{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 15680,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006377551020408163,
      "grad_norm": 0.6313766146507682,
      "learning_rate": 9.999998394288208e-06,
      "loss": 0.33850436210632323,
      "num_input_tokens_seen": 309624,
      "step": 5,
      "train_runtime": 66.1477,
      "train_tokens_per_second": 4680.794
    },
    {
      "epoch": 0.0012755102040816326,
      "grad_norm": 0.6205915413500387,
      "learning_rate": 9.999991871085821e-06,
      "loss": 0.3195484161376953,
      "num_input_tokens_seen": 617984,
      "step": 10,
      "train_runtime": 131.5668,
      "train_tokens_per_second": 4697.11
    },
    {
      "epoch": 0.001913265306122449,
      "grad_norm": 0.6369759482202858,
      "learning_rate": 9.999980330042391e-06,
      "loss": 0.31821389198303224,
      "num_input_tokens_seen": 922216,
      "step": 15,
      "train_runtime": 189.9263,
      "train_tokens_per_second": 4855.651
    },
    {
      "epoch": 0.002551020408163265,
      "grad_norm": 0.6101008388950145,
      "learning_rate": 9.999963771169503e-06,
      "loss": 0.3285261392593384,
      "num_input_tokens_seen": 1228440,
      "step": 20,
      "train_runtime": 255.8324,
      "train_tokens_per_second": 4801.737
    },
    {
      "epoch": 0.0031887755102040817,
      "grad_norm": 0.6834911121613362,
      "learning_rate": 9.999942194483775e-06,
      "loss": 0.3178420543670654,
      "num_input_tokens_seen": 1534256,
      "step": 25,
      "train_runtime": 314.5148,
      "train_tokens_per_second": 4878.168
    },
    {
      "epoch": 0.003826530612244898,
      "grad_norm": 0.6170487360838941,
      "learning_rate": 9.999915600006857e-06,
      "loss": 0.3137950897216797,
      "num_input_tokens_seen": 1841184,
      "step": 30,
      "train_runtime": 370.1851,
      "train_tokens_per_second": 4973.685
    },
    {
      "epoch": 0.004464285714285714,
      "grad_norm": 0.6716564497994858,
      "learning_rate": 9.999883987765443e-06,
      "loss": 0.30161247253417967,
      "num_input_tokens_seen": 2134520,
      "step": 35,
      "train_runtime": 444.1605,
      "train_tokens_per_second": 4805.74
    },
    {
      "epoch": 0.00510204081632653,
      "grad_norm": 0.6523834223862264,
      "learning_rate": 9.999847357791256e-06,
      "loss": 0.3267825126647949,
      "num_input_tokens_seen": 2437936,
      "step": 40,
      "train_runtime": 512.2405,
      "train_tokens_per_second": 4759.358
    },
    {
      "epoch": 0.005739795918367347,
      "grad_norm": 0.6398987507232226,
      "learning_rate": 9.999805710121055e-06,
      "loss": 0.3393638849258423,
      "num_input_tokens_seen": 2754040,
      "step": 45,
      "train_runtime": 571.3353,
      "train_tokens_per_second": 4820.357
    },
    {
      "epoch": 0.006377551020408163,
      "grad_norm": 0.6106921304806688,
      "learning_rate": 9.99975904479664e-06,
      "loss": 0.3237131118774414,
      "num_input_tokens_seen": 3041032,
      "step": 50,
      "train_runtime": 641.5805,
      "train_tokens_per_second": 4739.907
    },
    {
      "epoch": 0.00701530612244898,
      "grad_norm": 0.6958236841111044,
      "learning_rate": 9.999707361864841e-06,
      "loss": 0.36842975616455076,
      "num_input_tokens_seen": 3355376,
      "step": 55,
      "train_runtime": 689.5653,
      "train_tokens_per_second": 4865.929
    },
    {
      "epoch": 0.007653061224489796,
      "grad_norm": 0.7335802456853094,
      "learning_rate": 9.999650661377527e-06,
      "loss": 0.3665752410888672,
      "num_input_tokens_seen": 3655680,
      "step": 60,
      "train_runtime": 748.2584,
      "train_tokens_per_second": 4885.585
    },
    {
      "epoch": 0.008290816326530613,
      "grad_norm": 0.6788516722261737,
      "learning_rate": 9.999588943391597e-06,
      "loss": 0.3658756256103516,
      "num_input_tokens_seen": 3965648,
      "step": 65,
      "train_runtime": 821.1636,
      "train_tokens_per_second": 4829.303
    },
    {
      "epoch": 0.008928571428571428,
      "grad_norm": 0.6773708513395791,
      "learning_rate": 9.999522207968992e-06,
      "loss": 0.32802519798278806,
      "num_input_tokens_seen": 4276160,
      "step": 70,
      "train_runtime": 874.2068,
      "train_tokens_per_second": 4891.474
    },
    {
      "epoch": 0.009566326530612245,
      "grad_norm": 0.618622603818562,
      "learning_rate": 9.999450455176686e-06,
      "loss": 0.3478161573410034,
      "num_input_tokens_seen": 4584392,
      "step": 75,
      "train_runtime": 936.5082,
      "train_tokens_per_second": 4895.197
    },
    {
      "epoch": 0.01020408163265306,
      "grad_norm": 0.6260901752075039,
      "learning_rate": 9.99937368508669e-06,
      "loss": 0.3187378168106079,
      "num_input_tokens_seen": 4891664,
      "step": 80,
      "train_runtime": 1004.7186,
      "train_tokens_per_second": 4868.691
    },
    {
      "epoch": 0.010841836734693877,
      "grad_norm": 0.667190746463301,
      "learning_rate": 9.999291897776043e-06,
      "loss": 0.32980921268463137,
      "num_input_tokens_seen": 5189440,
      "step": 85,
      "train_runtime": 1078.1879,
      "train_tokens_per_second": 4813.113
    },
    {
      "epoch": 0.011479591836734694,
      "grad_norm": 0.5750586926863843,
      "learning_rate": 9.999205093326825e-06,
      "loss": 0.33111636638641356,
      "num_input_tokens_seen": 5485064,
      "step": 90,
      "train_runtime": 1146.8163,
      "train_tokens_per_second": 4782.862
    },
    {
      "epoch": 0.01211734693877551,
      "grad_norm": 0.6017076025749912,
      "learning_rate": 9.999113271826154e-06,
      "loss": 0.31922378540039065,
      "num_input_tokens_seen": 5780584,
      "step": 95,
      "train_runtime": 1208.5778,
      "train_tokens_per_second": 4782.964
    },
    {
      "epoch": 0.012755102040816327,
      "grad_norm": 0.606460864307619,
      "learning_rate": 9.999016433366178e-06,
      "loss": 0.34738271236419677,
      "num_input_tokens_seen": 6084648,
      "step": 100,
      "train_runtime": 1276.8553,
      "train_tokens_per_second": 4765.339
    },
    {
      "epoch": 0.013392857142857142,
      "grad_norm": 0.6197516341828899,
      "learning_rate": 9.998914578044079e-06,
      "loss": 0.3142980098724365,
      "num_input_tokens_seen": 6390464,
      "step": 105,
      "train_runtime": 1337.4726,
      "train_tokens_per_second": 4778.015
    },
    {
      "epoch": 0.01403061224489796,
      "grad_norm": 0.6235817964061284,
      "learning_rate": 9.998807705962079e-06,
      "loss": 0.3406642436981201,
      "num_input_tokens_seen": 6691224,
      "step": 110,
      "train_runtime": 1407.56,
      "train_tokens_per_second": 4753.775
    },
    {
      "epoch": 0.014668367346938776,
      "grad_norm": 0.677622003307496,
      "learning_rate": 9.99869581722743e-06,
      "loss": 0.3410198211669922,
      "num_input_tokens_seen": 6997664,
      "step": 115,
      "train_runtime": 1478.1146,
      "train_tokens_per_second": 4734.182
    },
    {
      "epoch": 0.015306122448979591,
      "grad_norm": 0.6649293990093726,
      "learning_rate": 9.99857891195242e-06,
      "loss": 0.34250473976135254,
      "num_input_tokens_seen": 7304528,
      "step": 120,
      "train_runtime": 1539.9315,
      "train_tokens_per_second": 4743.411
    },
    {
      "epoch": 0.01594387755102041,
      "grad_norm": 0.7071355973636468,
      "learning_rate": 9.998456990254371e-06,
      "loss": 0.32828989028930666,
      "num_input_tokens_seen": 7613176,
      "step": 125,
      "train_runtime": 1611.0141,
      "train_tokens_per_second": 4725.704
    },
    {
      "epoch": 0.016581632653061226,
      "grad_norm": 0.6158828508586639,
      "learning_rate": 9.998330052255642e-06,
      "loss": 0.3207663059234619,
      "num_input_tokens_seen": 7931376,
      "step": 130,
      "train_runtime": 1661.8304,
      "train_tokens_per_second": 4772.675
    },
    {
      "epoch": 0.01721938775510204,
      "grad_norm": 0.6507793141596079,
      "learning_rate": 9.998198098083623e-06,
      "loss": 0.3120201349258423,
      "num_input_tokens_seen": 8220976,
      "step": 135,
      "train_runtime": 1716.4662,
      "train_tokens_per_second": 4789.477
    },
    {
      "epoch": 0.017857142857142856,
      "grad_norm": 0.6480114718063784,
      "learning_rate": 9.998061127870738e-06,
      "loss": 0.3055912733078003,
      "num_input_tokens_seen": 8512112,
      "step": 140,
      "train_runtime": 1792.2391,
      "train_tokens_per_second": 4749.429
    },
    {
      "epoch": 0.018494897959183673,
      "grad_norm": 0.6852857295771357,
      "learning_rate": 9.997919141754448e-06,
      "loss": 0.35830059051513674,
      "num_input_tokens_seen": 8802400,
      "step": 145,
      "train_runtime": 1856.7624,
      "train_tokens_per_second": 4740.725
    },
    {
      "epoch": 0.01913265306122449,
      "grad_norm": 0.6199941796324793,
      "learning_rate": 9.997772139877245e-06,
      "loss": 0.3327900171279907,
      "num_input_tokens_seen": 9110000,
      "step": 150,
      "train_runtime": 1933.1163,
      "train_tokens_per_second": 4712.598
    },
    {
      "epoch": 0.019770408163265307,
      "grad_norm": 0.6153251276990073,
      "learning_rate": 9.997620122386658e-06,
      "loss": 0.33315272331237794,
      "num_input_tokens_seen": 9414456,
      "step": 155,
      "train_runtime": 1998.4404,
      "train_tokens_per_second": 4710.901
    },
    {
      "epoch": 0.02040816326530612,
      "grad_norm": 0.6646714549122126,
      "learning_rate": 9.997463089435242e-06,
      "loss": 0.32426862716674804,
      "num_input_tokens_seen": 9726616,
      "step": 160,
      "train_runtime": 2058.6526,
      "train_tokens_per_second": 4724.749
    },
    {
      "epoch": 0.021045918367346938,
      "grad_norm": 0.6534691087987078,
      "learning_rate": 9.997301041180597e-06,
      "loss": 0.342034649848938,
      "num_input_tokens_seen": 10030152,
      "step": 165,
      "train_runtime": 2128.5744,
      "train_tokens_per_second": 4712.145
    },
    {
      "epoch": 0.021683673469387755,
      "grad_norm": 0.6909544282185469,
      "learning_rate": 9.997133977785342e-06,
      "loss": 0.3254104137420654,
      "num_input_tokens_seen": 10335744,
      "step": 170,
      "train_runtime": 2180.5109,
      "train_tokens_per_second": 4740.056
    },
    {
      "epoch": 0.022321428571428572,
      "grad_norm": 0.6415788664349892,
      "learning_rate": 9.996961899417145e-06,
      "loss": 0.35123443603515625,
      "num_input_tokens_seen": 10643144,
      "step": 175,
      "train_runtime": 2249.0893,
      "train_tokens_per_second": 4732.202
    },
    {
      "epoch": 0.02295918367346939,
      "grad_norm": 0.6318044218192264,
      "learning_rate": 9.996784806248694e-06,
      "loss": 0.3360785961151123,
      "num_input_tokens_seen": 10946704,
      "step": 180,
      "train_runtime": 2314.4156,
      "train_tokens_per_second": 4729.792
    },
    {
      "epoch": 0.023596938775510203,
      "grad_norm": 0.6258613065343058,
      "learning_rate": 9.996602698457716e-06,
      "loss": 0.3184788465499878,
      "num_input_tokens_seen": 11253984,
      "step": 185,
      "train_runtime": 2390.5895,
      "train_tokens_per_second": 4707.619
    },
    {
      "epoch": 0.02423469387755102,
      "grad_norm": 0.6236546663558538,
      "learning_rate": 9.996415576226967e-06,
      "loss": 0.3205930233001709,
      "num_input_tokens_seen": 11557352,
      "step": 190,
      "train_runtime": 2455.2743,
      "train_tokens_per_second": 4707.153
    },
    {
      "epoch": 0.024872448979591837,
      "grad_norm": 0.6211891202484879,
      "learning_rate": 9.996223439744238e-06,
      "loss": 0.2992743492126465,
      "num_input_tokens_seen": 11867464,
      "step": 195,
      "train_runtime": 2522.985,
      "train_tokens_per_second": 4703.739
    },
    {
      "epoch": 0.025510204081632654,
      "grad_norm": 0.6677730693527612,
      "learning_rate": 9.996026289202352e-06,
      "loss": 0.3238770246505737,
      "num_input_tokens_seen": 12178464,
      "step": 200,
      "train_runtime": 2579.1454,
      "train_tokens_per_second": 4721.899
    },
    {
      "epoch": 0.02614795918367347,
      "grad_norm": 0.5823244571655314,
      "learning_rate": 9.995824124799164e-06,
      "loss": 0.36721694469451904,
      "num_input_tokens_seen": 12485272,
      "step": 205,
      "train_runtime": 2642.3033,
      "train_tokens_per_second": 4725.147
    },
    {
      "epoch": 0.026785714285714284,
      "grad_norm": 0.5931239174350026,
      "learning_rate": 9.995616946737558e-06,
      "loss": 0.3346077919006348,
      "num_input_tokens_seen": 12788440,
      "step": 210,
      "train_runtime": 2705.3052,
      "train_tokens_per_second": 4727.171
    },
    {
      "epoch": 0.0274234693877551,
      "grad_norm": 0.6489082273588189,
      "learning_rate": 9.995404755225454e-06,
      "loss": 0.3342667818069458,
      "num_input_tokens_seen": 13084544,
      "step": 215,
      "train_runtime": 2783.4989,
      "train_tokens_per_second": 4700.754
    },
    {
      "epoch": 0.02806122448979592,
      "grad_norm": 0.5670967952124059,
      "learning_rate": 9.995187550475799e-06,
      "loss": 0.30646519660949706,
      "num_input_tokens_seen": 13382224,
      "step": 220,
      "train_runtime": 2846.279,
      "train_tokens_per_second": 4701.656
    },
    {
      "epoch": 0.028698979591836735,
      "grad_norm": 0.6528085049465088,
      "learning_rate": 9.994965332706574e-06,
      "loss": 0.30767092704772947,
      "num_input_tokens_seen": 13672600,
      "step": 225,
      "train_runtime": 2913.7946,
      "train_tokens_per_second": 4692.369
    },
    {
      "epoch": 0.029336734693877552,
      "grad_norm": 0.6560986976249802,
      "learning_rate": 9.994738102140791e-06,
      "loss": 0.34403650760650634,
      "num_input_tokens_seen": 13985608,
      "step": 230,
      "train_runtime": 2982.5514,
      "train_tokens_per_second": 4689.142
    },
    {
      "epoch": 0.029974489795918366,
      "grad_norm": 0.655839503346995,
      "learning_rate": 9.994505859006489e-06,
      "loss": 0.32221179008483886,
      "num_input_tokens_seen": 14272056,
      "step": 235,
      "train_runtime": 3055.4689,
      "train_tokens_per_second": 4670.987
    },
    {
      "epoch": 0.030612244897959183,
      "grad_norm": 0.6214811118379946,
      "learning_rate": 9.994268603536743e-06,
      "loss": 0.3470602035522461,
      "num_input_tokens_seen": 14577080,
      "step": 240,
      "train_runtime": 3111.7544,
      "train_tokens_per_second": 4684.521
    },
    {
      "epoch": 0.03125,
      "grad_norm": 0.5494146630666381,
      "learning_rate": 9.994026335969654e-06,
      "loss": 0.32835218906402586,
      "num_input_tokens_seen": 14874720,
      "step": 245,
      "train_runtime": 3184.6531,
      "train_tokens_per_second": 4670.751
    },
    {
      "epoch": 0.03188775510204082,
      "grad_norm": 0.6047196528111599,
      "learning_rate": 9.993779056548357e-06,
      "loss": 0.32953906059265137,
      "num_input_tokens_seen": 15175768,
      "step": 250,
      "train_runtime": 3247.5928,
      "train_tokens_per_second": 4672.928
    },
    {
      "epoch": 0.032525510204081634,
      "grad_norm": 0.6743699455037632,
      "learning_rate": 9.99352676552101e-06,
      "loss": 0.33337490558624266,
      "num_input_tokens_seen": 15469696,
      "step": 255,
      "train_runtime": 3304.0945,
      "train_tokens_per_second": 4681.978
    },
    {
      "epoch": 0.03316326530612245,
      "grad_norm": 0.5933691290415959,
      "learning_rate": 9.993269463140805e-06,
      "loss": 0.3574438333511353,
      "num_input_tokens_seen": 15779624,
      "step": 260,
      "train_runtime": 3361.674,
      "train_tokens_per_second": 4693.978
    },
    {
      "epoch": 0.03380102040816327,
      "grad_norm": 0.6327740999108824,
      "learning_rate": 9.993007149665968e-06,
      "loss": 0.33641748428344725,
      "num_input_tokens_seen": 16077880,
      "step": 265,
      "train_runtime": 3431.2529,
      "train_tokens_per_second": 4685.717
    },
    {
      "epoch": 0.03443877551020408,
      "grad_norm": 0.5737650225224535,
      "learning_rate": 9.992739825359743e-06,
      "loss": 0.3066434383392334,
      "num_input_tokens_seen": 16374368,
      "step": 270,
      "train_runtime": 3504.5078,
      "train_tokens_per_second": 4672.373
    },
    {
      "epoch": 0.035076530612244895,
      "grad_norm": 0.6111088036267571,
      "learning_rate": 9.992467490490411e-06,
      "loss": 0.3326112747192383,
      "num_input_tokens_seen": 16685616,
      "step": 275,
      "train_runtime": 3574.2741,
      "train_tokens_per_second": 4668.253
    },
    {
      "epoch": 0.03571428571428571,
      "grad_norm": 0.6361151501047907,
      "learning_rate": 9.99219014533128e-06,
      "loss": 0.34957115650177,
      "num_input_tokens_seen": 17007744,
      "step": 280,
      "train_runtime": 3623.1821,
      "train_tokens_per_second": 4694.145
    },
    {
      "epoch": 0.03635204081632653,
      "grad_norm": 0.6791204972403216,
      "learning_rate": 9.991907790160686e-06,
      "loss": 0.33728978633880613,
      "num_input_tokens_seen": 17322144,
      "step": 285,
      "train_runtime": 3695.6144,
      "train_tokens_per_second": 4687.216
    },
    {
      "epoch": 0.036989795918367346,
      "grad_norm": 0.6933069106222348,
      "learning_rate": 9.991620425261987e-06,
      "loss": 0.33463027477264407,
      "num_input_tokens_seen": 17633104,
      "step": 290,
      "train_runtime": 3762.1203,
      "train_tokens_per_second": 4687.012
    },
    {
      "epoch": 0.03762755102040816,
      "grad_norm": 0.6497657790060584,
      "learning_rate": 9.99132805092358e-06,
      "loss": 0.33530759811401367,
      "num_input_tokens_seen": 17934608,
      "step": 295,
      "train_runtime": 3820.0982,
      "train_tokens_per_second": 4694.803
    },
    {
      "epoch": 0.03826530612244898,
      "grad_norm": 0.6882412546792613,
      "learning_rate": 9.99103066743888e-06,
      "loss": 0.36384072303771975,
      "num_input_tokens_seen": 18229648,
      "step": 300,
      "train_runtime": 3891.6162,
      "train_tokens_per_second": 4684.339
    },
    {
      "epoch": 0.0389030612244898,
      "grad_norm": 0.6601675334042054,
      "learning_rate": 9.990728275106332e-06,
      "loss": 0.32778329849243165,
      "num_input_tokens_seen": 18535656,
      "step": 305,
      "train_runtime": 3959.4362,
      "train_tokens_per_second": 4681.388
    },
    {
      "epoch": 0.039540816326530615,
      "grad_norm": 0.646885641152173,
      "learning_rate": 9.990420874229408e-06,
      "loss": 0.3228672981262207,
      "num_input_tokens_seen": 18843904,
      "step": 310,
      "train_runtime": 4026.317,
      "train_tokens_per_second": 4680.184
    },
    {
      "epoch": 0.04017857142857143,
      "grad_norm": 0.5931343448102652,
      "learning_rate": 9.990108465116606e-06,
      "loss": 0.32550835609436035,
      "num_input_tokens_seen": 19145520,
      "step": 315,
      "train_runtime": 4100.4082,
      "train_tokens_per_second": 4669.174
    },
    {
      "epoch": 0.04081632653061224,
      "grad_norm": 0.6677742695692401,
      "learning_rate": 9.989791048081452e-06,
      "loss": 0.3454264163970947,
      "num_input_tokens_seen": 19447640,
      "step": 320,
      "train_runtime": 4165.5338,
      "train_tokens_per_second": 4668.703
    },
    {
      "epoch": 0.04145408163265306,
      "grad_norm": 0.6533746396017394,
      "learning_rate": 9.989468623442494e-06,
      "loss": 0.36098618507385255,
      "num_input_tokens_seen": 19749576,
      "step": 325,
      "train_runtime": 4228.9644,
      "train_tokens_per_second": 4670.074
    },
    {
      "epoch": 0.042091836734693876,
      "grad_norm": 0.668231205428044,
      "learning_rate": 9.989141191523309e-06,
      "loss": 0.3276120662689209,
      "num_input_tokens_seen": 20035464,
      "step": 330,
      "train_runtime": 4285.975,
      "train_tokens_per_second": 4674.657
    },
    {
      "epoch": 0.04272959183673469,
      "grad_norm": 0.7191606014130846,
      "learning_rate": 9.988808752652498e-06,
      "loss": 0.3200536251068115,
      "num_input_tokens_seen": 20327480,
      "step": 335,
      "train_runtime": 4357.5244,
      "train_tokens_per_second": 4664.915
    },
    {
      "epoch": 0.04336734693877551,
      "grad_norm": 0.6495192286239109,
      "learning_rate": 9.988471307163684e-06,
      "loss": 0.30787248611450196,
      "num_input_tokens_seen": 20620464,
      "step": 340,
      "train_runtime": 4419.6411,
      "train_tokens_per_second": 4665.642
    },
    {
      "epoch": 0.04400510204081633,
      "grad_norm": 0.6033660537585547,
      "learning_rate": 9.988128855395522e-06,
      "loss": 0.3384589195251465,
      "num_input_tokens_seen": 20928144,
      "step": 345,
      "train_runtime": 4486.7509,
      "train_tokens_per_second": 4664.432
    },
    {
      "epoch": 0.044642857142857144,
      "grad_norm": 0.6270276756841308,
      "learning_rate": 9.987781397691683e-06,
      "loss": 0.3512593746185303,
      "num_input_tokens_seen": 21241216,
      "step": 350,
      "train_runtime": 4555.1403,
      "train_tokens_per_second": 4663.131
    },
    {
      "epoch": 0.04528061224489796,
      "grad_norm": 0.6379085152276468,
      "learning_rate": 9.987428934400864e-06,
      "loss": 0.33603107929229736,
      "num_input_tokens_seen": 21542400,
      "step": 355,
      "train_runtime": 4629.5182,
      "train_tokens_per_second": 4653.27
    },
    {
      "epoch": 0.04591836734693878,
      "grad_norm": 0.6445613765371325,
      "learning_rate": 9.98707146587679e-06,
      "loss": 0.3387459754943848,
      "num_input_tokens_seen": 21842576,
      "step": 360,
      "train_runtime": 4697.6725,
      "train_tokens_per_second": 4649.659
    },
    {
      "epoch": 0.046556122448979595,
      "grad_norm": 0.6049239039573681,
      "learning_rate": 9.986708992478202e-06,
      "loss": 0.3367520809173584,
      "num_input_tokens_seen": 22156256,
      "step": 365,
      "train_runtime": 4755.0115,
      "train_tokens_per_second": 4659.559
    },
    {
      "epoch": 0.047193877551020405,
      "grad_norm": 0.6640689638548932,
      "learning_rate": 9.98634151456887e-06,
      "loss": 0.3515306949615479,
      "num_input_tokens_seen": 22468688,
      "step": 370,
      "train_runtime": 4821.2083,
      "train_tokens_per_second": 4660.385
    },
    {
      "epoch": 0.04783163265306122,
      "grad_norm": 0.5714771338985628,
      "learning_rate": 9.985969032517582e-06,
      "loss": 0.3335010528564453,
      "num_input_tokens_seen": 22779688,
      "step": 375,
      "train_runtime": 4888.5125,
      "train_tokens_per_second": 4659.84
    },
    {
      "epoch": 0.04846938775510204,
      "grad_norm": 0.6375577903834478,
      "learning_rate": 9.985591546698151e-06,
      "loss": 0.33521523475646975,
      "num_input_tokens_seen": 23085304,
      "step": 380,
      "train_runtime": 4953.1284,
      "train_tokens_per_second": 4660.752
    },
    {
      "epoch": 0.049107142857142856,
      "grad_norm": 0.6750011275254378,
      "learning_rate": 9.98520905748941e-06,
      "loss": 0.34557697772979734,
      "num_input_tokens_seen": 23388632,
      "step": 385,
      "train_runtime": 5012.5547,
      "train_tokens_per_second": 4666.01
    },
    {
      "epoch": 0.04974489795918367,
      "grad_norm": 0.6307576116085186,
      "learning_rate": 9.984821565275214e-06,
      "loss": 0.31669833660125735,
      "num_input_tokens_seen": 23702200,
      "step": 390,
      "train_runtime": 5064.9324,
      "train_tokens_per_second": 4679.668
    },
    {
      "epoch": 0.05038265306122449,
      "grad_norm": 0.7090277542261124,
      "learning_rate": 9.984429070444437e-06,
      "loss": 0.3379413843154907,
      "num_input_tokens_seen": 23994432,
      "step": 395,
      "train_runtime": 5138.7592,
      "train_tokens_per_second": 4669.305
    },
    {
      "epoch": 0.05102040816326531,
      "grad_norm": 0.6241354831363873,
      "learning_rate": 9.984031573390977e-06,
      "loss": 0.33236329555511473,
      "num_input_tokens_seen": 24295472,
      "step": 400,
      "train_runtime": 5199.699,
      "train_tokens_per_second": 4672.477
    },
    {
      "epoch": 0.051658163265306124,
      "grad_norm": 0.6334452015043983,
      "learning_rate": 9.983629074513749e-06,
      "loss": 0.361618185043335,
      "num_input_tokens_seen": 24605608,
      "step": 405,
      "train_runtime": 5279.4591,
      "train_tokens_per_second": 4660.63
    },
    {
      "epoch": 0.05229591836734694,
      "grad_norm": 0.6075619104856002,
      "learning_rate": 9.983221574216688e-06,
      "loss": 0.33792781829833984,
      "num_input_tokens_seen": 24917816,
      "step": 410,
      "train_runtime": 5356.0183,
      "train_tokens_per_second": 4652.302
    },
    {
      "epoch": 0.05293367346938776,
      "grad_norm": 0.5927953294938682,
      "learning_rate": 9.98280907290875e-06,
      "loss": 0.34446349143981936,
      "num_input_tokens_seen": 25215896,
      "step": 415,
      "train_runtime": 5422.7067,
      "train_tokens_per_second": 4650.057
    },
    {
      "epoch": 0.05357142857142857,
      "grad_norm": 0.6349671362506129,
      "learning_rate": 9.98239157100391e-06,
      "loss": 0.31869771480560305,
      "num_input_tokens_seen": 25511528,
      "step": 420,
      "train_runtime": 5486.2099,
      "train_tokens_per_second": 4650.119
    },
    {
      "epoch": 0.054209183673469385,
      "grad_norm": 0.6363245190083693,
      "learning_rate": 9.981969068921157e-06,
      "loss": 0.32179903984069824,
      "num_input_tokens_seen": 25805496,
      "step": 425,
      "train_runtime": 5562.0273,
      "train_tokens_per_second": 4639.585
    },
    {
      "epoch": 0.0548469387755102,
      "grad_norm": 0.5953777633043515,
      "learning_rate": 9.981541567084506e-06,
      "loss": 0.32957944869995115,
      "num_input_tokens_seen": 26108872,
      "step": 430,
      "train_runtime": 5614.2716,
      "train_tokens_per_second": 4650.447
    },
    {
      "epoch": 0.05548469387755102,
      "grad_norm": 0.55458880364536,
      "learning_rate": 9.98110906592298e-06,
      "loss": 0.3697628974914551,
      "num_input_tokens_seen": 26433456,
      "step": 435,
      "train_runtime": 5675.2963,
      "train_tokens_per_second": 4657.635
    },
    {
      "epoch": 0.05612244897959184,
      "grad_norm": 0.6582635660637224,
      "learning_rate": 9.980671565870628e-06,
      "loss": 0.35071864128112795,
      "num_input_tokens_seen": 26732128,
      "step": 440,
      "train_runtime": 5750.8954,
      "train_tokens_per_second": 4648.342
    },
    {
      "epoch": 0.056760204081632654,
      "grad_norm": 0.5742208718122617,
      "learning_rate": 9.98022906736651e-06,
      "loss": 0.3407757759094238,
      "num_input_tokens_seen": 27028024,
      "step": 445,
      "train_runtime": 5819.2972,
      "train_tokens_per_second": 4644.551
    },
    {
      "epoch": 0.05739795918367347,
      "grad_norm": 0.5761437244964501,
      "learning_rate": 9.979781570854704e-06,
      "loss": 0.34862961769104006,
      "num_input_tokens_seen": 27341456,
      "step": 450,
      "train_runtime": 5879.7817,
      "train_tokens_per_second": 4650.08
    },
    {
      "epoch": 0.05803571428571429,
      "grad_norm": 0.5981064953687432,
      "learning_rate": 9.979329076784305e-06,
      "loss": 0.361210036277771,
      "num_input_tokens_seen": 27637416,
      "step": 455,
      "train_runtime": 5947.1589,
      "train_tokens_per_second": 4647.163
    },
    {
      "epoch": 0.058673469387755105,
      "grad_norm": 0.649912764397423,
      "learning_rate": 9.978871585609422e-06,
      "loss": 0.33243255615234374,
      "num_input_tokens_seen": 27936360,
      "step": 460,
      "train_runtime": 6019.5129,
      "train_tokens_per_second": 4640.967
    },
    {
      "epoch": 0.05931122448979592,
      "grad_norm": 0.643413112020552,
      "learning_rate": 9.978409097789178e-06,
      "loss": 0.3394655704498291,
      "num_input_tokens_seen": 28246160,
      "step": 465,
      "train_runtime": 6070.7082,
      "train_tokens_per_second": 4652.861
    },
    {
      "epoch": 0.05994897959183673,
      "grad_norm": 0.6137203637301495,
      "learning_rate": 9.977941613787714e-06,
      "loss": 0.3474451541900635,
      "num_input_tokens_seen": 28549672,
      "step": 470,
      "train_runtime": 6144.1843,
      "train_tokens_per_second": 4646.617
    },
    {
      "epoch": 0.06058673469387755,
      "grad_norm": 0.6349170869369642,
      "learning_rate": 9.977469134074181e-06,
      "loss": 0.3277826070785522,
      "num_input_tokens_seen": 28850928,
      "step": 475,
      "train_runtime": 6217.4577,
      "train_tokens_per_second": 4640.31
    },
    {
      "epoch": 0.061224489795918366,
      "grad_norm": 0.7111752400586683,
      "learning_rate": 9.976991659122748e-06,
      "loss": 0.3291780471801758,
      "num_input_tokens_seen": 29155440,
      "step": 480,
      "train_runtime": 6284.6269,
      "train_tokens_per_second": 4639.168
    },
    {
      "epoch": 0.06186224489795918,
      "grad_norm": 0.5912100762259809,
      "learning_rate": 9.976509189412591e-06,
      "loss": 0.3117806434631348,
      "num_input_tokens_seen": 29454808,
      "step": 485,
      "train_runtime": 6344.9385,
      "train_tokens_per_second": 4642.253
    },
    {
      "epoch": 0.0625,
      "grad_norm": 0.593222482776597,
      "learning_rate": 9.976021725427905e-06,
      "loss": 0.3248849153518677,
      "num_input_tokens_seen": 29753184,
      "step": 490,
      "train_runtime": 6414.4261,
      "train_tokens_per_second": 4638.48
    },
    {
      "epoch": 0.06313775510204081,
      "grad_norm": 0.5757997254457702,
      "learning_rate": 9.975529267657892e-06,
      "loss": 0.3143364191055298,
      "num_input_tokens_seen": 30053816,
      "step": 495,
      "train_runtime": 6487.1309,
      "train_tokens_per_second": 4632.836
    },
    {
      "epoch": 0.06377551020408163,
      "grad_norm": 0.6593953309542427,
      "learning_rate": 9.97503181659677e-06,
      "loss": 0.3483597755432129,
      "num_input_tokens_seen": 30370832,
      "step": 500,
      "train_runtime": 6537.8396,
      "train_tokens_per_second": 4645.393
    },
    {
      "epoch": 0.06441326530612244,
      "grad_norm": 0.6286079628698305,
      "learning_rate": 9.974529372743762e-06,
      "loss": 0.3520472288131714,
      "num_input_tokens_seen": 30677376,
      "step": 505,
      "train_runtime": 6590.1231,
      "train_tokens_per_second": 4655.054
    },
    {
      "epoch": 0.06505102040816327,
      "grad_norm": 0.6742459446609242,
      "learning_rate": 9.97402193660311e-06,
      "loss": 0.32537987232208254,
      "num_input_tokens_seen": 30984552,
      "step": 510,
      "train_runtime": 6655.3612,
      "train_tokens_per_second": 4655.578
    },
    {
      "epoch": 0.06568877551020408,
      "grad_norm": 0.6515613566569659,
      "learning_rate": 9.973509508684061e-06,
      "loss": 0.3295437812805176,
      "num_input_tokens_seen": 31290592,
      "step": 515,
      "train_runtime": 6721.7448,
      "train_tokens_per_second": 4655.129
    },
    {
      "epoch": 0.0663265306122449,
      "grad_norm": 0.6002794411888607,
      "learning_rate": 9.972992089500868e-06,
      "loss": 0.3507524013519287,
      "num_input_tokens_seen": 31581920,
      "step": 520,
      "train_runtime": 6792.6906,
      "train_tokens_per_second": 4649.398
    },
    {
      "epoch": 0.06696428571428571,
      "grad_norm": 0.6998894513007168,
      "learning_rate": 9.972469679572802e-06,
      "loss": 0.33227179050445554,
      "num_input_tokens_seen": 31883032,
      "step": 525,
      "train_runtime": 6849.6859,
      "train_tokens_per_second": 4654.671
    },
    {
      "epoch": 0.06760204081632654,
      "grad_norm": 0.651718521446083,
      "learning_rate": 9.971942279424136e-06,
      "loss": 0.3360892772674561,
      "num_input_tokens_seen": 32197784,
      "step": 530,
      "train_runtime": 6908.1131,
      "train_tokens_per_second": 4660.865
    },
    {
      "epoch": 0.06823979591836735,
      "grad_norm": 0.6460825483711362,
      "learning_rate": 9.971409889584151e-06,
      "loss": 0.34338998794555664,
      "num_input_tokens_seen": 32504160,
      "step": 535,
      "train_runtime": 6977.8177,
      "train_tokens_per_second": 4658.213
    },
    {
      "epoch": 0.06887755102040816,
      "grad_norm": 0.6497177932901091,
      "learning_rate": 9.970872510587142e-06,
      "loss": 0.34417474269866943,
      "num_input_tokens_seen": 32804584,
      "step": 540,
      "train_runtime": 7042.265,
      "train_tokens_per_second": 4658.243
    },
    {
      "epoch": 0.06951530612244898,
      "grad_norm": 0.6393545300470731,
      "learning_rate": 9.970330142972403e-06,
      "loss": 0.35027050971984863,
      "num_input_tokens_seen": 33113264,
      "step": 545,
      "train_runtime": 7118.1472,
      "train_tokens_per_second": 4651.95
    },
    {
      "epoch": 0.07015306122448979,
      "grad_norm": 0.6698249362905612,
      "learning_rate": 9.969782787284237e-06,
      "loss": 0.3356395483016968,
      "num_input_tokens_seen": 33425640,
      "step": 550,
      "train_runtime": 7179.3723,
      "train_tokens_per_second": 4655.789
    },
    {
      "epoch": 0.07079081632653061,
      "grad_norm": 0.6702548205068602,
      "learning_rate": 9.969230444071956e-06,
      "loss": 0.3323835849761963,
      "num_input_tokens_seen": 33730696,
      "step": 555,
      "train_runtime": 7237.5872,
      "train_tokens_per_second": 4660.489
    },
    {
      "epoch": 0.07142857142857142,
      "grad_norm": 0.6123175171096529,
      "learning_rate": 9.968673113889874e-06,
      "loss": 0.34821453094482424,
      "num_input_tokens_seen": 34035864,
      "step": 560,
      "train_runtime": 7294.7884,
      "train_tokens_per_second": 4665.778
    },
    {
      "epoch": 0.07206632653061225,
      "grad_norm": 0.6720244851960792,
      "learning_rate": 9.968110797297312e-06,
      "loss": 0.3442867279052734,
      "num_input_tokens_seen": 34341288,
      "step": 565,
      "train_runtime": 7365.9536,
      "train_tokens_per_second": 4662.165
    },
    {
      "epoch": 0.07270408163265306,
      "grad_norm": 0.6160081517048929,
      "learning_rate": 9.967543494858591e-06,
      "loss": 0.35814208984375,
      "num_input_tokens_seen": 34651224,
      "step": 570,
      "train_runtime": 7433.406,
      "train_tokens_per_second": 4661.554
    },
    {
      "epoch": 0.07334183673469388,
      "grad_norm": 0.6217454360955795,
      "learning_rate": 9.966971207143041e-06,
      "loss": 0.34238247871398925,
      "num_input_tokens_seen": 34960960,
      "step": 575,
      "train_runtime": 7494.761,
      "train_tokens_per_second": 4664.72
    },
    {
      "epoch": 0.07397959183673469,
      "grad_norm": 1.2904884702226334,
      "learning_rate": 9.966393934724992e-06,
      "loss": 0.3418263912200928,
      "num_input_tokens_seen": 35271552,
      "step": 580,
      "train_runtime": 7551.6917,
      "train_tokens_per_second": 4670.682
    },
    {
      "epoch": 0.07461734693877552,
      "grad_norm": 0.6324057925636936,
      "learning_rate": 9.965811678183777e-06,
      "loss": 0.330104398727417,
      "num_input_tokens_seen": 35581248,
      "step": 585,
      "train_runtime": 7620.6631,
      "train_tokens_per_second": 4669.049
    },
    {
      "epoch": 0.07525510204081633,
      "grad_norm": 0.646585573794272,
      "learning_rate": 9.965224438103733e-06,
      "loss": 0.34459099769592283,
      "num_input_tokens_seen": 35896192,
      "step": 590,
      "train_runtime": 7684.1699,
      "train_tokens_per_second": 4671.447
    },
    {
      "epoch": 0.07589285714285714,
      "grad_norm": 0.6258608008360923,
      "learning_rate": 9.964632215074193e-06,
      "loss": 0.3336080551147461,
      "num_input_tokens_seen": 36209472,
      "step": 595,
      "train_runtime": 7752.3812,
      "train_tokens_per_second": 4670.755
    },
    {
      "epoch": 0.07653061224489796,
      "grad_norm": 0.5619516673673918,
      "learning_rate": 9.964035009689498e-06,
      "loss": 0.33786745071411134,
      "num_input_tokens_seen": 36513008,
      "step": 600,
      "train_runtime": 7813.4726,
      "train_tokens_per_second": 4673.083
    },
    {
      "epoch": 0.07716836734693877,
      "grad_norm": 0.6998511212309851,
      "learning_rate": 9.963432822548982e-06,
      "loss": 0.32400827407836913,
      "num_input_tokens_seen": 36799952,
      "step": 605,
      "train_runtime": 7869.9921,
      "train_tokens_per_second": 4675.983
    },
    {
      "epoch": 0.0778061224489796,
      "grad_norm": 0.6340485994796795,
      "learning_rate": 9.962825654256984e-06,
      "loss": 0.33402822017669676,
      "num_input_tokens_seen": 37101008,
      "step": 610,
      "train_runtime": 7926.1372,
      "train_tokens_per_second": 4680.844
    },
    {
      "epoch": 0.0784438775510204,
      "grad_norm": 0.6514619233256685,
      "learning_rate": 9.962213505422838e-06,
      "loss": 0.3186351776123047,
      "num_input_tokens_seen": 37406416,
      "step": 615,
      "train_runtime": 8002.9603,
      "train_tokens_per_second": 4674.072
    },
    {
      "epoch": 0.07908163265306123,
      "grad_norm": 0.6353041852541652,
      "learning_rate": 9.96159637666088e-06,
      "loss": 0.3287219047546387,
      "num_input_tokens_seen": 37700984,
      "step": 620,
      "train_runtime": 8071.9407,
      "train_tokens_per_second": 4670.622
    },
    {
      "epoch": 0.07971938775510204,
      "grad_norm": 0.7428472192357477,
      "learning_rate": 9.96097426859044e-06,
      "loss": 0.309468412399292,
      "num_input_tokens_seen": 37974384,
      "step": 625,
      "train_runtime": 8140.39,
      "train_tokens_per_second": 4664.934
    },
    {
      "epoch": 0.08035714285714286,
      "grad_norm": 0.563023413948935,
      "learning_rate": 9.960347181835848e-06,
      "loss": 0.31683375835418703,
      "num_input_tokens_seen": 38282248,
      "step": 630,
      "train_runtime": 8205.1443,
      "train_tokens_per_second": 4665.64
    },
    {
      "epoch": 0.08099489795918367,
      "grad_norm": 0.6463783859310616,
      "learning_rate": 9.95971511702643e-06,
      "loss": 0.32001914978027346,
      "num_input_tokens_seen": 38580048,
      "step": 635,
      "train_runtime": 8269.7685,
      "train_tokens_per_second": 4665.191
    },
    {
      "epoch": 0.08163265306122448,
      "grad_norm": 0.6863907881842608,
      "learning_rate": 9.959078074796505e-06,
      "loss": 0.3427865028381348,
      "num_input_tokens_seen": 38888080,
      "step": 640,
      "train_runtime": 8324.1349,
      "train_tokens_per_second": 4671.726
    },
    {
      "epoch": 0.08227040816326531,
      "grad_norm": 0.6166462935595397,
      "learning_rate": 9.958436055785391e-06,
      "loss": 0.3504422664642334,
      "num_input_tokens_seen": 39201064,
      "step": 645,
      "train_runtime": 8402.0196,
      "train_tokens_per_second": 4665.672
    },
    {
      "epoch": 0.08290816326530612,
      "grad_norm": 0.6129701573859874,
      "learning_rate": 9.9577890606374e-06,
      "loss": 0.3311141014099121,
      "num_input_tokens_seen": 39506496,
      "step": 650,
      "train_runtime": 8457.9418,
      "train_tokens_per_second": 4670.935
    },
    {
      "epoch": 0.08354591836734694,
      "grad_norm": 0.6576165531867053,
      "learning_rate": 9.957137090001834e-06,
      "loss": 0.34262380599975584,
      "num_input_tokens_seen": 39787552,
      "step": 655,
      "train_runtime": 8511.4495,
      "train_tokens_per_second": 4674.592
    },
    {
      "epoch": 0.08418367346938775,
      "grad_norm": 0.6597706109347632,
      "learning_rate": 9.956480144532992e-06,
      "loss": 0.3301884889602661,
      "num_input_tokens_seen": 40090352,
      "step": 660,
      "train_runtime": 8580.9114,
      "train_tokens_per_second": 4672.039
    },
    {
      "epoch": 0.08482142857142858,
      "grad_norm": 0.8245169872586329,
      "learning_rate": 9.955818224890166e-06,
      "loss": 0.3425352334976196,
      "num_input_tokens_seen": 40405072,
      "step": 665,
      "train_runtime": 8639.845,
      "train_tokens_per_second": 4676.597
    },
    {
      "epoch": 0.08545918367346939,
      "grad_norm": 0.62100333481103,
      "learning_rate": 9.955151331737638e-06,
      "loss": 0.3391813516616821,
      "num_input_tokens_seen": 40722368,
      "step": 670,
      "train_runtime": 8702.4887,
      "train_tokens_per_second": 4679.393
    },
    {
      "epoch": 0.08609693877551021,
      "grad_norm": 0.5885127232768026,
      "learning_rate": 9.954479465744681e-06,
      "loss": 0.32563443183898927,
      "num_input_tokens_seen": 41015040,
      "step": 675,
      "train_runtime": 8766.1622,
      "train_tokens_per_second": 4678.791
    },
    {
      "epoch": 0.08673469387755102,
      "grad_norm": 0.642935608039282,
      "learning_rate": 9.95380262758556e-06,
      "loss": 0.3731600522994995,
      "num_input_tokens_seen": 41315736,
      "step": 680,
      "train_runtime": 8828.3355,
      "train_tokens_per_second": 4679.901
    },
    {
      "epoch": 0.08737244897959184,
      "grad_norm": 0.6561126308399429,
      "learning_rate": 9.953120817939532e-06,
      "loss": 0.3367309093475342,
      "num_input_tokens_seen": 41621136,
      "step": 685,
      "train_runtime": 8895.075,
      "train_tokens_per_second": 4679.121
    },
    {
      "epoch": 0.08801020408163265,
      "grad_norm": 0.6485319029413061,
      "learning_rate": 9.952434037490835e-06,
      "loss": 0.3516701698303223,
      "num_input_tokens_seen": 41931728,
      "step": 690,
      "train_runtime": 8962.3701,
      "train_tokens_per_second": 4678.643
    },
    {
      "epoch": 0.08864795918367346,
      "grad_norm": 0.5933925996798751,
      "learning_rate": 9.951742286928706e-06,
      "loss": 0.32941617965698244,
      "num_input_tokens_seen": 42235328,
      "step": 695,
      "train_runtime": 9021.7432,
      "train_tokens_per_second": 4681.504
    },
    {
      "epoch": 0.08928571428571429,
      "grad_norm": 0.564562833510779,
      "learning_rate": 9.951045566947361e-06,
      "loss": 0.31849193572998047,
      "num_input_tokens_seen": 42539752,
      "step": 700,
      "train_runtime": 9082.1865,
      "train_tokens_per_second": 4683.867
    },
    {
      "epoch": 0.0899234693877551,
      "grad_norm": 0.617759203056016,
      "learning_rate": 9.950343878246011e-06,
      "loss": 0.3369591236114502,
      "num_input_tokens_seen": 42844680,
      "step": 705,
      "train_runtime": 9151.1138,
      "train_tokens_per_second": 4681.909
    },
    {
      "epoch": 0.09056122448979592,
      "grad_norm": 0.6718067036117873,
      "learning_rate": 9.949637221528847e-06,
      "loss": 0.3505253314971924,
      "num_input_tokens_seen": 43149992,
      "step": 710,
      "train_runtime": 9214.6197,
      "train_tokens_per_second": 4682.775
    },
    {
      "epoch": 0.09119897959183673,
      "grad_norm": 0.7213035397634782,
      "learning_rate": 9.94892559750505e-06,
      "loss": 0.34964261054992674,
      "num_input_tokens_seen": 43455200,
      "step": 715,
      "train_runtime": 9285.6942,
      "train_tokens_per_second": 4679.801
    },
    {
      "epoch": 0.09183673469387756,
      "grad_norm": 0.6616448628629455,
      "learning_rate": 9.948209006888782e-06,
      "loss": 0.34442615509033203,
      "num_input_tokens_seen": 43756952,
      "step": 720,
      "train_runtime": 9350.4408,
      "train_tokens_per_second": 4679.667
    },
    {
      "epoch": 0.09247448979591837,
      "grad_norm": 0.641629063439199,
      "learning_rate": 9.947487450399196e-06,
      "loss": 0.37545912265777587,
      "num_input_tokens_seen": 44071424,
      "step": 725,
      "train_runtime": 9400.2472,
      "train_tokens_per_second": 4688.326
    },
    {
      "epoch": 0.09311224489795919,
      "grad_norm": 0.6240625238331874,
      "learning_rate": 9.946760928760419e-06,
      "loss": 0.3321858882904053,
      "num_input_tokens_seen": 44374592,
      "step": 730,
      "train_runtime": 9467.356,
      "train_tokens_per_second": 4687.116
    },
    {
      "epoch": 0.09375,
      "grad_norm": 0.6322018195102153,
      "learning_rate": 9.946029442701572e-06,
      "loss": 0.32565453052520754,
      "num_input_tokens_seen": 44671440,
      "step": 735,
      "train_runtime": 9523.3977,
      "train_tokens_per_second": 4690.704
    },
    {
      "epoch": 0.09438775510204081,
      "grad_norm": 0.6605175435876213,
      "learning_rate": 9.945292992956747e-06,
      "loss": 0.3345797538757324,
      "num_input_tokens_seen": 44976088,
      "step": 740,
      "train_runtime": 9594.0464,
      "train_tokens_per_second": 4687.916
    },
    {
      "epoch": 0.09502551020408163,
      "grad_norm": 0.5972935827923552,
      "learning_rate": 9.944551580265025e-06,
      "loss": 0.34259605407714844,
      "num_input_tokens_seen": 45280296,
      "step": 745,
      "train_runtime": 9666.5333,
      "train_tokens_per_second": 4684.233
    },
    {
      "epoch": 0.09566326530612244,
      "grad_norm": 0.6512752099965131,
      "learning_rate": 9.943805205370467e-06,
      "loss": 0.35463581085205076,
      "num_input_tokens_seen": 45583616,
      "step": 750,
      "train_runtime": 9730.0762,
      "train_tokens_per_second": 4684.816
    },
    {
      "epoch": 0.09630102040816327,
      "grad_norm": 0.6564157072566779,
      "learning_rate": 9.94305386902211e-06,
      "loss": 0.3260321140289307,
      "num_input_tokens_seen": 45876952,
      "step": 755,
      "train_runtime": 9798.0296,
      "train_tokens_per_second": 4682.263
    },
    {
      "epoch": 0.09693877551020408,
      "grad_norm": 0.6346769074928412,
      "learning_rate": 9.942297571973973e-06,
      "loss": 0.33512492179870607,
      "num_input_tokens_seen": 46178848,
      "step": 760,
      "train_runtime": 9868.2039,
      "train_tokens_per_second": 4679.56
    },
    {
      "epoch": 0.0975765306122449,
      "grad_norm": 0.5622200359560194,
      "learning_rate": 9.941536314985054e-06,
      "loss": 0.32372226715087893,
      "num_input_tokens_seen": 46489520,
      "step": 765,
      "train_runtime": 9943.8403,
      "train_tokens_per_second": 4675.208
    },
    {
      "epoch": 0.09821428571428571,
      "grad_norm": 0.652691796843303,
      "learning_rate": 9.940770098819329e-06,
      "loss": 0.3503239631652832,
      "num_input_tokens_seen": 46802232,
      "step": 770,
      "train_runtime": 9993.5689,
      "train_tokens_per_second": 4683.235
    },
    {
      "epoch": 0.09885204081632654,
      "grad_norm": 0.6504801242347312,
      "learning_rate": 9.939998924245742e-06,
      "loss": 0.32001385688781736,
      "num_input_tokens_seen": 47109088,
      "step": 775,
      "train_runtime": 10055.9232,
      "train_tokens_per_second": 4684.71
    },
    {
      "epoch": 0.09948979591836735,
      "grad_norm": 0.6290107603055541,
      "learning_rate": 9.93922279203823e-06,
      "loss": 0.3270054578781128,
      "num_input_tokens_seen": 47396392,
      "step": 780,
      "train_runtime": 10115.628,
      "train_tokens_per_second": 4685.462
    },
    {
      "epoch": 0.10012755102040816,
      "grad_norm": 0.5984760115235302,
      "learning_rate": 9.938441702975689e-06,
      "loss": 0.33510923385620117,
      "num_input_tokens_seen": 47691224,
      "step": 785,
      "train_runtime": 10171.722,
      "train_tokens_per_second": 4688.609
    },
    {
      "epoch": 0.10076530612244898,
      "grad_norm": 0.5945590755184339,
      "learning_rate": 9.937655657842001e-06,
      "loss": 0.33897085189819337,
      "num_input_tokens_seen": 48000248,
      "step": 790,
      "train_runtime": 10244.0159,
      "train_tokens_per_second": 4685.687
    },
    {
      "epoch": 0.10140306122448979,
      "grad_norm": 0.6403831240042489,
      "learning_rate": 9.936864657426014e-06,
      "loss": 0.3564270496368408,
      "num_input_tokens_seen": 48316296,
      "step": 795,
      "train_runtime": 10293.9437,
      "train_tokens_per_second": 4693.662
    },
    {
      "epoch": 0.10204081632653061,
      "grad_norm": 0.6446524416085734,
      "learning_rate": 9.936068702521553e-06,
      "loss": 0.3363536834716797,
      "num_input_tokens_seen": 48619456,
      "step": 800,
      "train_runtime": 10357.3747,
      "train_tokens_per_second": 4694.187
    },
    {
      "epoch": 0.10267857142857142,
      "grad_norm": 0.5714798655252041,
      "learning_rate": 9.935267793927416e-06,
      "loss": 0.3507491111755371,
      "num_input_tokens_seen": 48924672,
      "step": 805,
      "train_runtime": 10423.3082,
      "train_tokens_per_second": 4693.776
    },
    {
      "epoch": 0.10331632653061225,
      "grad_norm": 0.5951054701936696,
      "learning_rate": 9.93446193244737e-06,
      "loss": 0.3112443447113037,
      "num_input_tokens_seen": 49225280,
      "step": 810,
      "train_runtime": 10484.5803,
      "train_tokens_per_second": 4695.017
    },
    {
      "epoch": 0.10395408163265306,
      "grad_norm": 0.5913459917729529,
      "learning_rate": 9.933651118890151e-06,
      "loss": 0.3627671718597412,
      "num_input_tokens_seen": 49527824,
      "step": 815,
      "train_runtime": 10553.6081,
      "train_tokens_per_second": 4692.975
    },
    {
      "epoch": 0.10459183673469388,
      "grad_norm": 0.6067586558019212,
      "learning_rate": 9.93283535406947e-06,
      "loss": 0.31316590309143066,
      "num_input_tokens_seen": 49818424,
      "step": 820,
      "train_runtime": 10609.2781,
      "train_tokens_per_second": 4695.741
    },
    {
      "epoch": 0.10522959183673469,
      "grad_norm": 0.5900034325851385,
      "learning_rate": 9.932014638804001e-06,
      "loss": 0.3390337467193604,
      "num_input_tokens_seen": 50120392,
      "step": 825,
      "train_runtime": 10679.1642,
      "train_tokens_per_second": 4693.288
    },
    {
      "epoch": 0.10586734693877552,
      "grad_norm": 0.6118873763765331,
      "learning_rate": 9.931188973917393e-06,
      "loss": 0.3528244972229004,
      "num_input_tokens_seen": 50437328,
      "step": 830,
      "train_runtime": 10749.3443,
      "train_tokens_per_second": 4692.131
    },
    {
      "epoch": 0.10650510204081633,
      "grad_norm": 0.6191303656527237,
      "learning_rate": 9.930358360238255e-06,
      "loss": 0.3476259708404541,
      "num_input_tokens_seen": 50739776,
      "step": 835,
      "train_runtime": 10821.0403,
      "train_tokens_per_second": 4688.992
    },
    {
      "epoch": 0.10714285714285714,
      "grad_norm": 0.9321027342684852,
      "learning_rate": 9.929522798600168e-06,
      "loss": 0.32042813301086426,
      "num_input_tokens_seen": 51054616,
      "step": 840,
      "train_runtime": 10882.3019,
      "train_tokens_per_second": 4691.527
    },
    {
      "epoch": 0.10778061224489796,
      "grad_norm": 0.612929250179954,
      "learning_rate": 9.928682289841674e-06,
      "loss": 0.33142428398132323,
      "num_input_tokens_seen": 51357944,
      "step": 845,
      "train_runtime": 10949.2079,
      "train_tokens_per_second": 4690.562
    },
    {
      "epoch": 0.10841836734693877,
      "grad_norm": 0.6443381175533973,
      "learning_rate": 9.927836834806284e-06,
      "loss": 0.35215234756469727,
      "num_input_tokens_seen": 51662968,
      "step": 850,
      "train_runtime": 11020.3408,
      "train_tokens_per_second": 4687.965
    },
    {
      "epoch": 0.1090561224489796,
      "grad_norm": 0.6980326547970849,
      "learning_rate": 9.926986434342471e-06,
      "loss": 0.3072350025177002,
      "num_input_tokens_seen": 51955408,
      "step": 855,
      "train_runtime": 11084.8998,
      "train_tokens_per_second": 4687.044
    },
    {
      "epoch": 0.1096938775510204,
      "grad_norm": 0.6265378803336521,
      "learning_rate": 9.926131089303672e-06,
      "loss": 0.33462932109832766,
      "num_input_tokens_seen": 52261056,
      "step": 860,
      "train_runtime": 11146.8728,
      "train_tokens_per_second": 4688.405
    },
    {
      "epoch": 0.11033163265306123,
      "grad_norm": 0.6485933375531521,
      "learning_rate": 9.925270800548285e-06,
      "loss": 0.32468724250793457,
      "num_input_tokens_seen": 52553456,
      "step": 865,
      "train_runtime": 11211.2365,
      "train_tokens_per_second": 4687.57
    },
    {
      "epoch": 0.11096938775510204,
      "grad_norm": 0.753232518571963,
      "learning_rate": 9.924405568939667e-06,
      "loss": 0.38157219886779786,
      "num_input_tokens_seen": 52859568,
      "step": 870,
      "train_runtime": 11268.6986,
      "train_tokens_per_second": 4690.832
    },
    {
      "epoch": 0.11160714285714286,
      "grad_norm": 0.6538272459043324,
      "learning_rate": 9.923535395346145e-06,
      "loss": 0.31001625061035154,
      "num_input_tokens_seen": 53163928,
      "step": 875,
      "train_runtime": 11336.7117,
      "train_tokens_per_second": 4689.537
    },
    {
      "epoch": 0.11224489795918367,
      "grad_norm": 0.5755487815405781,
      "learning_rate": 9.92266028064099e-06,
      "loss": 0.30125041007995607,
      "num_input_tokens_seen": 53474152,
      "step": 880,
      "train_runtime": 11403.7345,
      "train_tokens_per_second": 4689.179
    },
    {
      "epoch": 0.11288265306122448,
      "grad_norm": 0.679482039375145,
      "learning_rate": 9.921780225702448e-06,
      "loss": 0.32144479751586913,
      "num_input_tokens_seen": 53781448,
      "step": 885,
      "train_runtime": 11477.1682,
      "train_tokens_per_second": 4685.951
    },
    {
      "epoch": 0.11352040816326531,
      "grad_norm": 0.6019576388093533,
      "learning_rate": 9.920895231413714e-06,
      "loss": 0.3461263418197632,
      "num_input_tokens_seen": 54101144,
      "step": 890,
      "train_runtime": 11539.6978,
      "train_tokens_per_second": 4688.263
    },
    {
      "epoch": 0.11415816326530612,
      "grad_norm": 0.6273273601475747,
      "learning_rate": 9.92000529866294e-06,
      "loss": 0.3378733158111572,
      "num_input_tokens_seen": 54410120,
      "step": 895,
      "train_runtime": 11606.4535,
      "train_tokens_per_second": 4687.92
    },
    {
      "epoch": 0.11479591836734694,
      "grad_norm": 0.6373721839889133,
      "learning_rate": 9.919110428343235e-06,
      "loss": 0.3378178596496582,
      "num_input_tokens_seen": 54718832,
      "step": 900,
      "train_runtime": 11659.9144,
      "train_tokens_per_second": 4692.902
    },
    {
      "epoch": 0.11543367346938775,
      "grad_norm": 0.5372536134204118,
      "learning_rate": 9.918210621352667e-06,
      "loss": 0.3274667978286743,
      "num_input_tokens_seen": 55019664,
      "step": 905,
      "train_runtime": 11730.9735,
      "train_tokens_per_second": 4690.119
    },
    {
      "epoch": 0.11607142857142858,
      "grad_norm": 0.6244634917562096,
      "learning_rate": 9.917305878594251e-06,
      "loss": 0.35554866790771483,
      "num_input_tokens_seen": 55329784,
      "step": 910,
      "train_runtime": 11809.9727,
      "train_tokens_per_second": 4685.005
    },
    {
      "epoch": 0.11670918367346939,
      "grad_norm": 0.6503819855008445,
      "learning_rate": 9.916396200975966e-06,
      "loss": 0.3127890110015869,
      "num_input_tokens_seen": 55623544,
      "step": 915,
      "train_runtime": 11871.2639,
      "train_tokens_per_second": 4685.562
    },
    {
      "epoch": 0.11734693877551021,
      "grad_norm": 0.6252291437789995,
      "learning_rate": 9.91548158941073e-06,
      "loss": 0.33112030029296874,
      "num_input_tokens_seen": 55927480,
      "step": 920,
      "train_runtime": 11940.2567,
      "train_tokens_per_second": 4683.943
    },
    {
      "epoch": 0.11798469387755102,
      "grad_norm": 0.5367371258827875,
      "learning_rate": 9.914562044816424e-06,
      "loss": 0.3396120548248291,
      "num_input_tokens_seen": 56238632,
      "step": 925,
      "train_runtime": 12013.7771,
      "train_tokens_per_second": 4681.178
    },
    {
      "epoch": 0.11862244897959184,
      "grad_norm": 0.6196964334710047,
      "learning_rate": 9.913637568115873e-06,
      "loss": 0.34828312397003175,
      "num_input_tokens_seen": 56523656,
      "step": 930,
      "train_runtime": 12080.8434,
      "train_tokens_per_second": 4678.784
    },
    {
      "epoch": 0.11926020408163265,
      "grad_norm": 0.6757997370063455,
      "learning_rate": 9.912708160236854e-06,
      "loss": 0.3335379123687744,
      "num_input_tokens_seen": 56827640,
      "step": 935,
      "train_runtime": 12133.9846,
      "train_tokens_per_second": 4683.345
    },
    {
      "epoch": 0.11989795918367346,
      "grad_norm": 0.5945713330525164,
      "learning_rate": 9.911773822112094e-06,
      "loss": 0.3137048244476318,
      "num_input_tokens_seen": 57124504,
      "step": 940,
      "train_runtime": 12198.6931,
      "train_tokens_per_second": 4682.838
    },
    {
      "epoch": 0.12053571428571429,
      "grad_norm": 0.5437143823809053,
      "learning_rate": 9.910834554679266e-06,
      "loss": 0.3561995029449463,
      "num_input_tokens_seen": 57425768,
      "step": 945,
      "train_runtime": 12271.2348,
      "train_tokens_per_second": 4679.706
    },
    {
      "epoch": 0.1211734693877551,
      "grad_norm": 0.7265574889311718,
      "learning_rate": 9.909890358880991e-06,
      "loss": 0.35443639755249023,
      "num_input_tokens_seen": 57725000,
      "step": 950,
      "train_runtime": 12346.5447,
      "train_tokens_per_second": 4675.397
    },
    {
      "epoch": 0.12181122448979592,
      "grad_norm": 0.579079551444378,
      "learning_rate": 9.908941235664835e-06,
      "loss": 0.3275070428848267,
      "num_input_tokens_seen": 58031336,
      "step": 955,
      "train_runtime": 12403.8628,
      "train_tokens_per_second": 4678.489
    },
    {
      "epoch": 0.12244897959183673,
      "grad_norm": 0.6174899647909609,
      "learning_rate": 9.907987185983307e-06,
      "loss": 0.3348393440246582,
      "num_input_tokens_seen": 58330568,
      "step": 960,
      "train_runtime": 12479.2646,
      "train_tokens_per_second": 4674.199
    },
    {
      "epoch": 0.12308673469387756,
      "grad_norm": 0.5622889563305987,
      "learning_rate": 9.907028210793868e-06,
      "loss": 0.32388012409210204,
      "num_input_tokens_seen": 58623792,
      "step": 965,
      "train_runtime": 12550.4246,
      "train_tokens_per_second": 4671.06
    },
    {
      "epoch": 0.12372448979591837,
      "grad_norm": 0.5555181465350426,
      "learning_rate": 9.906064311058911e-06,
      "loss": 0.3385571002960205,
      "num_input_tokens_seen": 58940704,
      "step": 970,
      "train_runtime": 12619.0253,
      "train_tokens_per_second": 4670.781
    },
    {
      "epoch": 0.12436224489795919,
      "grad_norm": 0.6177073026704814,
      "learning_rate": 9.90509548774578e-06,
      "loss": 0.31595683097839355,
      "num_input_tokens_seen": 59253592,
      "step": 975,
      "train_runtime": 12676.2346,
      "train_tokens_per_second": 4674.384
    },
    {
      "epoch": 0.125,
      "grad_norm": 0.6474714592562275,
      "learning_rate": 9.904121741826757e-06,
      "loss": 0.31098217964172364,
      "num_input_tokens_seen": 59551672,
      "step": 980,
      "train_runtime": 12733.5429,
      "train_tokens_per_second": 4676.756
    },
    {
      "epoch": 0.1256377551020408,
      "grad_norm": 0.6164468862965365,
      "learning_rate": 9.903143074279061e-06,
      "loss": 0.3462122678756714,
      "num_input_tokens_seen": 59863016,
      "step": 985,
      "train_runtime": 12790.1496,
      "train_tokens_per_second": 4680.4
    },
    {
      "epoch": 0.12627551020408162,
      "grad_norm": 0.5735618832789965,
      "learning_rate": 9.902159486084857e-06,
      "loss": 0.3320111513137817,
      "num_input_tokens_seen": 60161872,
      "step": 990,
      "train_runtime": 12856.4346,
      "train_tokens_per_second": 4679.514
    },
    {
      "epoch": 0.12691326530612246,
      "grad_norm": 0.6992565730851574,
      "learning_rate": 9.90117097823124e-06,
      "loss": 0.3359437704086304,
      "num_input_tokens_seen": 60465968,
      "step": 995,
      "train_runtime": 12924.8758,
      "train_tokens_per_second": 4678.263
    },
    {
      "epoch": 0.12755102040816327,
      "grad_norm": 0.6217414292705282,
      "learning_rate": 9.900177551710252e-06,
      "loss": 0.3319170713424683,
      "num_input_tokens_seen": 60768400,
      "step": 1000,
      "train_runtime": 12994.399,
      "train_tokens_per_second": 4676.507
    },
    {
      "epoch": 0.12818877551020408,
      "grad_norm": 0.5924633670456982,
      "learning_rate": 9.899179207518863e-06,
      "loss": 0.3651275157928467,
      "num_input_tokens_seen": 61080280,
      "step": 1005,
      "train_runtime": 13058.3122,
      "train_tokens_per_second": 4677.502
    },
    {
      "epoch": 0.1288265306122449,
      "grad_norm": 0.5687355197613791,
      "learning_rate": 9.89817594665898e-06,
      "loss": 0.33693101406097414,
      "num_input_tokens_seen": 61398592,
      "step": 1010,
      "train_runtime": 13130.7397,
      "train_tokens_per_second": 4675.943
    },
    {
      "epoch": 0.12946428571428573,
      "grad_norm": 0.6466288372316616,
      "learning_rate": 9.897167770137449e-06,
      "loss": 0.32893297672271726,
      "num_input_tokens_seen": 61709984,
      "step": 1015,
      "train_runtime": 13193.1442,
      "train_tokens_per_second": 4677.428
    },
    {
      "epoch": 0.13010204081632654,
      "grad_norm": 0.6322209107485859,
      "learning_rate": 9.89615467896604e-06,
      "loss": 0.3443321943283081,
      "num_input_tokens_seen": 62022072,
      "step": 1020,
      "train_runtime": 13255.5256,
      "train_tokens_per_second": 4678.96
    },
    {
      "epoch": 0.13073979591836735,
      "grad_norm": 0.5974934839984513,
      "learning_rate": 9.895136674161466e-06,
      "loss": 0.31420562267303465,
      "num_input_tokens_seen": 62326000,
      "step": 1025,
      "train_runtime": 13321.4397,
      "train_tokens_per_second": 4678.623
    },
    {
      "epoch": 0.13137755102040816,
      "grad_norm": 0.6123161071863797,
      "learning_rate": 9.894113756745362e-06,
      "loss": 0.3463484287261963,
      "num_input_tokens_seen": 62637088,
      "step": 1030,
      "train_runtime": 13392.564,
      "train_tokens_per_second": 4677.005
    },
    {
      "epoch": 0.13201530612244897,
      "grad_norm": 0.6059368270099965,
      "learning_rate": 9.893085927744302e-06,
      "loss": 0.3330828189849854,
      "num_input_tokens_seen": 62939296,
      "step": 1035,
      "train_runtime": 13456.54,
      "train_tokens_per_second": 4677.227
    },
    {
      "epoch": 0.1326530612244898,
      "grad_norm": 0.5889600903987154,
      "learning_rate": 9.892053188189778e-06,
      "loss": 0.3249910831451416,
      "num_input_tokens_seen": 63241312,
      "step": 1040,
      "train_runtime": 13531.3937,
      "train_tokens_per_second": 4673.673
    },
    {
      "epoch": 0.13329081632653061,
      "grad_norm": 0.593460422531129,
      "learning_rate": 9.89101553911822e-06,
      "loss": 0.332883358001709,
      "num_input_tokens_seen": 63542296,
      "step": 1045,
      "train_runtime": 13600.5526,
      "train_tokens_per_second": 4672.038
    },
    {
      "epoch": 0.13392857142857142,
      "grad_norm": 0.6279699500325815,
      "learning_rate": 9.889972981570984e-06,
      "loss": 0.32372775077819826,
      "num_input_tokens_seen": 63841688,
      "step": 1050,
      "train_runtime": 13664.5375,
      "train_tokens_per_second": 4672.071
    },
    {
      "epoch": 0.13456632653061223,
      "grad_norm": 0.5991600083612287,
      "learning_rate": 9.888925516594344e-06,
      "loss": 0.33316926956176757,
      "num_input_tokens_seen": 64152120,
      "step": 1055,
      "train_runtime": 13732.2053,
      "train_tokens_per_second": 4671.655
    },
    {
      "epoch": 0.13520408163265307,
      "grad_norm": 0.599830369848699,
      "learning_rate": 9.887873145239505e-06,
      "loss": 0.34979519844055174,
      "num_input_tokens_seen": 64463160,
      "step": 1060,
      "train_runtime": 13799.2665,
      "train_tokens_per_second": 4671.492
    },
    {
      "epoch": 0.13584183673469388,
      "grad_norm": 0.636347589464702,
      "learning_rate": 9.886815868562596e-06,
      "loss": 0.3216874122619629,
      "num_input_tokens_seen": 64768464,
      "step": 1065,
      "train_runtime": 13860.3494,
      "train_tokens_per_second": 4672.932
    },
    {
      "epoch": 0.1364795918367347,
      "grad_norm": 0.5370893072761296,
      "learning_rate": 9.885753687624671e-06,
      "loss": 0.31194138526916504,
      "num_input_tokens_seen": 65065256,
      "step": 1070,
      "train_runtime": 13918.2591,
      "train_tokens_per_second": 4674.813
    },
    {
      "epoch": 0.1371173469387755,
      "grad_norm": 0.651574208195918,
      "learning_rate": 9.884686603491697e-06,
      "loss": 0.34079148769378664,
      "num_input_tokens_seen": 65374104,
      "step": 1075,
      "train_runtime": 13979.5939,
      "train_tokens_per_second": 4676.395
    },
    {
      "epoch": 0.1377551020408163,
      "grad_norm": 0.6508794125806138,
      "learning_rate": 9.883614617234574e-06,
      "loss": 0.33709142208099363,
      "num_input_tokens_seen": 65675264,
      "step": 1080,
      "train_runtime": 14047.0217,
      "train_tokens_per_second": 4675.387
    },
    {
      "epoch": 0.13839285714285715,
      "grad_norm": 0.5785407358443382,
      "learning_rate": 9.882537729929111e-06,
      "loss": 0.33071603775024416,
      "num_input_tokens_seen": 65976824,
      "step": 1085,
      "train_runtime": 14110.0194,
      "train_tokens_per_second": 4675.885
    },
    {
      "epoch": 0.13903061224489796,
      "grad_norm": 0.6086267377627738,
      "learning_rate": 9.88145594265604e-06,
      "loss": 0.33824639320373534,
      "num_input_tokens_seen": 66274184,
      "step": 1090,
      "train_runtime": 14157.1032,
      "train_tokens_per_second": 4681.338
    },
    {
      "epoch": 0.13966836734693877,
      "grad_norm": 0.7124070129387011,
      "learning_rate": 9.880369256501009e-06,
      "loss": 0.3228901863098145,
      "num_input_tokens_seen": 66556608,
      "step": 1095,
      "train_runtime": 14210.6632,
      "train_tokens_per_second": 4683.568
    },
    {
      "epoch": 0.14030612244897958,
      "grad_norm": 0.6202223061956437,
      "learning_rate": 9.879277672554586e-06,
      "loss": 0.3071782350540161,
      "num_input_tokens_seen": 66858872,
      "step": 1100,
      "train_runtime": 14268.9491,
      "train_tokens_per_second": 4685.62
    },
    {
      "epoch": 0.14094387755102042,
      "grad_norm": 0.6261694646491559,
      "learning_rate": 9.878181191912251e-06,
      "loss": 0.3458325624465942,
      "num_input_tokens_seen": 67163584,
      "step": 1105,
      "train_runtime": 14326.1045,
      "train_tokens_per_second": 4688.196
    },
    {
      "epoch": 0.14158163265306123,
      "grad_norm": 0.5500372706011527,
      "learning_rate": 9.877079815674397e-06,
      "loss": 0.3350841045379639,
      "num_input_tokens_seen": 67468600,
      "step": 1110,
      "train_runtime": 14393.6743,
      "train_tokens_per_second": 4687.379
    },
    {
      "epoch": 0.14221938775510204,
      "grad_norm": 0.5755528320511979,
      "learning_rate": 9.875973544946335e-06,
      "loss": 0.3193055152893066,
      "num_input_tokens_seen": 67777000,
      "step": 1115,
      "train_runtime": 14458.4182,
      "train_tokens_per_second": 4687.719
    },
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 0.5630465939482919,
      "learning_rate": 9.874862380838284e-06,
      "loss": 0.29361796379089355,
      "num_input_tokens_seen": 68091832,
      "step": 1120,
      "train_runtime": 14522.3737,
      "train_tokens_per_second": 4688.754
    },
    {
      "epoch": 0.1434948979591837,
      "grad_norm": 0.5648592974298217,
      "learning_rate": 9.873746324465372e-06,
      "loss": 0.3391995429992676,
      "num_input_tokens_seen": 68401992,
      "step": 1125,
      "train_runtime": 14581.5859,
      "train_tokens_per_second": 4690.984
    },
    {
      "epoch": 0.1441326530612245,
      "grad_norm": 0.5907326903199505,
      "learning_rate": 9.872625376947642e-06,
      "loss": 0.3495980739593506,
      "num_input_tokens_seen": 68711720,
      "step": 1130,
      "train_runtime": 14646.5288,
      "train_tokens_per_second": 4691.331
    },
    {
      "epoch": 0.1447704081632653,
      "grad_norm": 0.6745907973502699,
      "learning_rate": 9.871499539410045e-06,
      "loss": 0.3603889226913452,
      "num_input_tokens_seen": 69022608,
      "step": 1135,
      "train_runtime": 14703.0569,
      "train_tokens_per_second": 4694.439
    },
    {
      "epoch": 0.14540816326530612,
      "grad_norm": 0.623210573264961,
      "learning_rate": 9.870368812982433e-06,
      "loss": 0.33861393928527833,
      "num_input_tokens_seen": 69342064,
      "step": 1140,
      "train_runtime": 14760.0119,
      "train_tokens_per_second": 4697.968
    },
    {
      "epoch": 0.14604591836734693,
      "grad_norm": 0.6487015160167615,
      "learning_rate": 9.869233198799572e-06,
      "loss": 0.35953919887542723,
      "num_input_tokens_seen": 69648560,
      "step": 1145,
      "train_runtime": 14827.9375,
      "train_tokens_per_second": 4697.117
    },
    {
      "epoch": 0.14668367346938777,
      "grad_norm": 0.6052028325524192,
      "learning_rate": 9.86809269800113e-06,
      "loss": 0.3488070011138916,
      "num_input_tokens_seen": 69946440,
      "step": 1150,
      "train_runtime": 14890.0099,
      "train_tokens_per_second": 4697.542
    },
    {
      "epoch": 0.14732142857142858,
      "grad_norm": 0.600872342321137,
      "learning_rate": 9.86694731173168e-06,
      "loss": 0.3258484363555908,
      "num_input_tokens_seen": 70251592,
      "step": 1155,
      "train_runtime": 14953.6021,
      "train_tokens_per_second": 4697.971
    },
    {
      "epoch": 0.14795918367346939,
      "grad_norm": 0.5252518214439542,
      "learning_rate": 9.865797041140695e-06,
      "loss": 0.32013731002807616,
      "num_input_tokens_seen": 70563112,
      "step": 1160,
      "train_runtime": 15008.0415,
      "train_tokens_per_second": 4701.687
    },
    {
      "epoch": 0.1485969387755102,
      "grad_norm": 0.6192588481743379,
      "learning_rate": 9.864641887382552e-06,
      "loss": 0.34639954566955566,
      "num_input_tokens_seen": 70878168,
      "step": 1165,
      "train_runtime": 15079.1516,
      "train_tokens_per_second": 4700.408
    },
    {
      "epoch": 0.14923469387755103,
      "grad_norm": 0.6911373002192646,
      "learning_rate": 9.863481851616528e-06,
      "loss": 0.34975519180297854,
      "num_input_tokens_seen": 71184248,
      "step": 1170,
      "train_runtime": 15143.6133,
      "train_tokens_per_second": 4700.612
    },
    {
      "epoch": 0.14987244897959184,
      "grad_norm": 0.571469570715032,
      "learning_rate": 9.862316935006803e-06,
      "loss": 0.34158849716186523,
      "num_input_tokens_seen": 71487968,
      "step": 1175,
      "train_runtime": 15203.7592,
      "train_tokens_per_second": 4701.993
    },
    {
      "epoch": 0.15051020408163265,
      "grad_norm": 0.551149256404094,
      "learning_rate": 9.86114713872245e-06,
      "loss": 0.3300424575805664,
      "num_input_tokens_seen": 71793392,
      "step": 1180,
      "train_runtime": 15279.8415,
      "train_tokens_per_second": 4698.569
    },
    {
      "epoch": 0.15114795918367346,
      "grad_norm": 0.6198455562277363,
      "learning_rate": 9.85997246393744e-06,
      "loss": 0.3214893102645874,
      "num_input_tokens_seen": 72085568,
      "step": 1185,
      "train_runtime": 15340.7072,
      "train_tokens_per_second": 4698.973
    },
    {
      "epoch": 0.15178571428571427,
      "grad_norm": 0.6330587851850584,
      "learning_rate": 9.858792911830643e-06,
      "loss": 0.34440088272094727,
      "num_input_tokens_seen": 72398520,
      "step": 1190,
      "train_runtime": 15408.0866,
      "train_tokens_per_second": 4698.735
    },
    {
      "epoch": 0.1524234693877551,
      "grad_norm": 0.6445338213269326,
      "learning_rate": 9.857608483585823e-06,
      "loss": 0.32814226150512693,
      "num_input_tokens_seen": 72690904,
      "step": 1195,
      "train_runtime": 15478.9022,
      "train_tokens_per_second": 4696.128
    },
    {
      "epoch": 0.15306122448979592,
      "grad_norm": 0.6491893481187504,
      "learning_rate": 9.856419180391636e-06,
      "loss": 0.3319169759750366,
      "num_input_tokens_seen": 72992704,
      "step": 1200,
      "train_runtime": 15556.4263,
      "train_tokens_per_second": 4692.125
    },
    {
      "epoch": 0.15369897959183673,
      "grad_norm": 0.6766123969843775,
      "learning_rate": 9.855225003441629e-06,
      "loss": 0.3418698310852051,
      "num_input_tokens_seen": 73289816,
      "step": 1205,
      "train_runtime": 15618.4499,
      "train_tokens_per_second": 4692.515
    },
    {
      "epoch": 0.15433673469387754,
      "grad_norm": 0.6074843890700258,
      "learning_rate": 9.85402595393424e-06,
      "loss": 0.33366372585296633,
      "num_input_tokens_seen": 73600072,
      "step": 1210,
      "train_runtime": 15685.2806,
      "train_tokens_per_second": 4692.302
    },
    {
      "epoch": 0.15497448979591838,
      "grad_norm": 0.5381123213921943,
      "learning_rate": 9.852822033072805e-06,
      "loss": 0.33455376625061034,
      "num_input_tokens_seen": 73906704,
      "step": 1215,
      "train_runtime": 15755.0043,
      "train_tokens_per_second": 4690.999
    },
    {
      "epoch": 0.1556122448979592,
      "grad_norm": 0.604250713337634,
      "learning_rate": 9.85161324206554e-06,
      "loss": 0.3582612991333008,
      "num_input_tokens_seen": 74201976,
      "step": 1220,
      "train_runtime": 15831.4985,
      "train_tokens_per_second": 4686.984
    },
    {
      "epoch": 0.15625,
      "grad_norm": 0.6200113895012692,
      "learning_rate": 9.850399582125548e-06,
      "loss": 0.3431835174560547,
      "num_input_tokens_seen": 74500840,
      "step": 1225,
      "train_runtime": 15885.0875,
      "train_tokens_per_second": 4689.986
    },
    {
      "epoch": 0.1568877551020408,
      "grad_norm": 0.673570817712159,
      "learning_rate": 9.849181054470827e-06,
      "loss": 0.34818758964538576,
      "num_input_tokens_seen": 74809560,
      "step": 1230,
      "train_runtime": 15952.0925,
      "train_tokens_per_second": 4689.639
    },
    {
      "epoch": 0.15752551020408162,
      "grad_norm": 0.7610110401468843,
      "learning_rate": 9.84795766032425e-06,
      "loss": 0.3170036792755127,
      "num_input_tokens_seen": 75113704,
      "step": 1235,
      "train_runtime": 16020.717,
      "train_tokens_per_second": 4688.536
    },
    {
      "epoch": 0.15816326530612246,
      "grad_norm": 0.6754346268170676,
      "learning_rate": 9.84672940091358e-06,
      "loss": 0.3286417007446289,
      "num_input_tokens_seen": 75401424,
      "step": 1240,
      "train_runtime": 16090.0586,
      "train_tokens_per_second": 4686.212
    },
    {
      "epoch": 0.15880102040816327,
      "grad_norm": 0.5720135423973471,
      "learning_rate": 9.845496277471463e-06,
      "loss": 0.3288563251495361,
      "num_input_tokens_seen": 75706752,
      "step": 1245,
      "train_runtime": 16149.7182,
      "train_tokens_per_second": 4687.806
    },
    {
      "epoch": 0.15943877551020408,
      "grad_norm": 0.5815285081831328,
      "learning_rate": 9.84425829123542e-06,
      "loss": 0.3585826873779297,
      "num_input_tokens_seen": 76007288,
      "step": 1250,
      "train_runtime": 16208.6122,
      "train_tokens_per_second": 4689.315
    },
    {
      "epoch": 0.1600765306122449,
      "grad_norm": 0.6005471155824291,
      "learning_rate": 9.84301544344786e-06,
      "loss": 0.34567928314208984,
      "num_input_tokens_seen": 76320024,
      "step": 1255,
      "train_runtime": 16268.9419,
      "train_tokens_per_second": 4691.149
    },
    {
      "epoch": 0.16071428571428573,
      "grad_norm": 0.6226023100698392,
      "learning_rate": 9.841767735356069e-06,
      "loss": 0.3292954683303833,
      "num_input_tokens_seen": 76603280,
      "step": 1260,
      "train_runtime": 16332.3032,
      "train_tokens_per_second": 4690.293
    },
    {
      "epoch": 0.16135204081632654,
      "grad_norm": 0.5842796545660951,
      "learning_rate": 9.840515168212208e-06,
      "loss": 0.3470223665237427,
      "num_input_tokens_seen": 76911416,
      "step": 1265,
      "train_runtime": 16388.5587,
      "train_tokens_per_second": 4692.995
    },
    {
      "epoch": 0.16198979591836735,
      "grad_norm": 0.5892994958787647,
      "learning_rate": 9.839257743273313e-06,
      "loss": 0.35660953521728517,
      "num_input_tokens_seen": 77218272,
      "step": 1270,
      "train_runtime": 16452.1484,
      "train_tokens_per_second": 4693.507
    },
    {
      "epoch": 0.16262755102040816,
      "grad_norm": 0.5684784352513662,
      "learning_rate": 9.8379954618013e-06,
      "loss": 0.31610519886016847,
      "num_input_tokens_seen": 77514744,
      "step": 1275,
      "train_runtime": 16520.7862,
      "train_tokens_per_second": 4691.952
    },
    {
      "epoch": 0.16326530612244897,
      "grad_norm": 0.5213398598954231,
      "learning_rate": 9.836728325062956e-06,
      "loss": 0.3151726484298706,
      "num_input_tokens_seen": 77823208,
      "step": 1280,
      "train_runtime": 16587.8748,
      "train_tokens_per_second": 4691.572
    },
    {
      "epoch": 0.1639030612244898,
      "grad_norm": 0.6029850909966052,
      "learning_rate": 9.83545633432994e-06,
      "loss": 0.34789628982543946,
      "num_input_tokens_seen": 78133400,
      "step": 1285,
      "train_runtime": 16657.7724,
      "train_tokens_per_second": 4690.507
    },
    {
      "epoch": 0.16454081632653061,
      "grad_norm": 0.5855806050135156,
      "learning_rate": 9.834179490878789e-06,
      "loss": 0.31041646003723145,
      "num_input_tokens_seen": 78430640,
      "step": 1290,
      "train_runtime": 16706.6472,
      "train_tokens_per_second": 4694.577
    },
    {
      "epoch": 0.16517857142857142,
      "grad_norm": 0.6483103953840245,
      "learning_rate": 9.832897795990896e-06,
      "loss": 0.3281932592391968,
      "num_input_tokens_seen": 78732032,
      "step": 1295,
      "train_runtime": 16772.0203,
      "train_tokens_per_second": 4694.249
    },
    {
      "epoch": 0.16581632653061223,
      "grad_norm": 0.5765423183133421,
      "learning_rate": 9.83161125095254e-06,
      "loss": 0.32477402687072754,
      "num_input_tokens_seen": 79043112,
      "step": 1300,
      "train_runtime": 16842.4359,
      "train_tokens_per_second": 4693.093
    },
    {
      "epoch": 0.16645408163265307,
      "grad_norm": 0.6425529652746363,
      "learning_rate": 9.830319857054853e-06,
      "loss": 0.31569385528564453,
      "num_input_tokens_seen": 79352128,
      "step": 1305,
      "train_runtime": 16908.6278,
      "train_tokens_per_second": 4692.996
    },
    {
      "epoch": 0.16709183673469388,
      "grad_norm": 0.601514067974229,
      "learning_rate": 9.82902361559384e-06,
      "loss": 0.3569095849990845,
      "num_input_tokens_seen": 79658200,
      "step": 1310,
      "train_runtime": 16972.1093,
      "train_tokens_per_second": 4693.477
    },
    {
      "epoch": 0.1677295918367347,
      "grad_norm": 0.5844899308913513,
      "learning_rate": 9.82772252787037e-06,
      "loss": 0.3371304988861084,
      "num_input_tokens_seen": 79970224,
      "step": 1315,
      "train_runtime": 17032.1991,
      "train_tokens_per_second": 4695.238
    },
    {
      "epoch": 0.1683673469387755,
      "grad_norm": 0.6295624315198685,
      "learning_rate": 9.826416595190177e-06,
      "loss": 0.35589327812194826,
      "num_input_tokens_seen": 80254064,
      "step": 1320,
      "train_runtime": 17099.1341,
      "train_tokens_per_second": 4693.458
    },
    {
      "epoch": 0.1690051020408163,
      "grad_norm": 0.6064368753336963,
      "learning_rate": 9.825105818863854e-06,
      "loss": 0.31534266471862793,
      "num_input_tokens_seen": 80554360,
      "step": 1325,
      "train_runtime": 17170.128,
      "train_tokens_per_second": 4691.541
    },
    {
      "epoch": 0.16964285714285715,
      "grad_norm": 0.6218906934525381,
      "learning_rate": 9.823790200206857e-06,
      "loss": 0.35923516750335693,
      "num_input_tokens_seen": 80855544,
      "step": 1330,
      "train_runtime": 17238.4699,
      "train_tokens_per_second": 4690.413
    },
    {
      "epoch": 0.17028061224489796,
      "grad_norm": 0.6919246313636518,
      "learning_rate": 9.822469740539503e-06,
      "loss": 0.3390108585357666,
      "num_input_tokens_seen": 81157864,
      "step": 1335,
      "train_runtime": 17298.2101,
      "train_tokens_per_second": 4691.691
    },
    {
      "epoch": 0.17091836734693877,
      "grad_norm": 0.6586039302284977,
      "learning_rate": 9.821144441186964e-06,
      "loss": 0.36199450492858887,
      "num_input_tokens_seen": 81472488,
      "step": 1340,
      "train_runtime": 17359.5203,
      "train_tokens_per_second": 4693.245
    },
    {
      "epoch": 0.17155612244897958,
      "grad_norm": 0.6101100898122199,
      "learning_rate": 9.819814303479268e-06,
      "loss": 0.34079747200012206,
      "num_input_tokens_seen": 81782968,
      "step": 1345,
      "train_runtime": 17435.2492,
      "train_tokens_per_second": 4690.668
    },
    {
      "epoch": 0.17219387755102042,
      "grad_norm": 0.687974682643214,
      "learning_rate": 9.818479328751303e-06,
      "loss": 0.32731170654296876,
      "num_input_tokens_seen": 82079256,
      "step": 1350,
      "train_runtime": 17497.4705,
      "train_tokens_per_second": 4690.921
    },
    {
      "epoch": 0.17283163265306123,
      "grad_norm": 0.6398125509922725,
      "learning_rate": 9.817139518342811e-06,
      "loss": 0.3492893218994141,
      "num_input_tokens_seen": 82394552,
      "step": 1355,
      "train_runtime": 17561.8393,
      "train_tokens_per_second": 4691.681
    },
    {
      "epoch": 0.17346938775510204,
      "grad_norm": 0.6023525103112162,
      "learning_rate": 9.815794873598384e-06,
      "loss": 0.3357325315475464,
      "num_input_tokens_seen": 82700328,
      "step": 1360,
      "train_runtime": 17623.788,
      "train_tokens_per_second": 4692.54
    },
    {
      "epoch": 0.17410714285714285,
      "grad_norm": 0.5927411576027568,
      "learning_rate": 9.814445395867467e-06,
      "loss": 0.33596770763397216,
      "num_input_tokens_seen": 82995568,
      "step": 1365,
      "train_runtime": 17682.3461,
      "train_tokens_per_second": 4693.697
    },
    {
      "epoch": 0.1747448979591837,
      "grad_norm": 0.5869024102535607,
      "learning_rate": 9.813091086504356e-06,
      "loss": 0.3234375476837158,
      "num_input_tokens_seen": 83286056,
      "step": 1370,
      "train_runtime": 17752.7155,
      "train_tokens_per_second": 4691.454
    },
    {
      "epoch": 0.1753826530612245,
      "grad_norm": 0.6770217851493542,
      "learning_rate": 9.811731946868192e-06,
      "loss": 0.34646997451782224,
      "num_input_tokens_seen": 83594600,
      "step": 1375,
      "train_runtime": 17825.1559,
      "train_tokens_per_second": 4689.698
    },
    {
      "epoch": 0.1760204081632653,
      "grad_norm": 0.6158790602590283,
      "learning_rate": 9.810367978322971e-06,
      "loss": 0.3418739318847656,
      "num_input_tokens_seen": 83902040,
      "step": 1380,
      "train_runtime": 17901.1437,
      "train_tokens_per_second": 4686.965
    },
    {
      "epoch": 0.17665816326530612,
      "grad_norm": 0.635968825879491,
      "learning_rate": 9.808999182237527e-06,
      "loss": 0.3338493347167969,
      "num_input_tokens_seen": 84187696,
      "step": 1385,
      "train_runtime": 17963.0468,
      "train_tokens_per_second": 4686.716
    },
    {
      "epoch": 0.17729591836734693,
      "grad_norm": 0.6397866500946762,
      "learning_rate": 9.807625559985546e-06,
      "loss": 0.35117316246032715,
      "num_input_tokens_seen": 84486344,
      "step": 1390,
      "train_runtime": 18026.1294,
      "train_tokens_per_second": 4686.882
    },
    {
      "epoch": 0.17793367346938777,
      "grad_norm": 0.6953159641397932,
      "learning_rate": 9.806247112945548e-06,
      "loss": 0.34419639110565187,
      "num_input_tokens_seen": 84789144,
      "step": 1395,
      "train_runtime": 18096.0387,
      "train_tokens_per_second": 4685.509
    },
    {
      "epoch": 0.17857142857142858,
      "grad_norm": 0.5786554718477309,
      "learning_rate": 9.804863842500906e-06,
      "loss": 0.32353553771972654,
      "num_input_tokens_seen": 85088672,
      "step": 1400,
      "train_runtime": 18171.3592,
      "train_tokens_per_second": 4682.571
    },
    {
      "epoch": 0.17920918367346939,
      "grad_norm": 0.6394331538844329,
      "learning_rate": 9.803475750039828e-06,
      "loss": 0.33654494285583497,
      "num_input_tokens_seen": 85395520,
      "step": 1405,
      "train_runtime": 18233.9007,
      "train_tokens_per_second": 4683.338
    },
    {
      "epoch": 0.1798469387755102,
      "grad_norm": 0.6519198773120587,
      "learning_rate": 9.80208283695536e-06,
      "loss": 0.35454869270324707,
      "num_input_tokens_seen": 85693248,
      "step": 1410,
      "train_runtime": 18297.3619,
      "train_tokens_per_second": 4683.366
    },
    {
      "epoch": 0.18048469387755103,
      "grad_norm": 0.583077400062921,
      "learning_rate": 9.80068510464539e-06,
      "loss": 0.34549190998077395,
      "num_input_tokens_seen": 85996488,
      "step": 1415,
      "train_runtime": 18359.2733,
      "train_tokens_per_second": 4684.09
    },
    {
      "epoch": 0.18112244897959184,
      "grad_norm": 0.559112413986124,
      "learning_rate": 9.799282554512637e-06,
      "loss": 0.3202133417129517,
      "num_input_tokens_seen": 86298048,
      "step": 1420,
      "train_runtime": 18435.7238,
      "train_tokens_per_second": 4681.023
    },
    {
      "epoch": 0.18176020408163265,
      "grad_norm": 0.6115506518189591,
      "learning_rate": 9.797875187964661e-06,
      "loss": 0.32825522422790526,
      "num_input_tokens_seen": 86604512,
      "step": 1425,
      "train_runtime": 18492.8336,
      "train_tokens_per_second": 4683.139
    },
    {
      "epoch": 0.18239795918367346,
      "grad_norm": 0.7063801060093161,
      "learning_rate": 9.79646300641385e-06,
      "loss": 0.3472906589508057,
      "num_input_tokens_seen": 86906176,
      "step": 1430,
      "train_runtime": 18562.0904,
      "train_tokens_per_second": 4681.918
    },
    {
      "epoch": 0.18303571428571427,
      "grad_norm": 0.9402045630578656,
      "learning_rate": 9.795046011277427e-06,
      "loss": 0.3585078716278076,
      "num_input_tokens_seen": 87224464,
      "step": 1435,
      "train_runtime": 18617.0363,
      "train_tokens_per_second": 4685.196
    },
    {
      "epoch": 0.1836734693877551,
      "grad_norm": 0.6208106233758729,
      "learning_rate": 9.793624203977446e-06,
      "loss": 0.3277950048446655,
      "num_input_tokens_seen": 87529408,
      "step": 1440,
      "train_runtime": 18681.5971,
      "train_tokens_per_second": 4685.328
    },
    {
      "epoch": 0.18431122448979592,
      "grad_norm": 0.7787654832597471,
      "learning_rate": 9.792197585940792e-06,
      "loss": 0.32544598579406736,
      "num_input_tokens_seen": 87829336,
      "step": 1445,
      "train_runtime": 18738.8119,
      "train_tokens_per_second": 4687.028
    },
    {
      "epoch": 0.18494897959183673,
      "grad_norm": 0.5738666671588774,
      "learning_rate": 9.790766158599172e-06,
      "loss": 0.35848388671875,
      "num_input_tokens_seen": 88140344,
      "step": 1450,
      "train_runtime": 18813.001,
      "train_tokens_per_second": 4685.076
    },
    {
      "epoch": 0.18558673469387754,
      "grad_norm": 0.6769713603304937,
      "learning_rate": 9.789329923389128e-06,
      "loss": 0.346897554397583,
      "num_input_tokens_seen": 88442888,
      "step": 1455,
      "train_runtime": 18877.8546,
      "train_tokens_per_second": 4685.007
    },
    {
      "epoch": 0.18622448979591838,
      "grad_norm": 0.6827615245423831,
      "learning_rate": 9.787888881752018e-06,
      "loss": 0.34128966331481936,
      "num_input_tokens_seen": 88741608,
      "step": 1460,
      "train_runtime": 18946.3371,
      "train_tokens_per_second": 4683.84
    },
    {
      "epoch": 0.1868622448979592,
      "grad_norm": 0.609570747387964,
      "learning_rate": 9.78644303513403e-06,
      "loss": 0.3022934913635254,
      "num_input_tokens_seen": 89034512,
      "step": 1465,
      "train_runtime": 19012.8709,
      "train_tokens_per_second": 4682.855
    },
    {
      "epoch": 0.1875,
      "grad_norm": 0.5839733500263187,
      "learning_rate": 9.784992384986173e-06,
      "loss": 0.33053126335144045,
      "num_input_tokens_seen": 89346616,
      "step": 1470,
      "train_runtime": 19070.2625,
      "train_tokens_per_second": 4685.128
    },
    {
      "epoch": 0.1881377551020408,
      "grad_norm": 0.6349108071927354,
      "learning_rate": 9.783536932764273e-06,
      "loss": 0.31446478366851804,
      "num_input_tokens_seen": 89639184,
      "step": 1475,
      "train_runtime": 19139.0289,
      "train_tokens_per_second": 4683.581
    },
    {
      "epoch": 0.18877551020408162,
      "grad_norm": 0.7009626312104598,
      "learning_rate": 9.782076679928981e-06,
      "loss": 0.3313866138458252,
      "num_input_tokens_seen": 89941344,
      "step": 1480,
      "train_runtime": 19207.9576,
      "train_tokens_per_second": 4682.504
    },
    {
      "epoch": 0.18941326530612246,
      "grad_norm": 0.6103496584672967,
      "learning_rate": 9.78061162794576e-06,
      "loss": 0.3180057525634766,
      "num_input_tokens_seen": 90246672,
      "step": 1485,
      "train_runtime": 19270.919,
      "train_tokens_per_second": 4683.05
    },
    {
      "epoch": 0.19005102040816327,
      "grad_norm": 0.657551239573293,
      "learning_rate": 9.779141778284893e-06,
      "loss": 0.35165953636169434,
      "num_input_tokens_seen": 90543072,
      "step": 1490,
      "train_runtime": 19336.0767,
      "train_tokens_per_second": 4682.598
    },
    {
      "epoch": 0.19068877551020408,
      "grad_norm": 0.6078249652219899,
      "learning_rate": 9.77766713242148e-06,
      "loss": 0.33533573150634766,
      "num_input_tokens_seen": 90861600,
      "step": 1495,
      "train_runtime": 19391.6258,
      "train_tokens_per_second": 4685.61
    },
    {
      "epoch": 0.1913265306122449,
      "grad_norm": 0.6052964539905235,
      "learning_rate": 9.776187691835424e-06,
      "loss": 0.343070387840271,
      "num_input_tokens_seen": 91165096,
      "step": 1500,
      "train_runtime": 19463.7243,
      "train_tokens_per_second": 4683.846
    },
    {
      "epoch": 0.19196428571428573,
      "grad_norm": 0.6040610134539512,
      "learning_rate": 9.774703458011453e-06,
      "loss": 0.3406609058380127,
      "num_input_tokens_seen": 91469480,
      "step": 1505,
      "train_runtime": 19531.31,
      "train_tokens_per_second": 4683.223
    },
    {
      "epoch": 0.19260204081632654,
      "grad_norm": 0.6099501886236799,
      "learning_rate": 9.773214432439098e-06,
      "loss": 0.348543381690979,
      "num_input_tokens_seen": 91781056,
      "step": 1510,
      "train_runtime": 19601.2663,
      "train_tokens_per_second": 4682.404
    },
    {
      "epoch": 0.19323979591836735,
      "grad_norm": 0.6071308410923287,
      "learning_rate": 9.771720616612697e-06,
      "loss": 0.3520734071731567,
      "num_input_tokens_seen": 92087768,
      "step": 1515,
      "train_runtime": 19673.3854,
      "train_tokens_per_second": 4680.83
    },
    {
      "epoch": 0.19387755102040816,
      "grad_norm": 0.6263835890108274,
      "learning_rate": 9.770222012031404e-06,
      "loss": 0.3443084716796875,
      "num_input_tokens_seen": 92386408,
      "step": 1520,
      "train_runtime": 19737.0164,
      "train_tokens_per_second": 4680.87
    },
    {
      "epoch": 0.19451530612244897,
      "grad_norm": 0.5864280482644441,
      "learning_rate": 9.768718620199168e-06,
      "loss": 0.3578033447265625,
      "num_input_tokens_seen": 92703640,
      "step": 1525,
      "train_runtime": 19795.5394,
      "train_tokens_per_second": 4683.057
    },
    {
      "epoch": 0.1951530612244898,
      "grad_norm": 0.6311396478788095,
      "learning_rate": 9.76721044262475e-06,
      "loss": 0.3278501510620117,
      "num_input_tokens_seen": 93006512,
      "step": 1530,
      "train_runtime": 19862.0833,
      "train_tokens_per_second": 4682.616
    },
    {
      "epoch": 0.19579081632653061,
      "grad_norm": 0.5855843370148337,
      "learning_rate": 9.765697480821714e-06,
      "loss": 0.33351426124572753,
      "num_input_tokens_seen": 93316656,
      "step": 1535,
      "train_runtime": 19925.5238,
      "train_tokens_per_second": 4683.272
    },
    {
      "epoch": 0.19642857142857142,
      "grad_norm": 0.6686340218559972,
      "learning_rate": 9.764179736308419e-06,
      "loss": 0.3357309103012085,
      "num_input_tokens_seen": 93635280,
      "step": 1540,
      "train_runtime": 19987.1797,
      "train_tokens_per_second": 4684.767
    },
    {
      "epoch": 0.19706632653061223,
      "grad_norm": 0.6698909869610306,
      "learning_rate": 9.762657210608028e-06,
      "loss": 0.33041768074035643,
      "num_input_tokens_seen": 93939712,
      "step": 1545,
      "train_runtime": 20039.3999,
      "train_tokens_per_second": 4687.751
    },
    {
      "epoch": 0.19770408163265307,
      "grad_norm": 0.5995295470856815,
      "learning_rate": 9.761129905248505e-06,
      "loss": 0.3290611505508423,
      "num_input_tokens_seen": 94249200,
      "step": 1550,
      "train_runtime": 20106.1318,
      "train_tokens_per_second": 4687.585
    },
    {
      "epoch": 0.19834183673469388,
      "grad_norm": 0.6519370050719231,
      "learning_rate": 9.759597821762603e-06,
      "loss": 0.36706769466400146,
      "num_input_tokens_seen": 94561768,
      "step": 1555,
      "train_runtime": 20165.6648,
      "train_tokens_per_second": 4689.246
    },
    {
      "epoch": 0.1989795918367347,
      "grad_norm": 0.6705660271899463,
      "learning_rate": 9.758060961687879e-06,
      "loss": 0.3317730188369751,
      "num_input_tokens_seen": 94858616,
      "step": 1560,
      "train_runtime": 20235.7289,
      "train_tokens_per_second": 4687.68
    },
    {
      "epoch": 0.1996173469387755,
      "grad_norm": 0.5125415703099178,
      "learning_rate": 9.756519326566677e-06,
      "loss": 0.35649838447570803,
      "num_input_tokens_seen": 95146448,
      "step": 1565,
      "train_runtime": 20303.9973,
      "train_tokens_per_second": 4686.094
    },
    {
      "epoch": 0.2002551020408163,
      "grad_norm": 0.5664448645100401,
      "learning_rate": 9.754972917946138e-06,
      "loss": 0.34983410835266116,
      "num_input_tokens_seen": 95445928,
      "step": 1570,
      "train_runtime": 20363.9882,
      "train_tokens_per_second": 4686.996
    },
    {
      "epoch": 0.20089285714285715,
      "grad_norm": 0.658224122786756,
      "learning_rate": 9.753421737378188e-06,
      "loss": 0.33331751823425293,
      "num_input_tokens_seen": 95743248,
      "step": 1575,
      "train_runtime": 20437.5432,
      "train_tokens_per_second": 4684.675
    },
    {
      "epoch": 0.20153061224489796,
      "grad_norm": 0.6531108135828381,
      "learning_rate": 9.751865786419546e-06,
      "loss": 0.3314680099487305,
      "num_input_tokens_seen": 96051904,
      "step": 1580,
      "train_runtime": 20508.0466,
      "train_tokens_per_second": 4683.62
    },
    {
      "epoch": 0.20216836734693877,
      "grad_norm": 0.5643789475087291,
      "learning_rate": 9.750305066631717e-06,
      "loss": 0.3457353115081787,
      "num_input_tokens_seen": 96357912,
      "step": 1585,
      "train_runtime": 20575.3858,
      "train_tokens_per_second": 4683.164
    },
    {
      "epoch": 0.20280612244897958,
      "grad_norm": 0.6512259971234623,
      "learning_rate": 9.748739579580995e-06,
      "loss": 0.36769120693206786,
      "num_input_tokens_seen": 96661584,
      "step": 1590,
      "train_runtime": 20640.7671,
      "train_tokens_per_second": 4683.042
    },
    {
      "epoch": 0.20344387755102042,
      "grad_norm": 0.6158621899309391,
      "learning_rate": 9.747169326838451e-06,
      "loss": 0.328265118598938,
      "num_input_tokens_seen": 96963800,
      "step": 1595,
      "train_runtime": 20720.1999,
      "train_tokens_per_second": 4679.675
    },
    {
      "epoch": 0.20408163265306123,
      "grad_norm": 0.6446817922675011,
      "learning_rate": 9.74559430997995e-06,
      "loss": 0.3056048393249512,
      "num_input_tokens_seen": 97270552,
      "step": 1600,
      "train_runtime": 20781.181,
      "train_tokens_per_second": 4680.704
    },
    {
      "epoch": 0.20471938775510204,
      "grad_norm": 0.5889398725102091,
      "learning_rate": 9.744014530586124e-06,
      "loss": 0.3354539632797241,
      "num_input_tokens_seen": 97574656,
      "step": 1605,
      "train_runtime": 20844.5544,
      "train_tokens_per_second": 4681.062
    },
    {
      "epoch": 0.20535714285714285,
      "grad_norm": 0.6617314093110237,
      "learning_rate": 9.742429990242394e-06,
      "loss": 0.3385885715484619,
      "num_input_tokens_seen": 97875664,
      "step": 1610,
      "train_runtime": 20912.0733,
      "train_tokens_per_second": 4680.342
    },
    {
      "epoch": 0.2059948979591837,
      "grad_norm": 0.5622388851008245,
      "learning_rate": 9.740840690538961e-06,
      "loss": 0.33730876445770264,
      "num_input_tokens_seen": 98188344,
      "step": 1615,
      "train_runtime": 20991.9156,
      "train_tokens_per_second": 4677.436
    },
    {
      "epoch": 0.2066326530612245,
      "grad_norm": 0.6184018191793977,
      "learning_rate": 9.739246633070796e-06,
      "loss": 0.3308307409286499,
      "num_input_tokens_seen": 98483392,
      "step": 1620,
      "train_runtime": 21058.0192,
      "train_tokens_per_second": 4676.764
    },
    {
      "epoch": 0.2072704081632653,
      "grad_norm": 0.6692617822427062,
      "learning_rate": 9.737647819437645e-06,
      "loss": 0.33433582782745364,
      "num_input_tokens_seen": 98797736,
      "step": 1625,
      "train_runtime": 21117.4021,
      "train_tokens_per_second": 4678.499
    },
    {
      "epoch": 0.20790816326530612,
      "grad_norm": 0.6180732480184044,
      "learning_rate": 9.73604425124403e-06,
      "loss": 0.3261995792388916,
      "num_input_tokens_seen": 99095392,
      "step": 1630,
      "train_runtime": 21186.3963,
      "train_tokens_per_second": 4677.312
    },
    {
      "epoch": 0.20854591836734693,
      "grad_norm": 0.672907040535661,
      "learning_rate": 9.734435930099246e-06,
      "loss": 0.3504668951034546,
      "num_input_tokens_seen": 99402384,
      "step": 1635,
      "train_runtime": 21264.2669,
      "train_tokens_per_second": 4674.621
    },
    {
      "epoch": 0.20918367346938777,
      "grad_norm": 0.6468683924248895,
      "learning_rate": 9.732822857617352e-06,
      "loss": 0.3475662708282471,
      "num_input_tokens_seen": 99720720,
      "step": 1640,
      "train_runtime": 21336.2991,
      "train_tokens_per_second": 4673.759
    },
    {
      "epoch": 0.20982142857142858,
      "grad_norm": 0.7137629298946826,
      "learning_rate": 9.731205035417183e-06,
      "loss": 0.32748582363128664,
      "num_input_tokens_seen": 100025296,
      "step": 1645,
      "train_runtime": 21409.0905,
      "train_tokens_per_second": 4672.095
    },
    {
      "epoch": 0.21045918367346939,
      "grad_norm": 0.5841812857235733,
      "learning_rate": 9.729582465122334e-06,
      "loss": 0.30248146057128905,
      "num_input_tokens_seen": 100324064,
      "step": 1650,
      "train_runtime": 21472.1864,
      "train_tokens_per_second": 4672.28
    },
    {
      "epoch": 0.2110969387755102,
      "grad_norm": 0.6216174551790111,
      "learning_rate": 9.727955148361166e-06,
      "loss": 0.3440696239471436,
      "num_input_tokens_seen": 100629136,
      "step": 1655,
      "train_runtime": 21544.5045,
      "train_tokens_per_second": 4670.757
    },
    {
      "epoch": 0.21173469387755103,
      "grad_norm": 0.6870197385899494,
      "learning_rate": 9.72632308676681e-06,
      "loss": 0.35681972503662107,
      "num_input_tokens_seen": 100933584,
      "step": 1660,
      "train_runtime": 21619.5784,
      "train_tokens_per_second": 4668.619
    },
    {
      "epoch": 0.21237244897959184,
      "grad_norm": 0.6033968390142657,
      "learning_rate": 9.724686281977146e-06,
      "loss": 0.339033842086792,
      "num_input_tokens_seen": 101232000,
      "step": 1665,
      "train_runtime": 21672.925,
      "train_tokens_per_second": 4670.897
    },
    {
      "epoch": 0.21301020408163265,
      "grad_norm": 0.6605680912767716,
      "learning_rate": 9.72304473563483e-06,
      "loss": 0.32911860942840576,
      "num_input_tokens_seen": 101532152,
      "step": 1670,
      "train_runtime": 21743.3698,
      "train_tokens_per_second": 4669.568
    },
    {
      "epoch": 0.21364795918367346,
      "grad_norm": 0.6057055273564611,
      "learning_rate": 9.721398449387265e-06,
      "loss": 0.34564208984375,
      "num_input_tokens_seen": 101844896,
      "step": 1675,
      "train_runtime": 21802.5161,
      "train_tokens_per_second": 4671.245
    },
    {
      "epoch": 0.21428571428571427,
      "grad_norm": 0.6712000744681125,
      "learning_rate": 9.719747424886613e-06,
      "loss": 0.32876513004302976,
      "num_input_tokens_seen": 102143320,
      "step": 1680,
      "train_runtime": 21865.6942,
      "train_tokens_per_second": 4671.396
    },
    {
      "epoch": 0.2149234693877551,
      "grad_norm": 0.5678300769230327,
      "learning_rate": 9.718091663789795e-06,
      "loss": 0.32980966567993164,
      "num_input_tokens_seen": 102464416,
      "step": 1685,
      "train_runtime": 21921.2594,
      "train_tokens_per_second": 4674.203
    },
    {
      "epoch": 0.21556122448979592,
      "grad_norm": 0.6077964209882442,
      "learning_rate": 9.716431167758482e-06,
      "loss": 0.3364400386810303,
      "num_input_tokens_seen": 102758392,
      "step": 1690,
      "train_runtime": 21990.5214,
      "train_tokens_per_second": 4672.849
    },
    {
      "epoch": 0.21619897959183673,
      "grad_norm": 0.6554874702286015,
      "learning_rate": 9.714765938459097e-06,
      "loss": 0.31924474239349365,
      "num_input_tokens_seen": 103060904,
      "step": 1695,
      "train_runtime": 22063.839,
      "train_tokens_per_second": 4671.032
    },
    {
      "epoch": 0.21683673469387754,
      "grad_norm": 0.6453871857802653,
      "learning_rate": 9.713095977562813e-06,
      "loss": 0.33878040313720703,
      "num_input_tokens_seen": 103368528,
      "step": 1700,
      "train_runtime": 22130.9977,
      "train_tokens_per_second": 4670.758
    },
    {
      "epoch": 0.21747448979591838,
      "grad_norm": 0.6426887219110298,
      "learning_rate": 9.711421286745554e-06,
      "loss": 0.3179375171661377,
      "num_input_tokens_seen": 103666416,
      "step": 1705,
      "train_runtime": 22196.6409,
      "train_tokens_per_second": 4670.365
    },
    {
      "epoch": 0.2181122448979592,
      "grad_norm": 0.60196172861106,
      "learning_rate": 9.70974186768799e-06,
      "loss": 0.3452445507049561,
      "num_input_tokens_seen": 103951664,
      "step": 1710,
      "train_runtime": 22255.5419,
      "train_tokens_per_second": 4670.822
    },
    {
      "epoch": 0.21875,
      "grad_norm": 0.6132093034250564,
      "learning_rate": 9.708057722075532e-06,
      "loss": 0.34422879219055175,
      "num_input_tokens_seen": 104255176,
      "step": 1715,
      "train_runtime": 22319.7294,
      "train_tokens_per_second": 4670.987
    },
    {
      "epoch": 0.2193877551020408,
      "grad_norm": 1.0482197968622364,
      "learning_rate": 9.706368851598343e-06,
      "loss": 0.34129197597503663,
      "num_input_tokens_seen": 104552216,
      "step": 1720,
      "train_runtime": 22386.2925,
      "train_tokens_per_second": 4670.368
    },
    {
      "epoch": 0.22002551020408162,
      "grad_norm": 0.5256163199980716,
      "learning_rate": 9.704675257951316e-06,
      "loss": 0.32835357189178466,
      "num_input_tokens_seen": 104868176,
      "step": 1725,
      "train_runtime": 22447.7966,
      "train_tokens_per_second": 4671.647
    },
    {
      "epoch": 0.22066326530612246,
      "grad_norm": 0.6695721720659025,
      "learning_rate": 9.702976942834096e-06,
      "loss": 0.33469085693359374,
      "num_input_tokens_seen": 105160960,
      "step": 1730,
      "train_runtime": 22509.1775,
      "train_tokens_per_second": 4671.915
    },
    {
      "epoch": 0.22130102040816327,
      "grad_norm": 0.6230385357847109,
      "learning_rate": 9.701273907951058e-06,
      "loss": 0.3191352844238281,
      "num_input_tokens_seen": 105470256,
      "step": 1735,
      "train_runtime": 22574.5748,
      "train_tokens_per_second": 4672.082
    },
    {
      "epoch": 0.22193877551020408,
      "grad_norm": 0.6248704925874972,
      "learning_rate": 9.699566155011317e-06,
      "loss": 0.3543462038040161,
      "num_input_tokens_seen": 105780200,
      "step": 1740,
      "train_runtime": 22641.1173,
      "train_tokens_per_second": 4672.04
    },
    {
      "epoch": 0.2225765306122449,
      "grad_norm": 0.5450021457057649,
      "learning_rate": 9.697853685728721e-06,
      "loss": 0.32406888008117674,
      "num_input_tokens_seen": 106078048,
      "step": 1745,
      "train_runtime": 22702.7513,
      "train_tokens_per_second": 4672.475
    },
    {
      "epoch": 0.22321428571428573,
      "grad_norm": 0.6523177995021735,
      "learning_rate": 9.696136501821858e-06,
      "loss": 0.32779037952423096,
      "num_input_tokens_seen": 106370016,
      "step": 1750,
      "train_runtime": 22770.0901,
      "train_tokens_per_second": 4671.48
    },
    {
      "epoch": 0.22385204081632654,
      "grad_norm": 0.7197164756320799,
      "learning_rate": 9.694414605014035e-06,
      "loss": 0.3535364389419556,
      "num_input_tokens_seen": 106684184,
      "step": 1755,
      "train_runtime": 22839.957,
      "train_tokens_per_second": 4670.945
    },
    {
      "epoch": 0.22448979591836735,
      "grad_norm": 0.5899732558505255,
      "learning_rate": 9.6926879970333e-06,
      "loss": 0.32790746688842776,
      "num_input_tokens_seen": 106998216,
      "step": 1760,
      "train_runtime": 22893.992,
      "train_tokens_per_second": 4673.637
    },
    {
      "epoch": 0.22512755102040816,
      "grad_norm": 0.6168161384769302,
      "learning_rate": 9.690956679612422e-06,
      "loss": 0.33826274871826173,
      "num_input_tokens_seen": 107306360,
      "step": 1765,
      "train_runtime": 22951.1765,
      "train_tokens_per_second": 4675.419
    },
    {
      "epoch": 0.22576530612244897,
      "grad_norm": 0.6104637422756916,
      "learning_rate": 9.6892206544889e-06,
      "loss": 0.32278287410736084,
      "num_input_tokens_seen": 107609480,
      "step": 1770,
      "train_runtime": 23017.8681,
      "train_tokens_per_second": 4675.041
    },
    {
      "epoch": 0.2264030612244898,
      "grad_norm": 0.6512960385084632,
      "learning_rate": 9.687479923404959e-06,
      "loss": 0.3486739158630371,
      "num_input_tokens_seen": 107926024,
      "step": 1775,
      "train_runtime": 23072.1009,
      "train_tokens_per_second": 4677.772
    },
    {
      "epoch": 0.22704081632653061,
      "grad_norm": 0.46664293177956334,
      "learning_rate": 9.68573448810754e-06,
      "loss": 0.31470379829406736,
      "num_input_tokens_seen": 108233416,
      "step": 1780,
      "train_runtime": 23136.5152,
      "train_tokens_per_second": 4678.034
    },
    {
      "epoch": 0.22767857142857142,
      "grad_norm": 0.6314733786111536,
      "learning_rate": 9.683984350348313e-06,
      "loss": 0.32955155372619627,
      "num_input_tokens_seen": 108542880,
      "step": 1785,
      "train_runtime": 23206.5379,
      "train_tokens_per_second": 4677.254
    },
    {
      "epoch": 0.22831632653061223,
      "grad_norm": 0.6113132858996045,
      "learning_rate": 9.68222951188366e-06,
      "loss": 0.3453743696212769,
      "num_input_tokens_seen": 108843736,
      "step": 1790,
      "train_runtime": 23273.2837,
      "train_tokens_per_second": 4676.767
    },
    {
      "epoch": 0.22895408163265307,
      "grad_norm": 0.5834108473728911,
      "learning_rate": 9.680469974474685e-06,
      "loss": 0.32249295711517334,
      "num_input_tokens_seen": 109150272,
      "step": 1795,
      "train_runtime": 23346.8051,
      "train_tokens_per_second": 4675.17
    },
    {
      "epoch": 0.22959183673469388,
      "grad_norm": 0.58938960320997,
      "learning_rate": 9.67870573988721e-06,
      "loss": 0.3216817855834961,
      "num_input_tokens_seen": 109455200,
      "step": 1800,
      "train_runtime": 23409.455,
      "train_tokens_per_second": 4675.683
    },
    {
      "epoch": 0.2302295918367347,
      "grad_norm": 0.8446716184001303,
      "learning_rate": 9.676936809891763e-06,
      "loss": 0.3525111436843872,
      "num_input_tokens_seen": 109764480,
      "step": 1805,
      "train_runtime": 23483.604,
      "train_tokens_per_second": 4674.09
    },
    {
      "epoch": 0.2308673469387755,
      "grad_norm": 0.615189710384464,
      "learning_rate": 9.675163186263591e-06,
      "loss": 0.3223212003707886,
      "num_input_tokens_seen": 110066288,
      "step": 1810,
      "train_runtime": 23554.5749,
      "train_tokens_per_second": 4672.82
    },
    {
      "epoch": 0.2315051020408163,
      "grad_norm": 0.5754384451688632,
      "learning_rate": 9.673384870782649e-06,
      "loss": 0.31503829956054685,
      "num_input_tokens_seen": 110362264,
      "step": 1815,
      "train_runtime": 23629.0737,
      "train_tokens_per_second": 4670.613
    },
    {
      "epoch": 0.23214285714285715,
      "grad_norm": 0.6350773841658875,
      "learning_rate": 9.6716018652336e-06,
      "loss": 0.350568962097168,
      "num_input_tokens_seen": 110668672,
      "step": 1820,
      "train_runtime": 23685.1886,
      "train_tokens_per_second": 4672.484
    },
    {
      "epoch": 0.23278061224489796,
      "grad_norm": 0.6722719077492426,
      "learning_rate": 9.669814171405818e-06,
      "loss": 0.34694905281066896,
      "num_input_tokens_seen": 110968352,
      "step": 1825,
      "train_runtime": 23751.2713,
      "train_tokens_per_second": 4672.102
    },
    {
      "epoch": 0.23341836734693877,
      "grad_norm": 0.6228912303661778,
      "learning_rate": 9.668021791093372e-06,
      "loss": 0.3096019744873047,
      "num_input_tokens_seen": 111282464,
      "step": 1830,
      "train_runtime": 23806.8118,
      "train_tokens_per_second": 4674.396
    },
    {
      "epoch": 0.23405612244897958,
      "grad_norm": 0.6542588610104904,
      "learning_rate": 9.666224726095048e-06,
      "loss": 0.3228870153427124,
      "num_input_tokens_seen": 111590336,
      "step": 1835,
      "train_runtime": 23870.8875,
      "train_tokens_per_second": 4674.746
    },
    {
      "epoch": 0.23469387755102042,
      "grad_norm": 0.6585031873202504,
      "learning_rate": 9.66442297821432e-06,
      "loss": 0.3584311485290527,
      "num_input_tokens_seen": 111899696,
      "step": 1840,
      "train_runtime": 23936.6995,
      "train_tokens_per_second": 4674.817
    },
    {
      "epoch": 0.23533163265306123,
      "grad_norm": 0.5898295862093664,
      "learning_rate": 9.662616549259374e-06,
      "loss": 0.30771589279174805,
      "num_input_tokens_seen": 112185968,
      "step": 1845,
      "train_runtime": 24008.3018,
      "train_tokens_per_second": 4672.799
    },
    {
      "epoch": 0.23596938775510204,
      "grad_norm": 0.6508941366898215,
      "learning_rate": 9.660805441043082e-06,
      "loss": 0.36535985469818116,
      "num_input_tokens_seen": 112485216,
      "step": 1850,
      "train_runtime": 24072.8985,
      "train_tokens_per_second": 4672.691
    },
    {
      "epoch": 0.23660714285714285,
      "grad_norm": 0.5513057418407049,
      "learning_rate": 9.65898965538302e-06,
      "loss": 0.31926376819610597,
      "num_input_tokens_seen": 112792024,
      "step": 1855,
      "train_runtime": 24142.1832,
      "train_tokens_per_second": 4671.989
    },
    {
      "epoch": 0.2372448979591837,
      "grad_norm": 0.5416953303992011,
      "learning_rate": 9.657169194101457e-06,
      "loss": 0.3331429958343506,
      "num_input_tokens_seen": 113094072,
      "step": 1860,
      "train_runtime": 24214.7571,
      "train_tokens_per_second": 4670.461
    },
    {
      "epoch": 0.2378826530612245,
      "grad_norm": 0.6195037185120192,
      "learning_rate": 9.655344059025351e-06,
      "loss": 0.339237904548645,
      "num_input_tokens_seen": 113410272,
      "step": 1865,
      "train_runtime": 24277.9126,
      "train_tokens_per_second": 4671.335
    },
    {
      "epoch": 0.2385204081632653,
      "grad_norm": 0.6658985533340913,
      "learning_rate": 9.653514251986354e-06,
      "loss": 0.34667558670043946,
      "num_input_tokens_seen": 113705304,
      "step": 1870,
      "train_runtime": 24338.1994,
      "train_tokens_per_second": 4671.886
    },
    {
      "epoch": 0.23915816326530612,
      "grad_norm": 0.6200509903536591,
      "learning_rate": 9.651679774820804e-06,
      "loss": 0.321960973739624,
      "num_input_tokens_seen": 114015568,
      "step": 1875,
      "train_runtime": 24388.0746,
      "train_tokens_per_second": 4675.054
    },
    {
      "epoch": 0.23979591836734693,
      "grad_norm": 0.636091566053549,
      "learning_rate": 9.649840629369728e-06,
      "loss": 0.3509210109710693,
      "num_input_tokens_seen": 114325720,
      "step": 1880,
      "train_runtime": 24461.5857,
      "train_tokens_per_second": 4673.684
    },
    {
      "epoch": 0.24043367346938777,
      "grad_norm": 0.657535257785578,
      "learning_rate": 9.647996817478835e-06,
      "loss": 0.3434359073638916,
      "num_input_tokens_seen": 114632136,
      "step": 1885,
      "train_runtime": 24537.3443,
      "train_tokens_per_second": 4671.742
    },
    {
      "epoch": 0.24107142857142858,
      "grad_norm": 0.6140700552709698,
      "learning_rate": 9.646148340998521e-06,
      "loss": 0.32828240394592284,
      "num_input_tokens_seen": 114922240,
      "step": 1890,
      "train_runtime": 24601.3517,
      "train_tokens_per_second": 4671.379
    },
    {
      "epoch": 0.24170918367346939,
      "grad_norm": 0.6578277822186156,
      "learning_rate": 9.64429520178386e-06,
      "loss": 0.35318818092346194,
      "num_input_tokens_seen": 115215224,
      "step": 1895,
      "train_runtime": 24666.4565,
      "train_tokens_per_second": 4670.927
    },
    {
      "epoch": 0.2423469387755102,
      "grad_norm": 0.6233192565315776,
      "learning_rate": 9.642437401694608e-06,
      "loss": 0.335646915435791,
      "num_input_tokens_seen": 115527552,
      "step": 1900,
      "train_runtime": 24740.1503,
      "train_tokens_per_second": 4669.638
    },
    {
      "epoch": 0.24298469387755103,
      "grad_norm": 0.5723546415005316,
      "learning_rate": 9.640574942595195e-06,
      "loss": 0.32998104095458985,
      "num_input_tokens_seen": 115821808,
      "step": 1905,
      "train_runtime": 24803.1445,
      "train_tokens_per_second": 4669.642
    },
    {
      "epoch": 0.24362244897959184,
      "grad_norm": 0.6652879717246634,
      "learning_rate": 9.638707826354732e-06,
      "loss": 0.33642148971557617,
      "num_input_tokens_seen": 116115216,
      "step": 1910,
      "train_runtime": 24866.3469,
      "train_tokens_per_second": 4669.573
    },
    {
      "epoch": 0.24426020408163265,
      "grad_norm": 0.584264535542241,
      "learning_rate": 9.636836054846997e-06,
      "loss": 0.32282869815826415,
      "num_input_tokens_seen": 116415800,
      "step": 1915,
      "train_runtime": 24928.176,
      "train_tokens_per_second": 4670.049
    },
    {
      "epoch": 0.24489795918367346,
      "grad_norm": 0.597573024851043,
      "learning_rate": 9.634959629950444e-06,
      "loss": 0.30709261894226075,
      "num_input_tokens_seen": 116717552,
      "step": 1920,
      "train_runtime": 25000.1179,
      "train_tokens_per_second": 4668.68
    },
    {
      "epoch": 0.24553571428571427,
      "grad_norm": 0.6656629131945074,
      "learning_rate": 9.6330785535482e-06,
      "loss": 0.3267780303955078,
      "num_input_tokens_seen": 117006136,
      "step": 1925,
      "train_runtime": 25067.316,
      "train_tokens_per_second": 4667.677
    },
    {
      "epoch": 0.2461734693877551,
      "grad_norm": 0.7050617650662212,
      "learning_rate": 9.631192827528054e-06,
      "loss": 0.3533444404602051,
      "num_input_tokens_seen": 117313624,
      "step": 1930,
      "train_runtime": 25140.9951,
      "train_tokens_per_second": 4666.228
    },
    {
      "epoch": 0.24681122448979592,
      "grad_norm": 0.6287617293269191,
      "learning_rate": 9.629302453782462e-06,
      "loss": 0.31642513275146483,
      "num_input_tokens_seen": 117618136,
      "step": 1935,
      "train_runtime": 25214.6032,
      "train_tokens_per_second": 4664.683
    },
    {
      "epoch": 0.24744897959183673,
      "grad_norm": 0.6302693162573803,
      "learning_rate": 9.62740743420855e-06,
      "loss": 0.35641942024230955,
      "num_input_tokens_seen": 117914104,
      "step": 1940,
      "train_runtime": 25277.4877,
      "train_tokens_per_second": 4664.787
    },
    {
      "epoch": 0.24808673469387754,
      "grad_norm": 0.6156501910538401,
      "learning_rate": 9.625507770708098e-06,
      "loss": 0.3540866136550903,
      "num_input_tokens_seen": 118221680,
      "step": 1945,
      "train_runtime": 25333.6354,
      "train_tokens_per_second": 4666.59
    },
    {
      "epoch": 0.24872448979591838,
      "grad_norm": 0.6189556769637666,
      "learning_rate": 9.623603465187554e-06,
      "loss": 0.3389142036437988,
      "num_input_tokens_seen": 118522328,
      "step": 1950,
      "train_runtime": 25401.2332,
      "train_tokens_per_second": 4666.007
    },
    {
      "epoch": 0.2493622448979592,
      "grad_norm": 0.581304482769978,
      "learning_rate": 9.621694519558021e-06,
      "loss": 0.3108387470245361,
      "num_input_tokens_seen": 118829640,
      "step": 1955,
      "train_runtime": 25467.7789,
      "train_tokens_per_second": 4665.882
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.572105158541402,
      "learning_rate": 9.619780935735258e-06,
      "loss": 0.3287662982940674,
      "num_input_tokens_seen": 119132240,
      "step": 1960,
      "train_runtime": 25537.5091,
      "train_tokens_per_second": 4664.991
    },
    {
      "epoch": 0.25063775510204084,
      "grad_norm": 0.689536161742216,
      "learning_rate": 9.617862715639685e-06,
      "loss": 0.32274279594421384,
      "num_input_tokens_seen": 119434136,
      "step": 1965,
      "train_runtime": 25614.07,
      "train_tokens_per_second": 4662.833
    },
    {
      "epoch": 0.2512755102040816,
      "grad_norm": 0.5757971447214045,
      "learning_rate": 9.615939861196364e-06,
      "loss": 0.34737372398376465,
      "num_input_tokens_seen": 119756744,
      "step": 1970,
      "train_runtime": 25685.048,
      "train_tokens_per_second": 4662.508
    },
    {
      "epoch": 0.25191326530612246,
      "grad_norm": 0.6088554653419755,
      "learning_rate": 9.614012374335014e-06,
      "loss": 0.3462496757507324,
      "num_input_tokens_seen": 120074472,
      "step": 1975,
      "train_runtime": 25738.5391,
      "train_tokens_per_second": 4665.163
    },
    {
      "epoch": 0.25255102040816324,
      "grad_norm": 0.5165799129994694,
      "learning_rate": 9.612080256990006e-06,
      "loss": 0.30801560878753664,
      "num_input_tokens_seen": 120371440,
      "step": 1980,
      "train_runtime": 25797.2104,
      "train_tokens_per_second": 4666.064
    },
    {
      "epoch": 0.2531887755102041,
      "grad_norm": 0.685217918137659,
      "learning_rate": 9.610143511100354e-06,
      "loss": 0.3369981527328491,
      "num_input_tokens_seen": 120668168,
      "step": 1985,
      "train_runtime": 25858.7819,
      "train_tokens_per_second": 4666.429
    },
    {
      "epoch": 0.2538265306122449,
      "grad_norm": 0.6602263616886537,
      "learning_rate": 9.608202138609714e-06,
      "loss": 0.3087838411331177,
      "num_input_tokens_seen": 120962704,
      "step": 1990,
      "train_runtime": 25925.2431,
      "train_tokens_per_second": 4665.827
    },
    {
      "epoch": 0.2544642857142857,
      "grad_norm": 0.6210613558847153,
      "learning_rate": 9.606256141466394e-06,
      "loss": 0.3179054260253906,
      "num_input_tokens_seen": 121257544,
      "step": 1995,
      "train_runtime": 25991.6959,
      "train_tokens_per_second": 4665.242
    },
    {
      "epoch": 0.25510204081632654,
      "grad_norm": 0.627194515529005,
      "learning_rate": 9.604305521623335e-06,
      "loss": 0.33051357269287107,
      "num_input_tokens_seen": 121558016,
      "step": 2000,
      "train_runtime": 26057.4945,
      "train_tokens_per_second": 4664.992
    },
    {
      "epoch": 0.2557397959183674,
      "grad_norm": 0.6114595189736588,
      "learning_rate": 9.602350281038123e-06,
      "loss": 0.3401334762573242,
      "num_input_tokens_seen": 121861496,
      "step": 2005,
      "train_runtime": 26130.2559,
      "train_tokens_per_second": 4663.617
    },
    {
      "epoch": 0.25637755102040816,
      "grad_norm": 0.6621780413858983,
      "learning_rate": 9.600390421672976e-06,
      "loss": 0.3431340932846069,
      "num_input_tokens_seen": 122161632,
      "step": 2010,
      "train_runtime": 26188.5267,
      "train_tokens_per_second": 4664.7
    },
    {
      "epoch": 0.257015306122449,
      "grad_norm": 0.7067543408043822,
      "learning_rate": 9.59842594549475e-06,
      "loss": 0.3796668291091919,
      "num_input_tokens_seen": 122475304,
      "step": 2015,
      "train_runtime": 26247.8272,
      "train_tokens_per_second": 4666.112
    },
    {
      "epoch": 0.2576530612244898,
      "grad_norm": 1.433248066928353,
      "learning_rate": 9.596456854474933e-06,
      "loss": 0.3507995128631592,
      "num_input_tokens_seen": 122785040,
      "step": 2020,
      "train_runtime": 26312.7337,
      "train_tokens_per_second": 4666.373
    },
    {
      "epoch": 0.2582908163265306,
      "grad_norm": 0.7614373423772619,
      "learning_rate": 9.594483150589647e-06,
      "loss": 0.3719475746154785,
      "num_input_tokens_seen": 123072088,
      "step": 2025,
      "train_runtime": 26369.887,
      "train_tokens_per_second": 4667.145
    },
    {
      "epoch": 0.25892857142857145,
      "grad_norm": 0.5916339291000087,
      "learning_rate": 9.592504835819642e-06,
      "loss": 0.32286806106567384,
      "num_input_tokens_seen": 123369592,
      "step": 2030,
      "train_runtime": 26441.9834,
      "train_tokens_per_second": 4665.671
    },
    {
      "epoch": 0.25956632653061223,
      "grad_norm": 0.5242712990169962,
      "learning_rate": 9.590521912150294e-06,
      "loss": 0.3350060939788818,
      "num_input_tokens_seen": 123680416,
      "step": 2035,
      "train_runtime": 26518.023,
      "train_tokens_per_second": 4664.013
    },
    {
      "epoch": 0.2602040816326531,
      "grad_norm": 0.6586684421360529,
      "learning_rate": 9.588534381571606e-06,
      "loss": 0.3625217914581299,
      "num_input_tokens_seen": 123991008,
      "step": 2040,
      "train_runtime": 26586.6716,
      "train_tokens_per_second": 4663.653
    },
    {
      "epoch": 0.26084183673469385,
      "grad_norm": 0.6898137997632707,
      "learning_rate": 9.586542246078203e-06,
      "loss": 0.3413558006286621,
      "num_input_tokens_seen": 124293272,
      "step": 2045,
      "train_runtime": 26654.3873,
      "train_tokens_per_second": 4663.145
    },
    {
      "epoch": 0.2614795918367347,
      "grad_norm": 0.5822105996114149,
      "learning_rate": 9.584545507669333e-06,
      "loss": 0.32697596549987795,
      "num_input_tokens_seen": 124601432,
      "step": 2050,
      "train_runtime": 26722.4206,
      "train_tokens_per_second": 4662.805
    },
    {
      "epoch": 0.26211734693877553,
      "grad_norm": 0.6422340553091131,
      "learning_rate": 9.582544168348864e-06,
      "loss": 0.32872767448425294,
      "num_input_tokens_seen": 124908560,
      "step": 2055,
      "train_runtime": 26790.9196,
      "train_tokens_per_second": 4662.347
    },
    {
      "epoch": 0.2627551020408163,
      "grad_norm": 0.6240089031569784,
      "learning_rate": 9.580538230125275e-06,
      "loss": 0.31921873092651365,
      "num_input_tokens_seen": 125204664,
      "step": 2060,
      "train_runtime": 26852.1852,
      "train_tokens_per_second": 4662.736
    },
    {
      "epoch": 0.26339285714285715,
      "grad_norm": 0.6815815083946033,
      "learning_rate": 9.57852769501167e-06,
      "loss": 0.3307302951812744,
      "num_input_tokens_seen": 125506080,
      "step": 2065,
      "train_runtime": 26914.4019,
      "train_tokens_per_second": 4663.157
    },
    {
      "epoch": 0.26403061224489793,
      "grad_norm": 0.5968652405541623,
      "learning_rate": 9.576512565025759e-06,
      "loss": 0.3403089284896851,
      "num_input_tokens_seen": 125807424,
      "step": 2070,
      "train_runtime": 26990.1978,
      "train_tokens_per_second": 4661.226
    },
    {
      "epoch": 0.26466836734693877,
      "grad_norm": 0.5939180111595969,
      "learning_rate": 9.574492842189865e-06,
      "loss": 0.3476198673248291,
      "num_input_tokens_seen": 126122304,
      "step": 2075,
      "train_runtime": 27053.9764,
      "train_tokens_per_second": 4661.877
    },
    {
      "epoch": 0.2653061224489796,
      "grad_norm": 0.6215273804787333,
      "learning_rate": 9.572468528530923e-06,
      "loss": 0.3389333724975586,
      "num_input_tokens_seen": 126428616,
      "step": 2080,
      "train_runtime": 27118.5249,
      "train_tokens_per_second": 4662.076
    },
    {
      "epoch": 0.2659438775510204,
      "grad_norm": 1.7346451945661905,
      "learning_rate": 9.570439626080472e-06,
      "loss": 0.3335702419281006,
      "num_input_tokens_seen": 126732032,
      "step": 2085,
      "train_runtime": 27193.5009,
      "train_tokens_per_second": 4660.379
    },
    {
      "epoch": 0.26658163265306123,
      "grad_norm": 0.5789927273759574,
      "learning_rate": 9.568406136874657e-06,
      "loss": 0.3421825647354126,
      "num_input_tokens_seen": 127033448,
      "step": 2090,
      "train_runtime": 27252.3425,
      "train_tokens_per_second": 4661.377
    },
    {
      "epoch": 0.26721938775510207,
      "grad_norm": 0.6284209400592989,
      "learning_rate": 9.566368062954227e-06,
      "loss": 0.348024582862854,
      "num_input_tokens_seen": 127334352,
      "step": 2095,
      "train_runtime": 27323.7215,
      "train_tokens_per_second": 4660.213
    },
    {
      "epoch": 0.26785714285714285,
      "grad_norm": 0.7275169918126387,
      "learning_rate": 9.564325406364532e-06,
      "loss": 0.32784428596496584,
      "num_input_tokens_seen": 127628856,
      "step": 2100,
      "train_runtime": 27391.1009,
      "train_tokens_per_second": 4659.501
    },
    {
      "epoch": 0.2684948979591837,
      "grad_norm": 0.685588875445038,
      "learning_rate": 9.562278169155518e-06,
      "loss": 0.31891841888427735,
      "num_input_tokens_seen": 127932352,
      "step": 2105,
      "train_runtime": 27465.9802,
      "train_tokens_per_second": 4657.848
    },
    {
      "epoch": 0.26913265306122447,
      "grad_norm": 0.6368294637105282,
      "learning_rate": 9.560226353381735e-06,
      "loss": 0.3372203826904297,
      "num_input_tokens_seen": 128220496,
      "step": 2110,
      "train_runtime": 27522.3695,
      "train_tokens_per_second": 4658.774
    },
    {
      "epoch": 0.2697704081632653,
      "grad_norm": 0.5977870469094814,
      "learning_rate": 9.55816996110232e-06,
      "loss": 0.31939680576324464,
      "num_input_tokens_seen": 128517880,
      "step": 2115,
      "train_runtime": 27587.7468,
      "train_tokens_per_second": 4658.513
    },
    {
      "epoch": 0.27040816326530615,
      "grad_norm": 0.6149500125523669,
      "learning_rate": 9.556108994381007e-06,
      "loss": 0.3233501434326172,
      "num_input_tokens_seen": 128818952,
      "step": 2120,
      "train_runtime": 27638.8016,
      "train_tokens_per_second": 4660.801
    },
    {
      "epoch": 0.2710459183673469,
      "grad_norm": 0.6985257716826092,
      "learning_rate": 9.55404345528612e-06,
      "loss": 0.35180096626281737,
      "num_input_tokens_seen": 129110440,
      "step": 2125,
      "train_runtime": 27692.8255,
      "train_tokens_per_second": 4662.234
    },
    {
      "epoch": 0.27168367346938777,
      "grad_norm": 0.6387223819839052,
      "learning_rate": 9.551973345890572e-06,
      "loss": 0.33643279075622556,
      "num_input_tokens_seen": 129406776,
      "step": 2130,
      "train_runtime": 27746.3988,
      "train_tokens_per_second": 4663.912
    },
    {
      "epoch": 0.27232142857142855,
      "grad_norm": 0.6630014602117915,
      "learning_rate": 9.549898668271862e-06,
      "loss": 0.37986087799072266,
      "num_input_tokens_seen": 129723592,
      "step": 2135,
      "train_runtime": 27810.6656,
      "train_tokens_per_second": 4664.527
    },
    {
      "epoch": 0.2729591836734694,
      "grad_norm": 0.6998665321005898,
      "learning_rate": 9.547819424512075e-06,
      "loss": 0.3238245964050293,
      "num_input_tokens_seen": 130027232,
      "step": 2140,
      "train_runtime": 27876.3893,
      "train_tokens_per_second": 4664.422
    },
    {
      "epoch": 0.2735969387755102,
      "grad_norm": 0.6482975563513065,
      "learning_rate": 9.545735616697875e-06,
      "loss": 0.3344579696655273,
      "num_input_tokens_seen": 130331312,
      "step": 2145,
      "train_runtime": 27941.891,
      "train_tokens_per_second": 4664.37
    },
    {
      "epoch": 0.274234693877551,
      "grad_norm": 0.5734214162164026,
      "learning_rate": 9.54364724692051e-06,
      "loss": 0.32822456359863283,
      "num_input_tokens_seen": 130631744,
      "step": 2150,
      "train_runtime": 28015.028,
      "train_tokens_per_second": 4662.917
    },
    {
      "epoch": 0.27487244897959184,
      "grad_norm": 0.6719178969699857,
      "learning_rate": 9.541554317275807e-06,
      "loss": 0.33201870918273924,
      "num_input_tokens_seen": 130932192,
      "step": 2155,
      "train_runtime": 28089.5669,
      "train_tokens_per_second": 4661.239
    },
    {
      "epoch": 0.2755102040816326,
      "grad_norm": 0.6440642242563434,
      "learning_rate": 9.539456829864164e-06,
      "loss": 0.3319772005081177,
      "num_input_tokens_seen": 131236424,
      "step": 2160,
      "train_runtime": 28156.2279,
      "train_tokens_per_second": 4661.009
    },
    {
      "epoch": 0.27614795918367346,
      "grad_norm": 0.5985989929529333,
      "learning_rate": 9.537354786790558e-06,
      "loss": 0.3271048069000244,
      "num_input_tokens_seen": 131538720,
      "step": 2165,
      "train_runtime": 28220.416,
      "train_tokens_per_second": 4661.119
    },
    {
      "epoch": 0.2767857142857143,
      "grad_norm": 0.6101031230908778,
      "learning_rate": 9.535248190164537e-06,
      "loss": 0.3490221261978149,
      "num_input_tokens_seen": 131847192,
      "step": 2170,
      "train_runtime": 28273.268,
      "train_tokens_per_second": 4663.316
    },
    {
      "epoch": 0.2774234693877551,
      "grad_norm": 0.6233698493259191,
      "learning_rate": 9.533137042100214e-06,
      "loss": 0.3301276206970215,
      "num_input_tokens_seen": 132153528,
      "step": 2175,
      "train_runtime": 28338.7671,
      "train_tokens_per_second": 4663.348
    },
    {
      "epoch": 0.2780612244897959,
      "grad_norm": 0.675889434483717,
      "learning_rate": 9.531021344716275e-06,
      "loss": 0.34771027565002444,
      "num_input_tokens_seen": 132455944,
      "step": 2180,
      "train_runtime": 28400.3994,
      "train_tokens_per_second": 4663.876
    },
    {
      "epoch": 0.27869897959183676,
      "grad_norm": 0.6074855332165697,
      "learning_rate": 9.528901100135971e-06,
      "loss": 0.32829039096832274,
      "num_input_tokens_seen": 132755232,
      "step": 2185,
      "train_runtime": 28469.4423,
      "train_tokens_per_second": 4663.078
    },
    {
      "epoch": 0.27933673469387754,
      "grad_norm": 0.6712975298444689,
      "learning_rate": 9.526776310487116e-06,
      "loss": 0.3137925863265991,
      "num_input_tokens_seen": 133059576,
      "step": 2190,
      "train_runtime": 28526.9266,
      "train_tokens_per_second": 4664.35
    },
    {
      "epoch": 0.2799744897959184,
      "grad_norm": 0.6352671046160449,
      "learning_rate": 9.524646977902083e-06,
      "loss": 0.33733036518096926,
      "num_input_tokens_seen": 133364448,
      "step": 2195,
      "train_runtime": 28595.4686,
      "train_tokens_per_second": 4663.832
    },
    {
      "epoch": 0.28061224489795916,
      "grad_norm": 0.6130702477666586,
      "learning_rate": 9.522513104517807e-06,
      "loss": 0.34655616283416746,
      "num_input_tokens_seen": 133671040,
      "step": 2200,
      "train_runtime": 28640.8713,
      "train_tokens_per_second": 4667.143
    },
    {
      "epoch": 0.28125,
      "grad_norm": 0.6405320405199912,
      "learning_rate": 9.52037469247578e-06,
      "loss": 0.3232752561569214,
      "num_input_tokens_seen": 133967984,
      "step": 2205,
      "train_runtime": 28694.0307,
      "train_tokens_per_second": 4668.845
    },
    {
      "epoch": 0.28188775510204084,
      "grad_norm": 0.6421967037212513,
      "learning_rate": 9.518231743922049e-06,
      "loss": 0.3547585248947144,
      "num_input_tokens_seen": 134265280,
      "step": 2210,
      "train_runtime": 28756.9357,
      "train_tokens_per_second": 4668.97
    },
    {
      "epoch": 0.2825255102040816,
      "grad_norm": 0.5754881277928156,
      "learning_rate": 9.516084261007207e-06,
      "loss": 0.344525146484375,
      "num_input_tokens_seen": 134583128,
      "step": 2215,
      "train_runtime": 28815.9492,
      "train_tokens_per_second": 4670.439
    },
    {
      "epoch": 0.28316326530612246,
      "grad_norm": 0.6544882715223516,
      "learning_rate": 9.513932245886409e-06,
      "loss": 0.32808513641357423,
      "num_input_tokens_seen": 134877664,
      "step": 2220,
      "train_runtime": 28875.3166,
      "train_tokens_per_second": 4671.037
    },
    {
      "epoch": 0.28380102040816324,
      "grad_norm": 0.6069905880793696,
      "learning_rate": 9.511775700719347e-06,
      "loss": 0.3475784778594971,
      "num_input_tokens_seen": 135163120,
      "step": 2225,
      "train_runtime": 28938.2272,
      "train_tokens_per_second": 4670.746
    },
    {
      "epoch": 0.2844387755102041,
      "grad_norm": 0.6417998679063434,
      "learning_rate": 9.50961462767027e-06,
      "loss": 0.33773374557495117,
      "num_input_tokens_seen": 135475328,
      "step": 2230,
      "train_runtime": 28996.3294,
      "train_tokens_per_second": 4672.154
    },
    {
      "epoch": 0.2850765306122449,
      "grad_norm": 0.62007247529229,
      "learning_rate": 9.507449028907961e-06,
      "loss": 0.3306095123291016,
      "num_input_tokens_seen": 135785712,
      "step": 2235,
      "train_runtime": 29071.8744,
      "train_tokens_per_second": 4670.69
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 0.607967783035985,
      "learning_rate": 9.505278906605752e-06,
      "loss": 0.34046273231506347,
      "num_input_tokens_seen": 136081960,
      "step": 2240,
      "train_runtime": 29132.4061,
      "train_tokens_per_second": 4671.154
    },
    {
      "epoch": 0.28635204081632654,
      "grad_norm": 0.6926293415406519,
      "learning_rate": 9.503104262941515e-06,
      "loss": 0.30557568073272706,
      "num_input_tokens_seen": 136392784,
      "step": 2245,
      "train_runtime": 29193.0625,
      "train_tokens_per_second": 4672.096
    },
    {
      "epoch": 0.2869897959183674,
      "grad_norm": 0.5863241339230645,
      "learning_rate": 9.500925100097652e-06,
      "loss": 0.3169878005981445,
      "num_input_tokens_seen": 136707440,
      "step": 2250,
      "train_runtime": 29254.6306,
      "train_tokens_per_second": 4673.019
    },
    {
      "epoch": 0.28762755102040816,
      "grad_norm": 0.64560871812282,
      "learning_rate": 9.498741420261109e-06,
      "loss": 0.3424893856048584,
      "num_input_tokens_seen": 137016104,
      "step": 2255,
      "train_runtime": 29320.6932,
      "train_tokens_per_second": 4673.017
    },
    {
      "epoch": 0.288265306122449,
      "grad_norm": 0.622639387104968,
      "learning_rate": 9.496553225623358e-06,
      "loss": 0.34136936664581297,
      "num_input_tokens_seen": 137319680,
      "step": 2260,
      "train_runtime": 29387.2361,
      "train_tokens_per_second": 4672.766
    },
    {
      "epoch": 0.2889030612244898,
      "grad_norm": 0.672758711986229,
      "learning_rate": 9.494360518380405e-06,
      "loss": 0.3330190896987915,
      "num_input_tokens_seen": 137617704,
      "step": 2265,
      "train_runtime": 29459.0509,
      "train_tokens_per_second": 4671.491
    },
    {
      "epoch": 0.2895408163265306,
      "grad_norm": 0.5953712510342588,
      "learning_rate": 9.49216330073279e-06,
      "loss": 0.3248729705810547,
      "num_input_tokens_seen": 137913424,
      "step": 2270,
      "train_runtime": 29525.4431,
      "train_tokens_per_second": 4671.003
    },
    {
      "epoch": 0.29017857142857145,
      "grad_norm": 0.5790091177853225,
      "learning_rate": 9.489961574885567e-06,
      "loss": 0.3313121795654297,
      "num_input_tokens_seen": 138217224,
      "step": 2275,
      "train_runtime": 29592.5955,
      "train_tokens_per_second": 4670.669
    },
    {
      "epoch": 0.29081632653061223,
      "grad_norm": 0.6349700782333549,
      "learning_rate": 9.487755343048328e-06,
      "loss": 0.33457436561584475,
      "num_input_tokens_seen": 138511400,
      "step": 2280,
      "train_runtime": 29659.6408,
      "train_tokens_per_second": 4670.03
    },
    {
      "epoch": 0.2914540816326531,
      "grad_norm": 0.6414273351448437,
      "learning_rate": 9.485544607435177e-06,
      "loss": 0.33912661075592043,
      "num_input_tokens_seen": 138805184,
      "step": 2285,
      "train_runtime": 29725.8377,
      "train_tokens_per_second": 4669.513
    },
    {
      "epoch": 0.29209183673469385,
      "grad_norm": 0.5509349800442801,
      "learning_rate": 9.483329370264743e-06,
      "loss": 0.33886494636535647,
      "num_input_tokens_seen": 139096920,
      "step": 2290,
      "train_runtime": 29795.2282,
      "train_tokens_per_second": 4668.429
    },
    {
      "epoch": 0.2927295918367347,
      "grad_norm": 0.6461902552788196,
      "learning_rate": 9.48110963376017e-06,
      "loss": 0.3412975549697876,
      "num_input_tokens_seen": 139391024,
      "step": 2295,
      "train_runtime": 29868.3203,
      "train_tokens_per_second": 4666.852
    },
    {
      "epoch": 0.29336734693877553,
      "grad_norm": 0.6252669208659251,
      "learning_rate": 9.47888540014912e-06,
      "loss": 0.3495046615600586,
      "num_input_tokens_seen": 139696504,
      "step": 2300,
      "train_runtime": 29933.1969,
      "train_tokens_per_second": 4666.942
    },
    {
      "epoch": 0.2940051020408163,
      "grad_norm": 0.5924907844107669,
      "learning_rate": 9.476656671663766e-06,
      "loss": 0.32523541450500487,
      "num_input_tokens_seen": 139993480,
      "step": 2305,
      "train_runtime": 30002.813,
      "train_tokens_per_second": 4666.012
    },
    {
      "epoch": 0.29464285714285715,
      "grad_norm": 0.7090138453110337,
      "learning_rate": 9.474423450540794e-06,
      "loss": 0.31680908203125,
      "num_input_tokens_seen": 140293864,
      "step": 2310,
      "train_runtime": 30063.5095,
      "train_tokens_per_second": 4666.583
    },
    {
      "epoch": 0.29528061224489793,
      "grad_norm": 0.641100675957291,
      "learning_rate": 9.472185739021396e-06,
      "loss": 0.35444614887237547,
      "num_input_tokens_seen": 140604936,
      "step": 2315,
      "train_runtime": 30123.6805,
      "train_tokens_per_second": 4667.588
    },
    {
      "epoch": 0.29591836734693877,
      "grad_norm": 0.6112601789105732,
      "learning_rate": 9.469943539351272e-06,
      "loss": 0.3213543653488159,
      "num_input_tokens_seen": 140904200,
      "step": 2320,
      "train_runtime": 30187.5952,
      "train_tokens_per_second": 4667.619
    },
    {
      "epoch": 0.2965561224489796,
      "grad_norm": 0.5906407618036358,
      "learning_rate": 9.467696853780625e-06,
      "loss": 0.34745540618896487,
      "num_input_tokens_seen": 141205224,
      "step": 2325,
      "train_runtime": 30254.6717,
      "train_tokens_per_second": 4667.22
    },
    {
      "epoch": 0.2971938775510204,
      "grad_norm": 0.6386828330282118,
      "learning_rate": 9.465445684564165e-06,
      "loss": 0.35611257553100584,
      "num_input_tokens_seen": 141513120,
      "step": 2330,
      "train_runtime": 30328.8833,
      "train_tokens_per_second": 4665.952
    },
    {
      "epoch": 0.29783163265306123,
      "grad_norm": 0.6725992439797863,
      "learning_rate": 9.463190033961092e-06,
      "loss": 0.3520667314529419,
      "num_input_tokens_seen": 141822048,
      "step": 2335,
      "train_runtime": 30398.8053,
      "train_tokens_per_second": 4665.382
    },
    {
      "epoch": 0.29846938775510207,
      "grad_norm": 0.6144161947870468,
      "learning_rate": 9.46092990423511e-06,
      "loss": 0.3467228889465332,
      "num_input_tokens_seen": 142127800,
      "step": 2340,
      "train_runtime": 30459.6782,
      "train_tokens_per_second": 4666.097
    },
    {
      "epoch": 0.29910714285714285,
      "grad_norm": 0.610908675314489,
      "learning_rate": 9.45866529765442e-06,
      "loss": 0.33246798515319825,
      "num_input_tokens_seen": 142446600,
      "step": 2345,
      "train_runtime": 30522.0312,
      "train_tokens_per_second": 4667.009
    },
    {
      "epoch": 0.2997448979591837,
      "grad_norm": 0.5213697942791907,
      "learning_rate": 9.456396216491713e-06,
      "loss": 0.34263951778411866,
      "num_input_tokens_seen": 142756616,
      "step": 2350,
      "train_runtime": 30587.5466,
      "train_tokens_per_second": 4667.148
    },
    {
      "epoch": 0.30038265306122447,
      "grad_norm": 0.6634250372068631,
      "learning_rate": 9.454122663024167e-06,
      "loss": 0.3423199415206909,
      "num_input_tokens_seen": 143058944,
      "step": 2355,
      "train_runtime": 30658.3181,
      "train_tokens_per_second": 4666.236
    },
    {
      "epoch": 0.3010204081632653,
      "grad_norm": 0.5854785785977943,
      "learning_rate": 9.451844639533453e-06,
      "loss": 0.320352840423584,
      "num_input_tokens_seen": 143361264,
      "step": 2360,
      "train_runtime": 30729.0955,
      "train_tokens_per_second": 4665.327
    },
    {
      "epoch": 0.30165816326530615,
      "grad_norm": 0.6102082780565505,
      "learning_rate": 9.44956214830573e-06,
      "loss": 0.3393766641616821,
      "num_input_tokens_seen": 143666552,
      "step": 2365,
      "train_runtime": 30793.6136,
      "train_tokens_per_second": 4665.466
    },
    {
      "epoch": 0.3022959183673469,
      "grad_norm": 1.0665172501048106,
      "learning_rate": 9.447275191631631e-06,
      "loss": 0.3327906608581543,
      "num_input_tokens_seen": 143979752,
      "step": 2370,
      "train_runtime": 30848.9958,
      "train_tokens_per_second": 4667.243
    },
    {
      "epoch": 0.30293367346938777,
      "grad_norm": 0.6647329108943871,
      "learning_rate": 9.444983771806281e-06,
      "loss": 0.3404208183288574,
      "num_input_tokens_seen": 144290592,
      "step": 2375,
      "train_runtime": 30899.6308,
      "train_tokens_per_second": 4669.654
    },
    {
      "epoch": 0.30357142857142855,
      "grad_norm": 0.623910979008559,
      "learning_rate": 9.44268789112928e-06,
      "loss": 0.30806612968444824,
      "num_input_tokens_seen": 144589824,
      "step": 2380,
      "train_runtime": 30971.8011,
      "train_tokens_per_second": 4668.434
    },
    {
      "epoch": 0.3042091836734694,
      "grad_norm": 0.591747167472421,
      "learning_rate": 9.440387551904705e-06,
      "loss": 0.3120733261108398,
      "num_input_tokens_seen": 144895296,
      "step": 2385,
      "train_runtime": 31033.0481,
      "train_tokens_per_second": 4669.064
    },
    {
      "epoch": 0.3048469387755102,
      "grad_norm": 0.6050224036140351,
      "learning_rate": 9.438082756441103e-06,
      "loss": 0.3376578092575073,
      "num_input_tokens_seen": 145203024,
      "step": 2390,
      "train_runtime": 31092.6318,
      "train_tokens_per_second": 4670.014
    },
    {
      "epoch": 0.305484693877551,
      "grad_norm": 0.594862069410769,
      "learning_rate": 9.4357735070515e-06,
      "loss": 0.33997819423675535,
      "num_input_tokens_seen": 145517392,
      "step": 2395,
      "train_runtime": 31162.2023,
      "train_tokens_per_second": 4669.676
    },
    {
      "epoch": 0.30612244897959184,
      "grad_norm": 0.6898138261948322,
      "learning_rate": 9.433459806053392e-06,
      "loss": 0.3264634132385254,
      "num_input_tokens_seen": 145811144,
      "step": 2400,
      "train_runtime": 31224.2055,
      "train_tokens_per_second": 4669.811
    },
    {
      "epoch": 0.3067602040816326,
      "grad_norm": 0.5453565880179001,
      "learning_rate": 9.431141655768733e-06,
      "loss": 0.347784423828125,
      "num_input_tokens_seen": 146115888,
      "step": 2405,
      "train_runtime": 31284.4086,
      "train_tokens_per_second": 4670.566
    },
    {
      "epoch": 0.30739795918367346,
      "grad_norm": 0.5593261854840652,
      "learning_rate": 9.428819058523953e-06,
      "loss": 0.3411884784698486,
      "num_input_tokens_seen": 146418576,
      "step": 2410,
      "train_runtime": 31359.232,
      "train_tokens_per_second": 4669.074
    },
    {
      "epoch": 0.3080357142857143,
      "grad_norm": 0.611737519404148,
      "learning_rate": 9.42649201664994e-06,
      "loss": 0.33095855712890626,
      "num_input_tokens_seen": 146713496,
      "step": 2415,
      "train_runtime": 31423.2437,
      "train_tokens_per_second": 4668.948
    },
    {
      "epoch": 0.3086734693877551,
      "grad_norm": 0.6356188069601292,
      "learning_rate": 9.424160532482042e-06,
      "loss": 0.3445183277130127,
      "num_input_tokens_seen": 147025448,
      "step": 2420,
      "train_runtime": 31481.6979,
      "train_tokens_per_second": 4670.188
    },
    {
      "epoch": 0.3093112244897959,
      "grad_norm": 0.5914621764135959,
      "learning_rate": 9.421824608360068e-06,
      "loss": 0.35089085102081297,
      "num_input_tokens_seen": 147333872,
      "step": 2425,
      "train_runtime": 31546.0157,
      "train_tokens_per_second": 4670.443
    },
    {
      "epoch": 0.30994897959183676,
      "grad_norm": 0.6453748491997388,
      "learning_rate": 9.419484246628279e-06,
      "loss": 0.32887113094329834,
      "num_input_tokens_seen": 147634248,
      "step": 2430,
      "train_runtime": 31609.5632,
      "train_tokens_per_second": 4670.556
    },
    {
      "epoch": 0.31058673469387754,
      "grad_norm": 0.5664137633617244,
      "learning_rate": 9.417139449635394e-06,
      "loss": 0.33138527870178225,
      "num_input_tokens_seen": 147924552,
      "step": 2435,
      "train_runtime": 31680.6953,
      "train_tokens_per_second": 4669.233
    },
    {
      "epoch": 0.3112244897959184,
      "grad_norm": 0.6066526860583623,
      "learning_rate": 9.414790219734578e-06,
      "loss": 0.32412359714508054,
      "num_input_tokens_seen": 148233704,
      "step": 2440,
      "train_runtime": 31745.772,
      "train_tokens_per_second": 4669.4
    },
    {
      "epoch": 0.31186224489795916,
      "grad_norm": 0.6447124007824722,
      "learning_rate": 9.41243655928345e-06,
      "loss": 0.32866530418395995,
      "num_input_tokens_seen": 148534120,
      "step": 2445,
      "train_runtime": 31793.5393,
      "train_tokens_per_second": 4671.833
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.7494477452485521,
      "learning_rate": 9.410078470644069e-06,
      "loss": 0.3428959846496582,
      "num_input_tokens_seen": 148828048,
      "step": 2450,
      "train_runtime": 31870.1946,
      "train_tokens_per_second": 4669.819
    },
    {
      "epoch": 0.31313775510204084,
      "grad_norm": 0.6417954136398213,
      "learning_rate": 9.407715956182942e-06,
      "loss": 0.3676748275756836,
      "num_input_tokens_seen": 149136848,
      "step": 2455,
      "train_runtime": 31943.0656,
      "train_tokens_per_second": 4668.833
    },
    {
      "epoch": 0.3137755102040816,
      "grad_norm": 0.6350797663441539,
      "learning_rate": 9.405349018271021e-06,
      "loss": 0.30690693855285645,
      "num_input_tokens_seen": 149445896,
      "step": 2460,
      "train_runtime": 32006.1498,
      "train_tokens_per_second": 4669.287
    },
    {
      "epoch": 0.31441326530612246,
      "grad_norm": 0.6864100318642214,
      "learning_rate": 9.40297765928369e-06,
      "loss": 0.35631861686706545,
      "num_input_tokens_seen": 149745824,
      "step": 2465,
      "train_runtime": 32071.9141,
      "train_tokens_per_second": 4669.064
    },
    {
      "epoch": 0.31505102040816324,
      "grad_norm": 0.6590135341609927,
      "learning_rate": 9.400601881600776e-06,
      "loss": 0.3548426628112793,
      "num_input_tokens_seen": 150051240,
      "step": 2470,
      "train_runtime": 32141.3372,
      "train_tokens_per_second": 4668.482
    },
    {
      "epoch": 0.3156887755102041,
      "grad_norm": 0.5599258217777506,
      "learning_rate": 9.398221687606535e-06,
      "loss": 0.30653972625732423,
      "num_input_tokens_seen": 150352384,
      "step": 2475,
      "train_runtime": 32203.4824,
      "train_tokens_per_second": 4668.824
    },
    {
      "epoch": 0.3163265306122449,
      "grad_norm": 0.6606285229487573,
      "learning_rate": 9.395837079689661e-06,
      "loss": 0.3408294439315796,
      "num_input_tokens_seen": 150646432,
      "step": 2480,
      "train_runtime": 32276.5113,
      "train_tokens_per_second": 4667.37
    },
    {
      "epoch": 0.3169642857142857,
      "grad_norm": 0.6955671485681683,
      "learning_rate": 9.393448060243272e-06,
      "loss": 0.33505294322967527,
      "num_input_tokens_seen": 150940672,
      "step": 2485,
      "train_runtime": 32350.0257,
      "train_tokens_per_second": 4665.859
    },
    {
      "epoch": 0.31760204081632654,
      "grad_norm": 0.6461058416579228,
      "learning_rate": 9.391054631664918e-06,
      "loss": 0.31870713233947756,
      "num_input_tokens_seen": 151244112,
      "step": 2490,
      "train_runtime": 32413.3615,
      "train_tokens_per_second": 4666.104
    },
    {
      "epoch": 0.3182397959183674,
      "grad_norm": 0.6324593473340161,
      "learning_rate": 9.388656796356569e-06,
      "loss": 0.3533759117126465,
      "num_input_tokens_seen": 151543264,
      "step": 2495,
      "train_runtime": 32486.3477,
      "train_tokens_per_second": 4664.829
    },
    {
      "epoch": 0.31887755102040816,
      "grad_norm": 0.7787696882323114,
      "learning_rate": 9.386254556724622e-06,
      "loss": 0.33388419151306153,
      "num_input_tokens_seen": 151833704,
      "step": 2500,
      "train_runtime": 32544.6127,
      "train_tokens_per_second": 4665.402
    },
    {
      "epoch": 0.319515306122449,
      "grad_norm": 0.5990995821779157,
      "learning_rate": 9.383847915179892e-06,
      "loss": 0.2864094257354736,
      "num_input_tokens_seen": 152130248,
      "step": 2505,
      "train_runtime": 32607.4554,
      "train_tokens_per_second": 4665.505
    },
    {
      "epoch": 0.3201530612244898,
      "grad_norm": 0.6419322189850525,
      "learning_rate": 9.381436874137613e-06,
      "loss": 0.3475051403045654,
      "num_input_tokens_seen": 152442352,
      "step": 2510,
      "train_runtime": 32673.4188,
      "train_tokens_per_second": 4665.638
    },
    {
      "epoch": 0.3207908163265306,
      "grad_norm": 0.6505549961954357,
      "learning_rate": 9.37902143601743e-06,
      "loss": 0.3537461757659912,
      "num_input_tokens_seen": 152760032,
      "step": 2515,
      "train_runtime": 32719.699,
      "train_tokens_per_second": 4668.748
    },
    {
      "epoch": 0.32142857142857145,
      "grad_norm": 0.5861846328820182,
      "learning_rate": 9.376601603243408e-06,
      "loss": 0.348602819442749,
      "num_input_tokens_seen": 153068600,
      "step": 2520,
      "train_runtime": 32770.6385,
      "train_tokens_per_second": 4670.907
    },
    {
      "epoch": 0.32206632653061223,
      "grad_norm": 0.7079839564574756,
      "learning_rate": 9.374177378244014e-06,
      "loss": 0.3363666534423828,
      "num_input_tokens_seen": 153377984,
      "step": 2525,
      "train_runtime": 32849.9291,
      "train_tokens_per_second": 4669.051
    },
    {
      "epoch": 0.3227040816326531,
      "grad_norm": 0.5879781830033334,
      "learning_rate": 9.37174876345213e-06,
      "loss": 0.3585888624191284,
      "num_input_tokens_seen": 153689064,
      "step": 2530,
      "train_runtime": 32914.019,
      "train_tokens_per_second": 4669.41
    },
    {
      "epoch": 0.32334183673469385,
      "grad_norm": 0.6540182742323833,
      "learning_rate": 9.369315761305039e-06,
      "loss": 0.31468565464019777,
      "num_input_tokens_seen": 153984552,
      "step": 2535,
      "train_runtime": 32979.5771,
      "train_tokens_per_second": 4669.088
    },
    {
      "epoch": 0.3239795918367347,
      "grad_norm": 0.6650660841122097,
      "learning_rate": 9.36687837424443e-06,
      "loss": 0.34476327896118164,
      "num_input_tokens_seen": 154287496,
      "step": 2540,
      "train_runtime": 33054.6929,
      "train_tokens_per_second": 4667.643
    },
    {
      "epoch": 0.32461734693877553,
      "grad_norm": 0.6233755553317001,
      "learning_rate": 9.364436604716389e-06,
      "loss": 0.3406651973724365,
      "num_input_tokens_seen": 154585272,
      "step": 2545,
      "train_runtime": 33123.6683,
      "train_tokens_per_second": 4666.913
    },
    {
      "epoch": 0.3252551020408163,
      "grad_norm": 0.6101525641330187,
      "learning_rate": 9.361990455171405e-06,
      "loss": 0.3554771661758423,
      "num_input_tokens_seen": 154886344,
      "step": 2550,
      "train_runtime": 33192.3732,
      "train_tokens_per_second": 4666.323
    },
    {
      "epoch": 0.32589285714285715,
      "grad_norm": 0.5682004310503971,
      "learning_rate": 9.359539928064358e-06,
      "loss": 0.32879574298858644,
      "num_input_tokens_seen": 155183904,
      "step": 2555,
      "train_runtime": 33265.43,
      "train_tokens_per_second": 4665.02
    },
    {
      "epoch": 0.32653061224489793,
      "grad_norm": 0.6046645904565979,
      "learning_rate": 9.357085025854525e-06,
      "loss": 0.32778658866882326,
      "num_input_tokens_seen": 155479896,
      "step": 2560,
      "train_runtime": 33327.2187,
      "train_tokens_per_second": 4665.253
    },
    {
      "epoch": 0.32716836734693877,
      "grad_norm": 0.6581990983962149,
      "learning_rate": 9.354625751005568e-06,
      "loss": 0.3326746702194214,
      "num_input_tokens_seen": 155769320,
      "step": 2565,
      "train_runtime": 33392.0899,
      "train_tokens_per_second": 4664.857
    },
    {
      "epoch": 0.3278061224489796,
      "grad_norm": 0.5969263683150732,
      "learning_rate": 9.352162105985544e-06,
      "loss": 0.3629491090774536,
      "num_input_tokens_seen": 156069912,
      "step": 2570,
      "train_runtime": 33461.8702,
      "train_tokens_per_second": 4664.112
    },
    {
      "epoch": 0.3284438775510204,
      "grad_norm": 0.5808380876202471,
      "learning_rate": 9.349694093266893e-06,
      "loss": 0.35565996170043945,
      "num_input_tokens_seen": 156382408,
      "step": 2575,
      "train_runtime": 33522.2671,
      "train_tokens_per_second": 4665.031
    },
    {
      "epoch": 0.32908163265306123,
      "grad_norm": 0.5636273327318048,
      "learning_rate": 9.347221715326437e-06,
      "loss": 0.31146578788757323,
      "num_input_tokens_seen": 156682336,
      "step": 2580,
      "train_runtime": 33587.2443,
      "train_tokens_per_second": 4664.936
    },
    {
      "epoch": 0.32971938775510207,
      "grad_norm": 0.6487848791397187,
      "learning_rate": 9.344744974645382e-06,
      "loss": 0.3354040145874023,
      "num_input_tokens_seen": 156989784,
      "step": 2585,
      "train_runtime": 33659.8888,
      "train_tokens_per_second": 4664.002
    },
    {
      "epoch": 0.33035714285714285,
      "grad_norm": 0.5585759467138597,
      "learning_rate": 9.342263873709307e-06,
      "loss": 0.3347164630889893,
      "num_input_tokens_seen": 157290608,
      "step": 2590,
      "train_runtime": 33726.6933,
      "train_tokens_per_second": 4663.683
    },
    {
      "epoch": 0.3309948979591837,
      "grad_norm": 0.6994474755761214,
      "learning_rate": 9.339778415008171e-06,
      "loss": 0.33094253540039065,
      "num_input_tokens_seen": 157596152,
      "step": 2595,
      "train_runtime": 33784.6265,
      "train_tokens_per_second": 4664.73
    },
    {
      "epoch": 0.33163265306122447,
      "grad_norm": 0.5525052017458213,
      "learning_rate": 9.337288601036307e-06,
      "loss": 0.31957478523254396,
      "num_input_tokens_seen": 157906568,
      "step": 2600,
      "train_runtime": 33836.3773,
      "train_tokens_per_second": 4666.769
    },
    {
      "epoch": 0.3322704081632653,
      "grad_norm": 0.6278856592683223,
      "learning_rate": 9.334794434292416e-06,
      "loss": 0.35186176300048827,
      "num_input_tokens_seen": 158220672,
      "step": 2605,
      "train_runtime": 33895.5559,
      "train_tokens_per_second": 4667.888
    },
    {
      "epoch": 0.33290816326530615,
      "grad_norm": 0.6876677412592418,
      "learning_rate": 9.33229591727957e-06,
      "loss": 0.3774047136306763,
      "num_input_tokens_seen": 158532136,
      "step": 2610,
      "train_runtime": 33966.8312,
      "train_tokens_per_second": 4667.263
    },
    {
      "epoch": 0.3335459183673469,
      "grad_norm": 0.5117917866687411,
      "learning_rate": 9.329793052505203e-06,
      "loss": 0.31768646240234377,
      "num_input_tokens_seen": 158820376,
      "step": 2615,
      "train_runtime": 34036.0631,
      "train_tokens_per_second": 4666.238
    },
    {
      "epoch": 0.33418367346938777,
      "grad_norm": 0.6339635444309031,
      "learning_rate": 9.327285842481114e-06,
      "loss": 0.32222609519958495,
      "num_input_tokens_seen": 159119232,
      "step": 2620,
      "train_runtime": 34103.306,
      "train_tokens_per_second": 4665.801
    },
    {
      "epoch": 0.33482142857142855,
      "grad_norm": 0.7031918289873753,
      "learning_rate": 9.324774289723469e-06,
      "loss": 0.33108365535736084,
      "num_input_tokens_seen": 159423616,
      "step": 2625,
      "train_runtime": 34164.9634,
      "train_tokens_per_second": 4666.29
    },
    {
      "epoch": 0.3354591836734694,
      "grad_norm": 0.5939881258178996,
      "learning_rate": 9.322258396752781e-06,
      "loss": 0.311852240562439,
      "num_input_tokens_seen": 159729448,
      "step": 2630,
      "train_runtime": 34226.2164,
      "train_tokens_per_second": 4666.874
    },
    {
      "epoch": 0.3360969387755102,
      "grad_norm": 0.6185057311864506,
      "learning_rate": 9.319738166093925e-06,
      "loss": 0.33764641284942626,
      "num_input_tokens_seen": 160035936,
      "step": 2635,
      "train_runtime": 34291.2022,
      "train_tokens_per_second": 4666.968
    },
    {
      "epoch": 0.336734693877551,
      "grad_norm": 0.5838681290380003,
      "learning_rate": 9.31721360027613e-06,
      "loss": 0.35221052169799805,
      "num_input_tokens_seen": 160344704,
      "step": 2640,
      "train_runtime": 34346.869,
      "train_tokens_per_second": 4668.394
    },
    {
      "epoch": 0.33737244897959184,
      "grad_norm": 0.5943851267565616,
      "learning_rate": 9.314684701832974e-06,
      "loss": 0.3359955072402954,
      "num_input_tokens_seen": 160656000,
      "step": 2645,
      "train_runtime": 34405.4803,
      "train_tokens_per_second": 4669.489
    },
    {
      "epoch": 0.3380102040816326,
      "grad_norm": 0.6619312646741659,
      "learning_rate": 9.312151473302381e-06,
      "loss": 0.3358307838439941,
      "num_input_tokens_seen": 160975440,
      "step": 2650,
      "train_runtime": 34459.3594,
      "train_tokens_per_second": 4671.458
    },
    {
      "epoch": 0.33864795918367346,
      "grad_norm": 1.537017500320144,
      "learning_rate": 9.309613917226627e-06,
      "loss": 0.32498579025268554,
      "num_input_tokens_seen": 161287256,
      "step": 2655,
      "train_runtime": 34513.7246,
      "train_tokens_per_second": 4673.134
    },
    {
      "epoch": 0.3392857142857143,
      "grad_norm": 0.6704002948164608,
      "learning_rate": 9.307072036152324e-06,
      "loss": 0.3317585468292236,
      "num_input_tokens_seen": 161591088,
      "step": 2660,
      "train_runtime": 34569.154,
      "train_tokens_per_second": 4674.43
    },
    {
      "epoch": 0.3399234693877551,
      "grad_norm": 3.357026823966067,
      "learning_rate": 9.304525832630426e-06,
      "loss": 0.35298595428466795,
      "num_input_tokens_seen": 161899864,
      "step": 2665,
      "train_runtime": 34615.6725,
      "train_tokens_per_second": 4677.068
    },
    {
      "epoch": 0.3405612244897959,
      "grad_norm": 0.6276999935784682,
      "learning_rate": 9.30197530921623e-06,
      "loss": 0.3207338809967041,
      "num_input_tokens_seen": 162190664,
      "step": 2670,
      "train_runtime": 34674.1465,
      "train_tokens_per_second": 4677.568
    },
    {
      "epoch": 0.34119897959183676,
      "grad_norm": 0.6447054948284444,
      "learning_rate": 9.29942046846936e-06,
      "loss": 0.36296284198760986,
      "num_input_tokens_seen": 162492848,
      "step": 2675,
      "train_runtime": 34736.6134,
      "train_tokens_per_second": 4677.855
    },
    {
      "epoch": 0.34183673469387754,
      "grad_norm": 0.6659259383281735,
      "learning_rate": 9.29686131295378e-06,
      "loss": 0.34287004470825194,
      "num_input_tokens_seen": 162799640,
      "step": 2680,
      "train_runtime": 34800.4481,
      "train_tokens_per_second": 4678.09
    },
    {
      "epoch": 0.3424744897959184,
      "grad_norm": 0.6160996870715756,
      "learning_rate": 9.294297845237778e-06,
      "loss": 0.32772340774536135,
      "num_input_tokens_seen": 163091384,
      "step": 2685,
      "train_runtime": 34870.4575,
      "train_tokens_per_second": 4677.065
    },
    {
      "epoch": 0.34311224489795916,
      "grad_norm": 0.6637960980537871,
      "learning_rate": 9.291730067893978e-06,
      "loss": 0.33747942447662355,
      "num_input_tokens_seen": 163408824,
      "step": 2690,
      "train_runtime": 34936.4167,
      "train_tokens_per_second": 4677.321
    },
    {
      "epoch": 0.34375,
      "grad_norm": 0.5806253007995478,
      "learning_rate": 9.28915798349932e-06,
      "loss": 0.34053735733032225,
      "num_input_tokens_seen": 163712008,
      "step": 2695,
      "train_runtime": 34996.0742,
      "train_tokens_per_second": 4678.011
    },
    {
      "epoch": 0.34438775510204084,
      "grad_norm": 0.5519737963229622,
      "learning_rate": 9.28658159463507e-06,
      "loss": 0.3610675573348999,
      "num_input_tokens_seen": 164021160,
      "step": 2700,
      "train_runtime": 35058.4891,
      "train_tokens_per_second": 4678.501
    },
    {
      "epoch": 0.3450255102040816,
      "grad_norm": 0.5769031799916639,
      "learning_rate": 9.284000903886818e-06,
      "loss": 0.3434394598007202,
      "num_input_tokens_seen": 164325600,
      "step": 2705,
      "train_runtime": 35124.4865,
      "train_tokens_per_second": 4678.377
    },
    {
      "epoch": 0.34566326530612246,
      "grad_norm": 0.5964397528249339,
      "learning_rate": 9.281415913844465e-06,
      "loss": 0.3025984287261963,
      "num_input_tokens_seen": 164618672,
      "step": 2710,
      "train_runtime": 35187.2744,
      "train_tokens_per_second": 4678.358
    },
    {
      "epoch": 0.34630102040816324,
      "grad_norm": 0.6144875198789141,
      "learning_rate": 9.278826627102228e-06,
      "loss": 0.32732648849487306,
      "num_input_tokens_seen": 164918320,
      "step": 2715,
      "train_runtime": 35259.9225,
      "train_tokens_per_second": 4677.217
    },
    {
      "epoch": 0.3469387755102041,
      "grad_norm": 0.6339683127694683,
      "learning_rate": 9.276233046258637e-06,
      "loss": 0.35390415191650393,
      "num_input_tokens_seen": 165221192,
      "step": 2720,
      "train_runtime": 35331.5231,
      "train_tokens_per_second": 4676.311
    },
    {
      "epoch": 0.3475765306122449,
      "grad_norm": 0.7106694950438467,
      "learning_rate": 9.273635173916535e-06,
      "loss": 0.3319819927215576,
      "num_input_tokens_seen": 165519432,
      "step": 2725,
      "train_runtime": 35402.4405,
      "train_tokens_per_second": 4675.368
    },
    {
      "epoch": 0.3482142857142857,
      "grad_norm": 0.6325425007324904,
      "learning_rate": 9.271033012683063e-06,
      "loss": 0.32244267463684084,
      "num_input_tokens_seen": 165814768,
      "step": 2730,
      "train_runtime": 35468.3018,
      "train_tokens_per_second": 4675.013
    },
    {
      "epoch": 0.34885204081632654,
      "grad_norm": 0.738569970816405,
      "learning_rate": 9.268426565169677e-06,
      "loss": 0.36324148178100585,
      "num_input_tokens_seen": 166118744,
      "step": 2735,
      "train_runtime": 35539.3616,
      "train_tokens_per_second": 4674.219
    },
    {
      "epoch": 0.3494897959183674,
      "grad_norm": 0.5759820580206395,
      "learning_rate": 9.265815833992126e-06,
      "loss": 0.3284599304199219,
      "num_input_tokens_seen": 166410816,
      "step": 2740,
      "train_runtime": 35608.415,
      "train_tokens_per_second": 4673.356
    },
    {
      "epoch": 0.35012755102040816,
      "grad_norm": 0.5748435489146855,
      "learning_rate": 9.263200821770462e-06,
      "loss": 0.3294307470321655,
      "num_input_tokens_seen": 166702848,
      "step": 2745,
      "train_runtime": 35678.5207,
      "train_tokens_per_second": 4672.359
    },
    {
      "epoch": 0.350765306122449,
      "grad_norm": 0.6659830395525751,
      "learning_rate": 9.260581531129032e-06,
      "loss": 0.3194013833999634,
      "num_input_tokens_seen": 167006248,
      "step": 2750,
      "train_runtime": 35747.2172,
      "train_tokens_per_second": 4671.867
    },
    {
      "epoch": 0.3514030612244898,
      "grad_norm": 0.6210573361812904,
      "learning_rate": 9.257957964696475e-06,
      "loss": 0.3320168972015381,
      "num_input_tokens_seen": 167315752,
      "step": 2755,
      "train_runtime": 35814.7093,
      "train_tokens_per_second": 4671.705
    },
    {
      "epoch": 0.3520408163265306,
      "grad_norm": 0.632148633969262,
      "learning_rate": 9.255330125105728e-06,
      "loss": 0.3254489183425903,
      "num_input_tokens_seen": 167627704,
      "step": 2760,
      "train_runtime": 35883.4015,
      "train_tokens_per_second": 4671.455
    },
    {
      "epoch": 0.35267857142857145,
      "grad_norm": 0.5862505482561928,
      "learning_rate": 9.252698014994008e-06,
      "loss": 0.35499675273895265,
      "num_input_tokens_seen": 167929672,
      "step": 2765,
      "train_runtime": 35954.0454,
      "train_tokens_per_second": 4670.675
    },
    {
      "epoch": 0.35331632653061223,
      "grad_norm": 0.5614888584027603,
      "learning_rate": 9.250061637002822e-06,
      "loss": 0.3320456504821777,
      "num_input_tokens_seen": 168226976,
      "step": 2770,
      "train_runtime": 36021.9545,
      "train_tokens_per_second": 4670.123
    },
    {
      "epoch": 0.3539540816326531,
      "grad_norm": 0.6631468104613814,
      "learning_rate": 9.24742099377796e-06,
      "loss": 0.3739363193511963,
      "num_input_tokens_seen": 168540912,
      "step": 2775,
      "train_runtime": 36085.8513,
      "train_tokens_per_second": 4670.554
    },
    {
      "epoch": 0.35459183673469385,
      "grad_norm": 0.572270159485271,
      "learning_rate": 9.244776087969492e-06,
      "loss": 0.33939266204833984,
      "num_input_tokens_seen": 168838760,
      "step": 2780,
      "train_runtime": 36158.9856,
      "train_tokens_per_second": 4669.344
    },
    {
      "epoch": 0.3552295918367347,
      "grad_norm": 0.5855835377406302,
      "learning_rate": 9.242126922231763e-06,
      "loss": 0.3214082717895508,
      "num_input_tokens_seen": 169136064,
      "step": 2785,
      "train_runtime": 36220.962,
      "train_tokens_per_second": 4669.563
    },
    {
      "epoch": 0.35586734693877553,
      "grad_norm": 0.6277419797401431,
      "learning_rate": 9.239473499223399e-06,
      "loss": 0.335541296005249,
      "num_input_tokens_seen": 169438592,
      "step": 2790,
      "train_runtime": 36286.8167,
      "train_tokens_per_second": 4669.426
    },
    {
      "epoch": 0.3565051020408163,
      "grad_norm": 0.695117959320036,
      "learning_rate": 9.236815821607295e-06,
      "loss": 0.35208282470703123,
      "num_input_tokens_seen": 169753176,
      "step": 2795,
      "train_runtime": 36346.4815,
      "train_tokens_per_second": 4670.416
    },
    {
      "epoch": 0.35714285714285715,
      "grad_norm": 0.6389643526772001,
      "learning_rate": 9.234153892050616e-06,
      "loss": 0.35019755363464355,
      "num_input_tokens_seen": 170057392,
      "step": 2800,
      "train_runtime": 36419.0912,
      "train_tokens_per_second": 4669.457
    },
    {
      "epoch": 0.35778061224489793,
      "grad_norm": 0.5914066202080734,
      "learning_rate": 9.231487713224793e-06,
      "loss": 0.32256550788879396,
      "num_input_tokens_seen": 170370160,
      "step": 2805,
      "train_runtime": 36487.9242,
      "train_tokens_per_second": 4669.22
    },
    {
      "epoch": 0.35841836734693877,
      "grad_norm": 0.6221792543769145,
      "learning_rate": 9.228817287805523e-06,
      "loss": 0.35767345428466796,
      "num_input_tokens_seen": 170668320,
      "step": 2810,
      "train_runtime": 36543.2233,
      "train_tokens_per_second": 4670.314
    },
    {
      "epoch": 0.3590561224489796,
      "grad_norm": 0.645679524738949,
      "learning_rate": 9.226142618472765e-06,
      "loss": 0.3371246814727783,
      "num_input_tokens_seen": 170973560,
      "step": 2815,
      "train_runtime": 36602.9245,
      "train_tokens_per_second": 4671.036
    },
    {
      "epoch": 0.3596938775510204,
      "grad_norm": 0.6052849458401102,
      "learning_rate": 9.223463707910736e-06,
      "loss": 0.32080750465393065,
      "num_input_tokens_seen": 171263632,
      "step": 2820,
      "train_runtime": 36657.9381,
      "train_tokens_per_second": 4671.938
    },
    {
      "epoch": 0.36033163265306123,
      "grad_norm": 0.5936493011922163,
      "learning_rate": 9.22078055880791e-06,
      "loss": 0.3442652463912964,
      "num_input_tokens_seen": 171562192,
      "step": 2825,
      "train_runtime": 36725.7365,
      "train_tokens_per_second": 4671.443
    },
    {
      "epoch": 0.36096938775510207,
      "grad_norm": 0.6263734728465141,
      "learning_rate": 9.218093173857014e-06,
      "loss": 0.317061185836792,
      "num_input_tokens_seen": 171858536,
      "step": 2830,
      "train_runtime": 36790.3942,
      "train_tokens_per_second": 4671.288
    },
    {
      "epoch": 0.36160714285714285,
      "grad_norm": 0.6645303163373804,
      "learning_rate": 9.21540155575503e-06,
      "loss": 0.37257728576660154,
      "num_input_tokens_seen": 172165192,
      "step": 2835,
      "train_runtime": 36847.811,
      "train_tokens_per_second": 4672.332
    },
    {
      "epoch": 0.3622448979591837,
      "grad_norm": 0.6587149142814338,
      "learning_rate": 9.212705707203177e-06,
      "loss": 0.3401576042175293,
      "num_input_tokens_seen": 172474456,
      "step": 2840,
      "train_runtime": 36913.8231,
      "train_tokens_per_second": 4672.354
    },
    {
      "epoch": 0.36288265306122447,
      "grad_norm": 0.6048189666141288,
      "learning_rate": 9.210005630906936e-06,
      "loss": 0.35530571937561034,
      "num_input_tokens_seen": 172794016,
      "step": 2845,
      "train_runtime": 36987.6805,
      "train_tokens_per_second": 4671.664
    },
    {
      "epoch": 0.3635204081632653,
      "grad_norm": 0.7282644912069675,
      "learning_rate": 9.207301329576017e-06,
      "loss": 0.3370820999145508,
      "num_input_tokens_seen": 173098728,
      "step": 2850,
      "train_runtime": 37059.1021,
      "train_tokens_per_second": 4670.883
    },
    {
      "epoch": 0.36415816326530615,
      "grad_norm": 0.6453836226269013,
      "learning_rate": 9.204592805924378e-06,
      "loss": 0.3250255584716797,
      "num_input_tokens_seen": 173405192,
      "step": 2855,
      "train_runtime": 37128.7211,
      "train_tokens_per_second": 4670.379
    },
    {
      "epoch": 0.3647959183673469,
      "grad_norm": 0.6142423379629992,
      "learning_rate": 9.201880062670207e-06,
      "loss": 0.35407509803771975,
      "num_input_tokens_seen": 173709184,
      "step": 2860,
      "train_runtime": 37195.9411,
      "train_tokens_per_second": 4670.111
    },
    {
      "epoch": 0.36543367346938777,
      "grad_norm": 0.5836807662560142,
      "learning_rate": 9.199163102535937e-06,
      "loss": 0.35567197799682615,
      "num_input_tokens_seen": 174005904,
      "step": 2865,
      "train_runtime": 37273.1105,
      "train_tokens_per_second": 4668.403
    },
    {
      "epoch": 0.36607142857142855,
      "grad_norm": 0.5665829230515435,
      "learning_rate": 9.196441928248222e-06,
      "loss": 0.3214565277099609,
      "num_input_tokens_seen": 174307712,
      "step": 2870,
      "train_runtime": 37341.2654,
      "train_tokens_per_second": 4667.965
    },
    {
      "epoch": 0.3667091836734694,
      "grad_norm": 0.6183474998074492,
      "learning_rate": 9.193716542537956e-06,
      "loss": 0.33814573287963867,
      "num_input_tokens_seen": 174623384,
      "step": 2875,
      "train_runtime": 37408.3943,
      "train_tokens_per_second": 4668.027
    },
    {
      "epoch": 0.3673469387755102,
      "grad_norm": 0.6280789635059695,
      "learning_rate": 9.19098694814025e-06,
      "loss": 0.3294132947921753,
      "num_input_tokens_seen": 174922208,
      "step": 2880,
      "train_runtime": 37471.0044,
      "train_tokens_per_second": 4668.202
    },
    {
      "epoch": 0.367984693877551,
      "grad_norm": 0.5877021053310618,
      "learning_rate": 9.188253147794443e-06,
      "loss": 0.351007080078125,
      "num_input_tokens_seen": 175229896,
      "step": 2885,
      "train_runtime": 37537.9516,
      "train_tokens_per_second": 4668.073
    },
    {
      "epoch": 0.36862244897959184,
      "grad_norm": 0.6256900617275598,
      "learning_rate": 9.185515144244097e-06,
      "loss": 0.34029929637908934,
      "num_input_tokens_seen": 175532040,
      "step": 2890,
      "train_runtime": 37597.9508,
      "train_tokens_per_second": 4668.66
    },
    {
      "epoch": 0.3692602040816326,
      "grad_norm": 0.6588914210325635,
      "learning_rate": 9.182772940236986e-06,
      "loss": 0.3353889942169189,
      "num_input_tokens_seen": 175835776,
      "step": 2895,
      "train_runtime": 37668.146,
      "train_tokens_per_second": 4668.023
    },
    {
      "epoch": 0.36989795918367346,
      "grad_norm": 0.6313523605519256,
      "learning_rate": 9.180026538525106e-06,
      "loss": 0.31960842609405515,
      "num_input_tokens_seen": 176141928,
      "step": 2900,
      "train_runtime": 37728.1665,
      "train_tokens_per_second": 4668.712
    },
    {
      "epoch": 0.3705357142857143,
      "grad_norm": 0.573107384113324,
      "learning_rate": 9.177275941864663e-06,
      "loss": 0.35578694343566897,
      "num_input_tokens_seen": 176446368,
      "step": 2905,
      "train_runtime": 37794.3189,
      "train_tokens_per_second": 4668.595
    },
    {
      "epoch": 0.3711734693877551,
      "grad_norm": 0.5731345850938927,
      "learning_rate": 9.174521153016072e-06,
      "loss": 0.33559885025024416,
      "num_input_tokens_seen": 176756824,
      "step": 2910,
      "train_runtime": 37853.7472,
      "train_tokens_per_second": 4669.467
    },
    {
      "epoch": 0.3718112244897959,
      "grad_norm": 0.6182752939068035,
      "learning_rate": 9.171762174743954e-06,
      "loss": 0.33328685760498045,
      "num_input_tokens_seen": 177065320,
      "step": 2915,
      "train_runtime": 37918.1928,
      "train_tokens_per_second": 4669.667
    },
    {
      "epoch": 0.37244897959183676,
      "grad_norm": 0.6799748268215919,
      "learning_rate": 9.168999009817141e-06,
      "loss": 0.33100380897521975,
      "num_input_tokens_seen": 177368072,
      "step": 2920,
      "train_runtime": 37988.2744,
      "train_tokens_per_second": 4669.022
    },
    {
      "epoch": 0.37308673469387754,
      "grad_norm": 0.6506459692910762,
      "learning_rate": 9.166231661008658e-06,
      "loss": 0.3455332040786743,
      "num_input_tokens_seen": 177684296,
      "step": 2925,
      "train_runtime": 38063.2205,
      "train_tokens_per_second": 4668.136
    },
    {
      "epoch": 0.3737244897959184,
      "grad_norm": 0.6286667442429432,
      "learning_rate": 9.163460131095735e-06,
      "loss": 0.33367528915405276,
      "num_input_tokens_seen": 177984512,
      "step": 2930,
      "train_runtime": 38132.8945,
      "train_tokens_per_second": 4667.48
    },
    {
      "epoch": 0.37436224489795916,
      "grad_norm": 0.5712602431075823,
      "learning_rate": 9.160684422859794e-06,
      "loss": 0.3466202259063721,
      "num_input_tokens_seen": 178290584,
      "step": 2935,
      "train_runtime": 38203.2929,
      "train_tokens_per_second": 4666.89
    },
    {
      "epoch": 0.375,
      "grad_norm": 0.6526649634821595,
      "learning_rate": 9.157904539086454e-06,
      "loss": 0.3054463863372803,
      "num_input_tokens_seen": 178587696,
      "step": 2940,
      "train_runtime": 38255.7101,
      "train_tokens_per_second": 4668.262
    },
    {
      "epoch": 0.37563775510204084,
      "grad_norm": 0.6184464389304304,
      "learning_rate": 9.15512048256552e-06,
      "loss": 0.32589786052703856,
      "num_input_tokens_seen": 178896720,
      "step": 2945,
      "train_runtime": 38317.4065,
      "train_tokens_per_second": 4668.811
    },
    {
      "epoch": 0.3762755102040816,
      "grad_norm": 0.5552739239070901,
      "learning_rate": 9.15233225609099e-06,
      "loss": 0.3620042324066162,
      "num_input_tokens_seen": 179201000,
      "step": 2950,
      "train_runtime": 38378.5683,
      "train_tokens_per_second": 4669.299
    },
    {
      "epoch": 0.37691326530612246,
      "grad_norm": 0.5162166500595451,
      "learning_rate": 9.149539862461044e-06,
      "loss": 0.2974902629852295,
      "num_input_tokens_seen": 179506648,
      "step": 2955,
      "train_runtime": 38446.2337,
      "train_tokens_per_second": 4669.031
    },
    {
      "epoch": 0.37755102040816324,
      "grad_norm": 0.6620949196358052,
      "learning_rate": 9.14674330447804e-06,
      "loss": 0.30786125659942626,
      "num_input_tokens_seen": 179794472,
      "step": 2960,
      "train_runtime": 38511.9579,
      "train_tokens_per_second": 4668.536
    },
    {
      "epoch": 0.3781887755102041,
      "grad_norm": 0.6700230669142001,
      "learning_rate": 9.143942584948525e-06,
      "loss": 0.34150445461273193,
      "num_input_tokens_seen": 180093352,
      "step": 2965,
      "train_runtime": 38580.8417,
      "train_tokens_per_second": 4667.948
    },
    {
      "epoch": 0.3788265306122449,
      "grad_norm": 0.5478533505985292,
      "learning_rate": 9.141137706683212e-06,
      "loss": 0.30869088172912595,
      "num_input_tokens_seen": 180392080,
      "step": 2970,
      "train_runtime": 38650.0598,
      "train_tokens_per_second": 4667.317
    },
    {
      "epoch": 0.3794642857142857,
      "grad_norm": 0.678337655081668,
      "learning_rate": 9.138328672496994e-06,
      "loss": 0.3172877073287964,
      "num_input_tokens_seen": 180697368,
      "step": 2975,
      "train_runtime": 38714.4903,
      "train_tokens_per_second": 4667.435
    },
    {
      "epoch": 0.38010204081632654,
      "grad_norm": 0.6415863337052933,
      "learning_rate": 9.135515485208933e-06,
      "loss": 0.36992654800415037,
      "num_input_tokens_seen": 181003232,
      "step": 2980,
      "train_runtime": 38782.0693,
      "train_tokens_per_second": 4667.189
    },
    {
      "epoch": 0.3807397959183674,
      "grad_norm": 0.6732183984497285,
      "learning_rate": 9.132698147642258e-06,
      "loss": 0.36478912830352783,
      "num_input_tokens_seen": 181304680,
      "step": 2985,
      "train_runtime": 38844.9148,
      "train_tokens_per_second": 4667.398
    },
    {
      "epoch": 0.38137755102040816,
      "grad_norm": 0.6655835113305426,
      "learning_rate": 9.129876662624366e-06,
      "loss": 0.3404196262359619,
      "num_input_tokens_seen": 181618216,
      "step": 2990,
      "train_runtime": 38914.3985,
      "train_tokens_per_second": 4667.121
    },
    {
      "epoch": 0.382015306122449,
      "grad_norm": 0.6170004441491723,
      "learning_rate": 9.127051032986814e-06,
      "loss": 0.377862811088562,
      "num_input_tokens_seen": 181925896,
      "step": 2995,
      "train_runtime": 38971.9278,
      "train_tokens_per_second": 4668.127
    },
    {
      "epoch": 0.3826530612244898,
      "grad_norm": 0.6059184841126629,
      "learning_rate": 9.124221261565316e-06,
      "loss": 0.3464046478271484,
      "num_input_tokens_seen": 182226640,
      "step": 3000,
      "train_runtime": 39042.6605,
      "train_tokens_per_second": 4667.372
    },
    {
      "epoch": 0.3832908163265306,
      "grad_norm": 0.6876833630306839,
      "learning_rate": 9.121387351199748e-06,
      "loss": 0.33391451835632324,
      "num_input_tokens_seen": 182541464,
      "step": 3005,
      "train_runtime": 39101.7645,
      "train_tokens_per_second": 4668.369
    },
    {
      "epoch": 0.38392857142857145,
      "grad_norm": 0.6239294662858951,
      "learning_rate": 9.118549304734136e-06,
      "loss": 0.3343007564544678,
      "num_input_tokens_seen": 182840416,
      "step": 3010,
      "train_runtime": 39157.4459,
      "train_tokens_per_second": 4669.365
    },
    {
      "epoch": 0.38456632653061223,
      "grad_norm": 0.6787382490837304,
      "learning_rate": 9.115707125016658e-06,
      "loss": 0.32906546592712405,
      "num_input_tokens_seen": 183150872,
      "step": 3015,
      "train_runtime": 39227.2893,
      "train_tokens_per_second": 4668.966
    },
    {
      "epoch": 0.3852040816326531,
      "grad_norm": 0.6087883139818103,
      "learning_rate": 9.112860814899637e-06,
      "loss": 0.3491034984588623,
      "num_input_tokens_seen": 183456848,
      "step": 3020,
      "train_runtime": 39306.7877,
      "train_tokens_per_second": 4667.307
    },
    {
      "epoch": 0.38584183673469385,
      "grad_norm": 0.5756683852291705,
      "learning_rate": 9.110010377239552e-06,
      "loss": 0.33542647361755373,
      "num_input_tokens_seen": 183754136,
      "step": 3025,
      "train_runtime": 39373.2272,
      "train_tokens_per_second": 4666.982
    },
    {
      "epoch": 0.3864795918367347,
      "grad_norm": 0.6047218819178658,
      "learning_rate": 9.107155814897007e-06,
      "loss": 0.3634347438812256,
      "num_input_tokens_seen": 184055568,
      "step": 3030,
      "train_runtime": 39439.871,
      "train_tokens_per_second": 4666.739
    },
    {
      "epoch": 0.38711734693877553,
      "grad_norm": 0.6724971852548446,
      "learning_rate": 9.10429713073676e-06,
      "loss": 0.3365731716156006,
      "num_input_tokens_seen": 184350472,
      "step": 3035,
      "train_runtime": 39504.5952,
      "train_tokens_per_second": 4666.558
    },
    {
      "epoch": 0.3877551020408163,
      "grad_norm": 0.592126040117668,
      "learning_rate": 9.101434327627696e-06,
      "loss": 0.3240904569625854,
      "num_input_tokens_seen": 184649680,
      "step": 3040,
      "train_runtime": 39580.6974,
      "train_tokens_per_second": 4665.145
    },
    {
      "epoch": 0.38839285714285715,
      "grad_norm": 0.7763590285508554,
      "learning_rate": 9.098567408442844e-06,
      "loss": 0.3177307605743408,
      "num_input_tokens_seen": 184945944,
      "step": 3045,
      "train_runtime": 39641.093,
      "train_tokens_per_second": 4665.511
    },
    {
      "epoch": 0.38903061224489793,
      "grad_norm": 0.5875868470111111,
      "learning_rate": 9.095696376059353e-06,
      "loss": 0.3542469024658203,
      "num_input_tokens_seen": 185251256,
      "step": 3050,
      "train_runtime": 39703.0058,
      "train_tokens_per_second": 4665.925
    },
    {
      "epoch": 0.38966836734693877,
      "grad_norm": 0.5847860295978273,
      "learning_rate": 9.092821233358507e-06,
      "loss": 0.35657784938812254,
      "num_input_tokens_seen": 185551480,
      "step": 3055,
      "train_runtime": 39764.1848,
      "train_tokens_per_second": 4666.297
    },
    {
      "epoch": 0.3903061224489796,
      "grad_norm": 0.6271717615017474,
      "learning_rate": 9.08994198322571e-06,
      "loss": 0.3421341419219971,
      "num_input_tokens_seen": 185856064,
      "step": 3060,
      "train_runtime": 39831.0054,
      "train_tokens_per_second": 4666.115
    },
    {
      "epoch": 0.3909438775510204,
      "grad_norm": 0.5714588255841407,
      "learning_rate": 9.087058628550492e-06,
      "loss": 0.3498190402984619,
      "num_input_tokens_seen": 186159872,
      "step": 3065,
      "train_runtime": 39901.5872,
      "train_tokens_per_second": 4665.475
    },
    {
      "epoch": 0.39158163265306123,
      "grad_norm": 0.562812539277608,
      "learning_rate": 9.084171172226502e-06,
      "loss": 0.3342771291732788,
      "num_input_tokens_seen": 186472704,
      "step": 3070,
      "train_runtime": 39967.9694,
      "train_tokens_per_second": 4665.554
    },
    {
      "epoch": 0.39221938775510207,
      "grad_norm": 0.5631366168298503,
      "learning_rate": 9.081279617151502e-06,
      "loss": 0.3412167549133301,
      "num_input_tokens_seen": 186778176,
      "step": 3075,
      "train_runtime": 40033.8552,
      "train_tokens_per_second": 4665.506
    },
    {
      "epoch": 0.39285714285714285,
      "grad_norm": 0.705040663198948,
      "learning_rate": 9.078383966227369e-06,
      "loss": 0.3425647735595703,
      "num_input_tokens_seen": 187063736,
      "step": 3080,
      "train_runtime": 40092.033,
      "train_tokens_per_second": 4665.858
    },
    {
      "epoch": 0.3934948979591837,
      "grad_norm": 0.6215548340584263,
      "learning_rate": 9.075484222360094e-06,
      "loss": 0.3367135047912598,
      "num_input_tokens_seen": 187368624,
      "step": 3085,
      "train_runtime": 40157.096,
      "train_tokens_per_second": 4665.891
    },
    {
      "epoch": 0.39413265306122447,
      "grad_norm": 0.5527858614708204,
      "learning_rate": 9.07258038845977e-06,
      "loss": 0.3358824014663696,
      "num_input_tokens_seen": 187663328,
      "step": 3090,
      "train_runtime": 40220.8318,
      "train_tokens_per_second": 4665.824
    },
    {
      "epoch": 0.3947704081632653,
      "grad_norm": 1.2400382462560369,
      "learning_rate": 9.069672467440598e-06,
      "loss": 0.33083035945892336,
      "num_input_tokens_seen": 187973304,
      "step": 3095,
      "train_runtime": 40292.7091,
      "train_tokens_per_second": 4665.194
    },
    {
      "epoch": 0.39540816326530615,
      "grad_norm": 0.580912158692128,
      "learning_rate": 9.066760462220878e-06,
      "loss": 0.33884828090667723,
      "num_input_tokens_seen": 188282120,
      "step": 3100,
      "train_runtime": 40359.1714,
      "train_tokens_per_second": 4665.163
    },
    {
      "epoch": 0.3960459183673469,
      "grad_norm": 0.6333973041936676,
      "learning_rate": 9.063844375723014e-06,
      "loss": 0.3614191055297852,
      "num_input_tokens_seen": 188588584,
      "step": 3105,
      "train_runtime": 40412.2684,
      "train_tokens_per_second": 4666.617
    },
    {
      "epoch": 0.39668367346938777,
      "grad_norm": 0.6616804860116007,
      "learning_rate": 9.060924210873501e-06,
      "loss": 0.3382896423339844,
      "num_input_tokens_seen": 188877960,
      "step": 3110,
      "train_runtime": 40479.3542,
      "train_tokens_per_second": 4666.032
    },
    {
      "epoch": 0.39732142857142855,
      "grad_norm": 0.5940013837269007,
      "learning_rate": 9.057999970602926e-06,
      "loss": 0.3252332448959351,
      "num_input_tokens_seen": 189183784,
      "step": 3115,
      "train_runtime": 40532.2194,
      "train_tokens_per_second": 4667.491
    },
    {
      "epoch": 0.3979591836734694,
      "grad_norm": 0.6260211613428975,
      "learning_rate": 9.055071657845973e-06,
      "loss": 0.32281057834625243,
      "num_input_tokens_seen": 189493632,
      "step": 3120,
      "train_runtime": 40607.1692,
      "train_tokens_per_second": 4666.507
    },
    {
      "epoch": 0.3985969387755102,
      "grad_norm": 0.5737762890912685,
      "learning_rate": 9.052139275541404e-06,
      "loss": 0.30558199882507325,
      "num_input_tokens_seen": 189783792,
      "step": 3125,
      "train_runtime": 40679.3976,
      "train_tokens_per_second": 4665.354
    },
    {
      "epoch": 0.399234693877551,
      "grad_norm": 0.6283015593723065,
      "learning_rate": 9.049202826632072e-06,
      "loss": 0.3246393442153931,
      "num_input_tokens_seen": 190085928,
      "step": 3130,
      "train_runtime": 40750.9828,
      "train_tokens_per_second": 4664.573
    },
    {
      "epoch": 0.39987244897959184,
      "grad_norm": 0.6689106619691115,
      "learning_rate": 9.046262314064907e-06,
      "loss": 0.324275541305542,
      "num_input_tokens_seen": 190382216,
      "step": 3135,
      "train_runtime": 40817.4109,
      "train_tokens_per_second": 4664.24
    },
    {
      "epoch": 0.4005102040816326,
      "grad_norm": 0.6360393990753626,
      "learning_rate": 9.04331774079092e-06,
      "loss": 0.3147127151489258,
      "num_input_tokens_seen": 190675216,
      "step": 3140,
      "train_runtime": 40881.2044,
      "train_tokens_per_second": 4664.129
    },
    {
      "epoch": 0.40114795918367346,
      "grad_norm": 0.6214882977851226,
      "learning_rate": 9.040369109765196e-06,
      "loss": 0.32260823249816895,
      "num_input_tokens_seen": 190965104,
      "step": 3145,
      "train_runtime": 40955.7806,
      "train_tokens_per_second": 4662.714
    },
    {
      "epoch": 0.4017857142857143,
      "grad_norm": 0.6637829015701485,
      "learning_rate": 9.037416423946891e-06,
      "loss": 0.3365804195404053,
      "num_input_tokens_seen": 191261352,
      "step": 3150,
      "train_runtime": 41016.5369,
      "train_tokens_per_second": 4663.03
    },
    {
      "epoch": 0.4024234693877551,
      "grad_norm": 0.6892121705707139,
      "learning_rate": 9.034459686299232e-06,
      "loss": 0.33006772994995115,
      "num_input_tokens_seen": 191568288,
      "step": 3155,
      "train_runtime": 41082.9564,
      "train_tokens_per_second": 4662.963
    },
    {
      "epoch": 0.4030612244897959,
      "grad_norm": 0.6292349021817905,
      "learning_rate": 9.031498899789513e-06,
      "loss": 0.3405275344848633,
      "num_input_tokens_seen": 191879040,
      "step": 3160,
      "train_runtime": 41159.2321,
      "train_tokens_per_second": 4661.871
    },
    {
      "epoch": 0.40369897959183676,
      "grad_norm": 0.6416600064075941,
      "learning_rate": 9.028534067389087e-06,
      "loss": 0.34804649353027345,
      "num_input_tokens_seen": 192184496,
      "step": 3165,
      "train_runtime": 41217.9089,
      "train_tokens_per_second": 4662.645
    },
    {
      "epoch": 0.40433673469387754,
      "grad_norm": 0.6028801651675489,
      "learning_rate": 9.025565192073374e-06,
      "loss": 0.31906704902648925,
      "num_input_tokens_seen": 192501256,
      "step": 3170,
      "train_runtime": 41282.7892,
      "train_tokens_per_second": 4662.991
    },
    {
      "epoch": 0.4049744897959184,
      "grad_norm": 0.6013871902386567,
      "learning_rate": 9.022592276821843e-06,
      "loss": 0.3202676773071289,
      "num_input_tokens_seen": 192801104,
      "step": 3175,
      "train_runtime": 41350.7709,
      "train_tokens_per_second": 4662.576
    },
    {
      "epoch": 0.40561224489795916,
      "grad_norm": 0.6532415703338951,
      "learning_rate": 9.019615324618027e-06,
      "loss": 0.3559399604797363,
      "num_input_tokens_seen": 193101872,
      "step": 3180,
      "train_runtime": 41407.7088,
      "train_tokens_per_second": 4663.428
    },
    {
      "epoch": 0.40625,
      "grad_norm": 0.6191157615724391,
      "learning_rate": 9.016634338449504e-06,
      "loss": 0.3301278591156006,
      "num_input_tokens_seen": 193401672,
      "step": 3185,
      "train_runtime": 41472.1512,
      "train_tokens_per_second": 4663.411
    },
    {
      "epoch": 0.40688775510204084,
      "grad_norm": 0.6973825057090168,
      "learning_rate": 9.0136493213079e-06,
      "loss": 0.3401746988296509,
      "num_input_tokens_seen": 193709256,
      "step": 3190,
      "train_runtime": 41541.9209,
      "train_tokens_per_second": 4662.983
    },
    {
      "epoch": 0.4075255102040816,
      "grad_norm": 0.6821771969199714,
      "learning_rate": 9.01066027618889e-06,
      "loss": 0.324747896194458,
      "num_input_tokens_seen": 194004880,
      "step": 3195,
      "train_runtime": 41607.2159,
      "train_tokens_per_second": 4662.77
    },
    {
      "epoch": 0.40816326530612246,
      "grad_norm": 0.6134365426069813,
      "learning_rate": 9.007667206092188e-06,
      "loss": 0.3389841318130493,
      "num_input_tokens_seen": 194301520,
      "step": 3200,
      "train_runtime": 41682.5693,
      "train_tokens_per_second": 4661.457
    },
    {
      "epoch": 0.40880102040816324,
      "grad_norm": 0.6142967231252621,
      "learning_rate": 9.00467011402155e-06,
      "loss": 0.3330172061920166,
      "num_input_tokens_seen": 194608512,
      "step": 3205,
      "train_runtime": 41760.7855,
      "train_tokens_per_second": 4660.078
    },
    {
      "epoch": 0.4094387755102041,
      "grad_norm": 0.6090999730172338,
      "learning_rate": 9.001669002984767e-06,
      "loss": 0.3154860019683838,
      "num_input_tokens_seen": 194893256,
      "step": 3210,
      "train_runtime": 41832.6509,
      "train_tokens_per_second": 4658.879
    },
    {
      "epoch": 0.4100765306122449,
      "grad_norm": 0.6169265613712519,
      "learning_rate": 8.998663875993666e-06,
      "loss": 0.3542327880859375,
      "num_input_tokens_seen": 195193368,
      "step": 3215,
      "train_runtime": 41896.08,
      "train_tokens_per_second": 4658.989
    },
    {
      "epoch": 0.4107142857142857,
      "grad_norm": 0.5498958620113921,
      "learning_rate": 8.995654736064096e-06,
      "loss": 0.32598466873168946,
      "num_input_tokens_seen": 195489920,
      "step": 3220,
      "train_runtime": 41960.6643,
      "train_tokens_per_second": 4658.885
    },
    {
      "epoch": 0.41135204081632654,
      "grad_norm": 0.6081149945341389,
      "learning_rate": 8.992641586215944e-06,
      "loss": 0.32182776927948,
      "num_input_tokens_seen": 195793216,
      "step": 3225,
      "train_runtime": 42028.6884,
      "train_tokens_per_second": 4658.561
    },
    {
      "epoch": 0.4119897959183674,
      "grad_norm": 0.5562314290478565,
      "learning_rate": 8.989624429473116e-06,
      "loss": 0.34074854850769043,
      "num_input_tokens_seen": 196100528,
      "step": 3230,
      "train_runtime": 42092.2974,
      "train_tokens_per_second": 4658.822
    },
    {
      "epoch": 0.41262755102040816,
      "grad_norm": 0.628203657068124,
      "learning_rate": 8.986603268863536e-06,
      "loss": 0.3286905765533447,
      "num_input_tokens_seen": 196393664,
      "step": 3235,
      "train_runtime": 42156.5376,
      "train_tokens_per_second": 4658.676
    },
    {
      "epoch": 0.413265306122449,
      "grad_norm": 0.6241058236960567,
      "learning_rate": 8.983578107419153e-06,
      "loss": 0.34261574745178225,
      "num_input_tokens_seen": 196697800,
      "step": 3240,
      "train_runtime": 42218.0436,
      "train_tokens_per_second": 4659.093
    },
    {
      "epoch": 0.4139030612244898,
      "grad_norm": 0.6728497794200269,
      "learning_rate": 8.980548948175926e-06,
      "loss": 0.3467216491699219,
      "num_input_tokens_seen": 197005264,
      "step": 3245,
      "train_runtime": 42279.8914,
      "train_tokens_per_second": 4659.55
    },
    {
      "epoch": 0.4145408163265306,
      "grad_norm": 0.8621514340505042,
      "learning_rate": 8.97751579417383e-06,
      "loss": 0.3394784688949585,
      "num_input_tokens_seen": 197297408,
      "step": 3250,
      "train_runtime": 42350.6875,
      "train_tokens_per_second": 4658.659
    },
    {
      "epoch": 0.41517857142857145,
      "grad_norm": 0.6153003291823951,
      "learning_rate": 8.974478648456845e-06,
      "loss": 0.34244298934936523,
      "num_input_tokens_seen": 197610888,
      "step": 3255,
      "train_runtime": 42424.4,
      "train_tokens_per_second": 4657.954
    },
    {
      "epoch": 0.41581632653061223,
      "grad_norm": 0.5896156619261562,
      "learning_rate": 8.971437514072959e-06,
      "loss": 0.33392794132232667,
      "num_input_tokens_seen": 197904104,
      "step": 3260,
      "train_runtime": 42490.4267,
      "train_tokens_per_second": 4657.616
    },
    {
      "epoch": 0.4164540816326531,
      "grad_norm": 0.5514546498909295,
      "learning_rate": 8.968392394074164e-06,
      "loss": 0.3191607713699341,
      "num_input_tokens_seen": 198223760,
      "step": 3265,
      "train_runtime": 42544.331,
      "train_tokens_per_second": 4659.229
    },
    {
      "epoch": 0.41709183673469385,
      "grad_norm": 0.5114826217649231,
      "learning_rate": 8.965343291516448e-06,
      "loss": 0.324001145362854,
      "num_input_tokens_seen": 198520776,
      "step": 3270,
      "train_runtime": 42613.6229,
      "train_tokens_per_second": 4658.622
    },
    {
      "epoch": 0.4177295918367347,
      "grad_norm": 0.5906218319794891,
      "learning_rate": 8.962290209459801e-06,
      "loss": 0.33489012718200684,
      "num_input_tokens_seen": 198828432,
      "step": 3275,
      "train_runtime": 42672.0773,
      "train_tokens_per_second": 4659.451
    },
    {
      "epoch": 0.41836734693877553,
      "grad_norm": 0.8141909804684863,
      "learning_rate": 8.959233150968203e-06,
      "loss": 0.3072980403900146,
      "num_input_tokens_seen": 199125336,
      "step": 3280,
      "train_runtime": 42730.6336,
      "train_tokens_per_second": 4660.014
    },
    {
      "epoch": 0.4190051020408163,
      "grad_norm": 0.5693135294695517,
      "learning_rate": 8.956172119109625e-06,
      "loss": 0.33353633880615235,
      "num_input_tokens_seen": 199438904,
      "step": 3285,
      "train_runtime": 42799.3366,
      "train_tokens_per_second": 4659.86
    },
    {
      "epoch": 0.41964285714285715,
      "grad_norm": 0.5614417137817604,
      "learning_rate": 8.953107116956027e-06,
      "loss": 0.34106614589691164,
      "num_input_tokens_seen": 199752248,
      "step": 3290,
      "train_runtime": 42867.8006,
      "train_tokens_per_second": 4659.727
    },
    {
      "epoch": 0.42028061224489793,
      "grad_norm": 0.5830425466578484,
      "learning_rate": 8.950038147583353e-06,
      "loss": 0.34395198822021483,
      "num_input_tokens_seen": 200059536,
      "step": 3295,
      "train_runtime": 42936.7429,
      "train_tokens_per_second": 4659.402
    },
    {
      "epoch": 0.42091836734693877,
      "grad_norm": 0.5922092142397991,
      "learning_rate": 8.946965214071528e-06,
      "loss": 0.33737869262695314,
      "num_input_tokens_seen": 200353952,
      "step": 3300,
      "train_runtime": 43008.6939,
      "train_tokens_per_second": 4658.452
    },
    {
      "epoch": 0.4215561224489796,
      "grad_norm": 0.7632206076268352,
      "learning_rate": 8.943888319504456e-06,
      "loss": 0.35701279640197753,
      "num_input_tokens_seen": 200664080,
      "step": 3305,
      "train_runtime": 43070.3424,
      "train_tokens_per_second": 4658.985
    },
    {
      "epoch": 0.4221938775510204,
      "grad_norm": 0.6028949445797422,
      "learning_rate": 8.940807466970014e-06,
      "loss": 0.3449193000793457,
      "num_input_tokens_seen": 200978760,
      "step": 3310,
      "train_runtime": 43136.8719,
      "train_tokens_per_second": 4659.094
    },
    {
      "epoch": 0.42283163265306123,
      "grad_norm": 0.5907954504190535,
      "learning_rate": 8.937722659560054e-06,
      "loss": 0.34102916717529297,
      "num_input_tokens_seen": 201280168,
      "step": 3315,
      "train_runtime": 43212.1164,
      "train_tokens_per_second": 4657.957
    },
    {
      "epoch": 0.42346938775510207,
      "grad_norm": 0.5920307060294568,
      "learning_rate": 8.934633900370395e-06,
      "loss": 0.33057827949523927,
      "num_input_tokens_seen": 201590040,
      "step": 3320,
      "train_runtime": 43270.9762,
      "train_tokens_per_second": 4658.782
    },
    {
      "epoch": 0.42410714285714285,
      "grad_norm": 0.6581209511524866,
      "learning_rate": 8.931541192500822e-06,
      "loss": 0.34342374801635744,
      "num_input_tokens_seen": 201892176,
      "step": 3325,
      "train_runtime": 43330.4223,
      "train_tokens_per_second": 4659.363
    },
    {
      "epoch": 0.4247448979591837,
      "grad_norm": 0.5803011854198395,
      "learning_rate": 8.928444539055086e-06,
      "loss": 0.3427203893661499,
      "num_input_tokens_seen": 202205768,
      "step": 3330,
      "train_runtime": 43397.0785,
      "train_tokens_per_second": 4659.433
    },
    {
      "epoch": 0.42538265306122447,
      "grad_norm": 0.6432927089595954,
      "learning_rate": 8.92534394314089e-06,
      "loss": 0.3572682857513428,
      "num_input_tokens_seen": 202521176,
      "step": 3335,
      "train_runtime": 43460.8207,
      "train_tokens_per_second": 4659.856
    },
    {
      "epoch": 0.4260204081632653,
      "grad_norm": 0.7242477526819633,
      "learning_rate": 8.922239407869904e-06,
      "loss": 0.3468234300613403,
      "num_input_tokens_seen": 202832992,
      "step": 3340,
      "train_runtime": 43525.8644,
      "train_tokens_per_second": 4660.057
    },
    {
      "epoch": 0.42665816326530615,
      "grad_norm": 0.6999456822183165,
      "learning_rate": 8.919130936357743e-06,
      "loss": 0.35007805824279786,
      "num_input_tokens_seen": 203131248,
      "step": 3345,
      "train_runtime": 43589.7982,
      "train_tokens_per_second": 4660.064
    },
    {
      "epoch": 0.4272959183673469,
      "grad_norm": 0.6101904475075166,
      "learning_rate": 8.916018531723975e-06,
      "loss": 0.32661712169647217,
      "num_input_tokens_seen": 203441272,
      "step": 3350,
      "train_runtime": 43658.7128,
      "train_tokens_per_second": 4659.809
    },
    {
      "epoch": 0.42793367346938777,
      "grad_norm": 0.6130959340037874,
      "learning_rate": 8.91290219709212e-06,
      "loss": 0.3261438846588135,
      "num_input_tokens_seen": 203750872,
      "step": 3355,
      "train_runtime": 43729.0432,
      "train_tokens_per_second": 4659.395
    },
    {
      "epoch": 0.42857142857142855,
      "grad_norm": 0.6493482786986688,
      "learning_rate": 8.909781935589629e-06,
      "loss": 0.3584270000457764,
      "num_input_tokens_seen": 204067096,
      "step": 3360,
      "train_runtime": 43794.0055,
      "train_tokens_per_second": 4659.704
    },
    {
      "epoch": 0.4292091836734694,
      "grad_norm": 0.5889032217179043,
      "learning_rate": 8.906657750347909e-06,
      "loss": 0.33870534896850585,
      "num_input_tokens_seen": 204383624,
      "step": 3365,
      "train_runtime": 43844.6231,
      "train_tokens_per_second": 4661.544
    },
    {
      "epoch": 0.4298469387755102,
      "grad_norm": 0.5419021099122644,
      "learning_rate": 8.903529644502297e-06,
      "loss": 0.3180680751800537,
      "num_input_tokens_seen": 204692960,
      "step": 3370,
      "train_runtime": 43916.3662,
      "train_tokens_per_second": 4660.972
    },
    {
      "epoch": 0.430484693877551,
      "grad_norm": 0.621544249114791,
      "learning_rate": 8.900397621192063e-06,
      "loss": 0.33640327453613283,
      "num_input_tokens_seen": 204997448,
      "step": 3375,
      "train_runtime": 43991.6741,
      "train_tokens_per_second": 4659.915
    },
    {
      "epoch": 0.43112244897959184,
      "grad_norm": 0.6050089551962087,
      "learning_rate": 8.897261683560413e-06,
      "loss": 0.33349683284759524,
      "num_input_tokens_seen": 205302480,
      "step": 3380,
      "train_runtime": 44065.1786,
      "train_tokens_per_second": 4659.064
    },
    {
      "epoch": 0.4317602040816326,
      "grad_norm": 0.754840342504173,
      "learning_rate": 8.89412183475448e-06,
      "loss": 0.36110312938690187,
      "num_input_tokens_seen": 205609000,
      "step": 3385,
      "train_runtime": 44128.4128,
      "train_tokens_per_second": 4659.334
    },
    {
      "epoch": 0.43239795918367346,
      "grad_norm": 0.5990457935156054,
      "learning_rate": 8.890978077925323e-06,
      "loss": 0.3305992126464844,
      "num_input_tokens_seen": 205906960,
      "step": 3390,
      "train_runtime": 44200.6081,
      "train_tokens_per_second": 4658.464
    },
    {
      "epoch": 0.4330357142857143,
      "grad_norm": 0.5857853241662915,
      "learning_rate": 8.887830416227918e-06,
      "loss": 0.33739218711853025,
      "num_input_tokens_seen": 206210696,
      "step": 3395,
      "train_runtime": 44275.8896,
      "train_tokens_per_second": 4657.404
    },
    {
      "epoch": 0.4336734693877551,
      "grad_norm": 0.6383005925200967,
      "learning_rate": 8.884678852821165e-06,
      "loss": 0.3457144021987915,
      "num_input_tokens_seen": 206510688,
      "step": 3400,
      "train_runtime": 44334.6533,
      "train_tokens_per_second": 4657.997
    },
    {
      "epoch": 0.4343112244897959,
      "grad_norm": 0.6268344964822126,
      "learning_rate": 8.881523390867876e-06,
      "loss": 0.3492433547973633,
      "num_input_tokens_seen": 206826344,
      "step": 3405,
      "train_runtime": 44403.9141,
      "train_tokens_per_second": 4657.84
    },
    {
      "epoch": 0.43494897959183676,
      "grad_norm": 0.6896730273877322,
      "learning_rate": 8.878364033534781e-06,
      "loss": 0.3448661804199219,
      "num_input_tokens_seen": 207133176,
      "step": 3410,
      "train_runtime": 44469.4236,
      "train_tokens_per_second": 4657.879
    },
    {
      "epoch": 0.43558673469387754,
      "grad_norm": 0.5781296268129277,
      "learning_rate": 8.875200783992514e-06,
      "loss": 0.3526355266571045,
      "num_input_tokens_seen": 207443552,
      "step": 3415,
      "train_runtime": 44536.3626,
      "train_tokens_per_second": 4657.847
    },
    {
      "epoch": 0.4362244897959184,
      "grad_norm": 0.5847382276580055,
      "learning_rate": 8.872033645415617e-06,
      "loss": 0.3426992654800415,
      "num_input_tokens_seen": 207753784,
      "step": 3420,
      "train_runtime": 44595.5073,
      "train_tokens_per_second": 4658.626
    },
    {
      "epoch": 0.43686224489795916,
      "grad_norm": 0.5844254850337194,
      "learning_rate": 8.868862620982534e-06,
      "loss": 0.32737252712249754,
      "num_input_tokens_seen": 208057576,
      "step": 3425,
      "train_runtime": 44666.7211,
      "train_tokens_per_second": 4658.0
    },
    {
      "epoch": 0.4375,
      "grad_norm": 0.5693357038747994,
      "learning_rate": 8.86568771387561e-06,
      "loss": 0.35897324085235593,
      "num_input_tokens_seen": 208378240,
      "step": 3430,
      "train_runtime": 44738.6746,
      "train_tokens_per_second": 4657.676
    },
    {
      "epoch": 0.43813775510204084,
      "grad_norm": 0.6292235196885807,
      "learning_rate": 8.862508927281086e-06,
      "loss": 0.3397051334381104,
      "num_input_tokens_seen": 208677608,
      "step": 3435,
      "train_runtime": 44806.0687,
      "train_tokens_per_second": 4657.351
    },
    {
      "epoch": 0.4387755102040816,
      "grad_norm": 0.6656699103580476,
      "learning_rate": 8.859326264389098e-06,
      "loss": 0.3317503690719604,
      "num_input_tokens_seen": 208979040,
      "step": 3440,
      "train_runtime": 44868.7902,
      "train_tokens_per_second": 4657.559
    },
    {
      "epoch": 0.43941326530612246,
      "grad_norm": 0.5913088524568678,
      "learning_rate": 8.856139728393667e-06,
      "loss": 0.32828850746154786,
      "num_input_tokens_seen": 209282536,
      "step": 3445,
      "train_runtime": 44931.7285,
      "train_tokens_per_second": 4657.79
    },
    {
      "epoch": 0.44005102040816324,
      "grad_norm": 0.7367234724225616,
      "learning_rate": 8.852949322492708e-06,
      "loss": 0.3224157810211182,
      "num_input_tokens_seen": 209589696,
      "step": 3450,
      "train_runtime": 45001.3185,
      "train_tokens_per_second": 4657.412
    },
    {
      "epoch": 0.4406887755102041,
      "grad_norm": 0.6518804929508688,
      "learning_rate": 8.849755049888014e-06,
      "loss": 0.3120415687561035,
      "num_input_tokens_seen": 209895168,
      "step": 3455,
      "train_runtime": 45053.3682,
      "train_tokens_per_second": 4658.812
    },
    {
      "epoch": 0.4413265306122449,
      "grad_norm": 0.6329772560530746,
      "learning_rate": 8.84655691378526e-06,
      "loss": 0.35138235092163084,
      "num_input_tokens_seen": 210209408,
      "step": 3460,
      "train_runtime": 45121.5113,
      "train_tokens_per_second": 4658.74
    },
    {
      "epoch": 0.4419642857142857,
      "grad_norm": 0.6083189363119031,
      "learning_rate": 8.843354917394e-06,
      "loss": 0.3106915712356567,
      "num_input_tokens_seen": 210511320,
      "step": 3465,
      "train_runtime": 45185.4567,
      "train_tokens_per_second": 4658.829
    },
    {
      "epoch": 0.44260204081632654,
      "grad_norm": 0.6485137944506096,
      "learning_rate": 8.840149063927663e-06,
      "loss": 0.3368816375732422,
      "num_input_tokens_seen": 210811088,
      "step": 3470,
      "train_runtime": 45254.2994,
      "train_tokens_per_second": 4658.366
    },
    {
      "epoch": 0.4432397959183674,
      "grad_norm": 0.6092074936218119,
      "learning_rate": 8.836939356603542e-06,
      "loss": 0.339404559135437,
      "num_input_tokens_seen": 211116448,
      "step": 3475,
      "train_runtime": 45315.305,
      "train_tokens_per_second": 4658.833
    },
    {
      "epoch": 0.44387755102040816,
      "grad_norm": 0.5633065601403158,
      "learning_rate": 8.833725798642809e-06,
      "loss": 0.3408853530883789,
      "num_input_tokens_seen": 211422040,
      "step": 3480,
      "train_runtime": 45375.3562,
      "train_tokens_per_second": 4659.402
    },
    {
      "epoch": 0.444515306122449,
      "grad_norm": 0.6201839375568762,
      "learning_rate": 8.830508393270487e-06,
      "loss": 0.3162368297576904,
      "num_input_tokens_seen": 211719672,
      "step": 3485,
      "train_runtime": 45443.2728,
      "train_tokens_per_second": 4658.988
    },
    {
      "epoch": 0.4451530612244898,
      "grad_norm": 0.5651255280682073,
      "learning_rate": 8.827287143715472e-06,
      "loss": 0.3251905679702759,
      "num_input_tokens_seen": 212028616,
      "step": 3490,
      "train_runtime": 45509.5767,
      "train_tokens_per_second": 4658.989
    },
    {
      "epoch": 0.4457908163265306,
      "grad_norm": 0.6887442211325981,
      "learning_rate": 8.824062053210512e-06,
      "loss": 0.33139405250549314,
      "num_input_tokens_seen": 212317040,
      "step": 3495,
      "train_runtime": 45582.7374,
      "train_tokens_per_second": 4657.839
    },
    {
      "epoch": 0.44642857142857145,
      "grad_norm": 0.6587881500321826,
      "learning_rate": 8.820833124992207e-06,
      "loss": 0.33382697105407716,
      "num_input_tokens_seen": 212617704,
      "step": 3500,
      "train_runtime": 45649.8867,
      "train_tokens_per_second": 4657.574
    },
    {
      "epoch": 0.44706632653061223,
      "grad_norm": 0.5922595852872055,
      "learning_rate": 8.817600362301018e-06,
      "loss": 0.33900742530822753,
      "num_input_tokens_seen": 212928504,
      "step": 3505,
      "train_runtime": 45698.8399,
      "train_tokens_per_second": 4659.385
    },
    {
      "epoch": 0.4477040816326531,
      "grad_norm": 0.6856015401514906,
      "learning_rate": 8.814363768381242e-06,
      "loss": 0.34448800086975095,
      "num_input_tokens_seen": 213231112,
      "step": 3510,
      "train_runtime": 45756.1646,
      "train_tokens_per_second": 4660.161
    },
    {
      "epoch": 0.44834183673469385,
      "grad_norm": 0.5798228631240803,
      "learning_rate": 8.81112334648103e-06,
      "loss": 0.31914894580841063,
      "num_input_tokens_seen": 213533896,
      "step": 3515,
      "train_runtime": 45826.3218,
      "train_tokens_per_second": 4659.634
    },
    {
      "epoch": 0.4489795918367347,
      "grad_norm": 0.5974213947846045,
      "learning_rate": 8.807879099852372e-06,
      "loss": 0.3257153034210205,
      "num_input_tokens_seen": 213833744,
      "step": 3520,
      "train_runtime": 45886.9257,
      "train_tokens_per_second": 4660.015
    },
    {
      "epoch": 0.44961734693877553,
      "grad_norm": 0.6616558421164055,
      "learning_rate": 8.804631031751096e-06,
      "loss": 0.3289602041244507,
      "num_input_tokens_seen": 214140608,
      "step": 3525,
      "train_runtime": 45955.8381,
      "train_tokens_per_second": 4659.704
    },
    {
      "epoch": 0.4502551020408163,
      "grad_norm": 0.5493349107092007,
      "learning_rate": 8.801379145436866e-06,
      "loss": 0.32906880378723147,
      "num_input_tokens_seen": 214444440,
      "step": 3530,
      "train_runtime": 46018.3078,
      "train_tokens_per_second": 4659.981
    },
    {
      "epoch": 0.45089285714285715,
      "grad_norm": 0.5863711108445256,
      "learning_rate": 8.798123444173174e-06,
      "loss": 0.3326809167861938,
      "num_input_tokens_seen": 214764408,
      "step": 3535,
      "train_runtime": 46079.5232,
      "train_tokens_per_second": 4660.734
    },
    {
      "epoch": 0.45153061224489793,
      "grad_norm": 0.5692545931440874,
      "learning_rate": 8.794863931227346e-06,
      "loss": 0.3165797233581543,
      "num_input_tokens_seen": 215066664,
      "step": 3540,
      "train_runtime": 46138.4384,
      "train_tokens_per_second": 4661.334
    },
    {
      "epoch": 0.45216836734693877,
      "grad_norm": 0.590150731385412,
      "learning_rate": 8.79160060987053e-06,
      "loss": 0.32453289031982424,
      "num_input_tokens_seen": 215375520,
      "step": 3545,
      "train_runtime": 46197.8852,
      "train_tokens_per_second": 4662.021
    },
    {
      "epoch": 0.4528061224489796,
      "grad_norm": 0.6298776588000911,
      "learning_rate": 8.788333483377699e-06,
      "loss": 0.33482890129089354,
      "num_input_tokens_seen": 215684472,
      "step": 3550,
      "train_runtime": 46264.3918,
      "train_tokens_per_second": 4661.997
    },
    {
      "epoch": 0.4534438775510204,
      "grad_norm": 0.7164484415230423,
      "learning_rate": 8.785062555027637e-06,
      "loss": 0.32947793006896975,
      "num_input_tokens_seen": 215978352,
      "step": 3555,
      "train_runtime": 46332.379,
      "train_tokens_per_second": 4661.499
    },
    {
      "epoch": 0.45408163265306123,
      "grad_norm": 0.6005379480462399,
      "learning_rate": 8.781787828102958e-06,
      "loss": 0.3068972587585449,
      "num_input_tokens_seen": 216279984,
      "step": 3560,
      "train_runtime": 46396.3682,
      "train_tokens_per_second": 4661.571
    },
    {
      "epoch": 0.45471938775510207,
      "grad_norm": 0.628236753497089,
      "learning_rate": 8.778509305890069e-06,
      "loss": 0.3233363151550293,
      "num_input_tokens_seen": 216587936,
      "step": 3565,
      "train_runtime": 46461.1093,
      "train_tokens_per_second": 4661.704
    },
    {
      "epoch": 0.45535714285714285,
      "grad_norm": 0.6604944018759947,
      "learning_rate": 8.775226991679205e-06,
      "loss": 0.31778683662414553,
      "num_input_tokens_seen": 216883960,
      "step": 3570,
      "train_runtime": 46527.9806,
      "train_tokens_per_second": 4661.366
    },
    {
      "epoch": 0.4559948979591837,
      "grad_norm": 0.6045429187096011,
      "learning_rate": 8.77194088876439e-06,
      "loss": 0.359822940826416,
      "num_input_tokens_seen": 217190424,
      "step": 3575,
      "train_runtime": 46600.9421,
      "train_tokens_per_second": 4660.644
    },
    {
      "epoch": 0.45663265306122447,
      "grad_norm": 0.6093154195881341,
      "learning_rate": 8.768651000443463e-06,
      "loss": 0.3432753086090088,
      "num_input_tokens_seen": 217492848,
      "step": 3580,
      "train_runtime": 46668.3348,
      "train_tokens_per_second": 4660.394
    },
    {
      "epoch": 0.4572704081632653,
      "grad_norm": 0.6020313159495515,
      "learning_rate": 8.765357330018056e-06,
      "loss": 0.3330278158187866,
      "num_input_tokens_seen": 217802264,
      "step": 3585,
      "train_runtime": 46733.4455,
      "train_tokens_per_second": 4660.522
    },
    {
      "epoch": 0.45790816326530615,
      "grad_norm": 0.592677521696768,
      "learning_rate": 8.762059880793595e-06,
      "loss": 0.31818888187408445,
      "num_input_tokens_seen": 218101968,
      "step": 3590,
      "train_runtime": 46802.4783,
      "train_tokens_per_second": 4660.052
    },
    {
      "epoch": 0.4585459183673469,
      "grad_norm": 0.6090140714152729,
      "learning_rate": 8.758758656079303e-06,
      "loss": 0.34148919582366943,
      "num_input_tokens_seen": 218407680,
      "step": 3595,
      "train_runtime": 46858.3375,
      "train_tokens_per_second": 4661.021
    },
    {
      "epoch": 0.45918367346938777,
      "grad_norm": 0.58287892611014,
      "learning_rate": 8.755453659188186e-06,
      "loss": 0.34990203380584717,
      "num_input_tokens_seen": 218713272,
      "step": 3600,
      "train_runtime": 46923.1988,
      "train_tokens_per_second": 4661.09
    },
    {
      "epoch": 0.45982142857142855,
      "grad_norm": 0.5833916162586771,
      "learning_rate": 8.752144893437045e-06,
      "loss": 0.3203828573226929,
      "num_input_tokens_seen": 219022144,
      "step": 3605,
      "train_runtime": 46993.4779,
      "train_tokens_per_second": 4660.692
    },
    {
      "epoch": 0.4604591836734694,
      "grad_norm": 0.5956637639825637,
      "learning_rate": 8.748832362146454e-06,
      "loss": 0.32918357849121094,
      "num_input_tokens_seen": 219326448,
      "step": 3610,
      "train_runtime": 47068.5692,
      "train_tokens_per_second": 4659.722
    },
    {
      "epoch": 0.4610969387755102,
      "grad_norm": 0.664499572011986,
      "learning_rate": 8.745516068640767e-06,
      "loss": 0.3219877243041992,
      "num_input_tokens_seen": 219624856,
      "step": 3615,
      "train_runtime": 47131.7287,
      "train_tokens_per_second": 4659.809
    },
    {
      "epoch": 0.461734693877551,
      "grad_norm": 0.6569114261470931,
      "learning_rate": 8.742196016248121e-06,
      "loss": 0.3326474666595459,
      "num_input_tokens_seen": 219925872,
      "step": 3620,
      "train_runtime": 47186.9921,
      "train_tokens_per_second": 4660.731
    },
    {
      "epoch": 0.46237244897959184,
      "grad_norm": 0.7617400027936144,
      "learning_rate": 8.738872208300417e-06,
      "loss": 0.3078127384185791,
      "num_input_tokens_seen": 220214592,
      "step": 3625,
      "train_runtime": 47254.858,
      "train_tokens_per_second": 4660.147
    },
    {
      "epoch": 0.4630102040816326,
      "grad_norm": 0.6347466685079263,
      "learning_rate": 8.735544648133331e-06,
      "loss": 0.3348182439804077,
      "num_input_tokens_seen": 220521248,
      "step": 3630,
      "train_runtime": 47320.6913,
      "train_tokens_per_second": 4660.144
    },
    {
      "epoch": 0.46364795918367346,
      "grad_norm": 0.6211229675216676,
      "learning_rate": 8.7322133390863e-06,
      "loss": 0.35152287483215333,
      "num_input_tokens_seen": 220810352,
      "step": 3635,
      "train_runtime": 47394.397,
      "train_tokens_per_second": 4658.997
    },
    {
      "epoch": 0.4642857142857143,
      "grad_norm": 0.7067748043120594,
      "learning_rate": 8.728878284502526e-06,
      "loss": 0.3194485902786255,
      "num_input_tokens_seen": 221106400,
      "step": 3640,
      "train_runtime": 47461.2057,
      "train_tokens_per_second": 4658.676
    },
    {
      "epoch": 0.4649234693877551,
      "grad_norm": 0.6362789992815582,
      "learning_rate": 8.725539487728969e-06,
      "loss": 0.33799190521240235,
      "num_input_tokens_seen": 221420712,
      "step": 3645,
      "train_runtime": 47526.8595,
      "train_tokens_per_second": 4658.854
    },
    {
      "epoch": 0.4655612244897959,
      "grad_norm": 0.5918305492650939,
      "learning_rate": 8.722196952116346e-06,
      "loss": 0.31893162727355956,
      "num_input_tokens_seen": 221725936,
      "step": 3650,
      "train_runtime": 47592.967,
      "train_tokens_per_second": 4658.796
    },
    {
      "epoch": 0.46619897959183676,
      "grad_norm": 0.6187216060217506,
      "learning_rate": 8.718850681019125e-06,
      "loss": 0.3435813426971436,
      "num_input_tokens_seen": 222021640,
      "step": 3655,
      "train_runtime": 47658.7045,
      "train_tokens_per_second": 4658.575
    },
    {
      "epoch": 0.46683673469387754,
      "grad_norm": 0.6007532120518013,
      "learning_rate": 8.715500677795519e-06,
      "loss": 0.3304643392562866,
      "num_input_tokens_seen": 222314448,
      "step": 3660,
      "train_runtime": 47737.5544,
      "train_tokens_per_second": 4657.014
    },
    {
      "epoch": 0.4674744897959184,
      "grad_norm": 0.6159209527989902,
      "learning_rate": 8.712146945807494e-06,
      "loss": 0.3191221714019775,
      "num_input_tokens_seen": 222623648,
      "step": 3665,
      "train_runtime": 47795.5768,
      "train_tokens_per_second": 4657.83
    },
    {
      "epoch": 0.46811224489795916,
      "grad_norm": 0.6731481661966066,
      "learning_rate": 8.708789488420753e-06,
      "loss": 0.32197272777557373,
      "num_input_tokens_seen": 222922432,
      "step": 3670,
      "train_runtime": 47870.6142,
      "train_tokens_per_second": 4656.77
    },
    {
      "epoch": 0.46875,
      "grad_norm": 0.6757831485969278,
      "learning_rate": 8.705428309004737e-06,
      "loss": 0.31526567935943606,
      "num_input_tokens_seen": 223229936,
      "step": 3675,
      "train_runtime": 47938.8509,
      "train_tokens_per_second": 4656.556
    },
    {
      "epoch": 0.46938775510204084,
      "grad_norm": 0.7112337740632515,
      "learning_rate": 8.702063410932628e-06,
      "loss": 0.3370952606201172,
      "num_input_tokens_seen": 223528488,
      "step": 3680,
      "train_runtime": 48014.0837,
      "train_tokens_per_second": 4655.478
    },
    {
      "epoch": 0.4700255102040816,
      "grad_norm": 0.6470080319086245,
      "learning_rate": 8.698694797581335e-06,
      "loss": 0.30045714378356936,
      "num_input_tokens_seen": 223821368,
      "step": 3685,
      "train_runtime": 48088.7204,
      "train_tokens_per_second": 4654.342
    },
    {
      "epoch": 0.47066326530612246,
      "grad_norm": 0.6107567931135042,
      "learning_rate": 8.695322472331497e-06,
      "loss": 0.34927806854248045,
      "num_input_tokens_seen": 224128912,
      "step": 3690,
      "train_runtime": 48154.5956,
      "train_tokens_per_second": 4654.362
    },
    {
      "epoch": 0.47130102040816324,
      "grad_norm": 0.5961638372408326,
      "learning_rate": 8.691946438567476e-06,
      "loss": 0.3163304328918457,
      "num_input_tokens_seen": 224441784,
      "step": 3695,
      "train_runtime": 48208.6294,
      "train_tokens_per_second": 4655.635
    },
    {
      "epoch": 0.4719387755102041,
      "grad_norm": 0.6182783846051844,
      "learning_rate": 8.688566699677362e-06,
      "loss": 0.33351778984069824,
      "num_input_tokens_seen": 224754568,
      "step": 3700,
      "train_runtime": 48261.291,
      "train_tokens_per_second": 4657.036
    },
    {
      "epoch": 0.4725765306122449,
      "grad_norm": 0.6391520103763577,
      "learning_rate": 8.685183259052951e-06,
      "loss": 0.32537355422973635,
      "num_input_tokens_seen": 225050008,
      "step": 3705,
      "train_runtime": 48328.626,
      "train_tokens_per_second": 4656.661
    },
    {
      "epoch": 0.4732142857142857,
      "grad_norm": 0.6602857375485054,
      "learning_rate": 8.68179612008977e-06,
      "loss": 0.3055689334869385,
      "num_input_tokens_seen": 225357160,
      "step": 3710,
      "train_runtime": 48391.4814,
      "train_tokens_per_second": 4656.959
    },
    {
      "epoch": 0.47385204081632654,
      "grad_norm": 0.547557456378693,
      "learning_rate": 8.678405286187046e-06,
      "loss": 0.32492766380310056,
      "num_input_tokens_seen": 225664864,
      "step": 3715,
      "train_runtime": 48447.0383,
      "train_tokens_per_second": 4657.97
    },
    {
      "epoch": 0.4744897959183674,
      "grad_norm": 0.5719022452845505,
      "learning_rate": 8.675010760747717e-06,
      "loss": 0.31012744903564454,
      "num_input_tokens_seen": 225978576,
      "step": 3720,
      "train_runtime": 48511.6481,
      "train_tokens_per_second": 4658.233
    },
    {
      "epoch": 0.47512755102040816,
      "grad_norm": 0.5936112059215125,
      "learning_rate": 8.671612547178428e-06,
      "loss": 0.35015435218811036,
      "num_input_tokens_seen": 226288696,
      "step": 3725,
      "train_runtime": 48576.1779,
      "train_tokens_per_second": 4658.429
    },
    {
      "epoch": 0.475765306122449,
      "grad_norm": 0.605088400054027,
      "learning_rate": 8.668210648889523e-06,
      "loss": 0.3316817045211792,
      "num_input_tokens_seen": 226587976,
      "step": 3730,
      "train_runtime": 48644.9087,
      "train_tokens_per_second": 4658.0
    },
    {
      "epoch": 0.4764030612244898,
      "grad_norm": 0.7170426138582027,
      "learning_rate": 8.664805069295046e-06,
      "loss": 0.31617088317871095,
      "num_input_tokens_seen": 226879712,
      "step": 3735,
      "train_runtime": 48707.6769,
      "train_tokens_per_second": 4657.987
    },
    {
      "epoch": 0.4770408163265306,
      "grad_norm": 0.5225170151134908,
      "learning_rate": 8.661395811812732e-06,
      "loss": 0.3001605987548828,
      "num_input_tokens_seen": 227183312,
      "step": 3740,
      "train_runtime": 48768.7211,
      "train_tokens_per_second": 4658.382
    },
    {
      "epoch": 0.47767857142857145,
      "grad_norm": 0.657497902829899,
      "learning_rate": 8.657982879864007e-06,
      "loss": 0.32870779037475584,
      "num_input_tokens_seen": 227483608,
      "step": 3745,
      "train_runtime": 48836.0865,
      "train_tokens_per_second": 4658.105
    },
    {
      "epoch": 0.47831632653061223,
      "grad_norm": 0.7152430489686709,
      "learning_rate": 8.654566276873992e-06,
      "loss": 0.3120705604553223,
      "num_input_tokens_seen": 227777824,
      "step": 3750,
      "train_runtime": 48905.4388,
      "train_tokens_per_second": 4657.515
    },
    {
      "epoch": 0.4789540816326531,
      "grad_norm": 0.6011045497669353,
      "learning_rate": 8.651146006271483e-06,
      "loss": 0.33093512058258057,
      "num_input_tokens_seen": 228081080,
      "step": 3755,
      "train_runtime": 48977.8418,
      "train_tokens_per_second": 4656.822
    },
    {
      "epoch": 0.47959183673469385,
      "grad_norm": 0.6013894891716157,
      "learning_rate": 8.64772207148896e-06,
      "loss": 0.33283772468566897,
      "num_input_tokens_seen": 228389848,
      "step": 3760,
      "train_runtime": 49050.6235,
      "train_tokens_per_second": 4656.207
    },
    {
      "epoch": 0.4802295918367347,
      "grad_norm": 0.6456911475911199,
      "learning_rate": 8.644294475962583e-06,
      "loss": 0.33153154850006106,
      "num_input_tokens_seen": 228686136,
      "step": 3765,
      "train_runtime": 49115.0176,
      "train_tokens_per_second": 4656.135
    },
    {
      "epoch": 0.48086734693877553,
      "grad_norm": 0.5802838347857451,
      "learning_rate": 8.640863223132183e-06,
      "loss": 0.3430925846099854,
      "num_input_tokens_seen": 228987808,
      "step": 3770,
      "train_runtime": 49175.4255,
      "train_tokens_per_second": 4656.55
    },
    {
      "epoch": 0.4815051020408163,
      "grad_norm": 0.5944701201279429,
      "learning_rate": 8.63742831644126e-06,
      "loss": 0.3659370899200439,
      "num_input_tokens_seen": 229298856,
      "step": 3775,
      "train_runtime": 49242.4393,
      "train_tokens_per_second": 4656.529
    },
    {
      "epoch": 0.48214285714285715,
      "grad_norm": 0.6364665335788862,
      "learning_rate": 8.633989759336984e-06,
      "loss": 0.34424939155578616,
      "num_input_tokens_seen": 229600528,
      "step": 3780,
      "train_runtime": 49310.6682,
      "train_tokens_per_second": 4656.204
    },
    {
      "epoch": 0.48278061224489793,
      "grad_norm": 0.7089879214400359,
      "learning_rate": 8.630547555270187e-06,
      "loss": 0.3644766330718994,
      "num_input_tokens_seen": 229905632,
      "step": 3785,
      "train_runtime": 49372.6658,
      "train_tokens_per_second": 4656.537
    },
    {
      "epoch": 0.48341836734693877,
      "grad_norm": 0.6271171405674849,
      "learning_rate": 8.627101707695365e-06,
      "loss": 0.32773098945617674,
      "num_input_tokens_seen": 230215496,
      "step": 3790,
      "train_runtime": 49435.6049,
      "train_tokens_per_second": 4656.876
    },
    {
      "epoch": 0.4840561224489796,
      "grad_norm": 0.5629236117996881,
      "learning_rate": 8.62365222007066e-06,
      "loss": 0.3321268320083618,
      "num_input_tokens_seen": 230515376,
      "step": 3795,
      "train_runtime": 49493.6597,
      "train_tokens_per_second": 4657.473
    },
    {
      "epoch": 0.4846938775510204,
      "grad_norm": 0.5580576774068381,
      "learning_rate": 8.620199095857877e-06,
      "loss": 0.3428853988647461,
      "num_input_tokens_seen": 230814064,
      "step": 3800,
      "train_runtime": 49564.9336,
      "train_tokens_per_second": 4656.802
    },
    {
      "epoch": 0.48533163265306123,
      "grad_norm": 0.6734770527934579,
      "learning_rate": 8.616742338522467e-06,
      "loss": 0.3223045587539673,
      "num_input_tokens_seen": 231106232,
      "step": 3805,
      "train_runtime": 49624.0942,
      "train_tokens_per_second": 4657.138
    },
    {
      "epoch": 0.48596938775510207,
      "grad_norm": 0.6027423188513089,
      "learning_rate": 8.613281951533527e-06,
      "loss": 0.30772149562835693,
      "num_input_tokens_seen": 231418568,
      "step": 3810,
      "train_runtime": 49691.9282,
      "train_tokens_per_second": 4657.066
    },
    {
      "epoch": 0.48660714285714285,
      "grad_norm": 0.620730028536161,
      "learning_rate": 8.609817938363799e-06,
      "loss": 0.3534107685089111,
      "num_input_tokens_seen": 231724240,
      "step": 3815,
      "train_runtime": 49749.4493,
      "train_tokens_per_second": 4657.825
    },
    {
      "epoch": 0.4872448979591837,
      "grad_norm": 0.6023564046787745,
      "learning_rate": 8.60635030248966e-06,
      "loss": 0.3601125001907349,
      "num_input_tokens_seen": 232033184,
      "step": 3820,
      "train_runtime": 49820.1508,
      "train_tokens_per_second": 4657.416
    },
    {
      "epoch": 0.48788265306122447,
      "grad_norm": 0.5820284956858566,
      "learning_rate": 8.602879047391127e-06,
      "loss": 0.3287564992904663,
      "num_input_tokens_seen": 232332512,
      "step": 3825,
      "train_runtime": 49887.8581,
      "train_tokens_per_second": 4657.095
    },
    {
      "epoch": 0.4885204081632653,
      "grad_norm": 0.5435262720901459,
      "learning_rate": 8.599404176551843e-06,
      "loss": 0.33783981800079343,
      "num_input_tokens_seen": 232642256,
      "step": 3830,
      "train_runtime": 49961.0696,
      "train_tokens_per_second": 4656.471
    },
    {
      "epoch": 0.48915816326530615,
      "grad_norm": 0.5898348531043288,
      "learning_rate": 8.595925693459086e-06,
      "loss": 0.32908217906951903,
      "num_input_tokens_seen": 232936208,
      "step": 3835,
      "train_runtime": 50029.4144,
      "train_tokens_per_second": 4655.985
    },
    {
      "epoch": 0.4897959183673469,
      "grad_norm": 0.6777176694788429,
      "learning_rate": 8.59244360160376e-06,
      "loss": 0.3408887624740601,
      "num_input_tokens_seen": 233254512,
      "step": 3840,
      "train_runtime": 50082.2953,
      "train_tokens_per_second": 4657.425
    },
    {
      "epoch": 0.49043367346938777,
      "grad_norm": 0.6490802984424683,
      "learning_rate": 8.588957904480381e-06,
      "loss": 0.3298821449279785,
      "num_input_tokens_seen": 233551432,
      "step": 3845,
      "train_runtime": 50153.4482,
      "train_tokens_per_second": 4656.737
    },
    {
      "epoch": 0.49107142857142855,
      "grad_norm": 0.644982223462375,
      "learning_rate": 8.585468605587094e-06,
      "loss": 0.3687849521636963,
      "num_input_tokens_seen": 233836184,
      "step": 3850,
      "train_runtime": 50221.8557,
      "train_tokens_per_second": 4656.064
    },
    {
      "epoch": 0.4917091836734694,
      "grad_norm": 0.6266169141222809,
      "learning_rate": 8.581975708425652e-06,
      "loss": 0.34029111862182615,
      "num_input_tokens_seen": 234149936,
      "step": 3855,
      "train_runtime": 50293.7681,
      "train_tokens_per_second": 4655.645
    },
    {
      "epoch": 0.4923469387755102,
      "grad_norm": 0.574291085519132,
      "learning_rate": 8.578479216501422e-06,
      "loss": 0.32351346015930177,
      "num_input_tokens_seen": 234445496,
      "step": 3860,
      "train_runtime": 50366.3962,
      "train_tokens_per_second": 4654.8
    },
    {
      "epoch": 0.492984693877551,
      "grad_norm": 0.5552501971461397,
      "learning_rate": 8.574979133323378e-06,
      "loss": 0.3432778835296631,
      "num_input_tokens_seen": 234765680,
      "step": 3865,
      "train_runtime": 50433.1949,
      "train_tokens_per_second": 4654.983
    },
    {
      "epoch": 0.49362244897959184,
      "grad_norm": 0.5766227232875598,
      "learning_rate": 8.571475462404098e-06,
      "loss": 0.3516408443450928,
      "num_input_tokens_seen": 235072624,
      "step": 3870,
      "train_runtime": 50496.9399,
      "train_tokens_per_second": 4655.186
    },
    {
      "epoch": 0.4942602040816326,
      "grad_norm": 0.6372292409084065,
      "learning_rate": 8.567968207259759e-06,
      "loss": 0.3464944839477539,
      "num_input_tokens_seen": 235380696,
      "step": 3875,
      "train_runtime": 50554.7986,
      "train_tokens_per_second": 4655.952
    },
    {
      "epoch": 0.49489795918367346,
      "grad_norm": 0.6023486727101804,
      "learning_rate": 8.564457371410139e-06,
      "loss": 0.3397261619567871,
      "num_input_tokens_seen": 235692760,
      "step": 3880,
      "train_runtime": 50612.5342,
      "train_tokens_per_second": 4656.806
    },
    {
      "epoch": 0.4955357142857143,
      "grad_norm": 0.6552499797293001,
      "learning_rate": 8.560942958378604e-06,
      "loss": 0.34970316886901853,
      "num_input_tokens_seen": 236001480,
      "step": 3885,
      "train_runtime": 50676.4157,
      "train_tokens_per_second": 4657.028
    },
    {
      "epoch": 0.4961734693877551,
      "grad_norm": 0.5336817905647219,
      "learning_rate": 8.557424971692117e-06,
      "loss": 0.35130982398986815,
      "num_input_tokens_seen": 236306648,
      "step": 3890,
      "train_runtime": 50747.1448,
      "train_tokens_per_second": 4656.551
    },
    {
      "epoch": 0.4968112244897959,
      "grad_norm": 0.5527697429127871,
      "learning_rate": 8.55390341488122e-06,
      "loss": 0.3193791389465332,
      "num_input_tokens_seen": 236604488,
      "step": 3895,
      "train_runtime": 50810.3713,
      "train_tokens_per_second": 4656.618
    },
    {
      "epoch": 0.49744897959183676,
      "grad_norm": 0.6175124900145508,
      "learning_rate": 8.550378291480041e-06,
      "loss": 0.3339862823486328,
      "num_input_tokens_seen": 236905912,
      "step": 3900,
      "train_runtime": 50871.7258,
      "train_tokens_per_second": 4656.927
    },
    {
      "epoch": 0.49808673469387754,
      "grad_norm": 0.6822366523915481,
      "learning_rate": 8.54684960502629e-06,
      "loss": 0.32977321147918703,
      "num_input_tokens_seen": 237197104,
      "step": 3905,
      "train_runtime": 50934.5909,
      "train_tokens_per_second": 4656.896
    },
    {
      "epoch": 0.4987244897959184,
      "grad_norm": 0.6573747525944167,
      "learning_rate": 8.543317359061247e-06,
      "loss": 0.32253079414367675,
      "num_input_tokens_seen": 237505528,
      "step": 3910,
      "train_runtime": 50996.1654,
      "train_tokens_per_second": 4657.321
    },
    {
      "epoch": 0.49936224489795916,
      "grad_norm": 0.6080138587125168,
      "learning_rate": 8.539781557129772e-06,
      "loss": 0.3567155361175537,
      "num_input_tokens_seen": 237813768,
      "step": 3915,
      "train_runtime": 51057.7262,
      "train_tokens_per_second": 4657.743
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.6502081197077322,
      "learning_rate": 8.536242202780285e-06,
      "loss": 0.34138736724853513,
      "num_input_tokens_seen": 238128160,
      "step": 3920,
      "train_runtime": 51116.414,
      "train_tokens_per_second": 4658.546
    },
    {
      "epoch": 0.5006377551020408,
      "grad_norm": 0.6123351154071554,
      "learning_rate": 8.532699299564777e-06,
      "loss": 0.33693814277648926,
      "num_input_tokens_seen": 238424680,
      "step": 3925,
      "train_runtime": 51189.0324,
      "train_tokens_per_second": 4657.73
    },
    {
      "epoch": 0.5012755102040817,
      "grad_norm": 0.5962592524709964,
      "learning_rate": 8.529152851038798e-06,
      "loss": 0.3420262813568115,
      "num_input_tokens_seen": 238728160,
      "step": 3930,
      "train_runtime": 51248.7541,
      "train_tokens_per_second": 4658.224
    },
    {
      "epoch": 0.5019132653061225,
      "grad_norm": 0.6452137863780961,
      "learning_rate": 8.52560286076146e-06,
      "loss": 0.3414464473724365,
      "num_input_tokens_seen": 239026280,
      "step": 3935,
      "train_runtime": 51312.8956,
      "train_tokens_per_second": 4658.211
    },
    {
      "epoch": 0.5025510204081632,
      "grad_norm": 0.5286085457622516,
      "learning_rate": 8.522049332295421e-06,
      "loss": 0.3040300369262695,
      "num_input_tokens_seen": 239330112,
      "step": 3940,
      "train_runtime": 51375.0974,
      "train_tokens_per_second": 4658.485
    },
    {
      "epoch": 0.5031887755102041,
      "grad_norm": 0.5907977823780266,
      "learning_rate": 8.5184922692069e-06,
      "loss": 0.30737330913543703,
      "num_input_tokens_seen": 239623416,
      "step": 3945,
      "train_runtime": 51442.1079,
      "train_tokens_per_second": 4658.118
    },
    {
      "epoch": 0.5038265306122449,
      "grad_norm": 0.6253963561292852,
      "learning_rate": 8.514931675065654e-06,
      "loss": 0.34375686645507814,
      "num_input_tokens_seen": 239949088,
      "step": 3950,
      "train_runtime": 51497.1449,
      "train_tokens_per_second": 4659.464
    },
    {
      "epoch": 0.5044642857142857,
      "grad_norm": 0.5817830506603652,
      "learning_rate": 8.51136755344499e-06,
      "loss": 0.3413503646850586,
      "num_input_tokens_seen": 240249752,
      "step": 3955,
      "train_runtime": 51570.0528,
      "train_tokens_per_second": 4658.707
    },
    {
      "epoch": 0.5051020408163265,
      "grad_norm": 0.6140954907942193,
      "learning_rate": 8.507799907921754e-06,
      "loss": 0.34429771900177003,
      "num_input_tokens_seen": 240561680,
      "step": 3960,
      "train_runtime": 51634.5519,
      "train_tokens_per_second": 4658.928
    },
    {
      "epoch": 0.5057397959183674,
      "grad_norm": 0.6027612668326606,
      "learning_rate": 8.504228742076325e-06,
      "loss": 0.2996108055114746,
      "num_input_tokens_seen": 240872848,
      "step": 3965,
      "train_runtime": 51701.6938,
      "train_tokens_per_second": 4658.897
    },
    {
      "epoch": 0.5063775510204082,
      "grad_norm": 0.5832533965594051,
      "learning_rate": 8.500654059492618e-06,
      "loss": 0.3469754934310913,
      "num_input_tokens_seen": 241173696,
      "step": 3970,
      "train_runtime": 51772.4157,
      "train_tokens_per_second": 4658.343
    },
    {
      "epoch": 0.5070153061224489,
      "grad_norm": 0.598068625125809,
      "learning_rate": 8.497075863758079e-06,
      "loss": 0.33405184745788574,
      "num_input_tokens_seen": 241476848,
      "step": 3975,
      "train_runtime": 51826.4034,
      "train_tokens_per_second": 4659.34
    },
    {
      "epoch": 0.5076530612244898,
      "grad_norm": 0.6090274485920489,
      "learning_rate": 8.493494158463674e-06,
      "loss": 0.34064507484436035,
      "num_input_tokens_seen": 241769888,
      "step": 3980,
      "train_runtime": 51889.3028,
      "train_tokens_per_second": 4659.34
    },
    {
      "epoch": 0.5082908163265306,
      "grad_norm": 0.5764020632951243,
      "learning_rate": 8.489908947203897e-06,
      "loss": 0.32036867141723635,
      "num_input_tokens_seen": 242074176,
      "step": 3985,
      "train_runtime": 51952.744,
      "train_tokens_per_second": 4659.507
    },
    {
      "epoch": 0.5089285714285714,
      "grad_norm": 0.7242092475659759,
      "learning_rate": 8.486320233576757e-06,
      "loss": 0.33285179138183596,
      "num_input_tokens_seen": 242362680,
      "step": 3990,
      "train_runtime": 52017.5364,
      "train_tokens_per_second": 4659.249
    },
    {
      "epoch": 0.5095663265306123,
      "grad_norm": 0.6547031028417888,
      "learning_rate": 8.482728021183777e-06,
      "loss": 0.3297649145126343,
      "num_input_tokens_seen": 242654240,
      "step": 3995,
      "train_runtime": 52078.2091,
      "train_tokens_per_second": 4659.42
    },
    {
      "epoch": 0.5102040816326531,
      "grad_norm": 0.6358861563754288,
      "learning_rate": 8.479132313629996e-06,
      "loss": 0.3276358127593994,
      "num_input_tokens_seen": 242954720,
      "step": 4000,
      "train_runtime": 52148.0392,
      "train_tokens_per_second": 4658.943
    },
    {
      "epoch": 0.5108418367346939,
      "grad_norm": 0.5814639689939518,
      "learning_rate": 8.475533114523954e-06,
      "loss": 0.3316447019577026,
      "num_input_tokens_seen": 243266952,
      "step": 4005,
      "train_runtime": 52208.9272,
      "train_tokens_per_second": 4659.49
    },
    {
      "epoch": 0.5114795918367347,
      "grad_norm": 0.5811701828500324,
      "learning_rate": 8.471930427477704e-06,
      "loss": 0.34278225898742676,
      "num_input_tokens_seen": 243562512,
      "step": 4010,
      "train_runtime": 52274.7962,
      "train_tokens_per_second": 4659.272
    },
    {
      "epoch": 0.5121173469387755,
      "grad_norm": 0.8127426829512003,
      "learning_rate": 8.468324256106789e-06,
      "loss": 0.3474293231964111,
      "num_input_tokens_seen": 243867856,
      "step": 4015,
      "train_runtime": 52319.5304,
      "train_tokens_per_second": 4661.125
    },
    {
      "epoch": 0.5127551020408163,
      "grad_norm": 0.6079638254587282,
      "learning_rate": 8.464714604030256e-06,
      "loss": 0.3270159482955933,
      "num_input_tokens_seen": 244155792,
      "step": 4020,
      "train_runtime": 52377.0182,
      "train_tokens_per_second": 4661.506
    },
    {
      "epoch": 0.5133928571428571,
      "grad_norm": 0.6174658791855551,
      "learning_rate": 8.461101474870642e-06,
      "loss": 0.32132925987243655,
      "num_input_tokens_seen": 244454032,
      "step": 4025,
      "train_runtime": 52436.0585,
      "train_tokens_per_second": 4661.945
    },
    {
      "epoch": 0.514030612244898,
      "grad_norm": 0.6180605320426141,
      "learning_rate": 8.457484872253976e-06,
      "loss": 0.33786468505859374,
      "num_input_tokens_seen": 244756848,
      "step": 4030,
      "train_runtime": 52510.39,
      "train_tokens_per_second": 4661.113
    },
    {
      "epoch": 0.5146683673469388,
      "grad_norm": 0.6256830321428409,
      "learning_rate": 8.453864799809772e-06,
      "loss": 0.3316938877105713,
      "num_input_tokens_seen": 245066784,
      "step": 4035,
      "train_runtime": 52566.0158,
      "train_tokens_per_second": 4662.076
    },
    {
      "epoch": 0.5153061224489796,
      "grad_norm": 0.5528575316724593,
      "learning_rate": 8.450241261171022e-06,
      "loss": 0.3056922435760498,
      "num_input_tokens_seen": 245349960,
      "step": 4040,
      "train_runtime": 52633.982,
      "train_tokens_per_second": 4661.436
    },
    {
      "epoch": 0.5159438775510204,
      "grad_norm": 0.5701067986347147,
      "learning_rate": 8.446614259974201e-06,
      "loss": 0.3409639835357666,
      "num_input_tokens_seen": 245655704,
      "step": 4045,
      "train_runtime": 52695.1388,
      "train_tokens_per_second": 4661.829
    },
    {
      "epoch": 0.5165816326530612,
      "grad_norm": 0.5770814423382815,
      "learning_rate": 8.44298379985926e-06,
      "loss": 0.32842864990234377,
      "num_input_tokens_seen": 245951352,
      "step": 4050,
      "train_runtime": 52762.8471,
      "train_tokens_per_second": 4661.45
    },
    {
      "epoch": 0.517219387755102,
      "grad_norm": 0.583521188324923,
      "learning_rate": 8.439349884469621e-06,
      "loss": 0.32632737159729003,
      "num_input_tokens_seen": 246260984,
      "step": 4055,
      "train_runtime": 52838.9104,
      "train_tokens_per_second": 4660.599
    },
    {
      "epoch": 0.5178571428571429,
      "grad_norm": 0.5831683900063419,
      "learning_rate": 8.435712517452167e-06,
      "loss": 0.3498189926147461,
      "num_input_tokens_seen": 246580168,
      "step": 4060,
      "train_runtime": 52901.2621,
      "train_tokens_per_second": 4661.14
    },
    {
      "epoch": 0.5184948979591837,
      "grad_norm": 0.5860177840547365,
      "learning_rate": 8.432071702457253e-06,
      "loss": 0.32159740924835206,
      "num_input_tokens_seen": 246871632,
      "step": 4065,
      "train_runtime": 52971.5142,
      "train_tokens_per_second": 4660.46
    },
    {
      "epoch": 0.5191326530612245,
      "grad_norm": 0.6673405050973865,
      "learning_rate": 8.428427443138689e-06,
      "loss": 0.3185391426086426,
      "num_input_tokens_seen": 247174896,
      "step": 4070,
      "train_runtime": 53039.7347,
      "train_tokens_per_second": 4660.183
    },
    {
      "epoch": 0.5197704081632653,
      "grad_norm": 0.6459087480582097,
      "learning_rate": 8.424779743153747e-06,
      "loss": 0.3615418910980225,
      "num_input_tokens_seen": 247484000,
      "step": 4075,
      "train_runtime": 53117.8875,
      "train_tokens_per_second": 4659.146
    },
    {
      "epoch": 0.5204081632653061,
      "grad_norm": 0.6223924077912255,
      "learning_rate": 8.421128606163145e-06,
      "loss": 0.3403435230255127,
      "num_input_tokens_seen": 247782864,
      "step": 4080,
      "train_runtime": 53187.3932,
      "train_tokens_per_second": 4658.677
    },
    {
      "epoch": 0.5210459183673469,
      "grad_norm": 0.5813458681204556,
      "learning_rate": 8.417474035831055e-06,
      "loss": 0.3410142183303833,
      "num_input_tokens_seen": 248091320,
      "step": 4085,
      "train_runtime": 53263.6908,
      "train_tokens_per_second": 4657.794
    },
    {
      "epoch": 0.5216836734693877,
      "grad_norm": 0.6309334729105832,
      "learning_rate": 8.413816035825095e-06,
      "loss": 0.3152641296386719,
      "num_input_tokens_seen": 248381832,
      "step": 4090,
      "train_runtime": 53331.3353,
      "train_tokens_per_second": 4657.334
    },
    {
      "epoch": 0.5223214285714286,
      "grad_norm": 0.6204939129556473,
      "learning_rate": 8.410154609816324e-06,
      "loss": 0.3589243173599243,
      "num_input_tokens_seen": 248693944,
      "step": 4095,
      "train_runtime": 53394.4663,
      "train_tokens_per_second": 4657.673
    },
    {
      "epoch": 0.5229591836734694,
      "grad_norm": 0.6698042327208841,
      "learning_rate": 8.406489761479235e-06,
      "loss": 0.3444381237030029,
      "num_input_tokens_seen": 248989360,
      "step": 4100,
      "train_runtime": 53467.622,
      "train_tokens_per_second": 4656.825
    },
    {
      "epoch": 0.5235969387755102,
      "grad_norm": 0.650372763522801,
      "learning_rate": 8.402821494491762e-06,
      "loss": 0.3539252519607544,
      "num_input_tokens_seen": 249294192,
      "step": 4105,
      "train_runtime": 53525.2602,
      "train_tokens_per_second": 4657.505
    },
    {
      "epoch": 0.5242346938775511,
      "grad_norm": 1.0162701733297357,
      "learning_rate": 8.399149812535269e-06,
      "loss": 0.3387748718261719,
      "num_input_tokens_seen": 249599112,
      "step": 4110,
      "train_runtime": 53575.3695,
      "train_tokens_per_second": 4658.841
    },
    {
      "epoch": 0.5248724489795918,
      "grad_norm": 0.6059781273687016,
      "learning_rate": 8.395474719294541e-06,
      "loss": 0.3216937780380249,
      "num_input_tokens_seen": 249894528,
      "step": 4115,
      "train_runtime": 53633.1185,
      "train_tokens_per_second": 4659.332
    },
    {
      "epoch": 0.5255102040816326,
      "grad_norm": 0.6072738267997749,
      "learning_rate": 8.391796218457794e-06,
      "loss": 0.3277886390686035,
      "num_input_tokens_seen": 250198456,
      "step": 4120,
      "train_runtime": 53688.9453,
      "train_tokens_per_second": 4660.148
    },
    {
      "epoch": 0.5261479591836735,
      "grad_norm": 0.643469498851059,
      "learning_rate": 8.388114313716658e-06,
      "loss": 0.3698796510696411,
      "num_input_tokens_seen": 250505920,
      "step": 4125,
      "train_runtime": 53755.0758,
      "train_tokens_per_second": 4660.135
    },
    {
      "epoch": 0.5267857142857143,
      "grad_norm": 0.6000460110686265,
      "learning_rate": 8.384429008766183e-06,
      "loss": 0.34650416374206544,
      "num_input_tokens_seen": 250820328,
      "step": 4130,
      "train_runtime": 53817.1049,
      "train_tokens_per_second": 4660.606
    },
    {
      "epoch": 0.5274234693877551,
      "grad_norm": 0.6271119719393099,
      "learning_rate": 8.380740307304831e-06,
      "loss": 0.31575334072113037,
      "num_input_tokens_seen": 251122984,
      "step": 4135,
      "train_runtime": 53874.8542,
      "train_tokens_per_second": 4661.228
    },
    {
      "epoch": 0.5280612244897959,
      "grad_norm": 0.6179729508877755,
      "learning_rate": 8.377048213034469e-06,
      "loss": 0.3536393642425537,
      "num_input_tokens_seen": 251421264,
      "step": 4140,
      "train_runtime": 53949.6247,
      "train_tokens_per_second": 4660.297
    },
    {
      "epoch": 0.5286989795918368,
      "grad_norm": 0.6090342687776809,
      "learning_rate": 8.373352729660373e-06,
      "loss": 0.35466365814208983,
      "num_input_tokens_seen": 251730568,
      "step": 4145,
      "train_runtime": 54022.3122,
      "train_tokens_per_second": 4659.752
    },
    {
      "epoch": 0.5293367346938775,
      "grad_norm": 0.5617293491551522,
      "learning_rate": 8.369653860891218e-06,
      "loss": 0.3340321063995361,
      "num_input_tokens_seen": 252036272,
      "step": 4150,
      "train_runtime": 54088.5695,
      "train_tokens_per_second": 4659.696
    },
    {
      "epoch": 0.5299744897959183,
      "grad_norm": 0.5556872361827987,
      "learning_rate": 8.365951610439078e-06,
      "loss": 0.3348118543624878,
      "num_input_tokens_seen": 252339592,
      "step": 4155,
      "train_runtime": 54144.2236,
      "train_tokens_per_second": 4660.508
    },
    {
      "epoch": 0.5306122448979592,
      "grad_norm": 0.6216737414602903,
      "learning_rate": 8.362245982019419e-06,
      "loss": 0.3383298397064209,
      "num_input_tokens_seen": 252647808,
      "step": 4160,
      "train_runtime": 54195.123,
      "train_tokens_per_second": 4661.818
    },
    {
      "epoch": 0.53125,
      "grad_norm": 0.580657736022884,
      "learning_rate": 8.358536979351098e-06,
      "loss": 0.3288172721862793,
      "num_input_tokens_seen": 252955704,
      "step": 4165,
      "train_runtime": 54255.6952,
      "train_tokens_per_second": 4662.289
    },
    {
      "epoch": 0.5318877551020408,
      "grad_norm": 0.7030259938621994,
      "learning_rate": 8.35482460615636e-06,
      "loss": 0.33261590003967284,
      "num_input_tokens_seen": 253256544,
      "step": 4170,
      "train_runtime": 54331.9484,
      "train_tokens_per_second": 4661.282
    },
    {
      "epoch": 0.5325255102040817,
      "grad_norm": 0.5128588950764369,
      "learning_rate": 8.351108866160827e-06,
      "loss": 0.3089122295379639,
      "num_input_tokens_seen": 253562032,
      "step": 4175,
      "train_runtime": 54395.0345,
      "train_tokens_per_second": 4661.492
    },
    {
      "epoch": 0.5331632653061225,
      "grad_norm": 0.5722884092944742,
      "learning_rate": 8.347389763093507e-06,
      "loss": 0.3515080213546753,
      "num_input_tokens_seen": 253862712,
      "step": 4180,
      "train_runtime": 54455.6775,
      "train_tokens_per_second": 4661.823
    },
    {
      "epoch": 0.5338010204081632,
      "grad_norm": 0.7257019076609187,
      "learning_rate": 8.34366730068678e-06,
      "loss": 0.3021442651748657,
      "num_input_tokens_seen": 254178096,
      "step": 4185,
      "train_runtime": 54522.9841,
      "train_tokens_per_second": 4661.852
    },
    {
      "epoch": 0.5344387755102041,
      "grad_norm": 0.618143234739904,
      "learning_rate": 8.339941482676394e-06,
      "loss": 0.3398129940032959,
      "num_input_tokens_seen": 254472504,
      "step": 4190,
      "train_runtime": 54596.9337,
      "train_tokens_per_second": 4660.93
    },
    {
      "epoch": 0.5350765306122449,
      "grad_norm": 0.6401486806740152,
      "learning_rate": 8.336212312801469e-06,
      "loss": 0.33443522453308105,
      "num_input_tokens_seen": 254780208,
      "step": 4195,
      "train_runtime": 54646.4304,
      "train_tokens_per_second": 4662.339
    },
    {
      "epoch": 0.5357142857142857,
      "grad_norm": 0.5832593997249361,
      "learning_rate": 8.332479794804489e-06,
      "loss": 0.32670321464538576,
      "num_input_tokens_seen": 255088280,
      "step": 4200,
      "train_runtime": 54722.0812,
      "train_tokens_per_second": 4661.524
    },
    {
      "epoch": 0.5363520408163265,
      "grad_norm": 0.6243660356934823,
      "learning_rate": 8.328743932431293e-06,
      "loss": 0.3391838312149048,
      "num_input_tokens_seen": 255398952,
      "step": 4205,
      "train_runtime": 54782.111,
      "train_tokens_per_second": 4662.087
    },
    {
      "epoch": 0.5369897959183674,
      "grad_norm": 0.595061368256599,
      "learning_rate": 8.325004729431083e-06,
      "loss": 0.33151419162750245,
      "num_input_tokens_seen": 255709048,
      "step": 4210,
      "train_runtime": 54840.549,
      "train_tokens_per_second": 4662.773
    },
    {
      "epoch": 0.5376275510204082,
      "grad_norm": 0.6589900199646409,
      "learning_rate": 8.32126218955641e-06,
      "loss": 0.32023797035217283,
      "num_input_tokens_seen": 256007488,
      "step": 4215,
      "train_runtime": 54913.8178,
      "train_tokens_per_second": 4661.987
    },
    {
      "epoch": 0.5382653061224489,
      "grad_norm": 0.5688734747502078,
      "learning_rate": 8.317516316563172e-06,
      "loss": 0.33050549030303955,
      "num_input_tokens_seen": 256312784,
      "step": 4220,
      "train_runtime": 54976.3523,
      "train_tokens_per_second": 4662.237
    },
    {
      "epoch": 0.5389030612244898,
      "grad_norm": 0.6100360803622228,
      "learning_rate": 8.313767114210615e-06,
      "loss": 0.3239306449890137,
      "num_input_tokens_seen": 256610864,
      "step": 4225,
      "train_runtime": 55043.8192,
      "train_tokens_per_second": 4661.938
    },
    {
      "epoch": 0.5395408163265306,
      "grad_norm": 0.6329495549647348,
      "learning_rate": 8.310014586261329e-06,
      "loss": 0.32516295909881593,
      "num_input_tokens_seen": 256905128,
      "step": 4230,
      "train_runtime": 55117.4109,
      "train_tokens_per_second": 4661.052
    },
    {
      "epoch": 0.5401785714285714,
      "grad_norm": 0.5788937358176532,
      "learning_rate": 8.306258736481231e-06,
      "loss": 0.321623706817627,
      "num_input_tokens_seen": 257220440,
      "step": 4235,
      "train_runtime": 55167.3251,
      "train_tokens_per_second": 4662.551
    },
    {
      "epoch": 0.5408163265306123,
      "grad_norm": 0.6166029813273239,
      "learning_rate": 8.302499568639583e-06,
      "loss": 0.3558704376220703,
      "num_input_tokens_seen": 257529992,
      "step": 4240,
      "train_runtime": 55227.6824,
      "train_tokens_per_second": 4663.06
    },
    {
      "epoch": 0.5414540816326531,
      "grad_norm": 0.5913914207581168,
      "learning_rate": 8.298737086508973e-06,
      "loss": 0.3280658721923828,
      "num_input_tokens_seen": 257830008,
      "step": 4245,
      "train_runtime": 55291.8955,
      "train_tokens_per_second": 4663.071
    },
    {
      "epoch": 0.5420918367346939,
      "grad_norm": 0.5743530312209789,
      "learning_rate": 8.294971293865315e-06,
      "loss": 0.32333903312683104,
      "num_input_tokens_seen": 258134680,
      "step": 4250,
      "train_runtime": 55363.3988,
      "train_tokens_per_second": 4662.551
    },
    {
      "epoch": 0.5427295918367347,
      "grad_norm": 0.5742413186288639,
      "learning_rate": 8.29120219448784e-06,
      "loss": 0.337364649772644,
      "num_input_tokens_seen": 258440416,
      "step": 4255,
      "train_runtime": 55434.4475,
      "train_tokens_per_second": 4662.091
    },
    {
      "epoch": 0.5433673469387755,
      "grad_norm": 0.6395149410799369,
      "learning_rate": 8.287429792159108e-06,
      "loss": 0.33406877517700195,
      "num_input_tokens_seen": 258733896,
      "step": 4260,
      "train_runtime": 55507.5119,
      "train_tokens_per_second": 4661.241
    },
    {
      "epoch": 0.5440051020408163,
      "grad_norm": 0.6090939041497327,
      "learning_rate": 8.283654090664986e-06,
      "loss": 0.32870163917541506,
      "num_input_tokens_seen": 259042576,
      "step": 4265,
      "train_runtime": 55576.4895,
      "train_tokens_per_second": 4661.01
    },
    {
      "epoch": 0.5446428571428571,
      "grad_norm": 0.6445881345465262,
      "learning_rate": 8.279875093794652e-06,
      "loss": 0.3245696544647217,
      "num_input_tokens_seen": 259348728,
      "step": 4270,
      "train_runtime": 55644.9794,
      "train_tokens_per_second": 4660.775
    },
    {
      "epoch": 0.545280612244898,
      "grad_norm": 0.603234032732469,
      "learning_rate": 8.276092805340596e-06,
      "loss": 0.35511810779571534,
      "num_input_tokens_seen": 259651088,
      "step": 4275,
      "train_runtime": 55706.7744,
      "train_tokens_per_second": 4661.033
    },
    {
      "epoch": 0.5459183673469388,
      "grad_norm": 0.6818970361684568,
      "learning_rate": 8.27230722909861e-06,
      "loss": 0.3598794460296631,
      "num_input_tokens_seen": 259958224,
      "step": 4280,
      "train_runtime": 55768.5734,
      "train_tokens_per_second": 4661.375
    },
    {
      "epoch": 0.5465561224489796,
      "grad_norm": 0.6505327242995605,
      "learning_rate": 8.268518368867781e-06,
      "loss": 0.33161163330078125,
      "num_input_tokens_seen": 260251048,
      "step": 4285,
      "train_runtime": 55833.9926,
      "train_tokens_per_second": 4661.158
    },
    {
      "epoch": 0.5471938775510204,
      "grad_norm": 0.6111247630488971,
      "learning_rate": 8.264726228450495e-06,
      "loss": 0.3262234449386597,
      "num_input_tokens_seen": 260554448,
      "step": 4290,
      "train_runtime": 55905.2245,
      "train_tokens_per_second": 4660.646
    },
    {
      "epoch": 0.5478316326530612,
      "grad_norm": 0.560187020542292,
      "learning_rate": 8.260930811652433e-06,
      "loss": 0.30494074821472167,
      "num_input_tokens_seen": 260860120,
      "step": 4295,
      "train_runtime": 55974.5718,
      "train_tokens_per_second": 4660.333
    },
    {
      "epoch": 0.548469387755102,
      "grad_norm": 0.6556376956868236,
      "learning_rate": 8.257132122282557e-06,
      "loss": 0.3426995754241943,
      "num_input_tokens_seen": 261152392,
      "step": 4300,
      "train_runtime": 56040.4177,
      "train_tokens_per_second": 4660.072
    },
    {
      "epoch": 0.5491071428571429,
      "grad_norm": 1.2172845913803712,
      "learning_rate": 8.253330164153118e-06,
      "loss": 0.35701465606689453,
      "num_input_tokens_seen": 261458776,
      "step": 4305,
      "train_runtime": 56103.0496,
      "train_tokens_per_second": 4660.331
    },
    {
      "epoch": 0.5497448979591837,
      "grad_norm": 0.6008391817753075,
      "learning_rate": 8.24952494107965e-06,
      "loss": 0.3199004650115967,
      "num_input_tokens_seen": 261748456,
      "step": 4310,
      "train_runtime": 56174.9237,
      "train_tokens_per_second": 4659.525
    },
    {
      "epoch": 0.5503826530612245,
      "grad_norm": 0.6256053884342111,
      "learning_rate": 8.245716456880954e-06,
      "loss": 0.32810063362121583,
      "num_input_tokens_seen": 262052304,
      "step": 4315,
      "train_runtime": 56244.5277,
      "train_tokens_per_second": 4659.161
    },
    {
      "epoch": 0.5510204081632653,
      "grad_norm": 0.5756480132256692,
      "learning_rate": 8.241904715379115e-06,
      "loss": 0.30528922080993653,
      "num_input_tokens_seen": 262339992,
      "step": 4320,
      "train_runtime": 56310.5372,
      "train_tokens_per_second": 4658.808
    },
    {
      "epoch": 0.5516581632653061,
      "grad_norm": 0.7898428422930921,
      "learning_rate": 8.23808972039948e-06,
      "loss": 0.3237664222717285,
      "num_input_tokens_seen": 262641960,
      "step": 4325,
      "train_runtime": 56385.1224,
      "train_tokens_per_second": 4658.001
    },
    {
      "epoch": 0.5522959183673469,
      "grad_norm": 0.6306552080612552,
      "learning_rate": 8.234271475770662e-06,
      "loss": 0.345841646194458,
      "num_input_tokens_seen": 262946000,
      "step": 4330,
      "train_runtime": 56445.266,
      "train_tokens_per_second": 4658.424
    },
    {
      "epoch": 0.5529336734693877,
      "grad_norm": 0.6090517723682343,
      "learning_rate": 8.230449985324538e-06,
      "loss": 0.35211758613586425,
      "num_input_tokens_seen": 263244136,
      "step": 4335,
      "train_runtime": 56497.7456,
      "train_tokens_per_second": 4659.374
    },
    {
      "epoch": 0.5535714285714286,
      "grad_norm": 0.6726399178350145,
      "learning_rate": 8.226625252896239e-06,
      "loss": 0.3599707126617432,
      "num_input_tokens_seen": 263558408,
      "step": 4340,
      "train_runtime": 56558.7384,
      "train_tokens_per_second": 4659.906
    },
    {
      "epoch": 0.5542091836734694,
      "grad_norm": 0.6354251802729429,
      "learning_rate": 8.222797282324152e-06,
      "loss": 0.34988815784454347,
      "num_input_tokens_seen": 263867960,
      "step": 4345,
      "train_runtime": 56624.2854,
      "train_tokens_per_second": 4659.979
    },
    {
      "epoch": 0.5548469387755102,
      "grad_norm": 0.5856782387382612,
      "learning_rate": 8.218966077449912e-06,
      "loss": 0.31808695793151853,
      "num_input_tokens_seen": 264172056,
      "step": 4350,
      "train_runtime": 56694.3273,
      "train_tokens_per_second": 4659.585
    },
    {
      "epoch": 0.5554846938775511,
      "grad_norm": 0.6182070325857266,
      "learning_rate": 8.215131642118401e-06,
      "loss": 0.32405288219451905,
      "num_input_tokens_seen": 264484880,
      "step": 4355,
      "train_runtime": 56748.976,
      "train_tokens_per_second": 4660.611
    },
    {
      "epoch": 0.5561224489795918,
      "grad_norm": 0.6757059490288188,
      "learning_rate": 8.211293980177744e-06,
      "loss": 0.34016482830047606,
      "num_input_tokens_seen": 264784920,
      "step": 4360,
      "train_runtime": 56812.2188,
      "train_tokens_per_second": 4660.704
    },
    {
      "epoch": 0.5567602040816326,
      "grad_norm": 0.6197881883465047,
      "learning_rate": 8.2074530954793e-06,
      "loss": 0.3228370904922485,
      "num_input_tokens_seen": 265087688,
      "step": 4365,
      "train_runtime": 56878.7717,
      "train_tokens_per_second": 4660.573
    },
    {
      "epoch": 0.5573979591836735,
      "grad_norm": 0.7030021254750853,
      "learning_rate": 8.203608991877669e-06,
      "loss": 0.33140063285827637,
      "num_input_tokens_seen": 265373784,
      "step": 4370,
      "train_runtime": 56943.0178,
      "train_tokens_per_second": 4660.339
    },
    {
      "epoch": 0.5580357142857143,
      "grad_norm": 0.5653275803125225,
      "learning_rate": 8.199761673230674e-06,
      "loss": 0.3155759334564209,
      "num_input_tokens_seen": 265686896,
      "step": 4375,
      "train_runtime": 57007.7207,
      "train_tokens_per_second": 4660.542
    },
    {
      "epoch": 0.5586734693877551,
      "grad_norm": 0.6345660325552052,
      "learning_rate": 8.195911143399368e-06,
      "loss": 0.3315152168273926,
      "num_input_tokens_seen": 265979352,
      "step": 4380,
      "train_runtime": 57074.253,
      "train_tokens_per_second": 4660.234
    },
    {
      "epoch": 0.5593112244897959,
      "grad_norm": 0.5627906489416599,
      "learning_rate": 8.192057406248028e-06,
      "loss": 0.3102290153503418,
      "num_input_tokens_seen": 266271928,
      "step": 4385,
      "train_runtime": 57134.1088,
      "train_tokens_per_second": 4660.472
    },
    {
      "epoch": 0.5599489795918368,
      "grad_norm": 0.6191145572605166,
      "learning_rate": 8.18820046564415e-06,
      "loss": 0.33108863830566404,
      "num_input_tokens_seen": 266577400,
      "step": 4390,
      "train_runtime": 57191.0161,
      "train_tokens_per_second": 4661.176
    },
    {
      "epoch": 0.5605867346938775,
      "grad_norm": 0.5960521522944929,
      "learning_rate": 8.184340325458439e-06,
      "loss": 0.3300520658493042,
      "num_input_tokens_seen": 266880024,
      "step": 4395,
      "train_runtime": 57250.013,
      "train_tokens_per_second": 4661.659
    },
    {
      "epoch": 0.5612244897959183,
      "grad_norm": 0.5831245845951161,
      "learning_rate": 8.180476989564818e-06,
      "loss": 0.3358687162399292,
      "num_input_tokens_seen": 267187080,
      "step": 4400,
      "train_runtime": 57312.1615,
      "train_tokens_per_second": 4661.961
    },
    {
      "epoch": 0.5618622448979592,
      "grad_norm": 0.566574126965116,
      "learning_rate": 8.176610461840415e-06,
      "loss": 0.33371782302856445,
      "num_input_tokens_seen": 267483488,
      "step": 4405,
      "train_runtime": 57382.3321,
      "train_tokens_per_second": 4661.426
    },
    {
      "epoch": 0.5625,
      "grad_norm": 0.6386464348029305,
      "learning_rate": 8.17274074616556e-06,
      "loss": 0.34981279373168944,
      "num_input_tokens_seen": 267778744,
      "step": 4410,
      "train_runtime": 57439.8382,
      "train_tokens_per_second": 4661.899
    },
    {
      "epoch": 0.5631377551020408,
      "grad_norm": 0.6265982841558887,
      "learning_rate": 8.16886784642378e-06,
      "loss": 0.3423975229263306,
      "num_input_tokens_seen": 268078000,
      "step": 4415,
      "train_runtime": 57512.1122,
      "train_tokens_per_second": 4661.244
    },
    {
      "epoch": 0.5637755102040817,
      "grad_norm": 0.6213852571279617,
      "learning_rate": 8.164991766501804e-06,
      "loss": 0.3394098997116089,
      "num_input_tokens_seen": 268384416,
      "step": 4420,
      "train_runtime": 57582.4629,
      "train_tokens_per_second": 4660.871
    },
    {
      "epoch": 0.5644132653061225,
      "grad_norm": 0.6432331774316757,
      "learning_rate": 8.16111251028955e-06,
      "loss": 0.315293288230896,
      "num_input_tokens_seen": 268675008,
      "step": 4425,
      "train_runtime": 57650.6913,
      "train_tokens_per_second": 4660.395
    },
    {
      "epoch": 0.5650510204081632,
      "grad_norm": 0.6143902666347689,
      "learning_rate": 8.157230081680118e-06,
      "loss": 0.32767605781555176,
      "num_input_tokens_seen": 268986712,
      "step": 4430,
      "train_runtime": 57710.5038,
      "train_tokens_per_second": 4660.966
    },
    {
      "epoch": 0.5656887755102041,
      "grad_norm": 0.5563706398159549,
      "learning_rate": 8.153344484569798e-06,
      "loss": 0.34493412971496584,
      "num_input_tokens_seen": 269287992,
      "step": 4435,
      "train_runtime": 57778.1765,
      "train_tokens_per_second": 4660.722
    },
    {
      "epoch": 0.5663265306122449,
      "grad_norm": 0.6450219066178285,
      "learning_rate": 8.14945572285806e-06,
      "loss": 0.34345855712890627,
      "num_input_tokens_seen": 269589840,
      "step": 4440,
      "train_runtime": 57836.4873,
      "train_tokens_per_second": 4661.242
    },
    {
      "epoch": 0.5669642857142857,
      "grad_norm": 0.6177947769403859,
      "learning_rate": 8.145563800447547e-06,
      "loss": 0.3244961738586426,
      "num_input_tokens_seen": 269897656,
      "step": 4445,
      "train_runtime": 57894.8442,
      "train_tokens_per_second": 4661.86
    },
    {
      "epoch": 0.5676020408163265,
      "grad_norm": 0.6674494880750335,
      "learning_rate": 8.141668721244073e-06,
      "loss": 0.33897535800933837,
      "num_input_tokens_seen": 270205176,
      "step": 4450,
      "train_runtime": 57957.129,
      "train_tokens_per_second": 4662.156
    },
    {
      "epoch": 0.5682397959183674,
      "grad_norm": 0.6399848039940615,
      "learning_rate": 8.137770489156624e-06,
      "loss": 0.32180759906768797,
      "num_input_tokens_seen": 270504888,
      "step": 4455,
      "train_runtime": 58018.3077,
      "train_tokens_per_second": 4662.406
    },
    {
      "epoch": 0.5688775510204082,
      "grad_norm": 0.578493062949667,
      "learning_rate": 8.133869108097349e-06,
      "loss": 0.31485824584960936,
      "num_input_tokens_seen": 270817528,
      "step": 4460,
      "train_runtime": 58078.3445,
      "train_tokens_per_second": 4662.969
    },
    {
      "epoch": 0.5695153061224489,
      "grad_norm": 0.626719750726604,
      "learning_rate": 8.129964581981554e-06,
      "loss": 0.3301039934158325,
      "num_input_tokens_seen": 271113568,
      "step": 4465,
      "train_runtime": 58144.7076,
      "train_tokens_per_second": 4662.739
    },
    {
      "epoch": 0.5701530612244898,
      "grad_norm": 0.6595078436656858,
      "learning_rate": 8.126056914727705e-06,
      "loss": 0.34999229907989504,
      "num_input_tokens_seen": 271422464,
      "step": 4470,
      "train_runtime": 58203.5296,
      "train_tokens_per_second": 4663.333
    },
    {
      "epoch": 0.5707908163265306,
      "grad_norm": 0.6950368995093762,
      "learning_rate": 8.122146110257419e-06,
      "loss": 0.343953537940979,
      "num_input_tokens_seen": 271721272,
      "step": 4475,
      "train_runtime": 58267.0338,
      "train_tokens_per_second": 4663.379
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 0.6180965739348507,
      "learning_rate": 8.11823217249546e-06,
      "loss": 0.3216872692108154,
      "num_input_tokens_seen": 272032224,
      "step": 4480,
      "train_runtime": 58338.5044,
      "train_tokens_per_second": 4662.996
    },
    {
      "epoch": 0.5720663265306123,
      "grad_norm": 0.5269746753228531,
      "learning_rate": 8.114315105369739e-06,
      "loss": 0.34439973831176757,
      "num_input_tokens_seen": 272342792,
      "step": 4485,
      "train_runtime": 58395.7931,
      "train_tokens_per_second": 4663.74
    },
    {
      "epoch": 0.5727040816326531,
      "grad_norm": 0.643538393285888,
      "learning_rate": 8.110394912811308e-06,
      "loss": 0.3304832935333252,
      "num_input_tokens_seen": 272647752,
      "step": 4490,
      "train_runtime": 58453.5655,
      "train_tokens_per_second": 4664.348
    },
    {
      "epoch": 0.5733418367346939,
      "grad_norm": 0.5933974245973419,
      "learning_rate": 8.106471598754353e-06,
      "loss": 0.3181920528411865,
      "num_input_tokens_seen": 272952752,
      "step": 4495,
      "train_runtime": 58511.3847,
      "train_tokens_per_second": 4664.951
    },
    {
      "epoch": 0.5739795918367347,
      "grad_norm": 0.6545169706021291,
      "learning_rate": 8.102545167136191e-06,
      "loss": 0.3237417221069336,
      "num_input_tokens_seen": 273259992,
      "step": 4500,
      "train_runtime": 58568.2779,
      "train_tokens_per_second": 4665.665
    },
    {
      "epoch": 0.5746173469387755,
      "grad_norm": 0.5676528809317268,
      "learning_rate": 8.098615621897272e-06,
      "loss": 0.3160714626312256,
      "num_input_tokens_seen": 273546088,
      "step": 4505,
      "train_runtime": 58635.3904,
      "train_tokens_per_second": 4665.205
    },
    {
      "epoch": 0.5752551020408163,
      "grad_norm": 0.5457134346155577,
      "learning_rate": 8.094682966981172e-06,
      "loss": 0.3343855142593384,
      "num_input_tokens_seen": 273846816,
      "step": 4510,
      "train_runtime": 58701.6981,
      "train_tokens_per_second": 4665.058
    },
    {
      "epoch": 0.5758928571428571,
      "grad_norm": 0.6021407602482653,
      "learning_rate": 8.090747206334582e-06,
      "loss": 0.30679736137390134,
      "num_input_tokens_seen": 274148336,
      "step": 4515,
      "train_runtime": 58763.4119,
      "train_tokens_per_second": 4665.29
    },
    {
      "epoch": 0.576530612244898,
      "grad_norm": 0.6363908329815029,
      "learning_rate": 8.086808343907313e-06,
      "loss": 0.3471099853515625,
      "num_input_tokens_seen": 274448928,
      "step": 4520,
      "train_runtime": 58814.3425,
      "train_tokens_per_second": 4666.361
    },
    {
      "epoch": 0.5771683673469388,
      "grad_norm": 0.5719847354063033,
      "learning_rate": 8.082866383652288e-06,
      "loss": 0.31707210540771485,
      "num_input_tokens_seen": 274760712,
      "step": 4525,
      "train_runtime": 58878.8956,
      "train_tokens_per_second": 4666.54
    },
    {
      "epoch": 0.5778061224489796,
      "grad_norm": 0.5920024664008375,
      "learning_rate": 8.07892132952554e-06,
      "loss": 0.3436100482940674,
      "num_input_tokens_seen": 275066192,
      "step": 4530,
      "train_runtime": 58954.14,
      "train_tokens_per_second": 4665.765
    },
    {
      "epoch": 0.5784438775510204,
      "grad_norm": 0.6934804408029586,
      "learning_rate": 8.074973185486206e-06,
      "loss": 0.3058030128479004,
      "num_input_tokens_seen": 275363072,
      "step": 4535,
      "train_runtime": 59010.6635,
      "train_tokens_per_second": 4666.327
    },
    {
      "epoch": 0.5790816326530612,
      "grad_norm": 0.5999272012793515,
      "learning_rate": 8.071021955496526e-06,
      "loss": 0.34128503799438475,
      "num_input_tokens_seen": 275663040,
      "step": 4540,
      "train_runtime": 59078.8766,
      "train_tokens_per_second": 4666.017
    },
    {
      "epoch": 0.579719387755102,
      "grad_norm": 0.5679709245220098,
      "learning_rate": 8.067067643521834e-06,
      "loss": 0.3404224872589111,
      "num_input_tokens_seen": 275975912,
      "step": 4545,
      "train_runtime": 59135.0954,
      "train_tokens_per_second": 4666.872
    },
    {
      "epoch": 0.5803571428571429,
      "grad_norm": 0.664053925117659,
      "learning_rate": 8.063110253530557e-06,
      "loss": 0.3541407108306885,
      "num_input_tokens_seen": 276277248,
      "step": 4550,
      "train_runtime": 59204.7878,
      "train_tokens_per_second": 4666.468
    },
    {
      "epoch": 0.5809948979591837,
      "grad_norm": 1.4570603403507616,
      "learning_rate": 8.059149789494215e-06,
      "loss": 0.32312707901000975,
      "num_input_tokens_seen": 276556904,
      "step": 4555,
      "train_runtime": 59269.0289,
      "train_tokens_per_second": 4666.128
    },
    {
      "epoch": 0.5816326530612245,
      "grad_norm": 0.6241014199416344,
      "learning_rate": 8.055186255387409e-06,
      "loss": 0.3319542646408081,
      "num_input_tokens_seen": 276857984,
      "step": 4560,
      "train_runtime": 59344.3973,
      "train_tokens_per_second": 4665.276
    },
    {
      "epoch": 0.5822704081632653,
      "grad_norm": 0.6554170093325126,
      "learning_rate": 8.051219655187818e-06,
      "loss": 0.3470942974090576,
      "num_input_tokens_seen": 277177000,
      "step": 4565,
      "train_runtime": 59406.8529,
      "train_tokens_per_second": 4665.741
    },
    {
      "epoch": 0.5829081632653061,
      "grad_norm": 0.7260344299114648,
      "learning_rate": 8.04724999287621e-06,
      "loss": 0.33258297443389895,
      "num_input_tokens_seen": 277478560,
      "step": 4570,
      "train_runtime": 59479.697,
      "train_tokens_per_second": 4665.097
    },
    {
      "epoch": 0.5835459183673469,
      "grad_norm": 0.5684570794452062,
      "learning_rate": 8.043277272436415e-06,
      "loss": 0.3166254997253418,
      "num_input_tokens_seen": 277781656,
      "step": 4575,
      "train_runtime": 59541.3895,
      "train_tokens_per_second": 4665.354
    },
    {
      "epoch": 0.5841836734693877,
      "grad_norm": 0.6025307167484164,
      "learning_rate": 8.039301497855331e-06,
      "loss": 0.3375540733337402,
      "num_input_tokens_seen": 278074920,
      "step": 4580,
      "train_runtime": 59605.1089,
      "train_tokens_per_second": 4665.287
    },
    {
      "epoch": 0.5848214285714286,
      "grad_norm": 0.6353710410028152,
      "learning_rate": 8.035322673122934e-06,
      "loss": 0.33584496974945066,
      "num_input_tokens_seen": 278385592,
      "step": 4585,
      "train_runtime": 59672.671,
      "train_tokens_per_second": 4665.211
    },
    {
      "epoch": 0.5854591836734694,
      "grad_norm": 0.5749678021350256,
      "learning_rate": 8.031340802232246e-06,
      "loss": 0.3267554044723511,
      "num_input_tokens_seen": 278690728,
      "step": 4590,
      "train_runtime": 59743.6738,
      "train_tokens_per_second": 4664.774
    },
    {
      "epoch": 0.5860969387755102,
      "grad_norm": 0.6351825933927931,
      "learning_rate": 8.027355889179356e-06,
      "loss": 0.33541412353515626,
      "num_input_tokens_seen": 278997912,
      "step": 4595,
      "train_runtime": 59803.038,
      "train_tokens_per_second": 4665.28
    },
    {
      "epoch": 0.5867346938775511,
      "grad_norm": 0.6412443966760725,
      "learning_rate": 8.0233679379634e-06,
      "loss": 0.3400087833404541,
      "num_input_tokens_seen": 279300264,
      "step": 4600,
      "train_runtime": 59876.0884,
      "train_tokens_per_second": 4664.638
    },
    {
      "epoch": 0.5873724489795918,
      "grad_norm": 0.5716880259212794,
      "learning_rate": 8.019376952586567e-06,
      "loss": 0.3251744270324707,
      "num_input_tokens_seen": 279603888,
      "step": 4605,
      "train_runtime": 59946.8549,
      "train_tokens_per_second": 4664.196
    },
    {
      "epoch": 0.5880102040816326,
      "grad_norm": 0.6149445066272312,
      "learning_rate": 8.01538293705409e-06,
      "loss": 0.34337458610534666,
      "num_input_tokens_seen": 279902336,
      "step": 4610,
      "train_runtime": 60009.1595,
      "train_tokens_per_second": 4664.327
    },
    {
      "epoch": 0.5886479591836735,
      "grad_norm": 0.6434618638162447,
      "learning_rate": 8.011385895374239e-06,
      "loss": 0.3512176752090454,
      "num_input_tokens_seen": 280206616,
      "step": 4615,
      "train_runtime": 60079.2758,
      "train_tokens_per_second": 4663.948
    },
    {
      "epoch": 0.5892857142857143,
      "grad_norm": 0.5872883115017938,
      "learning_rate": 8.007385831558329e-06,
      "loss": 0.3329361915588379,
      "num_input_tokens_seen": 280506992,
      "step": 4620,
      "train_runtime": 60150.8363,
      "train_tokens_per_second": 4663.393
    },
    {
      "epoch": 0.5899234693877551,
      "grad_norm": 0.554868862354082,
      "learning_rate": 8.003382749620704e-06,
      "loss": 0.3576455593109131,
      "num_input_tokens_seen": 280808544,
      "step": 4625,
      "train_runtime": 60230.8059,
      "train_tokens_per_second": 4662.208
    },
    {
      "epoch": 0.5905612244897959,
      "grad_norm": 0.5553576132935054,
      "learning_rate": 7.99937665357873e-06,
      "loss": 0.3338657855987549,
      "num_input_tokens_seen": 281115176,
      "step": 4630,
      "train_runtime": 60292.4288,
      "train_tokens_per_second": 4662.529
    },
    {
      "epoch": 0.5911989795918368,
      "grad_norm": 0.5674501935313173,
      "learning_rate": 7.995367547452811e-06,
      "loss": 0.30156421661376953,
      "num_input_tokens_seen": 281402488,
      "step": 4635,
      "train_runtime": 60357.7686,
      "train_tokens_per_second": 4662.241
    },
    {
      "epoch": 0.5918367346938775,
      "grad_norm": 0.5802911571097128,
      "learning_rate": 7.991355435266362e-06,
      "loss": 0.334645938873291,
      "num_input_tokens_seen": 281710624,
      "step": 4640,
      "train_runtime": 60421.3044,
      "train_tokens_per_second": 4662.439
    },
    {
      "epoch": 0.5924744897959183,
      "grad_norm": 0.6349513802375457,
      "learning_rate": 7.987340321045816e-06,
      "loss": 0.3482946872711182,
      "num_input_tokens_seen": 282012216,
      "step": 4645,
      "train_runtime": 60485.0048,
      "train_tokens_per_second": 4662.515
    },
    {
      "epoch": 0.5931122448979592,
      "grad_norm": 0.5573471217513042,
      "learning_rate": 7.983322208820624e-06,
      "loss": 0.33052778244018555,
      "num_input_tokens_seen": 282306456,
      "step": 4650,
      "train_runtime": 60555.3502,
      "train_tokens_per_second": 4661.957
    },
    {
      "epoch": 0.59375,
      "grad_norm": 0.6602647431954177,
      "learning_rate": 7.97930110262324e-06,
      "loss": 0.375149130821228,
      "num_input_tokens_seen": 282621480,
      "step": 4655,
      "train_runtime": 60609.0158,
      "train_tokens_per_second": 4663.027
    },
    {
      "epoch": 0.5943877551020408,
      "grad_norm": 0.6595985885235768,
      "learning_rate": 7.975277006489126e-06,
      "loss": 0.3693304300308228,
      "num_input_tokens_seen": 282926240,
      "step": 4660,
      "train_runtime": 60663.4131,
      "train_tokens_per_second": 4663.869
    },
    {
      "epoch": 0.5950255102040817,
      "grad_norm": 0.5684116290932937,
      "learning_rate": 7.971249924456742e-06,
      "loss": 0.3131096839904785,
      "num_input_tokens_seen": 283222512,
      "step": 4665,
      "train_runtime": 60732.2967,
      "train_tokens_per_second": 4663.458
    },
    {
      "epoch": 0.5956632653061225,
      "grad_norm": 0.6221996859576358,
      "learning_rate": 7.967219860567547e-06,
      "loss": 0.3460294008255005,
      "num_input_tokens_seen": 283518112,
      "step": 4670,
      "train_runtime": 60790.6214,
      "train_tokens_per_second": 4663.846
    },
    {
      "epoch": 0.5963010204081632,
      "grad_norm": 0.6320786766100532,
      "learning_rate": 7.963186818865993e-06,
      "loss": 0.33358302116394045,
      "num_input_tokens_seen": 283805976,
      "step": 4675,
      "train_runtime": 60864.1233,
      "train_tokens_per_second": 4662.944
    },
    {
      "epoch": 0.5969387755102041,
      "grad_norm": 0.6376403103388871,
      "learning_rate": 7.959150803399519e-06,
      "loss": 0.33843064308166504,
      "num_input_tokens_seen": 284115784,
      "step": 4680,
      "train_runtime": 60931.736,
      "train_tokens_per_second": 4662.854
    },
    {
      "epoch": 0.5975765306122449,
      "grad_norm": 0.6224191853678193,
      "learning_rate": 7.955111818218544e-06,
      "loss": 0.33420662879943847,
      "num_input_tokens_seen": 284410904,
      "step": 4685,
      "train_runtime": 61012.3022,
      "train_tokens_per_second": 4661.534
    },
    {
      "epoch": 0.5982142857142857,
      "grad_norm": 0.6193691805387954,
      "learning_rate": 7.951069867376477e-06,
      "loss": 0.3390709638595581,
      "num_input_tokens_seen": 284725216,
      "step": 4690,
      "train_runtime": 61079.6857,
      "train_tokens_per_second": 4661.537
    },
    {
      "epoch": 0.5988520408163265,
      "grad_norm": 0.646227711037228,
      "learning_rate": 7.947024954929696e-06,
      "loss": 0.35746586322784424,
      "num_input_tokens_seen": 285031680,
      "step": 4695,
      "train_runtime": 61143.8269,
      "train_tokens_per_second": 4661.659
    },
    {
      "epoch": 0.5994897959183674,
      "grad_norm": 0.6740064328926865,
      "learning_rate": 7.942977084937552e-06,
      "loss": 0.32452802658081054,
      "num_input_tokens_seen": 285330936,
      "step": 4700,
      "train_runtime": 61207.4163,
      "train_tokens_per_second": 4661.705
    },
    {
      "epoch": 0.6001275510204082,
      "grad_norm": 0.6473059340262228,
      "learning_rate": 7.938926261462366e-06,
      "loss": 0.3147572040557861,
      "num_input_tokens_seen": 285641568,
      "step": 4705,
      "train_runtime": 61269.392,
      "train_tokens_per_second": 4662.06
    },
    {
      "epoch": 0.6007653061224489,
      "grad_norm": 0.6293561266830001,
      "learning_rate": 7.934872488569423e-06,
      "loss": 0.3319873809814453,
      "num_input_tokens_seen": 285940128,
      "step": 4710,
      "train_runtime": 61341.2258,
      "train_tokens_per_second": 4661.467
    },
    {
      "epoch": 0.6014030612244898,
      "grad_norm": 0.6755563851553431,
      "learning_rate": 7.930815770326968e-06,
      "loss": 0.3462022304534912,
      "num_input_tokens_seen": 286236152,
      "step": 4715,
      "train_runtime": 61406.0953,
      "train_tokens_per_second": 4661.364
    },
    {
      "epoch": 0.6020408163265306,
      "grad_norm": 0.5858959645393711,
      "learning_rate": 7.926756110806197e-06,
      "loss": 0.30179471969604493,
      "num_input_tokens_seen": 286540184,
      "step": 4720,
      "train_runtime": 61461.7461,
      "train_tokens_per_second": 4662.09
    },
    {
      "epoch": 0.6026785714285714,
      "grad_norm": 0.6666598087218537,
      "learning_rate": 7.922693514081267e-06,
      "loss": 0.33141958713531494,
      "num_input_tokens_seen": 286833736,
      "step": 4725,
      "train_runtime": 61520.0469,
      "train_tokens_per_second": 4662.443
    },
    {
      "epoch": 0.6033163265306123,
      "grad_norm": 0.6391942747513293,
      "learning_rate": 7.918627984229274e-06,
      "loss": 0.338878870010376,
      "num_input_tokens_seen": 287138016,
      "step": 4730,
      "train_runtime": 61581.6349,
      "train_tokens_per_second": 4662.722
    },
    {
      "epoch": 0.6039540816326531,
      "grad_norm": 0.6113485497108972,
      "learning_rate": 7.914559525330262e-06,
      "loss": 0.35114235877990724,
      "num_input_tokens_seen": 287450896,
      "step": 4735,
      "train_runtime": 61643.7397,
      "train_tokens_per_second": 4663.1
    },
    {
      "epoch": 0.6045918367346939,
      "grad_norm": 0.5644556364920948,
      "learning_rate": 7.910488141467215e-06,
      "loss": 0.330631947517395,
      "num_input_tokens_seen": 287761752,
      "step": 4740,
      "train_runtime": 61701.008,
      "train_tokens_per_second": 4663.81
    },
    {
      "epoch": 0.6052295918367347,
      "grad_norm": 0.6426886023485954,
      "learning_rate": 7.906413836726049e-06,
      "loss": 0.3229456901550293,
      "num_input_tokens_seen": 288062320,
      "step": 4745,
      "train_runtime": 61769.8193,
      "train_tokens_per_second": 4663.48
    },
    {
      "epoch": 0.6058673469387755,
      "grad_norm": 0.5588105265523678,
      "learning_rate": 7.902336615195614e-06,
      "loss": 0.3303380012512207,
      "num_input_tokens_seen": 288357504,
      "step": 4750,
      "train_runtime": 61832.314,
      "train_tokens_per_second": 4663.541
    },
    {
      "epoch": 0.6065051020408163,
      "grad_norm": 0.6260031107978368,
      "learning_rate": 7.89825648096769e-06,
      "loss": 0.33974905014038087,
      "num_input_tokens_seen": 288654072,
      "step": 4755,
      "train_runtime": 61907.0899,
      "train_tokens_per_second": 4662.698
    },
    {
      "epoch": 0.6071428571428571,
      "grad_norm": 0.5458423897744933,
      "learning_rate": 7.89417343813697e-06,
      "loss": 0.3210566282272339,
      "num_input_tokens_seen": 288951872,
      "step": 4760,
      "train_runtime": 61969.6499,
      "train_tokens_per_second": 4662.797
    },
    {
      "epoch": 0.607780612244898,
      "grad_norm": 0.6370883128603102,
      "learning_rate": 7.890087490801077e-06,
      "loss": 0.32726516723632815,
      "num_input_tokens_seen": 289243832,
      "step": 4765,
      "train_runtime": 62039.7788,
      "train_tokens_per_second": 4662.232
    },
    {
      "epoch": 0.6084183673469388,
      "grad_norm": 0.5310342852708402,
      "learning_rate": 7.885998643060544e-06,
      "loss": 0.3283411979675293,
      "num_input_tokens_seen": 289548232,
      "step": 4770,
      "train_runtime": 62098.7189,
      "train_tokens_per_second": 4662.709
    },
    {
      "epoch": 0.6090561224489796,
      "grad_norm": 0.5635723415844118,
      "learning_rate": 7.881906899018815e-06,
      "loss": 0.3176594257354736,
      "num_input_tokens_seen": 289868800,
      "step": 4775,
      "train_runtime": 62148.1586,
      "train_tokens_per_second": 4664.157
    },
    {
      "epoch": 0.6096938775510204,
      "grad_norm": 0.5869010517099058,
      "learning_rate": 7.87781226278224e-06,
      "loss": 0.33094594478607176,
      "num_input_tokens_seen": 290163856,
      "step": 4780,
      "train_runtime": 62209.3692,
      "train_tokens_per_second": 4664.311
    },
    {
      "epoch": 0.6103316326530612,
      "grad_norm": 0.572120310946249,
      "learning_rate": 7.873714738460075e-06,
      "loss": 0.3263236999511719,
      "num_input_tokens_seen": 290462384,
      "step": 4785,
      "train_runtime": 62277.5925,
      "train_tokens_per_second": 4663.995
    },
    {
      "epoch": 0.610969387755102,
      "grad_norm": 0.5822636382201469,
      "learning_rate": 7.86961433016447e-06,
      "loss": 0.3235758304595947,
      "num_input_tokens_seen": 290775128,
      "step": 4790,
      "train_runtime": 62349.5125,
      "train_tokens_per_second": 4663.631
    },
    {
      "epoch": 0.6116071428571429,
      "grad_norm": 0.6029053979543979,
      "learning_rate": 7.865511042010469e-06,
      "loss": 0.31907784938812256,
      "num_input_tokens_seen": 291066008,
      "step": 4795,
      "train_runtime": 62412.6057,
      "train_tokens_per_second": 4663.577
    },
    {
      "epoch": 0.6122448979591837,
      "grad_norm": 0.6364330502640352,
      "learning_rate": 7.861404878116014e-06,
      "loss": 0.3359909772872925,
      "num_input_tokens_seen": 291355224,
      "step": 4800,
      "train_runtime": 62484.2112,
      "train_tokens_per_second": 4662.862
    },
    {
      "epoch": 0.6128826530612245,
      "grad_norm": 0.5671585607212491,
      "learning_rate": 7.857295842601921e-06,
      "loss": 0.3385308265686035,
      "num_input_tokens_seen": 291666904,
      "step": 4805,
      "train_runtime": 62531.9872,
      "train_tokens_per_second": 4664.283
    },
    {
      "epoch": 0.6135204081632653,
      "grad_norm": 0.6162061836917074,
      "learning_rate": 7.853183939591898e-06,
      "loss": 0.3188408136367798,
      "num_input_tokens_seen": 291954960,
      "step": 4810,
      "train_runtime": 62595.0014,
      "train_tokens_per_second": 4664.19
    },
    {
      "epoch": 0.6141581632653061,
      "grad_norm": 0.5758849290254675,
      "learning_rate": 7.849069173212526e-06,
      "loss": 0.33262321949005125,
      "num_input_tokens_seen": 292263992,
      "step": 4815,
      "train_runtime": 62670.2416,
      "train_tokens_per_second": 4663.521
    },
    {
      "epoch": 0.6147959183673469,
      "grad_norm": 0.6268541643214147,
      "learning_rate": 7.844951547593261e-06,
      "loss": 0.3587038516998291,
      "num_input_tokens_seen": 292570608,
      "step": 4820,
      "train_runtime": 62743.6215,
      "train_tokens_per_second": 4662.954
    },
    {
      "epoch": 0.6154336734693877,
      "grad_norm": 0.781103749729596,
      "learning_rate": 7.840831066866423e-06,
      "loss": 0.34357218742370604,
      "num_input_tokens_seen": 292889008,
      "step": 4825,
      "train_runtime": 62802.8569,
      "train_tokens_per_second": 4663.626
    },
    {
      "epoch": 0.6160714285714286,
      "grad_norm": 0.5723834748914789,
      "learning_rate": 7.83670773516721e-06,
      "loss": 0.30017411708831787,
      "num_input_tokens_seen": 293192336,
      "step": 4830,
      "train_runtime": 62868.64,
      "train_tokens_per_second": 4663.571
    },
    {
      "epoch": 0.6167091836734694,
      "grad_norm": 0.7844624121572441,
      "learning_rate": 7.83258155663367e-06,
      "loss": 0.34011092185974123,
      "num_input_tokens_seen": 293504568,
      "step": 4835,
      "train_runtime": 62917.6662,
      "train_tokens_per_second": 4664.899
    },
    {
      "epoch": 0.6173469387755102,
      "grad_norm": 0.6637915760976678,
      "learning_rate": 7.828452535406707e-06,
      "loss": 0.32102341651916505,
      "num_input_tokens_seen": 293801056,
      "step": 4840,
      "train_runtime": 62980.9822,
      "train_tokens_per_second": 4664.917
    },
    {
      "epoch": 0.6179846938775511,
      "grad_norm": 0.6612890525763778,
      "learning_rate": 7.82432067563009e-06,
      "loss": 0.28260757923126223,
      "num_input_tokens_seen": 294104640,
      "step": 4845,
      "train_runtime": 63044.6342,
      "train_tokens_per_second": 4665.023
    },
    {
      "epoch": 0.6186224489795918,
      "grad_norm": 0.561487135445506,
      "learning_rate": 7.820185981450422e-06,
      "loss": 0.3284600019454956,
      "num_input_tokens_seen": 294420608,
      "step": 4850,
      "train_runtime": 63112.8408,
      "train_tokens_per_second": 4664.987
    },
    {
      "epoch": 0.6192602040816326,
      "grad_norm": 0.6578404198864279,
      "learning_rate": 7.816048457017163e-06,
      "loss": 0.35275702476501464,
      "num_input_tokens_seen": 294729432,
      "step": 4855,
      "train_runtime": 63182.353,
      "train_tokens_per_second": 4664.743
    },
    {
      "epoch": 0.6198979591836735,
      "grad_norm": 0.6628087083893931,
      "learning_rate": 7.811908106482601e-06,
      "loss": 0.32400710582733155,
      "num_input_tokens_seen": 295026480,
      "step": 4860,
      "train_runtime": 63253.1576,
      "train_tokens_per_second": 4664.217
    },
    {
      "epoch": 0.6205357142857143,
      "grad_norm": 0.6062819712448757,
      "learning_rate": 7.807764934001875e-06,
      "loss": 0.3197571039199829,
      "num_input_tokens_seen": 295335136,
      "step": 4865,
      "train_runtime": 63318.8303,
      "train_tokens_per_second": 4664.254
    },
    {
      "epoch": 0.6211734693877551,
      "grad_norm": 0.554016057626016,
      "learning_rate": 7.803618943732943e-06,
      "loss": 0.30578064918518066,
      "num_input_tokens_seen": 295621456,
      "step": 4870,
      "train_runtime": 63382.3028,
      "train_tokens_per_second": 4664.101
    },
    {
      "epoch": 0.6218112244897959,
      "grad_norm": 0.673440426442395,
      "learning_rate": 7.799470139836593e-06,
      "loss": 0.33724274635314944,
      "num_input_tokens_seen": 295928456,
      "step": 4875,
      "train_runtime": 63449.7801,
      "train_tokens_per_second": 4663.979
    },
    {
      "epoch": 0.6224489795918368,
      "grad_norm": 0.598591390005993,
      "learning_rate": 7.795318526476448e-06,
      "loss": 0.3247530460357666,
      "num_input_tokens_seen": 296232464,
      "step": 4880,
      "train_runtime": 63503.3866,
      "train_tokens_per_second": 4664.829
    },
    {
      "epoch": 0.6230867346938775,
      "grad_norm": 0.6207353679082063,
      "learning_rate": 7.791164107818933e-06,
      "loss": 0.3155781030654907,
      "num_input_tokens_seen": 296533328,
      "step": 4885,
      "train_runtime": 63564.6937,
      "train_tokens_per_second": 4665.063
    },
    {
      "epoch": 0.6237244897959183,
      "grad_norm": 0.5600020735357846,
      "learning_rate": 7.787006888033304e-06,
      "loss": 0.3392642021179199,
      "num_input_tokens_seen": 296836440,
      "step": 4890,
      "train_runtime": 63634.6068,
      "train_tokens_per_second": 4664.701
    },
    {
      "epoch": 0.6243622448979592,
      "grad_norm": 0.621301757055225,
      "learning_rate": 7.782846871291619e-06,
      "loss": 0.33809504508972166,
      "num_input_tokens_seen": 297135096,
      "step": 4895,
      "train_runtime": 63693.8066,
      "train_tokens_per_second": 4665.055
    },
    {
      "epoch": 0.625,
      "grad_norm": 0.5581457268079149,
      "learning_rate": 7.778684061768744e-06,
      "loss": 0.33953664302825926,
      "num_input_tokens_seen": 297429472,
      "step": 4900,
      "train_runtime": 63760.2905,
      "train_tokens_per_second": 4664.807
    },
    {
      "epoch": 0.6256377551020408,
      "grad_norm": 0.6240505820939524,
      "learning_rate": 7.774518463642351e-06,
      "loss": 0.33265602588653564,
      "num_input_tokens_seen": 297713592,
      "step": 4905,
      "train_runtime": 63818.8102,
      "train_tokens_per_second": 4664.982
    },
    {
      "epoch": 0.6262755102040817,
      "grad_norm": 0.5886198862837947,
      "learning_rate": 7.770350081092906e-06,
      "loss": 0.3128589391708374,
      "num_input_tokens_seen": 298015744,
      "step": 4910,
      "train_runtime": 63886.5387,
      "train_tokens_per_second": 4664.766
    },
    {
      "epoch": 0.6269132653061225,
      "grad_norm": 0.7028247548263732,
      "learning_rate": 7.766178918303675e-06,
      "loss": 0.32480950355529786,
      "num_input_tokens_seen": 298324856,
      "step": 4915,
      "train_runtime": 63949.751,
      "train_tokens_per_second": 4664.989
    },
    {
      "epoch": 0.6275510204081632,
      "grad_norm": 0.609308928617577,
      "learning_rate": 7.762004979460711e-06,
      "loss": 0.3458104610443115,
      "num_input_tokens_seen": 298626488,
      "step": 4920,
      "train_runtime": 64014.9401,
      "train_tokens_per_second": 4664.95
    },
    {
      "epoch": 0.6281887755102041,
      "grad_norm": 0.601149571843337,
      "learning_rate": 7.75782826875285e-06,
      "loss": 0.32567148208618163,
      "num_input_tokens_seen": 298934448,
      "step": 4925,
      "train_runtime": 64090.7438,
      "train_tokens_per_second": 4664.237
    },
    {
      "epoch": 0.6288265306122449,
      "grad_norm": 0.6905427443909213,
      "learning_rate": 7.753648790371716e-06,
      "loss": 0.3416671991348267,
      "num_input_tokens_seen": 299242480,
      "step": 4930,
      "train_runtime": 64156.6957,
      "train_tokens_per_second": 4664.244
    },
    {
      "epoch": 0.6294642857142857,
      "grad_norm": 0.6074075357499965,
      "learning_rate": 7.749466548511706e-06,
      "loss": 0.32350864410400393,
      "num_input_tokens_seen": 299546696,
      "step": 4935,
      "train_runtime": 64224.297,
      "train_tokens_per_second": 4664.071
    },
    {
      "epoch": 0.6301020408163265,
      "grad_norm": 0.54837645000057,
      "learning_rate": 7.745281547369992e-06,
      "loss": 0.3141555070877075,
      "num_input_tokens_seen": 299857664,
      "step": 4940,
      "train_runtime": 64281.7126,
      "train_tokens_per_second": 4664.743
    },
    {
      "epoch": 0.6307397959183674,
      "grad_norm": 0.656812618336775,
      "learning_rate": 7.741093791146517e-06,
      "loss": 0.33570907115936277,
      "num_input_tokens_seen": 300162488,
      "step": 4945,
      "train_runtime": 64349.2952,
      "train_tokens_per_second": 4664.581
    },
    {
      "epoch": 0.6313775510204082,
      "grad_norm": 0.7271739146461138,
      "learning_rate": 7.736903284043985e-06,
      "loss": 0.31252236366271974,
      "num_input_tokens_seen": 300470800,
      "step": 4950,
      "train_runtime": 64410.7323,
      "train_tokens_per_second": 4664.918
    },
    {
      "epoch": 0.6320153061224489,
      "grad_norm": 0.5592254871328908,
      "learning_rate": 7.73271003026786e-06,
      "loss": 0.32289905548095704,
      "num_input_tokens_seen": 300771088,
      "step": 4955,
      "train_runtime": 64485.8894,
      "train_tokens_per_second": 4664.138
    },
    {
      "epoch": 0.6326530612244898,
      "grad_norm": 0.5909667799778385,
      "learning_rate": 7.72851403402637e-06,
      "loss": 0.3148157596588135,
      "num_input_tokens_seen": 301087160,
      "step": 4960,
      "train_runtime": 64546.7075,
      "train_tokens_per_second": 4664.64
    },
    {
      "epoch": 0.6332908163265306,
      "grad_norm": 0.6001910064087375,
      "learning_rate": 7.724315299530485e-06,
      "loss": 0.3344589710235596,
      "num_input_tokens_seen": 301394496,
      "step": 4965,
      "train_runtime": 64614.713,
      "train_tokens_per_second": 4664.487
    },
    {
      "epoch": 0.6339285714285714,
      "grad_norm": 0.686313433188379,
      "learning_rate": 7.720113830993934e-06,
      "loss": 0.346239709854126,
      "num_input_tokens_seen": 301692832,
      "step": 4970,
      "train_runtime": 64681.1812,
      "train_tokens_per_second": 4664.306
    },
    {
      "epoch": 0.6345663265306123,
      "grad_norm": 0.5807770199122811,
      "learning_rate": 7.71590963263318e-06,
      "loss": 0.3092792987823486,
      "num_input_tokens_seen": 301993848,
      "step": 4975,
      "train_runtime": 64758.1342,
      "train_tokens_per_second": 4663.412
    },
    {
      "epoch": 0.6352040816326531,
      "grad_norm": 0.5960635267156836,
      "learning_rate": 7.711702708667432e-06,
      "loss": 0.3285883903503418,
      "num_input_tokens_seen": 302288504,
      "step": 4980,
      "train_runtime": 64828.5858,
      "train_tokens_per_second": 4662.889
    },
    {
      "epoch": 0.6358418367346939,
      "grad_norm": 0.6264412065564793,
      "learning_rate": 7.70749306331863e-06,
      "loss": 0.3152578353881836,
      "num_input_tokens_seen": 302596976,
      "step": 4985,
      "train_runtime": 64894.2577,
      "train_tokens_per_second": 4662.924
    },
    {
      "epoch": 0.6364795918367347,
      "grad_norm": 0.6384532801785069,
      "learning_rate": 7.703280700811447e-06,
      "loss": 0.3586283683776855,
      "num_input_tokens_seen": 302885624,
      "step": 4990,
      "train_runtime": 64956.1119,
      "train_tokens_per_second": 4662.927
    },
    {
      "epoch": 0.6371173469387755,
      "grad_norm": 0.650443987703099,
      "learning_rate": 7.699065625373285e-06,
      "loss": 0.34339160919189454,
      "num_input_tokens_seen": 303170080,
      "step": 4995,
      "train_runtime": 65022.0927,
      "train_tokens_per_second": 4662.57
    },
    {
      "epoch": 0.6377551020408163,
      "grad_norm": 0.6388349023596219,
      "learning_rate": 7.694847841234268e-06,
      "loss": 0.31816675662994387,
      "num_input_tokens_seen": 303462440,
      "step": 5000,
      "train_runtime": 65090.1892,
      "train_tokens_per_second": 4662.184
    },
    {
      "epoch": 0.6383928571428571,
      "grad_norm": 0.5935913817839399,
      "learning_rate": 7.690627352627232e-06,
      "loss": 0.356259822845459,
      "num_input_tokens_seen": 303774280,
      "step": 5005,
      "train_runtime": 65151.9451,
      "train_tokens_per_second": 4662.551
    },
    {
      "epoch": 0.639030612244898,
      "grad_norm": 0.6480781837963139,
      "learning_rate": 7.686404163787734e-06,
      "loss": 0.3103009223937988,
      "num_input_tokens_seen": 304067248,
      "step": 5010,
      "train_runtime": 65223.1825,
      "train_tokens_per_second": 4661.95
    },
    {
      "epoch": 0.6396683673469388,
      "grad_norm": 0.6012006012262137,
      "learning_rate": 7.68217827895404e-06,
      "loss": 0.3121501445770264,
      "num_input_tokens_seen": 304370400,
      "step": 5015,
      "train_runtime": 65275.136,
      "train_tokens_per_second": 4662.884
    },
    {
      "epoch": 0.6403061224489796,
      "grad_norm": 0.6712143913991747,
      "learning_rate": 7.677949702367122e-06,
      "loss": 0.3567068576812744,
      "num_input_tokens_seen": 304674832,
      "step": 5020,
      "train_runtime": 65339.7216,
      "train_tokens_per_second": 4662.934
    },
    {
      "epoch": 0.6409438775510204,
      "grad_norm": 0.5599610051996647,
      "learning_rate": 7.673718438270649e-06,
      "loss": 0.3278343200683594,
      "num_input_tokens_seen": 304981880,
      "step": 5025,
      "train_runtime": 65411.5493,
      "train_tokens_per_second": 4662.508
    },
    {
      "epoch": 0.6415816326530612,
      "grad_norm": 0.6239344933483976,
      "learning_rate": 7.66948449091099e-06,
      "loss": 0.3229900598526001,
      "num_input_tokens_seen": 305285928,
      "step": 5030,
      "train_runtime": 65477.2048,
      "train_tokens_per_second": 4662.476
    },
    {
      "epoch": 0.642219387755102,
      "grad_norm": 0.6098449426813721,
      "learning_rate": 7.66524786453721e-06,
      "loss": 0.3233805656433105,
      "num_input_tokens_seen": 305584664,
      "step": 5035,
      "train_runtime": 65549.2606,
      "train_tokens_per_second": 4661.909
    },
    {
      "epoch": 0.6428571428571429,
      "grad_norm": 0.6247649176462382,
      "learning_rate": 7.661008563401056e-06,
      "loss": 0.3556224346160889,
      "num_input_tokens_seen": 305868632,
      "step": 5040,
      "train_runtime": 65606.5949,
      "train_tokens_per_second": 4662.163
    },
    {
      "epoch": 0.6434948979591837,
      "grad_norm": 0.5993651751501048,
      "learning_rate": 7.656766591756964e-06,
      "loss": 0.34767487049102785,
      "num_input_tokens_seen": 306179728,
      "step": 5045,
      "train_runtime": 65669.0743,
      "train_tokens_per_second": 4662.465
    },
    {
      "epoch": 0.6441326530612245,
      "grad_norm": 0.639117877395612,
      "learning_rate": 7.652521953862051e-06,
      "loss": 0.33482675552368163,
      "num_input_tokens_seen": 306489360,
      "step": 5050,
      "train_runtime": 65716.7925,
      "train_tokens_per_second": 4663.791
    },
    {
      "epoch": 0.6447704081632653,
      "grad_norm": 0.5849993879590735,
      "learning_rate": 7.648274653976102e-06,
      "loss": 0.3288681745529175,
      "num_input_tokens_seen": 306789736,
      "step": 5055,
      "train_runtime": 65778.6707,
      "train_tokens_per_second": 4663.97
    },
    {
      "epoch": 0.6454081632653061,
      "grad_norm": 0.6338086296242965,
      "learning_rate": 7.644024696361587e-06,
      "loss": 0.32494688034057617,
      "num_input_tokens_seen": 307097616,
      "step": 5060,
      "train_runtime": 65836.927,
      "train_tokens_per_second": 4664.519
    },
    {
      "epoch": 0.6460459183673469,
      "grad_norm": 0.5876557918206017,
      "learning_rate": 7.639772085283629e-06,
      "loss": 0.33415367603302004,
      "num_input_tokens_seen": 307408752,
      "step": 5065,
      "train_runtime": 65898.5938,
      "train_tokens_per_second": 4664.876
    },
    {
      "epoch": 0.6466836734693877,
      "grad_norm": 0.5839685640993281,
      "learning_rate": 7.635516825010022e-06,
      "loss": 0.3211622714996338,
      "num_input_tokens_seen": 307705200,
      "step": 5070,
      "train_runtime": 65964.8574,
      "train_tokens_per_second": 4664.684
    },
    {
      "epoch": 0.6473214285714286,
      "grad_norm": 0.5313809049144229,
      "learning_rate": 7.631258919811215e-06,
      "loss": 0.3212575435638428,
      "num_input_tokens_seen": 308018488,
      "step": 5075,
      "train_runtime": 66023.7313,
      "train_tokens_per_second": 4665.269
    },
    {
      "epoch": 0.6479591836734694,
      "grad_norm": 0.7271067216633091,
      "learning_rate": 7.626998373960317e-06,
      "loss": 0.3299903869628906,
      "num_input_tokens_seen": 308322376,
      "step": 5080,
      "train_runtime": 66094.8533,
      "train_tokens_per_second": 4664.847
    },
    {
      "epoch": 0.6485969387755102,
      "grad_norm": 0.7099319100446666,
      "learning_rate": 7.6227351917330835e-06,
      "loss": 0.3272510290145874,
      "num_input_tokens_seen": 308624608,
      "step": 5085,
      "train_runtime": 66157.0424,
      "train_tokens_per_second": 4665.03
    },
    {
      "epoch": 0.6492346938775511,
      "grad_norm": 0.5611417491839693,
      "learning_rate": 7.618469377407911e-06,
      "loss": 0.35759220123291013,
      "num_input_tokens_seen": 308935880,
      "step": 5090,
      "train_runtime": 66212.2113,
      "train_tokens_per_second": 4665.844
    },
    {
      "epoch": 0.6498724489795918,
      "grad_norm": 0.6029794330260951,
      "learning_rate": 7.614200935265845e-06,
      "loss": 0.3380232572555542,
      "num_input_tokens_seen": 309228488,
      "step": 5095,
      "train_runtime": 66282.9979,
      "train_tokens_per_second": 4665.276
    },
    {
      "epoch": 0.6505102040816326,
      "grad_norm": 0.5994525877243074,
      "learning_rate": 7.609929869590566e-06,
      "loss": 0.3147365808486938,
      "num_input_tokens_seen": 309541768,
      "step": 5100,
      "train_runtime": 66351.6034,
      "train_tokens_per_second": 4665.174
    },
    {
      "epoch": 0.6511479591836735,
      "grad_norm": 0.6222623980322116,
      "learning_rate": 7.605656184668385e-06,
      "loss": 0.32534070014953614,
      "num_input_tokens_seen": 309852264,
      "step": 5105,
      "train_runtime": 66411.8441,
      "train_tokens_per_second": 4665.618
    },
    {
      "epoch": 0.6517857142857143,
      "grad_norm": 0.5874904480880205,
      "learning_rate": 7.6013798847882455e-06,
      "loss": 0.3319838285446167,
      "num_input_tokens_seen": 310164128,
      "step": 5110,
      "train_runtime": 66464.1894,
      "train_tokens_per_second": 4666.635
    },
    {
      "epoch": 0.6524234693877551,
      "grad_norm": 0.5605939289385559,
      "learning_rate": 7.597100974241711e-06,
      "loss": 0.33452045917510986,
      "num_input_tokens_seen": 310465392,
      "step": 5115,
      "train_runtime": 66537.1495,
      "train_tokens_per_second": 4666.046
    },
    {
      "epoch": 0.6530612244897959,
      "grad_norm": 0.6209280952740241,
      "learning_rate": 7.592819457322967e-06,
      "loss": 0.3549307346343994,
      "num_input_tokens_seen": 310764976,
      "step": 5120,
      "train_runtime": 66599.5403,
      "train_tokens_per_second": 4666.173
    },
    {
      "epoch": 0.6536989795918368,
      "grad_norm": 0.5731115356851546,
      "learning_rate": 7.588535338328816e-06,
      "loss": 0.32546820640563967,
      "num_input_tokens_seen": 311068176,
      "step": 5125,
      "train_runtime": 66670.1991,
      "train_tokens_per_second": 4665.775
    },
    {
      "epoch": 0.6543367346938775,
      "grad_norm": 0.5160733758277285,
      "learning_rate": 7.584248621558669e-06,
      "loss": 0.33570246696472167,
      "num_input_tokens_seen": 311375264,
      "step": 5130,
      "train_runtime": 66733.5287,
      "train_tokens_per_second": 4665.949
    },
    {
      "epoch": 0.6549744897959183,
      "grad_norm": 0.6280358766100539,
      "learning_rate": 7.579959311314549e-06,
      "loss": 0.3122812509536743,
      "num_input_tokens_seen": 311663640,
      "step": 5135,
      "train_runtime": 66807.4591,
      "train_tokens_per_second": 4665.102
    },
    {
      "epoch": 0.6556122448979592,
      "grad_norm": 0.7342922788385822,
      "learning_rate": 7.575667411901074e-06,
      "loss": 0.33909385204315184,
      "num_input_tokens_seen": 311971384,
      "step": 5140,
      "train_runtime": 66862.4758,
      "train_tokens_per_second": 4665.866
    },
    {
      "epoch": 0.65625,
      "grad_norm": 0.5504399102589533,
      "learning_rate": 7.571372927625469e-06,
      "loss": 0.30200581550598143,
      "num_input_tokens_seen": 312278304,
      "step": 5145,
      "train_runtime": 66931.2224,
      "train_tokens_per_second": 4665.66
    },
    {
      "epoch": 0.6568877551020408,
      "grad_norm": 0.6350235883558237,
      "learning_rate": 7.567075862797546e-06,
      "loss": 0.3463855504989624,
      "num_input_tokens_seen": 312583576,
      "step": 5150,
      "train_runtime": 67003.1506,
      "train_tokens_per_second": 4665.207
    },
    {
      "epoch": 0.6575255102040817,
      "grad_norm": 0.5878594793890254,
      "learning_rate": 7.562776221729709e-06,
      "loss": 0.3354689121246338,
      "num_input_tokens_seen": 312893440,
      "step": 5155,
      "train_runtime": 67062.7977,
      "train_tokens_per_second": 4665.678
    },
    {
      "epoch": 0.6581632653061225,
      "grad_norm": 0.6261798997073404,
      "learning_rate": 7.558474008736951e-06,
      "loss": 0.3523919105529785,
      "num_input_tokens_seen": 313192904,
      "step": 5160,
      "train_runtime": 67123.8409,
      "train_tokens_per_second": 4665.897
    },
    {
      "epoch": 0.6588010204081632,
      "grad_norm": 0.6377921783909494,
      "learning_rate": 7.554169228136842e-06,
      "loss": 0.3369914054870605,
      "num_input_tokens_seen": 313489048,
      "step": 5165,
      "train_runtime": 67187.7626,
      "train_tokens_per_second": 4665.865
    },
    {
      "epoch": 0.6594387755102041,
      "grad_norm": 0.7885707396439451,
      "learning_rate": 7.549861884249529e-06,
      "loss": 0.3480203151702881,
      "num_input_tokens_seen": 313797432,
      "step": 5170,
      "train_runtime": 67259.0989,
      "train_tokens_per_second": 4665.502
    },
    {
      "epoch": 0.6600765306122449,
      "grad_norm": 0.557473276764076,
      "learning_rate": 7.545551981397733e-06,
      "loss": 0.3440972566604614,
      "num_input_tokens_seen": 314115728,
      "step": 5175,
      "train_runtime": 67332.3398,
      "train_tokens_per_second": 4665.154
    },
    {
      "epoch": 0.6607142857142857,
      "grad_norm": 1.2055867539620164,
      "learning_rate": 7.541239523906742e-06,
      "loss": 0.3330733060836792,
      "num_input_tokens_seen": 314426512,
      "step": 5180,
      "train_runtime": 67394.0315,
      "train_tokens_per_second": 4665.495
    },
    {
      "epoch": 0.6613520408163265,
      "grad_norm": 0.6684195088007945,
      "learning_rate": 7.536924516104411e-06,
      "loss": 0.31601767539978026,
      "num_input_tokens_seen": 314716736,
      "step": 5185,
      "train_runtime": 67469.1985,
      "train_tokens_per_second": 4664.599
    },
    {
      "epoch": 0.6619897959183674,
      "grad_norm": 0.5894649896209727,
      "learning_rate": 7.532606962321149e-06,
      "loss": 0.32779479026794434,
      "num_input_tokens_seen": 315014520,
      "step": 5190,
      "train_runtime": 67526.9463,
      "train_tokens_per_second": 4665.019
    },
    {
      "epoch": 0.6626275510204082,
      "grad_norm": 0.6098349370529698,
      "learning_rate": 7.528286866889924e-06,
      "loss": 0.3065429925918579,
      "num_input_tokens_seen": 315320568,
      "step": 5195,
      "train_runtime": 67595.9776,
      "train_tokens_per_second": 4664.783
    },
    {
      "epoch": 0.6632653061224489,
      "grad_norm": 0.6309166508502535,
      "learning_rate": 7.523964234146251e-06,
      "loss": 0.34712815284729004,
      "num_input_tokens_seen": 315634520,
      "step": 5200,
      "train_runtime": 67649.3854,
      "train_tokens_per_second": 4665.741
    },
    {
      "epoch": 0.6639030612244898,
      "grad_norm": 0.600054875455631,
      "learning_rate": 7.519639068428199e-06,
      "loss": 0.3406857967376709,
      "num_input_tokens_seen": 315938296,
      "step": 5205,
      "train_runtime": 67714.4417,
      "train_tokens_per_second": 4665.745
    },
    {
      "epoch": 0.6645408163265306,
      "grad_norm": 0.6260107068106353,
      "learning_rate": 7.515311374076368e-06,
      "loss": 0.3166226625442505,
      "num_input_tokens_seen": 316243072,
      "step": 5210,
      "train_runtime": 67782.659,
      "train_tokens_per_second": 4665.545
    },
    {
      "epoch": 0.6651785714285714,
      "grad_norm": 0.7558778469353974,
      "learning_rate": 7.5109811554339054e-06,
      "loss": 0.33198843002319334,
      "num_input_tokens_seen": 316550280,
      "step": 5215,
      "train_runtime": 67850.8903,
      "train_tokens_per_second": 4665.381
    },
    {
      "epoch": 0.6658163265306123,
      "grad_norm": 0.6794956607022652,
      "learning_rate": 7.506648416846485e-06,
      "loss": 0.3436640739440918,
      "num_input_tokens_seen": 316868448,
      "step": 5220,
      "train_runtime": 67902.4544,
      "train_tokens_per_second": 4666.524
    },
    {
      "epoch": 0.6664540816326531,
      "grad_norm": 0.588239386568709,
      "learning_rate": 7.502313162662316e-06,
      "loss": 0.3301560878753662,
      "num_input_tokens_seen": 317168256,
      "step": 5225,
      "train_runtime": 67975.9438,
      "train_tokens_per_second": 4665.89
    },
    {
      "epoch": 0.6670918367346939,
      "grad_norm": 0.5473296248429648,
      "learning_rate": 7.4979753972321265e-06,
      "loss": 0.31626310348510744,
      "num_input_tokens_seen": 317480320,
      "step": 5230,
      "train_runtime": 68048.371,
      "train_tokens_per_second": 4665.509
    },
    {
      "epoch": 0.6677295918367347,
      "grad_norm": 0.5500807486411848,
      "learning_rate": 7.493635124909167e-06,
      "loss": 0.31904003620147703,
      "num_input_tokens_seen": 317786664,
      "step": 5235,
      "train_runtime": 68128.2384,
      "train_tokens_per_second": 4664.537
    },
    {
      "epoch": 0.6683673469387755,
      "grad_norm": 0.5811028943455897,
      "learning_rate": 7.489292350049205e-06,
      "loss": 0.32049336433410647,
      "num_input_tokens_seen": 318090448,
      "step": 5240,
      "train_runtime": 68185.3268,
      "train_tokens_per_second": 4665.087
    },
    {
      "epoch": 0.6690051020408163,
      "grad_norm": 0.5623988371433525,
      "learning_rate": 7.484947077010519e-06,
      "loss": 0.3046008348464966,
      "num_input_tokens_seen": 318407360,
      "step": 5245,
      "train_runtime": 68249.2821,
      "train_tokens_per_second": 4665.358
    },
    {
      "epoch": 0.6696428571428571,
      "grad_norm": 0.6776051509696445,
      "learning_rate": 7.480599310153891e-06,
      "loss": 0.3047685146331787,
      "num_input_tokens_seen": 318686592,
      "step": 5250,
      "train_runtime": 68313.5643,
      "train_tokens_per_second": 4665.056
    },
    {
      "epoch": 0.670280612244898,
      "grad_norm": 0.6224288425840941,
      "learning_rate": 7.476249053842613e-06,
      "loss": 0.3419510364532471,
      "num_input_tokens_seen": 318984160,
      "step": 5255,
      "train_runtime": 68380.1767,
      "train_tokens_per_second": 4664.863
    },
    {
      "epoch": 0.6709183673469388,
      "grad_norm": 0.6286778320000875,
      "learning_rate": 7.471896312442467e-06,
      "loss": 0.32067461013793946,
      "num_input_tokens_seen": 319279088,
      "step": 5260,
      "train_runtime": 68451.2372,
      "train_tokens_per_second": 4664.329
    },
    {
      "epoch": 0.6715561224489796,
      "grad_norm": 0.5408996274044002,
      "learning_rate": 7.467541090321735e-06,
      "loss": 0.3092599153518677,
      "num_input_tokens_seen": 319582768,
      "step": 5265,
      "train_runtime": 68517.0909,
      "train_tokens_per_second": 4664.278
    },
    {
      "epoch": 0.6721938775510204,
      "grad_norm": 0.6550156616436309,
      "learning_rate": 7.463183391851186e-06,
      "loss": 0.34105687141418456,
      "num_input_tokens_seen": 319887592,
      "step": 5270,
      "train_runtime": 68565.1059,
      "train_tokens_per_second": 4665.458
    },
    {
      "epoch": 0.6728316326530612,
      "grad_norm": 0.5640935858174217,
      "learning_rate": 7.4588232214040744e-06,
      "loss": 0.3312492370605469,
      "num_input_tokens_seen": 320179160,
      "step": 5275,
      "train_runtime": 68640.17,
      "train_tokens_per_second": 4664.603
    },
    {
      "epoch": 0.673469387755102,
      "grad_norm": 0.5802278666882329,
      "learning_rate": 7.454460583356138e-06,
      "loss": 0.31220436096191406,
      "num_input_tokens_seen": 320468448,
      "step": 5280,
      "train_runtime": 68701.3544,
      "train_tokens_per_second": 4664.66
    },
    {
      "epoch": 0.6741071428571429,
      "grad_norm": 0.5742901619279837,
      "learning_rate": 7.4500954820855854e-06,
      "loss": 0.32859785556793214,
      "num_input_tokens_seen": 320776112,
      "step": 5285,
      "train_runtime": 68763.7941,
      "train_tokens_per_second": 4664.898
    },
    {
      "epoch": 0.6747448979591837,
      "grad_norm": 0.6296472556685766,
      "learning_rate": 7.445727921973104e-06,
      "loss": 0.3388023853302002,
      "num_input_tokens_seen": 321083768,
      "step": 5290,
      "train_runtime": 68843.7298,
      "train_tokens_per_second": 4663.951
    },
    {
      "epoch": 0.6753826530612245,
      "grad_norm": 0.5939938217312132,
      "learning_rate": 7.441357907401841e-06,
      "loss": 0.349368953704834,
      "num_input_tokens_seen": 321388456,
      "step": 5295,
      "train_runtime": 68920.4066,
      "train_tokens_per_second": 4663.183
    },
    {
      "epoch": 0.6760204081632653,
      "grad_norm": 0.5950419311161232,
      "learning_rate": 7.436985442757415e-06,
      "loss": 0.3301686763763428,
      "num_input_tokens_seen": 321699936,
      "step": 5300,
      "train_runtime": 68976.6748,
      "train_tokens_per_second": 4663.895
    },
    {
      "epoch": 0.6766581632653061,
      "grad_norm": 0.5886652925647744,
      "learning_rate": 7.432610532427899e-06,
      "loss": 0.3316835403442383,
      "num_input_tokens_seen": 322016888,
      "step": 5305,
      "train_runtime": 69027.8101,
      "train_tokens_per_second": 4665.031
    },
    {
      "epoch": 0.6772959183673469,
      "grad_norm": 0.6483338428234111,
      "learning_rate": 7.428233180803821e-06,
      "loss": 0.3472487688064575,
      "num_input_tokens_seen": 322323416,
      "step": 5310,
      "train_runtime": 69087.3584,
      "train_tokens_per_second": 4665.447
    },
    {
      "epoch": 0.6779336734693877,
      "grad_norm": 0.6604829029071831,
      "learning_rate": 7.423853392278158e-06,
      "loss": 0.3495924949645996,
      "num_input_tokens_seen": 322618320,
      "step": 5315,
      "train_runtime": 69151.3099,
      "train_tokens_per_second": 4665.397
    },
    {
      "epoch": 0.6785714285714286,
      "grad_norm": 0.625851074036591,
      "learning_rate": 7.419471171246334e-06,
      "loss": 0.3318294286727905,
      "num_input_tokens_seen": 322920944,
      "step": 5320,
      "train_runtime": 69206.4594,
      "train_tokens_per_second": 4666.052
    },
    {
      "epoch": 0.6792091836734694,
      "grad_norm": 0.6787277188188651,
      "learning_rate": 7.415086522106215e-06,
      "loss": 0.33146181106567385,
      "num_input_tokens_seen": 323229392,
      "step": 5325,
      "train_runtime": 69267.3088,
      "train_tokens_per_second": 4666.406
    },
    {
      "epoch": 0.6798469387755102,
      "grad_norm": 0.588898335172044,
      "learning_rate": 7.410699449258102e-06,
      "loss": 0.3333503246307373,
      "num_input_tokens_seen": 323539944,
      "step": 5330,
      "train_runtime": 69330.8453,
      "train_tokens_per_second": 4666.609
    },
    {
      "epoch": 0.6804846938775511,
      "grad_norm": 0.6651399649625767,
      "learning_rate": 7.406309957104727e-06,
      "loss": 0.31888589859008787,
      "num_input_tokens_seen": 323845648,
      "step": 5335,
      "train_runtime": 69392.707,
      "train_tokens_per_second": 4666.854
    },
    {
      "epoch": 0.6811224489795918,
      "grad_norm": 0.6132995379294711,
      "learning_rate": 7.401918050051256e-06,
      "loss": 0.32655973434448243,
      "num_input_tokens_seen": 324152360,
      "step": 5340,
      "train_runtime": 69458.2621,
      "train_tokens_per_second": 4666.865
    },
    {
      "epoch": 0.6817602040816326,
      "grad_norm": 0.5914369333981617,
      "learning_rate": 7.39752373250527e-06,
      "loss": 0.3464339256286621,
      "num_input_tokens_seen": 324461336,
      "step": 5345,
      "train_runtime": 69528.0201,
      "train_tokens_per_second": 4666.627
    },
    {
      "epoch": 0.6823979591836735,
      "grad_norm": 0.6206028765459817,
      "learning_rate": 7.393127008876777e-06,
      "loss": 0.3387403726577759,
      "num_input_tokens_seen": 324765688,
      "step": 5350,
      "train_runtime": 69593.4509,
      "train_tokens_per_second": 4666.613
    },
    {
      "epoch": 0.6830357142857143,
      "grad_norm": 0.6124027013893614,
      "learning_rate": 7.388727883578195e-06,
      "loss": 0.33055853843688965,
      "num_input_tokens_seen": 325067576,
      "step": 5355,
      "train_runtime": 69659.8013,
      "train_tokens_per_second": 4666.502
    },
    {
      "epoch": 0.6836734693877551,
      "grad_norm": 0.578899436811348,
      "learning_rate": 7.384326361024352e-06,
      "loss": 0.3409270763397217,
      "num_input_tokens_seen": 325377208,
      "step": 5360,
      "train_runtime": 69725.547,
      "train_tokens_per_second": 4666.542
    },
    {
      "epoch": 0.6843112244897959,
      "grad_norm": 1.4905814522448033,
      "learning_rate": 7.3799224456324865e-06,
      "loss": 0.34844231605529785,
      "num_input_tokens_seen": 325686680,
      "step": 5365,
      "train_runtime": 69780.2504,
      "train_tokens_per_second": 4667.319
    },
    {
      "epoch": 0.6849489795918368,
      "grad_norm": 0.6700794882530573,
      "learning_rate": 7.375516141822232e-06,
      "loss": 0.34676122665405273,
      "num_input_tokens_seen": 325995200,
      "step": 5370,
      "train_runtime": 69836.2209,
      "train_tokens_per_second": 4667.996
    },
    {
      "epoch": 0.6855867346938775,
      "grad_norm": 0.6015316796392,
      "learning_rate": 7.371107454015626e-06,
      "loss": 0.3110495090484619,
      "num_input_tokens_seen": 326303576,
      "step": 5375,
      "train_runtime": 69906.6207,
      "train_tokens_per_second": 4667.706
    },
    {
      "epoch": 0.6862244897959183,
      "grad_norm": 0.5671787721416277,
      "learning_rate": 7.3666963866370886e-06,
      "loss": 0.35372376441955566,
      "num_input_tokens_seen": 326611640,
      "step": 5380,
      "train_runtime": 69973.4664,
      "train_tokens_per_second": 4667.65
    },
    {
      "epoch": 0.6868622448979592,
      "grad_norm": 0.5416291571428138,
      "learning_rate": 7.3622829441134395e-06,
      "loss": 0.327644419670105,
      "num_input_tokens_seen": 326919968,
      "step": 5385,
      "train_runtime": 70040.409,
      "train_tokens_per_second": 4667.591
    },
    {
      "epoch": 0.6875,
      "grad_norm": 0.6808143576663714,
      "learning_rate": 7.357867130873875e-06,
      "loss": 0.3390071392059326,
      "num_input_tokens_seen": 327232832,
      "step": 5390,
      "train_runtime": 70100.1573,
      "train_tokens_per_second": 4668.076
    },
    {
      "epoch": 0.6881377551020408,
      "grad_norm": 0.6059110568690664,
      "learning_rate": 7.353448951349971e-06,
      "loss": 0.32433485984802246,
      "num_input_tokens_seen": 327541008,
      "step": 5395,
      "train_runtime": 70165.5977,
      "train_tokens_per_second": 4668.114
    },
    {
      "epoch": 0.6887755102040817,
      "grad_norm": 0.7204965819899145,
      "learning_rate": 7.349028409975681e-06,
      "loss": 0.3259130477905273,
      "num_input_tokens_seen": 327827272,
      "step": 5400,
      "train_runtime": 70237.2177,
      "train_tokens_per_second": 4667.43
    },
    {
      "epoch": 0.6894132653061225,
      "grad_norm": 0.6122643776935173,
      "learning_rate": 7.344605511187322e-06,
      "loss": 0.327258038520813,
      "num_input_tokens_seen": 328118504,
      "step": 5405,
      "train_runtime": 70293.2273,
      "train_tokens_per_second": 4667.854
    },
    {
      "epoch": 0.6900510204081632,
      "grad_norm": 0.5563082324746123,
      "learning_rate": 7.3401802594235894e-06,
      "loss": 0.3184638023376465,
      "num_input_tokens_seen": 328421136,
      "step": 5410,
      "train_runtime": 70363.2588,
      "train_tokens_per_second": 4667.509
    },
    {
      "epoch": 0.6906887755102041,
      "grad_norm": 0.5963140377954498,
      "learning_rate": 7.3357526591255265e-06,
      "loss": 0.327569317817688,
      "num_input_tokens_seen": 328718016,
      "step": 5415,
      "train_runtime": 70433.2862,
      "train_tokens_per_second": 4667.083
    },
    {
      "epoch": 0.6913265306122449,
      "grad_norm": 0.583905619885943,
      "learning_rate": 7.331322714736543e-06,
      "loss": 0.33679890632629395,
      "num_input_tokens_seen": 329015224,
      "step": 5420,
      "train_runtime": 70503.2202,
      "train_tokens_per_second": 4666.669
    },
    {
      "epoch": 0.6919642857142857,
      "grad_norm": 0.6177694828222271,
      "learning_rate": 7.326890430702396e-06,
      "loss": 0.3472728252410889,
      "num_input_tokens_seen": 329314184,
      "step": 5425,
      "train_runtime": 70571.6455,
      "train_tokens_per_second": 4666.381
    },
    {
      "epoch": 0.6926020408163265,
      "grad_norm": 0.5902678991465765,
      "learning_rate": 7.3224558114711905e-06,
      "loss": 0.3267884016036987,
      "num_input_tokens_seen": 329621696,
      "step": 5430,
      "train_runtime": 70641.1305,
      "train_tokens_per_second": 4666.144
    },
    {
      "epoch": 0.6932397959183674,
      "grad_norm": 0.7533725371491875,
      "learning_rate": 7.31801886149338e-06,
      "loss": 0.30158908367156984,
      "num_input_tokens_seen": 329919672,
      "step": 5435,
      "train_runtime": 70709.638,
      "train_tokens_per_second": 4665.837
    },
    {
      "epoch": 0.6938775510204082,
      "grad_norm": 0.666889462817546,
      "learning_rate": 7.313579585221752e-06,
      "loss": 0.34727749824523924,
      "num_input_tokens_seen": 330233896,
      "step": 5440,
      "train_runtime": 70764.7404,
      "train_tokens_per_second": 4666.645
    },
    {
      "epoch": 0.6945153061224489,
      "grad_norm": 0.5907073754530744,
      "learning_rate": 7.309137987111428e-06,
      "loss": 0.33589582443237304,
      "num_input_tokens_seen": 330530192,
      "step": 5445,
      "train_runtime": 70827.9949,
      "train_tokens_per_second": 4666.66
    },
    {
      "epoch": 0.6951530612244898,
      "grad_norm": 0.6530789790493967,
      "learning_rate": 7.304694071619866e-06,
      "loss": 0.3435753345489502,
      "num_input_tokens_seen": 330839440,
      "step": 5450,
      "train_runtime": 70882.9173,
      "train_tokens_per_second": 4667.407
    },
    {
      "epoch": 0.6957908163265306,
      "grad_norm": 0.5808913398797967,
      "learning_rate": 7.300247843206844e-06,
      "loss": 0.33095269203186034,
      "num_input_tokens_seen": 331125736,
      "step": 5455,
      "train_runtime": 70938.9735,
      "train_tokens_per_second": 4667.755
    },
    {
      "epoch": 0.6964285714285714,
      "grad_norm": 0.657713456241302,
      "learning_rate": 7.2957993063344615e-06,
      "loss": 0.3572857856750488,
      "num_input_tokens_seen": 331437480,
      "step": 5460,
      "train_runtime": 70996.2714,
      "train_tokens_per_second": 4668.379
    },
    {
      "epoch": 0.6970663265306123,
      "grad_norm": 0.661292725378582,
      "learning_rate": 7.291348465467136e-06,
      "loss": 0.32533111572265627,
      "num_input_tokens_seen": 331737480,
      "step": 5465,
      "train_runtime": 71066.994,
      "train_tokens_per_second": 4667.954
    },
    {
      "epoch": 0.6977040816326531,
      "grad_norm": 0.5561707413448671,
      "learning_rate": 7.286895325071599e-06,
      "loss": 0.3380601167678833,
      "num_input_tokens_seen": 332047296,
      "step": 5470,
      "train_runtime": 71130.3039,
      "train_tokens_per_second": 4668.155
    },
    {
      "epoch": 0.6983418367346939,
      "grad_norm": 0.5911161863962754,
      "learning_rate": 7.282439889616887e-06,
      "loss": 0.30793912410736085,
      "num_input_tokens_seen": 332348792,
      "step": 5475,
      "train_runtime": 71197.7994,
      "train_tokens_per_second": 4667.964
    },
    {
      "epoch": 0.6989795918367347,
      "grad_norm": 0.5788008200790318,
      "learning_rate": 7.277982163574339e-06,
      "loss": 0.33248081207275393,
      "num_input_tokens_seen": 332647496,
      "step": 5480,
      "train_runtime": 71256.7339,
      "train_tokens_per_second": 4668.296
    },
    {
      "epoch": 0.6996173469387755,
      "grad_norm": 0.563191763270536,
      "learning_rate": 7.273522151417598e-06,
      "loss": 0.327709436416626,
      "num_input_tokens_seen": 332958368,
      "step": 5485,
      "train_runtime": 71329.611,
      "train_tokens_per_second": 4667.884
    },
    {
      "epoch": 0.7002551020408163,
      "grad_norm": 0.6206240771212169,
      "learning_rate": 7.2690598576225954e-06,
      "loss": 0.327768611907959,
      "num_input_tokens_seen": 333253504,
      "step": 5490,
      "train_runtime": 71406.3473,
      "train_tokens_per_second": 4667.001
    },
    {
      "epoch": 0.7008928571428571,
      "grad_norm": 0.6607129218814294,
      "learning_rate": 7.264595286667554e-06,
      "loss": 0.3356297254562378,
      "num_input_tokens_seen": 333529432,
      "step": 5495,
      "train_runtime": 71473.3755,
      "train_tokens_per_second": 4666.485
    },
    {
      "epoch": 0.701530612244898,
      "grad_norm": 0.6302316526555676,
      "learning_rate": 7.2601284430329836e-06,
      "loss": 0.31958589553833006,
      "num_input_tokens_seen": 333837952,
      "step": 5500,
      "train_runtime": 71540.7581,
      "train_tokens_per_second": 4666.402
    },
    {
      "epoch": 0.7021683673469388,
      "grad_norm": 0.582175265602265,
      "learning_rate": 7.255659331201673e-06,
      "loss": 0.31284639835357664,
      "num_input_tokens_seen": 334139480,
      "step": 5505,
      "train_runtime": 71609.7302,
      "train_tokens_per_second": 4666.118
    },
    {
      "epoch": 0.7028061224489796,
      "grad_norm": 0.5642191190377086,
      "learning_rate": 7.251187955658691e-06,
      "loss": 0.3290869235992432,
      "num_input_tokens_seen": 334432600,
      "step": 5510,
      "train_runtime": 71672.4885,
      "train_tokens_per_second": 4666.122
    },
    {
      "epoch": 0.7034438775510204,
      "grad_norm": 0.5920364480819984,
      "learning_rate": 7.246714320891371e-06,
      "loss": 0.3378335237503052,
      "num_input_tokens_seen": 334738240,
      "step": 5515,
      "train_runtime": 71744.6332,
      "train_tokens_per_second": 4665.69
    },
    {
      "epoch": 0.7040816326530612,
      "grad_norm": 0.6125949109455139,
      "learning_rate": 7.24223843138932e-06,
      "loss": 0.32679245471954343,
      "num_input_tokens_seen": 335033688,
      "step": 5520,
      "train_runtime": 71814.7444,
      "train_tokens_per_second": 4665.249
    },
    {
      "epoch": 0.704719387755102,
      "grad_norm": 0.6221386376333026,
      "learning_rate": 7.237760291644405e-06,
      "loss": 0.3354485988616943,
      "num_input_tokens_seen": 335339536,
      "step": 5525,
      "train_runtime": 71874.4703,
      "train_tokens_per_second": 4665.628
    },
    {
      "epoch": 0.7053571428571429,
      "grad_norm": 0.5862948755770035,
      "learning_rate": 7.233279906150752e-06,
      "loss": 0.3378009796142578,
      "num_input_tokens_seen": 335617824,
      "step": 5530,
      "train_runtime": 71940.2307,
      "train_tokens_per_second": 4665.231
    },
    {
      "epoch": 0.7059948979591837,
      "grad_norm": 0.590531810670554,
      "learning_rate": 7.228797279404744e-06,
      "loss": 0.3126326322555542,
      "num_input_tokens_seen": 335926560,
      "step": 5535,
      "train_runtime": 72005.3302,
      "train_tokens_per_second": 4665.301
    },
    {
      "epoch": 0.7066326530612245,
      "grad_norm": 0.5867686757091221,
      "learning_rate": 7.224312415905005e-06,
      "loss": 0.3219478130340576,
      "num_input_tokens_seen": 336238896,
      "step": 5540,
      "train_runtime": 72075.2057,
      "train_tokens_per_second": 4665.112
    },
    {
      "epoch": 0.7072704081632653,
      "grad_norm": 0.7170878987217519,
      "learning_rate": 7.219825320152411e-06,
      "loss": 0.31927881240844724,
      "num_input_tokens_seen": 336545928,
      "step": 5545,
      "train_runtime": 72141.8066,
      "train_tokens_per_second": 4665.061
    },
    {
      "epoch": 0.7079081632653061,
      "grad_norm": 0.6497445986568611,
      "learning_rate": 7.2153359966500765e-06,
      "loss": 0.33097076416015625,
      "num_input_tokens_seen": 336863184,
      "step": 5550,
      "train_runtime": 72205.9275,
      "train_tokens_per_second": 4665.312
    },
    {
      "epoch": 0.7085459183673469,
      "grad_norm": 0.5740950852414692,
      "learning_rate": 7.210844449903352e-06,
      "loss": 0.3214916229248047,
      "num_input_tokens_seen": 337164816,
      "step": 5555,
      "train_runtime": 72272.0232,
      "train_tokens_per_second": 4665.219
    },
    {
      "epoch": 0.7091836734693877,
      "grad_norm": 0.6232369971662685,
      "learning_rate": 7.206350684419814e-06,
      "loss": 0.3146527767181396,
      "num_input_tokens_seen": 337478608,
      "step": 5560,
      "train_runtime": 72329.9707,
      "train_tokens_per_second": 4665.82
    },
    {
      "epoch": 0.7098214285714286,
      "grad_norm": 0.6216524464851271,
      "learning_rate": 7.2018547047092745e-06,
      "loss": 0.3411190748214722,
      "num_input_tokens_seen": 337792616,
      "step": 5565,
      "train_runtime": 72400.4977,
      "train_tokens_per_second": 4665.612
    },
    {
      "epoch": 0.7104591836734694,
      "grad_norm": 0.6316057403682567,
      "learning_rate": 7.1973565152837635e-06,
      "loss": 0.3246105194091797,
      "num_input_tokens_seen": 338097400,
      "step": 5570,
      "train_runtime": 72459.0139,
      "train_tokens_per_second": 4666.05
    },
    {
      "epoch": 0.7110969387755102,
      "grad_norm": 0.5859982217550813,
      "learning_rate": 7.192856120657524e-06,
      "loss": 0.33641302585601807,
      "num_input_tokens_seen": 338401176,
      "step": 5575,
      "train_runtime": 72532.5175,
      "train_tokens_per_second": 4665.51
    },
    {
      "epoch": 0.7117346938775511,
      "grad_norm": 0.6600579934426793,
      "learning_rate": 7.18835352534702e-06,
      "loss": 0.30387799739837645,
      "num_input_tokens_seen": 338686568,
      "step": 5580,
      "train_runtime": 72603.8349,
      "train_tokens_per_second": 4664.858
    },
    {
      "epoch": 0.7123724489795918,
      "grad_norm": 0.5854797426337409,
      "learning_rate": 7.183848733870917e-06,
      "loss": 0.319439697265625,
      "num_input_tokens_seen": 338995184,
      "step": 5585,
      "train_runtime": 72667.0423,
      "train_tokens_per_second": 4665.047
    },
    {
      "epoch": 0.7130102040816326,
      "grad_norm": 0.5689084429663139,
      "learning_rate": 7.179341750750092e-06,
      "loss": 0.3119250059127808,
      "num_input_tokens_seen": 339304344,
      "step": 5590,
      "train_runtime": 72730.0804,
      "train_tokens_per_second": 4665.255
    },
    {
      "epoch": 0.7136479591836735,
      "grad_norm": 0.6906071777710958,
      "learning_rate": 7.174832580507614e-06,
      "loss": 0.37570886611938475,
      "num_input_tokens_seen": 339600024,
      "step": 5595,
      "train_runtime": 72789.0288,
      "train_tokens_per_second": 4665.539
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 0.5688465678450141,
      "learning_rate": 7.170321227668752e-06,
      "loss": 0.3130634784698486,
      "num_input_tokens_seen": 339897768,
      "step": 5600,
      "train_runtime": 72855.4263,
      "train_tokens_per_second": 4665.373
    },
    {
      "epoch": 0.7149234693877551,
      "grad_norm": 0.5687010721380742,
      "learning_rate": 7.165807696760964e-06,
      "loss": 0.3177204608917236,
      "num_input_tokens_seen": 340201488,
      "step": 5605,
      "train_runtime": 72921.782,
      "train_tokens_per_second": 4665.293
    },
    {
      "epoch": 0.7155612244897959,
      "grad_norm": 0.5668284353788912,
      "learning_rate": 7.16129199231389e-06,
      "loss": 0.31652255058288575,
      "num_input_tokens_seen": 340513480,
      "step": 5610,
      "train_runtime": 72986.5017,
      "train_tokens_per_second": 4665.431
    },
    {
      "epoch": 0.7161989795918368,
      "grad_norm": 0.6065589618769056,
      "learning_rate": 7.15677411885936e-06,
      "loss": 0.32903294563293456,
      "num_input_tokens_seen": 340814592,
      "step": 5615,
      "train_runtime": 73055.5021,
      "train_tokens_per_second": 4665.146
    },
    {
      "epoch": 0.7168367346938775,
      "grad_norm": 0.6199070267833474,
      "learning_rate": 7.152254080931372e-06,
      "loss": 0.34875319004058836,
      "num_input_tokens_seen": 341125176,
      "step": 5620,
      "train_runtime": 73125.7921,
      "train_tokens_per_second": 4664.909
    },
    {
      "epoch": 0.7174744897959183,
      "grad_norm": 0.6459436729232075,
      "learning_rate": 7.1477318830661e-06,
      "loss": 0.36013131141662597,
      "num_input_tokens_seen": 341427632,
      "step": 5625,
      "train_runtime": 73194.3893,
      "train_tokens_per_second": 4664.669
    },
    {
      "epoch": 0.7181122448979592,
      "grad_norm": 0.5738189271833315,
      "learning_rate": 7.143207529801888e-06,
      "loss": 0.3343724250793457,
      "num_input_tokens_seen": 341719568,
      "step": 5630,
      "train_runtime": 73260.3066,
      "train_tokens_per_second": 4664.457
    },
    {
      "epoch": 0.71875,
      "grad_norm": 0.5858170490383494,
      "learning_rate": 7.1386810256792374e-06,
      "loss": 0.3392931938171387,
      "num_input_tokens_seen": 342019008,
      "step": 5635,
      "train_runtime": 73333.7142,
      "train_tokens_per_second": 4663.871
    },
    {
      "epoch": 0.7193877551020408,
      "grad_norm": 0.6160116996822735,
      "learning_rate": 7.134152375240812e-06,
      "loss": 0.29196462631225584,
      "num_input_tokens_seen": 342315160,
      "step": 5640,
      "train_runtime": 73405.6238,
      "train_tokens_per_second": 4663.337
    },
    {
      "epoch": 0.7200255102040817,
      "grad_norm": 0.5570448929611207,
      "learning_rate": 7.1296215830314295e-06,
      "loss": 0.3171334505081177,
      "num_input_tokens_seen": 342614960,
      "step": 5645,
      "train_runtime": 73466.3518,
      "train_tokens_per_second": 4663.563
    },
    {
      "epoch": 0.7206632653061225,
      "grad_norm": 0.6103563782721396,
      "learning_rate": 7.125088653598057e-06,
      "loss": 0.3042132377624512,
      "num_input_tokens_seen": 342912480,
      "step": 5650,
      "train_runtime": 73530.6953,
      "train_tokens_per_second": 4663.528
    },
    {
      "epoch": 0.7213010204081632,
      "grad_norm": 0.5953809118782077,
      "learning_rate": 7.1205535914898035e-06,
      "loss": 0.3171056270599365,
      "num_input_tokens_seen": 343216744,
      "step": 5655,
      "train_runtime": 73595.9657,
      "train_tokens_per_second": 4663.527
    },
    {
      "epoch": 0.7219387755102041,
      "grad_norm": 0.624785105512535,
      "learning_rate": 7.116016401257921e-06,
      "loss": 0.3005784034729004,
      "num_input_tokens_seen": 343522224,
      "step": 5660,
      "train_runtime": 73659.3765,
      "train_tokens_per_second": 4663.659
    },
    {
      "epoch": 0.7225765306122449,
      "grad_norm": 0.6666384460772958,
      "learning_rate": 7.1114770874558e-06,
      "loss": 0.33109331130981445,
      "num_input_tokens_seen": 343836648,
      "step": 5665,
      "train_runtime": 73724.3917,
      "train_tokens_per_second": 4663.811
    },
    {
      "epoch": 0.7232142857142857,
      "grad_norm": 0.6200346099959192,
      "learning_rate": 7.1069356546389555e-06,
      "loss": 0.3131404399871826,
      "num_input_tokens_seen": 344124648,
      "step": 5670,
      "train_runtime": 73798.7997,
      "train_tokens_per_second": 4663.011
    },
    {
      "epoch": 0.7238520408163265,
      "grad_norm": 0.6876751716263368,
      "learning_rate": 7.1023921073650325e-06,
      "loss": 0.34466369152069093,
      "num_input_tokens_seen": 344412648,
      "step": 5675,
      "train_runtime": 73855.4871,
      "train_tokens_per_second": 4663.332
    },
    {
      "epoch": 0.7244897959183674,
      "grad_norm": 0.6087284744869788,
      "learning_rate": 7.0978464501938e-06,
      "loss": 0.3497285842895508,
      "num_input_tokens_seen": 344714768,
      "step": 5680,
      "train_runtime": 73922.9284,
      "train_tokens_per_second": 4663.164
    },
    {
      "epoch": 0.7251275510204082,
      "grad_norm": 0.5739136579301118,
      "learning_rate": 7.093298687687141e-06,
      "loss": 0.3339118242263794,
      "num_input_tokens_seen": 345026816,
      "step": 5685,
      "train_runtime": 73988.7422,
      "train_tokens_per_second": 4663.234
    },
    {
      "epoch": 0.7257653061224489,
      "grad_norm": 0.692778186737699,
      "learning_rate": 7.088748824409053e-06,
      "loss": 0.36696739196777345,
      "num_input_tokens_seen": 345326376,
      "step": 5690,
      "train_runtime": 74052.1506,
      "train_tokens_per_second": 4663.286
    },
    {
      "epoch": 0.7264030612244898,
      "grad_norm": 0.6108096494683619,
      "learning_rate": 7.084196864925643e-06,
      "loss": 0.3174098491668701,
      "num_input_tokens_seen": 345631752,
      "step": 5695,
      "train_runtime": 74115.11,
      "train_tokens_per_second": 4663.445
    },
    {
      "epoch": 0.7270408163265306,
      "grad_norm": 0.6099136171760403,
      "learning_rate": 7.079642813805118e-06,
      "loss": 0.33266453742980956,
      "num_input_tokens_seen": 345940288,
      "step": 5700,
      "train_runtime": 74177.7298,
      "train_tokens_per_second": 4663.668
    },
    {
      "epoch": 0.7276785714285714,
      "grad_norm": 0.6145960848559486,
      "learning_rate": 7.075086675617788e-06,
      "loss": 0.33375253677368166,
      "num_input_tokens_seen": 346246208,
      "step": 5705,
      "train_runtime": 74241.1258,
      "train_tokens_per_second": 4663.806
    },
    {
      "epoch": 0.7283163265306123,
      "grad_norm": 0.6647919581563738,
      "learning_rate": 7.070528454936054e-06,
      "loss": 0.3297909736633301,
      "num_input_tokens_seen": 346546152,
      "step": 5710,
      "train_runtime": 74299.9081,
      "train_tokens_per_second": 4664.153
    },
    {
      "epoch": 0.7289540816326531,
      "grad_norm": 0.6745637918147991,
      "learning_rate": 7.065968156334413e-06,
      "loss": 0.3185250759124756,
      "num_input_tokens_seen": 346831096,
      "step": 5715,
      "train_runtime": 74366.2627,
      "train_tokens_per_second": 4663.823
    },
    {
      "epoch": 0.7295918367346939,
      "grad_norm": 0.5538137665844209,
      "learning_rate": 7.061405784389438e-06,
      "loss": 0.32861080169677737,
      "num_input_tokens_seen": 347131824,
      "step": 5720,
      "train_runtime": 74417.165,
      "train_tokens_per_second": 4664.674
    },
    {
      "epoch": 0.7302295918367347,
      "grad_norm": 0.6141795221101976,
      "learning_rate": 7.05684134367979e-06,
      "loss": 0.33352041244506836,
      "num_input_tokens_seen": 347433720,
      "step": 5725,
      "train_runtime": 74481.5452,
      "train_tokens_per_second": 4664.695
    },
    {
      "epoch": 0.7308673469387755,
      "grad_norm": 0.6315137917901033,
      "learning_rate": 7.052274838786204e-06,
      "loss": 0.3256068706512451,
      "num_input_tokens_seen": 347739456,
      "step": 5730,
      "train_runtime": 74545.587,
      "train_tokens_per_second": 4664.789
    },
    {
      "epoch": 0.7315051020408163,
      "grad_norm": 0.5899822147597998,
      "learning_rate": 7.047706274291488e-06,
      "loss": 0.3421637535095215,
      "num_input_tokens_seen": 348042144,
      "step": 5735,
      "train_runtime": 74611.9649,
      "train_tokens_per_second": 4664.696
    },
    {
      "epoch": 0.7321428571428571,
      "grad_norm": 0.6383420146332125,
      "learning_rate": 7.043135654780513e-06,
      "loss": 0.3239234447479248,
      "num_input_tokens_seen": 348349184,
      "step": 5740,
      "train_runtime": 74673.1392,
      "train_tokens_per_second": 4664.986
    },
    {
      "epoch": 0.732780612244898,
      "grad_norm": 0.6647982665082789,
      "learning_rate": 7.038562984840216e-06,
      "loss": 0.3468273878097534,
      "num_input_tokens_seen": 348657576,
      "step": 5745,
      "train_runtime": 74743.6028,
      "train_tokens_per_second": 4664.715
    },
    {
      "epoch": 0.7334183673469388,
      "grad_norm": 0.6360341258895988,
      "learning_rate": 7.033988269059593e-06,
      "loss": 0.35527334213256834,
      "num_input_tokens_seen": 348959448,
      "step": 5750,
      "train_runtime": 74818.5321,
      "train_tokens_per_second": 4664.078
    },
    {
      "epoch": 0.7340561224489796,
      "grad_norm": 0.6551149745352199,
      "learning_rate": 7.029411512029687e-06,
      "loss": 0.3083681583404541,
      "num_input_tokens_seen": 349278664,
      "step": 5755,
      "train_runtime": 74884.9479,
      "train_tokens_per_second": 4664.204
    },
    {
      "epoch": 0.7346938775510204,
      "grad_norm": 0.5597358503159139,
      "learning_rate": 7.024832718343594e-06,
      "loss": 0.3165610313415527,
      "num_input_tokens_seen": 349589600,
      "step": 5760,
      "train_runtime": 74936.2695,
      "train_tokens_per_second": 4665.159
    },
    {
      "epoch": 0.7353316326530612,
      "grad_norm": 0.602928020388534,
      "learning_rate": 7.020251892596456e-06,
      "loss": 0.30977587699890136,
      "num_input_tokens_seen": 349870128,
      "step": 5765,
      "train_runtime": 74992.9054,
      "train_tokens_per_second": 4665.376
    },
    {
      "epoch": 0.735969387755102,
      "grad_norm": 0.559042510700635,
      "learning_rate": 7.015669039385449e-06,
      "loss": 0.33393781185150145,
      "num_input_tokens_seen": 350175688,
      "step": 5770,
      "train_runtime": 75048.7015,
      "train_tokens_per_second": 4665.979
    },
    {
      "epoch": 0.7366071428571429,
      "grad_norm": 0.6140008788178031,
      "learning_rate": 7.011084163309786e-06,
      "loss": 0.36183881759643555,
      "num_input_tokens_seen": 350475216,
      "step": 5775,
      "train_runtime": 75116.1318,
      "train_tokens_per_second": 4665.778
    },
    {
      "epoch": 0.7372448979591837,
      "grad_norm": 0.6520628271953471,
      "learning_rate": 7.006497268970712e-06,
      "loss": 0.33647539615631106,
      "num_input_tokens_seen": 350786336,
      "step": 5780,
      "train_runtime": 75182.9294,
      "train_tokens_per_second": 4665.771
    },
    {
      "epoch": 0.7378826530612245,
      "grad_norm": 0.6233474505973957,
      "learning_rate": 7.001908360971495e-06,
      "loss": 0.3370200634002686,
      "num_input_tokens_seen": 351099720,
      "step": 5785,
      "train_runtime": 75237.035,
      "train_tokens_per_second": 4666.581
    },
    {
      "epoch": 0.7385204081632653,
      "grad_norm": 0.7470598154815499,
      "learning_rate": 6.997317443917424e-06,
      "loss": 0.3196087121963501,
      "num_input_tokens_seen": 351392808,
      "step": 5790,
      "train_runtime": 75302.5827,
      "train_tokens_per_second": 4666.411
    },
    {
      "epoch": 0.7391581632653061,
      "grad_norm": 0.5461156408706147,
      "learning_rate": 6.9927245224158066e-06,
      "loss": 0.34132537841796873,
      "num_input_tokens_seen": 351698480,
      "step": 5795,
      "train_runtime": 75374.0898,
      "train_tokens_per_second": 4666.039
    },
    {
      "epoch": 0.7397959183673469,
      "grad_norm": 0.548528853606226,
      "learning_rate": 6.988129601075961e-06,
      "loss": 0.33192293643951415,
      "num_input_tokens_seen": 352000720,
      "step": 5800,
      "train_runtime": 75441.4729,
      "train_tokens_per_second": 4665.878
    },
    {
      "epoch": 0.7404336734693877,
      "grad_norm": 0.7104822501217407,
      "learning_rate": 6.983532684509208e-06,
      "loss": 0.34473247528076173,
      "num_input_tokens_seen": 352294224,
      "step": 5805,
      "train_runtime": 75506.1737,
      "train_tokens_per_second": 4665.767
    },
    {
      "epoch": 0.7410714285714286,
      "grad_norm": 0.5885968025978827,
      "learning_rate": 6.978933777328878e-06,
      "loss": 0.32946100234985354,
      "num_input_tokens_seen": 352604544,
      "step": 5810,
      "train_runtime": 75582.1313,
      "train_tokens_per_second": 4665.184
    },
    {
      "epoch": 0.7417091836734694,
      "grad_norm": 0.5862463096375644,
      "learning_rate": 6.974332884150292e-06,
      "loss": 0.33791561126708985,
      "num_input_tokens_seen": 352914192,
      "step": 5815,
      "train_runtime": 75655.7869,
      "train_tokens_per_second": 4664.735
    },
    {
      "epoch": 0.7423469387755102,
      "grad_norm": 0.6387843424603685,
      "learning_rate": 6.96973000959077e-06,
      "loss": 0.3198561191558838,
      "num_input_tokens_seen": 353228696,
      "step": 5820,
      "train_runtime": 75718.2625,
      "train_tokens_per_second": 4665.04
    },
    {
      "epoch": 0.7429846938775511,
      "grad_norm": 0.6156698937848678,
      "learning_rate": 6.965125158269619e-06,
      "loss": 0.3436471462249756,
      "num_input_tokens_seen": 353528736,
      "step": 5825,
      "train_runtime": 75785.1764,
      "train_tokens_per_second": 4664.88
    },
    {
      "epoch": 0.7436224489795918,
      "grad_norm": 0.6463215182094665,
      "learning_rate": 6.9605183348081265e-06,
      "loss": 0.35364034175872805,
      "num_input_tokens_seen": 353839872,
      "step": 5830,
      "train_runtime": 75838.2181,
      "train_tokens_per_second": 4665.72
    },
    {
      "epoch": 0.7442602040816326,
      "grad_norm": 0.5705940916617028,
      "learning_rate": 6.955909543829564e-06,
      "loss": 0.34197463989257815,
      "num_input_tokens_seen": 354148056,
      "step": 5835,
      "train_runtime": 75908.2897,
      "train_tokens_per_second": 4665.473
    },
    {
      "epoch": 0.7448979591836735,
      "grad_norm": 0.6150639786222485,
      "learning_rate": 6.951298789959172e-06,
      "loss": 0.32314419746398926,
      "num_input_tokens_seen": 354445568,
      "step": 5840,
      "train_runtime": 75979.1681,
      "train_tokens_per_second": 4665.036
    },
    {
      "epoch": 0.7455357142857143,
      "grad_norm": 0.5764764170987056,
      "learning_rate": 6.946686077824167e-06,
      "loss": 0.3016700267791748,
      "num_input_tokens_seen": 354726728,
      "step": 5845,
      "train_runtime": 76043.4819,
      "train_tokens_per_second": 4664.788
    },
    {
      "epoch": 0.7461734693877551,
      "grad_norm": 0.6244869964936728,
      "learning_rate": 6.942071412053728e-06,
      "loss": 0.3447637319564819,
      "num_input_tokens_seen": 355041640,
      "step": 5850,
      "train_runtime": 76102.5316,
      "train_tokens_per_second": 4665.307
    },
    {
      "epoch": 0.7468112244897959,
      "grad_norm": 0.5437386597279267,
      "learning_rate": 6.937454797278991e-06,
      "loss": 0.30813610553741455,
      "num_input_tokens_seen": 355341584,
      "step": 5855,
      "train_runtime": 76165.1398,
      "train_tokens_per_second": 4665.41
    },
    {
      "epoch": 0.7474489795918368,
      "grad_norm": 0.6129276896456994,
      "learning_rate": 6.932836238133054e-06,
      "loss": 0.332719612121582,
      "num_input_tokens_seen": 355644760,
      "step": 5860,
      "train_runtime": 76242.1853,
      "train_tokens_per_second": 4664.672
    },
    {
      "epoch": 0.7480867346938775,
      "grad_norm": 0.6254378993439251,
      "learning_rate": 6.928215739250963e-06,
      "loss": 0.3112212657928467,
      "num_input_tokens_seen": 355938224,
      "step": 5865,
      "train_runtime": 76316.5034,
      "train_tokens_per_second": 4663.974
    },
    {
      "epoch": 0.7487244897959183,
      "grad_norm": 0.7044378334523771,
      "learning_rate": 6.923593305269711e-06,
      "loss": 0.3458071708679199,
      "num_input_tokens_seen": 356230344,
      "step": 5870,
      "train_runtime": 76378.7704,
      "train_tokens_per_second": 4663.997
    },
    {
      "epoch": 0.7493622448979592,
      "grad_norm": 0.5746317383052223,
      "learning_rate": 6.9189689408282345e-06,
      "loss": 0.3587623119354248,
      "num_input_tokens_seen": 356541344,
      "step": 5875,
      "train_runtime": 76446.2596,
      "train_tokens_per_second": 4663.948
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5839156599948164,
      "learning_rate": 6.914342650567405e-06,
      "loss": 0.3423443794250488,
      "num_input_tokens_seen": 356855496,
      "step": 5880,
      "train_runtime": 76514.8126,
      "train_tokens_per_second": 4663.875
    },
    {
      "epoch": 0.7506377551020408,
      "grad_norm": 0.6067382079199007,
      "learning_rate": 6.9097144391300285e-06,
      "loss": 0.3316153287887573,
      "num_input_tokens_seen": 357168744,
      "step": 5885,
      "train_runtime": 76567.3764,
      "train_tokens_per_second": 4664.764
    },
    {
      "epoch": 0.7512755102040817,
      "grad_norm": 0.6851279100024733,
      "learning_rate": 6.905084311160837e-06,
      "loss": 0.3190103054046631,
      "num_input_tokens_seen": 357471984,
      "step": 5890,
      "train_runtime": 76635.8642,
      "train_tokens_per_second": 4664.552
    },
    {
      "epoch": 0.7519132653061225,
      "grad_norm": 0.6147269666210299,
      "learning_rate": 6.90045227130649e-06,
      "loss": 0.3483097076416016,
      "num_input_tokens_seen": 357782784,
      "step": 5895,
      "train_runtime": 76693.7951,
      "train_tokens_per_second": 4665.081
    },
    {
      "epoch": 0.7525510204081632,
      "grad_norm": 0.564602641320379,
      "learning_rate": 6.8958183242155615e-06,
      "loss": 0.323588228225708,
      "num_input_tokens_seen": 358089800,
      "step": 5900,
      "train_runtime": 76751.1814,
      "train_tokens_per_second": 4665.593
    },
    {
      "epoch": 0.7531887755102041,
      "grad_norm": 0.5601331958131273,
      "learning_rate": 6.891182474538539e-06,
      "loss": 0.33099820613861086,
      "num_input_tokens_seen": 358399920,
      "step": 5905,
      "train_runtime": 76813.0904,
      "train_tokens_per_second": 4665.87
    },
    {
      "epoch": 0.7538265306122449,
      "grad_norm": 0.7381218893687573,
      "learning_rate": 6.886544726927826e-06,
      "loss": 0.32611565589904784,
      "num_input_tokens_seen": 358705672,
      "step": 5910,
      "train_runtime": 76879.3853,
      "train_tokens_per_second": 4665.824
    },
    {
      "epoch": 0.7544642857142857,
      "grad_norm": 0.5983243405906501,
      "learning_rate": 6.881905086037721e-06,
      "loss": 0.3351559638977051,
      "num_input_tokens_seen": 359014616,
      "step": 5915,
      "train_runtime": 76946.6056,
      "train_tokens_per_second": 4665.763
    },
    {
      "epoch": 0.7551020408163265,
      "grad_norm": 0.6525149194121894,
      "learning_rate": 6.877263556524432e-06,
      "loss": 0.31889750957489016,
      "num_input_tokens_seen": 359320984,
      "step": 5920,
      "train_runtime": 77011.3336,
      "train_tokens_per_second": 4665.82
    },
    {
      "epoch": 0.7557397959183674,
      "grad_norm": 0.5305086071161054,
      "learning_rate": 6.872620143046056e-06,
      "loss": 0.29836535453796387,
      "num_input_tokens_seen": 359619040,
      "step": 5925,
      "train_runtime": 77078.6369,
      "train_tokens_per_second": 4665.612
    },
    {
      "epoch": 0.7563775510204082,
      "grad_norm": 0.664885483277217,
      "learning_rate": 6.867974850262582e-06,
      "loss": 0.32633788585662843,
      "num_input_tokens_seen": 359932688,
      "step": 5930,
      "train_runtime": 77129.9165,
      "train_tokens_per_second": 4666.577
    },
    {
      "epoch": 0.7570153061224489,
      "grad_norm": 0.5575635399142616,
      "learning_rate": 6.863327682835887e-06,
      "loss": 0.3186999559402466,
      "num_input_tokens_seen": 360249184,
      "step": 5935,
      "train_runtime": 77189.587,
      "train_tokens_per_second": 4667.07
    },
    {
      "epoch": 0.7576530612244898,
      "grad_norm": 0.6726497256065533,
      "learning_rate": 6.858678645429728e-06,
      "loss": 0.34361996650695803,
      "num_input_tokens_seen": 360550856,
      "step": 5940,
      "train_runtime": 77263.2271,
      "train_tokens_per_second": 4666.526
    },
    {
      "epoch": 0.7582908163265306,
      "grad_norm": 0.6385074144866467,
      "learning_rate": 6.85402774270974e-06,
      "loss": 0.3265536308288574,
      "num_input_tokens_seen": 360848560,
      "step": 5945,
      "train_runtime": 77328.4218,
      "train_tokens_per_second": 4666.442
    },
    {
      "epoch": 0.7589285714285714,
      "grad_norm": 0.5575980924675723,
      "learning_rate": 6.849374979343426e-06,
      "loss": 0.31394202709198,
      "num_input_tokens_seen": 361158960,
      "step": 5950,
      "train_runtime": 77385.6518,
      "train_tokens_per_second": 4667.002
    },
    {
      "epoch": 0.7595663265306123,
      "grad_norm": 0.5600620882620891,
      "learning_rate": 6.84472036000016e-06,
      "loss": 0.3174293518066406,
      "num_input_tokens_seen": 361465312,
      "step": 5955,
      "train_runtime": 77447.6671,
      "train_tokens_per_second": 4667.22
    },
    {
      "epoch": 0.7602040816326531,
      "grad_norm": 0.6357556763963484,
      "learning_rate": 6.840063889351177e-06,
      "loss": 0.3175032138824463,
      "num_input_tokens_seen": 361767408,
      "step": 5960,
      "train_runtime": 77518.7869,
      "train_tokens_per_second": 4666.835
    },
    {
      "epoch": 0.7608418367346939,
      "grad_norm": 0.6142886014161878,
      "learning_rate": 6.835405572069572e-06,
      "loss": 0.3279739856719971,
      "num_input_tokens_seen": 362078816,
      "step": 5965,
      "train_runtime": 77598.5194,
      "train_tokens_per_second": 4666.053
    },
    {
      "epoch": 0.7614795918367347,
      "grad_norm": 0.5718797437710346,
      "learning_rate": 6.830745412830291e-06,
      "loss": 0.3344055414199829,
      "num_input_tokens_seen": 362387792,
      "step": 5970,
      "train_runtime": 77667.0341,
      "train_tokens_per_second": 4665.915
    },
    {
      "epoch": 0.7621173469387755,
      "grad_norm": 0.6464457092923884,
      "learning_rate": 6.826083416310129e-06,
      "loss": 0.32223589420318605,
      "num_input_tokens_seen": 362684544,
      "step": 5975,
      "train_runtime": 77742.0732,
      "train_tokens_per_second": 4665.229
    },
    {
      "epoch": 0.7627551020408163,
      "grad_norm": 0.6462567337229259,
      "learning_rate": 6.821419587187727e-06,
      "loss": 0.3272323369979858,
      "num_input_tokens_seen": 363005704,
      "step": 5980,
      "train_runtime": 77794.6777,
      "train_tokens_per_second": 4666.202
    },
    {
      "epoch": 0.7633928571428571,
      "grad_norm": 0.629682432967277,
      "learning_rate": 6.816753930143558e-06,
      "loss": 0.33120250701904297,
      "num_input_tokens_seen": 363314248,
      "step": 5985,
      "train_runtime": 77857.4932,
      "train_tokens_per_second": 4666.401
    },
    {
      "epoch": 0.764030612244898,
      "grad_norm": 0.6084422932129973,
      "learning_rate": 6.812086449859941e-06,
      "loss": 0.3483760833740234,
      "num_input_tokens_seen": 363612824,
      "step": 5990,
      "train_runtime": 77917.3347,
      "train_tokens_per_second": 4666.649
    },
    {
      "epoch": 0.7646683673469388,
      "grad_norm": 0.6086043126291493,
      "learning_rate": 6.807417151021015e-06,
      "loss": 0.3102273464202881,
      "num_input_tokens_seen": 363903136,
      "step": 5995,
      "train_runtime": 77986.8113,
      "train_tokens_per_second": 4666.214
    },
    {
      "epoch": 0.7653061224489796,
      "grad_norm": 0.5253091189133416,
      "learning_rate": 6.802746038312749e-06,
      "loss": 0.3180234909057617,
      "num_input_tokens_seen": 364214792,
      "step": 6000,
      "train_runtime": 78058.1001,
      "train_tokens_per_second": 4665.945
    },
    {
      "epoch": 0.7659438775510204,
      "grad_norm": 0.5865611367250618,
      "learning_rate": 6.79807311642293e-06,
      "loss": 0.308447003364563,
      "num_input_tokens_seen": 364516496,
      "step": 6005,
      "train_runtime": 78123.6172,
      "train_tokens_per_second": 4665.894
    },
    {
      "epoch": 0.7665816326530612,
      "grad_norm": 0.742585427058255,
      "learning_rate": 6.793398390041159e-06,
      "loss": 0.3338149070739746,
      "num_input_tokens_seen": 364812400,
      "step": 6010,
      "train_runtime": 78177.1426,
      "train_tokens_per_second": 4666.484
    },
    {
      "epoch": 0.767219387755102,
      "grad_norm": 0.6279573157717177,
      "learning_rate": 6.788721863858856e-06,
      "loss": 0.32828316688537595,
      "num_input_tokens_seen": 365116488,
      "step": 6015,
      "train_runtime": 78245.1961,
      "train_tokens_per_second": 4666.312
    },
    {
      "epoch": 0.7678571428571429,
      "grad_norm": 0.601824542408468,
      "learning_rate": 6.784043542569236e-06,
      "loss": 0.34796159267425536,
      "num_input_tokens_seen": 365417272,
      "step": 6020,
      "train_runtime": 78317.7469,
      "train_tokens_per_second": 4665.83
    },
    {
      "epoch": 0.7684948979591837,
      "grad_norm": 0.6150796694963792,
      "learning_rate": 6.7793634308673265e-06,
      "loss": 0.33051772117614747,
      "num_input_tokens_seen": 365734064,
      "step": 6025,
      "train_runtime": 78377.152,
      "train_tokens_per_second": 4666.335
    },
    {
      "epoch": 0.7691326530612245,
      "grad_norm": 0.6046368366820677,
      "learning_rate": 6.7746815334499426e-06,
      "loss": 0.36424710750579836,
      "num_input_tokens_seen": 366049464,
      "step": 6030,
      "train_runtime": 78448.8901,
      "train_tokens_per_second": 4666.089
    },
    {
      "epoch": 0.7697704081632653,
      "grad_norm": 0.5750524833227629,
      "learning_rate": 6.7699978550156954e-06,
      "loss": 0.32325961589813235,
      "num_input_tokens_seen": 366343576,
      "step": 6035,
      "train_runtime": 78515.9379,
      "train_tokens_per_second": 4665.85
    },
    {
      "epoch": 0.7704081632653061,
      "grad_norm": 0.5626232096953089,
      "learning_rate": 6.765312400264985e-06,
      "loss": 0.3155065059661865,
      "num_input_tokens_seen": 366636248,
      "step": 6040,
      "train_runtime": 78588.1266,
      "train_tokens_per_second": 4665.288
    },
    {
      "epoch": 0.7710459183673469,
      "grad_norm": 0.558817365723986,
      "learning_rate": 6.760625173899992e-06,
      "loss": 0.337229585647583,
      "num_input_tokens_seen": 366941976,
      "step": 6045,
      "train_runtime": 78664.9058,
      "train_tokens_per_second": 4664.621
    },
    {
      "epoch": 0.7716836734693877,
      "grad_norm": 0.6399197605548381,
      "learning_rate": 6.755936180624674e-06,
      "loss": 0.32468380928039553,
      "num_input_tokens_seen": 367248712,
      "step": 6050,
      "train_runtime": 78735.3628,
      "train_tokens_per_second": 4664.343
    },
    {
      "epoch": 0.7723214285714286,
      "grad_norm": 0.5470693399519958,
      "learning_rate": 6.751245425144765e-06,
      "loss": 0.32178239822387694,
      "num_input_tokens_seen": 367551760,
      "step": 6055,
      "train_runtime": 78809.3,
      "train_tokens_per_second": 4663.812
    },
    {
      "epoch": 0.7729591836734694,
      "grad_norm": 0.5914673884446321,
      "learning_rate": 6.746552912167766e-06,
      "loss": 0.32709217071533203,
      "num_input_tokens_seen": 367848936,
      "step": 6060,
      "train_runtime": 78872.5468,
      "train_tokens_per_second": 4663.84
    },
    {
      "epoch": 0.7735969387755102,
      "grad_norm": 0.5501746225450489,
      "learning_rate": 6.741858646402941e-06,
      "loss": 0.3186192035675049,
      "num_input_tokens_seen": 368142096,
      "step": 6065,
      "train_runtime": 78931.2452,
      "train_tokens_per_second": 4664.086
    },
    {
      "epoch": 0.7742346938775511,
      "grad_norm": 0.609968757706805,
      "learning_rate": 6.737162632561311e-06,
      "loss": 0.3359341621398926,
      "num_input_tokens_seen": 368445424,
      "step": 6070,
      "train_runtime": 78992.8181,
      "train_tokens_per_second": 4664.29
    },
    {
      "epoch": 0.7748724489795918,
      "grad_norm": 0.6369744096594077,
      "learning_rate": 6.732464875355657e-06,
      "loss": 0.31608850955963136,
      "num_input_tokens_seen": 368733496,
      "step": 6075,
      "train_runtime": 79057.457,
      "train_tokens_per_second": 4664.12
    },
    {
      "epoch": 0.7755102040816326,
      "grad_norm": 0.6393426913398693,
      "learning_rate": 6.7277653795005085e-06,
      "loss": 0.3017498254776001,
      "num_input_tokens_seen": 369034840,
      "step": 6080,
      "train_runtime": 79130.8288,
      "train_tokens_per_second": 4663.604
    },
    {
      "epoch": 0.7761479591836735,
      "grad_norm": 0.6038968306160026,
      "learning_rate": 6.723064149712131e-06,
      "loss": 0.32140769958496096,
      "num_input_tokens_seen": 369326688,
      "step": 6085,
      "train_runtime": 79203.3952,
      "train_tokens_per_second": 4663.016
    },
    {
      "epoch": 0.7767857142857143,
      "grad_norm": 0.5825328672269104,
      "learning_rate": 6.718361190708542e-06,
      "loss": 0.3158728122711182,
      "num_input_tokens_seen": 369618520,
      "step": 6090,
      "train_runtime": 79271.1472,
      "train_tokens_per_second": 4662.712
    },
    {
      "epoch": 0.7774234693877551,
      "grad_norm": 0.6176295264548202,
      "learning_rate": 6.713656507209489e-06,
      "loss": 0.33606772422790526,
      "num_input_tokens_seen": 369926744,
      "step": 6095,
      "train_runtime": 79348.8362,
      "train_tokens_per_second": 4662.031
    },
    {
      "epoch": 0.7780612244897959,
      "grad_norm": 0.6662682504949609,
      "learning_rate": 6.708950103936448e-06,
      "loss": 0.34429340362548827,
      "num_input_tokens_seen": 370215696,
      "step": 6100,
      "train_runtime": 79408.8203,
      "train_tokens_per_second": 4662.148
    },
    {
      "epoch": 0.7786989795918368,
      "grad_norm": 0.5750509168739135,
      "learning_rate": 6.704241985612625e-06,
      "loss": 0.3338142395019531,
      "num_input_tokens_seen": 370522096,
      "step": 6105,
      "train_runtime": 79476.6215,
      "train_tokens_per_second": 4662.026
    },
    {
      "epoch": 0.7793367346938775,
      "grad_norm": 0.6224476023884783,
      "learning_rate": 6.699532156962945e-06,
      "loss": 0.35149459838867186,
      "num_input_tokens_seen": 370830696,
      "step": 6110,
      "train_runtime": 79542.6319,
      "train_tokens_per_second": 4662.037
    },
    {
      "epoch": 0.7799744897959183,
      "grad_norm": 0.6121664876685937,
      "learning_rate": 6.6948206227140525e-06,
      "loss": 0.310459041595459,
      "num_input_tokens_seen": 371121648,
      "step": 6115,
      "train_runtime": 79610.8168,
      "train_tokens_per_second": 4661.699
    },
    {
      "epoch": 0.7806122448979592,
      "grad_norm": 0.7352469724085692,
      "learning_rate": 6.690107387594298e-06,
      "loss": 0.33449711799621584,
      "num_input_tokens_seen": 371420784,
      "step": 6120,
      "train_runtime": 79676.8981,
      "train_tokens_per_second": 4661.587
    },
    {
      "epoch": 0.78125,
      "grad_norm": 0.5947004628298128,
      "learning_rate": 6.685392456333744e-06,
      "loss": 0.34175596237182615,
      "num_input_tokens_seen": 371734544,
      "step": 6125,
      "train_runtime": 79741.5568,
      "train_tokens_per_second": 4661.742
    },
    {
      "epoch": 0.7818877551020408,
      "grad_norm": 0.8740188356488516,
      "learning_rate": 6.680675833664151e-06,
      "loss": 0.3424098491668701,
      "num_input_tokens_seen": 372041800,
      "step": 6130,
      "train_runtime": 79805.6374,
      "train_tokens_per_second": 4661.849
    },
    {
      "epoch": 0.7825255102040817,
      "grad_norm": 0.6536618358080417,
      "learning_rate": 6.67595752431898e-06,
      "loss": 0.3392969608306885,
      "num_input_tokens_seen": 372345688,
      "step": 6135,
      "train_runtime": 79870.8449,
      "train_tokens_per_second": 4661.847
    },
    {
      "epoch": 0.7831632653061225,
      "grad_norm": 0.5551735279679838,
      "learning_rate": 6.671237533033388e-06,
      "loss": 0.3433082580566406,
      "num_input_tokens_seen": 372652808,
      "step": 6140,
      "train_runtime": 79935.6229,
      "train_tokens_per_second": 4661.912
    },
    {
      "epoch": 0.7838010204081632,
      "grad_norm": 0.6893147550496415,
      "learning_rate": 6.66651586454421e-06,
      "loss": 0.3486523151397705,
      "num_input_tokens_seen": 372959856,
      "step": 6145,
      "train_runtime": 80009.8612,
      "train_tokens_per_second": 4661.424
    },
    {
      "epoch": 0.7844387755102041,
      "grad_norm": 0.6584184347838129,
      "learning_rate": 6.661792523589972e-06,
      "loss": 0.32809312343597413,
      "num_input_tokens_seen": 373252488,
      "step": 6150,
      "train_runtime": 80065.2693,
      "train_tokens_per_second": 4661.853
    },
    {
      "epoch": 0.7850765306122449,
      "grad_norm": 0.6952856515693691,
      "learning_rate": 6.65706751491088e-06,
      "loss": 0.31272296905517577,
      "num_input_tokens_seen": 373532416,
      "step": 6155,
      "train_runtime": 80123.0032,
      "train_tokens_per_second": 4661.987
    },
    {
      "epoch": 0.7857142857142857,
      "grad_norm": 0.598605271188881,
      "learning_rate": 6.652340843248806e-06,
      "loss": 0.32628664970397947,
      "num_input_tokens_seen": 373842496,
      "step": 6160,
      "train_runtime": 80185.1956,
      "train_tokens_per_second": 4662.238
    },
    {
      "epoch": 0.7863520408163265,
      "grad_norm": 0.6160544936173801,
      "learning_rate": 6.647612513347297e-06,
      "loss": 0.3464928150177002,
      "num_input_tokens_seen": 374144912,
      "step": 6165,
      "train_runtime": 80247.2719,
      "train_tokens_per_second": 4662.4
    },
    {
      "epoch": 0.7869897959183674,
      "grad_norm": 0.6015077295812333,
      "learning_rate": 6.642882529951561e-06,
      "loss": 0.3017390727996826,
      "num_input_tokens_seen": 374445840,
      "step": 6170,
      "train_runtime": 80308.8766,
      "train_tokens_per_second": 4662.571
    },
    {
      "epoch": 0.7876275510204082,
      "grad_norm": 0.5709503751639645,
      "learning_rate": 6.638150897808469e-06,
      "loss": 0.32829220294952394,
      "num_input_tokens_seen": 374751360,
      "step": 6175,
      "train_runtime": 80373.3675,
      "train_tokens_per_second": 4662.631
    },
    {
      "epoch": 0.7882653061224489,
      "grad_norm": 0.6255217183603037,
      "learning_rate": 6.6334176216665406e-06,
      "loss": 0.35643610954284666,
      "num_input_tokens_seen": 375070264,
      "step": 6180,
      "train_runtime": 80446.5686,
      "train_tokens_per_second": 4662.353
    },
    {
      "epoch": 0.7889030612244898,
      "grad_norm": 0.6107073525501082,
      "learning_rate": 6.628682706275953e-06,
      "loss": 0.3124289035797119,
      "num_input_tokens_seen": 375387304,
      "step": 6185,
      "train_runtime": 80516.9307,
      "train_tokens_per_second": 4662.216
    },
    {
      "epoch": 0.7895408163265306,
      "grad_norm": 0.524448236617867,
      "learning_rate": 6.623946156388523e-06,
      "loss": 0.31171629428863523,
      "num_input_tokens_seen": 375695664,
      "step": 6190,
      "train_runtime": 80579.8189,
      "train_tokens_per_second": 4662.404
    },
    {
      "epoch": 0.7901785714285714,
      "grad_norm": 0.5649584088837674,
      "learning_rate": 6.619207976757708e-06,
      "loss": 0.3073967218399048,
      "num_input_tokens_seen": 376000720,
      "step": 6195,
      "train_runtime": 80650.7899,
      "train_tokens_per_second": 4662.084
    },
    {
      "epoch": 0.7908163265306123,
      "grad_norm": 0.5932107758307361,
      "learning_rate": 6.614468172138603e-06,
      "loss": 0.33098201751708983,
      "num_input_tokens_seen": 376307968,
      "step": 6200,
      "train_runtime": 80718.4357,
      "train_tokens_per_second": 4661.983
    },
    {
      "epoch": 0.7914540816326531,
      "grad_norm": 0.7436937706248128,
      "learning_rate": 6.609726747287934e-06,
      "loss": 0.36113386154174804,
      "num_input_tokens_seen": 376615000,
      "step": 6205,
      "train_runtime": 80785.8571,
      "train_tokens_per_second": 4661.893
    },
    {
      "epoch": 0.7920918367346939,
      "grad_norm": 0.594063557630725,
      "learning_rate": 6.60498370696405e-06,
      "loss": 0.33744571208953855,
      "num_input_tokens_seen": 376925336,
      "step": 6210,
      "train_runtime": 80852.5824,
      "train_tokens_per_second": 4661.884
    },
    {
      "epoch": 0.7927295918367347,
      "grad_norm": 0.6226478270178896,
      "learning_rate": 6.600239055926923e-06,
      "loss": 0.32979750633239746,
      "num_input_tokens_seen": 377234936,
      "step": 6215,
      "train_runtime": 80919.1383,
      "train_tokens_per_second": 4661.875
    },
    {
      "epoch": 0.7933673469387755,
      "grad_norm": 0.5730911412772792,
      "learning_rate": 6.595492798938146e-06,
      "loss": 0.31255078315734863,
      "num_input_tokens_seen": 377545848,
      "step": 6220,
      "train_runtime": 80998.3094,
      "train_tokens_per_second": 4661.157
    },
    {
      "epoch": 0.7940051020408163,
      "grad_norm": 0.5987700994193833,
      "learning_rate": 6.5907449407609145e-06,
      "loss": 0.31581578254699705,
      "num_input_tokens_seen": 377855984,
      "step": 6225,
      "train_runtime": 81053.4701,
      "train_tokens_per_second": 4661.811
    },
    {
      "epoch": 0.7946428571428571,
      "grad_norm": 0.6665221316079585,
      "learning_rate": 6.585995486160038e-06,
      "loss": 0.34901602268218995,
      "num_input_tokens_seen": 378147568,
      "step": 6230,
      "train_runtime": 81127.3373,
      "train_tokens_per_second": 4661.161
    },
    {
      "epoch": 0.795280612244898,
      "grad_norm": 0.5583686682811075,
      "learning_rate": 6.581244439901926e-06,
      "loss": 0.32438206672668457,
      "num_input_tokens_seen": 378444336,
      "step": 6235,
      "train_runtime": 81201.6598,
      "train_tokens_per_second": 4660.549
    },
    {
      "epoch": 0.7959183673469388,
      "grad_norm": 0.6452049423092018,
      "learning_rate": 6.576491806754583e-06,
      "loss": 0.33559203147888184,
      "num_input_tokens_seen": 378743920,
      "step": 6240,
      "train_runtime": 81276.0583,
      "train_tokens_per_second": 4659.969
    },
    {
      "epoch": 0.7965561224489796,
      "grad_norm": 0.6689246592368139,
      "learning_rate": 6.571737591487611e-06,
      "loss": 0.3278643131256104,
      "num_input_tokens_seen": 379029696,
      "step": 6245,
      "train_runtime": 81347.8642,
      "train_tokens_per_second": 4659.369
    },
    {
      "epoch": 0.7971938775510204,
      "grad_norm": 0.627908531821963,
      "learning_rate": 6.566981798872196e-06,
      "loss": 0.3245952844619751,
      "num_input_tokens_seen": 379328424,
      "step": 6250,
      "train_runtime": 81402.2374,
      "train_tokens_per_second": 4659.926
    },
    {
      "epoch": 0.7978316326530612,
      "grad_norm": 0.5954047788072642,
      "learning_rate": 6.562224433681108e-06,
      "loss": 0.3304758071899414,
      "num_input_tokens_seen": 379631176,
      "step": 6255,
      "train_runtime": 81470.7054,
      "train_tokens_per_second": 4659.726
    },
    {
      "epoch": 0.798469387755102,
      "grad_norm": 0.6123728605895805,
      "learning_rate": 6.557465500688696e-06,
      "loss": 0.32650887966156006,
      "num_input_tokens_seen": 379935264,
      "step": 6260,
      "train_runtime": 81529.9905,
      "train_tokens_per_second": 4660.068
    },
    {
      "epoch": 0.7991071428571429,
      "grad_norm": 0.5951134259675872,
      "learning_rate": 6.55270500467088e-06,
      "loss": 0.332712984085083,
      "num_input_tokens_seen": 380227320,
      "step": 6265,
      "train_runtime": 81590.636,
      "train_tokens_per_second": 4660.183
    },
    {
      "epoch": 0.7997448979591837,
      "grad_norm": 0.6662929709457899,
      "learning_rate": 6.5479429504051526e-06,
      "loss": 0.3316300392150879,
      "num_input_tokens_seen": 380549496,
      "step": 6270,
      "train_runtime": 81653.2872,
      "train_tokens_per_second": 4660.553
    },
    {
      "epoch": 0.8003826530612245,
      "grad_norm": 0.6135397745851576,
      "learning_rate": 6.543179342670565e-06,
      "loss": 0.33006997108459474,
      "num_input_tokens_seen": 380858120,
      "step": 6275,
      "train_runtime": 81712.2338,
      "train_tokens_per_second": 4660.968
    },
    {
      "epoch": 0.8010204081632653,
      "grad_norm": 0.5700410729663047,
      "learning_rate": 6.53841418624773e-06,
      "loss": 0.3384885311126709,
      "num_input_tokens_seen": 381172088,
      "step": 6280,
      "train_runtime": 81769.3396,
      "train_tokens_per_second": 4661.553
    },
    {
      "epoch": 0.8016581632653061,
      "grad_norm": 0.6802979132660769,
      "learning_rate": 6.53364748591882e-06,
      "loss": 0.31926870346069336,
      "num_input_tokens_seen": 381480256,
      "step": 6285,
      "train_runtime": 81844.5595,
      "train_tokens_per_second": 4661.034
    },
    {
      "epoch": 0.8022959183673469,
      "grad_norm": 0.5752546558654944,
      "learning_rate": 6.528879246467546e-06,
      "loss": 0.3026753902435303,
      "num_input_tokens_seen": 381790184,
      "step": 6290,
      "train_runtime": 81924.4838,
      "train_tokens_per_second": 4660.27
    },
    {
      "epoch": 0.8029336734693877,
      "grad_norm": 0.5934807419403207,
      "learning_rate": 6.524109472679172e-06,
      "loss": 0.32082395553588866,
      "num_input_tokens_seen": 382099120,
      "step": 6295,
      "train_runtime": 81985.387,
      "train_tokens_per_second": 4660.576
    },
    {
      "epoch": 0.8035714285714286,
      "grad_norm": 0.5961198320827925,
      "learning_rate": 6.519338169340499e-06,
      "loss": 0.3002665042877197,
      "num_input_tokens_seen": 382391256,
      "step": 6300,
      "train_runtime": 82055.3024,
      "train_tokens_per_second": 4660.165
    },
    {
      "epoch": 0.8042091836734694,
      "grad_norm": 0.663586778912891,
      "learning_rate": 6.514565341239861e-06,
      "loss": 0.35150899887084963,
      "num_input_tokens_seen": 382694608,
      "step": 6305,
      "train_runtime": 82129.8251,
      "train_tokens_per_second": 4659.63
    },
    {
      "epoch": 0.8048469387755102,
      "grad_norm": 0.557004826692435,
      "learning_rate": 6.509790993167129e-06,
      "loss": 0.3224548578262329,
      "num_input_tokens_seen": 382998672,
      "step": 6310,
      "train_runtime": 82197.7839,
      "train_tokens_per_second": 4659.477
    },
    {
      "epoch": 0.8054846938775511,
      "grad_norm": 0.6436670145346937,
      "learning_rate": 6.505015129913689e-06,
      "loss": 0.32113525867462156,
      "num_input_tokens_seen": 383300808,
      "step": 6315,
      "train_runtime": 82263.0179,
      "train_tokens_per_second": 4659.455
    },
    {
      "epoch": 0.8061224489795918,
      "grad_norm": 0.6013563442523447,
      "learning_rate": 6.5002377562724585e-06,
      "loss": 0.31710638999938967,
      "num_input_tokens_seen": 383597304,
      "step": 6320,
      "train_runtime": 82326.0654,
      "train_tokens_per_second": 4659.488
    },
    {
      "epoch": 0.8067602040816326,
      "grad_norm": 0.7658646999119676,
      "learning_rate": 6.495458877037861e-06,
      "loss": 0.3188472747802734,
      "num_input_tokens_seen": 383907680,
      "step": 6325,
      "train_runtime": 82382.2507,
      "train_tokens_per_second": 4660.078
    },
    {
      "epoch": 0.8073979591836735,
      "grad_norm": 0.6187017459527038,
      "learning_rate": 6.490678497005839e-06,
      "loss": 0.3157522678375244,
      "num_input_tokens_seen": 384204280,
      "step": 6330,
      "train_runtime": 82453.6683,
      "train_tokens_per_second": 4659.638
    },
    {
      "epoch": 0.8080357142857143,
      "grad_norm": 0.6037166420742117,
      "learning_rate": 6.485896620973836e-06,
      "loss": 0.33968181610107423,
      "num_input_tokens_seen": 384514248,
      "step": 6335,
      "train_runtime": 82518.7284,
      "train_tokens_per_second": 4659.721
    },
    {
      "epoch": 0.8086734693877551,
      "grad_norm": 0.6461015234705619,
      "learning_rate": 6.481113253740799e-06,
      "loss": 0.31825881004333495,
      "num_input_tokens_seen": 384818136,
      "step": 6340,
      "train_runtime": 82591.5828,
      "train_tokens_per_second": 4659.29
    },
    {
      "epoch": 0.8093112244897959,
      "grad_norm": 0.6236157216796101,
      "learning_rate": 6.4763284001071715e-06,
      "loss": 0.32604708671569826,
      "num_input_tokens_seen": 385109072,
      "step": 6345,
      "train_runtime": 82662.8348,
      "train_tokens_per_second": 4658.793
    },
    {
      "epoch": 0.8099489795918368,
      "grad_norm": 0.617790913547995,
      "learning_rate": 6.4715420648748875e-06,
      "loss": 0.3351900577545166,
      "num_input_tokens_seen": 385425072,
      "step": 6350,
      "train_runtime": 82708.7407,
      "train_tokens_per_second": 4660.028
    },
    {
      "epoch": 0.8105867346938775,
      "grad_norm": 0.6086106861541603,
      "learning_rate": 6.46675425284737e-06,
      "loss": 0.31918950080871583,
      "num_input_tokens_seen": 385746512,
      "step": 6355,
      "train_runtime": 82770.1291,
      "train_tokens_per_second": 4660.456
    },
    {
      "epoch": 0.8112244897959183,
      "grad_norm": 0.5931590625891464,
      "learning_rate": 6.461964968829521e-06,
      "loss": 0.33324904441833497,
      "num_input_tokens_seen": 386044952,
      "step": 6360,
      "train_runtime": 82838.0062,
      "train_tokens_per_second": 4660.24
    },
    {
      "epoch": 0.8118622448979592,
      "grad_norm": 0.6622398128993591,
      "learning_rate": 6.457174217627722e-06,
      "loss": 0.3209887981414795,
      "num_input_tokens_seen": 386342120,
      "step": 6365,
      "train_runtime": 82905.0683,
      "train_tokens_per_second": 4660.054
    },
    {
      "epoch": 0.8125,
      "grad_norm": 0.6323717940288658,
      "learning_rate": 6.452382004049829e-06,
      "loss": 0.33527679443359376,
      "num_input_tokens_seen": 386646536,
      "step": 6370,
      "train_runtime": 82976.4238,
      "train_tokens_per_second": 4659.716
    },
    {
      "epoch": 0.8131377551020408,
      "grad_norm": 0.569865981444307,
      "learning_rate": 6.447588332905159e-06,
      "loss": 0.3298801422119141,
      "num_input_tokens_seen": 386949672,
      "step": 6375,
      "train_runtime": 83034.7558,
      "train_tokens_per_second": 4660.093
    },
    {
      "epoch": 0.8137755102040817,
      "grad_norm": 0.6043021041929536,
      "learning_rate": 6.4427932090044975e-06,
      "loss": 0.31019001007080077,
      "num_input_tokens_seen": 387249864,
      "step": 6380,
      "train_runtime": 83095.7845,
      "train_tokens_per_second": 4660.283
    },
    {
      "epoch": 0.8144132653061225,
      "grad_norm": 0.6370069942377361,
      "learning_rate": 6.437996637160086e-06,
      "loss": 0.30964839458465576,
      "num_input_tokens_seen": 387560656,
      "step": 6385,
      "train_runtime": 83159.147,
      "train_tokens_per_second": 4660.469
    },
    {
      "epoch": 0.8150510204081632,
      "grad_norm": 0.6015292072296982,
      "learning_rate": 6.433198622185621e-06,
      "loss": 0.3111919403076172,
      "num_input_tokens_seen": 387871280,
      "step": 6390,
      "train_runtime": 83225.6121,
      "train_tokens_per_second": 4660.48
    },
    {
      "epoch": 0.8156887755102041,
      "grad_norm": 0.6271936646021271,
      "learning_rate": 6.428399168896244e-06,
      "loss": 0.32283749580383303,
      "num_input_tokens_seen": 388172504,
      "step": 6395,
      "train_runtime": 83297.3809,
      "train_tokens_per_second": 4660.081
    },
    {
      "epoch": 0.8163265306122449,
      "grad_norm": 0.5838022923142084,
      "learning_rate": 6.423598282108541e-06,
      "loss": 0.3457418203353882,
      "num_input_tokens_seen": 388470624,
      "step": 6400,
      "train_runtime": 83363.089,
      "train_tokens_per_second": 4659.984
    },
    {
      "epoch": 0.8169642857142857,
      "grad_norm": 0.6815142257501933,
      "learning_rate": 6.41879596664054e-06,
      "loss": 0.3063065767288208,
      "num_input_tokens_seen": 388756200,
      "step": 6405,
      "train_runtime": 83431.8978,
      "train_tokens_per_second": 4659.563
    },
    {
      "epoch": 0.8176020408163265,
      "grad_norm": 0.6545595194246681,
      "learning_rate": 6.413992227311695e-06,
      "loss": 0.3302279472351074,
      "num_input_tokens_seen": 389061280,
      "step": 6410,
      "train_runtime": 83500.2499,
      "train_tokens_per_second": 4659.403
    },
    {
      "epoch": 0.8182397959183674,
      "grad_norm": 0.7018484712197044,
      "learning_rate": 6.409187068942899e-06,
      "loss": 0.34242465496063235,
      "num_input_tokens_seen": 389359968,
      "step": 6415,
      "train_runtime": 83569.2757,
      "train_tokens_per_second": 4659.128
    },
    {
      "epoch": 0.8188775510204082,
      "grad_norm": 0.6389283896565507,
      "learning_rate": 6.4043804963564616e-06,
      "loss": 0.3474923610687256,
      "num_input_tokens_seen": 389657360,
      "step": 6420,
      "train_runtime": 83635.9142,
      "train_tokens_per_second": 4658.972
    },
    {
      "epoch": 0.8195153061224489,
      "grad_norm": 0.7848232670416885,
      "learning_rate": 6.399572514376114e-06,
      "loss": 0.32995893955230715,
      "num_input_tokens_seen": 389956520,
      "step": 6425,
      "train_runtime": 83704.6178,
      "train_tokens_per_second": 4658.722
    },
    {
      "epoch": 0.8201530612244898,
      "grad_norm": 0.5588125579339422,
      "learning_rate": 6.394763127827001e-06,
      "loss": 0.34971914291381834,
      "num_input_tokens_seen": 390262136,
      "step": 6430,
      "train_runtime": 83769.1158,
      "train_tokens_per_second": 4658.783
    },
    {
      "epoch": 0.8207908163265306,
      "grad_norm": 0.5891738832949114,
      "learning_rate": 6.389952341535681e-06,
      "loss": 0.31062195301055906,
      "num_input_tokens_seen": 390557592,
      "step": 6435,
      "train_runtime": 83829.2771,
      "train_tokens_per_second": 4658.964
    },
    {
      "epoch": 0.8214285714285714,
      "grad_norm": 0.6444968160053899,
      "learning_rate": 6.385140160330113e-06,
      "loss": 0.3330610036849976,
      "num_input_tokens_seen": 390863264,
      "step": 6440,
      "train_runtime": 83900.8256,
      "train_tokens_per_second": 4658.634
    },
    {
      "epoch": 0.8220663265306123,
      "grad_norm": 0.6268657531488134,
      "learning_rate": 6.380326589039653e-06,
      "loss": 0.3594696521759033,
      "num_input_tokens_seen": 391172112,
      "step": 6445,
      "train_runtime": 83963.7395,
      "train_tokens_per_second": 4658.822
    },
    {
      "epoch": 0.8227040816326531,
      "grad_norm": 0.5742508108223557,
      "learning_rate": 6.375511632495062e-06,
      "loss": 0.3546117305755615,
      "num_input_tokens_seen": 391471568,
      "step": 6450,
      "train_runtime": 84028.32,
      "train_tokens_per_second": 4658.805
    },
    {
      "epoch": 0.8233418367346939,
      "grad_norm": 0.5760720307562832,
      "learning_rate": 6.370695295528482e-06,
      "loss": 0.31263294219970705,
      "num_input_tokens_seen": 391776424,
      "step": 6455,
      "train_runtime": 84107.7425,
      "train_tokens_per_second": 4658.03
    },
    {
      "epoch": 0.8239795918367347,
      "grad_norm": 0.6013164979326018,
      "learning_rate": 6.365877582973442e-06,
      "loss": 0.31137876510620116,
      "num_input_tokens_seen": 392070888,
      "step": 6460,
      "train_runtime": 84172.343,
      "train_tokens_per_second": 4657.954
    },
    {
      "epoch": 0.8246173469387755,
      "grad_norm": 0.6567154381584758,
      "learning_rate": 6.361058499664856e-06,
      "loss": 0.3098623514175415,
      "num_input_tokens_seen": 392367136,
      "step": 6465,
      "train_runtime": 84242.7761,
      "train_tokens_per_second": 4657.576
    },
    {
      "epoch": 0.8252551020408163,
      "grad_norm": 0.5720345799826225,
      "learning_rate": 6.3562380504390095e-06,
      "loss": 0.33841404914855955,
      "num_input_tokens_seen": 392670760,
      "step": 6470,
      "train_runtime": 84306.1391,
      "train_tokens_per_second": 4657.677
    },
    {
      "epoch": 0.8258928571428571,
      "grad_norm": 0.5465239630144308,
      "learning_rate": 6.35141624013356e-06,
      "loss": 0.3283223628997803,
      "num_input_tokens_seen": 392969760,
      "step": 6475,
      "train_runtime": 84371.6456,
      "train_tokens_per_second": 4657.605
    },
    {
      "epoch": 0.826530612244898,
      "grad_norm": 0.5962936472643869,
      "learning_rate": 6.34659307358753e-06,
      "loss": 0.3258841037750244,
      "num_input_tokens_seen": 393274232,
      "step": 6480,
      "train_runtime": 84439.421,
      "train_tokens_per_second": 4657.472
    },
    {
      "epoch": 0.8271683673469388,
      "grad_norm": 0.5841123087828858,
      "learning_rate": 6.341768555641306e-06,
      "loss": 0.3115019083023071,
      "num_input_tokens_seen": 393577920,
      "step": 6485,
      "train_runtime": 84514.3281,
      "train_tokens_per_second": 4656.937
    },
    {
      "epoch": 0.8278061224489796,
      "grad_norm": 0.6525095785159231,
      "learning_rate": 6.3369426911366275e-06,
      "loss": 0.33711268901824953,
      "num_input_tokens_seen": 393884184,
      "step": 6490,
      "train_runtime": 84572.3308,
      "train_tokens_per_second": 4657.365
    },
    {
      "epoch": 0.8284438775510204,
      "grad_norm": 2.880153799858662,
      "learning_rate": 6.332115484916586e-06,
      "loss": 0.32443020343780515,
      "num_input_tokens_seen": 394188088,
      "step": 6495,
      "train_runtime": 84650.3639,
      "train_tokens_per_second": 4656.661
    },
    {
      "epoch": 0.8290816326530612,
      "grad_norm": 0.5806111511308769,
      "learning_rate": 6.327286941825621e-06,
      "loss": 0.32749388217926023,
      "num_input_tokens_seen": 394492024,
      "step": 6500,
      "train_runtime": 84718.2783,
      "train_tokens_per_second": 4656.516
    },
    {
      "epoch": 0.829719387755102,
      "grad_norm": 0.6463155758573111,
      "learning_rate": 6.322457066709511e-06,
      "loss": 0.3317603588104248,
      "num_input_tokens_seen": 394788840,
      "step": 6505,
      "train_runtime": 84780.2793,
      "train_tokens_per_second": 4656.612
    },
    {
      "epoch": 0.8303571428571429,
      "grad_norm": 0.5919753293216771,
      "learning_rate": 6.317625864415374e-06,
      "loss": 0.333458685874939,
      "num_input_tokens_seen": 395087728,
      "step": 6510,
      "train_runtime": 84844.7953,
      "train_tokens_per_second": 4656.594
    },
    {
      "epoch": 0.8309948979591837,
      "grad_norm": 0.5666796302862332,
      "learning_rate": 6.312793339791662e-06,
      "loss": 0.3259648323059082,
      "num_input_tokens_seen": 395396384,
      "step": 6515,
      "train_runtime": 84907.0552,
      "train_tokens_per_second": 4656.814
    },
    {
      "epoch": 0.8316326530612245,
      "grad_norm": 0.5467432627293126,
      "learning_rate": 6.307959497688146e-06,
      "loss": 0.3173699140548706,
      "num_input_tokens_seen": 395713712,
      "step": 6520,
      "train_runtime": 84961.0482,
      "train_tokens_per_second": 4657.59
    },
    {
      "epoch": 0.8322704081632653,
      "grad_norm": 0.5785933416138389,
      "learning_rate": 6.303124342955928e-06,
      "loss": 0.30993361473083497,
      "num_input_tokens_seen": 396025264,
      "step": 6525,
      "train_runtime": 85025.9291,
      "train_tokens_per_second": 4657.7
    },
    {
      "epoch": 0.8329081632653061,
      "grad_norm": 0.6734163619536873,
      "learning_rate": 6.29828788044742e-06,
      "loss": 0.35101368427276614,
      "num_input_tokens_seen": 396326840,
      "step": 6530,
      "train_runtime": 85090.5541,
      "train_tokens_per_second": 4657.707
    },
    {
      "epoch": 0.8335459183673469,
      "grad_norm": 0.5907392688328745,
      "learning_rate": 6.293450115016353e-06,
      "loss": 0.3599194765090942,
      "num_input_tokens_seen": 396631624,
      "step": 6535,
      "train_runtime": 85165.0064,
      "train_tokens_per_second": 4657.214
    },
    {
      "epoch": 0.8341836734693877,
      "grad_norm": 0.5737048833343137,
      "learning_rate": 6.288611051517761e-06,
      "loss": 0.30787858963012693,
      "num_input_tokens_seen": 396939384,
      "step": 6540,
      "train_runtime": 85226.8383,
      "train_tokens_per_second": 4657.446
    },
    {
      "epoch": 0.8348214285714286,
      "grad_norm": 0.627008135161107,
      "learning_rate": 6.283770694807983e-06,
      "loss": 0.33837013244628905,
      "num_input_tokens_seen": 397242280,
      "step": 6545,
      "train_runtime": 85285.1229,
      "train_tokens_per_second": 4657.814
    },
    {
      "epoch": 0.8354591836734694,
      "grad_norm": 0.5817710958056629,
      "learning_rate": 6.2789290497446546e-06,
      "loss": 0.32402629852294923,
      "num_input_tokens_seen": 397544432,
      "step": 6550,
      "train_runtime": 85338.2856,
      "train_tokens_per_second": 4658.453
    },
    {
      "epoch": 0.8360969387755102,
      "grad_norm": 0.5508573650397668,
      "learning_rate": 6.274086121186704e-06,
      "loss": 0.3160156011581421,
      "num_input_tokens_seen": 397846616,
      "step": 6555,
      "train_runtime": 85400.0057,
      "train_tokens_per_second": 4658.625
    },
    {
      "epoch": 0.8367346938775511,
      "grad_norm": 0.5698976746345229,
      "learning_rate": 6.269241913994348e-06,
      "loss": 0.3336780071258545,
      "num_input_tokens_seen": 398146648,
      "step": 6560,
      "train_runtime": 85464.9343,
      "train_tokens_per_second": 4658.597
    },
    {
      "epoch": 0.8373724489795918,
      "grad_norm": 0.58908453323151,
      "learning_rate": 6.264396433029089e-06,
      "loss": 0.31669230461120607,
      "num_input_tokens_seen": 398463520,
      "step": 6565,
      "train_runtime": 85526.4788,
      "train_tokens_per_second": 4658.949
    },
    {
      "epoch": 0.8380102040816326,
      "grad_norm": 0.5893894659322871,
      "learning_rate": 6.2595496831537025e-06,
      "loss": 0.33449468612670896,
      "num_input_tokens_seen": 398761768,
      "step": 6570,
      "train_runtime": 85602.5773,
      "train_tokens_per_second": 4658.292
    },
    {
      "epoch": 0.8386479591836735,
      "grad_norm": 0.5854698219730254,
      "learning_rate": 6.254701669232243e-06,
      "loss": 0.3276800632476807,
      "num_input_tokens_seen": 399063288,
      "step": 6575,
      "train_runtime": 85668.9507,
      "train_tokens_per_second": 4658.202
    },
    {
      "epoch": 0.8392857142857143,
      "grad_norm": 0.6087896103292258,
      "learning_rate": 6.24985239613003e-06,
      "loss": 0.3333900451660156,
      "num_input_tokens_seen": 399362432,
      "step": 6580,
      "train_runtime": 85729.04,
      "train_tokens_per_second": 4658.427
    },
    {
      "epoch": 0.8399234693877551,
      "grad_norm": 0.6761284159315526,
      "learning_rate": 6.24500186871365e-06,
      "loss": 0.3394115447998047,
      "num_input_tokens_seen": 399677872,
      "step": 6585,
      "train_runtime": 85790.9397,
      "train_tokens_per_second": 4658.742
    },
    {
      "epoch": 0.8405612244897959,
      "grad_norm": 0.6011641987499475,
      "learning_rate": 6.240150091850941e-06,
      "loss": 0.30324516296386717,
      "num_input_tokens_seen": 399979672,
      "step": 6590,
      "train_runtime": 85859.4699,
      "train_tokens_per_second": 4658.539
    },
    {
      "epoch": 0.8411989795918368,
      "grad_norm": 0.6934634797968782,
      "learning_rate": 6.235297070411004e-06,
      "loss": 0.30373287200927734,
      "num_input_tokens_seen": 400268008,
      "step": 6595,
      "train_runtime": 85913.442,
      "train_tokens_per_second": 4658.968
    },
    {
      "epoch": 0.8418367346938775,
      "grad_norm": 0.6383056643165729,
      "learning_rate": 6.230442809264186e-06,
      "loss": 0.32518959045410156,
      "num_input_tokens_seen": 400566488,
      "step": 6600,
      "train_runtime": 85986.5719,
      "train_tokens_per_second": 4658.477
    },
    {
      "epoch": 0.8424744897959183,
      "grad_norm": 0.6064798600689645,
      "learning_rate": 6.225587313282071e-06,
      "loss": 0.3347949981689453,
      "num_input_tokens_seen": 400862976,
      "step": 6605,
      "train_runtime": 86053.6311,
      "train_tokens_per_second": 4658.292
    },
    {
      "epoch": 0.8431122448979592,
      "grad_norm": 0.5976829840641008,
      "learning_rate": 6.220730587337496e-06,
      "loss": 0.3171133041381836,
      "num_input_tokens_seen": 401154544,
      "step": 6610,
      "train_runtime": 86112.3978,
      "train_tokens_per_second": 4658.499
    },
    {
      "epoch": 0.84375,
      "grad_norm": 0.9827168306828648,
      "learning_rate": 6.21587263630452e-06,
      "loss": 0.318402099609375,
      "num_input_tokens_seen": 401472632,
      "step": 6615,
      "train_runtime": 86171.9306,
      "train_tokens_per_second": 4658.972
    },
    {
      "epoch": 0.8443877551020408,
      "grad_norm": 0.7070664354739608,
      "learning_rate": 6.2110134650584355e-06,
      "loss": 0.3207444429397583,
      "num_input_tokens_seen": 401768440,
      "step": 6620,
      "train_runtime": 86239.9675,
      "train_tokens_per_second": 4658.727
    },
    {
      "epoch": 0.8450255102040817,
      "grad_norm": 0.6534442138463878,
      "learning_rate": 6.2061530784757625e-06,
      "loss": 0.32271835803985593,
      "num_input_tokens_seen": 402067584,
      "step": 6625,
      "train_runtime": 86305.4398,
      "train_tokens_per_second": 4658.659
    },
    {
      "epoch": 0.8456632653061225,
      "grad_norm": 0.6086535906849888,
      "learning_rate": 6.201291481434238e-06,
      "loss": 0.3248939275741577,
      "num_input_tokens_seen": 402378216,
      "step": 6630,
      "train_runtime": 86372.6655,
      "train_tokens_per_second": 4658.629
    },
    {
      "epoch": 0.8463010204081632,
      "grad_norm": 0.612229472029973,
      "learning_rate": 6.196428678812816e-06,
      "loss": 0.3274501323699951,
      "num_input_tokens_seen": 402690616,
      "step": 6635,
      "train_runtime": 86451.2395,
      "train_tokens_per_second": 4658.009
    },
    {
      "epoch": 0.8469387755102041,
      "grad_norm": 0.6880105099037892,
      "learning_rate": 6.191564675491653e-06,
      "loss": 0.3477298736572266,
      "num_input_tokens_seen": 402988264,
      "step": 6640,
      "train_runtime": 86515.0636,
      "train_tokens_per_second": 4658.013
    },
    {
      "epoch": 0.8475765306122449,
      "grad_norm": 0.5832688242674897,
      "learning_rate": 6.1866994763521215e-06,
      "loss": 0.3256032466888428,
      "num_input_tokens_seen": 403287792,
      "step": 6645,
      "train_runtime": 86585.9563,
      "train_tokens_per_second": 4657.658
    },
    {
      "epoch": 0.8482142857142857,
      "grad_norm": 0.7112080174273683,
      "learning_rate": 6.1818330862767864e-06,
      "loss": 0.3299060344696045,
      "num_input_tokens_seen": 403598584,
      "step": 6650,
      "train_runtime": 86662.1619,
      "train_tokens_per_second": 4657.149
    },
    {
      "epoch": 0.8488520408163265,
      "grad_norm": 0.6657637345303508,
      "learning_rate": 6.176965510149409e-06,
      "loss": 0.3210991621017456,
      "num_input_tokens_seen": 403901864,
      "step": 6655,
      "train_runtime": 86722.7716,
      "train_tokens_per_second": 4657.391
    },
    {
      "epoch": 0.8494897959183674,
      "grad_norm": 0.6561772299058288,
      "learning_rate": 6.172096752854943e-06,
      "loss": 0.325441312789917,
      "num_input_tokens_seen": 404221656,
      "step": 6660,
      "train_runtime": 86777.3834,
      "train_tokens_per_second": 4658.145
    },
    {
      "epoch": 0.8501275510204082,
      "grad_norm": 0.6170286687454818,
      "learning_rate": 6.1672268192795285e-06,
      "loss": 0.33939661979675295,
      "num_input_tokens_seen": 404524552,
      "step": 6665,
      "train_runtime": 86846.2906,
      "train_tokens_per_second": 4657.937
    },
    {
      "epoch": 0.8507653061224489,
      "grad_norm": 0.5836196547961641,
      "learning_rate": 6.16235571431048e-06,
      "loss": 0.33211517333984375,
      "num_input_tokens_seen": 404845408,
      "step": 6670,
      "train_runtime": 86901.9065,
      "train_tokens_per_second": 4658.648
    },
    {
      "epoch": 0.8514030612244898,
      "grad_norm": 0.6182432750234188,
      "learning_rate": 6.157483442836294e-06,
      "loss": 0.34074184894561765,
      "num_input_tokens_seen": 405152056,
      "step": 6675,
      "train_runtime": 86966.0434,
      "train_tokens_per_second": 4658.739
    },
    {
      "epoch": 0.8520408163265306,
      "grad_norm": 0.5778241670275283,
      "learning_rate": 6.1526100097466345e-06,
      "loss": 0.3007451295852661,
      "num_input_tokens_seen": 405452704,
      "step": 6680,
      "train_runtime": 87032.5202,
      "train_tokens_per_second": 4658.635
    },
    {
      "epoch": 0.8526785714285714,
      "grad_norm": 0.5814159265196439,
      "learning_rate": 6.147735419932334e-06,
      "loss": 0.3128827571868896,
      "num_input_tokens_seen": 405763768,
      "step": 6685,
      "train_runtime": 87102.6208,
      "train_tokens_per_second": 4658.456
    },
    {
      "epoch": 0.8533163265306123,
      "grad_norm": 0.6343133127276688,
      "learning_rate": 6.142859678285378e-06,
      "loss": 0.329839825630188,
      "num_input_tokens_seen": 406067088,
      "step": 6690,
      "train_runtime": 87170.6739,
      "train_tokens_per_second": 4658.299
    },
    {
      "epoch": 0.8539540816326531,
      "grad_norm": 0.6425824705170711,
      "learning_rate": 6.13798278969892e-06,
      "loss": 0.33551831245422364,
      "num_input_tokens_seen": 406381664,
      "step": 6695,
      "train_runtime": 87221.5598,
      "train_tokens_per_second": 4659.188
    },
    {
      "epoch": 0.8545918367346939,
      "grad_norm": 0.6150998116233627,
      "learning_rate": 6.133104759067257e-06,
      "loss": 0.3395432949066162,
      "num_input_tokens_seen": 406688472,
      "step": 6700,
      "train_runtime": 87287.9773,
      "train_tokens_per_second": 4659.158
    },
    {
      "epoch": 0.8552295918367347,
      "grad_norm": 0.7722656530742228,
      "learning_rate": 6.1282255912858315e-06,
      "loss": 0.31645817756652833,
      "num_input_tokens_seen": 406991872,
      "step": 6705,
      "train_runtime": 87361.6625,
      "train_tokens_per_second": 4658.701
    },
    {
      "epoch": 0.8558673469387755,
      "grad_norm": 0.5749535580614875,
      "learning_rate": 6.1233452912512295e-06,
      "loss": 0.35752263069152834,
      "num_input_tokens_seen": 407305736,
      "step": 6710,
      "train_runtime": 87430.1312,
      "train_tokens_per_second": 4658.643
    },
    {
      "epoch": 0.8565051020408163,
      "grad_norm": 0.6054773718675978,
      "learning_rate": 6.118463863861174e-06,
      "loss": 0.32277803421020507,
      "num_input_tokens_seen": 407606096,
      "step": 6715,
      "train_runtime": 87488.3826,
      "train_tokens_per_second": 4658.974
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 0.5502253748394416,
      "learning_rate": 6.113581314014517e-06,
      "loss": 0.3247419595718384,
      "num_input_tokens_seen": 407909024,
      "step": 6720,
      "train_runtime": 87553.7945,
      "train_tokens_per_second": 4658.953
    },
    {
      "epoch": 0.857780612244898,
      "grad_norm": 0.6238722159992686,
      "learning_rate": 6.10869764661124e-06,
      "loss": 0.3387800931930542,
      "num_input_tokens_seen": 408218712,
      "step": 6725,
      "train_runtime": 87622.8176,
      "train_tokens_per_second": 4658.817
    },
    {
      "epoch": 0.8584183673469388,
      "grad_norm": 0.5910120036735711,
      "learning_rate": 6.103812866552444e-06,
      "loss": 0.31037344932556155,
      "num_input_tokens_seen": 408515880,
      "step": 6730,
      "train_runtime": 87688.7316,
      "train_tokens_per_second": 4658.704
    },
    {
      "epoch": 0.8590561224489796,
      "grad_norm": 0.628719461401233,
      "learning_rate": 6.0989269787403445e-06,
      "loss": 0.32601146697998046,
      "num_input_tokens_seen": 408812200,
      "step": 6735,
      "train_runtime": 87754.4398,
      "train_tokens_per_second": 4658.593
    },
    {
      "epoch": 0.8596938775510204,
      "grad_norm": 0.6335840002068116,
      "learning_rate": 6.094039988078273e-06,
      "loss": 0.33951191902160643,
      "num_input_tokens_seen": 409120600,
      "step": 6740,
      "train_runtime": 87803.9024,
      "train_tokens_per_second": 4659.481
    },
    {
      "epoch": 0.8603316326530612,
      "grad_norm": 0.5953512093094517,
      "learning_rate": 6.089151899470668e-06,
      "loss": 0.32826507091522217,
      "num_input_tokens_seen": 409431152,
      "step": 6745,
      "train_runtime": 87866.9323,
      "train_tokens_per_second": 4659.673
    },
    {
      "epoch": 0.860969387755102,
      "grad_norm": 0.5994422268905771,
      "learning_rate": 6.084262717823064e-06,
      "loss": 0.3277494192123413,
      "num_input_tokens_seen": 409740144,
      "step": 6750,
      "train_runtime": 87933.6284,
      "train_tokens_per_second": 4659.652
    },
    {
      "epoch": 0.8616071428571429,
      "grad_norm": 0.6159714059053882,
      "learning_rate": 6.079372448042098e-06,
      "loss": 0.3410026550292969,
      "num_input_tokens_seen": 410051832,
      "step": 6755,
      "train_runtime": 87995.8201,
      "train_tokens_per_second": 4659.901
    },
    {
      "epoch": 0.8622448979591837,
      "grad_norm": 0.6726111143451388,
      "learning_rate": 6.074481095035498e-06,
      "loss": 0.34213569164276125,
      "num_input_tokens_seen": 410351752,
      "step": 6760,
      "train_runtime": 88043.06,
      "train_tokens_per_second": 4660.807
    },
    {
      "epoch": 0.8628826530612245,
      "grad_norm": 0.5894010754354287,
      "learning_rate": 6.069588663712077e-06,
      "loss": 0.3385669946670532,
      "num_input_tokens_seen": 410653368,
      "step": 6765,
      "train_runtime": 88115.9985,
      "train_tokens_per_second": 4660.372
    },
    {
      "epoch": 0.8635204081632653,
      "grad_norm": 1.1067748071567547,
      "learning_rate": 6.064695158981732e-06,
      "loss": 0.33953258991241453,
      "num_input_tokens_seen": 410953864,
      "step": 6770,
      "train_runtime": 88194.0368,
      "train_tokens_per_second": 4659.656
    },
    {
      "epoch": 0.8641581632653061,
      "grad_norm": 0.6652045901148732,
      "learning_rate": 6.059800585755436e-06,
      "loss": 0.33167431354522703,
      "num_input_tokens_seen": 411263728,
      "step": 6775,
      "train_runtime": 88252.7673,
      "train_tokens_per_second": 4660.066
    },
    {
      "epoch": 0.8647959183673469,
      "grad_norm": 0.6505117316280133,
      "learning_rate": 6.054904948945238e-06,
      "loss": 0.3388916015625,
      "num_input_tokens_seen": 411574240,
      "step": 6780,
      "train_runtime": 88321.8296,
      "train_tokens_per_second": 4659.938
    },
    {
      "epoch": 0.8654336734693877,
      "grad_norm": 0.5780226702180542,
      "learning_rate": 6.050008253464247e-06,
      "loss": 0.3238846778869629,
      "num_input_tokens_seen": 411886128,
      "step": 6785,
      "train_runtime": 88384.0158,
      "train_tokens_per_second": 4660.188
    },
    {
      "epoch": 0.8660714285714286,
      "grad_norm": 0.6348347745882694,
      "learning_rate": 6.045110504226643e-06,
      "loss": 0.31720457077026365,
      "num_input_tokens_seen": 412184528,
      "step": 6790,
      "train_runtime": 88453.1393,
      "train_tokens_per_second": 4659.92
    },
    {
      "epoch": 0.8667091836734694,
      "grad_norm": 0.6159052094836616,
      "learning_rate": 6.0402117061476574e-06,
      "loss": 0.32110598087310793,
      "num_input_tokens_seen": 412488296,
      "step": 6795,
      "train_runtime": 88517.6452,
      "train_tokens_per_second": 4659.956
    },
    {
      "epoch": 0.8673469387755102,
      "grad_norm": 0.583082714162748,
      "learning_rate": 6.035311864143576e-06,
      "loss": 0.3190933704376221,
      "num_input_tokens_seen": 412792448,
      "step": 6800,
      "train_runtime": 88584.3884,
      "train_tokens_per_second": 4659.878
    },
    {
      "epoch": 0.8679846938775511,
      "grad_norm": 0.6329054180616354,
      "learning_rate": 6.030410983131733e-06,
      "loss": 0.3117242336273193,
      "num_input_tokens_seen": 413089200,
      "step": 6805,
      "train_runtime": 88648.4546,
      "train_tokens_per_second": 4659.858
    },
    {
      "epoch": 0.8686224489795918,
      "grad_norm": 1.0193393634445236,
      "learning_rate": 6.025509068030504e-06,
      "loss": 0.3140235424041748,
      "num_input_tokens_seen": 413392544,
      "step": 6810,
      "train_runtime": 88715.7465,
      "train_tokens_per_second": 4659.743
    },
    {
      "epoch": 0.8692602040816326,
      "grad_norm": 0.5683846750466159,
      "learning_rate": 6.020606123759305e-06,
      "loss": 0.33549013137817385,
      "num_input_tokens_seen": 413708784,
      "step": 6815,
      "train_runtime": 88778.9398,
      "train_tokens_per_second": 4659.988
    },
    {
      "epoch": 0.8698979591836735,
      "grad_norm": 0.5996200287821946,
      "learning_rate": 6.0157021552385805e-06,
      "loss": 0.32853450775146487,
      "num_input_tokens_seen": 414010272,
      "step": 6820,
      "train_runtime": 88848.909,
      "train_tokens_per_second": 4659.711
    },
    {
      "epoch": 0.8705357142857143,
      "grad_norm": 0.6096990676793227,
      "learning_rate": 6.010797167389808e-06,
      "loss": 0.34289889335632323,
      "num_input_tokens_seen": 414315712,
      "step": 6825,
      "train_runtime": 88900.2642,
      "train_tokens_per_second": 4660.455
    },
    {
      "epoch": 0.8711734693877551,
      "grad_norm": 0.6158292907241294,
      "learning_rate": 6.005891165135484e-06,
      "loss": 0.324583101272583,
      "num_input_tokens_seen": 414619040,
      "step": 6830,
      "train_runtime": 88974.2695,
      "train_tokens_per_second": 4659.988
    },
    {
      "epoch": 0.8718112244897959,
      "grad_norm": 0.62656675666996,
      "learning_rate": 6.000984153399123e-06,
      "loss": 0.3294857978820801,
      "num_input_tokens_seen": 414921552,
      "step": 6835,
      "train_runtime": 89043.8204,
      "train_tokens_per_second": 4659.746
    },
    {
      "epoch": 0.8724489795918368,
      "grad_norm": 0.5738738266668524,
      "learning_rate": 5.996076137105258e-06,
      "loss": 0.31545100212097166,
      "num_input_tokens_seen": 415222344,
      "step": 6840,
      "train_runtime": 89106.6037,
      "train_tokens_per_second": 4659.838
    },
    {
      "epoch": 0.8730867346938775,
      "grad_norm": 0.61010245641856,
      "learning_rate": 5.991167121179422e-06,
      "loss": 0.3479300498962402,
      "num_input_tokens_seen": 415531568,
      "step": 6845,
      "train_runtime": 89179.902,
      "train_tokens_per_second": 4659.475
    },
    {
      "epoch": 0.8737244897959183,
      "grad_norm": 0.6002850636159796,
      "learning_rate": 5.986257110548157e-06,
      "loss": 0.3386786937713623,
      "num_input_tokens_seen": 415827216,
      "step": 6850,
      "train_runtime": 89247.8163,
      "train_tokens_per_second": 4659.242
    },
    {
      "epoch": 0.8743622448979592,
      "grad_norm": 0.6307622924247351,
      "learning_rate": 5.981346110139001e-06,
      "loss": 0.35399162769317627,
      "num_input_tokens_seen": 416142632,
      "step": 6855,
      "train_runtime": 89324.0385,
      "train_tokens_per_second": 4658.798
    },
    {
      "epoch": 0.875,
      "grad_norm": 0.5723339817358812,
      "learning_rate": 5.976434124880487e-06,
      "loss": 0.3396456718444824,
      "num_input_tokens_seen": 416454960,
      "step": 6860,
      "train_runtime": 89384.0452,
      "train_tokens_per_second": 4659.164
    },
    {
      "epoch": 0.8756377551020408,
      "grad_norm": 0.6082194962582594,
      "learning_rate": 5.971521159702136e-06,
      "loss": 0.35317091941833495,
      "num_input_tokens_seen": 416763704,
      "step": 6865,
      "train_runtime": 89445.7172,
      "train_tokens_per_second": 4659.404
    },
    {
      "epoch": 0.8762755102040817,
      "grad_norm": 0.6157072471215066,
      "learning_rate": 5.966607219534448e-06,
      "loss": 0.3282179355621338,
      "num_input_tokens_seen": 417073280,
      "step": 6870,
      "train_runtime": 89516.3285,
      "train_tokens_per_second": 4659.187
    },
    {
      "epoch": 0.8769132653061225,
      "grad_norm": 0.5589296381663512,
      "learning_rate": 5.9616923093089095e-06,
      "loss": 0.3264785289764404,
      "num_input_tokens_seen": 417380120,
      "step": 6875,
      "train_runtime": 89569.3142,
      "train_tokens_per_second": 4659.856
    },
    {
      "epoch": 0.8775510204081632,
      "grad_norm": 0.6297249186199603,
      "learning_rate": 5.956776433957974e-06,
      "loss": 0.3343611240386963,
      "num_input_tokens_seen": 417687864,
      "step": 6880,
      "train_runtime": 89635.6819,
      "train_tokens_per_second": 4659.839
    },
    {
      "epoch": 0.8781887755102041,
      "grad_norm": 0.6026776729970337,
      "learning_rate": 5.951859598415063e-06,
      "loss": 0.33244986534118653,
      "num_input_tokens_seen": 417982568,
      "step": 6885,
      "train_runtime": 89698.3117,
      "train_tokens_per_second": 4659.871
    },
    {
      "epoch": 0.8788265306122449,
      "grad_norm": 0.5892975746978071,
      "learning_rate": 5.94694180761457e-06,
      "loss": 0.34135990142822265,
      "num_input_tokens_seen": 418288616,
      "step": 6890,
      "train_runtime": 89773.1174,
      "train_tokens_per_second": 4659.397
    },
    {
      "epoch": 0.8794642857142857,
      "grad_norm": 0.5675595345165033,
      "learning_rate": 5.942023066491839e-06,
      "loss": 0.32839064598083495,
      "num_input_tokens_seen": 418606240,
      "step": 6895,
      "train_runtime": 89835.7312,
      "train_tokens_per_second": 4659.685
    },
    {
      "epoch": 0.8801020408163265,
      "grad_norm": 0.6193483409474435,
      "learning_rate": 5.937103379983169e-06,
      "loss": 0.32634963989257815,
      "num_input_tokens_seen": 418910840,
      "step": 6900,
      "train_runtime": 89910.0006,
      "train_tokens_per_second": 4659.224
    },
    {
      "epoch": 0.8807397959183674,
      "grad_norm": 0.638309417243741,
      "learning_rate": 5.93218275302581e-06,
      "loss": 0.3340644836425781,
      "num_input_tokens_seen": 419219648,
      "step": 6905,
      "train_runtime": 89969.7123,
      "train_tokens_per_second": 4659.564
    },
    {
      "epoch": 0.8813775510204082,
      "grad_norm": 0.5340941556168116,
      "learning_rate": 5.927261190557955e-06,
      "loss": 0.31465539932250974,
      "num_input_tokens_seen": 419522896,
      "step": 6910,
      "train_runtime": 90040.3697,
      "train_tokens_per_second": 4659.276
    },
    {
      "epoch": 0.8820153061224489,
      "grad_norm": 0.5585091825607283,
      "learning_rate": 5.922338697518736e-06,
      "loss": 0.31777148246765136,
      "num_input_tokens_seen": 419833632,
      "step": 6915,
      "train_runtime": 90108.003,
      "train_tokens_per_second": 4659.227
    },
    {
      "epoch": 0.8826530612244898,
      "grad_norm": 0.6203310454487292,
      "learning_rate": 5.917415278848217e-06,
      "loss": 0.34000697135925295,
      "num_input_tokens_seen": 420149256,
      "step": 6920,
      "train_runtime": 90185.906,
      "train_tokens_per_second": 4658.702
    },
    {
      "epoch": 0.8832908163265306,
      "grad_norm": 0.6183876680824408,
      "learning_rate": 5.9124909394873955e-06,
      "loss": 0.3393259048461914,
      "num_input_tokens_seen": 420454576,
      "step": 6925,
      "train_runtime": 90243.9503,
      "train_tokens_per_second": 4659.089
    },
    {
      "epoch": 0.8839285714285714,
      "grad_norm": 0.6763192784941738,
      "learning_rate": 5.907565684378187e-06,
      "loss": 0.3432659387588501,
      "num_input_tokens_seen": 420748952,
      "step": 6930,
      "train_runtime": 90317.9918,
      "train_tokens_per_second": 4658.529
    },
    {
      "epoch": 0.8845663265306123,
      "grad_norm": 0.5468502596718098,
      "learning_rate": 5.902639518463429e-06,
      "loss": 0.3281343698501587,
      "num_input_tokens_seen": 421061672,
      "step": 6935,
      "train_runtime": 90378.2197,
      "train_tokens_per_second": 4658.884
    },
    {
      "epoch": 0.8852040816326531,
      "grad_norm": 0.6173701173906413,
      "learning_rate": 5.897712446686875e-06,
      "loss": 0.3060306072235107,
      "num_input_tokens_seen": 421354016,
      "step": 6940,
      "train_runtime": 90443.2914,
      "train_tokens_per_second": 4658.765
    },
    {
      "epoch": 0.8858418367346939,
      "grad_norm": 0.6116926654712084,
      "learning_rate": 5.892784473993184e-06,
      "loss": 0.32022268772125245,
      "num_input_tokens_seen": 421664096,
      "step": 6945,
      "train_runtime": 90505.8102,
      "train_tokens_per_second": 4658.973
    },
    {
      "epoch": 0.8864795918367347,
      "grad_norm": 0.6536503421323548,
      "learning_rate": 5.887855605327921e-06,
      "loss": 0.3388242721557617,
      "num_input_tokens_seen": 421977680,
      "step": 6950,
      "train_runtime": 90564.4915,
      "train_tokens_per_second": 4659.416
    },
    {
      "epoch": 0.8871173469387755,
      "grad_norm": 0.6197043239112039,
      "learning_rate": 5.882925845637551e-06,
      "loss": 0.3288001298904419,
      "num_input_tokens_seen": 422278024,
      "step": 6955,
      "train_runtime": 90635.1128,
      "train_tokens_per_second": 4659.1
    },
    {
      "epoch": 0.8877551020408163,
      "grad_norm": 0.6085897437814526,
      "learning_rate": 5.877995199869433e-06,
      "loss": 0.3016238212585449,
      "num_input_tokens_seen": 422579688,
      "step": 6960,
      "train_runtime": 90709.3597,
      "train_tokens_per_second": 4658.612
    },
    {
      "epoch": 0.8883928571428571,
      "grad_norm": 0.5691093492945718,
      "learning_rate": 5.87306367297181e-06,
      "loss": 0.3010798454284668,
      "num_input_tokens_seen": 422882808,
      "step": 6965,
      "train_runtime": 90766.9754,
      "train_tokens_per_second": 4658.994
    },
    {
      "epoch": 0.889030612244898,
      "grad_norm": 0.633966372540404,
      "learning_rate": 5.868131269893818e-06,
      "loss": 0.3299619197845459,
      "num_input_tokens_seen": 423187696,
      "step": 6970,
      "train_runtime": 90829.1946,
      "train_tokens_per_second": 4659.159
    },
    {
      "epoch": 0.8896683673469388,
      "grad_norm": 0.6862812288031181,
      "learning_rate": 5.863197995585468e-06,
      "loss": 0.3324789047241211,
      "num_input_tokens_seen": 423489032,
      "step": 6975,
      "train_runtime": 90904.8533,
      "train_tokens_per_second": 4658.597
    },
    {
      "epoch": 0.8903061224489796,
      "grad_norm": 0.6129910170597797,
      "learning_rate": 5.858263854997642e-06,
      "loss": 0.3130115270614624,
      "num_input_tokens_seen": 423791448,
      "step": 6980,
      "train_runtime": 90980.1013,
      "train_tokens_per_second": 4658.067
    },
    {
      "epoch": 0.8909438775510204,
      "grad_norm": 0.6889373904018948,
      "learning_rate": 5.853328853082098e-06,
      "loss": 0.33639523983001707,
      "num_input_tokens_seen": 424099240,
      "step": 6985,
      "train_runtime": 91053.8683,
      "train_tokens_per_second": 4657.674
    },
    {
      "epoch": 0.8915816326530612,
      "grad_norm": 0.6487956986698731,
      "learning_rate": 5.848392994791452e-06,
      "loss": 0.31321704387664795,
      "num_input_tokens_seen": 424398816,
      "step": 6990,
      "train_runtime": 91123.7521,
      "train_tokens_per_second": 4657.39
    },
    {
      "epoch": 0.892219387755102,
      "grad_norm": 0.5752884642905449,
      "learning_rate": 5.843456285079185e-06,
      "loss": 0.32476589679718015,
      "num_input_tokens_seen": 424692664,
      "step": 6995,
      "train_runtime": 91186.7653,
      "train_tokens_per_second": 4657.394
    },
    {
      "epoch": 0.8928571428571429,
      "grad_norm": 0.8369275993174212,
      "learning_rate": 5.838518728899627e-06,
      "loss": 0.3245981216430664,
      "num_input_tokens_seen": 424992552,
      "step": 7000,
      "train_runtime": 91246.3735,
      "train_tokens_per_second": 4657.638
    },
    {
      "epoch": 0.8934948979591837,
      "grad_norm": 0.6212390365806565,
      "learning_rate": 5.833580331207964e-06,
      "loss": 0.3225821256637573,
      "num_input_tokens_seen": 425287872,
      "step": 7005,
      "train_runtime": 91316.1681,
      "train_tokens_per_second": 4657.312
    },
    {
      "epoch": 0.8941326530612245,
      "grad_norm": 0.5958943094933631,
      "learning_rate": 5.828641096960223e-06,
      "loss": 0.34339883327484133,
      "num_input_tokens_seen": 425592264,
      "step": 7010,
      "train_runtime": 91387.0629,
      "train_tokens_per_second": 4657.03
    },
    {
      "epoch": 0.8947704081632653,
      "grad_norm": 0.6745530956803485,
      "learning_rate": 5.823701031113267e-06,
      "loss": 0.32695627212524414,
      "num_input_tokens_seen": 425906136,
      "step": 7015,
      "train_runtime": 91446.0903,
      "train_tokens_per_second": 4657.456
    },
    {
      "epoch": 0.8954081632653061,
      "grad_norm": 0.5292787632217651,
      "learning_rate": 5.8187601386248004e-06,
      "loss": 0.32545430660247804,
      "num_input_tokens_seen": 426201056,
      "step": 7020,
      "train_runtime": 91507.8066,
      "train_tokens_per_second": 4657.538
    },
    {
      "epoch": 0.8960459183673469,
      "grad_norm": 0.5968900855812483,
      "learning_rate": 5.8138184244533516e-06,
      "loss": 0.3182164430618286,
      "num_input_tokens_seen": 426497408,
      "step": 7025,
      "train_runtime": 91579.0022,
      "train_tokens_per_second": 4657.153
    },
    {
      "epoch": 0.8966836734693877,
      "grad_norm": 0.57990914735019,
      "learning_rate": 5.808875893558277e-06,
      "loss": 0.31878080368041994,
      "num_input_tokens_seen": 426794296,
      "step": 7030,
      "train_runtime": 91645.3438,
      "train_tokens_per_second": 4657.021
    },
    {
      "epoch": 0.8973214285714286,
      "grad_norm": 0.5994009927046162,
      "learning_rate": 5.8039325508997524e-06,
      "loss": 0.31372628211975095,
      "num_input_tokens_seen": 427098144,
      "step": 7035,
      "train_runtime": 91710.4468,
      "train_tokens_per_second": 4657.028
    },
    {
      "epoch": 0.8979591836734694,
      "grad_norm": 0.58470949819841,
      "learning_rate": 5.798988401438765e-06,
      "loss": 0.3209484577178955,
      "num_input_tokens_seen": 427409712,
      "step": 7040,
      "train_runtime": 91769.2455,
      "train_tokens_per_second": 4657.44
    },
    {
      "epoch": 0.8985969387755102,
      "grad_norm": 0.5459566895258812,
      "learning_rate": 5.7940434501371175e-06,
      "loss": 0.2963754177093506,
      "num_input_tokens_seen": 427722720,
      "step": 7045,
      "train_runtime": 91832.4817,
      "train_tokens_per_second": 4657.641
    },
    {
      "epoch": 0.8992346938775511,
      "grad_norm": 0.5464202688657641,
      "learning_rate": 5.78909770195741e-06,
      "loss": 0.336498761177063,
      "num_input_tokens_seen": 428037344,
      "step": 7050,
      "train_runtime": 91896.4776,
      "train_tokens_per_second": 4657.821
    },
    {
      "epoch": 0.8998724489795918,
      "grad_norm": 0.5918581612652531,
      "learning_rate": 5.7841511618630484e-06,
      "loss": 0.3361975193023682,
      "num_input_tokens_seen": 428338744,
      "step": 7055,
      "train_runtime": 91955.0584,
      "train_tokens_per_second": 4658.131
    },
    {
      "epoch": 0.9005102040816326,
      "grad_norm": 0.5989399830763436,
      "learning_rate": 5.779203834818232e-06,
      "loss": 0.3048173189163208,
      "num_input_tokens_seen": 428636752,
      "step": 7060,
      "train_runtime": 92025.6383,
      "train_tokens_per_second": 4657.797
    },
    {
      "epoch": 0.9011479591836735,
      "grad_norm": 0.6211553615360176,
      "learning_rate": 5.774255725787946e-06,
      "loss": 0.3191343069076538,
      "num_input_tokens_seen": 428944496,
      "step": 7065,
      "train_runtime": 92094.017,
      "train_tokens_per_second": 4657.68
    },
    {
      "epoch": 0.9017857142857143,
      "grad_norm": 0.5780687534165849,
      "learning_rate": 5.769306839737968e-06,
      "loss": 0.3296975135803223,
      "num_input_tokens_seen": 429250792,
      "step": 7070,
      "train_runtime": 92155.0756,
      "train_tokens_per_second": 4657.918
    },
    {
      "epoch": 0.9024234693877551,
      "grad_norm": 0.7428599286309785,
      "learning_rate": 5.764357181634847e-06,
      "loss": 0.3285045623779297,
      "num_input_tokens_seen": 429563984,
      "step": 7075,
      "train_runtime": 92218.342,
      "train_tokens_per_second": 4658.119
    },
    {
      "epoch": 0.9030612244897959,
      "grad_norm": 0.6293129087083874,
      "learning_rate": 5.759406756445911e-06,
      "loss": 0.34009225368499757,
      "num_input_tokens_seen": 429866120,
      "step": 7080,
      "train_runtime": 92289.2336,
      "train_tokens_per_second": 4657.814
    },
    {
      "epoch": 0.9036989795918368,
      "grad_norm": 0.5932264976811832,
      "learning_rate": 5.754455569139258e-06,
      "loss": 0.3277864456176758,
      "num_input_tokens_seen": 430171480,
      "step": 7085,
      "train_runtime": 92349.0988,
      "train_tokens_per_second": 4658.102
    },
    {
      "epoch": 0.9043367346938775,
      "grad_norm": 0.6211237918901299,
      "learning_rate": 5.74950362468375e-06,
      "loss": 0.33977360725402833,
      "num_input_tokens_seen": 430471856,
      "step": 7090,
      "train_runtime": 92423.4907,
      "train_tokens_per_second": 4657.602
    },
    {
      "epoch": 0.9049744897959183,
      "grad_norm": 0.601155659697981,
      "learning_rate": 5.74455092804901e-06,
      "loss": 0.3126554250717163,
      "num_input_tokens_seen": 430765624,
      "step": 7095,
      "train_runtime": 92484.8366,
      "train_tokens_per_second": 4657.689
    },
    {
      "epoch": 0.9056122448979592,
      "grad_norm": 0.5659404830215208,
      "learning_rate": 5.739597484205415e-06,
      "loss": 0.3299864768981934,
      "num_input_tokens_seen": 431077520,
      "step": 7100,
      "train_runtime": 92553.5679,
      "train_tokens_per_second": 4657.6
    },
    {
      "epoch": 0.90625,
      "grad_norm": 0.6158019873620275,
      "learning_rate": 5.734643298124091e-06,
      "loss": 0.3100090980529785,
      "num_input_tokens_seen": 431392784,
      "step": 7105,
      "train_runtime": 92610.8809,
      "train_tokens_per_second": 4658.122
    },
    {
      "epoch": 0.9068877551020408,
      "grad_norm": 0.7116486718196253,
      "learning_rate": 5.7296883747769096e-06,
      "loss": 0.34330143928527834,
      "num_input_tokens_seen": 431698568,
      "step": 7110,
      "train_runtime": 92676.596,
      "train_tokens_per_second": 4658.119
    },
    {
      "epoch": 0.9075255102040817,
      "grad_norm": 0.6347882352327057,
      "learning_rate": 5.724732719136481e-06,
      "loss": 0.33743948936462403,
      "num_input_tokens_seen": 432004008,
      "step": 7115,
      "train_runtime": 92742.449,
      "train_tokens_per_second": 4658.104
    },
    {
      "epoch": 0.9081632653061225,
      "grad_norm": 1.3582835645543267,
      "learning_rate": 5.719776336176156e-06,
      "loss": 0.3645359516143799,
      "num_input_tokens_seen": 432315232,
      "step": 7120,
      "train_runtime": 92804.8603,
      "train_tokens_per_second": 4658.325
    },
    {
      "epoch": 0.9088010204081632,
      "grad_norm": 0.6302302722909888,
      "learning_rate": 5.714819230870008e-06,
      "loss": 0.3343202114105225,
      "num_input_tokens_seen": 432623952,
      "step": 7125,
      "train_runtime": 92879.3354,
      "train_tokens_per_second": 4657.914
    },
    {
      "epoch": 0.9094387755102041,
      "grad_norm": 0.6264369369260477,
      "learning_rate": 5.709861408192838e-06,
      "loss": 0.3022934436798096,
      "num_input_tokens_seen": 432923000,
      "step": 7130,
      "train_runtime": 92953.8411,
      "train_tokens_per_second": 4657.398
    },
    {
      "epoch": 0.9100765306122449,
      "grad_norm": 0.6432488478492226,
      "learning_rate": 5.704902873120169e-06,
      "loss": 0.3636758327484131,
      "num_input_tokens_seen": 433223992,
      "step": 7135,
      "train_runtime": 93023.3667,
      "train_tokens_per_second": 4657.152
    },
    {
      "epoch": 0.9107142857142857,
      "grad_norm": 0.5829344402778385,
      "learning_rate": 5.699943630628237e-06,
      "loss": 0.3100710391998291,
      "num_input_tokens_seen": 433527848,
      "step": 7140,
      "train_runtime": 93081.8656,
      "train_tokens_per_second": 4657.49
    },
    {
      "epoch": 0.9113520408163265,
      "grad_norm": 0.5526469566314375,
      "learning_rate": 5.694983685693987e-06,
      "loss": 0.33217525482177734,
      "num_input_tokens_seen": 433827296,
      "step": 7145,
      "train_runtime": 93142.0513,
      "train_tokens_per_second": 4657.695
    },
    {
      "epoch": 0.9119897959183674,
      "grad_norm": 0.5780058168652707,
      "learning_rate": 5.690023043295071e-06,
      "loss": 0.3236429452896118,
      "num_input_tokens_seen": 434126264,
      "step": 7150,
      "train_runtime": 93216.6326,
      "train_tokens_per_second": 4657.176
    },
    {
      "epoch": 0.9126275510204082,
      "grad_norm": 0.6048496623171362,
      "learning_rate": 5.6850617084098416e-06,
      "loss": 0.34194958209991455,
      "num_input_tokens_seen": 434426760,
      "step": 7155,
      "train_runtime": 93285.3704,
      "train_tokens_per_second": 4656.966
    },
    {
      "epoch": 0.9132653061224489,
      "grad_norm": 0.6488646015967442,
      "learning_rate": 5.6800996860173404e-06,
      "loss": 0.3341176509857178,
      "num_input_tokens_seen": 434731896,
      "step": 7160,
      "train_runtime": 93337.1224,
      "train_tokens_per_second": 4657.653
    },
    {
      "epoch": 0.9139030612244898,
      "grad_norm": 0.611836929281971,
      "learning_rate": 5.675136981097309e-06,
      "loss": 0.316843056678772,
      "num_input_tokens_seen": 435030744,
      "step": 7165,
      "train_runtime": 93395.832,
      "train_tokens_per_second": 4657.925
    },
    {
      "epoch": 0.9145408163265306,
      "grad_norm": 0.5603054379719846,
      "learning_rate": 5.670173598630163e-06,
      "loss": 0.31385073661804197,
      "num_input_tokens_seen": 435328592,
      "step": 7170,
      "train_runtime": 93459.4387,
      "train_tokens_per_second": 4657.941
    },
    {
      "epoch": 0.9151785714285714,
      "grad_norm": 0.5619751681038885,
      "learning_rate": 5.6652095435970074e-06,
      "loss": 0.331546688079834,
      "num_input_tokens_seen": 435645672,
      "step": 7175,
      "train_runtime": 93520.8527,
      "train_tokens_per_second": 4658.273
    },
    {
      "epoch": 0.9158163265306123,
      "grad_norm": 0.5899417449963834,
      "learning_rate": 5.6602448209796165e-06,
      "loss": 0.30263442993164064,
      "num_input_tokens_seen": 435933128,
      "step": 7180,
      "train_runtime": 93589.0984,
      "train_tokens_per_second": 4657.948
    },
    {
      "epoch": 0.9164540816326531,
      "grad_norm": 0.6110417187873899,
      "learning_rate": 5.655279435760436e-06,
      "loss": 0.33107576370239256,
      "num_input_tokens_seen": 436241608,
      "step": 7185,
      "train_runtime": 93662.2867,
      "train_tokens_per_second": 4657.602
    },
    {
      "epoch": 0.9170918367346939,
      "grad_norm": 0.617808942550835,
      "learning_rate": 5.650313392922578e-06,
      "loss": 0.350404691696167,
      "num_input_tokens_seen": 436545304,
      "step": 7190,
      "train_runtime": 93725.158,
      "train_tokens_per_second": 4657.717
    },
    {
      "epoch": 0.9177295918367347,
      "grad_norm": 0.6129782951042175,
      "learning_rate": 5.64534669744981e-06,
      "loss": 0.30472793579101565,
      "num_input_tokens_seen": 436846352,
      "step": 7195,
      "train_runtime": 93792.2876,
      "train_tokens_per_second": 4657.594
    },
    {
      "epoch": 0.9183673469387755,
      "grad_norm": 0.5835829941484578,
      "learning_rate": 5.640379354326563e-06,
      "loss": 0.3401020050048828,
      "num_input_tokens_seen": 437151096,
      "step": 7200,
      "train_runtime": 93852.1533,
      "train_tokens_per_second": 4657.87
    },
    {
      "epoch": 0.9190051020408163,
      "grad_norm": 0.6207580203306774,
      "learning_rate": 5.63541136853791e-06,
      "loss": 0.3340470314025879,
      "num_input_tokens_seen": 437457208,
      "step": 7205,
      "train_runtime": 93920.5666,
      "train_tokens_per_second": 4657.736
    },
    {
      "epoch": 0.9196428571428571,
      "grad_norm": 0.6195876235301718,
      "learning_rate": 5.630442745069568e-06,
      "loss": 0.3098969221115112,
      "num_input_tokens_seen": 437757936,
      "step": 7210,
      "train_runtime": 93989.8753,
      "train_tokens_per_second": 4657.501
    },
    {
      "epoch": 0.920280612244898,
      "grad_norm": 0.5827833159501568,
      "learning_rate": 5.625473488907905e-06,
      "loss": 0.3573422908782959,
      "num_input_tokens_seen": 438057032,
      "step": 7215,
      "train_runtime": 94045.464,
      "train_tokens_per_second": 4657.928
    },
    {
      "epoch": 0.9209183673469388,
      "grad_norm": 0.6279376837393513,
      "learning_rate": 5.620503605039912e-06,
      "loss": 0.33810203075408934,
      "num_input_tokens_seen": 438355976,
      "step": 7220,
      "train_runtime": 94117.5499,
      "train_tokens_per_second": 4657.537
    },
    {
      "epoch": 0.9215561224489796,
      "grad_norm": 0.6528797216296661,
      "learning_rate": 5.615533098453215e-06,
      "loss": 0.3303686141967773,
      "num_input_tokens_seen": 438655912,
      "step": 7225,
      "train_runtime": 94193.2575,
      "train_tokens_per_second": 4656.978
    },
    {
      "epoch": 0.9221938775510204,
      "grad_norm": 0.5335016037971426,
      "learning_rate": 5.610561974136065e-06,
      "loss": 0.30500807762146,
      "num_input_tokens_seen": 438963328,
      "step": 7230,
      "train_runtime": 94250.7251,
      "train_tokens_per_second": 4657.4
    },
    {
      "epoch": 0.9228316326530612,
      "grad_norm": 0.5680252169060411,
      "learning_rate": 5.605590237077332e-06,
      "loss": 0.30874583721160886,
      "num_input_tokens_seen": 439259696,
      "step": 7235,
      "train_runtime": 94326.3534,
      "train_tokens_per_second": 4656.808
    },
    {
      "epoch": 0.923469387755102,
      "grad_norm": 0.6134626347075202,
      "learning_rate": 5.600617892266503e-06,
      "loss": 0.33132448196411135,
      "num_input_tokens_seen": 439564888,
      "step": 7240,
      "train_runtime": 94395.6222,
      "train_tokens_per_second": 4656.624
    },
    {
      "epoch": 0.9241071428571429,
      "grad_norm": 0.7101929570256023,
      "learning_rate": 5.5956449446936715e-06,
      "loss": 0.33731722831726074,
      "num_input_tokens_seen": 439871352,
      "step": 7245,
      "train_runtime": 94463.7725,
      "train_tokens_per_second": 4656.508
    },
    {
      "epoch": 0.9247448979591837,
      "grad_norm": 0.6208386726137252,
      "learning_rate": 5.590671399349539e-06,
      "loss": 0.30220632553100585,
      "num_input_tokens_seen": 440175224,
      "step": 7250,
      "train_runtime": 94529.0352,
      "train_tokens_per_second": 4656.508
    },
    {
      "epoch": 0.9253826530612245,
      "grad_norm": 0.6057749008208727,
      "learning_rate": 5.585697261225404e-06,
      "loss": 0.33912243843078616,
      "num_input_tokens_seen": 440483888,
      "step": 7255,
      "train_runtime": 94602.9872,
      "train_tokens_per_second": 4656.131
    },
    {
      "epoch": 0.9260204081632653,
      "grad_norm": 0.6436432236051591,
      "learning_rate": 5.58072253531316e-06,
      "loss": 0.3079514980316162,
      "num_input_tokens_seen": 440792440,
      "step": 7260,
      "train_runtime": 94666.454,
      "train_tokens_per_second": 4656.269
    },
    {
      "epoch": 0.9266581632653061,
      "grad_norm": 0.611831918626776,
      "learning_rate": 5.575747226605298e-06,
      "loss": 0.3464851140975952,
      "num_input_tokens_seen": 441096904,
      "step": 7265,
      "train_runtime": 94732.0455,
      "train_tokens_per_second": 4656.259
    },
    {
      "epoch": 0.9272959183673469,
      "grad_norm": 0.5782058210719649,
      "learning_rate": 5.570771340094883e-06,
      "loss": 0.3171382427215576,
      "num_input_tokens_seen": 441390952,
      "step": 7270,
      "train_runtime": 94795.1636,
      "train_tokens_per_second": 4656.26
    },
    {
      "epoch": 0.9279336734693877,
      "grad_norm": 0.6445610490656555,
      "learning_rate": 5.565794880775565e-06,
      "loss": 0.3248934268951416,
      "num_input_tokens_seen": 441705680,
      "step": 7275,
      "train_runtime": 94858.8705,
      "train_tokens_per_second": 4656.451
    },
    {
      "epoch": 0.9285714285714286,
      "grad_norm": 0.6618988924445185,
      "learning_rate": 5.560817853641568e-06,
      "loss": 0.34633026123046873,
      "num_input_tokens_seen": 442011840,
      "step": 7280,
      "train_runtime": 94929.2671,
      "train_tokens_per_second": 4656.223
    },
    {
      "epoch": 0.9292091836734694,
      "grad_norm": 0.6099102932835904,
      "learning_rate": 5.5558402636876885e-06,
      "loss": 0.3423534154891968,
      "num_input_tokens_seen": 442315264,
      "step": 7285,
      "train_runtime": 94997.9199,
      "train_tokens_per_second": 4656.052
    },
    {
      "epoch": 0.9298469387755102,
      "grad_norm": 0.5917154518525781,
      "learning_rate": 5.550862115909287e-06,
      "loss": 0.3233125925064087,
      "num_input_tokens_seen": 442602968,
      "step": 7290,
      "train_runtime": 95058.8301,
      "train_tokens_per_second": 4656.095
    },
    {
      "epoch": 0.9304846938775511,
      "grad_norm": 0.6617977932000766,
      "learning_rate": 5.545883415302276e-06,
      "loss": 0.3399376630783081,
      "num_input_tokens_seen": 442912720,
      "step": 7295,
      "train_runtime": 95131.7112,
      "train_tokens_per_second": 4655.784
    },
    {
      "epoch": 0.9311224489795918,
      "grad_norm": 0.6503030440623012,
      "learning_rate": 5.540904166863136e-06,
      "loss": 0.3292389869689941,
      "num_input_tokens_seen": 443214392,
      "step": 7300,
      "train_runtime": 95200.9687,
      "train_tokens_per_second": 4655.566
    },
    {
      "epoch": 0.9317602040816326,
      "grad_norm": 0.6114322068753539,
      "learning_rate": 5.535924375588887e-06,
      "loss": 0.32429957389831543,
      "num_input_tokens_seen": 443518360,
      "step": 7305,
      "train_runtime": 95266.3136,
      "train_tokens_per_second": 4655.563
    },
    {
      "epoch": 0.9323979591836735,
      "grad_norm": 0.5958960141965831,
      "learning_rate": 5.530944046477098e-06,
      "loss": 0.34727861881256106,
      "num_input_tokens_seen": 443824776,
      "step": 7310,
      "train_runtime": 95334.9067,
      "train_tokens_per_second": 4655.428
    },
    {
      "epoch": 0.9330357142857143,
      "grad_norm": 0.6220915816417448,
      "learning_rate": 5.525963184525877e-06,
      "loss": 0.31817402839660647,
      "num_input_tokens_seen": 444132048,
      "step": 7315,
      "train_runtime": 95394.9115,
      "train_tokens_per_second": 4655.721
    },
    {
      "epoch": 0.9336734693877551,
      "grad_norm": 0.640954113641302,
      "learning_rate": 5.5209817947338685e-06,
      "loss": 0.33860623836517334,
      "num_input_tokens_seen": 444440104,
      "step": 7320,
      "train_runtime": 95472.0678,
      "train_tokens_per_second": 4655.185
    },
    {
      "epoch": 0.9343112244897959,
      "grad_norm": 0.6926038866488701,
      "learning_rate": 5.5159998821002425e-06,
      "loss": 0.3216454029083252,
      "num_input_tokens_seen": 444730064,
      "step": 7325,
      "train_runtime": 95544.2943,
      "train_tokens_per_second": 4654.7
    },
    {
      "epoch": 0.9349489795918368,
      "grad_norm": 0.6395627568665867,
      "learning_rate": 5.511017451624698e-06,
      "loss": 0.3319781064987183,
      "num_input_tokens_seen": 445022400,
      "step": 7330,
      "train_runtime": 95605.4663,
      "train_tokens_per_second": 4654.78
    },
    {
      "epoch": 0.9355867346938775,
      "grad_norm": 0.5273091656155408,
      "learning_rate": 5.506034508307453e-06,
      "loss": 0.32101285457611084,
      "num_input_tokens_seen": 445313984,
      "step": 7335,
      "train_runtime": 95666.8777,
      "train_tokens_per_second": 4654.84
    },
    {
      "epoch": 0.9362244897959183,
      "grad_norm": 0.658261886396777,
      "learning_rate": 5.501051057149234e-06,
      "loss": 0.34318544864654543,
      "num_input_tokens_seen": 445611808,
      "step": 7340,
      "train_runtime": 95739.3601,
      "train_tokens_per_second": 4654.426
    },
    {
      "epoch": 0.9368622448979592,
      "grad_norm": 0.7703843129622262,
      "learning_rate": 5.496067103151288e-06,
      "loss": 0.31019752025604247,
      "num_input_tokens_seen": 445903928,
      "step": 7345,
      "train_runtime": 95799.6466,
      "train_tokens_per_second": 4654.547
    },
    {
      "epoch": 0.9375,
      "grad_norm": 0.6187082091632335,
      "learning_rate": 5.491082651315359e-06,
      "loss": 0.32934820652008057,
      "num_input_tokens_seen": 446217904,
      "step": 7350,
      "train_runtime": 95872.4724,
      "train_tokens_per_second": 4654.286
    },
    {
      "epoch": 0.9381377551020408,
      "grad_norm": 0.5609898508690694,
      "learning_rate": 5.4860977066436915e-06,
      "loss": 0.33247902393341067,
      "num_input_tokens_seen": 446539136,
      "step": 7355,
      "train_runtime": 95934.8268,
      "train_tokens_per_second": 4654.609
    },
    {
      "epoch": 0.9387755102040817,
      "grad_norm": 0.5558736498241871,
      "learning_rate": 5.481112274139026e-06,
      "loss": 0.3279747247695923,
      "num_input_tokens_seen": 446828760,
      "step": 7360,
      "train_runtime": 96004.602,
      "train_tokens_per_second": 4654.243
    },
    {
      "epoch": 0.9394132653061225,
      "grad_norm": 0.5841949904362537,
      "learning_rate": 5.476126358804594e-06,
      "loss": 0.3111677885055542,
      "num_input_tokens_seen": 447125808,
      "step": 7365,
      "train_runtime": 96072.9392,
      "train_tokens_per_second": 4654.024
    },
    {
      "epoch": 0.9400510204081632,
      "grad_norm": 0.5572463984794219,
      "learning_rate": 5.471139965644108e-06,
      "loss": 0.3141068458557129,
      "num_input_tokens_seen": 447421784,
      "step": 7370,
      "train_runtime": 96139.9287,
      "train_tokens_per_second": 4653.86
    },
    {
      "epoch": 0.9406887755102041,
      "grad_norm": 0.5856299318596453,
      "learning_rate": 5.46615309966176e-06,
      "loss": 0.32782411575317383,
      "num_input_tokens_seen": 447734880,
      "step": 7375,
      "train_runtime": 96194.1054,
      "train_tokens_per_second": 4654.494
    },
    {
      "epoch": 0.9413265306122449,
      "grad_norm": 0.5876070137236294,
      "learning_rate": 5.461165765862223e-06,
      "loss": 0.36186442375183103,
      "num_input_tokens_seen": 448036040,
      "step": 7380,
      "train_runtime": 96264.4631,
      "train_tokens_per_second": 4654.221
    },
    {
      "epoch": 0.9419642857142857,
      "grad_norm": 0.5452430461863326,
      "learning_rate": 5.456177969250632e-06,
      "loss": 0.3144784688949585,
      "num_input_tokens_seen": 448339064,
      "step": 7385,
      "train_runtime": 96330.5057,
      "train_tokens_per_second": 4654.175
    },
    {
      "epoch": 0.9426020408163265,
      "grad_norm": 0.5500309785987467,
      "learning_rate": 5.451189714832588e-06,
      "loss": 0.33200769424438475,
      "num_input_tokens_seen": 448634848,
      "step": 7390,
      "train_runtime": 96397.1529,
      "train_tokens_per_second": 4654.026
    },
    {
      "epoch": 0.9432397959183674,
      "grad_norm": 0.6129476571360919,
      "learning_rate": 5.446201007614156e-06,
      "loss": 0.352520227432251,
      "num_input_tokens_seen": 448932160,
      "step": 7395,
      "train_runtime": 96461.6202,
      "train_tokens_per_second": 4653.998
    },
    {
      "epoch": 0.9438775510204082,
      "grad_norm": 0.6186157828132597,
      "learning_rate": 5.441211852601849e-06,
      "loss": 0.3336463928222656,
      "num_input_tokens_seen": 449228936,
      "step": 7400,
      "train_runtime": 96525.9913,
      "train_tokens_per_second": 4653.969
    },
    {
      "epoch": 0.9445153061224489,
      "grad_norm": 0.5636015344138771,
      "learning_rate": 5.4362222548026325e-06,
      "loss": 0.33085658550262453,
      "num_input_tokens_seen": 449545320,
      "step": 7405,
      "train_runtime": 96591.8618,
      "train_tokens_per_second": 4654.07
    },
    {
      "epoch": 0.9451530612244898,
      "grad_norm": 0.62816717573118,
      "learning_rate": 5.43123221922392e-06,
      "loss": 0.3475645065307617,
      "num_input_tokens_seen": 449867272,
      "step": 7410,
      "train_runtime": 96653.3194,
      "train_tokens_per_second": 4654.442
    },
    {
      "epoch": 0.9457908163265306,
      "grad_norm": 0.6210180550702343,
      "learning_rate": 5.426241750873557e-06,
      "loss": 0.33647613525390624,
      "num_input_tokens_seen": 450173248,
      "step": 7415,
      "train_runtime": 96729.9668,
      "train_tokens_per_second": 4653.917
    },
    {
      "epoch": 0.9464285714285714,
      "grad_norm": 0.6061292817661059,
      "learning_rate": 5.42125085475983e-06,
      "loss": 0.30979266166687014,
      "num_input_tokens_seen": 450477304,
      "step": 7420,
      "train_runtime": 96795.8443,
      "train_tokens_per_second": 4653.891
    },
    {
      "epoch": 0.9470663265306123,
      "grad_norm": 0.5654865729852379,
      "learning_rate": 5.4162595358914475e-06,
      "loss": 0.31366941928863523,
      "num_input_tokens_seen": 450766696,
      "step": 7425,
      "train_runtime": 96861.3468,
      "train_tokens_per_second": 4653.731
    },
    {
      "epoch": 0.9477040816326531,
      "grad_norm": 0.5746923763393963,
      "learning_rate": 5.41126779927755e-06,
      "loss": 0.2894803524017334,
      "num_input_tokens_seen": 451078288,
      "step": 7430,
      "train_runtime": 96935.906,
      "train_tokens_per_second": 4653.366
    },
    {
      "epoch": 0.9483418367346939,
      "grad_norm": 0.622467448620151,
      "learning_rate": 5.4062756499276945e-06,
      "loss": 0.3030913591384888,
      "num_input_tokens_seen": 451375832,
      "step": 7435,
      "train_runtime": 97000.5206,
      "train_tokens_per_second": 4653.334
    },
    {
      "epoch": 0.9489795918367347,
      "grad_norm": 0.6974679560372611,
      "learning_rate": 5.401283092851848e-06,
      "loss": 0.32454471588134765,
      "num_input_tokens_seen": 451683696,
      "step": 7440,
      "train_runtime": 97066.0221,
      "train_tokens_per_second": 4653.366
    },
    {
      "epoch": 0.9496173469387755,
      "grad_norm": 0.6730161110801747,
      "learning_rate": 5.396290133060394e-06,
      "loss": 0.33219094276428224,
      "num_input_tokens_seen": 451988336,
      "step": 7445,
      "train_runtime": 97124.5383,
      "train_tokens_per_second": 4653.699
    },
    {
      "epoch": 0.9502551020408163,
      "grad_norm": 0.7152501057074385,
      "learning_rate": 5.391296775564114e-06,
      "loss": 0.3516712188720703,
      "num_input_tokens_seen": 452279800,
      "step": 7450,
      "train_runtime": 97193.9254,
      "train_tokens_per_second": 4653.375
    },
    {
      "epoch": 0.9508928571428571,
      "grad_norm": 0.6018966871088204,
      "learning_rate": 5.386303025374192e-06,
      "loss": 0.3267823219299316,
      "num_input_tokens_seen": 452569432,
      "step": 7455,
      "train_runtime": 97249.267,
      "train_tokens_per_second": 4653.705
    },
    {
      "epoch": 0.951530612244898,
      "grad_norm": 0.5884060900415792,
      "learning_rate": 5.381308887502204e-06,
      "loss": 0.32781100273132324,
      "num_input_tokens_seen": 452875960,
      "step": 7460,
      "train_runtime": 97320.3665,
      "train_tokens_per_second": 4653.455
    },
    {
      "epoch": 0.9521683673469388,
      "grad_norm": 0.588400128425506,
      "learning_rate": 5.376314366960118e-06,
      "loss": 0.31268844604492185,
      "num_input_tokens_seen": 453174224,
      "step": 7465,
      "train_runtime": 97382.9657,
      "train_tokens_per_second": 4653.527
    },
    {
      "epoch": 0.9528061224489796,
      "grad_norm": 0.6161681381907058,
      "learning_rate": 5.371319468760283e-06,
      "loss": 0.3634037017822266,
      "num_input_tokens_seen": 453475112,
      "step": 7470,
      "train_runtime": 97452.1415,
      "train_tokens_per_second": 4653.311
    },
    {
      "epoch": 0.9534438775510204,
      "grad_norm": 0.6035718300929784,
      "learning_rate": 5.366324197915426e-06,
      "loss": 0.35582818984985354,
      "num_input_tokens_seen": 453791072,
      "step": 7475,
      "train_runtime": 97522.84,
      "train_tokens_per_second": 4653.177
    },
    {
      "epoch": 0.9540816326530612,
      "grad_norm": 0.5479412766278943,
      "learning_rate": 5.361328559438655e-06,
      "loss": 0.32398262023925783,
      "num_input_tokens_seen": 454089208,
      "step": 7480,
      "train_runtime": 97590.6123,
      "train_tokens_per_second": 4653.001
    },
    {
      "epoch": 0.954719387755102,
      "grad_norm": 0.6203335233591618,
      "learning_rate": 5.3563325583434376e-06,
      "loss": 0.3262059688568115,
      "num_input_tokens_seen": 454392432,
      "step": 7485,
      "train_runtime": 97667.0425,
      "train_tokens_per_second": 4652.464
    },
    {
      "epoch": 0.9553571428571429,
      "grad_norm": 0.6510802818047046,
      "learning_rate": 5.351336199643612e-06,
      "loss": 0.3240252733230591,
      "num_input_tokens_seen": 454693328,
      "step": 7490,
      "train_runtime": 97744.0395,
      "train_tokens_per_second": 4651.878
    },
    {
      "epoch": 0.9559948979591837,
      "grad_norm": 0.6274483889915605,
      "learning_rate": 5.346339488353374e-06,
      "loss": 0.34021596908569335,
      "num_input_tokens_seen": 454996912,
      "step": 7495,
      "train_runtime": 97808.5332,
      "train_tokens_per_second": 4651.914
    },
    {
      "epoch": 0.9566326530612245,
      "grad_norm": 0.7061927649586376,
      "learning_rate": 5.341342429487271e-06,
      "loss": 0.31961989402770996,
      "num_input_tokens_seen": 455298984,
      "step": 7500,
      "train_runtime": 97874.0191,
      "train_tokens_per_second": 4651.888
    },
    {
      "epoch": 0.9572704081632653,
      "grad_norm": 0.5841115410135582,
      "learning_rate": 5.336345028060199e-06,
      "loss": 0.3159759283065796,
      "num_input_tokens_seen": 455607960,
      "step": 7505,
      "train_runtime": 97949.3064,
      "train_tokens_per_second": 4651.467
    },
    {
      "epoch": 0.9579081632653061,
      "grad_norm": 0.6978605355910545,
      "learning_rate": 5.331347289087402e-06,
      "loss": 0.33841490745544434,
      "num_input_tokens_seen": 455902144,
      "step": 7510,
      "train_runtime": 98008.61,
      "train_tokens_per_second": 4651.654
    },
    {
      "epoch": 0.9585459183673469,
      "grad_norm": 0.6327166338301546,
      "learning_rate": 5.326349217584459e-06,
      "loss": 0.30366837978363037,
      "num_input_tokens_seen": 456206152,
      "step": 7515,
      "train_runtime": 98072.4036,
      "train_tokens_per_second": 4651.728
    },
    {
      "epoch": 0.9591836734693877,
      "grad_norm": 0.6578129664300764,
      "learning_rate": 5.321350818567285e-06,
      "loss": 0.31577486991882325,
      "num_input_tokens_seen": 456504376,
      "step": 7520,
      "train_runtime": 98144.1948,
      "train_tokens_per_second": 4651.364
    },
    {
      "epoch": 0.9598214285714286,
      "grad_norm": 0.6901553111021399,
      "learning_rate": 5.3163520970521194e-06,
      "loss": 0.31000518798828125,
      "num_input_tokens_seen": 456809144,
      "step": 7525,
      "train_runtime": 98204.6031,
      "train_tokens_per_second": 4651.606
    },
    {
      "epoch": 0.9604591836734694,
      "grad_norm": 0.6419223040805333,
      "learning_rate": 5.311353058055533e-06,
      "loss": 0.319872522354126,
      "num_input_tokens_seen": 457106664,
      "step": 7530,
      "train_runtime": 98279.4697,
      "train_tokens_per_second": 4651.09
    },
    {
      "epoch": 0.9610969387755102,
      "grad_norm": 0.6003832927960608,
      "learning_rate": 5.306353706594407e-06,
      "loss": 0.3212242126464844,
      "num_input_tokens_seen": 457401712,
      "step": 7535,
      "train_runtime": 98348.7658,
      "train_tokens_per_second": 4650.813
    },
    {
      "epoch": 0.9617346938775511,
      "grad_norm": 0.5707879823933953,
      "learning_rate": 5.3013540476859404e-06,
      "loss": 0.3323201179504395,
      "num_input_tokens_seen": 457706640,
      "step": 7540,
      "train_runtime": 98422.6659,
      "train_tokens_per_second": 4650.419
    },
    {
      "epoch": 0.9623724489795918,
      "grad_norm": 0.5935247624471444,
      "learning_rate": 5.2963540863476404e-06,
      "loss": 0.28949751853942873,
      "num_input_tokens_seen": 458015200,
      "step": 7545,
      "train_runtime": 98492.8037,
      "train_tokens_per_second": 4650.24
    },
    {
      "epoch": 0.9630102040816326,
      "grad_norm": 0.6348233190150885,
      "learning_rate": 5.291353827597317e-06,
      "loss": 0.3134197473526001,
      "num_input_tokens_seen": 458318128,
      "step": 7550,
      "train_runtime": 98559.2024,
      "train_tokens_per_second": 4650.181
    },
    {
      "epoch": 0.9636479591836735,
      "grad_norm": 0.542306457023443,
      "learning_rate": 5.28635327645308e-06,
      "loss": 0.3190800428390503,
      "num_input_tokens_seen": 458630192,
      "step": 7555,
      "train_runtime": 98617.3868,
      "train_tokens_per_second": 4650.602
    },
    {
      "epoch": 0.9642857142857143,
      "grad_norm": 0.6234870704223687,
      "learning_rate": 5.281352437933331e-06,
      "loss": 0.32537388801574707,
      "num_input_tokens_seen": 458941680,
      "step": 7560,
      "train_runtime": 98674.0139,
      "train_tokens_per_second": 4651.09
    },
    {
      "epoch": 0.9649234693877551,
      "grad_norm": 0.6563923185368536,
      "learning_rate": 5.276351317056762e-06,
      "loss": 0.33820252418518065,
      "num_input_tokens_seen": 459239448,
      "step": 7565,
      "train_runtime": 98749.203,
      "train_tokens_per_second": 4650.564
    },
    {
      "epoch": 0.9655612244897959,
      "grad_norm": 0.573388055090737,
      "learning_rate": 5.271349918842344e-06,
      "loss": 0.32301039695739747,
      "num_input_tokens_seen": 459535592,
      "step": 7570,
      "train_runtime": 98815.9993,
      "train_tokens_per_second": 4650.417
    },
    {
      "epoch": 0.9661989795918368,
      "grad_norm": 0.6346907070870014,
      "learning_rate": 5.266348248309332e-06,
      "loss": 0.34938960075378417,
      "num_input_tokens_seen": 459839392,
      "step": 7575,
      "train_runtime": 98880.3976,
      "train_tokens_per_second": 4650.461
    },
    {
      "epoch": 0.9668367346938775,
      "grad_norm": 0.5673390742304041,
      "learning_rate": 5.2613463104772535e-06,
      "loss": 0.32122604846954345,
      "num_input_tokens_seen": 460145248,
      "step": 7580,
      "train_runtime": 98951.9034,
      "train_tokens_per_second": 4650.191
    },
    {
      "epoch": 0.9674744897959183,
      "grad_norm": 0.5863174042631039,
      "learning_rate": 5.256344110365896e-06,
      "loss": 0.3423152446746826,
      "num_input_tokens_seen": 460450360,
      "step": 7585,
      "train_runtime": 99008.4551,
      "train_tokens_per_second": 4650.617
    },
    {
      "epoch": 0.9681122448979592,
      "grad_norm": 0.6017504751228284,
      "learning_rate": 5.251341652995325e-06,
      "loss": 0.32942781448364256,
      "num_input_tokens_seen": 460751656,
      "step": 7590,
      "train_runtime": 99082.0659,
      "train_tokens_per_second": 4650.202
    },
    {
      "epoch": 0.96875,
      "grad_norm": 0.5750574521144,
      "learning_rate": 5.246338943385849e-06,
      "loss": 0.3291962146759033,
      "num_input_tokens_seen": 461062480,
      "step": 7595,
      "train_runtime": 99140.8353,
      "train_tokens_per_second": 4650.581
    },
    {
      "epoch": 0.9693877551020408,
      "grad_norm": 0.6221715700005203,
      "learning_rate": 5.24133598655804e-06,
      "loss": 0.31624937057495117,
      "num_input_tokens_seen": 461362320,
      "step": 7600,
      "train_runtime": 99201.305,
      "train_tokens_per_second": 4650.769
    },
    {
      "epoch": 0.9700255102040817,
      "grad_norm": 0.6957054548286925,
      "learning_rate": 5.2363327875327115e-06,
      "loss": 0.35048661231994627,
      "num_input_tokens_seen": 461662608,
      "step": 7605,
      "train_runtime": 99258.0619,
      "train_tokens_per_second": 4651.135
    },
    {
      "epoch": 0.9706632653061225,
      "grad_norm": 0.6123494402701751,
      "learning_rate": 5.231329351330927e-06,
      "loss": 0.307574200630188,
      "num_input_tokens_seen": 461951816,
      "step": 7610,
      "train_runtime": 99313.3687,
      "train_tokens_per_second": 4651.457
    },
    {
      "epoch": 0.9713010204081632,
      "grad_norm": 0.6034186500704106,
      "learning_rate": 5.226325682973984e-06,
      "loss": 0.33936357498168945,
      "num_input_tokens_seen": 462266560,
      "step": 7615,
      "train_runtime": 99369.7077,
      "train_tokens_per_second": 4651.987
    },
    {
      "epoch": 0.9719387755102041,
      "grad_norm": 0.6239223500534754,
      "learning_rate": 5.221321787483409e-06,
      "loss": 0.30183818340301516,
      "num_input_tokens_seen": 462553424,
      "step": 7620,
      "train_runtime": 99428.1103,
      "train_tokens_per_second": 4652.139
    },
    {
      "epoch": 0.9725765306122449,
      "grad_norm": 0.5498370673286883,
      "learning_rate": 5.216317669880965e-06,
      "loss": 0.3296989440917969,
      "num_input_tokens_seen": 462874408,
      "step": 7625,
      "train_runtime": 99494.731,
      "train_tokens_per_second": 4652.25
    },
    {
      "epoch": 0.9732142857142857,
      "grad_norm": 0.6322795699391487,
      "learning_rate": 5.21131333518863e-06,
      "loss": 0.3277259826660156,
      "num_input_tokens_seen": 463178672,
      "step": 7630,
      "train_runtime": 99550.4143,
      "train_tokens_per_second": 4652.705
    },
    {
      "epoch": 0.9738520408163265,
      "grad_norm": 0.6111160155773336,
      "learning_rate": 5.206308788428604e-06,
      "loss": 0.32791035175323485,
      "num_input_tokens_seen": 463487760,
      "step": 7635,
      "train_runtime": 99611.746,
      "train_tokens_per_second": 4652.943
    },
    {
      "epoch": 0.9744897959183674,
      "grad_norm": 0.6136903235756072,
      "learning_rate": 5.201304034623303e-06,
      "loss": 0.3223903656005859,
      "num_input_tokens_seen": 463788224,
      "step": 7640,
      "train_runtime": 99671.5435,
      "train_tokens_per_second": 4653.166
    },
    {
      "epoch": 0.9751275510204082,
      "grad_norm": 0.6220422479739535,
      "learning_rate": 5.1962990787953436e-06,
      "loss": 0.34218950271606446,
      "num_input_tokens_seen": 464093352,
      "step": 7645,
      "train_runtime": 99736.9533,
      "train_tokens_per_second": 4653.174
    },
    {
      "epoch": 0.9757653061224489,
      "grad_norm": 0.6215687010622566,
      "learning_rate": 5.191293925967548e-06,
      "loss": 0.3228957414627075,
      "num_input_tokens_seen": 464402904,
      "step": 7650,
      "train_runtime": 99806.9438,
      "train_tokens_per_second": 4653.012
    },
    {
      "epoch": 0.9764030612244898,
      "grad_norm": 0.6863971797846118,
      "learning_rate": 5.186288581162939e-06,
      "loss": 0.3238239049911499,
      "num_input_tokens_seen": 464703560,
      "step": 7655,
      "train_runtime": 99884.4102,
      "train_tokens_per_second": 4652.413
    },
    {
      "epoch": 0.9770408163265306,
      "grad_norm": 0.6250189379927238,
      "learning_rate": 5.181283049404728e-06,
      "loss": 0.327274751663208,
      "num_input_tokens_seen": 464999088,
      "step": 7660,
      "train_runtime": 99950.2681,
      "train_tokens_per_second": 4652.305
    },
    {
      "epoch": 0.9776785714285714,
      "grad_norm": 0.5657739843144446,
      "learning_rate": 5.1762773357163175e-06,
      "loss": 0.3050973415374756,
      "num_input_tokens_seen": 465311888,
      "step": 7665,
      "train_runtime": 100010.5888,
      "train_tokens_per_second": 4652.626
    },
    {
      "epoch": 0.9783163265306123,
      "grad_norm": 0.6021250645420273,
      "learning_rate": 5.171271445121286e-06,
      "loss": 0.3465284824371338,
      "num_input_tokens_seen": 465625424,
      "step": 7670,
      "train_runtime": 100063.5896,
      "train_tokens_per_second": 4653.295
    },
    {
      "epoch": 0.9789540816326531,
      "grad_norm": 0.6648857296020965,
      "learning_rate": 5.166265382643402e-06,
      "loss": 0.34261624813079833,
      "num_input_tokens_seen": 465924320,
      "step": 7675,
      "train_runtime": 100118.6981,
      "train_tokens_per_second": 4653.719
    },
    {
      "epoch": 0.9795918367346939,
      "grad_norm": 0.5425209533084114,
      "learning_rate": 5.161259153306592e-06,
      "loss": 0.32228875160217285,
      "num_input_tokens_seen": 466225272,
      "step": 7680,
      "train_runtime": 100185.4733,
      "train_tokens_per_second": 4653.621
    },
    {
      "epoch": 0.9802295918367347,
      "grad_norm": 0.6594876001926052,
      "learning_rate": 5.156252762134959e-06,
      "loss": 0.3042541742324829,
      "num_input_tokens_seen": 466522448,
      "step": 7685,
      "train_runtime": 100245.252,
      "train_tokens_per_second": 4653.811
    },
    {
      "epoch": 0.9808673469387755,
      "grad_norm": 0.6120901688455826,
      "learning_rate": 5.1512462141527675e-06,
      "loss": 0.326982307434082,
      "num_input_tokens_seen": 466829432,
      "step": 7690,
      "train_runtime": 100306.9353,
      "train_tokens_per_second": 4654.01
    },
    {
      "epoch": 0.9815051020408163,
      "grad_norm": 0.72869624956535,
      "learning_rate": 5.146239514384437e-06,
      "loss": 0.3118090867996216,
      "num_input_tokens_seen": 467125040,
      "step": 7695,
      "train_runtime": 100365.7346,
      "train_tokens_per_second": 4654.228
    },
    {
      "epoch": 0.9821428571428571,
      "grad_norm": 0.5858297135389228,
      "learning_rate": 5.14123266785454e-06,
      "loss": 0.31283886432647706,
      "num_input_tokens_seen": 467430568,
      "step": 7700,
      "train_runtime": 100428.1672,
      "train_tokens_per_second": 4654.377
    },
    {
      "epoch": 0.982780612244898,
      "grad_norm": 0.6616219119550749,
      "learning_rate": 5.136225679587797e-06,
      "loss": 0.3235911846160889,
      "num_input_tokens_seen": 467731488,
      "step": 7705,
      "train_runtime": 100488.6765,
      "train_tokens_per_second": 4654.569
    },
    {
      "epoch": 0.9834183673469388,
      "grad_norm": 0.6475714179562695,
      "learning_rate": 5.131218554609072e-06,
      "loss": 0.3317469596862793,
      "num_input_tokens_seen": 468049016,
      "step": 7710,
      "train_runtime": 100540.8673,
      "train_tokens_per_second": 4655.311
    },
    {
      "epoch": 0.9840561224489796,
      "grad_norm": 0.6104648998710915,
      "learning_rate": 5.126211297943362e-06,
      "loss": 0.3359741449356079,
      "num_input_tokens_seen": 468358808,
      "step": 7715,
      "train_runtime": 100613.2756,
      "train_tokens_per_second": 4655.04
    },
    {
      "epoch": 0.9846938775510204,
      "grad_norm": 0.6925028543025519,
      "learning_rate": 5.121203914615799e-06,
      "loss": 0.3237119197845459,
      "num_input_tokens_seen": 468666280,
      "step": 7720,
      "train_runtime": 100675.1044,
      "train_tokens_per_second": 4655.235
    },
    {
      "epoch": 0.9853316326530612,
      "grad_norm": 0.720995537625908,
      "learning_rate": 5.116196409651645e-06,
      "loss": 0.33925719261169435,
      "num_input_tokens_seen": 468960272,
      "step": 7725,
      "train_runtime": 100741.7138,
      "train_tokens_per_second": 4655.075
    },
    {
      "epoch": 0.985969387755102,
      "grad_norm": 0.5152183498459357,
      "learning_rate": 5.111188788076278e-06,
      "loss": 0.3303224086761475,
      "num_input_tokens_seen": 469255544,
      "step": 7730,
      "train_runtime": 100816.4365,
      "train_tokens_per_second": 4654.554
    },
    {
      "epoch": 0.9866071428571429,
      "grad_norm": 0.5392899174732972,
      "learning_rate": 5.106181054915196e-06,
      "loss": 0.29469146728515627,
      "num_input_tokens_seen": 469554832,
      "step": 7735,
      "train_runtime": 100888.5535,
      "train_tokens_per_second": 4654.193
    },
    {
      "epoch": 0.9872448979591837,
      "grad_norm": 0.5888347307774385,
      "learning_rate": 5.10117321519401e-06,
      "loss": 0.3403113126754761,
      "num_input_tokens_seen": 469860760,
      "step": 7740,
      "train_runtime": 100959.0582,
      "train_tokens_per_second": 4653.973
    },
    {
      "epoch": 0.9878826530612245,
      "grad_norm": 0.6171581041053468,
      "learning_rate": 5.0961652739384356e-06,
      "loss": 0.35056252479553224,
      "num_input_tokens_seen": 470173952,
      "step": 7745,
      "train_runtime": 101020.7346,
      "train_tokens_per_second": 4654.232
    },
    {
      "epoch": 0.9885204081632653,
      "grad_norm": 0.9065682412688728,
      "learning_rate": 5.091157236174295e-06,
      "loss": 0.2968605995178223,
      "num_input_tokens_seen": 470483144,
      "step": 7750,
      "train_runtime": 101074.947,
      "train_tokens_per_second": 4654.795
    },
    {
      "epoch": 0.9891581632653061,
      "grad_norm": 0.6732971859874851,
      "learning_rate": 5.086149106927499e-06,
      "loss": 0.31652278900146485,
      "num_input_tokens_seen": 470768664,
      "step": 7755,
      "train_runtime": 101147.1087,
      "train_tokens_per_second": 4654.297
    },
    {
      "epoch": 0.9897959183673469,
      "grad_norm": 0.6221879977048489,
      "learning_rate": 5.081140891224061e-06,
      "loss": 0.3348564624786377,
      "num_input_tokens_seen": 471069616,
      "step": 7760,
      "train_runtime": 101208.9241,
      "train_tokens_per_second": 4654.428
    },
    {
      "epoch": 0.9904336734693877,
      "grad_norm": 0.6458286263649597,
      "learning_rate": 5.076132594090068e-06,
      "loss": 0.32708206176757815,
      "num_input_tokens_seen": 471374728,
      "step": 7765,
      "train_runtime": 101276.3606,
      "train_tokens_per_second": 4654.341
    },
    {
      "epoch": 0.9910714285714286,
      "grad_norm": 0.6250983995368763,
      "learning_rate": 5.071124220551703e-06,
      "loss": 0.3107258081436157,
      "num_input_tokens_seen": 471671400,
      "step": 7770,
      "train_runtime": 101331.795,
      "train_tokens_per_second": 4654.723
    },
    {
      "epoch": 0.9917091836734694,
      "grad_norm": 0.596583920106894,
      "learning_rate": 5.066115775635213e-06,
      "loss": 0.3611313343048096,
      "num_input_tokens_seen": 471987080,
      "step": 7775,
      "train_runtime": 101388.9103,
      "train_tokens_per_second": 4655.214
    },
    {
      "epoch": 0.9923469387755102,
      "grad_norm": 0.5404534855849914,
      "learning_rate": 5.061107264366926e-06,
      "loss": 0.3070375442504883,
      "num_input_tokens_seen": 472305712,
      "step": 7780,
      "train_runtime": 101456.6814,
      "train_tokens_per_second": 4655.245
    },
    {
      "epoch": 0.9929846938775511,
      "grad_norm": 0.619493258101051,
      "learning_rate": 5.05609869177323e-06,
      "loss": 0.32773163318634035,
      "num_input_tokens_seen": 472612736,
      "step": 7785,
      "train_runtime": 101513.4314,
      "train_tokens_per_second": 4655.667
    },
    {
      "epoch": 0.9936224489795918,
      "grad_norm": 0.6073300169522049,
      "learning_rate": 5.05109006288058e-06,
      "loss": 0.3287439107894897,
      "num_input_tokens_seen": 472917496,
      "step": 7790,
      "train_runtime": 101585.616,
      "train_tokens_per_second": 4655.359
    },
    {
      "epoch": 0.9942602040816326,
      "grad_norm": 0.5976543699814326,
      "learning_rate": 5.046081382715484e-06,
      "loss": 0.32365686893463136,
      "num_input_tokens_seen": 473218584,
      "step": 7795,
      "train_runtime": 101661.1465,
      "train_tokens_per_second": 4654.862
    },
    {
      "epoch": 0.9948979591836735,
      "grad_norm": 0.5794707914008376,
      "learning_rate": 5.0410726563045005e-06,
      "loss": 0.3211040496826172,
      "num_input_tokens_seen": 473517592,
      "step": 7800,
      "train_runtime": 101731.9811,
      "train_tokens_per_second": 4654.56
    },
    {
      "epoch": 0.9955357142857143,
      "grad_norm": 0.6662957115139384,
      "learning_rate": 5.036063888674239e-06,
      "loss": 0.32327837944030763,
      "num_input_tokens_seen": 473813360,
      "step": 7805,
      "train_runtime": 101793.3238,
      "train_tokens_per_second": 4654.66
    },
    {
      "epoch": 0.9961734693877551,
      "grad_norm": 0.6987883027560683,
      "learning_rate": 5.031055084851349e-06,
      "loss": 0.30079605579376223,
      "num_input_tokens_seen": 474124592,
      "step": 7810,
      "train_runtime": 101858.0502,
      "train_tokens_per_second": 4654.758
    },
    {
      "epoch": 0.9968112244897959,
      "grad_norm": 4.253525166511764,
      "learning_rate": 5.0260462498625084e-06,
      "loss": 0.3088978290557861,
      "num_input_tokens_seen": 474430976,
      "step": 7815,
      "train_runtime": 101916.7207,
      "train_tokens_per_second": 4655.085
    },
    {
      "epoch": 0.9974489795918368,
      "grad_norm": 0.586371936059209,
      "learning_rate": 5.021037388734439e-06,
      "loss": 0.33654088973999025,
      "num_input_tokens_seen": 474733096,
      "step": 7820,
      "train_runtime": 101978.5564,
      "train_tokens_per_second": 4655.225
    },
    {
      "epoch": 0.9980867346938775,
      "grad_norm": 0.576815230469967,
      "learning_rate": 5.016028506493881e-06,
      "loss": 0.29360170364379884,
      "num_input_tokens_seen": 475031800,
      "step": 7825,
      "train_runtime": 102041.6684,
      "train_tokens_per_second": 4655.273
    },
    {
      "epoch": 0.9987244897959183,
      "grad_norm": 0.5791985939268749,
      "learning_rate": 5.011019608167596e-06,
      "loss": 0.3257938861846924,
      "num_input_tokens_seen": 475328464,
      "step": 7830,
      "train_runtime": 102114.9461,
      "train_tokens_per_second": 4654.837
    },
    {
      "epoch": 0.9993622448979592,
      "grad_norm": 0.5999869417308511,
      "learning_rate": 5.006010698782365e-06,
      "loss": 0.3119340896606445,
      "num_input_tokens_seen": 475626552,
      "step": 7835,
      "train_runtime": 102182.8039,
      "train_tokens_per_second": 4654.663
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.9166941734572396,
      "learning_rate": 5.001001783364978e-06,
      "loss": 0.30375056266784667,
      "num_input_tokens_seen": 475942160,
      "step": 7840,
      "train_runtime": 102242.4953,
      "train_tokens_per_second": 4655.033
    },
    {
      "epoch": 1.000637755102041,
      "grad_norm": 0.6247612870740253,
      "learning_rate": 4.995992866942232e-06,
      "loss": 0.267057466506958,
      "num_input_tokens_seen": 476239480,
      "step": 7845,
      "train_runtime": 102333.1184,
      "train_tokens_per_second": 4653.816
    },
    {
      "epoch": 1.0012755102040816,
      "grad_norm": 0.6909996036907871,
      "learning_rate": 4.990983954540925e-06,
      "loss": 0.2515033960342407,
      "num_input_tokens_seen": 476545208,
      "step": 7850,
      "train_runtime": 102395.6601,
      "train_tokens_per_second": 4653.959
    },
    {
      "epoch": 1.0019132653061225,
      "grad_norm": 0.6642170263412407,
      "learning_rate": 4.985975051187848e-06,
      "loss": 0.28331613540649414,
      "num_input_tokens_seen": 476859168,
      "step": 7855,
      "train_runtime": 102455.6737,
      "train_tokens_per_second": 4654.297
    },
    {
      "epoch": 1.0025510204081634,
      "grad_norm": 0.5862386325889741,
      "learning_rate": 4.980966161909785e-06,
      "loss": 0.24894070625305176,
      "num_input_tokens_seen": 477158016,
      "step": 7860,
      "train_runtime": 102531.3424,
      "train_tokens_per_second": 4653.777
    },
    {
      "epoch": 1.003188775510204,
      "grad_norm": 0.6820283813961971,
      "learning_rate": 4.975957291733511e-06,
      "loss": 0.2599456310272217,
      "num_input_tokens_seen": 477467600,
      "step": 7865,
      "train_runtime": 102600.8077,
      "train_tokens_per_second": 4653.644
    },
    {
      "epoch": 1.003826530612245,
      "grad_norm": 0.6293553093183993,
      "learning_rate": 4.970948445685772e-06,
      "loss": 0.26632611751556395,
      "num_input_tokens_seen": 477774896,
      "step": 7870,
      "train_runtime": 102674.6458,
      "train_tokens_per_second": 4653.29
    },
    {
      "epoch": 1.0044642857142858,
      "grad_norm": 0.6733067324704437,
      "learning_rate": 4.965939628793295e-06,
      "loss": 0.25829689502716063,
      "num_input_tokens_seen": 478085696,
      "step": 7875,
      "train_runtime": 102745.0125,
      "train_tokens_per_second": 4653.128
    },
    {
      "epoch": 1.0051020408163265,
      "grad_norm": 0.7376486939890683,
      "learning_rate": 4.960930846082783e-06,
      "loss": 0.27562432289123534,
      "num_input_tokens_seen": 478395512,
      "step": 7880,
      "train_runtime": 102807.5283,
      "train_tokens_per_second": 4653.312
    },
    {
      "epoch": 1.0057397959183674,
      "grad_norm": 0.6553939980833222,
      "learning_rate": 4.955922102580894e-06,
      "loss": 0.28523437976837157,
      "num_input_tokens_seen": 478696016,
      "step": 7885,
      "train_runtime": 102883.5291,
      "train_tokens_per_second": 4652.795
    },
    {
      "epoch": 1.0063775510204083,
      "grad_norm": 0.6084703407447201,
      "learning_rate": 4.9509134033142525e-06,
      "loss": 0.2536736488342285,
      "num_input_tokens_seen": 479003496,
      "step": 7890,
      "train_runtime": 102950.3634,
      "train_tokens_per_second": 4652.762
    },
    {
      "epoch": 1.007015306122449,
      "grad_norm": 0.7395301219386818,
      "learning_rate": 4.945904753309441e-06,
      "loss": 0.2690150260925293,
      "num_input_tokens_seen": 479311408,
      "step": 7895,
      "train_runtime": 103014.9331,
      "train_tokens_per_second": 4652.834
    },
    {
      "epoch": 1.0076530612244898,
      "grad_norm": 0.5996989201007621,
      "learning_rate": 4.940896157592985e-06,
      "loss": 0.24584484100341797,
      "num_input_tokens_seen": 479633944,
      "step": 7900,
      "train_runtime": 103081.9048,
      "train_tokens_per_second": 4652.94
    },
    {
      "epoch": 1.0082908163265305,
      "grad_norm": 1.0123091077906727,
      "learning_rate": 4.935887621191364e-06,
      "loss": 0.25741584300994874,
      "num_input_tokens_seen": 479943672,
      "step": 7905,
      "train_runtime": 103153.4164,
      "train_tokens_per_second": 4652.717
    },
    {
      "epoch": 1.0089285714285714,
      "grad_norm": 0.638985555424403,
      "learning_rate": 4.930879149130993e-06,
      "loss": 0.25302848815917967,
      "num_input_tokens_seen": 480246768,
      "step": 7910,
      "train_runtime": 103212.3781,
      "train_tokens_per_second": 4652.996
    },
    {
      "epoch": 1.0095663265306123,
      "grad_norm": 0.9761997263307471,
      "learning_rate": 4.925870746438223e-06,
      "loss": 0.2453629970550537,
      "num_input_tokens_seen": 480555560,
      "step": 7915,
      "train_runtime": 103273.0034,
      "train_tokens_per_second": 4653.254
    },
    {
      "epoch": 1.010204081632653,
      "grad_norm": 0.5582028468211174,
      "learning_rate": 4.920862418139335e-06,
      "loss": 0.24580702781677247,
      "num_input_tokens_seen": 480874576,
      "step": 7920,
      "train_runtime": 103341.5607,
      "train_tokens_per_second": 4653.254
    },
    {
      "epoch": 1.0108418367346939,
      "grad_norm": 0.692965875206244,
      "learning_rate": 4.91585416926054e-06,
      "loss": 0.25847864151000977,
      "num_input_tokens_seen": 481165672,
      "step": 7925,
      "train_runtime": 103402.9572,
      "train_tokens_per_second": 4653.307
    },
    {
      "epoch": 1.0114795918367347,
      "grad_norm": 0.6521915504350331,
      "learning_rate": 4.91084600482796e-06,
      "loss": 0.28442893028259275,
      "num_input_tokens_seen": 481472544,
      "step": 7930,
      "train_runtime": 103471.1104,
      "train_tokens_per_second": 4653.207
    },
    {
      "epoch": 1.0121173469387754,
      "grad_norm": 0.6593458337833973,
      "learning_rate": 4.905837929867643e-06,
      "loss": 0.2587409496307373,
      "num_input_tokens_seen": 481783768,
      "step": 7935,
      "train_runtime": 103536.0132,
      "train_tokens_per_second": 4653.297
    },
    {
      "epoch": 1.0127551020408163,
      "grad_norm": 0.6336318443270292,
      "learning_rate": 4.900829949405539e-06,
      "loss": 0.2645289421081543,
      "num_input_tokens_seen": 482093968,
      "step": 7940,
      "train_runtime": 103611.1224,
      "train_tokens_per_second": 4652.917
    },
    {
      "epoch": 1.0133928571428572,
      "grad_norm": 0.6657497957556802,
      "learning_rate": 4.895822068467505e-06,
      "loss": 0.2553886651992798,
      "num_input_tokens_seen": 482395560,
      "step": 7945,
      "train_runtime": 103673.5821,
      "train_tokens_per_second": 4653.023
    },
    {
      "epoch": 1.0140306122448979,
      "grad_norm": 0.6812110362287225,
      "learning_rate": 4.890814292079303e-06,
      "loss": 0.2639014482498169,
      "num_input_tokens_seen": 482696040,
      "step": 7950,
      "train_runtime": 103738.4539,
      "train_tokens_per_second": 4653.01
    },
    {
      "epoch": 1.0146683673469388,
      "grad_norm": 0.6321603411543906,
      "learning_rate": 4.885806625266585e-06,
      "loss": 0.2515778303146362,
      "num_input_tokens_seen": 482997952,
      "step": 7955,
      "train_runtime": 103814.8986,
      "train_tokens_per_second": 4652.492
    },
    {
      "epoch": 1.0153061224489797,
      "grad_norm": 0.5741318142930434,
      "learning_rate": 4.8807990730548905e-06,
      "loss": 0.2634756088256836,
      "num_input_tokens_seen": 483299448,
      "step": 7960,
      "train_runtime": 103880.1013,
      "train_tokens_per_second": 4652.474
    },
    {
      "epoch": 1.0159438775510203,
      "grad_norm": 0.635321293778533,
      "learning_rate": 4.8757916404696545e-06,
      "loss": 0.2525560617446899,
      "num_input_tokens_seen": 483598280,
      "step": 7965,
      "train_runtime": 103950.5884,
      "train_tokens_per_second": 4652.194
    },
    {
      "epoch": 1.0165816326530612,
      "grad_norm": 0.6340507571071121,
      "learning_rate": 4.870784332536181e-06,
      "loss": 0.26147069931030276,
      "num_input_tokens_seen": 483902832,
      "step": 7970,
      "train_runtime": 104010.0218,
      "train_tokens_per_second": 4652.464
    },
    {
      "epoch": 1.0172193877551021,
      "grad_norm": 0.637133557237528,
      "learning_rate": 4.865777154279653e-06,
      "loss": 0.2916694164276123,
      "num_input_tokens_seen": 484213240,
      "step": 7975,
      "train_runtime": 104071.9412,
      "train_tokens_per_second": 4652.678
    },
    {
      "epoch": 1.0178571428571428,
      "grad_norm": 0.689867539623888,
      "learning_rate": 4.8607701107251274e-06,
      "loss": 0.2569180965423584,
      "num_input_tokens_seen": 484520120,
      "step": 7980,
      "train_runtime": 104147.976,
      "train_tokens_per_second": 4652.228
    },
    {
      "epoch": 1.0184948979591837,
      "grad_norm": 0.6790640266183066,
      "learning_rate": 4.855763206897516e-06,
      "loss": 0.26433415412902833,
      "num_input_tokens_seen": 484814000,
      "step": 7985,
      "train_runtime": 104219.3971,
      "train_tokens_per_second": 4651.86
    },
    {
      "epoch": 1.0191326530612246,
      "grad_norm": 0.6665199792341517,
      "learning_rate": 4.850756447821602e-06,
      "loss": 0.25423283576965333,
      "num_input_tokens_seen": 485110200,
      "step": 7990,
      "train_runtime": 104276.1594,
      "train_tokens_per_second": 4652.168
    },
    {
      "epoch": 1.0197704081632653,
      "grad_norm": 0.6364233064926358,
      "learning_rate": 4.845749838522016e-06,
      "loss": 0.2718416929244995,
      "num_input_tokens_seen": 485420576,
      "step": 7995,
      "train_runtime": 104343.8851,
      "train_tokens_per_second": 4652.123
    },
    {
      "epoch": 1.0204081632653061,
      "grad_norm": 0.6202741303914555,
      "learning_rate": 4.84074338402324e-06,
      "loss": 0.26037960052490233,
      "num_input_tokens_seen": 485740608,
      "step": 8000,
      "train_runtime": 104401.1017,
      "train_tokens_per_second": 4652.639
    },
    {
      "epoch": 1.021045918367347,
      "grad_norm": 0.7220910599114735,
      "learning_rate": 4.835737089349599e-06,
      "loss": 0.2492842674255371,
      "num_input_tokens_seen": 486046816,
      "step": 8005,
      "train_runtime": 104462.7113,
      "train_tokens_per_second": 4652.826
    },
    {
      "epoch": 1.0216836734693877,
      "grad_norm": 0.5840777439565414,
      "learning_rate": 4.8307309595252624e-06,
      "loss": 0.261875057220459,
      "num_input_tokens_seen": 486355088,
      "step": 8010,
      "train_runtime": 104527.5586,
      "train_tokens_per_second": 4652.889
    },
    {
      "epoch": 1.0223214285714286,
      "grad_norm": 0.666076705817103,
      "learning_rate": 4.82572499957423e-06,
      "loss": 0.24945347309112548,
      "num_input_tokens_seen": 486668640,
      "step": 8015,
      "train_runtime": 104588.7746,
      "train_tokens_per_second": 4653.163
    },
    {
      "epoch": 1.0229591836734695,
      "grad_norm": 0.856542958405885,
      "learning_rate": 4.820719214520329e-06,
      "loss": 0.2600055694580078,
      "num_input_tokens_seen": 486972232,
      "step": 8020,
      "train_runtime": 104643.5721,
      "train_tokens_per_second": 4653.628
    },
    {
      "epoch": 1.0235969387755102,
      "grad_norm": 0.6381585023747365,
      "learning_rate": 4.815713609387221e-06,
      "loss": 0.2600249290466309,
      "num_input_tokens_seen": 487258464,
      "step": 8025,
      "train_runtime": 104715.3174,
      "train_tokens_per_second": 4653.173
    },
    {
      "epoch": 1.024234693877551,
      "grad_norm": 0.6715307356212113,
      "learning_rate": 4.810708189198374e-06,
      "loss": 0.25761544704437256,
      "num_input_tokens_seen": 487564392,
      "step": 8030,
      "train_runtime": 104780.5311,
      "train_tokens_per_second": 4653.196
    },
    {
      "epoch": 1.0248724489795917,
      "grad_norm": 0.643898777376009,
      "learning_rate": 4.80570295897708e-06,
      "loss": 0.26144454479217527,
      "num_input_tokens_seen": 487865256,
      "step": 8035,
      "train_runtime": 104848.4722,
      "train_tokens_per_second": 4653.051
    },
    {
      "epoch": 1.0255102040816326,
      "grad_norm": 0.7028248442260028,
      "learning_rate": 4.800697923746437e-06,
      "loss": 0.2451169013977051,
      "num_input_tokens_seen": 488171848,
      "step": 8040,
      "train_runtime": 104917.1839,
      "train_tokens_per_second": 4652.926
    },
    {
      "epoch": 1.0261479591836735,
      "grad_norm": 0.6465862682800327,
      "learning_rate": 4.795693088529345e-06,
      "loss": 0.24808902740478517,
      "num_input_tokens_seen": 488475712,
      "step": 8045,
      "train_runtime": 104982.6553,
      "train_tokens_per_second": 4652.918
    },
    {
      "epoch": 1.0267857142857142,
      "grad_norm": 0.5944052821225811,
      "learning_rate": 4.790688458348509e-06,
      "loss": 0.23900938034057617,
      "num_input_tokens_seen": 488771280,
      "step": 8050,
      "train_runtime": 105051.8384,
      "train_tokens_per_second": 4652.668
    },
    {
      "epoch": 1.027423469387755,
      "grad_norm": 0.6708975172355104,
      "learning_rate": 4.7856840382264234e-06,
      "loss": 0.2610027313232422,
      "num_input_tokens_seen": 489065984,
      "step": 8055,
      "train_runtime": 105120.8691,
      "train_tokens_per_second": 4652.416
    },
    {
      "epoch": 1.028061224489796,
      "grad_norm": 0.656067227588874,
      "learning_rate": 4.780679833185372e-06,
      "loss": 0.2779052257537842,
      "num_input_tokens_seen": 489372736,
      "step": 8060,
      "train_runtime": 105177.9651,
      "train_tokens_per_second": 4652.807
    },
    {
      "epoch": 1.0286989795918366,
      "grad_norm": 0.6408509861089134,
      "learning_rate": 4.775675848247427e-06,
      "loss": 0.246592378616333,
      "num_input_tokens_seen": 489656160,
      "step": 8065,
      "train_runtime": 105249.934,
      "train_tokens_per_second": 4652.318
    },
    {
      "epoch": 1.0293367346938775,
      "grad_norm": 0.6385842298683069,
      "learning_rate": 4.770672088434433e-06,
      "loss": 0.2661409854888916,
      "num_input_tokens_seen": 489956800,
      "step": 8070,
      "train_runtime": 105319.2908,
      "train_tokens_per_second": 4652.109
    },
    {
      "epoch": 1.0299744897959184,
      "grad_norm": 0.6751569338827333,
      "learning_rate": 4.765668558768015e-06,
      "loss": 0.25893139839172363,
      "num_input_tokens_seen": 490255024,
      "step": 8075,
      "train_runtime": 105378.8446,
      "train_tokens_per_second": 4652.31
    },
    {
      "epoch": 1.030612244897959,
      "grad_norm": 0.7029112914351848,
      "learning_rate": 4.760665264269566e-06,
      "loss": 0.2638129234313965,
      "num_input_tokens_seen": 490550736,
      "step": 8080,
      "train_runtime": 105450.2212,
      "train_tokens_per_second": 4651.965
    },
    {
      "epoch": 1.03125,
      "grad_norm": 0.6062632654229719,
      "learning_rate": 4.755662209960238e-06,
      "loss": 0.2742192506790161,
      "num_input_tokens_seen": 490851432,
      "step": 8085,
      "train_runtime": 105512.6796,
      "train_tokens_per_second": 4652.061
    },
    {
      "epoch": 1.031887755102041,
      "grad_norm": 0.620424732783558,
      "learning_rate": 4.750659400860945e-06,
      "loss": 0.2618445634841919,
      "num_input_tokens_seen": 491159256,
      "step": 8090,
      "train_runtime": 105568.4628,
      "train_tokens_per_second": 4652.519
    },
    {
      "epoch": 1.0325255102040816,
      "grad_norm": 0.6392574834281675,
      "learning_rate": 4.74565684199236e-06,
      "loss": 0.2552893400192261,
      "num_input_tokens_seen": 491470608,
      "step": 8095,
      "train_runtime": 105636.7798,
      "train_tokens_per_second": 4652.457
    },
    {
      "epoch": 1.0331632653061225,
      "grad_norm": 0.6726663259450165,
      "learning_rate": 4.740654538374896e-06,
      "loss": 0.2529830694198608,
      "num_input_tokens_seen": 491763560,
      "step": 8100,
      "train_runtime": 105696.9829,
      "train_tokens_per_second": 4652.579
    },
    {
      "epoch": 1.0338010204081634,
      "grad_norm": 0.6587618385624907,
      "learning_rate": 4.735652495028714e-06,
      "loss": 0.25743880271911623,
      "num_input_tokens_seen": 492064832,
      "step": 8105,
      "train_runtime": 105773.7668,
      "train_tokens_per_second": 4652.05
    },
    {
      "epoch": 1.034438775510204,
      "grad_norm": 0.5783179768407403,
      "learning_rate": 4.7306507169737176e-06,
      "loss": 0.2751675367355347,
      "num_input_tokens_seen": 492377080,
      "step": 8110,
      "train_runtime": 105839.6023,
      "train_tokens_per_second": 4652.106
    },
    {
      "epoch": 1.035076530612245,
      "grad_norm": 0.6465319699013996,
      "learning_rate": 4.725649209229537e-06,
      "loss": 0.25722560882568357,
      "num_input_tokens_seen": 492681472,
      "step": 8115,
      "train_runtime": 105912.9296,
      "train_tokens_per_second": 4651.759
    },
    {
      "epoch": 1.0357142857142858,
      "grad_norm": 0.6642526932627694,
      "learning_rate": 4.720647976815536e-06,
      "loss": 0.24139046669006348,
      "num_input_tokens_seen": 492970464,
      "step": 8120,
      "train_runtime": 105970.3804,
      "train_tokens_per_second": 4651.965
    },
    {
      "epoch": 1.0363520408163265,
      "grad_norm": 0.6473252402579394,
      "learning_rate": 4.7156470247508e-06,
      "loss": 0.275927734375,
      "num_input_tokens_seen": 493277120,
      "step": 8125,
      "train_runtime": 106037.1042,
      "train_tokens_per_second": 4651.929
    },
    {
      "epoch": 1.0369897959183674,
      "grad_norm": 0.6084817317793819,
      "learning_rate": 4.710646358054133e-06,
      "loss": 0.2354154109954834,
      "num_input_tokens_seen": 493583608,
      "step": 8130,
      "train_runtime": 106110.0221,
      "train_tokens_per_second": 4651.621
    },
    {
      "epoch": 1.0376275510204083,
      "grad_norm": 0.7212498914109421,
      "learning_rate": 4.705645981744055e-06,
      "loss": 0.258892822265625,
      "num_input_tokens_seen": 493890512,
      "step": 8135,
      "train_runtime": 106183.9086,
      "train_tokens_per_second": 4651.275
    },
    {
      "epoch": 1.038265306122449,
      "grad_norm": 0.6546681839807147,
      "learning_rate": 4.700645900838793e-06,
      "loss": 0.25546367168426515,
      "num_input_tokens_seen": 494183480,
      "step": 8140,
      "train_runtime": 106256.4337,
      "train_tokens_per_second": 4650.857
    },
    {
      "epoch": 1.0389030612244898,
      "grad_norm": 0.5896754902838178,
      "learning_rate": 4.695646120356275e-06,
      "loss": 0.2416550874710083,
      "num_input_tokens_seen": 494496488,
      "step": 8145,
      "train_runtime": 106323.3221,
      "train_tokens_per_second": 4650.875
    },
    {
      "epoch": 1.0395408163265305,
      "grad_norm": 0.5810570907446073,
      "learning_rate": 4.69064664531413e-06,
      "loss": 0.24819722175598144,
      "num_input_tokens_seen": 494818504,
      "step": 8150,
      "train_runtime": 106382.0337,
      "train_tokens_per_second": 4651.335
    },
    {
      "epoch": 1.0401785714285714,
      "grad_norm": 0.5770375167186163,
      "learning_rate": 4.685647480729684e-06,
      "loss": 0.23823702335357666,
      "num_input_tokens_seen": 495124752,
      "step": 8155,
      "train_runtime": 106455.573,
      "train_tokens_per_second": 4650.999
    },
    {
      "epoch": 1.0408163265306123,
      "grad_norm": 0.5969216247175317,
      "learning_rate": 4.680648631619945e-06,
      "loss": 0.266727876663208,
      "num_input_tokens_seen": 495417592,
      "step": 8160,
      "train_runtime": 106525.7383,
      "train_tokens_per_second": 4650.684
    },
    {
      "epoch": 1.041454081632653,
      "grad_norm": 0.7016300683844784,
      "learning_rate": 4.675650103001606e-06,
      "loss": 0.2604006052017212,
      "num_input_tokens_seen": 495721688,
      "step": 8165,
      "train_runtime": 106582.1881,
      "train_tokens_per_second": 4651.074
    },
    {
      "epoch": 1.0420918367346939,
      "grad_norm": 0.6467472959587393,
      "learning_rate": 4.670651899891043e-06,
      "loss": 0.2684458017349243,
      "num_input_tokens_seen": 496029336,
      "step": 8170,
      "train_runtime": 106639.8096,
      "train_tokens_per_second": 4651.446
    },
    {
      "epoch": 1.0427295918367347,
      "grad_norm": 0.6566571084817839,
      "learning_rate": 4.6656540273042986e-06,
      "loss": 0.23414976596832277,
      "num_input_tokens_seen": 496320248,
      "step": 8175,
      "train_runtime": 106705.718,
      "train_tokens_per_second": 4651.299
    },
    {
      "epoch": 1.0433673469387754,
      "grad_norm": 0.6642033019512137,
      "learning_rate": 4.66065649025709e-06,
      "loss": 0.26569809913635256,
      "num_input_tokens_seen": 496622736,
      "step": 8180,
      "train_runtime": 106782.1967,
      "train_tokens_per_second": 4650.801
    },
    {
      "epoch": 1.0440051020408163,
      "grad_norm": 0.6433544310559931,
      "learning_rate": 4.655659293764793e-06,
      "loss": 0.27560110092163087,
      "num_input_tokens_seen": 496940704,
      "step": 8185,
      "train_runtime": 106835.9992,
      "train_tokens_per_second": 4651.435
    },
    {
      "epoch": 1.0446428571428572,
      "grad_norm": 0.6801894221346246,
      "learning_rate": 4.6506624428424405e-06,
      "loss": 0.2756011724472046,
      "num_input_tokens_seen": 497235288,
      "step": 8190,
      "train_runtime": 106905.8988,
      "train_tokens_per_second": 4651.149
    },
    {
      "epoch": 1.0452806122448979,
      "grad_norm": 0.6593686921006262,
      "learning_rate": 4.645665942504728e-06,
      "loss": 0.27366573810577394,
      "num_input_tokens_seen": 497531328,
      "step": 8195,
      "train_runtime": 106972.2557,
      "train_tokens_per_second": 4651.031
    },
    {
      "epoch": 1.0459183673469388,
      "grad_norm": 0.7013843354836571,
      "learning_rate": 4.640669797765987e-06,
      "loss": 0.2521559715270996,
      "num_input_tokens_seen": 497820552,
      "step": 8200,
      "train_runtime": 107039.6241,
      "train_tokens_per_second": 4650.806
    },
    {
      "epoch": 1.0465561224489797,
      "grad_norm": 0.5962201167638378,
      "learning_rate": 4.6356740136402e-06,
      "loss": 0.25814969539642335,
      "num_input_tokens_seen": 498130520,
      "step": 8205,
      "train_runtime": 107095.0698,
      "train_tokens_per_second": 4651.293
    },
    {
      "epoch": 1.0471938775510203,
      "grad_norm": 0.6937696700908256,
      "learning_rate": 4.630678595140985e-06,
      "loss": 0.2693226099014282,
      "num_input_tokens_seen": 498428464,
      "step": 8210,
      "train_runtime": 107163.8015,
      "train_tokens_per_second": 4651.09
    },
    {
      "epoch": 1.0478316326530612,
      "grad_norm": 0.6359586973652941,
      "learning_rate": 4.625683547281592e-06,
      "loss": 0.2454256534576416,
      "num_input_tokens_seen": 498741112,
      "step": 8215,
      "train_runtime": 107237.407,
      "train_tokens_per_second": 4650.813
    },
    {
      "epoch": 1.0484693877551021,
      "grad_norm": 0.6222516562474432,
      "learning_rate": 4.620688875074902e-06,
      "loss": 0.2583413362503052,
      "num_input_tokens_seen": 499042560,
      "step": 8220,
      "train_runtime": 107303.6836,
      "train_tokens_per_second": 4650.75
    },
    {
      "epoch": 1.0491071428571428,
      "grad_norm": 0.684295054601306,
      "learning_rate": 4.615694583533418e-06,
      "loss": 0.24875664710998535,
      "num_input_tokens_seen": 499340840,
      "step": 8225,
      "train_runtime": 107362.1364,
      "train_tokens_per_second": 4650.996
    },
    {
      "epoch": 1.0497448979591837,
      "grad_norm": 0.6651829513990113,
      "learning_rate": 4.610700677669259e-06,
      "loss": 0.2657691717147827,
      "num_input_tokens_seen": 499653984,
      "step": 8230,
      "train_runtime": 107419.1222,
      "train_tokens_per_second": 4651.444
    },
    {
      "epoch": 1.0503826530612246,
      "grad_norm": 0.6262488827007487,
      "learning_rate": 4.605707162494157e-06,
      "loss": 0.26746602058410646,
      "num_input_tokens_seen": 499955880,
      "step": 8235,
      "train_runtime": 107484.7661,
      "train_tokens_per_second": 4651.412
    },
    {
      "epoch": 1.0510204081632653,
      "grad_norm": 0.6310229490441964,
      "learning_rate": 4.600714043019456e-06,
      "loss": 0.25636651515960696,
      "num_input_tokens_seen": 500260776,
      "step": 8240,
      "train_runtime": 107544.1602,
      "train_tokens_per_second": 4651.678
    },
    {
      "epoch": 1.0516581632653061,
      "grad_norm": 0.6879500520531826,
      "learning_rate": 4.5957213242560985e-06,
      "loss": 0.2619192600250244,
      "num_input_tokens_seen": 500560728,
      "step": 8245,
      "train_runtime": 107609.1779,
      "train_tokens_per_second": 4651.655
    },
    {
      "epoch": 1.052295918367347,
      "grad_norm": 0.5514238162725921,
      "learning_rate": 4.5907290112146265e-06,
      "loss": 0.24567170143127443,
      "num_input_tokens_seen": 500864232,
      "step": 8250,
      "train_runtime": 107665.2845,
      "train_tokens_per_second": 4652.049
    },
    {
      "epoch": 1.0529336734693877,
      "grad_norm": 0.6918867101607463,
      "learning_rate": 4.585737108905178e-06,
      "loss": 0.25845234394073485,
      "num_input_tokens_seen": 501171944,
      "step": 8255,
      "train_runtime": 107714.5333,
      "train_tokens_per_second": 4652.779
    },
    {
      "epoch": 1.0535714285714286,
      "grad_norm": 0.6568866025069288,
      "learning_rate": 4.5807456223374695e-06,
      "loss": 0.26011834144592283,
      "num_input_tokens_seen": 501482344,
      "step": 8260,
      "train_runtime": 107784.813,
      "train_tokens_per_second": 4652.625
    },
    {
      "epoch": 1.0542091836734695,
      "grad_norm": 0.6772907381209184,
      "learning_rate": 4.57575455652081e-06,
      "loss": 0.24192533493041993,
      "num_input_tokens_seen": 501767928,
      "step": 8265,
      "train_runtime": 107859.0094,
      "train_tokens_per_second": 4652.072
    },
    {
      "epoch": 1.0548469387755102,
      "grad_norm": 0.6817981326124604,
      "learning_rate": 4.570763916464084e-06,
      "loss": 0.2520601749420166,
      "num_input_tokens_seen": 502063992,
      "step": 8270,
      "train_runtime": 107921.5358,
      "train_tokens_per_second": 4652.121
    },
    {
      "epoch": 1.055484693877551,
      "grad_norm": 0.5940719591783477,
      "learning_rate": 4.565773707175743e-06,
      "loss": 0.27521862983703616,
      "num_input_tokens_seen": 502375752,
      "step": 8275,
      "train_runtime": 107975.6393,
      "train_tokens_per_second": 4652.677
    },
    {
      "epoch": 1.0561224489795917,
      "grad_norm": 0.6494867279933533,
      "learning_rate": 4.560783933663814e-06,
      "loss": 0.2762296676635742,
      "num_input_tokens_seen": 502676192,
      "step": 8280,
      "train_runtime": 108035.3648,
      "train_tokens_per_second": 4652.886
    },
    {
      "epoch": 1.0567602040816326,
      "grad_norm": 0.629753785819062,
      "learning_rate": 4.5557946009358846e-06,
      "loss": 0.2601089239120483,
      "num_input_tokens_seen": 502984856,
      "step": 8285,
      "train_runtime": 108105.2293,
      "train_tokens_per_second": 4652.734
    },
    {
      "epoch": 1.0573979591836735,
      "grad_norm": 0.5748847670556543,
      "learning_rate": 4.550805713999094e-06,
      "loss": 0.23802318572998046,
      "num_input_tokens_seen": 503296512,
      "step": 8290,
      "train_runtime": 108169.2907,
      "train_tokens_per_second": 4652.86
    },
    {
      "epoch": 1.0580357142857142,
      "grad_norm": 0.6811748430518816,
      "learning_rate": 4.545817277860143e-06,
      "loss": 0.27293121814727783,
      "num_input_tokens_seen": 503588512,
      "step": 8295,
      "train_runtime": 108232.862,
      "train_tokens_per_second": 4652.824
    },
    {
      "epoch": 1.058673469387755,
      "grad_norm": 0.6694897094057808,
      "learning_rate": 4.540829297525272e-06,
      "loss": 0.27684831619262695,
      "num_input_tokens_seen": 503900680,
      "step": 8300,
      "train_runtime": 108288.235,
      "train_tokens_per_second": 4653.328
    },
    {
      "epoch": 1.059311224489796,
      "grad_norm": 0.6332558484893667,
      "learning_rate": 4.53584177800027e-06,
      "loss": 0.24174211025238038,
      "num_input_tokens_seen": 504207016,
      "step": 8305,
      "train_runtime": 108344.5029,
      "train_tokens_per_second": 4653.739
    },
    {
      "epoch": 1.0599489795918366,
      "grad_norm": 0.6255828542219396,
      "learning_rate": 4.530854724290461e-06,
      "loss": 0.24687228202819825,
      "num_input_tokens_seen": 504518576,
      "step": 8310,
      "train_runtime": 108417.3121,
      "train_tokens_per_second": 4653.487
    },
    {
      "epoch": 1.0605867346938775,
      "grad_norm": 0.6754211592346366,
      "learning_rate": 4.5258681414007005e-06,
      "loss": 0.2555249691009521,
      "num_input_tokens_seen": 504814448,
      "step": 8315,
      "train_runtime": 108485.7262,
      "train_tokens_per_second": 4653.28
    },
    {
      "epoch": 1.0612244897959184,
      "grad_norm": 0.6111824380618206,
      "learning_rate": 4.5208820343353726e-06,
      "loss": 0.24149341583251954,
      "num_input_tokens_seen": 505109024,
      "step": 8320,
      "train_runtime": 108560.1597,
      "train_tokens_per_second": 4652.803
    },
    {
      "epoch": 1.061862244897959,
      "grad_norm": 0.6921188319092784,
      "learning_rate": 4.515896408098387e-06,
      "loss": 0.24489209651947022,
      "num_input_tokens_seen": 505399152,
      "step": 8325,
      "train_runtime": 108632.4622,
      "train_tokens_per_second": 4652.377
    },
    {
      "epoch": 1.0625,
      "grad_norm": 0.6457386674680605,
      "learning_rate": 4.510911267693164e-06,
      "loss": 0.27535414695739746,
      "num_input_tokens_seen": 505692008,
      "step": 8330,
      "train_runtime": 108694.8052,
      "train_tokens_per_second": 4652.403
    },
    {
      "epoch": 1.063137755102041,
      "grad_norm": 0.6404589113203911,
      "learning_rate": 4.505926618122641e-06,
      "loss": 0.25455851554870607,
      "num_input_tokens_seen": 506000168,
      "step": 8335,
      "train_runtime": 108761.9998,
      "train_tokens_per_second": 4652.362
    },
    {
      "epoch": 1.0637755102040816,
      "grad_norm": 0.6726710709845345,
      "learning_rate": 4.5009424643892644e-06,
      "loss": 0.2708827018737793,
      "num_input_tokens_seen": 506306912,
      "step": 8340,
      "train_runtime": 108815.3123,
      "train_tokens_per_second": 4652.901
    },
    {
      "epoch": 1.0644132653061225,
      "grad_norm": 0.6698963462362696,
      "learning_rate": 4.4959588114949785e-06,
      "loss": 0.24709620475769042,
      "num_input_tokens_seen": 506598448,
      "step": 8345,
      "train_runtime": 108871.1825,
      "train_tokens_per_second": 4653.191
    },
    {
      "epoch": 1.0650510204081634,
      "grad_norm": 0.6153694586832253,
      "learning_rate": 4.490975664441227e-06,
      "loss": 0.24962987899780273,
      "num_input_tokens_seen": 506905000,
      "step": 8350,
      "train_runtime": 108933.4304,
      "train_tokens_per_second": 4653.347
    },
    {
      "epoch": 1.065688775510204,
      "grad_norm": 0.620077961138061,
      "learning_rate": 4.4859930282289475e-06,
      "loss": 0.2546701431274414,
      "num_input_tokens_seen": 507222320,
      "step": 8355,
      "train_runtime": 108988.7403,
      "train_tokens_per_second": 4653.897
    },
    {
      "epoch": 1.066326530612245,
      "grad_norm": 0.6231083614697688,
      "learning_rate": 4.48101090785856e-06,
      "loss": 0.28885631561279296,
      "num_input_tokens_seen": 507517176,
      "step": 8360,
      "train_runtime": 109062.4286,
      "train_tokens_per_second": 4653.456
    },
    {
      "epoch": 1.0669642857142858,
      "grad_norm": 0.7449931510092558,
      "learning_rate": 4.4760293083299736e-06,
      "loss": 0.26137232780456543,
      "num_input_tokens_seen": 507804944,
      "step": 8365,
      "train_runtime": 109128.8225,
      "train_tokens_per_second": 4653.261
    },
    {
      "epoch": 1.0676020408163265,
      "grad_norm": 0.6686152755009991,
      "learning_rate": 4.471048234642572e-06,
      "loss": 0.2702186584472656,
      "num_input_tokens_seen": 508107408,
      "step": 8370,
      "train_runtime": 109188.7672,
      "train_tokens_per_second": 4653.477
    },
    {
      "epoch": 1.0682397959183674,
      "grad_norm": 0.7020326980677524,
      "learning_rate": 4.466067691795208e-06,
      "loss": 0.26437246799468994,
      "num_input_tokens_seen": 508414640,
      "step": 8375,
      "train_runtime": 109261.8547,
      "train_tokens_per_second": 4653.176
    },
    {
      "epoch": 1.068877551020408,
      "grad_norm": 0.6688995113612123,
      "learning_rate": 4.4610876847862034e-06,
      "loss": 0.2574469566345215,
      "num_input_tokens_seen": 508717400,
      "step": 8380,
      "train_runtime": 109335.2266,
      "train_tokens_per_second": 4652.822
    },
    {
      "epoch": 1.069515306122449,
      "grad_norm": 0.714734676236501,
      "learning_rate": 4.456108218613346e-06,
      "loss": 0.27723989486694334,
      "num_input_tokens_seen": 509023408,
      "step": 8385,
      "train_runtime": 109394.6985,
      "train_tokens_per_second": 4653.09
    },
    {
      "epoch": 1.0701530612244898,
      "grad_norm": 0.5704584998462686,
      "learning_rate": 4.451129298273877e-06,
      "loss": 0.2590214252471924,
      "num_input_tokens_seen": 509332808,
      "step": 8390,
      "train_runtime": 109456.5926,
      "train_tokens_per_second": 4653.286
    },
    {
      "epoch": 1.0707908163265305,
      "grad_norm": 0.6512911702869287,
      "learning_rate": 4.4461509287644885e-06,
      "loss": 0.25503556728363036,
      "num_input_tokens_seen": 509641560,
      "step": 8395,
      "train_runtime": 109520.4782,
      "train_tokens_per_second": 4653.391
    },
    {
      "epoch": 1.0714285714285714,
      "grad_norm": 0.5976966356347777,
      "learning_rate": 4.441173115081327e-06,
      "loss": 0.23296217918395995,
      "num_input_tokens_seen": 509941488,
      "step": 8400,
      "train_runtime": 109585.1984,
      "train_tokens_per_second": 4653.379
    },
    {
      "epoch": 1.0720663265306123,
      "grad_norm": 0.6228044749405278,
      "learning_rate": 4.43619586221997e-06,
      "loss": 0.27005853652954104,
      "num_input_tokens_seen": 510234680,
      "step": 8405,
      "train_runtime": 109642.8463,
      "train_tokens_per_second": 4653.607
    },
    {
      "epoch": 1.072704081632653,
      "grad_norm": 0.63348561639179,
      "learning_rate": 4.4312191751754415e-06,
      "loss": 0.23705193996429444,
      "num_input_tokens_seen": 510535432,
      "step": 8410,
      "train_runtime": 109715.4691,
      "train_tokens_per_second": 4653.268
    },
    {
      "epoch": 1.0733418367346939,
      "grad_norm": 0.6841082125372675,
      "learning_rate": 4.426243058942196e-06,
      "loss": 0.2547664403915405,
      "num_input_tokens_seen": 510829424,
      "step": 8415,
      "train_runtime": 109774.5772,
      "train_tokens_per_second": 4653.44
    },
    {
      "epoch": 1.0739795918367347,
      "grad_norm": 0.7232820006970838,
      "learning_rate": 4.42126751851411e-06,
      "loss": 0.2672074556350708,
      "num_input_tokens_seen": 511132368,
      "step": 8420,
      "train_runtime": 109841.1571,
      "train_tokens_per_second": 4653.377
    },
    {
      "epoch": 1.0746173469387754,
      "grad_norm": 0.7090149309099536,
      "learning_rate": 4.4162925588844895e-06,
      "loss": 0.269744348526001,
      "num_input_tokens_seen": 511428392,
      "step": 8425,
      "train_runtime": 109906.826,
      "train_tokens_per_second": 4653.291
    },
    {
      "epoch": 1.0752551020408163,
      "grad_norm": 0.6697337167265367,
      "learning_rate": 4.4113181850460504e-06,
      "loss": 0.25969829559326174,
      "num_input_tokens_seen": 511735696,
      "step": 8430,
      "train_runtime": 109975.9882,
      "train_tokens_per_second": 4653.158
    },
    {
      "epoch": 1.0758928571428572,
      "grad_norm": 0.6955769703739513,
      "learning_rate": 4.406344401990927e-06,
      "loss": 0.2720786571502686,
      "num_input_tokens_seen": 512041384,
      "step": 8435,
      "train_runtime": 110030.3589,
      "train_tokens_per_second": 4653.637
    },
    {
      "epoch": 1.0765306122448979,
      "grad_norm": 0.6990517466345093,
      "learning_rate": 4.401371214710658e-06,
      "loss": 0.2635928153991699,
      "num_input_tokens_seen": 512335736,
      "step": 8440,
      "train_runtime": 110084.6351,
      "train_tokens_per_second": 4654.017
    },
    {
      "epoch": 1.0771683673469388,
      "grad_norm": 0.5871716823716583,
      "learning_rate": 4.396398628196181e-06,
      "loss": 0.24565181732177735,
      "num_input_tokens_seen": 512636824,
      "step": 8445,
      "train_runtime": 110158.4388,
      "train_tokens_per_second": 4653.632
    },
    {
      "epoch": 1.0778061224489797,
      "grad_norm": 0.6538259620328585,
      "learning_rate": 4.391426647437836e-06,
      "loss": 0.2619929313659668,
      "num_input_tokens_seen": 512931896,
      "step": 8450,
      "train_runtime": 110223.0242,
      "train_tokens_per_second": 4653.582
    },
    {
      "epoch": 1.0784438775510203,
      "grad_norm": 0.626992678120448,
      "learning_rate": 4.386455277425355e-06,
      "loss": 0.24539623260498047,
      "num_input_tokens_seen": 513237216,
      "step": 8455,
      "train_runtime": 110282.9833,
      "train_tokens_per_second": 4653.821
    },
    {
      "epoch": 1.0790816326530612,
      "grad_norm": 0.6565855097817582,
      "learning_rate": 4.381484523147852e-06,
      "loss": 0.2395007610321045,
      "num_input_tokens_seen": 513545312,
      "step": 8460,
      "train_runtime": 110356.3949,
      "train_tokens_per_second": 4653.517
    },
    {
      "epoch": 1.0797193877551021,
      "grad_norm": 0.6354154110897268,
      "learning_rate": 4.376514389593827e-06,
      "loss": 0.24681005477905274,
      "num_input_tokens_seen": 513836768,
      "step": 8465,
      "train_runtime": 110425.6141,
      "train_tokens_per_second": 4653.239
    },
    {
      "epoch": 1.0803571428571428,
      "grad_norm": 0.7171363243436465,
      "learning_rate": 4.3715448817511565e-06,
      "loss": 0.26409809589385985,
      "num_input_tokens_seen": 514138696,
      "step": 8470,
      "train_runtime": 110492.6387,
      "train_tokens_per_second": 4653.149
    },
    {
      "epoch": 1.0809948979591837,
      "grad_norm": 0.7197280803089543,
      "learning_rate": 4.3665760046070885e-06,
      "loss": 0.25276215076446534,
      "num_input_tokens_seen": 514447760,
      "step": 8475,
      "train_runtime": 110562.2261,
      "train_tokens_per_second": 4653.016
    },
    {
      "epoch": 1.0816326530612246,
      "grad_norm": 0.6764093081611164,
      "learning_rate": 4.361607763148236e-06,
      "loss": 0.24439802169799804,
      "num_input_tokens_seen": 514743056,
      "step": 8480,
      "train_runtime": 110626.314,
      "train_tokens_per_second": 4652.989
    },
    {
      "epoch": 1.0822704081632653,
      "grad_norm": 0.7445548995509881,
      "learning_rate": 4.356640162360582e-06,
      "loss": 0.2473681688308716,
      "num_input_tokens_seen": 515046088,
      "step": 8485,
      "train_runtime": 110697.5579,
      "train_tokens_per_second": 4652.732
    },
    {
      "epoch": 1.0829081632653061,
      "grad_norm": 0.6605282556895079,
      "learning_rate": 4.3516732072294545e-06,
      "loss": 0.25981926918029785,
      "num_input_tokens_seen": 515354392,
      "step": 8490,
      "train_runtime": 110768.4954,
      "train_tokens_per_second": 4652.536
    },
    {
      "epoch": 1.083545918367347,
      "grad_norm": 0.78980964959287,
      "learning_rate": 4.346706902739544e-06,
      "loss": 0.2604411840438843,
      "num_input_tokens_seen": 515653976,
      "step": 8495,
      "train_runtime": 110826.1551,
      "train_tokens_per_second": 4652.818
    },
    {
      "epoch": 1.0841836734693877,
      "grad_norm": 0.6105232803409607,
      "learning_rate": 4.3417412538748824e-06,
      "loss": 0.2681291580200195,
      "num_input_tokens_seen": 515960096,
      "step": 8500,
      "train_runtime": 110894.6781,
      "train_tokens_per_second": 4652.704
    },
    {
      "epoch": 1.0848214285714286,
      "grad_norm": 0.7533094617059788,
      "learning_rate": 4.336776265618844e-06,
      "loss": 0.24369583129882813,
      "num_input_tokens_seen": 516259976,
      "step": 8505,
      "train_runtime": 110971.1032,
      "train_tokens_per_second": 4652.202
    },
    {
      "epoch": 1.0854591836734695,
      "grad_norm": 0.7094543382418285,
      "learning_rate": 4.331811942954143e-06,
      "loss": 0.2504972696304321,
      "num_input_tokens_seen": 516557400,
      "step": 8510,
      "train_runtime": 111025.1931,
      "train_tokens_per_second": 4652.614
    },
    {
      "epoch": 1.0860969387755102,
      "grad_norm": 0.5659879062873587,
      "learning_rate": 4.3268482908628245e-06,
      "loss": 0.25997779369354246,
      "num_input_tokens_seen": 516859248,
      "step": 8515,
      "train_runtime": 111085.0512,
      "train_tokens_per_second": 4652.825
    },
    {
      "epoch": 1.086734693877551,
      "grad_norm": 0.6567166938858798,
      "learning_rate": 4.321885314326258e-06,
      "loss": 0.23948345184326172,
      "num_input_tokens_seen": 517168080,
      "step": 8520,
      "train_runtime": 111154.8731,
      "train_tokens_per_second": 4652.68
    },
    {
      "epoch": 1.087372448979592,
      "grad_norm": 0.6788260721034282,
      "learning_rate": 4.316923018325137e-06,
      "loss": 0.25132384300231936,
      "num_input_tokens_seen": 517462696,
      "step": 8525,
      "train_runtime": 111221.6272,
      "train_tokens_per_second": 4652.537
    },
    {
      "epoch": 1.0880102040816326,
      "grad_norm": 0.6899299468505758,
      "learning_rate": 4.3119614078394756e-06,
      "loss": 0.2746640682220459,
      "num_input_tokens_seen": 517771872,
      "step": 8530,
      "train_runtime": 111296.2163,
      "train_tokens_per_second": 4652.197
    },
    {
      "epoch": 1.0886479591836735,
      "grad_norm": 0.6958708744235721,
      "learning_rate": 4.3070004878485925e-06,
      "loss": 0.25914363861083983,
      "num_input_tokens_seen": 518070000,
      "step": 8535,
      "train_runtime": 111359.8351,
      "train_tokens_per_second": 4652.216
    },
    {
      "epoch": 1.0892857142857142,
      "grad_norm": 0.6236966250731748,
      "learning_rate": 4.30204026333112e-06,
      "loss": 0.2593933343887329,
      "num_input_tokens_seen": 518386712,
      "step": 8540,
      "train_runtime": 111422.2229,
      "train_tokens_per_second": 4652.454
    },
    {
      "epoch": 1.089923469387755,
      "grad_norm": 0.6224503051450146,
      "learning_rate": 4.297080739264987e-06,
      "loss": 0.2581711053848267,
      "num_input_tokens_seen": 518698160,
      "step": 8545,
      "train_runtime": 111474.3926,
      "train_tokens_per_second": 4653.07
    },
    {
      "epoch": 1.090561224489796,
      "grad_norm": 0.6347803948249341,
      "learning_rate": 4.292121920627423e-06,
      "loss": 0.25314912796020506,
      "num_input_tokens_seen": 518996176,
      "step": 8550,
      "train_runtime": 111544.4488,
      "train_tokens_per_second": 4652.819
    },
    {
      "epoch": 1.0911989795918366,
      "grad_norm": 0.775808725416068,
      "learning_rate": 4.287163812394952e-06,
      "loss": 0.25922203063964844,
      "num_input_tokens_seen": 519299072,
      "step": 8555,
      "train_runtime": 111618.4393,
      "train_tokens_per_second": 4652.449
    },
    {
      "epoch": 1.0918367346938775,
      "grad_norm": 0.6314140806578885,
      "learning_rate": 4.282206419543379e-06,
      "loss": 0.2555856227874756,
      "num_input_tokens_seen": 519593872,
      "step": 8560,
      "train_runtime": 111695.7856,
      "train_tokens_per_second": 4651.866
    },
    {
      "epoch": 1.0924744897959184,
      "grad_norm": 0.6713664786987801,
      "learning_rate": 4.2772497470477934e-06,
      "loss": 0.2524415493011475,
      "num_input_tokens_seen": 519899040,
      "step": 8565,
      "train_runtime": 111756.2816,
      "train_tokens_per_second": 4652.079
    },
    {
      "epoch": 1.093112244897959,
      "grad_norm": 0.632549152859099,
      "learning_rate": 4.272293799882565e-06,
      "loss": 0.25772848129272463,
      "num_input_tokens_seen": 520204488,
      "step": 8570,
      "train_runtime": 111822.7613,
      "train_tokens_per_second": 4652.045
    },
    {
      "epoch": 1.09375,
      "grad_norm": 0.6935953187354543,
      "learning_rate": 4.267338583021331e-06,
      "loss": 0.26155643463134765,
      "num_input_tokens_seen": 520506248,
      "step": 8575,
      "train_runtime": 111880.4115,
      "train_tokens_per_second": 4652.345
    },
    {
      "epoch": 1.094387755102041,
      "grad_norm": 0.6267017654146606,
      "learning_rate": 4.262384101436997e-06,
      "loss": 0.2655954360961914,
      "num_input_tokens_seen": 520817536,
      "step": 8580,
      "train_runtime": 111952.1936,
      "train_tokens_per_second": 4652.142
    },
    {
      "epoch": 1.0950255102040816,
      "grad_norm": 0.6545632677163803,
      "learning_rate": 4.257430360101734e-06,
      "loss": 0.25361220836639403,
      "num_input_tokens_seen": 521116136,
      "step": 8585,
      "train_runtime": 112021.71,
      "train_tokens_per_second": 4651.921
    },
    {
      "epoch": 1.0956632653061225,
      "grad_norm": 0.6773920458249225,
      "learning_rate": 4.252477363986963e-06,
      "loss": 0.2651734590530396,
      "num_input_tokens_seen": 521413200,
      "step": 8590,
      "train_runtime": 112083.5453,
      "train_tokens_per_second": 4652.005
    },
    {
      "epoch": 1.0963010204081634,
      "grad_norm": 0.6760247929377783,
      "learning_rate": 4.247525118063366e-06,
      "loss": 0.25967419147491455,
      "num_input_tokens_seen": 521719080,
      "step": 8595,
      "train_runtime": 112146.6083,
      "train_tokens_per_second": 4652.116
    },
    {
      "epoch": 1.096938775510204,
      "grad_norm": 0.7361384427827368,
      "learning_rate": 4.242573627300866e-06,
      "loss": 0.2496013879776001,
      "num_input_tokens_seen": 522027776,
      "step": 8600,
      "train_runtime": 112211.59,
      "train_tokens_per_second": 4652.173
    },
    {
      "epoch": 1.097576530612245,
      "grad_norm": 0.7112095971701625,
      "learning_rate": 4.237622896668628e-06,
      "loss": 0.23990926742553711,
      "num_input_tokens_seen": 522328536,
      "step": 8605,
      "train_runtime": 112275.6717,
      "train_tokens_per_second": 4652.197
    },
    {
      "epoch": 1.0982142857142858,
      "grad_norm": 0.5728848513613037,
      "learning_rate": 4.232672931135058e-06,
      "loss": 0.2499545097351074,
      "num_input_tokens_seen": 522618664,
      "step": 8610,
      "train_runtime": 112343.064,
      "train_tokens_per_second": 4651.989
    },
    {
      "epoch": 1.0988520408163265,
      "grad_norm": 0.6471874739768435,
      "learning_rate": 4.227723735667793e-06,
      "loss": 0.25888404846191404,
      "num_input_tokens_seen": 522921736,
      "step": 8615,
      "train_runtime": 112405.616,
      "train_tokens_per_second": 4652.096
    },
    {
      "epoch": 1.0994897959183674,
      "grad_norm": 0.6875386080205632,
      "learning_rate": 4.222775315233693e-06,
      "loss": 0.2819283723831177,
      "num_input_tokens_seen": 523233512,
      "step": 8620,
      "train_runtime": 112478.6201,
      "train_tokens_per_second": 4651.849
    },
    {
      "epoch": 1.100127551020408,
      "grad_norm": 0.6569749537869434,
      "learning_rate": 4.217827674798845e-06,
      "loss": 0.2621577739715576,
      "num_input_tokens_seen": 523541032,
      "step": 8625,
      "train_runtime": 112549.288,
      "train_tokens_per_second": 4651.66
    },
    {
      "epoch": 1.100765306122449,
      "grad_norm": 0.6790870912016186,
      "learning_rate": 4.2128808193285555e-06,
      "loss": 0.26994221210479735,
      "num_input_tokens_seen": 523846904,
      "step": 8630,
      "train_runtime": 112623.9832,
      "train_tokens_per_second": 4651.291
    },
    {
      "epoch": 1.1014030612244898,
      "grad_norm": 0.6686336627874682,
      "learning_rate": 4.2079347537873325e-06,
      "loss": 0.262078070640564,
      "num_input_tokens_seen": 524142576,
      "step": 8635,
      "train_runtime": 112698.2326,
      "train_tokens_per_second": 4650.85
    },
    {
      "epoch": 1.1020408163265305,
      "grad_norm": 0.6913949851570558,
      "learning_rate": 4.2029894831389036e-06,
      "loss": 0.2705439329147339,
      "num_input_tokens_seen": 524450280,
      "step": 8640,
      "train_runtime": 112766.3885,
      "train_tokens_per_second": 4650.768
    },
    {
      "epoch": 1.1026785714285714,
      "grad_norm": 0.644982167172311,
      "learning_rate": 4.198045012346192e-06,
      "loss": 0.25299763679504395,
      "num_input_tokens_seen": 524764680,
      "step": 8645,
      "train_runtime": 112839.1068,
      "train_tokens_per_second": 4650.557
    },
    {
      "epoch": 1.1033163265306123,
      "grad_norm": 0.6974639447763437,
      "learning_rate": 4.1931013463713165e-06,
      "loss": 0.2784655809402466,
      "num_input_tokens_seen": 525060152,
      "step": 8650,
      "train_runtime": 112901.3259,
      "train_tokens_per_second": 4650.611
    },
    {
      "epoch": 1.103954081632653,
      "grad_norm": 0.6297322783406425,
      "learning_rate": 4.188158490175596e-06,
      "loss": 0.23096976280212403,
      "num_input_tokens_seen": 525366264,
      "step": 8655,
      "train_runtime": 112965.6203,
      "train_tokens_per_second": 4650.674
    },
    {
      "epoch": 1.1045918367346939,
      "grad_norm": 0.621717322580187,
      "learning_rate": 4.1832164487195295e-06,
      "loss": 0.24602036476135253,
      "num_input_tokens_seen": 525679368,
      "step": 8660,
      "train_runtime": 113032.0236,
      "train_tokens_per_second": 4650.712
    },
    {
      "epoch": 1.1052295918367347,
      "grad_norm": 0.6702723562719789,
      "learning_rate": 4.1782752269627995e-06,
      "loss": 0.2634816408157349,
      "num_input_tokens_seen": 525989064,
      "step": 8665,
      "train_runtime": 113087.4765,
      "train_tokens_per_second": 4651.17
    },
    {
      "epoch": 1.1058673469387754,
      "grad_norm": 0.6655949634928194,
      "learning_rate": 4.17333482986427e-06,
      "loss": 0.25561840534210206,
      "num_input_tokens_seen": 526300736,
      "step": 8670,
      "train_runtime": 113160.9482,
      "train_tokens_per_second": 4650.904
    },
    {
      "epoch": 1.1065051020408163,
      "grad_norm": 0.6482019607965435,
      "learning_rate": 4.16839526238197e-06,
      "loss": 0.2742323875427246,
      "num_input_tokens_seen": 526603144,
      "step": 8675,
      "train_runtime": 113230.2081,
      "train_tokens_per_second": 4650.73
    },
    {
      "epoch": 1.1071428571428572,
      "grad_norm": 0.6421331929913958,
      "learning_rate": 4.1634565294731045e-06,
      "loss": 0.2627676248550415,
      "num_input_tokens_seen": 526911344,
      "step": 8680,
      "train_runtime": 113300.3539,
      "train_tokens_per_second": 4650.571
    },
    {
      "epoch": 1.1077806122448979,
      "grad_norm": 0.6216744146333252,
      "learning_rate": 4.158518636094037e-06,
      "loss": 0.2641258478164673,
      "num_input_tokens_seen": 527229720,
      "step": 8685,
      "train_runtime": 113361.3858,
      "train_tokens_per_second": 4650.876
    },
    {
      "epoch": 1.1084183673469388,
      "grad_norm": 0.6039791858554615,
      "learning_rate": 4.153581587200287e-06,
      "loss": 0.2487846851348877,
      "num_input_tokens_seen": 527532776,
      "step": 8690,
      "train_runtime": 113438.9355,
      "train_tokens_per_second": 4650.368
    },
    {
      "epoch": 1.1090561224489797,
      "grad_norm": 0.6777894782617317,
      "learning_rate": 4.148645387746526e-06,
      "loss": 0.2532509803771973,
      "num_input_tokens_seen": 527834776,
      "step": 8695,
      "train_runtime": 113498.0376,
      "train_tokens_per_second": 4650.607
    },
    {
      "epoch": 1.1096938775510203,
      "grad_norm": 0.8143053554422732,
      "learning_rate": 4.143710042686579e-06,
      "loss": 0.25780043601989744,
      "num_input_tokens_seen": 528137192,
      "step": 8700,
      "train_runtime": 113570.0771,
      "train_tokens_per_second": 4650.32
    },
    {
      "epoch": 1.1103316326530612,
      "grad_norm": 0.6394011415019081,
      "learning_rate": 4.138775556973406e-06,
      "loss": 0.26498196125030515,
      "num_input_tokens_seen": 528440456,
      "step": 8705,
      "train_runtime": 113642.3309,
      "train_tokens_per_second": 4650.032
    },
    {
      "epoch": 1.1109693877551021,
      "grad_norm": 0.6489234851554295,
      "learning_rate": 4.133841935559109e-06,
      "loss": 0.26532301902770994,
      "num_input_tokens_seen": 528746080,
      "step": 8710,
      "train_runtime": 113705.612,
      "train_tokens_per_second": 4650.132
    },
    {
      "epoch": 1.1116071428571428,
      "grad_norm": 0.5899919397086718,
      "learning_rate": 4.1289091833949235e-06,
      "loss": 0.24462313652038575,
      "num_input_tokens_seen": 529028272,
      "step": 8715,
      "train_runtime": 113760.9322,
      "train_tokens_per_second": 4650.351
    },
    {
      "epoch": 1.1122448979591837,
      "grad_norm": 0.6681595352961904,
      "learning_rate": 4.123977305431209e-06,
      "loss": 0.26124532222747804,
      "num_input_tokens_seen": 529326792,
      "step": 8720,
      "train_runtime": 113828.2974,
      "train_tokens_per_second": 4650.221
    },
    {
      "epoch": 1.1128826530612246,
      "grad_norm": 0.7967346989172764,
      "learning_rate": 4.119046306617449e-06,
      "loss": 0.26501011848449707,
      "num_input_tokens_seen": 529624760,
      "step": 8725,
      "train_runtime": 113893.5473,
      "train_tokens_per_second": 4650.174
    },
    {
      "epoch": 1.1135204081632653,
      "grad_norm": 0.6203710374313256,
      "learning_rate": 4.114116191902248e-06,
      "loss": 0.28026351928710935,
      "num_input_tokens_seen": 529934352,
      "step": 8730,
      "train_runtime": 113964.0747,
      "train_tokens_per_second": 4650.012
    },
    {
      "epoch": 1.1141581632653061,
      "grad_norm": 0.6771760971082005,
      "learning_rate": 4.109186966233315e-06,
      "loss": 0.276874041557312,
      "num_input_tokens_seen": 530240440,
      "step": 8735,
      "train_runtime": 114033.2417,
      "train_tokens_per_second": 4649.876
    },
    {
      "epoch": 1.114795918367347,
      "grad_norm": 0.6677140771253915,
      "learning_rate": 4.104258634557478e-06,
      "loss": 0.2529959201812744,
      "num_input_tokens_seen": 530544176,
      "step": 8740,
      "train_runtime": 114107.8513,
      "train_tokens_per_second": 4649.498
    },
    {
      "epoch": 1.1154336734693877,
      "grad_norm": 0.5849309371137589,
      "learning_rate": 4.09933120182066e-06,
      "loss": 0.24083590507507324,
      "num_input_tokens_seen": 530825680,
      "step": 8745,
      "train_runtime": 114179.7829,
      "train_tokens_per_second": 4649.034
    },
    {
      "epoch": 1.1160714285714286,
      "grad_norm": 0.6986256981069222,
      "learning_rate": 4.094404672967882e-06,
      "loss": 0.2577672958374023,
      "num_input_tokens_seen": 531129632,
      "step": 8750,
      "train_runtime": 114244.6756,
      "train_tokens_per_second": 4649.054
    },
    {
      "epoch": 1.1167091836734695,
      "grad_norm": 0.6291116675381879,
      "learning_rate": 4.089479052943261e-06,
      "loss": 0.2546741008758545,
      "num_input_tokens_seen": 531421264,
      "step": 8755,
      "train_runtime": 114311.4113,
      "train_tokens_per_second": 4648.891
    },
    {
      "epoch": 1.1173469387755102,
      "grad_norm": 0.6683424189424261,
      "learning_rate": 4.084554346690003e-06,
      "loss": 0.2510077476501465,
      "num_input_tokens_seen": 531722392,
      "step": 8760,
      "train_runtime": 114382.5873,
      "train_tokens_per_second": 4648.631
    },
    {
      "epoch": 1.117984693877551,
      "grad_norm": 0.6811331218594819,
      "learning_rate": 4.079630559150392e-06,
      "loss": 0.2494870662689209,
      "num_input_tokens_seen": 532031576,
      "step": 8765,
      "train_runtime": 114458.0853,
      "train_tokens_per_second": 4648.266
    },
    {
      "epoch": 1.118622448979592,
      "grad_norm": 0.6409957446589616,
      "learning_rate": 4.074707695265792e-06,
      "loss": 0.2718391418457031,
      "num_input_tokens_seen": 532346776,
      "step": 8770,
      "train_runtime": 114514.0097,
      "train_tokens_per_second": 4648.748
    },
    {
      "epoch": 1.1192602040816326,
      "grad_norm": 0.6196177887451563,
      "learning_rate": 4.069785759976645e-06,
      "loss": 0.24614083766937256,
      "num_input_tokens_seen": 532659184,
      "step": 8775,
      "train_runtime": 114570.0017,
      "train_tokens_per_second": 4649.203
    },
    {
      "epoch": 1.1198979591836735,
      "grad_norm": 0.6312075905748146,
      "learning_rate": 4.0648647582224504e-06,
      "loss": 0.23855199813842773,
      "num_input_tokens_seen": 532964288,
      "step": 8780,
      "train_runtime": 114643.7731,
      "train_tokens_per_second": 4648.873
    },
    {
      "epoch": 1.1205357142857142,
      "grad_norm": 0.6693286564819559,
      "learning_rate": 4.059944694941783e-06,
      "loss": 0.24307601451873778,
      "num_input_tokens_seen": 533257496,
      "step": 8785,
      "train_runtime": 114711.8264,
      "train_tokens_per_second": 4648.671
    },
    {
      "epoch": 1.121173469387755,
      "grad_norm": 0.6018047291772123,
      "learning_rate": 4.0550255750722674e-06,
      "loss": 0.2722148895263672,
      "num_input_tokens_seen": 533572624,
      "step": 8790,
      "train_runtime": 114778.5459,
      "train_tokens_per_second": 4648.714
    },
    {
      "epoch": 1.121811224489796,
      "grad_norm": 0.6571055183840769,
      "learning_rate": 4.050107403550582e-06,
      "loss": 0.25997745990753174,
      "num_input_tokens_seen": 533874248,
      "step": 8795,
      "train_runtime": 114844.7632,
      "train_tokens_per_second": 4648.66
    },
    {
      "epoch": 1.1224489795918366,
      "grad_norm": 0.697352468549373,
      "learning_rate": 4.0451901853124596e-06,
      "loss": 0.2595637083053589,
      "num_input_tokens_seen": 534163736,
      "step": 8800,
      "train_runtime": 114909.7614,
      "train_tokens_per_second": 4648.55
    },
    {
      "epoch": 1.1230867346938775,
      "grad_norm": 0.6641710553181707,
      "learning_rate": 4.04027392529267e-06,
      "loss": 0.24140620231628418,
      "num_input_tokens_seen": 534459272,
      "step": 8805,
      "train_runtime": 114977.4312,
      "train_tokens_per_second": 4648.384
    },
    {
      "epoch": 1.1237244897959184,
      "grad_norm": 0.6204822347961743,
      "learning_rate": 4.035358628425022e-06,
      "loss": 0.26332621574401854,
      "num_input_tokens_seen": 534773128,
      "step": 8810,
      "train_runtime": 115027.5058,
      "train_tokens_per_second": 4649.089
    },
    {
      "epoch": 1.124362244897959,
      "grad_norm": 0.6160923511236912,
      "learning_rate": 4.030444299642363e-06,
      "loss": 0.25716571807861327,
      "num_input_tokens_seen": 535081112,
      "step": 8815,
      "train_runtime": 115097.1997,
      "train_tokens_per_second": 4648.95
    },
    {
      "epoch": 1.125,
      "grad_norm": 0.5877869159249574,
      "learning_rate": 4.02553094387656e-06,
      "loss": 0.25978541374206543,
      "num_input_tokens_seen": 535374072,
      "step": 8820,
      "train_runtime": 115167.309,
      "train_tokens_per_second": 4648.664
    },
    {
      "epoch": 1.125637755102041,
      "grad_norm": 0.6586013411058614,
      "learning_rate": 4.020618566058514e-06,
      "loss": 0.2957251787185669,
      "num_input_tokens_seen": 535689208,
      "step": 8825,
      "train_runtime": 115218.2257,
      "train_tokens_per_second": 4649.344
    },
    {
      "epoch": 1.1262755102040816,
      "grad_norm": 0.6367456147035857,
      "learning_rate": 4.015707171118136e-06,
      "loss": 0.24315261840820312,
      "num_input_tokens_seen": 535993008,
      "step": 8830,
      "train_runtime": 115283.5317,
      "train_tokens_per_second": 4649.346
    },
    {
      "epoch": 1.1269132653061225,
      "grad_norm": 0.6844291597487553,
      "learning_rate": 4.010796763984355e-06,
      "loss": 0.26010923385620116,
      "num_input_tokens_seen": 536287536,
      "step": 8835,
      "train_runtime": 115349.1198,
      "train_tokens_per_second": 4649.256
    },
    {
      "epoch": 1.1275510204081634,
      "grad_norm": 0.7441505374728936,
      "learning_rate": 4.005887349585106e-06,
      "loss": 0.2572501182556152,
      "num_input_tokens_seen": 536601616,
      "step": 8840,
      "train_runtime": 115410.3241,
      "train_tokens_per_second": 4649.511
    },
    {
      "epoch": 1.128188775510204,
      "grad_norm": 0.6688566228964161,
      "learning_rate": 4.000978932847332e-06,
      "loss": 0.24611384868621827,
      "num_input_tokens_seen": 536903984,
      "step": 8845,
      "train_runtime": 115483.9981,
      "train_tokens_per_second": 4649.163
    },
    {
      "epoch": 1.128826530612245,
      "grad_norm": 0.8668765078160117,
      "learning_rate": 3.996071518696969e-06,
      "loss": 0.253182053565979,
      "num_input_tokens_seen": 537200032,
      "step": 8850,
      "train_runtime": 115549.4825,
      "train_tokens_per_second": 4649.091
    },
    {
      "epoch": 1.1294642857142858,
      "grad_norm": 0.6354563329873995,
      "learning_rate": 3.99116511205895e-06,
      "loss": 0.2547412395477295,
      "num_input_tokens_seen": 537511520,
      "step": 8855,
      "train_runtime": 115612.0733,
      "train_tokens_per_second": 4649.268
    },
    {
      "epoch": 1.1301020408163265,
      "grad_norm": 0.6073506202891805,
      "learning_rate": 3.986259717857201e-06,
      "loss": 0.25578083992004397,
      "num_input_tokens_seen": 537814664,
      "step": 8860,
      "train_runtime": 115669.384,
      "train_tokens_per_second": 4649.585
    },
    {
      "epoch": 1.1307397959183674,
      "grad_norm": 0.6024339634852962,
      "learning_rate": 3.9813553410146225e-06,
      "loss": 0.24348747730255127,
      "num_input_tokens_seen": 538115864,
      "step": 8865,
      "train_runtime": 115732.8856,
      "train_tokens_per_second": 4649.637
    },
    {
      "epoch": 1.131377551020408,
      "grad_norm": 0.5755948047363689,
      "learning_rate": 3.9764519864531026e-06,
      "loss": 0.25777478218078614,
      "num_input_tokens_seen": 538415760,
      "step": 8870,
      "train_runtime": 115806.3956,
      "train_tokens_per_second": 4649.275
    },
    {
      "epoch": 1.132015306122449,
      "grad_norm": 0.766417172131927,
      "learning_rate": 3.971549659093499e-06,
      "loss": 0.2522728681564331,
      "num_input_tokens_seen": 538716616,
      "step": 8875,
      "train_runtime": 115869.7015,
      "train_tokens_per_second": 4649.331
    },
    {
      "epoch": 1.1326530612244898,
      "grad_norm": 0.7528524832008608,
      "learning_rate": 3.966648363855637e-06,
      "loss": 0.24546959400177001,
      "num_input_tokens_seen": 539024984,
      "step": 8880,
      "train_runtime": 115938.9831,
      "train_tokens_per_second": 4649.213
    },
    {
      "epoch": 1.1332908163265305,
      "grad_norm": 0.6316767517043722,
      "learning_rate": 3.961748105658312e-06,
      "loss": 0.2723077774047852,
      "num_input_tokens_seen": 539325304,
      "step": 8885,
      "train_runtime": 116009.2966,
      "train_tokens_per_second": 4648.983
    },
    {
      "epoch": 1.1339285714285714,
      "grad_norm": 0.5980594623638233,
      "learning_rate": 3.956848889419276e-06,
      "loss": 0.25560190677642824,
      "num_input_tokens_seen": 539631960,
      "step": 8890,
      "train_runtime": 116083.711,
      "train_tokens_per_second": 4648.645
    },
    {
      "epoch": 1.1345663265306123,
      "grad_norm": 0.5682418130757153,
      "learning_rate": 3.951950720055231e-06,
      "loss": 0.24716031551361084,
      "num_input_tokens_seen": 539923576,
      "step": 8895,
      "train_runtime": 116150.1354,
      "train_tokens_per_second": 4648.497
    },
    {
      "epoch": 1.135204081632653,
      "grad_norm": 0.6739247553620884,
      "learning_rate": 3.947053602481834e-06,
      "loss": 0.24239625930786132,
      "num_input_tokens_seen": 540207120,
      "step": 8900,
      "train_runtime": 116219.8851,
      "train_tokens_per_second": 4648.147
    },
    {
      "epoch": 1.1358418367346939,
      "grad_norm": 0.6048463044809329,
      "learning_rate": 3.9421575416136866e-06,
      "loss": 0.2518806457519531,
      "num_input_tokens_seen": 540519408,
      "step": 8905,
      "train_runtime": 116287.0634,
      "train_tokens_per_second": 4648.147
    },
    {
      "epoch": 1.1364795918367347,
      "grad_norm": 0.6345051902281528,
      "learning_rate": 3.937262542364324e-06,
      "loss": 0.23644161224365234,
      "num_input_tokens_seen": 540821840,
      "step": 8910,
      "train_runtime": 116354.9095,
      "train_tokens_per_second": 4648.036
    },
    {
      "epoch": 1.1371173469387754,
      "grad_norm": 0.6535233933728755,
      "learning_rate": 3.932368609646223e-06,
      "loss": 0.24362196922302246,
      "num_input_tokens_seen": 541125600,
      "step": 8915,
      "train_runtime": 116424.3936,
      "train_tokens_per_second": 4647.871
    },
    {
      "epoch": 1.1377551020408163,
      "grad_norm": 0.6229053726587129,
      "learning_rate": 3.927475748370784e-06,
      "loss": 0.2587940216064453,
      "num_input_tokens_seen": 541430368,
      "step": 8920,
      "train_runtime": 116504.0721,
      "train_tokens_per_second": 4647.309
    },
    {
      "epoch": 1.1383928571428572,
      "grad_norm": 0.6531128073896356,
      "learning_rate": 3.922583963448336e-06,
      "loss": 0.2494271755218506,
      "num_input_tokens_seen": 541730480,
      "step": 8925,
      "train_runtime": 116573.9245,
      "train_tokens_per_second": 4647.098
    },
    {
      "epoch": 1.1390306122448979,
      "grad_norm": 0.6282383024166348,
      "learning_rate": 3.9176932597881286e-06,
      "loss": 0.24280223846435547,
      "num_input_tokens_seen": 542040136,
      "step": 8930,
      "train_runtime": 116638.2435,
      "train_tokens_per_second": 4647.19
    },
    {
      "epoch": 1.1396683673469388,
      "grad_norm": 0.5540607193383991,
      "learning_rate": 3.912803642298323e-06,
      "loss": 0.2466367721557617,
      "num_input_tokens_seen": 542342080,
      "step": 8935,
      "train_runtime": 116701.0867,
      "train_tokens_per_second": 4647.275
    },
    {
      "epoch": 1.1403061224489797,
      "grad_norm": 0.5938538544519124,
      "learning_rate": 3.90791511588599e-06,
      "loss": 0.2512343883514404,
      "num_input_tokens_seen": 542652920,
      "step": 8940,
      "train_runtime": 116778.1641,
      "train_tokens_per_second": 4646.87
    },
    {
      "epoch": 1.1409438775510203,
      "grad_norm": 0.6292093673254473,
      "learning_rate": 3.903027685457112e-06,
      "loss": 0.23715415000915527,
      "num_input_tokens_seen": 542959840,
      "step": 8945,
      "train_runtime": 116841.1911,
      "train_tokens_per_second": 4646.99
    },
    {
      "epoch": 1.1415816326530612,
      "grad_norm": 0.7342036154377143,
      "learning_rate": 3.898141355916563e-06,
      "loss": 0.26971964836120604,
      "num_input_tokens_seen": 543258480,
      "step": 8950,
      "train_runtime": 116908.5041,
      "train_tokens_per_second": 4646.869
    },
    {
      "epoch": 1.1422193877551021,
      "grad_norm": 0.6446318464742593,
      "learning_rate": 3.893256132168117e-06,
      "loss": 0.24709455966949462,
      "num_input_tokens_seen": 543565848,
      "step": 8955,
      "train_runtime": 116979.643,
      "train_tokens_per_second": 4646.67
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 0.6946970284297688,
      "learning_rate": 3.888372019114438e-06,
      "loss": 0.27740399837493895,
      "num_input_tokens_seen": 543874712,
      "step": 8960,
      "train_runtime": 117053.892,
      "train_tokens_per_second": 4646.362
    },
    {
      "epoch": 1.1434948979591837,
      "grad_norm": 0.6225922465374178,
      "learning_rate": 3.8834890216570715e-06,
      "loss": 0.24256086349487305,
      "num_input_tokens_seen": 544181184,
      "step": 8965,
      "train_runtime": 117118.3172,
      "train_tokens_per_second": 4646.422
    },
    {
      "epoch": 1.1441326530612246,
      "grad_norm": 0.5767601583617836,
      "learning_rate": 3.878607144696449e-06,
      "loss": 0.26886234283447263,
      "num_input_tokens_seen": 544499280,
      "step": 8970,
      "train_runtime": 117179.5076,
      "train_tokens_per_second": 4646.711
    },
    {
      "epoch": 1.1447704081632653,
      "grad_norm": 0.7203369868395961,
      "learning_rate": 3.873726393131877e-06,
      "loss": 0.24144067764282226,
      "num_input_tokens_seen": 544791656,
      "step": 8975,
      "train_runtime": 117244.177,
      "train_tokens_per_second": 4646.641
    },
    {
      "epoch": 1.1454081632653061,
      "grad_norm": 0.6785159121697449,
      "learning_rate": 3.868846771861528e-06,
      "loss": 0.25340542793273924,
      "num_input_tokens_seen": 545087336,
      "step": 8980,
      "train_runtime": 117314.6175,
      "train_tokens_per_second": 4646.372
    },
    {
      "epoch": 1.146045918367347,
      "grad_norm": 0.6621496294951646,
      "learning_rate": 3.86396828578244e-06,
      "loss": 0.2459954261779785,
      "num_input_tokens_seen": 545394440,
      "step": 8985,
      "train_runtime": 117381.6402,
      "train_tokens_per_second": 4646.335
    },
    {
      "epoch": 1.1466836734693877,
      "grad_norm": 0.6228571824192749,
      "learning_rate": 3.859090939790521e-06,
      "loss": 0.26397042274475097,
      "num_input_tokens_seen": 545693688,
      "step": 8990,
      "train_runtime": 117450.9288,
      "train_tokens_per_second": 4646.142
    },
    {
      "epoch": 1.1473214285714286,
      "grad_norm": 0.6434743028346319,
      "learning_rate": 3.854214738780523e-06,
      "loss": 0.25994985103607177,
      "num_input_tokens_seen": 545998160,
      "step": 8995,
      "train_runtime": 117516.2834,
      "train_tokens_per_second": 4646.149
    },
    {
      "epoch": 1.1479591836734695,
      "grad_norm": 0.6480926801760627,
      "learning_rate": 3.849339687646055e-06,
      "loss": 0.26729607582092285,
      "num_input_tokens_seen": 546317448,
      "step": 9000,
      "train_runtime": 117591.7755,
      "train_tokens_per_second": 4645.881
    },
    {
      "epoch": 1.1485969387755102,
      "grad_norm": 0.6329400342074115,
      "learning_rate": 3.844465791279575e-06,
      "loss": 0.23711986541748048,
      "num_input_tokens_seen": 546603024,
      "step": 9005,
      "train_runtime": 117658.2574,
      "train_tokens_per_second": 4645.683
    },
    {
      "epoch": 1.149234693877551,
      "grad_norm": 0.6319202274699528,
      "learning_rate": 3.839593054572374e-06,
      "loss": 0.25211658477783205,
      "num_input_tokens_seen": 546895864,
      "step": 9010,
      "train_runtime": 117726.7789,
      "train_tokens_per_second": 4645.467
    },
    {
      "epoch": 1.149872448979592,
      "grad_norm": 0.7173512648576292,
      "learning_rate": 3.834721482414586e-06,
      "loss": 0.23781025409698486,
      "num_input_tokens_seen": 547189024,
      "step": 9015,
      "train_runtime": 117793.5964,
      "train_tokens_per_second": 4645.321
    },
    {
      "epoch": 1.1505102040816326,
      "grad_norm": 0.6742994082947601,
      "learning_rate": 3.829851079695174e-06,
      "loss": 0.2751111268997192,
      "num_input_tokens_seen": 547503640,
      "step": 9020,
      "train_runtime": 117865.0062,
      "train_tokens_per_second": 4645.176
    },
    {
      "epoch": 1.1511479591836735,
      "grad_norm": 0.7062489346148081,
      "learning_rate": 3.824981851301924e-06,
      "loss": 0.2480980396270752,
      "num_input_tokens_seen": 547811248,
      "step": 9025,
      "train_runtime": 117922.3768,
      "train_tokens_per_second": 4645.524
    },
    {
      "epoch": 1.1517857142857142,
      "grad_norm": 0.6754832083900386,
      "learning_rate": 3.820113802121453e-06,
      "loss": 0.2552891731262207,
      "num_input_tokens_seen": 548116528,
      "step": 9030,
      "train_runtime": 117987.8041,
      "train_tokens_per_second": 4645.535
    },
    {
      "epoch": 1.152423469387755,
      "grad_norm": 0.6442899344507199,
      "learning_rate": 3.815246937039182e-06,
      "loss": 0.25223641395568847,
      "num_input_tokens_seen": 548427008,
      "step": 9035,
      "train_runtime": 118052.8714,
      "train_tokens_per_second": 4645.605
    },
    {
      "epoch": 1.153061224489796,
      "grad_norm": 0.6499806078846673,
      "learning_rate": 3.810381260939354e-06,
      "loss": 0.2701232433319092,
      "num_input_tokens_seen": 548739432,
      "step": 9040,
      "train_runtime": 118115.4248,
      "train_tokens_per_second": 4645.79
    },
    {
      "epoch": 1.1536989795918366,
      "grad_norm": 0.7314891698229721,
      "learning_rate": 3.8055167787050134e-06,
      "loss": 0.25761349201202394,
      "num_input_tokens_seen": 549036680,
      "step": 9045,
      "train_runtime": 118171.7805,
      "train_tokens_per_second": 4646.09
    },
    {
      "epoch": 1.1543367346938775,
      "grad_norm": 0.683669749994871,
      "learning_rate": 3.8006534952180062e-06,
      "loss": 0.296785306930542,
      "num_input_tokens_seen": 549349696,
      "step": 9050,
      "train_runtime": 118233.3656,
      "train_tokens_per_second": 4646.317
    },
    {
      "epoch": 1.1549744897959184,
      "grad_norm": 0.7175970091232914,
      "learning_rate": 3.79579141535898e-06,
      "loss": 0.26458353996276857,
      "num_input_tokens_seen": 549661760,
      "step": 9055,
      "train_runtime": 118300.6297,
      "train_tokens_per_second": 4646.313
    },
    {
      "epoch": 1.155612244897959,
      "grad_norm": 0.6390365058333171,
      "learning_rate": 3.7909305440073707e-06,
      "loss": 0.2598105907440186,
      "num_input_tokens_seen": 549966736,
      "step": 9060,
      "train_runtime": 118362.0592,
      "train_tokens_per_second": 4646.478
    },
    {
      "epoch": 1.15625,
      "grad_norm": 0.6406087564810814,
      "learning_rate": 3.786070886041401e-06,
      "loss": 0.2434762239456177,
      "num_input_tokens_seen": 550258880,
      "step": 9065,
      "train_runtime": 118440.6712,
      "train_tokens_per_second": 4645.861
    },
    {
      "epoch": 1.156887755102041,
      "grad_norm": 0.567035901087959,
      "learning_rate": 3.7812124463380766e-06,
      "loss": 0.2483445882797241,
      "num_input_tokens_seen": 550571984,
      "step": 9070,
      "train_runtime": 118514.6154,
      "train_tokens_per_second": 4645.604
    },
    {
      "epoch": 1.1575255102040816,
      "grad_norm": 0.5703637135265959,
      "learning_rate": 3.7763552297731845e-06,
      "loss": 0.25247507095336913,
      "num_input_tokens_seen": 550878272,
      "step": 9075,
      "train_runtime": 118586.3175,
      "train_tokens_per_second": 4645.378
    },
    {
      "epoch": 1.1581632653061225,
      "grad_norm": 0.6406293192824949,
      "learning_rate": 3.771499241221277e-06,
      "loss": 0.28102898597717285,
      "num_input_tokens_seen": 551189928,
      "step": 9080,
      "train_runtime": 118641.8266,
      "train_tokens_per_second": 4645.831
    },
    {
      "epoch": 1.1588010204081634,
      "grad_norm": 0.6269056057353551,
      "learning_rate": 3.766644485555679e-06,
      "loss": 0.25998654365539553,
      "num_input_tokens_seen": 551490536,
      "step": 9085,
      "train_runtime": 118713.6419,
      "train_tokens_per_second": 4645.553
    },
    {
      "epoch": 1.159438775510204,
      "grad_norm": 0.6138490858247232,
      "learning_rate": 3.7617909676484788e-06,
      "loss": 0.24906582832336427,
      "num_input_tokens_seen": 551800888,
      "step": 9090,
      "train_runtime": 118786.3654,
      "train_tokens_per_second": 4645.322
    },
    {
      "epoch": 1.160076530612245,
      "grad_norm": 0.6859516693849371,
      "learning_rate": 3.7569386923705173e-06,
      "loss": 0.25821330547332766,
      "num_input_tokens_seen": 552112808,
      "step": 9095,
      "train_runtime": 118853.748,
      "train_tokens_per_second": 4645.313
    },
    {
      "epoch": 1.1607142857142858,
      "grad_norm": 0.6240387119145479,
      "learning_rate": 3.752087664591394e-06,
      "loss": 0.23380570411682128,
      "num_input_tokens_seen": 552415496,
      "step": 9100,
      "train_runtime": 118902.4853,
      "train_tokens_per_second": 4645.954
    },
    {
      "epoch": 1.1613520408163265,
      "grad_norm": 0.5647994495841705,
      "learning_rate": 3.7472378891794537e-06,
      "loss": 0.27939319610595703,
      "num_input_tokens_seen": 552715160,
      "step": 9105,
      "train_runtime": 118976.8579,
      "train_tokens_per_second": 4645.569
    },
    {
      "epoch": 1.1619897959183674,
      "grad_norm": 0.6673699722318139,
      "learning_rate": 3.7423893710017826e-06,
      "loss": 0.2548710107803345,
      "num_input_tokens_seen": 553027504,
      "step": 9110,
      "train_runtime": 119040.4117,
      "train_tokens_per_second": 4645.712
    },
    {
      "epoch": 1.162627551020408,
      "grad_norm": 0.7397290422057428,
      "learning_rate": 3.7375421149242102e-06,
      "loss": 0.27686965465545654,
      "num_input_tokens_seen": 553319712,
      "step": 9115,
      "train_runtime": 119106.5825,
      "train_tokens_per_second": 4645.585
    },
    {
      "epoch": 1.163265306122449,
      "grad_norm": 0.6110700112790147,
      "learning_rate": 3.7326961258112963e-06,
      "loss": 0.27753410339355467,
      "num_input_tokens_seen": 553618416,
      "step": 9120,
      "train_runtime": 119166.667,
      "train_tokens_per_second": 4645.749
    },
    {
      "epoch": 1.1639030612244898,
      "grad_norm": 0.7417733503562559,
      "learning_rate": 3.7278514085263275e-06,
      "loss": 0.2638828754425049,
      "num_input_tokens_seen": 553920520,
      "step": 9125,
      "train_runtime": 119237.3321,
      "train_tokens_per_second": 4645.529
    },
    {
      "epoch": 1.1645408163265305,
      "grad_norm": 0.7167049053623508,
      "learning_rate": 3.7230079679313157e-06,
      "loss": 0.2771193742752075,
      "num_input_tokens_seen": 554225448,
      "step": 9130,
      "train_runtime": 119293.7618,
      "train_tokens_per_second": 4645.888
    },
    {
      "epoch": 1.1651785714285714,
      "grad_norm": 0.5945102517876708,
      "learning_rate": 3.7181658088869954e-06,
      "loss": 0.25243654251098635,
      "num_input_tokens_seen": 554527792,
      "step": 9135,
      "train_runtime": 119361.2962,
      "train_tokens_per_second": 4645.792
    },
    {
      "epoch": 1.1658163265306123,
      "grad_norm": 0.6982078029262607,
      "learning_rate": 3.7133249362528075e-06,
      "loss": 0.27841687202453613,
      "num_input_tokens_seen": 554827168,
      "step": 9140,
      "train_runtime": 119420.2954,
      "train_tokens_per_second": 4646.004
    },
    {
      "epoch": 1.166454081632653,
      "grad_norm": 0.5554748498950471,
      "learning_rate": 3.708485354886906e-06,
      "loss": 0.23977646827697754,
      "num_input_tokens_seen": 555129232,
      "step": 9145,
      "train_runtime": 119487.802,
      "train_tokens_per_second": 4645.907
    },
    {
      "epoch": 1.1670918367346939,
      "grad_norm": 0.6681976381838787,
      "learning_rate": 3.7036470696461535e-06,
      "loss": 0.25014777183532716,
      "num_input_tokens_seen": 555422128,
      "step": 9150,
      "train_runtime": 119557.2032,
      "train_tokens_per_second": 4645.66
    },
    {
      "epoch": 1.1677295918367347,
      "grad_norm": 0.6624244457089297,
      "learning_rate": 3.698810085386101e-06,
      "loss": 0.23124778270721436,
      "num_input_tokens_seen": 555712664,
      "step": 9155,
      "train_runtime": 119624.4585,
      "train_tokens_per_second": 4645.477
    },
    {
      "epoch": 1.1683673469387754,
      "grad_norm": 0.645178968549477,
      "learning_rate": 3.693974406961004e-06,
      "loss": 0.23919620513916015,
      "num_input_tokens_seen": 556008712,
      "step": 9160,
      "train_runtime": 119694.5366,
      "train_tokens_per_second": 4645.23
    },
    {
      "epoch": 1.1690051020408163,
      "grad_norm": 0.6807500370413369,
      "learning_rate": 3.689140039223802e-06,
      "loss": 0.25184340476989747,
      "num_input_tokens_seen": 556316288,
      "step": 9165,
      "train_runtime": 119760.5645,
      "train_tokens_per_second": 4645.238
    },
    {
      "epoch": 1.1696428571428572,
      "grad_norm": 0.7270478809642932,
      "learning_rate": 3.6843069870261196e-06,
      "loss": 0.24816541671752929,
      "num_input_tokens_seen": 556629264,
      "step": 9170,
      "train_runtime": 119823.4727,
      "train_tokens_per_second": 4645.411
    },
    {
      "epoch": 1.1702806122448979,
      "grad_norm": 0.735604362472043,
      "learning_rate": 3.6794752552182654e-06,
      "loss": 0.2609705448150635,
      "num_input_tokens_seen": 556930672,
      "step": 9175,
      "train_runtime": 119900.2521,
      "train_tokens_per_second": 4644.95
    },
    {
      "epoch": 1.1709183673469388,
      "grad_norm": 0.6239171600231348,
      "learning_rate": 3.674644848649217e-06,
      "loss": 0.2458580732345581,
      "num_input_tokens_seen": 557224256,
      "step": 9180,
      "train_runtime": 119961.9083,
      "train_tokens_per_second": 4645.01
    },
    {
      "epoch": 1.1715561224489797,
      "grad_norm": 0.6786664998580014,
      "learning_rate": 3.669815772166625e-06,
      "loss": 0.2798786163330078,
      "num_input_tokens_seen": 557531800,
      "step": 9185,
      "train_runtime": 120027.734,
      "train_tokens_per_second": 4645.025
    },
    {
      "epoch": 1.1721938775510203,
      "grad_norm": 0.6526246976544205,
      "learning_rate": 3.6649880306168073e-06,
      "loss": 0.24728355407714844,
      "num_input_tokens_seen": 557835696,
      "step": 9190,
      "train_runtime": 120094.4711,
      "train_tokens_per_second": 4644.974
    },
    {
      "epoch": 1.1728316326530612,
      "grad_norm": 0.6941040048296104,
      "learning_rate": 3.6601616288447352e-06,
      "loss": 0.24804656505584716,
      "num_input_tokens_seen": 558143680,
      "step": 9195,
      "train_runtime": 120165.5013,
      "train_tokens_per_second": 4644.791
    },
    {
      "epoch": 1.1734693877551021,
      "grad_norm": 0.7422109498183178,
      "learning_rate": 3.655336571694044e-06,
      "loss": 0.23939886093139648,
      "num_input_tokens_seen": 558435040,
      "step": 9200,
      "train_runtime": 120228.5567,
      "train_tokens_per_second": 4644.779
    },
    {
      "epoch": 1.1741071428571428,
      "grad_norm": 0.6202743505662963,
      "learning_rate": 3.6505128640070152e-06,
      "loss": 0.24841651916503907,
      "num_input_tokens_seen": 558748312,
      "step": 9205,
      "train_runtime": 120297.9674,
      "train_tokens_per_second": 4644.703
    },
    {
      "epoch": 1.1747448979591837,
      "grad_norm": 0.6412255282172917,
      "learning_rate": 3.6456905106245743e-06,
      "loss": 0.2559295654296875,
      "num_input_tokens_seen": 559058256,
      "step": 9210,
      "train_runtime": 120361.1424,
      "train_tokens_per_second": 4644.84
    },
    {
      "epoch": 1.1753826530612246,
      "grad_norm": 0.6891076562375162,
      "learning_rate": 3.6408695163862895e-06,
      "loss": 0.2709405183792114,
      "num_input_tokens_seen": 559360120,
      "step": 9215,
      "train_runtime": 120430.5611,
      "train_tokens_per_second": 4644.669
    },
    {
      "epoch": 1.1760204081632653,
      "grad_norm": 0.602780498095672,
      "learning_rate": 3.6360498861303693e-06,
      "loss": 0.2672729969024658,
      "num_input_tokens_seen": 559663176,
      "step": 9220,
      "train_runtime": 120497.372,
      "train_tokens_per_second": 4644.609
    },
    {
      "epoch": 1.1766581632653061,
      "grad_norm": 0.644008402645429,
      "learning_rate": 3.631231624693645e-06,
      "loss": 0.26173014640808107,
      "num_input_tokens_seen": 559969824,
      "step": 9225,
      "train_runtime": 120557.0325,
      "train_tokens_per_second": 4644.854
    },
    {
      "epoch": 1.177295918367347,
      "grad_norm": 0.6350871203103394,
      "learning_rate": 3.6264147369115788e-06,
      "loss": 0.2579504013061523,
      "num_input_tokens_seen": 560272344,
      "step": 9230,
      "train_runtime": 120627.1844,
      "train_tokens_per_second": 4644.661
    },
    {
      "epoch": 1.1779336734693877,
      "grad_norm": 0.6485451973793857,
      "learning_rate": 3.621599227618258e-06,
      "loss": 0.24934630393981932,
      "num_input_tokens_seen": 560567064,
      "step": 9235,
      "train_runtime": 120698.6906,
      "train_tokens_per_second": 4644.351
    },
    {
      "epoch": 1.1785714285714286,
      "grad_norm": 0.6544192643141078,
      "learning_rate": 3.6167851016463762e-06,
      "loss": 0.2503244161605835,
      "num_input_tokens_seen": 560860256,
      "step": 9240,
      "train_runtime": 120759.48,
      "train_tokens_per_second": 4644.441
    },
    {
      "epoch": 1.1792091836734695,
      "grad_norm": 0.6241115845616516,
      "learning_rate": 3.61197236382725e-06,
      "loss": 0.24509780406951903,
      "num_input_tokens_seen": 561172000,
      "step": 9245,
      "train_runtime": 120818.411,
      "train_tokens_per_second": 4644.756
    },
    {
      "epoch": 1.1798469387755102,
      "grad_norm": 0.6458798203659332,
      "learning_rate": 3.6071610189907975e-06,
      "loss": 0.26144001483917234,
      "num_input_tokens_seen": 561472240,
      "step": 9250,
      "train_runtime": 120880.7071,
      "train_tokens_per_second": 4644.846
    },
    {
      "epoch": 1.180484693877551,
      "grad_norm": 0.7538210148514893,
      "learning_rate": 3.602351071965536e-06,
      "loss": 0.2761153936386108,
      "num_input_tokens_seen": 561781768,
      "step": 9255,
      "train_runtime": 120946.7927,
      "train_tokens_per_second": 4644.867
    },
    {
      "epoch": 1.181122448979592,
      "grad_norm": 0.669378573702209,
      "learning_rate": 3.597542527578586e-06,
      "loss": 0.25878252983093264,
      "num_input_tokens_seen": 562087760,
      "step": 9260,
      "train_runtime": 121006.0154,
      "train_tokens_per_second": 4645.122
    },
    {
      "epoch": 1.1817602040816326,
      "grad_norm": 0.6577261689258893,
      "learning_rate": 3.5927353906556583e-06,
      "loss": 0.282159686088562,
      "num_input_tokens_seen": 562399704,
      "step": 9265,
      "train_runtime": 121068.6595,
      "train_tokens_per_second": 4645.296
    },
    {
      "epoch": 1.1823979591836735,
      "grad_norm": 0.615019101145986,
      "learning_rate": 3.587929666021048e-06,
      "loss": 0.23938889503479005,
      "num_input_tokens_seen": 562710120,
      "step": 9270,
      "train_runtime": 121137.3838,
      "train_tokens_per_second": 4645.223
    },
    {
      "epoch": 1.1830357142857142,
      "grad_norm": 0.6166695210589391,
      "learning_rate": 3.5831253584976377e-06,
      "loss": 0.24919326305389405,
      "num_input_tokens_seen": 563009640,
      "step": 9275,
      "train_runtime": 121203.932,
      "train_tokens_per_second": 4645.143
    },
    {
      "epoch": 1.183673469387755,
      "grad_norm": 0.6383444618754819,
      "learning_rate": 3.578322472906882e-06,
      "loss": 0.25875191688537597,
      "num_input_tokens_seen": 563313456,
      "step": 9280,
      "train_runtime": 121267.3525,
      "train_tokens_per_second": 4645.219
    },
    {
      "epoch": 1.184311224489796,
      "grad_norm": 0.5956095629968433,
      "learning_rate": 3.5735210140688163e-06,
      "loss": 0.2727609872817993,
      "num_input_tokens_seen": 563624904,
      "step": 9285,
      "train_runtime": 121334.2716,
      "train_tokens_per_second": 4645.224
    },
    {
      "epoch": 1.1849489795918366,
      "grad_norm": 0.6636736374676456,
      "learning_rate": 3.568720986802039e-06,
      "loss": 0.2516184329986572,
      "num_input_tokens_seen": 563914608,
      "step": 9290,
      "train_runtime": 121404.3413,
      "train_tokens_per_second": 4644.93
    },
    {
      "epoch": 1.1855867346938775,
      "grad_norm": 0.6929999269615047,
      "learning_rate": 3.56392239592371e-06,
      "loss": 0.24294590950012207,
      "num_input_tokens_seen": 564227864,
      "step": 9295,
      "train_runtime": 121457.275,
      "train_tokens_per_second": 4645.484
    },
    {
      "epoch": 1.1862244897959184,
      "grad_norm": 0.6008541880041068,
      "learning_rate": 3.559125246249552e-06,
      "loss": 0.23285303115844727,
      "num_input_tokens_seen": 564530856,
      "step": 9300,
      "train_runtime": 121523.8302,
      "train_tokens_per_second": 4645.433
    },
    {
      "epoch": 1.186862244897959,
      "grad_norm": 0.6051562898635172,
      "learning_rate": 3.554329542593841e-06,
      "loss": 0.24616332054138185,
      "num_input_tokens_seen": 564840936,
      "step": 9305,
      "train_runtime": 121590.8913,
      "train_tokens_per_second": 4645.421
    },
    {
      "epoch": 1.1875,
      "grad_norm": 0.6042571130671134,
      "learning_rate": 3.549535289769399e-06,
      "loss": 0.2721709728240967,
      "num_input_tokens_seen": 565145832,
      "step": 9310,
      "train_runtime": 121646.9229,
      "train_tokens_per_second": 4645.788
    },
    {
      "epoch": 1.188137755102041,
      "grad_norm": 0.5940346288462378,
      "learning_rate": 3.544742492587593e-06,
      "loss": 0.25292162895202636,
      "num_input_tokens_seen": 565450120,
      "step": 9315,
      "train_runtime": 121718.5439,
      "train_tokens_per_second": 4645.554
    },
    {
      "epoch": 1.1887755102040816,
      "grad_norm": 0.6234099120937138,
      "learning_rate": 3.5399511558583323e-06,
      "loss": 0.24716696739196778,
      "num_input_tokens_seen": 565760960,
      "step": 9320,
      "train_runtime": 121778.0047,
      "train_tokens_per_second": 4645.839
    },
    {
      "epoch": 1.1894132653061225,
      "grad_norm": 0.6822500842747656,
      "learning_rate": 3.5351612843900555e-06,
      "loss": 0.27025156021118163,
      "num_input_tokens_seen": 566059608,
      "step": 9325,
      "train_runtime": 121838.3169,
      "train_tokens_per_second": 4645.99
    },
    {
      "epoch": 1.1900510204081634,
      "grad_norm": 0.618687190311368,
      "learning_rate": 3.5303728829897336e-06,
      "loss": 0.24979643821716307,
      "num_input_tokens_seen": 566349128,
      "step": 9330,
      "train_runtime": 121907.8723,
      "train_tokens_per_second": 4645.714
    },
    {
      "epoch": 1.190688775510204,
      "grad_norm": 0.7218385310383179,
      "learning_rate": 3.5255859564628637e-06,
      "loss": 0.2737593650817871,
      "num_input_tokens_seen": 566657304,
      "step": 9335,
      "train_runtime": 121976.982,
      "train_tokens_per_second": 4645.608
    },
    {
      "epoch": 1.191326530612245,
      "grad_norm": 0.6787468317957424,
      "learning_rate": 3.5208005096134567e-06,
      "loss": 0.27191920280456544,
      "num_input_tokens_seen": 566961464,
      "step": 9340,
      "train_runtime": 122041.4244,
      "train_tokens_per_second": 4645.648
    },
    {
      "epoch": 1.1919642857142858,
      "grad_norm": 0.719032768526688,
      "learning_rate": 3.516016547244047e-06,
      "loss": 0.281893253326416,
      "num_input_tokens_seen": 567275376,
      "step": 9345,
      "train_runtime": 122106.7763,
      "train_tokens_per_second": 4645.732
    },
    {
      "epoch": 1.1926020408163265,
      "grad_norm": 0.6239259777646281,
      "learning_rate": 3.5112340741556737e-06,
      "loss": 0.23227195739746093,
      "num_input_tokens_seen": 567570008,
      "step": 9350,
      "train_runtime": 122176.3341,
      "train_tokens_per_second": 4645.499
    },
    {
      "epoch": 1.1932397959183674,
      "grad_norm": 0.6608478022282359,
      "learning_rate": 3.5064530951478822e-06,
      "loss": 0.25134925842285155,
      "num_input_tokens_seen": 567862344,
      "step": 9355,
      "train_runtime": 122239.5036,
      "train_tokens_per_second": 4645.49
    },
    {
      "epoch": 1.193877551020408,
      "grad_norm": 0.6791779666007302,
      "learning_rate": 3.501673615018717e-06,
      "loss": 0.257344913482666,
      "num_input_tokens_seen": 568165880,
      "step": 9360,
      "train_runtime": 122316.6048,
      "train_tokens_per_second": 4645.043
    },
    {
      "epoch": 1.194515306122449,
      "grad_norm": 0.7097955397102591,
      "learning_rate": 3.496895638564724e-06,
      "loss": 0.2674292802810669,
      "num_input_tokens_seen": 568478840,
      "step": 9365,
      "train_runtime": 122364.2219,
      "train_tokens_per_second": 4645.793
    },
    {
      "epoch": 1.1951530612244898,
      "grad_norm": 0.5477342861602996,
      "learning_rate": 3.492119170580933e-06,
      "loss": 0.26298508644104,
      "num_input_tokens_seen": 568791504,
      "step": 9370,
      "train_runtime": 122426.5837,
      "train_tokens_per_second": 4645.98
    },
    {
      "epoch": 1.1957908163265305,
      "grad_norm": 0.5994955894784763,
      "learning_rate": 3.4873442158608638e-06,
      "loss": 0.23578338623046874,
      "num_input_tokens_seen": 569086712,
      "step": 9375,
      "train_runtime": 122496.8953,
      "train_tokens_per_second": 4645.724
    },
    {
      "epoch": 1.1964285714285714,
      "grad_norm": 0.6632625907484253,
      "learning_rate": 3.4825707791965213e-06,
      "loss": 0.24221220016479492,
      "num_input_tokens_seen": 569390656,
      "step": 9380,
      "train_runtime": 122568.9851,
      "train_tokens_per_second": 4645.471
    },
    {
      "epoch": 1.1970663265306123,
      "grad_norm": 0.6133307012409632,
      "learning_rate": 3.477798865378375e-06,
      "loss": 0.25411770343780515,
      "num_input_tokens_seen": 569697440,
      "step": 9385,
      "train_runtime": 122630.5745,
      "train_tokens_per_second": 4645.639
    },
    {
      "epoch": 1.197704081632653,
      "grad_norm": 0.6641846049590382,
      "learning_rate": 3.4730284791953793e-06,
      "loss": 0.2513757228851318,
      "num_input_tokens_seen": 570015832,
      "step": 9390,
      "train_runtime": 122700.3538,
      "train_tokens_per_second": 4645.592
    },
    {
      "epoch": 1.1983418367346939,
      "grad_norm": 0.6300890210144545,
      "learning_rate": 3.4682596254349487e-06,
      "loss": 0.2546231746673584,
      "num_input_tokens_seen": 570318520,
      "step": 9395,
      "train_runtime": 122769.3257,
      "train_tokens_per_second": 4645.448
    },
    {
      "epoch": 1.1989795918367347,
      "grad_norm": 0.6797534089242993,
      "learning_rate": 3.4634923088829586e-06,
      "loss": 0.2677803039550781,
      "num_input_tokens_seen": 570618864,
      "step": 9400,
      "train_runtime": 122836.4089,
      "train_tokens_per_second": 4645.356
    },
    {
      "epoch": 1.1996173469387754,
      "grad_norm": 0.6052770201065073,
      "learning_rate": 3.4587265343237474e-06,
      "loss": 0.2581784725189209,
      "num_input_tokens_seen": 570930248,
      "step": 9405,
      "train_runtime": 122899.9562,
      "train_tokens_per_second": 4645.488
    },
    {
      "epoch": 1.2002551020408163,
      "grad_norm": 0.605556935207266,
      "learning_rate": 3.4539623065401014e-06,
      "loss": 0.26486501693725584,
      "num_input_tokens_seen": 571244312,
      "step": 9410,
      "train_runtime": 122964.309,
      "train_tokens_per_second": 4645.611
    },
    {
      "epoch": 1.2008928571428572,
      "grad_norm": 0.5901276245521979,
      "learning_rate": 3.4491996303132553e-06,
      "loss": 0.26423025131225586,
      "num_input_tokens_seen": 571546144,
      "step": 9415,
      "train_runtime": 123029.5844,
      "train_tokens_per_second": 4645.599
    },
    {
      "epoch": 1.2015306122448979,
      "grad_norm": 0.8140654353032616,
      "learning_rate": 3.444438510422889e-06,
      "loss": 0.26038737297058107,
      "num_input_tokens_seen": 571850512,
      "step": 9420,
      "train_runtime": 123100.4514,
      "train_tokens_per_second": 4645.397
    },
    {
      "epoch": 1.2021683673469388,
      "grad_norm": 0.6415253190625197,
      "learning_rate": 3.4396789516471152e-06,
      "loss": 0.25277788639068605,
      "num_input_tokens_seen": 572162672,
      "step": 9425,
      "train_runtime": 123172.7106,
      "train_tokens_per_second": 4645.206
    },
    {
      "epoch": 1.2028061224489797,
      "grad_norm": 0.6498992505212711,
      "learning_rate": 3.4349209587624876e-06,
      "loss": 0.26095411777496336,
      "num_input_tokens_seen": 572478112,
      "step": 9430,
      "train_runtime": 123227.8961,
      "train_tokens_per_second": 4645.686
    },
    {
      "epoch": 1.2034438775510203,
      "grad_norm": 0.6384547583756184,
      "learning_rate": 3.4301645365439827e-06,
      "loss": 0.2607147216796875,
      "num_input_tokens_seen": 572768984,
      "step": 9435,
      "train_runtime": 123292.0608,
      "train_tokens_per_second": 4645.627
    },
    {
      "epoch": 1.2040816326530612,
      "grad_norm": 0.5492421097654189,
      "learning_rate": 3.4254096897650026e-06,
      "loss": 0.24343760013580323,
      "num_input_tokens_seen": 573065496,
      "step": 9440,
      "train_runtime": 123361.9626,
      "train_tokens_per_second": 4645.399
    },
    {
      "epoch": 1.2047193877551021,
      "grad_norm": 0.7492384906913376,
      "learning_rate": 3.4206564231973664e-06,
      "loss": 0.2628458499908447,
      "num_input_tokens_seen": 573373352,
      "step": 9445,
      "train_runtime": 123425.8893,
      "train_tokens_per_second": 4645.487
    },
    {
      "epoch": 1.2053571428571428,
      "grad_norm": 0.6848727653628047,
      "learning_rate": 3.4159047416113133e-06,
      "loss": 0.2423173427581787,
      "num_input_tokens_seen": 573665624,
      "step": 9450,
      "train_runtime": 123490.8524,
      "train_tokens_per_second": 4645.41
    },
    {
      "epoch": 1.2059948979591837,
      "grad_norm": 0.6151650299218879,
      "learning_rate": 3.4111546497754834e-06,
      "loss": 0.2351745843887329,
      "num_input_tokens_seen": 573958232,
      "step": 9455,
      "train_runtime": 123558.1197,
      "train_tokens_per_second": 4645.249
    },
    {
      "epoch": 1.2066326530612246,
      "grad_norm": 0.643481328726907,
      "learning_rate": 3.406406152456926e-06,
      "loss": 0.27458600997924804,
      "num_input_tokens_seen": 574272616,
      "step": 9460,
      "train_runtime": 123617.9173,
      "train_tokens_per_second": 4645.545
    },
    {
      "epoch": 1.2072704081632653,
      "grad_norm": 0.6546095475979791,
      "learning_rate": 3.4016592544210937e-06,
      "loss": 0.25125956535339355,
      "num_input_tokens_seen": 574572872,
      "step": 9465,
      "train_runtime": 123679.1006,
      "train_tokens_per_second": 4645.675
    },
    {
      "epoch": 1.2079081632653061,
      "grad_norm": 0.645162808449116,
      "learning_rate": 3.3969139604318257e-06,
      "loss": 0.27151367664337156,
      "num_input_tokens_seen": 574889856,
      "step": 9470,
      "train_runtime": 123747.1769,
      "train_tokens_per_second": 4645.681
    },
    {
      "epoch": 1.208545918367347,
      "grad_norm": 0.718029624025104,
      "learning_rate": 3.3921702752513576e-06,
      "loss": 0.266917085647583,
      "num_input_tokens_seen": 575194832,
      "step": 9475,
      "train_runtime": 123819.5178,
      "train_tokens_per_second": 4645.429
    },
    {
      "epoch": 1.2091836734693877,
      "grad_norm": 0.6053928629417192,
      "learning_rate": 3.3874282036403106e-06,
      "loss": 0.261753511428833,
      "num_input_tokens_seen": 575499512,
      "step": 9480,
      "train_runtime": 123880.1935,
      "train_tokens_per_second": 4645.614
    },
    {
      "epoch": 1.2098214285714286,
      "grad_norm": 0.6720324410157728,
      "learning_rate": 3.3826877503576805e-06,
      "loss": 0.24433953762054444,
      "num_input_tokens_seen": 575788032,
      "step": 9485,
      "train_runtime": 123953.4641,
      "train_tokens_per_second": 4645.195
    },
    {
      "epoch": 1.2104591836734695,
      "grad_norm": 0.7291148842700406,
      "learning_rate": 3.3779489201608468e-06,
      "loss": 0.2573875427246094,
      "num_input_tokens_seen": 576091944,
      "step": 9490,
      "train_runtime": 124027.906,
      "train_tokens_per_second": 4644.857
    },
    {
      "epoch": 1.2110969387755102,
      "grad_norm": 0.7062895668021288,
      "learning_rate": 3.373211717805557e-06,
      "loss": 0.23786654472351074,
      "num_input_tokens_seen": 576380232,
      "step": 9495,
      "train_runtime": 124092.8276,
      "train_tokens_per_second": 4644.751
    },
    {
      "epoch": 1.211734693877551,
      "grad_norm": 0.7542692179269392,
      "learning_rate": 3.3684761480459218e-06,
      "loss": 0.2816202640533447,
      "num_input_tokens_seen": 576690824,
      "step": 9500,
      "train_runtime": 124158.2954,
      "train_tokens_per_second": 4644.803
    },
    {
      "epoch": 1.212372448979592,
      "grad_norm": 0.6082962015511804,
      "learning_rate": 3.363742215634416e-06,
      "loss": 0.2572930335998535,
      "num_input_tokens_seen": 577004104,
      "step": 9505,
      "train_runtime": 124209.677,
      "train_tokens_per_second": 4645.404
    },
    {
      "epoch": 1.2130102040816326,
      "grad_norm": 0.6063825112611065,
      "learning_rate": 3.3590099253218745e-06,
      "loss": 0.2601757526397705,
      "num_input_tokens_seen": 577309832,
      "step": 9510,
      "train_runtime": 124279.3024,
      "train_tokens_per_second": 4645.261
    },
    {
      "epoch": 1.2136479591836735,
      "grad_norm": 0.7259437327933773,
      "learning_rate": 3.3542792818574793e-06,
      "loss": 0.26034154891967776,
      "num_input_tokens_seen": 577625664,
      "step": 9515,
      "train_runtime": 124342.6287,
      "train_tokens_per_second": 4645.436
    },
    {
      "epoch": 1.2142857142857142,
      "grad_norm": 0.7434337151323354,
      "learning_rate": 3.3495502899887624e-06,
      "loss": 0.24712867736816407,
      "num_input_tokens_seen": 577925920,
      "step": 9520,
      "train_runtime": 124401.1151,
      "train_tokens_per_second": 4645.665
    },
    {
      "epoch": 1.214923469387755,
      "grad_norm": 0.6421712597067616,
      "learning_rate": 3.344822954461595e-06,
      "loss": 0.2680546760559082,
      "num_input_tokens_seen": 578241024,
      "step": 9525,
      "train_runtime": 124450.2257,
      "train_tokens_per_second": 4646.364
    },
    {
      "epoch": 1.215561224489796,
      "grad_norm": 0.6202226500554692,
      "learning_rate": 3.34009728002019e-06,
      "loss": 0.26010828018188475,
      "num_input_tokens_seen": 578558576,
      "step": 9530,
      "train_runtime": 124496.961,
      "train_tokens_per_second": 4647.17
    },
    {
      "epoch": 1.2161989795918366,
      "grad_norm": 0.6090945321495278,
      "learning_rate": 3.3353732714070933e-06,
      "loss": 0.26022019386291506,
      "num_input_tokens_seen": 578861152,
      "step": 9535,
      "train_runtime": 124565.817,
      "train_tokens_per_second": 4647.031
    },
    {
      "epoch": 1.2168367346938775,
      "grad_norm": 0.7723797409337256,
      "learning_rate": 3.3306509333631755e-06,
      "loss": 0.26360206604003905,
      "num_input_tokens_seen": 579170880,
      "step": 9540,
      "train_runtime": 124628.3368,
      "train_tokens_per_second": 4647.185
    },
    {
      "epoch": 1.2174744897959184,
      "grad_norm": 0.6023917274244316,
      "learning_rate": 3.325930270627632e-06,
      "loss": 0.24070911407470702,
      "num_input_tokens_seen": 579462640,
      "step": 9545,
      "train_runtime": 124701.877,
      "train_tokens_per_second": 4646.784
    },
    {
      "epoch": 1.218112244897959,
      "grad_norm": 0.6116123405130297,
      "learning_rate": 3.3212112879379806e-06,
      "loss": 0.2483356475830078,
      "num_input_tokens_seen": 579752112,
      "step": 9550,
      "train_runtime": 124764.4211,
      "train_tokens_per_second": 4646.774
    },
    {
      "epoch": 1.21875,
      "grad_norm": 0.6800150427630616,
      "learning_rate": 3.3164939900300467e-06,
      "loss": 0.2761533737182617,
      "num_input_tokens_seen": 580059640,
      "step": 9555,
      "train_runtime": 124821.1157,
      "train_tokens_per_second": 4647.128
    },
    {
      "epoch": 1.219387755102041,
      "grad_norm": 0.6919701122421402,
      "learning_rate": 3.31177838163797e-06,
      "loss": 0.26920223236083984,
      "num_input_tokens_seen": 580366720,
      "step": 9560,
      "train_runtime": 124885.381,
      "train_tokens_per_second": 4647.195
    },
    {
      "epoch": 1.2200255102040816,
      "grad_norm": 0.6820769296505718,
      "learning_rate": 3.3070644674941938e-06,
      "loss": 0.23745627403259278,
      "num_input_tokens_seen": 580669264,
      "step": 9565,
      "train_runtime": 124952.154,
      "train_tokens_per_second": 4647.133
    },
    {
      "epoch": 1.2206632653061225,
      "grad_norm": 0.6766487773518557,
      "learning_rate": 3.302352252329456e-06,
      "loss": 0.26337151527404784,
      "num_input_tokens_seen": 580976232,
      "step": 9570,
      "train_runtime": 125027.1775,
      "train_tokens_per_second": 4646.8
    },
    {
      "epoch": 1.2213010204081634,
      "grad_norm": 0.5956951042470227,
      "learning_rate": 3.2976417408727978e-06,
      "loss": 0.273178243637085,
      "num_input_tokens_seen": 581296456,
      "step": 9575,
      "train_runtime": 125087.8872,
      "train_tokens_per_second": 4647.104
    },
    {
      "epoch": 1.221938775510204,
      "grad_norm": 0.6402258920372919,
      "learning_rate": 3.2929329378515466e-06,
      "loss": 0.261471152305603,
      "num_input_tokens_seen": 581614240,
      "step": 9580,
      "train_runtime": 125147.3849,
      "train_tokens_per_second": 4647.434
    },
    {
      "epoch": 1.222576530612245,
      "grad_norm": 0.7057805386972357,
      "learning_rate": 3.288225847991312e-06,
      "loss": 0.26575982570648193,
      "num_input_tokens_seen": 581908472,
      "step": 9585,
      "train_runtime": 125222.18,
      "train_tokens_per_second": 4647.008
    },
    {
      "epoch": 1.2232142857142858,
      "grad_norm": 0.6614193883668372,
      "learning_rate": 3.2835204760159883e-06,
      "loss": 0.24958672523498535,
      "num_input_tokens_seen": 582199800,
      "step": 9590,
      "train_runtime": 125289.5903,
      "train_tokens_per_second": 4646.833
    },
    {
      "epoch": 1.2238520408163265,
      "grad_norm": 0.7693666183824296,
      "learning_rate": 3.278816826647747e-06,
      "loss": 0.24088971614837645,
      "num_input_tokens_seen": 582494648,
      "step": 9595,
      "train_runtime": 125346.1978,
      "train_tokens_per_second": 4647.087
    },
    {
      "epoch": 1.2244897959183674,
      "grad_norm": 0.7672272633233124,
      "learning_rate": 3.2741149046070274e-06,
      "loss": 0.27124848365783694,
      "num_input_tokens_seen": 582806568,
      "step": 9600,
      "train_runtime": 125395.4622,
      "train_tokens_per_second": 4647.748
    },
    {
      "epoch": 1.225127551020408,
      "grad_norm": 0.6894928761161964,
      "learning_rate": 3.269414714612534e-06,
      "loss": 0.2765587329864502,
      "num_input_tokens_seen": 583109984,
      "step": 9605,
      "train_runtime": 125464.3099,
      "train_tokens_per_second": 4647.616
    },
    {
      "epoch": 1.225765306122449,
      "grad_norm": 0.8045788710717332,
      "learning_rate": 3.2647162613812435e-06,
      "loss": 0.25553967952728274,
      "num_input_tokens_seen": 583420288,
      "step": 9610,
      "train_runtime": 125520.7611,
      "train_tokens_per_second": 4647.998
    },
    {
      "epoch": 1.2264030612244898,
      "grad_norm": 0.7071738400477622,
      "learning_rate": 3.2600195496283716e-06,
      "loss": 0.2552331447601318,
      "num_input_tokens_seen": 583728232,
      "step": 9615,
      "train_runtime": 125588.4357,
      "train_tokens_per_second": 4647.946
    },
    {
      "epoch": 1.2270408163265305,
      "grad_norm": 0.6120604863160233,
      "learning_rate": 3.2553245840674037e-06,
      "loss": 0.2471709966659546,
      "num_input_tokens_seen": 584022528,
      "step": 9620,
      "train_runtime": 125658.4643,
      "train_tokens_per_second": 4647.697
    },
    {
      "epoch": 1.2276785714285714,
      "grad_norm": 0.6604253917564401,
      "learning_rate": 3.2506313694100643e-06,
      "loss": 0.26244239807128905,
      "num_input_tokens_seen": 584327616,
      "step": 9625,
      "train_runtime": 125718.9925,
      "train_tokens_per_second": 4647.887
    },
    {
      "epoch": 1.2283163265306123,
      "grad_norm": 0.6232415973266351,
      "learning_rate": 3.245939910366319e-06,
      "loss": 0.2675912618637085,
      "num_input_tokens_seen": 584631752,
      "step": 9630,
      "train_runtime": 125772.9344,
      "train_tokens_per_second": 4648.311
    },
    {
      "epoch": 1.228954081632653,
      "grad_norm": 0.6437369355610519,
      "learning_rate": 3.2412502116443783e-06,
      "loss": 0.25380563735961914,
      "num_input_tokens_seen": 584940920,
      "step": 9635,
      "train_runtime": 125844.4155,
      "train_tokens_per_second": 4648.128
    },
    {
      "epoch": 1.2295918367346939,
      "grad_norm": 0.6309497737441148,
      "learning_rate": 3.236562277950682e-06,
      "loss": 0.2619779586791992,
      "num_input_tokens_seen": 585235728,
      "step": 9640,
      "train_runtime": 125910.2575,
      "train_tokens_per_second": 4648.039
    },
    {
      "epoch": 1.2302295918367347,
      "grad_norm": 0.6611933619652276,
      "learning_rate": 3.231876113989897e-06,
      "loss": 0.2630010604858398,
      "num_input_tokens_seen": 585544368,
      "step": 9645,
      "train_runtime": 125975.5673,
      "train_tokens_per_second": 4648.079
    },
    {
      "epoch": 1.2308673469387754,
      "grad_norm": 0.6810986272814734,
      "learning_rate": 3.227191724464918e-06,
      "loss": 0.24690713882446289,
      "num_input_tokens_seen": 585837104,
      "step": 9650,
      "train_runtime": 126037.5642,
      "train_tokens_per_second": 4648.115
    },
    {
      "epoch": 1.2315051020408163,
      "grad_norm": 0.6961524232829194,
      "learning_rate": 3.2225091140768537e-06,
      "loss": 0.23948249816894532,
      "num_input_tokens_seen": 586144912,
      "step": 9655,
      "train_runtime": 126104.616,
      "train_tokens_per_second": 4648.085
    },
    {
      "epoch": 1.2321428571428572,
      "grad_norm": 0.650424688729444,
      "learning_rate": 3.2178282875250356e-06,
      "loss": 0.243941068649292,
      "num_input_tokens_seen": 586457568,
      "step": 9660,
      "train_runtime": 126173.5276,
      "train_tokens_per_second": 4648.024
    },
    {
      "epoch": 1.2327806122448979,
      "grad_norm": 0.5876253311521507,
      "learning_rate": 3.213149249506997e-06,
      "loss": 0.26530961990356444,
      "num_input_tokens_seen": 586767512,
      "step": 9665,
      "train_runtime": 126249.395,
      "train_tokens_per_second": 4647.686
    },
    {
      "epoch": 1.2334183673469388,
      "grad_norm": 0.673696068504565,
      "learning_rate": 3.208472004718479e-06,
      "loss": 0.2521230697631836,
      "num_input_tokens_seen": 587068624,
      "step": 9670,
      "train_runtime": 126306.4449,
      "train_tokens_per_second": 4647.97
    },
    {
      "epoch": 1.2340561224489797,
      "grad_norm": 0.6279495619779293,
      "learning_rate": 3.2037965578534245e-06,
      "loss": 0.27018778324127196,
      "num_input_tokens_seen": 587380544,
      "step": 9675,
      "train_runtime": 126380.0099,
      "train_tokens_per_second": 4647.733
    },
    {
      "epoch": 1.2346938775510203,
      "grad_norm": 0.7179734053191383,
      "learning_rate": 3.199122913603972e-06,
      "loss": 0.24798645973205566,
      "num_input_tokens_seen": 587682416,
      "step": 9680,
      "train_runtime": 126449.5054,
      "train_tokens_per_second": 4647.566
    },
    {
      "epoch": 1.2353316326530612,
      "grad_norm": 0.6291555011100273,
      "learning_rate": 3.1944510766604473e-06,
      "loss": 0.253771448135376,
      "num_input_tokens_seen": 587994000,
      "step": 9685,
      "train_runtime": 126521.3881,
      "train_tokens_per_second": 4647.388
    },
    {
      "epoch": 1.2359693877551021,
      "grad_norm": 0.706946272636358,
      "learning_rate": 3.1897810517113657e-06,
      "loss": 0.2575406551361084,
      "num_input_tokens_seen": 588303784,
      "step": 9690,
      "train_runtime": 126569.4031,
      "train_tokens_per_second": 4648.073
    },
    {
      "epoch": 1.2366071428571428,
      "grad_norm": 0.7003097570759911,
      "learning_rate": 3.1851128434434263e-06,
      "loss": 0.2514672756195068,
      "num_input_tokens_seen": 588603984,
      "step": 9695,
      "train_runtime": 126628.0385,
      "train_tokens_per_second": 4648.291
    },
    {
      "epoch": 1.2372448979591837,
      "grad_norm": 0.6753710085012944,
      "learning_rate": 3.1804464565414993e-06,
      "loss": 0.26751532554626467,
      "num_input_tokens_seen": 588907392,
      "step": 9700,
      "train_runtime": 126689.8644,
      "train_tokens_per_second": 4648.418
    },
    {
      "epoch": 1.2378826530612246,
      "grad_norm": 0.6405753566997168,
      "learning_rate": 3.17578189568863e-06,
      "loss": 0.2393946886062622,
      "num_input_tokens_seen": 589200744,
      "step": 9705,
      "train_runtime": 126744.5972,
      "train_tokens_per_second": 4648.725
    },
    {
      "epoch": 1.2385204081632653,
      "grad_norm": 0.6168794223813908,
      "learning_rate": 3.1711191655660327e-06,
      "loss": 0.256895112991333,
      "num_input_tokens_seen": 589514904,
      "step": 9710,
      "train_runtime": 126815.9873,
      "train_tokens_per_second": 4648.585
    },
    {
      "epoch": 1.2391581632653061,
      "grad_norm": 0.7566616034104741,
      "learning_rate": 3.1664582708530796e-06,
      "loss": 0.2531773805618286,
      "num_input_tokens_seen": 589802968,
      "step": 9715,
      "train_runtime": 126881.4978,
      "train_tokens_per_second": 4648.455
    },
    {
      "epoch": 1.239795918367347,
      "grad_norm": 0.6371920115656248,
      "learning_rate": 3.161799216227307e-06,
      "loss": 0.2313767671585083,
      "num_input_tokens_seen": 590100256,
      "step": 9720,
      "train_runtime": 126950.3212,
      "train_tokens_per_second": 4648.277
    },
    {
      "epoch": 1.2404336734693877,
      "grad_norm": 0.6850969299624414,
      "learning_rate": 3.157142006364401e-06,
      "loss": 0.256287145614624,
      "num_input_tokens_seen": 590398160,
      "step": 9725,
      "train_runtime": 127015.3617,
      "train_tokens_per_second": 4648.242
    },
    {
      "epoch": 1.2410714285714286,
      "grad_norm": 0.6707570764989049,
      "learning_rate": 3.1524866459381966e-06,
      "loss": 0.25388460159301757,
      "num_input_tokens_seen": 590694312,
      "step": 9730,
      "train_runtime": 127083.5671,
      "train_tokens_per_second": 4648.078
    },
    {
      "epoch": 1.2417091836734695,
      "grad_norm": 0.6804430561708349,
      "learning_rate": 3.147833139620671e-06,
      "loss": 0.25645253658294676,
      "num_input_tokens_seen": 590994960,
      "step": 9735,
      "train_runtime": 127140.3685,
      "train_tokens_per_second": 4648.366
    },
    {
      "epoch": 1.2423469387755102,
      "grad_norm": 0.6681750437913,
      "learning_rate": 3.143181492081947e-06,
      "loss": 0.26023178100585936,
      "num_input_tokens_seen": 591304168,
      "step": 9740,
      "train_runtime": 127188.4334,
      "train_tokens_per_second": 4649.04
    },
    {
      "epoch": 1.242984693877551,
      "grad_norm": 0.557440274643686,
      "learning_rate": 3.1385317079902743e-06,
      "loss": 0.26649203300476076,
      "num_input_tokens_seen": 591618200,
      "step": 9745,
      "train_runtime": 127245.3409,
      "train_tokens_per_second": 4649.429
    },
    {
      "epoch": 1.243622448979592,
      "grad_norm": 0.6415297135469716,
      "learning_rate": 3.133883792012037e-06,
      "loss": 0.24660344123840333,
      "num_input_tokens_seen": 591928872,
      "step": 9750,
      "train_runtime": 127307.7603,
      "train_tokens_per_second": 4649.59
    },
    {
      "epoch": 1.2442602040816326,
      "grad_norm": 0.6213889480690334,
      "learning_rate": 3.1292377488117466e-06,
      "loss": 0.2687695026397705,
      "num_input_tokens_seen": 592234376,
      "step": 9755,
      "train_runtime": 127364.8152,
      "train_tokens_per_second": 4649.906
    },
    {
      "epoch": 1.2448979591836735,
      "grad_norm": 0.6495897733022374,
      "learning_rate": 3.124593583052026e-06,
      "loss": 0.25377483367919923,
      "num_input_tokens_seen": 592531608,
      "step": 9760,
      "train_runtime": 127437.5012,
      "train_tokens_per_second": 4649.586
    },
    {
      "epoch": 1.2455357142857142,
      "grad_norm": 0.6523830276643983,
      "learning_rate": 3.119951299393625e-06,
      "loss": 0.2609021902084351,
      "num_input_tokens_seen": 592827608,
      "step": 9765,
      "train_runtime": 127497.901,
      "train_tokens_per_second": 4649.705
    },
    {
      "epoch": 1.246173469387755,
      "grad_norm": 0.7011963327521072,
      "learning_rate": 3.115310902495396e-06,
      "loss": 0.26243650913238525,
      "num_input_tokens_seen": 593137488,
      "step": 9770,
      "train_runtime": 127571.704,
      "train_tokens_per_second": 4649.444
    },
    {
      "epoch": 1.246811224489796,
      "grad_norm": 0.6816688394845224,
      "learning_rate": 3.1106723970143016e-06,
      "loss": 0.273140025138855,
      "num_input_tokens_seen": 593451248,
      "step": 9775,
      "train_runtime": 127642.3561,
      "train_tokens_per_second": 4649.329
    },
    {
      "epoch": 1.2474489795918366,
      "grad_norm": 0.6954608888802939,
      "learning_rate": 3.10603578760541e-06,
      "loss": 0.24771556854248047,
      "num_input_tokens_seen": 593735344,
      "step": 9780,
      "train_runtime": 127713.5722,
      "train_tokens_per_second": 4648.96
    },
    {
      "epoch": 1.2480867346938775,
      "grad_norm": 0.6552008274039413,
      "learning_rate": 3.101401078921878e-06,
      "loss": 0.2547657251358032,
      "num_input_tokens_seen": 594041320,
      "step": 9785,
      "train_runtime": 127777.3076,
      "train_tokens_per_second": 4649.036
    },
    {
      "epoch": 1.2487244897959184,
      "grad_norm": 0.7631631481534996,
      "learning_rate": 3.0967682756149607e-06,
      "loss": 0.2581759452819824,
      "num_input_tokens_seen": 594341352,
      "step": 9790,
      "train_runtime": 127849.4879,
      "train_tokens_per_second": 4648.758
    },
    {
      "epoch": 1.249362244897959,
      "grad_norm": 0.7153650277500097,
      "learning_rate": 3.0921373823340028e-06,
      "loss": 0.25716280937194824,
      "num_input_tokens_seen": 594646352,
      "step": 9795,
      "train_runtime": 127917.3968,
      "train_tokens_per_second": 4648.675
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.7737620565303293,
      "learning_rate": 3.0875084037264237e-06,
      "loss": 0.25070748329162595,
      "num_input_tokens_seen": 594951600,
      "step": 9800,
      "train_runtime": 127979.9371,
      "train_tokens_per_second": 4648.788
    },
    {
      "epoch": 1.250637755102041,
      "grad_norm": 0.6105011346752942,
      "learning_rate": 3.0828813444377304e-06,
      "loss": 0.26685409545898436,
      "num_input_tokens_seen": 595265464,
      "step": 9805,
      "train_runtime": 128040.6729,
      "train_tokens_per_second": 4649.034
    },
    {
      "epoch": 1.2512755102040816,
      "grad_norm": 0.6801389888878429,
      "learning_rate": 3.0782562091114997e-06,
      "loss": 0.24711506366729735,
      "num_input_tokens_seen": 595563768,
      "step": 9810,
      "train_runtime": 128103.6441,
      "train_tokens_per_second": 4649.077
    },
    {
      "epoch": 1.2519132653061225,
      "grad_norm": 0.6358161259128543,
      "learning_rate": 3.0736330023893774e-06,
      "loss": 0.27412877082824705,
      "num_input_tokens_seen": 595873808,
      "step": 9815,
      "train_runtime": 128150.8662,
      "train_tokens_per_second": 4649.784
    },
    {
      "epoch": 1.2525510204081631,
      "grad_norm": 0.547957933485934,
      "learning_rate": 3.0690117289110732e-06,
      "loss": 0.2447650909423828,
      "num_input_tokens_seen": 596168104,
      "step": 9820,
      "train_runtime": 128214.4639,
      "train_tokens_per_second": 4649.773
    },
    {
      "epoch": 1.253188775510204,
      "grad_norm": 0.687675566564068,
      "learning_rate": 3.0643923933143603e-06,
      "loss": 0.27698750495910646,
      "num_input_tokens_seen": 596477128,
      "step": 9825,
      "train_runtime": 128274.5149,
      "train_tokens_per_second": 4650.005
    },
    {
      "epoch": 1.253826530612245,
      "grad_norm": 0.6494597349522581,
      "learning_rate": 3.059775000235063e-06,
      "loss": 0.24812660217285157,
      "num_input_tokens_seen": 596775888,
      "step": 9830,
      "train_runtime": 128334.0069,
      "train_tokens_per_second": 4650.177
    },
    {
      "epoch": 1.2544642857142856,
      "grad_norm": 0.6742342248112834,
      "learning_rate": 3.0551595543070567e-06,
      "loss": 0.25735490322113036,
      "num_input_tokens_seen": 597083000,
      "step": 9835,
      "train_runtime": 128394.0745,
      "train_tokens_per_second": 4650.394
    },
    {
      "epoch": 1.2551020408163265,
      "grad_norm": 0.588132941763026,
      "learning_rate": 3.050546060162267e-06,
      "loss": 0.25598070621490476,
      "num_input_tokens_seen": 597384416,
      "step": 9840,
      "train_runtime": 128458.1201,
      "train_tokens_per_second": 4650.422
    },
    {
      "epoch": 1.2557397959183674,
      "grad_norm": 0.6523181185468291,
      "learning_rate": 3.045934522430655e-06,
      "loss": 0.2531641960144043,
      "num_input_tokens_seen": 597683408,
      "step": 9845,
      "train_runtime": 128519.4913,
      "train_tokens_per_second": 4650.527
    },
    {
      "epoch": 1.256377551020408,
      "grad_norm": 0.5186578824755282,
      "learning_rate": 3.0413249457402206e-06,
      "loss": 0.24352428913116456,
      "num_input_tokens_seen": 597978680,
      "step": 9850,
      "train_runtime": 128575.6032,
      "train_tokens_per_second": 4650.794
    },
    {
      "epoch": 1.257015306122449,
      "grad_norm": 0.7497706288807056,
      "learning_rate": 3.0367173347169978e-06,
      "loss": 0.2518972396850586,
      "num_input_tokens_seen": 598278904,
      "step": 9855,
      "train_runtime": 128644.3456,
      "train_tokens_per_second": 4650.643
    },
    {
      "epoch": 1.2576530612244898,
      "grad_norm": 0.6682936358424924,
      "learning_rate": 3.0321116939850435e-06,
      "loss": 0.25064496994018554,
      "num_input_tokens_seen": 598588448,
      "step": 9860,
      "train_runtime": 128715.656,
      "train_tokens_per_second": 4650.471
    },
    {
      "epoch": 1.2582908163265305,
      "grad_norm": 0.6569544347037569,
      "learning_rate": 3.0275080281664417e-06,
      "loss": 0.26858522891998293,
      "num_input_tokens_seen": 598894360,
      "step": 9865,
      "train_runtime": 128776.5363,
      "train_tokens_per_second": 4650.648
    },
    {
      "epoch": 1.2589285714285714,
      "grad_norm": 0.6280506510343941,
      "learning_rate": 3.0229063418812932e-06,
      "loss": 0.2568094253540039,
      "num_input_tokens_seen": 599191976,
      "step": 9870,
      "train_runtime": 128849.6351,
      "train_tokens_per_second": 4650.32
    },
    {
      "epoch": 1.2595663265306123,
      "grad_norm": 1.0130878857610142,
      "learning_rate": 3.0183066397477103e-06,
      "loss": 0.24946315288543702,
      "num_input_tokens_seen": 599499472,
      "step": 9875,
      "train_runtime": 128907.73,
      "train_tokens_per_second": 4650.609
    },
    {
      "epoch": 1.260204081632653,
      "grad_norm": 0.7316748022706298,
      "learning_rate": 3.013708926381814e-06,
      "loss": 0.2662359237670898,
      "num_input_tokens_seen": 599806944,
      "step": 9880,
      "train_runtime": 128977.9497,
      "train_tokens_per_second": 4650.461
    },
    {
      "epoch": 1.2608418367346939,
      "grad_norm": 0.651018892310976,
      "learning_rate": 3.0091132063977344e-06,
      "loss": 0.2581866502761841,
      "num_input_tokens_seen": 600115496,
      "step": 9885,
      "train_runtime": 129042.3842,
      "train_tokens_per_second": 4650.53
    },
    {
      "epoch": 1.2614795918367347,
      "grad_norm": 0.7103080293781776,
      "learning_rate": 3.004519484407594e-06,
      "loss": 0.25688824653625486,
      "num_input_tokens_seen": 600426616,
      "step": 9890,
      "train_runtime": 129107.0702,
      "train_tokens_per_second": 4650.61
    },
    {
      "epoch": 1.2621173469387754,
      "grad_norm": 0.6152615903240626,
      "learning_rate": 2.9999277650215164e-06,
      "loss": 0.25172607898712157,
      "num_input_tokens_seen": 600719552,
      "step": 9895,
      "train_runtime": 129175.3695,
      "train_tokens_per_second": 4650.419
    },
    {
      "epoch": 1.2627551020408163,
      "grad_norm": 0.6428804398792916,
      "learning_rate": 2.9953380528476105e-06,
      "loss": 0.25584452152252196,
      "num_input_tokens_seen": 601024464,
      "step": 9900,
      "train_runtime": 129248.2615,
      "train_tokens_per_second": 4650.155
    },
    {
      "epoch": 1.2633928571428572,
      "grad_norm": 0.5992685996482658,
      "learning_rate": 2.9907503524919734e-06,
      "loss": 0.26361684799194335,
      "num_input_tokens_seen": 601324424,
      "step": 9905,
      "train_runtime": 129318.0209,
      "train_tokens_per_second": 4649.966
    },
    {
      "epoch": 1.2640306122448979,
      "grad_norm": 0.5960767762245135,
      "learning_rate": 2.9861646685586843e-06,
      "loss": 0.26084651947021487,
      "num_input_tokens_seen": 601633216,
      "step": 9910,
      "train_runtime": 129390.1912,
      "train_tokens_per_second": 4649.759
    },
    {
      "epoch": 1.2646683673469388,
      "grad_norm": 0.6304029794461745,
      "learning_rate": 2.981581005649795e-06,
      "loss": 0.25563340187072753,
      "num_input_tokens_seen": 601941432,
      "step": 9915,
      "train_runtime": 129450.1832,
      "train_tokens_per_second": 4649.985
    },
    {
      "epoch": 1.2653061224489797,
      "grad_norm": 0.7046465445006264,
      "learning_rate": 2.9769993683653324e-06,
      "loss": 0.2499422550201416,
      "num_input_tokens_seen": 602250352,
      "step": 9920,
      "train_runtime": 129518.73,
      "train_tokens_per_second": 4649.909
    },
    {
      "epoch": 1.2659438775510203,
      "grad_norm": 0.6054572425287019,
      "learning_rate": 2.9724197613032917e-06,
      "loss": 0.24319024085998536,
      "num_input_tokens_seen": 602558568,
      "step": 9925,
      "train_runtime": 129579.8484,
      "train_tokens_per_second": 4650.095
    },
    {
      "epoch": 1.2665816326530612,
      "grad_norm": 0.6476194312179154,
      "learning_rate": 2.9678421890596254e-06,
      "loss": 0.25522913932800295,
      "num_input_tokens_seen": 602868024,
      "step": 9930,
      "train_runtime": 129647.6093,
      "train_tokens_per_second": 4650.051
    },
    {
      "epoch": 1.2672193877551021,
      "grad_norm": 0.6681534094124768,
      "learning_rate": 2.9632666562282477e-06,
      "loss": 0.26373605728149413,
      "num_input_tokens_seen": 603170320,
      "step": 9935,
      "train_runtime": 129708.6046,
      "train_tokens_per_second": 4650.195
    },
    {
      "epoch": 1.2678571428571428,
      "grad_norm": 0.6172682304373681,
      "learning_rate": 2.9586931674010265e-06,
      "loss": 0.2531527757644653,
      "num_input_tokens_seen": 603477864,
      "step": 9940,
      "train_runtime": 129774.7676,
      "train_tokens_per_second": 4650.194
    },
    {
      "epoch": 1.2684948979591837,
      "grad_norm": 0.5602088394574272,
      "learning_rate": 2.9541217271677747e-06,
      "loss": 0.2425062894821167,
      "num_input_tokens_seen": 603787784,
      "step": 9945,
      "train_runtime": 129839.2639,
      "train_tokens_per_second": 4650.271
    },
    {
      "epoch": 1.2691326530612246,
      "grad_norm": 0.6393547227516018,
      "learning_rate": 2.949552340116254e-06,
      "loss": 0.2609687089920044,
      "num_input_tokens_seen": 604096976,
      "step": 9950,
      "train_runtime": 129915.6675,
      "train_tokens_per_second": 4649.916
    },
    {
      "epoch": 1.2697704081632653,
      "grad_norm": 0.7057184622242589,
      "learning_rate": 2.9449850108321644e-06,
      "loss": 0.242216157913208,
      "num_input_tokens_seen": 604392392,
      "step": 9955,
      "train_runtime": 129982.4602,
      "train_tokens_per_second": 4649.8
    },
    {
      "epoch": 1.2704081632653061,
      "grad_norm": 0.6099371831529727,
      "learning_rate": 2.9404197438991377e-06,
      "loss": 0.26979479789733884,
      "num_input_tokens_seen": 604699384,
      "step": 9960,
      "train_runtime": 130048.0207,
      "train_tokens_per_second": 4649.816
    },
    {
      "epoch": 1.271045918367347,
      "grad_norm": 0.7060903353195286,
      "learning_rate": 2.935856543898737e-06,
      "loss": 0.2459017515182495,
      "num_input_tokens_seen": 605005040,
      "step": 9965,
      "train_runtime": 130111.253,
      "train_tokens_per_second": 4649.906
    },
    {
      "epoch": 1.2716836734693877,
      "grad_norm": 0.6778245237757761,
      "learning_rate": 2.9312954154104563e-06,
      "loss": 0.26631999015808105,
      "num_input_tokens_seen": 605306128,
      "step": 9970,
      "train_runtime": 130177.6962,
      "train_tokens_per_second": 4649.845
    },
    {
      "epoch": 1.2723214285714286,
      "grad_norm": 0.5990361048893761,
      "learning_rate": 2.9267363630117028e-06,
      "loss": 0.2491915225982666,
      "num_input_tokens_seen": 605616512,
      "step": 9975,
      "train_runtime": 130245.6472,
      "train_tokens_per_second": 4649.802
    },
    {
      "epoch": 1.2729591836734695,
      "grad_norm": 0.7182257701001821,
      "learning_rate": 2.9221793912778053e-06,
      "loss": 0.2490309476852417,
      "num_input_tokens_seen": 605908128,
      "step": 9980,
      "train_runtime": 130304.8941,
      "train_tokens_per_second": 4649.926
    },
    {
      "epoch": 1.2735969387755102,
      "grad_norm": 0.6271430490639972,
      "learning_rate": 2.9176245047820064e-06,
      "loss": 0.2644538640975952,
      "num_input_tokens_seen": 606204472,
      "step": 9985,
      "train_runtime": 130360.7128,
      "train_tokens_per_second": 4650.208
    },
    {
      "epoch": 1.274234693877551,
      "grad_norm": 0.6897428999300005,
      "learning_rate": 2.913071708095446e-06,
      "loss": 0.2759737730026245,
      "num_input_tokens_seen": 606511672,
      "step": 9990,
      "train_runtime": 130420.8642,
      "train_tokens_per_second": 4650.419
    },
    {
      "epoch": 1.274872448979592,
      "grad_norm": 0.6135000369081007,
      "learning_rate": 2.908521005787177e-06,
      "loss": 0.25093646049499513,
      "num_input_tokens_seen": 606807536,
      "step": 9995,
      "train_runtime": 130488.9034,
      "train_tokens_per_second": 4650.262
    },
    {
      "epoch": 1.2755102040816326,
      "grad_norm": 0.6637867383854398,
      "learning_rate": 2.903972402424151e-06,
      "loss": 0.26477274894714353,
      "num_input_tokens_seen": 607106096,
      "step": 10000,
      "train_runtime": 130563.6249,
      "train_tokens_per_second": 4649.887
    },
    {
      "epoch": 1.2761479591836735,
      "grad_norm": 0.7321868174113467,
      "learning_rate": 2.899425902571203e-06,
      "loss": 0.28002426624298093,
      "num_input_tokens_seen": 607424872,
      "step": 10005,
      "train_runtime": 130618.45,
      "train_tokens_per_second": 4650.376
    },
    {
      "epoch": 1.2767857142857144,
      "grad_norm": 0.7247413141361364,
      "learning_rate": 2.8948815107910653e-06,
      "loss": 0.25679717063903806,
      "num_input_tokens_seen": 607733480,
      "step": 10010,
      "train_runtime": 130684.2643,
      "train_tokens_per_second": 4650.395
    },
    {
      "epoch": 1.277423469387755,
      "grad_norm": 0.6190331531886365,
      "learning_rate": 2.8903392316443534e-06,
      "loss": 0.257327938079834,
      "num_input_tokens_seen": 608038432,
      "step": 10015,
      "train_runtime": 130756.388,
      "train_tokens_per_second": 4650.162
    },
    {
      "epoch": 1.278061224489796,
      "grad_norm": 0.6102981167242966,
      "learning_rate": 2.885799069689561e-06,
      "loss": 0.25208125114440916,
      "num_input_tokens_seen": 608354768,
      "step": 10020,
      "train_runtime": 130813.2447,
      "train_tokens_per_second": 4650.559
    },
    {
      "epoch": 1.2786989795918369,
      "grad_norm": 0.6294305335884982,
      "learning_rate": 2.8812610294830568e-06,
      "loss": 0.2639681100845337,
      "num_input_tokens_seen": 608646632,
      "step": 10025,
      "train_runtime": 130882.8881,
      "train_tokens_per_second": 4650.315
    },
    {
      "epoch": 1.2793367346938775,
      "grad_norm": 0.7137119012445925,
      "learning_rate": 2.8767251155790826e-06,
      "loss": 0.25915184020996096,
      "num_input_tokens_seen": 608954568,
      "step": 10030,
      "train_runtime": 130950.7914,
      "train_tokens_per_second": 4650.255
    },
    {
      "epoch": 1.2799744897959184,
      "grad_norm": 0.6110036230769141,
      "learning_rate": 2.8721913325297434e-06,
      "loss": 0.27418949604034426,
      "num_input_tokens_seen": 609265952,
      "step": 10035,
      "train_runtime": 131009.8313,
      "train_tokens_per_second": 4650.536
    },
    {
      "epoch": 1.280612244897959,
      "grad_norm": 0.6725197054712593,
      "learning_rate": 2.8676596848850084e-06,
      "loss": 0.25756547451019285,
      "num_input_tokens_seen": 609583120,
      "step": 10040,
      "train_runtime": 131071.6354,
      "train_tokens_per_second": 4650.763
    },
    {
      "epoch": 1.28125,
      "grad_norm": 0.6011573752497347,
      "learning_rate": 2.8631301771927024e-06,
      "loss": 0.219584321975708,
      "num_input_tokens_seen": 609891536,
      "step": 10045,
      "train_runtime": 131137.8367,
      "train_tokens_per_second": 4650.767
    },
    {
      "epoch": 1.281887755102041,
      "grad_norm": 0.6476099551009382,
      "learning_rate": 2.858602813998501e-06,
      "loss": 0.261248779296875,
      "num_input_tokens_seen": 610191608,
      "step": 10050,
      "train_runtime": 131209.2128,
      "train_tokens_per_second": 4650.524
    },
    {
      "epoch": 1.2825255102040816,
      "grad_norm": 0.7285425158352521,
      "learning_rate": 2.8540775998459336e-06,
      "loss": 0.2619506597518921,
      "num_input_tokens_seen": 610492744,
      "step": 10055,
      "train_runtime": 131276.3329,
      "train_tokens_per_second": 4650.44
    },
    {
      "epoch": 1.2831632653061225,
      "grad_norm": 0.649914020532788,
      "learning_rate": 2.849554539276361e-06,
      "loss": 0.2608958721160889,
      "num_input_tokens_seen": 610807288,
      "step": 10060,
      "train_runtime": 131350.7727,
      "train_tokens_per_second": 4650.199
    },
    {
      "epoch": 1.2838010204081631,
      "grad_norm": 0.6279072982442597,
      "learning_rate": 2.845033636828998e-06,
      "loss": 0.2728137016296387,
      "num_input_tokens_seen": 611113336,
      "step": 10065,
      "train_runtime": 131424.7595,
      "train_tokens_per_second": 4649.91
    },
    {
      "epoch": 1.284438775510204,
      "grad_norm": 0.6576997380235776,
      "learning_rate": 2.840514897040884e-06,
      "loss": 0.25149002075195315,
      "num_input_tokens_seen": 611424664,
      "step": 10070,
      "train_runtime": 131491.3931,
      "train_tokens_per_second": 4649.922
    },
    {
      "epoch": 1.285076530612245,
      "grad_norm": 0.7135138265820079,
      "learning_rate": 2.835998324446887e-06,
      "loss": 0.24344327449798583,
      "num_input_tokens_seen": 611722296,
      "step": 10075,
      "train_runtime": 131560.8457,
      "train_tokens_per_second": 4649.729
    },
    {
      "epoch": 1.2857142857142856,
      "grad_norm": 0.8300117316787209,
      "learning_rate": 2.831483923579703e-06,
      "loss": 0.21739587783813477,
      "num_input_tokens_seen": 612008712,
      "step": 10080,
      "train_runtime": 131630.3069,
      "train_tokens_per_second": 4649.451
    },
    {
      "epoch": 1.2863520408163265,
      "grad_norm": 0.7170568738190197,
      "learning_rate": 2.8269716989698547e-06,
      "loss": 0.2605217456817627,
      "num_input_tokens_seen": 612309344,
      "step": 10085,
      "train_runtime": 131699.8802,
      "train_tokens_per_second": 4649.278
    },
    {
      "epoch": 1.2869897959183674,
      "grad_norm": 0.690429723103565,
      "learning_rate": 2.822461655145669e-06,
      "loss": 0.2618260383605957,
      "num_input_tokens_seen": 612608216,
      "step": 10090,
      "train_runtime": 131763.1517,
      "train_tokens_per_second": 4649.314
    },
    {
      "epoch": 1.287627551020408,
      "grad_norm": 0.7211370361617034,
      "learning_rate": 2.817953796633289e-06,
      "loss": 0.24970576763153077,
      "num_input_tokens_seen": 612908504,
      "step": 10095,
      "train_runtime": 131831.2232,
      "train_tokens_per_second": 4649.191
    },
    {
      "epoch": 1.288265306122449,
      "grad_norm": 0.7259032001730406,
      "learning_rate": 2.813448127956672e-06,
      "loss": 0.2589538335800171,
      "num_input_tokens_seen": 613202592,
      "step": 10100,
      "train_runtime": 131894.2426,
      "train_tokens_per_second": 4649.199
    },
    {
      "epoch": 1.2889030612244898,
      "grad_norm": 0.6197107246379037,
      "learning_rate": 2.8089446536375642e-06,
      "loss": 0.2536892890930176,
      "num_input_tokens_seen": 613511544,
      "step": 10105,
      "train_runtime": 131961.9325,
      "train_tokens_per_second": 4649.155
    },
    {
      "epoch": 1.2895408163265305,
      "grad_norm": 0.7105835763581693,
      "learning_rate": 2.8044433781955196e-06,
      "loss": 0.2439013957977295,
      "num_input_tokens_seen": 613812360,
      "step": 10110,
      "train_runtime": 132024.5481,
      "train_tokens_per_second": 4649.229
    },
    {
      "epoch": 1.2901785714285714,
      "grad_norm": 0.5749126725639873,
      "learning_rate": 2.7999443061478827e-06,
      "loss": 0.2278393268585205,
      "num_input_tokens_seen": 614108216,
      "step": 10115,
      "train_runtime": 132091.8692,
      "train_tokens_per_second": 4649.099
    },
    {
      "epoch": 1.2908163265306123,
      "grad_norm": 0.7989012983361106,
      "learning_rate": 2.795447442009786e-06,
      "loss": 0.2468031167984009,
      "num_input_tokens_seen": 614409664,
      "step": 10120,
      "train_runtime": 132156.0444,
      "train_tokens_per_second": 4649.123
    },
    {
      "epoch": 1.291454081632653,
      "grad_norm": 0.67816707037487,
      "learning_rate": 2.7909527902941467e-06,
      "loss": 0.26167936325073243,
      "num_input_tokens_seen": 614716896,
      "step": 10125,
      "train_runtime": 132223.5742,
      "train_tokens_per_second": 4649.072
    },
    {
      "epoch": 1.2920918367346939,
      "grad_norm": 0.6771644170164809,
      "learning_rate": 2.786460355511662e-06,
      "loss": 0.2604956150054932,
      "num_input_tokens_seen": 615013432,
      "step": 10130,
      "train_runtime": 132282.9754,
      "train_tokens_per_second": 4649.226
    },
    {
      "epoch": 1.2927295918367347,
      "grad_norm": 0.5849170247320123,
      "learning_rate": 2.7819701421708047e-06,
      "loss": 0.2588627338409424,
      "num_input_tokens_seen": 615303208,
      "step": 10135,
      "train_runtime": 132344.7601,
      "train_tokens_per_second": 4649.245
    },
    {
      "epoch": 1.2933673469387754,
      "grad_norm": 0.6647315408989721,
      "learning_rate": 2.7774821547778176e-06,
      "loss": 0.276265287399292,
      "num_input_tokens_seen": 615628832,
      "step": 10140,
      "train_runtime": 132389.603,
      "train_tokens_per_second": 4650.13
    },
    {
      "epoch": 1.2940051020408163,
      "grad_norm": 0.6733416147279704,
      "learning_rate": 2.772996397836704e-06,
      "loss": 0.25055019855499266,
      "num_input_tokens_seen": 615931752,
      "step": 10145,
      "train_runtime": 132448.6172,
      "train_tokens_per_second": 4650.345
    },
    {
      "epoch": 1.2946428571428572,
      "grad_norm": 0.6500325488391534,
      "learning_rate": 2.768512875849242e-06,
      "loss": 0.23627872467041017,
      "num_input_tokens_seen": 616243688,
      "step": 10150,
      "train_runtime": 132496.0352,
      "train_tokens_per_second": 4651.035
    },
    {
      "epoch": 1.2952806122448979,
      "grad_norm": 0.6266059224504728,
      "learning_rate": 2.7640315933149568e-06,
      "loss": 0.26852166652679443,
      "num_input_tokens_seen": 616547944,
      "step": 10155,
      "train_runtime": 132564.6291,
      "train_tokens_per_second": 4650.923
    },
    {
      "epoch": 1.2959183673469388,
      "grad_norm": 0.6994851934152331,
      "learning_rate": 2.7595525547311257e-06,
      "loss": 0.24700803756713868,
      "num_input_tokens_seen": 616845272,
      "step": 10160,
      "train_runtime": 132636.8062,
      "train_tokens_per_second": 4650.634
    },
    {
      "epoch": 1.2965561224489797,
      "grad_norm": 0.6365087964133738,
      "learning_rate": 2.755075764592777e-06,
      "loss": 0.26582865715026854,
      "num_input_tokens_seen": 617160472,
      "step": 10165,
      "train_runtime": 132701.6567,
      "train_tokens_per_second": 4650.737
    },
    {
      "epoch": 1.2971938775510203,
      "grad_norm": 0.5938124532129865,
      "learning_rate": 2.750601227392687e-06,
      "loss": 0.24859161376953126,
      "num_input_tokens_seen": 617456352,
      "step": 10170,
      "train_runtime": 132771.0572,
      "train_tokens_per_second": 4650.534
    },
    {
      "epoch": 1.2978316326530612,
      "grad_norm": 0.6937636225671621,
      "learning_rate": 2.7461289476213617e-06,
      "loss": 0.2627084732055664,
      "num_input_tokens_seen": 617764112,
      "step": 10175,
      "train_runtime": 132829.6461,
      "train_tokens_per_second": 4650.8
    },
    {
      "epoch": 1.2984693877551021,
      "grad_norm": 0.6882085993566683,
      "learning_rate": 2.741658929767045e-06,
      "loss": 0.24822802543640138,
      "num_input_tokens_seen": 618072856,
      "step": 10180,
      "train_runtime": 132887.3005,
      "train_tokens_per_second": 4651.106
    },
    {
      "epoch": 1.2991071428571428,
      "grad_norm": 0.648652873862695,
      "learning_rate": 2.7371911783157178e-06,
      "loss": 0.23618090152740479,
      "num_input_tokens_seen": 618365920,
      "step": 10185,
      "train_runtime": 132947.2677,
      "train_tokens_per_second": 4651.212
    },
    {
      "epoch": 1.2997448979591837,
      "grad_norm": 0.5628727217052052,
      "learning_rate": 2.7327256977510764e-06,
      "loss": 0.261117959022522,
      "num_input_tokens_seen": 618681128,
      "step": 10190,
      "train_runtime": 133002.6092,
      "train_tokens_per_second": 4651.647
    },
    {
      "epoch": 1.3003826530612246,
      "grad_norm": 0.6734165895986504,
      "learning_rate": 2.7282624925545432e-06,
      "loss": 0.2443925380706787,
      "num_input_tokens_seen": 618987336,
      "step": 10195,
      "train_runtime": 133049.8715,
      "train_tokens_per_second": 4652.296
    },
    {
      "epoch": 1.3010204081632653,
      "grad_norm": 0.6491868597447439,
      "learning_rate": 2.723801567205256e-06,
      "loss": 0.2537305593490601,
      "num_input_tokens_seen": 619294688,
      "step": 10200,
      "train_runtime": 133121.5278,
      "train_tokens_per_second": 4652.1
    },
    {
      "epoch": 1.3016581632653061,
      "grad_norm": 0.6428935089414552,
      "learning_rate": 2.7193429261800657e-06,
      "loss": 0.2404731035232544,
      "num_input_tokens_seen": 619599992,
      "step": 10205,
      "train_runtime": 133185.5983,
      "train_tokens_per_second": 4652.155
    },
    {
      "epoch": 1.302295918367347,
      "grad_norm": 0.6936701268816926,
      "learning_rate": 2.7148865739535303e-06,
      "loss": 0.2664501667022705,
      "num_input_tokens_seen": 619904240,
      "step": 10210,
      "train_runtime": 133242.1964,
      "train_tokens_per_second": 4652.462
    },
    {
      "epoch": 1.3029336734693877,
      "grad_norm": 0.629156863618074,
      "learning_rate": 2.710432514997909e-06,
      "loss": 0.27101132869720457,
      "num_input_tokens_seen": 620218168,
      "step": 10215,
      "train_runtime": 133305.0068,
      "train_tokens_per_second": 4652.625
    },
    {
      "epoch": 1.3035714285714286,
      "grad_norm": 0.8439382359438976,
      "learning_rate": 2.7059807537831624e-06,
      "loss": 0.2491471290588379,
      "num_input_tokens_seen": 620530952,
      "step": 10220,
      "train_runtime": 133368.8263,
      "train_tokens_per_second": 4652.744
    },
    {
      "epoch": 1.3042091836734695,
      "grad_norm": 0.6380148848448121,
      "learning_rate": 2.7015312947769436e-06,
      "loss": 0.25763468742370604,
      "num_input_tokens_seen": 620831664,
      "step": 10225,
      "train_runtime": 133438.3092,
      "train_tokens_per_second": 4652.574
    },
    {
      "epoch": 1.3048469387755102,
      "grad_norm": 0.7221871794117705,
      "learning_rate": 2.6970841424445947e-06,
      "loss": 0.26262674331665037,
      "num_input_tokens_seen": 621126216,
      "step": 10230,
      "train_runtime": 133505.9152,
      "train_tokens_per_second": 4652.425
    },
    {
      "epoch": 1.305484693877551,
      "grad_norm": 0.6411570328012643,
      "learning_rate": 2.692639301249144e-06,
      "loss": 0.2711535692214966,
      "num_input_tokens_seen": 621431064,
      "step": 10235,
      "train_runtime": 133572.2515,
      "train_tokens_per_second": 4652.396
    },
    {
      "epoch": 1.306122448979592,
      "grad_norm": 0.7123075157071713,
      "learning_rate": 2.6881967756513004e-06,
      "loss": 0.24485445022583008,
      "num_input_tokens_seen": 621748664,
      "step": 10240,
      "train_runtime": 133619.3022,
      "train_tokens_per_second": 4653.135
    },
    {
      "epoch": 1.3067602040816326,
      "grad_norm": 0.6380666943265844,
      "learning_rate": 2.6837565701094505e-06,
      "loss": 0.26869773864746094,
      "num_input_tokens_seen": 622049712,
      "step": 10245,
      "train_runtime": 133690.8125,
      "train_tokens_per_second": 4652.898
    },
    {
      "epoch": 1.3073979591836735,
      "grad_norm": 0.7029422356592782,
      "learning_rate": 2.679318689079645e-06,
      "loss": 0.2626228094100952,
      "num_input_tokens_seen": 622359040,
      "step": 10250,
      "train_runtime": 133762.0832,
      "train_tokens_per_second": 4652.731
    },
    {
      "epoch": 1.3080357142857144,
      "grad_norm": 0.65230916271705,
      "learning_rate": 2.6748831370156148e-06,
      "loss": 0.24830842018127441,
      "num_input_tokens_seen": 622654288,
      "step": 10255,
      "train_runtime": 133827.315,
      "train_tokens_per_second": 4652.67
    },
    {
      "epoch": 1.308673469387755,
      "grad_norm": 0.6305512068807968,
      "learning_rate": 2.6704499183687404e-06,
      "loss": 0.24921224117279053,
      "num_input_tokens_seen": 622960992,
      "step": 10260,
      "train_runtime": 133891.5144,
      "train_tokens_per_second": 4652.729
    },
    {
      "epoch": 1.309311224489796,
      "grad_norm": 0.6641317675834468,
      "learning_rate": 2.6660190375880656e-06,
      "loss": 0.24191765785217284,
      "num_input_tokens_seen": 623271960,
      "step": 10265,
      "train_runtime": 133963.8275,
      "train_tokens_per_second": 4652.539
    },
    {
      "epoch": 1.3099489795918369,
      "grad_norm": 0.620307548738313,
      "learning_rate": 2.661590499120296e-06,
      "loss": 0.2493054151535034,
      "num_input_tokens_seen": 623578472,
      "step": 10270,
      "train_runtime": 134019.4072,
      "train_tokens_per_second": 4652.897
    },
    {
      "epoch": 1.3105867346938775,
      "grad_norm": 0.6408356538606649,
      "learning_rate": 2.657164307409773e-06,
      "loss": 0.2380746841430664,
      "num_input_tokens_seen": 623877328,
      "step": 10275,
      "train_runtime": 134078.8456,
      "train_tokens_per_second": 4653.063
    },
    {
      "epoch": 1.3112244897959184,
      "grad_norm": 0.5961639808305982,
      "learning_rate": 2.6527404668984907e-06,
      "loss": 0.24416801929473878,
      "num_input_tokens_seen": 624169576,
      "step": 10280,
      "train_runtime": 134146.2055,
      "train_tokens_per_second": 4652.905
    },
    {
      "epoch": 1.311862244897959,
      "grad_norm": 0.606658146504834,
      "learning_rate": 2.648318982026082e-06,
      "loss": 0.23076577186584474,
      "num_input_tokens_seen": 624471232,
      "step": 10285,
      "train_runtime": 134198.9787,
      "train_tokens_per_second": 4653.323
    },
    {
      "epoch": 1.3125,
      "grad_norm": 0.7003322418402668,
      "learning_rate": 2.6438998572298157e-06,
      "loss": 0.25791611671447756,
      "num_input_tokens_seen": 624766680,
      "step": 10290,
      "train_runtime": 134268.7884,
      "train_tokens_per_second": 4653.104
    },
    {
      "epoch": 1.313137755102041,
      "grad_norm": 0.7379463589403175,
      "learning_rate": 2.639483096944593e-06,
      "loss": 0.2841819763183594,
      "num_input_tokens_seen": 625069280,
      "step": 10295,
      "train_runtime": 134325.7322,
      "train_tokens_per_second": 4653.384
    },
    {
      "epoch": 1.3137755102040816,
      "grad_norm": 0.7038679723358398,
      "learning_rate": 2.63506870560294e-06,
      "loss": 0.25959005355834963,
      "num_input_tokens_seen": 625385304,
      "step": 10300,
      "train_runtime": 134388.9597,
      "train_tokens_per_second": 4653.547
    },
    {
      "epoch": 1.3144132653061225,
      "grad_norm": 0.694339505158264,
      "learning_rate": 2.6306566876350072e-06,
      "loss": 0.28363423347473143,
      "num_input_tokens_seen": 625697000,
      "step": 10305,
      "train_runtime": 134456.6965,
      "train_tokens_per_second": 4653.521
    },
    {
      "epoch": 1.3150510204081631,
      "grad_norm": 0.6729179093091422,
      "learning_rate": 2.626247047468563e-06,
      "loss": 0.2525624752044678,
      "num_input_tokens_seen": 625983968,
      "step": 10310,
      "train_runtime": 134518.7724,
      "train_tokens_per_second": 4653.506
    },
    {
      "epoch": 1.315688775510204,
      "grad_norm": 0.6307130113578302,
      "learning_rate": 2.6218397895289893e-06,
      "loss": 0.25787949562072754,
      "num_input_tokens_seen": 626278856,
      "step": 10315,
      "train_runtime": 134586.7912,
      "train_tokens_per_second": 4653.346
    },
    {
      "epoch": 1.316326530612245,
      "grad_norm": 0.6396057526222012,
      "learning_rate": 2.617434918239278e-06,
      "loss": 0.2836592197418213,
      "num_input_tokens_seen": 626594600,
      "step": 10320,
      "train_runtime": 134652.054,
      "train_tokens_per_second": 4653.435
    },
    {
      "epoch": 1.3169642857142856,
      "grad_norm": 0.6579783302043273,
      "learning_rate": 2.6130324380200234e-06,
      "loss": 0.27531752586364744,
      "num_input_tokens_seen": 626907120,
      "step": 10325,
      "train_runtime": 134709.5533,
      "train_tokens_per_second": 4653.769
    },
    {
      "epoch": 1.3176020408163265,
      "grad_norm": 0.7121021764571803,
      "learning_rate": 2.6086323532894225e-06,
      "loss": 0.2745412826538086,
      "num_input_tokens_seen": 627215384,
      "step": 10330,
      "train_runtime": 134771.8803,
      "train_tokens_per_second": 4653.904
    },
    {
      "epoch": 1.3182397959183674,
      "grad_norm": 0.6775781147529878,
      "learning_rate": 2.604234668463268e-06,
      "loss": 0.2636935949325562,
      "num_input_tokens_seen": 627522696,
      "step": 10335,
      "train_runtime": 134848.2587,
      "train_tokens_per_second": 4653.547
    },
    {
      "epoch": 1.318877551020408,
      "grad_norm": 0.676427120925447,
      "learning_rate": 2.5998393879549444e-06,
      "loss": 0.24675984382629396,
      "num_input_tokens_seen": 627822656,
      "step": 10340,
      "train_runtime": 134911.3137,
      "train_tokens_per_second": 4653.595
    },
    {
      "epoch": 1.319515306122449,
      "grad_norm": 0.6446057835307174,
      "learning_rate": 2.5954465161754226e-06,
      "loss": 0.2533129692077637,
      "num_input_tokens_seen": 628140568,
      "step": 10345,
      "train_runtime": 134974.9025,
      "train_tokens_per_second": 4653.758
    },
    {
      "epoch": 1.3201530612244898,
      "grad_norm": 0.6817642612737654,
      "learning_rate": 2.5910560575332524e-06,
      "loss": 0.2645681381225586,
      "num_input_tokens_seen": 628449208,
      "step": 10350,
      "train_runtime": 135037.885,
      "train_tokens_per_second": 4653.873
    },
    {
      "epoch": 1.3207908163265305,
      "grad_norm": 0.6776014241789975,
      "learning_rate": 2.5866680164345704e-06,
      "loss": 0.26526970863342286,
      "num_input_tokens_seen": 628754088,
      "step": 10355,
      "train_runtime": 135094.8645,
      "train_tokens_per_second": 4654.167
    },
    {
      "epoch": 1.3214285714285714,
      "grad_norm": 0.5642825013265818,
      "learning_rate": 2.5822823972830825e-06,
      "loss": 0.2514237403869629,
      "num_input_tokens_seen": 629068856,
      "step": 10360,
      "train_runtime": 135155.0177,
      "train_tokens_per_second": 4654.425
    },
    {
      "epoch": 1.3220663265306123,
      "grad_norm": 0.7327301705078337,
      "learning_rate": 2.5778992044800597e-06,
      "loss": 0.26770195960998533,
      "num_input_tokens_seen": 629366976,
      "step": 10365,
      "train_runtime": 135212.7967,
      "train_tokens_per_second": 4654.641
    },
    {
      "epoch": 1.322704081632653,
      "grad_norm": 0.5960025519414979,
      "learning_rate": 2.573518442424343e-06,
      "loss": 0.24228177070617676,
      "num_input_tokens_seen": 629667240,
      "step": 10370,
      "train_runtime": 135276.1603,
      "train_tokens_per_second": 4654.68
    },
    {
      "epoch": 1.3233418367346939,
      "grad_norm": 0.6116470066239749,
      "learning_rate": 2.569140115512338e-06,
      "loss": 0.2634902000427246,
      "num_input_tokens_seen": 629965416,
      "step": 10375,
      "train_runtime": 135348.6113,
      "train_tokens_per_second": 4654.391
    },
    {
      "epoch": 1.3239795918367347,
      "grad_norm": 0.5870005842251248,
      "learning_rate": 2.564764228137996e-06,
      "loss": 0.24268980026245118,
      "num_input_tokens_seen": 630276384,
      "step": 10380,
      "train_runtime": 135418.5572,
      "train_tokens_per_second": 4654.284
    },
    {
      "epoch": 1.3246173469387754,
      "grad_norm": 0.6140813594421831,
      "learning_rate": 2.5603907846928277e-06,
      "loss": 0.23833558559417725,
      "num_input_tokens_seen": 630580696,
      "step": 10385,
      "train_runtime": 135486.4945,
      "train_tokens_per_second": 4654.196
    },
    {
      "epoch": 1.3252551020408163,
      "grad_norm": 0.7194897793225727,
      "learning_rate": 2.556019789565889e-06,
      "loss": 0.2554591655731201,
      "num_input_tokens_seen": 630884048,
      "step": 10390,
      "train_runtime": 135558.9181,
      "train_tokens_per_second": 4653.947
    },
    {
      "epoch": 1.3258928571428572,
      "grad_norm": 0.6523988729726404,
      "learning_rate": 2.5516512471437783e-06,
      "loss": 0.2595228910446167,
      "num_input_tokens_seen": 631185680,
      "step": 10395,
      "train_runtime": 135630.3377,
      "train_tokens_per_second": 4653.72
    },
    {
      "epoch": 1.3265306122448979,
      "grad_norm": 0.6408525997044796,
      "learning_rate": 2.547285161810634e-06,
      "loss": 0.24489250183105468,
      "num_input_tokens_seen": 631496624,
      "step": 10400,
      "train_runtime": 135693.988,
      "train_tokens_per_second": 4653.829
    },
    {
      "epoch": 1.3271683673469388,
      "grad_norm": 0.6287652104536363,
      "learning_rate": 2.5429215379481267e-06,
      "loss": 0.23495521545410156,
      "num_input_tokens_seen": 631786144,
      "step": 10405,
      "train_runtime": 135751.3601,
      "train_tokens_per_second": 4653.995
    },
    {
      "epoch": 1.3278061224489797,
      "grad_norm": 0.749636468899271,
      "learning_rate": 2.5385603799354574e-06,
      "loss": 0.25196335315704343,
      "num_input_tokens_seen": 632093448,
      "step": 10410,
      "train_runtime": 135826.3662,
      "train_tokens_per_second": 4653.687
    },
    {
      "epoch": 1.3284438775510203,
      "grad_norm": 0.6425794870646088,
      "learning_rate": 2.534201692149354e-06,
      "loss": 0.2630856275558472,
      "num_input_tokens_seen": 632396136,
      "step": 10415,
      "train_runtime": 135882.1028,
      "train_tokens_per_second": 4654.006
    },
    {
      "epoch": 1.3290816326530612,
      "grad_norm": 0.6860730297327605,
      "learning_rate": 2.5298454789640636e-06,
      "loss": 0.2595140218734741,
      "num_input_tokens_seen": 632701096,
      "step": 10420,
      "train_runtime": 135941.9066,
      "train_tokens_per_second": 4654.202
    },
    {
      "epoch": 1.3297193877551021,
      "grad_norm": 0.6770836255647433,
      "learning_rate": 2.5254917447513503e-06,
      "loss": 0.24976706504821777,
      "num_input_tokens_seen": 632999184,
      "step": 10425,
      "train_runtime": 136015.0857,
      "train_tokens_per_second": 4653.89
    },
    {
      "epoch": 1.3303571428571428,
      "grad_norm": 0.6479162926817887,
      "learning_rate": 2.5211404938804917e-06,
      "loss": 0.25724124908447266,
      "num_input_tokens_seen": 633291624,
      "step": 10430,
      "train_runtime": 136085.2051,
      "train_tokens_per_second": 4653.641
    },
    {
      "epoch": 1.3309948979591837,
      "grad_norm": 0.6257966867323908,
      "learning_rate": 2.5167917307182675e-06,
      "loss": 0.24438064098358153,
      "num_input_tokens_seen": 633604200,
      "step": 10435,
      "train_runtime": 136156.77,
      "train_tokens_per_second": 4653.49
    },
    {
      "epoch": 1.3316326530612246,
      "grad_norm": 0.6305823586995404,
      "learning_rate": 2.5124454596289704e-06,
      "loss": 0.2546777009963989,
      "num_input_tokens_seen": 633907848,
      "step": 10440,
      "train_runtime": 136218.761,
      "train_tokens_per_second": 4653.602
    },
    {
      "epoch": 1.3322704081632653,
      "grad_norm": 0.70414678915359,
      "learning_rate": 2.508101684974387e-06,
      "loss": 0.2731082677841187,
      "num_input_tokens_seen": 634214488,
      "step": 10445,
      "train_runtime": 136277.663,
      "train_tokens_per_second": 4653.84
    },
    {
      "epoch": 1.3329081632653061,
      "grad_norm": 0.7045600039751864,
      "learning_rate": 2.5037604111137935e-06,
      "loss": 0.25175440311431885,
      "num_input_tokens_seen": 634512472,
      "step": 10450,
      "train_runtime": 136334.7747,
      "train_tokens_per_second": 4654.077
    },
    {
      "epoch": 1.333545918367347,
      "grad_norm": 0.6982239045524724,
      "learning_rate": 2.4994216424039637e-06,
      "loss": 0.26717848777770997,
      "num_input_tokens_seen": 634811112,
      "step": 10455,
      "train_runtime": 136402.5798,
      "train_tokens_per_second": 4653.952
    },
    {
      "epoch": 1.3341836734693877,
      "grad_norm": 0.7241038788461323,
      "learning_rate": 2.495085383199159e-06,
      "loss": 0.24141874313354492,
      "num_input_tokens_seen": 635107792,
      "step": 10460,
      "train_runtime": 136470.642,
      "train_tokens_per_second": 4653.805
    },
    {
      "epoch": 1.3348214285714286,
      "grad_norm": 0.5690722032425513,
      "learning_rate": 2.4907516378511137e-06,
      "loss": 0.2344221591949463,
      "num_input_tokens_seen": 635413496,
      "step": 10465,
      "train_runtime": 136536.9124,
      "train_tokens_per_second": 4653.785
    },
    {
      "epoch": 1.3354591836734695,
      "grad_norm": 0.6611650346510646,
      "learning_rate": 2.486420410709043e-06,
      "loss": 0.2744720935821533,
      "num_input_tokens_seen": 635717640,
      "step": 10470,
      "train_runtime": 136602.6341,
      "train_tokens_per_second": 4653.773
    },
    {
      "epoch": 1.3360969387755102,
      "grad_norm": 0.6479139327116292,
      "learning_rate": 2.482091706119642e-06,
      "loss": 0.2511033296585083,
      "num_input_tokens_seen": 636030608,
      "step": 10475,
      "train_runtime": 136662.1909,
      "train_tokens_per_second": 4654.035
    },
    {
      "epoch": 1.336734693877551,
      "grad_norm": 0.678178923807823,
      "learning_rate": 2.4777655284270623e-06,
      "loss": 0.25283474922180177,
      "num_input_tokens_seen": 636337536,
      "step": 10480,
      "train_runtime": 136732.7215,
      "train_tokens_per_second": 4653.879
    },
    {
      "epoch": 1.337372448979592,
      "grad_norm": 0.6802857394272956,
      "learning_rate": 2.473441881972927e-06,
      "loss": 0.2502155065536499,
      "num_input_tokens_seen": 636632528,
      "step": 10485,
      "train_runtime": 136801.8893,
      "train_tokens_per_second": 4653.682
    },
    {
      "epoch": 1.3380102040816326,
      "grad_norm": 0.7264578611546987,
      "learning_rate": 2.4691207710963185e-06,
      "loss": 0.27263717651367186,
      "num_input_tokens_seen": 636940976,
      "step": 10490,
      "train_runtime": 136871.2442,
      "train_tokens_per_second": 4653.578
    },
    {
      "epoch": 1.3386479591836735,
      "grad_norm": 0.6467516275814501,
      "learning_rate": 2.464802200133772e-06,
      "loss": 0.25528175830841066,
      "num_input_tokens_seen": 637252544,
      "step": 10495,
      "train_runtime": 136937.207,
      "train_tokens_per_second": 4653.611
    },
    {
      "epoch": 1.3392857142857144,
      "grad_norm": 0.7639110842016443,
      "learning_rate": 2.4604861734192755e-06,
      "loss": 0.2494715690612793,
      "num_input_tokens_seen": 637538400,
      "step": 10500,
      "train_runtime": 137009.1181,
      "train_tokens_per_second": 4653.255
    },
    {
      "epoch": 1.339923469387755,
      "grad_norm": 0.6978560512446076,
      "learning_rate": 2.456172695284263e-06,
      "loss": 0.2574897289276123,
      "num_input_tokens_seen": 637829512,
      "step": 10505,
      "train_runtime": 137074.4399,
      "train_tokens_per_second": 4653.162
    },
    {
      "epoch": 1.340561224489796,
      "grad_norm": 0.6433184043998549,
      "learning_rate": 2.4518617700576116e-06,
      "loss": 0.25566754341125486,
      "num_input_tokens_seen": 638144032,
      "step": 10510,
      "train_runtime": 137126.6969,
      "train_tokens_per_second": 4653.682
    },
    {
      "epoch": 1.3411989795918369,
      "grad_norm": 0.6613905419455521,
      "learning_rate": 2.447553402065637e-06,
      "loss": 0.23982324600219726,
      "num_input_tokens_seen": 638449872,
      "step": 10515,
      "train_runtime": 137195.1138,
      "train_tokens_per_second": 4653.59
    },
    {
      "epoch": 1.3418367346938775,
      "grad_norm": 0.6470415769391574,
      "learning_rate": 2.4432475956320824e-06,
      "loss": 0.24281463623046876,
      "num_input_tokens_seen": 638754928,
      "step": 10520,
      "train_runtime": 137261.0735,
      "train_tokens_per_second": 4653.577
    },
    {
      "epoch": 1.3424744897959184,
      "grad_norm": 0.6013572112951462,
      "learning_rate": 2.4389443550781304e-06,
      "loss": 0.2543042182922363,
      "num_input_tokens_seen": 639060400,
      "step": 10525,
      "train_runtime": 137321.1544,
      "train_tokens_per_second": 4653.765
    },
    {
      "epoch": 1.343112244897959,
      "grad_norm": 0.615071899800615,
      "learning_rate": 2.4346436847223836e-06,
      "loss": 0.24138739109039306,
      "num_input_tokens_seen": 639352320,
      "step": 10530,
      "train_runtime": 137385.6249,
      "train_tokens_per_second": 4653.706
    },
    {
      "epoch": 1.34375,
      "grad_norm": 0.6328537978193586,
      "learning_rate": 2.4303455888808618e-06,
      "loss": 0.253510856628418,
      "num_input_tokens_seen": 639673216,
      "step": 10535,
      "train_runtime": 137453.9003,
      "train_tokens_per_second": 4653.729
    },
    {
      "epoch": 1.344387755102041,
      "grad_norm": 0.651441410672442,
      "learning_rate": 2.4260500718670036e-06,
      "loss": 0.2616002559661865,
      "num_input_tokens_seen": 639980632,
      "step": 10540,
      "train_runtime": 137531.0913,
      "train_tokens_per_second": 4653.352
    },
    {
      "epoch": 1.3450255102040816,
      "grad_norm": 0.7015196917719295,
      "learning_rate": 2.4217571379916673e-06,
      "loss": 0.2384871006011963,
      "num_input_tokens_seen": 640294192,
      "step": 10545,
      "train_runtime": 137599.4791,
      "train_tokens_per_second": 4653.318
    },
    {
      "epoch": 1.3456632653061225,
      "grad_norm": 0.6998072802468709,
      "learning_rate": 2.4174667915631038e-06,
      "loss": 0.27579896450042723,
      "num_input_tokens_seen": 640588232,
      "step": 10550,
      "train_runtime": 137663.6147,
      "train_tokens_per_second": 4653.286
    },
    {
      "epoch": 1.3463010204081631,
      "grad_norm": 0.7016327537117313,
      "learning_rate": 2.413179036886978e-06,
      "loss": 0.2730722188949585,
      "num_input_tokens_seen": 640899440,
      "step": 10555,
      "train_runtime": 137716.4348,
      "train_tokens_per_second": 4653.761
    },
    {
      "epoch": 1.346938775510204,
      "grad_norm": 0.5847915314701873,
      "learning_rate": 2.408893878266355e-06,
      "loss": 0.26175551414489745,
      "num_input_tokens_seen": 641202424,
      "step": 10560,
      "train_runtime": 137776.9655,
      "train_tokens_per_second": 4653.916
    },
    {
      "epoch": 1.347576530612245,
      "grad_norm": 0.6487385023535115,
      "learning_rate": 2.404611320001686e-06,
      "loss": 0.263177752494812,
      "num_input_tokens_seen": 641499128,
      "step": 10565,
      "train_runtime": 137851.8385,
      "train_tokens_per_second": 4653.541
    },
    {
      "epoch": 1.3482142857142856,
      "grad_norm": 0.6904179368056829,
      "learning_rate": 2.4003313663908182e-06,
      "loss": 0.2511106491088867,
      "num_input_tokens_seen": 641811424,
      "step": 10570,
      "train_runtime": 137918.2592,
      "train_tokens_per_second": 4653.564
    },
    {
      "epoch": 1.3488520408163265,
      "grad_norm": 0.7734205389206501,
      "learning_rate": 2.3960540217289856e-06,
      "loss": 0.2438887119293213,
      "num_input_tokens_seen": 642110896,
      "step": 10575,
      "train_runtime": 137975.1196,
      "train_tokens_per_second": 4653.817
    },
    {
      "epoch": 1.3494897959183674,
      "grad_norm": 0.6849424327462471,
      "learning_rate": 2.3917792903088013e-06,
      "loss": 0.2553779125213623,
      "num_input_tokens_seen": 642416424,
      "step": 10580,
      "train_runtime": 138026.6713,
      "train_tokens_per_second": 4654.292
    },
    {
      "epoch": 1.350127551020408,
      "grad_norm": 0.682346840735774,
      "learning_rate": 2.387507176420256e-06,
      "loss": 0.2522465229034424,
      "num_input_tokens_seen": 642710552,
      "step": 10585,
      "train_runtime": 138101.3125,
      "train_tokens_per_second": 4653.906
    },
    {
      "epoch": 1.350765306122449,
      "grad_norm": 0.680619519077353,
      "learning_rate": 2.383237684350716e-06,
      "loss": 0.24289708137512206,
      "num_input_tokens_seen": 643003000,
      "step": 10590,
      "train_runtime": 138167.4873,
      "train_tokens_per_second": 4653.794
    },
    {
      "epoch": 1.3514030612244898,
      "grad_norm": 0.7758551088163684,
      "learning_rate": 2.378970818384914e-06,
      "loss": 0.2685849666595459,
      "num_input_tokens_seen": 643309952,
      "step": 10595,
      "train_runtime": 138237.0251,
      "train_tokens_per_second": 4653.673
    },
    {
      "epoch": 1.3520408163265305,
      "grad_norm": 0.6099978672505727,
      "learning_rate": 2.374706582804948e-06,
      "loss": 0.2731654167175293,
      "num_input_tokens_seen": 643622504,
      "step": 10600,
      "train_runtime": 138308.5133,
      "train_tokens_per_second": 4653.528
    },
    {
      "epoch": 1.3526785714285714,
      "grad_norm": 0.6176015901979797,
      "learning_rate": 2.3704449818902763e-06,
      "loss": 0.23430299758911133,
      "num_input_tokens_seen": 643922296,
      "step": 10605,
      "train_runtime": 138371.2462,
      "train_tokens_per_second": 4653.585
    },
    {
      "epoch": 1.3533163265306123,
      "grad_norm": 0.6312026294529185,
      "learning_rate": 2.366186019917713e-06,
      "loss": 0.26416237354278566,
      "num_input_tokens_seen": 644226648,
      "step": 10610,
      "train_runtime": 138447.6608,
      "train_tokens_per_second": 4653.214
    },
    {
      "epoch": 1.353954081632653,
      "grad_norm": 0.5697447749801084,
      "learning_rate": 2.3619297011614236e-06,
      "loss": 0.24305322170257568,
      "num_input_tokens_seen": 644529424,
      "step": 10615,
      "train_runtime": 138517.5458,
      "train_tokens_per_second": 4653.053
    },
    {
      "epoch": 1.3545918367346939,
      "grad_norm": 0.703943199557105,
      "learning_rate": 2.3576760298929236e-06,
      "loss": 0.27020907402038574,
      "num_input_tokens_seen": 644823984,
      "step": 10620,
      "train_runtime": 138580.3714,
      "train_tokens_per_second": 4653.069
    },
    {
      "epoch": 1.3552295918367347,
      "grad_norm": 0.6371976780201754,
      "learning_rate": 2.353425010381063e-06,
      "loss": 0.2463529586791992,
      "num_input_tokens_seen": 645127368,
      "step": 10625,
      "train_runtime": 138648.3282,
      "train_tokens_per_second": 4652.976
    },
    {
      "epoch": 1.3558673469387754,
      "grad_norm": 0.6849575221807079,
      "learning_rate": 2.349176646892045e-06,
      "loss": 0.2500694990158081,
      "num_input_tokens_seen": 645431280,
      "step": 10630,
      "train_runtime": 138714.7105,
      "train_tokens_per_second": 4652.94
    },
    {
      "epoch": 1.3565051020408163,
      "grad_norm": 0.6739018605585745,
      "learning_rate": 2.344930943689393e-06,
      "loss": 0.25265026092529297,
      "num_input_tokens_seen": 645752504,
      "step": 10635,
      "train_runtime": 138767.4436,
      "train_tokens_per_second": 4653.487
    },
    {
      "epoch": 1.3571428571428572,
      "grad_norm": 0.7320819590462384,
      "learning_rate": 2.3406879050339664e-06,
      "loss": 0.25464611053466796,
      "num_input_tokens_seen": 646053008,
      "step": 10640,
      "train_runtime": 138836.5189,
      "train_tokens_per_second": 4653.336
    },
    {
      "epoch": 1.3577806122448979,
      "grad_norm": 0.6326393653192357,
      "learning_rate": 2.3364475351839573e-06,
      "loss": 0.2856419086456299,
      "num_input_tokens_seen": 646363800,
      "step": 10645,
      "train_runtime": 138902.8072,
      "train_tokens_per_second": 4653.353
    },
    {
      "epoch": 1.3584183673469388,
      "grad_norm": 0.6238707170767187,
      "learning_rate": 2.3322098383948666e-06,
      "loss": 0.25574593544006347,
      "num_input_tokens_seen": 646662400,
      "step": 10650,
      "train_runtime": 138981.0839,
      "train_tokens_per_second": 4652.881
    },
    {
      "epoch": 1.3590561224489797,
      "grad_norm": 0.7119462373413934,
      "learning_rate": 2.327974818919521e-06,
      "loss": 0.2599836826324463,
      "num_input_tokens_seen": 646961552,
      "step": 10655,
      "train_runtime": 139043.9753,
      "train_tokens_per_second": 4652.928
    },
    {
      "epoch": 1.3596938775510203,
      "grad_norm": 0.6576389455281506,
      "learning_rate": 2.323742481008059e-06,
      "loss": 0.26797871589660643,
      "num_input_tokens_seen": 647269032,
      "step": 10660,
      "train_runtime": 139098.9088,
      "train_tokens_per_second": 4653.301
    },
    {
      "epoch": 1.3603316326530612,
      "grad_norm": 0.6732629641000283,
      "learning_rate": 2.3195128289079265e-06,
      "loss": 0.2578679323196411,
      "num_input_tokens_seen": 647568480,
      "step": 10665,
      "train_runtime": 139159.9933,
      "train_tokens_per_second": 4653.41
    },
    {
      "epoch": 1.3609693877551021,
      "grad_norm": 0.6746025544320785,
      "learning_rate": 2.315285866863875e-06,
      "loss": 0.27208359241485597,
      "num_input_tokens_seen": 647874144,
      "step": 10670,
      "train_runtime": 139222.0315,
      "train_tokens_per_second": 4653.532
    },
    {
      "epoch": 1.3616071428571428,
      "grad_norm": 0.9209266628908638,
      "learning_rate": 2.3110615991179556e-06,
      "loss": 0.24118657112121583,
      "num_input_tokens_seen": 648176336,
      "step": 10675,
      "train_runtime": 139276.9936,
      "train_tokens_per_second": 4653.865
    },
    {
      "epoch": 1.3622448979591837,
      "grad_norm": 0.6300570060400691,
      "learning_rate": 2.3068400299095172e-06,
      "loss": 0.2557355880737305,
      "num_input_tokens_seen": 648491096,
      "step": 10680,
      "train_runtime": 139343.1151,
      "train_tokens_per_second": 4653.916
    },
    {
      "epoch": 1.3628826530612246,
      "grad_norm": 0.6171430512983377,
      "learning_rate": 2.302621163475198e-06,
      "loss": 0.24427404403686523,
      "num_input_tokens_seen": 648799472,
      "step": 10685,
      "train_runtime": 139413.2653,
      "train_tokens_per_second": 4653.786
    },
    {
      "epoch": 1.3635204081632653,
      "grad_norm": 0.7335552082699521,
      "learning_rate": 2.2984050040489266e-06,
      "loss": 0.2685335636138916,
      "num_input_tokens_seen": 649093176,
      "step": 10690,
      "train_runtime": 139479.0576,
      "train_tokens_per_second": 4653.696
    },
    {
      "epoch": 1.3641581632653061,
      "grad_norm": 0.7072866477484043,
      "learning_rate": 2.2941915558619123e-06,
      "loss": 0.24063777923583984,
      "num_input_tokens_seen": 649381256,
      "step": 10695,
      "train_runtime": 139542.8225,
      "train_tokens_per_second": 4653.634
    },
    {
      "epoch": 1.364795918367347,
      "grad_norm": 0.6944594337005354,
      "learning_rate": 2.2899808231426447e-06,
      "loss": 0.25992212295532224,
      "num_input_tokens_seen": 649682888,
      "step": 10700,
      "train_runtime": 139609.0242,
      "train_tokens_per_second": 4653.588
    },
    {
      "epoch": 1.3654336734693877,
      "grad_norm": 0.6766058412661532,
      "learning_rate": 2.28577281011689e-06,
      "loss": 0.26885318756103516,
      "num_input_tokens_seen": 649995064,
      "step": 10705,
      "train_runtime": 139660.4965,
      "train_tokens_per_second": 4654.108
    },
    {
      "epoch": 1.3660714285714286,
      "grad_norm": 0.7018897965557243,
      "learning_rate": 2.281567521007678e-06,
      "loss": 0.23794171810150147,
      "num_input_tokens_seen": 650299792,
      "step": 10710,
      "train_runtime": 139727.3155,
      "train_tokens_per_second": 4654.063
    },
    {
      "epoch": 1.3667091836734695,
      "grad_norm": 0.6458270881545367,
      "learning_rate": 2.277364960035316e-06,
      "loss": 0.24845671653747559,
      "num_input_tokens_seen": 650599768,
      "step": 10715,
      "train_runtime": 139795.6731,
      "train_tokens_per_second": 4653.934
    },
    {
      "epoch": 1.3673469387755102,
      "grad_norm": 0.6378253214235245,
      "learning_rate": 2.2731651314173665e-06,
      "loss": 0.2722214937210083,
      "num_input_tokens_seen": 650901360,
      "step": 10720,
      "train_runtime": 139861.637,
      "train_tokens_per_second": 4653.895
    },
    {
      "epoch": 1.367984693877551,
      "grad_norm": 0.6439971278285374,
      "learning_rate": 2.2689680393686457e-06,
      "loss": 0.25092372894287107,
      "num_input_tokens_seen": 651208024,
      "step": 10725,
      "train_runtime": 139931.0735,
      "train_tokens_per_second": 4653.777
    },
    {
      "epoch": 1.368622448979592,
      "grad_norm": 0.67511136748276,
      "learning_rate": 2.2647736881012337e-06,
      "loss": 0.26037437915802003,
      "num_input_tokens_seen": 651509616,
      "step": 10730,
      "train_runtime": 139994.8024,
      "train_tokens_per_second": 4653.813
    },
    {
      "epoch": 1.3692602040816326,
      "grad_norm": 0.6325514970286513,
      "learning_rate": 2.2605820818244563e-06,
      "loss": 0.2595388650894165,
      "num_input_tokens_seen": 651812552,
      "step": 10735,
      "train_runtime": 140062.0458,
      "train_tokens_per_second": 4653.741
    },
    {
      "epoch": 1.3698979591836735,
      "grad_norm": 0.666538231347289,
      "learning_rate": 2.256393224744878e-06,
      "loss": 0.247703218460083,
      "num_input_tokens_seen": 652123936,
      "step": 10740,
      "train_runtime": 140115.296,
      "train_tokens_per_second": 4654.195
    },
    {
      "epoch": 1.3705357142857144,
      "grad_norm": 0.6245066028156089,
      "learning_rate": 2.252207121066311e-06,
      "loss": 0.2506424903869629,
      "num_input_tokens_seen": 652436208,
      "step": 10745,
      "train_runtime": 140172.2333,
      "train_tokens_per_second": 4654.532
    },
    {
      "epoch": 1.371173469387755,
      "grad_norm": 0.6897518255700242,
      "learning_rate": 2.248023774989804e-06,
      "loss": 0.2539947032928467,
      "num_input_tokens_seen": 652728840,
      "step": 10750,
      "train_runtime": 140244.505,
      "train_tokens_per_second": 4654.22
    },
    {
      "epoch": 1.371811224489796,
      "grad_norm": 0.6652062345999039,
      "learning_rate": 2.2438431907136365e-06,
      "loss": 0.25039238929748536,
      "num_input_tokens_seen": 653037248,
      "step": 10755,
      "train_runtime": 140317.5554,
      "train_tokens_per_second": 4653.995
    },
    {
      "epoch": 1.3724489795918369,
      "grad_norm": 0.6998307506209887,
      "learning_rate": 2.2396653724333167e-06,
      "loss": 0.2588667392730713,
      "num_input_tokens_seen": 653331840,
      "step": 10760,
      "train_runtime": 140382.2853,
      "train_tokens_per_second": 4653.948
    },
    {
      "epoch": 1.3730867346938775,
      "grad_norm": 0.7066375590823599,
      "learning_rate": 2.235490324341577e-06,
      "loss": 0.25351243019104003,
      "num_input_tokens_seen": 653619864,
      "step": 10765,
      "train_runtime": 140447.3579,
      "train_tokens_per_second": 4653.842
    },
    {
      "epoch": 1.3737244897959184,
      "grad_norm": 0.6894289922417881,
      "learning_rate": 2.2313180506283695e-06,
      "loss": 0.24595603942871094,
      "num_input_tokens_seen": 653918392,
      "step": 10770,
      "train_runtime": 140515.882,
      "train_tokens_per_second": 4653.697
    },
    {
      "epoch": 1.374362244897959,
      "grad_norm": 0.6348190895126634,
      "learning_rate": 2.227148555480863e-06,
      "loss": 0.25271408557891845,
      "num_input_tokens_seen": 654231808,
      "step": 10775,
      "train_runtime": 140578.6986,
      "train_tokens_per_second": 4653.847
    },
    {
      "epoch": 1.375,
      "grad_norm": 0.7041985937569952,
      "learning_rate": 2.222981843083436e-06,
      "loss": 0.2630213022232056,
      "num_input_tokens_seen": 654537768,
      "step": 10780,
      "train_runtime": 140648.6004,
      "train_tokens_per_second": 4653.71
    },
    {
      "epoch": 1.375637755102041,
      "grad_norm": 0.6473438090805954,
      "learning_rate": 2.2188179176176767e-06,
      "loss": 0.24607279300689697,
      "num_input_tokens_seen": 654839688,
      "step": 10785,
      "train_runtime": 140706.8133,
      "train_tokens_per_second": 4653.93
    },
    {
      "epoch": 1.3762755102040816,
      "grad_norm": 0.7402374340188662,
      "learning_rate": 2.2146567832623737e-06,
      "loss": 0.2657738924026489,
      "num_input_tokens_seen": 655155736,
      "step": 10790,
      "train_runtime": 140771.7221,
      "train_tokens_per_second": 4654.029
    },
    {
      "epoch": 1.3769132653061225,
      "grad_norm": 0.6499488706520828,
      "learning_rate": 2.2104984441935167e-06,
      "loss": 0.26737127304077146,
      "num_input_tokens_seen": 655448320,
      "step": 10795,
      "train_runtime": 140838.6502,
      "train_tokens_per_second": 4653.895
    },
    {
      "epoch": 1.3775510204081631,
      "grad_norm": 0.7711023442574942,
      "learning_rate": 2.20634290458429e-06,
      "loss": 0.2794210910797119,
      "num_input_tokens_seen": 655752952,
      "step": 10800,
      "train_runtime": 140908.7662,
      "train_tokens_per_second": 4653.741
    },
    {
      "epoch": 1.378188775510204,
      "grad_norm": 0.6198423134424618,
      "learning_rate": 2.2021901686050684e-06,
      "loss": 0.2476513147354126,
      "num_input_tokens_seen": 656054200,
      "step": 10805,
      "train_runtime": 140970.6449,
      "train_tokens_per_second": 4653.836
    },
    {
      "epoch": 1.378826530612245,
      "grad_norm": 0.6489766549170461,
      "learning_rate": 2.1980402404234074e-06,
      "loss": 0.26006624698638914,
      "num_input_tokens_seen": 656343272,
      "step": 10810,
      "train_runtime": 141034.9838,
      "train_tokens_per_second": 4653.762
    },
    {
      "epoch": 1.3794642857142856,
      "grad_norm": 0.6124169753289999,
      "learning_rate": 2.1938931242040557e-06,
      "loss": 0.23834261894226075,
      "num_input_tokens_seen": 656653912,
      "step": 10815,
      "train_runtime": 141102.7794,
      "train_tokens_per_second": 4653.728
    },
    {
      "epoch": 1.3801020408163265,
      "grad_norm": 0.6864530445874658,
      "learning_rate": 2.1897488241089336e-06,
      "loss": 0.24825513362884521,
      "num_input_tokens_seen": 656939456,
      "step": 10820,
      "train_runtime": 141170.4554,
      "train_tokens_per_second": 4653.519
    },
    {
      "epoch": 1.3807397959183674,
      "grad_norm": 0.6558066319681096,
      "learning_rate": 2.1856073442971322e-06,
      "loss": 0.23897290229797363,
      "num_input_tokens_seen": 657245400,
      "step": 10825,
      "train_runtime": 141246.3502,
      "train_tokens_per_second": 4653.185
    },
    {
      "epoch": 1.381377551020408,
      "grad_norm": 0.6315243823307106,
      "learning_rate": 2.181468688924916e-06,
      "loss": 0.25669231414794924,
      "num_input_tokens_seen": 657542320,
      "step": 10830,
      "train_runtime": 141315.8761,
      "train_tokens_per_second": 4652.997
    },
    {
      "epoch": 1.382015306122449,
      "grad_norm": 0.679056503801347,
      "learning_rate": 2.1773328621457196e-06,
      "loss": 0.266129207611084,
      "num_input_tokens_seen": 657848208,
      "step": 10835,
      "train_runtime": 141388.7871,
      "train_tokens_per_second": 4652.761
    },
    {
      "epoch": 1.3826530612244898,
      "grad_norm": 0.6481972340374259,
      "learning_rate": 2.1731998681101293e-06,
      "loss": 0.2581286907196045,
      "num_input_tokens_seen": 658141328,
      "step": 10840,
      "train_runtime": 141452.578,
      "train_tokens_per_second": 4652.735
    },
    {
      "epoch": 1.3832908163265305,
      "grad_norm": 0.5779596087218133,
      "learning_rate": 2.169069710965892e-06,
      "loss": 0.2503084659576416,
      "num_input_tokens_seen": 658449416,
      "step": 10845,
      "train_runtime": 141518.4324,
      "train_tokens_per_second": 4652.747
    },
    {
      "epoch": 1.3839285714285714,
      "grad_norm": 0.6133966098993952,
      "learning_rate": 2.1649423948579157e-06,
      "loss": 0.2516158580780029,
      "num_input_tokens_seen": 658752104,
      "step": 10850,
      "train_runtime": 141584.6078,
      "train_tokens_per_second": 4652.71
    },
    {
      "epoch": 1.3845663265306123,
      "grad_norm": 0.767753818892085,
      "learning_rate": 2.160817923928244e-06,
      "loss": 0.24149253368377685,
      "num_input_tokens_seen": 659068832,
      "step": 10855,
      "train_runtime": 141644.9478,
      "train_tokens_per_second": 4652.964
    },
    {
      "epoch": 1.385204081632653,
      "grad_norm": 0.592673489479143,
      "learning_rate": 2.156696302316074e-06,
      "loss": 0.24596431255340576,
      "num_input_tokens_seen": 659370344,
      "step": 10860,
      "train_runtime": 141708.6494,
      "train_tokens_per_second": 4653.0
    },
    {
      "epoch": 1.3858418367346939,
      "grad_norm": 0.6587773413475897,
      "learning_rate": 2.1525775341577404e-06,
      "loss": 0.2650430679321289,
      "num_input_tokens_seen": 659675272,
      "step": 10865,
      "train_runtime": 141780.046,
      "train_tokens_per_second": 4652.808
    },
    {
      "epoch": 1.3864795918367347,
      "grad_norm": 0.6304895019907499,
      "learning_rate": 2.1484616235867152e-06,
      "loss": 0.2624351501464844,
      "num_input_tokens_seen": 659984064,
      "step": 10870,
      "train_runtime": 141839.4022,
      "train_tokens_per_second": 4653.038
    },
    {
      "epoch": 1.3871173469387754,
      "grad_norm": 0.7080954736257979,
      "learning_rate": 2.1443485747336014e-06,
      "loss": 0.2529263973236084,
      "num_input_tokens_seen": 660283112,
      "step": 10875,
      "train_runtime": 141912.2568,
      "train_tokens_per_second": 4652.756
    },
    {
      "epoch": 1.3877551020408163,
      "grad_norm": 0.7192499873806749,
      "learning_rate": 2.140238391726131e-06,
      "loss": 0.23541016578674318,
      "num_input_tokens_seen": 660575688,
      "step": 10880,
      "train_runtime": 141975.9961,
      "train_tokens_per_second": 4652.728
    },
    {
      "epoch": 1.3883928571428572,
      "grad_norm": 0.6555123158847614,
      "learning_rate": 2.1361310786891596e-06,
      "loss": 0.24966943264007568,
      "num_input_tokens_seen": 660891000,
      "step": 10885,
      "train_runtime": 142026.5793,
      "train_tokens_per_second": 4653.291
    },
    {
      "epoch": 1.3890306122448979,
      "grad_norm": 0.6341630278590069,
      "learning_rate": 2.1320266397446655e-06,
      "loss": 0.24406077861785888,
      "num_input_tokens_seen": 661181296,
      "step": 10890,
      "train_runtime": 142100.5999,
      "train_tokens_per_second": 4652.91
    },
    {
      "epoch": 1.3896683673469388,
      "grad_norm": 0.6522383502190727,
      "learning_rate": 2.1279250790117327e-06,
      "loss": 0.24395380020141602,
      "num_input_tokens_seen": 661486088,
      "step": 10895,
      "train_runtime": 142172.7724,
      "train_tokens_per_second": 4652.692
    },
    {
      "epoch": 1.3903061224489797,
      "grad_norm": 0.6315288970380799,
      "learning_rate": 2.123826400606571e-06,
      "loss": 0.24505012035369872,
      "num_input_tokens_seen": 661776696,
      "step": 10900,
      "train_runtime": 142246.1174,
      "train_tokens_per_second": 4652.336
    },
    {
      "epoch": 1.3909438775510203,
      "grad_norm": 0.7412865776148488,
      "learning_rate": 2.119730608642489e-06,
      "loss": 0.2468052625656128,
      "num_input_tokens_seen": 662075896,
      "step": 10905,
      "train_runtime": 142312.876,
      "train_tokens_per_second": 4652.256
    },
    {
      "epoch": 1.3915816326530612,
      "grad_norm": 0.7447682594698796,
      "learning_rate": 2.1156377072298972e-06,
      "loss": 0.2510690450668335,
      "num_input_tokens_seen": 662382448,
      "step": 10910,
      "train_runtime": 142384.1467,
      "train_tokens_per_second": 4652.08
    },
    {
      "epoch": 1.3922193877551021,
      "grad_norm": 0.6465834278051001,
      "learning_rate": 2.111547700476308e-06,
      "loss": 0.26249260902404786,
      "num_input_tokens_seen": 662690784,
      "step": 10915,
      "train_runtime": 142439.5463,
      "train_tokens_per_second": 4652.435
    },
    {
      "epoch": 1.3928571428571428,
      "grad_norm": 0.7033734542030893,
      "learning_rate": 2.107460592486333e-06,
      "loss": 0.268072772026062,
      "num_input_tokens_seen": 662980848,
      "step": 10920,
      "train_runtime": 142504.2991,
      "train_tokens_per_second": 4652.357
    },
    {
      "epoch": 1.3934948979591837,
      "grad_norm": 0.6292468927908194,
      "learning_rate": 2.1033763873616675e-06,
      "loss": 0.24447717666625976,
      "num_input_tokens_seen": 663286584,
      "step": 10925,
      "train_runtime": 142579.6955,
      "train_tokens_per_second": 4652.041
    },
    {
      "epoch": 1.3941326530612246,
      "grad_norm": 0.5931620942430261,
      "learning_rate": 2.0992950892010948e-06,
      "loss": 0.2554739475250244,
      "num_input_tokens_seen": 663595984,
      "step": 10930,
      "train_runtime": 142638.9585,
      "train_tokens_per_second": 4652.277
    },
    {
      "epoch": 1.3947704081632653,
      "grad_norm": 0.6163909609686915,
      "learning_rate": 2.0952167021004887e-06,
      "loss": 0.2546344995498657,
      "num_input_tokens_seen": 663913360,
      "step": 10935,
      "train_runtime": 142709.5336,
      "train_tokens_per_second": 4652.2
    },
    {
      "epoch": 1.3954081632653061,
      "grad_norm": 0.5882756757092259,
      "learning_rate": 2.0911412301527895e-06,
      "loss": 0.248978590965271,
      "num_input_tokens_seen": 664212632,
      "step": 10940,
      "train_runtime": 142778.2066,
      "train_tokens_per_second": 4652.059
    },
    {
      "epoch": 1.396045918367347,
      "grad_norm": 0.6274000873800156,
      "learning_rate": 2.08706867744802e-06,
      "loss": 0.24985527992248535,
      "num_input_tokens_seen": 664511248,
      "step": 10945,
      "train_runtime": 142842.6126,
      "train_tokens_per_second": 4652.052
    },
    {
      "epoch": 1.3966836734693877,
      "grad_norm": 0.6184724192923571,
      "learning_rate": 2.0829990480732715e-06,
      "loss": 0.25580124855041503,
      "num_input_tokens_seen": 664799920,
      "step": 10950,
      "train_runtime": 142908.8862,
      "train_tokens_per_second": 4651.915
    },
    {
      "epoch": 1.3973214285714286,
      "grad_norm": 0.7150837740716233,
      "learning_rate": 2.0789323461127015e-06,
      "loss": 0.258500075340271,
      "num_input_tokens_seen": 665103792,
      "step": 10955,
      "train_runtime": 142971.6498,
      "train_tokens_per_second": 4651.998
    },
    {
      "epoch": 1.3979591836734695,
      "grad_norm": 0.7029402338819539,
      "learning_rate": 2.0748685756475286e-06,
      "loss": 0.2610204219818115,
      "num_input_tokens_seen": 665399664,
      "step": 10960,
      "train_runtime": 143034.9047,
      "train_tokens_per_second": 4652.009
    },
    {
      "epoch": 1.3985969387755102,
      "grad_norm": 0.6880501359533734,
      "learning_rate": 2.0708077407560305e-06,
      "loss": 0.24911460876464844,
      "num_input_tokens_seen": 665695688,
      "step": 10965,
      "train_runtime": 143093.5604,
      "train_tokens_per_second": 4652.171
    },
    {
      "epoch": 1.399234693877551,
      "grad_norm": 0.7276087711781211,
      "learning_rate": 2.06674984551354e-06,
      "loss": 0.2696921110153198,
      "num_input_tokens_seen": 666000336,
      "step": 10970,
      "train_runtime": 143150.2238,
      "train_tokens_per_second": 4652.458
    },
    {
      "epoch": 1.399872448979592,
      "grad_norm": 0.6611084736675233,
      "learning_rate": 2.062694893992436e-06,
      "loss": 0.24796009063720703,
      "num_input_tokens_seen": 666303304,
      "step": 10975,
      "train_runtime": 143209.2762,
      "train_tokens_per_second": 4652.655
    },
    {
      "epoch": 1.4005102040816326,
      "grad_norm": 0.6789059871274964,
      "learning_rate": 2.0586428902621474e-06,
      "loss": 0.24918723106384277,
      "num_input_tokens_seen": 666622728,
      "step": 10980,
      "train_runtime": 143254.657,
      "train_tokens_per_second": 4653.411
    },
    {
      "epoch": 1.4011479591836735,
      "grad_norm": 0.6725874595012141,
      "learning_rate": 2.0545938383891427e-06,
      "loss": 0.2662821292877197,
      "num_input_tokens_seen": 666917368,
      "step": 10985,
      "train_runtime": 143316.1471,
      "train_tokens_per_second": 4653.47
    },
    {
      "epoch": 1.4017857142857144,
      "grad_norm": 0.653781330360485,
      "learning_rate": 2.050547742436929e-06,
      "loss": 0.23631629943847657,
      "num_input_tokens_seen": 667217976,
      "step": 10990,
      "train_runtime": 143388.8015,
      "train_tokens_per_second": 4653.208
    },
    {
      "epoch": 1.402423469387755,
      "grad_norm": 0.6200075478649029,
      "learning_rate": 2.0465046064660437e-06,
      "loss": 0.2509471416473389,
      "num_input_tokens_seen": 667515672,
      "step": 10995,
      "train_runtime": 143458.2482,
      "train_tokens_per_second": 4653.031
    },
    {
      "epoch": 1.403061224489796,
      "grad_norm": 0.604581929551146,
      "learning_rate": 2.0424644345340554e-06,
      "loss": 0.24612851142883302,
      "num_input_tokens_seen": 667817512,
      "step": 11000,
      "train_runtime": 143523.1041,
      "train_tokens_per_second": 4653.031
    },
    {
      "epoch": 1.4036989795918369,
      "grad_norm": 0.678247158757741,
      "learning_rate": 2.038427230695565e-06,
      "loss": 0.24648277759552,
      "num_input_tokens_seen": 668120112,
      "step": 11005,
      "train_runtime": 143592.6223,
      "train_tokens_per_second": 4652.886
    },
    {
      "epoch": 1.4043367346938775,
      "grad_norm": 0.6286842547424614,
      "learning_rate": 2.034392999002182e-06,
      "loss": 0.2571901321411133,
      "num_input_tokens_seen": 668429272,
      "step": 11010,
      "train_runtime": 143668.67,
      "train_tokens_per_second": 4652.575
    },
    {
      "epoch": 1.4049744897959184,
      "grad_norm": 0.6550917364941298,
      "learning_rate": 2.0303617435025407e-06,
      "loss": 0.24138829708099366,
      "num_input_tokens_seen": 668739184,
      "step": 11015,
      "train_runtime": 143739.0058,
      "train_tokens_per_second": 4652.454
    },
    {
      "epoch": 1.405612244897959,
      "grad_norm": 0.6006425158343822,
      "learning_rate": 2.0263334682422923e-06,
      "loss": 0.2457888126373291,
      "num_input_tokens_seen": 669040392,
      "step": 11020,
      "train_runtime": 143815.1956,
      "train_tokens_per_second": 4652.084
    },
    {
      "epoch": 1.40625,
      "grad_norm": 0.7155943726564788,
      "learning_rate": 2.0223081772640867e-06,
      "loss": 0.2669305086135864,
      "num_input_tokens_seen": 669337248,
      "step": 11025,
      "train_runtime": 143873.3566,
      "train_tokens_per_second": 4652.267
    },
    {
      "epoch": 1.406887755102041,
      "grad_norm": 0.5918915902802782,
      "learning_rate": 2.0182858746075873e-06,
      "loss": 0.24875364303588868,
      "num_input_tokens_seen": 669647448,
      "step": 11030,
      "train_runtime": 143939.8178,
      "train_tokens_per_second": 4652.274
    },
    {
      "epoch": 1.4075255102040816,
      "grad_norm": 0.6224308265696982,
      "learning_rate": 2.014266564309455e-06,
      "loss": 0.2550758123397827,
      "num_input_tokens_seen": 669939544,
      "step": 11035,
      "train_runtime": 144002.6143,
      "train_tokens_per_second": 4652.273
    },
    {
      "epoch": 1.4081632653061225,
      "grad_norm": 0.7045550189140467,
      "learning_rate": 2.010250250403349e-06,
      "loss": 0.2659616947174072,
      "num_input_tokens_seen": 670236248,
      "step": 11040,
      "train_runtime": 144060.0709,
      "train_tokens_per_second": 4652.478
    },
    {
      "epoch": 1.4088010204081631,
      "grad_norm": 0.6146701009290687,
      "learning_rate": 2.00623693691992e-06,
      "loss": 0.2394233226776123,
      "num_input_tokens_seen": 670550248,
      "step": 11045,
      "train_runtime": 144126.109,
      "train_tokens_per_second": 4652.524
    },
    {
      "epoch": 1.409438775510204,
      "grad_norm": 0.6217305449765971,
      "learning_rate": 2.0022266278868086e-06,
      "loss": 0.2595332145690918,
      "num_input_tokens_seen": 670855080,
      "step": 11050,
      "train_runtime": 144187.8508,
      "train_tokens_per_second": 4652.646
    },
    {
      "epoch": 1.410076530612245,
      "grad_norm": 0.7312246103580766,
      "learning_rate": 1.9982193273286404e-06,
      "loss": 0.2402665615081787,
      "num_input_tokens_seen": 671159224,
      "step": 11055,
      "train_runtime": 144237.0253,
      "train_tokens_per_second": 4653.169
    },
    {
      "epoch": 1.4107142857142856,
      "grad_norm": 0.6980968981401449,
      "learning_rate": 1.9942150392670207e-06,
      "loss": 0.24990315437316896,
      "num_input_tokens_seen": 671456704,
      "step": 11060,
      "train_runtime": 144307.5285,
      "train_tokens_per_second": 4652.957
    },
    {
      "epoch": 1.4113520408163265,
      "grad_norm": 0.8088300387825774,
      "learning_rate": 1.990213767720533e-06,
      "loss": 0.2636767148971558,
      "num_input_tokens_seen": 671772664,
      "step": 11065,
      "train_runtime": 144366.5253,
      "train_tokens_per_second": 4653.244
    },
    {
      "epoch": 1.4119897959183674,
      "grad_norm": 0.6487090457377199,
      "learning_rate": 1.9862155167047324e-06,
      "loss": 0.26303236484527587,
      "num_input_tokens_seen": 672075568,
      "step": 11070,
      "train_runtime": 144433.1382,
      "train_tokens_per_second": 4653.195
    },
    {
      "epoch": 1.412627551020408,
      "grad_norm": 0.7040276445864658,
      "learning_rate": 1.982220290232143e-06,
      "loss": 0.2502783536911011,
      "num_input_tokens_seen": 672377576,
      "step": 11075,
      "train_runtime": 144500.5412,
      "train_tokens_per_second": 4653.115
    },
    {
      "epoch": 1.413265306122449,
      "grad_norm": 0.6810987286282223,
      "learning_rate": 1.9782280923122555e-06,
      "loss": 0.2444309711456299,
      "num_input_tokens_seen": 672674632,
      "step": 11080,
      "train_runtime": 144566.3023,
      "train_tokens_per_second": 4653.053
    },
    {
      "epoch": 1.4139030612244898,
      "grad_norm": 0.6463127960515297,
      "learning_rate": 1.9742389269515137e-06,
      "loss": 0.2458259105682373,
      "num_input_tokens_seen": 672988200,
      "step": 11085,
      "train_runtime": 144636.961,
      "train_tokens_per_second": 4652.948
    },
    {
      "epoch": 1.4145408163265305,
      "grad_norm": 0.6103837085208516,
      "learning_rate": 1.970252798153329e-06,
      "loss": 0.22437753677368164,
      "num_input_tokens_seen": 673289928,
      "step": 11090,
      "train_runtime": 144699.2516,
      "train_tokens_per_second": 4653.03
    },
    {
      "epoch": 1.4151785714285714,
      "grad_norm": 0.6710805945631285,
      "learning_rate": 1.9662697099180607e-06,
      "loss": 0.2727480888366699,
      "num_input_tokens_seen": 673609528,
      "step": 11095,
      "train_runtime": 144753.367,
      "train_tokens_per_second": 4653.498
    },
    {
      "epoch": 1.4158163265306123,
      "grad_norm": 0.6650466747935465,
      "learning_rate": 1.9622896662430098e-06,
      "loss": 0.24592394828796388,
      "num_input_tokens_seen": 673911976,
      "step": 11100,
      "train_runtime": 144816.7329,
      "train_tokens_per_second": 4653.55
    },
    {
      "epoch": 1.416454081632653,
      "grad_norm": 0.6881162712221625,
      "learning_rate": 1.9583126711224342e-06,
      "loss": 0.25740480422973633,
      "num_input_tokens_seen": 674217488,
      "step": 11105,
      "train_runtime": 144879.2297,
      "train_tokens_per_second": 4653.652
    },
    {
      "epoch": 1.4170918367346939,
      "grad_norm": 0.6304863826762142,
      "learning_rate": 1.954338728547526e-06,
      "loss": 0.25806460380554197,
      "num_input_tokens_seen": 674528768,
      "step": 11110,
      "train_runtime": 144947.6353,
      "train_tokens_per_second": 4653.603
    },
    {
      "epoch": 1.4177295918367347,
      "grad_norm": 0.5838760323917608,
      "learning_rate": 1.9503678425064097e-06,
      "loss": 0.26084744930267334,
      "num_input_tokens_seen": 674833976,
      "step": 11115,
      "train_runtime": 145021.509,
      "train_tokens_per_second": 4653.337
    },
    {
      "epoch": 1.4183673469387754,
      "grad_norm": 0.6044228807138501,
      "learning_rate": 1.9464000169841495e-06,
      "loss": 0.2530702829360962,
      "num_input_tokens_seen": 675144408,
      "step": 11120,
      "train_runtime": 145090.0515,
      "train_tokens_per_second": 4653.278
    },
    {
      "epoch": 1.4190051020408163,
      "grad_norm": 0.6851616617139263,
      "learning_rate": 1.942435255962736e-06,
      "loss": 0.2522618055343628,
      "num_input_tokens_seen": 675444376,
      "step": 11125,
      "train_runtime": 145158.6162,
      "train_tokens_per_second": 4653.147
    },
    {
      "epoch": 1.4196428571428572,
      "grad_norm": 0.659989561463371,
      "learning_rate": 1.938473563421082e-06,
      "loss": 0.2615769624710083,
      "num_input_tokens_seen": 675753496,
      "step": 11130,
      "train_runtime": 145222.3305,
      "train_tokens_per_second": 4653.234
    },
    {
      "epoch": 1.4202806122448979,
      "grad_norm": 0.610403521010447,
      "learning_rate": 1.934514943335024e-06,
      "loss": 0.25546267032623293,
      "num_input_tokens_seen": 676051376,
      "step": 11135,
      "train_runtime": 145295.6783,
      "train_tokens_per_second": 4652.935
    },
    {
      "epoch": 1.4209183673469388,
      "grad_norm": 0.6489582829306019,
      "learning_rate": 1.930559399677313e-06,
      "loss": 0.26701812744140624,
      "num_input_tokens_seen": 676348080,
      "step": 11140,
      "train_runtime": 145370.4233,
      "train_tokens_per_second": 4652.584
    },
    {
      "epoch": 1.4215561224489797,
      "grad_norm": 0.696342377331813,
      "learning_rate": 1.9266069364176144e-06,
      "loss": 0.26505365371704104,
      "num_input_tokens_seen": 676648304,
      "step": 11145,
      "train_runtime": 145433.6569,
      "train_tokens_per_second": 4652.625
    },
    {
      "epoch": 1.4221938775510203,
      "grad_norm": 0.6049080136284425,
      "learning_rate": 1.9226575575225002e-06,
      "loss": 0.26203341484069825,
      "num_input_tokens_seen": 676951648,
      "step": 11150,
      "train_runtime": 145490.8333,
      "train_tokens_per_second": 4652.882
    },
    {
      "epoch": 1.4228316326530612,
      "grad_norm": 0.8259829756622332,
      "learning_rate": 1.918711266955448e-06,
      "loss": 0.26873183250427246,
      "num_input_tokens_seen": 677263584,
      "step": 11155,
      "train_runtime": 145557.7369,
      "train_tokens_per_second": 4652.886
    },
    {
      "epoch": 1.4234693877551021,
      "grad_norm": 0.65287980347922,
      "learning_rate": 1.914768068676836e-06,
      "loss": 0.24950363636016845,
      "num_input_tokens_seen": 677572232,
      "step": 11160,
      "train_runtime": 145630.1619,
      "train_tokens_per_second": 4652.692
    },
    {
      "epoch": 1.4241071428571428,
      "grad_norm": 0.7017597610220788,
      "learning_rate": 1.9108279666439394e-06,
      "loss": 0.2430121660232544,
      "num_input_tokens_seen": 677882176,
      "step": 11165,
      "train_runtime": 145695.7154,
      "train_tokens_per_second": 4652.726
    },
    {
      "epoch": 1.4247448979591837,
      "grad_norm": 0.6412105711113006,
      "learning_rate": 1.9068909648109258e-06,
      "loss": 0.270355224609375,
      "num_input_tokens_seen": 678173632,
      "step": 11170,
      "train_runtime": 145760.5805,
      "train_tokens_per_second": 4652.655
    },
    {
      "epoch": 1.4253826530612246,
      "grad_norm": 0.6807194539663797,
      "learning_rate": 1.9029570671288511e-06,
      "loss": 0.25338528156280515,
      "num_input_tokens_seen": 678474600,
      "step": 11175,
      "train_runtime": 145818.2662,
      "train_tokens_per_second": 4652.878
    },
    {
      "epoch": 1.4260204081632653,
      "grad_norm": 0.6991012794676906,
      "learning_rate": 1.899026277545658e-06,
      "loss": 0.25733702182769774,
      "num_input_tokens_seen": 678780864,
      "step": 11180,
      "train_runtime": 145883.5944,
      "train_tokens_per_second": 4652.894
    },
    {
      "epoch": 1.4266581632653061,
      "grad_norm": 0.6677068937815375,
      "learning_rate": 1.895098600006164e-06,
      "loss": 0.25646467208862306,
      "num_input_tokens_seen": 679088520,
      "step": 11185,
      "train_runtime": 145961.5774,
      "train_tokens_per_second": 4652.516
    },
    {
      "epoch": 1.427295918367347,
      "grad_norm": 0.6687378811532653,
      "learning_rate": 1.8911740384520717e-06,
      "loss": 0.25247950553894044,
      "num_input_tokens_seen": 679405536,
      "step": 11190,
      "train_runtime": 146027.8484,
      "train_tokens_per_second": 4652.575
    },
    {
      "epoch": 1.4279336734693877,
      "grad_norm": 0.638291105690982,
      "learning_rate": 1.8872525968219541e-06,
      "loss": 0.2619307041168213,
      "num_input_tokens_seen": 679708992,
      "step": 11195,
      "train_runtime": 146092.8956,
      "train_tokens_per_second": 4652.581
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.6510291957929055,
      "learning_rate": 1.8833342790512472e-06,
      "loss": 0.2534599304199219,
      "num_input_tokens_seen": 680014824,
      "step": 11200,
      "train_runtime": 146148.156,
      "train_tokens_per_second": 4652.914
    },
    {
      "epoch": 1.4292091836734695,
      "grad_norm": 0.6219358860383175,
      "learning_rate": 1.8794190890722575e-06,
      "loss": 0.25667262077331543,
      "num_input_tokens_seen": 680316552,
      "step": 11205,
      "train_runtime": 146209.7228,
      "train_tokens_per_second": 4653.019
    },
    {
      "epoch": 1.4298469387755102,
      "grad_norm": 0.6511492931917631,
      "learning_rate": 1.875507030814156e-06,
      "loss": 0.24242887496948243,
      "num_input_tokens_seen": 680617056,
      "step": 11210,
      "train_runtime": 146273.6837,
      "train_tokens_per_second": 4653.038
    },
    {
      "epoch": 1.430484693877551,
      "grad_norm": 0.6142662389207844,
      "learning_rate": 1.871598108202961e-06,
      "loss": 0.25613572597503664,
      "num_input_tokens_seen": 680913152,
      "step": 11215,
      "train_runtime": 146333.2509,
      "train_tokens_per_second": 4653.168
    },
    {
      "epoch": 1.431122448979592,
      "grad_norm": 0.6010152931977929,
      "learning_rate": 1.8676923251615503e-06,
      "loss": 0.24273557662963868,
      "num_input_tokens_seen": 681212248,
      "step": 11220,
      "train_runtime": 146400.5962,
      "train_tokens_per_second": 4653.07
    },
    {
      "epoch": 1.4317602040816326,
      "grad_norm": 0.7367522839094218,
      "learning_rate": 1.863789685609655e-06,
      "loss": 0.25536012649536133,
      "num_input_tokens_seen": 681517160,
      "step": 11225,
      "train_runtime": 146466.0992,
      "train_tokens_per_second": 4653.071
    },
    {
      "epoch": 1.4323979591836735,
      "grad_norm": 0.6386127756897391,
      "learning_rate": 1.8598901934638402e-06,
      "loss": 0.25140721797943116,
      "num_input_tokens_seen": 681824104,
      "step": 11230,
      "train_runtime": 146519.1214,
      "train_tokens_per_second": 4653.482
    },
    {
      "epoch": 1.4330357142857144,
      "grad_norm": 0.6559922692928403,
      "learning_rate": 1.8559938526375215e-06,
      "loss": 0.26719353199005125,
      "num_input_tokens_seen": 682111592,
      "step": 11235,
      "train_runtime": 146583.941,
      "train_tokens_per_second": 4653.386
    },
    {
      "epoch": 1.433673469387755,
      "grad_norm": 0.6614059379119014,
      "learning_rate": 1.8521006670409481e-06,
      "loss": 0.2607134819030762,
      "num_input_tokens_seen": 682419768,
      "step": 11240,
      "train_runtime": 146649.2813,
      "train_tokens_per_second": 4653.414
    },
    {
      "epoch": 1.434311224489796,
      "grad_norm": 0.6255632269000909,
      "learning_rate": 1.8482106405812046e-06,
      "loss": 0.2534268856048584,
      "num_input_tokens_seen": 682722376,
      "step": 11245,
      "train_runtime": 146715.2995,
      "train_tokens_per_second": 4653.382
    },
    {
      "epoch": 1.4349489795918369,
      "grad_norm": 0.628044852794189,
      "learning_rate": 1.8443237771622037e-06,
      "loss": 0.261114501953125,
      "num_input_tokens_seen": 683030000,
      "step": 11250,
      "train_runtime": 146786.4286,
      "train_tokens_per_second": 4653.223
    },
    {
      "epoch": 1.4355867346938775,
      "grad_norm": 0.6303084742948237,
      "learning_rate": 1.840440080684684e-06,
      "loss": 0.2725719451904297,
      "num_input_tokens_seen": 683344624,
      "step": 11255,
      "train_runtime": 146841.0765,
      "train_tokens_per_second": 4653.634
    },
    {
      "epoch": 1.4362244897959184,
      "grad_norm": 0.7133814236210575,
      "learning_rate": 1.8365595550462068e-06,
      "loss": 0.26194334030151367,
      "num_input_tokens_seen": 683647312,
      "step": 11260,
      "train_runtime": 146910.9089,
      "train_tokens_per_second": 4653.482
    },
    {
      "epoch": 1.436862244897959,
      "grad_norm": 0.6572877425943786,
      "learning_rate": 1.8326822041411524e-06,
      "loss": 0.25354886054992676,
      "num_input_tokens_seen": 683953976,
      "step": 11265,
      "train_runtime": 146967.1436,
      "train_tokens_per_second": 4653.788
    },
    {
      "epoch": 1.4375,
      "grad_norm": 0.6807081378467906,
      "learning_rate": 1.828808031860707e-06,
      "loss": 0.2550026893615723,
      "num_input_tokens_seen": 684245328,
      "step": 11270,
      "train_runtime": 147037.7816,
      "train_tokens_per_second": 4653.534
    },
    {
      "epoch": 1.438137755102041,
      "grad_norm": 0.6421792190220771,
      "learning_rate": 1.824937042092879e-06,
      "loss": 0.255743408203125,
      "num_input_tokens_seen": 684534456,
      "step": 11275,
      "train_runtime": 147100.7536,
      "train_tokens_per_second": 4653.507
    },
    {
      "epoch": 1.4387755102040816,
      "grad_norm": 0.6703049773622103,
      "learning_rate": 1.8210692387224776e-06,
      "loss": 0.2595668792724609,
      "num_input_tokens_seen": 684835696,
      "step": 11280,
      "train_runtime": 147152.7923,
      "train_tokens_per_second": 4653.909
    },
    {
      "epoch": 1.4394132653061225,
      "grad_norm": 0.705144171072784,
      "learning_rate": 1.8172046256311087e-06,
      "loss": 0.262544584274292,
      "num_input_tokens_seen": 685141560,
      "step": 11285,
      "train_runtime": 147220.3621,
      "train_tokens_per_second": 4653.851
    },
    {
      "epoch": 1.4400510204081631,
      "grad_norm": 0.5983619231553162,
      "learning_rate": 1.8133432066971823e-06,
      "loss": 0.24910640716552734,
      "num_input_tokens_seen": 685444368,
      "step": 11290,
      "train_runtime": 147293.6179,
      "train_tokens_per_second": 4653.592
    },
    {
      "epoch": 1.440688775510204,
      "grad_norm": 0.7044716748584612,
      "learning_rate": 1.8094849857959074e-06,
      "loss": 0.267687463760376,
      "num_input_tokens_seen": 685735024,
      "step": 11295,
      "train_runtime": 147363.6411,
      "train_tokens_per_second": 4653.353
    },
    {
      "epoch": 1.441326530612245,
      "grad_norm": 0.6729325667272261,
      "learning_rate": 1.8056299667992721e-06,
      "loss": 0.25908875465393066,
      "num_input_tokens_seen": 686029680,
      "step": 11300,
      "train_runtime": 147424.7146,
      "train_tokens_per_second": 4653.424
    },
    {
      "epoch": 1.4419642857142856,
      "grad_norm": 0.6602083938659958,
      "learning_rate": 1.8017781535760581e-06,
      "loss": 0.2524980068206787,
      "num_input_tokens_seen": 686323368,
      "step": 11305,
      "train_runtime": 147490.7589,
      "train_tokens_per_second": 4653.331
    },
    {
      "epoch": 1.4426020408163265,
      "grad_norm": 0.7239698072065522,
      "learning_rate": 1.7979295499918338e-06,
      "loss": 0.25349082946777346,
      "num_input_tokens_seen": 686619760,
      "step": 11310,
      "train_runtime": 147558.1911,
      "train_tokens_per_second": 4653.213
    },
    {
      "epoch": 1.4432397959183674,
      "grad_norm": 0.635097988576949,
      "learning_rate": 1.7940841599089364e-06,
      "loss": 0.26335620880126953,
      "num_input_tokens_seen": 686937152,
      "step": 11315,
      "train_runtime": 147611.6909,
      "train_tokens_per_second": 4653.677
    },
    {
      "epoch": 1.443877551020408,
      "grad_norm": 0.640691388944003,
      "learning_rate": 1.790241987186485e-06,
      "loss": 0.2568617582321167,
      "num_input_tokens_seen": 687234768,
      "step": 11320,
      "train_runtime": 147678.6639,
      "train_tokens_per_second": 4653.582
    },
    {
      "epoch": 1.444515306122449,
      "grad_norm": 0.6255519134411113,
      "learning_rate": 1.7864030356803685e-06,
      "loss": 0.2553276300430298,
      "num_input_tokens_seen": 687535408,
      "step": 11325,
      "train_runtime": 147737.6656,
      "train_tokens_per_second": 4653.758
    },
    {
      "epoch": 1.4451530612244898,
      "grad_norm": 0.6569709229834011,
      "learning_rate": 1.7825673092432433e-06,
      "loss": 0.26634774208068845,
      "num_input_tokens_seen": 687842240,
      "step": 11330,
      "train_runtime": 147795.4392,
      "train_tokens_per_second": 4654.015
    },
    {
      "epoch": 1.4457908163265305,
      "grad_norm": 0.6146477199018123,
      "learning_rate": 1.7787348117245278e-06,
      "loss": 0.24077410697937013,
      "num_input_tokens_seen": 688143488,
      "step": 11335,
      "train_runtime": 147851.7803,
      "train_tokens_per_second": 4654.279
    },
    {
      "epoch": 1.4464285714285714,
      "grad_norm": 0.6745603895723179,
      "learning_rate": 1.7749055469704013e-06,
      "loss": 0.24639506340026857,
      "num_input_tokens_seen": 688438208,
      "step": 11340,
      "train_runtime": 147925.3508,
      "train_tokens_per_second": 4653.957
    },
    {
      "epoch": 1.4470663265306123,
      "grad_norm": 0.6365832756715657,
      "learning_rate": 1.771079518823799e-06,
      "loss": 0.2675870418548584,
      "num_input_tokens_seen": 688750184,
      "step": 11345,
      "train_runtime": 147998.0674,
      "train_tokens_per_second": 4653.778
    },
    {
      "epoch": 1.447704081632653,
      "grad_norm": 0.6345086185106481,
      "learning_rate": 1.7672567311244065e-06,
      "loss": 0.24507179260253906,
      "num_input_tokens_seen": 689051608,
      "step": 11350,
      "train_runtime": 148067.9814,
      "train_tokens_per_second": 4653.617
    },
    {
      "epoch": 1.4483418367346939,
      "grad_norm": 0.7331991768030408,
      "learning_rate": 1.7634371877086587e-06,
      "loss": 0.251706600189209,
      "num_input_tokens_seen": 689338368,
      "step": 11355,
      "train_runtime": 148122.4903,
      "train_tokens_per_second": 4653.84
    },
    {
      "epoch": 1.4489795918367347,
      "grad_norm": 0.6022787473010074,
      "learning_rate": 1.7596208924097336e-06,
      "loss": 0.2331090450286865,
      "num_input_tokens_seen": 689645832,
      "step": 11360,
      "train_runtime": 148195.7569,
      "train_tokens_per_second": 4653.614
    },
    {
      "epoch": 1.4496173469387754,
      "grad_norm": 0.6283517904938899,
      "learning_rate": 1.7558078490575531e-06,
      "loss": 0.2676694393157959,
      "num_input_tokens_seen": 689953024,
      "step": 11365,
      "train_runtime": 148258.7314,
      "train_tokens_per_second": 4653.709
    },
    {
      "epoch": 1.4502551020408163,
      "grad_norm": 0.6577691490155231,
      "learning_rate": 1.7519980614787673e-06,
      "loss": 0.26604576110839845,
      "num_input_tokens_seen": 690265872,
      "step": 11370,
      "train_runtime": 148310.2138,
      "train_tokens_per_second": 4654.203
    },
    {
      "epoch": 1.4508928571428572,
      "grad_norm": 0.6751365630787896,
      "learning_rate": 1.748191533496766e-06,
      "loss": 0.2404193878173828,
      "num_input_tokens_seen": 690559424,
      "step": 11375,
      "train_runtime": 148381.5591,
      "train_tokens_per_second": 4653.944
    },
    {
      "epoch": 1.4515306122448979,
      "grad_norm": 0.6791540330159553,
      "learning_rate": 1.7443882689316693e-06,
      "loss": 0.261887264251709,
      "num_input_tokens_seen": 690867440,
      "step": 11380,
      "train_runtime": 148446.7869,
      "train_tokens_per_second": 4653.974
    },
    {
      "epoch": 1.4521683673469388,
      "grad_norm": 0.5696416042203014,
      "learning_rate": 1.7405882716003154e-06,
      "loss": 0.2412036657333374,
      "num_input_tokens_seen": 691179896,
      "step": 11385,
      "train_runtime": 148503.0122,
      "train_tokens_per_second": 4654.316
    },
    {
      "epoch": 1.4528061224489797,
      "grad_norm": 0.663334327792484,
      "learning_rate": 1.7367915453162653e-06,
      "loss": 0.2515883207321167,
      "num_input_tokens_seen": 691479832,
      "step": 11390,
      "train_runtime": 148575.0559,
      "train_tokens_per_second": 4654.078
    },
    {
      "epoch": 1.4534438775510203,
      "grad_norm": 0.7672916515860333,
      "learning_rate": 1.732998093889805e-06,
      "loss": 0.25553102493286134,
      "num_input_tokens_seen": 691770328,
      "step": 11395,
      "train_runtime": 148643.9058,
      "train_tokens_per_second": 4653.876
    },
    {
      "epoch": 1.4540816326530612,
      "grad_norm": 0.7613214609422131,
      "learning_rate": 1.7292079211279229e-06,
      "loss": 0.2637500286102295,
      "num_input_tokens_seen": 692047224,
      "step": 11400,
      "train_runtime": 148707.8063,
      "train_tokens_per_second": 4653.738
    },
    {
      "epoch": 1.4547193877551021,
      "grad_norm": 0.6793202080818056,
      "learning_rate": 1.7254210308343233e-06,
      "loss": 0.2611916303634644,
      "num_input_tokens_seen": 692351568,
      "step": 11405,
      "train_runtime": 148772.0315,
      "train_tokens_per_second": 4653.775
    },
    {
      "epoch": 1.4553571428571428,
      "grad_norm": 0.6934879394356063,
      "learning_rate": 1.721637426809415e-06,
      "loss": 0.2512598991394043,
      "num_input_tokens_seen": 692648960,
      "step": 11410,
      "train_runtime": 148844.8894,
      "train_tokens_per_second": 4653.495
    },
    {
      "epoch": 1.4559948979591837,
      "grad_norm": 0.75152296594556,
      "learning_rate": 1.7178571128503097e-06,
      "loss": 0.2514254093170166,
      "num_input_tokens_seen": 692945064,
      "step": 11415,
      "train_runtime": 148915.559,
      "train_tokens_per_second": 4653.275
    },
    {
      "epoch": 1.4566326530612246,
      "grad_norm": 0.6026216151819691,
      "learning_rate": 1.7140800927508162e-06,
      "loss": 0.26080052852630614,
      "num_input_tokens_seen": 693250712,
      "step": 11420,
      "train_runtime": 148980.5413,
      "train_tokens_per_second": 4653.297
    },
    {
      "epoch": 1.4572704081632653,
      "grad_norm": 0.6652573244887677,
      "learning_rate": 1.7103063703014372e-06,
      "loss": 0.25720174312591554,
      "num_input_tokens_seen": 693557616,
      "step": 11425,
      "train_runtime": 149051.2751,
      "train_tokens_per_second": 4653.148
    },
    {
      "epoch": 1.4579081632653061,
      "grad_norm": 0.7123151801360577,
      "learning_rate": 1.706535949289368e-06,
      "loss": 0.27224562168121336,
      "num_input_tokens_seen": 693865096,
      "step": 11430,
      "train_runtime": 149124.6564,
      "train_tokens_per_second": 4652.92
    },
    {
      "epoch": 1.458545918367347,
      "grad_norm": 0.7165055019879256,
      "learning_rate": 1.7027688334984881e-06,
      "loss": 0.26955580711364746,
      "num_input_tokens_seen": 694167488,
      "step": 11435,
      "train_runtime": 149194.6544,
      "train_tokens_per_second": 4652.764
    },
    {
      "epoch": 1.4591836734693877,
      "grad_norm": 0.6742270512617384,
      "learning_rate": 1.6990050267093621e-06,
      "loss": 0.24876112937927247,
      "num_input_tokens_seen": 694471352,
      "step": 11440,
      "train_runtime": 149246.5097,
      "train_tokens_per_second": 4653.183
    },
    {
      "epoch": 1.4598214285714286,
      "grad_norm": 0.6383675482091885,
      "learning_rate": 1.6952445326992327e-06,
      "loss": 0.23597230911254882,
      "num_input_tokens_seen": 694760392,
      "step": 11445,
      "train_runtime": 149313.7796,
      "train_tokens_per_second": 4653.023
    },
    {
      "epoch": 1.4604591836734695,
      "grad_norm": 0.7035456067026871,
      "learning_rate": 1.6914873552420185e-06,
      "loss": 0.26068577766418455,
      "num_input_tokens_seen": 695056072,
      "step": 11450,
      "train_runtime": 149390.9183,
      "train_tokens_per_second": 4652.599
    },
    {
      "epoch": 1.4610969387755102,
      "grad_norm": 0.6285270523273403,
      "learning_rate": 1.687733498108311e-06,
      "loss": 0.26918802261352537,
      "num_input_tokens_seen": 695365848,
      "step": 11455,
      "train_runtime": 149448.4664,
      "train_tokens_per_second": 4652.88
    },
    {
      "epoch": 1.461734693877551,
      "grad_norm": 0.6693220843063449,
      "learning_rate": 1.6839829650653632e-06,
      "loss": 0.24681711196899414,
      "num_input_tokens_seen": 695661496,
      "step": 11460,
      "train_runtime": 149511.3983,
      "train_tokens_per_second": 4652.899
    },
    {
      "epoch": 1.462372448979592,
      "grad_norm": 0.7651750699630233,
      "learning_rate": 1.6802357598771014e-06,
      "loss": 0.2575855016708374,
      "num_input_tokens_seen": 695960184,
      "step": 11465,
      "train_runtime": 149577.9152,
      "train_tokens_per_second": 4652.827
    },
    {
      "epoch": 1.4630102040816326,
      "grad_norm": 0.7036196087435692,
      "learning_rate": 1.6764918863041079e-06,
      "loss": 0.25083796977996825,
      "num_input_tokens_seen": 696258456,
      "step": 11470,
      "train_runtime": 149640.317,
      "train_tokens_per_second": 4652.88
    },
    {
      "epoch": 1.4636479591836735,
      "grad_norm": 0.6862839891742669,
      "learning_rate": 1.6727513481036167e-06,
      "loss": 0.23392910957336427,
      "num_input_tokens_seen": 696562944,
      "step": 11475,
      "train_runtime": 149709.7164,
      "train_tokens_per_second": 4652.757
    },
    {
      "epoch": 1.4642857142857144,
      "grad_norm": 0.6643195742481856,
      "learning_rate": 1.6690141490295259e-06,
      "loss": 0.2381272792816162,
      "num_input_tokens_seen": 696861776,
      "step": 11480,
      "train_runtime": 149781.0158,
      "train_tokens_per_second": 4652.537
    },
    {
      "epoch": 1.464923469387755,
      "grad_norm": 0.6700710365667815,
      "learning_rate": 1.6652802928323698e-06,
      "loss": 0.2412243127822876,
      "num_input_tokens_seen": 697176216,
      "step": 11485,
      "train_runtime": 149856.7681,
      "train_tokens_per_second": 4652.284
    },
    {
      "epoch": 1.465561224489796,
      "grad_norm": 0.6561042403096796,
      "learning_rate": 1.6615497832593365e-06,
      "loss": 0.25841598510742186,
      "num_input_tokens_seen": 697478792,
      "step": 11490,
      "train_runtime": 149923.1941,
      "train_tokens_per_second": 4652.241
    },
    {
      "epoch": 1.4661989795918369,
      "grad_norm": 0.6941070429451296,
      "learning_rate": 1.6578226240542528e-06,
      "loss": 0.2742449760437012,
      "num_input_tokens_seen": 697785848,
      "step": 11495,
      "train_runtime": 149988.4626,
      "train_tokens_per_second": 4652.263
    },
    {
      "epoch": 1.4668367346938775,
      "grad_norm": 0.6802527349106692,
      "learning_rate": 1.6540988189575823e-06,
      "loss": 0.2669424533843994,
      "num_input_tokens_seen": 698107216,
      "step": 11500,
      "train_runtime": 150056.271,
      "train_tokens_per_second": 4652.303
    },
    {
      "epoch": 1.4674744897959184,
      "grad_norm": 0.6649874981638693,
      "learning_rate": 1.6503783717064247e-06,
      "loss": 0.2652735710144043,
      "num_input_tokens_seen": 698415200,
      "step": 11505,
      "train_runtime": 150133.3092,
      "train_tokens_per_second": 4651.967
    },
    {
      "epoch": 1.468112244897959,
      "grad_norm": 0.7210063670504991,
      "learning_rate": 1.6466612860345083e-06,
      "loss": 0.2547138690948486,
      "num_input_tokens_seen": 698717160,
      "step": 11510,
      "train_runtime": 150202.1339,
      "train_tokens_per_second": 4651.846
    },
    {
      "epoch": 1.46875,
      "grad_norm": 0.6904115337156073,
      "learning_rate": 1.6429475656721878e-06,
      "loss": 0.2531946897506714,
      "num_input_tokens_seen": 699019328,
      "step": 11515,
      "train_runtime": 150273.788,
      "train_tokens_per_second": 4651.638
    },
    {
      "epoch": 1.469387755102041,
      "grad_norm": 0.6096103663871228,
      "learning_rate": 1.6392372143464407e-06,
      "loss": 0.24852967262268066,
      "num_input_tokens_seen": 699332520,
      "step": 11520,
      "train_runtime": 150326.2769,
      "train_tokens_per_second": 4652.098
    },
    {
      "epoch": 1.4700255102040816,
      "grad_norm": 0.83723132893415,
      "learning_rate": 1.635530235780865e-06,
      "loss": 0.2493694543838501,
      "num_input_tokens_seen": 699626568,
      "step": 11525,
      "train_runtime": 150399.2931,
      "train_tokens_per_second": 4651.794
    },
    {
      "epoch": 1.4706632653061225,
      "grad_norm": 0.6874689194377291,
      "learning_rate": 1.6318266336956713e-06,
      "loss": 0.239760160446167,
      "num_input_tokens_seen": 699925336,
      "step": 11530,
      "train_runtime": 150473.2629,
      "train_tokens_per_second": 4651.493
    },
    {
      "epoch": 1.4713010204081631,
      "grad_norm": 0.6081449887494256,
      "learning_rate": 1.6281264118076833e-06,
      "loss": 0.23812055587768555,
      "num_input_tokens_seen": 700232720,
      "step": 11535,
      "train_runtime": 150540.3556,
      "train_tokens_per_second": 4651.462
    },
    {
      "epoch": 1.471938775510204,
      "grad_norm": 0.6389000206547071,
      "learning_rate": 1.6244295738303323e-06,
      "loss": 0.2438938856124878,
      "num_input_tokens_seen": 700527648,
      "step": 11540,
      "train_runtime": 150611.8723,
      "train_tokens_per_second": 4651.211
    },
    {
      "epoch": 1.472576530612245,
      "grad_norm": 0.6718629266348322,
      "learning_rate": 1.6207361234736535e-06,
      "loss": 0.2975417375564575,
      "num_input_tokens_seen": 700843920,
      "step": 11545,
      "train_runtime": 150663.5359,
      "train_tokens_per_second": 4651.716
    },
    {
      "epoch": 1.4732142857142856,
      "grad_norm": 0.6449463224727301,
      "learning_rate": 1.6170460644442814e-06,
      "loss": 0.25352983474731444,
      "num_input_tokens_seen": 701144312,
      "step": 11550,
      "train_runtime": 150730.063,
      "train_tokens_per_second": 4651.655
    },
    {
      "epoch": 1.4738520408163265,
      "grad_norm": 0.7997272426587076,
      "learning_rate": 1.6133594004454512e-06,
      "loss": 0.26451525688171384,
      "num_input_tokens_seen": 701449824,
      "step": 11555,
      "train_runtime": 150799.9905,
      "train_tokens_per_second": 4651.524
    },
    {
      "epoch": 1.4744897959183674,
      "grad_norm": 0.6763084026613673,
      "learning_rate": 1.6096761351769813e-06,
      "loss": 0.26746971607208253,
      "num_input_tokens_seen": 701742128,
      "step": 11560,
      "train_runtime": 150867.234,
      "train_tokens_per_second": 4651.389
    },
    {
      "epoch": 1.475127551020408,
      "grad_norm": 0.6794598871157979,
      "learning_rate": 1.6059962723352912e-06,
      "loss": 0.2517268657684326,
      "num_input_tokens_seen": 702054784,
      "step": 11565,
      "train_runtime": 150929.2574,
      "train_tokens_per_second": 4651.549
    },
    {
      "epoch": 1.475765306122449,
      "grad_norm": 0.6899303692739722,
      "learning_rate": 1.6023198156133796e-06,
      "loss": 0.2649550437927246,
      "num_input_tokens_seen": 702354392,
      "step": 11570,
      "train_runtime": 150994.8582,
      "train_tokens_per_second": 4651.512
    },
    {
      "epoch": 1.4764030612244898,
      "grad_norm": 0.5932483402474683,
      "learning_rate": 1.5986467687008256e-06,
      "loss": 0.23227922916412352,
      "num_input_tokens_seen": 702659368,
      "step": 11575,
      "train_runtime": 151054.3276,
      "train_tokens_per_second": 4651.7
    },
    {
      "epoch": 1.4770408163265305,
      "grad_norm": 0.6628646805863575,
      "learning_rate": 1.5949771352837873e-06,
      "loss": 0.27329676151275634,
      "num_input_tokens_seen": 702974120,
      "step": 11580,
      "train_runtime": 151123.6404,
      "train_tokens_per_second": 4651.649
    },
    {
      "epoch": 1.4776785714285714,
      "grad_norm": 0.6005475806622863,
      "learning_rate": 1.5913109190450033e-06,
      "loss": 0.2482689619064331,
      "num_input_tokens_seen": 703274640,
      "step": 11585,
      "train_runtime": 151199.0932,
      "train_tokens_per_second": 4651.315
    },
    {
      "epoch": 1.4783163265306123,
      "grad_norm": 0.7087277704302467,
      "learning_rate": 1.5876481236637725e-06,
      "loss": 0.25067954063415526,
      "num_input_tokens_seen": 703571232,
      "step": 11590,
      "train_runtime": 151267.0529,
      "train_tokens_per_second": 4651.186
    },
    {
      "epoch": 1.478954081632653,
      "grad_norm": 0.6929832721752661,
      "learning_rate": 1.5839887528159664e-06,
      "loss": 0.24625492095947266,
      "num_input_tokens_seen": 703886312,
      "step": 11595,
      "train_runtime": 151335.5903,
      "train_tokens_per_second": 4651.162
    },
    {
      "epoch": 1.4795918367346939,
      "grad_norm": 0.7363768129383547,
      "learning_rate": 1.580332810174024e-06,
      "loss": 0.27662837505340576,
      "num_input_tokens_seen": 704198792,
      "step": 11600,
      "train_runtime": 151411.2335,
      "train_tokens_per_second": 4650.902
    },
    {
      "epoch": 1.4802295918367347,
      "grad_norm": 0.6949936888575473,
      "learning_rate": 1.5766802994069336e-06,
      "loss": 0.2493530035018921,
      "num_input_tokens_seen": 704515848,
      "step": 11605,
      "train_runtime": 151471.1362,
      "train_tokens_per_second": 4651.156
    },
    {
      "epoch": 1.4808673469387754,
      "grad_norm": 0.6292968059570132,
      "learning_rate": 1.573031224180247e-06,
      "loss": 0.27558221817016604,
      "num_input_tokens_seen": 704821864,
      "step": 11610,
      "train_runtime": 151535.6845,
      "train_tokens_per_second": 4651.194
    },
    {
      "epoch": 1.4815051020408163,
      "grad_norm": 0.6738011434557234,
      "learning_rate": 1.5693855881560654e-06,
      "loss": 0.23457353115081786,
      "num_input_tokens_seen": 705133552,
      "step": 11615,
      "train_runtime": 151602.2108,
      "train_tokens_per_second": 4651.209
    },
    {
      "epoch": 1.4821428571428572,
      "grad_norm": 0.662589033840902,
      "learning_rate": 1.5657433949930406e-06,
      "loss": 0.25937981605529786,
      "num_input_tokens_seen": 705442408,
      "step": 11620,
      "train_runtime": 151676.7926,
      "train_tokens_per_second": 4650.958
    },
    {
      "epoch": 1.4827806122448979,
      "grad_norm": 0.6699351629628751,
      "learning_rate": 1.5621046483463663e-06,
      "loss": 0.2675338268280029,
      "num_input_tokens_seen": 705756872,
      "step": 11625,
      "train_runtime": 151745.0479,
      "train_tokens_per_second": 4650.938
    },
    {
      "epoch": 1.4834183673469388,
      "grad_norm": 0.7252766587622275,
      "learning_rate": 1.5584693518677796e-06,
      "loss": 0.2641123294830322,
      "num_input_tokens_seen": 706064448,
      "step": 11630,
      "train_runtime": 151816.1304,
      "train_tokens_per_second": 4650.787
    },
    {
      "epoch": 1.4840561224489797,
      "grad_norm": 0.6411274114325384,
      "learning_rate": 1.554837509205554e-06,
      "loss": 0.24312682151794435,
      "num_input_tokens_seen": 706367832,
      "step": 11635,
      "train_runtime": 151882.446,
      "train_tokens_per_second": 4650.754
    },
    {
      "epoch": 1.4846938775510203,
      "grad_norm": 0.6423826592976337,
      "learning_rate": 1.5512091240044996e-06,
      "loss": 0.2624907732009888,
      "num_input_tokens_seen": 706681984,
      "step": 11640,
      "train_runtime": 151942.2284,
      "train_tokens_per_second": 4650.991
    },
    {
      "epoch": 1.4853316326530612,
      "grad_norm": 0.7663904546335204,
      "learning_rate": 1.5475841999059487e-06,
      "loss": 0.27055678367614744,
      "num_input_tokens_seen": 706987120,
      "step": 11645,
      "train_runtime": 152005.7348,
      "train_tokens_per_second": 4651.056
    },
    {
      "epoch": 1.4859693877551021,
      "grad_norm": 0.6313909324997609,
      "learning_rate": 1.5439627405477714e-06,
      "loss": 0.2421438217163086,
      "num_input_tokens_seen": 707300040,
      "step": 11650,
      "train_runtime": 152065.1961,
      "train_tokens_per_second": 4651.295
    },
    {
      "epoch": 1.4866071428571428,
      "grad_norm": 0.6534547954585155,
      "learning_rate": 1.5403447495643542e-06,
      "loss": 0.23953189849853515,
      "num_input_tokens_seen": 707605520,
      "step": 11655,
      "train_runtime": 152139.5543,
      "train_tokens_per_second": 4651.029
    },
    {
      "epoch": 1.4872448979591837,
      "grad_norm": 0.6527691401390453,
      "learning_rate": 1.5367302305866017e-06,
      "loss": 0.2567351579666138,
      "num_input_tokens_seen": 707906880,
      "step": 11660,
      "train_runtime": 152208.2664,
      "train_tokens_per_second": 4650.91
    },
    {
      "epoch": 1.4878826530612246,
      "grad_norm": 0.7298686287037895,
      "learning_rate": 1.5331191872419349e-06,
      "loss": 0.25123307704925535,
      "num_input_tokens_seen": 708213592,
      "step": 11665,
      "train_runtime": 152277.8543,
      "train_tokens_per_second": 4650.798
    },
    {
      "epoch": 1.4885204081632653,
      "grad_norm": 0.73646842290651,
      "learning_rate": 1.5295116231542935e-06,
      "loss": 0.2430964231491089,
      "num_input_tokens_seen": 708503720,
      "step": 11670,
      "train_runtime": 152344.4542,
      "train_tokens_per_second": 4650.67
    },
    {
      "epoch": 1.4891581632653061,
      "grad_norm": 0.6199818195617495,
      "learning_rate": 1.5259075419441144e-06,
      "loss": 0.25212526321411133,
      "num_input_tokens_seen": 708809960,
      "step": 11675,
      "train_runtime": 152411.4443,
      "train_tokens_per_second": 4650.635
    },
    {
      "epoch": 1.489795918367347,
      "grad_norm": 0.7303765404971818,
      "learning_rate": 1.5223069472283464e-06,
      "loss": 0.2657268762588501,
      "num_input_tokens_seen": 709116352,
      "step": 11680,
      "train_runtime": 152468.2353,
      "train_tokens_per_second": 4650.912
    },
    {
      "epoch": 1.4904336734693877,
      "grad_norm": 0.6133630983750509,
      "learning_rate": 1.5187098426204411e-06,
      "loss": 0.22965846061706544,
      "num_input_tokens_seen": 709399184,
      "step": 11685,
      "train_runtime": 152536.6366,
      "train_tokens_per_second": 4650.681
    },
    {
      "epoch": 1.4910714285714286,
      "grad_norm": 0.7161191464349163,
      "learning_rate": 1.5151162317303398e-06,
      "loss": 0.23941805362701415,
      "num_input_tokens_seen": 709692272,
      "step": 11690,
      "train_runtime": 152605.7257,
      "train_tokens_per_second": 4650.496
    },
    {
      "epoch": 1.4917091836734695,
      "grad_norm": 0.669251318517889,
      "learning_rate": 1.5115261181644842e-06,
      "loss": 0.26945855617523196,
      "num_input_tokens_seen": 710004984,
      "step": 11695,
      "train_runtime": 152682.019,
      "train_tokens_per_second": 4650.22
    },
    {
      "epoch": 1.4923469387755102,
      "grad_norm": 0.6460954619175574,
      "learning_rate": 1.5079395055258027e-06,
      "loss": 0.23934993743896485,
      "num_input_tokens_seen": 710313688,
      "step": 11700,
      "train_runtime": 152745.5762,
      "train_tokens_per_second": 4650.306
    },
    {
      "epoch": 1.492984693877551,
      "grad_norm": 0.613900227712934,
      "learning_rate": 1.5043563974137132e-06,
      "loss": 0.22617621421813966,
      "num_input_tokens_seen": 710618288,
      "step": 11705,
      "train_runtime": 152807.8819,
      "train_tokens_per_second": 4650.403
    },
    {
      "epoch": 1.493622448979592,
      "grad_norm": 0.675599640664044,
      "learning_rate": 1.5007767974241133e-06,
      "loss": 0.2418062686920166,
      "num_input_tokens_seen": 710926392,
      "step": 11710,
      "train_runtime": 152873.0827,
      "train_tokens_per_second": 4650.435
    },
    {
      "epoch": 1.4942602040816326,
      "grad_norm": 0.6174613887003167,
      "learning_rate": 1.497200709149383e-06,
      "loss": 0.2354290008544922,
      "num_input_tokens_seen": 711230608,
      "step": 11715,
      "train_runtime": 152933.8752,
      "train_tokens_per_second": 4650.576
    },
    {
      "epoch": 1.4948979591836735,
      "grad_norm": 0.697269811303447,
      "learning_rate": 1.493628136178376e-06,
      "loss": 0.2707733154296875,
      "num_input_tokens_seen": 711527456,
      "step": 11720,
      "train_runtime": 153003.793,
      "train_tokens_per_second": 4650.391
    },
    {
      "epoch": 1.4955357142857144,
      "grad_norm": 0.6802569943864075,
      "learning_rate": 1.49005908209642e-06,
      "loss": 0.2669675827026367,
      "num_input_tokens_seen": 711838440,
      "step": 11725,
      "train_runtime": 153067.2024,
      "train_tokens_per_second": 4650.496
    },
    {
      "epoch": 1.496173469387755,
      "grad_norm": 0.6672094633913374,
      "learning_rate": 1.4864935504853062e-06,
      "loss": 0.24235136508941652,
      "num_input_tokens_seen": 712150208,
      "step": 11730,
      "train_runtime": 153141.6004,
      "train_tokens_per_second": 4650.273
    },
    {
      "epoch": 1.496811224489796,
      "grad_norm": 0.6512272194630198,
      "learning_rate": 1.4829315449232984e-06,
      "loss": 0.2580772876739502,
      "num_input_tokens_seen": 712457024,
      "step": 11735,
      "train_runtime": 153204.1999,
      "train_tokens_per_second": 4650.375
    },
    {
      "epoch": 1.4974489795918369,
      "grad_norm": 0.7186619430371217,
      "learning_rate": 1.4793730689851187e-06,
      "loss": 0.2615934133529663,
      "num_input_tokens_seen": 712757864,
      "step": 11740,
      "train_runtime": 153269.0385,
      "train_tokens_per_second": 4650.371
    },
    {
      "epoch": 1.4980867346938775,
      "grad_norm": 0.6517705009255923,
      "learning_rate": 1.4758181262419425e-06,
      "loss": 0.24282798767089844,
      "num_input_tokens_seen": 713075888,
      "step": 11745,
      "train_runtime": 153334.0472,
      "train_tokens_per_second": 4650.473
    },
    {
      "epoch": 1.4987244897959184,
      "grad_norm": 0.6569295399291899,
      "learning_rate": 1.4722667202614037e-06,
      "loss": 0.23960046768188475,
      "num_input_tokens_seen": 713371464,
      "step": 11750,
      "train_runtime": 153388.3982,
      "train_tokens_per_second": 4650.752
    },
    {
      "epoch": 1.499362244897959,
      "grad_norm": 0.7019841676795132,
      "learning_rate": 1.4687188546075903e-06,
      "loss": 0.23960208892822266,
      "num_input_tokens_seen": 713657632,
      "step": 11755,
      "train_runtime": 153458.7883,
      "train_tokens_per_second": 4650.484
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.6403120523509626,
      "learning_rate": 1.4651745328410289e-06,
      "loss": 0.258426833152771,
      "num_input_tokens_seen": 713955144,
      "step": 11760,
      "train_runtime": 153527.5697,
      "train_tokens_per_second": 4650.338
    },
    {
      "epoch": 1.5006377551020407,
      "grad_norm": 0.6522180290527697,
      "learning_rate": 1.4616337585186941e-06,
      "loss": 0.24520792961120605,
      "num_input_tokens_seen": 714257392,
      "step": 11765,
      "train_runtime": 153594.7663,
      "train_tokens_per_second": 4650.272
    },
    {
      "epoch": 1.5012755102040818,
      "grad_norm": 0.6244342453355227,
      "learning_rate": 1.4580965351940045e-06,
      "loss": 0.26330649852752686,
      "num_input_tokens_seen": 714569056,
      "step": 11770,
      "train_runtime": 153664.5251,
      "train_tokens_per_second": 4650.189
    },
    {
      "epoch": 1.5019132653061225,
      "grad_norm": 0.6309084668490915,
      "learning_rate": 1.454562866416806e-06,
      "loss": 0.23243155479431152,
      "num_input_tokens_seen": 714881472,
      "step": 11775,
      "train_runtime": 153718.1096,
      "train_tokens_per_second": 4650.6
    },
    {
      "epoch": 1.5025510204081631,
      "grad_norm": 0.5808562626569077,
      "learning_rate": 1.4510327557333843e-06,
      "loss": 0.2522455930709839,
      "num_input_tokens_seen": 715192808,
      "step": 11780,
      "train_runtime": 153793.2709,
      "train_tokens_per_second": 4650.352
    },
    {
      "epoch": 1.5031887755102042,
      "grad_norm": 0.6563879128278611,
      "learning_rate": 1.4475062066864516e-06,
      "loss": 0.2767788410186768,
      "num_input_tokens_seen": 715515792,
      "step": 11785,
      "train_runtime": 153853.7861,
      "train_tokens_per_second": 4650.622
    },
    {
      "epoch": 1.503826530612245,
      "grad_norm": 0.6227574867093538,
      "learning_rate": 1.4439832228151457e-06,
      "loss": 0.24577155113220214,
      "num_input_tokens_seen": 715831936,
      "step": 11790,
      "train_runtime": 153901.7271,
      "train_tokens_per_second": 4651.227
    },
    {
      "epoch": 1.5044642857142856,
      "grad_norm": 0.6138943688581633,
      "learning_rate": 1.4404638076550276e-06,
      "loss": 0.2581317901611328,
      "num_input_tokens_seen": 716136120,
      "step": 11795,
      "train_runtime": 153953.6899,
      "train_tokens_per_second": 4651.633
    },
    {
      "epoch": 1.5051020408163265,
      "grad_norm": 0.6624156064238919,
      "learning_rate": 1.4369479647380758e-06,
      "loss": 0.2634751319885254,
      "num_input_tokens_seen": 716445392,
      "step": 11800,
      "train_runtime": 154022.9587,
      "train_tokens_per_second": 4651.549
    },
    {
      "epoch": 1.5057397959183674,
      "grad_norm": 0.6929581381180583,
      "learning_rate": 1.4334356975926843e-06,
      "loss": 0.2595194339752197,
      "num_input_tokens_seen": 716746224,
      "step": 11805,
      "train_runtime": 154097.722,
      "train_tokens_per_second": 4651.245
    },
    {
      "epoch": 1.506377551020408,
      "grad_norm": 0.6476555100531105,
      "learning_rate": 1.429927009743659e-06,
      "loss": 0.2642998218536377,
      "num_input_tokens_seen": 717062648,
      "step": 11810,
      "train_runtime": 154155.19,
      "train_tokens_per_second": 4651.563
    },
    {
      "epoch": 1.507015306122449,
      "grad_norm": 0.690826021203568,
      "learning_rate": 1.4264219047122125e-06,
      "loss": 0.26960577964782717,
      "num_input_tokens_seen": 717368720,
      "step": 11815,
      "train_runtime": 154213.5952,
      "train_tokens_per_second": 4651.786
    },
    {
      "epoch": 1.5076530612244898,
      "grad_norm": 0.6016910574060026,
      "learning_rate": 1.422920386015963e-06,
      "loss": 0.25219082832336426,
      "num_input_tokens_seen": 717673224,
      "step": 11820,
      "train_runtime": 154266.9301,
      "train_tokens_per_second": 4652.152
    },
    {
      "epoch": 1.5082908163265305,
      "grad_norm": 0.6534672221331839,
      "learning_rate": 1.4194224571689286e-06,
      "loss": 0.26119270324707033,
      "num_input_tokens_seen": 717975600,
      "step": 11825,
      "train_runtime": 154335.2884,
      "train_tokens_per_second": 4652.051
    },
    {
      "epoch": 1.5089285714285714,
      "grad_norm": 0.6185095518581548,
      "learning_rate": 1.4159281216815269e-06,
      "loss": 0.26681981086730955,
      "num_input_tokens_seen": 718269840,
      "step": 11830,
      "train_runtime": 154412.0099,
      "train_tokens_per_second": 4651.645
    },
    {
      "epoch": 1.5095663265306123,
      "grad_norm": 0.6684805621613861,
      "learning_rate": 1.412437383060563e-06,
      "loss": 0.24793519973754882,
      "num_input_tokens_seen": 718583704,
      "step": 11835,
      "train_runtime": 154471.8592,
      "train_tokens_per_second": 4651.875
    },
    {
      "epoch": 1.510204081632653,
      "grad_norm": 0.797425326744484,
      "learning_rate": 1.4089502448092418e-06,
      "loss": 0.25735931396484374,
      "num_input_tokens_seen": 718886464,
      "step": 11840,
      "train_runtime": 154537.0972,
      "train_tokens_per_second": 4651.87
    },
    {
      "epoch": 1.5108418367346939,
      "grad_norm": 0.7215197781217865,
      "learning_rate": 1.4054667104271497e-06,
      "loss": 0.2530500411987305,
      "num_input_tokens_seen": 719178912,
      "step": 11845,
      "train_runtime": 154602.6924,
      "train_tokens_per_second": 4651.788
    },
    {
      "epoch": 1.5114795918367347,
      "grad_norm": 0.5517203232441623,
      "learning_rate": 1.4019867834102525e-06,
      "loss": 0.25861315727233886,
      "num_input_tokens_seen": 719488832,
      "step": 11850,
      "train_runtime": 154666.1485,
      "train_tokens_per_second": 4651.883
    },
    {
      "epoch": 1.5121173469387754,
      "grad_norm": 0.6400167365239027,
      "learning_rate": 1.398510467250906e-06,
      "loss": 0.23546905517578126,
      "num_input_tokens_seen": 719789416,
      "step": 11855,
      "train_runtime": 154741.9525,
      "train_tokens_per_second": 4651.547
    },
    {
      "epoch": 1.5127551020408163,
      "grad_norm": 0.7072383194006815,
      "learning_rate": 1.3950377654378312e-06,
      "loss": 0.2624575853347778,
      "num_input_tokens_seen": 720099464,
      "step": 11860,
      "train_runtime": 154812.0911,
      "train_tokens_per_second": 4651.442
    },
    {
      "epoch": 1.5133928571428572,
      "grad_norm": 0.7617778889505888,
      "learning_rate": 1.3915686814561285e-06,
      "loss": 0.2599982023239136,
      "num_input_tokens_seen": 720395896,
      "step": 11865,
      "train_runtime": 154875.0287,
      "train_tokens_per_second": 4651.466
    },
    {
      "epoch": 1.5140306122448979,
      "grad_norm": 0.6801430846003712,
      "learning_rate": 1.3881032187872666e-06,
      "loss": 0.2580726623535156,
      "num_input_tokens_seen": 720701088,
      "step": 11870,
      "train_runtime": 154942.1964,
      "train_tokens_per_second": 4651.419
    },
    {
      "epoch": 1.5146683673469388,
      "grad_norm": 0.6725219117370117,
      "learning_rate": 1.3846413809090792e-06,
      "loss": 0.24667282104492189,
      "num_input_tokens_seen": 720984848,
      "step": 11875,
      "train_runtime": 155004.8064,
      "train_tokens_per_second": 4651.371
    },
    {
      "epoch": 1.5153061224489797,
      "grad_norm": 0.6333060000660075,
      "learning_rate": 1.381183171295762e-06,
      "loss": 0.2713295936584473,
      "num_input_tokens_seen": 721284416,
      "step": 11880,
      "train_runtime": 155077.6104,
      "train_tokens_per_second": 4651.119
    },
    {
      "epoch": 1.5159438775510203,
      "grad_norm": 0.6418711574137776,
      "learning_rate": 1.3777285934178697e-06,
      "loss": 0.26169524192810056,
      "num_input_tokens_seen": 721595160,
      "step": 11885,
      "train_runtime": 155140.4977,
      "train_tokens_per_second": 4651.237
    },
    {
      "epoch": 1.5165816326530612,
      "grad_norm": 0.6462644081520978,
      "learning_rate": 1.3742776507423127e-06,
      "loss": 0.2720316171646118,
      "num_input_tokens_seen": 721886792,
      "step": 11890,
      "train_runtime": 155203.1296,
      "train_tokens_per_second": 4651.239
    },
    {
      "epoch": 1.5172193877551021,
      "grad_norm": 0.6265701983248271,
      "learning_rate": 1.3708303467323536e-06,
      "loss": 0.24444787502288817,
      "num_input_tokens_seen": 722191392,
      "step": 11895,
      "train_runtime": 155260.1119,
      "train_tokens_per_second": 4651.493
    },
    {
      "epoch": 1.5178571428571428,
      "grad_norm": 0.6080293985446179,
      "learning_rate": 1.367386684847602e-06,
      "loss": 0.24946393966674804,
      "num_input_tokens_seen": 722495968,
      "step": 11900,
      "train_runtime": 155325.7968,
      "train_tokens_per_second": 4651.487
    },
    {
      "epoch": 1.5184948979591837,
      "grad_norm": 0.7042361456981605,
      "learning_rate": 1.3639466685440133e-06,
      "loss": 0.2545601844787598,
      "num_input_tokens_seen": 722787536,
      "step": 11905,
      "train_runtime": 155403.6674,
      "train_tokens_per_second": 4651.033
    },
    {
      "epoch": 1.5191326530612246,
      "grad_norm": 0.6526395755457752,
      "learning_rate": 1.360510301273884e-06,
      "loss": 0.24752330780029297,
      "num_input_tokens_seen": 723079568,
      "step": 11910,
      "train_runtime": 155475.8836,
      "train_tokens_per_second": 4650.751
    },
    {
      "epoch": 1.5197704081632653,
      "grad_norm": 0.6920301297712828,
      "learning_rate": 1.3570775864858487e-06,
      "loss": 0.26047818660736083,
      "num_input_tokens_seen": 723382784,
      "step": 11915,
      "train_runtime": 155545.369,
      "train_tokens_per_second": 4650.622
    },
    {
      "epoch": 1.5204081632653061,
      "grad_norm": 0.7825761711802505,
      "learning_rate": 1.3536485276248768e-06,
      "loss": 0.24889864921569824,
      "num_input_tokens_seen": 723681840,
      "step": 11920,
      "train_runtime": 155607.433,
      "train_tokens_per_second": 4650.689
    },
    {
      "epoch": 1.521045918367347,
      "grad_norm": 0.6399928528672232,
      "learning_rate": 1.350223128132268e-06,
      "loss": 0.24691829681396485,
      "num_input_tokens_seen": 723983624,
      "step": 11925,
      "train_runtime": 155683.011,
      "train_tokens_per_second": 4650.37
    },
    {
      "epoch": 1.5216836734693877,
      "grad_norm": 0.728504362584846,
      "learning_rate": 1.3468013914456518e-06,
      "loss": 0.2515766143798828,
      "num_input_tokens_seen": 724282784,
      "step": 11930,
      "train_runtime": 155746.2417,
      "train_tokens_per_second": 4650.403
    },
    {
      "epoch": 1.5223214285714286,
      "grad_norm": 0.6493723109247684,
      "learning_rate": 1.3433833209989743e-06,
      "loss": 0.2518887996673584,
      "num_input_tokens_seen": 724577608,
      "step": 11935,
      "train_runtime": 155806.9747,
      "train_tokens_per_second": 4650.482
    },
    {
      "epoch": 1.5229591836734695,
      "grad_norm": 0.7027434732171063,
      "learning_rate": 1.3399689202225136e-06,
      "loss": 0.2504230737686157,
      "num_input_tokens_seen": 724872224,
      "step": 11940,
      "train_runtime": 155874.443,
      "train_tokens_per_second": 4650.36
    },
    {
      "epoch": 1.5235969387755102,
      "grad_norm": 0.6535899077544529,
      "learning_rate": 1.3365581925428594e-06,
      "loss": 0.253280782699585,
      "num_input_tokens_seen": 725180384,
      "step": 11945,
      "train_runtime": 155947.0663,
      "train_tokens_per_second": 4650.17
    },
    {
      "epoch": 1.524234693877551,
      "grad_norm": 0.707458938724632,
      "learning_rate": 1.3331511413829113e-06,
      "loss": 0.2489948272705078,
      "num_input_tokens_seen": 725483576,
      "step": 11950,
      "train_runtime": 156020.2085,
      "train_tokens_per_second": 4649.933
    },
    {
      "epoch": 1.524872448979592,
      "grad_norm": 0.8008707159298643,
      "learning_rate": 1.3297477701618833e-06,
      "loss": 0.2944218635559082,
      "num_input_tokens_seen": 725791648,
      "step": 11955,
      "train_runtime": 156091.9103,
      "train_tokens_per_second": 4649.771
    },
    {
      "epoch": 1.5255102040816326,
      "grad_norm": 0.6746720094362882,
      "learning_rate": 1.3263480822953011e-06,
      "loss": 0.24970026016235353,
      "num_input_tokens_seen": 726100440,
      "step": 11960,
      "train_runtime": 156155.1141,
      "train_tokens_per_second": 4649.867
    },
    {
      "epoch": 1.5261479591836735,
      "grad_norm": 0.7606976589129846,
      "learning_rate": 1.3229520811949836e-06,
      "loss": 0.2613847732543945,
      "num_input_tokens_seen": 726418864,
      "step": 11965,
      "train_runtime": 156213.7931,
      "train_tokens_per_second": 4650.158
    },
    {
      "epoch": 1.5267857142857144,
      "grad_norm": 0.673824089849578,
      "learning_rate": 1.3195597702690577e-06,
      "loss": 0.2561951160430908,
      "num_input_tokens_seen": 726713968,
      "step": 11970,
      "train_runtime": 156279.0687,
      "train_tokens_per_second": 4650.104
    },
    {
      "epoch": 1.527423469387755,
      "grad_norm": 0.6521630410082774,
      "learning_rate": 1.3161711529219434e-06,
      "loss": 0.24800543785095214,
      "num_input_tokens_seen": 726998080,
      "step": 11975,
      "train_runtime": 156344.8422,
      "train_tokens_per_second": 4649.965
    },
    {
      "epoch": 1.5280612244897958,
      "grad_norm": 0.664011278667914,
      "learning_rate": 1.3127862325543555e-06,
      "loss": 0.2613114356994629,
      "num_input_tokens_seen": 727314256,
      "step": 11980,
      "train_runtime": 156408.8857,
      "train_tokens_per_second": 4650.083
    },
    {
      "epoch": 1.5286989795918369,
      "grad_norm": 0.6800994302773266,
      "learning_rate": 1.3094050125632973e-06,
      "loss": 0.24103777408599852,
      "num_input_tokens_seen": 727614136,
      "step": 11985,
      "train_runtime": 156482.4272,
      "train_tokens_per_second": 4649.814
    },
    {
      "epoch": 1.5293367346938775,
      "grad_norm": 0.6575959516635664,
      "learning_rate": 1.30602749634206e-06,
      "loss": 0.26449012756347656,
      "num_input_tokens_seen": 727924752,
      "step": 11990,
      "train_runtime": 156534.2532,
      "train_tokens_per_second": 4650.259
    },
    {
      "epoch": 1.5299744897959182,
      "grad_norm": 0.6090825136182979,
      "learning_rate": 1.302653687280216e-06,
      "loss": 0.25784077644348147,
      "num_input_tokens_seen": 728228968,
      "step": 11995,
      "train_runtime": 156601.5897,
      "train_tokens_per_second": 4650.202
    },
    {
      "epoch": 1.5306122448979593,
      "grad_norm": 0.6375662939031023,
      "learning_rate": 1.299283588763619e-06,
      "loss": 0.2542006492614746,
      "num_input_tokens_seen": 728525984,
      "step": 12000,
      "train_runtime": 156670.3864,
      "train_tokens_per_second": 4650.055
    },
    {
      "epoch": 1.53125,
      "grad_norm": 0.6810512872276492,
      "learning_rate": 1.2959172041743985e-06,
      "loss": 0.2613798141479492,
      "num_input_tokens_seen": 728835224,
      "step": 12005,
      "train_runtime": 156732.5491,
      "train_tokens_per_second": 4650.184
    },
    {
      "epoch": 1.5318877551020407,
      "grad_norm": 0.733040943616195,
      "learning_rate": 1.2925545368909558e-06,
      "loss": 0.2360307455062866,
      "num_input_tokens_seen": 729121784,
      "step": 12010,
      "train_runtime": 156803.9519,
      "train_tokens_per_second": 4649.894
    },
    {
      "epoch": 1.5325255102040818,
      "grad_norm": 0.6912753074644812,
      "learning_rate": 1.289195590287965e-06,
      "loss": 0.25995163917541503,
      "num_input_tokens_seen": 729424080,
      "step": 12015,
      "train_runtime": 156864.734,
      "train_tokens_per_second": 4650.02
    },
    {
      "epoch": 1.5331632653061225,
      "grad_norm": 0.7037050729121367,
      "learning_rate": 1.285840367736358e-06,
      "loss": 0.25633811950683594,
      "num_input_tokens_seen": 729730768,
      "step": 12020,
      "train_runtime": 156924.8548,
      "train_tokens_per_second": 4650.192
    },
    {
      "epoch": 1.5338010204081631,
      "grad_norm": 0.6941207469555901,
      "learning_rate": 1.2824888726033391e-06,
      "loss": 0.2491147756576538,
      "num_input_tokens_seen": 730035304,
      "step": 12025,
      "train_runtime": 156992.9708,
      "train_tokens_per_second": 4650.115
    },
    {
      "epoch": 1.5344387755102042,
      "grad_norm": 0.6863693055640985,
      "learning_rate": 1.2791411082523692e-06,
      "loss": 0.2503811836242676,
      "num_input_tokens_seen": 730325824,
      "step": 12030,
      "train_runtime": 157058.949,
      "train_tokens_per_second": 4650.011
    },
    {
      "epoch": 1.535076530612245,
      "grad_norm": 0.6773444450347682,
      "learning_rate": 1.2757970780431594e-06,
      "loss": 0.27198493480682373,
      "num_input_tokens_seen": 730630584,
      "step": 12035,
      "train_runtime": 157134.4856,
      "train_tokens_per_second": 4649.715
    },
    {
      "epoch": 1.5357142857142856,
      "grad_norm": 0.7069076912015884,
      "learning_rate": 1.2724567853316788e-06,
      "loss": 0.24685401916503907,
      "num_input_tokens_seen": 730926424,
      "step": 12040,
      "train_runtime": 157212.3694,
      "train_tokens_per_second": 4649.293
    },
    {
      "epoch": 1.5363520408163265,
      "grad_norm": 0.6719480918536438,
      "learning_rate": 1.2691202334701474e-06,
      "loss": 0.2603030204772949,
      "num_input_tokens_seen": 731245296,
      "step": 12045,
      "train_runtime": 157282.4769,
      "train_tokens_per_second": 4649.248
    },
    {
      "epoch": 1.5369897959183674,
      "grad_norm": 0.6809489913946712,
      "learning_rate": 1.2657874258070247e-06,
      "loss": 0.24165427684783936,
      "num_input_tokens_seen": 731549024,
      "step": 12050,
      "train_runtime": 157352.4093,
      "train_tokens_per_second": 4649.112
    },
    {
      "epoch": 1.537627551020408,
      "grad_norm": 0.665830841555118,
      "learning_rate": 1.2624583656870153e-06,
      "loss": 0.24589719772338867,
      "num_input_tokens_seen": 731854376,
      "step": 12055,
      "train_runtime": 157413.2736,
      "train_tokens_per_second": 4649.255
    },
    {
      "epoch": 1.538265306122449,
      "grad_norm": 0.7004659243635903,
      "learning_rate": 1.2591330564510683e-06,
      "loss": 0.24840226173400878,
      "num_input_tokens_seen": 732146608,
      "step": 12060,
      "train_runtime": 157481.9292,
      "train_tokens_per_second": 4649.083
    },
    {
      "epoch": 1.5389030612244898,
      "grad_norm": 0.67867125984092,
      "learning_rate": 1.2558115014363592e-06,
      "loss": 0.2419034481048584,
      "num_input_tokens_seen": 732451520,
      "step": 12065,
      "train_runtime": 157558.0906,
      "train_tokens_per_second": 4648.771
    },
    {
      "epoch": 1.5395408163265305,
      "grad_norm": 0.6589380234833201,
      "learning_rate": 1.2524937039763025e-06,
      "loss": 0.25946569442749023,
      "num_input_tokens_seen": 732764640,
      "step": 12070,
      "train_runtime": 157617.0796,
      "train_tokens_per_second": 4649.018
    },
    {
      "epoch": 1.5401785714285714,
      "grad_norm": 0.6869336535457288,
      "learning_rate": 1.2491796674005385e-06,
      "loss": 0.2480013370513916,
      "num_input_tokens_seen": 733067904,
      "step": 12075,
      "train_runtime": 157688.603,
      "train_tokens_per_second": 4648.833
    },
    {
      "epoch": 1.5408163265306123,
      "grad_norm": 0.6493326430733583,
      "learning_rate": 1.2458693950349349e-06,
      "loss": 0.27153429985046384,
      "num_input_tokens_seen": 733376560,
      "step": 12080,
      "train_runtime": 157757.158,
      "train_tokens_per_second": 4648.769
    },
    {
      "epoch": 1.541454081632653,
      "grad_norm": 0.7245698269749907,
      "learning_rate": 1.2425628902015818e-06,
      "loss": 0.24982929229736328,
      "num_input_tokens_seen": 733668536,
      "step": 12085,
      "train_runtime": 157814.4223,
      "train_tokens_per_second": 4648.932
    },
    {
      "epoch": 1.5420918367346939,
      "grad_norm": 0.6548672882769807,
      "learning_rate": 1.2392601562187867e-06,
      "loss": 0.25737121105194094,
      "num_input_tokens_seen": 733967608,
      "step": 12090,
      "train_runtime": 157886.1355,
      "train_tokens_per_second": 4648.715
    },
    {
      "epoch": 1.5427295918367347,
      "grad_norm": 0.626716168682283,
      "learning_rate": 1.2359611964010747e-06,
      "loss": 0.26229133605957033,
      "num_input_tokens_seen": 734268424,
      "step": 12095,
      "train_runtime": 157953.518,
      "train_tokens_per_second": 4648.636
    },
    {
      "epoch": 1.5433673469387754,
      "grad_norm": 0.7023623923339372,
      "learning_rate": 1.2326660140591835e-06,
      "loss": 0.24214916229248046,
      "num_input_tokens_seen": 734570048,
      "step": 12100,
      "train_runtime": 158020.4129,
      "train_tokens_per_second": 4648.577
    },
    {
      "epoch": 1.5440051020408163,
      "grad_norm": 0.6458540208307123,
      "learning_rate": 1.2293746125000538e-06,
      "loss": 0.2654332876205444,
      "num_input_tokens_seen": 734878848,
      "step": 12105,
      "train_runtime": 158080.2557,
      "train_tokens_per_second": 4648.771
    },
    {
      "epoch": 1.5446428571428572,
      "grad_norm": 0.6321348613604179,
      "learning_rate": 1.2260869950268422e-06,
      "loss": 0.25348896980285646,
      "num_input_tokens_seen": 735169440,
      "step": 12110,
      "train_runtime": 158152.9672,
      "train_tokens_per_second": 4648.471
    },
    {
      "epoch": 1.5452806122448979,
      "grad_norm": 0.6215893303584566,
      "learning_rate": 1.2228031649389021e-06,
      "loss": 0.23989582061767578,
      "num_input_tokens_seen": 735470464,
      "step": 12115,
      "train_runtime": 158225.2378,
      "train_tokens_per_second": 4648.25
    },
    {
      "epoch": 1.5459183673469388,
      "grad_norm": 0.6630734645291291,
      "learning_rate": 1.2195231255317834e-06,
      "loss": 0.2431865692138672,
      "num_input_tokens_seen": 735778104,
      "step": 12120,
      "train_runtime": 158291.9873,
      "train_tokens_per_second": 4648.233
    },
    {
      "epoch": 1.5465561224489797,
      "grad_norm": 0.5979526128037281,
      "learning_rate": 1.2162468800972344e-06,
      "loss": 0.2632904052734375,
      "num_input_tokens_seen": 736087024,
      "step": 12125,
      "train_runtime": 158360.3683,
      "train_tokens_per_second": 4648.177
    },
    {
      "epoch": 1.5471938775510203,
      "grad_norm": 0.7078350457459534,
      "learning_rate": 1.2129744319232013e-06,
      "loss": 0.2559070587158203,
      "num_input_tokens_seen": 736400016,
      "step": 12130,
      "train_runtime": 158427.5125,
      "train_tokens_per_second": 4648.183
    },
    {
      "epoch": 1.5478316326530612,
      "grad_norm": 0.6967124468116201,
      "learning_rate": 1.2097057842938093e-06,
      "loss": 0.23843395709991455,
      "num_input_tokens_seen": 736714392,
      "step": 12135,
      "train_runtime": 158499.4134,
      "train_tokens_per_second": 4648.058
    },
    {
      "epoch": 1.5484693877551021,
      "grad_norm": 0.6847539702153206,
      "learning_rate": 1.206440940489375e-06,
      "loss": 0.2515080690383911,
      "num_input_tokens_seen": 737012256,
      "step": 12140,
      "train_runtime": 158557.999,
      "train_tokens_per_second": 4648.219
    },
    {
      "epoch": 1.5491071428571428,
      "grad_norm": 0.6025976955597792,
      "learning_rate": 1.203179903786401e-06,
      "loss": 0.2598711967468262,
      "num_input_tokens_seen": 737325784,
      "step": 12145,
      "train_runtime": 158620.8065,
      "train_tokens_per_second": 4648.355
    },
    {
      "epoch": 1.5497448979591837,
      "grad_norm": 0.6463825176551172,
      "learning_rate": 1.1999226774575618e-06,
      "loss": 0.24842023849487305,
      "num_input_tokens_seen": 737609096,
      "step": 12150,
      "train_runtime": 158687.6455,
      "train_tokens_per_second": 4648.182
    },
    {
      "epoch": 1.5503826530612246,
      "grad_norm": 0.6566695886080559,
      "learning_rate": 1.1966692647717127e-06,
      "loss": 0.2655647754669189,
      "num_input_tokens_seen": 737918056,
      "step": 12155,
      "train_runtime": 158750.0407,
      "train_tokens_per_second": 4648.302
    },
    {
      "epoch": 1.5510204081632653,
      "grad_norm": 0.603574418856841,
      "learning_rate": 1.1934196689938804e-06,
      "loss": 0.24559359550476073,
      "num_input_tokens_seen": 738225024,
      "step": 12160,
      "train_runtime": 158808.3636,
      "train_tokens_per_second": 4648.527
    },
    {
      "epoch": 1.5516581632653061,
      "grad_norm": 0.6730319896336483,
      "learning_rate": 1.1901738933852608e-06,
      "loss": 0.27126526832580566,
      "num_input_tokens_seen": 738522432,
      "step": 12165,
      "train_runtime": 158876.3494,
      "train_tokens_per_second": 4648.41
    },
    {
      "epoch": 1.552295918367347,
      "grad_norm": 0.7200139917166635,
      "learning_rate": 1.1869319412032171e-06,
      "loss": 0.23754267692565917,
      "num_input_tokens_seen": 738817336,
      "step": 12170,
      "train_runtime": 158941.2413,
      "train_tokens_per_second": 4648.368
    },
    {
      "epoch": 1.5529336734693877,
      "grad_norm": 0.6089938268511548,
      "learning_rate": 1.1836938157012746e-06,
      "loss": 0.23989269733428956,
      "num_input_tokens_seen": 739126784,
      "step": 12175,
      "train_runtime": 158990.99,
      "train_tokens_per_second": 4648.86
    },
    {
      "epoch": 1.5535714285714286,
      "grad_norm": 0.6955702361750687,
      "learning_rate": 1.180459520129118e-06,
      "loss": 0.2581769943237305,
      "num_input_tokens_seen": 739438768,
      "step": 12180,
      "train_runtime": 159052.7638,
      "train_tokens_per_second": 4649.016
    },
    {
      "epoch": 1.5542091836734695,
      "grad_norm": 0.6237399623312642,
      "learning_rate": 1.1772290577325895e-06,
      "loss": 0.26799817085266114,
      "num_input_tokens_seen": 739734376,
      "step": 12185,
      "train_runtime": 159125.1411,
      "train_tokens_per_second": 4648.759
    },
    {
      "epoch": 1.5548469387755102,
      "grad_norm": 1.0016838504443217,
      "learning_rate": 1.174002431753683e-06,
      "loss": 0.25603523254394533,
      "num_input_tokens_seen": 740027464,
      "step": 12190,
      "train_runtime": 159188.7785,
      "train_tokens_per_second": 4648.741
    },
    {
      "epoch": 1.555484693877551,
      "grad_norm": 0.6660423706946417,
      "learning_rate": 1.1707796454305437e-06,
      "loss": 0.23553390502929689,
      "num_input_tokens_seen": 740332856,
      "step": 12195,
      "train_runtime": 159257.0766,
      "train_tokens_per_second": 4648.665
    },
    {
      "epoch": 1.556122448979592,
      "grad_norm": 0.7377728105319677,
      "learning_rate": 1.167560701997462e-06,
      "loss": 0.23785796165466308,
      "num_input_tokens_seen": 740630472,
      "step": 12200,
      "train_runtime": 159322.3265,
      "train_tokens_per_second": 4648.63
    },
    {
      "epoch": 1.5567602040816326,
      "grad_norm": 0.6513154516722532,
      "learning_rate": 1.1643456046848745e-06,
      "loss": 0.2628336429595947,
      "num_input_tokens_seen": 740938432,
      "step": 12205,
      "train_runtime": 159377.7956,
      "train_tokens_per_second": 4648.944
    },
    {
      "epoch": 1.5573979591836735,
      "grad_norm": 0.7078302120378647,
      "learning_rate": 1.161134356719351e-06,
      "loss": 0.24759845733642577,
      "num_input_tokens_seen": 741239992,
      "step": 12210,
      "train_runtime": 159446.4784,
      "train_tokens_per_second": 4648.833
    },
    {
      "epoch": 1.5580357142857144,
      "grad_norm": 0.6849642774899249,
      "learning_rate": 1.1579269613236094e-06,
      "loss": 0.25596597194671633,
      "num_input_tokens_seen": 741543208,
      "step": 12215,
      "train_runtime": 159506.9488,
      "train_tokens_per_second": 4648.971
    },
    {
      "epoch": 1.558673469387755,
      "grad_norm": 0.6957177025296452,
      "learning_rate": 1.1547234217164904e-06,
      "loss": 0.24630370140075683,
      "num_input_tokens_seen": 741853032,
      "step": 12220,
      "train_runtime": 159567.633,
      "train_tokens_per_second": 4649.145
    },
    {
      "epoch": 1.5593112244897958,
      "grad_norm": 0.8110997041912901,
      "learning_rate": 1.1515237411129698e-06,
      "loss": 0.2798213005065918,
      "num_input_tokens_seen": 742146616,
      "step": 12225,
      "train_runtime": 159636.0723,
      "train_tokens_per_second": 4648.991
    },
    {
      "epoch": 1.5599489795918369,
      "grad_norm": 0.5939899974031979,
      "learning_rate": 1.148327922724155e-06,
      "loss": 0.2604410648345947,
      "num_input_tokens_seen": 742453680,
      "step": 12230,
      "train_runtime": 159700.2719,
      "train_tokens_per_second": 4649.045
    },
    {
      "epoch": 1.5605867346938775,
      "grad_norm": 0.6203420194491245,
      "learning_rate": 1.1451359697572684e-06,
      "loss": 0.24161162376403808,
      "num_input_tokens_seen": 742755152,
      "step": 12235,
      "train_runtime": 159772.5733,
      "train_tokens_per_second": 4648.828
    },
    {
      "epoch": 1.5612244897959182,
      "grad_norm": 0.6274593917407443,
      "learning_rate": 1.141947885415659e-06,
      "loss": 0.2437690258026123,
      "num_input_tokens_seen": 743055552,
      "step": 12240,
      "train_runtime": 159836.7285,
      "train_tokens_per_second": 4648.841
    },
    {
      "epoch": 1.5618622448979593,
      "grad_norm": 0.6859634198906147,
      "learning_rate": 1.1387636728987927e-06,
      "loss": 0.26891021728515624,
      "num_input_tokens_seen": 743348152,
      "step": 12245,
      "train_runtime": 159901.2782,
      "train_tokens_per_second": 4648.794
    },
    {
      "epoch": 1.5625,
      "grad_norm": 0.7543946247330733,
      "learning_rate": 1.135583335402249e-06,
      "loss": 0.2813580513000488,
      "num_input_tokens_seen": 743649088,
      "step": 12250,
      "train_runtime": 159975.996,
      "train_tokens_per_second": 4648.504
    },
    {
      "epoch": 1.5631377551020407,
      "grad_norm": 0.7375762686537727,
      "learning_rate": 1.1324068761177176e-06,
      "loss": 0.26078205108642577,
      "num_input_tokens_seen": 743956208,
      "step": 12255,
      "train_runtime": 160033.7913,
      "train_tokens_per_second": 4648.745
    },
    {
      "epoch": 1.5637755102040818,
      "grad_norm": 0.6320469685833705,
      "learning_rate": 1.129234298232999e-06,
      "loss": 0.23866262435913085,
      "num_input_tokens_seen": 744251880,
      "step": 12260,
      "train_runtime": 160105.4615,
      "train_tokens_per_second": 4648.51
    },
    {
      "epoch": 1.5644132653061225,
      "grad_norm": 0.7016221467512594,
      "learning_rate": 1.1260656049319957e-06,
      "loss": 0.25392653942108157,
      "num_input_tokens_seen": 744561384,
      "step": 12265,
      "train_runtime": 160180.8267,
      "train_tokens_per_second": 4648.255
    },
    {
      "epoch": 1.5650510204081631,
      "grad_norm": 0.6206845679523242,
      "learning_rate": 1.1229007993947122e-06,
      "loss": 0.22959496974945068,
      "num_input_tokens_seen": 744861632,
      "step": 12270,
      "train_runtime": 160256.3916,
      "train_tokens_per_second": 4647.937
    },
    {
      "epoch": 1.5656887755102042,
      "grad_norm": 0.635809068968366,
      "learning_rate": 1.119739884797253e-06,
      "loss": 0.2407773494720459,
      "num_input_tokens_seen": 745159520,
      "step": 12275,
      "train_runtime": 160328.2446,
      "train_tokens_per_second": 4647.712
    },
    {
      "epoch": 1.566326530612245,
      "grad_norm": 0.6135957879838808,
      "learning_rate": 1.116582864311816e-06,
      "loss": 0.2587071418762207,
      "num_input_tokens_seen": 745473160,
      "step": 12280,
      "train_runtime": 160398.0401,
      "train_tokens_per_second": 4647.645
    },
    {
      "epoch": 1.5669642857142856,
      "grad_norm": 0.6633182069955801,
      "learning_rate": 1.1134297411066925e-06,
      "loss": 0.25202369689941406,
      "num_input_tokens_seen": 745782512,
      "step": 12285,
      "train_runtime": 160458.8237,
      "train_tokens_per_second": 4647.812
    },
    {
      "epoch": 1.5676020408163265,
      "grad_norm": 0.7290278901024204,
      "learning_rate": 1.1102805183462605e-06,
      "loss": 0.2682202816009521,
      "num_input_tokens_seen": 746079024,
      "step": 12290,
      "train_runtime": 160531.6962,
      "train_tokens_per_second": 4647.55
    },
    {
      "epoch": 1.5682397959183674,
      "grad_norm": 0.680217591331704,
      "learning_rate": 1.1071351991909863e-06,
      "loss": 0.2658061504364014,
      "num_input_tokens_seen": 746387368,
      "step": 12295,
      "train_runtime": 160586.665,
      "train_tokens_per_second": 4647.879
    },
    {
      "epoch": 1.568877551020408,
      "grad_norm": 0.6669425109814144,
      "learning_rate": 1.1039937867974166e-06,
      "loss": 0.24722461700439452,
      "num_input_tokens_seen": 746695480,
      "step": 12300,
      "train_runtime": 160657.1925,
      "train_tokens_per_second": 4647.756
    },
    {
      "epoch": 1.569515306122449,
      "grad_norm": 0.6879081393123695,
      "learning_rate": 1.1008562843181796e-06,
      "loss": 0.24347903728485107,
      "num_input_tokens_seen": 747007624,
      "step": 12305,
      "train_runtime": 160707.9711,
      "train_tokens_per_second": 4648.23
    },
    {
      "epoch": 1.5701530612244898,
      "grad_norm": 0.6805344549483223,
      "learning_rate": 1.0977226949019743e-06,
      "loss": 0.25902271270751953,
      "num_input_tokens_seen": 747317696,
      "step": 12310,
      "train_runtime": 160774.188,
      "train_tokens_per_second": 4648.244
    },
    {
      "epoch": 1.5707908163265305,
      "grad_norm": 0.690352949680991,
      "learning_rate": 1.09459302169358e-06,
      "loss": 0.26403059959411623,
      "num_input_tokens_seen": 747607568,
      "step": 12315,
      "train_runtime": 160840.4773,
      "train_tokens_per_second": 4648.131
    },
    {
      "epoch": 1.5714285714285714,
      "grad_norm": 0.5855693473094103,
      "learning_rate": 1.0914672678338434e-06,
      "loss": 0.24433636665344238,
      "num_input_tokens_seen": 747925216,
      "step": 12320,
      "train_runtime": 160912.6931,
      "train_tokens_per_second": 4648.019
    },
    {
      "epoch": 1.5720663265306123,
      "grad_norm": 0.6064300330678128,
      "learning_rate": 1.088345436459673e-06,
      "loss": 0.253980016708374,
      "num_input_tokens_seen": 748233008,
      "step": 12325,
      "train_runtime": 160964.3216,
      "train_tokens_per_second": 4648.44
    },
    {
      "epoch": 1.572704081632653,
      "grad_norm": 0.6585216485835502,
      "learning_rate": 1.0852275307040454e-06,
      "loss": 0.2468017578125,
      "num_input_tokens_seen": 748531304,
      "step": 12330,
      "train_runtime": 161039.2322,
      "train_tokens_per_second": 4648.13
    },
    {
      "epoch": 1.5733418367346939,
      "grad_norm": 0.9329390462393788,
      "learning_rate": 1.082113553696001e-06,
      "loss": 0.2438680648803711,
      "num_input_tokens_seen": 748829480,
      "step": 12335,
      "train_runtime": 161106.0483,
      "train_tokens_per_second": 4648.053
    },
    {
      "epoch": 1.5739795918367347,
      "grad_norm": 0.6342739903657667,
      "learning_rate": 1.0790035085606282e-06,
      "loss": 0.2425311803817749,
      "num_input_tokens_seen": 749126768,
      "step": 12340,
      "train_runtime": 161177.8775,
      "train_tokens_per_second": 4647.826
    },
    {
      "epoch": 1.5746173469387754,
      "grad_norm": 0.6580289275192713,
      "learning_rate": 1.0758973984190764e-06,
      "loss": 0.2524994373321533,
      "num_input_tokens_seen": 749431224,
      "step": 12345,
      "train_runtime": 161244.4081,
      "train_tokens_per_second": 4647.797
    },
    {
      "epoch": 1.5752551020408163,
      "grad_norm": 0.6234385628360789,
      "learning_rate": 1.072795226388544e-06,
      "loss": 0.2606823444366455,
      "num_input_tokens_seen": 749740744,
      "step": 12350,
      "train_runtime": 161309.4357,
      "train_tokens_per_second": 4647.842
    },
    {
      "epoch": 1.5758928571428572,
      "grad_norm": 0.6702496133284668,
      "learning_rate": 1.0696969955822772e-06,
      "loss": 0.26224217414855955,
      "num_input_tokens_seen": 750049528,
      "step": 12355,
      "train_runtime": 161364.2733,
      "train_tokens_per_second": 4648.176
    },
    {
      "epoch": 1.5765306122448979,
      "grad_norm": 0.755795291137064,
      "learning_rate": 1.0666027091095672e-06,
      "loss": 0.26146037578582765,
      "num_input_tokens_seen": 750354768,
      "step": 12360,
      "train_runtime": 161422.5277,
      "train_tokens_per_second": 4648.389
    },
    {
      "epoch": 1.5771683673469388,
      "grad_norm": 0.6517319605002192,
      "learning_rate": 1.063512370075746e-06,
      "loss": 0.26197237968444825,
      "num_input_tokens_seen": 750665432,
      "step": 12365,
      "train_runtime": 161496.964,
      "train_tokens_per_second": 4648.171
    },
    {
      "epoch": 1.5778061224489797,
      "grad_norm": 0.651251326730073,
      "learning_rate": 1.060425981582186e-06,
      "loss": 0.2529475688934326,
      "num_input_tokens_seen": 750965168,
      "step": 12370,
      "train_runtime": 161567.0836,
      "train_tokens_per_second": 4648.008
    },
    {
      "epoch": 1.5784438775510203,
      "grad_norm": 0.7758776621921266,
      "learning_rate": 1.0573435467262922e-06,
      "loss": 0.2558449268341064,
      "num_input_tokens_seen": 751272352,
      "step": 12375,
      "train_runtime": 161630.7318,
      "train_tokens_per_second": 4648.079
    },
    {
      "epoch": 1.5790816326530612,
      "grad_norm": 0.7109361021397126,
      "learning_rate": 1.0542650686015039e-06,
      "loss": 0.25114145278930666,
      "num_input_tokens_seen": 751568720,
      "step": 12380,
      "train_runtime": 161690.0745,
      "train_tokens_per_second": 4648.206
    },
    {
      "epoch": 1.5797193877551021,
      "grad_norm": 0.7577522354548263,
      "learning_rate": 1.0511905502972885e-06,
      "loss": 0.2342022180557251,
      "num_input_tokens_seen": 751854528,
      "step": 12385,
      "train_runtime": 161756.5953,
      "train_tokens_per_second": 4648.061
    },
    {
      "epoch": 1.5803571428571428,
      "grad_norm": 0.6628391627410214,
      "learning_rate": 1.0481199948991416e-06,
      "loss": 0.2556438446044922,
      "num_input_tokens_seen": 752152672,
      "step": 12390,
      "train_runtime": 161827.3768,
      "train_tokens_per_second": 4647.87
    },
    {
      "epoch": 1.5809948979591837,
      "grad_norm": 0.5925759235289038,
      "learning_rate": 1.0450534054885765e-06,
      "loss": 0.24029622077941895,
      "num_input_tokens_seen": 752451632,
      "step": 12395,
      "train_runtime": 161894.0761,
      "train_tokens_per_second": 4647.802
    },
    {
      "epoch": 1.5816326530612246,
      "grad_norm": 0.6436167380481044,
      "learning_rate": 1.0419907851431321e-06,
      "loss": 0.24811995029449463,
      "num_input_tokens_seen": 752748848,
      "step": 12400,
      "train_runtime": 161967.8942,
      "train_tokens_per_second": 4647.519
    },
    {
      "epoch": 1.5822704081632653,
      "grad_norm": 0.5683408987596007,
      "learning_rate": 1.0389321369363636e-06,
      "loss": 0.252943229675293,
      "num_input_tokens_seen": 753048504,
      "step": 12405,
      "train_runtime": 162044.2892,
      "train_tokens_per_second": 4647.177
    },
    {
      "epoch": 1.5829081632653061,
      "grad_norm": 0.7459873593609917,
      "learning_rate": 1.0358774639378344e-06,
      "loss": 0.24510138034820556,
      "num_input_tokens_seen": 753350536,
      "step": 12410,
      "train_runtime": 162106.7656,
      "train_tokens_per_second": 4647.249
    },
    {
      "epoch": 1.583545918367347,
      "grad_norm": 0.6015788491277244,
      "learning_rate": 1.0328267692131227e-06,
      "loss": 0.23749666213989257,
      "num_input_tokens_seen": 753655064,
      "step": 12415,
      "train_runtime": 162181.177,
      "train_tokens_per_second": 4646.995
    },
    {
      "epoch": 1.5841836734693877,
      "grad_norm": 0.6346068440396071,
      "learning_rate": 1.0297800558238175e-06,
      "loss": 0.2700827598571777,
      "num_input_tokens_seen": 753965960,
      "step": 12420,
      "train_runtime": 162241.871,
      "train_tokens_per_second": 4647.172
    },
    {
      "epoch": 1.5848214285714286,
      "grad_norm": 0.6181126675184303,
      "learning_rate": 1.026737326827505e-06,
      "loss": 0.2491556167602539,
      "num_input_tokens_seen": 754261800,
      "step": 12425,
      "train_runtime": 162314.88,
      "train_tokens_per_second": 4646.905
    },
    {
      "epoch": 1.5854591836734695,
      "grad_norm": 0.6400739322880858,
      "learning_rate": 1.0236985852777748e-06,
      "loss": 0.22559471130371095,
      "num_input_tokens_seen": 754566128,
      "step": 12430,
      "train_runtime": 162386.383,
      "train_tokens_per_second": 4646.733
    },
    {
      "epoch": 1.5860969387755102,
      "grad_norm": 0.6611720183258928,
      "learning_rate": 1.020663834224222e-06,
      "loss": 0.25476698875427245,
      "num_input_tokens_seen": 754874792,
      "step": 12435,
      "train_runtime": 162439.4803,
      "train_tokens_per_second": 4647.114
    },
    {
      "epoch": 1.586734693877551,
      "grad_norm": 0.6717042386516047,
      "learning_rate": 1.017633076712426e-06,
      "loss": 0.2633042335510254,
      "num_input_tokens_seen": 755183832,
      "step": 12440,
      "train_runtime": 162507.3109,
      "train_tokens_per_second": 4647.076
    },
    {
      "epoch": 1.587372448979592,
      "grad_norm": 0.6448707254906307,
      "learning_rate": 1.0146063157839653e-06,
      "loss": 0.24064431190490723,
      "num_input_tokens_seen": 755483224,
      "step": 12445,
      "train_runtime": 162562.8587,
      "train_tokens_per_second": 4647.33
    },
    {
      "epoch": 1.5880102040816326,
      "grad_norm": 0.6782417280961641,
      "learning_rate": 1.0115835544764063e-06,
      "loss": 0.2448643207550049,
      "num_input_tokens_seen": 755793864,
      "step": 12450,
      "train_runtime": 162642.277,
      "train_tokens_per_second": 4646.971
    },
    {
      "epoch": 1.5886479591836735,
      "grad_norm": 0.730241214502829,
      "learning_rate": 1.0085647958233008e-06,
      "loss": 0.23660359382629395,
      "num_input_tokens_seen": 756095472,
      "step": 12455,
      "train_runtime": 162703.9159,
      "train_tokens_per_second": 4647.064
    },
    {
      "epoch": 1.5892857142857144,
      "grad_norm": 0.6339507181811259,
      "learning_rate": 1.0055500428541836e-06,
      "loss": 0.23841919898986816,
      "num_input_tokens_seen": 756410888,
      "step": 12460,
      "train_runtime": 162769.9956,
      "train_tokens_per_second": 4647.115
    },
    {
      "epoch": 1.589923469387755,
      "grad_norm": 0.7954503791909062,
      "learning_rate": 1.0025392985945703e-06,
      "loss": 0.25959603786468505,
      "num_input_tokens_seen": 756716856,
      "step": 12465,
      "train_runtime": 162823.6443,
      "train_tokens_per_second": 4647.463
    },
    {
      "epoch": 1.5905612244897958,
      "grad_norm": 0.6330852145188037,
      "learning_rate": 9.995325660659533e-07,
      "loss": 0.25395450592041013,
      "num_input_tokens_seen": 757020568,
      "step": 12470,
      "train_runtime": 162902.8794,
      "train_tokens_per_second": 4647.067
    },
    {
      "epoch": 1.5911989795918369,
      "grad_norm": 0.6568184414599136,
      "learning_rate": 9.965298482857993e-07,
      "loss": 0.25415830612182616,
      "num_input_tokens_seen": 757323152,
      "step": 12475,
      "train_runtime": 162964.734,
      "train_tokens_per_second": 4647.16
    },
    {
      "epoch": 1.5918367346938775,
      "grad_norm": 0.6240054378691067,
      "learning_rate": 9.935311482675413e-07,
      "loss": 0.2535807371139526,
      "num_input_tokens_seen": 757630000,
      "step": 12480,
      "train_runtime": 163028.3631,
      "train_tokens_per_second": 4647.228
    },
    {
      "epoch": 1.5924744897959182,
      "grad_norm": 0.6457252261362081,
      "learning_rate": 9.905364690205888e-07,
      "loss": 0.2578327417373657,
      "num_input_tokens_seen": 757940600,
      "step": 12485,
      "train_runtime": 163092.1774,
      "train_tokens_per_second": 4647.314
    },
    {
      "epoch": 1.5931122448979593,
      "grad_norm": 0.6839511232994004,
      "learning_rate": 9.875458135503113e-07,
      "loss": 0.26314606666564944,
      "num_input_tokens_seen": 758251840,
      "step": 12490,
      "train_runtime": 163157.0738,
      "train_tokens_per_second": 4647.373
    },
    {
      "epoch": 1.59375,
      "grad_norm": 0.663254212663678,
      "learning_rate": 9.845591848580376e-07,
      "loss": 0.23812398910522461,
      "num_input_tokens_seen": 758555896,
      "step": 12495,
      "train_runtime": 163235.6326,
      "train_tokens_per_second": 4646.999
    },
    {
      "epoch": 1.5943877551020407,
      "grad_norm": 0.6812542727199498,
      "learning_rate": 9.815765859410586e-07,
      "loss": 0.269971227645874,
      "num_input_tokens_seen": 758850688,
      "step": 12500,
      "train_runtime": 163302.0593,
      "train_tokens_per_second": 4646.914
    },
    {
      "epoch": 1.5950255102040818,
      "grad_norm": 0.6430649487312459,
      "learning_rate": 9.785980197926242e-07,
      "loss": 0.24490811824798583,
      "num_input_tokens_seen": 759157776,
      "step": 12505,
      "train_runtime": 163369.5905,
      "train_tokens_per_second": 4646.873
    },
    {
      "epoch": 1.5956632653061225,
      "grad_norm": 0.6769972255789957,
      "learning_rate": 9.7562348940193e-07,
      "loss": 0.25566740036010743,
      "num_input_tokens_seen": 759468720,
      "step": 12510,
      "train_runtime": 163438.4854,
      "train_tokens_per_second": 4646.817
    },
    {
      "epoch": 1.5963010204081631,
      "grad_norm": 0.6413108724848737,
      "learning_rate": 9.726529977541243e-07,
      "loss": 0.24758310317993165,
      "num_input_tokens_seen": 759783072,
      "step": 12515,
      "train_runtime": 163496.7421,
      "train_tokens_per_second": 4647.084
    },
    {
      "epoch": 1.5969387755102042,
      "grad_norm": 0.6876242906005937,
      "learning_rate": 9.696865478303075e-07,
      "loss": 0.24775145053863526,
      "num_input_tokens_seen": 760088384,
      "step": 12520,
      "train_runtime": 163559.2065,
      "train_tokens_per_second": 4647.176
    },
    {
      "epoch": 1.597576530612245,
      "grad_norm": 0.6154892690237602,
      "learning_rate": 9.667241426075152e-07,
      "loss": 0.26232051849365234,
      "num_input_tokens_seen": 760400232,
      "step": 12525,
      "train_runtime": 163628.499,
      "train_tokens_per_second": 4647.114
    },
    {
      "epoch": 1.5982142857142856,
      "grad_norm": 0.7051712926015147,
      "learning_rate": 9.637657850587289e-07,
      "loss": 0.25011744499206545,
      "num_input_tokens_seen": 760702064,
      "step": 12530,
      "train_runtime": 163695.1537,
      "train_tokens_per_second": 4647.065
    },
    {
      "epoch": 1.5988520408163265,
      "grad_norm": 0.6263122887546746,
      "learning_rate": 9.608114781528671e-07,
      "loss": 0.22640318870544435,
      "num_input_tokens_seen": 760993656,
      "step": 12535,
      "train_runtime": 163763.9541,
      "train_tokens_per_second": 4646.894
    },
    {
      "epoch": 1.5994897959183674,
      "grad_norm": 0.8739539461244864,
      "learning_rate": 9.578612248547836e-07,
      "loss": 0.25423121452331543,
      "num_input_tokens_seen": 761293168,
      "step": 12540,
      "train_runtime": 163831.0134,
      "train_tokens_per_second": 4646.82
    },
    {
      "epoch": 1.600127551020408,
      "grad_norm": 0.6684990455461551,
      "learning_rate": 9.549150281252633e-07,
      "loss": 0.24723145961761475,
      "num_input_tokens_seen": 761597688,
      "step": 12545,
      "train_runtime": 163894.8582,
      "train_tokens_per_second": 4646.867
    },
    {
      "epoch": 1.600765306122449,
      "grad_norm": 0.6423018731637105,
      "learning_rate": 9.519728909210202e-07,
      "loss": 0.2562461614608765,
      "num_input_tokens_seen": 761898112,
      "step": 12550,
      "train_runtime": 163966.9091,
      "train_tokens_per_second": 4646.658
    },
    {
      "epoch": 1.6014030612244898,
      "grad_norm": 0.6717204273912907,
      "learning_rate": 9.49034816194695e-07,
      "loss": 0.24855079650878906,
      "num_input_tokens_seen": 762201840,
      "step": 12555,
      "train_runtime": 164025.806,
      "train_tokens_per_second": 4646.841
    },
    {
      "epoch": 1.6020408163265305,
      "grad_norm": 0.6531453875530732,
      "learning_rate": 9.461008068948502e-07,
      "loss": 0.25662262439727784,
      "num_input_tokens_seen": 762511008,
      "step": 12560,
      "train_runtime": 164092.7065,
      "train_tokens_per_second": 4646.831
    },
    {
      "epoch": 1.6026785714285714,
      "grad_norm": 0.6070432186468975,
      "learning_rate": 9.431708659659694e-07,
      "loss": 0.2622831344604492,
      "num_input_tokens_seen": 762819600,
      "step": 12565,
      "train_runtime": 164154.4675,
      "train_tokens_per_second": 4646.962
    },
    {
      "epoch": 1.6033163265306123,
      "grad_norm": 0.73515273248632,
      "learning_rate": 9.402449963484528e-07,
      "loss": 0.2846518516540527,
      "num_input_tokens_seen": 763114568,
      "step": 12570,
      "train_runtime": 164218.8269,
      "train_tokens_per_second": 4646.937
    },
    {
      "epoch": 1.603954081632653,
      "grad_norm": 0.6320718832689852,
      "learning_rate": 9.373232009786154e-07,
      "loss": 0.25565176010131835,
      "num_input_tokens_seen": 763426824,
      "step": 12575,
      "train_runtime": 164290.0544,
      "train_tokens_per_second": 4646.823
    },
    {
      "epoch": 1.6045918367346939,
      "grad_norm": 0.6435414856314439,
      "learning_rate": 9.344054827886839e-07,
      "loss": 0.24167089462280272,
      "num_input_tokens_seen": 763735968,
      "step": 12580,
      "train_runtime": 164349.7934,
      "train_tokens_per_second": 4647.015
    },
    {
      "epoch": 1.6052295918367347,
      "grad_norm": 0.6775080813258296,
      "learning_rate": 9.314918447067877e-07,
      "loss": 0.25985682010650635,
      "num_input_tokens_seen": 764041712,
      "step": 12585,
      "train_runtime": 164415.7592,
      "train_tokens_per_second": 4647.01
    },
    {
      "epoch": 1.6058673469387754,
      "grad_norm": 0.6539921493558628,
      "learning_rate": 9.285822896569724e-07,
      "loss": 0.2366727590560913,
      "num_input_tokens_seen": 764362000,
      "step": 12590,
      "train_runtime": 164477.1793,
      "train_tokens_per_second": 4647.222
    },
    {
      "epoch": 1.6065051020408163,
      "grad_norm": 0.6916270860827861,
      "learning_rate": 9.256768205591754e-07,
      "loss": 0.2414870262145996,
      "num_input_tokens_seen": 764667152,
      "step": 12595,
      "train_runtime": 164546.0853,
      "train_tokens_per_second": 4647.131
    },
    {
      "epoch": 1.6071428571428572,
      "grad_norm": 0.5895810637455151,
      "learning_rate": 9.227754403292377e-07,
      "loss": 0.2509591579437256,
      "num_input_tokens_seen": 764968712,
      "step": 12600,
      "train_runtime": 164593.4889,
      "train_tokens_per_second": 4647.624
    },
    {
      "epoch": 1.6077806122448979,
      "grad_norm": 0.7110181866293258,
      "learning_rate": 9.198781518789007e-07,
      "loss": 0.25958638191223143,
      "num_input_tokens_seen": 765281152,
      "step": 12605,
      "train_runtime": 164650.7227,
      "train_tokens_per_second": 4647.906
    },
    {
      "epoch": 1.6084183673469388,
      "grad_norm": 0.6421934326938831,
      "learning_rate": 9.169849581157925e-07,
      "loss": 0.26078200340270996,
      "num_input_tokens_seen": 765588280,
      "step": 12610,
      "train_runtime": 164710.0853,
      "train_tokens_per_second": 4648.096
    },
    {
      "epoch": 1.6090561224489797,
      "grad_norm": 0.7019644430031099,
      "learning_rate": 9.140958619434353e-07,
      "loss": 0.24608101844787597,
      "num_input_tokens_seen": 765895232,
      "step": 12615,
      "train_runtime": 164769.8889,
      "train_tokens_per_second": 4648.272
    },
    {
      "epoch": 1.6096938775510203,
      "grad_norm": 0.6613627444193002,
      "learning_rate": 9.112108662612401e-07,
      "loss": 0.24444227218627929,
      "num_input_tokens_seen": 766209152,
      "step": 12620,
      "train_runtime": 164829.6191,
      "train_tokens_per_second": 4648.492
    },
    {
      "epoch": 1.6103316326530612,
      "grad_norm": 0.6655626500240223,
      "learning_rate": 9.083299739645007e-07,
      "loss": 0.25504388809204104,
      "num_input_tokens_seen": 766515456,
      "step": 12625,
      "train_runtime": 164890.7621,
      "train_tokens_per_second": 4648.626
    },
    {
      "epoch": 1.6109693877551021,
      "grad_norm": 0.6664621269846308,
      "learning_rate": 9.054531879443945e-07,
      "loss": 0.24010090827941893,
      "num_input_tokens_seen": 766811560,
      "step": 12630,
      "train_runtime": 164954.2327,
      "train_tokens_per_second": 4648.632
    },
    {
      "epoch": 1.6116071428571428,
      "grad_norm": 0.7012525846119625,
      "learning_rate": 9.025805110879765e-07,
      "loss": 0.23866443634033202,
      "num_input_tokens_seen": 767105144,
      "step": 12635,
      "train_runtime": 165026.8157,
      "train_tokens_per_second": 4648.367
    },
    {
      "epoch": 1.6122448979591837,
      "grad_norm": 0.6761558011488014,
      "learning_rate": 8.997119462781784e-07,
      "loss": 0.26564598083496094,
      "num_input_tokens_seen": 767407136,
      "step": 12640,
      "train_runtime": 165088.5375,
      "train_tokens_per_second": 4648.458
    },
    {
      "epoch": 1.6128826530612246,
      "grad_norm": 0.6257364716430434,
      "learning_rate": 8.968474963938062e-07,
      "loss": 0.2686514377593994,
      "num_input_tokens_seen": 767714440,
      "step": 12645,
      "train_runtime": 165147.3166,
      "train_tokens_per_second": 4648.664
    },
    {
      "epoch": 1.6135204081632653,
      "grad_norm": 0.6518750582262893,
      "learning_rate": 8.939871643095344e-07,
      "loss": 0.24472124576568605,
      "num_input_tokens_seen": 768028176,
      "step": 12650,
      "train_runtime": 165214.0288,
      "train_tokens_per_second": 4648.686
    },
    {
      "epoch": 1.6141581632653061,
      "grad_norm": 0.9517786978644486,
      "learning_rate": 8.911309528959072e-07,
      "loss": 0.23885109424591064,
      "num_input_tokens_seen": 768310720,
      "step": 12655,
      "train_runtime": 165278.177,
      "train_tokens_per_second": 4648.591
    },
    {
      "epoch": 1.614795918367347,
      "grad_norm": 0.6710515862145993,
      "learning_rate": 8.882788650193308e-07,
      "loss": 0.25642588138580324,
      "num_input_tokens_seen": 768609424,
      "step": 12660,
      "train_runtime": 165350.9736,
      "train_tokens_per_second": 4648.351
    },
    {
      "epoch": 1.6154336734693877,
      "grad_norm": 0.7388780230675929,
      "learning_rate": 8.854309035420772e-07,
      "loss": 0.24589982032775878,
      "num_input_tokens_seen": 768910864,
      "step": 12665,
      "train_runtime": 165424.8765,
      "train_tokens_per_second": 4648.097
    },
    {
      "epoch": 1.6160714285714286,
      "grad_norm": 0.5957853216469311,
      "learning_rate": 8.8258707132227e-07,
      "loss": 0.24904451370239258,
      "num_input_tokens_seen": 769224144,
      "step": 12670,
      "train_runtime": 165489.6155,
      "train_tokens_per_second": 4648.172
    },
    {
      "epoch": 1.6167091836734695,
      "grad_norm": 0.7132504903398998,
      "learning_rate": 8.797473712138976e-07,
      "loss": 0.2533883571624756,
      "num_input_tokens_seen": 769534752,
      "step": 12675,
      "train_runtime": 165557.229,
      "train_tokens_per_second": 4648.15
    },
    {
      "epoch": 1.6173469387755102,
      "grad_norm": 1.095703129087896,
      "learning_rate": 8.769118060667969e-07,
      "loss": 0.26964540481567384,
      "num_input_tokens_seen": 769849432,
      "step": 12680,
      "train_runtime": 165623.9959,
      "train_tokens_per_second": 4648.176
    },
    {
      "epoch": 1.617984693877551,
      "grad_norm": 0.5650373890406885,
      "learning_rate": 8.740803787266522e-07,
      "loss": 0.24618177413940429,
      "num_input_tokens_seen": 770147568,
      "step": 12685,
      "train_runtime": 165689.3216,
      "train_tokens_per_second": 4648.142
    },
    {
      "epoch": 1.618622448979592,
      "grad_norm": 0.691918460107524,
      "learning_rate": 8.712530920350026e-07,
      "loss": 0.2687231540679932,
      "num_input_tokens_seen": 770452992,
      "step": 12690,
      "train_runtime": 165756.7138,
      "train_tokens_per_second": 4648.095
    },
    {
      "epoch": 1.6192602040816326,
      "grad_norm": 0.6717102323006657,
      "learning_rate": 8.684299488292275e-07,
      "loss": 0.25114462375640867,
      "num_input_tokens_seen": 770747832,
      "step": 12695,
      "train_runtime": 165831.2316,
      "train_tokens_per_second": 4647.785
    },
    {
      "epoch": 1.6198979591836735,
      "grad_norm": 0.6485772922668345,
      "learning_rate": 8.656109519425454e-07,
      "loss": 0.24409866333007812,
      "num_input_tokens_seen": 771058800,
      "step": 12700,
      "train_runtime": 165893.6085,
      "train_tokens_per_second": 4647.911
    },
    {
      "epoch": 1.6205357142857144,
      "grad_norm": 0.6370709899706526,
      "learning_rate": 8.627961042040183e-07,
      "loss": 0.260186767578125,
      "num_input_tokens_seen": 771367112,
      "step": 12705,
      "train_runtime": 165957.67,
      "train_tokens_per_second": 4647.975
    },
    {
      "epoch": 1.621173469387755,
      "grad_norm": 0.6552549424628946,
      "learning_rate": 8.599854084385422e-07,
      "loss": 0.2518947601318359,
      "num_input_tokens_seen": 771673080,
      "step": 12710,
      "train_runtime": 166022.1598,
      "train_tokens_per_second": 4648.013
    },
    {
      "epoch": 1.6218112244897958,
      "grad_norm": 0.6712303761106847,
      "learning_rate": 8.571788674668469e-07,
      "loss": 0.2479722499847412,
      "num_input_tokens_seen": 771961968,
      "step": 12715,
      "train_runtime": 166094.98,
      "train_tokens_per_second": 4647.714
    },
    {
      "epoch": 1.6224489795918369,
      "grad_norm": 0.6403917027777989,
      "learning_rate": 8.543764841054919e-07,
      "loss": 0.2650299072265625,
      "num_input_tokens_seen": 772271248,
      "step": 12720,
      "train_runtime": 166157.5141,
      "train_tokens_per_second": 4647.826
    },
    {
      "epoch": 1.6230867346938775,
      "grad_norm": 0.6207948302831311,
      "learning_rate": 8.515782611668649e-07,
      "loss": 0.24235401153564454,
      "num_input_tokens_seen": 772580152,
      "step": 12725,
      "train_runtime": 166222.6592,
      "train_tokens_per_second": 4647.863
    },
    {
      "epoch": 1.6237244897959182,
      "grad_norm": 0.6235366276277365,
      "learning_rate": 8.487842014591779e-07,
      "loss": 0.2494971752166748,
      "num_input_tokens_seen": 772893608,
      "step": 12730,
      "train_runtime": 166298.9091,
      "train_tokens_per_second": 4647.617
    },
    {
      "epoch": 1.6243622448979593,
      "grad_norm": 0.6475420590957922,
      "learning_rate": 8.459943077864651e-07,
      "loss": 0.2487192392349243,
      "num_input_tokens_seen": 773206664,
      "step": 12735,
      "train_runtime": 166357.1118,
      "train_tokens_per_second": 4647.873
    },
    {
      "epoch": 1.625,
      "grad_norm": 0.6043118267700197,
      "learning_rate": 8.432085829485792e-07,
      "loss": 0.23452670574188234,
      "num_input_tokens_seen": 773511840,
      "step": 12740,
      "train_runtime": 166419.6854,
      "train_tokens_per_second": 4647.959
    },
    {
      "epoch": 1.6256377551020407,
      "grad_norm": 0.6258257029860365,
      "learning_rate": 8.404270297411904e-07,
      "loss": 0.23406319618225097,
      "num_input_tokens_seen": 773795056,
      "step": 12745,
      "train_runtime": 166489.5778,
      "train_tokens_per_second": 4647.709
    },
    {
      "epoch": 1.6262755102040818,
      "grad_norm": 0.6748373692673187,
      "learning_rate": 8.376496509557808e-07,
      "loss": 0.24109892845153807,
      "num_input_tokens_seen": 774086496,
      "step": 12750,
      "train_runtime": 166558.0852,
      "train_tokens_per_second": 4647.547
    },
    {
      "epoch": 1.6269132653061225,
      "grad_norm": 0.5912168780820537,
      "learning_rate": 8.348764493796447e-07,
      "loss": 0.2351205587387085,
      "num_input_tokens_seen": 774407344,
      "step": 12755,
      "train_runtime": 166612.9101,
      "train_tokens_per_second": 4647.943
    },
    {
      "epoch": 1.6275510204081631,
      "grad_norm": 0.6398902840648175,
      "learning_rate": 8.321074277958835e-07,
      "loss": 0.22835516929626465,
      "num_input_tokens_seen": 774713704,
      "step": 12760,
      "train_runtime": 166680.5548,
      "train_tokens_per_second": 4647.895
    },
    {
      "epoch": 1.6281887755102042,
      "grad_norm": 0.600158707800418,
      "learning_rate": 8.293425889834045e-07,
      "loss": 0.24194543361663817,
      "num_input_tokens_seen": 775024480,
      "step": 12765,
      "train_runtime": 166748.931,
      "train_tokens_per_second": 4647.853
    },
    {
      "epoch": 1.628826530612245,
      "grad_norm": 0.649943783495159,
      "learning_rate": 8.265819357169124e-07,
      "loss": 0.25502166748046873,
      "num_input_tokens_seen": 775311536,
      "step": 12770,
      "train_runtime": 166819.561,
      "train_tokens_per_second": 4647.606
    },
    {
      "epoch": 1.6294642857142856,
      "grad_norm": 0.6760333282071129,
      "learning_rate": 8.238254707669208e-07,
      "loss": 0.24874403476715087,
      "num_input_tokens_seen": 775627888,
      "step": 12775,
      "train_runtime": 166894.9868,
      "train_tokens_per_second": 4647.401
    },
    {
      "epoch": 1.6301020408163265,
      "grad_norm": 0.6916972987662771,
      "learning_rate": 8.210731968997331e-07,
      "loss": 0.2492523431777954,
      "num_input_tokens_seen": 775932416,
      "step": 12780,
      "train_runtime": 166959.6287,
      "train_tokens_per_second": 4647.425
    },
    {
      "epoch": 1.6307397959183674,
      "grad_norm": 0.6405934150651219,
      "learning_rate": 8.183251168774476e-07,
      "loss": 0.26210508346557615,
      "num_input_tokens_seen": 776250104,
      "step": 12785,
      "train_runtime": 167024.186,
      "train_tokens_per_second": 4647.531
    },
    {
      "epoch": 1.631377551020408,
      "grad_norm": 0.6984561798431718,
      "learning_rate": 8.155812334579532e-07,
      "loss": 0.26555054187774657,
      "num_input_tokens_seen": 776555176,
      "step": 12790,
      "train_runtime": 167088.2304,
      "train_tokens_per_second": 4647.576
    },
    {
      "epoch": 1.632015306122449,
      "grad_norm": 0.6725511432312967,
      "learning_rate": 8.128415493949326e-07,
      "loss": 0.24640944004058837,
      "num_input_tokens_seen": 776859304,
      "step": 12795,
      "train_runtime": 167158.2613,
      "train_tokens_per_second": 4647.448
    },
    {
      "epoch": 1.6326530612244898,
      "grad_norm": 0.7031564911090373,
      "learning_rate": 8.101060674378463e-07,
      "loss": 0.25874876976013184,
      "num_input_tokens_seen": 777159832,
      "step": 12800,
      "train_runtime": 167227.9464,
      "train_tokens_per_second": 4647.308
    },
    {
      "epoch": 1.6332908163265305,
      "grad_norm": 0.6419896032981751,
      "learning_rate": 8.073747903319418e-07,
      "loss": 0.24507718086242675,
      "num_input_tokens_seen": 777445384,
      "step": 12805,
      "train_runtime": 167288.2821,
      "train_tokens_per_second": 4647.339
    },
    {
      "epoch": 1.6339285714285714,
      "grad_norm": 0.6945349273462113,
      "learning_rate": 8.046477208182501e-07,
      "loss": 0.268278980255127,
      "num_input_tokens_seen": 777738600,
      "step": 12810,
      "train_runtime": 167357.3929,
      "train_tokens_per_second": 4647.172
    },
    {
      "epoch": 1.6345663265306123,
      "grad_norm": 0.6192297441163143,
      "learning_rate": 8.019248616335717e-07,
      "loss": 0.26207268238067627,
      "num_input_tokens_seen": 778042536,
      "step": 12815,
      "train_runtime": 167423.4732,
      "train_tokens_per_second": 4647.153
    },
    {
      "epoch": 1.635204081632653,
      "grad_norm": 0.6679164110253424,
      "learning_rate": 7.992062155104879e-07,
      "loss": 0.2402644395828247,
      "num_input_tokens_seen": 778347480,
      "step": 12820,
      "train_runtime": 167485.8631,
      "train_tokens_per_second": 4647.243
    },
    {
      "epoch": 1.6358418367346939,
      "grad_norm": 0.7196758258318408,
      "learning_rate": 7.964917851773496e-07,
      "loss": 0.263356351852417,
      "num_input_tokens_seen": 778652856,
      "step": 12825,
      "train_runtime": 167554.5656,
      "train_tokens_per_second": 4647.16
    },
    {
      "epoch": 1.6364795918367347,
      "grad_norm": 0.6713829114133978,
      "learning_rate": 7.937815733582771e-07,
      "loss": 0.26779518127441404,
      "num_input_tokens_seen": 778960296,
      "step": 12830,
      "train_runtime": 167614.0902,
      "train_tokens_per_second": 4647.344
    },
    {
      "epoch": 1.6371173469387754,
      "grad_norm": 0.5928561672777125,
      "learning_rate": 7.910755827731576e-07,
      "loss": 0.23808066844940184,
      "num_input_tokens_seen": 779262384,
      "step": 12835,
      "train_runtime": 167687.292,
      "train_tokens_per_second": 4647.117
    },
    {
      "epoch": 1.6377551020408163,
      "grad_norm": 0.6671776925554667,
      "learning_rate": 7.883738161376414e-07,
      "loss": 0.23737592697143556,
      "num_input_tokens_seen": 779563184,
      "step": 12840,
      "train_runtime": 167748.5715,
      "train_tokens_per_second": 4647.212
    },
    {
      "epoch": 1.6383928571428572,
      "grad_norm": 0.6871976813821795,
      "learning_rate": 7.856762761631398e-07,
      "loss": 0.24407224655151366,
      "num_input_tokens_seen": 779881240,
      "step": 12845,
      "train_runtime": 167818.4208,
      "train_tokens_per_second": 4647.173
    },
    {
      "epoch": 1.6390306122448979,
      "grad_norm": 0.7142196653106222,
      "learning_rate": 7.829829655568239e-07,
      "loss": 0.2409985065460205,
      "num_input_tokens_seen": 780172920,
      "step": 12850,
      "train_runtime": 167881.531,
      "train_tokens_per_second": 4647.163
    },
    {
      "epoch": 1.6396683673469388,
      "grad_norm": 0.7376532862214654,
      "learning_rate": 7.802938870216159e-07,
      "loss": 0.255547308921814,
      "num_input_tokens_seen": 780480104,
      "step": 12855,
      "train_runtime": 167948.5035,
      "train_tokens_per_second": 4647.139
    },
    {
      "epoch": 1.6403061224489797,
      "grad_norm": 0.6089057538564588,
      "learning_rate": 7.776090432561967e-07,
      "loss": 0.229118275642395,
      "num_input_tokens_seen": 780783224,
      "step": 12860,
      "train_runtime": 168011.6113,
      "train_tokens_per_second": 4647.198
    },
    {
      "epoch": 1.6409438775510203,
      "grad_norm": 0.7440044578666771,
      "learning_rate": 7.749284369549954e-07,
      "loss": 0.27794227600097654,
      "num_input_tokens_seen": 781090976,
      "step": 12865,
      "train_runtime": 168076.9357,
      "train_tokens_per_second": 4647.223
    },
    {
      "epoch": 1.6415816326530612,
      "grad_norm": 0.660081761218456,
      "learning_rate": 7.722520708081843e-07,
      "loss": 0.24514400959014893,
      "num_input_tokens_seen": 781398792,
      "step": 12870,
      "train_runtime": 168144.6422,
      "train_tokens_per_second": 4647.182
    },
    {
      "epoch": 1.6422193877551021,
      "grad_norm": 0.6957063767801462,
      "learning_rate": 7.695799475016846e-07,
      "loss": 0.2551210880279541,
      "num_input_tokens_seen": 781715312,
      "step": 12875,
      "train_runtime": 168195.7598,
      "train_tokens_per_second": 4647.652
    },
    {
      "epoch": 1.6428571428571428,
      "grad_norm": 0.6920126738404653,
      "learning_rate": 7.669120697171617e-07,
      "loss": 0.24247095584869385,
      "num_input_tokens_seen": 782018872,
      "step": 12880,
      "train_runtime": 168266.8482,
      "train_tokens_per_second": 4647.492
    },
    {
      "epoch": 1.6434948979591837,
      "grad_norm": 0.7161977483184672,
      "learning_rate": 7.642484401320138e-07,
      "loss": 0.2649207592010498,
      "num_input_tokens_seen": 782310112,
      "step": 12885,
      "train_runtime": 168338.2235,
      "train_tokens_per_second": 4647.252
    },
    {
      "epoch": 1.6441326530612246,
      "grad_norm": 0.645323956347639,
      "learning_rate": 7.615890614193788e-07,
      "loss": 0.2558600902557373,
      "num_input_tokens_seen": 782613848,
      "step": 12890,
      "train_runtime": 168402.0099,
      "train_tokens_per_second": 4647.295
    },
    {
      "epoch": 1.6447704081632653,
      "grad_norm": 0.7350749728534063,
      "learning_rate": 7.589339362481334e-07,
      "loss": 0.26041407585144044,
      "num_input_tokens_seen": 782912536,
      "step": 12895,
      "train_runtime": 168479.4939,
      "train_tokens_per_second": 4646.931
    },
    {
      "epoch": 1.6454081632653061,
      "grad_norm": 0.7315785775996593,
      "learning_rate": 7.562830672828775e-07,
      "loss": 0.2617289304733276,
      "num_input_tokens_seen": 783219264,
      "step": 12900,
      "train_runtime": 168536.6184,
      "train_tokens_per_second": 4647.176
    },
    {
      "epoch": 1.646045918367347,
      "grad_norm": 0.5986714401284035,
      "learning_rate": 7.536364571839439e-07,
      "loss": 0.26701955795288085,
      "num_input_tokens_seen": 783532200,
      "step": 12905,
      "train_runtime": 168590.2163,
      "train_tokens_per_second": 4647.554
    },
    {
      "epoch": 1.6466836734693877,
      "grad_norm": 0.6373404485150242,
      "learning_rate": 7.509941086073907e-07,
      "loss": 0.26680240631103513,
      "num_input_tokens_seen": 783844176,
      "step": 12910,
      "train_runtime": 168658.7672,
      "train_tokens_per_second": 4647.515
    },
    {
      "epoch": 1.6473214285714286,
      "grad_norm": 0.7478457373623948,
      "learning_rate": 7.483560242049998e-07,
      "loss": 0.2463916301727295,
      "num_input_tokens_seen": 784130784,
      "step": 12915,
      "train_runtime": 168728.7601,
      "train_tokens_per_second": 4647.286
    },
    {
      "epoch": 1.6479591836734695,
      "grad_norm": 0.6040051783701004,
      "learning_rate": 7.457222066242725e-07,
      "loss": 0.24036722183227538,
      "num_input_tokens_seen": 784447456,
      "step": 12920,
      "train_runtime": 168803.5961,
      "train_tokens_per_second": 4647.102
    },
    {
      "epoch": 1.6485969387755102,
      "grad_norm": 0.5724296435449022,
      "learning_rate": 7.430926585084292e-07,
      "loss": 0.24784183502197266,
      "num_input_tokens_seen": 784767912,
      "step": 12925,
      "train_runtime": 168865.3747,
      "train_tokens_per_second": 4647.299
    },
    {
      "epoch": 1.649234693877551,
      "grad_norm": 0.6161602208167504,
      "learning_rate": 7.404673824964048e-07,
      "loss": 0.25671067237854006,
      "num_input_tokens_seen": 785078864,
      "step": 12930,
      "train_runtime": 168920.8491,
      "train_tokens_per_second": 4647.614
    },
    {
      "epoch": 1.649872448979592,
      "grad_norm": 0.6653737470727212,
      "learning_rate": 7.37846381222847e-07,
      "loss": 0.2695626974105835,
      "num_input_tokens_seen": 785393048,
      "step": 12935,
      "train_runtime": 168984.6808,
      "train_tokens_per_second": 4647.717
    },
    {
      "epoch": 1.6505102040816326,
      "grad_norm": 0.6729359819046147,
      "learning_rate": 7.352296573181139e-07,
      "loss": 0.23758177757263182,
      "num_input_tokens_seen": 785691312,
      "step": 12940,
      "train_runtime": 169057.8521,
      "train_tokens_per_second": 4647.47
    },
    {
      "epoch": 1.6511479591836735,
      "grad_norm": 0.7259161159526286,
      "learning_rate": 7.326172134082704e-07,
      "loss": 0.26804699897766116,
      "num_input_tokens_seen": 785992592,
      "step": 12945,
      "train_runtime": 169125.7438,
      "train_tokens_per_second": 4647.386
    },
    {
      "epoch": 1.6517857142857144,
      "grad_norm": 0.6972611463595348,
      "learning_rate": 7.300090521150882e-07,
      "loss": 0.26202845573425293,
      "num_input_tokens_seen": 786295144,
      "step": 12950,
      "train_runtime": 169190.347,
      "train_tokens_per_second": 4647.4
    },
    {
      "epoch": 1.652423469387755,
      "grad_norm": 0.6996257601862602,
      "learning_rate": 7.274051760560364e-07,
      "loss": 0.2594417095184326,
      "num_input_tokens_seen": 786595296,
      "step": 12955,
      "train_runtime": 169251.5681,
      "train_tokens_per_second": 4647.492
    },
    {
      "epoch": 1.6530612244897958,
      "grad_norm": 0.6956483422072044,
      "learning_rate": 7.248055878442861e-07,
      "loss": 0.25502710342407225,
      "num_input_tokens_seen": 786887408,
      "step": 12960,
      "train_runtime": 169315.5845,
      "train_tokens_per_second": 4647.46
    },
    {
      "epoch": 1.6536989795918369,
      "grad_norm": 0.6318095173521857,
      "learning_rate": 7.222102900887102e-07,
      "loss": 0.28072850704193114,
      "num_input_tokens_seen": 787190392,
      "step": 12965,
      "train_runtime": 169378.8465,
      "train_tokens_per_second": 4647.513
    },
    {
      "epoch": 1.6543367346938775,
      "grad_norm": 0.797435690024026,
      "learning_rate": 7.196192853938672e-07,
      "loss": 0.2597139835357666,
      "num_input_tokens_seen": 787490568,
      "step": 12970,
      "train_runtime": 169444.4574,
      "train_tokens_per_second": 4647.485
    },
    {
      "epoch": 1.6549744897959182,
      "grad_norm": 0.6367133962237158,
      "learning_rate": 7.170325763600106e-07,
      "loss": 0.24894609451293945,
      "num_input_tokens_seen": 787785952,
      "step": 12975,
      "train_runtime": 169499.3648,
      "train_tokens_per_second": 4647.722
    },
    {
      "epoch": 1.6556122448979593,
      "grad_norm": 0.6834042418423427,
      "learning_rate": 7.144501655830882e-07,
      "loss": 0.2468048334121704,
      "num_input_tokens_seen": 788076952,
      "step": 12980,
      "train_runtime": 169563.6177,
      "train_tokens_per_second": 4647.677
    },
    {
      "epoch": 1.65625,
      "grad_norm": 0.700017295575168,
      "learning_rate": 7.118720556547259e-07,
      "loss": 0.2554357051849365,
      "num_input_tokens_seen": 788368824,
      "step": 12985,
      "train_runtime": 169620.6824,
      "train_tokens_per_second": 4647.834
    },
    {
      "epoch": 1.6568877551020407,
      "grad_norm": 0.7179057334335764,
      "learning_rate": 7.092982491622374e-07,
      "loss": 0.261881685256958,
      "num_input_tokens_seen": 788675568,
      "step": 12990,
      "train_runtime": 169678.38,
      "train_tokens_per_second": 4648.062
    },
    {
      "epoch": 1.6575255102040818,
      "grad_norm": 0.7314788021055493,
      "learning_rate": 7.067287486886181e-07,
      "loss": 0.24860353469848634,
      "num_input_tokens_seen": 788981344,
      "step": 12995,
      "train_runtime": 169746.6263,
      "train_tokens_per_second": 4647.994
    },
    {
      "epoch": 1.6581632653061225,
      "grad_norm": 0.6219051493099943,
      "learning_rate": 7.04163556812541e-07,
      "loss": 0.2610947132110596,
      "num_input_tokens_seen": 789282208,
      "step": 13000,
      "train_runtime": 169810.3115,
      "train_tokens_per_second": 4648.023
    },
    {
      "epoch": 1.6588010204081631,
      "grad_norm": 0.6735901059791817,
      "learning_rate": 7.016026761083556e-07,
      "loss": 0.2498173713684082,
      "num_input_tokens_seen": 789584376,
      "step": 13005,
      "train_runtime": 169882.9359,
      "train_tokens_per_second": 4647.815
    },
    {
      "epoch": 1.6594387755102042,
      "grad_norm": 0.6114395561094688,
      "learning_rate": 6.990461091460832e-07,
      "loss": 0.25535106658935547,
      "num_input_tokens_seen": 789879496,
      "step": 13010,
      "train_runtime": 169946.2966,
      "train_tokens_per_second": 4647.818
    },
    {
      "epoch": 1.660076530612245,
      "grad_norm": 0.652726195798795,
      "learning_rate": 6.964938584914188e-07,
      "loss": 0.2566126585006714,
      "num_input_tokens_seen": 790183328,
      "step": 13015,
      "train_runtime": 170018.965,
      "train_tokens_per_second": 4647.619
    },
    {
      "epoch": 1.6607142857142856,
      "grad_norm": 0.6508018780529824,
      "learning_rate": 6.939459267057236e-07,
      "loss": 0.2553213119506836,
      "num_input_tokens_seen": 790497336,
      "step": 13020,
      "train_runtime": 170076.9135,
      "train_tokens_per_second": 4647.881
    },
    {
      "epoch": 1.6613520408163265,
      "grad_norm": 0.654588286168501,
      "learning_rate": 6.914023163460248e-07,
      "loss": 0.2686570644378662,
      "num_input_tokens_seen": 790798448,
      "step": 13025,
      "train_runtime": 170145.0643,
      "train_tokens_per_second": 4647.79
    },
    {
      "epoch": 1.6619897959183674,
      "grad_norm": 0.6961943531662362,
      "learning_rate": 6.888630299650134e-07,
      "loss": 0.2502615213394165,
      "num_input_tokens_seen": 791104728,
      "step": 13030,
      "train_runtime": 170207.6932,
      "train_tokens_per_second": 4647.879
    },
    {
      "epoch": 1.662627551020408,
      "grad_norm": 0.6604878791794039,
      "learning_rate": 6.863280701110409e-07,
      "loss": 0.23832058906555176,
      "num_input_tokens_seen": 791402312,
      "step": 13035,
      "train_runtime": 170279.2635,
      "train_tokens_per_second": 4647.673
    },
    {
      "epoch": 1.663265306122449,
      "grad_norm": 0.7042894468943262,
      "learning_rate": 6.837974393281171e-07,
      "loss": 0.24954562187194823,
      "num_input_tokens_seen": 791703896,
      "step": 13040,
      "train_runtime": 170341.38,
      "train_tokens_per_second": 4647.749
    },
    {
      "epoch": 1.6639030612244898,
      "grad_norm": 0.6495592094015468,
      "learning_rate": 6.812711401559036e-07,
      "loss": 0.23029787540435792,
      "num_input_tokens_seen": 791997456,
      "step": 13045,
      "train_runtime": 170404.5112,
      "train_tokens_per_second": 4647.749
    },
    {
      "epoch": 1.6645408163265305,
      "grad_norm": 0.6606908330358849,
      "learning_rate": 6.787491751297215e-07,
      "loss": 0.2461026668548584,
      "num_input_tokens_seen": 792301680,
      "step": 13050,
      "train_runtime": 170468.167,
      "train_tokens_per_second": 4647.798
    },
    {
      "epoch": 1.6651785714285714,
      "grad_norm": 0.69760734388266,
      "learning_rate": 6.762315467805391e-07,
      "loss": 0.28578948974609375,
      "num_input_tokens_seen": 792607080,
      "step": 13055,
      "train_runtime": 170532.8667,
      "train_tokens_per_second": 4647.826
    },
    {
      "epoch": 1.6658163265306123,
      "grad_norm": 0.7601081293086912,
      "learning_rate": 6.737182576349682e-07,
      "loss": 0.246743106842041,
      "num_input_tokens_seen": 792914096,
      "step": 13060,
      "train_runtime": 170606.4979,
      "train_tokens_per_second": 4647.62
    },
    {
      "epoch": 1.666454081632653,
      "grad_norm": 0.6877040137064744,
      "learning_rate": 6.712093102152739e-07,
      "loss": 0.24052057266235352,
      "num_input_tokens_seen": 793216536,
      "step": 13065,
      "train_runtime": 170678.1463,
      "train_tokens_per_second": 4647.441
    },
    {
      "epoch": 1.6670918367346939,
      "grad_norm": 0.5969262966889857,
      "learning_rate": 6.687047070393599e-07,
      "loss": 0.2614285469055176,
      "num_input_tokens_seen": 793522384,
      "step": 13070,
      "train_runtime": 170744.8061,
      "train_tokens_per_second": 4647.417
    },
    {
      "epoch": 1.6677295918367347,
      "grad_norm": 0.6661205590578827,
      "learning_rate": 6.662044506207682e-07,
      "loss": 0.2567417144775391,
      "num_input_tokens_seen": 793828944,
      "step": 13075,
      "train_runtime": 170802.18,
      "train_tokens_per_second": 4647.651
    },
    {
      "epoch": 1.6683673469387754,
      "grad_norm": 0.6185885633000117,
      "learning_rate": 6.637085434686818e-07,
      "loss": 0.253106427192688,
      "num_input_tokens_seen": 794146080,
      "step": 13080,
      "train_runtime": 170857.761,
      "train_tokens_per_second": 4647.995
    },
    {
      "epoch": 1.6690051020408163,
      "grad_norm": 0.7594755059064203,
      "learning_rate": 6.612169880879182e-07,
      "loss": 0.24351756572723388,
      "num_input_tokens_seen": 794452296,
      "step": 13085,
      "train_runtime": 170922.5506,
      "train_tokens_per_second": 4648.025
    },
    {
      "epoch": 1.6696428571428572,
      "grad_norm": 0.8822950418836593,
      "learning_rate": 6.587297869789267e-07,
      "loss": 0.25382318496704104,
      "num_input_tokens_seen": 794756976,
      "step": 13090,
      "train_runtime": 170981.4459,
      "train_tokens_per_second": 4648.206
    },
    {
      "epoch": 1.6702806122448979,
      "grad_norm": 0.6204122292589561,
      "learning_rate": 6.562469426377876e-07,
      "loss": 0.25609397888183594,
      "num_input_tokens_seen": 795069352,
      "step": 13095,
      "train_runtime": 171052.2104,
      "train_tokens_per_second": 4648.109
    },
    {
      "epoch": 1.6709183673469388,
      "grad_norm": 0.6885263526936767,
      "learning_rate": 6.537684575562092e-07,
      "loss": 0.2575140237808228,
      "num_input_tokens_seen": 795365352,
      "step": 13100,
      "train_runtime": 171123.378,
      "train_tokens_per_second": 4647.906
    },
    {
      "epoch": 1.6715561224489797,
      "grad_norm": 0.5939617734955507,
      "learning_rate": 6.512943342215234e-07,
      "loss": 0.26410722732543945,
      "num_input_tokens_seen": 795678608,
      "step": 13105,
      "train_runtime": 171180.4204,
      "train_tokens_per_second": 4648.187
    },
    {
      "epoch": 1.6721938775510203,
      "grad_norm": 0.6700230393287288,
      "learning_rate": 6.48824575116686e-07,
      "loss": 0.2558436870574951,
      "num_input_tokens_seen": 795975296,
      "step": 13110,
      "train_runtime": 171248.6134,
      "train_tokens_per_second": 4648.069
    },
    {
      "epoch": 1.6728316326530612,
      "grad_norm": 0.6511984697457034,
      "learning_rate": 6.463591827202731e-07,
      "loss": 0.2514554262161255,
      "num_input_tokens_seen": 796281592,
      "step": 13115,
      "train_runtime": 171307.2427,
      "train_tokens_per_second": 4648.266
    },
    {
      "epoch": 1.6734693877551021,
      "grad_norm": 0.6649083733321107,
      "learning_rate": 6.43898159506478e-07,
      "loss": 0.2432316780090332,
      "num_input_tokens_seen": 796582856,
      "step": 13120,
      "train_runtime": 171376.2934,
      "train_tokens_per_second": 4648.151
    },
    {
      "epoch": 1.6741071428571428,
      "grad_norm": 0.6294188107652842,
      "learning_rate": 6.414415079451092e-07,
      "loss": 0.24376821517944336,
      "num_input_tokens_seen": 796872776,
      "step": 13125,
      "train_runtime": 171438.8505,
      "train_tokens_per_second": 4648.146
    },
    {
      "epoch": 1.6747448979591837,
      "grad_norm": 0.6789238026386295,
      "learning_rate": 6.389892305015882e-07,
      "loss": 0.2742983341217041,
      "num_input_tokens_seen": 797187792,
      "step": 13130,
      "train_runtime": 171506.428,
      "train_tokens_per_second": 4648.151
    },
    {
      "epoch": 1.6753826530612246,
      "grad_norm": 0.6076933171974049,
      "learning_rate": 6.365413296369466e-07,
      "loss": 0.24300801753997803,
      "num_input_tokens_seen": 797483784,
      "step": 13135,
      "train_runtime": 171572.1649,
      "train_tokens_per_second": 4648.095
    },
    {
      "epoch": 1.6760204081632653,
      "grad_norm": 0.7168667349383481,
      "learning_rate": 6.340978078078253e-07,
      "loss": 0.24329781532287598,
      "num_input_tokens_seen": 797783992,
      "step": 13140,
      "train_runtime": 171638.1145,
      "train_tokens_per_second": 4648.058
    },
    {
      "epoch": 1.6766581632653061,
      "grad_norm": 0.5899359019932424,
      "learning_rate": 6.316586674664654e-07,
      "loss": 0.24906373023986816,
      "num_input_tokens_seen": 798095032,
      "step": 13145,
      "train_runtime": 171710.1951,
      "train_tokens_per_second": 4647.919
    },
    {
      "epoch": 1.677295918367347,
      "grad_norm": 0.6557035081135042,
      "learning_rate": 6.292239110607185e-07,
      "loss": 0.2658334732055664,
      "num_input_tokens_seen": 798396536,
      "step": 13150,
      "train_runtime": 171785.7432,
      "train_tokens_per_second": 4647.63
    },
    {
      "epoch": 1.6779336734693877,
      "grad_norm": 0.6005148875851889,
      "learning_rate": 6.26793541034032e-07,
      "loss": 0.23490290641784667,
      "num_input_tokens_seen": 798675488,
      "step": 13155,
      "train_runtime": 171851.0286,
      "train_tokens_per_second": 4647.487
    },
    {
      "epoch": 1.6785714285714286,
      "grad_norm": 0.6255835844291253,
      "learning_rate": 6.243675598254501e-07,
      "loss": 0.24631378650665284,
      "num_input_tokens_seen": 798979112,
      "step": 13160,
      "train_runtime": 171915.5124,
      "train_tokens_per_second": 4647.51
    },
    {
      "epoch": 1.6792091836734695,
      "grad_norm": 0.6639968923924517,
      "learning_rate": 6.21945969869614e-07,
      "loss": 0.24266324043273926,
      "num_input_tokens_seen": 799286176,
      "step": 13165,
      "train_runtime": 171988.225,
      "train_tokens_per_second": 4647.331
    },
    {
      "epoch": 1.6798469387755102,
      "grad_norm": 0.6913054421720293,
      "learning_rate": 6.195287735967626e-07,
      "loss": 0.22390503883361818,
      "num_input_tokens_seen": 799599776,
      "step": 13170,
      "train_runtime": 172042.4176,
      "train_tokens_per_second": 4647.69
    },
    {
      "epoch": 1.680484693877551,
      "grad_norm": 0.6601408563471931,
      "learning_rate": 6.171159734327164e-07,
      "loss": 0.2543494462966919,
      "num_input_tokens_seen": 799903368,
      "step": 13175,
      "train_runtime": 172111.4657,
      "train_tokens_per_second": 4647.589
    },
    {
      "epoch": 1.681122448979592,
      "grad_norm": 0.7278103517343254,
      "learning_rate": 6.147075717988904e-07,
      "loss": 0.24501893520355225,
      "num_input_tokens_seen": 800207056,
      "step": 13180,
      "train_runtime": 172177.1603,
      "train_tokens_per_second": 4647.58
    },
    {
      "epoch": 1.6817602040816326,
      "grad_norm": 0.6882107332000177,
      "learning_rate": 6.12303571112286e-07,
      "loss": 0.2572981834411621,
      "num_input_tokens_seen": 800511624,
      "step": 13185,
      "train_runtime": 172242.6711,
      "train_tokens_per_second": 4647.58
    },
    {
      "epoch": 1.6823979591836735,
      "grad_norm": 0.6414566805891898,
      "learning_rate": 6.099039737854834e-07,
      "loss": 0.2360074996948242,
      "num_input_tokens_seen": 800811064,
      "step": 13190,
      "train_runtime": 172318.1387,
      "train_tokens_per_second": 4647.282
    },
    {
      "epoch": 1.6830357142857144,
      "grad_norm": 0.6378694853146921,
      "learning_rate": 6.075087822266462e-07,
      "loss": 0.22553706169128418,
      "num_input_tokens_seen": 801112472,
      "step": 13195,
      "train_runtime": 172376.5946,
      "train_tokens_per_second": 4647.455
    },
    {
      "epoch": 1.683673469387755,
      "grad_norm": 0.7353522949445532,
      "learning_rate": 6.051179988395172e-07,
      "loss": 0.24001829624176024,
      "num_input_tokens_seen": 801409320,
      "step": 13200,
      "train_runtime": 172446.868,
      "train_tokens_per_second": 4647.283
    },
    {
      "epoch": 1.6843112244897958,
      "grad_norm": 0.6460275443676332,
      "learning_rate": 6.027316260234146e-07,
      "loss": 0.25508794784545896,
      "num_input_tokens_seen": 801695944,
      "step": 13205,
      "train_runtime": 172510.3045,
      "train_tokens_per_second": 4647.235
    },
    {
      "epoch": 1.6849489795918369,
      "grad_norm": 0.6123069601635089,
      "learning_rate": 6.003496661732294e-07,
      "loss": 0.23457629680633546,
      "num_input_tokens_seen": 802004992,
      "step": 13210,
      "train_runtime": 172572.7687,
      "train_tokens_per_second": 4647.344
    },
    {
      "epoch": 1.6855867346938775,
      "grad_norm": 0.6576269523805083,
      "learning_rate": 5.979721216794248e-07,
      "loss": 0.24449772834777833,
      "num_input_tokens_seen": 802316736,
      "step": 13215,
      "train_runtime": 172632.7628,
      "train_tokens_per_second": 4647.535
    },
    {
      "epoch": 1.6862244897959182,
      "grad_norm": 0.6373869199709481,
      "learning_rate": 5.955989949280339e-07,
      "loss": 0.2484027624130249,
      "num_input_tokens_seen": 802621320,
      "step": 13220,
      "train_runtime": 172703.2604,
      "train_tokens_per_second": 4647.401
    },
    {
      "epoch": 1.6868622448979593,
      "grad_norm": 0.6765967013532039,
      "learning_rate": 5.932302883006546e-07,
      "loss": 0.24606828689575194,
      "num_input_tokens_seen": 802920632,
      "step": 13225,
      "train_runtime": 172767.4249,
      "train_tokens_per_second": 4647.408
    },
    {
      "epoch": 1.6875,
      "grad_norm": 0.6421432588198012,
      "learning_rate": 5.908660041744473e-07,
      "loss": 0.25984389781951905,
      "num_input_tokens_seen": 803215896,
      "step": 13230,
      "train_runtime": 172831.9334,
      "train_tokens_per_second": 4647.381
    },
    {
      "epoch": 1.6881377551020407,
      "grad_norm": 0.6845704620349297,
      "learning_rate": 5.88506144922139e-07,
      "loss": 0.2351299285888672,
      "num_input_tokens_seen": 803516888,
      "step": 13235,
      "train_runtime": 172894.4456,
      "train_tokens_per_second": 4647.442
    },
    {
      "epoch": 1.6887755102040818,
      "grad_norm": 0.7025820115997669,
      "learning_rate": 5.861507129120136e-07,
      "loss": 0.26414904594421384,
      "num_input_tokens_seen": 803810240,
      "step": 13240,
      "train_runtime": 172964.0166,
      "train_tokens_per_second": 4647.269
    },
    {
      "epoch": 1.6894132653061225,
      "grad_norm": 0.7472631460747936,
      "learning_rate": 5.83799710507909e-07,
      "loss": 0.25545492172241213,
      "num_input_tokens_seen": 804093920,
      "step": 13245,
      "train_runtime": 173031.3847,
      "train_tokens_per_second": 4647.099
    },
    {
      "epoch": 1.6900510204081631,
      "grad_norm": 0.7679617662934242,
      "learning_rate": 5.814531400692208e-07,
      "loss": 0.26322863101959226,
      "num_input_tokens_seen": 804396312,
      "step": 13250,
      "train_runtime": 173100.8754,
      "train_tokens_per_second": 4646.98
    },
    {
      "epoch": 1.6906887755102042,
      "grad_norm": 0.6114569152138319,
      "learning_rate": 5.791110039508991e-07,
      "loss": 0.24986505508422852,
      "num_input_tokens_seen": 804713896,
      "step": 13255,
      "train_runtime": 173153.1822,
      "train_tokens_per_second": 4647.41
    },
    {
      "epoch": 1.691326530612245,
      "grad_norm": 0.686443915314153,
      "learning_rate": 5.767733045034384e-07,
      "loss": 0.2752215385437012,
      "num_input_tokens_seen": 805030240,
      "step": 13260,
      "train_runtime": 173210.2933,
      "train_tokens_per_second": 4647.704
    },
    {
      "epoch": 1.6919642857142856,
      "grad_norm": 0.6551831629345426,
      "learning_rate": 5.744400440728826e-07,
      "loss": 0.25286269187927246,
      "num_input_tokens_seen": 805319600,
      "step": 13265,
      "train_runtime": 173287.5594,
      "train_tokens_per_second": 4647.302
    },
    {
      "epoch": 1.6926020408163265,
      "grad_norm": 0.7862394352964676,
      "learning_rate": 5.721112250008254e-07,
      "loss": 0.24343976974487305,
      "num_input_tokens_seen": 805599816,
      "step": 13270,
      "train_runtime": 173352.1217,
      "train_tokens_per_second": 4647.188
    },
    {
      "epoch": 1.6932397959183674,
      "grad_norm": 0.6541326760795353,
      "learning_rate": 5.697868496243958e-07,
      "loss": 0.24480876922607422,
      "num_input_tokens_seen": 805887848,
      "step": 13275,
      "train_runtime": 173419.0605,
      "train_tokens_per_second": 4647.055
    },
    {
      "epoch": 1.693877551020408,
      "grad_norm": 0.6778576794479275,
      "learning_rate": 5.674669202762684e-07,
      "loss": 0.25247907638549805,
      "num_input_tokens_seen": 806196384,
      "step": 13280,
      "train_runtime": 173492.1644,
      "train_tokens_per_second": 4646.875
    },
    {
      "epoch": 1.694515306122449,
      "grad_norm": 0.6439818130367023,
      "learning_rate": 5.651514392846535e-07,
      "loss": 0.2437495231628418,
      "num_input_tokens_seen": 806500696,
      "step": 13285,
      "train_runtime": 173561.0655,
      "train_tokens_per_second": 4646.784
    },
    {
      "epoch": 1.6951530612244898,
      "grad_norm": 0.6706718045627558,
      "learning_rate": 5.628404089732986e-07,
      "loss": 0.2544881343841553,
      "num_input_tokens_seen": 806810160,
      "step": 13290,
      "train_runtime": 173626.3889,
      "train_tokens_per_second": 4646.818
    },
    {
      "epoch": 1.6957908163265305,
      "grad_norm": 0.7304220260099141,
      "learning_rate": 5.605338316614839e-07,
      "loss": 0.25745701789855957,
      "num_input_tokens_seen": 807112440,
      "step": 13295,
      "train_runtime": 173694.5112,
      "train_tokens_per_second": 4646.735
    },
    {
      "epoch": 1.6964285714285714,
      "grad_norm": 0.6162790075864725,
      "learning_rate": 5.582317096640205e-07,
      "loss": 0.2437589645385742,
      "num_input_tokens_seen": 807413184,
      "step": 13300,
      "train_runtime": 173762.4513,
      "train_tokens_per_second": 4646.649
    },
    {
      "epoch": 1.6970663265306123,
      "grad_norm": 0.6547404051230347,
      "learning_rate": 5.559340452912488e-07,
      "loss": 0.24849872589111327,
      "num_input_tokens_seen": 807731504,
      "step": 13305,
      "train_runtime": 173814.3615,
      "train_tokens_per_second": 4647.093
    },
    {
      "epoch": 1.697704081632653,
      "grad_norm": 0.6101879213333442,
      "learning_rate": 5.536408408490357e-07,
      "loss": 0.23475873470306396,
      "num_input_tokens_seen": 808032920,
      "step": 13310,
      "train_runtime": 173878.1385,
      "train_tokens_per_second": 4647.122
    },
    {
      "epoch": 1.6983418367346939,
      "grad_norm": 0.6805653048642977,
      "learning_rate": 5.513520986387721e-07,
      "loss": 0.2626044511795044,
      "num_input_tokens_seen": 808338432,
      "step": 13315,
      "train_runtime": 173946.2483,
      "train_tokens_per_second": 4647.059
    },
    {
      "epoch": 1.6989795918367347,
      "grad_norm": 0.6795411098661497,
      "learning_rate": 5.490678209573702e-07,
      "loss": 0.2594637632369995,
      "num_input_tokens_seen": 808650944,
      "step": 13320,
      "train_runtime": 174006.1976,
      "train_tokens_per_second": 4647.254
    },
    {
      "epoch": 1.6996173469387754,
      "grad_norm": 0.7221133636236472,
      "learning_rate": 5.467880100972633e-07,
      "loss": 0.24765939712524415,
      "num_input_tokens_seen": 808959368,
      "step": 13325,
      "train_runtime": 174071.1629,
      "train_tokens_per_second": 4647.291
    },
    {
      "epoch": 1.7002551020408163,
      "grad_norm": 0.601381599602597,
      "learning_rate": 5.445126683463991e-07,
      "loss": 0.24539852142333984,
      "num_input_tokens_seen": 809263464,
      "step": 13330,
      "train_runtime": 174145.0622,
      "train_tokens_per_second": 4647.065
    },
    {
      "epoch": 1.7008928571428572,
      "grad_norm": 0.7462785409425798,
      "learning_rate": 5.422417979882416e-07,
      "loss": 0.24530429840087892,
      "num_input_tokens_seen": 809557232,
      "step": 13335,
      "train_runtime": 174219.7268,
      "train_tokens_per_second": 4646.76
    },
    {
      "epoch": 1.7015306122448979,
      "grad_norm": 0.6091027212542391,
      "learning_rate": 5.399754013017711e-07,
      "loss": 0.26610822677612306,
      "num_input_tokens_seen": 809849272,
      "step": 13340,
      "train_runtime": 174286.9811,
      "train_tokens_per_second": 4646.642
    },
    {
      "epoch": 1.7021683673469388,
      "grad_norm": 0.6836631433776763,
      "learning_rate": 5.377134805614714e-07,
      "loss": 0.2476370096206665,
      "num_input_tokens_seen": 810150448,
      "step": 13345,
      "train_runtime": 174343.4431,
      "train_tokens_per_second": 4646.865
    },
    {
      "epoch": 1.7028061224489797,
      "grad_norm": 0.656118211163798,
      "learning_rate": 5.354560380373391e-07,
      "loss": 0.23286876678466797,
      "num_input_tokens_seen": 810450680,
      "step": 13350,
      "train_runtime": 174406.4934,
      "train_tokens_per_second": 4646.907
    },
    {
      "epoch": 1.7034438775510203,
      "grad_norm": 0.6095295851422854,
      "learning_rate": 5.33203075994877e-07,
      "loss": 0.26010301113128664,
      "num_input_tokens_seen": 810750576,
      "step": 13355,
      "train_runtime": 174476.514,
      "train_tokens_per_second": 4646.761
    },
    {
      "epoch": 1.7040816326530612,
      "grad_norm": 0.6538707511126961,
      "learning_rate": 5.30954596695088e-07,
      "loss": 0.2614151477813721,
      "num_input_tokens_seen": 811061000,
      "step": 13360,
      "train_runtime": 174534.244,
      "train_tokens_per_second": 4647.002
    },
    {
      "epoch": 1.7047193877551021,
      "grad_norm": 0.6992112135245512,
      "learning_rate": 5.287106023944782e-07,
      "loss": 0.2446915864944458,
      "num_input_tokens_seen": 811364840,
      "step": 13365,
      "train_runtime": 174601.4086,
      "train_tokens_per_second": 4646.955
    },
    {
      "epoch": 1.7053571428571428,
      "grad_norm": 0.6749212694990641,
      "learning_rate": 5.264710953450536e-07,
      "loss": 0.2663325309753418,
      "num_input_tokens_seen": 811669248,
      "step": 13370,
      "train_runtime": 174670.908,
      "train_tokens_per_second": 4646.849
    },
    {
      "epoch": 1.7059948979591837,
      "grad_norm": 0.6756642574882481,
      "learning_rate": 5.242360777943151e-07,
      "loss": 0.26160728931427,
      "num_input_tokens_seen": 811973912,
      "step": 13375,
      "train_runtime": 174740.3209,
      "train_tokens_per_second": 4646.746
    },
    {
      "epoch": 1.7066326530612246,
      "grad_norm": 0.7266428847277792,
      "learning_rate": 5.220055519852596e-07,
      "loss": 0.2614679574966431,
      "num_input_tokens_seen": 812269376,
      "step": 13380,
      "train_runtime": 174812.2899,
      "train_tokens_per_second": 4646.523
    },
    {
      "epoch": 1.7072704081632653,
      "grad_norm": 0.6409602778513998,
      "learning_rate": 5.197795201563744e-07,
      "loss": 0.26145367622375487,
      "num_input_tokens_seen": 812574120,
      "step": 13385,
      "train_runtime": 174865.0983,
      "train_tokens_per_second": 4646.863
    },
    {
      "epoch": 1.7079081632653061,
      "grad_norm": 0.6362314502165928,
      "learning_rate": 5.17557984541639e-07,
      "loss": 0.2672783851623535,
      "num_input_tokens_seen": 812875616,
      "step": 13390,
      "train_runtime": 174921.8089,
      "train_tokens_per_second": 4647.08
    },
    {
      "epoch": 1.708545918367347,
      "grad_norm": 0.6602153513659031,
      "learning_rate": 5.153409473705196e-07,
      "loss": 0.24822840690612794,
      "num_input_tokens_seen": 813175688,
      "step": 13395,
      "train_runtime": 174987.0558,
      "train_tokens_per_second": 4647.062
    },
    {
      "epoch": 1.7091836734693877,
      "grad_norm": 0.6824710805889549,
      "learning_rate": 5.131284108679669e-07,
      "loss": 0.25509934425354003,
      "num_input_tokens_seen": 813475616,
      "step": 13400,
      "train_runtime": 175051.6624,
      "train_tokens_per_second": 4647.06
    },
    {
      "epoch": 1.7098214285714286,
      "grad_norm": 0.7214556204881735,
      "learning_rate": 5.109203772544164e-07,
      "loss": 0.25195741653442383,
      "num_input_tokens_seen": 813776688,
      "step": 13405,
      "train_runtime": 175111.7734,
      "train_tokens_per_second": 4647.184
    },
    {
      "epoch": 1.7104591836734695,
      "grad_norm": 0.6530731194100159,
      "learning_rate": 5.087168487457838e-07,
      "loss": 0.2547295570373535,
      "num_input_tokens_seen": 814086584,
      "step": 13410,
      "train_runtime": 175180.2361,
      "train_tokens_per_second": 4647.137
    },
    {
      "epoch": 1.7110969387755102,
      "grad_norm": 0.6734735506154492,
      "learning_rate": 5.065178275534649e-07,
      "loss": 0.24732956886291504,
      "num_input_tokens_seen": 814396000,
      "step": 13415,
      "train_runtime": 175251.0201,
      "train_tokens_per_second": 4647.026
    },
    {
      "epoch": 1.711734693877551,
      "grad_norm": 0.6588060391082898,
      "learning_rate": 5.043233158843286e-07,
      "loss": 0.2538780212402344,
      "num_input_tokens_seen": 814709408,
      "step": 13420,
      "train_runtime": 175321.622,
      "train_tokens_per_second": 4646.942
    },
    {
      "epoch": 1.712372448979592,
      "grad_norm": 0.6253785817459473,
      "learning_rate": 5.021333159407232e-07,
      "loss": 0.258031439781189,
      "num_input_tokens_seen": 815013680,
      "step": 13425,
      "train_runtime": 175388.9032,
      "train_tokens_per_second": 4646.894
    },
    {
      "epoch": 1.7130102040816326,
      "grad_norm": 0.6223563531524244,
      "learning_rate": 4.999478299204663e-07,
      "loss": 0.23726649284362794,
      "num_input_tokens_seen": 815325336,
      "step": 13430,
      "train_runtime": 175442.4964,
      "train_tokens_per_second": 4647.251
    },
    {
      "epoch": 1.7136479591836735,
      "grad_norm": 0.6174363850558218,
      "learning_rate": 4.977668600168428e-07,
      "loss": 0.23950729370117188,
      "num_input_tokens_seen": 815613392,
      "step": 13435,
      "train_runtime": 175518.5989,
      "train_tokens_per_second": 4646.877
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 0.5987978775881054,
      "learning_rate": 4.95590408418613e-07,
      "loss": 0.2627774953842163,
      "num_input_tokens_seen": 815921320,
      "step": 13440,
      "train_runtime": 175584.2231,
      "train_tokens_per_second": 4646.894
    },
    {
      "epoch": 1.714923469387755,
      "grad_norm": 0.6625060930478297,
      "learning_rate": 4.934184773099943e-07,
      "loss": 0.27510771751403806,
      "num_input_tokens_seen": 816239416,
      "step": 13445,
      "train_runtime": 175658.2589,
      "train_tokens_per_second": 4646.747
    },
    {
      "epoch": 1.7155612244897958,
      "grad_norm": 0.5766109387544881,
      "learning_rate": 4.912510688706729e-07,
      "loss": 0.2529017448425293,
      "num_input_tokens_seen": 816543240,
      "step": 13450,
      "train_runtime": 175723.552,
      "train_tokens_per_second": 4646.749
    },
    {
      "epoch": 1.7161989795918369,
      "grad_norm": 0.6261581832501326,
      "learning_rate": 4.89088185275794e-07,
      "loss": 0.2732319116592407,
      "num_input_tokens_seen": 816852968,
      "step": 13455,
      "train_runtime": 175785.6898,
      "train_tokens_per_second": 4646.868
    },
    {
      "epoch": 1.7168367346938775,
      "grad_norm": 0.6495123546009153,
      "learning_rate": 4.869298286959629e-07,
      "loss": 0.2589241981506348,
      "num_input_tokens_seen": 817160528,
      "step": 13460,
      "train_runtime": 175854.4415,
      "train_tokens_per_second": 4646.801
    },
    {
      "epoch": 1.7174744897959182,
      "grad_norm": 0.6897678748516881,
      "learning_rate": 4.847760012972402e-07,
      "loss": 0.24462056159973145,
      "num_input_tokens_seen": 817470512,
      "step": 13465,
      "train_runtime": 175920.5458,
      "train_tokens_per_second": 4646.817
    },
    {
      "epoch": 1.7181122448979593,
      "grad_norm": 0.5820599825977025,
      "learning_rate": 4.826267052411432e-07,
      "loss": 0.2470067024230957,
      "num_input_tokens_seen": 817764792,
      "step": 13470,
      "train_runtime": 175984.2991,
      "train_tokens_per_second": 4646.805
    },
    {
      "epoch": 1.71875,
      "grad_norm": 0.6199065448408366,
      "learning_rate": 4.804819426846402e-07,
      "loss": 0.2478943109512329,
      "num_input_tokens_seen": 818067224,
      "step": 13475,
      "train_runtime": 176052.2544,
      "train_tokens_per_second": 4646.73
    },
    {
      "epoch": 1.7193877551020407,
      "grad_norm": 0.7305909389713294,
      "learning_rate": 4.783417157801496e-07,
      "loss": 0.2814014911651611,
      "num_input_tokens_seen": 818371368,
      "step": 13480,
      "train_runtime": 176116.1352,
      "train_tokens_per_second": 4646.771
    },
    {
      "epoch": 1.7200255102040818,
      "grad_norm": 0.6601304986648187,
      "learning_rate": 4.7620602667553973e-07,
      "loss": 0.24475960731506347,
      "num_input_tokens_seen": 818675136,
      "step": 13485,
      "train_runtime": 176182.0275,
      "train_tokens_per_second": 4646.757
    },
    {
      "epoch": 1.7206632653061225,
      "grad_norm": 0.6585753514402448,
      "learning_rate": 4.7407487751412297e-07,
      "loss": 0.2473832368850708,
      "num_input_tokens_seen": 818969848,
      "step": 13490,
      "train_runtime": 176245.8929,
      "train_tokens_per_second": 4646.746
    },
    {
      "epoch": 1.7213010204081631,
      "grad_norm": 0.6785347715300538,
      "learning_rate": 4.719482704346567e-07,
      "loss": 0.25057687759399416,
      "num_input_tokens_seen": 819271288,
      "step": 13495,
      "train_runtime": 176315.113,
      "train_tokens_per_second": 4646.631
    },
    {
      "epoch": 1.7219387755102042,
      "grad_norm": 0.6355962626394969,
      "learning_rate": 4.6982620757133947e-07,
      "loss": 0.2580732345581055,
      "num_input_tokens_seen": 819576904,
      "step": 13500,
      "train_runtime": 176378.7128,
      "train_tokens_per_second": 4646.688
    },
    {
      "epoch": 1.722576530612245,
      "grad_norm": 0.7766392615848923,
      "learning_rate": 4.677086910538092e-07,
      "loss": 0.250787091255188,
      "num_input_tokens_seen": 819888528,
      "step": 13505,
      "train_runtime": 176444.6693,
      "train_tokens_per_second": 4646.717
    },
    {
      "epoch": 1.7232142857142856,
      "grad_norm": 0.6335216455614607,
      "learning_rate": 4.6559572300714227e-07,
      "loss": 0.24778370857238768,
      "num_input_tokens_seen": 820188984,
      "step": 13510,
      "train_runtime": 176509.2452,
      "train_tokens_per_second": 4646.72
    },
    {
      "epoch": 1.7238520408163265,
      "grad_norm": 0.7380125184980411,
      "learning_rate": 4.6348730555185017e-07,
      "loss": 0.26547608375549314,
      "num_input_tokens_seen": 820484616,
      "step": 13515,
      "train_runtime": 176568.408,
      "train_tokens_per_second": 4646.837
    },
    {
      "epoch": 1.7244897959183674,
      "grad_norm": 0.6265763025101634,
      "learning_rate": 4.6138344080387477e-07,
      "loss": 0.24309449195861815,
      "num_input_tokens_seen": 820793416,
      "step": 13520,
      "train_runtime": 176634.2089,
      "train_tokens_per_second": 4646.854
    },
    {
      "epoch": 1.725127551020408,
      "grad_norm": 0.60170752157238,
      "learning_rate": 4.5928413087459325e-07,
      "loss": 0.2588818073272705,
      "num_input_tokens_seen": 821094712,
      "step": 13525,
      "train_runtime": 176707.2394,
      "train_tokens_per_second": 4646.639
    },
    {
      "epoch": 1.725765306122449,
      "grad_norm": 0.7034269884118203,
      "learning_rate": 4.571893778708103e-07,
      "loss": 0.23107919692993165,
      "num_input_tokens_seen": 821399712,
      "step": 13530,
      "train_runtime": 176771.1188,
      "train_tokens_per_second": 4646.685
    },
    {
      "epoch": 1.7264030612244898,
      "grad_norm": 0.6511470149372105,
      "learning_rate": 4.550991838947555e-07,
      "loss": 0.265388560295105,
      "num_input_tokens_seen": 821707328,
      "step": 13535,
      "train_runtime": 176833.4494,
      "train_tokens_per_second": 4646.787
    },
    {
      "epoch": 1.7270408163265305,
      "grad_norm": 0.6710553403884467,
      "learning_rate": 4.5301355104408364e-07,
      "loss": 0.2338574171066284,
      "num_input_tokens_seen": 822009192,
      "step": 13540,
      "train_runtime": 176909.3498,
      "train_tokens_per_second": 4646.499
    },
    {
      "epoch": 1.7276785714285714,
      "grad_norm": 0.6798197272438135,
      "learning_rate": 4.509324814118754e-07,
      "loss": 0.25485005378723147,
      "num_input_tokens_seen": 822306680,
      "step": 13545,
      "train_runtime": 176964.9246,
      "train_tokens_per_second": 4646.721
    },
    {
      "epoch": 1.7283163265306123,
      "grad_norm": 0.7753455593426912,
      "learning_rate": 4.48855977086628e-07,
      "loss": 0.25541257858276367,
      "num_input_tokens_seen": 822605176,
      "step": 13550,
      "train_runtime": 177025.1618,
      "train_tokens_per_second": 4646.826
    },
    {
      "epoch": 1.728954081632653,
      "grad_norm": 0.7149215180122932,
      "learning_rate": 4.4678404015225774e-07,
      "loss": 0.24527978897094727,
      "num_input_tokens_seen": 822907840,
      "step": 13555,
      "train_runtime": 177099.1207,
      "train_tokens_per_second": 4646.595
    },
    {
      "epoch": 1.7295918367346939,
      "grad_norm": 0.6971028776253299,
      "learning_rate": 4.447166726881008e-07,
      "loss": 0.26441819667816163,
      "num_input_tokens_seen": 823217208,
      "step": 13560,
      "train_runtime": 177175.0466,
      "train_tokens_per_second": 4646.35
    },
    {
      "epoch": 1.7302295918367347,
      "grad_norm": 0.6703504694700919,
      "learning_rate": 4.4265387676890203e-07,
      "loss": 0.24485955238342286,
      "num_input_tokens_seen": 823522816,
      "step": 13565,
      "train_runtime": 177237.6068,
      "train_tokens_per_second": 4646.434
    },
    {
      "epoch": 1.7308673469387754,
      "grad_norm": 0.6428849614154608,
      "learning_rate": 4.4059565446482264e-07,
      "loss": 0.23918466567993163,
      "num_input_tokens_seen": 823824144,
      "step": 13570,
      "train_runtime": 177302.6242,
      "train_tokens_per_second": 4646.43
    },
    {
      "epoch": 1.7315051020408163,
      "grad_norm": 0.6392656161875652,
      "learning_rate": 4.385420078414321e-07,
      "loss": 0.2566037893295288,
      "num_input_tokens_seen": 824116440,
      "step": 13575,
      "train_runtime": 177365.3695,
      "train_tokens_per_second": 4646.434
    },
    {
      "epoch": 1.7321428571428572,
      "grad_norm": 0.6965722442977956,
      "learning_rate": 4.364929389597078e-07,
      "loss": 0.23905861377716064,
      "num_input_tokens_seen": 824420288,
      "step": 13580,
      "train_runtime": 177429.1114,
      "train_tokens_per_second": 4646.477
    },
    {
      "epoch": 1.7327806122448979,
      "grad_norm": 0.6787328147362667,
      "learning_rate": 4.344484498760343e-07,
      "loss": 0.25505044460296633,
      "num_input_tokens_seen": 824731064,
      "step": 13585,
      "train_runtime": 177489.3243,
      "train_tokens_per_second": 4646.652
    },
    {
      "epoch": 1.7334183673469388,
      "grad_norm": 0.6430044392578245,
      "learning_rate": 4.324085426421981e-07,
      "loss": 0.27058048248291017,
      "num_input_tokens_seen": 825043776,
      "step": 13590,
      "train_runtime": 177556.7657,
      "train_tokens_per_second": 4646.648
    },
    {
      "epoch": 1.7340561224489797,
      "grad_norm": 0.664261873904836,
      "learning_rate": 4.3037321930539e-07,
      "loss": 0.24852211475372316,
      "num_input_tokens_seen": 825343224,
      "step": 13595,
      "train_runtime": 177625.3829,
      "train_tokens_per_second": 4646.539
    },
    {
      "epoch": 1.7346938775510203,
      "grad_norm": 0.6328411981625071,
      "learning_rate": 4.283424819081988e-07,
      "loss": 0.24766051769256592,
      "num_input_tokens_seen": 825642352,
      "step": 13600,
      "train_runtime": 177694.6897,
      "train_tokens_per_second": 4646.41
    },
    {
      "epoch": 1.7353316326530612,
      "grad_norm": 0.6192323219119893,
      "learning_rate": 4.2631633248860937e-07,
      "loss": 0.24031505584716797,
      "num_input_tokens_seen": 825946616,
      "step": 13605,
      "train_runtime": 177760.2681,
      "train_tokens_per_second": 4646.407
    },
    {
      "epoch": 1.7359693877551021,
      "grad_norm": 0.6748273193909125,
      "learning_rate": 4.2429477308000676e-07,
      "loss": 0.24826493263244628,
      "num_input_tokens_seen": 826243296,
      "step": 13610,
      "train_runtime": 177827.8368,
      "train_tokens_per_second": 4646.31
    },
    {
      "epoch": 1.7366071428571428,
      "grad_norm": 0.7136297364370146,
      "learning_rate": 4.222778057111665e-07,
      "loss": 0.26136996746063235,
      "num_input_tokens_seen": 826537744,
      "step": 13615,
      "train_runtime": 177892.9233,
      "train_tokens_per_second": 4646.265
    },
    {
      "epoch": 1.7372448979591837,
      "grad_norm": 0.5820350102037467,
      "learning_rate": 4.202654324062544e-07,
      "loss": 0.2185586452484131,
      "num_input_tokens_seen": 826831632,
      "step": 13620,
      "train_runtime": 177947.7145,
      "train_tokens_per_second": 4646.486
    },
    {
      "epoch": 1.7378826530612246,
      "grad_norm": 0.6366344431603493,
      "learning_rate": 4.182576551848283e-07,
      "loss": 0.24439539909362792,
      "num_input_tokens_seen": 827132840,
      "step": 13625,
      "train_runtime": 178009.3383,
      "train_tokens_per_second": 4646.57
    },
    {
      "epoch": 1.7385204081632653,
      "grad_norm": 0.6833577349409844,
      "learning_rate": 4.1625447606183533e-07,
      "loss": 0.23791561126708985,
      "num_input_tokens_seen": 827439840,
      "step": 13630,
      "train_runtime": 178079.292,
      "train_tokens_per_second": 4646.469
    },
    {
      "epoch": 1.7391581632653061,
      "grad_norm": 0.6669176618264123,
      "learning_rate": 4.142558970476024e-07,
      "loss": 0.2288142204284668,
      "num_input_tokens_seen": 827753720,
      "step": 13635,
      "train_runtime": 178139.9289,
      "train_tokens_per_second": 4646.649
    },
    {
      "epoch": 1.739795918367347,
      "grad_norm": 0.6408512920375541,
      "learning_rate": 4.1226192014784393e-07,
      "loss": 0.2142507553100586,
      "num_input_tokens_seen": 828054520,
      "step": 13640,
      "train_runtime": 178199.167,
      "train_tokens_per_second": 4646.792
    },
    {
      "epoch": 1.7404336734693877,
      "grad_norm": 0.7204366097448385,
      "learning_rate": 4.102725473636571e-07,
      "loss": 0.24937853813171387,
      "num_input_tokens_seen": 828362584,
      "step": 13645,
      "train_runtime": 178260.0878,
      "train_tokens_per_second": 4646.932
    },
    {
      "epoch": 1.7410714285714286,
      "grad_norm": 0.7351000011253044,
      "learning_rate": 4.0828778069151363e-07,
      "loss": 0.27016596794128417,
      "num_input_tokens_seen": 828668024,
      "step": 13650,
      "train_runtime": 178338.4885,
      "train_tokens_per_second": 4646.602
    },
    {
      "epoch": 1.7417091836734695,
      "grad_norm": 0.6230140317665974,
      "learning_rate": 4.063076221232676e-07,
      "loss": 0.24024586677551268,
      "num_input_tokens_seen": 828968552,
      "step": 13655,
      "train_runtime": 178404.0612,
      "train_tokens_per_second": 4646.579
    },
    {
      "epoch": 1.7423469387755102,
      "grad_norm": 0.7086958836856496,
      "learning_rate": 4.043320736461448e-07,
      "loss": 0.27800915241241453,
      "num_input_tokens_seen": 829280800,
      "step": 13660,
      "train_runtime": 178464.4523,
      "train_tokens_per_second": 4646.756
    },
    {
      "epoch": 1.742984693877551,
      "grad_norm": 0.6161159829434065,
      "learning_rate": 4.0236113724274716e-07,
      "loss": 0.2446291923522949,
      "num_input_tokens_seen": 829588144,
      "step": 13665,
      "train_runtime": 178528.3859,
      "train_tokens_per_second": 4646.814
    },
    {
      "epoch": 1.743622448979592,
      "grad_norm": 0.6257870799616637,
      "learning_rate": 4.0039481489104636e-07,
      "loss": 0.2662153959274292,
      "num_input_tokens_seen": 829898896,
      "step": 13670,
      "train_runtime": 178590.2827,
      "train_tokens_per_second": 4646.943
    },
    {
      "epoch": 1.7442602040816326,
      "grad_norm": 0.6325520658994559,
      "learning_rate": 3.98433108564385e-07,
      "loss": 0.24277727603912352,
      "num_input_tokens_seen": 830197328,
      "step": 13675,
      "train_runtime": 178653.0888,
      "train_tokens_per_second": 4646.98
    },
    {
      "epoch": 1.7448979591836735,
      "grad_norm": 0.6674103850787549,
      "learning_rate": 3.964760202314716e-07,
      "loss": 0.24192190170288086,
      "num_input_tokens_seen": 830513592,
      "step": 13680,
      "train_runtime": 178708.4145,
      "train_tokens_per_second": 4647.311
    },
    {
      "epoch": 1.7455357142857144,
      "grad_norm": 0.663067930734789,
      "learning_rate": 3.9452355185638226e-07,
      "loss": 0.25597119331359863,
      "num_input_tokens_seen": 830815680,
      "step": 13685,
      "train_runtime": 178781.0334,
      "train_tokens_per_second": 4647.113
    },
    {
      "epoch": 1.746173469387755,
      "grad_norm": 0.7594095249949443,
      "learning_rate": 3.9257570539855237e-07,
      "loss": 0.2551027774810791,
      "num_input_tokens_seen": 831116968,
      "step": 13690,
      "train_runtime": 178833.74,
      "train_tokens_per_second": 4647.428
    },
    {
      "epoch": 1.7468112244897958,
      "grad_norm": 0.6802275542106344,
      "learning_rate": 3.9063248281278555e-07,
      "loss": 0.24135251045227052,
      "num_input_tokens_seen": 831412080,
      "step": 13695,
      "train_runtime": 178905.7841,
      "train_tokens_per_second": 4647.206
    },
    {
      "epoch": 1.7474489795918369,
      "grad_norm": 0.6075509637244532,
      "learning_rate": 3.8869388604924007e-07,
      "loss": 0.2383573532104492,
      "num_input_tokens_seen": 831711544,
      "step": 13700,
      "train_runtime": 178972.8335,
      "train_tokens_per_second": 4647.138
    },
    {
      "epoch": 1.7480867346938775,
      "grad_norm": 0.6119671525051149,
      "learning_rate": 3.8675991705343197e-07,
      "loss": 0.2385228157043457,
      "num_input_tokens_seen": 832007984,
      "step": 13705,
      "train_runtime": 179036.3298,
      "train_tokens_per_second": 4647.146
    },
    {
      "epoch": 1.7487244897959182,
      "grad_norm": 0.6284455385677916,
      "learning_rate": 3.848305777662342e-07,
      "loss": 0.2456441879272461,
      "num_input_tokens_seen": 832303496,
      "step": 13710,
      "train_runtime": 179102.8964,
      "train_tokens_per_second": 4647.069
    },
    {
      "epoch": 1.7493622448979593,
      "grad_norm": 0.5998925230170503,
      "learning_rate": 3.829058701238764e-07,
      "loss": 0.2562390327453613,
      "num_input_tokens_seen": 832603704,
      "step": 13715,
      "train_runtime": 179165.3503,
      "train_tokens_per_second": 4647.125
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.6185610623610748,
      "learning_rate": 3.8098579605793385e-07,
      "loss": 0.25684871673583987,
      "num_input_tokens_seen": 832912352,
      "step": 13720,
      "train_runtime": 179222.6214,
      "train_tokens_per_second": 4647.362
    },
    {
      "epoch": 1.7506377551020407,
      "grad_norm": 0.6887984162820846,
      "learning_rate": 3.790703574953353e-07,
      "loss": 0.25089006423950194,
      "num_input_tokens_seen": 833230920,
      "step": 13725,
      "train_runtime": 179288.8269,
      "train_tokens_per_second": 4647.422
    },
    {
      "epoch": 1.7512755102040818,
      "grad_norm": 0.6352271383759429,
      "learning_rate": 3.771595563583602e-07,
      "loss": 0.22961339950561524,
      "num_input_tokens_seen": 833535792,
      "step": 13730,
      "train_runtime": 179367.2834,
      "train_tokens_per_second": 4647.089
    },
    {
      "epoch": 1.7519132653061225,
      "grad_norm": 0.760425539674986,
      "learning_rate": 3.752533945646275e-07,
      "loss": 0.27449326515197753,
      "num_input_tokens_seen": 833842504,
      "step": 13735,
      "train_runtime": 179426.6537,
      "train_tokens_per_second": 4647.261
    },
    {
      "epoch": 1.7525510204081631,
      "grad_norm": 0.7291184134987161,
      "learning_rate": 3.733518740271047e-07,
      "loss": 0.24901988506317138,
      "num_input_tokens_seen": 834138640,
      "step": 13740,
      "train_runtime": 179490.3874,
      "train_tokens_per_second": 4647.261
    },
    {
      "epoch": 1.7531887755102042,
      "grad_norm": 0.660442184020874,
      "learning_rate": 3.7145499665410147e-07,
      "loss": 0.24527840614318847,
      "num_input_tokens_seen": 834425800,
      "step": 13745,
      "train_runtime": 179558.736,
      "train_tokens_per_second": 4647.091
    },
    {
      "epoch": 1.753826530612245,
      "grad_norm": 0.6594943264407748,
      "learning_rate": 3.695627643492655e-07,
      "loss": 0.2338253974914551,
      "num_input_tokens_seen": 834725488,
      "step": 13750,
      "train_runtime": 179623.282,
      "train_tokens_per_second": 4647.09
    },
    {
      "epoch": 1.7544642857142856,
      "grad_norm": 0.6688459721750367,
      "learning_rate": 3.6767517901158455e-07,
      "loss": 0.236647891998291,
      "num_input_tokens_seen": 835027360,
      "step": 13755,
      "train_runtime": 179682.3795,
      "train_tokens_per_second": 4647.241
    },
    {
      "epoch": 1.7551020408163265,
      "grad_norm": 0.6213010699853478,
      "learning_rate": 3.6579224253538213e-07,
      "loss": 0.2543475866317749,
      "num_input_tokens_seen": 835324880,
      "step": 13760,
      "train_runtime": 179744.6966,
      "train_tokens_per_second": 4647.285
    },
    {
      "epoch": 1.7557397959183674,
      "grad_norm": 0.7126939250044705,
      "learning_rate": 3.639139568103173e-07,
      "loss": 0.2533092021942139,
      "num_input_tokens_seen": 835637264,
      "step": 13765,
      "train_runtime": 179807.1156,
      "train_tokens_per_second": 4647.409
    },
    {
      "epoch": 1.756377551020408,
      "grad_norm": 0.7787011727512609,
      "learning_rate": 3.620403237213799e-07,
      "loss": 0.25517432689666747,
      "num_input_tokens_seen": 835941560,
      "step": 13770,
      "train_runtime": 179868.7194,
      "train_tokens_per_second": 4647.509
    },
    {
      "epoch": 1.757015306122449,
      "grad_norm": 0.7571677415678235,
      "learning_rate": 3.6017134514889196e-07,
      "loss": 0.26593616008758547,
      "num_input_tokens_seen": 836258672,
      "step": 13775,
      "train_runtime": 179923.4454,
      "train_tokens_per_second": 4647.858
    },
    {
      "epoch": 1.7576530612244898,
      "grad_norm": 0.6690829842295413,
      "learning_rate": 3.583070229685043e-07,
      "loss": 0.2549485206604004,
      "num_input_tokens_seen": 836561840,
      "step": 13780,
      "train_runtime": 179983.311,
      "train_tokens_per_second": 4647.997
    },
    {
      "epoch": 1.7582908163265305,
      "grad_norm": 0.7150546347355768,
      "learning_rate": 3.564473590511941e-07,
      "loss": 0.24862639904022216,
      "num_input_tokens_seen": 836865200,
      "step": 13785,
      "train_runtime": 180051.2278,
      "train_tokens_per_second": 4647.928
    },
    {
      "epoch": 1.7589285714285714,
      "grad_norm": 0.5953940498152313,
      "learning_rate": 3.5459235526326464e-07,
      "loss": 0.240686297416687,
      "num_input_tokens_seen": 837168000,
      "step": 13790,
      "train_runtime": 180115.1063,
      "train_tokens_per_second": 4647.961
    },
    {
      "epoch": 1.7595663265306123,
      "grad_norm": 0.6596747408033998,
      "learning_rate": 3.527420134663401e-07,
      "loss": 0.2507588863372803,
      "num_input_tokens_seen": 837463912,
      "step": 13795,
      "train_runtime": 180171.9969,
      "train_tokens_per_second": 4648.136
    },
    {
      "epoch": 1.760204081632653,
      "grad_norm": 0.7453931252493031,
      "learning_rate": 3.5089633551736946e-07,
      "loss": 0.246191143989563,
      "num_input_tokens_seen": 837767624,
      "step": 13800,
      "train_runtime": 180234.6778,
      "train_tokens_per_second": 4648.204
    },
    {
      "epoch": 1.7608418367346939,
      "grad_norm": 0.6343858395241456,
      "learning_rate": 3.4905532326861946e-07,
      "loss": 0.23849906921386718,
      "num_input_tokens_seen": 838076624,
      "step": 13805,
      "train_runtime": 180300.4707,
      "train_tokens_per_second": 4648.222
    },
    {
      "epoch": 1.7614795918367347,
      "grad_norm": 0.6993935655900692,
      "learning_rate": 3.4721897856767275e-07,
      "loss": 0.2654818773269653,
      "num_input_tokens_seen": 838387688,
      "step": 13810,
      "train_runtime": 180372.754,
      "train_tokens_per_second": 4648.084
    },
    {
      "epoch": 1.7621173469387754,
      "grad_norm": 0.7849860789012794,
      "learning_rate": 3.4538730325743174e-07,
      "loss": 0.24663257598876953,
      "num_input_tokens_seen": 838688192,
      "step": 13815,
      "train_runtime": 180446.2759,
      "train_tokens_per_second": 4647.855
    },
    {
      "epoch": 1.7627551020408163,
      "grad_norm": 0.6588620875640854,
      "learning_rate": 3.4356029917610833e-07,
      "loss": 0.24232873916625977,
      "num_input_tokens_seen": 838997520,
      "step": 13820,
      "train_runtime": 180511.1573,
      "train_tokens_per_second": 4647.898
    },
    {
      "epoch": 1.7633928571428572,
      "grad_norm": 0.6643840471979123,
      "learning_rate": 3.417379681572297e-07,
      "loss": 0.24702844619750977,
      "num_input_tokens_seen": 839296400,
      "step": 13825,
      "train_runtime": 180587.4613,
      "train_tokens_per_second": 4647.59
    },
    {
      "epoch": 1.7640306122448979,
      "grad_norm": 0.7288712257155641,
      "learning_rate": 3.3992031202963293e-07,
      "loss": 0.25472800731658934,
      "num_input_tokens_seen": 839608360,
      "step": 13830,
      "train_runtime": 180639.9779,
      "train_tokens_per_second": 4647.965
    },
    {
      "epoch": 1.7646683673469388,
      "grad_norm": 0.6195781054964941,
      "learning_rate": 3.3810733261746167e-07,
      "loss": 0.24146430492401122,
      "num_input_tokens_seen": 839922696,
      "step": 13835,
      "train_runtime": 180713.3854,
      "train_tokens_per_second": 4647.817
    },
    {
      "epoch": 1.7653061224489797,
      "grad_norm": 0.6629599072218642,
      "learning_rate": 3.362990317401688e-07,
      "loss": 0.2496337413787842,
      "num_input_tokens_seen": 840213632,
      "step": 13840,
      "train_runtime": 180780.084,
      "train_tokens_per_second": 4647.711
    },
    {
      "epoch": 1.7659438775510203,
      "grad_norm": 0.7104737196447849,
      "learning_rate": 3.344954112125087e-07,
      "loss": 0.23286681175231932,
      "num_input_tokens_seen": 840513384,
      "step": 13845,
      "train_runtime": 180858.3911,
      "train_tokens_per_second": 4647.356
    },
    {
      "epoch": 1.7665816326530612,
      "grad_norm": 0.6650838498230978,
      "learning_rate": 3.3269647284454253e-07,
      "loss": 0.24792745113372802,
      "num_input_tokens_seen": 840793696,
      "step": 13850,
      "train_runtime": 180928.1672,
      "train_tokens_per_second": 4647.113
    },
    {
      "epoch": 1.7672193877551021,
      "grad_norm": 0.7428204263279207,
      "learning_rate": 3.309022184416288e-07,
      "loss": 0.2583718776702881,
      "num_input_tokens_seen": 841104192,
      "step": 13855,
      "train_runtime": 180997.5697,
      "train_tokens_per_second": 4647.047
    },
    {
      "epoch": 1.7678571428571428,
      "grad_norm": 0.6503670677463549,
      "learning_rate": 3.2911264980442834e-07,
      "loss": 0.25423059463500974,
      "num_input_tokens_seen": 841415992,
      "step": 13860,
      "train_runtime": 181065.3529,
      "train_tokens_per_second": 4647.029
    },
    {
      "epoch": 1.7684948979591837,
      "grad_norm": 0.6359925475280099,
      "learning_rate": 3.273277687288978e-07,
      "loss": 0.2673047542572021,
      "num_input_tokens_seen": 841723464,
      "step": 13865,
      "train_runtime": 181131.2083,
      "train_tokens_per_second": 4647.037
    },
    {
      "epoch": 1.7691326530612246,
      "grad_norm": 0.6618450911892061,
      "learning_rate": 3.255475770062899e-07,
      "loss": 0.25228047370910645,
      "num_input_tokens_seen": 842018680,
      "step": 13870,
      "train_runtime": 181196.6132,
      "train_tokens_per_second": 4646.989
    },
    {
      "epoch": 1.7697704081632653,
      "grad_norm": 0.7361786903531883,
      "learning_rate": 3.237720764231517e-07,
      "loss": 0.27175297737121584,
      "num_input_tokens_seen": 842324096,
      "step": 13875,
      "train_runtime": 181260.2651,
      "train_tokens_per_second": 4647.042
    },
    {
      "epoch": 1.7704081632653061,
      "grad_norm": 0.6056380018779747,
      "learning_rate": 3.220012687613222e-07,
      "loss": 0.2552263975143433,
      "num_input_tokens_seen": 842624944,
      "step": 13880,
      "train_runtime": 181333.0954,
      "train_tokens_per_second": 4646.835
    },
    {
      "epoch": 1.771045918367347,
      "grad_norm": 0.7720243317578994,
      "learning_rate": 3.2023515579793e-07,
      "loss": 0.26123366355895994,
      "num_input_tokens_seen": 842936928,
      "step": 13885,
      "train_runtime": 181398.226,
      "train_tokens_per_second": 4646.886
    },
    {
      "epoch": 1.7716836734693877,
      "grad_norm": 0.6295963283423344,
      "learning_rate": 3.1847373930539326e-07,
      "loss": 0.24546425342559813,
      "num_input_tokens_seen": 843244832,
      "step": 13890,
      "train_runtime": 181454.3124,
      "train_tokens_per_second": 4647.147
    },
    {
      "epoch": 1.7723214285714286,
      "grad_norm": 0.6467099201436239,
      "learning_rate": 3.167170210514153e-07,
      "loss": 0.2333463191986084,
      "num_input_tokens_seen": 843537008,
      "step": 13895,
      "train_runtime": 181519.9392,
      "train_tokens_per_second": 4647.076
    },
    {
      "epoch": 1.7729591836734695,
      "grad_norm": 0.6329033059130065,
      "learning_rate": 3.14965002798987e-07,
      "loss": 0.2539116621017456,
      "num_input_tokens_seen": 843848648,
      "step": 13900,
      "train_runtime": 181589.5625,
      "train_tokens_per_second": 4647.011
    },
    {
      "epoch": 1.7735969387755102,
      "grad_norm": 0.6681516188241793,
      "learning_rate": 3.1321768630638073e-07,
      "loss": 0.23679733276367188,
      "num_input_tokens_seen": 844152576,
      "step": 13905,
      "train_runtime": 181647.8166,
      "train_tokens_per_second": 4647.194
    },
    {
      "epoch": 1.774234693877551,
      "grad_norm": 0.6513231584972891,
      "learning_rate": 3.1147507332715044e-07,
      "loss": 0.24175968170166015,
      "num_input_tokens_seen": 844450232,
      "step": 13910,
      "train_runtime": 181713.9919,
      "train_tokens_per_second": 4647.139
    },
    {
      "epoch": 1.774872448979592,
      "grad_norm": 0.6402550001567784,
      "learning_rate": 3.097371656101289e-07,
      "loss": 0.25646772384643557,
      "num_input_tokens_seen": 844753504,
      "step": 13915,
      "train_runtime": 181779.8747,
      "train_tokens_per_second": 4647.123
    },
    {
      "epoch": 1.7755102040816326,
      "grad_norm": 0.6702558880296042,
      "learning_rate": 3.080039648994304e-07,
      "loss": 0.2563156604766846,
      "num_input_tokens_seen": 845053312,
      "step": 13920,
      "train_runtime": 181842.9033,
      "train_tokens_per_second": 4647.161
    },
    {
      "epoch": 1.7761479591836735,
      "grad_norm": 0.6585912651253576,
      "learning_rate": 3.062754729344408e-07,
      "loss": 0.2489215612411499,
      "num_input_tokens_seen": 845353704,
      "step": 13925,
      "train_runtime": 181907.9236,
      "train_tokens_per_second": 4647.152
    },
    {
      "epoch": 1.7767857142857144,
      "grad_norm": 0.703042446598996,
      "learning_rate": 3.0455169144982257e-07,
      "loss": 0.24673690795898437,
      "num_input_tokens_seen": 845669088,
      "step": 13930,
      "train_runtime": 181977.421,
      "train_tokens_per_second": 4647.11
    },
    {
      "epoch": 1.777423469387755,
      "grad_norm": 0.6530625339120676,
      "learning_rate": 3.028326221755118e-07,
      "loss": 0.2515824317932129,
      "num_input_tokens_seen": 845963736,
      "step": 13935,
      "train_runtime": 182039.512,
      "train_tokens_per_second": 4647.144
    },
    {
      "epoch": 1.7780612244897958,
      "grad_norm": 0.6846638353735537,
      "learning_rate": 3.0111826683671407e-07,
      "loss": 0.29273903369903564,
      "num_input_tokens_seen": 846263064,
      "step": 13940,
      "train_runtime": 182101.1988,
      "train_tokens_per_second": 4647.213
    },
    {
      "epoch": 1.7786989795918369,
      "grad_norm": 0.6624536048396347,
      "learning_rate": 2.9940862715390483e-07,
      "loss": 0.2593369483947754,
      "num_input_tokens_seen": 846571384,
      "step": 13945,
      "train_runtime": 182176.6939,
      "train_tokens_per_second": 4646.98
    },
    {
      "epoch": 1.7793367346938775,
      "grad_norm": 0.6629850876763493,
      "learning_rate": 2.9770370484282665e-07,
      "loss": 0.24895977973937988,
      "num_input_tokens_seen": 846871040,
      "step": 13950,
      "train_runtime": 182251.9296,
      "train_tokens_per_second": 4646.705
    },
    {
      "epoch": 1.7799744897959182,
      "grad_norm": 0.7033528480998114,
      "learning_rate": 2.960035016144885e-07,
      "loss": 0.2705617666244507,
      "num_input_tokens_seen": 847170952,
      "step": 13955,
      "train_runtime": 182323.756,
      "train_tokens_per_second": 4646.52
    },
    {
      "epoch": 1.7806122448979593,
      "grad_norm": 0.6696348082058241,
      "learning_rate": 2.943080191751629e-07,
      "loss": 0.26220757961273194,
      "num_input_tokens_seen": 847480888,
      "step": 13960,
      "train_runtime": 182380.7598,
      "train_tokens_per_second": 4646.767
    },
    {
      "epoch": 1.78125,
      "grad_norm": 0.6381588890555563,
      "learning_rate": 2.9261725922638485e-07,
      "loss": 0.24409053325653077,
      "num_input_tokens_seen": 847780648,
      "step": 13965,
      "train_runtime": 182447.1243,
      "train_tokens_per_second": 4646.72
    },
    {
      "epoch": 1.7818877551020407,
      "grad_norm": 0.6943220822102325,
      "learning_rate": 2.9093122346494974e-07,
      "loss": 0.23370254039764404,
      "num_input_tokens_seen": 848066224,
      "step": 13970,
      "train_runtime": 182503.7318,
      "train_tokens_per_second": 4646.843
    },
    {
      "epoch": 1.7825255102040818,
      "grad_norm": 0.5779729833787437,
      "learning_rate": 2.8924991358291354e-07,
      "loss": 0.2403944969177246,
      "num_input_tokens_seen": 848364288,
      "step": 13975,
      "train_runtime": 182557.4179,
      "train_tokens_per_second": 4647.109
    },
    {
      "epoch": 1.7831632653061225,
      "grad_norm": 0.7469470835026416,
      "learning_rate": 2.875733312675855e-07,
      "loss": 0.2287726640701294,
      "num_input_tokens_seen": 848652288,
      "step": 13980,
      "train_runtime": 182618.8975,
      "train_tokens_per_second": 4647.122
    },
    {
      "epoch": 1.7838010204081631,
      "grad_norm": 0.7216525773887124,
      "learning_rate": 2.8590147820153513e-07,
      "loss": 0.2606127977371216,
      "num_input_tokens_seen": 848956280,
      "step": 13985,
      "train_runtime": 182683.4529,
      "train_tokens_per_second": 4647.144
    },
    {
      "epoch": 1.7844387755102042,
      "grad_norm": 0.6859899777720221,
      "learning_rate": 2.8423435606258485e-07,
      "loss": 0.25071115493774415,
      "num_input_tokens_seen": 849265056,
      "step": 13990,
      "train_runtime": 182740.7151,
      "train_tokens_per_second": 4647.377
    },
    {
      "epoch": 1.785076530612245,
      "grad_norm": 0.7838533272897487,
      "learning_rate": 2.8257196652380523e-07,
      "loss": 0.27875847816467286,
      "num_input_tokens_seen": 849562600,
      "step": 13995,
      "train_runtime": 182812.1603,
      "train_tokens_per_second": 4647.189
    },
    {
      "epoch": 1.7857142857142856,
      "grad_norm": 0.6870819284813265,
      "learning_rate": 2.809143112535212e-07,
      "loss": 0.25108349323272705,
      "num_input_tokens_seen": 849859784,
      "step": 14000,
      "train_runtime": 182873.9372,
      "train_tokens_per_second": 4647.244
    },
    {
      "epoch": 1.7863520408163265,
      "grad_norm": 0.6557445435384791,
      "learning_rate": 2.7926139191530697e-07,
      "loss": 0.24347131252288817,
      "num_input_tokens_seen": 850149912,
      "step": 14005,
      "train_runtime": 182942.3276,
      "train_tokens_per_second": 4647.092
    },
    {
      "epoch": 1.7869897959183674,
      "grad_norm": 0.6161919896676236,
      "learning_rate": 2.7761321016798183e-07,
      "loss": 0.25662524700164796,
      "num_input_tokens_seen": 850451656,
      "step": 14010,
      "train_runtime": 183007.8151,
      "train_tokens_per_second": 4647.078
    },
    {
      "epoch": 1.787627551020408,
      "grad_norm": 0.6480858234573649,
      "learning_rate": 2.7596976766560977e-07,
      "loss": 0.25433058738708497,
      "num_input_tokens_seen": 850768992,
      "step": 14015,
      "train_runtime": 183080.2664,
      "train_tokens_per_second": 4646.973
    },
    {
      "epoch": 1.788265306122449,
      "grad_norm": 0.6157549086187802,
      "learning_rate": 2.743310660575027e-07,
      "loss": 0.22808046340942384,
      "num_input_tokens_seen": 851067016,
      "step": 14020,
      "train_runtime": 183152.4548,
      "train_tokens_per_second": 4646.768
    },
    {
      "epoch": 1.7889030612244898,
      "grad_norm": 0.6577093841837045,
      "learning_rate": 2.726971069882101e-07,
      "loss": 0.2544532060623169,
      "num_input_tokens_seen": 851376720,
      "step": 14025,
      "train_runtime": 183229.1979,
      "train_tokens_per_second": 4646.512
    },
    {
      "epoch": 1.7895408163265305,
      "grad_norm": 0.6313778365859739,
      "learning_rate": 2.7106789209752513e-07,
      "loss": 0.23698220252990723,
      "num_input_tokens_seen": 851672720,
      "step": 14030,
      "train_runtime": 183296.8813,
      "train_tokens_per_second": 4646.411
    },
    {
      "epoch": 1.7901785714285714,
      "grad_norm": 0.6976681415900199,
      "learning_rate": 2.6944342302047785e-07,
      "loss": 0.2619039297103882,
      "num_input_tokens_seen": 851976424,
      "step": 14035,
      "train_runtime": 183368.148,
      "train_tokens_per_second": 4646.262
    },
    {
      "epoch": 1.7908163265306123,
      "grad_norm": 0.6318411761298479,
      "learning_rate": 2.678237013873375e-07,
      "loss": 0.26267783641815184,
      "num_input_tokens_seen": 852281432,
      "step": 14040,
      "train_runtime": 183423.3029,
      "train_tokens_per_second": 4646.528
    },
    {
      "epoch": 1.791454081632653,
      "grad_norm": 0.6047257626502369,
      "learning_rate": 2.66208728823607e-07,
      "loss": 0.2652120113372803,
      "num_input_tokens_seen": 852594760,
      "step": 14045,
      "train_runtime": 183498.0974,
      "train_tokens_per_second": 4646.341
    },
    {
      "epoch": 1.7920918367346939,
      "grad_norm": 0.7264580741551205,
      "learning_rate": 2.6459850695002454e-07,
      "loss": 0.24768829345703125,
      "num_input_tokens_seen": 852914064,
      "step": 14050,
      "train_runtime": 183548.1676,
      "train_tokens_per_second": 4646.813
    },
    {
      "epoch": 1.7927295918367347,
      "grad_norm": 0.7594343746620614,
      "learning_rate": 2.629930373825595e-07,
      "loss": 0.24914608001708985,
      "num_input_tokens_seen": 853226440,
      "step": 14055,
      "train_runtime": 183609.4309,
      "train_tokens_per_second": 4646.964
    },
    {
      "epoch": 1.7933673469387754,
      "grad_norm": 0.7112763675532762,
      "learning_rate": 2.6139232173241434e-07,
      "loss": 0.2694690227508545,
      "num_input_tokens_seen": 853536736,
      "step": 14060,
      "train_runtime": 183679.4922,
      "train_tokens_per_second": 4646.881
    },
    {
      "epoch": 1.7940051020408163,
      "grad_norm": 0.6551911498856521,
      "learning_rate": 2.5979636160601673e-07,
      "loss": 0.24811267852783203,
      "num_input_tokens_seen": 853826704,
      "step": 14065,
      "train_runtime": 183747.2023,
      "train_tokens_per_second": 4646.747
    },
    {
      "epoch": 1.7946428571428572,
      "grad_norm": 0.6838683823425431,
      "learning_rate": 2.5820515860502604e-07,
      "loss": 0.27320079803466796,
      "num_input_tokens_seen": 854143904,
      "step": 14070,
      "train_runtime": 183806.5216,
      "train_tokens_per_second": 4646.973
    },
    {
      "epoch": 1.7952806122448979,
      "grad_norm": 0.6286894231694805,
      "learning_rate": 2.5661871432632545e-07,
      "loss": 0.2347564935684204,
      "num_input_tokens_seen": 854451200,
      "step": 14075,
      "train_runtime": 183870.7025,
      "train_tokens_per_second": 4647.022
    },
    {
      "epoch": 1.7959183673469388,
      "grad_norm": 0.4986039569614931,
      "learning_rate": 2.550370303620214e-07,
      "loss": 0.20640251636505128,
      "num_input_tokens_seen": 854750472,
      "step": 14080,
      "train_runtime": 183938.071,
      "train_tokens_per_second": 4646.947
    },
    {
      "epoch": 1.7965561224489797,
      "grad_norm": 0.7190719301797829,
      "learning_rate": 2.534601082994437e-07,
      "loss": 0.2538137912750244,
      "num_input_tokens_seen": 855047024,
      "step": 14085,
      "train_runtime": 184005.8618,
      "train_tokens_per_second": 4646.847
    },
    {
      "epoch": 1.7971938775510203,
      "grad_norm": 0.6826560743610066,
      "learning_rate": 2.5188794972114684e-07,
      "loss": 0.2629712104797363,
      "num_input_tokens_seen": 855326880,
      "step": 14090,
      "train_runtime": 184074.28,
      "train_tokens_per_second": 4646.64
    },
    {
      "epoch": 1.7978316326530612,
      "grad_norm": 0.6121660757427955,
      "learning_rate": 2.503205562048994e-07,
      "loss": 0.23544673919677733,
      "num_input_tokens_seen": 855626608,
      "step": 14095,
      "train_runtime": 184150.4967,
      "train_tokens_per_second": 4646.344
    },
    {
      "epoch": 1.7984693877551021,
      "grad_norm": 0.7785284207508509,
      "learning_rate": 2.487579293236897e-07,
      "loss": 0.2749136447906494,
      "num_input_tokens_seen": 855935280,
      "step": 14100,
      "train_runtime": 184217.7303,
      "train_tokens_per_second": 4646.324
    },
    {
      "epoch": 1.7991071428571428,
      "grad_norm": 0.6536024201963503,
      "learning_rate": 2.4720007064572506e-07,
      "loss": 0.25101404190063475,
      "num_input_tokens_seen": 856237432,
      "step": 14105,
      "train_runtime": 184284.5597,
      "train_tokens_per_second": 4646.279
    },
    {
      "epoch": 1.7997448979591837,
      "grad_norm": 0.713273301359594,
      "learning_rate": 2.456469817344237e-07,
      "loss": 0.24380555152893066,
      "num_input_tokens_seen": 856528816,
      "step": 14110,
      "train_runtime": 184348.4076,
      "train_tokens_per_second": 4646.25
    },
    {
      "epoch": 1.8003826530612246,
      "grad_norm": 0.7100508414107232,
      "learning_rate": 2.4409866414841996e-07,
      "loss": 0.25036029815673827,
      "num_input_tokens_seen": 856837488,
      "step": 14115,
      "train_runtime": 184398.2947,
      "train_tokens_per_second": 4646.667
    },
    {
      "epoch": 1.8010204081632653,
      "grad_norm": 0.606344854788227,
      "learning_rate": 2.4255511944155764e-07,
      "loss": 0.2392751932144165,
      "num_input_tokens_seen": 857138856,
      "step": 14120,
      "train_runtime": 184466.5761,
      "train_tokens_per_second": 4646.581
    },
    {
      "epoch": 1.8016581632653061,
      "grad_norm": 0.5690719359798412,
      "learning_rate": 2.410163491628925e-07,
      "loss": 0.2524121761322021,
      "num_input_tokens_seen": 857460528,
      "step": 14125,
      "train_runtime": 184523.5605,
      "train_tokens_per_second": 4646.889
    },
    {
      "epoch": 1.802295918367347,
      "grad_norm": 0.6631517596741462,
      "learning_rate": 2.394823548566877e-07,
      "loss": 0.24928345680236816,
      "num_input_tokens_seen": 857764176,
      "step": 14130,
      "train_runtime": 184603.0616,
      "train_tokens_per_second": 4646.533
    },
    {
      "epoch": 1.8029336734693877,
      "grad_norm": 0.6523048042702136,
      "learning_rate": 2.3795313806241337e-07,
      "loss": 0.23600704669952394,
      "num_input_tokens_seen": 858062936,
      "step": 14135,
      "train_runtime": 184666.2406,
      "train_tokens_per_second": 4646.561
    },
    {
      "epoch": 1.8035714285714286,
      "grad_norm": 0.6837868788862734,
      "learning_rate": 2.364287003147453e-07,
      "loss": 0.24544596672058105,
      "num_input_tokens_seen": 858351512,
      "step": 14140,
      "train_runtime": 184732.4412,
      "train_tokens_per_second": 4646.458
    },
    {
      "epoch": 1.8042091836734695,
      "grad_norm": 0.6808156534706771,
      "learning_rate": 2.3490904314356412e-07,
      "loss": 0.2638862133026123,
      "num_input_tokens_seen": 858666680,
      "step": 14145,
      "train_runtime": 184807.5782,
      "train_tokens_per_second": 4646.274
    },
    {
      "epoch": 1.8048469387755102,
      "grad_norm": 0.7256553064605521,
      "learning_rate": 2.3339416807395154e-07,
      "loss": 0.25586857795715334,
      "num_input_tokens_seen": 858970664,
      "step": 14150,
      "train_runtime": 184873.5503,
      "train_tokens_per_second": 4646.26
    },
    {
      "epoch": 1.805484693877551,
      "grad_norm": 0.6270706782907929,
      "learning_rate": 2.3188407662618983e-07,
      "loss": 0.24245853424072267,
      "num_input_tokens_seen": 859261112,
      "step": 14155,
      "train_runtime": 184941.1489,
      "train_tokens_per_second": 4646.133
    },
    {
      "epoch": 1.806122448979592,
      "grad_norm": 0.64741133491633,
      "learning_rate": 2.3037877031576172e-07,
      "loss": 0.2581301212310791,
      "num_input_tokens_seen": 859577136,
      "step": 14160,
      "train_runtime": 185000.7743,
      "train_tokens_per_second": 4646.343
    },
    {
      "epoch": 1.8067602040816326,
      "grad_norm": 0.6632840804843064,
      "learning_rate": 2.2887825065334822e-07,
      "loss": 0.2506843566894531,
      "num_input_tokens_seen": 859880464,
      "step": 14165,
      "train_runtime": 185065.143,
      "train_tokens_per_second": 4646.366
    },
    {
      "epoch": 1.8073979591836735,
      "grad_norm": 0.713382029018524,
      "learning_rate": 2.2738251914482312e-07,
      "loss": 0.24305098056793212,
      "num_input_tokens_seen": 860178936,
      "step": 14170,
      "train_runtime": 185142.2931,
      "train_tokens_per_second": 4646.042
    },
    {
      "epoch": 1.8080357142857144,
      "grad_norm": 0.6850868170743648,
      "learning_rate": 2.258915772912601e-07,
      "loss": 0.23382172584533692,
      "num_input_tokens_seen": 860472656,
      "step": 14175,
      "train_runtime": 185210.4182,
      "train_tokens_per_second": 4645.919
    },
    {
      "epoch": 1.808673469387755,
      "grad_norm": 0.6509414843067686,
      "learning_rate": 2.2440542658892172e-07,
      "loss": 0.26403107643127444,
      "num_input_tokens_seen": 860776504,
      "step": 14180,
      "train_runtime": 185277.0564,
      "train_tokens_per_second": 4645.888
    },
    {
      "epoch": 1.8093112244897958,
      "grad_norm": 0.663937822628379,
      "learning_rate": 2.2292406852926385e-07,
      "loss": 0.25989222526550293,
      "num_input_tokens_seen": 861080704,
      "step": 14185,
      "train_runtime": 185342.5877,
      "train_tokens_per_second": 4645.887
    },
    {
      "epoch": 1.8099489795918369,
      "grad_norm": 0.7188686180245325,
      "learning_rate": 2.214475045989345e-07,
      "loss": 0.2608367919921875,
      "num_input_tokens_seen": 861392760,
      "step": 14190,
      "train_runtime": 185399.4962,
      "train_tokens_per_second": 4646.144
    },
    {
      "epoch": 1.8105867346938775,
      "grad_norm": 0.6708563723098995,
      "learning_rate": 2.199757362797672e-07,
      "loss": 0.24211552143096923,
      "num_input_tokens_seen": 861703280,
      "step": 14195,
      "train_runtime": 185469.4488,
      "train_tokens_per_second": 4646.066
    },
    {
      "epoch": 1.8112244897959182,
      "grad_norm": 0.706510013791032,
      "learning_rate": 2.1850876504878492e-07,
      "loss": 0.26641933917999266,
      "num_input_tokens_seen": 862005704,
      "step": 14200,
      "train_runtime": 185522.502,
      "train_tokens_per_second": 4646.367
    },
    {
      "epoch": 1.8118622448979593,
      "grad_norm": 0.6217409753033918,
      "learning_rate": 2.1704659237819558e-07,
      "loss": 0.2593922853469849,
      "num_input_tokens_seen": 862313352,
      "step": 14205,
      "train_runtime": 185594.2525,
      "train_tokens_per_second": 4646.229
    },
    {
      "epoch": 1.8125,
      "grad_norm": 0.6715338430285219,
      "learning_rate": 2.1558921973539092e-07,
      "loss": 0.22741270065307617,
      "num_input_tokens_seen": 862611096,
      "step": 14210,
      "train_runtime": 185656.4179,
      "train_tokens_per_second": 4646.277
    },
    {
      "epoch": 1.8131377551020407,
      "grad_norm": 0.6998027270839655,
      "learning_rate": 2.1413664858294713e-07,
      "loss": 0.2631678104400635,
      "num_input_tokens_seen": 862909136,
      "step": 14215,
      "train_runtime": 185707.6256,
      "train_tokens_per_second": 4646.6
    },
    {
      "epoch": 1.8137755102040818,
      "grad_norm": 0.5670888159968765,
      "learning_rate": 2.126888803786209e-07,
      "loss": 0.23754780292510985,
      "num_input_tokens_seen": 863205776,
      "step": 14220,
      "train_runtime": 185763.461,
      "train_tokens_per_second": 4646.801
    },
    {
      "epoch": 1.8144132653061225,
      "grad_norm": 0.7433984551662745,
      "learning_rate": 2.1124591657534776e-07,
      "loss": 0.2515824556350708,
      "num_input_tokens_seen": 863513144,
      "step": 14225,
      "train_runtime": 185823.9326,
      "train_tokens_per_second": 4646.943
    },
    {
      "epoch": 1.8150510204081631,
      "grad_norm": 0.6718313238722916,
      "learning_rate": 2.0980775862124326e-07,
      "loss": 0.26154017448425293,
      "num_input_tokens_seen": 863819928,
      "step": 14230,
      "train_runtime": 185886.3326,
      "train_tokens_per_second": 4647.033
    },
    {
      "epoch": 1.8156887755102042,
      "grad_norm": 0.7065683223653454,
      "learning_rate": 2.083744079595995e-07,
      "loss": 0.24450421333312988,
      "num_input_tokens_seen": 864123928,
      "step": 14235,
      "train_runtime": 185939.1605,
      "train_tokens_per_second": 4647.348
    },
    {
      "epoch": 1.816326530612245,
      "grad_norm": 0.6260478894014965,
      "learning_rate": 2.069458660288831e-07,
      "loss": 0.25495457649230957,
      "num_input_tokens_seen": 864431064,
      "step": 14240,
      "train_runtime": 185996.0951,
      "train_tokens_per_second": 4647.576
    },
    {
      "epoch": 1.8169642857142856,
      "grad_norm": 0.6957318363665137,
      "learning_rate": 2.0552213426273725e-07,
      "loss": 0.26381802558898926,
      "num_input_tokens_seen": 864729568,
      "step": 14245,
      "train_runtime": 186059.0462,
      "train_tokens_per_second": 4647.608
    },
    {
      "epoch": 1.8176020408163265,
      "grad_norm": 0.6389361327846603,
      "learning_rate": 2.0410321408997459e-07,
      "loss": 0.25765416622161863,
      "num_input_tokens_seen": 865040992,
      "step": 14250,
      "train_runtime": 186137.216,
      "train_tokens_per_second": 4647.33
    },
    {
      "epoch": 1.8182397959183674,
      "grad_norm": 0.6401658008989223,
      "learning_rate": 2.0268910693458154e-07,
      "loss": 0.25167059898376465,
      "num_input_tokens_seen": 865330088,
      "step": 14255,
      "train_runtime": 186201.8332,
      "train_tokens_per_second": 4647.269
    },
    {
      "epoch": 1.818877551020408,
      "grad_norm": 0.6473296193207879,
      "learning_rate": 2.0127981421571295e-07,
      "loss": 0.23990678787231445,
      "num_input_tokens_seen": 865633168,
      "step": 14260,
      "train_runtime": 186268.6883,
      "train_tokens_per_second": 4647.229
    },
    {
      "epoch": 1.819515306122449,
      "grad_norm": 0.7195140822124848,
      "learning_rate": 1.998753373476936e-07,
      "loss": 0.25752208232879636,
      "num_input_tokens_seen": 865932040,
      "step": 14265,
      "train_runtime": 186330.5301,
      "train_tokens_per_second": 4647.29
    },
    {
      "epoch": 1.8201530612244898,
      "grad_norm": 0.7001758138626799,
      "learning_rate": 1.9847567774001153e-07,
      "loss": 0.2669686794281006,
      "num_input_tokens_seen": 866232520,
      "step": 14270,
      "train_runtime": 186394.3236,
      "train_tokens_per_second": 4647.312
    },
    {
      "epoch": 1.8207908163265305,
      "grad_norm": 0.7296375689862533,
      "learning_rate": 1.9708083679732538e-07,
      "loss": 0.2516944885253906,
      "num_input_tokens_seen": 866539944,
      "step": 14275,
      "train_runtime": 186458.8822,
      "train_tokens_per_second": 4647.351
    },
    {
      "epoch": 1.8214285714285714,
      "grad_norm": 0.6677299264673359,
      "learning_rate": 1.9569081591945537e-07,
      "loss": 0.2641996622085571,
      "num_input_tokens_seen": 866835376,
      "step": 14280,
      "train_runtime": 186521.6387,
      "train_tokens_per_second": 4647.372
    },
    {
      "epoch": 1.8220663265306123,
      "grad_norm": 0.6780265762181436,
      "learning_rate": 1.9430561650138292e-07,
      "loss": 0.2752277374267578,
      "num_input_tokens_seen": 867149688,
      "step": 14285,
      "train_runtime": 186588.3087,
      "train_tokens_per_second": 4647.396
    },
    {
      "epoch": 1.822704081632653,
      "grad_norm": 0.700308185107038,
      "learning_rate": 1.9292523993325318e-07,
      "loss": 0.2481313943862915,
      "num_input_tokens_seen": 867451672,
      "step": 14290,
      "train_runtime": 186662.5271,
      "train_tokens_per_second": 4647.166
    },
    {
      "epoch": 1.8233418367346939,
      "grad_norm": 0.7158540700829921,
      "learning_rate": 1.9154968760037142e-07,
      "loss": 0.24785711765289306,
      "num_input_tokens_seen": 867746048,
      "step": 14295,
      "train_runtime": 186727.6649,
      "train_tokens_per_second": 4647.121
    },
    {
      "epoch": 1.8239795918367347,
      "grad_norm": 0.7064749893470933,
      "learning_rate": 1.9017896088319888e-07,
      "loss": 0.26390371322631834,
      "num_input_tokens_seen": 868050248,
      "step": 14300,
      "train_runtime": 186786.3421,
      "train_tokens_per_second": 4647.29
    },
    {
      "epoch": 1.8246173469387754,
      "grad_norm": 0.6555062357278937,
      "learning_rate": 1.8881306115735632e-07,
      "loss": 0.23825533390045167,
      "num_input_tokens_seen": 868353320,
      "step": 14305,
      "train_runtime": 186851.2879,
      "train_tokens_per_second": 4647.296
    },
    {
      "epoch": 1.8252551020408163,
      "grad_norm": 0.6445761263231499,
      "learning_rate": 1.8745198979361944e-07,
      "loss": 0.25106494426727294,
      "num_input_tokens_seen": 868666888,
      "step": 14310,
      "train_runtime": 186906.1346,
      "train_tokens_per_second": 4647.61
    },
    {
      "epoch": 1.8258928571428572,
      "grad_norm": 0.6356923129717615,
      "learning_rate": 1.8609574815791843e-07,
      "loss": 0.25816683769226073,
      "num_input_tokens_seen": 868959528,
      "step": 14315,
      "train_runtime": 186975.7484,
      "train_tokens_per_second": 4647.445
    },
    {
      "epoch": 1.8265306122448979,
      "grad_norm": 0.6946034954343195,
      "learning_rate": 1.8474433761133614e-07,
      "loss": 0.2542288780212402,
      "num_input_tokens_seen": 869269160,
      "step": 14320,
      "train_runtime": 187046.8974,
      "train_tokens_per_second": 4647.333
    },
    {
      "epoch": 1.8271683673469388,
      "grad_norm": 0.660378050821299,
      "learning_rate": 1.833977595101083e-07,
      "loss": 0.26385793685913084,
      "num_input_tokens_seen": 869579752,
      "step": 14325,
      "train_runtime": 187114.798,
      "train_tokens_per_second": 4647.306
    },
    {
      "epoch": 1.8278061224489797,
      "grad_norm": 0.6366486775794228,
      "learning_rate": 1.8205601520561945e-07,
      "loss": 0.24700777530670165,
      "num_input_tokens_seen": 869872912,
      "step": 14330,
      "train_runtime": 187182.5796,
      "train_tokens_per_second": 4647.189
    },
    {
      "epoch": 1.8284438775510203,
      "grad_norm": 0.6473337544815833,
      "learning_rate": 1.8071910604440356e-07,
      "loss": 0.2464057445526123,
      "num_input_tokens_seen": 870179712,
      "step": 14335,
      "train_runtime": 187246.6634,
      "train_tokens_per_second": 4647.237
    },
    {
      "epoch": 1.8290816326530612,
      "grad_norm": 0.5948782537013565,
      "learning_rate": 1.79387033368143e-07,
      "loss": 0.23783507347106933,
      "num_input_tokens_seen": 870472648,
      "step": 14340,
      "train_runtime": 187306.0704,
      "train_tokens_per_second": 4647.327
    },
    {
      "epoch": 1.8297193877551021,
      "grad_norm": 0.6901984740177014,
      "learning_rate": 1.7805979851366505e-07,
      "loss": 0.25328662395477297,
      "num_input_tokens_seen": 870770136,
      "step": 14345,
      "train_runtime": 187369.6636,
      "train_tokens_per_second": 4647.338
    },
    {
      "epoch": 1.8303571428571428,
      "grad_norm": 0.6756107635776969,
      "learning_rate": 1.7673740281294316e-07,
      "loss": 0.25124058723449705,
      "num_input_tokens_seen": 871079520,
      "step": 14350,
      "train_runtime": 187444.0672,
      "train_tokens_per_second": 4647.144
    },
    {
      "epoch": 1.8309948979591837,
      "grad_norm": 0.6127683238599845,
      "learning_rate": 1.7541984759309294e-07,
      "loss": 0.2524874687194824,
      "num_input_tokens_seen": 871377240,
      "step": 14355,
      "train_runtime": 187512.5228,
      "train_tokens_per_second": 4647.035
    },
    {
      "epoch": 1.8316326530612246,
      "grad_norm": 0.674476404412451,
      "learning_rate": 1.7410713417637338e-07,
      "loss": 0.2508760929107666,
      "num_input_tokens_seen": 871676128,
      "step": 14360,
      "train_runtime": 187577.2399,
      "train_tokens_per_second": 4647.025
    },
    {
      "epoch": 1.8322704081632653,
      "grad_norm": 0.7270189261309183,
      "learning_rate": 1.7279926388018564e-07,
      "loss": 0.24621005058288575,
      "num_input_tokens_seen": 871980776,
      "step": 14365,
      "train_runtime": 187641.0739,
      "train_tokens_per_second": 4647.068
    },
    {
      "epoch": 1.8329081632653061,
      "grad_norm": 0.6963106266766658,
      "learning_rate": 1.714962380170665e-07,
      "loss": 0.23934614658355713,
      "num_input_tokens_seen": 872279784,
      "step": 14370,
      "train_runtime": 187710.1285,
      "train_tokens_per_second": 4646.951
    },
    {
      "epoch": 1.833545918367347,
      "grad_norm": 0.6853932411913404,
      "learning_rate": 1.701980578946938e-07,
      "loss": 0.22678751945495607,
      "num_input_tokens_seen": 872592464,
      "step": 14375,
      "train_runtime": 187774.326,
      "train_tokens_per_second": 4647.028
    },
    {
      "epoch": 1.8341836734693877,
      "grad_norm": 0.7962057581032416,
      "learning_rate": 1.6890472481588426e-07,
      "loss": 0.2569415092468262,
      "num_input_tokens_seen": 872884192,
      "step": 14380,
      "train_runtime": 187848.7707,
      "train_tokens_per_second": 4646.739
    },
    {
      "epoch": 1.8348214285714286,
      "grad_norm": 0.6615958345699283,
      "learning_rate": 1.6761624007858524e-07,
      "loss": 0.24323534965515137,
      "num_input_tokens_seen": 873200056,
      "step": 14385,
      "train_runtime": 187899.9735,
      "train_tokens_per_second": 4647.154
    },
    {
      "epoch": 1.8354591836734695,
      "grad_norm": 0.6614228182106681,
      "learning_rate": 1.6633260497588233e-07,
      "loss": 0.25774874687194826,
      "num_input_tokens_seen": 873496664,
      "step": 14390,
      "train_runtime": 187974.3484,
      "train_tokens_per_second": 4646.893
    },
    {
      "epoch": 1.8360969387755102,
      "grad_norm": 0.6877463849143334,
      "learning_rate": 1.6505382079599341e-07,
      "loss": 0.26739423274993895,
      "num_input_tokens_seen": 873815000,
      "step": 14395,
      "train_runtime": 188039.779,
      "train_tokens_per_second": 4646.969
    },
    {
      "epoch": 1.836734693877551,
      "grad_norm": 0.7078027136954513,
      "learning_rate": 1.6377988882226691e-07,
      "loss": 0.2510525226593018,
      "num_input_tokens_seen": 874111704,
      "step": 14400,
      "train_runtime": 188116.0135,
      "train_tokens_per_second": 4646.663
    },
    {
      "epoch": 1.837372448979592,
      "grad_norm": 0.7580819643931567,
      "learning_rate": 1.6251081033318294e-07,
      "loss": 0.2626865148544312,
      "num_input_tokens_seen": 874416736,
      "step": 14405,
      "train_runtime": 188183.8824,
      "train_tokens_per_second": 4646.608
    },
    {
      "epoch": 1.8380102040816326,
      "grad_norm": 0.7085086490871588,
      "learning_rate": 1.6124658660234993e-07,
      "loss": 0.25091991424560545,
      "num_input_tokens_seen": 874708496,
      "step": 14410,
      "train_runtime": 188256.4407,
      "train_tokens_per_second": 4646.367
    },
    {
      "epoch": 1.8386479591836735,
      "grad_norm": 0.6446046434293724,
      "learning_rate": 1.5998721889850577e-07,
      "loss": 0.2551647901535034,
      "num_input_tokens_seen": 875023056,
      "step": 14415,
      "train_runtime": 188306.5667,
      "train_tokens_per_second": 4646.801
    },
    {
      "epoch": 1.8392857142857144,
      "grad_norm": 0.662134237887001,
      "learning_rate": 1.5873270848551227e-07,
      "loss": 0.2508412837982178,
      "num_input_tokens_seen": 875310232,
      "step": 14420,
      "train_runtime": 188373.5681,
      "train_tokens_per_second": 4646.672
    },
    {
      "epoch": 1.839923469387755,
      "grad_norm": 0.6112720261337682,
      "learning_rate": 1.5748305662236008e-07,
      "loss": 0.2600100994110107,
      "num_input_tokens_seen": 875617936,
      "step": 14425,
      "train_runtime": 188442.4026,
      "train_tokens_per_second": 4646.608
    },
    {
      "epoch": 1.8405612244897958,
      "grad_norm": 0.610175837551541,
      "learning_rate": 1.562382645631605e-07,
      "loss": 0.24649903774261475,
      "num_input_tokens_seen": 875913640,
      "step": 14430,
      "train_runtime": 188515.8723,
      "train_tokens_per_second": 4646.365
    },
    {
      "epoch": 1.8411989795918369,
      "grad_norm": 0.6658147878089529,
      "learning_rate": 1.549983335571509e-07,
      "loss": 0.24332361221313475,
      "num_input_tokens_seen": 876218840,
      "step": 14435,
      "train_runtime": 188572.153,
      "train_tokens_per_second": 4646.597
    },
    {
      "epoch": 1.8418367346938775,
      "grad_norm": 0.7215403228955583,
      "learning_rate": 1.537632648486864e-07,
      "loss": 0.2727149248123169,
      "num_input_tokens_seen": 876527088,
      "step": 14440,
      "train_runtime": 188635.6936,
      "train_tokens_per_second": 4646.666
    },
    {
      "epoch": 1.8424744897959182,
      "grad_norm": 0.6682756447039231,
      "learning_rate": 1.5253305967724674e-07,
      "loss": 0.24445388317108155,
      "num_input_tokens_seen": 876825528,
      "step": 14445,
      "train_runtime": 188693.5814,
      "train_tokens_per_second": 4646.822
    },
    {
      "epoch": 1.8431122448979593,
      "grad_norm": 0.6355429041547278,
      "learning_rate": 1.513077192774287e-07,
      "loss": 0.2287886619567871,
      "num_input_tokens_seen": 877131672,
      "step": 14450,
      "train_runtime": 188760.8044,
      "train_tokens_per_second": 4646.789
    },
    {
      "epoch": 1.84375,
      "grad_norm": 0.6808345607996471,
      "learning_rate": 1.500872448789459e-07,
      "loss": 0.2465895652770996,
      "num_input_tokens_seen": 877437424,
      "step": 14455,
      "train_runtime": 188816.2473,
      "train_tokens_per_second": 4647.044
    },
    {
      "epoch": 1.8443877551020407,
      "grad_norm": 0.6173045192997421,
      "learning_rate": 1.4887163770662905e-07,
      "loss": 0.2551894664764404,
      "num_input_tokens_seen": 877726048,
      "step": 14460,
      "train_runtime": 188873.6196,
      "train_tokens_per_second": 4647.161
    },
    {
      "epoch": 1.8450255102040818,
      "grad_norm": 0.6090481756128957,
      "learning_rate": 1.4766089898042678e-07,
      "loss": 0.2440091848373413,
      "num_input_tokens_seen": 878032896,
      "step": 14465,
      "train_runtime": 188930.2722,
      "train_tokens_per_second": 4647.391
    },
    {
      "epoch": 1.8456632653061225,
      "grad_norm": 0.6849560768714724,
      "learning_rate": 1.4645502991539883e-07,
      "loss": 0.24983086585998535,
      "num_input_tokens_seen": 878336096,
      "step": 14470,
      "train_runtime": 189001.2366,
      "train_tokens_per_second": 4647.251
    },
    {
      "epoch": 1.8463010204081631,
      "grad_norm": 0.6332439136947746,
      "learning_rate": 1.452540317217188e-07,
      "loss": 0.2324315309524536,
      "num_input_tokens_seen": 878647240,
      "step": 14475,
      "train_runtime": 189074.486,
      "train_tokens_per_second": 4647.096
    },
    {
      "epoch": 1.8469387755102042,
      "grad_norm": 0.6390851043551204,
      "learning_rate": 1.440579056046737e-07,
      "loss": 0.2568353176116943,
      "num_input_tokens_seen": 878950824,
      "step": 14480,
      "train_runtime": 189139.028,
      "train_tokens_per_second": 4647.115
    },
    {
      "epoch": 1.847576530612245,
      "grad_norm": 0.7183239161813246,
      "learning_rate": 1.428666527646577e-07,
      "loss": 0.2626702308654785,
      "num_input_tokens_seen": 879270336,
      "step": 14485,
      "train_runtime": 189198.8639,
      "train_tokens_per_second": 4647.334
    },
    {
      "epoch": 1.8482142857142856,
      "grad_norm": 0.7044687927889742,
      "learning_rate": 1.4168027439717735e-07,
      "loss": 0.25818753242492676,
      "num_input_tokens_seen": 879571280,
      "step": 14490,
      "train_runtime": 189262.0499,
      "train_tokens_per_second": 4647.373
    },
    {
      "epoch": 1.8488520408163265,
      "grad_norm": 0.640036435227342,
      "learning_rate": 1.4049877169284575e-07,
      "loss": 0.2519019842147827,
      "num_input_tokens_seen": 879869408,
      "step": 14495,
      "train_runtime": 189321.1195,
      "train_tokens_per_second": 4647.497
    },
    {
      "epoch": 1.8494897959183674,
      "grad_norm": 0.6266036999291648,
      "learning_rate": 1.3932214583738334e-07,
      "loss": 0.273012113571167,
      "num_input_tokens_seen": 880185176,
      "step": 14500,
      "train_runtime": 189386.1211,
      "train_tokens_per_second": 4647.57
    },
    {
      "epoch": 1.850127551020408,
      "grad_norm": 0.6474337326362705,
      "learning_rate": 1.3815039801161723e-07,
      "loss": 0.23094744682312013,
      "num_input_tokens_seen": 880486032,
      "step": 14505,
      "train_runtime": 189455.2359,
      "train_tokens_per_second": 4647.462
    },
    {
      "epoch": 1.850765306122449,
      "grad_norm": 0.6427624494514836,
      "learning_rate": 1.369835293914773e-07,
      "loss": 0.25781404972076416,
      "num_input_tokens_seen": 880787096,
      "step": 14510,
      "train_runtime": 189511.6112,
      "train_tokens_per_second": 4647.668
    },
    {
      "epoch": 1.8514030612244898,
      "grad_norm": 0.6415880011453653,
      "learning_rate": 1.358215411479974e-07,
      "loss": 0.2545320510864258,
      "num_input_tokens_seen": 881095592,
      "step": 14515,
      "train_runtime": 189587.434,
      "train_tokens_per_second": 4647.437
    },
    {
      "epoch": 1.8520408163265305,
      "grad_norm": 0.6401573982383106,
      "learning_rate": 1.3466443444731525e-07,
      "loss": 0.2468923568725586,
      "num_input_tokens_seen": 881389136,
      "step": 14520,
      "train_runtime": 189655.5039,
      "train_tokens_per_second": 4647.316
    },
    {
      "epoch": 1.8526785714285714,
      "grad_norm": 0.685404466407528,
      "learning_rate": 1.3351221045066698e-07,
      "loss": 0.25362067222595214,
      "num_input_tokens_seen": 881695144,
      "step": 14525,
      "train_runtime": 189721.7728,
      "train_tokens_per_second": 4647.306
    },
    {
      "epoch": 1.8533163265306123,
      "grad_norm": 0.6354264987157726,
      "learning_rate": 1.3236487031439037e-07,
      "loss": 0.2443603277206421,
      "num_input_tokens_seen": 882008248,
      "step": 14530,
      "train_runtime": 189787.2808,
      "train_tokens_per_second": 4647.352
    },
    {
      "epoch": 1.853954081632653,
      "grad_norm": 0.6704886175647126,
      "learning_rate": 1.3122241518992107e-07,
      "loss": 0.23978824615478517,
      "num_input_tokens_seen": 882314240,
      "step": 14535,
      "train_runtime": 189852.5231,
      "train_tokens_per_second": 4647.366
    },
    {
      "epoch": 1.8545918367346939,
      "grad_norm": 0.8442090166251053,
      "learning_rate": 1.3008484622379358e-07,
      "loss": 0.2655137300491333,
      "num_input_tokens_seen": 882616304,
      "step": 14540,
      "train_runtime": 189920.9843,
      "train_tokens_per_second": 4647.282
    },
    {
      "epoch": 1.8552295918367347,
      "grad_norm": 0.6764105086813371,
      "learning_rate": 1.2895216455763582e-07,
      "loss": 0.2473900079727173,
      "num_input_tokens_seen": 882923560,
      "step": 14545,
      "train_runtime": 189978.8193,
      "train_tokens_per_second": 4647.484
    },
    {
      "epoch": 1.8558673469387754,
      "grad_norm": 0.7022617311790411,
      "learning_rate": 1.278243713281757e-07,
      "loss": 0.2586209774017334,
      "num_input_tokens_seen": 883219664,
      "step": 14550,
      "train_runtime": 190056.7007,
      "train_tokens_per_second": 4647.138
    },
    {
      "epoch": 1.8565051020408163,
      "grad_norm": 0.5938136729185922,
      "learning_rate": 1.2670146766723013e-07,
      "loss": 0.234542179107666,
      "num_input_tokens_seen": 883530760,
      "step": 14555,
      "train_runtime": 190116.9986,
      "train_tokens_per_second": 4647.3
    },
    {
      "epoch": 1.8571428571428572,
      "grad_norm": 0.6633103907103941,
      "learning_rate": 1.2558345470171153e-07,
      "loss": 0.2434833526611328,
      "num_input_tokens_seen": 883836120,
      "step": 14560,
      "train_runtime": 190184.6109,
      "train_tokens_per_second": 4647.254
    },
    {
      "epoch": 1.8577806122448979,
      "grad_norm": 0.6651934099047041,
      "learning_rate": 1.244703335536257e-07,
      "loss": 0.2338038921356201,
      "num_input_tokens_seen": 884131064,
      "step": 14565,
      "train_runtime": 190246.0994,
      "train_tokens_per_second": 4647.302
    },
    {
      "epoch": 1.8584183673469388,
      "grad_norm": 0.6593021054152319,
      "learning_rate": 1.2336210534006576e-07,
      "loss": 0.254522705078125,
      "num_input_tokens_seen": 884427608,
      "step": 14570,
      "train_runtime": 190307.0874,
      "train_tokens_per_second": 4647.371
    },
    {
      "epoch": 1.8590561224489797,
      "grad_norm": 0.7006739894624194,
      "learning_rate": 1.2225877117321705e-07,
      "loss": 0.23550329208374024,
      "num_input_tokens_seen": 884738240,
      "step": 14575,
      "train_runtime": 190385.2489,
      "train_tokens_per_second": 4647.094
    },
    {
      "epoch": 1.8596938775510203,
      "grad_norm": 0.6381391847097326,
      "learning_rate": 1.2116033216035162e-07,
      "loss": 0.26369943618774416,
      "num_input_tokens_seen": 885046808,
      "step": 14580,
      "train_runtime": 190459.7978,
      "train_tokens_per_second": 4646.896
    },
    {
      "epoch": 1.8603316326530612,
      "grad_norm": 0.6705524383278435,
      "learning_rate": 1.20066789403831e-07,
      "loss": 0.26955347061157225,
      "num_input_tokens_seen": 885351872,
      "step": 14585,
      "train_runtime": 190517.1238,
      "train_tokens_per_second": 4647.099
    },
    {
      "epoch": 1.8609693877551021,
      "grad_norm": 0.6174540318052323,
      "learning_rate": 1.1897814400110064e-07,
      "loss": 0.24823312759399413,
      "num_input_tokens_seen": 885663712,
      "step": 14590,
      "train_runtime": 190575.0922,
      "train_tokens_per_second": 4647.321
    },
    {
      "epoch": 1.8616071428571428,
      "grad_norm": 0.6761567719825483,
      "learning_rate": 1.1789439704469219e-07,
      "loss": 0.2636047124862671,
      "num_input_tokens_seen": 885960504,
      "step": 14595,
      "train_runtime": 190639.4012,
      "train_tokens_per_second": 4647.311
    },
    {
      "epoch": 1.8622448979591837,
      "grad_norm": 0.6309923322609609,
      "learning_rate": 1.168155496222223e-07,
      "loss": 0.2618051290512085,
      "num_input_tokens_seen": 886267272,
      "step": 14600,
      "train_runtime": 190707.3016,
      "train_tokens_per_second": 4647.264
    },
    {
      "epoch": 1.8628826530612246,
      "grad_norm": 0.625295239253637,
      "learning_rate": 1.1574160281638935e-07,
      "loss": 0.2571128845214844,
      "num_input_tokens_seen": 886563408,
      "step": 14605,
      "train_runtime": 190762.3627,
      "train_tokens_per_second": 4647.476
    },
    {
      "epoch": 1.8635204081632653,
      "grad_norm": 0.6617635960001381,
      "learning_rate": 1.1467255770497343e-07,
      "loss": 0.24137578010559083,
      "num_input_tokens_seen": 886868720,
      "step": 14610,
      "train_runtime": 190831.106,
      "train_tokens_per_second": 4647.401
    },
    {
      "epoch": 1.8641581632653061,
      "grad_norm": 0.6745362192358028,
      "learning_rate": 1.1360841536083633e-07,
      "loss": 0.23988428115844726,
      "num_input_tokens_seen": 887181184,
      "step": 14615,
      "train_runtime": 190888.7338,
      "train_tokens_per_second": 4647.635
    },
    {
      "epoch": 1.864795918367347,
      "grad_norm": 0.6825878261964686,
      "learning_rate": 1.125491768519199e-07,
      "loss": 0.2605908393859863,
      "num_input_tokens_seen": 887479224,
      "step": 14620,
      "train_runtime": 190966.4323,
      "train_tokens_per_second": 4647.305
    },
    {
      "epoch": 1.8654336734693877,
      "grad_norm": 0.645239604402163,
      "learning_rate": 1.1149484324124326e-07,
      "loss": 0.25436906814575194,
      "num_input_tokens_seen": 887799912,
      "step": 14625,
      "train_runtime": 191035.1537,
      "train_tokens_per_second": 4647.312
    },
    {
      "epoch": 1.8660714285714286,
      "grad_norm": 0.7403347112064289,
      "learning_rate": 1.1044541558690281e-07,
      "loss": 0.25912206172943114,
      "num_input_tokens_seen": 888088832,
      "step": 14630,
      "train_runtime": 191092.3326,
      "train_tokens_per_second": 4647.433
    },
    {
      "epoch": 1.8667091836734695,
      "grad_norm": 0.7230370391364567,
      "learning_rate": 1.094008949420744e-07,
      "loss": 0.22946701049804688,
      "num_input_tokens_seen": 888399576,
      "step": 14635,
      "train_runtime": 191157.0422,
      "train_tokens_per_second": 4647.485
    },
    {
      "epoch": 1.8673469387755102,
      "grad_norm": 0.6907678499682646,
      "learning_rate": 1.0836128235500676e-07,
      "loss": 0.23314986228942872,
      "num_input_tokens_seen": 888707984,
      "step": 14640,
      "train_runtime": 191216.5543,
      "train_tokens_per_second": 4647.652
    },
    {
      "epoch": 1.867984693877551,
      "grad_norm": 0.6631760247916414,
      "learning_rate": 1.0732657886902308e-07,
      "loss": 0.2371466875076294,
      "num_input_tokens_seen": 889023480,
      "step": 14645,
      "train_runtime": 191281.7976,
      "train_tokens_per_second": 4647.716
    },
    {
      "epoch": 1.868622448979592,
      "grad_norm": 0.6625549576654546,
      "learning_rate": 1.0629678552252109e-07,
      "loss": 0.25177536010742185,
      "num_input_tokens_seen": 889315784,
      "step": 14650,
      "train_runtime": 191350.9212,
      "train_tokens_per_second": 4647.565
    },
    {
      "epoch": 1.8692602040816326,
      "grad_norm": 0.6557793654429599,
      "learning_rate": 1.0527190334897076e-07,
      "loss": 0.26426329612731936,
      "num_input_tokens_seen": 889620424,
      "step": 14655,
      "train_runtime": 191416.8843,
      "train_tokens_per_second": 4647.555
    },
    {
      "epoch": 1.8698979591836735,
      "grad_norm": 0.6724420927989058,
      "learning_rate": 1.0425193337691208e-07,
      "loss": 0.2512818813323975,
      "num_input_tokens_seen": 889919392,
      "step": 14660,
      "train_runtime": 191481.7376,
      "train_tokens_per_second": 4647.542
    },
    {
      "epoch": 1.8705357142857144,
      "grad_norm": 0.6773586930650444,
      "learning_rate": 1.0323687662995685e-07,
      "loss": 0.271722412109375,
      "num_input_tokens_seen": 890229504,
      "step": 14665,
      "train_runtime": 191548.2801,
      "train_tokens_per_second": 4647.546
    },
    {
      "epoch": 1.871173469387755,
      "grad_norm": 0.62826592172563,
      "learning_rate": 1.0222673412678518e-07,
      "loss": 0.23242733478546143,
      "num_input_tokens_seen": 890534280,
      "step": 14670,
      "train_runtime": 191616.1162,
      "train_tokens_per_second": 4647.492
    },
    {
      "epoch": 1.8718112244897958,
      "grad_norm": 0.6378366560239725,
      "learning_rate": 1.0122150688114507e-07,
      "loss": 0.25556354522705077,
      "num_input_tokens_seen": 890838848,
      "step": 14675,
      "train_runtime": 191678.2331,
      "train_tokens_per_second": 4647.574
    },
    {
      "epoch": 1.8724489795918369,
      "grad_norm": 0.6153509806232496,
      "learning_rate": 1.0022119590185287e-07,
      "loss": 0.25017662048339845,
      "num_input_tokens_seen": 891149616,
      "step": 14680,
      "train_runtime": 191746.3293,
      "train_tokens_per_second": 4647.545
    },
    {
      "epoch": 1.8730867346938775,
      "grad_norm": 0.671415910331799,
      "learning_rate": 9.922580219279055e-08,
      "loss": 0.25757060050964353,
      "num_input_tokens_seen": 891449720,
      "step": 14685,
      "train_runtime": 191823.8479,
      "train_tokens_per_second": 4647.231
    },
    {
      "epoch": 1.8737244897959182,
      "grad_norm": 0.7137048685934427,
      "learning_rate": 9.82353267529046e-08,
      "loss": 0.2548252582550049,
      "num_input_tokens_seen": 891747784,
      "step": 14690,
      "train_runtime": 191884.1037,
      "train_tokens_per_second": 4647.325
    },
    {
      "epoch": 1.8743622448979593,
      "grad_norm": 0.6596229862745624,
      "learning_rate": 9.72497705762071e-08,
      "loss": 0.2520712375640869,
      "num_input_tokens_seen": 892058120,
      "step": 14695,
      "train_runtime": 191947.7977,
      "train_tokens_per_second": 4647.4
    },
    {
      "epoch": 1.875,
      "grad_norm": 0.6780638735375085,
      "learning_rate": 9.626913465177246e-08,
      "loss": 0.26645793914794924,
      "num_input_tokens_seen": 892363192,
      "step": 14700,
      "train_runtime": 192009.6554,
      "train_tokens_per_second": 4647.491
    },
    {
      "epoch": 1.8756377551020407,
      "grad_norm": 0.6405694736902278,
      "learning_rate": 9.529341996373675e-08,
      "loss": 0.2566962718963623,
      "num_input_tokens_seen": 892669392,
      "step": 14705,
      "train_runtime": 192082.3114,
      "train_tokens_per_second": 4647.327
    },
    {
      "epoch": 1.8762755102040818,
      "grad_norm": 0.6509202472516895,
      "learning_rate": 9.432262749129784e-08,
      "loss": 0.24963088035583497,
      "num_input_tokens_seen": 892968720,
      "step": 14710,
      "train_runtime": 192156.2238,
      "train_tokens_per_second": 4647.098
    },
    {
      "epoch": 1.8769132653061225,
      "grad_norm": 0.778756285306421,
      "learning_rate": 9.335675820871415e-08,
      "loss": 0.24836084842681885,
      "num_input_tokens_seen": 893280344,
      "step": 14715,
      "train_runtime": 192216.0713,
      "train_tokens_per_second": 4647.272
    },
    {
      "epoch": 1.8775510204081631,
      "grad_norm": 0.7139360855592187,
      "learning_rate": 9.239581308530254e-08,
      "loss": 0.27693932056427,
      "num_input_tokens_seen": 893587816,
      "step": 14720,
      "train_runtime": 192280.2293,
      "train_tokens_per_second": 4647.32
    },
    {
      "epoch": 1.8781887755102042,
      "grad_norm": 0.9524393643486794,
      "learning_rate": 9.143979308543993e-08,
      "loss": 0.2538874626159668,
      "num_input_tokens_seen": 893900112,
      "step": 14725,
      "train_runtime": 192344.8959,
      "train_tokens_per_second": 4647.382
    },
    {
      "epoch": 1.878826530612245,
      "grad_norm": 0.6972297812402133,
      "learning_rate": 9.048869916855662e-08,
      "loss": 0.2388587236404419,
      "num_input_tokens_seen": 894202528,
      "step": 14730,
      "train_runtime": 192410.9123,
      "train_tokens_per_second": 4647.359
    },
    {
      "epoch": 1.8794642857142856,
      "grad_norm": 0.7200074370892064,
      "learning_rate": 8.954253228914356e-08,
      "loss": 0.24752345085144042,
      "num_input_tokens_seen": 894517032,
      "step": 14735,
      "train_runtime": 192485.7653,
      "train_tokens_per_second": 4647.185
    },
    {
      "epoch": 1.8801020408163265,
      "grad_norm": 0.7667717526084538,
      "learning_rate": 8.860129339674562e-08,
      "loss": 0.2614644289016724,
      "num_input_tokens_seen": 894810904,
      "step": 14740,
      "train_runtime": 192545.6707,
      "train_tokens_per_second": 4647.266
    },
    {
      "epoch": 1.8807397959183674,
      "grad_norm": 0.610970003049497,
      "learning_rate": 8.766498343596053e-08,
      "loss": 0.23589935302734374,
      "num_input_tokens_seen": 895101344,
      "step": 14745,
      "train_runtime": 192613.469,
      "train_tokens_per_second": 4647.138
    },
    {
      "epoch": 1.881377551020408,
      "grad_norm": 0.6752224221115799,
      "learning_rate": 8.67336033464411e-08,
      "loss": 0.2402803421020508,
      "num_input_tokens_seen": 895400528,
      "step": 14750,
      "train_runtime": 192686.7509,
      "train_tokens_per_second": 4646.923
    },
    {
      "epoch": 1.882015306122449,
      "grad_norm": 0.731006217203887,
      "learning_rate": 8.580715406289353e-08,
      "loss": 0.23293819427490234,
      "num_input_tokens_seen": 895706304,
      "step": 14755,
      "train_runtime": 192763.9837,
      "train_tokens_per_second": 4646.648
    },
    {
      "epoch": 1.8826530612244898,
      "grad_norm": 0.6650984421358715,
      "learning_rate": 8.488563651507242e-08,
      "loss": 0.2630185127258301,
      "num_input_tokens_seen": 896006184,
      "step": 14760,
      "train_runtime": 192829.1678,
      "train_tokens_per_second": 4646.632
    },
    {
      "epoch": 1.8832908163265305,
      "grad_norm": 0.6788993548832946,
      "learning_rate": 8.396905162778523e-08,
      "loss": 0.26844196319580077,
      "num_input_tokens_seen": 896313528,
      "step": 14765,
      "train_runtime": 192891.096,
      "train_tokens_per_second": 4646.734
    },
    {
      "epoch": 1.8839285714285714,
      "grad_norm": 0.6692467737558845,
      "learning_rate": 8.305740032089116e-08,
      "loss": 0.2632472991943359,
      "num_input_tokens_seen": 896619432,
      "step": 14770,
      "train_runtime": 192960.9219,
      "train_tokens_per_second": 4646.637
    },
    {
      "epoch": 1.8845663265306123,
      "grad_norm": 0.5692059577633254,
      "learning_rate": 8.215068350929334e-08,
      "loss": 0.25128650665283203,
      "num_input_tokens_seen": 896916752,
      "step": 14775,
      "train_runtime": 193037.4187,
      "train_tokens_per_second": 4646.336
    },
    {
      "epoch": 1.885204081632653,
      "grad_norm": 0.6830063880564934,
      "learning_rate": 8.124890210294667e-08,
      "loss": 0.24625537395477295,
      "num_input_tokens_seen": 897221008,
      "step": 14780,
      "train_runtime": 193101.1866,
      "train_tokens_per_second": 4646.377
    },
    {
      "epoch": 1.8858418367346939,
      "grad_norm": 0.724929616005513,
      "learning_rate": 8.035205700685167e-08,
      "loss": 0.2617891788482666,
      "num_input_tokens_seen": 897511264,
      "step": 14785,
      "train_runtime": 193167.4357,
      "train_tokens_per_second": 4646.287
    },
    {
      "epoch": 1.8864795918367347,
      "grad_norm": 0.6710932872068884,
      "learning_rate": 7.946014912105505e-08,
      "loss": 0.26091532707214354,
      "num_input_tokens_seen": 897807720,
      "step": 14790,
      "train_runtime": 193248.1877,
      "train_tokens_per_second": 4645.879
    },
    {
      "epoch": 1.8871173469387754,
      "grad_norm": 0.7016151066427102,
      "learning_rate": 7.857317934064857e-08,
      "loss": 0.2553201198577881,
      "num_input_tokens_seen": 898117528,
      "step": 14795,
      "train_runtime": 193317.5078,
      "train_tokens_per_second": 4645.816
    },
    {
      "epoch": 1.8877551020408163,
      "grad_norm": 0.6811844967660088,
      "learning_rate": 7.769114855576799e-08,
      "loss": 0.26304941177368163,
      "num_input_tokens_seen": 898417680,
      "step": 14800,
      "train_runtime": 193387.6582,
      "train_tokens_per_second": 4645.683
    },
    {
      "epoch": 1.8883928571428572,
      "grad_norm": 0.8877659045871802,
      "learning_rate": 7.6814057651593e-08,
      "loss": 0.25156078338623045,
      "num_input_tokens_seen": 898714720,
      "step": 14805,
      "train_runtime": 193464.5556,
      "train_tokens_per_second": 4645.371
    },
    {
      "epoch": 1.8890306122448979,
      "grad_norm": 0.6594008814729282,
      "learning_rate": 7.594190750834617e-08,
      "loss": 0.24910693168640136,
      "num_input_tokens_seen": 899011592,
      "step": 14810,
      "train_runtime": 193538.5333,
      "train_tokens_per_second": 4645.13
    },
    {
      "epoch": 1.8896683673469388,
      "grad_norm": 0.6314110620163867,
      "learning_rate": 7.507469900129016e-08,
      "loss": 0.25336446762084963,
      "num_input_tokens_seen": 899318184,
      "step": 14815,
      "train_runtime": 193593.9739,
      "train_tokens_per_second": 4645.383
    },
    {
      "epoch": 1.8903061224489797,
      "grad_norm": 0.6638875288172695,
      "learning_rate": 7.421243300072989e-08,
      "loss": 0.23741183280944825,
      "num_input_tokens_seen": 899618656,
      "step": 14820,
      "train_runtime": 193650.562,
      "train_tokens_per_second": 4645.577
    },
    {
      "epoch": 1.8909438775510203,
      "grad_norm": 0.6194066197151755,
      "learning_rate": 7.335511037200982e-08,
      "loss": 0.2510697841644287,
      "num_input_tokens_seen": 899924056,
      "step": 14825,
      "train_runtime": 193703.02,
      "train_tokens_per_second": 4645.896
    },
    {
      "epoch": 1.8915816326530612,
      "grad_norm": 0.6089635763799623,
      "learning_rate": 7.250273197551227e-08,
      "loss": 0.23379080295562743,
      "num_input_tokens_seen": 900226328,
      "step": 14830,
      "train_runtime": 193777.3724,
      "train_tokens_per_second": 4645.673
    },
    {
      "epoch": 1.8922193877551021,
      "grad_norm": 0.7337019039784103,
      "learning_rate": 7.165529866665855e-08,
      "loss": 0.23528714179992677,
      "num_input_tokens_seen": 900513536,
      "step": 14835,
      "train_runtime": 193837.193,
      "train_tokens_per_second": 4645.721
    },
    {
      "epoch": 1.8928571428571428,
      "grad_norm": 0.6454351181533083,
      "learning_rate": 7.081281129590777e-08,
      "loss": 0.2547890663146973,
      "num_input_tokens_seen": 900821760,
      "step": 14840,
      "train_runtime": 193909.6783,
      "train_tokens_per_second": 4645.574
    },
    {
      "epoch": 1.8934948979591837,
      "grad_norm": 0.7226144777369936,
      "learning_rate": 6.99752707087542e-08,
      "loss": 0.26011736392974855,
      "num_input_tokens_seen": 901126832,
      "step": 14845,
      "train_runtime": 193972.6126,
      "train_tokens_per_second": 4645.64
    },
    {
      "epoch": 1.8941326530612246,
      "grad_norm": 0.5872104501936667,
      "learning_rate": 6.91426777457288e-08,
      "loss": 0.25217404365539553,
      "num_input_tokens_seen": 901424120,
      "step": 14850,
      "train_runtime": 194044.4187,
      "train_tokens_per_second": 4645.452
    },
    {
      "epoch": 1.8947704081632653,
      "grad_norm": 0.7711445799170084,
      "learning_rate": 6.831503324239707e-08,
      "loss": 0.24703493118286132,
      "num_input_tokens_seen": 901724328,
      "step": 14855,
      "train_runtime": 194110.0007,
      "train_tokens_per_second": 4645.43
    },
    {
      "epoch": 1.8954081632653061,
      "grad_norm": 0.7023465986124341,
      "learning_rate": 6.749233802935684e-08,
      "loss": 0.2439436674118042,
      "num_input_tokens_seen": 902025920,
      "step": 14860,
      "train_runtime": 194169.7617,
      "train_tokens_per_second": 4645.553
    },
    {
      "epoch": 1.896045918367347,
      "grad_norm": 0.6630922976778474,
      "learning_rate": 6.667459293224155e-08,
      "loss": 0.2603579044342041,
      "num_input_tokens_seen": 902344136,
      "step": 14865,
      "train_runtime": 194218.4433,
      "train_tokens_per_second": 4646.027
    },
    {
      "epoch": 1.8966836734693877,
      "grad_norm": 0.673900051723132,
      "learning_rate": 6.586179877171473e-08,
      "loss": 0.24415109157562256,
      "num_input_tokens_seen": 902653128,
      "step": 14870,
      "train_runtime": 194274.2905,
      "train_tokens_per_second": 4646.282
    },
    {
      "epoch": 1.8973214285714286,
      "grad_norm": 0.6509213291700938,
      "learning_rate": 6.505395636347222e-08,
      "loss": 0.24298529624938964,
      "num_input_tokens_seen": 902969696,
      "step": 14875,
      "train_runtime": 194339.7776,
      "train_tokens_per_second": 4646.345
    },
    {
      "epoch": 1.8979591836734695,
      "grad_norm": 0.5752364867057913,
      "learning_rate": 6.425106651824054e-08,
      "loss": 0.24060254096984862,
      "num_input_tokens_seen": 903275896,
      "step": 14880,
      "train_runtime": 194400.783,
      "train_tokens_per_second": 4646.462
    },
    {
      "epoch": 1.8985969387755102,
      "grad_norm": 0.6258477215576815,
      "learning_rate": 6.345313004177511e-08,
      "loss": 0.23076014518737792,
      "num_input_tokens_seen": 903589936,
      "step": 14885,
      "train_runtime": 194467.4704,
      "train_tokens_per_second": 4646.484
    },
    {
      "epoch": 1.899234693877551,
      "grad_norm": 0.5958702299030119,
      "learning_rate": 6.266014773486207e-08,
      "loss": 0.23166513442993164,
      "num_input_tokens_seen": 903899464,
      "step": 14890,
      "train_runtime": 194532.9383,
      "train_tokens_per_second": 4646.511
    },
    {
      "epoch": 1.899872448979592,
      "grad_norm": 0.6304073581468033,
      "learning_rate": 6.187212039331314e-08,
      "loss": 0.24292712211608886,
      "num_input_tokens_seen": 904199632,
      "step": 14895,
      "train_runtime": 194597.5108,
      "train_tokens_per_second": 4646.512
    },
    {
      "epoch": 1.9005102040816326,
      "grad_norm": 0.6838106741225728,
      "learning_rate": 6.108904880797017e-08,
      "loss": 0.24226779937744142,
      "num_input_tokens_seen": 904499744,
      "step": 14900,
      "train_runtime": 194667.7083,
      "train_tokens_per_second": 4646.378
    },
    {
      "epoch": 1.9011479591836735,
      "grad_norm": 0.679098533124189,
      "learning_rate": 6.031093376469899e-08,
      "loss": 0.2677560091018677,
      "num_input_tokens_seen": 904810904,
      "step": 14905,
      "train_runtime": 194734.9704,
      "train_tokens_per_second": 4646.371
    },
    {
      "epoch": 1.9017857142857144,
      "grad_norm": 0.7047378809168635,
      "learning_rate": 5.9537776044393256e-08,
      "loss": 0.2479987621307373,
      "num_input_tokens_seen": 905119168,
      "step": 14910,
      "train_runtime": 194803.6285,
      "train_tokens_per_second": 4646.316
    },
    {
      "epoch": 1.902423469387755,
      "grad_norm": 0.5684364054960253,
      "learning_rate": 5.87695764229701e-08,
      "loss": 0.23430242538452148,
      "num_input_tokens_seen": 905425624,
      "step": 14915,
      "train_runtime": 194872.5299,
      "train_tokens_per_second": 4646.246
    },
    {
      "epoch": 1.9030612244897958,
      "grad_norm": 0.7559823323485224,
      "learning_rate": 5.80063356713717e-08,
      "loss": 0.26269304752349854,
      "num_input_tokens_seen": 905726560,
      "step": 14920,
      "train_runtime": 194930.7129,
      "train_tokens_per_second": 4646.403
    },
    {
      "epoch": 1.9036989795918369,
      "grad_norm": 0.6822043816304155,
      "learning_rate": 5.7248054555563704e-08,
      "loss": 0.2633235931396484,
      "num_input_tokens_seen": 906035824,
      "step": 14925,
      "train_runtime": 194986.1528,
      "train_tokens_per_second": 4646.668
    },
    {
      "epoch": 1.9043367346938775,
      "grad_norm": 0.6632641780306863,
      "learning_rate": 5.6494733836534035e-08,
      "loss": 0.24860916137695313,
      "num_input_tokens_seen": 906346096,
      "step": 14930,
      "train_runtime": 195049.3716,
      "train_tokens_per_second": 4646.752
    },
    {
      "epoch": 1.9049744897959182,
      "grad_norm": 0.6258973686369059,
      "learning_rate": 5.574637427029239e-08,
      "loss": 0.24390010833740233,
      "num_input_tokens_seen": 906648504,
      "step": 14935,
      "train_runtime": 195112.3756,
      "train_tokens_per_second": 4646.802
    },
    {
      "epoch": 1.9056122448979593,
      "grad_norm": 0.760002254228921,
      "learning_rate": 5.50029766078708e-08,
      "loss": 0.26183409690856935,
      "num_input_tokens_seen": 906951760,
      "step": 14940,
      "train_runtime": 195181.55,
      "train_tokens_per_second": 4646.708
    },
    {
      "epoch": 1.90625,
      "grad_norm": 0.6095984450226043,
      "learning_rate": 5.426454159531913e-08,
      "loss": 0.25227718353271483,
      "num_input_tokens_seen": 907262016,
      "step": 14945,
      "train_runtime": 195251.5453,
      "train_tokens_per_second": 4646.632
    },
    {
      "epoch": 1.9068877551020407,
      "grad_norm": 0.5567129695809843,
      "learning_rate": 5.3531069973709607e-08,
      "loss": 0.2548694610595703,
      "num_input_tokens_seen": 907561240,
      "step": 14950,
      "train_runtime": 195320.9545,
      "train_tokens_per_second": 4646.512
    },
    {
      "epoch": 1.9075255102040818,
      "grad_norm": 0.6926951020902639,
      "learning_rate": 5.280256247913229e-08,
      "loss": 0.26091346740722654,
      "num_input_tokens_seen": 907859912,
      "step": 14955,
      "train_runtime": 195392.9353,
      "train_tokens_per_second": 4646.329
    },
    {
      "epoch": 1.9081632653061225,
      "grad_norm": 0.6581359098639753,
      "learning_rate": 5.207901984269459e-08,
      "loss": 0.24626507759094238,
      "num_input_tokens_seen": 908156416,
      "step": 14960,
      "train_runtime": 195455.9186,
      "train_tokens_per_second": 4646.349
    },
    {
      "epoch": 1.9088010204081631,
      "grad_norm": 0.6771379588752399,
      "learning_rate": 5.136044279052288e-08,
      "loss": 0.260360860824585,
      "num_input_tokens_seen": 908460312,
      "step": 14965,
      "train_runtime": 195526.4192,
      "train_tokens_per_second": 4646.228
    },
    {
      "epoch": 1.9094387755102042,
      "grad_norm": 0.6564757017128225,
      "learning_rate": 5.0646832043758645e-08,
      "loss": 0.23393001556396484,
      "num_input_tokens_seen": 908759856,
      "step": 14970,
      "train_runtime": 195581.843,
      "train_tokens_per_second": 4646.443
    },
    {
      "epoch": 1.910076530612245,
      "grad_norm": 0.6505722919038079,
      "learning_rate": 4.993818831856123e-08,
      "loss": 0.25123028755187987,
      "num_input_tokens_seen": 909059128,
      "step": 14975,
      "train_runtime": 195651.7876,
      "train_tokens_per_second": 4646.311
    },
    {
      "epoch": 1.9107142857142856,
      "grad_norm": 0.6862658423869104,
      "learning_rate": 4.923451232610288e-08,
      "loss": 0.2689199686050415,
      "num_input_tokens_seen": 909368704,
      "step": 14980,
      "train_runtime": 195709.3519,
      "train_tokens_per_second": 4646.527
    },
    {
      "epoch": 1.9113520408163265,
      "grad_norm": 0.6118149479886968,
      "learning_rate": 4.8535804772572024e-08,
      "loss": 0.24207632541656493,
      "num_input_tokens_seen": 909668768,
      "step": 14985,
      "train_runtime": 195771.7263,
      "train_tokens_per_second": 4646.579
    },
    {
      "epoch": 1.9119897959183674,
      "grad_norm": 0.7068122232336729,
      "learning_rate": 4.7842066359170544e-08,
      "loss": 0.26170997619628905,
      "num_input_tokens_seen": 909982280,
      "step": 14990,
      "train_runtime": 195831.4708,
      "train_tokens_per_second": 4646.762
    },
    {
      "epoch": 1.912627551020408,
      "grad_norm": 0.6311868431963777,
      "learning_rate": 4.715329778211375e-08,
      "loss": 0.24396302700042724,
      "num_input_tokens_seen": 910291280,
      "step": 14995,
      "train_runtime": 195892.0501,
      "train_tokens_per_second": 4646.903
    },
    {
      "epoch": 1.913265306122449,
      "grad_norm": 0.6618045098131926,
      "learning_rate": 4.646949973262871e-08,
      "loss": 0.241226863861084,
      "num_input_tokens_seen": 910597016,
      "step": 15000,
      "train_runtime": 195954.5226,
      "train_tokens_per_second": 4646.981
    },
    {
      "epoch": 1.9139030612244898,
      "grad_norm": 0.6936883965561665,
      "learning_rate": 4.579067289695427e-08,
      "loss": 0.25010688304901124,
      "num_input_tokens_seen": 910889416,
      "step": 15005,
      "train_runtime": 196020.2285,
      "train_tokens_per_second": 4646.915
    },
    {
      "epoch": 1.9145408163265305,
      "grad_norm": 0.7035237706792213,
      "learning_rate": 4.511681795634049e-08,
      "loss": 0.24684762954711914,
      "num_input_tokens_seen": 911194392,
      "step": 15010,
      "train_runtime": 196090.9804,
      "train_tokens_per_second": 4646.794
    },
    {
      "epoch": 1.9151785714285714,
      "grad_norm": 0.6322354753921676,
      "learning_rate": 4.4447935587049185e-08,
      "loss": 0.24867401123046876,
      "num_input_tokens_seen": 911506240,
      "step": 15015,
      "train_runtime": 196159.3016,
      "train_tokens_per_second": 4646.765
    },
    {
      "epoch": 1.9158163265306123,
      "grad_norm": 0.6199826293536299,
      "learning_rate": 4.37840264603484e-08,
      "loss": 0.24369287490844727,
      "num_input_tokens_seen": 911812040,
      "step": 15020,
      "train_runtime": 196220.2107,
      "train_tokens_per_second": 4646.881
    },
    {
      "epoch": 1.916454081632653,
      "grad_norm": 0.6283653539917077,
      "learning_rate": 4.312509124251907e-08,
      "loss": 0.26245894432067873,
      "num_input_tokens_seen": 912118696,
      "step": 15025,
      "train_runtime": 196285.1403,
      "train_tokens_per_second": 4646.907
    },
    {
      "epoch": 1.9170918367346939,
      "grad_norm": 0.6365747752856089,
      "learning_rate": 4.247113059484775e-08,
      "loss": 0.24296834468841552,
      "num_input_tokens_seen": 912411904,
      "step": 15030,
      "train_runtime": 196359.8021,
      "train_tokens_per_second": 4646.633
    },
    {
      "epoch": 1.9177295918367347,
      "grad_norm": 0.6921456195408252,
      "learning_rate": 4.1822145173630033e-08,
      "loss": 0.23697316646575928,
      "num_input_tokens_seen": 912720144,
      "step": 15035,
      "train_runtime": 196425.7606,
      "train_tokens_per_second": 4646.642
    },
    {
      "epoch": 1.9183673469387754,
      "grad_norm": 0.6103621580852481,
      "learning_rate": 4.117813563016826e-08,
      "loss": 0.24823360443115233,
      "num_input_tokens_seen": 913034744,
      "step": 15040,
      "train_runtime": 196481.0788,
      "train_tokens_per_second": 4646.935
    },
    {
      "epoch": 1.9190051020408163,
      "grad_norm": 0.6223000336676402,
      "learning_rate": 4.0539102610770434e-08,
      "loss": 0.2661388397216797,
      "num_input_tokens_seen": 913344128,
      "step": 15045,
      "train_runtime": 196537.9916,
      "train_tokens_per_second": 4647.163
    },
    {
      "epoch": 1.9196428571428572,
      "grad_norm": 0.6176804790718705,
      "learning_rate": 3.990504675675133e-08,
      "loss": 0.24283137321472167,
      "num_input_tokens_seen": 913656280,
      "step": 15050,
      "train_runtime": 196599.1813,
      "train_tokens_per_second": 4647.305
    },
    {
      "epoch": 1.9202806122448979,
      "grad_norm": 0.692932533981291,
      "learning_rate": 3.927596870442973e-08,
      "loss": 0.2644121885299683,
      "num_input_tokens_seen": 913958416,
      "step": 15055,
      "train_runtime": 196655.93,
      "train_tokens_per_second": 4647.5
    },
    {
      "epoch": 1.9209183673469388,
      "grad_norm": 0.7009069930702748,
      "learning_rate": 3.8651869085130076e-08,
      "loss": 0.24419598579406737,
      "num_input_tokens_seen": 914260256,
      "step": 15060,
      "train_runtime": 196729.9289,
      "train_tokens_per_second": 4647.286
    },
    {
      "epoch": 1.9215561224489797,
      "grad_norm": 0.6427188050500241,
      "learning_rate": 3.8032748525179684e-08,
      "loss": 0.23569097518920898,
      "num_input_tokens_seen": 914553184,
      "step": 15065,
      "train_runtime": 196806.1105,
      "train_tokens_per_second": 4646.976
    },
    {
      "epoch": 1.9221938775510203,
      "grad_norm": 0.6405735410167481,
      "learning_rate": 3.741860764590877e-08,
      "loss": 0.24265587329864502,
      "num_input_tokens_seen": 914855288,
      "step": 15070,
      "train_runtime": 196880.4543,
      "train_tokens_per_second": 4646.755
    },
    {
      "epoch": 1.9228316326530612,
      "grad_norm": 0.6170910879559761,
      "learning_rate": 3.680944706365097e-08,
      "loss": 0.25319323539733884,
      "num_input_tokens_seen": 915159176,
      "step": 15075,
      "train_runtime": 196951.652,
      "train_tokens_per_second": 4646.618
    },
    {
      "epoch": 1.9234693877551021,
      "grad_norm": 0.6047867279145934,
      "learning_rate": 3.6205267389741724e-08,
      "loss": 0.23566744327545167,
      "num_input_tokens_seen": 915473424,
      "step": 15080,
      "train_runtime": 197013.4256,
      "train_tokens_per_second": 4646.757
    },
    {
      "epoch": 1.9241071428571428,
      "grad_norm": 0.643163464215507,
      "learning_rate": 3.560606923051768e-08,
      "loss": 0.23976473808288573,
      "num_input_tokens_seen": 915764344,
      "step": 15085,
      "train_runtime": 197083.9997,
      "train_tokens_per_second": 4646.569
    },
    {
      "epoch": 1.9247448979591837,
      "grad_norm": 0.6942785035430976,
      "learning_rate": 3.5011853187315035e-08,
      "loss": 0.23405833244323732,
      "num_input_tokens_seen": 916061512,
      "step": 15090,
      "train_runtime": 197148.4158,
      "train_tokens_per_second": 4646.558
    },
    {
      "epoch": 1.9253826530612246,
      "grad_norm": 0.6309596070222945,
      "learning_rate": 3.442261985647177e-08,
      "loss": 0.25694937705993653,
      "num_input_tokens_seen": 916381160,
      "step": 15095,
      "train_runtime": 197225.9566,
      "train_tokens_per_second": 4646.352
    },
    {
      "epoch": 1.9260204081632653,
      "grad_norm": 0.6690711283204608,
      "learning_rate": 3.383836982932542e-08,
      "loss": 0.2462223768234253,
      "num_input_tokens_seen": 916678328,
      "step": 15100,
      "train_runtime": 197291.473,
      "train_tokens_per_second": 4646.315
    },
    {
      "epoch": 1.9266581632653061,
      "grad_norm": 0.6414088969806848,
      "learning_rate": 3.325910369220975e-08,
      "loss": 0.25452375411987305,
      "num_input_tokens_seen": 916992400,
      "step": 15105,
      "train_runtime": 197346.9904,
      "train_tokens_per_second": 4646.599
    },
    {
      "epoch": 1.927295918367347,
      "grad_norm": 0.6832228334139179,
      "learning_rate": 3.268482202646084e-08,
      "loss": 0.2523342609405518,
      "num_input_tokens_seen": 917292000,
      "step": 15110,
      "train_runtime": 197411.991,
      "train_tokens_per_second": 4646.587
    },
    {
      "epoch": 1.9279336734693877,
      "grad_norm": 0.6914368072719113,
      "learning_rate": 3.211552540840934e-08,
      "loss": 0.26769270896911623,
      "num_input_tokens_seen": 917597016,
      "step": 15115,
      "train_runtime": 197482.0316,
      "train_tokens_per_second": 4646.484
    },
    {
      "epoch": 1.9285714285714286,
      "grad_norm": 0.7639182198577933,
      "learning_rate": 3.1551214409384355e-08,
      "loss": 0.23019323348999024,
      "num_input_tokens_seen": 917908152,
      "step": 15120,
      "train_runtime": 197558.7458,
      "train_tokens_per_second": 4646.254
    },
    {
      "epoch": 1.9292091836734695,
      "grad_norm": 0.661951476734585,
      "learning_rate": 3.09918895957112e-08,
      "loss": 0.2444523811340332,
      "num_input_tokens_seen": 918217408,
      "step": 15125,
      "train_runtime": 197626.8518,
      "train_tokens_per_second": 4646.218
    },
    {
      "epoch": 1.9298469387755102,
      "grad_norm": 0.7461359982111557,
      "learning_rate": 3.0437551528711974e-08,
      "loss": 0.2376544237136841,
      "num_input_tokens_seen": 918506528,
      "step": 15130,
      "train_runtime": 197693.0533,
      "train_tokens_per_second": 4646.124
    },
    {
      "epoch": 1.930484693877551,
      "grad_norm": 0.59542529064351,
      "learning_rate": 2.988820076470278e-08,
      "loss": 0.25788092613220215,
      "num_input_tokens_seen": 918805920,
      "step": 15135,
      "train_runtime": 197765.548,
      "train_tokens_per_second": 4645.935
    },
    {
      "epoch": 1.931122448979592,
      "grad_norm": 0.5968634469164296,
      "learning_rate": 2.934383785499595e-08,
      "loss": 0.23025188446044922,
      "num_input_tokens_seen": 919113720,
      "step": 15140,
      "train_runtime": 197818.2006,
      "train_tokens_per_second": 4646.255
    },
    {
      "epoch": 1.9317602040816326,
      "grad_norm": 0.7298214145553429,
      "learning_rate": 2.880446334589837e-08,
      "loss": 0.2734971046447754,
      "num_input_tokens_seen": 919419376,
      "step": 15145,
      "train_runtime": 197882.1511,
      "train_tokens_per_second": 4646.298
    },
    {
      "epoch": 1.9323979591836735,
      "grad_norm": 0.7018643224048751,
      "learning_rate": 2.8270077778708714e-08,
      "loss": 0.25088844299316404,
      "num_input_tokens_seen": 919729912,
      "step": 15150,
      "train_runtime": 197941.2509,
      "train_tokens_per_second": 4646.479
    },
    {
      "epoch": 1.9330357142857144,
      "grad_norm": 0.6706079135994369,
      "learning_rate": 2.7740681689721327e-08,
      "loss": 0.2477257490158081,
      "num_input_tokens_seen": 920027720,
      "step": 15155,
      "train_runtime": 198007.7054,
      "train_tokens_per_second": 4646.424
    },
    {
      "epoch": 1.933673469387755,
      "grad_norm": 0.6630686337450358,
      "learning_rate": 2.721627561022122e-08,
      "loss": 0.25936317443847656,
      "num_input_tokens_seen": 920323064,
      "step": 15160,
      "train_runtime": 198071.6653,
      "train_tokens_per_second": 4646.415
    },
    {
      "epoch": 1.9343112244897958,
      "grad_norm": 0.6035022817342884,
      "learning_rate": 2.6696860066487417e-08,
      "loss": 0.2339953899383545,
      "num_input_tokens_seen": 920627616,
      "step": 15165,
      "train_runtime": 198144.17,
      "train_tokens_per_second": 4646.251
    },
    {
      "epoch": 1.9349489795918369,
      "grad_norm": 0.7183830825847068,
      "learning_rate": 2.6182435579789055e-08,
      "loss": 0.2542128562927246,
      "num_input_tokens_seen": 920923736,
      "step": 15170,
      "train_runtime": 198208.6669,
      "train_tokens_per_second": 4646.233
    },
    {
      "epoch": 1.9355867346938775,
      "grad_norm": 0.6806673000923349,
      "learning_rate": 2.567300266638706e-08,
      "loss": 0.27027325630187987,
      "num_input_tokens_seen": 921217520,
      "step": 15175,
      "train_runtime": 198280.9546,
      "train_tokens_per_second": 4646.021
    },
    {
      "epoch": 1.9362244897959182,
      "grad_norm": 0.7614629532461339,
      "learning_rate": 2.516856183753358e-08,
      "loss": 0.2549232006072998,
      "num_input_tokens_seen": 921505632,
      "step": 15180,
      "train_runtime": 198341.0521,
      "train_tokens_per_second": 4646.066
    },
    {
      "epoch": 1.9368622448979593,
      "grad_norm": 0.7897218362116912,
      "learning_rate": 2.4669113599469774e-08,
      "loss": 0.24539024829864503,
      "num_input_tokens_seen": 921818320,
      "step": 15185,
      "train_runtime": 198407.6694,
      "train_tokens_per_second": 4646.082
    },
    {
      "epoch": 1.9375,
      "grad_norm": 0.6273112141121303,
      "learning_rate": 2.4174658453426368e-08,
      "loss": 0.24641995429992675,
      "num_input_tokens_seen": 922118720,
      "step": 15190,
      "train_runtime": 198475.3425,
      "train_tokens_per_second": 4646.011
    },
    {
      "epoch": 1.9381377551020407,
      "grad_norm": 0.800921008943011,
      "learning_rate": 2.3685196895624206e-08,
      "loss": 0.2644636631011963,
      "num_input_tokens_seen": 922420232,
      "step": 15195,
      "train_runtime": 198535.7621,
      "train_tokens_per_second": 4646.116
    },
    {
      "epoch": 1.9387755102040818,
      "grad_norm": 0.6749239767271911,
      "learning_rate": 2.3200729417272028e-08,
      "loss": 0.2815537452697754,
      "num_input_tokens_seen": 922733976,
      "step": 15200,
      "train_runtime": 198598.6298,
      "train_tokens_per_second": 4646.225
    },
    {
      "epoch": 1.9394132653061225,
      "grad_norm": 0.6502870759297176,
      "learning_rate": 2.2721256504567026e-08,
      "loss": 0.24536752700805664,
      "num_input_tokens_seen": 923030704,
      "step": 15205,
      "train_runtime": 198676.0273,
      "train_tokens_per_second": 4645.909
    },
    {
      "epoch": 1.9400510204081631,
      "grad_norm": 0.6721506499078248,
      "learning_rate": 2.2246778638692623e-08,
      "loss": 0.241164493560791,
      "num_input_tokens_seen": 923335856,
      "step": 15210,
      "train_runtime": 198752.0539,
      "train_tokens_per_second": 4645.667
    },
    {
      "epoch": 1.9406887755102042,
      "grad_norm": 0.601116855162163,
      "learning_rate": 2.177729629582237e-08,
      "loss": 0.26407551765441895,
      "num_input_tokens_seen": 923639016,
      "step": 15215,
      "train_runtime": 198811.5165,
      "train_tokens_per_second": 4645.802
    },
    {
      "epoch": 1.941326530612245,
      "grad_norm": 0.5823190210070482,
      "learning_rate": 2.1312809947113267e-08,
      "loss": 0.22718312740325927,
      "num_input_tokens_seen": 923939616,
      "step": 15220,
      "train_runtime": 198878.8625,
      "train_tokens_per_second": 4645.741
    },
    {
      "epoch": 1.9419642857142856,
      "grad_norm": 0.6932532957430269,
      "learning_rate": 2.0853320058710214e-08,
      "loss": 0.23678016662597656,
      "num_input_tokens_seen": 924234088,
      "step": 15225,
      "train_runtime": 198943.0174,
      "train_tokens_per_second": 4645.723
    },
    {
      "epoch": 1.9426020408163265,
      "grad_norm": 0.6411570772643133,
      "learning_rate": 2.0398827091743234e-08,
      "loss": 0.27280292510986326,
      "num_input_tokens_seen": 924543768,
      "step": 15230,
      "train_runtime": 199016.607,
      "train_tokens_per_second": 4645.561
    },
    {
      "epoch": 1.9432397959183674,
      "grad_norm": 0.6094249346498507,
      "learning_rate": 1.9949331502327475e-08,
      "loss": 0.23533403873443604,
      "num_input_tokens_seen": 924832536,
      "step": 15235,
      "train_runtime": 199087.1224,
      "train_tokens_per_second": 4645.366
    },
    {
      "epoch": 1.943877551020408,
      "grad_norm": 0.6801283977984018,
      "learning_rate": 1.950483374156431e-08,
      "loss": 0.26111152172088625,
      "num_input_tokens_seen": 925138328,
      "step": 15240,
      "train_runtime": 199153.887,
      "train_tokens_per_second": 4645.344
    },
    {
      "epoch": 1.944515306122449,
      "grad_norm": 0.6425962753034345,
      "learning_rate": 1.9065334255536916e-08,
      "loss": 0.25782184600830077,
      "num_input_tokens_seen": 925447200,
      "step": 15245,
      "train_runtime": 199225.8819,
      "train_tokens_per_second": 4645.216
    },
    {
      "epoch": 1.9451530612244898,
      "grad_norm": 0.6434454146016215,
      "learning_rate": 1.8630833485314693e-08,
      "loss": 0.24272508621215821,
      "num_input_tokens_seen": 925753016,
      "step": 15250,
      "train_runtime": 199302.7033,
      "train_tokens_per_second": 4644.96
    },
    {
      "epoch": 1.9457908163265305,
      "grad_norm": 0.6971730916606771,
      "learning_rate": 1.8201331866948834e-08,
      "loss": 0.25693254470825194,
      "num_input_tokens_seen": 926056384,
      "step": 15255,
      "train_runtime": 199373.5173,
      "train_tokens_per_second": 4644.831
    },
    {
      "epoch": 1.9464285714285714,
      "grad_norm": 0.6431297755968343,
      "learning_rate": 1.7776829831474553e-08,
      "loss": 0.2646693229675293,
      "num_input_tokens_seen": 926363824,
      "step": 15260,
      "train_runtime": 199440.0629,
      "train_tokens_per_second": 4644.823
    },
    {
      "epoch": 1.9470663265306123,
      "grad_norm": 0.6666066906242785,
      "learning_rate": 1.735732780490884e-08,
      "loss": 0.27986161708831786,
      "num_input_tokens_seen": 926674400,
      "step": 15265,
      "train_runtime": 199495.8617,
      "train_tokens_per_second": 4645.081
    },
    {
      "epoch": 1.947704081632653,
      "grad_norm": 0.6656222223883199,
      "learning_rate": 1.694282620825216e-08,
      "loss": 0.2596017122268677,
      "num_input_tokens_seen": 926972904,
      "step": 15270,
      "train_runtime": 199564.442,
      "train_tokens_per_second": 4644.98
    },
    {
      "epoch": 1.9483418367346939,
      "grad_norm": 0.6275419261464044,
      "learning_rate": 1.6533325457485093e-08,
      "loss": 0.25693569183349607,
      "num_input_tokens_seen": 927282680,
      "step": 15275,
      "train_runtime": 199629.5856,
      "train_tokens_per_second": 4645.016
    },
    {
      "epoch": 1.9489795918367347,
      "grad_norm": 0.733231245261694,
      "learning_rate": 1.612882596357057e-08,
      "loss": 0.2503585577011108,
      "num_input_tokens_seen": 927592416,
      "step": 15280,
      "train_runtime": 199700.9486,
      "train_tokens_per_second": 4644.907
    },
    {
      "epoch": 1.9496173469387754,
      "grad_norm": 0.6585430168053389,
      "learning_rate": 1.5729328132452203e-08,
      "loss": 0.24806904792785645,
      "num_input_tokens_seen": 927904256,
      "step": 15285,
      "train_runtime": 199764.2267,
      "train_tokens_per_second": 4644.997
    },
    {
      "epoch": 1.9502551020408163,
      "grad_norm": 0.6606777651635418,
      "learning_rate": 1.533483236505373e-08,
      "loss": 0.24865076541900635,
      "num_input_tokens_seen": 928217856,
      "step": 15290,
      "train_runtime": 199830.4145,
      "train_tokens_per_second": 4645.028
    },
    {
      "epoch": 1.9508928571428572,
      "grad_norm": 0.6117485454438667,
      "learning_rate": 1.494533905727902e-08,
      "loss": 0.26568968296051027,
      "num_input_tokens_seen": 928520184,
      "step": 15295,
      "train_runtime": 199898.8403,
      "train_tokens_per_second": 4644.95
    },
    {
      "epoch": 1.9515306122448979,
      "grad_norm": 0.7172290442315818,
      "learning_rate": 1.4560848600012056e-08,
      "loss": 0.24280576705932616,
      "num_input_tokens_seen": 928828280,
      "step": 15300,
      "train_runtime": 199960.7418,
      "train_tokens_per_second": 4645.053
    },
    {
      "epoch": 1.9521683673469388,
      "grad_norm": 0.5929442067738704,
      "learning_rate": 1.4181361379115854e-08,
      "loss": 0.25272064208984374,
      "num_input_tokens_seen": 929145448,
      "step": 15305,
      "train_runtime": 200030.8148,
      "train_tokens_per_second": 4645.012
    },
    {
      "epoch": 1.9528061224489797,
      "grad_norm": 0.6939548267696385,
      "learning_rate": 1.3806877775432436e-08,
      "loss": 0.24609718322753907,
      "num_input_tokens_seen": 929452648,
      "step": 15310,
      "train_runtime": 200084.4506,
      "train_tokens_per_second": 4645.302
    },
    {
      "epoch": 1.9534438775510203,
      "grad_norm": 0.7921844194217872,
      "learning_rate": 1.3437398164781734e-08,
      "loss": 0.24976670742034912,
      "num_input_tokens_seen": 929748480,
      "step": 15315,
      "train_runtime": 200149.5511,
      "train_tokens_per_second": 4645.269
    },
    {
      "epoch": 1.9540816326530612,
      "grad_norm": 0.6323729258574822,
      "learning_rate": 1.3072922917963249e-08,
      "loss": 0.24251127243041992,
      "num_input_tokens_seen": 930053288,
      "step": 15320,
      "train_runtime": 200213.486,
      "train_tokens_per_second": 4645.308
    },
    {
      "epoch": 1.9547193877551021,
      "grad_norm": 0.6906993758341127,
      "learning_rate": 1.2713452400752724e-08,
      "loss": 0.2625253677368164,
      "num_input_tokens_seen": 930369984,
      "step": 15325,
      "train_runtime": 200280.8044,
      "train_tokens_per_second": 4645.328
    },
    {
      "epoch": 1.9553571428571428,
      "grad_norm": 0.6574415884308985,
      "learning_rate": 1.2358986973903807e-08,
      "loss": 0.2642559051513672,
      "num_input_tokens_seen": 930684352,
      "step": 15330,
      "train_runtime": 200345.4531,
      "train_tokens_per_second": 4645.398
    },
    {
      "epoch": 1.9559948979591837,
      "grad_norm": 0.6779783906783353,
      "learning_rate": 1.2009526993147503e-08,
      "loss": 0.2565456390380859,
      "num_input_tokens_seen": 930994304,
      "step": 15335,
      "train_runtime": 200406.1223,
      "train_tokens_per_second": 4645.538
    },
    {
      "epoch": 1.9566326530612246,
      "grad_norm": 0.5707303816920276,
      "learning_rate": 1.1665072809191602e-08,
      "loss": 0.24982020854949952,
      "num_input_tokens_seen": 931306504,
      "step": 15340,
      "train_runtime": 200467.0895,
      "train_tokens_per_second": 4645.683
    },
    {
      "epoch": 1.9572704081632653,
      "grad_norm": 0.6364327412740145,
      "learning_rate": 1.132562476771959e-08,
      "loss": 0.25244736671447754,
      "num_input_tokens_seen": 931609672,
      "step": 15345,
      "train_runtime": 200521.3871,
      "train_tokens_per_second": 4645.937
    },
    {
      "epoch": 1.9579081632653061,
      "grad_norm": 0.6742588290680526,
      "learning_rate": 1.0991183209391188e-08,
      "loss": 0.24548232555389404,
      "num_input_tokens_seen": 931914232,
      "step": 15350,
      "train_runtime": 200590.7823,
      "train_tokens_per_second": 4645.848
    },
    {
      "epoch": 1.958545918367347,
      "grad_norm": 0.6639449792544483,
      "learning_rate": 1.0661748469842359e-08,
      "loss": 0.2334390640258789,
      "num_input_tokens_seen": 932218120,
      "step": 15355,
      "train_runtime": 200655.2846,
      "train_tokens_per_second": 4645.869
    },
    {
      "epoch": 1.9591836734693877,
      "grad_norm": 0.7667890659307897,
      "learning_rate": 1.0337320879683087e-08,
      "loss": 0.23780970573425292,
      "num_input_tokens_seen": 932530040,
      "step": 15360,
      "train_runtime": 200728.498,
      "train_tokens_per_second": 4645.728
    },
    {
      "epoch": 1.9598214285714286,
      "grad_norm": 1.3434161686822583,
      "learning_rate": 1.0017900764500155e-08,
      "loss": 0.255706787109375,
      "num_input_tokens_seen": 932840976,
      "step": 15365,
      "train_runtime": 200783.3764,
      "train_tokens_per_second": 4646.007
    },
    {
      "epoch": 1.9604591836734695,
      "grad_norm": 0.568162911645631,
      "learning_rate": 9.703488444853248e-09,
      "loss": 0.2541600942611694,
      "num_input_tokens_seen": 933147184,
      "step": 15370,
      "train_runtime": 200854.2584,
      "train_tokens_per_second": 4645.892
    },
    {
      "epoch": 1.9610969387755102,
      "grad_norm": 0.6462376565475885,
      "learning_rate": 9.39408423627719e-09,
      "loss": 0.24727683067321776,
      "num_input_tokens_seen": 933449616,
      "step": 15375,
      "train_runtime": 200912.2784,
      "train_tokens_per_second": 4646.056
    },
    {
      "epoch": 1.961734693877551,
      "grad_norm": 0.6670489638809874,
      "learning_rate": 9.089688449280265e-09,
      "loss": 0.2606107950210571,
      "num_input_tokens_seen": 933748368,
      "step": 15380,
      "train_runtime": 200986.7656,
      "train_tokens_per_second": 4645.82
    },
    {
      "epoch": 1.962372448979592,
      "grad_norm": 0.6731224891552309,
      "learning_rate": 8.79030138934589e-09,
      "loss": 0.25869710445404054,
      "num_input_tokens_seen": 934066280,
      "step": 15385,
      "train_runtime": 201032.061,
      "train_tokens_per_second": 4646.355
    },
    {
      "epoch": 1.9630102040816326,
      "grad_norm": 0.6351121907164077,
      "learning_rate": 8.49592335692928e-09,
      "loss": 0.23783254623413086,
      "num_input_tokens_seen": 934360528,
      "step": 15390,
      "train_runtime": 201092.7173,
      "train_tokens_per_second": 4646.417
    },
    {
      "epoch": 1.9636479591836735,
      "grad_norm": 0.682117467602623,
      "learning_rate": 8.206554647459675e-09,
      "loss": 0.22529397010803223,
      "num_input_tokens_seen": 934662128,
      "step": 15395,
      "train_runtime": 201167.2167,
      "train_tokens_per_second": 4646.195
    },
    {
      "epoch": 1.9642857142857144,
      "grad_norm": 0.6821335275257784,
      "learning_rate": 7.922195551338107e-09,
      "loss": 0.2720287084579468,
      "num_input_tokens_seen": 934954856,
      "step": 15400,
      "train_runtime": 201225.6993,
      "train_tokens_per_second": 4646.299
    },
    {
      "epoch": 1.964923469387755,
      "grad_norm": 0.662620995677526,
      "learning_rate": 7.642846353939637e-09,
      "loss": 0.2662359237670898,
      "num_input_tokens_seen": 935263264,
      "step": 15405,
      "train_runtime": 201304.3613,
      "train_tokens_per_second": 4646.016
    },
    {
      "epoch": 1.9655612244897958,
      "grad_norm": 0.6280068912895532,
      "learning_rate": 7.368507335610009e-09,
      "loss": 0.2379246711730957,
      "num_input_tokens_seen": 935557064,
      "step": 15410,
      "train_runtime": 201365.9686,
      "train_tokens_per_second": 4646.054
    },
    {
      "epoch": 1.9661989795918369,
      "grad_norm": 0.6099009764095411,
      "learning_rate": 7.099178771668436e-09,
      "loss": 0.24412119388580322,
      "num_input_tokens_seen": 935867560,
      "step": 15415,
      "train_runtime": 201429.8519,
      "train_tokens_per_second": 4646.121
    },
    {
      "epoch": 1.9668367346938775,
      "grad_norm": 0.6845277674559912,
      "learning_rate": 6.834860932403709e-09,
      "loss": 0.25436787605285643,
      "num_input_tokens_seen": 936164024,
      "step": 15420,
      "train_runtime": 201510.2903,
      "train_tokens_per_second": 4645.738
    },
    {
      "epoch": 1.9674744897959182,
      "grad_norm": 0.6636064340985193,
      "learning_rate": 6.575554083078084e-09,
      "loss": 0.2626820087432861,
      "num_input_tokens_seen": 936466568,
      "step": 15425,
      "train_runtime": 201582.2793,
      "train_tokens_per_second": 4645.58
    },
    {
      "epoch": 1.9681122448979593,
      "grad_norm": 0.7282240604628171,
      "learning_rate": 6.321258483924508e-09,
      "loss": 0.2702171802520752,
      "num_input_tokens_seen": 936768752,
      "step": 15430,
      "train_runtime": 201645.1336,
      "train_tokens_per_second": 4645.63
    },
    {
      "epoch": 1.96875,
      "grad_norm": 0.683843369114111,
      "learning_rate": 6.071974390144953e-09,
      "loss": 0.23894684314727782,
      "num_input_tokens_seen": 937071272,
      "step": 15435,
      "train_runtime": 201706.7922,
      "train_tokens_per_second": 4645.71
    },
    {
      "epoch": 1.9693877551020407,
      "grad_norm": 0.7116665504616505,
      "learning_rate": 5.827702051914852e-09,
      "loss": 0.25569274425506594,
      "num_input_tokens_seen": 937374224,
      "step": 15440,
      "train_runtime": 201782.1183,
      "train_tokens_per_second": 4645.477
    },
    {
      "epoch": 1.9700255102040818,
      "grad_norm": 0.7290339208563372,
      "learning_rate": 5.5884417143775574e-09,
      "loss": 0.2737967252731323,
      "num_input_tokens_seen": 937680296,
      "step": 15445,
      "train_runtime": 201837.285,
      "train_tokens_per_second": 4645.724
    },
    {
      "epoch": 1.9706632653061225,
      "grad_norm": 0.719496216823003,
      "learning_rate": 5.354193617648218e-09,
      "loss": 0.25485644340515134,
      "num_input_tokens_seen": 937965664,
      "step": 15450,
      "train_runtime": 201906.2543,
      "train_tokens_per_second": 4645.55
    },
    {
      "epoch": 1.9713010204081631,
      "grad_norm": 0.6817270705455376,
      "learning_rate": 5.124957996811009e-09,
      "loss": 0.2593040466308594,
      "num_input_tokens_seen": 938265008,
      "step": 15455,
      "train_runtime": 201970.522,
      "train_tokens_per_second": 4645.554
    },
    {
      "epoch": 1.9719387755102042,
      "grad_norm": 0.6383414887725855,
      "learning_rate": 4.900735081919128e-09,
      "loss": 0.23735318183898926,
      "num_input_tokens_seen": 938569376,
      "step": 15460,
      "train_runtime": 202033.0734,
      "train_tokens_per_second": 4645.622
    },
    {
      "epoch": 1.972576530612245,
      "grad_norm": 0.6458138943082795,
      "learning_rate": 4.6815250979970195e-09,
      "loss": 0.25776076316833496,
      "num_input_tokens_seen": 938878464,
      "step": 15465,
      "train_runtime": 202094.6504,
      "train_tokens_per_second": 4645.736
    },
    {
      "epoch": 1.9732142857142856,
      "grad_norm": 0.637196312980925,
      "learning_rate": 4.467328265037041e-09,
      "loss": 0.25017719268798827,
      "num_input_tokens_seen": 939168384,
      "step": 15470,
      "train_runtime": 202158.0587,
      "train_tokens_per_second": 4645.713
    },
    {
      "epoch": 1.9738520408163265,
      "grad_norm": 0.5942506586339228,
      "learning_rate": 4.258144798000019e-09,
      "loss": 0.2674983501434326,
      "num_input_tokens_seen": 939479720,
      "step": 15475,
      "train_runtime": 202228.6521,
      "train_tokens_per_second": 4645.631
    },
    {
      "epoch": 1.9744897959183674,
      "grad_norm": 0.6741715420082401,
      "learning_rate": 4.05397490681636e-09,
      "loss": 0.2481661319732666,
      "num_input_tokens_seen": 939776816,
      "step": 15480,
      "train_runtime": 202298.5971,
      "train_tokens_per_second": 4645.493
    },
    {
      "epoch": 1.975127551020408,
      "grad_norm": 0.7177061038203771,
      "learning_rate": 3.854818796385495e-09,
      "loss": 0.26866137981414795,
      "num_input_tokens_seen": 940072480,
      "step": 15485,
      "train_runtime": 202360.2348,
      "train_tokens_per_second": 4645.54
    },
    {
      "epoch": 1.975765306122449,
      "grad_norm": 0.6509472749608338,
      "learning_rate": 3.660676666573659e-09,
      "loss": 0.24293665885925292,
      "num_input_tokens_seen": 940372488,
      "step": 15490,
      "train_runtime": 202424.866,
      "train_tokens_per_second": 4645.538
    },
    {
      "epoch": 1.9764030612244898,
      "grad_norm": 0.6646487596865714,
      "learning_rate": 3.4715487122161107e-09,
      "loss": 0.2463254451751709,
      "num_input_tokens_seen": 940678120,
      "step": 15495,
      "train_runtime": 202490.7442,
      "train_tokens_per_second": 4645.536
    },
    {
      "epoch": 1.9770408163265305,
      "grad_norm": 0.6649270996945488,
      "learning_rate": 3.287435123116578e-09,
      "loss": 0.258991003036499,
      "num_input_tokens_seen": 940991336,
      "step": 15500,
      "train_runtime": 202554.0482,
      "train_tokens_per_second": 4645.631
    },
    {
      "epoch": 1.9776785714285714,
      "grad_norm": 0.687046950087734,
      "learning_rate": 3.1083360840455935e-09,
      "loss": 0.24148159027099608,
      "num_input_tokens_seen": 941285288,
      "step": 15505,
      "train_runtime": 202617.521,
      "train_tokens_per_second": 4645.626
    },
    {
      "epoch": 1.9783163265306123,
      "grad_norm": 0.7612710304971785,
      "learning_rate": 2.9342517747410482e-09,
      "loss": 0.25423603057861327,
      "num_input_tokens_seen": 941603816,
      "step": 15510,
      "train_runtime": 202668.4043,
      "train_tokens_per_second": 4646.032
    },
    {
      "epoch": 1.978954081632653,
      "grad_norm": 0.6790729073095124,
      "learning_rate": 2.7651823699093027e-09,
      "loss": 0.24060502052307128,
      "num_input_tokens_seen": 941927344,
      "step": 15515,
      "train_runtime": 202721.8213,
      "train_tokens_per_second": 4646.403
    },
    {
      "epoch": 1.9795918367346939,
      "grad_norm": 0.7419740535259216,
      "learning_rate": 2.6011280392235215e-09,
      "loss": 0.25852391719818113,
      "num_input_tokens_seen": 942228704,
      "step": 15520,
      "train_runtime": 202789.095,
      "train_tokens_per_second": 4646.348
    },
    {
      "epoch": 1.9802295918367347,
      "grad_norm": 0.6641922208558795,
      "learning_rate": 2.442088947323118e-09,
      "loss": 0.2551721572875977,
      "num_input_tokens_seen": 942534256,
      "step": 15525,
      "train_runtime": 202861.7971,
      "train_tokens_per_second": 4646.189
    },
    {
      "epoch": 1.9808673469387754,
      "grad_norm": 0.6034823957794491,
      "learning_rate": 2.288065253814864e-09,
      "loss": 0.2408912420272827,
      "num_input_tokens_seen": 942822104,
      "step": 15530,
      "train_runtime": 202925.0236,
      "train_tokens_per_second": 4646.16
    },
    {
      "epoch": 1.9815051020408163,
      "grad_norm": 0.6185290463513637,
      "learning_rate": 2.139057113272891e-09,
      "loss": 0.249487042427063,
      "num_input_tokens_seen": 943125064,
      "step": 15535,
      "train_runtime": 202986.8056,
      "train_tokens_per_second": 4646.238
    },
    {
      "epoch": 1.9821428571428572,
      "grad_norm": 0.700980111156108,
      "learning_rate": 1.995064675237024e-09,
      "loss": 0.24320669174194337,
      "num_input_tokens_seen": 943414656,
      "step": 15540,
      "train_runtime": 203052.9781,
      "train_tokens_per_second": 4646.15
    },
    {
      "epoch": 1.9827806122448979,
      "grad_norm": 0.6680659977178877,
      "learning_rate": 1.8560880842133366e-09,
      "loss": 0.2668965578079224,
      "num_input_tokens_seen": 943715632,
      "step": 15545,
      "train_runtime": 203109.5051,
      "train_tokens_per_second": 4646.339
    },
    {
      "epoch": 1.9834183673469388,
      "grad_norm": 0.6832056937466896,
      "learning_rate": 1.7221274796752618e-09,
      "loss": 0.2613052845001221,
      "num_input_tokens_seen": 944022192,
      "step": 15550,
      "train_runtime": 203180.7187,
      "train_tokens_per_second": 4646.219
    },
    {
      "epoch": 1.9840561224489797,
      "grad_norm": 0.6653955276996287,
      "learning_rate": 1.5931829960608158e-09,
      "loss": 0.266310453414917,
      "num_input_tokens_seen": 944328320,
      "step": 15555,
      "train_runtime": 203251.005,
      "train_tokens_per_second": 4646.119
    },
    {
      "epoch": 1.9846938775510203,
      "grad_norm": 0.6843844432253159,
      "learning_rate": 1.469254762775374e-09,
      "loss": 0.2567771911621094,
      "num_input_tokens_seen": 944620232,
      "step": 15560,
      "train_runtime": 203308.201,
      "train_tokens_per_second": 4646.248
    },
    {
      "epoch": 1.9853316326530612,
      "grad_norm": 0.6452788820576626,
      "learning_rate": 1.350342904188895e-09,
      "loss": 0.25018773078918455,
      "num_input_tokens_seen": 944927232,
      "step": 15565,
      "train_runtime": 203351.7131,
      "train_tokens_per_second": 4646.763
    },
    {
      "epoch": 1.9859693877551021,
      "grad_norm": 0.6391438789234974,
      "learning_rate": 1.2364475396386966e-09,
      "loss": 0.25844545364379884,
      "num_input_tokens_seen": 945234328,
      "step": 15570,
      "train_runtime": 203419.2949,
      "train_tokens_per_second": 4646.729
    },
    {
      "epoch": 1.9866071428571428,
      "grad_norm": 0.6599275936883742,
      "learning_rate": 1.1275687834266803e-09,
      "loss": 0.24469830989837646,
      "num_input_tokens_seen": 945532880,
      "step": 15575,
      "train_runtime": 203488.83,
      "train_tokens_per_second": 4646.608
    },
    {
      "epoch": 1.9872448979591837,
      "grad_norm": 0.6697241520164385,
      "learning_rate": 1.0237067448193306e-09,
      "loss": 0.24163453578948973,
      "num_input_tokens_seen": 945847360,
      "step": 15580,
      "train_runtime": 203561.7077,
      "train_tokens_per_second": 4646.49
    },
    {
      "epoch": 1.9878826530612246,
      "grad_norm": 0.6638025854331141,
      "learning_rate": 9.248615280499362e-10,
      "loss": 0.24226274490356445,
      "num_input_tokens_seen": 946141912,
      "step": 15585,
      "train_runtime": 203617.9354,
      "train_tokens_per_second": 4646.653
    },
    {
      "epoch": 1.9885204081632653,
      "grad_norm": 0.7521110076166722,
      "learning_rate": 8.310332323169246e-10,
      "loss": 0.2418217182159424,
      "num_input_tokens_seen": 946445056,
      "step": 15590,
      "train_runtime": 203690.9792,
      "train_tokens_per_second": 4646.475
    },
    {
      "epoch": 1.9891581632653061,
      "grad_norm": 0.7037770439883531,
      "learning_rate": 7.422219517833062e-10,
      "loss": 0.25137481689453123,
      "num_input_tokens_seen": 946746768,
      "step": 15595,
      "train_runtime": 203761.1565,
      "train_tokens_per_second": 4646.355
    },
    {
      "epoch": 1.989795918367347,
      "grad_norm": 0.647441479828051,
      "learning_rate": 6.584277755772306e-10,
      "loss": 0.24022195339202881,
      "num_input_tokens_seen": 947033408,
      "step": 15600,
      "train_runtime": 203821.8061,
      "train_tokens_per_second": 4646.379
    },
    {
      "epoch": 1.9904336734693877,
      "grad_norm": 0.6193009607138327,
      "learning_rate": 5.796507877919855e-10,
      "loss": 0.24291386604309081,
      "num_input_tokens_seen": 947333432,
      "step": 15605,
      "train_runtime": 203881.3778,
      "train_tokens_per_second": 4646.493
    },
    {
      "epoch": 1.9910714285714286,
      "grad_norm": 0.6491329994488889,
      "learning_rate": 5.058910674859973e-10,
      "loss": 0.25441684722900393,
      "num_input_tokens_seen": 947631536,
      "step": 15610,
      "train_runtime": 203947.025,
      "train_tokens_per_second": 4646.459
    },
    {
      "epoch": 1.9917091836734695,
      "grad_norm": 0.675203280757148,
      "learning_rate": 4.371486886822762e-10,
      "loss": 0.2432384490966797,
      "num_input_tokens_seen": 947946016,
      "step": 15615,
      "train_runtime": 204003.6215,
      "train_tokens_per_second": 4646.712
    },
    {
      "epoch": 1.9923469387755102,
      "grad_norm": 0.6856722346574126,
      "learning_rate": 3.7342372036841546e-10,
      "loss": 0.2673120737075806,
      "num_input_tokens_seen": 948251504,
      "step": 15620,
      "train_runtime": 204068.2295,
      "train_tokens_per_second": 4646.738
    },
    {
      "epoch": 1.992984693877551,
      "grad_norm": 0.6423156847670656,
      "learning_rate": 3.147162264971471e-10,
      "loss": 0.26491403579711914,
      "num_input_tokens_seen": 948559176,
      "step": 15625,
      "train_runtime": 204130.627,
      "train_tokens_per_second": 4646.824
    },
    {
      "epoch": 1.993622448979592,
      "grad_norm": 0.5998854673719396,
      "learning_rate": 2.610262659852314e-10,
      "loss": 0.23858168125152587,
      "num_input_tokens_seen": 948871280,
      "step": 15630,
      "train_runtime": 204191.073,
      "train_tokens_per_second": 4646.977
    },
    {
      "epoch": 1.9942602040816326,
      "grad_norm": 0.6171558060159329,
      "learning_rate": 2.123538927145674e-10,
      "loss": 0.25632877349853517,
      "num_input_tokens_seen": 949161856,
      "step": 15635,
      "train_runtime": 204260.3076,
      "train_tokens_per_second": 4646.825
    },
    {
      "epoch": 1.9948979591836735,
      "grad_norm": 0.6536324489667538,
      "learning_rate": 1.686991555310824e-10,
      "loss": 0.2401134490966797,
      "num_input_tokens_seen": 949457536,
      "step": 15640,
      "train_runtime": 204329.9453,
      "train_tokens_per_second": 4646.688
    },
    {
      "epoch": 1.9955357142857144,
      "grad_norm": 0.7366112659743931,
      "learning_rate": 1.3006209824584226e-10,
      "loss": 0.24632782936096193,
      "num_input_tokens_seen": 949760728,
      "step": 15645,
      "train_runtime": 204387.1487,
      "train_tokens_per_second": 4646.871
    },
    {
      "epoch": 1.996173469387755,
      "grad_norm": 0.7691263250269446,
      "learning_rate": 9.6442759632831e-11,
      "loss": 0.24630413055419922,
      "num_input_tokens_seen": 950074048,
      "step": 15650,
      "train_runtime": 204458.7634,
      "train_tokens_per_second": 4646.776
    },
    {
      "epoch": 1.9968112244897958,
      "grad_norm": 0.629024715586974,
      "learning_rate": 6.784117343228147e-11,
      "loss": 0.22919554710388185,
      "num_input_tokens_seen": 950380584,
      "step": 15655,
      "train_runtime": 204511.7495,
      "train_tokens_per_second": 4647.071
    },
    {
      "epoch": 1.9974489795918369,
      "grad_norm": 0.703394379431084,
      "learning_rate": 4.425736834789973e-11,
      "loss": 0.2392258644104004,
      "num_input_tokens_seen": 950672704,
      "step": 15660,
      "train_runtime": 204582.9692,
      "train_tokens_per_second": 4646.881
    },
    {
      "epoch": 1.9980867346938775,
      "grad_norm": 0.6532427160875364,
      "learning_rate": 2.5691368046865118e-11,
      "loss": 0.2364060640335083,
      "num_input_tokens_seen": 950979656,
      "step": 15665,
      "train_runtime": 204655.5746,
      "train_tokens_per_second": 4646.732
    },
    {
      "epoch": 1.9987244897959182,
      "grad_norm": 0.6688449806367598,
      "learning_rate": 1.214319116260576e-11,
      "loss": 0.24228816032409667,
      "num_input_tokens_seen": 951292560,
      "step": 15670,
      "train_runtime": 204718.0305,
      "train_tokens_per_second": 4646.843
    },
    {
      "epoch": 1.9993622448979593,
      "grad_norm": 0.7314343258292,
      "learning_rate": 3.6128512903577105e-12,
      "loss": 0.25652594566345216,
      "num_input_tokens_seen": 951600896,
      "step": 15675,
      "train_runtime": 204786.0422,
      "train_tokens_per_second": 4646.805
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.7020156799873025,
      "learning_rate": 1.0035699216093265e-13,
      "loss": 0.23354415893554686,
      "num_input_tokens_seen": 951884320,
      "step": 15680,
      "train_runtime": 204849.0219,
      "train_tokens_per_second": 4646.76
    },
    {
      "epoch": 2.0,
      "num_input_tokens_seen": 951884320,
      "step": 15680,
      "total_flos": 1760361553723392.0,
      "train_loss": 0.2931738243951481,
      "train_runtime": 204889.8677,
      "train_samples_per_second": 0.306,
      "train_steps_per_second": 0.077
    }
  ],
  "logging_steps": 5,
  "max_steps": 15680,
  "num_input_tokens_seen": 951884320,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1760361553723392.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}