": 716.0699,
      "eval_samples_per_second": 24.654,
      "eval_steps_per_second": 0.385,
      "step": 196
    },
    {
      "epoch": 0.28561072852482783,
      "grad_norm": 2.5623521157065654,
      "learning_rate": 1.9343088265805642e-05,
      "loss": 1.1911,
      "step": 197
    },
    {
      "epoch": 0.2870605291772381,
      "grad_norm": 2.035294740675946,
      "learning_rate": 1.9334680207204525e-05,
      "loss": 1.1487,
      "step": 198
    },
    {
      "epoch": 0.2885103298296484,
      "grad_norm": 2.1932399272399055,
      "learning_rate": 1.932622053239809e-05,
      "loss": 1.3344,
      "step": 199
    },
    {
      "epoch": 0.2899601304820587,
      "grad_norm": 2.2590202559155976,
      "learning_rate": 1.9317709288164198e-05,
      "loss": 1.1094,
      "step": 200
    },
    {
      "epoch": 0.291409931134469,
      "grad_norm": 2.203537652576977,
      "learning_rate": 1.930914652156585e-05,
      "loss": 1.2379,
      "step": 201
    },
    {
      "epoch": 0.2928597317868793,
      "grad_norm": 2.2580019067205295,
      "learning_rate": 1.930053227995095e-05,
      "loss": 1.1993,
      "step": 202
    },
    {
      "epoch": 0.2943095324392896,
      "grad_norm": 2.5973320927801646,
      "learning_rate": 1.929186661095203e-05,
      "loss": 1.2236,
      "step": 203
    },
    {
      "epoch": 0.2957593330916999,
      "grad_norm": 2.4523994495137806,
      "learning_rate": 1.9283149562485997e-05,
      "loss": 1.2818,
      "step": 204
    },
    {
      "epoch": 0.2972091337441102,
      "grad_norm": 2.5966445881746707,
      "learning_rate": 1.927438118275385e-05,
      "loss": 1.2673,
      "step": 205
    },
    {
      "epoch": 0.29865893439652047,
      "grad_norm": 2.4441252210590196,
      "learning_rate": 1.9265561520240434e-05,
      "loss": 1.1934,
      "step": 206
    },
    {
      "epoch": 0.30010873504893076,
      "grad_norm": 2.4192792506391907,
      "learning_rate": 1.925669062371415e-05,
      "loss": 1.2354,
      "step": 207
    },
    {
      "epoch": 0.30155853570134106,
      "grad_norm": 2.5207767010232907,
      "learning_rate": 1.924776854222672e-05,
      "loss": 1.1952,
      "step": 208
    },
    {
      "epoch": 0.30300833635375135,
      "grad_norm": 2.41838421422327,
      "learning_rate": 1.9238795325112867e-05,
      "loss": 1.2296,
      "step": 209
    },
    {
      "epoch": 0.30445813700616164,
      "grad_norm": 2.3039411142163115,
      "learning_rate": 1.9229771021990094e-05,
      "loss": 1.2105,
      "step": 210
    },
    {
      "epoch": 0.30445813700616164,
      "eval_loss": 2.021575927734375,
      "eval_runtime": 716.2969,
      "eval_samples_per_second": 24.646,
      "eval_steps_per_second": 0.385,
      "step": 210
    },
    {
      "epoch": 0.30590793765857194,
      "grad_norm": 2.4529224433188457,
      "learning_rate": 1.922069568275837e-05,
      "loss": 1.2347,
      "step": 211
    },
    {
      "epoch": 0.30735773831098223,
      "grad_norm": 2.4592105152550974,
      "learning_rate": 1.9211569357599868e-05,
      "loss": 1.1877,
      "step": 212
    },
    {
      "epoch": 0.3088075389633925,
      "grad_norm": 2.464162692134468,
      "learning_rate": 1.92023920969787e-05,
      "loss": 1.2716,
      "step": 213
    },
    {
      "epoch": 0.3102573396158028,
      "grad_norm": 2.219750042871641,
      "learning_rate": 1.9193163951640608e-05,
      "loss": 1.2256,
      "step": 214
    },
    {
      "epoch": 0.3117071402682131,
      "grad_norm": 2.3946063559221704,
      "learning_rate": 1.9183884972612718e-05,
      "loss": 1.2723,
      "step": 215
    },
    {
      "epoch": 0.3131569409206234,
      "grad_norm": 2.4836009406889943,
      "learning_rate": 1.9174555211203235e-05,
      "loss": 1.2202,
      "step": 216
    },
    {
      "epoch": 0.3146067415730337,
      "grad_norm": 2.2369185060358854,
      "learning_rate": 1.9165174719001162e-05,
      "loss": 1.2542,
      "step": 217
    },
    {
      "epoch": 0.316056542225444,
      "grad_norm": 2.394823954722132,
      "learning_rate": 1.9155743547876026e-05,
      "loss": 1.2572,
      "step": 218
    },
    {
      "epoch": 0.3175063428778543,
      "grad_norm": 3.104612085551012,
      "learning_rate": 1.9146261749977578e-05,
      "loss": 1.3067,
      "step": 219
    },
    {
      "epoch": 0.3189561435302646,
      "grad_norm": 2.3721820547760952,
      "learning_rate": 1.9136729377735514e-05,
      "loss": 1.2576,
      "step": 220
    },
    {
      "epoch": 0.3204059441826749,
      "grad_norm": 2.2956296446459734,
      "learning_rate": 1.9127146483859176e-05,
      "loss": 1.2722,
      "step": 221
    },
    {
      "epoch": 0.32185574483508517,
      "grad_norm": 2.3711808828082703,
      "learning_rate": 1.9117513121337273e-05,
      "loss": 1.2467,
      "step": 222
    },
    {
      "epoch": 0.32330554548749546,
      "grad_norm": 2.3139714936747016,
      "learning_rate": 1.9107829343437575e-05,
      "loss": 1.2483,
      "step": 223
    },
    {
      "epoch": 0.32475534613990575,
      "grad_norm": 2.3449813033437596,
      "learning_rate": 1.9098095203706625e-05,
      "loss": 1.1874,
      "step": 224
    },
    {
      "epoch": 0.32475534613990575,
      "eval_loss": 2.003979206085205,
      "eval_runtime": 712.4984,
      "eval_samples_per_second": 24.778,
      "eval_steps_per_second": 0.387,
      "step": 224
    },
    {
      "epoch": 0.32620514679231605,
      "grad_norm": 2.5348991321288845,
      "learning_rate": 1.9088310755969447e-05,
      "loss": 1.2421,
      "step": 225
    },
    {
      "epoch": 0.32765494744472634,
      "grad_norm": 2.3904744415704444,
      "learning_rate": 1.9078476054329237e-05,
      "loss": 1.2513,
      "step": 226
    },
    {
      "epoch": 0.32910474809713663,
      "grad_norm": 3.281811699518377,
      "learning_rate": 1.906859115316708e-05,
      "loss": 1.1977,
      "step": 227
    },
    {
      "epoch": 0.33055454874954693,
      "grad_norm": 2.3534265723608403,
      "learning_rate": 1.905865610714162e-05,
      "loss": 1.205,
      "step": 228
    },
    {
      "epoch": 0.3320043494019572,
      "grad_norm": 2.495818433885312,
      "learning_rate": 1.9048670971188805e-05,
      "loss": 1.1764,
      "step": 229
    },
    {
      "epoch": 0.3334541500543675,
      "grad_norm": 2.0801481539434103,
      "learning_rate": 1.9038635800521538e-05,
      "loss": 1.1663,
      "step": 230
    },
    {
      "epoch": 0.3349039507067778,
      "grad_norm": 2.0759288017658237,
      "learning_rate": 1.902855065062939e-05,
      "loss": 1.2117,
      "step": 231
    },
    {
      "epoch": 0.3363537513591881,
      "grad_norm": 2.8282939149622335,
      "learning_rate": 1.9018415577278295e-05,
      "loss": 1.3016,
      "step": 232
    },
    {
      "epoch": 0.3378035520115984,
      "grad_norm": 2.347979730470765,
      "learning_rate": 1.9008230636510244e-05,
      "loss": 1.1996,
      "step": 233
    },
    {
      "epoch": 0.3392533526640087,
      "grad_norm": 2.3496813048298875,
      "learning_rate": 1.8997995884642968e-05,
      "loss": 1.2773,
      "step": 234
    },
    {
      "epoch": 0.340703153316419,
      "grad_norm": 2.688004856175403,
      "learning_rate": 1.898771137826962e-05,
      "loss": 1.2736,
      "step": 235
    },
    {
      "epoch": 0.3421529539688293,
      "grad_norm": 2.319278611769484,
      "learning_rate": 1.8977377174258485e-05,
      "loss": 1.3212,
      "step": 236
    },
    {
      "epoch": 0.34360275462123957,
      "grad_norm": 2.1735594155443922,
      "learning_rate": 1.896699332975264e-05,
      "loss": 1.1559,
      "step": 237
    },
    {
      "epoch": 0.34505255527364986,
      "grad_norm": 2.2104275710129446,
      "learning_rate": 1.895655990216966e-05,
      "loss": 1.2081,
      "step": 238
    },
    {
      "epoch": 0.34505255527364986,
      "eval_loss": 2.0066256523132324,
      "eval_runtime": 719.7656,
      "eval_samples_per_second": 24.527,
      "eval_steps_per_second": 0.383,
      "step": 238
    },
    {
      "epoch": 0.34650235592606016,
      "grad_norm": 2.584856734083762,
      "learning_rate": 1.894607694920128e-05,
      "loss": 1.2883,
      "step": 239
    },
    {
      "epoch": 0.34795215657847045,
      "grad_norm": 2.2842535401392667,
      "learning_rate": 1.893554452881309e-05,
      "loss": 1.2681,
      "step": 240
    },
    {
      "epoch": 0.34940195723088074,
      "grad_norm": 2.4105767674218845,
      "learning_rate": 1.8924962699244216e-05,
      "loss": 1.2003,
      "step": 241
    },
    {
      "epoch": 0.35085175788329104,
      "grad_norm": 2.465986681467506,
      "learning_rate": 1.8914331519006985e-05,
      "loss": 1.2995,
      "step": 242
    },
    {
      "epoch": 0.35230155853570133,
      "grad_norm": 2.2957015953664595,
      "learning_rate": 1.8903651046886614e-05,
      "loss": 1.1952,
      "step": 243
    },
    {
      "epoch": 0.3537513591881116,
      "grad_norm": 2.4744430736249936,
      "learning_rate": 1.8892921341940873e-05,
      "loss": 1.1697,
      "step": 244
    },
    {
      "epoch": 0.3552011598405219,
      "grad_norm": 2.3194837643045148,
      "learning_rate": 1.8882142463499778e-05,
      "loss": 1.2855,
      "step": 245
    },
    {
      "epoch": 0.3566509604929322,
      "grad_norm": 2.349066845133006,
      "learning_rate": 1.8871314471165233e-05,
      "loss": 1.2822,
      "step": 246
    },
    {
      "epoch": 0.3581007611453425,
      "grad_norm": 2.278513796592711,
      "learning_rate": 1.8860437424810734e-05,
      "loss": 1.2772,
      "step": 247
    },
    {
      "epoch": 0.3595505617977528,
      "grad_norm": 2.1811177508003476,
      "learning_rate": 1.8849511384581015e-05,
      "loss": 1.2458,
      "step": 248
    },
    {
      "epoch": 0.3610003624501631,
      "grad_norm": 2.3698993895933302,
      "learning_rate": 1.8838536410891723e-05,
      "loss": 1.1253,
      "step": 249
    },
    {
      "epoch": 0.3624501631025734,
      "grad_norm": 2.6746777916268494,
      "learning_rate": 1.8827512564429076e-05,
      "loss": 1.2213,
      "step": 250
    },
    {
      "epoch": 0.3638999637549837,
      "grad_norm": 2.117952540048079,
      "learning_rate": 1.8816439906149543e-05,
      "loss": 1.1907,
      "step": 251
    },
    {
      "epoch": 0.365349764407394,
      "grad_norm": 2.253381492179174,
      "learning_rate": 1.8805318497279498e-05,
      "loss": 1.2437,
      "step": 252
    },
    {
      "epoch": 0.365349764407394,
      "eval_loss": 2.0089666843414307,
      "eval_runtime": 713.9777,
      "eval_samples_per_second": 24.726,
      "eval_steps_per_second": 0.387,
      "step": 252
    },
    {
      "epoch": 0.36679956505980427,
      "grad_norm": 2.4243289978651554,
      "learning_rate": 1.8794148399314884e-05,
      "loss": 1.2385,
      "step": 253
    },
    {
      "epoch": 0.36824936571221456,
      "grad_norm": 2.4423981823472176,
      "learning_rate": 1.8782929674020856e-05,
      "loss": 1.2935,
      "step": 254
    },
    {
      "epoch": 0.36969916636462485,
      "grad_norm": 2.402680219879031,
      "learning_rate": 1.877166238343147e-05,
      "loss": 1.3177,
      "step": 255
    },
    {
      "epoch": 0.37114896701703515,
      "grad_norm": 2.396729276257348,
      "learning_rate": 1.8760346589849323e-05,
      "loss": 1.2928,
      "step": 256
    },
    {
      "epoch": 0.37259876766944544,
      "grad_norm": 2.3260256072455476,
      "learning_rate": 1.8748982355845206e-05,
      "loss": 1.1893,
      "step": 257
    },
    {
      "epoch": 0.37404856832185573,
      "grad_norm": 2.4306729856247293,
      "learning_rate": 1.8737569744257754e-05,
      "loss": 1.2428,
      "step": 258
    },
    {
      "epoch": 0.37549836897426603,
      "grad_norm": 2.197910589391717,
      "learning_rate": 1.8726108818193124e-05,
      "loss": 1.3108,
      "step": 259
    },
    {
      "epoch": 0.3769481696266763,
      "grad_norm": 2.268397849281069,
      "learning_rate": 1.871459964102461e-05,
      "loss": 1.232,
      "step": 260
    },
    {
      "epoch": 0.3783979702790866,
      "grad_norm": 2.220929775157337,
      "learning_rate": 1.870304227639232e-05,
      "loss": 1.2202,
      "step": 261
    },
    {
      "epoch": 0.3798477709314969,
      "grad_norm": 2.2570419065330487,
      "learning_rate": 1.8691436788202823e-05,
      "loss": 1.1832,
      "step": 262
    },
    {
      "epoch": 0.3812975715839072,
      "grad_norm": 2.5438998641449504,
      "learning_rate": 1.8679783240628776e-05,
      "loss": 1.1904,
      "step": 263
    },
    {
      "epoch": 0.3827473722363175,
      "grad_norm": 2.4510584690373296,
      "learning_rate": 1.8668081698108576e-05,
      "loss": 1.239,
      "step": 264
    },
    {
      "epoch": 0.3841971728887278,
      "grad_norm": 2.227874164041286,
      "learning_rate": 1.8656332225346025e-05,
      "loss": 1.1998,
      "step": 265
    },
    {
      "epoch": 0.3856469735411381,
      "grad_norm": 2.223595341606881,
      "learning_rate": 1.8644534887309947e-05,
      "loss": 1.2595,
      "step": 266
    },
    {
      "epoch": 0.3856469735411381,
      "eval_loss": 2.0087993144989014,
      "eval_runtime": 713.8642,
      "eval_samples_per_second": 24.73,
      "eval_steps_per_second": 0.387,
      "step": 266
    },
    {
      "epoch": 0.3870967741935484,
      "grad_norm": 2.4042444102748632,
      "learning_rate": 1.863268974923383e-05,
      "loss": 1.2439,
      "step": 267
    },
    {
      "epoch": 0.38854657484595867,
      "grad_norm": 2.6112567084266787,
      "learning_rate": 1.8620796876615494e-05,
      "loss": 1.1048,
      "step": 268
    },
    {
      "epoch": 0.38999637549836896,
      "grad_norm": 2.644255870793031,
      "learning_rate": 1.8608856335216684e-05,
      "loss": 1.2906,
      "step": 269
    },
    {
      "epoch": 0.39144617615077926,
      "grad_norm": 2.7214305529280263,
      "learning_rate": 1.859686819106274e-05,
      "loss": 1.243,
      "step": 270
    },
    {
      "epoch": 0.39289597680318955,
      "grad_norm": 2.5150483457172172,
      "learning_rate": 1.8584832510442226e-05,
      "loss": 1.2055,
      "step": 271
    },
    {
      "epoch": 0.39434577745559984,
      "grad_norm": 2.385740943275113,
      "learning_rate": 1.8572749359906556e-05,
      "loss": 1.2034,
      "step": 272
    },
    {
      "epoch": 0.39579557810801014,
      "grad_norm": 2.382075892282483,
      "learning_rate": 1.8560618806269625e-05,
      "loss": 1.2246,
      "step": 273
    },
    {
      "epoch": 0.39724537876042043,
      "grad_norm": 2.4227217015044338,
      "learning_rate": 1.8548440916607454e-05,
      "loss": 1.2427,
      "step": 274
    },
    {
      "epoch": 0.3986951794128307,
      "grad_norm": 2.9972728091479826,
      "learning_rate": 1.8536215758257797e-05,
      "loss": 1.2079,
      "step": 275
    },
    {
      "epoch": 0.400144980065241,
      "grad_norm": 2.6689506511983536,
      "learning_rate": 1.8523943398819786e-05,
      "loss": 1.2883,
      "step": 276
    },
    {
      "epoch": 0.4015947807176513,
      "grad_norm": 2.3055782958086746,
      "learning_rate": 1.8511623906153554e-05,
      "loss": 1.2419,
      "step": 277
    },
    {
      "epoch": 0.4030445813700616,
      "grad_norm": 2.330479854689483,
      "learning_rate": 1.8499257348379857e-05,
      "loss": 1.2073,
      "step": 278
    },
    {
      "epoch": 0.4044943820224719,
      "grad_norm": 2.864666393669899,
      "learning_rate": 1.8486843793879697e-05,
      "loss": 1.2853,
      "step": 279
    },
    {
      "epoch": 0.4059441826748822,
      "grad_norm": 2.668588109461095,
      "learning_rate": 1.8474383311293943e-05,
      "loss": 1.286,
      "step": 280
    },
    {
      "epoch": 0.4059441826748822,
      "eval_loss": 2.000519275665283,
      "eval_runtime": 717.5373,
      "eval_samples_per_second": 24.604,
      "eval_steps_per_second": 0.385,
      "step": 280
    },
    {
      "epoch": 0.4073939833272925,
      "grad_norm": 2.6379903329270444,
      "learning_rate": 1.8461875969522956e-05,
      "loss": 1.1249,
      "step": 281
    },
    {
      "epoch": 0.4088437839797028,
      "grad_norm": 2.4907328503120527,
      "learning_rate": 1.8449321837726207e-05,
      "loss": 1.2542,
      "step": 282
    },
    {
      "epoch": 0.4102935846321131,
      "grad_norm": 2.4405089675298295,
      "learning_rate": 1.8436720985321885e-05,
      "loss": 1.1507,
      "step": 283
    },
    {
      "epoch": 0.41174338528452337,
      "grad_norm": 2.3901031159134485,
      "learning_rate": 1.842407348198653e-05,
      "loss": 1.193,
      "step": 284
    },
    {
      "epoch": 0.41319318593693366,
      "grad_norm": 2.168110515412198,
      "learning_rate": 1.8411379397654637e-05,
      "loss": 1.1848,
      "step": 285
    },
    {
      "epoch": 0.41464298658934395,
      "grad_norm": 2.52692901768871,
      "learning_rate": 1.8398638802518263e-05,
      "loss": 1.1762,
      "step": 286
    },
    {
      "epoch": 0.41609278724175425,
      "grad_norm": 2.1586963455592714,
      "learning_rate": 1.8385851767026656e-05,
      "loss": 1.1968,
      "step": 287
    },
    {
      "epoch": 0.41754258789416454,
      "grad_norm": 2.210697644855524,
      "learning_rate": 1.8373018361885856e-05,
      "loss": 1.2127,
      "step": 288
    },
    {
      "epoch": 0.41899238854657483,
      "grad_norm": 2.3261812221804994,
      "learning_rate": 1.8360138658058302e-05,
      "loss": 1.2233,
      "step": 289
    },
    {
      "epoch": 0.42044218919898513,
      "grad_norm": 2.269871376821584,
      "learning_rate": 1.8347212726762436e-05,
      "loss": 1.2976,
      "step": 290
    },
    {
      "epoch": 0.4218919898513954,
      "grad_norm": 2.2564260867493537,
      "learning_rate": 1.833424063947233e-05,
      "loss": 1.1639,
      "step": 291
    },
    {
      "epoch": 0.4233417905038057,
      "grad_norm": 2.4404563732338924,
      "learning_rate": 1.832122246791726e-05,
      "loss": 1.2327,
      "step": 292
    },
    {
      "epoch": 0.424791591156216,
      "grad_norm": 2.1321025410505876,
      "learning_rate": 1.8308158284081336e-05,
      "loss": 1.1994,
      "step": 293
    },
    {
      "epoch": 0.4262413918086263,
      "grad_norm": 2.6681361632347995,
      "learning_rate": 1.8295048160203087e-05,
      "loss": 1.1886,
      "step": 294
    },
    {
      "epoch": 0.4262413918086263,
      "eval_loss": 1.9848660230636597,
      "eval_runtime": 716.375,
      "eval_samples_per_second": 24.644,
      "eval_steps_per_second": 0.385,
      "step": 294
    },
    {
      "epoch": 0.4276911924610366,
      "grad_norm": 2.2139831739583133,
      "learning_rate": 1.8281892168775076e-05,
      "loss": 1.2248,
      "step": 295
    },
    {
      "epoch": 0.4291409931134469,
      "grad_norm": 2.164222132073749,
      "learning_rate": 1.8268690382543484e-05,
      "loss": 1.1648,
      "step": 296
    },
    {
      "epoch": 0.4305907937658572,
      "grad_norm": 2.556309504162271,
      "learning_rate": 1.8255442874507718e-05,
      "loss": 1.2023,
      "step": 297
    },
    {
      "epoch": 0.4320405944182675,
      "grad_norm": 2.2103665894506754,
      "learning_rate": 1.8242149717919993e-05,
      "loss": 1.2041,
      "step": 298
    },
    {
      "epoch": 0.43349039507067777,
      "grad_norm": 2.323485478062366,
      "learning_rate": 1.8228810986284965e-05,
      "loss": 1.2315,
      "step": 299
    },
    {
      "epoch": 0.43494019572308806,
      "grad_norm": 2.212911752094372,
      "learning_rate": 1.8215426753359268e-05,
      "loss": 1.2034,
      "step": 300
    },
    {
      "epoch": 0.43638999637549836,
      "grad_norm": 2.407312576528732,
      "learning_rate": 1.8201997093151152e-05,
      "loss": 1.2094,
      "step": 301
    },
    {
      "epoch": 0.43783979702790865,
      "grad_norm": 3.0462035880381255,
      "learning_rate": 1.8188522079920056e-05,
      "loss": 1.237,
      "step": 302
    },
    {
      "epoch": 0.43928959768031894,
      "grad_norm": 2.322808504857698,
      "learning_rate": 1.81750017881762e-05,
      "loss": 1.2477,
      "step": 303
    },
    {
      "epoch": 0.44073939833272924,
      "grad_norm": 2.445790702003767,
      "learning_rate": 1.816143629268016e-05,
      "loss": 1.3042,
      "step": 304
    },
    {
      "epoch": 0.44218919898513953,
      "grad_norm": 2.4881729799369,
      "learning_rate": 1.8147825668442482e-05,
      "loss": 1.1552,
      "step": 305
    },
    {
      "epoch": 0.4436389996375498,
      "grad_norm": 2.665893567084258,
      "learning_rate": 1.8134169990723246e-05,
      "loss": 1.2082,
      "step": 306
    },
    {
      "epoch": 0.4450888002899601,
      "grad_norm": 2.7429387866903867,
      "learning_rate": 1.8120469335031653e-05,
      "loss": 1.2368,
      "step": 307
    },
    {
      "epoch": 0.4465386009423704,
      "grad_norm": 2.6051004132852404,
      "learning_rate": 1.8106723777125614e-05,
      "loss": 1.2376,
      "step": 308
    },
    {
      "epoch": 0.4465386009423704,
      "eval_loss": 2.0027413368225098,
      "eval_runtime": 711.0849,
      "eval_samples_per_second": 24.827,
      "eval_steps_per_second": 0.388,
      "step": 308
    },
    {
      "epoch": 0.4479884015947807,
      "grad_norm": 2.0863176180303022,
      "learning_rate": 1.8092933393011325e-05,
      "loss": 1.2543,
      "step": 309
    },
    {
      "epoch": 0.449438202247191,
      "grad_norm": 2.8115611289920652,
      "learning_rate": 1.8079098258942846e-05,
      "loss": 1.2103,
      "step": 310
    },
    {
      "epoch": 0.4508880028996013,
      "grad_norm": 2.8369584917302597,
      "learning_rate": 1.8065218451421693e-05,
      "loss": 1.2365,
      "step": 311
    },
    {
      "epoch": 0.4523378035520116,
      "grad_norm": 2.3371897993999853,
      "learning_rate": 1.8051294047196393e-05,
      "loss": 1.2052,
      "step": 312
    },
    {
      "epoch": 0.4537876042044219,
      "grad_norm": 2.559553290401871,
      "learning_rate": 1.8037325123262074e-05,
      "loss": 1.1785,
      "step": 313
    },
    {
      "epoch": 0.4552374048568322,
      "grad_norm": 2.8783954770622393,
      "learning_rate": 1.8023311756860036e-05,
      "loss": 1.201,
      "step": 314
    },
    {
      "epoch": 0.45668720550924247,
      "grad_norm": 2.1924543240708387,
      "learning_rate": 1.8009254025477333e-05,
      "loss": 1.1785,
      "step": 315
    },
    {
      "epoch": 0.45813700616165276,
      "grad_norm": 2.6573851379174336,
      "learning_rate": 1.799515200684632e-05,
      "loss": 1.2181,
      "step": 316
    },
    {
      "epoch": 0.45958680681406305,
      "grad_norm": 2.3806677887148546,
      "learning_rate": 1.798100577894425e-05,
      "loss": 1.2181,
      "step": 317
    },
    {
      "epoch": 0.46103660746647335,
      "grad_norm": 2.4983147508026837,
      "learning_rate": 1.796681541999283e-05,
      "loss": 1.2565,
      "step": 318
    },
    {
      "epoch": 0.46248640811888364,
      "grad_norm": 2.4192445662254944,
      "learning_rate": 1.7952581008457783e-05,
      "loss": 1.2031,
      "step": 319
    },
    {
      "epoch": 0.46393620877129393,
      "grad_norm": 2.2785399538519147,
      "learning_rate": 1.793830262304843e-05,
      "loss": 1.0891,
      "step": 320
    },
    {
      "epoch": 0.4653860094237042,
      "grad_norm": 2.1780582182151713,
      "learning_rate": 1.7923980342717244e-05,
      "loss": 1.1708,
      "step": 321
    },
    {
      "epoch": 0.4668358100761145,
      "grad_norm": 2.173213036433812,
      "learning_rate": 1.7909614246659406e-05,
      "loss": 1.1784,
      "step": 322
    },
    {
      "epoch": 0.4668358100761145,
      "eval_loss": 2.0089244842529297,
      "eval_runtime": 714.9403,
      "eval_samples_per_second": 24.693,
      "eval_steps_per_second": 0.386,
      "step": 322
    },
    {
      "epoch": 0.4682856107285248,
      "grad_norm": 2.3944251055071977,
      "learning_rate": 1.7895204414312395e-05,
      "loss": 1.1972,
      "step": 323
    },
    {
      "epoch": 0.4697354113809351,
      "grad_norm": 2.395979308507948,
      "learning_rate": 1.7880750925355516e-05,
      "loss": 1.1666,
      "step": 324
    },
    {
      "epoch": 0.4711852120333454,
      "grad_norm": 2.357433825786659,
      "learning_rate": 1.7866253859709476e-05,
      "loss": 1.2056,
      "step": 325
    },
    {
      "epoch": 0.4726350126857557,
      "grad_norm": 2.5434968669586553,
      "learning_rate": 1.785171329753594e-05,
      "loss": 1.2169,
      "step": 326
    },
    {
      "epoch": 0.474084813338166,
      "grad_norm": 2.4888586925181304,
      "learning_rate": 1.783712931923709e-05,
      "loss": 1.188,
      "step": 327
    },
    {
      "epoch": 0.4755346139905763,
      "grad_norm": 1.966561200904472,
      "learning_rate": 1.7822502005455176e-05,
      "loss": 1.1953,
      "step": 328
    },
    {
      "epoch": 0.4769844146429866,
      "grad_norm": 2.616672502491357,
      "learning_rate": 1.780783143707207e-05,
      "loss": 1.155,
      "step": 329
    },
    {
      "epoch": 0.47843421529539687,
      "grad_norm": 2.615441357131915,
      "learning_rate": 1.7793117695208828e-05,
      "loss": 1.2002,
      "step": 330
    },
    {
      "epoch": 0.47988401594780716,
      "grad_norm": 2.0841782004825875,
      "learning_rate": 1.7778360861225224e-05,
      "loss": 1.2243,
      "step": 331
    },
    {
      "epoch": 0.48133381660021746,
      "grad_norm": 2.2306853560307527,
      "learning_rate": 1.7763561016719323e-05,
      "loss": 1.1294,
      "step": 332
    },
    {
      "epoch": 0.48278361725262775,
      "grad_norm": 2.874257809689897,
      "learning_rate": 1.7748718243527e-05,
      "loss": 1.1858,
      "step": 333
    },
    {
      "epoch": 0.48423341790503804,
      "grad_norm": 2.2920578238007487,
      "learning_rate": 1.773383262372152e-05,
      "loss": 1.147,
      "step": 334
    },
    {
      "epoch": 0.48568321855744834,
      "grad_norm": 2.4375836870713585,
      "learning_rate": 1.7718904239613066e-05,
      "loss": 1.1641,
      "step": 335
    },
    {
      "epoch": 0.48713301920985863,
      "grad_norm": 2.6603113847222315,
      "learning_rate": 1.7703933173748276e-05,
      "loss": 1.2878,
      "step": 336
    },
    {
      "epoch": 0.48713301920985863,
      "eval_loss": 1.9789317846298218,
      "eval_runtime": 712.0072,
      "eval_samples_per_second": 24.795,
      "eval_steps_per_second": 0.388,
      "step": 336
    },
    {
      "epoch": 0.4885828198622689,
      "grad_norm": 2.4684746930730426,
      "learning_rate": 1.7688919508909807e-05,
      "loss": 1.2239,
      "step": 337
    },
    {
      "epoch": 0.4900326205146792,
      "grad_norm": 2.3892705567846644,
      "learning_rate": 1.767386332811587e-05,
      "loss": 1.2126,
      "step": 338
    },
    {
      "epoch": 0.4914824211670895,
      "grad_norm": 2.1463363792452292,
      "learning_rate": 1.7658764714619762e-05,
      "loss": 1.211,
      "step": 339
    },
    {
      "epoch": 0.4929322218194998,
      "grad_norm": 2.1625727005942506,
      "learning_rate": 1.764362375190941e-05,
      "loss": 1.167,
      "step": 340
    },
    {
      "epoch": 0.4943820224719101,
      "grad_norm": 2.789636476103976,
      "learning_rate": 1.7628440523706914e-05,
      "loss": 1.2645,
      "step": 341
    },
    {
      "epoch": 0.4958318231243204,
      "grad_norm": 2.1464402978107398,
      "learning_rate": 1.761321511396809e-05,
      "loss": 1.2131,
      "step": 342
    },
    {
      "epoch": 0.4972816237767307,
      "grad_norm": 2.0606745977353382,
      "learning_rate": 1.759794760688198e-05,
      "loss": 1.2799,
      "step": 343
    },
    {
      "epoch": 0.498731424429141,
      "grad_norm": 2.0193305538486253,
      "learning_rate": 1.7582638086870426e-05,
      "loss": 1.1756,
      "step": 344
    },
    {
      "epoch": 0.5001812250815513,
      "grad_norm": 2.126626983279132,
      "learning_rate": 1.756728663858756e-05,
      "loss": 1.2258,
      "step": 345
    },
    {
      "epoch": 0.5016310257339616,
      "grad_norm": 2.1706504596142664,
      "learning_rate": 1.7551893346919363e-05,
      "loss": 1.1133,
      "step": 346
    },
    {
      "epoch": 0.5030808263863719,
      "grad_norm": 2.390523545308175,
      "learning_rate": 1.7536458296983198e-05,
      "loss": 1.2605,
      "step": 347
    },
    {
      "epoch": 0.5045306270387822,
      "grad_norm": 2.1518336568241914,
      "learning_rate": 1.752098157412732e-05,
      "loss": 1.1485,
      "step": 348
    },
    {
      "epoch": 0.5059804276911924,
      "grad_norm": 2.1919868427136406,
      "learning_rate": 1.750546326393042e-05,
      "loss": 1.2012,
      "step": 349
    },
    {
      "epoch": 0.5074302283436027,
      "grad_norm": 2.521393726555328,
      "learning_rate": 1.7489903452201144e-05,
      "loss": 1.2205,
      "step": 350
    },
    {
      "epoch": 0.5074302283436027,
      "eval_loss": 1.9919556379318237,
      "eval_runtime": 713.3734,
      "eval_samples_per_second": 24.747,
      "eval_steps_per_second": 0.387,
      "step": 350
    },
    {
      "epoch": 0.508880028996013,
      "grad_norm": 2.263077066914077,
      "learning_rate": 1.7474302224977628e-05,
      "loss": 1.1502,
      "step": 351
    },
    {
      "epoch": 0.5103298296484233,
      "grad_norm": 2.1103882146373367,
      "learning_rate": 1.7458659668527005e-05,
      "loss": 1.1642,
      "step": 352
    },
    {
      "epoch": 0.5117796303008336,
      "grad_norm": 2.35310551198277,
      "learning_rate": 1.7442975869344948e-05,
      "loss": 1.2524,
      "step": 353
    },
    {
      "epoch": 0.5132294309532439,
      "grad_norm": 2.0997946270908106,
      "learning_rate": 1.742725091415517e-05,
      "loss": 1.1435,
      "step": 354
    },
    {
      "epoch": 0.5146792316056542,
      "grad_norm": 3.009884148806102,
      "learning_rate": 1.7411484889908976e-05,
      "loss": 1.2117,
      "step": 355
    },
    {
      "epoch": 0.5161290322580645,
      "grad_norm": 2.257913355698772,
      "learning_rate": 1.7395677883784742e-05,
      "loss": 1.1994,
      "step": 356
    },
    {
      "epoch": 0.5175788329104748,
      "grad_norm": 2.406562007350679,
      "learning_rate": 1.737982998318746e-05,
      "loss": 1.2301,
      "step": 357
    },
    {
      "epoch": 0.5190286335628851,
      "grad_norm": 2.437542589113885,
      "learning_rate": 1.7363941275748263e-05,
      "loss": 1.2607,
      "step": 358
    },
    {
      "epoch": 0.5204784342152954,
      "grad_norm": 2.393949167489127,
      "learning_rate": 1.7348011849323904e-05,
      "loss": 1.1853,
      "step": 359
    },
    {
      "epoch": 0.5219282348677057,
      "grad_norm": 2.152676170467717,
      "learning_rate": 1.733204179199631e-05,
      "loss": 1.1637,
      "step": 360
    },
    {
      "epoch": 0.523378035520116,
      "grad_norm": 2.6641142182858357,
      "learning_rate": 1.7316031192072056e-05,
      "loss": 1.1186,
      "step": 361
    },
    {
      "epoch": 0.5248278361725263,
      "grad_norm": 2.244905473604366,
      "learning_rate": 1.729998013808192e-05,
      "loss": 1.3086,
      "step": 362
    },
    {
      "epoch": 0.5262776368249366,
      "grad_norm": 2.2591291620464773,
      "learning_rate": 1.7283888718780364e-05,
      "loss": 1.0992,
      "step": 363
    },
    {
      "epoch": 0.5277274374773469,
      "grad_norm": 2.3476864819690544,
      "learning_rate": 1.726775702314504e-05,
      "loss": 1.1904,
      "step": 364
    },
    {
      "epoch": 0.5277274374773469,
      "eval_loss": 2.020822048187256,
      "eval_runtime": 715.1033,
      "eval_samples_per_second": 24.687,
      "eval_steps_per_second": 0.386,
      "step": 364
    },
    {
      "epoch": 0.5291772381297571,
      "grad_norm": 2.140122284731223,
      "learning_rate": 1.7251585140376322e-05,
      "loss": 1.1584,
      "step": 365
    },
    {
      "epoch": 0.5306270387821674,
      "grad_norm": 2.2744702865444473,
      "learning_rate": 1.7235373159896792e-05,
      "loss": 1.2096,
      "step": 366
    },
    {
      "epoch": 0.5320768394345777,
      "grad_norm": 2.2480959485202283,
      "learning_rate": 1.7219121171350757e-05,
      "loss": 1.1729,
      "step": 367
    },
    {
      "epoch": 0.533526640086988,
      "grad_norm": 2.6499161696790865,
      "learning_rate": 1.720282926460374e-05,
      "loss": 1.141,
      "step": 368
    },
    {
      "epoch": 0.5349764407393983,
      "grad_norm": 2.378730650178595,
      "learning_rate": 1.7186497529742002e-05,
      "loss": 1.1865,
      "step": 369
    },
    {
      "epoch": 0.5364262413918086,
      "grad_norm": 2.250360368238434,
      "learning_rate": 1.717012605707203e-05,
      "loss": 1.173,
      "step": 370
    },
    {
      "epoch": 0.5378760420442189,
      "grad_norm": 2.193145432136091,
      "learning_rate": 1.7153714937120044e-05,
      "loss": 1.2687,
      "step": 371
    },
    {
      "epoch": 0.5393258426966292,
      "grad_norm": 2.3418755600653434,
      "learning_rate": 1.7137264260631493e-05,
      "loss": 1.1711,
      "step": 372
    },
    {
      "epoch": 0.5407756433490395,
      "grad_norm": 2.191873177509745,
      "learning_rate": 1.7120774118570548e-05,
      "loss": 1.2555,
      "step": 373
    },
    {
      "epoch": 0.5422254440014498,
      "grad_norm": 2.264051113844044,
      "learning_rate": 1.7104244602119613e-05,
      "loss": 1.1594,
      "step": 374
    },
    {
      "epoch": 0.5436752446538601,
      "grad_norm": 2.27267337740226,
      "learning_rate": 1.708767580267881e-05,
      "loss": 1.1568,
      "step": 375
    },
    {
      "epoch": 0.5451250453062704,
      "grad_norm": 2.474824943569311,
      "learning_rate": 1.7071067811865477e-05,
      "loss": 1.2217,
      "step": 376
    },
    {
      "epoch": 0.5465748459586807,
      "grad_norm": 2.174098387892562,
      "learning_rate": 1.705442072151366e-05,
      "loss": 1.1953,
      "step": 377
    },
    {
      "epoch": 0.548024646611091,
      "grad_norm": 2.40804063416524,
      "learning_rate": 1.7037734623673616e-05,
      "loss": 1.2254,
      "step": 378
    },
    {
      "epoch": 0.548024646611091,
      "eval_loss": 2.0017855167388916,
      "eval_runtime": 718.0572,
      "eval_samples_per_second": 24.586,
      "eval_steps_per_second": 0.384,
      "step": 378
    },
    {
      "epoch": 0.5494744472635013,
      "grad_norm": 2.300206338606927,
      "learning_rate": 1.7021009610611276e-05,
      "loss": 1.2334,
      "step": 379
    },
    {
      "epoch": 0.5509242479159115,
      "grad_norm": 2.1393703813187948,
      "learning_rate": 1.7004245774807764e-05,
      "loss": 1.2304,
      "step": 380
    },
    {
      "epoch": 0.5523740485683218,
      "grad_norm": 2.223495231050503,
      "learning_rate": 1.698744320895888e-05,
      "loss": 1.1546,
      "step": 381
    },
    {
      "epoch": 0.5538238492207321,
      "grad_norm": 2.1944650479865686,
      "learning_rate": 1.697060200597457e-05,
      "loss": 1.1929,
      "step": 382
    },
    {
      "epoch": 0.5552736498731424,
      "grad_norm": 2.01722969353647,
      "learning_rate": 1.6953722258978437e-05,
      "loss": 1.218,
      "step": 383
    },
    {
      "epoch": 0.5567234505255527,
      "grad_norm": 2.015154622044917,
      "learning_rate": 1.6936804061307196e-05,
      "loss": 1.2038,
      "step": 384
    },
    {
      "epoch": 0.558173251177963,
      "grad_norm": 2.266107098932082,
      "learning_rate": 1.6919847506510194e-05,
      "loss": 1.2435,
      "step": 385
    },
    {
      "epoch": 0.5596230518303733,
      "grad_norm": 2.223963306994925,
      "learning_rate": 1.6902852688348866e-05,
      "loss": 1.1822,
      "step": 386
    },
    {
      "epoch": 0.5610728524827836,
      "grad_norm": 2.333787971452684,
      "learning_rate": 1.688581970079622e-05,
      "loss": 1.1172,
      "step": 387
    },
    {
      "epoch": 0.5625226531351939,
      "grad_norm": 2.1898205339482106,
      "learning_rate": 1.686874863803633e-05,
      "loss": 1.2313,
      "step": 388
    },
    {
      "epoch": 0.5639724537876042,
      "grad_norm": 2.1090903318884333,
      "learning_rate": 1.685163959446381e-05,
      "loss": 1.1324,
      "step": 389
    },
    {
      "epoch": 0.5654222544400145,
      "grad_norm": 2.286240978471785,
      "learning_rate": 1.6834492664683275e-05,
      "loss": 1.3325,
      "step": 390
    },
    {
      "epoch": 0.5668720550924248,
      "grad_norm": 2.550582728541338,
      "learning_rate": 1.6817307943508843e-05,
      "loss": 1.2902,
      "step": 391
    },
    {
      "epoch": 0.5683218557448351,
      "grad_norm": 2.8090305280543766,
      "learning_rate": 1.68000855259636e-05,
      "loss": 1.2044,
      "step": 392
    },
    {
      "epoch": 0.5683218557448351,
      "eval_loss": 2.000269889831543,
      "eval_runtime": 713.4226,
      "eval_samples_per_second": 24.746,
      "eval_steps_per_second": 0.387,
      "step": 392
    },
    {
      "epoch": 0.5697716563972454,
      "grad_norm": 2.3295591286030066,
      "learning_rate": 1.6782825507279067e-05,
      "loss": 1.1416,
      "step": 393
    },
    {
      "epoch": 0.5712214570496557,
      "grad_norm": 2.208937610414336,
      "learning_rate": 1.676552798289469e-05,
      "loss": 1.1502,
      "step": 394
    },
    {
      "epoch": 0.572671257702066,
      "grad_norm": 2.199628400811618,
      "learning_rate": 1.6748193048457295e-05,
      "loss": 1.1296,
      "step": 395
    },
    {
      "epoch": 0.5741210583544762,
      "grad_norm": 2.2851692358513613,
      "learning_rate": 1.6730820799820575e-05,
      "loss": 1.1851,
      "step": 396
    },
    {
      "epoch": 0.5755708590068865,
      "grad_norm": 2.1347868442215385,
      "learning_rate": 1.6713411333044547e-05,
      "loss": 1.0847,
      "step": 397
    },
    {
      "epoch": 0.5770206596592968,
      "grad_norm": 2.5079702588829784,
      "learning_rate": 1.669596474439503e-05,
      "loss": 1.1673,
      "step": 398
    },
    {
      "epoch": 0.5784704603117071,
      "grad_norm": 2.3489423647010637,
      "learning_rate": 1.66784811303431e-05,
      "loss": 1.1386,
      "step": 399
    },
    {
      "epoch": 0.5799202609641174,
      "grad_norm": 2.104289554107292,
      "learning_rate": 1.666096058756458e-05,
      "loss": 1.1404,
      "step": 400
    },
    {
      "epoch": 0.5813700616165277,
      "grad_norm": 1.9724796846665824,
      "learning_rate": 1.6643403212939474e-05,
      "loss": 1.189,
      "step": 401
    },
    {
      "epoch": 0.582819862268938,
      "grad_norm": 2.1421452773539076,
      "learning_rate": 1.6625809103551462e-05,
      "loss": 1.1489,
      "step": 402
    },
    {
      "epoch": 0.5842696629213483,
      "grad_norm": 2.156247856943813,
      "learning_rate": 1.660817835668734e-05,
      "loss": 1.2531,
      "step": 403
    },
    {
      "epoch": 0.5857194635737586,
      "grad_norm": 2.410640214209435,
      "learning_rate": 1.6590511069836498e-05,
      "loss": 1.1626,
      "step": 404
    },
    {
      "epoch": 0.5871692642261689,
      "grad_norm": 2.0688025461995343,
      "learning_rate": 1.6572807340690367e-05,
      "loss": 1.2028,
      "step": 405
    },
    {
      "epoch": 0.5886190648785792,
      "grad_norm": 2.3278885552777515,
      "learning_rate": 1.6555067267141896e-05,
      "loss": 1.2252,
      "step": 406
    },
    {
      "epoch": 0.5886190648785792,
      "eval_loss": 1.9900593757629395,
      "eval_runtime": 714.5915,
      "eval_samples_per_second": 24.705,
      "eval_steps_per_second": 0.386,
      "step": 406
    },
    {
      "epoch": 0.5900688655309895,
      "grad_norm": 2.209295197803796,
      "learning_rate": 1.6537290947284986e-05,
      "loss": 1.1622,
      "step": 407
    },
    {
      "epoch": 0.5915186661833998,
      "grad_norm": 2.616857207627912,
      "learning_rate": 1.651947847941398e-05,
      "loss": 1.2086,
      "step": 408
    },
    {
      "epoch": 0.5929684668358101,
      "grad_norm": 2.7268215902150668,
      "learning_rate": 1.650162996202309e-05,
      "loss": 1.2019,
      "step": 409
    },
    {
      "epoch": 0.5944182674882204,
      "grad_norm": 2.144309627920424,
      "learning_rate": 1.648374549380587e-05,
      "loss": 1.1862,
      "step": 410
    },
    {
      "epoch": 0.5958680681406306,
      "grad_norm": 2.641191030051115,
      "learning_rate": 1.6465825173654652e-05,
      "loss": 1.1346,
      "step": 411
    },
    {
      "epoch": 0.5973178687930409,
      "grad_norm": 2.235952164309796,
      "learning_rate": 1.644786910066003e-05,
      "loss": 1.1461,
      "step": 412
    },
    {
      "epoch": 0.5987676694454512,
      "grad_norm": 2.4039677628707206,
      "learning_rate": 1.6429877374110282e-05,
      "loss": 1.1145,
      "step": 413
    },
    {
      "epoch": 0.6002174700978615,
      "grad_norm": 2.1260494207875946,
      "learning_rate": 1.641185009349084e-05,
      "loss": 1.1592,
      "step": 414
    },
    {
      "epoch": 0.6016672707502718,
      "grad_norm": 2.2945325309346085,
      "learning_rate": 1.6393787358483717e-05,
      "loss": 1.1416,
      "step": 415
    },
    {
      "epoch": 0.6031170714026821,
      "grad_norm": 2.38633818865568,
      "learning_rate": 1.637568926896699e-05,
      "loss": 1.161,
      "step": 416
    },
    {
      "epoch": 0.6045668720550924,
      "grad_norm": 2.272191261485309,
      "learning_rate": 1.6357555925014227e-05,
      "loss": 1.2736,
      "step": 417
    },
    {
      "epoch": 0.6060166727075027,
      "grad_norm": 2.2999912621370817,
      "learning_rate": 1.633938742689392e-05,
      "loss": 1.1043,
      "step": 418
    },
    {
      "epoch": 0.607466473359913,
      "grad_norm": 2.283533711159278,
      "learning_rate": 1.632118387506896e-05,
      "loss": 1.1773,
      "step": 419
    },
    {
      "epoch": 0.6089162740123233,
      "grad_norm": 2.826611251718709,
      "learning_rate": 1.6302945370196074e-05,
      "loss": 1.2404,
      "step": 420
    },
    {
      "epoch": 0.6089162740123233,
      "eval_loss": 1.978614091873169,
      "eval_runtime": 711.575,
      "eval_samples_per_second": 24.81,
      "eval_steps_per_second": 0.388,
      "step": 420
    },
    {
      "epoch": 0.6103660746647336,
      "grad_norm": 2.3274085282410324,
      "learning_rate": 1.6284672013125237e-05,
      "loss": 1.2027,
      "step": 421
    },
    {
      "epoch": 0.6118158753171439,
      "grad_norm": 2.1968666001430335,
      "learning_rate": 1.626636390489916e-05,
      "loss": 1.1668,
      "step": 422
    },
    {
      "epoch": 0.6132656759695542,
      "grad_norm": 2.2475284250175624,
      "learning_rate": 1.6248021146752715e-05,
      "loss": 1.1604,
      "step": 423
    },
    {
      "epoch": 0.6147154766219645,
      "grad_norm": 2.4752875374347587,
      "learning_rate": 1.6229643840112355e-05,
      "loss": 1.2371,
      "step": 424
    },
    {
      "epoch": 0.6161652772743748,
      "grad_norm": 1.9756738881966343,
      "learning_rate": 1.6211232086595576e-05,
      "loss": 1.1237,
      "step": 425
    },
    {
      "epoch": 0.617615077926785,
      "grad_norm": 2.695611293232905,
      "learning_rate": 1.6192785988010354e-05,
      "loss": 1.2681,
      "step": 426
    },
    {
      "epoch": 0.6190648785791953,
      "grad_norm": 2.1902655450882493,
      "learning_rate": 1.617430564635456e-05,
      "loss": 1.1574,
      "step": 427
    },
    {
      "epoch": 0.6205146792316056,
      "grad_norm": 2.194164957713764,
      "learning_rate": 1.6155791163815438e-05,
      "loss": 1.1644,
      "step": 428
    },
    {
      "epoch": 0.6219644798840159,
      "grad_norm": 2.4819170490309084,
      "learning_rate": 1.613724264276899e-05,
      "loss": 1.0863,
      "step": 429
    },
    {
      "epoch": 0.6234142805364262,
      "grad_norm": 2.1758770916977896,
      "learning_rate": 1.6118660185779448e-05,
      "loss": 1.1776,
      "step": 430
    },
    {
      "epoch": 0.6248640811888365,
      "grad_norm": 2.229813202981371,
      "learning_rate": 1.6100043895598685e-05,
      "loss": 1.2324,
      "step": 431
    },
    {
      "epoch": 0.6263138818412468,
      "grad_norm": 2.1799309573192174,
      "learning_rate": 1.6081393875165665e-05,
      "loss": 1.2466,
      "step": 432
    },
    {
      "epoch": 0.6277636824936571,
      "grad_norm": 2.58310001914373,
      "learning_rate": 1.6062710227605856e-05,
      "loss": 1.2292,
      "step": 433
    },
    {
      "epoch": 0.6292134831460674,
      "grad_norm": 2.055905901998397,
      "learning_rate": 1.604399305623067e-05,
      "loss": 1.0537,
      "step": 434
    },
    {
      "epoch": 0.6292134831460674,
      "eval_loss": 1.9674261808395386,
      "eval_runtime": 714.5775,
      "eval_samples_per_second": 24.706,
      "eval_steps_per_second": 0.386,
      "step": 434
    },
    {
      "epoch": 0.6306632837984777,
      "grad_norm": 2.2611120818261945,
      "learning_rate": 1.602524246453689e-05,
      "loss": 1.1395,
      "step": 435
    },
    {
      "epoch": 0.632113084450888,
      "grad_norm": 2.0999108255026666,
      "learning_rate": 1.6006458556206087e-05,
      "loss": 1.1132,
      "step": 436
    },
    {
      "epoch": 0.6335628851032983,
      "grad_norm": 2.248804426138939,
      "learning_rate": 1.5987641435104075e-05,
      "loss": 1.1391,
      "step": 437
    },
    {
      "epoch": 0.6350126857557086,
      "grad_norm": 2.239697309792104,
      "learning_rate": 1.5968791205280306e-05,
      "loss": 1.2308,
      "step": 438
    },
    {
      "epoch": 0.6364624864081189,
      "grad_norm": 2.2089290667632357,
      "learning_rate": 1.5949907970967312e-05,
      "loss": 1.2353,
      "step": 439
    },
    {
      "epoch": 0.6379122870605292,
      "grad_norm": 2.335922666057784,
      "learning_rate": 1.593099183658012e-05,
      "loss": 1.2059,
      "step": 440
    },
    {
      "epoch": 0.6393620877129395,
      "grad_norm": 2.0999257928484814,
      "learning_rate": 1.5912042906715676e-05,
      "loss": 1.1368,
      "step": 441
    },
    {
      "epoch": 0.6408118883653497,
      "grad_norm": 2.2106106277905084,
      "learning_rate": 1.5893061286152275e-05,
      "loss": 1.1186,
      "step": 442
    },
    {
      "epoch": 0.64226168901776,
      "grad_norm": 2.345946104163717,
      "learning_rate": 1.5874047079848972e-05,
      "loss": 1.1831,
      "step": 443
    },
    {
      "epoch": 0.6437114896701703,
      "grad_norm": 2.1566896070354336,
      "learning_rate": 1.585500039294501e-05,
      "loss": 1.1363,
      "step": 444
    },
    {
      "epoch": 0.6451612903225806,
      "grad_norm": 2.1636489690680176,
      "learning_rate": 1.5835921330759233e-05,
      "loss": 1.15,
      "step": 445
    },
    {
      "epoch": 0.6466110909749909,
      "grad_norm": 2.1400337060185297,
      "learning_rate": 1.5816809998789493e-05,
      "loss": 1.2369,
      "step": 446
    },
    {
      "epoch": 0.6480608916274012,
      "grad_norm": 1.9968232310781666,
      "learning_rate": 1.579766650271209e-05,
      "loss": 1.127,
      "step": 447
    },
    {
      "epoch": 0.6495106922798115,
      "grad_norm": 2.242529765020952,
      "learning_rate": 1.5778490948381175e-05,
      "loss": 1.236,
      "step": 448
    },
    {
      "epoch": 0.6495106922798115,
      "eval_loss": 1.979479193687439,
      "eval_runtime": 723.1657,
      "eval_samples_per_second": 24.412,
      "eval_steps_per_second": 0.382,
      "step": 448
    },
    {
      "epoch": 0.6509604929322218,
      "grad_norm": 2.3170218471516457,
      "learning_rate": 1.575928344182816e-05,
      "loss": 1.2114,
      "step": 449
    },
    {
      "epoch": 0.6524102935846321,
      "grad_norm": 2.304799685091385,
      "learning_rate": 1.5740044089261137e-05,
      "loss": 1.1667,
      "step": 450
    },
    {
      "epoch": 0.6538600942370424,
      "grad_norm": 2.4403429889681654,
      "learning_rate": 1.5720772997064302e-05,
      "loss": 1.3027,
      "step": 451
    },
    {
      "epoch": 0.6553098948894527,
      "grad_norm": 2.1503430939702413,
      "learning_rate": 1.570147027179734e-05,
      "loss": 1.1712,
      "step": 452
    },
    {
      "epoch": 0.656759695541863,
      "grad_norm": 2.267564664572557,
      "learning_rate": 1.5682136020194863e-05,
      "loss": 1.3161,
      "step": 453
    },
    {
      "epoch": 0.6582094961942733,
      "grad_norm": 2.033719525182907,
      "learning_rate": 1.5662770349165802e-05,
      "loss": 1.1319,
      "step": 454
    },
    {
      "epoch": 0.6596592968466836,
      "grad_norm": 2.1230804090066475,
      "learning_rate": 1.5643373365792824e-05,
      "loss": 1.1831,
      "step": 455
    },
    {
      "epoch": 0.6611090974990939,
      "grad_norm": 2.026820775062979,
      "learning_rate": 1.562394517733174e-05,
      "loss": 1.1602,
      "step": 456
    },
    {
      "epoch": 0.6625588981515041,
      "grad_norm": 2.4139450089537537,
      "learning_rate": 1.5604485891210904e-05,
      "loss": 1.0363,
      "step": 457
    },
    {
      "epoch": 0.6640086988039144,
      "grad_norm": 3.503168551608323,
      "learning_rate": 1.5584995615030634e-05,
      "loss": 1.1414,
      "step": 458
    },
    {
      "epoch": 0.6654584994563247,
      "grad_norm": 2.0198765979429973,
      "learning_rate": 1.5565474456562598e-05,
      "loss": 1.2115,
      "step": 459
    },
    {
      "epoch": 0.666908300108735,
      "grad_norm": 2.121927829358635,
      "learning_rate": 1.554592252374923e-05,
      "loss": 1.1222,
      "step": 460
    },
    {
      "epoch": 0.6683581007611453,
      "grad_norm": 2.1921502059165063,
      "learning_rate": 1.5526339924703142e-05,
      "loss": 1.1067,
      "step": 461
    },
    {
      "epoch": 0.6698079014135556,
      "grad_norm": 2.3460831669206184,
      "learning_rate": 1.55067267677065e-05,
      "loss": 1.1882,
      "step": 462
    },
    {
      "epoch": 0.6698079014135556,
      "eval_loss": 1.9846652746200562,
      "eval_runtime": 711.1957,
      "eval_samples_per_second": 24.823,
      "eval_steps_per_second": 0.388,
      "step": 462
    },
    {
      "epoch": 0.6712577020659659,
      "grad_norm": 2.417788324868438,
      "learning_rate": 1.548708316121044e-05,
      "loss": 1.2034,
      "step": 463
    },
    {
      "epoch": 0.6727075027183762,
      "grad_norm": 2.230834124103965,
      "learning_rate": 1.5467409213834496e-05,
      "loss": 1.147,
      "step": 464
    },
    {
      "epoch": 0.6741573033707865,
      "grad_norm": 2.1222760456975776,
      "learning_rate": 1.544770503436593e-05,
      "loss": 1.1079,
      "step": 465
    },
    {
      "epoch": 0.6756071040231968,
      "grad_norm": 2.25849380928062,
      "learning_rate": 1.5427970731759214e-05,
      "loss": 1.1951,
      "step": 466
    },
    {
      "epoch": 0.6770569046756071,
      "grad_norm": 2.291065494079678,
      "learning_rate": 1.5408206415135353e-05,
      "loss": 1.1512,
      "step": 467
    },
    {
      "epoch": 0.6785067053280174,
      "grad_norm": 2.3272733179636798,
      "learning_rate": 1.5388412193781334e-05,
      "loss": 1.1371,
      "step": 468
    },
    {
      "epoch": 0.6799565059804277,
      "grad_norm": 2.302879294890553,
      "learning_rate": 1.5368588177149498e-05,
      "loss": 1.2611,
      "step": 469
    },
    {
      "epoch": 0.681406306632838,
      "grad_norm": 2.9102807940404465,
      "learning_rate": 1.5348734474856934e-05,
      "loss": 1.2399,
      "step": 470
    },
    {
      "epoch": 0.6828561072852483,
      "grad_norm": 2.0475244563015726,
      "learning_rate": 1.5328851196684883e-05,
      "loss": 1.1984,
      "step": 471
    },
    {
      "epoch": 0.6843059079376586,
      "grad_norm": 2.3264313051673073,
      "learning_rate": 1.5308938452578136e-05,
      "loss": 1.1506,
      "step": 472
    },
    {
      "epoch": 0.6857557085900688,
      "grad_norm": 2.2735466116684164,
      "learning_rate": 1.5288996352644395e-05,
      "loss": 1.12,
      "step": 473
    },
    {
      "epoch": 0.6872055092424791,
      "grad_norm": 2.234794097737822,
      "learning_rate": 1.52690250071537e-05,
      "loss": 1.1573,
      "step": 474
    },
    {
      "epoch": 0.6886553098948894,
      "grad_norm": 2.135917178038943,
      "learning_rate": 1.52490245265378e-05,
      "loss": 1.109,
      "step": 475
    },
    {
      "epoch": 0.6901051105472997,
      "grad_norm": 2.3845965710842427,
      "learning_rate": 1.5228995021389548e-05,
      "loss": 1.0961,
      "step": 476
    },
    {
      "epoch": 0.6901051105472997,
      "eval_loss": 1.9868731498718262,
      "eval_runtime": 715.8742,
      "eval_samples_per_second": 24.661,
      "eval_steps_per_second": 0.386,
      "step": 476
    },
    {
      "epoch": 0.69155491119971,
      "grad_norm": 2.2686159132678925,
      "learning_rate": 1.5208936602462285e-05,
      "loss": 1.1838,
      "step": 477
    },
    {
      "epoch": 0.6930047118521203,
      "grad_norm": 2.444687031640569,
      "learning_rate": 1.5188849380669241e-05,
      "loss": 1.1663,
      "step": 478
    },
    {
      "epoch": 0.6944545125045306,
      "grad_norm": 2.3498450574801057,
      "learning_rate": 1.5168733467082901e-05,
      "loss": 1.1753,
      "step": 479
    },
    {
      "epoch": 0.6959043131569409,
      "grad_norm": 2.546824435564423,
      "learning_rate": 1.5148588972934408e-05,
      "loss": 1.0448,
      "step": 480
    },
    {
      "epoch": 0.6973541138093512,
      "grad_norm": 2.135006484206936,
      "learning_rate": 1.5128416009612934e-05,
      "loss": 1.2456,
      "step": 481
    },
    {
      "epoch": 0.6988039144617615,
      "grad_norm": 2.229621367249865,
      "learning_rate": 1.510821468866508e-05,
      "loss": 1.2337,
      "step": 482
    },
    {
      "epoch": 0.7002537151141718,
      "grad_norm": 2.0819646719369485,
      "learning_rate": 1.5087985121794248e-05,
      "loss": 1.1973,
      "step": 483
    },
    {
      "epoch": 0.7017035157665821,
      "grad_norm": 2.435673157910559,
      "learning_rate": 1.5067727420860023e-05,
      "loss": 1.2324,
      "step": 484
    },
    {
      "epoch": 0.7031533164189924,
      "grad_norm": 2.3383365804565215,
      "learning_rate": 1.5047441697877563e-05,
      "loss": 1.1075,
      "step": 485
    },
    {
      "epoch": 0.7046031170714027,
      "grad_norm": 2.1756008091234547,
      "learning_rate": 1.5027128065016969e-05,
      "loss": 1.1523,
      "step": 486
    },
    {
      "epoch": 0.706052917723813,
      "grad_norm": 2.0606928667981355,
      "learning_rate": 1.5006786634602675e-05,
      "loss": 1.1203,
      "step": 487
    },
    {
      "epoch": 0.7075027183762232,
      "grad_norm": 2.1352374591354972,
      "learning_rate": 1.4986417519112813e-05,
      "loss": 1.2534,
      "step": 488
    },
    {
      "epoch": 0.7089525190286335,
      "grad_norm": 2.0212563596461073,
      "learning_rate": 1.4966020831178612e-05,
      "loss": 1.1452,
      "step": 489
    },
    {
      "epoch": 0.7104023196810438,
      "grad_norm": 2.040524798362301,
      "learning_rate": 1.4945596683583754e-05,
      "loss": 1.1156,
      "step": 490
    },
    {
      "epoch": 0.7104023196810438,
      "eval_loss": 1.9905571937561035,
      "eval_runtime": 709.5814,
      "eval_samples_per_second": 24.879,
      "eval_steps_per_second": 0.389,
      "step": 490
    },
    {
      "epoch": 0.7118521203334541,
      "grad_norm": 2.3055486733308896,
      "learning_rate": 1.4925145189263763e-05,
      "loss": 1.1592,
      "step": 491
    },
    {
      "epoch": 0.7133019209858644,
      "grad_norm": 2.3011237544754,
      "learning_rate": 1.4904666461305377e-05,
      "loss": 1.1636,
      "step": 492
    },
    {
      "epoch": 0.7147517216382747,
      "grad_norm": 2.0393388969907953,
      "learning_rate": 1.4884160612945924e-05,
      "loss": 1.1663,
      "step": 493
    },
    {
      "epoch": 0.716201522290685,
      "grad_norm": 2.2383416720304443,
      "learning_rate": 1.4863627757572687e-05,
      "loss": 1.1055,
      "step": 494
    },
    {
      "epoch": 0.7176513229430953,
      "grad_norm": 2.0640719299296832,
      "learning_rate": 1.4843068008722297e-05,
      "loss": 1.1148,
      "step": 495
    },
    {
      "epoch": 0.7191011235955056,
      "grad_norm": 2.340725413289667,
      "learning_rate": 1.4822481480080079e-05,
      "loss": 1.1181,
      "step": 496
    },
    {
      "epoch": 0.7205509242479159,
      "grad_norm": 2.1607049510983503,
      "learning_rate": 1.480186828547945e-05,
      "loss": 1.2522,
      "step": 497
    },
    {
      "epoch": 0.7220007249003262,
      "grad_norm": 1.999331131141336,
      "learning_rate": 1.4781228538901267e-05,
      "loss": 1.1209,
      "step": 498
    },
    {
      "epoch": 0.7234505255527365,
      "grad_norm": 2.1661977968512827,
      "learning_rate": 1.4760562354473215e-05,
      "loss": 1.151,
      "step": 499
    },
    {
      "epoch": 0.7249003262051468,
      "grad_norm": 1.9860757845580463,
      "learning_rate": 1.4739869846469155e-05,
      "loss": 1.2069,
      "step": 500
    },
    {
      "epoch": 0.7263501268575571,
      "grad_norm": 2.21927035991495,
      "learning_rate": 1.4719151129308518e-05,
      "loss": 1.0994,
      "step": 501
    },
    {
      "epoch": 0.7277999275099674,
      "grad_norm": 2.313203511158404,
      "learning_rate": 1.4698406317555652e-05,
      "loss": 1.1702,
      "step": 502
    },
    {
      "epoch": 0.7292497281623777,
      "grad_norm": 2.1954672455412574,
      "learning_rate": 1.4677635525919192e-05,
      "loss": 1.1579,
      "step": 503
    },
    {
      "epoch": 0.730699528814788,
      "grad_norm": 2.677387499759199,
      "learning_rate": 1.465683886925144e-05,
      "loss": 1.1873,
      "step": 504
    },
    {
      "epoch": 0.730699528814788,
      "eval_loss": 1.9812017679214478,
      "eval_runtime": 710.9637,
      "eval_samples_per_second": 24.831,
      "eval_steps_per_second": 0.388,
      "step": 504
    },
    {
      "epoch": 0.7321493294671982,
      "grad_norm": 2.3133463561317913,
      "learning_rate": 1.4636016462547702e-05,
      "loss": 1.0891,
      "step": 505
    },
    {
      "epoch": 0.7335991301196085,
      "grad_norm": 2.111132449002895,
      "learning_rate": 1.4615168420945689e-05,
      "loss": 1.086,
      "step": 506
    },
    {
      "epoch": 0.7350489307720188,
      "grad_norm": 2.3700337131740614,
      "learning_rate": 1.4594294859724847e-05,
      "loss": 1.1966,
      "step": 507
    },
    {
      "epoch": 0.7364987314244291,
      "grad_norm": 2.485889381254839,
      "learning_rate": 1.4573395894305736e-05,
      "loss": 1.1046,
      "step": 508
    },
    {
      "epoch": 0.7379485320768394,
      "grad_norm": 2.3003721504009533,
      "learning_rate": 1.4552471640249388e-05,
      "loss": 1.1975,
      "step": 509
    },
    {
      "epoch": 0.7393983327292497,
      "grad_norm": 2.4972036913465794,
      "learning_rate": 1.4531522213256675e-05,
      "loss": 1.1569,
      "step": 510
    },
    {
      "epoch": 0.74084813338166,
      "grad_norm": 2.286516454884241,
      "learning_rate": 1.451054772916765e-05,
      "loss": 1.0876,
      "step": 511
    },
    {
      "epoch": 0.7422979340340703,
      "grad_norm": 2.2787338482417008,
      "learning_rate": 1.4489548303960931e-05,
      "loss": 1.1803,
      "step": 512
    },
    {
      "epoch": 0.7437477346864806,
      "grad_norm": 2.1660812150951263,
      "learning_rate": 1.4468524053753047e-05,
      "loss": 1.075,
      "step": 513
    },
    {
      "epoch": 0.7451975353388909,
      "grad_norm": 2.1774740785775064,
      "learning_rate": 1.4447475094797796e-05,
      "loss": 1.1714,
      "step": 514
    },
    {
      "epoch": 0.7466473359913012,
      "grad_norm": 2.170361658231525,
      "learning_rate": 1.4426401543485598e-05,
      "loss": 1.1339,
      "step": 515
    },
    {
      "epoch": 0.7480971366437115,
      "grad_norm": 2.3629262234918587,
      "learning_rate": 1.440530351634286e-05,
      "loss": 1.2745,
      "step": 516
    },
    {
      "epoch": 0.7495469372961218,
      "grad_norm": 2.160542072799467,
      "learning_rate": 1.4384181130031336e-05,
      "loss": 1.1152,
      "step": 517
    },
    {
      "epoch": 0.7509967379485321,
      "grad_norm": 2.1361260213317586,
      "learning_rate": 1.4363034501347457e-05,
      "loss": 1.1941,
      "step": 518
    },
    {
      "epoch": 0.7509967379485321,
      "eval_loss": 1.9871357679367065,
      "eval_runtime": 710.8617,
      "eval_samples_per_second": 24.835,
      "eval_steps_per_second": 0.388,
      "step": 518
    },
    {
      "epoch": 0.7524465386009423,
      "grad_norm": 2.200844223799515,
      "learning_rate": 1.434186374722172e-05,
      "loss": 1.1127,
      "step": 519
    },
    {
      "epoch": 0.7538963392533526,
      "grad_norm": 2.432277075619356,
      "learning_rate": 1.4320668984718013e-05,
      "loss": 1.1803,
      "step": 520
    },
    {
      "epoch": 0.7553461399057629,
      "grad_norm": 2.254117413082917,
      "learning_rate": 1.4299450331032987e-05,
      "loss": 1.1254,
      "step": 521
    },
    {
      "epoch": 0.7567959405581732,
      "grad_norm": 2.2109113922377257,
      "learning_rate": 1.4278207903495388e-05,
      "loss": 1.1793,
      "step": 522
    },
    {
      "epoch": 0.7582457412105835,
      "grad_norm": 1.9678180477242784,
      "learning_rate": 1.4256941819565433e-05,
      "loss": 1.1196,
      "step": 523
    },
    {
      "epoch": 0.7596955418629938,
      "grad_norm": 2.0747843000396298,
      "learning_rate": 1.4235652196834135e-05,
      "loss": 1.1348,
      "step": 524
    },
    {
      "epoch": 0.7611453425154041,
      "grad_norm": 2.234651166990784,
      "learning_rate": 1.4214339153022676e-05,
      "loss": 1.1915,
      "step": 525
    },
    {
      "epoch": 0.7625951431678144,
      "grad_norm": 2.21294113762401,
      "learning_rate": 1.4193002805981731e-05,
      "loss": 1.2556,
      "step": 526
    },
    {
      "epoch": 0.7640449438202247,
      "grad_norm": 2.585514201827331,
      "learning_rate": 1.4171643273690849e-05,
      "loss": 1.1147,
      "step": 527
    },
    {
      "epoch": 0.765494744472635,
      "grad_norm": 2.2053464866208086,
      "learning_rate": 1.4150260674257767e-05,
      "loss": 1.1573,
      "step": 528
    },
    {
      "epoch": 0.7669445451250453,
      "grad_norm": 2.0931361818535312,
      "learning_rate": 1.412885512591778e-05,
      "loss": 1.1536,
      "step": 529
    },
    {
      "epoch": 0.7683943457774556,
      "grad_norm": 2.2051201841627646,
      "learning_rate": 1.4107426747033075e-05,
      "loss": 1.1675,
      "step": 530
    },
    {
      "epoch": 0.7698441464298659,
      "grad_norm": 2.0550925012272008,
      "learning_rate": 1.4085975656092082e-05,
      "loss": 1.1143,
      "step": 531
    },
    {
      "epoch": 0.7712939470822762,
      "grad_norm": 2.4604537506517965,
      "learning_rate": 1.406450197170882e-05,
      "loss": 1.2074,
      "step": 532
    },
    {
      "epoch": 0.7712939470822762,
      "eval_loss": 1.994682788848877,
      "eval_runtime": 713.5304,
      "eval_samples_per_second": 24.742,
      "eval_steps_per_second": 0.387,
      "step": 532
    },
    {
      "epoch": 0.7727437477346865,
      "grad_norm": 2.0695918984067507,
      "learning_rate": 1.4043005812622228e-05,
      "loss": 1.1537,
      "step": 533
    },
    {
      "epoch": 0.7741935483870968,
      "grad_norm": 2.1273125406104967,
      "learning_rate": 1.4021487297695539e-05,
      "loss": 1.1657,
      "step": 534
    },
    {
      "epoch": 0.775643349039507,
      "grad_norm": 2.0092150605216443,
      "learning_rate": 1.3999946545915586e-05,
      "loss": 1.1667,
      "step": 535
    },
    {
      "epoch": 0.7770931496919173,
      "grad_norm": 2.7218583845457927,
      "learning_rate": 1.3978383676392166e-05,
      "loss": 1.2018,
      "step": 536
    },
    {
      "epoch": 0.7785429503443276,
      "grad_norm": 2.2596021983293957,
      "learning_rate": 1.3956798808357377e-05,
      "loss": 1.1262,
      "step": 537
    },
    {
      "epoch": 0.7799927509967379,
      "grad_norm": 2.07517448483323,
      "learning_rate": 1.3935192061164957e-05,
      "loss": 1.2107,
      "step": 538
    },
    {
      "epoch": 0.7814425516491482,
      "grad_norm": 2.1963284140423345,
      "learning_rate": 1.3913563554289627e-05,
      "loss": 1.0785,
      "step": 539
    },
    {
      "epoch": 0.7828923523015585,
      "grad_norm": 2.144442013472242,
      "learning_rate": 1.3891913407326424e-05,
      "loss": 1.1898,
      "step": 540
    },
    {
      "epoch": 0.7843421529539688,
      "grad_norm": 2.2125840463885487,
      "learning_rate": 1.3870241739990047e-05,
      "loss": 1.1186,
      "step": 541
    },
    {
      "epoch": 0.7857919536063791,
      "grad_norm": 2.2955824257451094,
      "learning_rate": 1.3848548672114197e-05,
      "loss": 1.1003,
      "step": 542
    },
    {
      "epoch": 0.7872417542587894,
      "grad_norm": 2.382672226008031,
      "learning_rate": 1.3826834323650899e-05,
      "loss": 1.1276,
      "step": 543
    },
    {
      "epoch": 0.7886915549111997,
      "grad_norm": 2.383601480642944,
      "learning_rate": 1.3805098814669857e-05,
      "loss": 1.0979,
      "step": 544
    },
    {
      "epoch": 0.79014135556361,
      "grad_norm": 2.3740231906141425,
      "learning_rate": 1.3783342265357784e-05,
      "loss": 1.147,
      "step": 545
    },
    {
      "epoch": 0.7915911562160203,
      "grad_norm": 2.186666016437112,
      "learning_rate": 1.3761564796017724e-05,
      "loss": 1.0948,
      "step": 546
    },
    {
      "epoch": 0.7915911562160203,
      "eval_loss": 1.9701581001281738,
      "eval_runtime": 712.9952,
      "eval_samples_per_second": 24.76,
      "eval_steps_per_second": 0.387,
      "step": 546
    },
    {
      "epoch": 0.7930409568684306,
      "grad_norm": 2.2986049058358535,
      "learning_rate": 1.373976652706841e-05,
      "loss": 1.1506,
      "step": 547
    },
    {
      "epoch": 0.7944907575208409,
      "grad_norm": 2.04864649988135,
      "learning_rate": 1.3717947579043584e-05,
      "loss": 1.1045,
      "step": 548
    },
    {
      "epoch": 0.7959405581732512,
      "grad_norm": 2.4175787873310037,
      "learning_rate": 1.3696108072591333e-05,
      "loss": 1.165,
      "step": 549
    },
    {
      "epoch": 0.7973903588256614,
      "grad_norm": 2.3127626075063286,
      "learning_rate": 1.3674248128473417e-05,
      "loss": 1.1925,
      "step": 550
    },
    {
      "epoch": 0.7988401594780717,
      "grad_norm": 2.1603651426160924,
      "learning_rate": 1.3652367867564614e-05,
      "loss": 1.1401,
      "step": 551
    },
    {
      "epoch": 0.800289960130482,
      "grad_norm": 2.2373654630043878,
      "learning_rate": 1.3630467410852038e-05,
      "loss": 1.0854,
      "step": 552
    },
    {
      "epoch": 0.8017397607828923,
      "grad_norm": 2.202076972196098,
      "learning_rate": 1.3608546879434481e-05,
      "loss": 1.073,
      "step": 553
    },
    {
      "epoch": 0.8031895614353026,
      "grad_norm": 2.225354677700146,
      "learning_rate": 1.3586606394521729e-05,
      "loss": 1.1326,
      "step": 554
    },
    {
      "epoch": 0.8046393620877129,
      "grad_norm": 2.0414935461191126,
      "learning_rate": 1.3564646077433914e-05,
      "loss": 1.127,
      "step": 555
    },
    {
      "epoch": 0.8060891627401232,
      "grad_norm": 2.2792263916620494,
      "learning_rate": 1.3542666049600822e-05,
      "loss": 1.1714,
      "step": 556
    },
    {
      "epoch": 0.8075389633925335,
      "grad_norm": 2.4320217203612033,
      "learning_rate": 1.3520666432561233e-05,
      "loss": 1.092,
      "step": 557
    },
    {
      "epoch": 0.8089887640449438,
      "grad_norm": 2.006279500727713,
      "learning_rate": 1.349864734796224e-05,
      "loss": 1.1392,
      "step": 558
    },
    {
      "epoch": 0.8104385646973541,
      "grad_norm": 2.230371558777294,
      "learning_rate": 1.3476608917558587e-05,
      "loss": 1.1044,
      "step": 559
    },
    {
      "epoch": 0.8118883653497644,
      "grad_norm": 2.0735595287679316,
      "learning_rate": 1.3454551263211993e-05,
      "loss": 1.1178,
      "step": 560
    },
    {
      "epoch": 0.8118883653497644,
      "eval_loss": 1.9728071689605713,
      "eval_runtime": 719.8823,
      "eval_samples_per_second": 24.523,
      "eval_steps_per_second": 0.383,
      "step": 560
    },
    {
      "epoch": 0.8133381660021747,
      "grad_norm": 2.186656733520298,
      "learning_rate": 1.343247450689047e-05,
      "loss": 1.1921,
      "step": 561
    },
    {
      "epoch": 0.814787966654585,
      "grad_norm": 2.472372473314396,
      "learning_rate": 1.341037877066766e-05,
      "loss": 1.1543,
      "step": 562
    },
    {
      "epoch": 0.8162377673069953,
      "grad_norm": 2.575211089977401,
      "learning_rate": 1.338826417672215e-05,
      "loss": 1.2184,
      "step": 563
    },
    {
      "epoch": 0.8176875679594056,
      "grad_norm": 2.4846278783036384,
      "learning_rate": 1.3366130847336806e-05,
      "loss": 1.2012,
      "step": 564
    },
    {
      "epoch": 0.8191373686118159,
      "grad_norm": 2.5141709211491228,
      "learning_rate": 1.3343978904898087e-05,
      "loss": 1.1265,
      "step": 565
    },
    {
      "epoch": 0.8205871692642261,
      "grad_norm": 2.251338864418867,
      "learning_rate": 1.3321808471895374e-05,
      "loss": 1.1736,
      "step": 566
    },
    {
      "epoch": 0.8220369699166364,
      "grad_norm": 2.2312861520272267,
      "learning_rate": 1.3299619670920294e-05,
      "loss": 1.2135,
      "step": 567
    },
    {
      "epoch": 0.8234867705690467,
      "grad_norm": 1.965228630913633,
      "learning_rate": 1.3277412624666037e-05,
      "loss": 1.0578,
      "step": 568
    },
    {
      "epoch": 0.824936571221457,
      "grad_norm": 2.5599435245966187,
      "learning_rate": 1.3255187455926684e-05,
      "loss": 1.1575,
      "step": 569
    },
    {
      "epoch": 0.8263863718738673,
      "grad_norm": 2.228237509265747,
      "learning_rate": 1.323294428759652e-05,
      "loss": 1.131,
      "step": 570
    },
    {
      "epoch": 0.8278361725262776,
      "grad_norm": 2.0971949142528503,
      "learning_rate": 1.3210683242669366e-05,
      "loss": 1.2134,
      "step": 571
    },
    {
      "epoch": 0.8292859731786879,
      "grad_norm": 2.4090998063329994,
      "learning_rate": 1.3188404444237883e-05,
      "loss": 1.0766,
      "step": 572
    },
    {
      "epoch": 0.8307357738310982,
      "grad_norm": 2.0572122678505904,
      "learning_rate": 1.3166108015492905e-05,
      "loss": 1.1641,
      "step": 573
    },
    {
      "epoch": 0.8321855744835085,
      "grad_norm": 2.2239302206454896,
      "learning_rate": 1.3143794079722754e-05,
      "loss": 1.2372,
      "step": 574
    },
    {
      "epoch": 0.8321855744835085,
      "eval_loss": 1.9768322706222534,
      "eval_runtime": 718.4808,
      "eval_samples_per_second": 24.571,
      "eval_steps_per_second": 0.384,
      "step": 574
    },
    {
      "epoch": 0.8336353751359188,
      "grad_norm": 2.1449089284095835,
      "learning_rate": 1.3121462760312551e-05,
      "loss": 1.1875,
      "step": 575
    },
    {
      "epoch": 0.8350851757883291,
      "grad_norm": 2.1609020745584053,
      "learning_rate": 1.309911418074355e-05,
      "loss": 1.0783,
      "step": 576
    },
    {
      "epoch": 0.8365349764407394,
      "grad_norm": 2.203170469071651,
      "learning_rate": 1.3076748464592441e-05,
      "loss": 1.1835,
      "step": 577
    },
    {
      "epoch": 0.8379847770931497,
      "grad_norm": 2.1637812934943326,
      "learning_rate": 1.3054365735530666e-05,
      "loss": 1.1258,
      "step": 578
    },
    {
      "epoch": 0.83943457774556,
      "grad_norm": 2.1534446612450853,
      "learning_rate": 1.3031966117323745e-05,
      "loss": 1.2072,
      "step": 579
    },
    {
      "epoch": 0.8408843783979703,
      "grad_norm": 2.231603732801282,
      "learning_rate": 1.3009549733830582e-05,
      "loss": 1.085,
      "step": 580
    },
    {
      "epoch": 0.8423341790503805,
      "grad_norm": 2.1980905174338083,
      "learning_rate": 1.2987116709002794e-05,
      "loss": 1.1634,
      "step": 581
    },
    {
      "epoch": 0.8437839797027908,
      "grad_norm": 2.107706876497223,
      "learning_rate": 1.2964667166884005e-05,
      "loss": 1.0485,
      "step": 582
    },
    {
      "epoch": 0.8452337803552011,
      "grad_norm": 2.150156803318543,
      "learning_rate": 1.2942201231609185e-05,
      "loss": 1.1781,
      "step": 583
    },
    {
      "epoch": 0.8466835810076114,
      "grad_norm": 2.376389762401588,
      "learning_rate": 1.2919719027403935e-05,
      "loss": 1.0576,
      "step": 584
    },
    {
      "epoch": 0.8481333816600217,
      "grad_norm": 2.277091916668181,
      "learning_rate": 1.2897220678583826e-05,
      "loss": 1.1035,
      "step": 585
    },
    {
      "epoch": 0.849583182312432,
      "grad_norm": 2.3213498233933634,
      "learning_rate": 1.2874706309553697e-05,
      "loss": 1.0907,
      "step": 586
    },
    {
      "epoch": 0.8510329829648423,
      "grad_norm": 2.0118413972624607,
      "learning_rate": 1.285217604480697e-05,
      "loss": 1.0519,
      "step": 587
    },
    {
      "epoch": 0.8524827836172526,
      "grad_norm": 2.4524891597533176,
      "learning_rate": 1.2829630008924966e-05,
      "loss": 1.2318,
      "step": 588
    },
    {
      "epoch": 0.8524827836172526,
      "eval_loss": 1.987449049949646,
      "eval_runtime": 717.049,
      "eval_samples_per_second": 24.62,
      "eval_steps_per_second": 0.385,
      "step": 588
    },
    {
      "epoch": 0.8539325842696629,
      "grad_norm": 2.158955058630859,
      "learning_rate": 1.2807068326576211e-05,
      "loss": 1.127,
      "step": 589
    },
    {
      "epoch": 0.8553823849220732,
      "grad_norm": 2.3927164380538395,
      "learning_rate": 1.2784491122515743e-05,
      "loss": 1.1141,
      "step": 590
    },
    {
      "epoch": 0.8568321855744835,
      "grad_norm": 2.298564279963096,
      "learning_rate": 1.2761898521584436e-05,
      "loss": 1.1749,
      "step": 591
    },
    {
      "epoch": 0.8582819862268938,
      "grad_norm": 2.046841333838917,
      "learning_rate": 1.2739290648708298e-05,
      "loss": 1.1862,
      "step": 592
    },
    {
      "epoch": 0.8597317868793041,
      "grad_norm": 2.2713309315995622,
      "learning_rate": 1.271666762889778e-05,
      "loss": 1.0988,
      "step": 593
    },
    {
      "epoch": 0.8611815875317144,
      "grad_norm": 2.3805845518343896,
      "learning_rate": 1.2694029587247097e-05,
      "loss": 1.1876,
      "step": 594
    },
    {
      "epoch": 0.8626313881841247,
      "grad_norm": 2.3267613591546388,
      "learning_rate": 1.2671376648933511e-05,
      "loss": 1.228,
      "step": 595
    },
    {
      "epoch": 0.864081188836535,
      "grad_norm": 2.456641790889024,
      "learning_rate": 1.2648708939216671e-05,
      "loss": 1.1265,
      "step": 596
    },
    {
      "epoch": 0.8655309894889452,
      "grad_norm": 2.4175844363963113,
      "learning_rate": 1.2626026583437902e-05,
      "loss": 1.1308,
      "step": 597
    },
    {
      "epoch": 0.8669807901413555,
      "grad_norm": 2.1743867370816266,
      "learning_rate": 1.260332970701951e-05,
      "loss": 1.2075,
      "step": 598
    },
    {
      "epoch": 0.8684305907937658,
      "grad_norm": 2.024454705030007,
      "learning_rate": 1.2580618435464092e-05,
      "loss": 1.1327,
      "step": 599
    },
    {
      "epoch": 0.8698803914461761,
      "grad_norm": 2.1962456975987066,
      "learning_rate": 1.2557892894353848e-05,
      "loss": 1.0799,
      "step": 600
    },
    {
      "epoch": 0.8713301920985864,
      "grad_norm": 2.3202216455747657,
      "learning_rate": 1.2535153209349878e-05,
      "loss": 1.1399,
      "step": 601
    },
    {
      "epoch": 0.8727799927509967,
      "grad_norm": 1.9875458863552233,
      "learning_rate": 1.251239950619149e-05,
      "loss": 1.1356,
      "step": 602
    },
    {
      "epoch": 0.8727799927509967,
      "eval_loss": 1.9592655897140503,
      "eval_runtime": 713.115,
      "eval_samples_per_second": 24.756,
      "eval_steps_per_second": 0.387,
      "step": 602
    },
    {
      "epoch": 0.874229793403407,
      "grad_norm": 2.1293267635542543,
      "learning_rate": 1.2489631910695513e-05,
      "loss": 1.0188,
      "step": 603
    },
    {
      "epoch": 0.8756795940558173,
      "grad_norm": 2.0894266999465785,
      "learning_rate": 1.2466850548755586e-05,
      "loss": 1.0736,
      "step": 604
    },
    {
      "epoch": 0.8771293947082276,
      "grad_norm": 2.1836390564485617,
      "learning_rate": 1.2444055546341474e-05,
      "loss": 1.0383,
      "step": 605
    },
    {
      "epoch": 0.8785791953606379,
      "grad_norm": 2.109632418691755,
      "learning_rate": 1.242124702949836e-05,
      "loss": 1.1516,
      "step": 606
    },
    {
      "epoch": 0.8800289960130482,
      "grad_norm": 2.1115945719216773,
      "learning_rate": 1.239842512434617e-05,
      "loss": 1.109,
      "step": 607
    },
    {
      "epoch": 0.8814787966654585,
      "grad_norm": 2.6520432577176303,
      "learning_rate": 1.237558995707884e-05,
      "loss": 1.069,
      "step": 608
    },
    {
      "epoch": 0.8829285973178688,
      "grad_norm": 2.3447315617277167,
      "learning_rate": 1.2352741653963657e-05,
      "loss": 1.2223,
      "step": 609
    },
    {
      "epoch": 0.8843783979702791,
      "grad_norm": 2.1460706989623803,
      "learning_rate": 1.2329880341340533e-05,
      "loss": 1.0939,
      "step": 610
    },
    {
      "epoch": 0.8858281986226894,
      "grad_norm": 2.246701770370575,
      "learning_rate": 1.2307006145621326e-05,
      "loss": 1.1458,
      "step": 611
    },
    {
      "epoch": 0.8872779992750996,
      "grad_norm": 2.371069430412228,
      "learning_rate": 1.2284119193289114e-05,
      "loss": 1.193,
      "step": 612
    },
    {
      "epoch": 0.8887277999275099,
      "grad_norm": 2.4341945214071896,
      "learning_rate": 1.2261219610897535e-05,
      "loss": 1.1163,
      "step": 613
    },
    {
      "epoch": 0.8901776005799202,
      "grad_norm": 2.1403121332367445,
      "learning_rate": 1.223830752507004e-05,
      "loss": 1.1475,
      "step": 614
    },
    {
      "epoch": 0.8916274012323305,
      "grad_norm": 1.9768990530648036,
      "learning_rate": 1.2215383062499246e-05,
      "loss": 1.1393,
      "step": 615
    },
    {
      "epoch": 0.8930772018847408,
      "grad_norm": 2.2812772488409454,
      "learning_rate": 1.2192446349946181e-05,
      "loss": 1.275,
      "step": 616
    },
    {
      "epoch": 0.8930772018847408,
      "eval_loss": 1.9718694686889648,
      "eval_runtime": 712.1399,
      "eval_samples_per_second": 24.79,
      "eval_steps_per_second": 0.388,
      "step": 616
    },
    {
      "epoch": 0.8945270025371511,
      "grad_norm": 2.0838776441591738,
      "learning_rate": 1.2169497514239622e-05,
      "loss": 1.1022,
      "step": 617
    },
    {
      "epoch": 0.8959768031895614,
      "grad_norm": 2.055436398461139,
      "learning_rate": 1.2146536682275388e-05,
      "loss": 1.0527,
      "step": 618
    },
    {
      "epoch": 0.8974266038419717,
      "grad_norm": 2.2040994002601506,
      "learning_rate": 1.2123563981015615e-05,
      "loss": 1.1063,
      "step": 619
    },
    {
      "epoch": 0.898876404494382,
      "grad_norm": 2.020691006733176,
      "learning_rate": 1.2100579537488084e-05,
      "loss": 1.1567,
      "step": 620
    },
    {
      "epoch": 0.9003262051467923,
      "grad_norm": 2.4162642894108637,
      "learning_rate": 1.2077583478785497e-05,
      "loss": 1.1012,
      "step": 621
    },
    {
      "epoch": 0.9017760057992026,
      "grad_norm": 2.1650104706017834,
      "learning_rate": 1.2054575932064787e-05,
      "loss": 1.1902,
      "step": 622
    },
    {
      "epoch": 0.9032258064516129,
      "grad_norm": 1.9095006939868062,
      "learning_rate": 1.2031557024546407e-05,
      "loss": 1.0361,
      "step": 623
    },
    {
      "epoch": 0.9046756071040232,
      "grad_norm": 2.203179982627449,
      "learning_rate": 1.200852688351363e-05,
      "loss": 1.1131,
      "step": 624
    },
    {
      "epoch": 0.9061254077564335,
      "grad_norm": 2.060457425334721,
      "learning_rate": 1.1985485636311846e-05,
      "loss": 1.1264,
      "step": 625
    },
    {
      "epoch": 0.9075752084088438,
      "grad_norm": 2.4485675238329403,
      "learning_rate": 1.196243341034786e-05,
      "loss": 1.1357,
      "step": 626
    },
    {
      "epoch": 0.909025009061254,
      "grad_norm": 2.180367774596997,
      "learning_rate": 1.1939370333089175e-05,
      "loss": 1.2013,
      "step": 627
    },
    {
      "epoch": 0.9104748097136643,
      "grad_norm": 2.222915445489064,
      "learning_rate": 1.1916296532063308e-05,
      "loss": 1.1349,
      "step": 628
    },
    {
      "epoch": 0.9119246103660746,
      "grad_norm": 2.3347795993674225,
      "learning_rate": 1.189321213485706e-05,
      "loss": 1.1355,
      "step": 629
    },
    {
      "epoch": 0.9133744110184849,
      "grad_norm": 2.300879199013833,
      "learning_rate": 1.1870117269115833e-05,
      "loss": 1.1664,
      "step": 630
    },
    {
      "epoch": 0.9133744110184849,
      "eval_loss": 1.9626564979553223,
      "eval_runtime": 712.6303,
      "eval_samples_per_second": 24.773,
      "eval_steps_per_second": 0.387,
      "step": 630
    },
    {
      "epoch": 0.9148242116708952,
      "grad_norm": 2.2753273516872845,
      "learning_rate": 1.1847012062542914e-05,
      "loss": 1.2055,
      "step": 631
    },
    {
      "epoch": 0.9162740123233055,
      "grad_norm": 2.3172837045251544,
      "learning_rate": 1.1823896642898769e-05,
      "loss": 1.1371,
      "step": 632
    },
    {
      "epoch": 0.9177238129757158,
      "grad_norm": 2.1210493196147047,
      "learning_rate": 1.1800771138000334e-05,
      "loss": 1.1175,
      "step": 633
    },
    {
      "epoch": 0.9191736136281261,
      "grad_norm": 2.1359623685879305,
      "learning_rate": 1.1777635675720313e-05,
      "loss": 1.0994,
      "step": 634
    },
    {
      "epoch": 0.9206234142805364,
      "grad_norm": 1.93072192368431,
      "learning_rate": 1.175449038398648e-05,
      "loss": 1.1648,
      "step": 635
    },
    {
      "epoch": 0.9220732149329467,
      "grad_norm": 2.2234339241139063,
      "learning_rate": 1.1731335390780935e-05,
      "loss": 1.048,
      "step": 636
    },
    {
      "epoch": 0.923523015585357,
      "grad_norm": 2.452328995299883,
      "learning_rate": 1.1708170824139453e-05,
      "loss": 1.0724,
      "step": 637
    },
    {
      "epoch": 0.9249728162377673,
      "grad_norm": 2.3307725415549188,
      "learning_rate": 1.1684996812150726e-05,
      "loss": 1.1708,
      "step": 638
    },
    {
      "epoch": 0.9264226168901776,
      "grad_norm": 2.074172194332946,
      "learning_rate": 1.1661813482955682e-05,
      "loss": 1.1357,
      "step": 639
    },
    {
      "epoch": 0.9278724175425879,
      "grad_norm": 1.9694157236917276,
      "learning_rate": 1.1638620964746761e-05,
      "loss": 1.0663,
      "step": 640
    },
    {
      "epoch": 0.9293222181949982,
      "grad_norm": 2.4038365134722777,
      "learning_rate": 1.1615419385767223e-05,
      "loss": 1.1508,
      "step": 641
    },
    {
      "epoch": 0.9307720188474085,
      "grad_norm": 2.094518955914741,
      "learning_rate": 1.1592208874310426e-05,
      "loss": 1.1687,
      "step": 642
    },
    {
      "epoch": 0.9322218194998187,
      "grad_norm": 2.1627586148563136,
      "learning_rate": 1.1568989558719118e-05,
      "loss": 1.0259,
      "step": 643
    },
    {
      "epoch": 0.933671620152229,
      "grad_norm": 2.1568130227121523,
      "learning_rate": 1.1545761567384725e-05,
      "loss": 1.1062,
      "step": 644
    },
    {
      "epoch": 0.933671620152229,
      "eval_loss": 1.9537007808685303,
      "eval_runtime": 713.167,
      "eval_samples_per_second": 24.754,
      "eval_steps_per_second": 0.387,
      "step": 644
    },
    {
      "epoch": 0.9351214208046393,
      "grad_norm": 2.177052691313776,
      "learning_rate": 1.1522525028746665e-05,
      "loss": 1.187,
      "step": 645
    },
    {
      "epoch": 0.9365712214570496,
      "grad_norm": 2.310237953537198,
      "learning_rate": 1.1499280071291596e-05,
      "loss": 1.1035,
      "step": 646
    },
    {
      "epoch": 0.9380210221094599,
      "grad_norm": 2.2502346034533507,
      "learning_rate": 1.147602682355274e-05,
      "loss": 1.1438,
      "step": 647
    },
    {
      "epoch": 0.9394708227618702,
      "grad_norm": 2.201614071556744,
      "learning_rate": 1.1452765414109155e-05,
      "loss": 1.1702,
      "step": 648
    },
    {
      "epoch": 0.9409206234142805,
      "grad_norm": 2.282962146128519,
      "learning_rate": 1.1429495971585032e-05,
      "loss": 1.1251,
      "step": 649
    },
    {
      "epoch": 0.9423704240666908,
      "grad_norm": 2.081159532621975,
      "learning_rate": 1.1406218624648986e-05,
      "loss": 1.0626,
      "step": 650
    },
    {
      "epoch": 0.9438202247191011,
      "grad_norm": 2.0942013634152388,
      "learning_rate": 1.1382933502013325e-05,
      "loss": 1.1765,
      "step": 651
    },
    {
      "epoch": 0.9452700253715114,
      "grad_norm": 2.2353858341609123,
      "learning_rate": 1.1359640732433369e-05,
      "loss": 1.1733,
      "step": 652
    },
    {
      "epoch": 0.9467198260239217,
      "grad_norm": 2.030359060023634,
      "learning_rate": 1.133634044470671e-05,
      "loss": 1.0991,
      "step": 653
    },
    {
      "epoch": 0.948169626676332,
      "grad_norm": 2.0882509981340607,
      "learning_rate": 1.1313032767672525e-05,
      "loss": 1.2229,
      "step": 654
    },
    {
      "epoch": 0.9496194273287423,
      "grad_norm": 2.1344950764995834,
      "learning_rate": 1.1289717830210828e-05,
      "loss": 1.1069,
      "step": 655
    },
    {
      "epoch": 0.9510692279811526,
      "grad_norm": 2.1570234549786864,
      "learning_rate": 1.1266395761241809e-05,
      "loss": 1.1037,
      "step": 656
    },
    {
      "epoch": 0.9525190286335629,
      "grad_norm": 2.424170346430373,
      "learning_rate": 1.1243066689725062e-05,
      "loss": 1.0986,
      "step": 657
    },
    {
      "epoch": 0.9539688292859732,
      "grad_norm": 3.175534736084341,
      "learning_rate": 1.1219730744658921e-05,
      "loss": 1.092,
      "step": 658
    },
    {
      "epoch": 0.9539688292859732,
      "eval_loss": 1.9552119970321655,
      "eval_runtime": 715.0011,
      "eval_samples_per_second": 24.691,
      "eval_steps_per_second": 0.386,
      "step": 658
    },
    {
      "epoch": 0.9554186299383834,
      "grad_norm": 2.2811343427428024,
      "learning_rate": 1.1196388055079724e-05,
      "loss": 1.1491,
      "step": 659
    },
    {
      "epoch": 0.9568684305907937,
      "grad_norm": 2.057888403182058,
      "learning_rate": 1.1173038750061099e-05,
      "loss": 1.0762,
      "step": 660
    },
    {
      "epoch": 0.958318231243204,
      "grad_norm": 2.2600060184331494,
      "learning_rate": 1.1149682958713255e-05,
      "loss": 1.196,
      "step": 661
    },
    {
      "epoch": 0.9597680318956143,
      "grad_norm": 2.18171044788343,
      "learning_rate": 1.1126320810182271e-05,
      "loss": 1.0778,
      "step": 662
    },
    {
      "epoch": 0.9612178325480246,
      "grad_norm": 2.0450660637163116,
      "learning_rate": 1.1102952433649374e-05,
      "loss": 1.0727,
      "step": 663
    },
    {
      "epoch": 0.9626676332004349,
      "grad_norm": 2.204974408075549,
      "learning_rate": 1.107957795833023e-05,
      "loss": 1.0751,
      "step": 664
    },
    {
      "epoch": 0.9641174338528452,
      "grad_norm": 2.590163626713659,
      "learning_rate": 1.1056197513474227e-05,
      "loss": 1.108,
      "step": 665
    },
    {
      "epoch": 0.9655672345052555,
      "grad_norm": 2.205592113159895,
      "learning_rate": 1.1032811228363766e-05,
      "loss": 1.0991,
      "step": 666
    },
    {
      "epoch": 0.9670170351576658,
      "grad_norm": 1.9951568102868962,
      "learning_rate": 1.1009419232313536e-05,
      "loss": 1.0891,
      "step": 667
    },
    {
      "epoch": 0.9684668358100761,
      "grad_norm": 2.243426197206969,
      "learning_rate": 1.0986021654669807e-05,
      "loss": 1.1138,
      "step": 668
    },
    {
      "epoch": 0.9699166364624864,
      "grad_norm": 2.3372899568500514,
      "learning_rate": 1.0962618624809718e-05,
      "loss": 1.1549,
      "step": 669
    },
    {
      "epoch": 0.9713664371148967,
      "grad_norm": 2.051502780717789,
      "learning_rate": 1.0939210272140545e-05,
      "loss": 1.1093,
      "step": 670
    },
    {
      "epoch": 0.972816237767307,
      "grad_norm": 2.459046512891056,
      "learning_rate": 1.0915796726099008e-05,
      "loss": 1.1394,
      "step": 671
    },
    {
      "epoch": 0.9742660384197173,
      "grad_norm": 2.301284925658402,
      "learning_rate": 1.089237811615053e-05,
      "loss": 1.1812,
      "step": 672
    },
    {
      "epoch": 0.9742660384197173,
      "eval_loss": 1.9643124341964722,
      "eval_runtime": 707.6968,
      "eval_samples_per_second": 24.946,
      "eval_steps_per_second": 0.39,
      "step": 672
    },
    {
      "epoch": 0.9757158390721276,
      "grad_norm": 2.4687074035876675,
      "learning_rate": 1.0868954571788554e-05,
      "loss": 1.166,
      "step": 673
    },
    {
      "epoch": 0.9771656397245378,
      "grad_norm": 2.0857581917550294,
      "learning_rate": 1.0845526222533791e-05,
      "loss": 1.0818,
      "step": 674
    },
    {
      "epoch": 0.9786154403769481,
      "grad_norm": 2.1656788415535284,
      "learning_rate": 1.0822093197933528e-05,
      "loss": 1.0989,
      "step": 675
    },
    {
      "epoch": 0.9800652410293584,
      "grad_norm": 2.4751829301849457,
      "learning_rate": 1.07986556275609e-05,
      "loss": 1.1796,
      "step": 676
    },
    {
      "epoch": 0.9815150416817687,
      "grad_norm": 2.096671280684346,
      "learning_rate": 1.077521364101419e-05,
      "loss": 1.0937,
      "step": 677
    },
    {
      "epoch": 0.982964842334179,
      "grad_norm": 2.2584523725403085,
      "learning_rate": 1.0751767367916081e-05,
      "loss": 1.0891,
      "step": 678
    },
    {
      "epoch": 0.9844146429865893,
      "grad_norm": 2.240019992405247,
      "learning_rate": 1.0728316937912975e-05,
      "loss": 1.0762,
      "step": 679
    },
    {
      "epoch": 0.9858644436389996,
      "grad_norm": 2.1848259800290726,
      "learning_rate": 1.0704862480674251e-05,
      "loss": 1.1808,
      "step": 680
    },
    {
      "epoch": 0.9873142442914099,
      "grad_norm": 2.3103567570570065,
      "learning_rate": 1.0681404125891561e-05,
      "loss": 1.0629,
      "step": 681
    },
    {
      "epoch": 0.9887640449438202,
      "grad_norm": 2.4899089064845454,
      "learning_rate": 1.0657942003278107e-05,
      "loss": 1.1868,
      "step": 682
    },
    {
      "epoch": 0.9902138455962305,
      "grad_norm": 2.0374871709611164,
      "learning_rate": 1.0634476242567924e-05,
      "loss": 1.0669,
      "step": 683
    },
    {
      "epoch": 0.9916636462486408,
      "grad_norm": 2.2368915378481304,
      "learning_rate": 1.0611006973515165e-05,
      "loss": 1.1445,
      "step": 684
    },
    {
      "epoch": 0.9931134469010511,
      "grad_norm": 2.177429690749187,
      "learning_rate": 1.0587534325893383e-05,
      "loss": 1.1697,
      "step": 685
    },
    {
      "epoch": 0.9945632475534614,
      "grad_norm": 2.1970475841521275,
      "learning_rate": 1.0564058429494808e-05,
      "loss": 1.195,
      "step": 686
    },
    {
      "epoch": 0.9945632475534614,
      "eval_loss": 1.96137273311615,
      "eval_runtime": 715.8611,
      "eval_samples_per_second": 24.661,
      "eval_steps_per_second": 0.386,
      "step": 686
    },
    {
      "epoch": 0.9960130482058717,
      "grad_norm": 2.143687342321846,
      "learning_rate": 1.0540579414129644e-05,
      "loss": 1.1646,
      "step": 687
    },
    {
      "epoch": 0.997462848858282,
      "grad_norm": 2.4545719280313065,
      "learning_rate": 1.0517097409625335e-05,
      "loss": 1.2362,
      "step": 688
    },
    {
      "epoch": 0.9989126495106923,
      "grad_norm": 2.262885077378799,
      "learning_rate": 1.0493612545825851e-05,
      "loss": 1.0789,
      "step": 689
    },
    {
      "epoch": 1.0003624501631025,
      "grad_norm": 2.2905142670931498,
      "learning_rate": 1.0470124952590976e-05,
      "loss": 1.0639,
      "step": 690
    },
    {
      "epoch": 1.001812250815513,
      "grad_norm": 2.524422721180522,
      "learning_rate": 1.0446634759795588e-05,
      "loss": 0.9745,
      "step": 691
    },
    {
      "epoch": 1.0032620514679231,
      "grad_norm": 2.1088843871671603,
      "learning_rate": 1.0423142097328938e-05,
      "loss": 0.9725,
      "step": 692
    },
    {
      "epoch": 1.0047118521203335,
      "grad_norm": 2.0134889872231043,
      "learning_rate": 1.0399647095093926e-05,
      "loss": 0.8939,
      "step": 693
    },
    {
      "epoch": 1.0061616527727437,
      "grad_norm": 2.3846670769399627,
      "learning_rate": 1.0376149883006406e-05,
      "loss": 0.8668,
      "step": 694
    },
    {
      "epoch": 1.0076114534251541,
      "grad_norm": 2.757915499198385,
      "learning_rate": 1.0352650590994433e-05,
      "loss": 0.8438,
      "step": 695
    },
    {
      "epoch": 1.0090612540775643,
      "grad_norm": 2.7259792400980056,
      "learning_rate": 1.0329149348997576e-05,
      "loss": 0.9383,
      "step": 696
    },
    {
      "epoch": 1.0105110547299747,
      "grad_norm": 2.2682262169396084,
      "learning_rate": 1.0305646286966185e-05,
      "loss": 0.8162,
      "step": 697
    },
    {
      "epoch": 1.011960855382385,
      "grad_norm": 2.112720257288056,
      "learning_rate": 1.0282141534860662e-05,
      "loss": 0.8496,
      "step": 698
    },
    {
      "epoch": 1.0134106560347953,
      "grad_norm": 2.045060501976874,
      "learning_rate": 1.025863522265077e-05,
      "loss": 0.8717,
      "step": 699
    },
    {
      "epoch": 1.0148604566872055,
      "grad_norm": 2.2388848057175172,
      "learning_rate": 1.0235127480314888e-05,
      "loss": 0.872,
      "step": 700
    },
    {
      "epoch": 1.0148604566872055,
      "eval_loss": 1.9942997694015503,
      "eval_runtime": 714.9907,
      "eval_samples_per_second": 24.691,
      "eval_steps_per_second": 0.386,
      "step": 700
    },
    {
      "epoch": 1.0163102573396159,
      "grad_norm": 2.212449545902223,
      "learning_rate": 1.0211618437839313e-05,
      "loss": 0.7782,
      "step": 701
    },
    {
      "epoch": 1.017760057992026,
      "grad_norm": 2.3214572949015135,
      "learning_rate": 1.0188108225217516e-05,
      "loss": 0.901,
      "step": 702
    },
    {
      "epoch": 1.0192098586444365,
      "grad_norm": 2.178993569119099,
      "learning_rate": 1.0164596972449457e-05,
      "loss": 0.8226,
      "step": 703
    },
    {
      "epoch": 1.0206596592968467,
      "grad_norm": 2.4273816622299873,
      "learning_rate": 1.0141084809540826e-05,
      "loss": 0.9174,
      "step": 704
    },
    {
      "epoch": 1.022109459949257,
      "grad_norm": 2.152663263884313,
      "learning_rate": 1.0117571866502369e-05,
      "loss": 0.8247,
      "step": 705
    },
    {
      "epoch": 1.0235592606016672,
      "grad_norm": 2.3339012013061207,
      "learning_rate": 1.0094058273349125e-05,
      "loss": 0.859,
      "step": 706
    },
    {
      "epoch": 1.0250090612540776,
      "grad_norm": 2.1814985086443452,
      "learning_rate": 1.0070544160099742e-05,
      "loss": 0.8358,
      "step": 707
    },
    {
      "epoch": 1.0264588619064878,
      "grad_norm": 2.2230048297978544,
      "learning_rate": 1.0047029656775735e-05,
      "loss": 0.8346,
      "step": 708
    },
    {
      "epoch": 1.0279086625588982,
      "grad_norm": 2.473681629605397,
      "learning_rate": 1.0023514893400786e-05,
      "loss": 0.8307,
      "step": 709
    },
    {
      "epoch": 1.0293584632113084,
      "grad_norm": 2.1711887247941486,
      "learning_rate": 1e-05,
      "loss": 0.9156,
      "step": 710
    },
    {
      "epoch": 1.0308082638637188,
      "grad_norm": 2.2238475874879247,
      "learning_rate": 9.976485106599217e-06,
      "loss": 0.8822,
      "step": 711
    },
    {
      "epoch": 1.032258064516129,
      "grad_norm": 2.2090836214952634,
      "learning_rate": 9.952970343224268e-06,
      "loss": 0.8321,
      "step": 712
    },
    {
      "epoch": 1.0337078651685394,
      "grad_norm": 2.1337966839921343,
      "learning_rate": 9.929455839900261e-06,
      "loss": 0.8732,
      "step": 713
    },
    {
      "epoch": 1.0351576658209496,
      "grad_norm": 2.1425781472569043,
      "learning_rate": 9.90594172665088e-06,
      "loss": 0.8532,
      "step": 714
    },
    {
      "epoch": 1.0351576658209496,
      "eval_loss": 1.988709807395935,
      "eval_runtime": 713.9582,
      "eval_samples_per_second": 24.727,
      "eval_steps_per_second": 0.387,
      "step": 714
    },
    {
      "epoch": 1.03660746647336,
      "grad_norm": 2.163095860814875,
      "learning_rate": 9.882428133497636e-06,
      "loss": 0.8918,
      "step": 715
    },
    {
      "epoch": 1.0380572671257702,
      "grad_norm": 2.104095711217899,
      "learning_rate": 9.858915190459179e-06,
      "loss": 0.9539,
      "step": 716
    },
    {
      "epoch": 1.0395070677781806,
      "grad_norm": 2.078232337601294,
      "learning_rate": 9.835403027550545e-06,
      "loss": 0.8672,
      "step": 717
    },
    {
      "epoch": 1.0409568684305908,
      "grad_norm": 2.223770760882681,
      "learning_rate": 9.811891774782484e-06,
      "loss": 0.885,
      "step": 718
    },
    {
      "epoch": 1.0424066690830012,
      "grad_norm": 2.3424282146978084,
      "learning_rate": 9.788381562160689e-06,
      "loss": 0.7915,
      "step": 719
    },
    {
      "epoch": 1.0438564697354114,
      "grad_norm": 2.3705894293086023,
      "learning_rate": 9.764872519685112e-06,
      "loss": 0.7858,
      "step": 720
    },
    {
      "epoch": 1.0453062703878218,
      "grad_norm": 2.430893547589542,
      "learning_rate": 9.741364777349232e-06,
      "loss": 0.9259,
      "step": 721
    },
    {
      "epoch": 1.046756071040232,
      "grad_norm": 2.245049408881038,
      "learning_rate": 9.717858465139341e-06,
      "loss": 0.8903,
      "step": 722
    },
    {
      "epoch": 1.0482058716926423,
      "grad_norm": 2.1242742695460586,
      "learning_rate": 9.694353713033819e-06,
      "loss": 0.8625,
      "step": 723
    },
    {
      "epoch": 1.0496556723450525,
      "grad_norm": 2.277584724064431,
      "learning_rate": 9.670850651002427e-06,
      "loss": 0.9242,
      "step": 724
    },
    {
      "epoch": 1.051105472997463,
      "grad_norm": 2.1538113260294662,
      "learning_rate": 9.64734940900557e-06,
      "loss": 0.8605,
      "step": 725
    },
    {
      "epoch": 1.0525552736498731,
      "grad_norm": 2.0910715171275585,
      "learning_rate": 9.623850116993598e-06,
      "loss": 0.9176,
      "step": 726
    },
    {
      "epoch": 1.0540050743022835,
      "grad_norm": 2.062145224287595,
      "learning_rate": 9.600352904906076e-06,
      "loss": 0.8456,
      "step": 727
    },
    {
      "epoch": 1.0554548749546937,
      "grad_norm": 2.0316057207135954,
      "learning_rate": 9.576857902671067e-06,
      "loss": 0.8061,
      "step": 728
    },
    {
      "epoch": 1.0554548749546937,
      "eval_loss": 1.9934834241867065,
      "eval_runtime": 712.804,
      "eval_samples_per_second": 24.767,
      "eval_steps_per_second": 0.387,
      "step": 728
    },
    {
      "epoch": 1.056904675607104,
      "grad_norm": 2.0926820765854006,
      "learning_rate": 9.553365240204417e-06,
      "loss": 0.8736,
      "step": 729
    },
    {
      "epoch": 1.0583544762595143,
      "grad_norm": 1.95640992866782,
      "learning_rate": 9.529875047409027e-06,
      "loss": 0.8644,
      "step": 730
    },
    {
      "epoch": 1.0598042769119247,
      "grad_norm": 2.302518483080677,
      "learning_rate": 9.50638745417415e-06,
      "loss": 0.9145,
      "step": 731
    },
    {
      "epoch": 1.0612540775643349,
      "grad_norm": 2.027942197770537,
      "learning_rate": 9.482902590374665e-06,
      "loss": 0.7982,
      "step": 732
    },
    {
      "epoch": 1.0627038782167453,
      "grad_norm": 2.0700797703339666,
      "learning_rate": 9.459420585870358e-06,
      "loss": 0.801,
      "step": 733
    },
    {
      "epoch": 1.0641536788691555,
      "grad_norm": 2.150689330034752,
      "learning_rate": 9.435941570505192e-06,
      "loss": 0.8645,
      "step": 734
    },
    {
      "epoch": 1.0656034795215659,
      "grad_norm": 2.008114666507977,
      "learning_rate": 9.412465674106622e-06,
      "loss": 0.8016,
      "step": 735
    },
    {
      "epoch": 1.067053280173976,
      "grad_norm": 2.634711435347276,
      "learning_rate": 9.388993026484837e-06,
      "loss": 0.914,
      "step": 736
    },
    {
      "epoch": 1.0685030808263865,
      "grad_norm": 2.243143515998206,
      "learning_rate": 9.36552375743208e-06,
      "loss": 0.8579,
      "step": 737
    },
    {
      "epoch": 1.0699528814787966,
      "grad_norm": 1.9706389707874192,
      "learning_rate": 9.342057996721894e-06,
      "loss": 0.8281,
      "step": 738
    },
    {
      "epoch": 1.071402682131207,
      "grad_norm": 2.427417234593578,
      "learning_rate": 9.318595874108442e-06,
      "loss": 0.8465,
      "step": 739
    },
    {
      "epoch": 1.0728524827836172,
      "grad_norm": 2.1344664536092286,
      "learning_rate": 9.295137519325752e-06,
      "loss": 0.8584,
      "step": 740
    },
    {
      "epoch": 1.0743022834360276,
      "grad_norm": 2.2762676903421064,
      "learning_rate": 9.271683062087029e-06,
      "loss": 0.8644,
      "step": 741
    },
    {
      "epoch": 1.0757520840884378,
      "grad_norm": 2.2668370112309324,
      "learning_rate": 9.248232632083924e-06,
      "loss": 0.9604,
      "step": 742
    },
    {
      "epoch": 1.0757520840884378,
      "eval_loss": 1.9986748695373535,
      "eval_runtime": 712.8024,
      "eval_samples_per_second": 24.767,
      "eval_steps_per_second": 0.387,
      "step": 742
    },
    {
      "epoch": 1.0772018847408482,
      "grad_norm": 2.201958961704166,
      "learning_rate": 9.224786358985814e-06,
      "loss": 0.8856,
      "step": 743
    },
    {
      "epoch": 1.0786516853932584,
      "grad_norm": 2.08792286695628,
      "learning_rate": 9.201344372439103e-06,
      "loss": 0.8807,
      "step": 744
    },
    {
      "epoch": 1.0801014860456688,
      "grad_norm": 2.2509417072753175,
      "learning_rate": 9.177906802066474e-06,
      "loss": 0.9223,
      "step": 745
    },
    {
      "epoch": 1.081551286698079,
      "grad_norm": 1.9967757083906108,
      "learning_rate": 9.154473777466209e-06,
      "loss": 0.9472,
      "step": 746
    },
    {
      "epoch": 1.0830010873504894,
      "grad_norm": 2.1793914764736213,
      "learning_rate": 9.131045428211446e-06,
      "loss": 0.784,
      "step": 747
    },
    {
      "epoch": 1.0844508880028996,
      "grad_norm": 2.2125551111713913,
      "learning_rate": 9.10762188384947e-06,
      "loss": 0.8966,
      "step": 748
    },
    {
      "epoch": 1.08590068865531,
      "grad_norm": 2.034174923665976,
      "learning_rate": 9.084203273900994e-06,
      "loss": 0.8849,
      "step": 749
    },
    {
      "epoch": 1.0873504893077202,
      "grad_norm": 2.2123030789742146,
      "learning_rate": 9.060789727859457e-06,
      "loss": 0.8766,
      "step": 750
    },
    {
      "epoch": 1.0888002899601306,
      "grad_norm": 2.095370602012184,
      "learning_rate": 9.037381375190285e-06,
      "loss": 0.9076,
      "step": 751
    },
    {
      "epoch": 1.0902500906125407,
      "grad_norm": 2.1168230992914463,
      "learning_rate": 9.013978345330195e-06,
      "loss": 0.883,
      "step": 752
    },
    {
      "epoch": 1.0916998912649511,
      "grad_norm": 2.276605870525737,
      "learning_rate": 8.990580767686467e-06,
      "loss": 0.8753,
      "step": 753
    },
    {
      "epoch": 1.0931496919173613,
      "grad_norm": 2.061418158102787,
      "learning_rate": 8.967188771636237e-06,
      "loss": 0.887,
      "step": 754
    },
    {
      "epoch": 1.0945994925697717,
      "grad_norm": 2.2741196559985943,
      "learning_rate": 8.943802486525777e-06,
      "loss": 0.8704,
      "step": 755
    },
    {
      "epoch": 1.096049293222182,
      "grad_norm": 2.0362793939697994,
      "learning_rate": 8.920422041669774e-06,
      "loss": 0.8444,
      "step": 756
    },
    {
      "epoch": 1.096049293222182,
      "eval_loss": 1.9978744983673096,
      "eval_runtime": 718.7815,
      "eval_samples_per_second": 24.561,
      "eval_steps_per_second": 0.384,
      "step": 756
    },
    {
      "epoch": 1.0974990938745923,
      "grad_norm": 2.124570729502526,
      "learning_rate": 8.89704756635063e-06,
      "loss": 0.879,
      "step": 757
    },
    {
      "epoch": 1.0989488945270025,
      "grad_norm": 2.2175364644533406,
      "learning_rate": 8.873679189817732e-06,
      "loss": 0.8576,
      "step": 758
    },
    {
      "epoch": 1.100398695179413,
      "grad_norm": 2.163709158992549,
      "learning_rate": 8.850317041286744e-06,
      "loss": 0.8372,
      "step": 759
    },
    {
      "epoch": 1.101848495831823,
      "grad_norm": 1.9823655423982065,
      "learning_rate": 8.826961249938903e-06,
      "loss": 0.8406,
      "step": 760
    },
    {
      "epoch": 1.1032982964842335,
      "grad_norm": 2.51762541543677,
      "learning_rate": 8.803611944920278e-06,
      "loss": 0.8987,
      "step": 761
    },
    {
      "epoch": 1.1047480971366437,
      "grad_norm": 2.0521382984247016,
      "learning_rate": 8.78026925534108e-06,
      "loss": 0.8578,
      "step": 762
    },
    {
      "epoch": 1.106197897789054,
      "grad_norm": 2.317038370933627,
      "learning_rate": 8.756933310274943e-06,
      "loss": 0.8725,
      "step": 763
    },
    {
      "epoch": 1.1076476984414643,
      "grad_norm": 2.244045388790396,
      "learning_rate": 8.733604238758196e-06,
      "loss": 0.9115,
      "step": 764
    },
    {
      "epoch": 1.1090974990938747,
      "grad_norm": 2.291098848529536,
      "learning_rate": 8.710282169789173e-06,
      "loss": 0.9545,
      "step": 765
    },
    {
      "epoch": 1.1105472997462849,
      "grad_norm": 2.2709502642864767,
      "learning_rate": 8.68696723232748e-06,
      "loss": 0.9418,
      "step": 766
    },
    {
      "epoch": 1.1119971003986953,
      "grad_norm": 2.1682893909743948,
      "learning_rate": 8.663659555293292e-06,
      "loss": 0.7669,
      "step": 767
    },
    {
      "epoch": 1.1134469010511054,
      "grad_norm": 2.2998203793993315,
      "learning_rate": 8.640359267566633e-06,
      "loss": 0.844,
      "step": 768
    },
    {
      "epoch": 1.1148967017035158,
      "grad_norm": 2.425957326712206,
      "learning_rate": 8.617066497986676e-06,
      "loss": 0.9188,
      "step": 769
    },
    {
      "epoch": 1.116346502355926,
      "grad_norm": 2.222990290966565,
      "learning_rate": 8.59378137535102e-06,
      "loss": 0.8623,
      "step": 770
    },
    {
      "epoch": 1.116346502355926,
      "eval_loss": 1.9970474243164062,
      "eval_runtime": 716.063,
      "eval_samples_per_second": 24.654,
      "eval_steps_per_second": 0.385,
      "step": 770
    },
    {
      "epoch": 1.1177963030083364,
      "grad_norm": 2.031741606672598,
      "learning_rate": 8.57050402841497e-06,
      "loss": 0.9106,
      "step": 771
    },
    {
      "epoch": 1.1192461036607466,
      "grad_norm": 2.1285633927820613,
      "learning_rate": 8.54723458589085e-06,
      "loss": 0.7975,
      "step": 772
    },
    {
      "epoch": 1.120695904313157,
      "grad_norm": 2.0308499858360123,
      "learning_rate": 8.523973176447264e-06,
      "loss": 0.8771,
      "step": 773
    },
    {
      "epoch": 1.1221457049655672,
      "grad_norm": 2.4025650630737676,
      "learning_rate": 8.500719928708406e-06,
      "loss": 0.9195,
      "step": 774
    },
    {
      "epoch": 1.1235955056179776,
      "grad_norm": 2.053053642799504,
      "learning_rate": 8.477474971253338e-06,
      "loss": 0.831,
      "step": 775
    },
    {
      "epoch": 1.1250453062703878,
      "grad_norm": 2.093123056406655,
      "learning_rate": 8.454238432615275e-06,
      "loss": 0.8482,
      "step": 776
    },
    {
      "epoch": 1.1264951069227982,
      "grad_norm": 2.4190304512851277,
      "learning_rate": 8.431010441280884e-06,
      "loss": 0.8235,
      "step": 777
    },
    {
      "epoch": 1.1279449075752084,
      "grad_norm": 2.104031395395008,
      "learning_rate": 8.407791125689577e-06,
      "loss": 0.7997,
      "step": 778
    },
    {
      "epoch": 1.1293947082276188,
      "grad_norm": 2.416480566955371,
      "learning_rate": 8.384580614232779e-06,
      "loss": 0.8628,
      "step": 779
    },
    {
      "epoch": 1.130844508880029,
      "grad_norm": 2.511019073961137,
      "learning_rate": 8.36137903525324e-06,
      "loss": 0.862,
      "step": 780
    },
    {
      "epoch": 1.1322943095324394,
      "grad_norm": 2.17454275925511,
      "learning_rate": 8.338186517044321e-06,
      "loss": 0.9645,
      "step": 781
    },
    {
      "epoch": 1.1337441101848496,
      "grad_norm": 2.1586107575866778,
      "learning_rate": 8.315003187849276e-06,
      "loss": 0.8381,
      "step": 782
    },
    {
      "epoch": 1.13519391083726,
      "grad_norm": 2.2761558860701454,
      "learning_rate": 8.29182917586055e-06,
      "loss": 0.7825,
      "step": 783
    },
    {
      "epoch": 1.1366437114896701,
      "grad_norm": 2.206844928639924,
      "learning_rate": 8.268664609219066e-06,
      "loss": 0.8842,
      "step": 784
    },
    {
      "epoch": 1.1366437114896701,
      "eval_loss": 1.9962694644927979,
      "eval_runtime": 718.1704,
      "eval_samples_per_second": 24.582,
      "eval_steps_per_second": 0.384,
      "step": 784
    },
    {
      "epoch": 1.1380935121420805,
      "grad_norm": 2.1540177237345723,
      "learning_rate": 8.245509616013527e-06,
      "loss": 0.8358,
      "step": 785
    },
    {
      "epoch": 1.1395433127944907,
      "grad_norm": 2.188546471502365,
      "learning_rate": 8.222364324279689e-06,
      "loss": 0.8413,
      "step": 786
    },
    {
      "epoch": 1.1409931134469011,
      "grad_norm": 2.2742971619683883,
      "learning_rate": 8.199228861999668e-06,
      "loss": 0.9081,
      "step": 787
    },
    {
      "epoch": 1.1424429140993113,
      "grad_norm": 2.4825375836535417,
      "learning_rate": 8.176103357101235e-06,
      "loss": 0.8751,
      "step": 788
    },
    {
      "epoch": 1.1438927147517217,
      "grad_norm": 2.3940410280187883,
      "learning_rate": 8.152987937457088e-06,
      "loss": 0.8414,
      "step": 789
    },
    {
      "epoch": 1.145342515404132,
      "grad_norm": 2.1299456223849322,
      "learning_rate": 8.129882730884168e-06,
      "loss": 0.8751,
      "step": 790
    },
    {
      "epoch": 1.1467923160565423,
      "grad_norm": 2.055692624643807,
      "learning_rate": 8.106787865142945e-06,
      "loss": 0.7687,
      "step": 791
    },
    {
      "epoch": 1.1482421167089525,
      "grad_norm": 2.2243992229333043,
      "learning_rate": 8.083703467936696e-06,
      "loss": 0.8313,
      "step": 792
    },
    {
      "epoch": 1.149691917361363,
      "grad_norm": 2.1503935839219057,
      "learning_rate": 8.060629666910827e-06,
      "loss": 0.8461,
      "step": 793
    },
    {
      "epoch": 1.151141718013773,
      "grad_norm": 2.2975679655704764,
      "learning_rate": 8.037566589652141e-06,
      "loss": 0.8667,
      "step": 794
    },
    {
      "epoch": 1.1525915186661835,
      "grad_norm": 2.1060486405315335,
      "learning_rate": 8.014514363688156e-06,
      "loss": 0.8296,
      "step": 795
    },
    {
      "epoch": 1.1540413193185937,
      "grad_norm": 2.1330972925646945,
      "learning_rate": 7.991473116486372e-06,
      "loss": 0.8936,
      "step": 796
    },
    {
      "epoch": 1.155491119971004,
      "grad_norm": 2.0773464296736934,
      "learning_rate": 7.968442975453598e-06,
      "loss": 0.8496,
      "step": 797
    },
    {
      "epoch": 1.1569409206234142,
      "grad_norm": 2.108334672914265,
      "learning_rate": 7.945424067935216e-06,
      "loss": 0.8485,
      "step": 798
    },
    {
      "epoch": 1.1569409206234142,
      "eval_loss": 1.995725154876709,
      "eval_runtime": 718.4729,
      "eval_samples_per_second": 24.572,
      "eval_steps_per_second": 0.384,
      "step": 798
    },
    {
      "epoch": 1.1583907212758247,
      "grad_norm": 2.084517567269502,
      "learning_rate": 7.922416521214505e-06,
      "loss": 0.8953,
      "step": 799
    },
    {
      "epoch": 1.1598405219282348,
      "grad_norm": 2.07952759443931,
      "learning_rate": 7.899420462511921e-06,
      "loss": 0.8923,
      "step": 800
    },
    {
      "epoch": 1.1612903225806452,
      "grad_norm": 2.3548728204828455,
      "learning_rate": 7.876436018984387e-06,
      "loss": 0.7979,
      "step": 801
    },
    {
      "epoch": 1.1627401232330554,
      "grad_norm": 2.428442543356957,
      "learning_rate": 7.853463317724614e-06,
      "loss": 0.9285,
      "step": 802
    },
    {
      "epoch": 1.1641899238854658,
      "grad_norm": 2.1545027106165677,
      "learning_rate": 7.83050248576038e-06,
      "loss": 0.838,
      "step": 803
    },
    {
      "epoch": 1.165639724537876,
      "grad_norm": 2.057576500116282,
      "learning_rate": 7.807553650053824e-06,
      "loss": 0.8271,
      "step": 804
    },
    {
      "epoch": 1.1670895251902864,
      "grad_norm": 2.210050668577785,
      "learning_rate": 7.784616937500757e-06,
      "loss": 0.8159,
      "step": 805
    },
    {
      "epoch": 1.1685393258426966,
      "grad_norm": 2.2742830791231867,
      "learning_rate": 7.761692474929961e-06,
      "loss": 0.831,
      "step": 806
    },
    {
      "epoch": 1.169989126495107,
      "grad_norm": 2.0205690560671576,
      "learning_rate": 7.73878038910247e-06,
      "loss": 0.7679,
      "step": 807
    },
    {
      "epoch": 1.1714389271475172,
      "grad_norm": 2.1676184835472556,
      "learning_rate": 7.715880806710888e-06,
      "loss": 0.9305,
      "step": 808
    },
    {
      "epoch": 1.1728887277999276,
      "grad_norm": 2.3993659314188482,
      "learning_rate": 7.692993854378678e-06,
      "loss": 0.8603,
      "step": 809
    },
    {
      "epoch": 1.1743385284523378,
      "grad_norm": 2.2302819877411473,
      "learning_rate": 7.670119658659469e-06,
      "loss": 0.8089,
      "step": 810
    },
    {
      "epoch": 1.1757883291047482,
      "grad_norm": 2.182787594232779,
      "learning_rate": 7.647258346036348e-06,
      "loss": 0.852,
      "step": 811
    },
    {
      "epoch": 1.1772381297571584,
      "grad_norm": 2.1686763913924363,
      "learning_rate": 7.6244100429211644e-06,
      "loss": 0.846,
      "step": 812
    },
    {
      "epoch": 1.1772381297571584,
      "eval_loss": 1.9976224899291992,
      "eval_runtime": 718.956,
      "eval_samples_per_second": 24.555,
      "eval_steps_per_second": 0.384,
      "step": 812
    },
    {
      "epoch": 1.1786879304095688,
      "grad_norm": 2.2558805796923043,
      "learning_rate": 7.601574875653836e-06,
      "loss": 0.8595,
      "step": 813
    },
    {
      "epoch": 1.180137731061979,
      "grad_norm": 2.4545945051684717,
      "learning_rate": 7.578752970501641e-06,
      "loss": 0.9004,
      "step": 814
    },
    {
      "epoch": 1.1815875317143893,
      "grad_norm": 2.320614980099812,
      "learning_rate": 7.555944453658528e-06,
      "loss": 0.8624,
      "step": 815
    },
    {
      "epoch": 1.1830373323667995,
      "grad_norm": 2.171534965481249,
      "learning_rate": 7.533149451244416e-06,
      "loss": 0.8068,
      "step": 816
    },
    {
      "epoch": 1.18448713301921,
      "grad_norm": 2.0894322310858913,
      "learning_rate": 7.510368089304489e-06,
      "loss": 0.8627,
      "step": 817
    },
    {
      "epoch": 1.1859369336716201,
      "grad_norm": 2.229336942699049,
      "learning_rate": 7.487600493808513e-06,
      "loss": 0.9392,
      "step": 818
    },
    {
      "epoch": 1.1873867343240305,
      "grad_norm": 2.1454217288375648,
      "learning_rate": 7.464846790650126e-06,
      "loss": 0.9056,
      "step": 819
    },
    {
      "epoch": 1.1888365349764407,
      "grad_norm": 2.1859031330177072,
      "learning_rate": 7.442107105646156e-06,
      "loss": 0.8359,
      "step": 820
    },
    {
      "epoch": 1.190286335628851,
      "grad_norm": 2.186177817226838,
      "learning_rate": 7.419381564535913e-06,
      "loss": 0.7719,
      "step": 821
    },
    {
      "epoch": 1.1917361362812613,
      "grad_norm": 2.1901417533502943,
      "learning_rate": 7.396670292980492e-06,
      "loss": 0.8149,
      "step": 822
    },
    {
      "epoch": 1.1931859369336717,
      "grad_norm": 2.200931214522664,
      "learning_rate": 7.373973416562101e-06,
      "loss": 0.846,
      "step": 823
    },
    {
      "epoch": 1.1946357375860819,
      "grad_norm": 2.1636651179824007,
      "learning_rate": 7.35129106078333e-06,
      "loss": 0.8392,
      "step": 824
    },
    {
      "epoch": 1.1960855382384923,
      "grad_norm": 2.0786109482633,
      "learning_rate": 7.328623351066494e-06,
      "loss": 0.818,
      "step": 825
    },
    {
      "epoch": 1.1975353388909025,
      "grad_norm": 2.2931337894550823,
      "learning_rate": 7.30597041275291e-06,
      "loss": 0.8665,
      "step": 826
    },
    {
      "epoch": 1.1975353388909025,
      "eval_loss": 2.005251884460449,
      "eval_runtime": 713.6306,
      "eval_samples_per_second": 24.738,
      "eval_steps_per_second": 0.387,
      "step": 826
    },
    {
      "epoch": 1.1989851395433129,
      "grad_norm": 2.1184239600705297,
      "learning_rate": 7.283332371102222e-06,
      "loss": 0.9272,
      "step": 827
    },
    {
      "epoch": 1.200434940195723,
      "grad_norm": 2.00553425555541,
      "learning_rate": 7.260709351291706e-06,
      "loss": 0.8026,
      "step": 828
    },
    {
      "epoch": 1.2018847408481335,
      "grad_norm": 2.2950133469880276,
      "learning_rate": 7.238101478415564e-06,
      "loss": 0.9522,
      "step": 829
    },
    {
      "epoch": 1.2033345415005436,
      "grad_norm": 2.8518472616743775,
      "learning_rate": 7.215508877484259e-06,
      "loss": 0.9323,
      "step": 830
    },
    {
      "epoch": 1.204784342152954,
      "grad_norm": 2.385481688048741,
      "learning_rate": 7.192931673423793e-06,
      "loss": 0.7776,
      "step": 831
    },
    {
      "epoch": 1.2062341428053642,
      "grad_norm": 2.3709910988324654,
      "learning_rate": 7.170369991075035e-06,
      "loss": 0.8075,
      "step": 832
    },
    {
      "epoch": 1.2076839434577746,
      "grad_norm": 2.392018383380877,
      "learning_rate": 7.147823955193034e-06,
      "loss": 0.8792,
      "step": 833
    },
    {
      "epoch": 1.2091337441101848,
      "grad_norm": 2.3946591046479324,
      "learning_rate": 7.125293690446307e-06,
      "loss": 0.8957,
      "step": 834
    },
    {
      "epoch": 1.2105835447625952,
      "grad_norm": 2.216448209995848,
      "learning_rate": 7.102779321416177e-06,
      "loss": 0.9366,
      "step": 835
    },
    {
      "epoch": 1.2120333454150054,
      "grad_norm": 2.558060314543385,
      "learning_rate": 7.08028097259607e-06,
      "loss": 0.957,
      "step": 836
    },
    {
      "epoch": 1.2134831460674158,
      "grad_norm": 2.2996062634686445,
      "learning_rate": 7.057798768390819e-06,
      "loss": 0.8851,
      "step": 837
    },
    {
      "epoch": 1.214932946719826,
      "grad_norm": 2.0875631171739846,
      "learning_rate": 7.035332833115998e-06,
      "loss": 0.8279,
      "step": 838
    },
    {
      "epoch": 1.2163827473722364,
      "grad_norm": 2.0861261926703674,
      "learning_rate": 7.01288329099721e-06,
      "loss": 0.8786,
      "step": 839
    },
    {
      "epoch": 1.2178325480246466,
      "grad_norm": 2.166945325837133,
      "learning_rate": 6.990450266169424e-06,
      "loss": 0.8676,
      "step": 840
    },
    {
      "epoch": 1.2178325480246466,
      "eval_loss": 2.004594564437866,
      "eval_runtime": 717.4731,
      "eval_samples_per_second": 24.606,
      "eval_steps_per_second": 0.385,
      "step": 840
    },
    {
      "epoch": 1.219282348677057,
      "grad_norm": 2.04243854688211,
      "learning_rate": 6.968033882676262e-06,
      "loss": 0.8094,
      "step": 841
    },
    {
      "epoch": 1.2207321493294672,
      "grad_norm": 2.1825813025386673,
      "learning_rate": 6.945634264469338e-06,
      "loss": 0.8545,
      "step": 842
    },
    {
      "epoch": 1.2221819499818776,
      "grad_norm": 2.355689342600703,
      "learning_rate": 6.923251535407559e-06,
      "loss": 0.8469,
      "step": 843
    },
    {
      "epoch": 1.2236317506342878,
      "grad_norm": 2.2130134504505903,
      "learning_rate": 6.900885819256449e-06,
      "loss": 0.7574,
      "step": 844
    },
    {
      "epoch": 1.2250815512866982,
      "grad_norm": 2.0768868346623273,
      "learning_rate": 6.878537239687448e-06,
      "loss": 0.8569,
      "step": 845
    },
    {
      "epoch": 1.2265313519391083,
      "grad_norm": 2.320385634420862,
      "learning_rate": 6.85620592027725e-06,
      "loss": 0.8999,
      "step": 846
    },
    {
      "epoch": 1.2279811525915187,
      "grad_norm": 2.37187750788015,
      "learning_rate": 6.833891984507097e-06,
      "loss": 0.8124,
      "step": 847
    },
    {
      "epoch": 1.229430953243929,
      "grad_norm": 2.2060471898810237,
      "learning_rate": 6.811595555762118e-06,
      "loss": 0.8991,
      "step": 848
    },
    {
      "epoch": 1.2308807538963393,
      "grad_norm": 2.1988571431075625,
      "learning_rate": 6.789316757330638e-06,
      "loss": 0.9304,
      "step": 849
    },
    {
      "epoch": 1.2323305545487495,
      "grad_norm": 2.0478198625384287,
      "learning_rate": 6.767055712403481e-06,
      "loss": 0.8371,
      "step": 850
    },
    {
      "epoch": 1.23378035520116,
      "grad_norm": 2.5814211124353856,
      "learning_rate": 6.74481254407332e-06,
      "loss": 0.9187,
      "step": 851
    },
    {
      "epoch": 1.23523015585357,
      "grad_norm": 2.2901629328107336,
      "learning_rate": 6.722587375333967e-06,
      "loss": 0.851,
      "step": 852
    },
    {
      "epoch": 1.2366799565059805,
      "grad_norm": 2.177830849725486,
      "learning_rate": 6.700380329079711e-06,
      "loss": 0.9085,
      "step": 853
    },
    {
      "epoch": 1.2381297571583907,
      "grad_norm": 2.1298754464694234,
      "learning_rate": 6.678191528104629e-06,
      "loss": 0.9365,
      "step": 854
    },
    {
      "epoch": 1.2381297571583907,
      "eval_loss": 2.01713228225708,
      "eval_runtime": 712.3702,
      "eval_samples_per_second": 24.782,
      "eval_steps_per_second": 0.387,
      "step": 854
    },
    {
      "epoch": 1.239579557810801,
      "grad_norm": 2.2747840086260696,
      "learning_rate": 6.656021095101919e-06,
      "loss": 0.7696,
      "step": 855
    },
    {
      "epoch": 1.2410293584632113,
      "grad_norm": 2.071973814133778,
      "learning_rate": 6.633869152663197e-06,
      "loss": 0.9447,
      "step": 856
    },
    {
      "epoch": 1.2424791591156217,
      "grad_norm": 2.2262582142639764,
      "learning_rate": 6.6117358232778494e-06,
      "loss": 0.8169,
      "step": 857
    },
    {
      "epoch": 1.2439289597680319,
      "grad_norm": 2.705184019947187,
      "learning_rate": 6.58962122933234e-06,
      "loss": 0.8884,
      "step": 858
    },
    {
      "epoch": 1.2453787604204423,
      "grad_norm": 2.017940423659092,
      "learning_rate": 6.567525493109533e-06,
      "loss": 0.79,
      "step": 859
    },
    {
      "epoch": 1.2468285610728524,
      "grad_norm": 2.4136072207468193,
      "learning_rate": 6.545448736788009e-06,
      "loss": 0.8945,
      "step": 860
    },
    {
      "epoch": 1.2482783617252629,
      "grad_norm": 2.1952109255564713,
      "learning_rate": 6.523391082441417e-06,
      "loss": 0.9153,
      "step": 861
    },
    {
      "epoch": 1.249728162377673,
      "grad_norm": 2.2668689977593197,
      "learning_rate": 6.501352652037765e-06,
      "loss": 0.8706,
      "step": 862
    },
    {
      "epoch": 1.2511779630300834,
      "grad_norm": 2.2380414315255077,
      "learning_rate": 6.47933356743877e-06,
      "loss": 0.8855,
      "step": 863
    },
    {
      "epoch": 1.2526277636824936,
      "grad_norm": 2.2708779635902205,
      "learning_rate": 6.457333950399181e-06,
      "loss": 0.9241,
      "step": 864
    },
    {
      "epoch": 1.254077564334904,
      "grad_norm": 2.1367778049519095,
      "learning_rate": 6.435353922566088e-06,
      "loss": 0.8715,
      "step": 865
    },
    {
      "epoch": 1.2555273649873142,
      "grad_norm": 2.1482439241763953,
      "learning_rate": 6.413393605478275e-06,
      "loss": 0.9199,
      "step": 866
    },
    {
      "epoch": 1.2569771656397246,
      "grad_norm": 2.130806402211673,
      "learning_rate": 6.391453120565525e-06,
      "loss": 0.8473,
      "step": 867
    },
    {
      "epoch": 1.2584269662921348,
      "grad_norm": 2.292713851550391,
      "learning_rate": 6.369532589147967e-06,
      "loss": 0.852,
      "step": 868
    },
    {
      "epoch": 1.2584269662921348,
      "eval_loss": 2.004816770553589,
      "eval_runtime": 713.3336,
      "eval_samples_per_second": 24.749,
      "eval_steps_per_second": 0.387,
      "step": 868
    },
    {
      "epoch": 1.2598767669445452,
      "grad_norm": 2.128171326841611,
      "learning_rate": 6.34763213243539e-06,
      "loss": 0.8863,
      "step": 869
    },
    {
      "epoch": 1.2613265675969554,
      "grad_norm": 2.1587604555413193,
      "learning_rate": 6.325751871526589e-06,
      "loss": 0.8443,
      "step": 870
    },
    {
      "epoch": 1.2627763682493658,
      "grad_norm": 2.4458195035246235,
      "learning_rate": 6.303891927408667e-06,
      "loss": 0.9638,
      "step": 871
    },
    {
      "epoch": 1.264226168901776,
      "grad_norm": 2.1703670577536167,
      "learning_rate": 6.282052420956417e-06,
      "loss": 0.8722,
      "step": 872
    },
    {
      "epoch": 1.2656759695541864,
      "grad_norm": 2.227233136350377,
      "learning_rate": 6.260233472931591e-06,
      "loss": 0.8217,
      "step": 873
    },
    {
      "epoch": 1.2671257702065966,
      "grad_norm": 2.845506725529196,
      "learning_rate": 6.238435203982278e-06,
      "loss": 0.8732,
      "step": 874
    },
    {
      "epoch": 1.268575570859007,
      "grad_norm": 2.3180215743367234,
      "learning_rate": 6.21665773464222e-06,
      "loss": 0.9219,
      "step": 875
    },
    {
      "epoch": 1.2700253715114171,
      "grad_norm": 2.3473281039548204,
      "learning_rate": 6.194901185330145e-06,
      "loss": 0.8781,
      "step": 876
    },
    {
      "epoch": 1.2714751721638275,
      "grad_norm": 2.2559456175352963,
      "learning_rate": 6.173165676349103e-06,
      "loss": 0.8533,
      "step": 877
    },
    {
      "epoch": 1.2729249728162377,
      "grad_norm": 2.2200488072007025,
      "learning_rate": 6.1514513278858055e-06,
      "loss": 0.8193,
      "step": 878
    },
    {
      "epoch": 1.2743747734686481,
      "grad_norm": 2.3118479051308367,
      "learning_rate": 6.129758260009956e-06,
      "loss": 0.8791,
      "step": 879
    },
    {
      "epoch": 1.2758245741210583,
      "grad_norm": 2.659974879482613,
      "learning_rate": 6.1080865926735814e-06,
      "loss": 0.8464,
      "step": 880
    },
    {
      "epoch": 1.2772743747734687,
      "grad_norm": 2.276682462999749,
      "learning_rate": 6.08643644571038e-06,
      "loss": 0.8413,
      "step": 881
    },
    {
      "epoch": 1.278724175425879,
      "grad_norm": 2.2726009827313445,
      "learning_rate": 6.064807938835046e-06,
      "loss": 0.7916,
      "step": 882
    },
    {
      "epoch": 1.278724175425879,
      "eval_loss": 1.9852372407913208,
      "eval_runtime": 713.9468,
      "eval_samples_per_second": 24.727,
      "eval_steps_per_second": 0.387,
      "step": 882
    },
    {
      "epoch": 1.2801739760782893,
      "grad_norm": 2.0962960726692237,
      "learning_rate": 6.043201191642629e-06,
      "loss": 0.8685,
      "step": 883
    },
    {
      "epoch": 1.2816237767306995,
      "grad_norm": 2.0048955605930434,
      "learning_rate": 6.021616323607834e-06,
      "loss": 0.8645,
      "step": 884
    },
    {
      "epoch": 1.28307357738311,
      "grad_norm": 2.481752223924236,
      "learning_rate": 6.000053454084416e-06,
      "loss": 0.8156,
      "step": 885
    },
    {
      "epoch": 1.28452337803552,
      "grad_norm": 2.308264264933887,
      "learning_rate": 5.978512702304461e-06,
      "loss": 0.9074,
      "step": 886
    },
    {
      "epoch": 1.2859731786879305,
      "grad_norm": 2.271177516699977,
      "learning_rate": 5.956994187377772e-06,
      "loss": 0.8062,
      "step": 887
    },
    {
      "epoch": 1.2874229793403407,
      "grad_norm": 2.0134320244475914,
      "learning_rate": 5.935498028291183e-06,
      "loss": 0.8088,
      "step": 888
    },
    {
      "epoch": 1.288872779992751,
      "grad_norm": 1.9940590250550827,
      "learning_rate": 5.914024343907921e-06,
      "loss": 0.9295,
      "step": 889
    },
    {
      "epoch": 1.2903225806451613,
      "grad_norm": 2.200027288390014,
      "learning_rate": 5.892573252966926e-06,
      "loss": 0.827,
      "step": 890
    },
    {
      "epoch": 1.2917723812975717,
      "grad_norm": 2.1606876260786487,
      "learning_rate": 5.871144874082222e-06,
      "loss": 0.8506,
      "step": 891
    },
    {
      "epoch": 1.2932221819499818,
      "grad_norm": 2.2411354289721457,
      "learning_rate": 5.849739325742234e-06,
      "loss": 0.8581,
      "step": 892
    },
    {
      "epoch": 1.2946719826023922,
      "grad_norm": 2.1648554588178186,
      "learning_rate": 5.828356726309154e-06,
      "loss": 0.9069,
      "step": 893
    },
    {
      "epoch": 1.2961217832548024,
      "grad_norm": 2.1357175729821125,
      "learning_rate": 5.806997194018272e-06,
      "loss": 0.8577,
      "step": 894
    },
    {
      "epoch": 1.2975715839072128,
      "grad_norm": 2.236716810169864,
      "learning_rate": 5.785660846977329e-06,
      "loss": 0.9094,
      "step": 895
    },
    {
      "epoch": 1.299021384559623,
      "grad_norm": 2.2513732080898716,
      "learning_rate": 5.7643478031658676e-06,
      "loss": 0.8906,
      "step": 896
    },
    {
      "epoch": 1.299021384559623,
      "eval_loss": 1.9898792505264282,
      "eval_runtime": 715.0349,
      "eval_samples_per_second": 24.69,
      "eval_steps_per_second": 0.386,
      "step": 896
    },
    {
      "epoch": 1.3004711852120334,
      "grad_norm": 2.3141235736550487,
      "learning_rate": 5.743058180434573e-06,
      "loss": 0.8315,
      "step": 897
    },
    {
      "epoch": 1.3019209858644436,
      "grad_norm": 2.0626506943605527,
      "learning_rate": 5.7217920965046115e-06,
      "loss": 0.8522,
      "step": 898
    },
    {
      "epoch": 1.303370786516854,
      "grad_norm": 2.133863353026807,
      "learning_rate": 5.700549668967013e-06,
      "loss": 0.8185,
      "step": 899
    },
    {
      "epoch": 1.3048205871692642,
      "grad_norm": 2.102587399141849,
      "learning_rate": 5.679331015281984e-06,
      "loss": 0.8653,
      "step": 900
    },
    {
      "epoch": 1.3062703878216746,
      "grad_norm": 2.1873631708085894,
      "learning_rate": 5.6581362527782825e-06,
      "loss": 0.8902,
      "step": 901
    },
    {
      "epoch": 1.3077201884740848,
      "grad_norm": 2.3131481728406293,
      "learning_rate": 5.636965498652544e-06,
      "loss": 0.8386,
      "step": 902
    },
    {
      "epoch": 1.3091699891264952,
      "grad_norm": 2.0945913695070555,
      "learning_rate": 5.615818869968666e-06,
      "loss": 0.8888,
      "step": 903
    },
    {
      "epoch": 1.3106197897789054,
      "grad_norm": 2.1585237201304004,
      "learning_rate": 5.594696483657138e-06,
      "loss": 0.8434,
      "step": 904
    },
    {
      "epoch": 1.3120695904313158,
      "grad_norm": 2.1681373769535384,
      "learning_rate": 5.573598456514406e-06,
      "loss": 0.7834,
      "step": 905
    },
    {
      "epoch": 1.313519391083726,
      "grad_norm": 2.2344550532719114,
      "learning_rate": 5.552524905202208e-06,
      "loss": 0.8404,
      "step": 906
    },
    {
      "epoch": 1.3149691917361364,
      "grad_norm": 2.1955307677560714,
      "learning_rate": 5.531475946246953e-06,
      "loss": 0.9281,
      "step": 907
    },
    {
      "epoch": 1.3164189923885465,
      "grad_norm": 2.5134422415937694,
      "learning_rate": 5.5104516960390695e-06,
      "loss": 0.8596,
      "step": 908
    },
    {
      "epoch": 1.317868793040957,
      "grad_norm": 2.13524443286974,
      "learning_rate": 5.489452270832354e-06,
      "loss": 0.8612,
      "step": 909
    },
    {
      "epoch": 1.3193185936933671,
      "grad_norm": 2.144450180853689,
      "learning_rate": 5.468477786743334e-06,
      "loss": 0.8455,
      "step": 910
    },
    {
      "epoch": 1.3193185936933671,
      "eval_loss": 1.9904202222824097,
      "eval_runtime": 715.7186,
      "eval_samples_per_second": 24.666,
      "eval_steps_per_second": 0.386,
      "step": 910
    },
    {
      "epoch": 1.3207683943457775,
      "grad_norm": 2.4998284879111243,
      "learning_rate": 5.447528359750617e-06,
      "loss": 0.8478,
      "step": 911
    },
    {
      "epoch": 1.3222181949981877,
      "grad_norm": 2.021568861904417,
      "learning_rate": 5.4266041056942644e-06,
      "loss": 0.8534,
      "step": 912
    },
    {
      "epoch": 1.3236679956505981,
      "grad_norm": 1.985089718577658,
      "learning_rate": 5.405705140275153e-06,
      "loss": 0.8598,
      "step": 913
    },
    {
      "epoch": 1.3251177963030083,
      "grad_norm": 2.1794190486608347,
      "learning_rate": 5.384831579054313e-06,
      "loss": 0.8406,
      "step": 914
    },
    {
      "epoch": 1.3265675969554187,
      "grad_norm": 2.231841777780217,
      "learning_rate": 5.363983537452298e-06,
      "loss": 0.842,
      "step": 915
    },
    {
      "epoch": 1.3280173976078289,
      "grad_norm": 2.225347720836563,
      "learning_rate": 5.343161130748564e-06,
      "loss": 0.7903,
      "step": 916
    },
    {
      "epoch": 1.3294671982602393,
      "grad_norm": 2.3232641250930737,
      "learning_rate": 5.3223644740808075e-06,
      "loss": 0.8781,
      "step": 917
    },
    {
      "epoch": 1.3309169989126495,
      "grad_norm": 1.9364149914472255,
      "learning_rate": 5.3015936824443526e-06,
      "loss": 0.8483,
      "step": 918
    },
    {
      "epoch": 1.3323667995650599,
      "grad_norm": 2.412452528282832,
      "learning_rate": 5.2808488706914855e-06,
      "loss": 0.8905,
      "step": 919
    },
    {
      "epoch": 1.33381660021747,
      "grad_norm": 2.2338265193497997,
      "learning_rate": 5.260130153530847e-06,
      "loss": 0.8256,
      "step": 920
    },
    {
      "epoch": 1.3352664008698805,
      "grad_norm": 2.3312910506199116,
      "learning_rate": 5.239437645526788e-06,
      "loss": 0.8378,
      "step": 921
    },
    {
      "epoch": 1.3367162015222906,
      "grad_norm": 2.2921394438367666,
      "learning_rate": 5.218771461098733e-06,
      "loss": 0.8277,
      "step": 922
    },
    {
      "epoch": 1.338166002174701,
      "grad_norm": 2.370990785536441,
      "learning_rate": 5.198131714520554e-06,
      "loss": 0.8607,
      "step": 923
    },
    {
      "epoch": 1.3396158028271112,
      "grad_norm": 2.237371918887581,
      "learning_rate": 5.177518519919924e-06,
      "loss": 0.7918,
      "step": 924
    },
    {
      "epoch": 1.3396158028271112,
      "eval_loss": 1.9944241046905518,
      "eval_runtime": 713.7331,
      "eval_samples_per_second": 24.735,
      "eval_steps_per_second": 0.387,
      "step": 924
    },
    {
      "epoch": 1.3410656034795216,
      "grad_norm": 2.1827542776775792,
      "learning_rate": 5.156931991277707e-06,
      "loss": 0.8524,
      "step": 925
    },
    {
      "epoch": 1.3425154041319318,
      "grad_norm": 2.443492220217258,
      "learning_rate": 5.136372242427316e-06,
      "loss": 0.7748,
      "step": 926
    },
    {
      "epoch": 1.3439652047843422,
      "grad_norm": 2.062916566525091,
      "learning_rate": 5.11583938705408e-06,
      "loss": 0.8073,
      "step": 927
    },
    {
      "epoch": 1.3454150054367524,
      "grad_norm": 2.0489653477918552,
      "learning_rate": 5.095333538694624e-06,
      "loss": 0.8264,
      "step": 928
    },
    {
      "epoch": 1.3468648060891628,
      "grad_norm": 2.1321873410493604,
      "learning_rate": 5.074854810736238e-06,
      "loss": 0.8665,
      "step": 929
    },
    {
      "epoch": 1.348314606741573,
      "grad_norm": 2.0815327828672876,
      "learning_rate": 5.054403316416247e-06,
      "loss": 0.9055,
      "step": 930
    },
    {
      "epoch": 1.3497644073939834,
      "grad_norm": 2.157848957858876,
      "learning_rate": 5.033979168821393e-06,
      "loss": 0.8507,
      "step": 931
    },
    {
      "epoch": 1.3512142080463936,
      "grad_norm": 2.223973992687024,
      "learning_rate": 5.0135824808871905e-06,
      "loss": 0.8813,
      "step": 932
    },
    {
      "epoch": 1.352664008698804,
      "grad_norm": 2.427673348217411,
      "learning_rate": 4.993213365397329e-06,
      "loss": 0.8597,
      "step": 933
    },
    {
      "epoch": 1.3541138093512142,
      "grad_norm": 2.1162229746493857,
      "learning_rate": 4.972871934983032e-06,
      "loss": 0.8743,
      "step": 934
    },
    {
      "epoch": 1.3555636100036246,
      "grad_norm": 2.5179711289658533,
      "learning_rate": 4.952558302122442e-06,
      "loss": 0.8942,
      "step": 935
    },
    {
      "epoch": 1.3570134106560348,
      "grad_norm": 2.2002795077688297,
      "learning_rate": 4.932272579139981e-06,
      "loss": 0.8302,
      "step": 936
    },
    {
      "epoch": 1.3584632113084452,
      "grad_norm": 2.3468239446931873,
      "learning_rate": 4.912014878205757e-06,
      "loss": 0.8237,
      "step": 937
    },
    {
      "epoch": 1.3599130119608553,
      "grad_norm": 2.2242795575198673,
      "learning_rate": 4.891785311334924e-06,
      "loss": 0.86,
      "step": 938
    },
    {
      "epoch": 1.3599130119608553,
      "eval_loss": 1.989232063293457,
      "eval_runtime": 717.5267,
      "eval_samples_per_second": 24.604,
      "eval_steps_per_second": 0.385,
      "step": 938
    },
    {
      "epoch": 1.3613628126132657,
      "grad_norm": 2.2304831818740953,
      "learning_rate": 4.8715839903870685e-06,
      "loss": 0.8369,
      "step": 939
    },
    {
      "epoch": 1.362812613265676,
      "grad_norm": 2.276465719762189,
      "learning_rate": 4.851411027065596e-06,
      "loss": 0.9036,
      "step": 940
    },
    {
      "epoch": 1.3642624139180863,
      "grad_norm": 2.1959067263618186,
      "learning_rate": 4.8312665329170995e-06,
      "loss": 0.8673,
      "step": 941
    },
    {
      "epoch": 1.3657122145704965,
      "grad_norm": 2.214423540208915,
      "learning_rate": 4.811150619330759e-06,
      "loss": 0.8078,
      "step": 942
    },
    {
      "epoch": 1.367162015222907,
      "grad_norm": 2.2825587387004855,
      "learning_rate": 4.791063397537714e-06,
      "loss": 0.8488,
      "step": 943
    },
    {
      "epoch": 1.368611815875317,
      "grad_norm": 2.136400646596556,
      "learning_rate": 4.771004978610457e-06,
      "loss": 0.8285,
      "step": 944
    },
    {
      "epoch": 1.3700616165277275,
      "grad_norm": 2.0998897096222144,
      "learning_rate": 4.750975473462204e-06,
      "loss": 0.8652,
      "step": 945
    },
    {
      "epoch": 1.3715114171801377,
      "grad_norm": 2.1379285230361598,
      "learning_rate": 4.730974992846304e-06,
      "loss": 0.8496,
      "step": 946
    },
    {
      "epoch": 1.372961217832548,
      "grad_norm": 2.281333983914172,
      "learning_rate": 4.711003647355606e-06,
      "loss": 0.8818,
      "step": 947
    },
    {
      "epoch": 1.3744110184849583,
      "grad_norm": 2.1708244664441394,
      "learning_rate": 4.6910615474218675e-06,
      "loss": 0.8437,
      "step": 948
    },
    {
      "epoch": 1.3758608191373687,
      "grad_norm": 2.322523494523771,
      "learning_rate": 4.671148803315118e-06,
      "loss": 0.9043,
      "step": 949
    },
    {
      "epoch": 1.3773106197897789,
      "grad_norm": 2.4353745350977083,
      "learning_rate": 4.651265525143069e-06,
      "loss": 0.9258,
      "step": 950
    },
    {
      "epoch": 1.3787604204421893,
      "grad_norm": 2.214843923001071,
      "learning_rate": 4.631411822850507e-06,
      "loss": 0.7514,
      "step": 951
    },
    {
      "epoch": 1.3802102210945995,
      "grad_norm": 2.227960164312638,
      "learning_rate": 4.611587806218669e-06,
      "loss": 0.8275,
      "step": 952
    },
    {
      "epoch": 1.3802102210945995,
      "eval_loss": 1.998836636543274,
      "eval_runtime": 716.5656,
      "eval_samples_per_second": 24.637,
      "eval_steps_per_second": 0.385,
      "step": 952
    },
    {
      "epoch": 1.3816600217470099,
      "grad_norm": 2.3176111858923885,
      "learning_rate": 4.591793584864652e-06,
      "loss": 0.8383,
      "step": 953
    },
    {
      "epoch": 1.38310982239942,
      "grad_norm": 2.1539844922147044,
      "learning_rate": 4.572029268240787e-06,
      "loss": 0.9097,
      "step": 954
    },
    {
      "epoch": 1.3845596230518304,
      "grad_norm": 2.1091480911916447,
      "learning_rate": 4.5522949656340675e-06,
      "loss": 0.8115,
      "step": 955
    },
    {
      "epoch": 1.3860094237042406,
      "grad_norm": 2.1086830373073373,
      "learning_rate": 4.532590786165509e-06,
      "loss": 0.8956,
      "step": 956
    },
    {
      "epoch": 1.387459224356651,
      "grad_norm": 1.9642868133537161,
      "learning_rate": 4.512916838789559e-06,
      "loss": 0.7701,
      "step": 957
    },
    {
      "epoch": 1.3889090250090612,
      "grad_norm": 2.2964120293271337,
      "learning_rate": 4.4932732322935055e-06,
      "loss": 0.8559,
      "step": 958
    },
    {
      "epoch": 1.3903588256614716,
      "grad_norm": 2.285334805593311,
      "learning_rate": 4.473660075296861e-06,
      "loss": 0.8625,
      "step": 959
    },
    {
      "epoch": 1.3918086263138818,
      "grad_norm": 2.152315034077545,
      "learning_rate": 4.45407747625077e-06,
      "loss": 0.8864,
      "step": 960
    },
    {
      "epoch": 1.3932584269662922,
      "grad_norm": 1.943522905749773,
      "learning_rate": 4.434525543437407e-06,
      "loss": 0.8975,
      "step": 961
    },
    {
      "epoch": 1.3947082276187024,
      "grad_norm": 2.0784395299224085,
      "learning_rate": 4.41500438496937e-06,
      "loss": 0.8821,
      "step": 962
    },
    {
      "epoch": 1.3961580282711128,
      "grad_norm": 2.065064350504038,
      "learning_rate": 4.395514108789097e-06,
      "loss": 0.8426,
      "step": 963
    },
    {
      "epoch": 1.397607828923523,
      "grad_norm": 2.064862722582907,
      "learning_rate": 4.376054822668262e-06,
      "loss": 0.8106,
      "step": 964
    },
    {
      "epoch": 1.3990576295759334,
      "grad_norm": 2.117541354923039,
      "learning_rate": 4.356626634207177e-06,
      "loss": 0.8975,
      "step": 965
    },
    {
      "epoch": 1.4005074302283436,
      "grad_norm": 2.6971971203611114,
      "learning_rate": 4.3372296508342025e-06,
      "loss": 0.8633,
      "step": 966
    },
    {
      "epoch": 1.4005074302283436,
      "eval_loss": 1.9968363046646118,
      "eval_runtime": 721.0507,
      "eval_samples_per_second": 24.484,
      "eval_steps_per_second": 0.383,
      "step": 966
    },
    {
      "epoch": 1.401957230880754,
      "grad_norm": 2.3600953745236644,
      "learning_rate": 4.317863979805142e-06,
      "loss": 0.915,
      "step": 967
    },
    {
      "epoch": 1.4034070315331642,
      "grad_norm": 2.1490321926846057,
      "learning_rate": 4.298529728202661e-06,
      "loss": 0.828,
      "step": 968
    },
    {
      "epoch": 1.4048568321855746,
      "grad_norm": 2.4793536317967892,
      "learning_rate": 4.279227002935702e-06,
      "loss": 0.8776,
      "step": 969
    },
    {
      "epoch": 1.4063066328379847,
      "grad_norm": 2.0989922948450412,
      "learning_rate": 4.2599559107388645e-06,
      "loss": 0.9722,
      "step": 970
    },
    {
      "epoch": 1.4077564334903951,
      "grad_norm": 2.215063785189321,
      "learning_rate": 4.240716558171844e-06,
      "loss": 0.8146,
      "step": 971
    },
    {
      "epoch": 1.4092062341428053,
      "grad_norm": 2.201459796316733,
      "learning_rate": 4.221509051618827e-06,
      "loss": 0.9319,
      "step": 972
    },
    {
      "epoch": 1.4106560347952157,
      "grad_norm": 2.0959527843372068,
      "learning_rate": 4.20233349728791e-06,
      "loss": 0.8906,
      "step": 973
    },
    {
      "epoch": 1.412105835447626,
      "grad_norm": 2.1955643112434333,
      "learning_rate": 4.183190001210511e-06,
      "loss": 0.8081,
      "step": 974
    },
    {
      "epoch": 1.4135556361000363,
      "grad_norm": 2.2840675048390966,
      "learning_rate": 4.164078669240771e-06,
      "loss": 0.8209,
      "step": 975
    },
    {
      "epoch": 1.4150054367524465,
      "grad_norm": 2.2764351133663223,
      "learning_rate": 4.1449996070549895e-06,
      "loss": 0.8298,
      "step": 976
    },
    {
      "epoch": 1.416455237404857,
      "grad_norm": 2.295722544540531,
      "learning_rate": 4.125952920151027e-06,
      "loss": 0.8902,
      "step": 977
    },
    {
      "epoch": 1.417905038057267,
      "grad_norm": 2.462216921082119,
      "learning_rate": 4.10693871384773e-06,
      "loss": 0.7678,
      "step": 978
    },
    {
      "epoch": 1.4193548387096775,
      "grad_norm": 2.6255107607441825,
      "learning_rate": 4.08795709328433e-06,
      "loss": 0.8223,
      "step": 979
    },
    {
      "epoch": 1.4208046393620877,
      "grad_norm": 2.0680600546265224,
      "learning_rate": 4.0690081634198865e-06,
      "loss": 0.7676,
      "step": 980
    },
    {
      "epoch": 1.4208046393620877,
      "eval_loss": 1.9924441576004028,
      "eval_runtime": 719.1663,
      "eval_samples_per_second": 24.548,
      "eval_steps_per_second": 0.384,
      "step": 980
    },
    {
      "epoch": 1.422254440014498,
      "grad_norm": 2.223763699412959,
      "learning_rate": 4.050092029032691e-06,
      "loss": 0.8782,
      "step": 981
    },
    {
      "epoch": 1.4237042406669083,
      "grad_norm": 2.3703865724270807,
      "learning_rate": 4.031208794719695e-06,
      "loss": 0.8184,
      "step": 982
    },
    {
      "epoch": 1.4251540413193187,
      "grad_norm": 2.0820993504427325,
      "learning_rate": 4.012358564895926e-06,
      "loss": 0.7788,
      "step": 983
    },
    {
      "epoch": 1.4266038419717288,
      "grad_norm": 2.182154484215453,
      "learning_rate": 3.993541443793914e-06,
      "loss": 0.8639,
      "step": 984
    },
    {
      "epoch": 1.4280536426241393,
      "grad_norm": 2.409523726231329,
      "learning_rate": 3.974757535463114e-06,
      "loss": 0.8766,
      "step": 985
    },
    {
      "epoch": 1.4295034432765494,
      "grad_norm": 2.132848989874844,
      "learning_rate": 3.956006943769331e-06,
      "loss": 0.8087,
      "step": 986
    },
    {
      "epoch": 1.4309532439289598,
      "grad_norm": 2.139254148372024,
      "learning_rate": 3.937289772394146e-06,
      "loss": 0.8515,
      "step": 987
    },
    {
      "epoch": 1.43240304458137,
      "grad_norm": 2.2949999928130254,
      "learning_rate": 3.918606124834337e-06,
      "loss": 0.8123,
      "step": 988
    },
    {
      "epoch": 1.4338528452337804,
      "grad_norm": 2.439861946375941,
      "learning_rate": 3.899956104401316e-06,
      "loss": 0.8552,
      "step": 989
    },
    {
      "epoch": 1.4353026458861906,
      "grad_norm": 2.2007693692251316,
      "learning_rate": 3.881339814220555e-06,
      "loss": 0.8092,
      "step": 990
    },
    {
      "epoch": 1.436752446538601,
      "grad_norm": 1.983753925370567,
      "learning_rate": 3.862757357231015e-06,
      "loss": 0.8253,
      "step": 991
    },
    {
      "epoch": 1.4382022471910112,
      "grad_norm": 2.149042948716829,
      "learning_rate": 3.844208836184567e-06,
      "loss": 0.8423,
      "step": 992
    },
    {
      "epoch": 1.4396520478434216,
      "grad_norm": 2.1932217181992013,
      "learning_rate": 3.825694353645441e-06,
      "loss": 0.8798,
      "step": 993
    },
    {
      "epoch": 1.4411018484958318,
      "grad_norm": 2.245082514128648,
      "learning_rate": 3.8072140119896507e-06,
      "loss": 0.8361,
      "step": 994
    },
    {
      "epoch": 1.4411018484958318,
      "eval_loss": 1.9976791143417358,
      "eval_runtime": 714.6424,
      "eval_samples_per_second": 24.703,
      "eval_steps_per_second": 0.386,
      "step": 994
    },
    {
      "epoch": 1.4425516491482422,
      "grad_norm": 2.3394029705470505,
      "learning_rate": 3.788767913404425e-06,
      "loss": 0.7327,
      "step": 995
    },
    {
      "epoch": 1.4440014498006524,
      "grad_norm": 2.2767835337912596,
      "learning_rate": 3.770356159887646e-06,
      "loss": 0.8404,
      "step": 996
    },
    {
      "epoch": 1.4454512504530628,
      "grad_norm": 2.1537496205698132,
      "learning_rate": 3.7519788532472846e-06,
      "loss": 0.8438,
      "step": 997
    },
    {
      "epoch": 1.446901051105473,
      "grad_norm": 2.4299407163870685,
      "learning_rate": 3.733636095100838e-06,
      "loss": 0.8391,
      "step": 998
    },
    {
      "epoch": 1.4483508517578834,
      "grad_norm": 2.2836192981311445,
      "learning_rate": 3.7153279868747673e-06,
      "loss": 0.8472,
      "step": 999
    },
    {
      "epoch": 1.4498006524102935,
      "grad_norm": 2.1369123256096487,
      "learning_rate": 3.697054629803931e-06,
      "loss": 0.9587,
      "step": 1000
    },
    {
      "epoch": 1.451250453062704,
      "grad_norm": 2.234685844136932,
      "learning_rate": 3.6788161249310394e-06,
      "loss": 0.9122,
      "step": 1001
    },
    {
      "epoch": 1.4527002537151141,
      "grad_norm": 2.1596483606111447,
      "learning_rate": 3.6606125731060815e-06,
      "loss": 0.8065,
      "step": 1002
    },
    {
      "epoch": 1.4541500543675245,
      "grad_norm": 2.163266761692936,
      "learning_rate": 3.642444074985775e-06,
      "loss": 0.8123,
      "step": 1003
    },
    {
      "epoch": 1.4555998550199347,
      "grad_norm": 2.117622454297108,
      "learning_rate": 3.6243107310330117e-06,
      "loss": 0.8111,
      "step": 1004
    },
    {
      "epoch": 1.4570496556723451,
      "grad_norm": 2.282859821672258,
      "learning_rate": 3.606212641516287e-06,
      "loss": 0.8358,
      "step": 1005
    },
    {
      "epoch": 1.4584994563247553,
      "grad_norm": 2.179320610564975,
      "learning_rate": 3.5881499065091663e-06,
      "loss": 0.8216,
      "step": 1006
    },
    {
      "epoch": 1.4599492569771657,
      "grad_norm": 2.3127564197185664,
      "learning_rate": 3.5701226258897193e-06,
      "loss": 0.8239,
      "step": 1007
    },
    {
      "epoch": 1.461399057629576,
      "grad_norm": 2.144077031318143,
      "learning_rate": 3.552130899339975e-06,
      "loss": 0.8503,
      "step": 1008
    },
    {
      "epoch": 1.461399057629576,
      "eval_loss": 1.99373459815979,
      "eval_runtime": 716.9165,
      "eval_samples_per_second": 24.625,
      "eval_steps_per_second": 0.385,
      "step": 1008
    },
    {
      "epoch": 1.4628488582819863,
      "grad_norm": 2.3358768034607182,
      "learning_rate": 3.534174826345352e-06,
      "loss": 0.8549,
      "step": 1009
    },
    {
      "epoch": 1.4642986589343965,
      "grad_norm": 1.8967676710916461,
      "learning_rate": 3.5162545061941332e-06,
      "loss": 0.9225,
      "step": 1010
    },
    {
      "epoch": 1.4657484595868069,
      "grad_norm": 2.2202492800317923,
      "learning_rate": 3.498370037976908e-06,
      "loss": 0.8371,
      "step": 1011
    },
    {
      "epoch": 1.467198260239217,
      "grad_norm": 2.496008440334526,
      "learning_rate": 3.480521520586021e-06,
      "loss": 0.813,
      "step": 1012
    },
    {
      "epoch": 1.4686480608916275,
      "grad_norm": 2.311367289304339,
      "learning_rate": 3.462709052715014e-06,
      "loss": 0.8416,
      "step": 1013
    },
    {
      "epoch": 1.4700978615440377,
      "grad_norm": 2.516509968639446,
      "learning_rate": 3.4449327328581075e-06,
      "loss": 0.782,
      "step": 1014
    },
    {
      "epoch": 1.471547662196448,
      "grad_norm": 2.53088823966191,
      "learning_rate": 3.427192659309634e-06,
      "loss": 0.8192,
      "step": 1015
    },
    {
      "epoch": 1.4729974628488582,
      "grad_norm": 2.170088558151364,
      "learning_rate": 3.4094889301635036e-06,
      "loss": 0.8362,
      "step": 1016
    },
    {
      "epoch": 1.4744472635012686,
      "grad_norm": 2.4357224724765203,
      "learning_rate": 3.3918216433126637e-06,
      "loss": 0.8648,
      "step": 1017
    },
    {
      "epoch": 1.4758970641536788,
      "grad_norm": 2.1497166671485797,
      "learning_rate": 3.3741908964485415e-06,
      "loss": 0.9065,
      "step": 1018
    },
    {
      "epoch": 1.4773468648060892,
      "grad_norm": 2.2281462402784475,
      "learning_rate": 3.3565967870605278e-06,
      "loss": 0.8617,
      "step": 1019
    },
    {
      "epoch": 1.4787966654584994,
      "grad_norm": 2.291503746187098,
      "learning_rate": 3.339039412435422e-06,
      "loss": 0.8124,
      "step": 1020
    },
    {
      "epoch": 1.4802464661109098,
      "grad_norm": 2.5192333827624083,
      "learning_rate": 3.3215188696569025e-06,
      "loss": 0.8216,
      "step": 1021
    },
    {
      "epoch": 1.48169626676332,
      "grad_norm": 2.1265193940395575,
      "learning_rate": 3.304035255604974e-06,
      "loss": 0.806,
      "step": 1022
    },
    {
      "epoch": 1.48169626676332,
      "eval_loss": 1.9893933534622192,
      "eval_runtime": 715.6266,
      "eval_samples_per_second": 24.669,
      "eval_steps_per_second": 0.386,
      "step": 1022
    },
    {
      "epoch": 1.4831460674157304,
      "grad_norm": 2.2068207293271387,
      "learning_rate": 3.2865886669554546e-06,
      "loss": 0.8204,
      "step": 1023
    },
    {
      "epoch": 1.4845958680681406,
      "grad_norm": 2.142591486492342,
      "learning_rate": 3.2691792001794244e-06,
      "loss": 0.8194,
      "step": 1024
    },
    {
      "epoch": 1.486045668720551,
      "grad_norm": 2.014580968676955,
      "learning_rate": 3.2518069515427074e-06,
      "loss": 0.8018,
      "step": 1025
    },
    {
      "epoch": 1.4874954693729612,
      "grad_norm": 2.0896253989210005,
      "learning_rate": 3.2344720171053134e-06,
      "loss": 0.8271,
      "step": 1026
    },
    {
      "epoch": 1.4889452700253716,
      "grad_norm": 2.222439458671072,
      "learning_rate": 3.2171744927209357e-06,
      "loss": 0.8138,
      "step": 1027
    },
    {
      "epoch": 1.4903950706777818,
      "grad_norm": 2.322279854851702,
      "learning_rate": 3.199914474036403e-06,
      "loss": 0.8636,
      "step": 1028
    },
    {
      "epoch": 1.4918448713301922,
      "grad_norm": 2.0813223076707743,
      "learning_rate": 3.1826920564911614e-06,
      "loss": 0.8125,
      "step": 1029
    },
    {
      "epoch": 1.4932946719826024,
      "grad_norm": 2.291119596057406,
      "learning_rate": 3.1655073353167285e-06,
      "loss": 0.8321,
      "step": 1030
    },
    {
      "epoch": 1.4947444726350128,
      "grad_norm": 2.0761682423616734,
      "learning_rate": 3.1483604055361927e-06,
      "loss": 0.854,
      "step": 1031
    },
    {
      "epoch": 1.496194273287423,
      "grad_norm": 2.1868030646045593,
      "learning_rate": 3.1312513619636686e-06,
      "loss": 0.8275,
      "step": 1032
    },
    {
      "epoch": 1.4976440739398333,
      "grad_norm": 2.17950463615111,
      "learning_rate": 3.1141802992037804e-06,
      "loss": 0.9135,
      "step": 1033
    },
    {
      "epoch": 1.4990938745922435,
      "grad_norm": 2.178276445849971,
      "learning_rate": 3.09714731165114e-06,
      "loss": 0.9377,
      "step": 1034
    },
    {
      "epoch": 1.500543675244654,
      "grad_norm": 2.1569365147656603,
      "learning_rate": 3.0801524934898096e-06,
      "loss": 0.8312,
      "step": 1035
    },
    {
      "epoch": 1.5019934758970641,
      "grad_norm": 2.2947788820149926,
      "learning_rate": 3.0631959386928078e-06,
      "loss": 0.8814,
      "step": 1036
    },
    {
      "epoch": 1.5019934758970641,
      "eval_loss": 1.9883453845977783,
      "eval_runtime": 713.3696,
      "eval_samples_per_second": 24.747,
      "eval_steps_per_second": 0.387,
      "step": 1036
    },
    {
      "epoch": 1.5034432765494743,
      "grad_norm": 2.1892213820795217,
      "learning_rate": 3.0462777410215684e-06,
      "loss": 0.8716,
      "step": 1037
    },
    {
      "epoch": 1.5048930772018847,
      "grad_norm": 2.355781779017383,
      "learning_rate": 3.0293979940254316e-06,
      "loss": 0.8859,
      "step": 1038
    },
    {
      "epoch": 1.506342877854295,
      "grad_norm": 2.0357651901923686,
      "learning_rate": 3.012556791041121e-06,
      "loss": 0.8752,
      "step": 1039
    },
    {
      "epoch": 1.5077926785067053,
      "grad_norm": 2.4353010119049645,
      "learning_rate": 2.9957542251922354e-06,
      "loss": 0.8921,
      "step": 1040
    },
    {
      "epoch": 1.5092424791591155,
      "grad_norm": 2.4318802331223974,
      "learning_rate": 2.978990389388725e-06,
      "loss": 0.8723,
      "step": 1041
    },
    {
      "epoch": 1.5106922798115259,
      "grad_norm": 2.114409840473563,
      "learning_rate": 2.9622653763263874e-06,
      "loss": 0.8231,
      "step": 1042
    },
    {
      "epoch": 1.5121420804639363,
      "grad_norm": 2.261042839318657,
      "learning_rate": 2.9455792784863403e-06,
      "loss": 0.8527,
      "step": 1043
    },
    {
      "epoch": 1.5135918811163465,
      "grad_norm": 2.20506815324262,
      "learning_rate": 2.9289321881345257e-06,
      "loss": 0.8638,
      "step": 1044
    },
    {
      "epoch": 1.5150416817687566,
      "grad_norm": 2.2592928858692383,
      "learning_rate": 2.9123241973211937e-06,
      "loss": 0.8511,
      "step": 1045
    },
    {
      "epoch": 1.516491482421167,
      "grad_norm": 2.080676700055508,
      "learning_rate": 2.8957553978803897e-06,
      "loss": 0.831,
      "step": 1046
    },
    {
      "epoch": 1.5179412830735775,
      "grad_norm": 2.2368858450455353,
      "learning_rate": 2.8792258814294572e-06,
      "loss": 0.8939,
      "step": 1047
    },
    {
      "epoch": 1.5193910837259876,
      "grad_norm": 2.114949588969858,
      "learning_rate": 2.8627357393685117e-06,
      "loss": 0.8522,
      "step": 1048
    },
    {
      "epoch": 1.5208408843783978,
      "grad_norm": 2.328148106508165,
      "learning_rate": 2.8462850628799567e-06,
      "loss": 0.8842,
      "step": 1049
    },
    {
      "epoch": 1.5222906850308082,
      "grad_norm": 2.2557518326544774,
      "learning_rate": 2.829873942927971e-06,
      "loss": 0.885,
      "step": 1050
    },
    {
      "epoch": 1.5222906850308082,
      "eval_loss": 1.98646879196167,
      "eval_runtime": 715.8437,
      "eval_samples_per_second": 24.662,
      "eval_steps_per_second": 0.386,
      "step": 1050
    },
    {
      "epoch": 1.5237404856832186,
      "grad_norm": 2.2113811266627215,
      "learning_rate": 2.8135024702579996e-06,
      "loss": 0.754,
      "step": 1051
    },
    {
      "epoch": 1.5251902863356288,
      "grad_norm": 2.091212348836759,
      "learning_rate": 2.797170735396263e-06,
      "loss": 0.8895,
      "step": 1052
    },
    {
      "epoch": 1.526640086988039,
      "grad_norm": 2.1246345425374775,
      "learning_rate": 2.780878828649246e-06,
      "loss": 0.879,
      "step": 1053
    },
    {
      "epoch": 1.5280898876404494,
      "grad_norm": 2.1052231351447475,
      "learning_rate": 2.7646268401032086e-06,
      "loss": 0.788,
      "step": 1054
    },
    {
      "epoch": 1.5295396882928598,
      "grad_norm": 2.1844316775978756,
      "learning_rate": 2.7484148596236806e-06,
      "loss": 0.8782,
      "step": 1055
    },
    {
      "epoch": 1.53098948894527,
      "grad_norm": 2.090968007146545,
      "learning_rate": 2.732242976854962e-06,
      "loss": 0.8227,
      "step": 1056
    },
    {
      "epoch": 1.5324392895976802,
      "grad_norm": 2.055130187970517,
      "learning_rate": 2.716111281219639e-06,
      "loss": 0.9006,
      "step": 1057
    },
    {
      "epoch": 1.5338890902500906,
      "grad_norm": 2.2063539689679104,
      "learning_rate": 2.7000198619180795e-06,
      "loss": 0.8982,
      "step": 1058
    },
    {
      "epoch": 1.535338890902501,
      "grad_norm": 2.2922951881692506,
      "learning_rate": 2.6839688079279446e-06,
      "loss": 0.7586,
      "step": 1059
    },
    {
      "epoch": 1.5367886915549112,
      "grad_norm": 2.063209695931878,
      "learning_rate": 2.6679582080036957e-06,
      "loss": 0.8601,
      "step": 1060
    },
    {
      "epoch": 1.5382384922073213,
      "grad_norm": 2.1240009703648233,
      "learning_rate": 2.651988150676097e-06,
      "loss": 0.8352,
      "step": 1061
    },
    {
      "epoch": 1.5396882928597317,
      "grad_norm": 2.3114656969924448,
      "learning_rate": 2.636058724251739e-06,
      "loss": 0.8449,
      "step": 1062
    },
    {
      "epoch": 1.5411380935121421,
      "grad_norm": 2.229424829585977,
      "learning_rate": 2.620170016812539e-06,
      "loss": 0.8448,
      "step": 1063
    },
    {
      "epoch": 1.5425878941645523,
      "grad_norm": 2.3022373732918244,
      "learning_rate": 2.6043221162152632e-06,
      "loss": 0.8049,
      "step": 1064
    },
    {
      "epoch": 1.5425878941645523,
      "eval_loss": 1.9863100051879883,
      "eval_runtime": 717.7455,
      "eval_samples_per_second": 24.596,
      "eval_steps_per_second": 0.385,
      "step": 1064
    },
    {
      "epoch": 1.5440376948169625,
      "grad_norm": 2.254121269817127,
      "learning_rate": 2.5885151100910254e-06,
      "loss": 0.8037,
      "step": 1065
    },
    {
      "epoch": 1.545487495469373,
      "grad_norm": 2.0785608866143956,
      "learning_rate": 2.572749085844829e-06,
      "loss": 0.8643,
      "step": 1066
    },
    {
      "epoch": 1.5469372961217833,
      "grad_norm": 2.2785476639178746,
      "learning_rate": 2.5570241306550537e-06,
      "loss": 0.8522,
      "step": 1067
    },
    {
      "epoch": 1.5483870967741935,
      "grad_norm": 2.17251752212887,
      "learning_rate": 2.541340331472998e-06,
      "loss": 0.7991,
      "step": 1068
    },
    {
      "epoch": 1.5498368974266037,
      "grad_norm": 2.0323230913631884,
      "learning_rate": 2.5256977750223755e-06,
      "loss": 0.805,
      "step": 1069
    },
    {
      "epoch": 1.551286698079014,
      "grad_norm": 2.1164765648639605,
      "learning_rate": 2.510096547798857e-06,
      "loss": 0.8596,
      "step": 1070
    },
    {
      "epoch": 1.5527364987314245,
      "grad_norm": 2.0210833039152845,
      "learning_rate": 2.4945367360695828e-06,
      "loss": 0.8393,
      "step": 1071
    },
    {
      "epoch": 1.5541862993838347,
      "grad_norm": 2.0843362554059985,
      "learning_rate": 2.479018425872686e-06,
      "loss": 0.7969,
      "step": 1072
    },
    {
      "epoch": 1.5556361000362449,
      "grad_norm": 2.066878716298172,
      "learning_rate": 2.4635417030168064e-06,
      "loss": 0.856,
      "step": 1073
    },
    {
      "epoch": 1.5570859006886553,
      "grad_norm": 2.109778407626963,
      "learning_rate": 2.44810665308064e-06,
      "loss": 0.8615,
      "step": 1074
    },
    {
      "epoch": 1.5585357013410657,
      "grad_norm": 2.2444990200544974,
      "learning_rate": 2.432713361412444e-06,
      "loss": 0.7951,
      "step": 1075
    },
    {
      "epoch": 1.5599855019934759,
      "grad_norm": 1.9574347360926185,
      "learning_rate": 2.4173619131295754e-06,
      "loss": 0.8135,
      "step": 1076
    },
    {
      "epoch": 1.561435302645886,
      "grad_norm": 2.533904384650765,
      "learning_rate": 2.4020523931180207e-06,
      "loss": 0.8804,
      "step": 1077
    },
    {
      "epoch": 1.5628851032982964,
      "grad_norm": 2.3923025601716796,
      "learning_rate": 2.3867848860319143e-06,
      "loss": 0.8597,
      "step": 1078
    },
    {
      "epoch": 1.5628851032982964,
      "eval_loss": 1.9971156120300293,
      "eval_runtime": 715.3799,
      "eval_samples_per_second": 24.678,
      "eval_steps_per_second": 0.386,
      "step": 1078
    },
    {
      "epoch": 1.5643349039507068,
      "grad_norm": 2.097183514715348,
      "learning_rate": 2.371559476293086e-06,
      "loss": 0.8658,
      "step": 1079
    },
    {
      "epoch": 1.565784704603117,
      "grad_norm": 2.6489696043366777,
      "learning_rate": 2.3563762480905917e-06,
      "loss": 0.8858,
      "step": 1080
    },
    {
      "epoch": 1.5672345052555272,
      "grad_norm": 2.2629303537246512,
      "learning_rate": 2.341235285380242e-06,
      "loss": 0.8642,
      "step": 1081
    },
    {
      "epoch": 1.5686843059079376,
      "grad_norm": 2.437820827972319,
      "learning_rate": 2.3261366718841307e-06,
      "loss": 0.8254,
      "step": 1082
    },
    {
      "epoch": 1.570134106560348,
      "grad_norm": 2.1991078001026474,
      "learning_rate": 2.3110804910901928e-06,
      "loss": 0.8188,
      "step": 1083
    },
    {
      "epoch": 1.5715839072127582,
      "grad_norm": 2.183867050691279,
      "learning_rate": 2.296066826251725e-06,
      "loss": 0.8058,
      "step": 1084
    },
    {
      "epoch": 1.5730337078651684,
      "grad_norm": 1.980952710035457,
      "learning_rate": 2.281095760386939e-06,
      "loss": 0.8266,
      "step": 1085
    },
    {
      "epoch": 1.5744835085175788,
      "grad_norm": 2.24718029107916,
      "learning_rate": 2.266167376278481e-06,
      "loss": 0.8783,
      "step": 1086
    },
    {
      "epoch": 1.5759333091699892,
      "grad_norm": 2.2801638673061317,
      "learning_rate": 2.2512817564730017e-06,
      "loss": 0.861,
      "step": 1087
    },
    {
      "epoch": 1.5773831098223994,
      "grad_norm": 2.2398443276008226,
      "learning_rate": 2.23643898328068e-06,
      "loss": 0.8821,
      "step": 1088
    },
    {
      "epoch": 1.5788329104748096,
      "grad_norm": 2.1738646551629093,
      "learning_rate": 2.221639138774776e-06,
      "loss": 0.8281,
      "step": 1089
    },
    {
      "epoch": 1.58028271112722,
      "grad_norm": 2.1535350265994944,
      "learning_rate": 2.206882304791176e-06,
      "loss": 0.8508,
      "step": 1090
    },
    {
      "epoch": 1.5817325117796304,
      "grad_norm": 2.1633040053395325,
      "learning_rate": 2.192168562927933e-06,
      "loss": 0.8608,
      "step": 1091
    },
    {
      "epoch": 1.5831823124320406,
      "grad_norm": 2.176965267637412,
      "learning_rate": 2.177497994544828e-06,
      "loss": 0.8739,
      "step": 1092
    },
    {
      "epoch": 1.5831823124320406,
      "eval_loss": 1.995527982711792,
      "eval_runtime": 712.647,
      "eval_samples_per_second": 24.772,
      "eval_steps_per_second": 0.387,
      "step": 1092
    },
    {
      "epoch": 1.5846321130844507,
      "grad_norm": 2.1501357831318595,
      "learning_rate": 2.162870680762914e-06,
      "loss": 0.8208,
      "step": 1093
    },
    {
      "epoch": 1.5860819137368611,
      "grad_norm": 2.131079594500507,
      "learning_rate": 2.148286702464063e-06,
      "loss": 0.9174,
      "step": 1094
    },
    {
      "epoch": 1.5875317143892715,
      "grad_norm": 2.1340932615372377,
      "learning_rate": 2.133746140290527e-06,
      "loss": 0.9001,
      "step": 1095
    },
    {
      "epoch": 1.5889815150416817,
      "grad_norm": 2.1448367392031,
      "learning_rate": 2.119249074644485e-06,
      "loss": 0.8513,
      "step": 1096
    },
    {
      "epoch": 1.590431315694092,
      "grad_norm": 2.1547396128663086,
      "learning_rate": 2.1047955856876045e-06,
      "loss": 0.7978,
      "step": 1097
    },
    {
      "epoch": 1.5918811163465023,
      "grad_norm": 2.408902969917587,
      "learning_rate": 2.090385753340596e-06,
      "loss": 0.7652,
      "step": 1098
    },
    {
      "epoch": 1.5933309169989127,
      "grad_norm": 2.233883621337741,
      "learning_rate": 2.0760196572827618e-06,
      "loss": 0.8524,
      "step": 1099
    },
    {
      "epoch": 1.594780717651323,
      "grad_norm": 2.3352614173912274,
      "learning_rate": 2.0616973769515737e-06,
      "loss": 0.865,
      "step": 1100
    },
    {
      "epoch": 1.596230518303733,
      "grad_norm": 2.1609967957412075,
      "learning_rate": 2.0474189915422206e-06,
      "loss": 0.8399,
      "step": 1101
    },
    {
      "epoch": 1.5976803189561435,
      "grad_norm": 2.2091852650518202,
      "learning_rate": 2.033184580007177e-06,
      "loss": 0.8457,
      "step": 1102
    },
    {
      "epoch": 1.599130119608554,
      "grad_norm": 2.6324187678612456,
      "learning_rate": 2.0189942210557536e-06,
      "loss": 0.8728,
      "step": 1103
    },
    {
      "epoch": 1.600579920260964,
      "grad_norm": 2.185950688422059,
      "learning_rate": 2.0048479931536833e-06,
      "loss": 0.8236,
      "step": 1104
    },
    {
      "epoch": 1.6020297209133743,
      "grad_norm": 2.339979783614652,
      "learning_rate": 1.990745974522671e-06,
      "loss": 0.7762,
      "step": 1105
    },
    {
      "epoch": 1.6034795215657847,
      "grad_norm": 2.246416887978597,
      "learning_rate": 1.976688243139965e-06,
      "loss": 0.8193,
      "step": 1106
    },
    {
      "epoch": 1.6034795215657847,
      "eval_loss": 1.9948707818984985,
      "eval_runtime": 709.959,
      "eval_samples_per_second": 24.866,
      "eval_steps_per_second": 0.389,
      "step": 1106
    },
    {
      "epoch": 1.604929322218195,
      "grad_norm": 2.505760380289249,
      "learning_rate": 1.962674876737929e-06,
      "loss": 0.8216,
      "step": 1107
    },
    {
      "epoch": 1.6063791228706052,
      "grad_norm": 2.217237490048216,
      "learning_rate": 1.94870595280361e-06,
      "loss": 0.8879,
      "step": 1108
    },
    {
      "epoch": 1.6078289235230154,
      "grad_norm": 2.1143288373491935,
      "learning_rate": 1.9347815485783084e-06,
      "loss": 0.8941,
      "step": 1109
    },
    {
      "epoch": 1.6092787241754258,
      "grad_norm": 2.236870251337571,
      "learning_rate": 1.920901741057153e-06,
      "loss": 0.8319,
      "step": 1110
    },
    {
      "epoch": 1.6107285248278362,
      "grad_norm": 2.4129565940130617,
      "learning_rate": 1.907066606988679e-06,
      "loss": 0.838,
      "step": 1111
    },
    {
      "epoch": 1.6121783254802464,
      "grad_norm": 2.078552752338979,
      "learning_rate": 1.8932762228743884e-06,
      "loss": 0.8727,
      "step": 1112
    },
    {
      "epoch": 1.6136281261326566,
      "grad_norm": 2.042650870057603,
      "learning_rate": 1.8795306649683476e-06,
      "loss": 0.7706,
      "step": 1113
    },
    {
      "epoch": 1.615077926785067,
      "grad_norm": 2.416486231069537,
      "learning_rate": 1.8658300092767546e-06,
      "loss": 0.9601,
      "step": 1114
    },
    {
      "epoch": 1.6165277274374774,
      "grad_norm": 2.1906206704476188,
      "learning_rate": 1.8521743315575202e-06,
      "loss": 0.8151,
      "step": 1115
    },
    {
      "epoch": 1.6179775280898876,
      "grad_norm": 2.1781658622555033,
      "learning_rate": 1.8385637073198426e-06,
      "loss": 0.862,
      "step": 1116
    },
    {
      "epoch": 1.6194273287422978,
      "grad_norm": 2.3116677082332866,
      "learning_rate": 1.8249982118238042e-06,
      "loss": 0.7598,
      "step": 1117
    },
    {
      "epoch": 1.6208771293947082,
      "grad_norm": 2.125001956859173,
      "learning_rate": 1.8114779200799437e-06,
      "loss": 0.7785,
      "step": 1118
    },
    {
      "epoch": 1.6223269300471186,
      "grad_norm": 2.0668561397479173,
      "learning_rate": 1.798002906848848e-06,
      "loss": 0.8255,
      "step": 1119
    },
    {
      "epoch": 1.6237767306995288,
      "grad_norm": 2.244824181175537,
      "learning_rate": 1.7845732466407361e-06,
      "loss": 0.8762,
      "step": 1120
    },
    {
      "epoch": 1.6237767306995288,
      "eval_loss": 1.9928207397460938,
      "eval_runtime": 707.8587,
      "eval_samples_per_second": 24.94,
      "eval_steps_per_second": 0.39,
      "step": 1120
    },
    {
      "epoch": 1.625226531351939,
      "grad_norm": 2.1420128329571897,
      "learning_rate": 1.7711890137150368e-06,
      "loss": 0.8171,
      "step": 1121
    },
    {
      "epoch": 1.6266763320043494,
      "grad_norm": 2.5389880015362487,
      "learning_rate": 1.7578502820800048e-06,
      "loss": 0.8168,
      "step": 1122
    },
    {
      "epoch": 1.6281261326567598,
      "grad_norm": 2.2721157468307025,
      "learning_rate": 1.7445571254922877e-06,
      "loss": 0.7821,
      "step": 1123
    },
    {
      "epoch": 1.62957593330917,
      "grad_norm": 1.9952804297422466,
      "learning_rate": 1.731309617456518e-06,
      "loss": 0.8288,
      "step": 1124
    },
    {
      "epoch": 1.6310257339615801,
      "grad_norm": 2.197774147746833,
      "learning_rate": 1.7181078312249256e-06,
      "loss": 0.8491,
      "step": 1125
    },
    {
      "epoch": 1.6324755346139905,
      "grad_norm": 2.1777904566114996,
      "learning_rate": 1.7049518397969135e-06,
      "loss": 0.8761,
      "step": 1126
    },
    {
      "epoch": 1.633925335266401,
      "grad_norm": 2.315236780838234,
      "learning_rate": 1.6918417159186662e-06,
      "loss": 0.8695,
      "step": 1127
    },
    {
      "epoch": 1.6353751359188111,
      "grad_norm": 2.260461988641999,
      "learning_rate": 1.6787775320827438e-06,
      "loss": 0.8338,
      "step": 1128
    },
    {
      "epoch": 1.6368249365712213,
      "grad_norm": 1.950055589656603,
      "learning_rate": 1.6657593605276746e-06,
      "loss": 0.8762,
      "step": 1129
    },
    {
      "epoch": 1.6382747372236317,
      "grad_norm": 2.1921907550595336,
      "learning_rate": 1.652787273237565e-06,
      "loss": 0.9171,
      "step": 1130
    },
    {
      "epoch": 1.639724537876042,
      "grad_norm": 2.286322434257131,
      "learning_rate": 1.639861341941701e-06,
      "loss": 0.8793,
      "step": 1131
    },
    {
      "epoch": 1.6411743385284523,
      "grad_norm": 2.0161623291548656,
      "learning_rate": 1.6269816381141446e-06,
      "loss": 0.8504,
      "step": 1132
    },
    {
      "epoch": 1.6426241391808625,
      "grad_norm": 2.258004277314312,
      "learning_rate": 1.614148232973346e-06,
      "loss": 0.9225,
      "step": 1133
    },
    {
      "epoch": 1.6440739398332729,
      "grad_norm": 2.1444850995608418,
      "learning_rate": 1.601361197481741e-06,
      "loss": 0.8503,
      "step": 1134
    },
    {
      "epoch": 1.6440739398332729,
      "eval_loss": 1.9911795854568481,
      "eval_runtime": 718.5811,
      "eval_samples_per_second": 24.568,
      "eval_steps_per_second": 0.384,
      "step": 1134
    },
    {
      "epoch": 1.6455237404856833,
      "grad_norm": 1.9838758934901295,
      "learning_rate": 1.5886206023453642e-06,
      "loss": 0.8678,
      "step": 1135
    },
    {
      "epoch": 1.6469735411380935,
      "grad_norm": 2.0349957414114126,
      "learning_rate": 1.5759265180134709e-06,
      "loss": 0.8807,
      "step": 1136
    },
    {
      "epoch": 1.6484233417905036,
      "grad_norm": 2.4335095147609276,
      "learning_rate": 1.5632790146781162e-06,
      "loss": 0.8479,
      "step": 1137
    },
    {
      "epoch": 1.649873142442914,
      "grad_norm": 2.0235904556337676,
      "learning_rate": 1.5506781622737943e-06,
      "loss": 0.8655,
      "step": 1138
    },
    {
      "epoch": 1.6513229430953245,
      "grad_norm": 2.2903351112763315,
      "learning_rate": 1.538124030477044e-06,
      "loss": 0.8612,
      "step": 1139
    },
    {
      "epoch": 1.6527727437477346,
      "grad_norm": 2.3169079563538366,
      "learning_rate": 1.5256166887060575e-06,
      "loss": 0.8148,
      "step": 1140
    },
    {
      "epoch": 1.6542225444001448,
      "grad_norm": 2.1187961286361596,
      "learning_rate": 1.5131562061203053e-06,
      "loss": 0.7623,
      "step": 1141
    },
    {
      "epoch": 1.6556723450525552,
      "grad_norm": 2.2680015778132145,
      "learning_rate": 1.5007426516201451e-06,
      "loss": 0.8419,
      "step": 1142
    },
    {
      "epoch": 1.6571221457049656,
      "grad_norm": 2.2649121402149452,
      "learning_rate": 1.4883760938464487e-06,
      "loss": 0.8757,
      "step": 1143
    },
    {
      "epoch": 1.6585719463573758,
      "grad_norm": 2.2546937877039257,
      "learning_rate": 1.4760566011802168e-06,
      "loss": 0.8139,
      "step": 1144
    },
    {
      "epoch": 1.660021747009786,
      "grad_norm": 2.3029034702661053,
      "learning_rate": 1.4637842417422088e-06,
      "loss": 0.9081,
      "step": 1145
    },
    {
      "epoch": 1.6614715476621964,
      "grad_norm": 2.197485062919516,
      "learning_rate": 1.4515590833925508e-06,
      "loss": 0.8046,
      "step": 1146
    },
    {
      "epoch": 1.6629213483146068,
      "grad_norm": 2.1416849214845537,
      "learning_rate": 1.439381193730376e-06,
      "loss": 0.8518,
      "step": 1147
    },
    {
      "epoch": 1.664371148967017,
      "grad_norm": 2.2385859029506436,
      "learning_rate": 1.4272506400934461e-06,
      "loss": 0.9069,
      "step": 1148
    },
    {
      "epoch": 1.664371148967017,
      "eval_loss": 1.993903398513794,
      "eval_runtime": 716.1514,
      "eval_samples_per_second": 24.651,
      "eval_steps_per_second": 0.385,
      "step": 1148
    },
    {
      "epoch": 1.6658209496194272,
      "grad_norm": 2.065552623071493,
      "learning_rate": 1.415167489557775e-06,
      "loss": 0.8337,
      "step": 1149
    },
    {
      "epoch": 1.6672707502718376,
      "grad_norm": 2.3410757370054154,
      "learning_rate": 1.4031318089372614e-06,
      "loss": 0.8587,
      "step": 1150
    },
    {
      "epoch": 1.668720550924248,
      "grad_norm": 2.5260601516312677,
      "learning_rate": 1.3911436647833188e-06,
      "loss": 0.8834,
      "step": 1151
    },
    {
      "epoch": 1.6701703515766582,
      "grad_norm": 2.2669547472576244,
      "learning_rate": 1.379203123384507e-06,
      "loss": 0.8106,
      "step": 1152
    },
    {
      "epoch": 1.6716201522290683,
      "grad_norm": 2.014416690584,
      "learning_rate": 1.3673102507661673e-06,
      "loss": 0.8445,
      "step": 1153
    },
    {
      "epoch": 1.6730699528814788,
      "grad_norm": 2.228372925552681,
      "learning_rate": 1.3554651126900564e-06,
      "loss": 0.8121,
      "step": 1154
    },
    {
      "epoch": 1.6745197535338892,
      "grad_norm": 2.323473525141172,
      "learning_rate": 1.343667774653976e-06,
      "loss": 0.8126,
      "step": 1155
    },
    {
      "epoch": 1.6759695541862993,
      "grad_norm": 2.2382148932531605,
      "learning_rate": 1.331918301891425e-06,
      "loss": 0.8935,
      "step": 1156
    },
    {
      "epoch": 1.6774193548387095,
      "grad_norm": 2.5995648466790513,
      "learning_rate": 1.3202167593712279e-06,
      "loss": 0.8513,
      "step": 1157
    },
    {
      "epoch": 1.67886915549112,
      "grad_norm": 2.1997458595140644,
      "learning_rate": 1.3085632117971792e-06,
      "loss": 0.8016,
      "step": 1158
    },
    {
      "epoch": 1.6803189561435303,
      "grad_norm": 2.287684071966117,
      "learning_rate": 1.2969577236076814e-06,
      "loss": 0.8,
      "step": 1159
    },
    {
      "epoch": 1.6817687567959405,
      "grad_norm": 2.130632703860925,
      "learning_rate": 1.285400358975395e-06,
      "loss": 0.8031,
      "step": 1160
    },
    {
      "epoch": 1.6832185574483507,
      "grad_norm": 2.204789226027851,
      "learning_rate": 1.2738911818068811e-06,
      "loss": 0.8871,
      "step": 1161
    },
    {
      "epoch": 1.684668358100761,
      "grad_norm": 2.1437749614489476,
      "learning_rate": 1.2624302557422475e-06,
      "loss": 0.809,
      "step": 1162
    },
    {
      "epoch": 1.684668358100761,
      "eval_loss": 1.9906678199768066,
      "eval_runtime": 712.7549,
      "eval_samples_per_second": 24.769,
      "eval_steps_per_second": 0.387,
      "step": 1162
    },
    {
      "epoch": 1.6861181587531715,
      "grad_norm": 2.0589397935860325,
      "learning_rate": 1.2510176441547972e-06,
      "loss": 0.9041,
      "step": 1163
    },
    {
      "epoch": 1.6875679594055817,
      "grad_norm": 2.2607100679041134,
      "learning_rate": 1.2396534101506774e-06,
      "loss": 0.8123,
      "step": 1164
    },
    {
      "epoch": 1.6890177600579919,
      "grad_norm": 2.1011870880413204,
      "learning_rate": 1.228337616568529e-06,
      "loss": 0.8188,
      "step": 1165
    },
    {
      "epoch": 1.6904675607104023,
      "grad_norm": 2.2135679098733245,
      "learning_rate": 1.2170703259791472e-06,
      "loss": 0.8494,
      "step": 1166
    },
    {
      "epoch": 1.6919173613628127,
      "grad_norm": 2.4687920021298955,
      "learning_rate": 1.2058516006851206e-06,
      "loss": 0.8749,
      "step": 1167
    },
    {
      "epoch": 1.6933671620152229,
      "grad_norm": 2.1252548326750933,
      "learning_rate": 1.1946815027205027e-06,
      "loss": 0.8627,
      "step": 1168
    },
    {
      "epoch": 1.694816962667633,
      "grad_norm": 2.253460350104929,
      "learning_rate": 1.1835600938504577e-06,
      "loss": 0.824,
      "step": 1169
    },
    {
      "epoch": 1.6962667633200434,
      "grad_norm": 2.727681483955523,
      "learning_rate": 1.1724874355709259e-06,
      "loss": 0.764,
      "step": 1170
    },
    {
      "epoch": 1.6977165639724539,
      "grad_norm": 2.0802804535486312,
      "learning_rate": 1.1614635891082815e-06,
      "loss": 0.8997,
      "step": 1171
    },
    {
      "epoch": 1.699166364624864,
      "grad_norm": 2.275788805544603,
      "learning_rate": 1.1504886154189854e-06,
      "loss": 0.831,
      "step": 1172
    },
    {
      "epoch": 1.7006161652772742,
      "grad_norm": 2.2991559680958233,
      "learning_rate": 1.1395625751892659e-06,
      "loss": 0.7699,
      "step": 1173
    },
    {
      "epoch": 1.7020659659296846,
      "grad_norm": 2.180066338215803,
      "learning_rate": 1.1286855288347676e-06,
      "loss": 0.9128,
      "step": 1174
    },
    {
      "epoch": 1.703515766582095,
      "grad_norm": 2.111862812365535,
      "learning_rate": 1.1178575365002264e-06,
      "loss": 0.8688,
      "step": 1175
    },
    {
      "epoch": 1.7049655672345052,
      "grad_norm": 2.092554498056762,
      "learning_rate": 1.1070786580591287e-06,
      "loss": 0.8202,
      "step": 1176
    },
    {
      "epoch": 1.7049655672345052,
      "eval_loss": 1.9890180826187134,
      "eval_runtime": 714.7537,
      "eval_samples_per_second": 24.699,
      "eval_steps_per_second": 0.386,
      "step": 1176
    },
    {
      "epoch": 1.7064153678869154,
      "grad_norm": 1.9189718860006622,
      "learning_rate": 1.0963489531133874e-06,
      "loss": 0.9,
      "step": 1177
    },
    {
      "epoch": 1.7078651685393258,
      "grad_norm": 2.3132568748585376,
      "learning_rate": 1.085668480993015e-06,
      "loss": 0.8292,
      "step": 1178
    },
    {
      "epoch": 1.7093149691917362,
      "grad_norm": 2.0762161267470765,
      "learning_rate": 1.0750373007557858e-06,
      "loss": 0.8074,
      "step": 1179
    },
    {
      "epoch": 1.7107647698441464,
      "grad_norm": 2.1372075666488675,
      "learning_rate": 1.0644554711869115e-06,
      "loss": 0.882,
      "step": 1180
    },
    {
      "epoch": 1.7122145704965566,
      "grad_norm": 2.1827416181277193,
      "learning_rate": 1.053923050798723e-06,
      "loss": 0.8151,
      "step": 1181
    },
    {
      "epoch": 1.713664371148967,
      "grad_norm": 2.431917909355072,
      "learning_rate": 1.0434400978303427e-06,
      "loss": 0.8218,
      "step": 1182
    },
    {
      "epoch": 1.7151141718013774,
      "grad_norm": 2.19911411365956,
      "learning_rate": 1.0330066702473607e-06,
      "loss": 0.7863,
      "step": 1183
    },
    {
      "epoch": 1.7165639724537876,
      "grad_norm": 2.1841728799219293,
      "learning_rate": 1.0226228257415183e-06,
      "loss": 0.8314,
      "step": 1184
    },
    {
      "epoch": 1.7180137731061977,
      "grad_norm": 2.0872214556009365,
      "learning_rate": 1.0122886217303818e-06,
      "loss": 0.8029,
      "step": 1185
    },
    {
      "epoch": 1.7194635737586081,
      "grad_norm": 2.2017495340146804,
      "learning_rate": 1.0020041153570349e-06,
      "loss": 0.8426,
      "step": 1186
    },
    {
      "epoch": 1.7209133744110185,
      "grad_norm": 2.1255636815554735,
      "learning_rate": 9.917693634897552e-07,
      "loss": 0.8789,
      "step": 1187
    },
    {
      "epoch": 1.7223631750634287,
      "grad_norm": 2.221113889349983,
      "learning_rate": 9.815844227217065e-07,
      "loss": 0.8244,
      "step": 1188
    },
    {
      "epoch": 1.723812975715839,
      "grad_norm": 2.2151092303432867,
      "learning_rate": 9.714493493706134e-07,
      "loss": 0.8184,
      "step": 1189
    },
    {
      "epoch": 1.7252627763682493,
      "grad_norm": 2.4651075476067823,
      "learning_rate": 9.61364199478465e-07,
      "loss": 0.7985,
      "step": 1190
    },
    {
      "epoch": 1.7252627763682493,
      "eval_loss": 1.9917757511138916,
      "eval_runtime": 717.6744,
      "eval_samples_per_second": 24.599,
      "eval_steps_per_second": 0.385,
      "step": 1190
    },
    {
      "epoch": 1.7267125770206597,
      "grad_norm": 2.2182885391859934,
      "learning_rate": 9.513290288111943e-07,
      "loss": 0.7886,
      "step": 1191
    },
    {
      "epoch": 1.72816237767307,
      "grad_norm": 2.474545726399445,
      "learning_rate": 9.413438928583807e-07,
      "loss": 0.8342,
      "step": 1192
    },
    {
      "epoch": 1.72961217832548,
      "grad_norm": 2.1568518353098747,
      "learning_rate": 9.314088468329252e-07,
      "loss": 0.7544,
      "step": 1193
    },
    {
      "epoch": 1.7310619789778905,
      "grad_norm": 2.370091685119633,
      "learning_rate": 9.215239456707636e-07,
      "loss": 0.8593,
      "step": 1194
    },
    {
      "epoch": 1.732511779630301,
      "grad_norm": 2.232321988405892,
      "learning_rate": 9.116892440305547e-07,
      "loss": 0.8048,
      "step": 1195
    },
    {
      "epoch": 1.733961580282711,
      "grad_norm": 2.1694650557999764,
      "learning_rate": 9.01904796293378e-07,
      "loss": 0.8373,
      "step": 1196
    },
    {
      "epoch": 1.7354113809351213,
      "grad_norm": 2.087483452002179,
      "learning_rate": 8.921706565624289e-07,
      "loss": 0.8775,
      "step": 1197
    },
    {
      "epoch": 1.7368611815875317,
      "grad_norm": 2.202684319750977,
      "learning_rate": 8.824868786627305e-07,
      "loss": 0.8506,
      "step": 1198
    },
    {
      "epoch": 1.738310982239942,
      "grad_norm": 2.2141324673170204,
      "learning_rate": 8.728535161408258e-07,
      "loss": 0.8771,
      "step": 1199
    },
    {
      "epoch": 1.7397607828923523,
      "grad_norm": 2.3718743014411188,
      "learning_rate": 8.632706222644893e-07,
      "loss": 0.7526,
      "step": 1200
    },
    {
      "epoch": 1.7412105835447624,
      "grad_norm": 2.3219163996305134,
      "learning_rate": 8.537382500224256e-07,
      "loss": 0.8731,
      "step": 1201
    },
    {
      "epoch": 1.7426603841971728,
      "grad_norm": 2.299758685081104,
      "learning_rate": 8.442564521239782e-07,
      "loss": 0.8344,
      "step": 1202
    },
    {
      "epoch": 1.7441101848495832,
      "grad_norm": 2.0103841439395884,
      "learning_rate": 8.348252809988411e-07,
      "loss": 0.88,
      "step": 1203
    },
    {
      "epoch": 1.7455599855019934,
      "grad_norm": 2.172169884792372,
      "learning_rate": 8.254447887967698e-07,
      "loss": 0.8471,
      "step": 1204
    },
    {
      "epoch": 1.7455599855019934,
      "eval_loss": 1.9910231828689575,
      "eval_runtime": 717.4208,
      "eval_samples_per_second": 24.608,
      "eval_steps_per_second": 0.385,
      "step": 1204
    },
    {
      "epoch": 1.7470097861544036,
      "grad_norm": 2.2495622362779857,
      "learning_rate": 8.161150273872854e-07,
      "loss": 0.8648,
      "step": 1205
    },
    {
      "epoch": 1.748459586806814,
      "grad_norm": 2.438775080164551,
      "learning_rate": 8.06836048359394e-07,
      "loss": 0.8609,
      "step": 1206
    },
    {
      "epoch": 1.7499093874592244,
      "grad_norm": 2.1666041739450246,
      "learning_rate": 7.976079030213035e-07,
      "loss": 0.8738,
      "step": 1207
    },
    {
      "epoch": 1.7513591881116346,
      "grad_norm": 1.9460362247427663,
      "learning_rate": 7.884306424001309e-07,
      "loss": 0.811,
      "step": 1208
    },
    {
      "epoch": 1.7528089887640448,
      "grad_norm": 2.201811294271485,
      "learning_rate": 7.793043172416325e-07,
      "loss": 0.9,
      "step": 1209
    },
    {
      "epoch": 1.7542587894164552,
      "grad_norm": 2.013259352424008,
      "learning_rate": 7.70228978009907e-07,
      "loss": 0.8342,
      "step": 1210
    },
    {
      "epoch": 1.7557085900688656,
      "grad_norm": 2.012249811228104,
      "learning_rate": 7.612046748871327e-07,
      "loss": 0.7847,
      "step": 1211
    },
    {
      "epoch": 1.7571583907212758,
      "grad_norm": 2.139469615844435,
      "learning_rate": 7.522314577732836e-07,
      "loss": 0.7888,
      "step": 1212
    },
    {
      "epoch": 1.758608191373686,
      "grad_norm": 2.1571860599105745,
      "learning_rate": 7.433093762858501e-07,
      "loss": 0.7783,
      "step": 1213
    },
    {
      "epoch": 1.7600579920260964,
      "grad_norm": 2.1025889052142586,
      "learning_rate": 7.344384797595717e-07,
      "loss": 0.7974,
      "step": 1214
    },
    {
      "epoch": 1.7615077926785068,
      "grad_norm": 2.289124200502034,
      "learning_rate": 7.256188172461521e-07,
      "loss": 0.8587,
      "step": 1215
    },
    {
      "epoch": 1.762957593330917,
      "grad_norm": 2.253045606276788,
      "learning_rate": 7.16850437514004e-07,
      "loss": 0.8514,
      "step": 1216
    },
    {
      "epoch": 1.7644073939833271,
      "grad_norm": 2.2044439033249974,
      "learning_rate": 7.081333890479691e-07,
      "loss": 0.8682,
      "step": 1217
    },
    {
      "epoch": 1.7658571946357375,
      "grad_norm": 2.107090164151527,
      "learning_rate": 6.994677200490508e-07,
      "loss": 0.848,
      "step": 1218
    },
    {
      "epoch": 1.7658571946357375,
      "eval_loss": 1.9923357963562012,
      "eval_runtime": 718.25,
      "eval_samples_per_second": 24.579,
      "eval_steps_per_second": 0.384,
      "step": 1218
    },
    {
      "epoch": 1.767306995288148,
      "grad_norm": 2.121615700492474,
      "learning_rate": 6.908534784341514e-07,
      "loss": 0.9243,
      "step": 1219
    },
    {
      "epoch": 1.7687567959405581,
      "grad_norm": 2.3850807789005,
      "learning_rate": 6.822907118358047e-07,
      "loss": 0.8718,
      "step": 1220
    },
    {
      "epoch": 1.7702065965929683,
      "grad_norm": 2.192733536376332,
      "learning_rate": 6.737794676019105e-07,
      "loss": 0.8297,
      "step": 1221
    },
    {
      "epoch": 1.7716563972453787,
      "grad_norm": 2.184301444251066,
      "learning_rate": 6.65319792795479e-07,
      "loss": 0.7989,
      "step": 1222
    },
    {
      "epoch": 1.7731061978977891,
      "grad_norm": 2.100152598434965,
      "learning_rate": 6.569117341943621e-07,
      "loss": 0.8593,
      "step": 1223
    },
    {
      "epoch": 1.7745559985501993,
      "grad_norm": 1.9818040970369857,
      "learning_rate": 6.485553382910026e-07,
      "loss": 0.7911,
      "step": 1224
    },
    {
      "epoch": 1.7760057992026095,
      "grad_norm": 2.333905575009875,
      "learning_rate": 6.402506512921736e-07,
      "loss": 0.8578,
      "step": 1225
    },
    {
      "epoch": 1.7774555998550199,
      "grad_norm": 2.1435464691932196,
      "learning_rate": 6.319977191187232e-07,
      "loss": 0.7987,
      "step": 1226
    },
    {
      "epoch": 1.7789054005074303,
      "grad_norm": 2.1494631554321004,
      "learning_rate": 6.237965874053209e-07,
      "loss": 0.7623,
      "step": 1227
    },
    {
      "epoch": 1.7803552011598405,
      "grad_norm": 2.3698974558716634,
      "learning_rate": 6.15647301500203e-07,
      "loss": 0.8749,
      "step": 1228
    },
    {
      "epoch": 1.7818050018122507,
      "grad_norm": 2.407161515696355,
      "learning_rate": 6.07549906464926e-07,
      "loss": 0.8609,
      "step": 1229
    },
    {
      "epoch": 1.783254802464661,
      "grad_norm": 2.159757561806183,
      "learning_rate": 5.995044470741151e-07,
      "loss": 0.8922,
      "step": 1230
    },
    {
      "epoch": 1.7847046031170715,
      "grad_norm": 2.140417575242218,
      "learning_rate": 5.915109678152176e-07,
      "loss": 0.8016,
      "step": 1231
    },
    {
      "epoch": 1.7861544037694816,
      "grad_norm": 2.053526278885337,
      "learning_rate": 5.835695128882512e-07,
      "loss": 0.8441,
      "step": 1232
    },
    {
      "epoch": 1.7861544037694816,
      "eval_loss": 1.9908360242843628,
      "eval_runtime": 716.251,
      "eval_samples_per_second": 24.648,
      "eval_steps_per_second": 0.385,
      "step": 1232
    },
    {
      "epoch": 1.7876042044218918,
      "grad_norm": 2.046920553778616,
      "learning_rate": 5.756801262055722e-07,
      "loss": 0.7977,
      "step": 1233
    },
    {
      "epoch": 1.7890540050743022,
      "grad_norm": 2.2912293685686405,
      "learning_rate": 5.678428513916212e-07,
      "loss": 0.7718,
      "step": 1234
    },
    {
      "epoch": 1.7905038057267126,
      "grad_norm": 2.136686223279879,
      "learning_rate": 5.600577317826861e-07,
      "loss": 0.9005,
      "step": 1235
    },
    {
      "epoch": 1.7919536063791228,
      "grad_norm": 2.4793018869650276,
      "learning_rate": 5.523248104266577e-07,
      "loss": 0.8606,
      "step": 1236
    },
    {
      "epoch": 1.793403407031533,
      "grad_norm": 2.5630043892312195,
      "learning_rate": 5.446441300828043e-07,
      "loss": 0.8007,
      "step": 1237
    },
    {
      "epoch": 1.7948532076839434,
      "grad_norm": 2.061209185007648,
      "learning_rate": 5.370157332215209e-07,
      "loss": 0.8686,
      "step": 1238
    },
    {
      "epoch": 1.7963030083363538,
      "grad_norm": 2.3350870222523152,
      "learning_rate": 5.294396620241038e-07,
      "loss": 0.8666,
      "step": 1239
    },
    {
      "epoch": 1.797752808988764,
      "grad_norm": 2.1851469922748383,
      "learning_rate": 5.219159583825107e-07,
      "loss": 0.8512,
      "step": 1240
    },
    {
      "epoch": 1.7992026096411742,
      "grad_norm": 2.1336249054339707,
      "learning_rate": 5.144446638991363e-07,
      "loss": 0.8029,
      "step": 1241
    },
    {
      "epoch": 1.8006524102935846,
      "grad_norm": 2.1553186006182075,
      "learning_rate": 5.070258198865741e-07,
      "loss": 0.8405,
      "step": 1242
    },
    {
      "epoch": 1.802102210945995,
      "grad_norm": 2.1111657715404424,
      "learning_rate": 4.99659467367396e-07,
      "loss": 0.7992,
      "step": 1243
    },
    {
      "epoch": 1.8035520115984052,
      "grad_norm": 2.33024804054167,
      "learning_rate": 4.923456470739219e-07,
      "loss": 0.8308,
      "step": 1244
    },
    {
      "epoch": 1.8050018122508154,
      "grad_norm": 2.187622025284149,
      "learning_rate": 4.850843994479892e-07,
      "loss": 0.8451,
      "step": 1245
    },
    {
      "epoch": 1.8064516129032258,
      "grad_norm": 2.2488941698736356,
      "learning_rate": 4.778757646407361e-07,
      "loss": 0.8795,
      "step": 1246
    },
    {
      "epoch": 1.8064516129032258,
      "eval_loss": 1.9922864437103271,
      "eval_runtime": 717.7803,
      "eval_samples_per_second": 24.595,
      "eval_steps_per_second": 0.385,
      "step": 1246
    },
    {
      "epoch": 1.8079014135556362,
      "grad_norm": 2.523143520626568,
      "learning_rate": 4.707197825123844e-07,
      "loss": 0.8662,
      "step": 1247
    },
    {
      "epoch": 1.8093512142080463,
      "grad_norm": 2.054454547469118,
      "learning_rate": 4.636164926320064e-07,
      "loss": 0.7958,
      "step": 1248
    },
    {
      "epoch": 1.8108010148604565,
      "grad_norm": 2.2153270605643542,
      "learning_rate": 4.565659342773088e-07,
      "loss": 0.8692,
      "step": 1249
    },
    {
      "epoch": 1.812250815512867,
      "grad_norm": 2.142591881691151,
      "learning_rate": 4.495681464344259e-07,
      "loss": 0.8439,
      "step": 1250
    },
    {
      "epoch": 1.8137006161652773,
      "grad_norm": 2.3121882108109455,
      "learning_rate": 4.426231677976922e-07,
      "loss": 0.7906,
      "step": 1251
    },
    {
      "epoch": 1.8151504168176875,
      "grad_norm": 2.180802221775493,
      "learning_rate": 4.3573103676943784e-07,
      "loss": 0.8578,
      "step": 1252
    },
    {
      "epoch": 1.8166002174700977,
      "grad_norm": 2.320230114407269,
      "learning_rate": 4.2889179145976457e-07,
      "loss": 0.8545,
      "step": 1253
    },
    {
      "epoch": 1.818050018122508,
      "grad_norm": 2.3719493518316153,
      "learning_rate": 4.2210546968634803e-07,
      "loss": 0.8557,
      "step": 1254
    },
    {
      "epoch": 1.8194998187749185,
      "grad_norm": 2.493479221496669,
      "learning_rate": 4.153721089742213e-07,
      "loss": 0.9154,
      "step": 1255
    },
    {
      "epoch": 1.8209496194273287,
      "grad_norm": 2.0747833801399254,
      "learning_rate": 4.086917465555662e-07,
      "loss": 0.8529,
      "step": 1256
    },
    {
      "epoch": 1.8223994200797389,
      "grad_norm": 1.933887631090084,
      "learning_rate": 4.0206441936951556e-07,
      "loss": 0.7236,
      "step": 1257
    },
    {
      "epoch": 1.8238492207321493,
      "grad_norm": 2.170704103201851,
      "learning_rate": 3.9549016406193686e-07,
      "loss": 0.9103,
      "step": 1258
    },
    {
      "epoch": 1.8252990213845597,
      "grad_norm": 2.3362798049220728,
      "learning_rate": 3.889690169852378e-07,
      "loss": 0.8449,
      "step": 1259
    },
    {
      "epoch": 1.8267488220369699,
      "grad_norm": 2.276132943452303,
      "learning_rate": 3.8250101419816777e-07,
      "loss": 0.7784,
      "step": 1260
    },
    {
      "epoch": 1.8267488220369699,
      "eval_loss": 1.9929546117782593,
      "eval_runtime": 714.0203,
      "eval_samples_per_second": 24.725,
      "eval_steps_per_second": 0.387,
      "step": 1260
    },
    {
      "epoch": 1.82819862268938,
      "grad_norm": 2.1669846316286505,
      "learning_rate": 3.7608619146560996e-07,
      "loss": 0.8131,
      "step": 1261
    },
    {
      "epoch": 1.8296484233417905,
      "grad_norm": 1.971450538763261,
      "learning_rate": 3.6972458425838963e-07,
      "loss": 0.8581,
      "step": 1262
    },
    {
      "epoch": 1.8310982239942009,
      "grad_norm": 2.1710930519866505,
      "learning_rate": 3.6341622775307506e-07,
      "loss": 0.8425,
      "step": 1263
    },
    {
      "epoch": 1.832548024646611,
      "grad_norm": 2.2818712093684352,
      "learning_rate": 3.5716115683178563e-07,
      "loss": 0.7891,
      "step": 1264
    },
    {
      "epoch": 1.8339978252990212,
      "grad_norm": 2.2687382403391974,
      "learning_rate": 3.5095940608199875e-07,
      "loss": 0.8931,
      "step": 1265
    },
    {
      "epoch": 1.8354476259514316,
      "grad_norm": 2.0306979971189962,
      "learning_rate": 3.4481100979635304e-07,
      "loss": 0.8198,
      "step": 1266
    },
    {
      "epoch": 1.836897426603842,
      "grad_norm": 2.0911196887193824,
      "learning_rate": 3.3871600197246667e-07,
      "loss": 0.7904,
      "step": 1267
    },
    {
      "epoch": 1.8383472272562522,
      "grad_norm": 2.145949880159978,
      "learning_rate": 3.32674416312746e-07,
      "loss": 0.8366,
      "step": 1268
    },
    {
      "epoch": 1.8397970279086624,
      "grad_norm": 2.085556897438766,
      "learning_rate": 3.2668628622419707e-07,
      "loss": 0.8235,
      "step": 1269
    },
    {
      "epoch": 1.8412468285610728,
      "grad_norm": 2.163067756145073,
      "learning_rate": 3.207516448182435e-07,
      "loss": 0.833,
      "step": 1270
    },
    {
      "epoch": 1.8426966292134832,
      "grad_norm": 2.0890503431871372,
      "learning_rate": 3.1487052491054216e-07,
      "loss": 0.862,
      "step": 1271
    },
    {
      "epoch": 1.8441464298658934,
      "grad_norm": 2.120683177848187,
      "learning_rate": 3.090429590208033e-07,
      "loss": 0.8831,
      "step": 1272
    },
    {
      "epoch": 1.8455962305183036,
      "grad_norm": 2.257483000349536,
      "learning_rate": 3.032689793726107e-07,
      "loss": 0.7932,
      "step": 1273
    },
    {
      "epoch": 1.847046031170714,
      "grad_norm": 2.108524772901258,
      "learning_rate": 2.9754861789324074e-07,
      "loss": 0.7124,
      "step": 1274
    },
    {
      "epoch": 1.847046031170714,
      "eval_loss": 1.9929630756378174,
      "eval_runtime": 713.9371,
      "eval_samples_per_second": 24.728,
      "eval_steps_per_second": 0.387,
      "step": 1274
    },
    {
      "epoch": 1.8484958318231244,
      "grad_norm": 2.2069888080373112,
      "learning_rate": 2.91881906213487e-07,
      "loss": 0.8675,
      "step": 1275
    },
    {
      "epoch": 1.8499456324755346,
      "grad_norm": 2.173204022221021,
      "learning_rate": 2.862688756674881e-07,
      "loss": 0.8183,
      "step": 1276
    },
    {
      "epoch": 1.8513954331279447,
      "grad_norm": 2.189526444810239,
      "learning_rate": 2.8070955729255e-07,
      "loss": 0.785,
      "step": 1277
    },
    {
      "epoch": 1.8528452337803551,
      "grad_norm": 2.3771964846068316,
      "learning_rate": 2.752039818289809e-07,
      "loss": 0.8897,
      "step": 1278
    },
    {
      "epoch": 1.8542950344327656,
      "grad_norm": 2.110356795634518,
      "learning_rate": 2.697521797199099e-07,
      "loss": 0.8196,
      "step": 1279
    },
    {
      "epoch": 1.8557448350851757,
      "grad_norm": 2.1445568126409746,
      "learning_rate": 2.6435418111113277e-07,
      "loss": 0.8626,
      "step": 1280
    },
    {
      "epoch": 1.857194635737586,
      "grad_norm": 2.3612235079933055,
      "learning_rate": 2.5901001585093453e-07,
      "loss": 0.8382,
      "step": 1281
    },
    {
      "epoch": 1.8586444363899963,
      "grad_norm": 2.340763698089208,
      "learning_rate": 2.537197134899294e-07,
      "loss": 0.8254,
      "step": 1282
    },
    {
      "epoch": 1.8600942370424067,
      "grad_norm": 2.3010489544975363,
      "learning_rate": 2.4848330328089643e-07,
      "loss": 0.914,
      "step": 1283
    },
    {
      "epoch": 1.861544037694817,
      "grad_norm": 2.019453046767646,
      "learning_rate": 2.4330081417861527e-07,
      "loss": 0.8647,
      "step": 1284
    },
    {
      "epoch": 1.862993838347227,
      "grad_norm": 2.2937850144263687,
      "learning_rate": 2.3817227483971084e-07,
      "loss": 0.8768,
      "step": 1285
    },
    {
      "epoch": 1.8644436389996375,
      "grad_norm": 2.2136506520594197,
      "learning_rate": 2.3309771362249323e-07,
      "loss": 0.799,
      "step": 1286
    },
    {
      "epoch": 1.865893439652048,
      "grad_norm": 2.2799429303455723,
      "learning_rate": 2.280771585867969e-07,
      "loss": 0.8377,
      "step": 1287
    },
    {
      "epoch": 1.867343240304458,
      "grad_norm": 2.470863928651675,
      "learning_rate": 2.2311063749382743e-07,
      "loss": 0.8059,
      "step": 1288
    },
    {
      "epoch": 1.867343240304458,
      "eval_loss": 1.992066740989685,
      "eval_runtime": 711.7633,
      "eval_samples_per_second": 24.803,
      "eval_steps_per_second": 0.388,
      "step": 1288
    },
    {
      "epoch": 1.8687930409568683,
      "grad_norm": 2.22213354301716,
      "learning_rate": 2.1819817780601494e-07,
      "loss": 0.8232,
      "step": 1289
    },
    {
      "epoch": 1.8702428416092787,
      "grad_norm": 2.3427078842323246,
      "learning_rate": 2.1333980668685416e-07,
      "loss": 0.8255,
      "step": 1290
    },
    {
      "epoch": 1.871692642261689,
      "grad_norm": 2.158432434537126,
      "learning_rate": 2.085355510007514e-07,
      "loss": 0.8454,
      "step": 1291
    },
    {
      "epoch": 1.8731424429140993,
      "grad_norm": 2.1578965787718754,
      "learning_rate": 2.037854373128889e-07,
      "loss": 0.86,
      "step": 1292
    },
    {
      "epoch": 1.8745922435665094,
      "grad_norm": 2.1393091529675807,
      "learning_rate": 1.9908949188906513e-07,
      "loss": 0.8829,
      "step": 1293
    },
    {
      "epoch": 1.8760420442189198,
      "grad_norm": 2.222909926071255,
      "learning_rate": 1.9444774069555694e-07,
      "loss": 0.8283,
      "step": 1294
    },
    {
      "epoch": 1.8774918448713303,
      "grad_norm": 2.2535245896528315,
      "learning_rate": 1.898602093989721e-07,
      "loss": 0.7958,
      "step": 1295
    },
    {
      "epoch": 1.8789416455237404,
      "grad_norm": 1.9793513025034748,
      "learning_rate": 1.8532692336611037e-07,
      "loss": 0.83,
      "step": 1296
    },
    {
      "epoch": 1.8803914461761506,
      "grad_norm": 2.044094728130835,
      "learning_rate": 1.8084790766382032e-07,
      "loss": 0.7845,
      "step": 1297
    },
    {
      "epoch": 1.881841246828561,
      "grad_norm": 2.218645705281939,
      "learning_rate": 1.7642318705886284e-07,
      "loss": 0.825,
      "step": 1298
    },
    {
      "epoch": 1.8832910474809714,
      "grad_norm": 2.097587790697249,
      "learning_rate": 1.7205278601777565e-07,
      "loss": 0.7882,
      "step": 1299
    },
    {
      "epoch": 1.8847408481333816,
      "grad_norm": 2.100545696172868,
      "learning_rate": 1.677367287067322e-07,
      "loss": 0.7571,
      "step": 1300
    },
    {
      "epoch": 1.8861906487857918,
      "grad_norm": 2.3104562461778833,
      "learning_rate": 1.6347503899141303e-07,
      "loss": 0.8614,
      "step": 1301
    },
    {
      "epoch": 1.8876404494382022,
      "grad_norm": 2.5112678379774827,
      "learning_rate": 1.5926774043687366e-07,
      "loss": 0.8213,
      "step": 1302
    },
    {
      "epoch": 1.8876404494382022,
      "eval_loss": 1.9908733367919922,
      "eval_runtime": 712.1232,
      "eval_samples_per_second": 24.791,
      "eval_steps_per_second": 0.388,
      "step": 1302
    },
    {
      "epoch": 1.8890902500906126,
      "grad_norm": 2.2854104468065395,
      "learning_rate": 1.5511485630741452e-07,
      "loss": 0.8332,
      "step": 1303
    },
    {
      "epoch": 1.8905400507430228,
      "grad_norm": 2.14999861816316,
      "learning_rate": 1.510164095664457e-07,
      "loss": 0.8191,
      "step": 1304
    },
    {
      "epoch": 1.891989851395433,
      "grad_norm": 2.59779626841818,
      "learning_rate": 1.4697242287637025e-07,
      "loss": 0.8882,
      "step": 1305
    },
    {
      "epoch": 1.8934396520478434,
      "grad_norm": 2.212616111367292,
      "learning_rate": 1.4298291859845215e-07,
      "loss": 0.8626,
      "step": 1306
    },
    {
      "epoch": 1.8948894527002538,
      "grad_norm": 2.3719425347244423,
      "learning_rate": 1.3904791879269296e-07,
      "loss": 0.7726,
      "step": 1307
    },
    {
      "epoch": 1.896339253352664,
      "grad_norm": 2.4089239600460037,
      "learning_rate": 1.351674452177143e-07,
      "loss": 0.8951,
      "step": 1308
    },
    {
      "epoch": 1.8977890540050741,
      "grad_norm": 2.3779499546623932,
      "learning_rate": 1.313415193306322e-07,
      "loss": 0.8538,
      "step": 1309
    },
    {
      "epoch": 1.8992388546574845,
      "grad_norm": 2.1199931249891337,
      "learning_rate": 1.2757016228693964e-07,
      "loss": 0.7918,
      "step": 1310
    },
    {
      "epoch": 1.900688655309895,
      "grad_norm": 2.4960573304139255,
      "learning_rate": 1.2385339494039307e-07,
      "loss": 0.8524,
      "step": 1311
    },
    {
      "epoch": 1.9021384559623051,
      "grad_norm": 2.38840949774226,
      "learning_rate": 1.2019123784289488e-07,
      "loss": 0.8369,
      "step": 1312
    },
    {
      "epoch": 1.9035882566147153,
      "grad_norm": 2.3803287166667055,
      "learning_rate": 1.1658371124437796e-07,
      "loss": 0.8473,
      "step": 1313
    },
    {
      "epoch": 1.9050380572671257,
      "grad_norm": 2.2744794341611874,
      "learning_rate": 1.1303083509269452e-07,
      "loss": 0.9025,
      "step": 1314
    },
    {
      "epoch": 1.9064878579195361,
      "grad_norm": 2.2722169969170825,
      "learning_rate": 1.0953262903350858e-07,
      "loss": 0.8525,
      "step": 1315
    },
    {
      "epoch": 1.9079376585719463,
      "grad_norm": 2.0741552187285976,
      "learning_rate": 1.0608911241018594e-07,
      "loss": 0.8603,
      "step": 1316
    },
    {
      "epoch": 1.9079376585719463,
      "eval_loss": 1.9904147386550903,
      "eval_runtime": 710.5034,
      "eval_samples_per_second": 24.847,
      "eval_steps_per_second": 0.388,
      "step": 1316
    },
    {
      "epoch": 1.9093874592243565,
      "grad_norm": 2.416024759058928,
      "learning_rate": 1.0270030426368427e-07,
      "loss": 0.9295,
      "step": 1317
    },
    {
      "epoch": 1.910837259876767,
      "grad_norm": 2.512925732607704,
      "learning_rate": 9.936622333245105e-08,
      "loss": 0.8741,
      "step": 1318
    },
    {
      "epoch": 1.9122870605291773,
      "grad_norm": 2.3046332686313606,
      "learning_rate": 9.608688805232024e-08,
      "loss": 0.8094,
      "step": 1319
    },
    {
      "epoch": 1.9137368611815875,
      "grad_norm": 2.1750487437484702,
      "learning_rate": 9.286231655640799e-08,
      "loss": 0.8895,
      "step": 1320
    },
    {
      "epoch": 1.9151866618339977,
      "grad_norm": 2.0926427690000606,
      "learning_rate": 8.969252667501372e-08,
      "loss": 0.8494,
      "step": 1321
    },
    {
      "epoch": 1.916636462486408,
      "grad_norm": 2.1983298971224947,
      "learning_rate": 8.657753593552142e-08,
      "loss": 0.8366,
      "step": 1322
    },
    {
      "epoch": 1.9180862631388185,
      "grad_norm": 2.4296166867506592,
      "learning_rate": 8.351736156230305e-08,
      "loss": 0.8862,
      "step": 1323
    },
    {
      "epoch": 1.9195360637912287,
      "grad_norm": 2.39395990388658,
      "learning_rate": 8.051202047662187e-08,
      "loss": 0.8204,
      "step": 1324
    },
    {
      "epoch": 1.9209858644436388,
      "grad_norm": 1.968322916498447,
      "learning_rate": 7.75615292965426e-08,
      "loss": 0.9303,
      "step": 1325
    },
    {
      "epoch": 1.9224356650960492,
      "grad_norm": 2.1683820935329194,
      "learning_rate": 7.466590433683252e-08,
      "loss": 0.8229,
      "step": 1326
    },
    {
      "epoch": 1.9238854657484596,
      "grad_norm": 2.2048786282054107,
      "learning_rate": 7.182516160887943e-08,
      "loss": 0.7996,
      "step": 1327
    },
    {
      "epoch": 1.9253352664008698,
      "grad_norm": 2.0835582437546805,
      "learning_rate": 6.903931682059828e-08,
      "loss": 0.7705,
      "step": 1328
    },
    {
      "epoch": 1.92678506705328,
      "grad_norm": 2.1544450006600147,
      "learning_rate": 6.630838537634465e-08,
      "loss": 0.9109,
      "step": 1329
    },
    {
      "epoch": 1.9282348677056904,
      "grad_norm": 2.2591067844591723,
      "learning_rate": 6.363238237683034e-08,
      "loss": 0.8402,
      "step": 1330
    },
    {
      "epoch": 1.9282348677056904,
      "eval_loss": 1.9893231391906738,
      "eval_runtime": 716.3056,
      "eval_samples_per_second": 24.646,
      "eval_steps_per_second": 0.385,
      "step": 1330
    },
    {
      "epoch": 1.9296846683581008,
      "grad_norm": 2.455688107986527,
      "learning_rate": 6.101132261904009e-08,
      "loss": 0.8356,
      "step": 1331
    },
    {
      "epoch": 1.931134469010511,
      "grad_norm": 1.9961595959615874,
      "learning_rate": 5.8445220596149434e-08,
      "loss": 0.8111,
      "step": 1332
    },
    {
      "epoch": 1.9325842696629212,
      "grad_norm": 2.3879403592665773,
      "learning_rate": 5.593409049744591e-08,
      "loss": 0.8886,
      "step": 1333
    },
    {
      "epoch": 1.9340340703153316,
      "grad_norm": 2.347979631551465,
      "learning_rate": 5.347794620824576e-08,
      "loss": 0.9087,
      "step": 1334
    },
    {
      "epoch": 1.935483870967742,
      "grad_norm": 2.3196593422793397,
      "learning_rate": 5.1076801309823954e-08,
      "loss": 0.8701,
      "step": 1335
    },
    {
      "epoch": 1.9369336716201522,
      "grad_norm": 2.5055558778620113,
      "learning_rate": 4.8730669079335434e-08,
      "loss": 0.8979,
      "step": 1336
    },
    {
      "epoch": 1.9383834722725624,
      "grad_norm": 2.1676302569194723,
      "learning_rate": 4.643956248973958e-08,
      "loss": 0.7302,
      "step": 1337
    },
    {
      "epoch": 1.9398332729249728,
      "grad_norm": 2.3897454990290337,
      "learning_rate": 4.420349420973358e-08,
      "loss": 0.8005,
      "step": 1338
    },
    {
      "epoch": 1.9412830735773832,
      "grad_norm": 2.1956858503889145,
      "learning_rate": 4.202247660367809e-08,
      "loss": 0.7514,
      "step": 1339
    },
    {
      "epoch": 1.9427328742297933,
      "grad_norm": 2.2090594473952074,
      "learning_rate": 3.98965217315328e-08,
      "loss": 0.8707,
      "step": 1340
    },
    {
      "epoch": 1.9441826748822035,
      "grad_norm": 2.1339030059185924,
      "learning_rate": 3.7825641348784304e-08,
      "loss": 0.8175,
      "step": 1341
    },
    {
      "epoch": 1.945632475534614,
      "grad_norm": 2.1979642753594977,
      "learning_rate": 3.580984690638611e-08,
      "loss": 0.818,
      "step": 1342
    },
    {
      "epoch": 1.9470822761870243,
      "grad_norm": 2.349815276796271,
      "learning_rate": 3.3849149550694294e-08,
      "loss": 0.8233,
      "step": 1343
    },
    {
      "epoch": 1.9485320768394345,
      "grad_norm": 2.428446436701222,
      "learning_rate": 3.194356012340305e-08,
      "loss": 0.8492,
      "step": 1344
    },
    {
      "epoch": 1.9485320768394345,
      "eval_loss": 1.990075707435608,
      "eval_runtime": 717.9662,
      "eval_samples_per_second": 24.589,
      "eval_steps_per_second": 0.384,
      "step": 1344
    },
    {
      "epoch": 1.9499818774918447,
      "grad_norm": 2.3796373844188716,
      "learning_rate": 3.009308916148812e-08,
      "loss": 0.7978,
      "step": 1345
    },
    {
      "epoch": 1.951431678144255,
      "grad_norm": 2.0041837352782106,
      "learning_rate": 2.8297746897146816e-08,
      "loss": 0.854,
      "step": 1346
    },
    {
      "epoch": 1.9528814787966655,
      "grad_norm": 2.2530547828978817,
      "learning_rate": 2.6557543257741404e-08,
      "loss": 0.7767,
      "step": 1347
    },
    {
      "epoch": 1.9543312794490757,
      "grad_norm": 2.3912278167385494,
      "learning_rate": 2.4872487865744698e-08,
      "loss": 0.8501,
      "step": 1348
    },
    {
      "epoch": 1.9557810801014859,
      "grad_norm": 2.1973995583645762,
      "learning_rate": 2.3242590038687896e-08,
      "loss": 0.7526,
      "step": 1349
    },
    {
      "epoch": 1.9572308807538963,
      "grad_norm": 2.112060910154867,
      "learning_rate": 2.1667858789105046e-08,
      "loss": 0.7967,
      "step": 1350
    },
    {
      "epoch": 1.9586806814063067,
      "grad_norm": 2.1849056636414597,
      "learning_rate": 2.0148302824490872e-08,
      "loss": 0.8294,
      "step": 1351
    },
    {
      "epoch": 1.9601304820587169,
      "grad_norm": 2.3461606241071182,
      "learning_rate": 1.8683930547243044e-08,
      "loss": 0.7979,
      "step": 1352
    },
    {
      "epoch": 1.961580282711127,
      "grad_norm": 2.212577854404884,
      "learning_rate": 1.7274750054622204e-08,
      "loss": 0.8566,
      "step": 1353
    },
    {
      "epoch": 1.9630300833635375,
      "grad_norm": 2.155734143684328,
      "learning_rate": 1.592076913870644e-08,
      "loss": 0.8222,
      "step": 1354
    },
    {
      "epoch": 1.9644798840159479,
      "grad_norm": 2.2674443861659324,
      "learning_rate": 1.462199528634578e-08,
      "loss": 0.8573,
      "step": 1355
    },
    {
      "epoch": 1.965929684668358,
      "grad_norm": 2.214105333576144,
      "learning_rate": 1.3378435679122225e-08,
      "loss": 0.8202,
      "step": 1356
    },
    {
      "epoch": 1.9673794853207682,
      "grad_norm": 2.4818998203739886,
      "learning_rate": 1.21900971933131e-08,
      "loss": 0.8677,
      "step": 1357
    },
    {
      "epoch": 1.9688292859731786,
      "grad_norm": 2.3359980788198818,
      "learning_rate": 1.1056986399845537e-08,
      "loss": 0.843,
      "step": 1358
    },
    {
      "epoch": 1.9688292859731786,
      "eval_loss": 1.9899805784225464,
      "eval_runtime": 714.4129,
      "eval_samples_per_second": 24.711,
      "eval_steps_per_second": 0.386,
      "step": 1358
    },
    {
      "epoch": 1.970279086625589,
      "grad_norm": 2.357049735704998,
      "learning_rate": 9.979109564267619e-09,
      "loss": 0.866,
      "step": 1359
    },
    {
      "epoch": 1.9717288872779992,
      "grad_norm": 2.1648208349303353,
      "learning_rate": 8.956472646710624e-09,
      "loss": 0.813,
      "step": 1360
    },
    {
      "epoch": 1.9731786879304094,
      "grad_norm": 2.306804077346519,
      "learning_rate": 7.989081301853496e-09,
      "loss": 0.837,
      "step": 1361
    },
    {
      "epoch": 1.9746284885828198,
      "grad_norm": 2.3130294887932568,
      "learning_rate": 7.07694087889621e-09,
      "loss": 0.8433,
      "step": 1362
    },
    {
      "epoch": 1.9760782892352302,
      "grad_norm": 2.298575357573264,
      "learning_rate": 6.220056421530895e-09,
      "loss": 0.8197,
      "step": 1363
    },
    {
      "epoch": 1.9775280898876404,
      "grad_norm": 2.1227695765787167,
      "learning_rate": 5.418432667905204e-09,
      "loss": 0.871,
      "step": 1364
    },
    {
      "epoch": 1.9789778905400506,
      "grad_norm": 2.1411296244476596,
      "learning_rate": 4.672074050607878e-09,
      "loss": 0.8542,
      "step": 1365
    },
    {
      "epoch": 1.980427691192461,
      "grad_norm": 2.1074488434056406,
      "learning_rate": 3.9809846966343265e-09,
      "loss": 0.7862,
      "step": 1366
    },
    {
      "epoch": 1.9818774918448714,
      "grad_norm": 2.2685586697126197,
      "learning_rate": 3.345168427371093e-09,
      "loss": 0.8854,
      "step": 1367
    },
    {
      "epoch": 1.9833272924972816,
      "grad_norm": 2.3674256189165352,
      "learning_rate": 2.764628758570309e-09,
      "loss": 0.8155,
      "step": 1368
    },
    {
      "epoch": 1.9847770931496918,
      "grad_norm": 1.926523934505123,
      "learning_rate": 2.2393689003308296e-09,
      "loss": 0.9116,
      "step": 1369
    },
    {
      "epoch": 1.9862268938021022,
      "grad_norm": 2.0467717368763667,
      "learning_rate": 1.7693917570837938e-09,
      "loss": 0.7796,
      "step": 1370
    },
    {
      "epoch": 1.9876766944545126,
      "grad_norm": 2.093729017663572,
      "learning_rate": 1.3546999275715344e-09,
      "loss": 0.8335,
      "step": 1371
    },
    {
      "epoch": 1.9891264951069227,
      "grad_norm": 2.13305807864378,
      "learning_rate": 9.95295704835364e-10,
      "loss": 0.7784,
      "step": 1372
    },
    {
      "epoch": 1.9891264951069227,
      "eval_loss": 1.9901281595230103,
      "eval_runtime": 714.3259,
      "eval_samples_per_second": 24.714,
      "eval_steps_per_second": 0.386,
      "step": 1372
    },
    {
      "epoch": 1.990576295759333,
      "grad_norm": 2.185518676436933,
      "learning_rate": 6.911810762066929e-10,
      "loss": 0.8326,
      "step": 1373
    },
    {
      "epoch": 1.9920260964117433,
      "grad_norm": 2.055870573171998,
      "learning_rate": 4.4235772328815594e-10,
      "loss": 0.7941,
      "step": 1374
    },
    {
      "epoch": 1.9934758970641537,
      "grad_norm": 2.1987540775939483,
      "learning_rate": 2.4882702195361265e-10,
      "loss": 0.8484,
      "step": 1375
    },
    {
      "epoch": 1.994925697716564,
      "grad_norm": 2.247386337310816,
      "learning_rate": 1.1059004233038296e-10,
      "loss": 0.8048,
      "step": 1376
    },
    {
      "epoch": 1.996375498368974,
      "grad_norm": 2.212082576318791,
      "learning_rate": 2.764754880146825e-11,
      "loss": 0.8428,
      "step": 1377
    },
    {
      "epoch": 1.9978252990213845,
      "grad_norm": 2.224473621074023,
      "learning_rate": 0.0,
      "loss": 0.7812,
      "step": 1378
    },
    {
      "epoch": 1.9978252990213845,
      "step": 1378,
      "total_flos": 869856505233408.0,
      "train_loss": 1.0392584558062699,
      "train_runtime": 118131.7407,
      "train_samples_per_second": 2.989,
      "train_steps_per_second": 0.012
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 1378,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 138,
  "total_flos": 869856505233408.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}