{
  "best_global_step": 75000,
  "best_metric": 0.33609068393707275,
  "best_model_checkpoint": "/workspace/MT_En_Multilingual/checkpoint-75000",
  "epoch": 3.0,
  "eval_steps": 5000,
  "global_step": 84789,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003538202637287791,
      "grad_norm": 1.5,
      "learning_rate": 4.9977515176118345e-05,
      "loss": 1.0859756469726562,
      "num_input_tokens_seen": 2861504,
      "step": 10,
      "train_runtime": 106.9904,
      "train_tokens_per_second": 26745.424
    },
    {
      "epoch": 0.0007076405274575581,
      "grad_norm": 0.90625,
      "learning_rate": 4.9952567580506e-05,
      "loss": 0.8924098968505859,
      "num_input_tokens_seen": 5659264,
      "step": 20,
      "train_runtime": 200.1545,
      "train_tokens_per_second": 28274.476
    },
    {
      "epoch": 0.0010614607911863373,
      "grad_norm": 0.828125,
      "learning_rate": 4.992765730738634e-05,
      "loss": 0.8380861282348633,
      "num_input_tokens_seen": 8484416,
      "step": 30,
      "train_runtime": 298.1801,
      "train_tokens_per_second": 28453.997
    },
    {
      "epoch": 0.0014152810549151163,
      "grad_norm": 0.76953125,
      "learning_rate": 4.9902784263792476e-05,
      "loss": 0.8041961669921875,
      "num_input_tokens_seen": 11342464,
      "step": 40,
      "train_runtime": 395.2287,
      "train_tokens_per_second": 28698.484
    },
    {
      "epoch": 0.0017691013186438955,
      "grad_norm": 0.796875,
      "learning_rate": 4.987794835708133e-05,
      "loss": 0.7847923755645752,
      "num_input_tokens_seen": 14208000,
      "step": 50,
      "train_runtime": 492.4151,
      "train_tokens_per_second": 28853.703
    },
    {
      "epoch": 0.0021229215823726747,
      "grad_norm": 0.80859375,
      "learning_rate": 4.985314949493234e-05,
      "loss": 0.761699104309082,
      "num_input_tokens_seen": 17070848,
      "step": 60,
      "train_runtime": 590.3487,
      "train_tokens_per_second": 28916.55
    },
    {
      "epoch": 0.0024767418461014534,
      "grad_norm": 0.80078125,
      "learning_rate": 4.982838758534584e-05,
      "loss": 0.7455968856811523,
      "num_input_tokens_seen": 19876352,
      "step": 70,
      "train_runtime": 685.2112,
      "train_tokens_per_second": 29007.628
    },
    {
      "epoch": 0.0028305621098302326,
      "grad_norm": 0.73828125,
      "learning_rate": 4.980366253664179e-05,
      "loss": 0.7417703628540039,
      "num_input_tokens_seen": 22738816,
      "step": 80,
      "train_runtime": 783.207,
      "train_tokens_per_second": 29032.959
    },
    {
      "epoch": 0.0031843823735590118,
      "grad_norm": 0.76953125,
      "learning_rate": 4.977897425745825e-05,
      "loss": 0.7280791759490967,
      "num_input_tokens_seen": 25550720,
      "step": 90,
      "train_runtime": 882.6487,
      "train_tokens_per_second": 28947.778
    },
    {
      "epoch": 0.003538202637287791,
      "grad_norm": 0.84375,
      "learning_rate": 4.975432265674997e-05,
      "loss": 0.7208812713623047,
      "num_input_tokens_seen": 28315520,
      "step": 100,
      "train_runtime": 977.2817,
      "train_tokens_per_second": 28973.755
    },
    {
      "epoch": 0.0038920229010165697,
      "grad_norm": 0.76171875,
      "learning_rate": 4.972970764378705e-05,
      "loss": 0.7049215316772461,
      "num_input_tokens_seen": 31176512,
      "step": 110,
      "train_runtime": 1074.9122,
      "train_tokens_per_second": 29003.774
    },
    {
      "epoch": 0.004245843164745349,
      "grad_norm": 0.72265625,
      "learning_rate": 4.970512912815344e-05,
      "loss": 0.6969927787780762,
      "num_input_tokens_seen": 34090624,
      "step": 120,
      "train_runtime": 1177.7235,
      "train_tokens_per_second": 28946.205
    },
    {
      "epoch": 0.004599663428474128,
      "grad_norm": 0.78125,
      "learning_rate": 4.968058701974564e-05,
      "loss": 0.6911009311676025,
      "num_input_tokens_seen": 36877632,
      "step": 130,
      "train_runtime": 1272.0223,
      "train_tokens_per_second": 28991.34
    },
    {
      "epoch": 0.004953483692202907,
      "grad_norm": 0.7578125,
      "learning_rate": 4.96560812287712e-05,
      "loss": 0.6851213455200196,
      "num_input_tokens_seen": 39719808,
      "step": 140,
      "train_runtime": 1371.1951,
      "train_tokens_per_second": 28967.291
    },
    {
      "epoch": 0.005307303955931686,
      "grad_norm": 0.703125,
      "learning_rate": 4.963161166574748e-05,
      "loss": 0.6843182563781738,
      "num_input_tokens_seen": 42594048,
      "step": 150,
      "train_runtime": 1471.095,
      "train_tokens_per_second": 28953.976
    },
    {
      "epoch": 0.005661124219660465,
      "grad_norm": 0.75390625,
      "learning_rate": 4.960717824150013e-05,
      "loss": 0.6695240020751954,
      "num_input_tokens_seen": 45422080,
      "step": 160,
      "train_runtime": 1566.1419,
      "train_tokens_per_second": 29002.532
    },
    {
      "epoch": 0.006014944483389244,
      "grad_norm": 0.765625,
      "learning_rate": 4.9582780867161893e-05,
      "loss": 0.6712607383728028,
      "num_input_tokens_seen": 48225408,
      "step": 170,
      "train_runtime": 1660.613,
      "train_tokens_per_second": 29040.727
    },
    {
      "epoch": 0.0063687647471180235,
      "grad_norm": 0.703125,
      "learning_rate": 4.955841945417105e-05,
      "loss": 0.6608043670654297,
      "num_input_tokens_seen": 51081536,
      "step": 180,
      "train_runtime": 1756.5585,
      "train_tokens_per_second": 29080.465
    },
    {
      "epoch": 0.006722585010846802,
      "grad_norm": 0.7421875,
      "learning_rate": 4.953409391427024e-05,
      "loss": 0.6619660377502441,
      "num_input_tokens_seen": 53934336,
      "step": 190,
      "train_runtime": 1855.1051,
      "train_tokens_per_second": 29073.466
    },
    {
      "epoch": 0.007076405274575582,
      "grad_norm": 0.81640625,
      "learning_rate": 4.950980415950502e-05,
      "loss": 0.6573782444000245,
      "num_input_tokens_seen": 56792576,
      "step": 200,
      "train_runtime": 1952.6485,
      "train_tokens_per_second": 29084.894
    },
    {
      "epoch": 0.007430225538304361,
      "grad_norm": 0.74609375,
      "learning_rate": 4.9485550102222575e-05,
      "loss": 0.6446058273315429,
      "num_input_tokens_seen": 59683840,
      "step": 210,
      "train_runtime": 2051.8374,
      "train_tokens_per_second": 29087.997
    },
    {
      "epoch": 0.007784045802033139,
      "grad_norm": 0.734375,
      "learning_rate": 4.946133165507037e-05,
      "loss": 0.647282075881958,
      "num_input_tokens_seen": 62532608,
      "step": 220,
      "train_runtime": 2150.2306,
      "train_tokens_per_second": 29081.815
    },
    {
      "epoch": 0.008137866065761919,
      "grad_norm": 0.76171875,
      "learning_rate": 4.943714873099483e-05,
      "loss": 0.6425490379333496,
      "num_input_tokens_seen": 65382784,
      "step": 230,
      "train_runtime": 2247.0233,
      "train_tokens_per_second": 29097.51
    },
    {
      "epoch": 0.008491686329490699,
      "grad_norm": 0.72265625,
      "learning_rate": 4.9413001243240024e-05,
      "loss": 0.6406509399414062,
      "num_input_tokens_seen": 68271488,
      "step": 240,
      "train_runtime": 2342.3839,
      "train_tokens_per_second": 29146.157
    },
    {
      "epoch": 0.008845506593219477,
      "grad_norm": 0.703125,
      "learning_rate": 4.938888910534637e-05,
      "loss": 0.6345381736755371,
      "num_input_tokens_seen": 71153984,
      "step": 250,
      "train_runtime": 2441.4613,
      "train_tokens_per_second": 29144.014
    },
    {
      "epoch": 0.009199326856948256,
      "grad_norm": 0.7421875,
      "learning_rate": 4.936481223114932e-05,
      "loss": 0.6321357727050781,
      "num_input_tokens_seen": 73939776,
      "step": 260,
      "train_runtime": 2537.8979,
      "train_tokens_per_second": 29134.26
    },
    {
      "epoch": 0.009553147120677036,
      "grad_norm": 0.69921875,
      "learning_rate": 4.934077053477808e-05,
      "loss": 0.6258403778076171,
      "num_input_tokens_seen": 76772352,
      "step": 270,
      "train_runtime": 2637.1103,
      "train_tokens_per_second": 29112.302
    },
    {
      "epoch": 0.009906967384405814,
      "grad_norm": 0.70703125,
      "learning_rate": 4.931676393065431e-05,
      "loss": 0.6268699645996094,
      "num_input_tokens_seen": 79604032,
      "step": 280,
      "train_runtime": 2736.7819,
      "train_tokens_per_second": 29086.728
    },
    {
      "epoch": 0.010260787648134593,
      "grad_norm": 0.68359375,
      "learning_rate": 4.929279233349088e-05,
      "loss": 0.6168413162231445,
      "num_input_tokens_seen": 82418560,
      "step": 290,
      "train_runtime": 2831.5158,
      "train_tokens_per_second": 29107.576
    },
    {
      "epoch": 0.010614607911863373,
      "grad_norm": 0.73046875,
      "learning_rate": 4.926885565829051e-05,
      "loss": 0.619741678237915,
      "num_input_tokens_seen": 85216128,
      "step": 300,
      "train_runtime": 2922.6967,
      "train_tokens_per_second": 29156.679
    },
    {
      "epoch": 0.01096842817559215,
      "grad_norm": 0.69921875,
      "learning_rate": 4.924495382034461e-05,
      "loss": 0.6142029762268066,
      "num_input_tokens_seen": 88064064,
      "step": 310,
      "train_runtime": 3021.69,
      "train_tokens_per_second": 29143.977
    },
    {
      "epoch": 0.01132224843932093,
      "grad_norm": 0.7109375,
      "learning_rate": 4.9221086735231975e-05,
      "loss": 0.614790391921997,
      "num_input_tokens_seen": 90948800,
      "step": 320,
      "train_runtime": 3119.3955,
      "train_tokens_per_second": 29155.905
    },
    {
      "epoch": 0.01167606870304971,
      "grad_norm": 0.73046875,
      "learning_rate": 4.919725431881751e-05,
      "loss": 0.6131664276123047,
      "num_input_tokens_seen": 93845760,
      "step": 330,
      "train_runtime": 3218.1563,
      "train_tokens_per_second": 29161.343
    },
    {
      "epoch": 0.012029888966778488,
      "grad_norm": 0.70703125,
      "learning_rate": 4.917345648725101e-05,
      "loss": 0.6082728862762451,
      "num_input_tokens_seen": 96686912,
      "step": 340,
      "train_runtime": 3314.7421,
      "train_tokens_per_second": 29168.758
    },
    {
      "epoch": 0.012383709230507267,
      "grad_norm": 0.7265625,
      "learning_rate": 4.914969315696596e-05,
      "loss": 0.5998101711273194,
      "num_input_tokens_seen": 99527552,
      "step": 350,
      "train_runtime": 3411.9304,
      "train_tokens_per_second": 29170.452
    },
    {
      "epoch": 0.012737529494236047,
      "grad_norm": 0.72265625,
      "learning_rate": 4.912596424467818e-05,
      "loss": 0.6033230304718018,
      "num_input_tokens_seen": 102450560,
      "step": 360,
      "train_runtime": 3511.7131,
      "train_tokens_per_second": 29173.955
    },
    {
      "epoch": 0.013091349757964827,
      "grad_norm": 0.703125,
      "learning_rate": 4.910226966738475e-05,
      "loss": 0.5978434562683106,
      "num_input_tokens_seen": 105327808,
      "step": 370,
      "train_runtime": 3609.5991,
      "train_tokens_per_second": 29179.919
    },
    {
      "epoch": 0.013445170021693605,
      "grad_norm": 0.734375,
      "learning_rate": 4.9078609342362666e-05,
      "loss": 0.6012052536010742,
      "num_input_tokens_seen": 108158336,
      "step": 380,
      "train_runtime": 3705.6403,
      "train_tokens_per_second": 29187.489
    },
    {
      "epoch": 0.013798990285422384,
      "grad_norm": 0.7265625,
      "learning_rate": 4.905498318716775e-05,
      "loss": 0.596697187423706,
      "num_input_tokens_seen": 111012864,
      "step": 390,
      "train_runtime": 3803.8255,
      "train_tokens_per_second": 29184.531
    },
    {
      "epoch": 0.014152810549151164,
      "grad_norm": 0.69921875,
      "learning_rate": 4.9031391119633295e-05,
      "loss": 0.5964305400848389,
      "num_input_tokens_seen": 113838336,
      "step": 400,
      "train_runtime": 3899.6622,
      "train_tokens_per_second": 29191.846
    },
    {
      "epoch": 0.014506630812879942,
      "grad_norm": 0.72265625,
      "learning_rate": 4.9007833057869e-05,
      "loss": 0.5897239685058594,
      "num_input_tokens_seen": 116627008,
      "step": 410,
      "train_runtime": 3994.5521,
      "train_tokens_per_second": 29196.517
    },
    {
      "epoch": 0.014860451076608721,
      "grad_norm": 0.73046875,
      "learning_rate": 4.898430892025967e-05,
      "loss": 0.595780897140503,
      "num_input_tokens_seen": 119470080,
      "step": 420,
      "train_runtime": 4095.1828,
      "train_tokens_per_second": 29173.32
    },
    {
      "epoch": 0.015214271340337501,
      "grad_norm": 0.71484375,
      "learning_rate": 4.896081862546415e-05,
      "loss": 0.5898738861083984,
      "num_input_tokens_seen": 122313024,
      "step": 430,
      "train_runtime": 4192.2419,
      "train_tokens_per_second": 29176.042
    },
    {
      "epoch": 0.015568091604066279,
      "grad_norm": 0.71484375,
      "learning_rate": 4.8937362092414e-05,
      "loss": 0.5900569915771484,
      "num_input_tokens_seen": 125125248,
      "step": 440,
      "train_runtime": 4285.5509,
      "train_tokens_per_second": 29197.004
    },
    {
      "epoch": 0.01592191186779506,
      "grad_norm": 0.69921875,
      "learning_rate": 4.891393924031244e-05,
      "loss": 0.5830600738525391,
      "num_input_tokens_seen": 127990208,
      "step": 450,
      "train_runtime": 4383.6017,
      "train_tokens_per_second": 29197.5
    },
    {
      "epoch": 0.016275732131523838,
      "grad_norm": 0.70703125,
      "learning_rate": 4.8890549988633095e-05,
      "loss": 0.5887485027313233,
      "num_input_tokens_seen": 130801600,
      "step": 460,
      "train_runtime": 4476.705,
      "train_tokens_per_second": 29218.275
    },
    {
      "epoch": 0.016629552395252618,
      "grad_norm": 0.74609375,
      "learning_rate": 4.8867194257118907e-05,
      "loss": 0.5799453258514404,
      "num_input_tokens_seen": 133635712,
      "step": 470,
      "train_runtime": 4575.8324,
      "train_tokens_per_second": 29204.678
    },
    {
      "epoch": 0.016983372658981397,
      "grad_norm": 0.6640625,
      "learning_rate": 4.884387196578093e-05,
      "loss": 0.5826944828033447,
      "num_input_tokens_seen": 136496704,
      "step": 480,
      "train_runtime": 4674.0084,
      "train_tokens_per_second": 29203.35
    },
    {
      "epoch": 0.017337192922710173,
      "grad_norm": 0.69140625,
      "learning_rate": 4.882058303489718e-05,
      "loss": 0.5787400245666504,
      "num_input_tokens_seen": 139289984,
      "step": 490,
      "train_runtime": 4766.2023,
      "train_tokens_per_second": 29224.522
    },
    {
      "epoch": 0.017691013186438953,
      "grad_norm": 0.73046875,
      "learning_rate": 4.8797327385011496e-05,
      "loss": 0.5779561042785645,
      "num_input_tokens_seen": 142138496,
      "step": 500,
      "train_runtime": 4860.1689,
      "train_tokens_per_second": 29245.588
    },
    {
      "epoch": 0.018044833450167733,
      "grad_norm": 0.671875,
      "learning_rate": 4.8774104936932425e-05,
      "loss": 0.5814637184143067,
      "num_input_tokens_seen": 144959360,
      "step": 510,
      "train_runtime": 4956.0697,
      "train_tokens_per_second": 29248.854
    },
    {
      "epoch": 0.018398653713896512,
      "grad_norm": 0.71875,
      "learning_rate": 4.8750915611732076e-05,
      "loss": 0.5717947006225585,
      "num_input_tokens_seen": 147701312,
      "step": 520,
      "train_runtime": 5045.6694,
      "train_tokens_per_second": 29272.887
    },
    {
      "epoch": 0.018752473977625292,
      "grad_norm": 0.6796875,
      "learning_rate": 4.8727759330744986e-05,
      "loss": 0.57232666015625,
      "num_input_tokens_seen": 150572416,
      "step": 530,
      "train_runtime": 5145.6734,
      "train_tokens_per_second": 29261.946
    },
    {
      "epoch": 0.01910629424135407,
      "grad_norm": 0.6796875,
      "learning_rate": 4.870463601556696e-05,
      "loss": 0.5723038673400879,
      "num_input_tokens_seen": 153366080,
      "step": 540,
      "train_runtime": 5240.2786,
      "train_tokens_per_second": 29266.78
    },
    {
      "epoch": 0.019460114505082848,
      "grad_norm": 0.72265625,
      "learning_rate": 4.8681545588054075e-05,
      "loss": 0.5722018241882324,
      "num_input_tokens_seen": 156225920,
      "step": 550,
      "train_runtime": 5341.4112,
      "train_tokens_per_second": 29248.061
    },
    {
      "epoch": 0.019813934768811627,
      "grad_norm": 0.671875,
      "learning_rate": 4.8658487970321404e-05,
      "loss": 0.5680769920349121,
      "num_input_tokens_seen": 159087552,
      "step": 560,
      "train_runtime": 5441.0533,
      "train_tokens_per_second": 29238.374
    },
    {
      "epoch": 0.020167755032540407,
      "grad_norm": 0.68359375,
      "learning_rate": 4.863546308474209e-05,
      "loss": 0.5642250061035157,
      "num_input_tokens_seen": 161982784,
      "step": 570,
      "train_runtime": 5543.0086,
      "train_tokens_per_second": 29222.9
    },
    {
      "epoch": 0.020521575296269187,
      "grad_norm": 0.69921875,
      "learning_rate": 4.86124708539461e-05,
      "loss": 0.566973876953125,
      "num_input_tokens_seen": 164782400,
      "step": 580,
      "train_runtime": 5639.5169,
      "train_tokens_per_second": 29219.24
    },
    {
      "epoch": 0.020875395559997966,
      "grad_norm": 0.68359375,
      "learning_rate": 4.8589511200819216e-05,
      "loss": 0.5642753601074219,
      "num_input_tokens_seen": 167594880,
      "step": 590,
      "train_runtime": 5732.1297,
      "train_tokens_per_second": 29237.803
    },
    {
      "epoch": 0.021229215823726746,
      "grad_norm": 0.65234375,
      "learning_rate": 4.8566584048501926e-05,
      "loss": 0.5602854728698731,
      "num_input_tokens_seen": 170407296,
      "step": 600,
      "train_runtime": 5828.0015,
      "train_tokens_per_second": 29239.405
    },
    {
      "epoch": 0.021583036087455525,
      "grad_norm": 0.68359375,
      "learning_rate": 4.854368932038835e-05,
      "loss": 0.5584062576293946,
      "num_input_tokens_seen": 173246784,
      "step": 610,
      "train_runtime": 5924.6438,
      "train_tokens_per_second": 29241.721
    },
    {
      "epoch": 0.0219368563511843,
      "grad_norm": 0.70703125,
      "learning_rate": 4.8520826940125144e-05,
      "loss": 0.5647170066833496,
      "num_input_tokens_seen": 176162112,
      "step": 620,
      "train_runtime": 6025.6359,
      "train_tokens_per_second": 29235.439
    },
    {
      "epoch": 0.02229067661491308,
      "grad_norm": 0.6875,
      "learning_rate": 4.849799683161046e-05,
      "loss": 0.5587602615356445,
      "num_input_tokens_seen": 179001984,
      "step": 630,
      "train_runtime": 6120.8074,
      "train_tokens_per_second": 29244.832
    },
    {
      "epoch": 0.02264449687864186,
      "grad_norm": 0.6640625,
      "learning_rate": 4.8475198918992835e-05,
      "loss": 0.5582832813262939,
      "num_input_tokens_seen": 181931712,
      "step": 640,
      "train_runtime": 6222.1443,
      "train_tokens_per_second": 29239.391
    },
    {
      "epoch": 0.02299831714237064,
      "grad_norm": 0.69140625,
      "learning_rate": 4.845243312667023e-05,
      "loss": 0.5552670001983643,
      "num_input_tokens_seen": 184767872,
      "step": 650,
      "train_runtime": 6319.7456,
      "train_tokens_per_second": 29236.6
    },
    {
      "epoch": 0.02335213740609942,
      "grad_norm": 0.68359375,
      "learning_rate": 4.842969937928884e-05,
      "loss": 0.554955005645752,
      "num_input_tokens_seen": 187563008,
      "step": 660,
      "train_runtime": 6414.6287,
      "train_tokens_per_second": 29239.885
    },
    {
      "epoch": 0.0237059576698282,
      "grad_norm": 0.6484375,
      "learning_rate": 4.840699760174217e-05,
      "loss": 0.5562273979187011,
      "num_input_tokens_seen": 190415872,
      "step": 670,
      "train_runtime": 6511.0319,
      "train_tokens_per_second": 29245.114
    },
    {
      "epoch": 0.024059777933556976,
      "grad_norm": 0.62109375,
      "learning_rate": 4.8384327719169906e-05,
      "loss": 0.548237419128418,
      "num_input_tokens_seen": 193276160,
      "step": 680,
      "train_runtime": 6606.2242,
      "train_tokens_per_second": 29256.676
    },
    {
      "epoch": 0.024413598197285755,
      "grad_norm": 0.6953125,
      "learning_rate": 4.836168965695694e-05,
      "loss": 0.5568647384643555,
      "num_input_tokens_seen": 196100352,
      "step": 690,
      "train_runtime": 6703.1311,
      "train_tokens_per_second": 29255.038
    },
    {
      "epoch": 0.024767418461014535,
      "grad_norm": 0.703125,
      "learning_rate": 4.8339083340732304e-05,
      "loss": 0.5538945674896241,
      "num_input_tokens_seen": 198917632,
      "step": 700,
      "train_runtime": 6799.7012,
      "train_tokens_per_second": 29253.878
    },
    {
      "epoch": 0.025121238724743315,
      "grad_norm": 0.65234375,
      "learning_rate": 4.8316508696368154e-05,
      "loss": 0.5495064735412598,
      "num_input_tokens_seen": 201755584,
      "step": 710,
      "train_runtime": 6895.5809,
      "train_tokens_per_second": 29258.678
    },
    {
      "epoch": 0.025475058988472094,
      "grad_norm": 0.67578125,
      "learning_rate": 4.8293965649978714e-05,
      "loss": 0.5516688346862793,
      "num_input_tokens_seen": 204637376,
      "step": 720,
      "train_runtime": 6997.0307,
      "train_tokens_per_second": 29246.317
    },
    {
      "epoch": 0.025828879252200874,
      "grad_norm": 0.6953125,
      "learning_rate": 4.8271454127919364e-05,
      "loss": 0.5563485145568847,
      "num_input_tokens_seen": 207481920,
      "step": 730,
      "train_runtime": 7092.1158,
      "train_tokens_per_second": 29255.292
    },
    {
      "epoch": 0.026182699515929653,
      "grad_norm": 0.6640625,
      "learning_rate": 4.824897405678549e-05,
      "loss": 0.5479368209838867,
      "num_input_tokens_seen": 210324480,
      "step": 740,
      "train_runtime": 7191.5445,
      "train_tokens_per_second": 29246.079
    },
    {
      "epoch": 0.02653651977965843,
      "grad_norm": 0.66015625,
      "learning_rate": 4.8226525363411576e-05,
      "loss": 0.5478935718536377,
      "num_input_tokens_seen": 213174656,
      "step": 750,
      "train_runtime": 7288.099,
      "train_tokens_per_second": 29249.693
    },
    {
      "epoch": 0.02689034004338721,
      "grad_norm": 0.63671875,
      "learning_rate": 4.820410797487017e-05,
      "loss": 0.544947624206543,
      "num_input_tokens_seen": 216051136,
      "step": 760,
      "train_runtime": 7388.5322,
      "train_tokens_per_second": 29241.415
    },
    {
      "epoch": 0.02724416030711599,
      "grad_norm": 0.6640625,
      "learning_rate": 4.818172181847091e-05,
      "loss": 0.5453941345214843,
      "num_input_tokens_seen": 218912640,
      "step": 770,
      "train_runtime": 7487.1763,
      "train_tokens_per_second": 29238.344
    },
    {
      "epoch": 0.02759798057084477,
      "grad_norm": 0.68359375,
      "learning_rate": 4.81593668217595e-05,
      "loss": 0.5468230724334717,
      "num_input_tokens_seen": 221661120,
      "step": 780,
      "train_runtime": 7577.9253,
      "train_tokens_per_second": 29250.898
    },
    {
      "epoch": 0.027951800834573548,
      "grad_norm": 0.671875,
      "learning_rate": 4.813704291251675e-05,
      "loss": 0.5402215957641602,
      "num_input_tokens_seen": 224521408,
      "step": 790,
      "train_runtime": 7672.5981,
      "train_tokens_per_second": 29262.761
    },
    {
      "epoch": 0.028305621098302328,
      "grad_norm": 0.625,
      "learning_rate": 4.811475001875759e-05,
      "loss": 0.5399605751037597,
      "num_input_tokens_seen": 227384512,
      "step": 800,
      "train_runtime": 7772.4854,
      "train_tokens_per_second": 29255.058
    },
    {
      "epoch": 0.028659441362031104,
      "grad_norm": 0.65234375,
      "learning_rate": 4.8092488068730105e-05,
      "loss": 0.5373763561248779,
      "num_input_tokens_seen": 230168128,
      "step": 810,
      "train_runtime": 7867.1237,
      "train_tokens_per_second": 29256.961
    },
    {
      "epoch": 0.029013261625759883,
      "grad_norm": 0.68359375,
      "learning_rate": 4.807025699091452e-05,
      "loss": 0.5417660713195801,
      "num_input_tokens_seen": 233010944,
      "step": 820,
      "train_runtime": 7965.4807,
      "train_tokens_per_second": 29252.59
    },
    {
      "epoch": 0.029367081889488663,
      "grad_norm": 0.67578125,
      "learning_rate": 4.8048056714022325e-05,
      "loss": 0.5361145973205567,
      "num_input_tokens_seen": 235868608,
      "step": 830,
      "train_runtime": 8067.1626,
      "train_tokens_per_second": 29238.112
    },
    {
      "epoch": 0.029720902153217443,
      "grad_norm": 0.6328125,
      "learning_rate": 4.802588716699519e-05,
      "loss": 0.5446301460266113,
      "num_input_tokens_seen": 238670976,
      "step": 840,
      "train_runtime": 8162.6467,
      "train_tokens_per_second": 29239.41
    },
    {
      "epoch": 0.030074722416946222,
      "grad_norm": 0.6640625,
      "learning_rate": 4.8003748279004156e-05,
      "loss": 0.5387370109558105,
      "num_input_tokens_seen": 241592064,
      "step": 850,
      "train_runtime": 8264.6103,
      "train_tokens_per_second": 29232.118
    },
    {
      "epoch": 0.030428542680675002,
      "grad_norm": 0.67578125,
      "learning_rate": 4.798163997944854e-05,
      "loss": 0.5372297286987304,
      "num_input_tokens_seen": 244417792,
      "step": 860,
      "train_runtime": 8358.1194,
      "train_tokens_per_second": 29243.156
    },
    {
      "epoch": 0.030782362944403778,
      "grad_norm": 0.62890625,
      "learning_rate": 4.79595621979551e-05,
      "loss": 0.5457850933074951,
      "num_input_tokens_seen": 247252096,
      "step": 870,
      "train_runtime": 8454.4824,
      "train_tokens_per_second": 29245.09
    },
    {
      "epoch": 0.031136183208132558,
      "grad_norm": 0.65625,
      "learning_rate": 4.793751486437702e-05,
      "loss": 0.5330609321594239,
      "num_input_tokens_seen": 250084480,
      "step": 880,
      "train_runtime": 8550.6343,
      "train_tokens_per_second": 29247.477
    },
    {
      "epoch": 0.03149000347186134,
      "grad_norm": 0.6640625,
      "learning_rate": 4.7915497908793064e-05,
      "loss": 0.5324831962585449,
      "num_input_tokens_seen": 252907136,
      "step": 890,
      "train_runtime": 8647.9808,
      "train_tokens_per_second": 29244.646
    },
    {
      "epoch": 0.03184382373559012,
      "grad_norm": 0.66796875,
      "learning_rate": 4.7893511261506516e-05,
      "loss": 0.5301326751708985,
      "num_input_tokens_seen": 255730432,
      "step": 900,
      "train_runtime": 8745.0541,
      "train_tokens_per_second": 29242.865
    },
    {
      "epoch": 0.032197643999318896,
      "grad_norm": 0.6328125,
      "learning_rate": 4.787155485304435e-05,
      "loss": 0.5360992908477783,
      "num_input_tokens_seen": 258587072,
      "step": 910,
      "train_runtime": 8842.4166,
      "train_tokens_per_second": 29243.937
    },
    {
      "epoch": 0.032551464263047676,
      "grad_norm": 0.71484375,
      "learning_rate": 4.784962861415629e-05,
      "loss": 0.5338759422302246,
      "num_input_tokens_seen": 261403456,
      "step": 920,
      "train_runtime": 8938.3531,
      "train_tokens_per_second": 29245.148
    },
    {
      "epoch": 0.032905284526776456,
      "grad_norm": 0.69140625,
      "learning_rate": 4.7827732475813884e-05,
      "loss": 0.5312234878540039,
      "num_input_tokens_seen": 264245888,
      "step": 930,
      "train_runtime": 9035.3917,
      "train_tokens_per_second": 29245.648
    },
    {
      "epoch": 0.033259104790505235,
      "grad_norm": 0.66796875,
      "learning_rate": 4.7805866369209576e-05,
      "loss": 0.5281752109527588,
      "num_input_tokens_seen": 267068608,
      "step": 940,
      "train_runtime": 9132.2426,
      "train_tokens_per_second": 29244.581
    },
    {
      "epoch": 0.033612925054234015,
      "grad_norm": 0.6328125,
      "learning_rate": 4.778403022575583e-05,
      "loss": 0.530787992477417,
      "num_input_tokens_seen": 269879616,
      "step": 950,
      "train_runtime": 9226.0258,
      "train_tokens_per_second": 29251.99
    },
    {
      "epoch": 0.033966745317962795,
      "grad_norm": 0.640625,
      "learning_rate": 4.7762223977084195e-05,
      "loss": 0.5359733581542969,
      "num_input_tokens_seen": 272747456,
      "step": 960,
      "train_runtime": 9326.7801,
      "train_tokens_per_second": 29243.474
    },
    {
      "epoch": 0.03432056558169157,
      "grad_norm": 0.65625,
      "learning_rate": 4.774044755504444e-05,
      "loss": 0.5251823902130127,
      "num_input_tokens_seen": 275555264,
      "step": 970,
      "train_runtime": 9422.2686,
      "train_tokens_per_second": 29245.108
    },
    {
      "epoch": 0.03467438584542035,
      "grad_norm": 0.58984375,
      "learning_rate": 4.7718700891703616e-05,
      "loss": 0.5232102870941162,
      "num_input_tokens_seen": 278383808,
      "step": 980,
      "train_runtime": 9516.932,
      "train_tokens_per_second": 29251.423
    },
    {
      "epoch": 0.035028206109149126,
      "grad_norm": 0.65234375,
      "learning_rate": 4.7696983919345215e-05,
      "loss": 0.5289015769958496,
      "num_input_tokens_seen": 281244672,
      "step": 990,
      "train_runtime": 9617.3733,
      "train_tokens_per_second": 29243.398
    },
    {
      "epoch": 0.035382026372877906,
      "grad_norm": 0.6328125,
      "learning_rate": 4.7675296570468216e-05,
      "loss": 0.5262557029724121,
      "num_input_tokens_seen": 284070144,
      "step": 1000,
      "train_runtime": 9714.088,
      "train_tokens_per_second": 29243.11
    },
    {
      "epoch": 0.035735846636606686,
      "grad_norm": 0.609375,
      "learning_rate": 4.76536387777863e-05,
      "loss": 0.5245723724365234,
      "num_input_tokens_seen": 286892224,
      "step": 1010,
      "train_runtime": 9809.3995,
      "train_tokens_per_second": 29246.665
    },
    {
      "epoch": 0.036089666900335465,
      "grad_norm": 0.640625,
      "learning_rate": 4.7632010474226915e-05,
      "loss": 0.5246537685394287,
      "num_input_tokens_seen": 289760768,
      "step": 1020,
      "train_runtime": 9910.5607,
      "train_tokens_per_second": 29237.576
    },
    {
      "epoch": 0.036443487164064245,
      "grad_norm": 0.65234375,
      "learning_rate": 4.761041159293035e-05,
      "loss": 0.5270932197570801,
      "num_input_tokens_seen": 292580992,
      "step": 1030,
      "train_runtime": 10006.494,
      "train_tokens_per_second": 29239.111
    },
    {
      "epoch": 0.036797307427793025,
      "grad_norm": 0.625,
      "learning_rate": 4.7588842067249e-05,
      "loss": 0.524391508102417,
      "num_input_tokens_seen": 295430272,
      "step": 1040,
      "train_runtime": 10106.4067,
      "train_tokens_per_second": 29231.979
    },
    {
      "epoch": 0.037151127691521804,
      "grad_norm": 0.63671875,
      "learning_rate": 4.756730183074637e-05,
      "loss": 0.5262633323669433,
      "num_input_tokens_seen": 298248704,
      "step": 1050,
      "train_runtime": 10201.3863,
      "train_tokens_per_second": 29236.095
    },
    {
      "epoch": 0.037504947955250584,
      "grad_norm": 0.64453125,
      "learning_rate": 4.7545790817196314e-05,
      "loss": 0.5237319946289063,
      "num_input_tokens_seen": 301097088,
      "step": 1060,
      "train_runtime": 10300.4084,
      "train_tokens_per_second": 29231.568
    },
    {
      "epoch": 0.03785876821897936,
      "grad_norm": 0.66015625,
      "learning_rate": 4.752430896058212e-05,
      "loss": 0.5183588027954101,
      "num_input_tokens_seen": 303881344,
      "step": 1070,
      "train_runtime": 10393.8696,
      "train_tokens_per_second": 29236.594
    },
    {
      "epoch": 0.03821258848270814,
      "grad_norm": 0.640625,
      "learning_rate": 4.750285619509567e-05,
      "loss": 0.5271801471710205,
      "num_input_tokens_seen": 306701696,
      "step": 1080,
      "train_runtime": 10490.9209,
      "train_tokens_per_second": 29234.964
    },
    {
      "epoch": 0.03856640874643692,
      "grad_norm": 0.6484375,
      "learning_rate": 4.7481432455136644e-05,
      "loss": 0.5265097618103027,
      "num_input_tokens_seen": 309561344,
      "step": 1090,
      "train_runtime": 10590.9189,
      "train_tokens_per_second": 29228.941
    },
    {
      "epoch": 0.038920229010165695,
      "grad_norm": 0.63671875,
      "learning_rate": 4.7460037675311584e-05,
      "loss": 0.52159423828125,
      "num_input_tokens_seen": 312447872,
      "step": 1100,
      "train_runtime": 10691.3702,
      "train_tokens_per_second": 29224.306
    },
    {
      "epoch": 0.039274049273894475,
      "grad_norm": 0.62890625,
      "learning_rate": 4.7438671790433126e-05,
      "loss": 0.5192754745483399,
      "num_input_tokens_seen": 315278400,
      "step": 1110,
      "train_runtime": 10788.4584,
      "train_tokens_per_second": 29223.675
    },
    {
      "epoch": 0.039627869537623255,
      "grad_norm": 0.609375,
      "learning_rate": 4.741733473551915e-05,
      "loss": 0.5120278835296631,
      "num_input_tokens_seen": 318156160,
      "step": 1120,
      "train_runtime": 10887.3016,
      "train_tokens_per_second": 29222.683
    },
    {
      "epoch": 0.039981689801352034,
      "grad_norm": 0.65625,
      "learning_rate": 4.7396026445791966e-05,
      "loss": 0.5187259674072265,
      "num_input_tokens_seen": 321007808,
      "step": 1130,
      "train_runtime": 10985.723,
      "train_tokens_per_second": 29220.454
    },
    {
      "epoch": 0.040335510065080814,
      "grad_norm": 0.671875,
      "learning_rate": 4.737474685667742e-05,
      "loss": 0.5205765724182129,
      "num_input_tokens_seen": 323854016,
      "step": 1140,
      "train_runtime": 11083.4517,
      "train_tokens_per_second": 29219.599
    },
    {
      "epoch": 0.04068933032880959,
      "grad_norm": 0.65234375,
      "learning_rate": 4.7353495903804165e-05,
      "loss": 0.5147505760192871,
      "num_input_tokens_seen": 326739072,
      "step": 1150,
      "train_runtime": 11185.6144,
      "train_tokens_per_second": 29210.651
    },
    {
      "epoch": 0.04104315059253837,
      "grad_norm": 0.66015625,
      "learning_rate": 4.733227352300277e-05,
      "loss": 0.5157520294189453,
      "num_input_tokens_seen": 329584000,
      "step": 1160,
      "train_runtime": 11282.661,
      "train_tokens_per_second": 29211.549
    },
    {
      "epoch": 0.04139697085626715,
      "grad_norm": 0.63671875,
      "learning_rate": 4.731107965030496e-05,
      "loss": 0.5139248847961426,
      "num_input_tokens_seen": 332447488,
      "step": 1170,
      "train_runtime": 11381.2166,
      "train_tokens_per_second": 29210.189
    },
    {
      "epoch": 0.04175079111999593,
      "grad_norm": 0.65234375,
      "learning_rate": 4.728991422194278e-05,
      "loss": 0.5168056488037109,
      "num_input_tokens_seen": 335291008,
      "step": 1180,
      "train_runtime": 11479.3888,
      "train_tokens_per_second": 29208.089
    },
    {
      "epoch": 0.04210461138372471,
      "grad_norm": 0.65625,
      "learning_rate": 4.726877717434773e-05,
      "loss": 0.5130697250366211,
      "num_input_tokens_seen": 338140352,
      "step": 1190,
      "train_runtime": 11578.4203,
      "train_tokens_per_second": 29204.36
    },
    {
      "epoch": 0.04245843164745349,
      "grad_norm": 0.62890625,
      "learning_rate": 4.724766844415013e-05,
      "loss": 0.515099573135376,
      "num_input_tokens_seen": 340990272,
      "step": 1200,
      "train_runtime": 11677.4394,
      "train_tokens_per_second": 29200.774
    },
    {
      "epoch": 0.04281225191118227,
      "grad_norm": 0.66015625,
      "learning_rate": 4.722658796817813e-05,
      "loss": 0.5136539936065674,
      "num_input_tokens_seen": 343838208,
      "step": 1210,
      "train_runtime": 11775.0554,
      "train_tokens_per_second": 29200.56
    },
    {
      "epoch": 0.04316607217491105,
      "grad_norm": 0.65234375,
      "learning_rate": 4.7205535683457044e-05,
      "loss": 0.5174936294555664,
      "num_input_tokens_seen": 346657792,
      "step": 1220,
      "train_runtime": 11873.8516,
      "train_tokens_per_second": 29195.059
    },
    {
      "epoch": 0.04351989243863982,
      "grad_norm": 0.62109375,
      "learning_rate": 4.7184511527208484e-05,
      "loss": 0.5219906806945801,
      "num_input_tokens_seen": 349406848,
      "step": 1230,
      "train_runtime": 11966.1139,
      "train_tokens_per_second": 29199.693
    },
    {
      "epoch": 0.0438737127023686,
      "grad_norm": 0.61328125,
      "learning_rate": 4.7163515436849644e-05,
      "loss": 0.5125075817108155,
      "num_input_tokens_seen": 352205120,
      "step": 1240,
      "train_runtime": 12058.3514,
      "train_tokens_per_second": 29208.397
    },
    {
      "epoch": 0.04422753296609738,
      "grad_norm": 0.625,
      "learning_rate": 4.714254734999245e-05,
      "loss": 0.5103229999542236,
      "num_input_tokens_seen": 355032320,
      "step": 1250,
      "train_runtime": 12156.101,
      "train_tokens_per_second": 29206.101
    },
    {
      "epoch": 0.04458135322982616,
      "grad_norm": 0.62109375,
      "learning_rate": 4.712160720444284e-05,
      "loss": 0.5108553886413574,
      "num_input_tokens_seen": 357865728,
      "step": 1260,
      "train_runtime": 12251.671,
      "train_tokens_per_second": 29209.544
    },
    {
      "epoch": 0.04493517349355494,
      "grad_norm": 0.6015625,
      "learning_rate": 4.710069493819992e-05,
      "loss": 0.511588191986084,
      "num_input_tokens_seen": 360726208,
      "step": 1270,
      "train_runtime": 12347.9074,
      "train_tokens_per_second": 29213.55
    },
    {
      "epoch": 0.04528899375728372,
      "grad_norm": 0.6328125,
      "learning_rate": 4.70798104894553e-05,
      "loss": 0.5121116638183594,
      "num_input_tokens_seen": 363523776,
      "step": 1280,
      "train_runtime": 12443.8328,
      "train_tokens_per_second": 29213.168
    },
    {
      "epoch": 0.0456428140210125,
      "grad_norm": 0.6015625,
      "learning_rate": 4.705895379659219e-05,
      "loss": 0.5085021495819092,
      "num_input_tokens_seen": 366363840,
      "step": 1290,
      "train_runtime": 12540.1612,
      "train_tokens_per_second": 29215.242
    },
    {
      "epoch": 0.04599663428474128,
      "grad_norm": 0.62109375,
      "learning_rate": 4.7038124798184766e-05,
      "loss": 0.5098957061767578,
      "num_input_tokens_seen": 369254272,
      "step": 1300,
      "train_runtime": 12639.1347,
      "train_tokens_per_second": 29215.155
    },
    {
      "epoch": 0.04635045454847006,
      "grad_norm": 0.66015625,
      "learning_rate": 4.7017323432997304e-05,
      "loss": 0.5119119167327881,
      "num_input_tokens_seen": 372060416,
      "step": 1310,
      "train_runtime": 12732.7391,
      "train_tokens_per_second": 29220.768
    },
    {
      "epoch": 0.04670427481219884,
      "grad_norm": 0.6171875,
      "learning_rate": 4.6996549639983506e-05,
      "loss": 0.5074324607849121,
      "num_input_tokens_seen": 374916672,
      "step": 1320,
      "train_runtime": 12831.2097,
      "train_tokens_per_second": 29219.121
    },
    {
      "epoch": 0.04705809507592762,
      "grad_norm": 0.63671875,
      "learning_rate": 4.697580335828569e-05,
      "loss": 0.5119407653808594,
      "num_input_tokens_seen": 377688768,
      "step": 1330,
      "train_runtime": 12924.7716,
      "train_tokens_per_second": 29222.084
    },
    {
      "epoch": 0.0474119153396564,
      "grad_norm": 0.60546875,
      "learning_rate": 4.6955084527234076e-05,
      "loss": 0.5067808151245117,
      "num_input_tokens_seen": 380504704,
      "step": 1340,
      "train_runtime": 13021.8795,
      "train_tokens_per_second": 29220.414
    },
    {
      "epoch": 0.04776573560338518,
      "grad_norm": 0.62109375,
      "learning_rate": 4.6934393086346034e-05,
      "loss": 0.5052962303161621,
      "num_input_tokens_seen": 383362752,
      "step": 1350,
      "train_runtime": 13119.4099,
      "train_tokens_per_second": 29221.036
    },
    {
      "epoch": 0.04811955586711395,
      "grad_norm": 0.6328125,
      "learning_rate": 4.6913728975325324e-05,
      "loss": 0.5087544441223144,
      "num_input_tokens_seen": 386168256,
      "step": 1360,
      "train_runtime": 13214.8992,
      "train_tokens_per_second": 29222.187
    },
    {
      "epoch": 0.04847337613084273,
      "grad_norm": 0.6171875,
      "learning_rate": 4.6893092134061393e-05,
      "loss": 0.5076364517211914,
      "num_input_tokens_seen": 389013440,
      "step": 1370,
      "train_runtime": 13309.6751,
      "train_tokens_per_second": 29227.869
    },
    {
      "epoch": 0.04882719639457151,
      "grad_norm": 0.63671875,
      "learning_rate": 4.687248250262859e-05,
      "loss": 0.5082167625427246,
      "num_input_tokens_seen": 391862656,
      "step": 1380,
      "train_runtime": 13407.5273,
      "train_tokens_per_second": 29227.064
    },
    {
      "epoch": 0.04918101665830029,
      "grad_norm": 0.6484375,
      "learning_rate": 4.685190002128548e-05,
      "loss": 0.5068390846252442,
      "num_input_tokens_seen": 394694016,
      "step": 1390,
      "train_runtime": 13504.5689,
      "train_tokens_per_second": 29226.702
    },
    {
      "epoch": 0.04953483692202907,
      "grad_norm": 0.61328125,
      "learning_rate": 4.6831344630474114e-05,
      "loss": 0.5044609546661377,
      "num_input_tokens_seen": 397559168,
      "step": 1400,
      "train_runtime": 13603.1765,
      "train_tokens_per_second": 29225.466
    },
    {
      "epoch": 0.04988865718575785,
      "grad_norm": 0.63671875,
      "learning_rate": 4.6810816270819276e-05,
      "loss": 0.5071157455444336,
      "num_input_tokens_seen": 400339968,
      "step": 1410,
      "train_runtime": 13696.3087,
      "train_tokens_per_second": 29229.771
    },
    {
      "epoch": 0.05024247744948663,
      "grad_norm": 0.61328125,
      "learning_rate": 4.679031488312777e-05,
      "loss": 0.503534984588623,
      "num_input_tokens_seen": 403147712,
      "step": 1420,
      "train_runtime": 13788.2059,
      "train_tokens_per_second": 29238.591
    },
    {
      "epoch": 0.05059629771321541,
      "grad_norm": 0.6015625,
      "learning_rate": 4.6769840408387717e-05,
      "loss": 0.50531005859375,
      "num_input_tokens_seen": 405921984,
      "step": 1430,
      "train_runtime": 13882.0958,
      "train_tokens_per_second": 29240.685
    },
    {
      "epoch": 0.05095011797694419,
      "grad_norm": 0.609375,
      "learning_rate": 4.674939278776787e-05,
      "loss": 0.5033230781555176,
      "num_input_tokens_seen": 408771328,
      "step": 1440,
      "train_runtime": 13981.6214,
      "train_tokens_per_second": 29236.332
    },
    {
      "epoch": 0.05130393824067297,
      "grad_norm": 0.61328125,
      "learning_rate": 4.672897196261683e-05,
      "loss": 0.497056770324707,
      "num_input_tokens_seen": 411631360,
      "step": 1450,
      "train_runtime": 14076.8573,
      "train_tokens_per_second": 29241.709
    },
    {
      "epoch": 0.05165775850440175,
      "grad_norm": 0.60546875,
      "learning_rate": 4.670857787446238e-05,
      "loss": 0.5002260208129883,
      "num_input_tokens_seen": 414472512,
      "step": 1460,
      "train_runtime": 14174.0097,
      "train_tokens_per_second": 29241.726
    },
    {
      "epoch": 0.05201157876813053,
      "grad_norm": 0.60546875,
      "learning_rate": 4.668821046501082e-05,
      "loss": 0.5049727439880372,
      "num_input_tokens_seen": 417288512,
      "step": 1470,
      "train_runtime": 14267.9159,
      "train_tokens_per_second": 29246.634
    },
    {
      "epoch": 0.05236539903185931,
      "grad_norm": 0.6171875,
      "learning_rate": 4.6667869676146194e-05,
      "loss": 0.5048169136047364,
      "num_input_tokens_seen": 420114432,
      "step": 1480,
      "train_runtime": 14364.9711,
      "train_tokens_per_second": 29245.755
    },
    {
      "epoch": 0.05271921929558808,
      "grad_norm": 0.62890625,
      "learning_rate": 4.6647555449929645e-05,
      "loss": 0.5045809268951416,
      "num_input_tokens_seen": 423016960,
      "step": 1490,
      "train_runtime": 14464.8637,
      "train_tokens_per_second": 29244.448
    },
    {
      "epoch": 0.05307303955931686,
      "grad_norm": 0.625,
      "learning_rate": 4.662726772859869e-05,
      "loss": 0.5020414352416992,
      "num_input_tokens_seen": 425864448,
      "step": 1500,
      "train_runtime": 14561.586,
      "train_tokens_per_second": 29245.746
    },
    {
      "epoch": 0.05342685982304564,
      "grad_norm": 0.6015625,
      "learning_rate": 4.660700645456655e-05,
      "loss": 0.5014071941375733,
      "num_input_tokens_seen": 428695872,
      "step": 1510,
      "train_runtime": 14660.0558,
      "train_tokens_per_second": 29242.445
    },
    {
      "epoch": 0.05378068008677442,
      "grad_norm": 0.6484375,
      "learning_rate": 4.658677157042149e-05,
      "loss": 0.5045583724975586,
      "num_input_tokens_seen": 431512576,
      "step": 1520,
      "train_runtime": 14756.1916,
      "train_tokens_per_second": 29242.815
    },
    {
      "epoch": 0.0541345003505032,
      "grad_norm": 0.609375,
      "learning_rate": 4.656656301892605e-05,
      "loss": 0.5063568592071533,
      "num_input_tokens_seen": 434367744,
      "step": 1530,
      "train_runtime": 14854.9388,
      "train_tokens_per_second": 29240.628
    },
    {
      "epoch": 0.05448832061423198,
      "grad_norm": 0.62890625,
      "learning_rate": 4.6546380743016465e-05,
      "loss": 0.5017670631408692,
      "num_input_tokens_seen": 437205952,
      "step": 1540,
      "train_runtime": 14953.2632,
      "train_tokens_per_second": 29238.163
    },
    {
      "epoch": 0.05484214087796076,
      "grad_norm": 0.62890625,
      "learning_rate": 4.652622468580193e-05,
      "loss": 0.4997280120849609,
      "num_input_tokens_seen": 440075648,
      "step": 1550,
      "train_runtime": 15053.6898,
      "train_tokens_per_second": 29233.74
    },
    {
      "epoch": 0.05519596114168954,
      "grad_norm": 0.625,
      "learning_rate": 4.650609479056392e-05,
      "loss": 0.4976215839385986,
      "num_input_tokens_seen": 442901888,
      "step": 1560,
      "train_runtime": 15149.6237,
      "train_tokens_per_second": 29235.174
    },
    {
      "epoch": 0.055549781405418316,
      "grad_norm": 0.60546875,
      "learning_rate": 4.648599100075556e-05,
      "loss": 0.5020614624023437,
      "num_input_tokens_seen": 445732352,
      "step": 1570,
      "train_runtime": 15245.6944,
      "train_tokens_per_second": 29236.605
    },
    {
      "epoch": 0.055903601669147096,
      "grad_norm": 0.5859375,
      "learning_rate": 4.6465913260000945e-05,
      "loss": 0.4983253002166748,
      "num_input_tokens_seen": 448596480,
      "step": 1580,
      "train_runtime": 15345.71,
      "train_tokens_per_second": 29232.696
    },
    {
      "epoch": 0.056257421932875876,
      "grad_norm": 0.625,
      "learning_rate": 4.644586151209444e-05,
      "loss": 0.4939885139465332,
      "num_input_tokens_seen": 451430272,
      "step": 1590,
      "train_runtime": 15443.4348,
      "train_tokens_per_second": 29231.209
    },
    {
      "epoch": 0.056611242196604655,
      "grad_norm": 0.609375,
      "learning_rate": 4.6425835701000084e-05,
      "loss": 0.49611196517944334,
      "num_input_tokens_seen": 454233152,
      "step": 1600,
      "train_runtime": 15535.6015,
      "train_tokens_per_second": 29238.208
    },
    {
      "epoch": 0.05696506246033343,
      "grad_norm": 0.6484375,
      "learning_rate": 4.640583577085084e-05,
      "loss": 0.4994488716125488,
      "num_input_tokens_seen": 457093248,
      "step": 1610,
      "train_runtime": 15632.9592,
      "train_tokens_per_second": 29239.074
    },
    {
      "epoch": 0.05731888272406221,
      "grad_norm": 0.625,
      "learning_rate": 4.638586166594806e-05,
      "loss": 0.4968722343444824,
      "num_input_tokens_seen": 459936448,
      "step": 1620,
      "train_runtime": 15729.9043,
      "train_tokens_per_second": 29239.622
    },
    {
      "epoch": 0.05767270298779099,
      "grad_norm": 0.578125,
      "learning_rate": 4.6365913330760726e-05,
      "loss": 0.49298572540283203,
      "num_input_tokens_seen": 462743744,
      "step": 1630,
      "train_runtime": 15826.5709,
      "train_tokens_per_second": 29238.408
    },
    {
      "epoch": 0.05802652325151977,
      "grad_norm": 0.62890625,
      "learning_rate": 4.6345990709924855e-05,
      "loss": 0.498442554473877,
      "num_input_tokens_seen": 465583680,
      "step": 1640,
      "train_runtime": 15923.9936,
      "train_tokens_per_second": 29237.872
    },
    {
      "epoch": 0.058380343515248546,
      "grad_norm": 0.609375,
      "learning_rate": 4.632609374824284e-05,
      "loss": 0.49790544509887696,
      "num_input_tokens_seen": 468450816,
      "step": 1650,
      "train_runtime": 16022.7153,
      "train_tokens_per_second": 29236.668
    },
    {
      "epoch": 0.058734163778977326,
      "grad_norm": 0.6015625,
      "learning_rate": 4.630622239068285e-05,
      "loss": 0.49540348052978517,
      "num_input_tokens_seen": 471258944,
      "step": 1660,
      "train_runtime": 16119.9432,
      "train_tokens_per_second": 29234.529
    },
    {
      "epoch": 0.059087984042706106,
      "grad_norm": 0.578125,
      "learning_rate": 4.628637658237808e-05,
      "loss": 0.4915264129638672,
      "num_input_tokens_seen": 474096320,
      "step": 1670,
      "train_runtime": 16219.2057,
      "train_tokens_per_second": 29230.551
    },
    {
      "epoch": 0.059441804306434885,
      "grad_norm": 0.609375,
      "learning_rate": 4.626655626862625e-05,
      "loss": 0.49178447723388674,
      "num_input_tokens_seen": 476921472,
      "step": 1680,
      "train_runtime": 16315.9306,
      "train_tokens_per_second": 29230.418
    },
    {
      "epoch": 0.059795624570163665,
      "grad_norm": 0.59375,
      "learning_rate": 4.624676139488888e-05,
      "loss": 0.49329314231872556,
      "num_input_tokens_seen": 479802176,
      "step": 1690,
      "train_runtime": 16416.7699,
      "train_tokens_per_second": 29226.345
    },
    {
      "epoch": 0.060149444833892445,
      "grad_norm": 0.609375,
      "learning_rate": 4.6226991906790686e-05,
      "loss": 0.49400696754455564,
      "num_input_tokens_seen": 482621056,
      "step": 1700,
      "train_runtime": 16512.7673,
      "train_tokens_per_second": 29227.146
    },
    {
      "epoch": 0.060503265097621224,
      "grad_norm": 0.61328125,
      "learning_rate": 4.620724775011897e-05,
      "loss": 0.49402532577514646,
      "num_input_tokens_seen": 485433280,
      "step": 1710,
      "train_runtime": 16606.7295,
      "train_tokens_per_second": 29231.119
    },
    {
      "epoch": 0.060857085361350004,
      "grad_norm": 0.59375,
      "learning_rate": 4.618752887082297e-05,
      "loss": 0.48993425369262694,
      "num_input_tokens_seen": 488321600,
      "step": 1720,
      "train_runtime": 16706.35,
      "train_tokens_per_second": 29229.7
    },
    {
      "epoch": 0.06121090562507878,
      "grad_norm": 0.63671875,
      "learning_rate": 4.616783521501325e-05,
      "loss": 0.4922123908996582,
      "num_input_tokens_seen": 491185728,
      "step": 1730,
      "train_runtime": 16806.1119,
      "train_tokens_per_second": 29226.613
    },
    {
      "epoch": 0.061564725888807556,
      "grad_norm": 0.6171875,
      "learning_rate": 4.614816672896108e-05,
      "loss": 0.49462089538574217,
      "num_input_tokens_seen": 494032640,
      "step": 1740,
      "train_runtime": 16903.6141,
      "train_tokens_per_second": 29226.45
    },
    {
      "epoch": 0.061918546152536336,
      "grad_norm": 0.6015625,
      "learning_rate": 4.612852335909782e-05,
      "loss": 0.49348812103271483,
      "num_input_tokens_seen": 496898944,
      "step": 1750,
      "train_runtime": 16999.5825,
      "train_tokens_per_second": 29230.067
    },
    {
      "epoch": 0.062272366416265115,
      "grad_norm": 0.60546875,
      "learning_rate": 4.6108905052014323e-05,
      "loss": 0.49039106369018554,
      "num_input_tokens_seen": 499705920,
      "step": 1760,
      "train_runtime": 17092.9555,
      "train_tokens_per_second": 29234.612
    },
    {
      "epoch": 0.0626261866799939,
      "grad_norm": 0.609375,
      "learning_rate": 4.608931175446027e-05,
      "loss": 0.48816871643066406,
      "num_input_tokens_seen": 502554560,
      "step": 1770,
      "train_runtime": 17189.6723,
      "train_tokens_per_second": 29235.843
    },
    {
      "epoch": 0.06298000694372267,
      "grad_norm": 0.62109375,
      "learning_rate": 4.606974341334367e-05,
      "loss": 0.4953399658203125,
      "num_input_tokens_seen": 505327296,
      "step": 1780,
      "train_runtime": 17284.2912,
      "train_tokens_per_second": 29236.217
    },
    {
      "epoch": 0.06333382720745145,
      "grad_norm": 0.60546875,
      "learning_rate": 4.605019997573011e-05,
      "loss": 0.48953962326049805,
      "num_input_tokens_seen": 508246208,
      "step": 1790,
      "train_runtime": 17385.6776,
      "train_tokens_per_second": 29233.615
    },
    {
      "epoch": 0.06368764747118023,
      "grad_norm": 0.6015625,
      "learning_rate": 4.603068138884229e-05,
      "loss": 0.4889236927032471,
      "num_input_tokens_seen": 511120896,
      "step": 1800,
      "train_runtime": 17485.6064,
      "train_tokens_per_second": 29230.95
    },
    {
      "epoch": 0.06404146773490901,
      "grad_norm": 0.6171875,
      "learning_rate": 4.6011187600059345e-05,
      "loss": 0.49122133255004885,
      "num_input_tokens_seen": 513971072,
      "step": 1810,
      "train_runtime": 17580.9085,
      "train_tokens_per_second": 29234.614
    },
    {
      "epoch": 0.06439528799863779,
      "grad_norm": 0.62109375,
      "learning_rate": 4.599171855691629e-05,
      "loss": 0.4832293510437012,
      "num_input_tokens_seen": 516781632,
      "step": 1820,
      "train_runtime": 17677.2972,
      "train_tokens_per_second": 29234.199
    },
    {
      "epoch": 0.06474910826236657,
      "grad_norm": 0.6015625,
      "learning_rate": 4.597227420710335e-05,
      "loss": 0.48655948638916013,
      "num_input_tokens_seen": 519587648,
      "step": 1830,
      "train_runtime": 17773.3819,
      "train_tokens_per_second": 29234.034
    },
    {
      "epoch": 0.06510292852609535,
      "grad_norm": 0.609375,
      "learning_rate": 4.595285449846551e-05,
      "loss": 0.4880838394165039,
      "num_input_tokens_seen": 522408064,
      "step": 1840,
      "train_runtime": 17869.6001,
      "train_tokens_per_second": 29234.458
    },
    {
      "epoch": 0.06545674878982413,
      "grad_norm": 0.60546875,
      "learning_rate": 4.593345937900178e-05,
      "loss": 0.48679656982421876,
      "num_input_tokens_seen": 525223808,
      "step": 1850,
      "train_runtime": 17963.8289,
      "train_tokens_per_second": 29237.854
    },
    {
      "epoch": 0.06581056905355291,
      "grad_norm": 0.609375,
      "learning_rate": 4.591408879686472e-05,
      "loss": 0.4835360050201416,
      "num_input_tokens_seen": 527996032,
      "step": 1860,
      "train_runtime": 18055.9966,
      "train_tokens_per_second": 29242.143
    },
    {
      "epoch": 0.06616438931728169,
      "grad_norm": 0.58984375,
      "learning_rate": 4.5894742700359775e-05,
      "loss": 0.48793563842773435,
      "num_input_tokens_seen": 530825152,
      "step": 1870,
      "train_runtime": 18149.8895,
      "train_tokens_per_second": 29246.743
    },
    {
      "epoch": 0.06651820958101047,
      "grad_norm": 0.625,
      "learning_rate": 4.587542103794477e-05,
      "loss": 0.48436679840087893,
      "num_input_tokens_seen": 533647104,
      "step": 1880,
      "train_runtime": 18247.2758,
      "train_tokens_per_second": 29245.303
    },
    {
      "epoch": 0.06687202984473925,
      "grad_norm": 0.62109375,
      "learning_rate": 4.5856123758229247e-05,
      "loss": 0.48836069107055663,
      "num_input_tokens_seen": 536478464,
      "step": 1890,
      "train_runtime": 18342.7489,
      "train_tokens_per_second": 29247.441
    },
    {
      "epoch": 0.06722585010846803,
      "grad_norm": 0.6171875,
      "learning_rate": 4.5836850809973993e-05,
      "loss": 0.4861155033111572,
      "num_input_tokens_seen": 539351808,
      "step": 1900,
      "train_runtime": 18441.8215,
      "train_tokens_per_second": 29246.124
    },
    {
      "epoch": 0.06757967037219681,
      "grad_norm": 0.625,
      "learning_rate": 4.5817602142090385e-05,
      "loss": 0.4900822639465332,
      "num_input_tokens_seen": 542161408,
      "step": 1910,
      "train_runtime": 18539.2971,
      "train_tokens_per_second": 29243.903
    },
    {
      "epoch": 0.06793349063592559,
      "grad_norm": 0.60546875,
      "learning_rate": 4.579837770363989e-05,
      "loss": 0.48744726181030273,
      "num_input_tokens_seen": 544951744,
      "step": 1920,
      "train_runtime": 18630.5695,
      "train_tokens_per_second": 29250.407
    },
    {
      "epoch": 0.06828731089965437,
      "grad_norm": 0.6015625,
      "learning_rate": 4.57791774438334e-05,
      "loss": 0.4854752063751221,
      "num_input_tokens_seen": 547750400,
      "step": 1930,
      "train_runtime": 18726.7333,
      "train_tokens_per_second": 29249.65
    },
    {
      "epoch": 0.06864113116338313,
      "grad_norm": 0.58984375,
      "learning_rate": 4.576000131203078e-05,
      "loss": 0.49063758850097655,
      "num_input_tokens_seen": 550594112,
      "step": 1940,
      "train_runtime": 18825.4178,
      "train_tokens_per_second": 29247.378
    },
    {
      "epoch": 0.06899495142711191,
      "grad_norm": 0.59765625,
      "learning_rate": 4.574084925774023e-05,
      "loss": 0.4877795696258545,
      "num_input_tokens_seen": 553487040,
      "step": 1950,
      "train_runtime": 18926.0416,
      "train_tokens_per_second": 29244.733
    },
    {
      "epoch": 0.0693487716908407,
      "grad_norm": 0.609375,
      "learning_rate": 4.5721721230617795e-05,
      "loss": 0.4819538116455078,
      "num_input_tokens_seen": 556323008,
      "step": 1960,
      "train_runtime": 19025.3704,
      "train_tokens_per_second": 29241.113
    },
    {
      "epoch": 0.06970259195456947,
      "grad_norm": 0.625,
      "learning_rate": 4.57026171804667e-05,
      "loss": 0.48713436126708987,
      "num_input_tokens_seen": 559121728,
      "step": 1970,
      "train_runtime": 19121.9829,
      "train_tokens_per_second": 29239.736
    },
    {
      "epoch": 0.07005641221829825,
      "grad_norm": 0.63671875,
      "learning_rate": 4.568353705723692e-05,
      "loss": 0.47935781478881834,
      "num_input_tokens_seen": 562017152,
      "step": 1980,
      "train_runtime": 19221.5884,
      "train_tokens_per_second": 29238.851
    },
    {
      "epoch": 0.07041023248202703,
      "grad_norm": 0.6171875,
      "learning_rate": 4.566448081102455e-05,
      "loss": 0.48120651245117185,
      "num_input_tokens_seen": 564825344,
      "step": 1990,
      "train_runtime": 19315.8829,
      "train_tokens_per_second": 29241.498
    },
    {
      "epoch": 0.07076405274575581,
      "grad_norm": 0.59765625,
      "learning_rate": 4.564544839207128e-05,
      "loss": 0.47998952865600586,
      "num_input_tokens_seen": 567674176,
      "step": 2000,
      "train_runtime": 19412.9958,
      "train_tokens_per_second": 29241.967
    },
    {
      "epoch": 0.07111787300948459,
      "grad_norm": 0.60546875,
      "learning_rate": 4.562643975076387e-05,
      "loss": 0.48445773124694824,
      "num_input_tokens_seen": 570516160,
      "step": 2010,
      "train_runtime": 19511.5222,
      "train_tokens_per_second": 29239.962
    },
    {
      "epoch": 0.07147169327321337,
      "grad_norm": 0.6171875,
      "learning_rate": 4.560745483763357e-05,
      "loss": 0.48499622344970705,
      "num_input_tokens_seen": 573346944,
      "step": 2020,
      "train_runtime": 19609.9296,
      "train_tokens_per_second": 29237.583
    },
    {
      "epoch": 0.07182551353694215,
      "grad_norm": 0.62109375,
      "learning_rate": 4.5588493603355595e-05,
      "loss": 0.4852696418762207,
      "num_input_tokens_seen": 576276288,
      "step": 2030,
      "train_runtime": 19713.7756,
      "train_tokens_per_second": 29232.162
    },
    {
      "epoch": 0.07217933380067093,
      "grad_norm": 0.59375,
      "learning_rate": 4.556955599874859e-05,
      "loss": 0.48374109268188475,
      "num_input_tokens_seen": 579141184,
      "step": 2040,
      "train_runtime": 19813.6656,
      "train_tokens_per_second": 29229.381
    },
    {
      "epoch": 0.07253315406439971,
      "grad_norm": 0.60546875,
      "learning_rate": 4.555064197477409e-05,
      "loss": 0.48391122817993165,
      "num_input_tokens_seen": 582017088,
      "step": 2050,
      "train_runtime": 19913.7142,
      "train_tokens_per_second": 29226.948
    },
    {
      "epoch": 0.07288697432812849,
      "grad_norm": 0.6171875,
      "learning_rate": 4.5531751482536e-05,
      "loss": 0.4845563888549805,
      "num_input_tokens_seen": 584861632,
      "step": 2060,
      "train_runtime": 20014.4872,
      "train_tokens_per_second": 29221.914
    },
    {
      "epoch": 0.07324079459185727,
      "grad_norm": 0.62109375,
      "learning_rate": 4.5512884473280024e-05,
      "loss": 0.48123531341552733,
      "num_input_tokens_seen": 587725440,
      "step": 2070,
      "train_runtime": 20110.808,
      "train_tokens_per_second": 29224.357
    },
    {
      "epoch": 0.07359461485558605,
      "grad_norm": 0.6015625,
      "learning_rate": 4.549404089839322e-05,
      "loss": 0.48112030029296876,
      "num_input_tokens_seen": 590559040,
      "step": 2080,
      "train_runtime": 20208.3055,
      "train_tokens_per_second": 29223.58
    },
    {
      "epoch": 0.07394843511931483,
      "grad_norm": 0.6015625,
      "learning_rate": 4.547522070940335e-05,
      "loss": 0.4818718433380127,
      "num_input_tokens_seen": 593424512,
      "step": 2090,
      "train_runtime": 20307.0415,
      "train_tokens_per_second": 29222.598
    },
    {
      "epoch": 0.07430225538304361,
      "grad_norm": 0.59375,
      "learning_rate": 4.545642385797848e-05,
      "loss": 0.4805013656616211,
      "num_input_tokens_seen": 596225088,
      "step": 2100,
      "train_runtime": 20398.6211,
      "train_tokens_per_second": 29228.696
    },
    {
      "epoch": 0.07465607564677239,
      "grad_norm": 0.5859375,
      "learning_rate": 4.543765029592637e-05,
      "loss": 0.48232307434082033,
      "num_input_tokens_seen": 599054400,
      "step": 2110,
      "train_runtime": 20492.0291,
      "train_tokens_per_second": 29233.533
    },
    {
      "epoch": 0.07500989591050117,
      "grad_norm": 0.6328125,
      "learning_rate": 4.541889997519403e-05,
      "loss": 0.478058910369873,
      "num_input_tokens_seen": 601905792,
      "step": 2120,
      "train_runtime": 20592.0345,
      "train_tokens_per_second": 29230.03
    },
    {
      "epoch": 0.07536371617422995,
      "grad_norm": 0.58203125,
      "learning_rate": 4.5400172847867095e-05,
      "loss": 0.48308491706848145,
      "num_input_tokens_seen": 604781696,
      "step": 2130,
      "train_runtime": 20691.8944,
      "train_tokens_per_second": 29227.952
    },
    {
      "epoch": 0.07571753643795873,
      "grad_norm": 0.6015625,
      "learning_rate": 4.5381468866169466e-05,
      "loss": 0.4799912929534912,
      "num_input_tokens_seen": 607602112,
      "step": 2140,
      "train_runtime": 20785.9732,
      "train_tokens_per_second": 29231.353
    },
    {
      "epoch": 0.0760713567016875,
      "grad_norm": 0.6171875,
      "learning_rate": 4.5362787982462616e-05,
      "loss": 0.4795804023742676,
      "num_input_tokens_seen": 610399360,
      "step": 2150,
      "train_runtime": 20883.6315,
      "train_tokens_per_second": 29228.602
    },
    {
      "epoch": 0.07642517696541629,
      "grad_norm": 0.6171875,
      "learning_rate": 4.5344130149245275e-05,
      "loss": 0.47710742950439455,
      "num_input_tokens_seen": 613185408,
      "step": 2160,
      "train_runtime": 20976.3577,
      "train_tokens_per_second": 29232.215
    },
    {
      "epoch": 0.07677899722914507,
      "grad_norm": 0.60546875,
      "learning_rate": 4.5325495319152715e-05,
      "loss": 0.4809116363525391,
      "num_input_tokens_seen": 616021440,
      "step": 2170,
      "train_runtime": 21072.8929,
      "train_tokens_per_second": 29232.884
    },
    {
      "epoch": 0.07713281749287385,
      "grad_norm": 0.59375,
      "learning_rate": 4.530688344495644e-05,
      "loss": 0.47991437911987306,
      "num_input_tokens_seen": 618854080,
      "step": 2180,
      "train_runtime": 21168.2486,
      "train_tokens_per_second": 29235.016
    },
    {
      "epoch": 0.07748663775660261,
      "grad_norm": 0.6328125,
      "learning_rate": 4.528829447956357e-05,
      "loss": 0.48104305267333985,
      "num_input_tokens_seen": 621675904,
      "step": 2190,
      "train_runtime": 21264.1246,
      "train_tokens_per_second": 29235.904
    },
    {
      "epoch": 0.07784045802033139,
      "grad_norm": 0.59375,
      "learning_rate": 4.526972837601633e-05,
      "loss": 0.47821741104125975,
      "num_input_tokens_seen": 624492288,
      "step": 2200,
      "train_runtime": 21359.2133,
      "train_tokens_per_second": 29237.607
    },
    {
      "epoch": 0.07819427828406017,
      "grad_norm": 0.609375,
      "learning_rate": 4.525118508749165e-05,
      "loss": 0.48128366470336914,
      "num_input_tokens_seen": 627315200,
      "step": 2210,
      "train_runtime": 21456.7457,
      "train_tokens_per_second": 29236.269
    },
    {
      "epoch": 0.07854809854778895,
      "grad_norm": 0.58203125,
      "learning_rate": 4.5232664567300546e-05,
      "loss": 0.4781455993652344,
      "num_input_tokens_seen": 630126080,
      "step": 2220,
      "train_runtime": 21554.1663,
      "train_tokens_per_second": 29234.537
    },
    {
      "epoch": 0.07890191881151773,
      "grad_norm": 0.60546875,
      "learning_rate": 4.521416676888773e-05,
      "loss": 0.47599353790283205,
      "num_input_tokens_seen": 632958400,
      "step": 2230,
      "train_runtime": 21649.7073,
      "train_tokens_per_second": 29236.349
    },
    {
      "epoch": 0.07925573907524651,
      "grad_norm": 0.61328125,
      "learning_rate": 4.519569164583107e-05,
      "loss": 0.47762494087219237,
      "num_input_tokens_seen": 635800064,
      "step": 2240,
      "train_runtime": 21749.5715,
      "train_tokens_per_second": 29232.763
    },
    {
      "epoch": 0.07960955933897529,
      "grad_norm": 0.5859375,
      "learning_rate": 4.517723915184109e-05,
      "loss": 0.47615938186645507,
      "num_input_tokens_seen": 638624768,
      "step": 2250,
      "train_runtime": 21843.5959,
      "train_tokens_per_second": 29236.247
    },
    {
      "epoch": 0.07996337960270407,
      "grad_norm": 0.6171875,
      "learning_rate": 4.5158809240760506e-05,
      "loss": 0.47748804092407227,
      "num_input_tokens_seen": 641490688,
      "step": 2260,
      "train_runtime": 21943.3106,
      "train_tokens_per_second": 29233.998
    },
    {
      "epoch": 0.08031719986643285,
      "grad_norm": 0.5859375,
      "learning_rate": 4.514040186656375e-05,
      "loss": 0.4740769863128662,
      "num_input_tokens_seen": 644356096,
      "step": 2270,
      "train_runtime": 22038.8263,
      "train_tokens_per_second": 29237.314
    },
    {
      "epoch": 0.08067102013016163,
      "grad_norm": 0.60546875,
      "learning_rate": 4.512201698335644e-05,
      "loss": 0.4734847068786621,
      "num_input_tokens_seen": 647194560,
      "step": 2280,
      "train_runtime": 22137.4514,
      "train_tokens_per_second": 29235.279
    },
    {
      "epoch": 0.08102484039389041,
      "grad_norm": 0.59765625,
      "learning_rate": 4.510365454537496e-05,
      "loss": 0.4742901802062988,
      "num_input_tokens_seen": 650039296,
      "step": 2290,
      "train_runtime": 22233.3403,
      "train_tokens_per_second": 29237.141
    },
    {
      "epoch": 0.08137866065761919,
      "grad_norm": 0.58984375,
      "learning_rate": 4.5085314506985945e-05,
      "loss": 0.4732800483703613,
      "num_input_tokens_seen": 652945088,
      "step": 2300,
      "train_runtime": 22333.4529,
      "train_tokens_per_second": 29236.191
    },
    {
      "epoch": 0.08173248092134797,
      "grad_norm": 0.6015625,
      "learning_rate": 4.50669968226858e-05,
      "loss": 0.47473607063293455,
      "num_input_tokens_seen": 655788096,
      "step": 2310,
      "train_runtime": 22430.2258,
      "train_tokens_per_second": 29236.803
    },
    {
      "epoch": 0.08208630118507675,
      "grad_norm": 0.6015625,
      "learning_rate": 4.504870144710027e-05,
      "loss": 0.4761085510253906,
      "num_input_tokens_seen": 658622528,
      "step": 2320,
      "train_runtime": 22526.6623,
      "train_tokens_per_second": 29237.466
    },
    {
      "epoch": 0.08244012144880553,
      "grad_norm": 0.61328125,
      "learning_rate": 4.5030428334983884e-05,
      "loss": 0.478547477722168,
      "num_input_tokens_seen": 661542720,
      "step": 2330,
      "train_runtime": 22630.0996,
      "train_tokens_per_second": 29232.868
    },
    {
      "epoch": 0.0827939417125343,
      "grad_norm": 0.578125,
      "learning_rate": 4.501217744121959e-05,
      "loss": 0.4752105712890625,
      "num_input_tokens_seen": 664371264,
      "step": 2340,
      "train_runtime": 22725.6682,
      "train_tokens_per_second": 29234.4
    },
    {
      "epoch": 0.08314776197626308,
      "grad_norm": 0.5703125,
      "learning_rate": 4.499394872081821e-05,
      "loss": 0.47428164482116697,
      "num_input_tokens_seen": 667226688,
      "step": 2350,
      "train_runtime": 22822.4492,
      "train_tokens_per_second": 29235.543
    },
    {
      "epoch": 0.08350158223999186,
      "grad_norm": 0.60546875,
      "learning_rate": 4.4975742128918e-05,
      "loss": 0.4785414695739746,
      "num_input_tokens_seen": 670128512,
      "step": 2360,
      "train_runtime": 22921.1048,
      "train_tokens_per_second": 29236.309
    },
    {
      "epoch": 0.08385540250372064,
      "grad_norm": 0.59765625,
      "learning_rate": 4.495755762078418e-05,
      "loss": 0.47456836700439453,
      "num_input_tokens_seen": 672977984,
      "step": 2370,
      "train_runtime": 23016.0796,
      "train_tokens_per_second": 29239.471
    },
    {
      "epoch": 0.08420922276744942,
      "grad_norm": 0.63671875,
      "learning_rate": 4.49393951518085e-05,
      "loss": 0.47785263061523436,
      "num_input_tokens_seen": 675826496,
      "step": 2380,
      "train_runtime": 23113.1051,
      "train_tokens_per_second": 29239.97
    },
    {
      "epoch": 0.0845630430311782,
      "grad_norm": 0.5859375,
      "learning_rate": 4.4921254677508716e-05,
      "loss": 0.47466516494750977,
      "num_input_tokens_seen": 678657856,
      "step": 2390,
      "train_runtime": 23207.4088,
      "train_tokens_per_second": 29243.155
    },
    {
      "epoch": 0.08491686329490698,
      "grad_norm": 0.61328125,
      "learning_rate": 4.490313615352821e-05,
      "loss": 0.47445173263549806,
      "num_input_tokens_seen": 681497152,
      "step": 2400,
      "train_runtime": 23302.754,
      "train_tokens_per_second": 29245.348
    },
    {
      "epoch": 0.08527068355863576,
      "grad_norm": 0.609375,
      "learning_rate": 4.48850395356355e-05,
      "loss": 0.4685967445373535,
      "num_input_tokens_seen": 684355968,
      "step": 2410,
      "train_runtime": 23399.8478,
      "train_tokens_per_second": 29246.172
    },
    {
      "epoch": 0.08562450382236454,
      "grad_norm": 0.625,
      "learning_rate": 4.486696477972375e-05,
      "loss": 0.4712062358856201,
      "num_input_tokens_seen": 687187904,
      "step": 2420,
      "train_runtime": 23494.1246,
      "train_tokens_per_second": 29249.351
    },
    {
      "epoch": 0.08597832408609332,
      "grad_norm": 0.6015625,
      "learning_rate": 4.484891184181041e-05,
      "loss": 0.47417125701904295,
      "num_input_tokens_seen": 689986048,
      "step": 2430,
      "train_runtime": 23586.5638,
      "train_tokens_per_second": 29253.352
    },
    {
      "epoch": 0.0863321443498221,
      "grad_norm": 0.60546875,
      "learning_rate": 4.483088067803662e-05,
      "loss": 0.47461824417114257,
      "num_input_tokens_seen": 692802496,
      "step": 2440,
      "train_runtime": 23683.4359,
      "train_tokens_per_second": 29252.618
    },
    {
      "epoch": 0.08668596461355087,
      "grad_norm": 0.58203125,
      "learning_rate": 4.481287124466697e-05,
      "loss": 0.47368178367614744,
      "num_input_tokens_seen": 695674432,
      "step": 2450,
      "train_runtime": 23781.7637,
      "train_tokens_per_second": 29252.432
    },
    {
      "epoch": 0.08703978487727965,
      "grad_norm": 0.56640625,
      "learning_rate": 4.479488349808885e-05,
      "loss": 0.4781245231628418,
      "num_input_tokens_seen": 698575488,
      "step": 2460,
      "train_runtime": 23881.6686,
      "train_tokens_per_second": 29251.536
    },
    {
      "epoch": 0.08739360514100843,
      "grad_norm": 0.578125,
      "learning_rate": 4.4776917394812114e-05,
      "loss": 0.47491955757141113,
      "num_input_tokens_seen": 701434496,
      "step": 2470,
      "train_runtime": 23981.4477,
      "train_tokens_per_second": 29249.047
    },
    {
      "epoch": 0.0877474254047372,
      "grad_norm": 0.58203125,
      "learning_rate": 4.475897289146862e-05,
      "loss": 0.4757704734802246,
      "num_input_tokens_seen": 704291840,
      "step": 2480,
      "train_runtime": 24081.307,
      "train_tokens_per_second": 29246.413
    },
    {
      "epoch": 0.08810124566846599,
      "grad_norm": 0.58203125,
      "learning_rate": 4.4741049944811806e-05,
      "loss": 0.47130985260009767,
      "num_input_tokens_seen": 707117184,
      "step": 2490,
      "train_runtime": 24178.3855,
      "train_tokens_per_second": 29245.84
    },
    {
      "epoch": 0.08845506593219477,
      "grad_norm": 0.61328125,
      "learning_rate": 4.472314851171621e-05,
      "loss": 0.47155261039733887,
      "num_input_tokens_seen": 709938944,
      "step": 2500,
      "train_runtime": 24275.8201,
      "train_tokens_per_second": 29244.695
    },
    {
      "epoch": 0.08880888619592354,
      "grad_norm": 0.57421875,
      "learning_rate": 4.4705268549177084e-05,
      "loss": 0.46933469772338865,
      "num_input_tokens_seen": 712816832,
      "step": 2510,
      "train_runtime": 24374.0398,
      "train_tokens_per_second": 29244.919
    },
    {
      "epoch": 0.08916270645965232,
      "grad_norm": 0.60546875,
      "learning_rate": 4.468741001430989e-05,
      "loss": 0.4701040267944336,
      "num_input_tokens_seen": 715680704,
      "step": 2520,
      "train_runtime": 24472.2469,
      "train_tokens_per_second": 29244.585
    },
    {
      "epoch": 0.0895165267233811,
      "grad_norm": 0.59765625,
      "learning_rate": 4.466957286434997e-05,
      "loss": 0.4729485511779785,
      "num_input_tokens_seen": 718503552,
      "step": 2530,
      "train_runtime": 24567.1367,
      "train_tokens_per_second": 29246.532
    },
    {
      "epoch": 0.08987034698710988,
      "grad_norm": 0.59375,
      "learning_rate": 4.4651757056652e-05,
      "loss": 0.47142491340637205,
      "num_input_tokens_seen": 721313088,
      "step": 2540,
      "train_runtime": 24663.3361,
      "train_tokens_per_second": 29246.371
    },
    {
      "epoch": 0.09022416725083866,
      "grad_norm": 0.5859375,
      "learning_rate": 4.463396254868968e-05,
      "loss": 0.47236042022705077,
      "num_input_tokens_seen": 724171456,
      "step": 2550,
      "train_runtime": 24762.8515,
      "train_tokens_per_second": 29244.268
    },
    {
      "epoch": 0.09057798751456744,
      "grad_norm": 0.6171875,
      "learning_rate": 4.461618929805519e-05,
      "loss": 0.4721323013305664,
      "num_input_tokens_seen": 727007232,
      "step": 2560,
      "train_runtime": 24861.3916,
      "train_tokens_per_second": 29242.419
    },
    {
      "epoch": 0.09093180777829622,
      "grad_norm": 0.58984375,
      "learning_rate": 4.459843726245888e-05,
      "loss": 0.47497129440307617,
      "num_input_tokens_seen": 729885696,
      "step": 2570,
      "train_runtime": 24958.1264,
      "train_tokens_per_second": 29244.411
    },
    {
      "epoch": 0.091285628042025,
      "grad_norm": 0.57421875,
      "learning_rate": 4.458070639972875e-05,
      "loss": 0.4680130958557129,
      "num_input_tokens_seen": 732750016,
      "step": 2580,
      "train_runtime": 25055.1221,
      "train_tokens_per_second": 29245.518
    },
    {
      "epoch": 0.09163944830575378,
      "grad_norm": 0.578125,
      "learning_rate": 4.456299666781007e-05,
      "loss": 0.47108659744262693,
      "num_input_tokens_seen": 735551168,
      "step": 2590,
      "train_runtime": 25148.25,
      "train_tokens_per_second": 29248.603
    },
    {
      "epoch": 0.09199326856948256,
      "grad_norm": 0.58984375,
      "learning_rate": 4.4545308024764984e-05,
      "loss": 0.4726726531982422,
      "num_input_tokens_seen": 738383808,
      "step": 2600,
      "train_runtime": 25248.3029,
      "train_tokens_per_second": 29244.889
    },
    {
      "epoch": 0.09234708883321134,
      "grad_norm": 0.58984375,
      "learning_rate": 4.452764042877207e-05,
      "loss": 0.4747706413269043,
      "num_input_tokens_seen": 741271168,
      "step": 2610,
      "train_runtime": 25346.8555,
      "train_tokens_per_second": 29245.094
    },
    {
      "epoch": 0.09270090909694012,
      "grad_norm": 0.61328125,
      "learning_rate": 4.45099938381259e-05,
      "loss": 0.4734789848327637,
      "num_input_tokens_seen": 744088896,
      "step": 2620,
      "train_runtime": 25443.0629,
      "train_tokens_per_second": 29245.256
    },
    {
      "epoch": 0.0930547293606689,
      "grad_norm": 0.59375,
      "learning_rate": 4.449236821123667e-05,
      "loss": 0.47025070190429685,
      "num_input_tokens_seen": 746930688,
      "step": 2630,
      "train_runtime": 25538.782,
      "train_tokens_per_second": 29246.919
    },
    {
      "epoch": 0.09340854962439768,
      "grad_norm": 0.56640625,
      "learning_rate": 4.447476350662976e-05,
      "loss": 0.4715579032897949,
      "num_input_tokens_seen": 749821440,
      "step": 2640,
      "train_runtime": 25638.5953,
      "train_tokens_per_second": 29245.808
    },
    {
      "epoch": 0.09376236988812646,
      "grad_norm": 0.5703125,
      "learning_rate": 4.4457179682945346e-05,
      "loss": 0.4647881031036377,
      "num_input_tokens_seen": 752690176,
      "step": 2650,
      "train_runtime": 25737.8341,
      "train_tokens_per_second": 29244.503
    },
    {
      "epoch": 0.09411619015185524,
      "grad_norm": 0.5703125,
      "learning_rate": 4.443961669893798e-05,
      "loss": 0.46997499465942383,
      "num_input_tokens_seen": 755524736,
      "step": 2660,
      "train_runtime": 25834.6908,
      "train_tokens_per_second": 29244.582
    },
    {
      "epoch": 0.09447001041558402,
      "grad_norm": 0.6015625,
      "learning_rate": 4.4422074513476155e-05,
      "loss": 0.47669157981872556,
      "num_input_tokens_seen": 758300864,
      "step": 2670,
      "train_runtime": 25928.472,
      "train_tokens_per_second": 29245.875
    },
    {
      "epoch": 0.0948238306793128,
      "grad_norm": 0.59375,
      "learning_rate": 4.4404553085541955e-05,
      "loss": 0.465960693359375,
      "num_input_tokens_seen": 761173376,
      "step": 2680,
      "train_runtime": 26024.1535,
      "train_tokens_per_second": 29248.728
    },
    {
      "epoch": 0.09517765094304158,
      "grad_norm": 0.58984375,
      "learning_rate": 4.438705237423063e-05,
      "loss": 0.4664862632751465,
      "num_input_tokens_seen": 764039808,
      "step": 2690,
      "train_runtime": 26119.9455,
      "train_tokens_per_second": 29251.202
    },
    {
      "epoch": 0.09553147120677036,
      "grad_norm": 0.57421875,
      "learning_rate": 4.436957233875017e-05,
      "loss": 0.4669026374816895,
      "num_input_tokens_seen": 766838592,
      "step": 2700,
      "train_runtime": 26214.9104,
      "train_tokens_per_second": 29252.001
    },
    {
      "epoch": 0.09588529147049912,
      "grad_norm": 0.5859375,
      "learning_rate": 4.4352112938420956e-05,
      "loss": 0.47040748596191406,
      "num_input_tokens_seen": 769722496,
      "step": 2710,
      "train_runtime": 26314.5255,
      "train_tokens_per_second": 29250.86
    },
    {
      "epoch": 0.0962391117342279,
      "grad_norm": 0.59765625,
      "learning_rate": 4.433467413267529e-05,
      "loss": 0.46546106338500975,
      "num_input_tokens_seen": 772516544,
      "step": 2720,
      "train_runtime": 26410.2051,
      "train_tokens_per_second": 29250.683
    },
    {
      "epoch": 0.09659293199795668,
      "grad_norm": 0.59765625,
      "learning_rate": 4.431725588105708e-05,
      "loss": 0.4701519966125488,
      "num_input_tokens_seen": 775288064,
      "step": 2730,
      "train_runtime": 26505.124,
      "train_tokens_per_second": 29250.498
    },
    {
      "epoch": 0.09694675226168546,
      "grad_norm": 0.6015625,
      "learning_rate": 4.4299858143221377e-05,
      "loss": 0.467894458770752,
      "num_input_tokens_seen": 778139008,
      "step": 2740,
      "train_runtime": 26602.7883,
      "train_tokens_per_second": 29250.28
    },
    {
      "epoch": 0.09730057252541424,
      "grad_norm": 0.59765625,
      "learning_rate": 4.4282480878934065e-05,
      "loss": 0.47024030685424806,
      "num_input_tokens_seen": 781009600,
      "step": 2750,
      "train_runtime": 26700.9732,
      "train_tokens_per_second": 29250.23
    },
    {
      "epoch": 0.09765439278914302,
      "grad_norm": 0.58984375,
      "learning_rate": 4.4265124048071346e-05,
      "loss": 0.4699270248413086,
      "num_input_tokens_seen": 783864896,
      "step": 2760,
      "train_runtime": 26803.2228,
      "train_tokens_per_second": 29245.173
    },
    {
      "epoch": 0.0980082130528718,
      "grad_norm": 0.56640625,
      "learning_rate": 4.4247787610619477e-05,
      "loss": 0.4677596092224121,
      "num_input_tokens_seen": 786664128,
      "step": 2770,
      "train_runtime": 26897.5824,
      "train_tokens_per_second": 29246.648
    },
    {
      "epoch": 0.09836203331660058,
      "grad_norm": 0.60546875,
      "learning_rate": 4.42304715266743e-05,
      "loss": 0.4651189804077148,
      "num_input_tokens_seen": 789491136,
      "step": 2780,
      "train_runtime": 26994.9562,
      "train_tokens_per_second": 29245.876
    },
    {
      "epoch": 0.09871585358032936,
      "grad_norm": 0.57421875,
      "learning_rate": 4.421317575644092e-05,
      "loss": 0.46714029312133787,
      "num_input_tokens_seen": 792327488,
      "step": 2790,
      "train_runtime": 27092.0228,
      "train_tokens_per_second": 29245.786
    },
    {
      "epoch": 0.09906967384405814,
      "grad_norm": 0.5859375,
      "learning_rate": 4.419590026023325e-05,
      "loss": 0.46109347343444823,
      "num_input_tokens_seen": 795147904,
      "step": 2800,
      "train_runtime": 27191.1165,
      "train_tokens_per_second": 29242.93
    },
    {
      "epoch": 0.09942349410778692,
      "grad_norm": 0.56640625,
      "learning_rate": 4.417864499847368e-05,
      "loss": 0.46106829643249514,
      "num_input_tokens_seen": 797988288,
      "step": 2810,
      "train_runtime": 27286.3586,
      "train_tokens_per_second": 29244.954
    },
    {
      "epoch": 0.0997773143715157,
      "grad_norm": 0.5859375,
      "learning_rate": 4.4161409931692676e-05,
      "loss": 0.4673023223876953,
      "num_input_tokens_seen": 800824704,
      "step": 2820,
      "train_runtime": 27382.2401,
      "train_tokens_per_second": 29246.136
    },
    {
      "epoch": 0.10013113463524448,
      "grad_norm": 0.63671875,
      "learning_rate": 4.414419502052841e-05,
      "loss": 0.4656674861907959,
      "num_input_tokens_seen": 803713152,
      "step": 2830,
      "train_runtime": 27482.7755,
      "train_tokens_per_second": 29244.25
    },
    {
      "epoch": 0.10048495489897326,
      "grad_norm": 0.6171875,
      "learning_rate": 4.412700022572637e-05,
      "loss": 0.4688971996307373,
      "num_input_tokens_seen": 806536128,
      "step": 2840,
      "train_runtime": 27579.6745,
      "train_tokens_per_second": 29243.86
    },
    {
      "epoch": 0.10083877516270204,
      "grad_norm": 0.61328125,
      "learning_rate": 4.410982550813902e-05,
      "loss": 0.46872291564941404,
      "num_input_tokens_seen": 809369600,
      "step": 2850,
      "train_runtime": 27674.8775,
      "train_tokens_per_second": 29245.643
    },
    {
      "epoch": 0.10119259542643082,
      "grad_norm": 0.57421875,
      "learning_rate": 4.409267082872535e-05,
      "loss": 0.4665546417236328,
      "num_input_tokens_seen": 812267648,
      "step": 2860,
      "train_runtime": 27777.6491,
      "train_tokens_per_second": 29241.771
    },
    {
      "epoch": 0.1015464156901596,
      "grad_norm": 0.5859375,
      "learning_rate": 4.407553614855059e-05,
      "loss": 0.4677140712738037,
      "num_input_tokens_seen": 815092928,
      "step": 2870,
      "train_runtime": 27874.7343,
      "train_tokens_per_second": 29241.281
    },
    {
      "epoch": 0.10190023595388838,
      "grad_norm": 0.609375,
      "learning_rate": 4.405842142878579e-05,
      "loss": 0.4661547183990479,
      "num_input_tokens_seen": 817994752,
      "step": 2880,
      "train_runtime": 27974.3735,
      "train_tokens_per_second": 29240.86
    },
    {
      "epoch": 0.10225405621761716,
      "grad_norm": 0.5859375,
      "learning_rate": 4.404132663070745e-05,
      "loss": 0.4628773212432861,
      "num_input_tokens_seen": 820838144,
      "step": 2890,
      "train_runtime": 28069.5642,
      "train_tokens_per_second": 29242.996
    },
    {
      "epoch": 0.10260787648134594,
      "grad_norm": 0.58203125,
      "learning_rate": 4.402425171569716e-05,
      "loss": 0.4591650485992432,
      "num_input_tokens_seen": 823684480,
      "step": 2900,
      "train_runtime": 28168.1984,
      "train_tokens_per_second": 29241.646
    },
    {
      "epoch": 0.10296169674507472,
      "grad_norm": 0.5703125,
      "learning_rate": 4.400719664524127e-05,
      "loss": 0.46363110542297364,
      "num_input_tokens_seen": 826506304,
      "step": 2910,
      "train_runtime": 28264.666,
      "train_tokens_per_second": 29241.68
    },
    {
      "epoch": 0.1033155170088035,
      "grad_norm": 0.5859375,
      "learning_rate": 4.399016138093044e-05,
      "loss": 0.4681601047515869,
      "num_input_tokens_seen": 829323264,
      "step": 2920,
      "train_runtime": 28362.069,
      "train_tokens_per_second": 29240.577
    },
    {
      "epoch": 0.10366933727253227,
      "grad_norm": 0.5703125,
      "learning_rate": 4.397314588445937e-05,
      "loss": 0.46255884170532224,
      "num_input_tokens_seen": 832161984,
      "step": 2930,
      "train_runtime": 28458.5227,
      "train_tokens_per_second": 29241.222
    },
    {
      "epoch": 0.10402315753626105,
      "grad_norm": 0.578125,
      "learning_rate": 4.395615011762637e-05,
      "loss": 0.4611948013305664,
      "num_input_tokens_seen": 834996416,
      "step": 2940,
      "train_runtime": 28554.2134,
      "train_tokens_per_second": 29242.494
    },
    {
      "epoch": 0.10437697779998983,
      "grad_norm": 0.578125,
      "learning_rate": 4.3939174042333057e-05,
      "loss": 0.4690437316894531,
      "num_input_tokens_seen": 837916736,
      "step": 2950,
      "train_runtime": 28654.9799,
      "train_tokens_per_second": 29241.575
    },
    {
      "epoch": 0.10473079806371861,
      "grad_norm": 0.609375,
      "learning_rate": 4.3922217620583904e-05,
      "loss": 0.4671497821807861,
      "num_input_tokens_seen": 840767040,
      "step": 2960,
      "train_runtime": 28751.904,
      "train_tokens_per_second": 29242.134
    },
    {
      "epoch": 0.10508461832744738,
      "grad_norm": 0.59765625,
      "learning_rate": 4.3905280814486025e-05,
      "loss": 0.4650600433349609,
      "num_input_tokens_seen": 843630912,
      "step": 2970,
      "train_runtime": 28851.8868,
      "train_tokens_per_second": 29240.06
    },
    {
      "epoch": 0.10543843859117616,
      "grad_norm": 0.6015625,
      "learning_rate": 4.388836358624867e-05,
      "loss": 0.46319026947021485,
      "num_input_tokens_seen": 846460864,
      "step": 2980,
      "train_runtime": 28948.2457,
      "train_tokens_per_second": 29240.489
    },
    {
      "epoch": 0.10579225885490494,
      "grad_norm": 0.5859375,
      "learning_rate": 4.3871465898182976e-05,
      "loss": 0.46334037780761717,
      "num_input_tokens_seen": 849315712,
      "step": 2990,
      "train_runtime": 29045.2387,
      "train_tokens_per_second": 29241.134
    },
    {
      "epoch": 0.10614607911863372,
      "grad_norm": 0.58984375,
      "learning_rate": 4.385458771270156e-05,
      "loss": 0.4656219005584717,
      "num_input_tokens_seen": 852171840,
      "step": 3000,
      "train_runtime": 29142.881,
      "train_tokens_per_second": 29241.167
    },
    {
      "epoch": 0.1064998993823625,
      "grad_norm": 0.58203125,
      "learning_rate": 4.3837728992318205e-05,
      "loss": 0.4621277332305908,
      "num_input_tokens_seen": 855081664,
      "step": 3010,
      "train_runtime": 29244.2836,
      "train_tokens_per_second": 29239.275
    },
    {
      "epoch": 0.10685371964609128,
      "grad_norm": 0.578125,
      "learning_rate": 4.382088969964746e-05,
      "loss": 0.4653298377990723,
      "num_input_tokens_seen": 857962944,
      "step": 3020,
      "train_runtime": 29345.035,
      "train_tokens_per_second": 29237.073
    },
    {
      "epoch": 0.10720753990982006,
      "grad_norm": 0.56640625,
      "learning_rate": 4.380406979740436e-05,
      "loss": 0.4639897346496582,
      "num_input_tokens_seen": 860796160,
      "step": 3030,
      "train_runtime": 29440.6292,
      "train_tokens_per_second": 29238.375
    },
    {
      "epoch": 0.10756136017354884,
      "grad_norm": 0.60546875,
      "learning_rate": 4.3787269248403994e-05,
      "loss": 0.4654397487640381,
      "num_input_tokens_seen": 863619072,
      "step": 3040,
      "train_runtime": 29539.103,
      "train_tokens_per_second": 29236.469
    },
    {
      "epoch": 0.10791518043727762,
      "grad_norm": 0.57421875,
      "learning_rate": 4.377048801556126e-05,
      "loss": 0.4612120628356934,
      "num_input_tokens_seen": 866453888,
      "step": 3050,
      "train_runtime": 29634.7419,
      "train_tokens_per_second": 29237.774
    },
    {
      "epoch": 0.1082690007010064,
      "grad_norm": 0.5859375,
      "learning_rate": 4.3753726061890446e-05,
      "loss": 0.4632655143737793,
      "num_input_tokens_seen": 869239744,
      "step": 3060,
      "train_runtime": 29729.1417,
      "train_tokens_per_second": 29238.642
    },
    {
      "epoch": 0.10862282096473518,
      "grad_norm": 0.58203125,
      "learning_rate": 4.373698335050488e-05,
      "loss": 0.4609369277954102,
      "num_input_tokens_seen": 872085568,
      "step": 3070,
      "train_runtime": 29828.317,
      "train_tokens_per_second": 29236.834
    },
    {
      "epoch": 0.10897664122846396,
      "grad_norm": 0.6015625,
      "learning_rate": 4.372025984461667e-05,
      "loss": 0.4588226318359375,
      "num_input_tokens_seen": 874937024,
      "step": 3080,
      "train_runtime": 29928.0732,
      "train_tokens_per_second": 29234.659
    },
    {
      "epoch": 0.10933046149219273,
      "grad_norm": 0.58984375,
      "learning_rate": 4.370355550753629e-05,
      "loss": 0.4575453281402588,
      "num_input_tokens_seen": 877781248,
      "step": 3090,
      "train_runtime": 30024.6702,
      "train_tokens_per_second": 29235.334
    },
    {
      "epoch": 0.10968428175592151,
      "grad_norm": 0.58203125,
      "learning_rate": 4.368687030267226e-05,
      "loss": 0.4613487720489502,
      "num_input_tokens_seen": 880572416,
      "step": 3100,
      "train_runtime": 30115.8335,
      "train_tokens_per_second": 29239.517
    },
    {
      "epoch": 0.1100381020196503,
      "grad_norm": 0.5625,
      "learning_rate": 4.367020419353081e-05,
      "loss": 0.4559629440307617,
      "num_input_tokens_seen": 883425984,
      "step": 3110,
      "train_runtime": 30213.5201,
      "train_tokens_per_second": 29239.426
    },
    {
      "epoch": 0.11039192228337907,
      "grad_norm": 0.5625,
      "learning_rate": 4.365355714371558e-05,
      "loss": 0.4625598430633545,
      "num_input_tokens_seen": 886238272,
      "step": 3120,
      "train_runtime": 30307.4264,
      "train_tokens_per_second": 29241.621
    },
    {
      "epoch": 0.11074574254710785,
      "grad_norm": 0.55078125,
      "learning_rate": 4.3636929116927235e-05,
      "loss": 0.4601325988769531,
      "num_input_tokens_seen": 889061760,
      "step": 3130,
      "train_runtime": 30405.4857,
      "train_tokens_per_second": 29240.176
    },
    {
      "epoch": 0.11109956281083663,
      "grad_norm": 0.5859375,
      "learning_rate": 4.362032007696314e-05,
      "loss": 0.45470199584960935,
      "num_input_tokens_seen": 891973888,
      "step": 3140,
      "train_runtime": 30506.5573,
      "train_tokens_per_second": 29238.759
    },
    {
      "epoch": 0.11145338307456541,
      "grad_norm": 0.578125,
      "learning_rate": 4.360372998771707e-05,
      "loss": 0.46184520721435546,
      "num_input_tokens_seen": 894815232,
      "step": 3150,
      "train_runtime": 30603.8801,
      "train_tokens_per_second": 29238.62
    },
    {
      "epoch": 0.11180720333829419,
      "grad_norm": 0.58203125,
      "learning_rate": 4.358715881317884e-05,
      "loss": 0.4563714027404785,
      "num_input_tokens_seen": 897675520,
      "step": 3160,
      "train_runtime": 30701.5098,
      "train_tokens_per_second": 29238.807
    },
    {
      "epoch": 0.11216102360202297,
      "grad_norm": 0.59765625,
      "learning_rate": 4.357060651743399e-05,
      "loss": 0.4601877212524414,
      "num_input_tokens_seen": 900471808,
      "step": 3170,
      "train_runtime": 30795.4976,
      "train_tokens_per_second": 29240.372
    },
    {
      "epoch": 0.11251484386575175,
      "grad_norm": 0.55859375,
      "learning_rate": 4.3554073064663454e-05,
      "loss": 0.45745162963867186,
      "num_input_tokens_seen": 903332224,
      "step": 3180,
      "train_runtime": 30893.007,
      "train_tokens_per_second": 29240.67
    },
    {
      "epoch": 0.11286866412948053,
      "grad_norm": 0.57421875,
      "learning_rate": 4.353755841914325e-05,
      "loss": 0.46396403312683104,
      "num_input_tokens_seen": 906127424,
      "step": 3190,
      "train_runtime": 30988.7253,
      "train_tokens_per_second": 29240.552
    },
    {
      "epoch": 0.11322248439320931,
      "grad_norm": 0.56640625,
      "learning_rate": 4.3521062545244116e-05,
      "loss": 0.46085448265075685,
      "num_input_tokens_seen": 908971136,
      "step": 3200,
      "train_runtime": 31087.1403,
      "train_tokens_per_second": 29239.458
    },
    {
      "epoch": 0.11357630465693809,
      "grad_norm": 0.56640625,
      "learning_rate": 4.350458540743126e-05,
      "loss": 0.4626326560974121,
      "num_input_tokens_seen": 911782080,
      "step": 3210,
      "train_runtime": 31181.442,
      "train_tokens_per_second": 29241.178
    },
    {
      "epoch": 0.11393012492066686,
      "grad_norm": 0.578125,
      "learning_rate": 4.3488126970263955e-05,
      "loss": 0.459613561630249,
      "num_input_tokens_seen": 914660992,
      "step": 3220,
      "train_runtime": 31280.785,
      "train_tokens_per_second": 29240.347
    },
    {
      "epoch": 0.11428394518439564,
      "grad_norm": 0.56640625,
      "learning_rate": 4.347168719839527e-05,
      "loss": 0.456086254119873,
      "num_input_tokens_seen": 917527296,
      "step": 3230,
      "train_runtime": 31376.2338,
      "train_tokens_per_second": 29242.748
    },
    {
      "epoch": 0.11463776544812442,
      "grad_norm": 0.578125,
      "learning_rate": 4.345526605657173e-05,
      "loss": 0.45608949661254883,
      "num_input_tokens_seen": 920384640,
      "step": 3240,
      "train_runtime": 31474.9727,
      "train_tokens_per_second": 29241.793
    },
    {
      "epoch": 0.1149915857118532,
      "grad_norm": 0.5703125,
      "learning_rate": 4.343886350963304e-05,
      "loss": 0.460129451751709,
      "num_input_tokens_seen": 923238016,
      "step": 3250,
      "train_runtime": 31574.2738,
      "train_tokens_per_second": 29240.198
    },
    {
      "epoch": 0.11534540597558197,
      "grad_norm": 0.58203125,
      "learning_rate": 4.3422479522511697e-05,
      "loss": 0.4605868339538574,
      "num_input_tokens_seen": 926139264,
      "step": 3260,
      "train_runtime": 31677.8043,
      "train_tokens_per_second": 29236.22
    },
    {
      "epoch": 0.11569922623931075,
      "grad_norm": 0.59375,
      "learning_rate": 4.340611406023272e-05,
      "loss": 0.4636395454406738,
      "num_input_tokens_seen": 928941120,
      "step": 3270,
      "train_runtime": 31770.5738,
      "train_tokens_per_second": 29239.041
    },
    {
      "epoch": 0.11605304650303953,
      "grad_norm": 0.58984375,
      "learning_rate": 4.338976708791336e-05,
      "loss": 0.45813422203063964,
      "num_input_tokens_seen": 931777472,
      "step": 3280,
      "train_runtime": 31867.5449,
      "train_tokens_per_second": 29239.073
    },
    {
      "epoch": 0.11640686676676831,
      "grad_norm": 0.5703125,
      "learning_rate": 4.337343857076272e-05,
      "loss": 0.45672240257263186,
      "num_input_tokens_seen": 934654080,
      "step": 3290,
      "train_runtime": 31967.5512,
      "train_tokens_per_second": 29237.588
    },
    {
      "epoch": 0.11676068703049709,
      "grad_norm": 0.578125,
      "learning_rate": 4.33571284740815e-05,
      "loss": 0.4613646984100342,
      "num_input_tokens_seen": 937466048,
      "step": 3300,
      "train_runtime": 32062.7104,
      "train_tokens_per_second": 29238.515
    },
    {
      "epoch": 0.11711450729422587,
      "grad_norm": 0.5859375,
      "learning_rate": 4.3340836763261675e-05,
      "loss": 0.46317329406738283,
      "num_input_tokens_seen": 940354688,
      "step": 3310,
      "train_runtime": 32165.0962,
      "train_tokens_per_second": 29235.252
    },
    {
      "epoch": 0.11746832755795465,
      "grad_norm": 0.59375,
      "learning_rate": 4.332456340378618e-05,
      "loss": 0.45850648880004885,
      "num_input_tokens_seen": 943173696,
      "step": 3320,
      "train_runtime": 32260.9528,
      "train_tokens_per_second": 29235.767
    },
    {
      "epoch": 0.11782214782168343,
      "grad_norm": 0.58203125,
      "learning_rate": 4.3308308361228586e-05,
      "loss": 0.4607564926147461,
      "num_input_tokens_seen": 946035008,
      "step": 3330,
      "train_runtime": 32359.7855,
      "train_tokens_per_second": 29234.897
    },
    {
      "epoch": 0.11817596808541221,
      "grad_norm": 0.609375,
      "learning_rate": 4.329207160125282e-05,
      "loss": 0.45890512466430666,
      "num_input_tokens_seen": 948837056,
      "step": 3340,
      "train_runtime": 32453.4983,
      "train_tokens_per_second": 29236.819
    },
    {
      "epoch": 0.11852978834914099,
      "grad_norm": 0.56640625,
      "learning_rate": 4.327585308961287e-05,
      "loss": 0.4617131233215332,
      "num_input_tokens_seen": 951684928,
      "step": 3350,
      "train_runtime": 32551.7129,
      "train_tokens_per_second": 29236.094
    },
    {
      "epoch": 0.11888360861286977,
      "grad_norm": 0.55078125,
      "learning_rate": 4.325965279215243e-05,
      "loss": 0.4547280311584473,
      "num_input_tokens_seen": 954490496,
      "step": 3360,
      "train_runtime": 32644.7359,
      "train_tokens_per_second": 29238.726
    },
    {
      "epoch": 0.11923742887659855,
      "grad_norm": 0.57421875,
      "learning_rate": 4.3243470674804686e-05,
      "loss": 0.45595827102661135,
      "num_input_tokens_seen": 957301120,
      "step": 3370,
      "train_runtime": 32739.4169,
      "train_tokens_per_second": 29240.017
    },
    {
      "epoch": 0.11959124914032733,
      "grad_norm": 0.59375,
      "learning_rate": 4.3227306703591904e-05,
      "loss": 0.4553481101989746,
      "num_input_tokens_seen": 960131584,
      "step": 3380,
      "train_runtime": 32834.606,
      "train_tokens_per_second": 29241.453
    },
    {
      "epoch": 0.11994506940405611,
      "grad_norm": 0.5546875,
      "learning_rate": 4.32111608446252e-05,
      "loss": 0.454563045501709,
      "num_input_tokens_seen": 962939648,
      "step": 3390,
      "train_runtime": 32926.6867,
      "train_tokens_per_second": 29244.96
    },
    {
      "epoch": 0.12029888966778489,
      "grad_norm": 0.55859375,
      "learning_rate": 4.319503306410426e-05,
      "loss": 0.4604179382324219,
      "num_input_tokens_seen": 965777408,
      "step": 3400,
      "train_runtime": 33019.0209,
      "train_tokens_per_second": 29249.123
    },
    {
      "epoch": 0.12065270993151367,
      "grad_norm": 0.58984375,
      "learning_rate": 4.317892332831699e-05,
      "loss": 0.4535073280334473,
      "num_input_tokens_seen": 968573504,
      "step": 3410,
      "train_runtime": 33111.6067,
      "train_tokens_per_second": 29251.782
    },
    {
      "epoch": 0.12100653019524245,
      "grad_norm": 0.58984375,
      "learning_rate": 4.316283160363922e-05,
      "loss": 0.4528909683227539,
      "num_input_tokens_seen": 971386240,
      "step": 3420,
      "train_runtime": 33205.2318,
      "train_tokens_per_second": 29254.012
    },
    {
      "epoch": 0.12136035045897123,
      "grad_norm": 0.5703125,
      "learning_rate": 4.314675785653447e-05,
      "loss": 0.4609466552734375,
      "num_input_tokens_seen": 974241984,
      "step": 3430,
      "train_runtime": 33304.4337,
      "train_tokens_per_second": 29252.621
    },
    {
      "epoch": 0.12171417072270001,
      "grad_norm": 0.55859375,
      "learning_rate": 4.3130702053553606e-05,
      "loss": 0.4554624557495117,
      "num_input_tokens_seen": 977055936,
      "step": 3440,
      "train_runtime": 33398.0832,
      "train_tokens_per_second": 29254.851
    },
    {
      "epoch": 0.12206799098642879,
      "grad_norm": 0.5703125,
      "learning_rate": 4.3114664161334546e-05,
      "loss": 0.45702285766601564,
      "num_input_tokens_seen": 979927168,
      "step": 3450,
      "train_runtime": 33495.4348,
      "train_tokens_per_second": 29255.544
    },
    {
      "epoch": 0.12242181125015757,
      "grad_norm": 0.58984375,
      "learning_rate": 4.3098644146601984e-05,
      "loss": 0.4516008377075195,
      "num_input_tokens_seen": 982789760,
      "step": 3460,
      "train_runtime": 33595.244,
      "train_tokens_per_second": 29253.836
    },
    {
      "epoch": 0.12277563151388635,
      "grad_norm": 0.5703125,
      "learning_rate": 4.30826419761671e-05,
      "loss": 0.4585239410400391,
      "num_input_tokens_seen": 985613760,
      "step": 3470,
      "train_runtime": 33690.508,
      "train_tokens_per_second": 29254.939
    },
    {
      "epoch": 0.12312945177761511,
      "grad_norm": 0.59375,
      "learning_rate": 4.30666576169273e-05,
      "loss": 0.4601104259490967,
      "num_input_tokens_seen": 988397824,
      "step": 3480,
      "train_runtime": 33780.5111,
      "train_tokens_per_second": 29259.41
    },
    {
      "epoch": 0.12348327204134389,
      "grad_norm": 0.65625,
      "learning_rate": 4.305069103586585e-05,
      "loss": 0.4570477485656738,
      "num_input_tokens_seen": 991241024,
      "step": 3490,
      "train_runtime": 33873.8417,
      "train_tokens_per_second": 29262.728
    },
    {
      "epoch": 0.12383709230507267,
      "grad_norm": 0.6015625,
      "learning_rate": 4.303474220005164e-05,
      "loss": 0.45610885620117186,
      "num_input_tokens_seen": 994065856,
      "step": 3500,
      "train_runtime": 33967.4752,
      "train_tokens_per_second": 29265.227
    },
    {
      "epoch": 0.12419091256880145,
      "grad_norm": 0.58203125,
      "learning_rate": 4.3018811076638944e-05,
      "loss": 0.4574295997619629,
      "num_input_tokens_seen": 996866560,
      "step": 3510,
      "train_runtime": 34063.3403,
      "train_tokens_per_second": 29265.085
    },
    {
      "epoch": 0.12454473283253023,
      "grad_norm": 0.5859375,
      "learning_rate": 4.300289763286704e-05,
      "loss": 0.45502915382385256,
      "num_input_tokens_seen": 999720064,
      "step": 3520,
      "train_runtime": 34159.3455,
      "train_tokens_per_second": 29266.371
    },
    {
      "epoch": 0.12489855309625901,
      "grad_norm": 0.6015625,
      "learning_rate": 4.298700183606e-05,
      "loss": 0.4610752105712891,
      "num_input_tokens_seen": 1002563968,
      "step": 3530,
      "train_runtime": 34257.4393,
      "train_tokens_per_second": 29265.584
    },
    {
      "epoch": 0.1252523733599878,
      "grad_norm": 0.58203125,
      "learning_rate": 4.297112365362637e-05,
      "loss": 0.4594162940979004,
      "num_input_tokens_seen": 1005402176,
      "step": 3540,
      "train_runtime": 34353.0211,
      "train_tokens_per_second": 29266.776
    },
    {
      "epoch": 0.12560619362371658,
      "grad_norm": 0.55078125,
      "learning_rate": 4.295526305305891e-05,
      "loss": 0.45300869941711425,
      "num_input_tokens_seen": 1008258880,
      "step": 3550,
      "train_runtime": 34449.2917,
      "train_tokens_per_second": 29267.913
    },
    {
      "epoch": 0.12596001388744535,
      "grad_norm": 0.6015625,
      "learning_rate": 4.293942000193429e-05,
      "loss": 0.45346708297729493,
      "num_input_tokens_seen": 1011047872,
      "step": 3560,
      "train_runtime": 34544.805,
      "train_tokens_per_second": 29267.726
    },
    {
      "epoch": 0.12631383415117414,
      "grad_norm": 0.58984375,
      "learning_rate": 4.2923594467912866e-05,
      "loss": 0.45587739944458006,
      "num_input_tokens_seen": 1013899392,
      "step": 3570,
      "train_runtime": 34639.7096,
      "train_tokens_per_second": 29269.858
    },
    {
      "epoch": 0.1266676544149029,
      "grad_norm": 0.578125,
      "learning_rate": 4.290778641873832e-05,
      "loss": 0.45630655288696287,
      "num_input_tokens_seen": 1016776256,
      "step": 3580,
      "train_runtime": 34736.7158,
      "train_tokens_per_second": 29270.938
    },
    {
      "epoch": 0.1270214746786317,
      "grad_norm": 0.5625,
      "learning_rate": 4.2891995822237455e-05,
      "loss": 0.45297880172729493,
      "num_input_tokens_seen": 1019636224,
      "step": 3590,
      "train_runtime": 34833.7922,
      "train_tokens_per_second": 29271.468
    },
    {
      "epoch": 0.12737529494236047,
      "grad_norm": 0.578125,
      "learning_rate": 4.28762226463199e-05,
      "loss": 0.45767717361450194,
      "num_input_tokens_seen": 1022491264,
      "step": 3600,
      "train_runtime": 34931.1145,
      "train_tokens_per_second": 29271.647
    },
    {
      "epoch": 0.12772911520608923,
      "grad_norm": 0.59375,
      "learning_rate": 4.286046685897781e-05,
      "loss": 0.4487922191619873,
      "num_input_tokens_seen": 1025290688,
      "step": 3610,
      "train_runtime": 35024.2945,
      "train_tokens_per_second": 29273.7
    },
    {
      "epoch": 0.12808293546981803,
      "grad_norm": 0.57421875,
      "learning_rate": 4.284472842828562e-05,
      "loss": 0.4598641872406006,
      "num_input_tokens_seen": 1028080896,
      "step": 3620,
      "train_runtime": 35115.3025,
      "train_tokens_per_second": 29277.29
    },
    {
      "epoch": 0.1284367557335468,
      "grad_norm": 0.5546875,
      "learning_rate": 4.282900732239977e-05,
      "loss": 0.45454702377319334,
      "num_input_tokens_seen": 1030951360,
      "step": 3630,
      "train_runtime": 35212.0358,
      "train_tokens_per_second": 29278.38
    },
    {
      "epoch": 0.12879057599727559,
      "grad_norm": 0.57421875,
      "learning_rate": 4.281330350955845e-05,
      "loss": 0.45328598022460936,
      "num_input_tokens_seen": 1033796352,
      "step": 3640,
      "train_runtime": 35309.0996,
      "train_tokens_per_second": 29278.468
    },
    {
      "epoch": 0.12914439626100435,
      "grad_norm": 0.578125,
      "learning_rate": 4.279761695808125e-05,
      "loss": 0.45642919540405275,
      "num_input_tokens_seen": 1036626624,
      "step": 3650,
      "train_runtime": 35405.8836,
      "train_tokens_per_second": 29278.372
    },
    {
      "epoch": 0.12949821652473315,
      "grad_norm": 0.56640625,
      "learning_rate": 4.278194763636904e-05,
      "loss": 0.45912914276123046,
      "num_input_tokens_seen": 1039479552,
      "step": 3660,
      "train_runtime": 35504.7294,
      "train_tokens_per_second": 29277.214
    },
    {
      "epoch": 0.1298520367884619,
      "grad_norm": 0.58984375,
      "learning_rate": 4.276629551290354e-05,
      "loss": 0.4563636779785156,
      "num_input_tokens_seen": 1042291904,
      "step": 3670,
      "train_runtime": 35600.895,
      "train_tokens_per_second": 29277.126
    },
    {
      "epoch": 0.1302058570521907,
      "grad_norm": 0.5703125,
      "learning_rate": 4.2750660556247175e-05,
      "loss": 0.45493526458740235,
      "num_input_tokens_seen": 1045103104,
      "step": 3680,
      "train_runtime": 35697.7762,
      "train_tokens_per_second": 29276.42
    },
    {
      "epoch": 0.13055967731591947,
      "grad_norm": 0.5703125,
      "learning_rate": 4.273504273504274e-05,
      "loss": 0.4566159725189209,
      "num_input_tokens_seen": 1047985280,
      "step": 3690,
      "train_runtime": 35795.4584,
      "train_tokens_per_second": 29277.046
    },
    {
      "epoch": 0.13091349757964826,
      "grad_norm": 0.5859375,
      "learning_rate": 4.271944201801317e-05,
      "loss": 0.4487285614013672,
      "num_input_tokens_seen": 1050893376,
      "step": 3700,
      "train_runtime": 35895.123,
      "train_tokens_per_second": 29276.773
    },
    {
      "epoch": 0.13126731784337703,
      "grad_norm": 0.57421875,
      "learning_rate": 4.270385837396127e-05,
      "loss": 0.4545039176940918,
      "num_input_tokens_seen": 1053785536,
      "step": 3710,
      "train_runtime": 35994.6059,
      "train_tokens_per_second": 29276.207
    },
    {
      "epoch": 0.13162113810710582,
      "grad_norm": 0.5859375,
      "learning_rate": 4.268829177176945e-05,
      "loss": 0.45177464485168456,
      "num_input_tokens_seen": 1056617280,
      "step": 3720,
      "train_runtime": 36088.1237,
      "train_tokens_per_second": 29278.809
    },
    {
      "epoch": 0.1319749583708346,
      "grad_norm": 0.5703125,
      "learning_rate": 4.2672742180399455e-05,
      "loss": 0.4510002136230469,
      "num_input_tokens_seen": 1059486912,
      "step": 3730,
      "train_runtime": 36186.0883,
      "train_tokens_per_second": 29278.846
    },
    {
      "epoch": 0.13232877863456338,
      "grad_norm": 0.58203125,
      "learning_rate": 4.265720956889213e-05,
      "loss": 0.4509366512298584,
      "num_input_tokens_seen": 1062342016,
      "step": 3740,
      "train_runtime": 36283.4094,
      "train_tokens_per_second": 29279.002
    },
    {
      "epoch": 0.13268259889829215,
      "grad_norm": 0.59375,
      "learning_rate": 4.2641693906367113e-05,
      "loss": 0.45595321655273435,
      "num_input_tokens_seen": 1065209920,
      "step": 3750,
      "train_runtime": 36384.0012,
      "train_tokens_per_second": 29276.877
    },
    {
      "epoch": 0.13303641916202094,
      "grad_norm": 0.57421875,
      "learning_rate": 4.2626195162022646e-05,
      "loss": 0.4538719654083252,
      "num_input_tokens_seen": 1068011200,
      "step": 3760,
      "train_runtime": 36477.8307,
      "train_tokens_per_second": 29278.364
    },
    {
      "epoch": 0.1333902394257497,
      "grad_norm": 0.5625,
      "learning_rate": 4.2610713305135255e-05,
      "loss": 0.45287380218505857,
      "num_input_tokens_seen": 1070900608,
      "step": 3770,
      "train_runtime": 36579.4611,
      "train_tokens_per_second": 29276.008
    },
    {
      "epoch": 0.1337440596894785,
      "grad_norm": 0.55078125,
      "learning_rate": 4.2595248305059546e-05,
      "loss": 0.4500288009643555,
      "num_input_tokens_seen": 1073795328,
      "step": 3780,
      "train_runtime": 36680.6327,
      "train_tokens_per_second": 29274.177
    },
    {
      "epoch": 0.13409787995320727,
      "grad_norm": 0.58203125,
      "learning_rate": 4.2579800131227916e-05,
      "loss": 0.4546367645263672,
      "num_input_tokens_seen": 1076604416,
      "step": 3790,
      "train_runtime": 36774.6455,
      "train_tokens_per_second": 29275.725
    },
    {
      "epoch": 0.13445170021693606,
      "grad_norm": 0.5703125,
      "learning_rate": 4.256436875315028e-05,
      "loss": 0.45539026260375975,
      "num_input_tokens_seen": 1079443520,
      "step": 3800,
      "train_runtime": 36867.7115,
      "train_tokens_per_second": 29278.832
    },
    {
      "epoch": 0.13480552048066483,
      "grad_norm": 0.58203125,
      "learning_rate": 4.2548954140413895e-05,
      "loss": 0.4538921356201172,
      "num_input_tokens_seen": 1082304064,
      "step": 3810,
      "train_runtime": 36964.165,
      "train_tokens_per_second": 29279.819
    },
    {
      "epoch": 0.13515934074439362,
      "grad_norm": 0.5703125,
      "learning_rate": 4.253355626268302e-05,
      "loss": 0.4505466938018799,
      "num_input_tokens_seen": 1085158976,
      "step": 3820,
      "train_runtime": 37061.8534,
      "train_tokens_per_second": 29279.674
    },
    {
      "epoch": 0.13551316100812238,
      "grad_norm": 0.609375,
      "learning_rate": 4.2518175089698716e-05,
      "loss": 0.4554553508758545,
      "num_input_tokens_seen": 1088037632,
      "step": 3830,
      "train_runtime": 37162.5735,
      "train_tokens_per_second": 29277.779
    },
    {
      "epoch": 0.13586698127185118,
      "grad_norm": 0.57421875,
      "learning_rate": 4.25028105912786e-05,
      "loss": 0.45158748626708983,
      "num_input_tokens_seen": 1090901952,
      "step": 3840,
      "train_runtime": 37265.3789,
      "train_tokens_per_second": 29273.873
    },
    {
      "epoch": 0.13622080153557994,
      "grad_norm": 0.59765625,
      "learning_rate": 4.2487462737316565e-05,
      "loss": 0.45415143966674804,
      "num_input_tokens_seen": 1093725824,
      "step": 3850,
      "train_runtime": 37362.0133,
      "train_tokens_per_second": 29273.739
    },
    {
      "epoch": 0.13657462179930874,
      "grad_norm": 0.5859375,
      "learning_rate": 4.2472131497782555e-05,
      "loss": 0.4521495819091797,
      "num_input_tokens_seen": 1096502720,
      "step": 3860,
      "train_runtime": 37453.3127,
      "train_tokens_per_second": 29276.522
    },
    {
      "epoch": 0.1369284420630375,
      "grad_norm": 0.58203125,
      "learning_rate": 4.245681684272231e-05,
      "loss": 0.4537048816680908,
      "num_input_tokens_seen": 1099356800,
      "step": 3870,
      "train_runtime": 37549.847,
      "train_tokens_per_second": 29277.264
    },
    {
      "epoch": 0.13728226232676627,
      "grad_norm": 0.57421875,
      "learning_rate": 4.244151874225712e-05,
      "loss": 0.45152950286865234,
      "num_input_tokens_seen": 1102169280,
      "step": 3880,
      "train_runtime": 37644.0084,
      "train_tokens_per_second": 29278.744
    },
    {
      "epoch": 0.13763608259049506,
      "grad_norm": 0.58984375,
      "learning_rate": 4.2426237166583596e-05,
      "loss": 0.44892234802246095,
      "num_input_tokens_seen": 1105003328,
      "step": 3890,
      "train_runtime": 37743.2785,
      "train_tokens_per_second": 29276.824
    },
    {
      "epoch": 0.13798990285422383,
      "grad_norm": 0.56640625,
      "learning_rate": 4.241097208597339e-05,
      "loss": 0.44681386947631835,
      "num_input_tokens_seen": 1107821184,
      "step": 3900,
      "train_runtime": 37837.5494,
      "train_tokens_per_second": 29278.354
    },
    {
      "epoch": 0.13834372311795262,
      "grad_norm": 0.5390625,
      "learning_rate": 4.2395723470773005e-05,
      "loss": 0.45065946578979493,
      "num_input_tokens_seen": 1110715968,
      "step": 3910,
      "train_runtime": 37936.5788,
      "train_tokens_per_second": 29278.232
    },
    {
      "epoch": 0.1386975433816814,
      "grad_norm": 0.5390625,
      "learning_rate": 4.238049129140347e-05,
      "loss": 0.45148382186889646,
      "num_input_tokens_seen": 1113568000,
      "step": 3920,
      "train_runtime": 38035.0983,
      "train_tokens_per_second": 29277.379
    },
    {
      "epoch": 0.13905136364541018,
      "grad_norm": 0.59765625,
      "learning_rate": 4.236527551836022e-05,
      "loss": 0.4460158348083496,
      "num_input_tokens_seen": 1116387136,
      "step": 3930,
      "train_runtime": 38129.4486,
      "train_tokens_per_second": 29278.869
    },
    {
      "epoch": 0.13940518390913895,
      "grad_norm": 0.5859375,
      "learning_rate": 4.235007612221274e-05,
      "loss": 0.447994327545166,
      "num_input_tokens_seen": 1119228928,
      "step": 3940,
      "train_runtime": 38223.5096,
      "train_tokens_per_second": 29281.166
    },
    {
      "epoch": 0.13975900417286774,
      "grad_norm": 0.56640625,
      "learning_rate": 4.2334893073604386e-05,
      "loss": 0.44850983619689944,
      "num_input_tokens_seen": 1122038912,
      "step": 3950,
      "train_runtime": 38316.52,
      "train_tokens_per_second": 29283.424
    },
    {
      "epoch": 0.1401128244365965,
      "grad_norm": 0.55859375,
      "learning_rate": 4.231972634325214e-05,
      "loss": 0.45399012565612795,
      "num_input_tokens_seen": 1124911232,
      "step": 3960,
      "train_runtime": 38416.7375,
      "train_tokens_per_second": 29281.8
    },
    {
      "epoch": 0.1404666447003253,
      "grad_norm": 0.55859375,
      "learning_rate": 4.230457590194635e-05,
      "loss": 0.4500166893005371,
      "num_input_tokens_seen": 1127771200,
      "step": 3970,
      "train_runtime": 38514.8297,
      "train_tokens_per_second": 29281.48
    },
    {
      "epoch": 0.14082046496405407,
      "grad_norm": 0.53125,
      "learning_rate": 4.228944172055053e-05,
      "loss": 0.4504100799560547,
      "num_input_tokens_seen": 1130658240,
      "step": 3980,
      "train_runtime": 38614.1783,
      "train_tokens_per_second": 29280.909
    },
    {
      "epoch": 0.14117428522778286,
      "grad_norm": 0.609375,
      "learning_rate": 4.22743237700011e-05,
      "loss": 0.451992130279541,
      "num_input_tokens_seen": 1133511744,
      "step": 3990,
      "train_runtime": 38712.8516,
      "train_tokens_per_second": 29279.986
    },
    {
      "epoch": 0.14152810549151162,
      "grad_norm": 0.5859375,
      "learning_rate": 4.225922202130716e-05,
      "loss": 0.4450711727142334,
      "num_input_tokens_seen": 1136350848,
      "step": 4000,
      "train_runtime": 38809.7896,
      "train_tokens_per_second": 29280.005
    },
    {
      "epoch": 0.14188192575524042,
      "grad_norm": 0.57421875,
      "learning_rate": 4.224413644555024e-05,
      "loss": 0.4510810375213623,
      "num_input_tokens_seen": 1139143808,
      "step": 4010,
      "train_runtime": 38903.4318,
      "train_tokens_per_second": 29281.319
    },
    {
      "epoch": 0.14223574601896918,
      "grad_norm": 0.625,
      "learning_rate": 4.222906701388411e-05,
      "loss": 0.45055713653564455,
      "num_input_tokens_seen": 1141969856,
      "step": 4020,
      "train_runtime": 39002.1782,
      "train_tokens_per_second": 29279.643
    },
    {
      "epoch": 0.14258956628269798,
      "grad_norm": 0.58984375,
      "learning_rate": 4.2214013697534466e-05,
      "loss": 0.45257959365844724,
      "num_input_tokens_seen": 1144803136,
      "step": 4030,
      "train_runtime": 39100.3578,
      "train_tokens_per_second": 29278.585
    },
    {
      "epoch": 0.14294338654642674,
      "grad_norm": 0.58203125,
      "learning_rate": 4.219897646779882e-05,
      "loss": 0.4558560371398926,
      "num_input_tokens_seen": 1147692672,
      "step": 4040,
      "train_runtime": 39202.9539,
      "train_tokens_per_second": 29275.668
    },
    {
      "epoch": 0.14329720681015554,
      "grad_norm": 0.55078125,
      "learning_rate": 4.2183955296046145e-05,
      "loss": 0.45154776573181155,
      "num_input_tokens_seen": 1150605952,
      "step": 4050,
      "train_runtime": 39305.4902,
      "train_tokens_per_second": 29273.416
    },
    {
      "epoch": 0.1436510270738843,
      "grad_norm": 0.55859375,
      "learning_rate": 4.2168950153716746e-05,
      "loss": 0.4491125106811523,
      "num_input_tokens_seen": 1153387008,
      "step": 4060,
      "train_runtime": 39398.6764,
      "train_tokens_per_second": 29274.765
    },
    {
      "epoch": 0.1440048473376131,
      "grad_norm": 0.5625,
      "learning_rate": 4.215396101232197e-05,
      "loss": 0.44748859405517577,
      "num_input_tokens_seen": 1156206208,
      "step": 4070,
      "train_runtime": 39492.9816,
      "train_tokens_per_second": 29276.245
    },
    {
      "epoch": 0.14435866760134186,
      "grad_norm": 0.58984375,
      "learning_rate": 4.213898784344398e-05,
      "loss": 0.4471442699432373,
      "num_input_tokens_seen": 1159107648,
      "step": 4080,
      "train_runtime": 39594.0109,
      "train_tokens_per_second": 29274.823
    },
    {
      "epoch": 0.14471248786507065,
      "grad_norm": 0.55859375,
      "learning_rate": 4.21240306187356e-05,
      "loss": 0.44981160163879397,
      "num_input_tokens_seen": 1161944576,
      "step": 4090,
      "train_runtime": 39691.417,
      "train_tokens_per_second": 29274.454
    },
    {
      "epoch": 0.14506630812879942,
      "grad_norm": 0.55859375,
      "learning_rate": 4.2109089309919967e-05,
      "loss": 0.4546297550201416,
      "num_input_tokens_seen": 1164828992,
      "step": 4100,
      "train_runtime": 39790.9805,
      "train_tokens_per_second": 29273.694
    },
    {
      "epoch": 0.14542012839252821,
      "grad_norm": 0.57421875,
      "learning_rate": 4.2094163888790445e-05,
      "loss": 0.4490847110748291,
      "num_input_tokens_seen": 1167660480,
      "step": 4110,
      "train_runtime": 39886.9364,
      "train_tokens_per_second": 29274.258
    },
    {
      "epoch": 0.14577394865625698,
      "grad_norm": 0.55859375,
      "learning_rate": 4.2079254327210294e-05,
      "loss": 0.445095157623291,
      "num_input_tokens_seen": 1170493056,
      "step": 4120,
      "train_runtime": 39985.3671,
      "train_tokens_per_second": 29273.035
    },
    {
      "epoch": 0.14612776891998575,
      "grad_norm": 0.57421875,
      "learning_rate": 4.206436059711249e-05,
      "loss": 0.45116581916809084,
      "num_input_tokens_seen": 1173303936,
      "step": 4130,
      "train_runtime": 40081.9337,
      "train_tokens_per_second": 29272.638
    },
    {
      "epoch": 0.14648158918371454,
      "grad_norm": 0.57421875,
      "learning_rate": 4.20494826704995e-05,
      "loss": 0.4513398170471191,
      "num_input_tokens_seen": 1176151616,
      "step": 4140,
      "train_runtime": 40181.3332,
      "train_tokens_per_second": 29271.095
    },
    {
      "epoch": 0.1468354094474433,
      "grad_norm": 0.578125,
      "learning_rate": 4.203462051944307e-05,
      "loss": 0.44718356132507325,
      "num_input_tokens_seen": 1179087552,
      "step": 4150,
      "train_runtime": 40285.7438,
      "train_tokens_per_second": 29268.109
    },
    {
      "epoch": 0.1471892297111721,
      "grad_norm": 0.57421875,
      "learning_rate": 4.201977411608398e-05,
      "loss": 0.4508021354675293,
      "num_input_tokens_seen": 1181852160,
      "step": 4160,
      "train_runtime": 40377.9135,
      "train_tokens_per_second": 29269.768
    },
    {
      "epoch": 0.14754304997490086,
      "grad_norm": 0.58203125,
      "learning_rate": 4.200494343263185e-05,
      "loss": 0.4480936527252197,
      "num_input_tokens_seen": 1184651008,
      "step": 4170,
      "train_runtime": 40473.0686,
      "train_tokens_per_second": 29270.106
    },
    {
      "epoch": 0.14789687023862966,
      "grad_norm": 0.5546875,
      "learning_rate": 4.1990128441364914e-05,
      "loss": 0.4518306732177734,
      "num_input_tokens_seen": 1187485120,
      "step": 4180,
      "train_runtime": 40570.2833,
      "train_tokens_per_second": 29269.826
    },
    {
      "epoch": 0.14825069050235842,
      "grad_norm": 0.59375,
      "learning_rate": 4.197532911462977e-05,
      "loss": 0.44388160705566404,
      "num_input_tokens_seen": 1190325248,
      "step": 4190,
      "train_runtime": 40669.8782,
      "train_tokens_per_second": 29267.982
    },
    {
      "epoch": 0.14860451076608722,
      "grad_norm": 0.58203125,
      "learning_rate": 4.196054542484125e-05,
      "loss": 0.4475285530090332,
      "num_input_tokens_seen": 1193193088,
      "step": 4200,
      "train_runtime": 40769.9021,
      "train_tokens_per_second": 29266.518
    },
    {
      "epoch": 0.14895833102981598,
      "grad_norm": 0.546875,
      "learning_rate": 4.1945777344482084e-05,
      "loss": 0.44704599380493165,
      "num_input_tokens_seen": 1196034560,
      "step": 4210,
      "train_runtime": 40866.0938,
      "train_tokens_per_second": 29267.161
    },
    {
      "epoch": 0.14931215129354478,
      "grad_norm": 0.5859375,
      "learning_rate": 4.19310248461028e-05,
      "loss": 0.44858689308166505,
      "num_input_tokens_seen": 1198912832,
      "step": 4220,
      "train_runtime": 40968.5924,
      "train_tokens_per_second": 29264.194
    },
    {
      "epoch": 0.14966597155727354,
      "grad_norm": 0.59375,
      "learning_rate": 4.1916287902321405e-05,
      "loss": 0.44658374786376953,
      "num_input_tokens_seen": 1201798272,
      "step": 4230,
      "train_runtime": 41069.8206,
      "train_tokens_per_second": 29262.321
    },
    {
      "epoch": 0.15001979182100234,
      "grad_norm": 0.59375,
      "learning_rate": 4.190156648582328e-05,
      "loss": 0.4517244338989258,
      "num_input_tokens_seen": 1204654784,
      "step": 4240,
      "train_runtime": 41165.7084,
      "train_tokens_per_second": 29263.55
    },
    {
      "epoch": 0.1503736120847311,
      "grad_norm": 0.6328125,
      "learning_rate": 4.188686056936087e-05,
      "loss": 0.447774600982666,
      "num_input_tokens_seen": 1207475008,
      "step": 4250,
      "train_runtime": 41260.3545,
      "train_tokens_per_second": 29264.775
    },
    {
      "epoch": 0.1507274323484599,
      "grad_norm": 0.5859375,
      "learning_rate": 4.187217012575352e-05,
      "loss": 0.4460301399230957,
      "num_input_tokens_seen": 1210316224,
      "step": 4260,
      "train_runtime": 41356.4397,
      "train_tokens_per_second": 29265.484
    },
    {
      "epoch": 0.15108125261218866,
      "grad_norm": 0.5703125,
      "learning_rate": 4.185749512788727e-05,
      "loss": 0.44805269241333007,
      "num_input_tokens_seen": 1213199552,
      "step": 4270,
      "train_runtime": 41454.8152,
      "train_tokens_per_second": 29265.588
    },
    {
      "epoch": 0.15143507287591745,
      "grad_norm": 0.6015625,
      "learning_rate": 4.184283554871462e-05,
      "loss": 0.4456193923950195,
      "num_input_tokens_seen": 1216018816,
      "step": 4280,
      "train_runtime": 41551.5081,
      "train_tokens_per_second": 29265.335
    },
    {
      "epoch": 0.15178889313964622,
      "grad_norm": 0.5859375,
      "learning_rate": 4.1828191361254344e-05,
      "loss": 0.4457979202270508,
      "num_input_tokens_seen": 1218848896,
      "step": 4290,
      "train_runtime": 41645.8463,
      "train_tokens_per_second": 29266.998
    },
    {
      "epoch": 0.152142713403375,
      "grad_norm": 0.5859375,
      "learning_rate": 4.181356253859127e-05,
      "loss": 0.4439018726348877,
      "num_input_tokens_seen": 1221640448,
      "step": 4300,
      "train_runtime": 41736.532,
      "train_tokens_per_second": 29270.291
    },
    {
      "epoch": 0.15249653366710378,
      "grad_norm": 0.578125,
      "learning_rate": 4.179894905387606e-05,
      "loss": 0.4440751075744629,
      "num_input_tokens_seen": 1224492544,
      "step": 4310,
      "train_runtime": 41833.8649,
      "train_tokens_per_second": 29270.366
    },
    {
      "epoch": 0.15285035393083257,
      "grad_norm": 0.5703125,
      "learning_rate": 4.178435088032502e-05,
      "loss": 0.4444403171539307,
      "num_input_tokens_seen": 1227310592,
      "step": 4320,
      "train_runtime": 41929.8447,
      "train_tokens_per_second": 29270.573
    },
    {
      "epoch": 0.15320417419456134,
      "grad_norm": 0.58984375,
      "learning_rate": 4.176976799121989e-05,
      "loss": 0.4419120788574219,
      "num_input_tokens_seen": 1230119616,
      "step": 4330,
      "train_runtime": 42026.5755,
      "train_tokens_per_second": 29270.042
    },
    {
      "epoch": 0.15355799445829013,
      "grad_norm": 0.59765625,
      "learning_rate": 4.1755200359907657e-05,
      "loss": 0.4464087009429932,
      "num_input_tokens_seen": 1232923648,
      "step": 4340,
      "train_runtime": 42120.9797,
      "train_tokens_per_second": 29271.011
    },
    {
      "epoch": 0.1539118147220189,
      "grad_norm": 0.55859375,
      "learning_rate": 4.174064795980028e-05,
      "loss": 0.4458340644836426,
      "num_input_tokens_seen": 1235791296,
      "step": 4350,
      "train_runtime": 42217.6846,
      "train_tokens_per_second": 29271.887
    },
    {
      "epoch": 0.1542656349857477,
      "grad_norm": 0.58984375,
      "learning_rate": 4.17261107643746e-05,
      "loss": 0.45078067779541015,
      "num_input_tokens_seen": 1238657856,
      "step": 4360,
      "train_runtime": 42318.7128,
      "train_tokens_per_second": 29269.743
    },
    {
      "epoch": 0.15461945524947646,
      "grad_norm": 0.5703125,
      "learning_rate": 4.171158874717204e-05,
      "loss": 0.4443507671356201,
      "num_input_tokens_seen": 1241488704,
      "step": 4370,
      "train_runtime": 42414.3771,
      "train_tokens_per_second": 29270.469
    },
    {
      "epoch": 0.15497327551320522,
      "grad_norm": 0.5703125,
      "learning_rate": 4.169708188179844e-05,
      "loss": 0.44482998847961425,
      "num_input_tokens_seen": 1244296576,
      "step": 4380,
      "train_runtime": 42507.6357,
      "train_tokens_per_second": 29272.307
    },
    {
      "epoch": 0.15532709577693402,
      "grad_norm": 0.58203125,
      "learning_rate": 4.1682590141923846e-05,
      "loss": 0.44451127052307127,
      "num_input_tokens_seen": 1247104320,
      "step": 4390,
      "train_runtime": 42602.7056,
      "train_tokens_per_second": 29272.89
    },
    {
      "epoch": 0.15568091604066278,
      "grad_norm": 0.57421875,
      "learning_rate": 4.1668113501282335e-05,
      "loss": 0.4432711601257324,
      "num_input_tokens_seen": 1249909376,
      "step": 4400,
      "train_runtime": 42698.7455,
      "train_tokens_per_second": 29272.742
    },
    {
      "epoch": 0.15603473630439157,
      "grad_norm": 0.578125,
      "learning_rate": 4.165365193367178e-05,
      "loss": 0.4427494049072266,
      "num_input_tokens_seen": 1252730048,
      "step": 4410,
      "train_runtime": 42794.217,
      "train_tokens_per_second": 29273.349
    },
    {
      "epoch": 0.15638855656812034,
      "grad_norm": 0.5859375,
      "learning_rate": 4.163920541295369e-05,
      "loss": 0.44465007781982424,
      "num_input_tokens_seen": 1255548928,
      "step": 4420,
      "train_runtime": 42890.1209,
      "train_tokens_per_second": 29273.616
    },
    {
      "epoch": 0.15674237683184913,
      "grad_norm": 0.5703125,
      "learning_rate": 4.1624773913052946e-05,
      "loss": 0.44714765548706054,
      "num_input_tokens_seen": 1258380608,
      "step": 4430,
      "train_runtime": 42988.7589,
      "train_tokens_per_second": 29272.318
    },
    {
      "epoch": 0.1570961970955779,
      "grad_norm": 0.6171875,
      "learning_rate": 4.161035740795769e-05,
      "loss": 0.4462038516998291,
      "num_input_tokens_seen": 1261217344,
      "step": 4440,
      "train_runtime": 43084.4082,
      "train_tokens_per_second": 29273.173
    },
    {
      "epoch": 0.1574500173593067,
      "grad_norm": 0.5859375,
      "learning_rate": 4.1595955871719055e-05,
      "loss": 0.44342823028564454,
      "num_input_tokens_seen": 1264081024,
      "step": 4450,
      "train_runtime": 43184.5049,
      "train_tokens_per_second": 29271.634
    },
    {
      "epoch": 0.15780383762303546,
      "grad_norm": 0.6015625,
      "learning_rate": 4.158156927845101e-05,
      "loss": 0.4425454616546631,
      "num_input_tokens_seen": 1266929920,
      "step": 4460,
      "train_runtime": 43280.4797,
      "train_tokens_per_second": 29272.548
    },
    {
      "epoch": 0.15815765788676425,
      "grad_norm": 0.56640625,
      "learning_rate": 4.156719760233016e-05,
      "loss": 0.44738616943359377,
      "num_input_tokens_seen": 1269772864,
      "step": 4470,
      "train_runtime": 43376.465,
      "train_tokens_per_second": 29273.314
    },
    {
      "epoch": 0.15851147815049302,
      "grad_norm": 0.5703125,
      "learning_rate": 4.155284081759552e-05,
      "loss": 0.4473750114440918,
      "num_input_tokens_seen": 1272572224,
      "step": 4480,
      "train_runtime": 43470.5255,
      "train_tokens_per_second": 29274.369
    },
    {
      "epoch": 0.1588652984142218,
      "grad_norm": 0.60546875,
      "learning_rate": 4.1538498898548356e-05,
      "loss": 0.44532470703125,
      "num_input_tokens_seen": 1275388928,
      "step": 4490,
      "train_runtime": 43567.8188,
      "train_tokens_per_second": 29273.647
    },
    {
      "epoch": 0.15921911867795058,
      "grad_norm": 0.5859375,
      "learning_rate": 4.1524171819552e-05,
      "loss": 0.4446286201477051,
      "num_input_tokens_seen": 1278251456,
      "step": 4500,
      "train_runtime": 43665.5674,
      "train_tokens_per_second": 29273.671
    },
    {
      "epoch": 0.15957293894167937,
      "grad_norm": 0.58203125,
      "learning_rate": 4.15098595550316e-05,
      "loss": 0.44756202697753905,
      "num_input_tokens_seen": 1281085120,
      "step": 4510,
      "train_runtime": 43761.8508,
      "train_tokens_per_second": 29274.016
    },
    {
      "epoch": 0.15992675920540814,
      "grad_norm": 0.58203125,
      "learning_rate": 4.1495562079474e-05,
      "loss": 0.44554719924926756,
      "num_input_tokens_seen": 1283928256,
      "step": 4520,
      "train_runtime": 43857.6516,
      "train_tokens_per_second": 29274.898
    },
    {
      "epoch": 0.16028057946913693,
      "grad_norm": 0.5546875,
      "learning_rate": 4.148127936742749e-05,
      "loss": 0.4479020118713379,
      "num_input_tokens_seen": 1286786880,
      "step": 4530,
      "train_runtime": 43956.8482,
      "train_tokens_per_second": 29273.866
    },
    {
      "epoch": 0.1606343997328657,
      "grad_norm": 0.5859375,
      "learning_rate": 4.146701139350166e-05,
      "loss": 0.4463022232055664,
      "num_input_tokens_seen": 1289581760,
      "step": 4540,
      "train_runtime": 44052.4808,
      "train_tokens_per_second": 29273.76
    },
    {
      "epoch": 0.1609882199965945,
      "grad_norm": 0.56640625,
      "learning_rate": 4.1452758132367196e-05,
      "loss": 0.4495826721191406,
      "num_input_tokens_seen": 1292423360,
      "step": 4550,
      "train_runtime": 44149.2431,
      "train_tokens_per_second": 29273.964
    },
    {
      "epoch": 0.16134204026032326,
      "grad_norm": 0.56640625,
      "learning_rate": 4.1438519558755656e-05,
      "loss": 0.44225711822509767,
      "num_input_tokens_seen": 1295337216,
      "step": 4560,
      "train_runtime": 44250.0336,
      "train_tokens_per_second": 29273.135
    },
    {
      "epoch": 0.16169586052405205,
      "grad_norm": 0.60546875,
      "learning_rate": 4.1424295647459336e-05,
      "loss": 0.43920583724975587,
      "num_input_tokens_seen": 1298182656,
      "step": 4570,
      "train_runtime": 44347.5545,
      "train_tokens_per_second": 29272.925
    },
    {
      "epoch": 0.16204968078778081,
      "grad_norm": 0.578125,
      "learning_rate": 4.141008637333106e-05,
      "loss": 0.4452220916748047,
      "num_input_tokens_seen": 1301050048,
      "step": 4580,
      "train_runtime": 44446.4972,
      "train_tokens_per_second": 29272.274
    },
    {
      "epoch": 0.1624035010515096,
      "grad_norm": 0.55078125,
      "learning_rate": 4.1395891711283974e-05,
      "loss": 0.4469286918640137,
      "num_input_tokens_seen": 1303916544,
      "step": 4590,
      "train_runtime": 44545.9198,
      "train_tokens_per_second": 29271.29
    },
    {
      "epoch": 0.16275732131523837,
      "grad_norm": 0.5859375,
      "learning_rate": 4.1381711636291395e-05,
      "loss": 0.44227113723754885,
      "num_input_tokens_seen": 1306714944,
      "step": 4600,
      "train_runtime": 44641.1509,
      "train_tokens_per_second": 29271.533
    },
    {
      "epoch": 0.16311114157896717,
      "grad_norm": 0.57421875,
      "learning_rate": 4.1367546123386604e-05,
      "loss": 0.4420786380767822,
      "num_input_tokens_seen": 1309505600,
      "step": 4610,
      "train_runtime": 44737.226,
      "train_tokens_per_second": 29271.05
    },
    {
      "epoch": 0.16346496184269593,
      "grad_norm": 0.546875,
      "learning_rate": 4.1353395147662673e-05,
      "loss": 0.4446345329284668,
      "num_input_tokens_seen": 1312348416,
      "step": 4620,
      "train_runtime": 44834.9468,
      "train_tokens_per_second": 29270.658
    },
    {
      "epoch": 0.16381878210642473,
      "grad_norm": 0.56640625,
      "learning_rate": 4.133925868427225e-05,
      "loss": 0.44669132232666015,
      "num_input_tokens_seen": 1315210752,
      "step": 4630,
      "train_runtime": 44934.1588,
      "train_tokens_per_second": 29269.731
    },
    {
      "epoch": 0.1641726023701535,
      "grad_norm": 0.59765625,
      "learning_rate": 4.132513670842744e-05,
      "loss": 0.44263653755187987,
      "num_input_tokens_seen": 1318022080,
      "step": 4640,
      "train_runtime": 45027.9453,
      "train_tokens_per_second": 29271.202
    },
    {
      "epoch": 0.16452642263388226,
      "grad_norm": 0.57421875,
      "learning_rate": 4.1311029195399534e-05,
      "loss": 0.4447168350219727,
      "num_input_tokens_seen": 1320899072,
      "step": 4650,
      "train_runtime": 45129.8031,
      "train_tokens_per_second": 29268.886
    },
    {
      "epoch": 0.16488024289761105,
      "grad_norm": 0.59375,
      "learning_rate": 4.129693612051892e-05,
      "loss": 0.44136433601379393,
      "num_input_tokens_seen": 1323750912,
      "step": 4660,
      "train_runtime": 45227.172,
      "train_tokens_per_second": 29268.93
    },
    {
      "epoch": 0.16523406316133982,
      "grad_norm": 0.59765625,
      "learning_rate": 4.1282857459174826e-05,
      "loss": 0.4422027587890625,
      "num_input_tokens_seen": 1326532416,
      "step": 4670,
      "train_runtime": 45320.6147,
      "train_tokens_per_second": 29269.956
    },
    {
      "epoch": 0.1655878834250686,
      "grad_norm": 0.6015625,
      "learning_rate": 4.1268793186815184e-05,
      "loss": 0.44470810890197754,
      "num_input_tokens_seen": 1329344384,
      "step": 4680,
      "train_runtime": 45416.7577,
      "train_tokens_per_second": 29269.909
    },
    {
      "epoch": 0.16594170368879738,
      "grad_norm": 0.58203125,
      "learning_rate": 4.1254743278946456e-05,
      "loss": 0.4466115951538086,
      "num_input_tokens_seen": 1332233664,
      "step": 4690,
      "train_runtime": 45517.0491,
      "train_tokens_per_second": 29268.894
    },
    {
      "epoch": 0.16629552395252617,
      "grad_norm": 0.56640625,
      "learning_rate": 4.1240707711133394e-05,
      "loss": 0.44083662033081056,
      "num_input_tokens_seen": 1335070720,
      "step": 4700,
      "train_runtime": 45613.8844,
      "train_tokens_per_second": 29268.955
    },
    {
      "epoch": 0.16664934421625494,
      "grad_norm": 0.56640625,
      "learning_rate": 4.122668645899893e-05,
      "loss": 0.4424298286437988,
      "num_input_tokens_seen": 1338002176,
      "step": 4710,
      "train_runtime": 45716.1136,
      "train_tokens_per_second": 29267.627
    },
    {
      "epoch": 0.16700316447998373,
      "grad_norm": 0.55859375,
      "learning_rate": 4.1212679498223975e-05,
      "loss": 0.4458757400512695,
      "num_input_tokens_seen": 1340848640,
      "step": 4720,
      "train_runtime": 45813.5276,
      "train_tokens_per_second": 29267.527
    },
    {
      "epoch": 0.1673569847437125,
      "grad_norm": 0.5625,
      "learning_rate": 4.1198686804547215e-05,
      "loss": 0.4416823387145996,
      "num_input_tokens_seen": 1343707776,
      "step": 4730,
      "train_runtime": 45910.019,
      "train_tokens_per_second": 29268.291
    },
    {
      "epoch": 0.1677108050074413,
      "grad_norm": 0.609375,
      "learning_rate": 4.118470835376499e-05,
      "loss": 0.44687299728393554,
      "num_input_tokens_seen": 1346536704,
      "step": 4740,
      "train_runtime": 46006.9392,
      "train_tokens_per_second": 29268.122
    },
    {
      "epoch": 0.16806462527117005,
      "grad_norm": 0.578125,
      "learning_rate": 4.117074412173107e-05,
      "loss": 0.4451130867004395,
      "num_input_tokens_seen": 1349381632,
      "step": 4750,
      "train_runtime": 46106.5344,
      "train_tokens_per_second": 29266.603
    },
    {
      "epoch": 0.16841844553489885,
      "grad_norm": 0.59765625,
      "learning_rate": 4.115679408435648e-05,
      "loss": 0.44522705078125,
      "num_input_tokens_seen": 1352231744,
      "step": 4760,
      "train_runtime": 46204.559,
      "train_tokens_per_second": 29266.197
    },
    {
      "epoch": 0.1687722657986276,
      "grad_norm": 0.54296875,
      "learning_rate": 4.114285821760937e-05,
      "loss": 0.4441199779510498,
      "num_input_tokens_seen": 1355066048,
      "step": 4770,
      "train_runtime": 46297.5464,
      "train_tokens_per_second": 29268.636
    },
    {
      "epoch": 0.1691260860623564,
      "grad_norm": 0.59375,
      "learning_rate": 4.11289364975148e-05,
      "loss": 0.4475994110107422,
      "num_input_tokens_seen": 1357933376,
      "step": 4780,
      "train_runtime": 46395.4123,
      "train_tokens_per_second": 29268.699
    },
    {
      "epoch": 0.16947990632608517,
      "grad_norm": 0.58203125,
      "learning_rate": 4.111502890015456e-05,
      "loss": 0.4435115337371826,
      "num_input_tokens_seen": 1360776320,
      "step": 4790,
      "train_runtime": 46488.6889,
      "train_tokens_per_second": 29271.127
    },
    {
      "epoch": 0.16983372658981397,
      "grad_norm": 0.55078125,
      "learning_rate": 4.1101135401667056e-05,
      "loss": 0.4493249893188477,
      "num_input_tokens_seen": 1363593088,
      "step": 4800,
      "train_runtime": 46585.3372,
      "train_tokens_per_second": 29270.864
    },
    {
      "epoch": 0.17018754685354273,
      "grad_norm": 0.56640625,
      "learning_rate": 4.108725597824708e-05,
      "loss": 0.4432815074920654,
      "num_input_tokens_seen": 1366410176,
      "step": 4810,
      "train_runtime": 46680.3784,
      "train_tokens_per_second": 29271.617
    },
    {
      "epoch": 0.17054136711727153,
      "grad_norm": 0.58203125,
      "learning_rate": 4.107339060614564e-05,
      "loss": 0.44433069229125977,
      "num_input_tokens_seen": 1369257856,
      "step": 4820,
      "train_runtime": 46777.5863,
      "train_tokens_per_second": 29271.665
    },
    {
      "epoch": 0.1708951873810003,
      "grad_norm": 0.6328125,
      "learning_rate": 4.1059539261669825e-05,
      "loss": 0.43774995803833006,
      "num_input_tokens_seen": 1372139584,
      "step": 4830,
      "train_runtime": 46879.2941,
      "train_tokens_per_second": 29269.63
    },
    {
      "epoch": 0.17124900764472908,
      "grad_norm": 0.58203125,
      "learning_rate": 4.104570192118262e-05,
      "loss": 0.4434209823608398,
      "num_input_tokens_seen": 1374939520,
      "step": 4840,
      "train_runtime": 46975.9637,
      "train_tokens_per_second": 29269.001
    },
    {
      "epoch": 0.17160282790845785,
      "grad_norm": 0.56640625,
      "learning_rate": 4.1031878561102714e-05,
      "loss": 0.44222288131713866,
      "num_input_tokens_seen": 1377782592,
      "step": 4850,
      "train_runtime": 47071.7365,
      "train_tokens_per_second": 29269.848
    },
    {
      "epoch": 0.17195664817218664,
      "grad_norm": 0.59765625,
      "learning_rate": 4.1018069157904385e-05,
      "loss": 0.4392059326171875,
      "num_input_tokens_seen": 1380642112,
      "step": 4860,
      "train_runtime": 47168.9594,
      "train_tokens_per_second": 29270.141
    },
    {
      "epoch": 0.1723104684359154,
      "grad_norm": 0.55859375,
      "learning_rate": 4.100427368811727e-05,
      "loss": 0.440401554107666,
      "num_input_tokens_seen": 1383498688,
      "step": 4870,
      "train_runtime": 47266.3978,
      "train_tokens_per_second": 29270.237
    },
    {
      "epoch": 0.1726642886996442,
      "grad_norm": 0.5703125,
      "learning_rate": 4.099049212832622e-05,
      "loss": 0.444883918762207,
      "num_input_tokens_seen": 1386405760,
      "step": 4880,
      "train_runtime": 47367.0689,
      "train_tokens_per_second": 29269.402
    },
    {
      "epoch": 0.17301810896337297,
      "grad_norm": 0.56640625,
      "learning_rate": 4.0976724455171155e-05,
      "loss": 0.4421547412872314,
      "num_input_tokens_seen": 1389235520,
      "step": 4890,
      "train_runtime": 47464.2533,
      "train_tokens_per_second": 29269.09
    },
    {
      "epoch": 0.17337192922710173,
      "grad_norm": 0.5859375,
      "learning_rate": 4.096297064534688e-05,
      "loss": 0.4385939598083496,
      "num_input_tokens_seen": 1392041408,
      "step": 4900,
      "train_runtime": 47560.1357,
      "train_tokens_per_second": 29269.08
    },
    {
      "epoch": 0.17372574949083053,
      "grad_norm": 0.55078125,
      "learning_rate": 4.0949230675602904e-05,
      "loss": 0.4424541473388672,
      "num_input_tokens_seen": 1394892672,
      "step": 4910,
      "train_runtime": 47656.6316,
      "train_tokens_per_second": 29269.645
    },
    {
      "epoch": 0.1740795697545593,
      "grad_norm": 0.5703125,
      "learning_rate": 4.09355045227433e-05,
      "loss": 0.4425206184387207,
      "num_input_tokens_seen": 1397753408,
      "step": 4920,
      "train_runtime": 47755.9347,
      "train_tokens_per_second": 29268.685
    },
    {
      "epoch": 0.1744333900182881,
      "grad_norm": 0.55078125,
      "learning_rate": 4.092179216362654e-05,
      "loss": 0.44448113441467285,
      "num_input_tokens_seen": 1400560384,
      "step": 4930,
      "train_runtime": 47849.8214,
      "train_tokens_per_second": 29269.919
    },
    {
      "epoch": 0.17478721028201685,
      "grad_norm": 0.6015625,
      "learning_rate": 4.090809357516532e-05,
      "loss": 0.4441685199737549,
      "num_input_tokens_seen": 1403432320,
      "step": 4940,
      "train_runtime": 47949.1995,
      "train_tokens_per_second": 29269.15
    },
    {
      "epoch": 0.17514103054574565,
      "grad_norm": 0.58203125,
      "learning_rate": 4.089440873432638e-05,
      "loss": 0.4443987846374512,
      "num_input_tokens_seen": 1406244736,
      "step": 4950,
      "train_runtime": 48041.564,
      "train_tokens_per_second": 29271.419
    },
    {
      "epoch": 0.1754948508094744,
      "grad_norm": 0.59765625,
      "learning_rate": 4.088073761813037e-05,
      "loss": 0.43969202041625977,
      "num_input_tokens_seen": 1409138176,
      "step": 4960,
      "train_runtime": 48142.9848,
      "train_tokens_per_second": 29269.855
    },
    {
      "epoch": 0.1758486710732032,
      "grad_norm": 0.57421875,
      "learning_rate": 4.086708020365172e-05,
      "loss": 0.4387206077575684,
      "num_input_tokens_seen": 1412013888,
      "step": 4970,
      "train_runtime": 48244.0273,
      "train_tokens_per_second": 29268.16
    },
    {
      "epoch": 0.17620249133693197,
      "grad_norm": 0.578125,
      "learning_rate": 4.0853436468018354e-05,
      "loss": 0.44150686264038086,
      "num_input_tokens_seen": 1414840448,
      "step": 4980,
      "train_runtime": 48340.4585,
      "train_tokens_per_second": 29268.246
    },
    {
      "epoch": 0.17655631160066076,
      "grad_norm": 0.55859375,
      "learning_rate": 4.0839806388411686e-05,
      "loss": 0.4403355598449707,
      "num_input_tokens_seen": 1417739520,
      "step": 4990,
      "train_runtime": 48439.1089,
      "train_tokens_per_second": 29268.489
    },
    {
      "epoch": 0.17691013186438953,
      "grad_norm": 0.5546875,
      "learning_rate": 4.0826189942066346e-05,
      "loss": 0.44421892166137694,
      "num_input_tokens_seen": 1420613056,
      "step": 5000,
      "train_runtime": 48537.5967,
      "train_tokens_per_second": 29268.302
    },
    {
      "epoch": 0.17691013186438953,
      "eval_loss": 0.36652371287345886,
      "eval_runtime": 8.4272,
      "eval_samples_per_second": 473.231,
      "eval_steps_per_second": 3.797,
      "num_input_tokens_seen": 1420613056,
      "step": 5000
    },
    {
      "epoch": 0.17726395212811832,
      "grad_norm": 0.5625,
      "learning_rate": 4.081258710627008e-05,
      "loss": 0.43889927864074707,
      "num_input_tokens_seen": 1423456576,
      "step": 5010,
      "train_runtime": 48665.0882,
      "train_tokens_per_second": 29250.056
    },
    {
      "epoch": 0.1776177723918471,
      "grad_norm": 0.5546875,
      "learning_rate": 4.0798997858363557e-05,
      "loss": 0.4373011589050293,
      "num_input_tokens_seen": 1426246848,
      "step": 5020,
      "train_runtime": 48756.7438,
      "train_tokens_per_second": 29252.299
    },
    {
      "epoch": 0.17797159265557588,
      "grad_norm": 0.578125,
      "learning_rate": 4.078542217574024e-05,
      "loss": 0.43915085792541503,
      "num_input_tokens_seen": 1429053568,
      "step": 5030,
      "train_runtime": 48852.5806,
      "train_tokens_per_second": 29252.366
    },
    {
      "epoch": 0.17832541291930465,
      "grad_norm": 0.56640625,
      "learning_rate": 4.0771860035846196e-05,
      "loss": 0.44334979057312013,
      "num_input_tokens_seen": 1431894464,
      "step": 5040,
      "train_runtime": 48949.4283,
      "train_tokens_per_second": 29252.527
    },
    {
      "epoch": 0.17867923318303344,
      "grad_norm": 0.58984375,
      "learning_rate": 4.0758311416179965e-05,
      "loss": 0.4382296562194824,
      "num_input_tokens_seen": 1434778752,
      "step": 5050,
      "train_runtime": 49049.0857,
      "train_tokens_per_second": 29251.896
    },
    {
      "epoch": 0.1790330534467622,
      "grad_norm": 0.57421875,
      "learning_rate": 4.0744776294292386e-05,
      "loss": 0.43564801216125487,
      "num_input_tokens_seen": 1437638720,
      "step": 5060,
      "train_runtime": 49148.0198,
      "train_tokens_per_second": 29251.203
    },
    {
      "epoch": 0.179386873710491,
      "grad_norm": 0.59375,
      "learning_rate": 4.073125464778646e-05,
      "loss": 0.44344005584716795,
      "num_input_tokens_seen": 1440541568,
      "step": 5070,
      "train_runtime": 49250.8401,
      "train_tokens_per_second": 29249.076
    },
    {
      "epoch": 0.17974069397421977,
      "grad_norm": 0.5859375,
      "learning_rate": 4.071774645431717e-05,
      "loss": 0.43868389129638674,
      "num_input_tokens_seen": 1443293440,
      "step": 5080,
      "train_runtime": 49345.8647,
      "train_tokens_per_second": 29248.519
    },
    {
      "epoch": 0.18009451423794856,
      "grad_norm": 0.57421875,
      "learning_rate": 4.070425169159135e-05,
      "loss": 0.44031376838684083,
      "num_input_tokens_seen": 1446114368,
      "step": 5090,
      "train_runtime": 49442.8729,
      "train_tokens_per_second": 29248.187
    },
    {
      "epoch": 0.18044833450167733,
      "grad_norm": 0.57421875,
      "learning_rate": 4.069077033736751e-05,
      "loss": 0.437284517288208,
      "num_input_tokens_seen": 1448938560,
      "step": 5100,
      "train_runtime": 49540.5367,
      "train_tokens_per_second": 29247.535
    },
    {
      "epoch": 0.18080215476540612,
      "grad_norm": 0.5859375,
      "learning_rate": 4.06773023694557e-05,
      "loss": 0.4442091941833496,
      "num_input_tokens_seen": 1451746496,
      "step": 5110,
      "train_runtime": 49636.0598,
      "train_tokens_per_second": 29247.819
    },
    {
      "epoch": 0.18115597502913489,
      "grad_norm": 0.58984375,
      "learning_rate": 4.066384776571732e-05,
      "loss": 0.4413691520690918,
      "num_input_tokens_seen": 1454587072,
      "step": 5120,
      "train_runtime": 49732.0646,
      "train_tokens_per_second": 29248.475
    },
    {
      "epoch": 0.18150979529286368,
      "grad_norm": 0.60546875,
      "learning_rate": 4.065040650406504e-05,
      "loss": 0.44220633506774903,
      "num_input_tokens_seen": 1457448256,
      "step": 5130,
      "train_runtime": 49831.3365,
      "train_tokens_per_second": 29247.625
    },
    {
      "epoch": 0.18186361555659245,
      "grad_norm": 0.60546875,
      "learning_rate": 4.0636978562462576e-05,
      "loss": 0.44106149673461914,
      "num_input_tokens_seen": 1460298880,
      "step": 5140,
      "train_runtime": 49929.737,
      "train_tokens_per_second": 29247.077
    },
    {
      "epoch": 0.1822174358203212,
      "grad_norm": 0.56640625,
      "learning_rate": 4.062356391892456e-05,
      "loss": 0.44039249420166016,
      "num_input_tokens_seen": 1463134592,
      "step": 5150,
      "train_runtime": 50028.8472,
      "train_tokens_per_second": 29245.819
    },
    {
      "epoch": 0.18257125608405,
      "grad_norm": 0.578125,
      "learning_rate": 4.0610162551516395e-05,
      "loss": 0.43768110275268557,
      "num_input_tokens_seen": 1466017024,
      "step": 5160,
      "train_runtime": 50129.5574,
      "train_tokens_per_second": 29244.563
    },
    {
      "epoch": 0.18292507634777877,
      "grad_norm": 0.5625,
      "learning_rate": 4.059677443835412e-05,
      "loss": 0.4380631923675537,
      "num_input_tokens_seen": 1468844032,
      "step": 5170,
      "train_runtime": 50223.7226,
      "train_tokens_per_second": 29246.021
    },
    {
      "epoch": 0.18327889661150756,
      "grad_norm": 0.578125,
      "learning_rate": 4.058339955760423e-05,
      "loss": 0.44063806533813477,
      "num_input_tokens_seen": 1471720064,
      "step": 5180,
      "train_runtime": 50325.8734,
      "train_tokens_per_second": 29243.806
    },
    {
      "epoch": 0.18363271687523633,
      "grad_norm": 0.5546875,
      "learning_rate": 4.0570037887483535e-05,
      "loss": 0.4378070831298828,
      "num_input_tokens_seen": 1474620736,
      "step": 5190,
      "train_runtime": 50426.891,
      "train_tokens_per_second": 29242.745
    },
    {
      "epoch": 0.18398653713896512,
      "grad_norm": 0.56640625,
      "learning_rate": 4.0556689406259025e-05,
      "loss": 0.4409457206726074,
      "num_input_tokens_seen": 1477467008,
      "step": 5200,
      "train_runtime": 50523.3874,
      "train_tokens_per_second": 29243.229
    },
    {
      "epoch": 0.1843403574026939,
      "grad_norm": 0.58203125,
      "learning_rate": 4.054335409224771e-05,
      "loss": 0.43840694427490234,
      "num_input_tokens_seen": 1480353088,
      "step": 5210,
      "train_runtime": 50622.3633,
      "train_tokens_per_second": 29243.066
    },
    {
      "epoch": 0.18469417766642268,
      "grad_norm": 0.55859375,
      "learning_rate": 4.053003192381646e-05,
      "loss": 0.43650293350219727,
      "num_input_tokens_seen": 1483264448,
      "step": 5220,
      "train_runtime": 50721.6647,
      "train_tokens_per_second": 29243.213
    },
    {
      "epoch": 0.18504799793015145,
      "grad_norm": 0.61328125,
      "learning_rate": 4.051672287938189e-05,
      "loss": 0.4394537925720215,
      "num_input_tokens_seen": 1486090944,
      "step": 5230,
      "train_runtime": 50818.5913,
      "train_tokens_per_second": 29243.057
    },
    {
      "epoch": 0.18540181819388024,
      "grad_norm": 0.5703125,
      "learning_rate": 4.050342693741019e-05,
      "loss": 0.43662214279174805,
      "num_input_tokens_seen": 1488925888,
      "step": 5240,
      "train_runtime": 50914.7782,
      "train_tokens_per_second": 29243.492
    },
    {
      "epoch": 0.185755638457609,
      "grad_norm": 0.57421875,
      "learning_rate": 4.049014407641699e-05,
      "loss": 0.44120097160339355,
      "num_input_tokens_seen": 1491699968,
      "step": 5250,
      "train_runtime": 51007.8139,
      "train_tokens_per_second": 29244.538
    },
    {
      "epoch": 0.1861094587213378,
      "grad_norm": 0.55859375,
      "learning_rate": 4.047687427496717e-05,
      "loss": 0.43669471740722654,
      "num_input_tokens_seen": 1494559296,
      "step": 5260,
      "train_runtime": 51107.3415,
      "train_tokens_per_second": 29243.534
    },
    {
      "epoch": 0.18646327898506657,
      "grad_norm": 0.5546875,
      "learning_rate": 4.046361751167479e-05,
      "loss": 0.43963027000427246,
      "num_input_tokens_seen": 1497424320,
      "step": 5270,
      "train_runtime": 51201.6252,
      "train_tokens_per_second": 29245.64
    },
    {
      "epoch": 0.18681709924879536,
      "grad_norm": 0.5703125,
      "learning_rate": 4.045037376520292e-05,
      "loss": 0.4410958766937256,
      "num_input_tokens_seen": 1500251456,
      "step": 5280,
      "train_runtime": 51299.4315,
      "train_tokens_per_second": 29244.992
    },
    {
      "epoch": 0.18717091951252413,
      "grad_norm": 0.58203125,
      "learning_rate": 4.043714301426344e-05,
      "loss": 0.44168972969055176,
      "num_input_tokens_seen": 1503083520,
      "step": 5290,
      "train_runtime": 51397.8517,
      "train_tokens_per_second": 29244.092
    },
    {
      "epoch": 0.18752473977625292,
      "grad_norm": 0.5546875,
      "learning_rate": 4.042392523761696e-05,
      "loss": 0.43744945526123047,
      "num_input_tokens_seen": 1505860864,
      "step": 5300,
      "train_runtime": 51492.3483,
      "train_tokens_per_second": 29244.362
    },
    {
      "epoch": 0.18787856003998168,
      "grad_norm": 0.5625,
      "learning_rate": 4.041072041407267e-05,
      "loss": 0.4394649028778076,
      "num_input_tokens_seen": 1508762624,
      "step": 5310,
      "train_runtime": 51592.3092,
      "train_tokens_per_second": 29243.944
    },
    {
      "epoch": 0.18823238030371048,
      "grad_norm": 0.59765625,
      "learning_rate": 4.039752852248815e-05,
      "loss": 0.43597002029418946,
      "num_input_tokens_seen": 1511563328,
      "step": 5320,
      "train_runtime": 51688.0104,
      "train_tokens_per_second": 29243.984
    },
    {
      "epoch": 0.18858620056743924,
      "grad_norm": 0.546875,
      "learning_rate": 4.0384349541769286e-05,
      "loss": 0.4423725128173828,
      "num_input_tokens_seen": 1514407936,
      "step": 5330,
      "train_runtime": 51785.3292,
      "train_tokens_per_second": 29243.957
    },
    {
      "epoch": 0.18894002083116804,
      "grad_norm": 0.59375,
      "learning_rate": 4.037118345087011e-05,
      "loss": 0.4436190605163574,
      "num_input_tokens_seen": 1517231040,
      "step": 5340,
      "train_runtime": 51882.7902,
      "train_tokens_per_second": 29243.436
    },
    {
      "epoch": 0.1892938410948968,
      "grad_norm": 0.578125,
      "learning_rate": 4.0358030228792636e-05,
      "loss": 0.4411949157714844,
      "num_input_tokens_seen": 1520078656,
      "step": 5350,
      "train_runtime": 51981.5454,
      "train_tokens_per_second": 29242.66
    },
    {
      "epoch": 0.1896476613586256,
      "grad_norm": 0.5859375,
      "learning_rate": 4.034488985458673e-05,
      "loss": 0.4380805492401123,
      "num_input_tokens_seen": 1522890176,
      "step": 5360,
      "train_runtime": 52074.5408,
      "train_tokens_per_second": 29244.428
    },
    {
      "epoch": 0.19000148162235436,
      "grad_norm": 0.5625,
      "learning_rate": 4.033176230735001e-05,
      "loss": 0.4371777057647705,
      "num_input_tokens_seen": 1525755264,
      "step": 5370,
      "train_runtime": 52171.7969,
      "train_tokens_per_second": 29244.829
    },
    {
      "epoch": 0.19035530188608316,
      "grad_norm": 0.5703125,
      "learning_rate": 4.0318647566227626e-05,
      "loss": 0.44295578002929686,
      "num_input_tokens_seen": 1528616832,
      "step": 5380,
      "train_runtime": 52268.8176,
      "train_tokens_per_second": 29245.292
    },
    {
      "epoch": 0.19070912214981192,
      "grad_norm": 0.59375,
      "learning_rate": 4.0305545610412205e-05,
      "loss": 0.4352247714996338,
      "num_input_tokens_seen": 1531522944,
      "step": 5390,
      "train_runtime": 52371.9608,
      "train_tokens_per_second": 29243.185
    },
    {
      "epoch": 0.19106294241354072,
      "grad_norm": 0.578125,
      "learning_rate": 4.029245641914365e-05,
      "loss": 0.44030084609985354,
      "num_input_tokens_seen": 1534331328,
      "step": 5400,
      "train_runtime": 52468.6826,
      "train_tokens_per_second": 29242.803
    },
    {
      "epoch": 0.19141676267726948,
      "grad_norm": 0.5703125,
      "learning_rate": 4.027937997170904e-05,
      "loss": 0.4353508949279785,
      "num_input_tokens_seen": 1537209664,
      "step": 5410,
      "train_runtime": 52567.6264,
      "train_tokens_per_second": 29242.516
    },
    {
      "epoch": 0.19177058294099825,
      "grad_norm": 0.55859375,
      "learning_rate": 4.026631624744247e-05,
      "loss": 0.43828487396240234,
      "num_input_tokens_seen": 1540037376,
      "step": 5420,
      "train_runtime": 52665.1925,
      "train_tokens_per_second": 29242.035
    },
    {
      "epoch": 0.19212440320472704,
      "grad_norm": 0.56640625,
      "learning_rate": 4.025326522572493e-05,
      "loss": 0.4378074645996094,
      "num_input_tokens_seen": 1542879168,
      "step": 5430,
      "train_runtime": 52762.7136,
      "train_tokens_per_second": 29241.846
    },
    {
      "epoch": 0.1924782234684558,
      "grad_norm": 0.56640625,
      "learning_rate": 4.024022688598415e-05,
      "loss": 0.4380806922912598,
      "num_input_tokens_seen": 1545750912,
      "step": 5440,
      "train_runtime": 52861.7227,
      "train_tokens_per_second": 29241.403
    },
    {
      "epoch": 0.1928320437321846,
      "grad_norm": 0.578125,
      "learning_rate": 4.0227201207694494e-05,
      "loss": 0.43492374420166013,
      "num_input_tokens_seen": 1548581376,
      "step": 5450,
      "train_runtime": 52957.6312,
      "train_tokens_per_second": 29241.893
    },
    {
      "epoch": 0.19318586399591336,
      "grad_norm": 0.58203125,
      "learning_rate": 4.021418817037677e-05,
      "loss": 0.4354818344116211,
      "num_input_tokens_seen": 1551461504,
      "step": 5460,
      "train_runtime": 53057.2016,
      "train_tokens_per_second": 29241.299
    },
    {
      "epoch": 0.19353968425964216,
      "grad_norm": 0.6015625,
      "learning_rate": 4.0201187753598174e-05,
      "loss": 0.4353951454162598,
      "num_input_tokens_seen": 1554299584,
      "step": 5470,
      "train_runtime": 53155.0779,
      "train_tokens_per_second": 29240.849
    },
    {
      "epoch": 0.19389350452337092,
      "grad_norm": 0.59765625,
      "learning_rate": 4.018819993697208e-05,
      "loss": 0.43374977111816404,
      "num_input_tokens_seen": 1557140416,
      "step": 5480,
      "train_runtime": 53249.8961,
      "train_tokens_per_second": 29242.131
    },
    {
      "epoch": 0.19424732478709972,
      "grad_norm": 0.5703125,
      "learning_rate": 4.017522470015793e-05,
      "loss": 0.43536815643310545,
      "num_input_tokens_seen": 1560041024,
      "step": 5490,
      "train_runtime": 53348.7549,
      "train_tokens_per_second": 29242.314
    },
    {
      "epoch": 0.19460114505082848,
      "grad_norm": 0.625,
      "learning_rate": 4.0162262022861144e-05,
      "loss": 0.4359905242919922,
      "num_input_tokens_seen": 1562892608,
      "step": 5500,
      "train_runtime": 53448.1659,
      "train_tokens_per_second": 29241.277
    },
    {
      "epoch": 0.19495496531455728,
      "grad_norm": 0.5625,
      "learning_rate": 4.0149311884832906e-05,
      "loss": 0.4387944221496582,
      "num_input_tokens_seen": 1565761536,
      "step": 5510,
      "train_runtime": 53546.1226,
      "train_tokens_per_second": 29241.362
    },
    {
      "epoch": 0.19530878557828604,
      "grad_norm": 0.55078125,
      "learning_rate": 4.0136374265870116e-05,
      "loss": 0.4379259586334229,
      "num_input_tokens_seen": 1568601984,
      "step": 5520,
      "train_runtime": 53641.3977,
      "train_tokens_per_second": 29242.377
    },
    {
      "epoch": 0.19566260584201484,
      "grad_norm": 0.58984375,
      "learning_rate": 4.0123449145815174e-05,
      "loss": 0.4340057849884033,
      "num_input_tokens_seen": 1571458304,
      "step": 5530,
      "train_runtime": 53735.775,
      "train_tokens_per_second": 29244.173
    },
    {
      "epoch": 0.1960164261057436,
      "grad_norm": 0.56640625,
      "learning_rate": 4.011053650455592e-05,
      "loss": 0.44149117469787597,
      "num_input_tokens_seen": 1574254144,
      "step": 5540,
      "train_runtime": 53829.3976,
      "train_tokens_per_second": 29245.249
    },
    {
      "epoch": 0.1963702463694724,
      "grad_norm": 0.5546875,
      "learning_rate": 4.0097636322025466e-05,
      "loss": 0.4321294784545898,
      "num_input_tokens_seen": 1577121152,
      "step": 5550,
      "train_runtime": 53926.6045,
      "train_tokens_per_second": 29245.697
    },
    {
      "epoch": 0.19672406663320116,
      "grad_norm": 0.58203125,
      "learning_rate": 4.008474857820206e-05,
      "loss": 0.43638410568237307,
      "num_input_tokens_seen": 1579922944,
      "step": 5560,
      "train_runtime": 54020.6927,
      "train_tokens_per_second": 29246.625
    },
    {
      "epoch": 0.19707788689692995,
      "grad_norm": 0.58203125,
      "learning_rate": 4.007187325310899e-05,
      "loss": 0.43534650802612307,
      "num_input_tokens_seen": 1582776832,
      "step": 5570,
      "train_runtime": 54117.0354,
      "train_tokens_per_second": 29247.294
    },
    {
      "epoch": 0.19743170716065872,
      "grad_norm": 0.54296875,
      "learning_rate": 4.00590103268144e-05,
      "loss": 0.4353529453277588,
      "num_input_tokens_seen": 1585641792,
      "step": 5580,
      "train_runtime": 54216.7602,
      "train_tokens_per_second": 29246.34
    },
    {
      "epoch": 0.19778552742438751,
      "grad_norm": 0.56640625,
      "learning_rate": 4.004615977943124e-05,
      "loss": 0.43984546661376955,
      "num_input_tokens_seen": 1588441088,
      "step": 5590,
      "train_runtime": 54311.1466,
      "train_tokens_per_second": 29247.055
    },
    {
      "epoch": 0.19813934768811628,
      "grad_norm": 0.58203125,
      "learning_rate": 4.0033321591117025e-05,
      "loss": 0.4320873260498047,
      "num_input_tokens_seen": 1591290368,
      "step": 5600,
      "train_runtime": 54407.6745,
      "train_tokens_per_second": 29247.535
    },
    {
      "epoch": 0.19849316795184507,
      "grad_norm": 0.5859375,
      "learning_rate": 4.002049574207381e-05,
      "loss": 0.4366870880126953,
      "num_input_tokens_seen": 1594144128,
      "step": 5610,
      "train_runtime": 54505.3077,
      "train_tokens_per_second": 29247.503
    },
    {
      "epoch": 0.19884698821557384,
      "grad_norm": 0.56640625,
      "learning_rate": 4.000768221254803e-05,
      "loss": 0.43494539260864257,
      "num_input_tokens_seen": 1597000064,
      "step": 5620,
      "train_runtime": 54601.603,
      "train_tokens_per_second": 29248.227
    },
    {
      "epoch": 0.19920080847930263,
      "grad_norm": 0.57421875,
      "learning_rate": 3.999488098283034e-05,
      "loss": 0.43576898574829104,
      "num_input_tokens_seen": 1599860288,
      "step": 5630,
      "train_runtime": 54702.1675,
      "train_tokens_per_second": 29246.744
    },
    {
      "epoch": 0.1995546287430314,
      "grad_norm": 0.55859375,
      "learning_rate": 3.9982092033255506e-05,
      "loss": 0.4365368366241455,
      "num_input_tokens_seen": 1602709120,
      "step": 5640,
      "train_runtime": 54799.7285,
      "train_tokens_per_second": 29246.662
    },
    {
      "epoch": 0.1999084490067602,
      "grad_norm": 0.55078125,
      "learning_rate": 3.996931534420232e-05,
      "loss": 0.43445682525634766,
      "num_input_tokens_seen": 1605536384,
      "step": 5650,
      "train_runtime": 54896.1811,
      "train_tokens_per_second": 29246.777
    },
    {
      "epoch": 0.20026226927048896,
      "grad_norm": 0.56640625,
      "learning_rate": 3.995655089609339e-05,
      "loss": 0.43600945472717284,
      "num_input_tokens_seen": 1608395072,
      "step": 5660,
      "train_runtime": 54992.376,
      "train_tokens_per_second": 29247.601
    },
    {
      "epoch": 0.20061608953421772,
      "grad_norm": 0.59765625,
      "learning_rate": 3.994379866939511e-05,
      "loss": 0.4321427345275879,
      "num_input_tokens_seen": 1611246528,
      "step": 5670,
      "train_runtime": 55090.1326,
      "train_tokens_per_second": 29247.461
    },
    {
      "epoch": 0.20096990979794652,
      "grad_norm": 0.58203125,
      "learning_rate": 3.993105864461745e-05,
      "loss": 0.43267669677734377,
      "num_input_tokens_seen": 1614097856,
      "step": 5680,
      "train_runtime": 55188.8137,
      "train_tokens_per_second": 29246.83
    },
    {
      "epoch": 0.20132373006167528,
      "grad_norm": 0.5703125,
      "learning_rate": 3.9918330802313866e-05,
      "loss": 0.4402439117431641,
      "num_input_tokens_seen": 1616895488,
      "step": 5690,
      "train_runtime": 55286.8025,
      "train_tokens_per_second": 29245.596
    },
    {
      "epoch": 0.20167755032540408,
      "grad_norm": 0.5625,
      "learning_rate": 3.9905615123081206e-05,
      "loss": 0.4373167991638184,
      "num_input_tokens_seen": 1619738624,
      "step": 5700,
      "train_runtime": 55383.7084,
      "train_tokens_per_second": 29245.76
    },
    {
      "epoch": 0.20203137058913284,
      "grad_norm": 0.57421875,
      "learning_rate": 3.989291158755953e-05,
      "loss": 0.43944807052612306,
      "num_input_tokens_seen": 1622599168,
      "step": 5710,
      "train_runtime": 55483.6791,
      "train_tokens_per_second": 29244.621
    },
    {
      "epoch": 0.20238519085286164,
      "grad_norm": 0.58984375,
      "learning_rate": 3.988022017643201e-05,
      "loss": 0.4356792449951172,
      "num_input_tokens_seen": 1625486464,
      "step": 5720,
      "train_runtime": 55583.5851,
      "train_tokens_per_second": 29244.002
    },
    {
      "epoch": 0.2027390111165904,
      "grad_norm": 0.578125,
      "learning_rate": 3.9867540870424826e-05,
      "loss": 0.4399091720581055,
      "num_input_tokens_seen": 1628304704,
      "step": 5730,
      "train_runtime": 55679.8726,
      "train_tokens_per_second": 29244.045
    },
    {
      "epoch": 0.2030928313803192,
      "grad_norm": 0.59765625,
      "learning_rate": 3.985487365030702e-05,
      "loss": 0.43842029571533203,
      "num_input_tokens_seen": 1631111616,
      "step": 5740,
      "train_runtime": 55774.3747,
      "train_tokens_per_second": 29244.821
    },
    {
      "epoch": 0.20344665164404796,
      "grad_norm": 0.55078125,
      "learning_rate": 3.984221849689036e-05,
      "loss": 0.43639116287231444,
      "num_input_tokens_seen": 1633982144,
      "step": 5750,
      "train_runtime": 55873.2106,
      "train_tokens_per_second": 29244.465
    },
    {
      "epoch": 0.20380047190777675,
      "grad_norm": 0.59765625,
      "learning_rate": 3.982957539102927e-05,
      "loss": 0.43750653266906736,
      "num_input_tokens_seen": 1636864512,
      "step": 5760,
      "train_runtime": 55970.1344,
      "train_tokens_per_second": 29245.32
    },
    {
      "epoch": 0.20415429217150552,
      "grad_norm": 0.57421875,
      "learning_rate": 3.981694431362065e-05,
      "loss": 0.43770227432250974,
      "num_input_tokens_seen": 1639718144,
      "step": 5770,
      "train_runtime": 56068.2638,
      "train_tokens_per_second": 29245.032
    },
    {
      "epoch": 0.2045081124352343,
      "grad_norm": 0.6015625,
      "learning_rate": 3.9804325245603786e-05,
      "loss": 0.44153947830200196,
      "num_input_tokens_seen": 1642537728,
      "step": 5780,
      "train_runtime": 56168.3941,
      "train_tokens_per_second": 29243.096
    },
    {
      "epoch": 0.20486193269896308,
      "grad_norm": 0.57421875,
      "learning_rate": 3.9791718167960226e-05,
      "loss": 0.4355274200439453,
      "num_input_tokens_seen": 1645398208,
      "step": 5790,
      "train_runtime": 56266.1962,
      "train_tokens_per_second": 29243.104
    },
    {
      "epoch": 0.20521575296269187,
      "grad_norm": 0.5625,
      "learning_rate": 3.9779123061713665e-05,
      "loss": 0.433730936050415,
      "num_input_tokens_seen": 1648256576,
      "step": 5800,
      "train_runtime": 56362.4708,
      "train_tokens_per_second": 29243.867
    },
    {
      "epoch": 0.20556957322642064,
      "grad_norm": 0.55859375,
      "learning_rate": 3.976653990792979e-05,
      "loss": 0.43136072158813477,
      "num_input_tokens_seen": 1651121344,
      "step": 5810,
      "train_runtime": 56462.3938,
      "train_tokens_per_second": 29242.851
    },
    {
      "epoch": 0.20592339349014943,
      "grad_norm": 0.5625,
      "learning_rate": 3.9753968687716206e-05,
      "loss": 0.4295208930969238,
      "num_input_tokens_seen": 1654019904,
      "step": 5820,
      "train_runtime": 56563.7663,
      "train_tokens_per_second": 29241.686
    },
    {
      "epoch": 0.2062772137538782,
      "grad_norm": 0.55078125,
      "learning_rate": 3.974140938222232e-05,
      "loss": 0.4371951580047607,
      "num_input_tokens_seen": 1656882560,
      "step": 5830,
      "train_runtime": 56660.9753,
      "train_tokens_per_second": 29242.041
    },
    {
      "epoch": 0.206631034017607,
      "grad_norm": 0.5859375,
      "learning_rate": 3.972886197263915e-05,
      "loss": 0.4320386409759521,
      "num_input_tokens_seen": 1659705024,
      "step": 5840,
      "train_runtime": 56757.8905,
      "train_tokens_per_second": 29241.838
    },
    {
      "epoch": 0.20698485428133576,
      "grad_norm": 0.578125,
      "learning_rate": 3.97163264401993e-05,
      "loss": 0.43626089096069337,
      "num_input_tokens_seen": 1662524416,
      "step": 5850,
      "train_runtime": 56855.4501,
      "train_tokens_per_second": 29241.25
    },
    {
      "epoch": 0.20733867454506455,
      "grad_norm": 0.578125,
      "learning_rate": 3.970380276617677e-05,
      "loss": 0.43509936332702637,
      "num_input_tokens_seen": 1665339712,
      "step": 5860,
      "train_runtime": 56949.3076,
      "train_tokens_per_second": 29242.493
    },
    {
      "epoch": 0.20769249480879332,
      "grad_norm": 0.57421875,
      "learning_rate": 3.96912909318869e-05,
      "loss": 0.4371882438659668,
      "num_input_tokens_seen": 1668169472,
      "step": 5870,
      "train_runtime": 57045.2908,
      "train_tokens_per_second": 29242.895
    },
    {
      "epoch": 0.2080463150725221,
      "grad_norm": 0.5859375,
      "learning_rate": 3.96787909186862e-05,
      "loss": 0.43482317924499514,
      "num_input_tokens_seen": 1671030400,
      "step": 5880,
      "train_runtime": 57144.5523,
      "train_tokens_per_second": 29242.165
    },
    {
      "epoch": 0.20840013533625087,
      "grad_norm": 0.5703125,
      "learning_rate": 3.9666302707972244e-05,
      "loss": 0.43056726455688477,
      "num_input_tokens_seen": 1673912576,
      "step": 5890,
      "train_runtime": 57244.6049,
      "train_tokens_per_second": 29241.403
    },
    {
      "epoch": 0.20875395559997967,
      "grad_norm": 0.5546875,
      "learning_rate": 3.965382628118358e-05,
      "loss": 0.43903384208679197,
      "num_input_tokens_seen": 1676753344,
      "step": 5900,
      "train_runtime": 57343.7692,
      "train_tokens_per_second": 29240.376
    },
    {
      "epoch": 0.20910777586370843,
      "grad_norm": 0.578125,
      "learning_rate": 3.964136161979959e-05,
      "loss": 0.4356219291687012,
      "num_input_tokens_seen": 1679636032,
      "step": 5910,
      "train_runtime": 57444.8029,
      "train_tokens_per_second": 29239.13
    },
    {
      "epoch": 0.20946159612743723,
      "grad_norm": 0.56640625,
      "learning_rate": 3.9628908705340406e-05,
      "loss": 0.42699484825134276,
      "num_input_tokens_seen": 1682448960,
      "step": 5920,
      "train_runtime": 57538.5227,
      "train_tokens_per_second": 29240.392
    },
    {
      "epoch": 0.209815416391166,
      "grad_norm": 0.578125,
      "learning_rate": 3.961646751936673e-05,
      "loss": 0.4383396625518799,
      "num_input_tokens_seen": 1685275328,
      "step": 5930,
      "train_runtime": 57636.5192,
      "train_tokens_per_second": 29239.714
    },
    {
      "epoch": 0.21016923665489476,
      "grad_norm": 0.5703125,
      "learning_rate": 3.960403804347979e-05,
      "loss": 0.43402700424194335,
      "num_input_tokens_seen": 1688108032,
      "step": 5940,
      "train_runtime": 57736.8732,
      "train_tokens_per_second": 29237.954
    },
    {
      "epoch": 0.21052305691862355,
      "grad_norm": 0.5703125,
      "learning_rate": 3.959162025932119e-05,
      "loss": 0.435057258605957,
      "num_input_tokens_seen": 1691020608,
      "step": 5950,
      "train_runtime": 57838.9724,
      "train_tokens_per_second": 29236.699
    },
    {
      "epoch": 0.21087687718235232,
      "grad_norm": 0.59375,
      "learning_rate": 3.95792141485728e-05,
      "loss": 0.43424162864685056,
      "num_input_tokens_seen": 1693874496,
      "step": 5960,
      "train_runtime": 57936.219,
      "train_tokens_per_second": 29236.884
    },
    {
      "epoch": 0.2112306974460811,
      "grad_norm": 0.55859375,
      "learning_rate": 3.956681969295664e-05,
      "loss": 0.4301346778869629,
      "num_input_tokens_seen": 1696687552,
      "step": 5970,
      "train_runtime": 58032.3767,
      "train_tokens_per_second": 29236.913
    },
    {
      "epoch": 0.21158451770980988,
      "grad_norm": 0.5625,
      "learning_rate": 3.955443687423479e-05,
      "loss": 0.43507919311523435,
      "num_input_tokens_seen": 1699554688,
      "step": 5980,
      "train_runtime": 58128.4636,
      "train_tokens_per_second": 29237.908
    },
    {
      "epoch": 0.21193833797353867,
      "grad_norm": 0.5625,
      "learning_rate": 3.954206567420924e-05,
      "loss": 0.4325693130493164,
      "num_input_tokens_seen": 1702368448,
      "step": 5990,
      "train_runtime": 58224.0348,
      "train_tokens_per_second": 29238.243
    },
    {
      "epoch": 0.21229215823726744,
      "grad_norm": 0.56640625,
      "learning_rate": 3.952970607472179e-05,
      "loss": 0.4390561103820801,
      "num_input_tokens_seen": 1705224320,
      "step": 6000,
      "train_runtime": 58321.2372,
      "train_tokens_per_second": 29238.48
    },
    {
      "epoch": 0.21264597850099623,
      "grad_norm": 0.59375,
      "learning_rate": 3.951735805765399e-05,
      "loss": 0.4354909896850586,
      "num_input_tokens_seen": 1708082560,
      "step": 6010,
      "train_runtime": 58421.3306,
      "train_tokens_per_second": 29237.31
    },
    {
      "epoch": 0.212999798764725,
      "grad_norm": 0.54296875,
      "learning_rate": 3.950502160492692e-05,
      "loss": 0.43860807418823244,
      "num_input_tokens_seen": 1710936896,
      "step": 6020,
      "train_runtime": 58520.0155,
      "train_tokens_per_second": 29236.781
    },
    {
      "epoch": 0.2133536190284538,
      "grad_norm": 0.58984375,
      "learning_rate": 3.9492696698501205e-05,
      "loss": 0.4326430320739746,
      "num_input_tokens_seen": 1713764544,
      "step": 6030,
      "train_runtime": 58617.7727,
      "train_tokens_per_second": 29236.262
    },
    {
      "epoch": 0.21370743929218255,
      "grad_norm": 0.54296875,
      "learning_rate": 3.9480383320376784e-05,
      "loss": 0.43421192169189454,
      "num_input_tokens_seen": 1716610688,
      "step": 6040,
      "train_runtime": 58712.4979,
      "train_tokens_per_second": 29237.569
    },
    {
      "epoch": 0.21406125955591135,
      "grad_norm": 0.57421875,
      "learning_rate": 3.94680814525929e-05,
      "loss": 0.4363199234008789,
      "num_input_tokens_seen": 1719404096,
      "step": 6050,
      "train_runtime": 58807.8177,
      "train_tokens_per_second": 29237.679
    },
    {
      "epoch": 0.21441507981964011,
      "grad_norm": 0.57421875,
      "learning_rate": 3.945579107722792e-05,
      "loss": 0.433974027633667,
      "num_input_tokens_seen": 1722238400,
      "step": 6060,
      "train_runtime": 58905.901,
      "train_tokens_per_second": 29237.112
    },
    {
      "epoch": 0.2147689000833689,
      "grad_norm": 0.60546875,
      "learning_rate": 3.9443512176399276e-05,
      "loss": 0.4350698947906494,
      "num_input_tokens_seen": 1725054016,
      "step": 6070,
      "train_runtime": 59005.6392,
      "train_tokens_per_second": 29235.409
    },
    {
      "epoch": 0.21512272034709767,
      "grad_norm": 0.5703125,
      "learning_rate": 3.9431244732263307e-05,
      "loss": 0.433062744140625,
      "num_input_tokens_seen": 1727897216,
      "step": 6080,
      "train_runtime": 59102.3186,
      "train_tokens_per_second": 29235.693
    },
    {
      "epoch": 0.21547654061082647,
      "grad_norm": 0.59375,
      "learning_rate": 3.941898872701519e-05,
      "loss": 0.43312692642211914,
      "num_input_tokens_seen": 1730726784,
      "step": 6090,
      "train_runtime": 59200.119,
      "train_tokens_per_second": 29235.191
    },
    {
      "epoch": 0.21583036087455523,
      "grad_norm": 0.578125,
      "learning_rate": 3.940674414288882e-05,
      "loss": 0.43694748878479006,
      "num_input_tokens_seen": 1733550144,
      "step": 6100,
      "train_runtime": 59294.8004,
      "train_tokens_per_second": 29236.124
    },
    {
      "epoch": 0.21618418113828403,
      "grad_norm": 0.55859375,
      "learning_rate": 3.939451096215668e-05,
      "loss": 0.43339242935180666,
      "num_input_tokens_seen": 1736353792,
      "step": 6110,
      "train_runtime": 59386.953,
      "train_tokens_per_second": 29237.967
    },
    {
      "epoch": 0.2165380014020128,
      "grad_norm": 0.5703125,
      "learning_rate": 3.938228916712978e-05,
      "loss": 0.43361706733703614,
      "num_input_tokens_seen": 1739209408,
      "step": 6120,
      "train_runtime": 59484.3381,
      "train_tokens_per_second": 29238.106
    },
    {
      "epoch": 0.21689182166574159,
      "grad_norm": 0.57421875,
      "learning_rate": 3.937007874015748e-05,
      "loss": 0.4330623626708984,
      "num_input_tokens_seen": 1742103104,
      "step": 6130,
      "train_runtime": 59584.1619,
      "train_tokens_per_second": 29237.687
    },
    {
      "epoch": 0.21724564192947035,
      "grad_norm": 0.546875,
      "learning_rate": 3.935787966362748e-05,
      "loss": 0.4316445827484131,
      "num_input_tokens_seen": 1744954176,
      "step": 6140,
      "train_runtime": 59682.0509,
      "train_tokens_per_second": 29237.504
    },
    {
      "epoch": 0.21759946219319914,
      "grad_norm": 0.578125,
      "learning_rate": 3.9345691919965595e-05,
      "loss": 0.4371340751647949,
      "num_input_tokens_seen": 1747763136,
      "step": 6150,
      "train_runtime": 59775.9221,
      "train_tokens_per_second": 29238.581
    },
    {
      "epoch": 0.2179532824569279,
      "grad_norm": 0.5625,
      "learning_rate": 3.9333515491635764e-05,
      "loss": 0.42518091201782227,
      "num_input_tokens_seen": 1750596096,
      "step": 6160,
      "train_runtime": 59872.1007,
      "train_tokens_per_second": 29238.929
    },
    {
      "epoch": 0.2183071027206567,
      "grad_norm": 0.6015625,
      "learning_rate": 3.932135036113987e-05,
      "loss": 0.4350085735321045,
      "num_input_tokens_seen": 1753441664,
      "step": 6170,
      "train_runtime": 59969.5217,
      "train_tokens_per_second": 29238.88
    },
    {
      "epoch": 0.21866092298438547,
      "grad_norm": 0.5703125,
      "learning_rate": 3.930919651101764e-05,
      "loss": 0.4320687770843506,
      "num_input_tokens_seen": 1756270976,
      "step": 6180,
      "train_runtime": 60063.3769,
      "train_tokens_per_second": 29240.297
    },
    {
      "epoch": 0.21901474324811424,
      "grad_norm": 0.5625,
      "learning_rate": 3.9297053923846576e-05,
      "loss": 0.42859630584716796,
      "num_input_tokens_seen": 1759120064,
      "step": 6190,
      "train_runtime": 60161.1552,
      "train_tokens_per_second": 29240.131
    },
    {
      "epoch": 0.21936856351184303,
      "grad_norm": 0.578125,
      "learning_rate": 3.928492258224183e-05,
      "loss": 0.4324918270111084,
      "num_input_tokens_seen": 1761969408,
      "step": 6200,
      "train_runtime": 60261.3726,
      "train_tokens_per_second": 29238.787
    },
    {
      "epoch": 0.2197223837755718,
      "grad_norm": 0.578125,
      "learning_rate": 3.927280246885609e-05,
      "loss": 0.4357311248779297,
      "num_input_tokens_seen": 1764794368,
      "step": 6210,
      "train_runtime": 60355.7351,
      "train_tokens_per_second": 29239.879
    },
    {
      "epoch": 0.2200762040393006,
      "grad_norm": 0.578125,
      "learning_rate": 3.9260693566379486e-05,
      "loss": 0.4329801559448242,
      "num_input_tokens_seen": 1767617984,
      "step": 6220,
      "train_runtime": 60450.1228,
      "train_tokens_per_second": 29240.933
    },
    {
      "epoch": 0.22043002430302935,
      "grad_norm": 0.5703125,
      "learning_rate": 3.924859585753948e-05,
      "loss": 0.43080663681030273,
      "num_input_tokens_seen": 1770441920,
      "step": 6230,
      "train_runtime": 60549.0,
      "train_tokens_per_second": 29239.821
    },
    {
      "epoch": 0.22078384456675815,
      "grad_norm": 0.57421875,
      "learning_rate": 3.923650932510079e-05,
      "loss": 0.42781476974487304,
      "num_input_tokens_seen": 1773221248,
      "step": 6240,
      "train_runtime": 60644.4675,
      "train_tokens_per_second": 29239.621
    },
    {
      "epoch": 0.2211376648304869,
      "grad_norm": 0.55859375,
      "learning_rate": 3.9224433951865215e-05,
      "loss": 0.43264055252075195,
      "num_input_tokens_seen": 1776083840,
      "step": 6250,
      "train_runtime": 60744.0645,
      "train_tokens_per_second": 29238.805
    },
    {
      "epoch": 0.2214914850942157,
      "grad_norm": 0.6015625,
      "learning_rate": 3.921236972067165e-05,
      "loss": 0.4291335105895996,
      "num_input_tokens_seen": 1778896128,
      "step": 6260,
      "train_runtime": 60840.254,
      "train_tokens_per_second": 29238.802
    },
    {
      "epoch": 0.22184530535794447,
      "grad_norm": 0.59765625,
      "learning_rate": 3.920031661439585e-05,
      "loss": 0.43102331161499025,
      "num_input_tokens_seen": 1781776256,
      "step": 6270,
      "train_runtime": 60940.0491,
      "train_tokens_per_second": 29238.182
    },
    {
      "epoch": 0.22219912562167327,
      "grad_norm": 0.5546875,
      "learning_rate": 3.918827461595045e-05,
      "loss": 0.43372297286987305,
      "num_input_tokens_seen": 1784618432,
      "step": 6280,
      "train_runtime": 61037.3337,
      "train_tokens_per_second": 29238.145
    },
    {
      "epoch": 0.22255294588540203,
      "grad_norm": 0.5703125,
      "learning_rate": 3.9176243708284746e-05,
      "loss": 0.430194616317749,
      "num_input_tokens_seen": 1787409536,
      "step": 6290,
      "train_runtime": 61130.3131,
      "train_tokens_per_second": 29239.332
    },
    {
      "epoch": 0.22290676614913082,
      "grad_norm": 0.5703125,
      "learning_rate": 3.9164223874384715e-05,
      "loss": 0.4324221134185791,
      "num_input_tokens_seen": 1790285248,
      "step": 6300,
      "train_runtime": 61229.6622,
      "train_tokens_per_second": 29238.856
    },
    {
      "epoch": 0.2232605864128596,
      "grad_norm": 0.57421875,
      "learning_rate": 3.91522150972728e-05,
      "loss": 0.4361156463623047,
      "num_input_tokens_seen": 1793080832,
      "step": 6310,
      "train_runtime": 61323.7765,
      "train_tokens_per_second": 29239.57
    },
    {
      "epoch": 0.22361440667658838,
      "grad_norm": 0.55859375,
      "learning_rate": 3.9140217360007896e-05,
      "loss": 0.4314689636230469,
      "num_input_tokens_seen": 1795923200,
      "step": 6320,
      "train_runtime": 61422.0999,
      "train_tokens_per_second": 29239.039
    },
    {
      "epoch": 0.22396822694031715,
      "grad_norm": 0.56640625,
      "learning_rate": 3.912823064568521e-05,
      "loss": 0.43456201553344725,
      "num_input_tokens_seen": 1798770432,
      "step": 6330,
      "train_runtime": 61518.3987,
      "train_tokens_per_second": 29239.552
    },
    {
      "epoch": 0.22432204720404594,
      "grad_norm": 0.5625,
      "learning_rate": 3.9116254937436155e-05,
      "loss": 0.43364620208740234,
      "num_input_tokens_seen": 1801678912,
      "step": 6340,
      "train_runtime": 61621.3967,
      "train_tokens_per_second": 29237.879
    },
    {
      "epoch": 0.2246758674677747,
      "grad_norm": 0.55859375,
      "learning_rate": 3.910429021842825e-05,
      "loss": 0.4326483726501465,
      "num_input_tokens_seen": 1804571712,
      "step": 6350,
      "train_runtime": 61723.6682,
      "train_tokens_per_second": 29236.301
    },
    {
      "epoch": 0.2250296877315035,
      "grad_norm": 0.57421875,
      "learning_rate": 3.9092336471865084e-05,
      "loss": 0.4279055595397949,
      "num_input_tokens_seen": 1807458880,
      "step": 6360,
      "train_runtime": 61822.638,
      "train_tokens_per_second": 29236.198
    },
    {
      "epoch": 0.22538350799523227,
      "grad_norm": 0.60546875,
      "learning_rate": 3.908039368098611e-05,
      "loss": 0.42912998199462893,
      "num_input_tokens_seen": 1810334720,
      "step": 6370,
      "train_runtime": 61922.865,
      "train_tokens_per_second": 29235.319
    },
    {
      "epoch": 0.22573732825896106,
      "grad_norm": 0.58203125,
      "learning_rate": 3.9068461829066633e-05,
      "loss": 0.42847251892089844,
      "num_input_tokens_seen": 1813165248,
      "step": 6380,
      "train_runtime": 62021.7351,
      "train_tokens_per_second": 29234.352
    },
    {
      "epoch": 0.22609114852268983,
      "grad_norm": 0.58984375,
      "learning_rate": 3.9056540899417656e-05,
      "loss": 0.43103508949279784,
      "num_input_tokens_seen": 1816047488,
      "step": 6390,
      "train_runtime": 62120.0229,
      "train_tokens_per_second": 29234.495
    },
    {
      "epoch": 0.22644496878641862,
      "grad_norm": 0.57421875,
      "learning_rate": 3.904463087538585e-05,
      "loss": 0.43492941856384276,
      "num_input_tokens_seen": 1818968000,
      "step": 6400,
      "train_runtime": 62221.6787,
      "train_tokens_per_second": 29233.67
    },
    {
      "epoch": 0.2267987890501474,
      "grad_norm": 0.5703125,
      "learning_rate": 3.903273174035336e-05,
      "loss": 0.4328557014465332,
      "num_input_tokens_seen": 1821808384,
      "step": 6410,
      "train_runtime": 62317.2662,
      "train_tokens_per_second": 29234.408
    },
    {
      "epoch": 0.22715260931387618,
      "grad_norm": 0.5546875,
      "learning_rate": 3.902084347773779e-05,
      "loss": 0.4325701236724854,
      "num_input_tokens_seen": 1824615104,
      "step": 6420,
      "train_runtime": 62407.9526,
      "train_tokens_per_second": 29236.901
    },
    {
      "epoch": 0.22750642957760495,
      "grad_norm": 0.56640625,
      "learning_rate": 3.900896607099207e-05,
      "loss": 0.4290501594543457,
      "num_input_tokens_seen": 1827493440,
      "step": 6430,
      "train_runtime": 62505.29,
      "train_tokens_per_second": 29237.42
    },
    {
      "epoch": 0.2278602498413337,
      "grad_norm": 0.58203125,
      "learning_rate": 3.899709950360437e-05,
      "loss": 0.43046369552612307,
      "num_input_tokens_seen": 1830364224,
      "step": 6440,
      "train_runtime": 62609.1841,
      "train_tokens_per_second": 29234.756
    },
    {
      "epoch": 0.2282140701050625,
      "grad_norm": 0.55859375,
      "learning_rate": 3.8985243759097997e-05,
      "loss": 0.4359885215759277,
      "num_input_tokens_seen": 1833173248,
      "step": 6450,
      "train_runtime": 62706.9275,
      "train_tokens_per_second": 29233.983
    },
    {
      "epoch": 0.22856789036879127,
      "grad_norm": 0.5625,
      "learning_rate": 3.897339882103129e-05,
      "loss": 0.4279576301574707,
      "num_input_tokens_seen": 1836004224,
      "step": 6460,
      "train_runtime": 62803.2408,
      "train_tokens_per_second": 29234.227
    },
    {
      "epoch": 0.22892171063252006,
      "grad_norm": 0.57421875,
      "learning_rate": 3.8961564672997544e-05,
      "loss": 0.4297611713409424,
      "num_input_tokens_seen": 1838892288,
      "step": 6470,
      "train_runtime": 62902.6571,
      "train_tokens_per_second": 29233.937
    },
    {
      "epoch": 0.22927553089624883,
      "grad_norm": 0.5859375,
      "learning_rate": 3.8949741298624924e-05,
      "loss": 0.43323631286621095,
      "num_input_tokens_seen": 1841783296,
      "step": 6480,
      "train_runtime": 63002.8296,
      "train_tokens_per_second": 29233.342
    },
    {
      "epoch": 0.22962935115997762,
      "grad_norm": 0.55859375,
      "learning_rate": 3.8937928681576305e-05,
      "loss": 0.4244824409484863,
      "num_input_tokens_seen": 1844600320,
      "step": 6490,
      "train_runtime": 63100.2299,
      "train_tokens_per_second": 29232.862
    },
    {
      "epoch": 0.2299831714237064,
      "grad_norm": 0.578125,
      "learning_rate": 3.8926126805549276e-05,
      "loss": 0.42909870147705076,
      "num_input_tokens_seen": 1847412800,
      "step": 6500,
      "train_runtime": 63196.8007,
      "train_tokens_per_second": 29232.695
    },
    {
      "epoch": 0.23033699168743518,
      "grad_norm": 0.57421875,
      "learning_rate": 3.891433565427596e-05,
      "loss": 0.42655296325683595,
      "num_input_tokens_seen": 1850266368,
      "step": 6510,
      "train_runtime": 63293.0039,
      "train_tokens_per_second": 29233.347
    },
    {
      "epoch": 0.23069081195116395,
      "grad_norm": 0.58203125,
      "learning_rate": 3.8902555211522964e-05,
      "loss": 0.4330574035644531,
      "num_input_tokens_seen": 1853080064,
      "step": 6520,
      "train_runtime": 63390.1764,
      "train_tokens_per_second": 29232.922
    },
    {
      "epoch": 0.23104463221489274,
      "grad_norm": 0.58984375,
      "learning_rate": 3.889078546109127e-05,
      "loss": 0.42397561073303225,
      "num_input_tokens_seen": 1855954496,
      "step": 6530,
      "train_runtime": 63488.2576,
      "train_tokens_per_second": 29233.036
    },
    {
      "epoch": 0.2313984524786215,
      "grad_norm": 0.5703125,
      "learning_rate": 3.887902638681616e-05,
      "loss": 0.434549617767334,
      "num_input_tokens_seen": 1858839360,
      "step": 6540,
      "train_runtime": 63590.8332,
      "train_tokens_per_second": 29231.247
    },
    {
      "epoch": 0.2317522727423503,
      "grad_norm": 0.5546875,
      "learning_rate": 3.886727797256707e-05,
      "loss": 0.4277657032012939,
      "num_input_tokens_seen": 1861636608,
      "step": 6550,
      "train_runtime": 63684.0417,
      "train_tokens_per_second": 29232.388
    },
    {
      "epoch": 0.23210609300607907,
      "grad_norm": 0.58203125,
      "learning_rate": 3.88555402022476e-05,
      "loss": 0.4328862190246582,
      "num_input_tokens_seen": 1864463936,
      "step": 6560,
      "train_runtime": 63779.9879,
      "train_tokens_per_second": 29232.742
    },
    {
      "epoch": 0.23245991326980786,
      "grad_norm": 0.59765625,
      "learning_rate": 3.884381305979528e-05,
      "loss": 0.43508000373840333,
      "num_input_tokens_seen": 1867305792,
      "step": 6570,
      "train_runtime": 63875.2377,
      "train_tokens_per_second": 29233.641
    },
    {
      "epoch": 0.23281373353353663,
      "grad_norm": 0.54296875,
      "learning_rate": 3.883209652918163e-05,
      "loss": 0.4378073692321777,
      "num_input_tokens_seen": 1870197312,
      "step": 6580,
      "train_runtime": 63976.6842,
      "train_tokens_per_second": 29232.483
    },
    {
      "epoch": 0.23316755379726542,
      "grad_norm": 0.546875,
      "learning_rate": 3.8820390594411935e-05,
      "loss": 0.43173627853393554,
      "num_input_tokens_seen": 1873037888,
      "step": 6590,
      "train_runtime": 64072.3747,
      "train_tokens_per_second": 29233.159
    },
    {
      "epoch": 0.23352137406099419,
      "grad_norm": 0.578125,
      "learning_rate": 3.880869523952524e-05,
      "loss": 0.43375453948974607,
      "num_input_tokens_seen": 1875862336,
      "step": 6600,
      "train_runtime": 64167.7465,
      "train_tokens_per_second": 29233.726
    },
    {
      "epoch": 0.23387519432472298,
      "grad_norm": 0.5625,
      "learning_rate": 3.879701044859422e-05,
      "loss": 0.4325495719909668,
      "num_input_tokens_seen": 1878708736,
      "step": 6610,
      "train_runtime": 64266.6201,
      "train_tokens_per_second": 29233.041
    },
    {
      "epoch": 0.23422901458845174,
      "grad_norm": 0.5859375,
      "learning_rate": 3.87853362057251e-05,
      "loss": 0.4317276954650879,
      "num_input_tokens_seen": 1881544192,
      "step": 6620,
      "train_runtime": 64363.588,
      "train_tokens_per_second": 29233.053
    },
    {
      "epoch": 0.23458283485218054,
      "grad_norm": 0.578125,
      "learning_rate": 3.8773672495057576e-05,
      "loss": 0.4251361846923828,
      "num_input_tokens_seen": 1884380800,
      "step": 6630,
      "train_runtime": 64461.4269,
      "train_tokens_per_second": 29232.688
    },
    {
      "epoch": 0.2349366551159093,
      "grad_norm": 0.578125,
      "learning_rate": 3.8762019300764674e-05,
      "loss": 0.43528404235839846,
      "num_input_tokens_seen": 1887308288,
      "step": 6640,
      "train_runtime": 64563.214,
      "train_tokens_per_second": 29231.945
    },
    {
      "epoch": 0.2352904753796381,
      "grad_norm": 0.5625,
      "learning_rate": 3.875037660705273e-05,
      "loss": 0.4310269832611084,
      "num_input_tokens_seen": 1890241536,
      "step": 6650,
      "train_runtime": 64667.2865,
      "train_tokens_per_second": 29230.259
    },
    {
      "epoch": 0.23564429564336686,
      "grad_norm": 0.5703125,
      "learning_rate": 3.873874439816127e-05,
      "loss": 0.43013877868652345,
      "num_input_tokens_seen": 1893055168,
      "step": 6660,
      "train_runtime": 64762.2709,
      "train_tokens_per_second": 29230.834
    },
    {
      "epoch": 0.23599811590709566,
      "grad_norm": 0.5703125,
      "learning_rate": 3.872712265836289e-05,
      "loss": 0.42960572242736816,
      "num_input_tokens_seen": 1895939264,
      "step": 6670,
      "train_runtime": 64863.0824,
      "train_tokens_per_second": 29229.867
    },
    {
      "epoch": 0.23635193617082442,
      "grad_norm": 0.5703125,
      "learning_rate": 3.8715511371963225e-05,
      "loss": 0.4291656494140625,
      "num_input_tokens_seen": 1898837952,
      "step": 6680,
      "train_runtime": 64963.6927,
      "train_tokens_per_second": 29229.218
    },
    {
      "epoch": 0.23670575643455322,
      "grad_norm": 0.55859375,
      "learning_rate": 3.87039105233008e-05,
      "loss": 0.4306159973144531,
      "num_input_tokens_seen": 1901653440,
      "step": 6690,
      "train_runtime": 65057.7854,
      "train_tokens_per_second": 29230.221
    },
    {
      "epoch": 0.23705957669828198,
      "grad_norm": 0.59375,
      "learning_rate": 3.8692320096746975e-05,
      "loss": 0.42613606452941893,
      "num_input_tokens_seen": 1904521472,
      "step": 6700,
      "train_runtime": 65157.0992,
      "train_tokens_per_second": 29229.685
    },
    {
      "epoch": 0.23741339696201075,
      "grad_norm": 0.5703125,
      "learning_rate": 3.868074007670589e-05,
      "loss": 0.4319624900817871,
      "num_input_tokens_seen": 1907357632,
      "step": 6710,
      "train_runtime": 65251.0927,
      "train_tokens_per_second": 29231.045
    },
    {
      "epoch": 0.23776721722573954,
      "grad_norm": 0.5859375,
      "learning_rate": 3.866917044761428e-05,
      "loss": 0.435207462310791,
      "num_input_tokens_seen": 1910141312,
      "step": 6720,
      "train_runtime": 65345.4286,
      "train_tokens_per_second": 29231.445
    },
    {
      "epoch": 0.2381210374894683,
      "grad_norm": 0.61328125,
      "learning_rate": 3.8657611193941486e-05,
      "loss": 0.42978782653808595,
      "num_input_tokens_seen": 1912992640,
      "step": 6730,
      "train_runtime": 65442.3277,
      "train_tokens_per_second": 29231.733
    },
    {
      "epoch": 0.2384748577531971,
      "grad_norm": 0.56640625,
      "learning_rate": 3.8646062300189315e-05,
      "loss": 0.43167934417724607,
      "num_input_tokens_seen": 1915814976,
      "step": 6740,
      "train_runtime": 65538.7604,
      "train_tokens_per_second": 29231.785
    },
    {
      "epoch": 0.23882867801692587,
      "grad_norm": 0.57421875,
      "learning_rate": 3.8634523750891984e-05,
      "loss": 0.43164815902709963,
      "num_input_tokens_seen": 1918640832,
      "step": 6750,
      "train_runtime": 65634.5568,
      "train_tokens_per_second": 29232.175
    },
    {
      "epoch": 0.23918249828065466,
      "grad_norm": 0.5703125,
      "learning_rate": 3.862299553061597e-05,
      "loss": 0.4300533294677734,
      "num_input_tokens_seen": 1921455040,
      "step": 6760,
      "train_runtime": 65730.437,
      "train_tokens_per_second": 29232.348
    },
    {
      "epoch": 0.23953631854438343,
      "grad_norm": 0.56640625,
      "learning_rate": 3.861147762396e-05,
      "loss": 0.42687139511108396,
      "num_input_tokens_seen": 1924308160,
      "step": 6770,
      "train_runtime": 65827.6392,
      "train_tokens_per_second": 29232.526
    },
    {
      "epoch": 0.23989013880811222,
      "grad_norm": 0.58203125,
      "learning_rate": 3.859997001555494e-05,
      "loss": 0.43641343116760256,
      "num_input_tokens_seen": 1927171840,
      "step": 6780,
      "train_runtime": 65927.4964,
      "train_tokens_per_second": 29231.686
    },
    {
      "epoch": 0.24024395907184098,
      "grad_norm": 0.55859375,
      "learning_rate": 3.8588472690063676e-05,
      "loss": 0.424938440322876,
      "num_input_tokens_seen": 1930048320,
      "step": 6790,
      "train_runtime": 66025.3438,
      "train_tokens_per_second": 29231.931
    },
    {
      "epoch": 0.24059777933556978,
      "grad_norm": 0.58203125,
      "learning_rate": 3.857698563218106e-05,
      "loss": 0.426741886138916,
      "num_input_tokens_seen": 1932894976,
      "step": 6800,
      "train_runtime": 66121.7659,
      "train_tokens_per_second": 29232.356
    },
    {
      "epoch": 0.24095159959929854,
      "grad_norm": 0.57421875,
      "learning_rate": 3.8565508826633836e-05,
      "loss": 0.43222684860229493,
      "num_input_tokens_seen": 1935687936,
      "step": 6810,
      "train_runtime": 66216.8666,
      "train_tokens_per_second": 29232.551
    },
    {
      "epoch": 0.24130541986302734,
      "grad_norm": 0.58203125,
      "learning_rate": 3.855404225818049e-05,
      "loss": 0.4261011600494385,
      "num_input_tokens_seen": 1938496576,
      "step": 6820,
      "train_runtime": 66309.3366,
      "train_tokens_per_second": 29234.142
    },
    {
      "epoch": 0.2416592401267561,
      "grad_norm": 0.57421875,
      "learning_rate": 3.8542585911611286e-05,
      "loss": 0.43096199035644533,
      "num_input_tokens_seen": 1941282368,
      "step": 6830,
      "train_runtime": 66406.0347,
      "train_tokens_per_second": 29233.523
    },
    {
      "epoch": 0.2420130603904849,
      "grad_norm": 0.55859375,
      "learning_rate": 3.853113977174803e-05,
      "loss": 0.43064117431640625,
      "num_input_tokens_seen": 1944067968,
      "step": 6840,
      "train_runtime": 66499.026,
      "train_tokens_per_second": 29234.533
    },
    {
      "epoch": 0.24236688065421366,
      "grad_norm": 0.5859375,
      "learning_rate": 3.851970382344411e-05,
      "loss": 0.4254364013671875,
      "num_input_tokens_seen": 1946858304,
      "step": 6850,
      "train_runtime": 66593.1838,
      "train_tokens_per_second": 29235.099
    },
    {
      "epoch": 0.24272070091794246,
      "grad_norm": 0.5859375,
      "learning_rate": 3.850827805158433e-05,
      "loss": 0.42681407928466797,
      "num_input_tokens_seen": 1949729472,
      "step": 6860,
      "train_runtime": 66692.3552,
      "train_tokens_per_second": 29234.677
    },
    {
      "epoch": 0.24307452118167122,
      "grad_norm": 0.57421875,
      "learning_rate": 3.8496862441084896e-05,
      "loss": 0.4354985237121582,
      "num_input_tokens_seen": 1952568000,
      "step": 6870,
      "train_runtime": 66791.1525,
      "train_tokens_per_second": 29233.932
    },
    {
      "epoch": 0.24342834144540001,
      "grad_norm": 0.5859375,
      "learning_rate": 3.848545697689328e-05,
      "loss": 0.43179874420166015,
      "num_input_tokens_seen": 1955436224,
      "step": 6880,
      "train_runtime": 66890.0468,
      "train_tokens_per_second": 29233.59
    },
    {
      "epoch": 0.24378216170912878,
      "grad_norm": 0.58984375,
      "learning_rate": 3.8474061643988136e-05,
      "loss": 0.43176898956298826,
      "num_input_tokens_seen": 1958284672,
      "step": 6890,
      "train_runtime": 66986.404,
      "train_tokens_per_second": 29234.062
    },
    {
      "epoch": 0.24413598197285757,
      "grad_norm": 0.578125,
      "learning_rate": 3.846267642737925e-05,
      "loss": 0.429611873626709,
      "num_input_tokens_seen": 1961151936,
      "step": 6900,
      "train_runtime": 67085.6245,
      "train_tokens_per_second": 29233.565
    },
    {
      "epoch": 0.24448980223658634,
      "grad_norm": 0.578125,
      "learning_rate": 3.8451301312107455e-05,
      "loss": 0.4352381706237793,
      "num_input_tokens_seen": 1963977536,
      "step": 6910,
      "train_runtime": 67178.4117,
      "train_tokens_per_second": 29235.248
    },
    {
      "epoch": 0.24484362250031513,
      "grad_norm": 0.6640625,
      "learning_rate": 3.843993628324451e-05,
      "loss": 0.43280868530273436,
      "num_input_tokens_seen": 1966882304,
      "step": 6920,
      "train_runtime": 67279.0431,
      "train_tokens_per_second": 29234.695
    },
    {
      "epoch": 0.2451974427640439,
      "grad_norm": 0.58203125,
      "learning_rate": 3.8428581325893034e-05,
      "loss": 0.42553396224975587,
      "num_input_tokens_seen": 1969747328,
      "step": 6930,
      "train_runtime": 67378.6292,
      "train_tokens_per_second": 29234.007
    },
    {
      "epoch": 0.2455512630277727,
      "grad_norm": 0.578125,
      "learning_rate": 3.8417236425186484e-05,
      "loss": 0.4305004119873047,
      "num_input_tokens_seen": 1972638400,
      "step": 6940,
      "train_runtime": 67478.0472,
      "train_tokens_per_second": 29233.78
    },
    {
      "epoch": 0.24590508329150146,
      "grad_norm": 0.55078125,
      "learning_rate": 3.840590156628895e-05,
      "loss": 0.4329527854919434,
      "num_input_tokens_seen": 1975386240,
      "step": 6950,
      "train_runtime": 67569.6218,
      "train_tokens_per_second": 29234.828
    },
    {
      "epoch": 0.24625890355523022,
      "grad_norm": 0.59765625,
      "learning_rate": 3.8394576734395205e-05,
      "loss": 0.4306673049926758,
      "num_input_tokens_seen": 1978244864,
      "step": 6960,
      "train_runtime": 67666.1616,
      "train_tokens_per_second": 29235.364
    },
    {
      "epoch": 0.24661272381895902,
      "grad_norm": 0.59375,
      "learning_rate": 3.838326191473054e-05,
      "loss": 0.4305530548095703,
      "num_input_tokens_seen": 1981100288,
      "step": 6970,
      "train_runtime": 67762.2022,
      "train_tokens_per_second": 29236.067
    },
    {
      "epoch": 0.24696654408268778,
      "grad_norm": 0.5625,
      "learning_rate": 3.837195709255069e-05,
      "loss": 0.4284849166870117,
      "num_input_tokens_seen": 1983949504,
      "step": 6980,
      "train_runtime": 67859.4366,
      "train_tokens_per_second": 29236.162
    },
    {
      "epoch": 0.24732036434641658,
      "grad_norm": 0.5703125,
      "learning_rate": 3.8360662253141796e-05,
      "loss": 0.4298586368560791,
      "num_input_tokens_seen": 1986802688,
      "step": 6990,
      "train_runtime": 67957.57,
      "train_tokens_per_second": 29235.929
    },
    {
      "epoch": 0.24767418461014534,
      "grad_norm": 0.58203125,
      "learning_rate": 3.834937738182029e-05,
      "loss": 0.4331483840942383,
      "num_input_tokens_seen": 1989645568,
      "step": 7000,
      "train_runtime": 68055.4779,
      "train_tokens_per_second": 29235.642
    },
    {
      "epoch": 0.24802800487387414,
      "grad_norm": 0.546875,
      "learning_rate": 3.833810246393281e-05,
      "loss": 0.4273562431335449,
      "num_input_tokens_seen": 1992492736,
      "step": 7010,
      "train_runtime": 68152.3788,
      "train_tokens_per_second": 29235.85
    },
    {
      "epoch": 0.2483818251376029,
      "grad_norm": 0.5703125,
      "learning_rate": 3.832683748485616e-05,
      "loss": 0.4271261215209961,
      "num_input_tokens_seen": 1995362560,
      "step": 7020,
      "train_runtime": 68253.171,
      "train_tokens_per_second": 29234.723
    },
    {
      "epoch": 0.2487356454013317,
      "grad_norm": 0.5546875,
      "learning_rate": 3.8315582429997184e-05,
      "loss": 0.42997045516967775,
      "num_input_tokens_seen": 1998209664,
      "step": 7030,
      "train_runtime": 68350.3865,
      "train_tokens_per_second": 29234.797
    },
    {
      "epoch": 0.24908946566506046,
      "grad_norm": 0.5625,
      "learning_rate": 3.830433728479272e-05,
      "loss": 0.43201465606689454,
      "num_input_tokens_seen": 2001065600,
      "step": 7040,
      "train_runtime": 68448.9739,
      "train_tokens_per_second": 29234.413
    },
    {
      "epoch": 0.24944328592878925,
      "grad_norm": 0.53515625,
      "learning_rate": 3.829310203470948e-05,
      "loss": 0.42966575622558595,
      "num_input_tokens_seen": 2003880192,
      "step": 7050,
      "train_runtime": 68545.3122,
      "train_tokens_per_second": 29234.387
    },
    {
      "epoch": 0.24979710619251802,
      "grad_norm": 0.59375,
      "learning_rate": 3.828187666524403e-05,
      "loss": 0.4268825531005859,
      "num_input_tokens_seen": 2006683904,
      "step": 7060,
      "train_runtime": 68638.518,
      "train_tokens_per_second": 29235.537
    },
    {
      "epoch": 0.2501509264562468,
      "grad_norm": 0.5859375,
      "learning_rate": 3.827066116192266e-05,
      "loss": 0.4267873764038086,
      "num_input_tokens_seen": 2009511232,
      "step": 7070,
      "train_runtime": 68734.7504,
      "train_tokens_per_second": 29235.739
    },
    {
      "epoch": 0.2505047467199756,
      "grad_norm": 0.59765625,
      "learning_rate": 3.825945551030135e-05,
      "loss": 0.4300816535949707,
      "num_input_tokens_seen": 2012319744,
      "step": 7080,
      "train_runtime": 68830.1799,
      "train_tokens_per_second": 29236.009
    },
    {
      "epoch": 0.25085856698370435,
      "grad_norm": 0.578125,
      "learning_rate": 3.824825969596561e-05,
      "loss": 0.4281140327453613,
      "num_input_tokens_seen": 2015164544,
      "step": 7090,
      "train_runtime": 68927.3037,
      "train_tokens_per_second": 29236.085
    },
    {
      "epoch": 0.25121238724743317,
      "grad_norm": 0.57421875,
      "learning_rate": 3.823707370453054e-05,
      "loss": 0.43103675842285155,
      "num_input_tokens_seen": 2018028928,
      "step": 7100,
      "train_runtime": 69026.2347,
      "train_tokens_per_second": 29235.68
    },
    {
      "epoch": 0.25156620751116193,
      "grad_norm": 0.56640625,
      "learning_rate": 3.8225897521640614e-05,
      "loss": 0.42860612869262693,
      "num_input_tokens_seen": 2020869120,
      "step": 7110,
      "train_runtime": 69121.0515,
      "train_tokens_per_second": 29236.666
    },
    {
      "epoch": 0.2519200277748907,
      "grad_norm": 0.5625,
      "learning_rate": 3.8214731132969675e-05,
      "loss": 0.429230785369873,
      "num_input_tokens_seen": 2023728320,
      "step": 7120,
      "train_runtime": 69220.8027,
      "train_tokens_per_second": 29235.84
    },
    {
      "epoch": 0.25227384803861946,
      "grad_norm": 0.56640625,
      "learning_rate": 3.820357452422084e-05,
      "loss": 0.4243276596069336,
      "num_input_tokens_seen": 2026488448,
      "step": 7130,
      "train_runtime": 69313.8973,
      "train_tokens_per_second": 29236.395
    },
    {
      "epoch": 0.2526276683023483,
      "grad_norm": 0.5703125,
      "learning_rate": 3.8192427681126445e-05,
      "loss": 0.42637834548950193,
      "num_input_tokens_seen": 2029323392,
      "step": 7140,
      "train_runtime": 69411.0809,
      "train_tokens_per_second": 29236.303
    },
    {
      "epoch": 0.25298148856607705,
      "grad_norm": 0.5703125,
      "learning_rate": 3.818129058944793e-05,
      "loss": 0.4294004440307617,
      "num_input_tokens_seen": 2032191488,
      "step": 7150,
      "train_runtime": 69508.8643,
      "train_tokens_per_second": 29236.436
    },
    {
      "epoch": 0.2533353088298058,
      "grad_norm": 0.55078125,
      "learning_rate": 3.817016323497578e-05,
      "loss": 0.42581949234008787,
      "num_input_tokens_seen": 2035036480,
      "step": 7160,
      "train_runtime": 69605.8708,
      "train_tokens_per_second": 29236.564
    },
    {
      "epoch": 0.2536891290935346,
      "grad_norm": 0.5859375,
      "learning_rate": 3.8159045603529455e-05,
      "loss": 0.42792572975158694,
      "num_input_tokens_seen": 2037872896,
      "step": 7170,
      "train_runtime": 69700.9018,
      "train_tokens_per_second": 29237.396
    },
    {
      "epoch": 0.2540429493572634,
      "grad_norm": 0.58203125,
      "learning_rate": 3.8147937680957334e-05,
      "loss": 0.42626428604125977,
      "num_input_tokens_seen": 2040756736,
      "step": 7180,
      "train_runtime": 69802.394,
      "train_tokens_per_second": 29236.2
    },
    {
      "epoch": 0.25439676962099217,
      "grad_norm": 0.56640625,
      "learning_rate": 3.813683945313658e-05,
      "loss": 0.4278706550598145,
      "num_input_tokens_seen": 2043658624,
      "step": 7190,
      "train_runtime": 69904.8704,
      "train_tokens_per_second": 29234.853
    },
    {
      "epoch": 0.25475058988472093,
      "grad_norm": 0.58984375,
      "learning_rate": 3.812575090597313e-05,
      "loss": 0.4257061004638672,
      "num_input_tokens_seen": 2046480768,
      "step": 7200,
      "train_runtime": 69998.9083,
      "train_tokens_per_second": 29235.896
    },
    {
      "epoch": 0.2551044101484497,
      "grad_norm": 0.56640625,
      "learning_rate": 3.811467202540156e-05,
      "loss": 0.4267751693725586,
      "num_input_tokens_seen": 2049320192,
      "step": 7210,
      "train_runtime": 70096.5692,
      "train_tokens_per_second": 29235.67
    },
    {
      "epoch": 0.25545823041217847,
      "grad_norm": 0.5703125,
      "learning_rate": 3.810360279738507e-05,
      "loss": 0.4266937732696533,
      "num_input_tokens_seen": 2052214592,
      "step": 7220,
      "train_runtime": 70197.7757,
      "train_tokens_per_second": 29234.752
    },
    {
      "epoch": 0.2558120506759073,
      "grad_norm": 0.60546875,
      "learning_rate": 3.809254320791535e-05,
      "loss": 0.42996883392333984,
      "num_input_tokens_seen": 2055053696,
      "step": 7230,
      "train_runtime": 70296.1801,
      "train_tokens_per_second": 29234.216
    },
    {
      "epoch": 0.25616587093963605,
      "grad_norm": 0.56640625,
      "learning_rate": 3.808149324301256e-05,
      "loss": 0.4278991222381592,
      "num_input_tokens_seen": 2057861696,
      "step": 7240,
      "train_runtime": 70391.1548,
      "train_tokens_per_second": 29234.663
    },
    {
      "epoch": 0.2565196912033648,
      "grad_norm": 0.57421875,
      "learning_rate": 3.807045288872522e-05,
      "loss": 0.43072357177734377,
      "num_input_tokens_seen": 2060795840,
      "step": 7250,
      "train_runtime": 70494.545,
      "train_tokens_per_second": 29233.409
    },
    {
      "epoch": 0.2568735114670936,
      "grad_norm": 0.578125,
      "learning_rate": 3.805942213113015e-05,
      "loss": 0.4289708137512207,
      "num_input_tokens_seen": 2063608512,
      "step": 7260,
      "train_runtime": 70591.2357,
      "train_tokens_per_second": 29233.211
    },
    {
      "epoch": 0.2572273317308224,
      "grad_norm": 0.5703125,
      "learning_rate": 3.8048400956332385e-05,
      "loss": 0.42472009658813475,
      "num_input_tokens_seen": 2066395712,
      "step": 7270,
      "train_runtime": 70684.7815,
      "train_tokens_per_second": 29233.955
    },
    {
      "epoch": 0.25758115199455117,
      "grad_norm": 0.58984375,
      "learning_rate": 3.803738935046512e-05,
      "loss": 0.42498350143432617,
      "num_input_tokens_seen": 2069224768,
      "step": 7280,
      "train_runtime": 70776.3714,
      "train_tokens_per_second": 29236.096
    },
    {
      "epoch": 0.25793497225827994,
      "grad_norm": 0.5859375,
      "learning_rate": 3.802638729968962e-05,
      "loss": 0.4275249481201172,
      "num_input_tokens_seen": 2072058432,
      "step": 7290,
      "train_runtime": 70871.7768,
      "train_tokens_per_second": 29236.722
    },
    {
      "epoch": 0.2582887925220087,
      "grad_norm": 0.59375,
      "learning_rate": 3.8015394790195145e-05,
      "loss": 0.4259969234466553,
      "num_input_tokens_seen": 2074911296,
      "step": 7300,
      "train_runtime": 70968.1521,
      "train_tokens_per_second": 29237.217
    },
    {
      "epoch": 0.2586426127857375,
      "grad_norm": 0.55078125,
      "learning_rate": 3.800441180819891e-05,
      "loss": 0.42400274276733396,
      "num_input_tokens_seen": 2077735040,
      "step": 7310,
      "train_runtime": 71061.1667,
      "train_tokens_per_second": 29238.685
    },
    {
      "epoch": 0.2589964330494663,
      "grad_norm": 0.546875,
      "learning_rate": 3.7993438339945965e-05,
      "loss": 0.4285571098327637,
      "num_input_tokens_seen": 2080523264,
      "step": 7320,
      "train_runtime": 71156.1282,
      "train_tokens_per_second": 29238.849
    },
    {
      "epoch": 0.25935025331319506,
      "grad_norm": 0.5703125,
      "learning_rate": 3.798247437170914e-05,
      "loss": 0.4280029296875,
      "num_input_tokens_seen": 2083436160,
      "step": 7330,
      "train_runtime": 71255.2592,
      "train_tokens_per_second": 29239.051
    },
    {
      "epoch": 0.2597040735769238,
      "grad_norm": 0.55859375,
      "learning_rate": 3.797151988978901e-05,
      "loss": 0.4295159339904785,
      "num_input_tokens_seen": 2086317440,
      "step": 7340,
      "train_runtime": 71355.8136,
      "train_tokens_per_second": 29238.226
    },
    {
      "epoch": 0.26005789384065264,
      "grad_norm": 0.5703125,
      "learning_rate": 3.796057488051377e-05,
      "loss": 0.4264115810394287,
      "num_input_tokens_seen": 2089149312,
      "step": 7350,
      "train_runtime": 71452.8076,
      "train_tokens_per_second": 29238.17
    },
    {
      "epoch": 0.2604117141043814,
      "grad_norm": 0.578125,
      "learning_rate": 3.794963933023918e-05,
      "loss": 0.43225893974304197,
      "num_input_tokens_seen": 2092033920,
      "step": 7360,
      "train_runtime": 71552.6994,
      "train_tokens_per_second": 29237.666
    },
    {
      "epoch": 0.2607655343681102,
      "grad_norm": 0.5703125,
      "learning_rate": 3.79387132253485e-05,
      "loss": 0.42646141052246095,
      "num_input_tokens_seen": 2094888320,
      "step": 7370,
      "train_runtime": 71648.1577,
      "train_tokens_per_second": 29238.551
    },
    {
      "epoch": 0.26111935463183894,
      "grad_norm": 0.5859375,
      "learning_rate": 3.792779655225243e-05,
      "loss": 0.4248672962188721,
      "num_input_tokens_seen": 2097696448,
      "step": 7380,
      "train_runtime": 71743.1746,
      "train_tokens_per_second": 29238.969
    },
    {
      "epoch": 0.26147317489556776,
      "grad_norm": 0.56640625,
      "learning_rate": 3.791688929738902e-05,
      "loss": 0.4293975830078125,
      "num_input_tokens_seen": 2100542080,
      "step": 7390,
      "train_runtime": 71840.467,
      "train_tokens_per_second": 29238.981
    },
    {
      "epoch": 0.2618269951592965,
      "grad_norm": 0.55859375,
      "learning_rate": 3.79059914472236e-05,
      "loss": 0.4282999515533447,
      "num_input_tokens_seen": 2103424704,
      "step": 7400,
      "train_runtime": 71939.9939,
      "train_tokens_per_second": 29238.6
    },
    {
      "epoch": 0.2621808154230253,
      "grad_norm": 0.5703125,
      "learning_rate": 3.7895102988248716e-05,
      "loss": 0.4268609046936035,
      "num_input_tokens_seen": 2106209600,
      "step": 7410,
      "train_runtime": 72030.6012,
      "train_tokens_per_second": 29240.483
    },
    {
      "epoch": 0.26253463568675406,
      "grad_norm": 0.55078125,
      "learning_rate": 3.7884223906984064e-05,
      "loss": 0.428282356262207,
      "num_input_tokens_seen": 2109084864,
      "step": 7420,
      "train_runtime": 72131.8065,
      "train_tokens_per_second": 29239.318
    },
    {
      "epoch": 0.2628884559504829,
      "grad_norm": 0.56640625,
      "learning_rate": 3.787335418997641e-05,
      "loss": 0.42298011779785155,
      "num_input_tokens_seen": 2111953856,
      "step": 7430,
      "train_runtime": 72229.0755,
      "train_tokens_per_second": 29239.663
    },
    {
      "epoch": 0.26324227621421165,
      "grad_norm": 0.55078125,
      "learning_rate": 3.786249382379952e-05,
      "loss": 0.42200241088867185,
      "num_input_tokens_seen": 2114753408,
      "step": 7440,
      "train_runtime": 72323.5023,
      "train_tokens_per_second": 29240.196
    },
    {
      "epoch": 0.2635960964779404,
      "grad_norm": 0.56640625,
      "learning_rate": 3.785164279505411e-05,
      "loss": 0.4272355556488037,
      "num_input_tokens_seen": 2117616384,
      "step": 7450,
      "train_runtime": 72422.6944,
      "train_tokens_per_second": 29239.68
    },
    {
      "epoch": 0.2639499167416692,
      "grad_norm": 0.55859375,
      "learning_rate": 3.7840801090367744e-05,
      "loss": 0.4236455917358398,
      "num_input_tokens_seen": 2120492864,
      "step": 7460,
      "train_runtime": 72522.5693,
      "train_tokens_per_second": 29239.075
    },
    {
      "epoch": 0.26430373700539794,
      "grad_norm": 0.55078125,
      "learning_rate": 3.782996869639479e-05,
      "loss": 0.42912778854370115,
      "num_input_tokens_seen": 2123302848,
      "step": 7470,
      "train_runtime": 72618.1433,
      "train_tokens_per_second": 29239.289
    },
    {
      "epoch": 0.26465755726912676,
      "grad_norm": 0.58203125,
      "learning_rate": 3.7819145599816354e-05,
      "loss": 0.42866220474243166,
      "num_input_tokens_seen": 2126127488,
      "step": 7480,
      "train_runtime": 72712.092,
      "train_tokens_per_second": 29240.356
    },
    {
      "epoch": 0.26501137753285553,
      "grad_norm": 0.59375,
      "learning_rate": 3.780833178734018e-05,
      "loss": 0.43070359230041505,
      "num_input_tokens_seen": 2129004864,
      "step": 7490,
      "train_runtime": 72811.6123,
      "train_tokens_per_second": 29239.908
    },
    {
      "epoch": 0.2653651977965843,
      "grad_norm": 0.56640625,
      "learning_rate": 3.77975272457006e-05,
      "loss": 0.4281341552734375,
      "num_input_tokens_seen": 2131798656,
      "step": 7500,
      "train_runtime": 72905.6241,
      "train_tokens_per_second": 29240.524
    },
    {
      "epoch": 0.26571901806031306,
      "grad_norm": 0.56640625,
      "learning_rate": 3.778673196165851e-05,
      "loss": 0.4266397476196289,
      "num_input_tokens_seen": 2134637504,
      "step": 7510,
      "train_runtime": 73002.9886,
      "train_tokens_per_second": 29240.413
    },
    {
      "epoch": 0.2660728383240419,
      "grad_norm": 0.58203125,
      "learning_rate": 3.7775945922001186e-05,
      "loss": 0.4280815601348877,
      "num_input_tokens_seen": 2137523456,
      "step": 7520,
      "train_runtime": 73102.4027,
      "train_tokens_per_second": 29240.126
    },
    {
      "epoch": 0.26642665858777065,
      "grad_norm": 0.5859375,
      "learning_rate": 3.776516911354236e-05,
      "loss": 0.42893023490905763,
      "num_input_tokens_seen": 2140406464,
      "step": 7530,
      "train_runtime": 73197.5663,
      "train_tokens_per_second": 29241.498
    },
    {
      "epoch": 0.2667804788514994,
      "grad_norm": 0.5703125,
      "learning_rate": 3.775440152312205e-05,
      "loss": 0.4321183204650879,
      "num_input_tokens_seen": 2143232512,
      "step": 7540,
      "train_runtime": 73292.9331,
      "train_tokens_per_second": 29242.008
    },
    {
      "epoch": 0.2671342991152282,
      "grad_norm": 0.5703125,
      "learning_rate": 3.774364313760652e-05,
      "loss": 0.42190303802490237,
      "num_input_tokens_seen": 2146056960,
      "step": 7550,
      "train_runtime": 73387.9673,
      "train_tokens_per_second": 29242.627
    },
    {
      "epoch": 0.267488119378957,
      "grad_norm": 0.57421875,
      "learning_rate": 3.7732893943888224e-05,
      "loss": 0.42957496643066406,
      "num_input_tokens_seen": 2148890048,
      "step": 7560,
      "train_runtime": 73483.3392,
      "train_tokens_per_second": 29243.228
    },
    {
      "epoch": 0.26784193964268577,
      "grad_norm": 0.56640625,
      "learning_rate": 3.772215392888574e-05,
      "loss": 0.42708258628845214,
      "num_input_tokens_seen": 2151736000,
      "step": 7570,
      "train_runtime": 73578.9924,
      "train_tokens_per_second": 29243.891
    },
    {
      "epoch": 0.26819575990641453,
      "grad_norm": 0.5859375,
      "learning_rate": 3.771142307954368e-05,
      "loss": 0.42136316299438475,
      "num_input_tokens_seen": 2154576640,
      "step": 7580,
      "train_runtime": 73674.3327,
      "train_tokens_per_second": 29244.603
    },
    {
      "epoch": 0.2685495801701433,
      "grad_norm": 0.57421875,
      "learning_rate": 3.770070138283264e-05,
      "loss": 0.42917208671569823,
      "num_input_tokens_seen": 2157471680,
      "step": 7590,
      "train_runtime": 73771.8775,
      "train_tokens_per_second": 29245.178
    },
    {
      "epoch": 0.2689034004338721,
      "grad_norm": 0.54296875,
      "learning_rate": 3.768998882574915e-05,
      "loss": 0.4248990058898926,
      "num_input_tokens_seen": 2160348288,
      "step": 7600,
      "train_runtime": 73874.2149,
      "train_tokens_per_second": 29243.604
    },
    {
      "epoch": 0.2692572206976009,
      "grad_norm": 0.578125,
      "learning_rate": 3.767928539531557e-05,
      "loss": 0.43067193031311035,
      "num_input_tokens_seen": 2163158784,
      "step": 7610,
      "train_runtime": 73969.3217,
      "train_tokens_per_second": 29243.999
    },
    {
      "epoch": 0.26961104096132965,
      "grad_norm": 0.57421875,
      "learning_rate": 3.7668591078580055e-05,
      "loss": 0.4271222114562988,
      "num_input_tokens_seen": 2166053632,
      "step": 7620,
      "train_runtime": 74072.7674,
      "train_tokens_per_second": 29242.24
    },
    {
      "epoch": 0.2699648612250584,
      "grad_norm": 0.5703125,
      "learning_rate": 3.765790586261647e-05,
      "loss": 0.42212657928466796,
      "num_input_tokens_seen": 2168858112,
      "step": 7630,
      "train_runtime": 74168.5013,
      "train_tokens_per_second": 29242.307
    },
    {
      "epoch": 0.27031868148878724,
      "grad_norm": 0.53125,
      "learning_rate": 3.7647229734524326e-05,
      "loss": 0.42572507858276365,
      "num_input_tokens_seen": 2171655104,
      "step": 7640,
      "train_runtime": 74261.4714,
      "train_tokens_per_second": 29243.362
    },
    {
      "epoch": 0.270672501752516,
      "grad_norm": 0.5859375,
      "learning_rate": 3.7636562681428744e-05,
      "loss": 0.42652573585510256,
      "num_input_tokens_seen": 2174459328,
      "step": 7650,
      "train_runtime": 74354.356,
      "train_tokens_per_second": 29244.545
    },
    {
      "epoch": 0.27102632201624477,
      "grad_norm": 0.56640625,
      "learning_rate": 3.7625904690480346e-05,
      "loss": 0.4243968963623047,
      "num_input_tokens_seen": 2177253312,
      "step": 7660,
      "train_runtime": 74448.7434,
      "train_tokens_per_second": 29244.997
    },
    {
      "epoch": 0.27138014227997354,
      "grad_norm": 0.5625,
      "learning_rate": 3.7615255748855224e-05,
      "loss": 0.42897806167602537,
      "num_input_tokens_seen": 2180137280,
      "step": 7670,
      "train_runtime": 74547.5464,
      "train_tokens_per_second": 29244.923
    },
    {
      "epoch": 0.27173396254370236,
      "grad_norm": 0.578125,
      "learning_rate": 3.7604615843754845e-05,
      "loss": 0.42595462799072265,
      "num_input_tokens_seen": 2182987840,
      "step": 7680,
      "train_runtime": 74645.6274,
      "train_tokens_per_second": 29244.685
    },
    {
      "epoch": 0.2720877828074311,
      "grad_norm": 0.578125,
      "learning_rate": 3.759398496240601e-05,
      "loss": 0.42676386833190916,
      "num_input_tokens_seen": 2185845312,
      "step": 7690,
      "train_runtime": 74741.6041,
      "train_tokens_per_second": 29245.363
    },
    {
      "epoch": 0.2724416030711599,
      "grad_norm": 0.55859375,
      "learning_rate": 3.7583363092060815e-05,
      "loss": 0.424839448928833,
      "num_input_tokens_seen": 2188683776,
      "step": 7700,
      "train_runtime": 74838.4827,
      "train_tokens_per_second": 29245.432
    },
    {
      "epoch": 0.27279542333488865,
      "grad_norm": 0.60546875,
      "learning_rate": 3.757275021999649e-05,
      "loss": 0.426145076751709,
      "num_input_tokens_seen": 2191521152,
      "step": 7710,
      "train_runtime": 74937.1507,
      "train_tokens_per_second": 29244.789
    },
    {
      "epoch": 0.2731492435986175,
      "grad_norm": 0.56640625,
      "learning_rate": 3.7562146333515445e-05,
      "loss": 0.43163356781005857,
      "num_input_tokens_seen": 2194422656,
      "step": 7720,
      "train_runtime": 75034.2576,
      "train_tokens_per_second": 29245.61
    },
    {
      "epoch": 0.27350306386234624,
      "grad_norm": 0.55859375,
      "learning_rate": 3.7551551419945167e-05,
      "loss": 0.42371454238891604,
      "num_input_tokens_seen": 2197244288,
      "step": 7730,
      "train_runtime": 75130.2214,
      "train_tokens_per_second": 29245.811
    },
    {
      "epoch": 0.273856884126075,
      "grad_norm": 0.5703125,
      "learning_rate": 3.7540965466638104e-05,
      "loss": 0.424973201751709,
      "num_input_tokens_seen": 2200139968,
      "step": 7740,
      "train_runtime": 75229.9568,
      "train_tokens_per_second": 29245.53
    },
    {
      "epoch": 0.27421070438980377,
      "grad_norm": 0.57421875,
      "learning_rate": 3.753038846097172e-05,
      "loss": 0.4214158058166504,
      "num_input_tokens_seen": 2203025792,
      "step": 7750,
      "train_runtime": 75328.5743,
      "train_tokens_per_second": 29245.553
    },
    {
      "epoch": 0.27456452465353254,
      "grad_norm": 0.5859375,
      "learning_rate": 3.751982039034827e-05,
      "loss": 0.42480859756469724,
      "num_input_tokens_seen": 2205819776,
      "step": 7760,
      "train_runtime": 75422.9713,
      "train_tokens_per_second": 29245.994
    },
    {
      "epoch": 0.27491834491726136,
      "grad_norm": 0.55078125,
      "learning_rate": 3.75092612421949e-05,
      "loss": 0.42972517013549805,
      "num_input_tokens_seen": 2208704192,
      "step": 7770,
      "train_runtime": 75522.0637,
      "train_tokens_per_second": 29245.814
    },
    {
      "epoch": 0.2752721651809901,
      "grad_norm": 0.578125,
      "learning_rate": 3.7498711003963475e-05,
      "loss": 0.43226261138916017,
      "num_input_tokens_seen": 2211600192,
      "step": 7780,
      "train_runtime": 75623.5631,
      "train_tokens_per_second": 29244.856
    },
    {
      "epoch": 0.2756259854447189,
      "grad_norm": 0.55078125,
      "learning_rate": 3.748816966313058e-05,
      "loss": 0.4253261566162109,
      "num_input_tokens_seen": 2214473088,
      "step": 7790,
      "train_runtime": 75723.3439,
      "train_tokens_per_second": 29244.259
    },
    {
      "epoch": 0.27597980570844766,
      "grad_norm": 0.5703125,
      "learning_rate": 3.7477637207197374e-05,
      "loss": 0.42342290878295896,
      "num_input_tokens_seen": 2217345664,
      "step": 7800,
      "train_runtime": 75825.6471,
      "train_tokens_per_second": 29242.687
    },
    {
      "epoch": 0.2763336259721765,
      "grad_norm": 0.5859375,
      "learning_rate": 3.7467113623689666e-05,
      "loss": 0.4228833198547363,
      "num_input_tokens_seen": 2220153792,
      "step": 7810,
      "train_runtime": 75920.6525,
      "train_tokens_per_second": 29243.081
    },
    {
      "epoch": 0.27668744623590524,
      "grad_norm": 0.58203125,
      "learning_rate": 3.745659890015768e-05,
      "loss": 0.419264030456543,
      "num_input_tokens_seen": 2222968896,
      "step": 7820,
      "train_runtime": 76017.0949,
      "train_tokens_per_second": 29243.013
    },
    {
      "epoch": 0.277041266499634,
      "grad_norm": 0.5859375,
      "learning_rate": 3.744609302417615e-05,
      "loss": 0.42331132888793943,
      "num_input_tokens_seen": 2225786688,
      "step": 7830,
      "train_runtime": 76114.9431,
      "train_tokens_per_second": 29242.44
    },
    {
      "epoch": 0.2773950867633628,
      "grad_norm": 0.58984375,
      "learning_rate": 3.7435595983344175e-05,
      "loss": 0.42845883369445803,
      "num_input_tokens_seen": 2228604288,
      "step": 7840,
      "train_runtime": 76213.185,
      "train_tokens_per_second": 29241.716
    },
    {
      "epoch": 0.2777489070270916,
      "grad_norm": 0.5625,
      "learning_rate": 3.7425107765285155e-05,
      "loss": 0.4245955467224121,
      "num_input_tokens_seen": 2231460224,
      "step": 7850,
      "train_runtime": 76310.4332,
      "train_tokens_per_second": 29241.876
    },
    {
      "epoch": 0.27810272729082036,
      "grad_norm": 0.58984375,
      "learning_rate": 3.741462835764676e-05,
      "loss": 0.4287534713745117,
      "num_input_tokens_seen": 2234335488,
      "step": 7860,
      "train_runtime": 76408.8776,
      "train_tokens_per_second": 29241.831
    },
    {
      "epoch": 0.2784565475545491,
      "grad_norm": 0.578125,
      "learning_rate": 3.740415774810088e-05,
      "loss": 0.4283725738525391,
      "num_input_tokens_seen": 2237156352,
      "step": 7870,
      "train_runtime": 76503.359,
      "train_tokens_per_second": 29242.59
    },
    {
      "epoch": 0.2788103678182779,
      "grad_norm": 0.55078125,
      "learning_rate": 3.739369592434351e-05,
      "loss": 0.42887067794799805,
      "num_input_tokens_seen": 2239989440,
      "step": 7880,
      "train_runtime": 76598.2932,
      "train_tokens_per_second": 29243.334
    },
    {
      "epoch": 0.2791641880820067,
      "grad_norm": 0.56640625,
      "learning_rate": 3.738324287409473e-05,
      "loss": 0.42405362129211427,
      "num_input_tokens_seen": 2242795072,
      "step": 7890,
      "train_runtime": 76694.3023,
      "train_tokens_per_second": 29243.308
    },
    {
      "epoch": 0.2795180083457355,
      "grad_norm": 0.5625,
      "learning_rate": 3.7372798585098644e-05,
      "loss": 0.4241333961486816,
      "num_input_tokens_seen": 2245659968,
      "step": 7900,
      "train_runtime": 76791.5857,
      "train_tokens_per_second": 29243.568
    },
    {
      "epoch": 0.27987182860946425,
      "grad_norm": 0.57421875,
      "learning_rate": 3.736236304512331e-05,
      "loss": 0.42028160095214845,
      "num_input_tokens_seen": 2248550016,
      "step": 7910,
      "train_runtime": 76890.7189,
      "train_tokens_per_second": 29243.452
    },
    {
      "epoch": 0.280225648873193,
      "grad_norm": 0.57421875,
      "learning_rate": 3.735193624196067e-05,
      "loss": 0.43004746437072755,
      "num_input_tokens_seen": 2251377408,
      "step": 7920,
      "train_runtime": 76989.4668,
      "train_tokens_per_second": 29242.668
    },
    {
      "epoch": 0.28057946913692183,
      "grad_norm": 0.57421875,
      "learning_rate": 3.7341518163426514e-05,
      "loss": 0.42587881088256835,
      "num_input_tokens_seen": 2254267840,
      "step": 7930,
      "train_runtime": 77090.1855,
      "train_tokens_per_second": 29241.956
    },
    {
      "epoch": 0.2809332894006506,
      "grad_norm": 0.5859375,
      "learning_rate": 3.73311087973604e-05,
      "loss": 0.42616939544677734,
      "num_input_tokens_seen": 2257108288,
      "step": 7940,
      "train_runtime": 77187.29,
      "train_tokens_per_second": 29241.968
    },
    {
      "epoch": 0.28128710966437936,
      "grad_norm": 0.57421875,
      "learning_rate": 3.732070813162561e-05,
      "loss": 0.42601823806762695,
      "num_input_tokens_seen": 2259932288,
      "step": 7950,
      "train_runtime": 77283.1284,
      "train_tokens_per_second": 29242.246
    },
    {
      "epoch": 0.28164092992810813,
      "grad_norm": 0.58203125,
      "learning_rate": 3.731031615410908e-05,
      "loss": 0.42812113761901854,
      "num_input_tokens_seen": 2262767296,
      "step": 7960,
      "train_runtime": 77382.4329,
      "train_tokens_per_second": 29241.356
    },
    {
      "epoch": 0.28199475019183695,
      "grad_norm": 0.5859375,
      "learning_rate": 3.729993285272132e-05,
      "loss": 0.4210846424102783,
      "num_input_tokens_seen": 2265615168,
      "step": 7970,
      "train_runtime": 77476.3212,
      "train_tokens_per_second": 29242.679
    },
    {
      "epoch": 0.2823485704555657,
      "grad_norm": 0.56640625,
      "learning_rate": 3.7289558215396414e-05,
      "loss": 0.426118278503418,
      "num_input_tokens_seen": 2268486528,
      "step": 7980,
      "train_runtime": 77576.4042,
      "train_tokens_per_second": 29241.965
    },
    {
      "epoch": 0.2827023907192945,
      "grad_norm": 0.58203125,
      "learning_rate": 3.727919223009191e-05,
      "loss": 0.4243349552154541,
      "num_input_tokens_seen": 2271343488,
      "step": 7990,
      "train_runtime": 77673.6338,
      "train_tokens_per_second": 29242.143
    },
    {
      "epoch": 0.28305621098302325,
      "grad_norm": 0.5859375,
      "learning_rate": 3.726883488478877e-05,
      "loss": 0.4312952995300293,
      "num_input_tokens_seen": 2274239872,
      "step": 8000,
      "train_runtime": 77775.909,
      "train_tokens_per_second": 29240.929
    },
    {
      "epoch": 0.283410031246752,
      "grad_norm": 0.58984375,
      "learning_rate": 3.7258486167491323e-05,
      "loss": 0.42570013999938966,
      "num_input_tokens_seen": 2277056192,
      "step": 8010,
      "train_runtime": 77874.6188,
      "train_tokens_per_second": 29240.03
    },
    {
      "epoch": 0.28376385151048084,
      "grad_norm": 0.56640625,
      "learning_rate": 3.724814606622721e-05,
      "loss": 0.4266170024871826,
      "num_input_tokens_seen": 2279883712,
      "step": 8020,
      "train_runtime": 77970.9895,
      "train_tokens_per_second": 29240.154
    },
    {
      "epoch": 0.2841176717742096,
      "grad_norm": 0.58984375,
      "learning_rate": 3.7237814569047294e-05,
      "loss": 0.42457008361816406,
      "num_input_tokens_seen": 2282755712,
      "step": 8030,
      "train_runtime": 78070.525,
      "train_tokens_per_second": 29239.661
    },
    {
      "epoch": 0.28447149203793837,
      "grad_norm": 0.5625,
      "learning_rate": 3.7227491664025656e-05,
      "loss": 0.42130188941955565,
      "num_input_tokens_seen": 2285597504,
      "step": 8040,
      "train_runtime": 78167.2235,
      "train_tokens_per_second": 29239.845
    },
    {
      "epoch": 0.28482531230166713,
      "grad_norm": 0.55859375,
      "learning_rate": 3.721717733925948e-05,
      "loss": 0.42447624206542967,
      "num_input_tokens_seen": 2288411904,
      "step": 8050,
      "train_runtime": 78262.5135,
      "train_tokens_per_second": 29240.205
    },
    {
      "epoch": 0.28517913256539595,
      "grad_norm": 0.59765625,
      "learning_rate": 3.720687158286904e-05,
      "loss": 0.42598509788513184,
      "num_input_tokens_seen": 2291227200,
      "step": 8060,
      "train_runtime": 78360.3018,
      "train_tokens_per_second": 29239.642
    },
    {
      "epoch": 0.2855329528291247,
      "grad_norm": 0.56640625,
      "learning_rate": 3.719657438299762e-05,
      "loss": 0.4228774070739746,
      "num_input_tokens_seen": 2294135296,
      "step": 8070,
      "train_runtime": 78461.2151,
      "train_tokens_per_second": 29239.1
    },
    {
      "epoch": 0.2858867730928535,
      "grad_norm": 0.57421875,
      "learning_rate": 3.7186285727811446e-05,
      "loss": 0.42594180107116697,
      "num_input_tokens_seen": 2296980672,
      "step": 8080,
      "train_runtime": 78556.6979,
      "train_tokens_per_second": 29239.781
    },
    {
      "epoch": 0.28624059335658225,
      "grad_norm": 0.57421875,
      "learning_rate": 3.717600560549967e-05,
      "loss": 0.42366781234741213,
      "num_input_tokens_seen": 2299847616,
      "step": 8090,
      "train_runtime": 78655.8803,
      "train_tokens_per_second": 29239.36
    },
    {
      "epoch": 0.2865944136203111,
      "grad_norm": 0.5625,
      "learning_rate": 3.716573400427426e-05,
      "loss": 0.42875213623046876,
      "num_input_tokens_seen": 2302721024,
      "step": 8100,
      "train_runtime": 78753.5738,
      "train_tokens_per_second": 29239.575
    },
    {
      "epoch": 0.28694823388403984,
      "grad_norm": 0.58203125,
      "learning_rate": 3.7155470912370004e-05,
      "loss": 0.42367944717407224,
      "num_input_tokens_seen": 2305532864,
      "step": 8110,
      "train_runtime": 78848.8859,
      "train_tokens_per_second": 29239.891
    },
    {
      "epoch": 0.2873020541477686,
      "grad_norm": 0.55859375,
      "learning_rate": 3.714521631804439e-05,
      "loss": 0.42185335159301757,
      "num_input_tokens_seen": 2308429184,
      "step": 8120,
      "train_runtime": 78950.3109,
      "train_tokens_per_second": 29239.013
    },
    {
      "epoch": 0.28765587441149737,
      "grad_norm": 0.546875,
      "learning_rate": 3.713497020957759e-05,
      "loss": 0.42450857162475586,
      "num_input_tokens_seen": 2311290368,
      "step": 8130,
      "train_runtime": 79048.0564,
      "train_tokens_per_second": 29239.054
    },
    {
      "epoch": 0.2880096946752262,
      "grad_norm": 0.546875,
      "learning_rate": 3.712473257527238e-05,
      "loss": 0.42467265129089354,
      "num_input_tokens_seen": 2314154880,
      "step": 8140,
      "train_runtime": 79144.97,
      "train_tokens_per_second": 29239.444
    },
    {
      "epoch": 0.28836351493895496,
      "grad_norm": 0.5625,
      "learning_rate": 3.711450340345412e-05,
      "loss": 0.4224575042724609,
      "num_input_tokens_seen": 2316930176,
      "step": 8150,
      "train_runtime": 79238.0483,
      "train_tokens_per_second": 29240.122
    },
    {
      "epoch": 0.2887173352026837,
      "grad_norm": 0.5625,
      "learning_rate": 3.710428268247067e-05,
      "loss": 0.4255655288696289,
      "num_input_tokens_seen": 2319758464,
      "step": 8160,
      "train_runtime": 79333.6058,
      "train_tokens_per_second": 29240.552
    },
    {
      "epoch": 0.2890711554664125,
      "grad_norm": 0.55078125,
      "learning_rate": 3.709407040069233e-05,
      "loss": 0.42122840881347656,
      "num_input_tokens_seen": 2322607680,
      "step": 8170,
      "train_runtime": 79428.5859,
      "train_tokens_per_second": 29241.458
    },
    {
      "epoch": 0.2894249757301413,
      "grad_norm": 0.5625,
      "learning_rate": 3.708386654651179e-05,
      "loss": 0.4254447460174561,
      "num_input_tokens_seen": 2325455936,
      "step": 8180,
      "train_runtime": 79524.0691,
      "train_tokens_per_second": 29242.165
    },
    {
      "epoch": 0.2897787959938701,
      "grad_norm": 0.59765625,
      "learning_rate": 3.707367110834409e-05,
      "loss": 0.42529659271240233,
      "num_input_tokens_seen": 2328385344,
      "step": 8190,
      "train_runtime": 79624.4541,
      "train_tokens_per_second": 29242.089
    },
    {
      "epoch": 0.29013261625759884,
      "grad_norm": 0.58203125,
      "learning_rate": 3.7063484074626555e-05,
      "loss": 0.42379035949707033,
      "num_input_tokens_seen": 2331216704,
      "step": 8200,
      "train_runtime": 79720.2646,
      "train_tokens_per_second": 29242.461
    },
    {
      "epoch": 0.2904864365213276,
      "grad_norm": 0.5703125,
      "learning_rate": 3.7053305433818725e-05,
      "loss": 0.419068717956543,
      "num_input_tokens_seen": 2334031296,
      "step": 8210,
      "train_runtime": 79815.4855,
      "train_tokens_per_second": 29242.838
    },
    {
      "epoch": 0.29084025678505643,
      "grad_norm": 0.58984375,
      "learning_rate": 3.704313517440232e-05,
      "loss": 0.4278595447540283,
      "num_input_tokens_seen": 2336879680,
      "step": 8220,
      "train_runtime": 79912.3628,
      "train_tokens_per_second": 29243.031
    },
    {
      "epoch": 0.2911940770487852,
      "grad_norm": 0.5703125,
      "learning_rate": 3.703297328488118e-05,
      "loss": 0.4246028423309326,
      "num_input_tokens_seen": 2339704960,
      "step": 8230,
      "train_runtime": 80007.126,
      "train_tokens_per_second": 29243.707
    },
    {
      "epoch": 0.29154789731251396,
      "grad_norm": 0.57421875,
      "learning_rate": 3.70228197537812e-05,
      "loss": 0.4211237907409668,
      "num_input_tokens_seen": 2342555712,
      "step": 8240,
      "train_runtime": 80105.1256,
      "train_tokens_per_second": 29243.518
    },
    {
      "epoch": 0.2919017175762427,
      "grad_norm": 0.57421875,
      "learning_rate": 3.7012674569650305e-05,
      "loss": 0.4255837917327881,
      "num_input_tokens_seen": 2345399232,
      "step": 8250,
      "train_runtime": 80202.132,
      "train_tokens_per_second": 29243.602
    },
    {
      "epoch": 0.2922555378399715,
      "grad_norm": 0.578125,
      "learning_rate": 3.700253772105835e-05,
      "loss": 0.4239809036254883,
      "num_input_tokens_seen": 2348243328,
      "step": 8260,
      "train_runtime": 80300.1518,
      "train_tokens_per_second": 29243.324
    },
    {
      "epoch": 0.2926093581037003,
      "grad_norm": 0.578125,
      "learning_rate": 3.699240919659711e-05,
      "loss": 0.42288665771484374,
      "num_input_tokens_seen": 2351143808,
      "step": 8270,
      "train_runtime": 80400.6049,
      "train_tokens_per_second": 29242.862
    },
    {
      "epoch": 0.2929631783674291,
      "grad_norm": 0.55078125,
      "learning_rate": 3.698228898488019e-05,
      "loss": 0.42441229820251464,
      "num_input_tokens_seen": 2354002048,
      "step": 8280,
      "train_runtime": 80498.8028,
      "train_tokens_per_second": 29242.696
    },
    {
      "epoch": 0.29331699863115784,
      "grad_norm": 0.54296875,
      "learning_rate": 3.6972177074543e-05,
      "loss": 0.4221537113189697,
      "num_input_tokens_seen": 2356863744,
      "step": 8290,
      "train_runtime": 80595.8715,
      "train_tokens_per_second": 29242.984
    },
    {
      "epoch": 0.2936708188948866,
      "grad_norm": 0.5703125,
      "learning_rate": 3.69620734542427e-05,
      "loss": 0.4187718391418457,
      "num_input_tokens_seen": 2359627648,
      "step": 8300,
      "train_runtime": 80690.6118,
      "train_tokens_per_second": 29242.902
    },
    {
      "epoch": 0.29402463915861543,
      "grad_norm": 0.5859375,
      "learning_rate": 3.695197811265811e-05,
      "loss": 0.4208488941192627,
      "num_input_tokens_seen": 2362501568,
      "step": 8310,
      "train_runtime": 80787.3404,
      "train_tokens_per_second": 29243.463
    },
    {
      "epoch": 0.2943784594223442,
      "grad_norm": 0.5625,
      "learning_rate": 3.6941891038489694e-05,
      "loss": 0.415517520904541,
      "num_input_tokens_seen": 2365346752,
      "step": 8320,
      "train_runtime": 80885.1793,
      "train_tokens_per_second": 29243.265
    },
    {
      "epoch": 0.29473227968607296,
      "grad_norm": 0.5546875,
      "learning_rate": 3.693181222045952e-05,
      "loss": 0.4225395679473877,
      "num_input_tokens_seen": 2368080704,
      "step": 8330,
      "train_runtime": 80974.9028,
      "train_tokens_per_second": 29244.625
    },
    {
      "epoch": 0.29508609994980173,
      "grad_norm": 0.56640625,
      "learning_rate": 3.692174164731113e-05,
      "loss": 0.42214412689208985,
      "num_input_tokens_seen": 2370936576,
      "step": 8340,
      "train_runtime": 81073.5741,
      "train_tokens_per_second": 29244.259
    },
    {
      "epoch": 0.29543992021353055,
      "grad_norm": 0.5625,
      "learning_rate": 3.6911679307809595e-05,
      "loss": 0.42295150756835936,
      "num_input_tokens_seen": 2373767424,
      "step": 8350,
      "train_runtime": 81169.8705,
      "train_tokens_per_second": 29244.44
    },
    {
      "epoch": 0.2957937404772593,
      "grad_norm": 0.5625,
      "learning_rate": 3.690162519074137e-05,
      "loss": 0.4286691665649414,
      "num_input_tokens_seen": 2376621184,
      "step": 8360,
      "train_runtime": 81267.2114,
      "train_tokens_per_second": 29244.527
    },
    {
      "epoch": 0.2961475607409881,
      "grad_norm": 0.5703125,
      "learning_rate": 3.689157928491431e-05,
      "loss": 0.4271433353424072,
      "num_input_tokens_seen": 2379502336,
      "step": 8370,
      "train_runtime": 81368.4083,
      "train_tokens_per_second": 29243.565
    },
    {
      "epoch": 0.29650138100471685,
      "grad_norm": 0.55859375,
      "learning_rate": 3.6881541579157566e-05,
      "loss": 0.42048940658569334,
      "num_input_tokens_seen": 2382356224,
      "step": 8380,
      "train_runtime": 81464.17,
      "train_tokens_per_second": 29244.221
    },
    {
      "epoch": 0.29685520126844567,
      "grad_norm": 0.57421875,
      "learning_rate": 3.687151206232154e-05,
      "loss": 0.4218634605407715,
      "num_input_tokens_seen": 2385200640,
      "step": 8390,
      "train_runtime": 81560.8545,
      "train_tokens_per_second": 29244.429
    },
    {
      "epoch": 0.29720902153217443,
      "grad_norm": 0.58984375,
      "learning_rate": 3.686149072327788e-05,
      "loss": 0.4210012435913086,
      "num_input_tokens_seen": 2388069312,
      "step": 8400,
      "train_runtime": 81663.6296,
      "train_tokens_per_second": 29242.753
    },
    {
      "epoch": 0.2975628417959032,
      "grad_norm": 0.59375,
      "learning_rate": 3.685147755091937e-05,
      "loss": 0.42243032455444335,
      "num_input_tokens_seen": 2390887552,
      "step": 8410,
      "train_runtime": 81757.8026,
      "train_tokens_per_second": 29243.54
    },
    {
      "epoch": 0.29791666205963196,
      "grad_norm": 0.56640625,
      "learning_rate": 3.684147253415992e-05,
      "loss": 0.4245555877685547,
      "num_input_tokens_seen": 2393791232,
      "step": 8420,
      "train_runtime": 81858.638,
      "train_tokens_per_second": 29242.989
    },
    {
      "epoch": 0.2982704823233608,
      "grad_norm": 0.609375,
      "learning_rate": 3.683147566193448e-05,
      "loss": 0.42319154739379883,
      "num_input_tokens_seen": 2396629440,
      "step": 8430,
      "train_runtime": 81956.1035,
      "train_tokens_per_second": 29242.843
    },
    {
      "epoch": 0.29862430258708955,
      "grad_norm": 0.55078125,
      "learning_rate": 3.6821486923199025e-05,
      "loss": 0.41884260177612304,
      "num_input_tokens_seen": 2399512128,
      "step": 8440,
      "train_runtime": 82052.7973,
      "train_tokens_per_second": 29243.514
    },
    {
      "epoch": 0.2989781228508183,
      "grad_norm": 0.55859375,
      "learning_rate": 3.681150630693046e-05,
      "loss": 0.42299470901489256,
      "num_input_tokens_seen": 2402339072,
      "step": 8450,
      "train_runtime": 82147.952,
      "train_tokens_per_second": 29244.053
    },
    {
      "epoch": 0.2993319431145471,
      "grad_norm": 0.57421875,
      "learning_rate": 3.6801533802126615e-05,
      "loss": 0.4188076972961426,
      "num_input_tokens_seen": 2405264896,
      "step": 8460,
      "train_runtime": 82247.3543,
      "train_tokens_per_second": 29244.283
    },
    {
      "epoch": 0.2996857633782759,
      "grad_norm": 0.58984375,
      "learning_rate": 3.679156939780617e-05,
      "loss": 0.4195985794067383,
      "num_input_tokens_seen": 2408088704,
      "step": 8470,
      "train_runtime": 82343.2734,
      "train_tokens_per_second": 29244.51
    },
    {
      "epoch": 0.30003958364200467,
      "grad_norm": 0.5703125,
      "learning_rate": 3.6781613083008594e-05,
      "loss": 0.42058448791503905,
      "num_input_tokens_seen": 2410963584,
      "step": 8480,
      "train_runtime": 82443.0464,
      "train_tokens_per_second": 29243.99
    },
    {
      "epoch": 0.30039340390573344,
      "grad_norm": 0.57421875,
      "learning_rate": 3.677166484679412e-05,
      "loss": 0.42074432373046877,
      "num_input_tokens_seen": 2413775104,
      "step": 8490,
      "train_runtime": 82542.0792,
      "train_tokens_per_second": 29242.965
    },
    {
      "epoch": 0.3007472241694622,
      "grad_norm": 0.56640625,
      "learning_rate": 3.676172467824368e-05,
      "loss": 0.4243292808532715,
      "num_input_tokens_seen": 2416579520,
      "step": 8500,
      "train_runtime": 82637.1342,
      "train_tokens_per_second": 29243.264
    },
    {
      "epoch": 0.30110104443319097,
      "grad_norm": 0.55859375,
      "learning_rate": 3.675179256645885e-05,
      "loss": 0.42053852081298826,
      "num_input_tokens_seen": 2419437504,
      "step": 8510,
      "train_runtime": 82734.1879,
      "train_tokens_per_second": 29243.503
    },
    {
      "epoch": 0.3014548646969198,
      "grad_norm": 0.56640625,
      "learning_rate": 3.674186850056181e-05,
      "loss": 0.42450647354125975,
      "num_input_tokens_seen": 2422266048,
      "step": 8520,
      "train_runtime": 82830.6654,
      "train_tokens_per_second": 29243.59
    },
    {
      "epoch": 0.30180868496064855,
      "grad_norm": 0.59375,
      "learning_rate": 3.67319524696953e-05,
      "loss": 0.42336201667785645,
      "num_input_tokens_seen": 2425138624,
      "step": 8530,
      "train_runtime": 82928.0559,
      "train_tokens_per_second": 29243.886
    },
    {
      "epoch": 0.3021625052243773,
      "grad_norm": 0.609375,
      "learning_rate": 3.6722044463022536e-05,
      "loss": 0.422471809387207,
      "num_input_tokens_seen": 2427948672,
      "step": 8540,
      "train_runtime": 83023.8189,
      "train_tokens_per_second": 29244.001
    },
    {
      "epoch": 0.3025163254881061,
      "grad_norm": 0.55859375,
      "learning_rate": 3.6712144469727214e-05,
      "loss": 0.4284646987915039,
      "num_input_tokens_seen": 2430787840,
      "step": 8550,
      "train_runtime": 83120.4144,
      "train_tokens_per_second": 29244.174
    },
    {
      "epoch": 0.3028701457518349,
      "grad_norm": 0.578125,
      "learning_rate": 3.67022524790134e-05,
      "loss": 0.42098326683044435,
      "num_input_tokens_seen": 2433624896,
      "step": 8560,
      "train_runtime": 83218.6414,
      "train_tokens_per_second": 29243.747
    },
    {
      "epoch": 0.3032239660155637,
      "grad_norm": 0.5546875,
      "learning_rate": 3.6692368480105546e-05,
      "loss": 0.4222285270690918,
      "num_input_tokens_seen": 2436540992,
      "step": 8570,
      "train_runtime": 83322.6143,
      "train_tokens_per_second": 29242.253
    },
    {
      "epoch": 0.30357778627929244,
      "grad_norm": 0.59375,
      "learning_rate": 3.6682492462248374e-05,
      "loss": 0.42070865631103516,
      "num_input_tokens_seen": 2439352320,
      "step": 8580,
      "train_runtime": 83418.0476,
      "train_tokens_per_second": 29242.501
    },
    {
      "epoch": 0.3039316065430212,
      "grad_norm": 0.58203125,
      "learning_rate": 3.667262441470689e-05,
      "loss": 0.41478118896484373,
      "num_input_tokens_seen": 2442149504,
      "step": 8590,
      "train_runtime": 83513.4444,
      "train_tokens_per_second": 29242.591
    },
    {
      "epoch": 0.30428542680675,
      "grad_norm": 0.5703125,
      "learning_rate": 3.6662764326766255e-05,
      "loss": 0.4226981163024902,
      "num_input_tokens_seen": 2445011904,
      "step": 8600,
      "train_runtime": 83609.2664,
      "train_tokens_per_second": 29243.312
    },
    {
      "epoch": 0.3046392470704788,
      "grad_norm": 0.5703125,
      "learning_rate": 3.665291218773185e-05,
      "loss": 0.4245340347290039,
      "num_input_tokens_seen": 2447813440,
      "step": 8610,
      "train_runtime": 83705.7213,
      "train_tokens_per_second": 29243.084
    },
    {
      "epoch": 0.30499306733420756,
      "grad_norm": 0.5859375,
      "learning_rate": 3.664306798692912e-05,
      "loss": 0.4257809638977051,
      "num_input_tokens_seen": 2450714688,
      "step": 8620,
      "train_runtime": 83805.2216,
      "train_tokens_per_second": 29242.983
    },
    {
      "epoch": 0.3053468875979363,
      "grad_norm": 0.5859375,
      "learning_rate": 3.6633231713703576e-05,
      "loss": 0.420089054107666,
      "num_input_tokens_seen": 2453575680,
      "step": 8630,
      "train_runtime": 83900.0412,
      "train_tokens_per_second": 29244.034
    },
    {
      "epoch": 0.30570070786166514,
      "grad_norm": 0.54296875,
      "learning_rate": 3.6623403357420745e-05,
      "loss": 0.41765193939208983,
      "num_input_tokens_seen": 2456404864,
      "step": 8640,
      "train_runtime": 83996.1649,
      "train_tokens_per_second": 29244.25
    },
    {
      "epoch": 0.3060545281253939,
      "grad_norm": 0.5859375,
      "learning_rate": 3.661358290746611e-05,
      "loss": 0.4180574893951416,
      "num_input_tokens_seen": 2459235200,
      "step": 8650,
      "train_runtime": 84090.3856,
      "train_tokens_per_second": 29245.141
    },
    {
      "epoch": 0.3064083483891227,
      "grad_norm": 0.55859375,
      "learning_rate": 3.6603770353245056e-05,
      "loss": 0.4231413841247559,
      "num_input_tokens_seen": 2462055168,
      "step": 8660,
      "train_runtime": 84186.2322,
      "train_tokens_per_second": 29245.342
    },
    {
      "epoch": 0.30676216865285144,
      "grad_norm": 0.61328125,
      "learning_rate": 3.659396568418286e-05,
      "loss": 0.4267998695373535,
      "num_input_tokens_seen": 2464862528,
      "step": 8670,
      "train_runtime": 84281.5514,
      "train_tokens_per_second": 29245.576
    },
    {
      "epoch": 0.30711598891658026,
      "grad_norm": 0.546875,
      "learning_rate": 3.658416888972459e-05,
      "loss": 0.42171216011047363,
      "num_input_tokens_seen": 2467719744,
      "step": 8680,
      "train_runtime": 84382.1167,
      "train_tokens_per_second": 29244.582
    },
    {
      "epoch": 0.30746980918030903,
      "grad_norm": 0.55078125,
      "learning_rate": 3.6574379959335106e-05,
      "loss": 0.42200021743774413,
      "num_input_tokens_seen": 2470527360,
      "step": 8690,
      "train_runtime": 84479.4688,
      "train_tokens_per_second": 29244.116
    },
    {
      "epoch": 0.3078236294440378,
      "grad_norm": 0.578125,
      "learning_rate": 3.6564598882498976e-05,
      "loss": 0.4245904922485352,
      "num_input_tokens_seen": 2473384512,
      "step": 8700,
      "train_runtime": 84576.925,
      "train_tokens_per_second": 29244.2
    },
    {
      "epoch": 0.30817744970776656,
      "grad_norm": 0.578125,
      "learning_rate": 3.655482564872043e-05,
      "loss": 0.4238411903381348,
      "num_input_tokens_seen": 2476265920,
      "step": 8710,
      "train_runtime": 84677.5276,
      "train_tokens_per_second": 29243.484
    },
    {
      "epoch": 0.3085312699714954,
      "grad_norm": 0.5859375,
      "learning_rate": 3.654506024752336e-05,
      "loss": 0.4213128566741943,
      "num_input_tokens_seen": 2479082048,
      "step": 8720,
      "train_runtime": 84774.4262,
      "train_tokens_per_second": 29243.277
    },
    {
      "epoch": 0.30888509023522415,
      "grad_norm": 0.5703125,
      "learning_rate": 3.653530266845121e-05,
      "loss": 0.4259797096252441,
      "num_input_tokens_seen": 2481906176,
      "step": 8730,
      "train_runtime": 84873.2286,
      "train_tokens_per_second": 29242.509
    },
    {
      "epoch": 0.3092389104989529,
      "grad_norm": 0.58203125,
      "learning_rate": 3.652555290106696e-05,
      "loss": 0.41959848403930666,
      "num_input_tokens_seen": 2484805632,
      "step": 8740,
      "train_runtime": 84974.4492,
      "train_tokens_per_second": 29241.797
    },
    {
      "epoch": 0.3095927307626817,
      "grad_norm": 0.6015625,
      "learning_rate": 3.6515810934953084e-05,
      "loss": 0.425579833984375,
      "num_input_tokens_seen": 2487667264,
      "step": 8750,
      "train_runtime": 85073.1587,
      "train_tokens_per_second": 29241.506
    },
    {
      "epoch": 0.30994655102641044,
      "grad_norm": 0.58984375,
      "learning_rate": 3.650607675971151e-05,
      "loss": 0.4237998962402344,
      "num_input_tokens_seen": 2490510976,
      "step": 8760,
      "train_runtime": 85169.1173,
      "train_tokens_per_second": 29241.949
    },
    {
      "epoch": 0.31030037129013927,
      "grad_norm": 0.5859375,
      "learning_rate": 3.649635036496351e-05,
      "loss": 0.42565183639526366,
      "num_input_tokens_seen": 2493334080,
      "step": 8770,
      "train_runtime": 85265.7504,
      "train_tokens_per_second": 29241.918
    },
    {
      "epoch": 0.31065419155386803,
      "grad_norm": 0.55859375,
      "learning_rate": 3.6486631740349746e-05,
      "loss": 0.420766544342041,
      "num_input_tokens_seen": 2496189760,
      "step": 8780,
      "train_runtime": 85364.0648,
      "train_tokens_per_second": 29241.693
    },
    {
      "epoch": 0.3110080118175968,
      "grad_norm": 0.5703125,
      "learning_rate": 3.647692087553018e-05,
      "loss": 0.4212955474853516,
      "num_input_tokens_seen": 2499008640,
      "step": 8790,
      "train_runtime": 85460.6011,
      "train_tokens_per_second": 29241.646
    },
    {
      "epoch": 0.31136183208132556,
      "grad_norm": 0.55859375,
      "learning_rate": 3.6467217760184005e-05,
      "loss": 0.41483302116394044,
      "num_input_tokens_seen": 2501886848,
      "step": 8800,
      "train_runtime": 85559.127,
      "train_tokens_per_second": 29241.613
    },
    {
      "epoch": 0.3117156523450544,
      "grad_norm": 0.5546875,
      "learning_rate": 3.6457522384009625e-05,
      "loss": 0.42199859619140623,
      "num_input_tokens_seen": 2504794944,
      "step": 8810,
      "train_runtime": 85659.5596,
      "train_tokens_per_second": 29241.277
    },
    {
      "epoch": 0.31206947260878315,
      "grad_norm": 0.59765625,
      "learning_rate": 3.644783473672462e-05,
      "loss": 0.4206065654754639,
      "num_input_tokens_seen": 2507562112,
      "step": 8820,
      "train_runtime": 85750.3248,
      "train_tokens_per_second": 29242.596
    },
    {
      "epoch": 0.3124232928725119,
      "grad_norm": 0.5625,
      "learning_rate": 3.643815480806568e-05,
      "loss": 0.42336406707763674,
      "num_input_tokens_seen": 2510362944,
      "step": 8830,
      "train_runtime": 85844.9471,
      "train_tokens_per_second": 29242.99
    },
    {
      "epoch": 0.3127771131362407,
      "grad_norm": 0.578125,
      "learning_rate": 3.6428482587788555e-05,
      "loss": 0.42230544090270994,
      "num_input_tokens_seen": 2513173952,
      "step": 8840,
      "train_runtime": 85941.2457,
      "train_tokens_per_second": 29242.931
    },
    {
      "epoch": 0.3131309333999695,
      "grad_norm": 0.58203125,
      "learning_rate": 3.641881806566803e-05,
      "loss": 0.42572808265686035,
      "num_input_tokens_seen": 2515977600,
      "step": 8850,
      "train_runtime": 86036.4721,
      "train_tokens_per_second": 29243.152
    },
    {
      "epoch": 0.31348475366369827,
      "grad_norm": 0.5703125,
      "learning_rate": 3.640916123149788e-05,
      "loss": 0.4210675239562988,
      "num_input_tokens_seen": 2518817344,
      "step": 8860,
      "train_runtime": 86130.2049,
      "train_tokens_per_second": 29244.298
    },
    {
      "epoch": 0.31383857392742703,
      "grad_norm": 0.5703125,
      "learning_rate": 3.639951207509079e-05,
      "loss": 0.42089319229125977,
      "num_input_tokens_seen": 2521707904,
      "step": 8870,
      "train_runtime": 86228.309,
      "train_tokens_per_second": 29244.548
    },
    {
      "epoch": 0.3141923941911558,
      "grad_norm": 0.57421875,
      "learning_rate": 3.6389870586278333e-05,
      "loss": 0.4201314926147461,
      "num_input_tokens_seen": 2524563456,
      "step": 8880,
      "train_runtime": 86329.0894,
      "train_tokens_per_second": 29243.485
    },
    {
      "epoch": 0.3145462144548846,
      "grad_norm": 0.56640625,
      "learning_rate": 3.6380236754910965e-05,
      "loss": 0.4212482452392578,
      "num_input_tokens_seen": 2527402176,
      "step": 8890,
      "train_runtime": 86426.2919,
      "train_tokens_per_second": 29243.441
    },
    {
      "epoch": 0.3149000347186134,
      "grad_norm": 0.55859375,
      "learning_rate": 3.6370610570857897e-05,
      "loss": 0.4239944934844971,
      "num_input_tokens_seen": 2530241984,
      "step": 8900,
      "train_runtime": 86522.0646,
      "train_tokens_per_second": 29243.893
    },
    {
      "epoch": 0.31525385498234215,
      "grad_norm": 0.59765625,
      "learning_rate": 3.6360992024007114e-05,
      "loss": 0.4248338222503662,
      "num_input_tokens_seen": 2533102720,
      "step": 8910,
      "train_runtime": 86618.7894,
      "train_tokens_per_second": 29244.264
    },
    {
      "epoch": 0.3156076752460709,
      "grad_norm": 0.56640625,
      "learning_rate": 3.6351381104265304e-05,
      "loss": 0.4176307678222656,
      "num_input_tokens_seen": 2535987776,
      "step": 8920,
      "train_runtime": 86721.3802,
      "train_tokens_per_second": 29242.936
    },
    {
      "epoch": 0.31596149550979974,
      "grad_norm": 0.5859375,
      "learning_rate": 3.634177780155783e-05,
      "loss": 0.42096424102783203,
      "num_input_tokens_seen": 2538791680,
      "step": 8930,
      "train_runtime": 86816.1753,
      "train_tokens_per_second": 29243.303
    },
    {
      "epoch": 0.3163153157735285,
      "grad_norm": 0.57421875,
      "learning_rate": 3.633218210582867e-05,
      "loss": 0.42520437240600584,
      "num_input_tokens_seen": 2541619776,
      "step": 8940,
      "train_runtime": 86910.3608,
      "train_tokens_per_second": 29244.152
    },
    {
      "epoch": 0.31666913603725727,
      "grad_norm": 0.5703125,
      "learning_rate": 3.6322594007040376e-05,
      "loss": 0.4216555118560791,
      "num_input_tokens_seen": 2544407552,
      "step": 8950,
      "train_runtime": 87002.4994,
      "train_tokens_per_second": 29245.224
    },
    {
      "epoch": 0.31702295630098604,
      "grad_norm": 0.5546875,
      "learning_rate": 3.631301349517403e-05,
      "loss": 0.41985092163085935,
      "num_input_tokens_seen": 2547242816,
      "step": 8960,
      "train_runtime": 87099.3623,
      "train_tokens_per_second": 29245.252
    },
    {
      "epoch": 0.31737677656471486,
      "grad_norm": 0.57421875,
      "learning_rate": 3.6303440560229216e-05,
      "loss": 0.42310395240783694,
      "num_input_tokens_seen": 2550115904,
      "step": 8970,
      "train_runtime": 87199.3646,
      "train_tokens_per_second": 29244.662
    },
    {
      "epoch": 0.3177305968284436,
      "grad_norm": 0.578125,
      "learning_rate": 3.629387519222395e-05,
      "loss": 0.41895322799682616,
      "num_input_tokens_seen": 2552975104,
      "step": 8980,
      "train_runtime": 87298.0183,
      "train_tokens_per_second": 29244.365
    },
    {
      "epoch": 0.3180844170921724,
      "grad_norm": 0.55859375,
      "learning_rate": 3.628431738119464e-05,
      "loss": 0.41408486366271974,
      "num_input_tokens_seen": 2555840832,
      "step": 8990,
      "train_runtime": 87397.4846,
      "train_tokens_per_second": 29243.872
    },
    {
      "epoch": 0.31843823735590115,
      "grad_norm": 0.5546875,
      "learning_rate": 3.62747671171961e-05,
      "loss": 0.42155799865722654,
      "num_input_tokens_seen": 2558715456,
      "step": 9000,
      "train_runtime": 87496.6752,
      "train_tokens_per_second": 29243.574
    },
    {
      "epoch": 0.3187920576196299,
      "grad_norm": 0.5703125,
      "learning_rate": 3.626522439030138e-05,
      "loss": 0.4224367618560791,
      "num_input_tokens_seen": 2561515072,
      "step": 9010,
      "train_runtime": 87591.7546,
      "train_tokens_per_second": 29243.792
    },
    {
      "epoch": 0.31914587788335874,
      "grad_norm": 0.578125,
      "learning_rate": 3.6255689190601863e-05,
      "loss": 0.42210540771484373,
      "num_input_tokens_seen": 2564361216,
      "step": 9020,
      "train_runtime": 87690.6392,
      "train_tokens_per_second": 29243.272
    },
    {
      "epoch": 0.3194996981470875,
      "grad_norm": 0.5546875,
      "learning_rate": 3.624616150820714e-05,
      "loss": 0.4253790855407715,
      "num_input_tokens_seen": 2567278592,
      "step": 9030,
      "train_runtime": 87790.0637,
      "train_tokens_per_second": 29243.385
    },
    {
      "epoch": 0.3198535184108163,
      "grad_norm": 0.5625,
      "learning_rate": 3.623664133324499e-05,
      "loss": 0.4170662879943848,
      "num_input_tokens_seen": 2570106688,
      "step": 9040,
      "train_runtime": 87887.6359,
      "train_tokens_per_second": 29243.097
    },
    {
      "epoch": 0.32020733867454504,
      "grad_norm": 0.57421875,
      "learning_rate": 3.622712865586131e-05,
      "loss": 0.4181173801422119,
      "num_input_tokens_seen": 2572995136,
      "step": 9050,
      "train_runtime": 87990.5347,
      "train_tokens_per_second": 29241.726
    },
    {
      "epoch": 0.32056115893827386,
      "grad_norm": 0.5625,
      "learning_rate": 3.621762346622014e-05,
      "loss": 0.4182613372802734,
      "num_input_tokens_seen": 2575895040,
      "step": 9060,
      "train_runtime": 88090.3509,
      "train_tokens_per_second": 29241.512
    },
    {
      "epoch": 0.3209149792020026,
      "grad_norm": 0.57421875,
      "learning_rate": 3.620812575450352e-05,
      "loss": 0.4163680553436279,
      "num_input_tokens_seen": 2578794624,
      "step": 9070,
      "train_runtime": 88187.2731,
      "train_tokens_per_second": 29242.254
    },
    {
      "epoch": 0.3212687994657314,
      "grad_norm": 0.57421875,
      "learning_rate": 3.6198635510911556e-05,
      "loss": 0.4187276840209961,
      "num_input_tokens_seen": 2581654976,
      "step": 9080,
      "train_runtime": 88282.4337,
      "train_tokens_per_second": 29243.133
    },
    {
      "epoch": 0.32162261972946016,
      "grad_norm": 0.55859375,
      "learning_rate": 3.618915272566228e-05,
      "loss": 0.414417028427124,
      "num_input_tokens_seen": 2584450176,
      "step": 9090,
      "train_runtime": 88374.169,
      "train_tokens_per_second": 29244.407
    },
    {
      "epoch": 0.321976439993189,
      "grad_norm": 0.55859375,
      "learning_rate": 3.6179677388991694e-05,
      "loss": 0.42461519241333007,
      "num_input_tokens_seen": 2587310016,
      "step": 9100,
      "train_runtime": 88471.3462,
      "train_tokens_per_second": 29244.61
    },
    {
      "epoch": 0.32233026025691774,
      "grad_norm": 0.57421875,
      "learning_rate": 3.617020949115366e-05,
      "loss": 0.4204258918762207,
      "num_input_tokens_seen": 2590123008,
      "step": 9110,
      "train_runtime": 88569.3179,
      "train_tokens_per_second": 29244.021
    },
    {
      "epoch": 0.3226840805206465,
      "grad_norm": 0.55078125,
      "learning_rate": 3.6160749022419886e-05,
      "loss": 0.41892280578613283,
      "num_input_tokens_seen": 2592997760,
      "step": 9120,
      "train_runtime": 88667.9342,
      "train_tokens_per_second": 29243.918
    },
    {
      "epoch": 0.3230379007843753,
      "grad_norm": 0.56640625,
      "learning_rate": 3.6151295973079887e-05,
      "loss": 0.41943182945251467,
      "num_input_tokens_seen": 2595862784,
      "step": 9130,
      "train_runtime": 88766.9948,
      "train_tokens_per_second": 29243.558
    },
    {
      "epoch": 0.3233917210481041,
      "grad_norm": 0.56640625,
      "learning_rate": 3.6141850333440934e-05,
      "loss": 0.42088727951049804,
      "num_input_tokens_seen": 2598674112,
      "step": 9140,
      "train_runtime": 88860.6608,
      "train_tokens_per_second": 29244.371
    },
    {
      "epoch": 0.32374554131183286,
      "grad_norm": 0.5625,
      "learning_rate": 3.613241209382803e-05,
      "loss": 0.42317800521850585,
      "num_input_tokens_seen": 2601500736,
      "step": 9150,
      "train_runtime": 88957.5696,
      "train_tokens_per_second": 29244.287
    },
    {
      "epoch": 0.32409936157556163,
      "grad_norm": 0.56640625,
      "learning_rate": 3.6122981244583834e-05,
      "loss": 0.42010927200317383,
      "num_input_tokens_seen": 2604376704,
      "step": 9160,
      "train_runtime": 89057.4102,
      "train_tokens_per_second": 29243.796
    },
    {
      "epoch": 0.3244531818392904,
      "grad_norm": 0.5546875,
      "learning_rate": 3.6113557776068644e-05,
      "loss": 0.42517943382263185,
      "num_input_tokens_seen": 2607175040,
      "step": 9170,
      "train_runtime": 89152.4463,
      "train_tokens_per_second": 29244.01
    },
    {
      "epoch": 0.3248070021030192,
      "grad_norm": 0.56640625,
      "learning_rate": 3.6104141678660386e-05,
      "loss": 0.4233957290649414,
      "num_input_tokens_seen": 2610071488,
      "step": 9180,
      "train_runtime": 89253.2289,
      "train_tokens_per_second": 29243.441
    },
    {
      "epoch": 0.325160822366748,
      "grad_norm": 0.58203125,
      "learning_rate": 3.6094732942754487e-05,
      "loss": 0.41938085556030275,
      "num_input_tokens_seen": 2612900480,
      "step": 9190,
      "train_runtime": 89350.3305,
      "train_tokens_per_second": 29243.322
    },
    {
      "epoch": 0.32551464263047675,
      "grad_norm": 0.578125,
      "learning_rate": 3.60853315587639e-05,
      "loss": 0.41816225051879885,
      "num_input_tokens_seen": 2615733952,
      "step": 9200,
      "train_runtime": 89446.7603,
      "train_tokens_per_second": 29243.473
    },
    {
      "epoch": 0.3258684628942055,
      "grad_norm": 0.55078125,
      "learning_rate": 3.607593751711909e-05,
      "loss": 0.41954669952392576,
      "num_input_tokens_seen": 2618562688,
      "step": 9210,
      "train_runtime": 89543.6609,
      "train_tokens_per_second": 29243.418
    },
    {
      "epoch": 0.32622228315793433,
      "grad_norm": 0.5703125,
      "learning_rate": 3.60665508082679e-05,
      "loss": 0.4246511459350586,
      "num_input_tokens_seen": 2621396928,
      "step": 9220,
      "train_runtime": 89641.0703,
      "train_tokens_per_second": 29243.258
    },
    {
      "epoch": 0.3265761034216631,
      "grad_norm": 0.57421875,
      "learning_rate": 3.6057171422675585e-05,
      "loss": 0.42350406646728517,
      "num_input_tokens_seen": 2624207232,
      "step": 9230,
      "train_runtime": 89733.9363,
      "train_tokens_per_second": 29244.312
    },
    {
      "epoch": 0.32692992368539187,
      "grad_norm": 0.578125,
      "learning_rate": 3.604779935082474e-05,
      "loss": 0.41873812675476074,
      "num_input_tokens_seen": 2627099328,
      "step": 9240,
      "train_runtime": 89833.7127,
      "train_tokens_per_second": 29244.025
    },
    {
      "epoch": 0.32728374394912063,
      "grad_norm": 0.58203125,
      "learning_rate": 3.603843458321526e-05,
      "loss": 0.4257979393005371,
      "num_input_tokens_seen": 2629939328,
      "step": 9250,
      "train_runtime": 89929.7697,
      "train_tokens_per_second": 29244.369
    },
    {
      "epoch": 0.32763756421284945,
      "grad_norm": 0.6328125,
      "learning_rate": 3.6029077110364355e-05,
      "loss": 0.42142295837402344,
      "num_input_tokens_seen": 2632770752,
      "step": 9260,
      "train_runtime": 90022.0638,
      "train_tokens_per_second": 29245.839
    },
    {
      "epoch": 0.3279913844765782,
      "grad_norm": 0.5859375,
      "learning_rate": 3.60197269228064e-05,
      "loss": 0.42386627197265625,
      "num_input_tokens_seen": 2635556544,
      "step": 9270,
      "train_runtime": 90116.2517,
      "train_tokens_per_second": 29246.185
    },
    {
      "epoch": 0.328345204740307,
      "grad_norm": 0.59765625,
      "learning_rate": 3.601038401109299e-05,
      "loss": 0.4232340812683105,
      "num_input_tokens_seen": 2638380544,
      "step": 9280,
      "train_runtime": 90210.204,
      "train_tokens_per_second": 29247.03
    },
    {
      "epoch": 0.32869902500403575,
      "grad_norm": 0.546875,
      "learning_rate": 3.6001048365792846e-05,
      "loss": 0.4203153133392334,
      "num_input_tokens_seen": 2641256448,
      "step": 9290,
      "train_runtime": 90309.1023,
      "train_tokens_per_second": 29246.846
    },
    {
      "epoch": 0.3290528452677645,
      "grad_norm": 0.57421875,
      "learning_rate": 3.599171997749182e-05,
      "loss": 0.4207463264465332,
      "num_input_tokens_seen": 2644147904,
      "step": 9300,
      "train_runtime": 90408.1626,
      "train_tokens_per_second": 29246.783
    },
    {
      "epoch": 0.32940666553149334,
      "grad_norm": 0.57421875,
      "learning_rate": 3.598239883679281e-05,
      "loss": 0.4229604721069336,
      "num_input_tokens_seen": 2646989184,
      "step": 9310,
      "train_runtime": 90505.9811,
      "train_tokens_per_second": 29246.566
    },
    {
      "epoch": 0.3297604857952221,
      "grad_norm": 0.578125,
      "learning_rate": 3.597308493431576e-05,
      "loss": 0.4221336364746094,
      "num_input_tokens_seen": 2649807872,
      "step": 9320,
      "train_runtime": 90601.9413,
      "train_tokens_per_second": 29246.701
    },
    {
      "epoch": 0.33011430605895087,
      "grad_norm": 0.58203125,
      "learning_rate": 3.596377826069758e-05,
      "loss": 0.4223775863647461,
      "num_input_tokens_seen": 2652620224,
      "step": 9330,
      "train_runtime": 90698.3669,
      "train_tokens_per_second": 29246.615
    },
    {
      "epoch": 0.33046812632267963,
      "grad_norm": 0.58984375,
      "learning_rate": 3.5954478806592155e-05,
      "loss": 0.42059326171875,
      "num_input_tokens_seen": 2655494400,
      "step": 9340,
      "train_runtime": 90797.1484,
      "train_tokens_per_second": 29246.452
    },
    {
      "epoch": 0.33082194658640846,
      "grad_norm": 0.56640625,
      "learning_rate": 3.594518656267024e-05,
      "loss": 0.4238277912139893,
      "num_input_tokens_seen": 2658351552,
      "step": 9350,
      "train_runtime": 90895.6542,
      "train_tokens_per_second": 29246.19
    },
    {
      "epoch": 0.3311757668501372,
      "grad_norm": 0.56640625,
      "learning_rate": 3.5935901519619496e-05,
      "loss": 0.42474069595336916,
      "num_input_tokens_seen": 2661200768,
      "step": 9360,
      "train_runtime": 90989.6174,
      "train_tokens_per_second": 29247.301
    },
    {
      "epoch": 0.331529587113866,
      "grad_norm": 0.58203125,
      "learning_rate": 3.5926623668144385e-05,
      "loss": 0.41840200424194335,
      "num_input_tokens_seen": 2664068544,
      "step": 9370,
      "train_runtime": 91088.3122,
      "train_tokens_per_second": 29247.095
    },
    {
      "epoch": 0.33188340737759475,
      "grad_norm": 0.57421875,
      "learning_rate": 3.5917352998966194e-05,
      "loss": 0.42109098434448244,
      "num_input_tokens_seen": 2666974464,
      "step": 9380,
      "train_runtime": 91190.4753,
      "train_tokens_per_second": 29246.195
    },
    {
      "epoch": 0.3322372276413236,
      "grad_norm": 0.55078125,
      "learning_rate": 3.5908089502822914e-05,
      "loss": 0.420253324508667,
      "num_input_tokens_seen": 2669785344,
      "step": 9390,
      "train_runtime": 91287.8546,
      "train_tokens_per_second": 29245.789
    },
    {
      "epoch": 0.33259104790505234,
      "grad_norm": 0.56640625,
      "learning_rate": 3.589883317046929e-05,
      "loss": 0.42046523094177246,
      "num_input_tokens_seen": 2672659712,
      "step": 9400,
      "train_runtime": 91386.8261,
      "train_tokens_per_second": 29245.569
    },
    {
      "epoch": 0.3329448681687811,
      "grad_norm": 0.57421875,
      "learning_rate": 3.5889583992676715e-05,
      "loss": 0.41854071617126465,
      "num_input_tokens_seen": 2675633856,
      "step": 9410,
      "train_runtime": 91495.4341,
      "train_tokens_per_second": 29243.359
    },
    {
      "epoch": 0.33329868843250987,
      "grad_norm": 0.5703125,
      "learning_rate": 3.5880341960233244e-05,
      "loss": 0.4192509651184082,
      "num_input_tokens_seen": 2678472576,
      "step": 9420,
      "train_runtime": 91591.6497,
      "train_tokens_per_second": 29243.633
    },
    {
      "epoch": 0.3336525086962387,
      "grad_norm": 0.58203125,
      "learning_rate": 3.58711070639435e-05,
      "loss": 0.4218461513519287,
      "num_input_tokens_seen": 2681330496,
      "step": 9430,
      "train_runtime": 91690.4315,
      "train_tokens_per_second": 29243.297
    },
    {
      "epoch": 0.33400632895996746,
      "grad_norm": 0.56640625,
      "learning_rate": 3.586187929462869e-05,
      "loss": 0.41759138107299804,
      "num_input_tokens_seen": 2684185664,
      "step": 9440,
      "train_runtime": 91787.4542,
      "train_tokens_per_second": 29243.492
    },
    {
      "epoch": 0.3343601492236962,
      "grad_norm": 0.57421875,
      "learning_rate": 3.585265864312651e-05,
      "loss": 0.42147026062011717,
      "num_input_tokens_seen": 2687038656,
      "step": 9450,
      "train_runtime": 91886.4182,
      "train_tokens_per_second": 29243.045
    },
    {
      "epoch": 0.334713969487425,
      "grad_norm": 0.5859375,
      "learning_rate": 3.584344510029118e-05,
      "loss": 0.4218087673187256,
      "num_input_tokens_seen": 2689887424,
      "step": 9460,
      "train_runtime": 91989.1945,
      "train_tokens_per_second": 29241.341
    },
    {
      "epoch": 0.3350677897511538,
      "grad_norm": 0.57421875,
      "learning_rate": 3.583423865699333e-05,
      "loss": 0.41876797676086425,
      "num_input_tokens_seen": 2692676672,
      "step": 9470,
      "train_runtime": 92083.7605,
      "train_tokens_per_second": 29241.602
    },
    {
      "epoch": 0.3354216100148826,
      "grad_norm": 0.58203125,
      "learning_rate": 3.5825039304119994e-05,
      "loss": 0.41921253204345704,
      "num_input_tokens_seen": 2695477440,
      "step": 9480,
      "train_runtime": 92177.7653,
      "train_tokens_per_second": 29242.165
    },
    {
      "epoch": 0.33577543027861134,
      "grad_norm": 0.59765625,
      "learning_rate": 3.581584703257461e-05,
      "loss": 0.4214989185333252,
      "num_input_tokens_seen": 2698277824,
      "step": 9490,
      "train_runtime": 92272.7061,
      "train_tokens_per_second": 29242.426
    },
    {
      "epoch": 0.3361292505423401,
      "grad_norm": 0.58203125,
      "learning_rate": 3.580666183327689e-05,
      "loss": 0.4211748123168945,
      "num_input_tokens_seen": 2701066624,
      "step": 9500,
      "train_runtime": 92368.0489,
      "train_tokens_per_second": 29242.435
    },
    {
      "epoch": 0.33648307080606893,
      "grad_norm": 0.57421875,
      "learning_rate": 3.5797483697162906e-05,
      "loss": 0.4170060157775879,
      "num_input_tokens_seen": 2703923072,
      "step": 9510,
      "train_runtime": 92466.5316,
      "train_tokens_per_second": 29242.181
    },
    {
      "epoch": 0.3368368910697977,
      "grad_norm": 0.6015625,
      "learning_rate": 3.5788312615184936e-05,
      "loss": 0.42041454315185545,
      "num_input_tokens_seen": 2706712320,
      "step": 9520,
      "train_runtime": 92559.468,
      "train_tokens_per_second": 29242.955
    },
    {
      "epoch": 0.33719071133352646,
      "grad_norm": 0.57421875,
      "learning_rate": 3.5779148578311476e-05,
      "loss": 0.4195714950561523,
      "num_input_tokens_seen": 2709565248,
      "step": 9530,
      "train_runtime": 92657.9203,
      "train_tokens_per_second": 29242.673
    },
    {
      "epoch": 0.3375445315972552,
      "grad_norm": 0.578125,
      "learning_rate": 3.5769991577527236e-05,
      "loss": 0.4205338001251221,
      "num_input_tokens_seen": 2712431488,
      "step": 9540,
      "train_runtime": 92757.7818,
      "train_tokens_per_second": 29242.091
    },
    {
      "epoch": 0.337898351860984,
      "grad_norm": 0.57421875,
      "learning_rate": 3.5760841603833034e-05,
      "loss": 0.4251713752746582,
      "num_input_tokens_seen": 2715215744,
      "step": 9550,
      "train_runtime": 92851.7507,
      "train_tokens_per_second": 29242.483
    },
    {
      "epoch": 0.3382521721247128,
      "grad_norm": 0.5859375,
      "learning_rate": 3.5751698648245814e-05,
      "loss": 0.42135062217712405,
      "num_input_tokens_seen": 2718075904,
      "step": 9560,
      "train_runtime": 92949.1261,
      "train_tokens_per_second": 29242.619
    },
    {
      "epoch": 0.3386059923884416,
      "grad_norm": 0.55859375,
      "learning_rate": 3.574256270179857e-05,
      "loss": 0.4234762191772461,
      "num_input_tokens_seen": 2720922240,
      "step": 9570,
      "train_runtime": 93044.3913,
      "train_tokens_per_second": 29243.27
    },
    {
      "epoch": 0.33895981265217034,
      "grad_norm": 0.5625,
      "learning_rate": 3.573343375554037e-05,
      "loss": 0.4181545257568359,
      "num_input_tokens_seen": 2723745536,
      "step": 9580,
      "train_runtime": 93141.7502,
      "train_tokens_per_second": 29243.014
    },
    {
      "epoch": 0.3393136329158991,
      "grad_norm": 0.5625,
      "learning_rate": 3.572431180053621e-05,
      "loss": 0.4175090789794922,
      "num_input_tokens_seen": 2726585024,
      "step": 9590,
      "train_runtime": 93237.7146,
      "train_tokens_per_second": 29243.37
    },
    {
      "epoch": 0.33966745317962793,
      "grad_norm": 0.58203125,
      "learning_rate": 3.571519682786711e-05,
      "loss": 0.42274131774902346,
      "num_input_tokens_seen": 2729407232,
      "step": 9600,
      "train_runtime": 93332.8181,
      "train_tokens_per_second": 29243.81
    },
    {
      "epoch": 0.3400212734433567,
      "grad_norm": 0.58203125,
      "learning_rate": 3.570608882862996e-05,
      "loss": 0.4224428653717041,
      "num_input_tokens_seen": 2732339136,
      "step": 9610,
      "train_runtime": 93435.4146,
      "train_tokens_per_second": 29243.078
    },
    {
      "epoch": 0.34037509370708546,
      "grad_norm": 0.56640625,
      "learning_rate": 3.569698779393757e-05,
      "loss": 0.41882858276367185,
      "num_input_tokens_seen": 2735181952,
      "step": 9620,
      "train_runtime": 93532.1919,
      "train_tokens_per_second": 29243.215
    },
    {
      "epoch": 0.34072891397081423,
      "grad_norm": 0.59375,
      "learning_rate": 3.568789371491859e-05,
      "loss": 0.4197474479675293,
      "num_input_tokens_seen": 2738008576,
      "step": 9630,
      "train_runtime": 93626.9676,
      "train_tokens_per_second": 29243.803
    },
    {
      "epoch": 0.34108273423454305,
      "grad_norm": 0.5703125,
      "learning_rate": 3.567880658271748e-05,
      "loss": 0.4182888031005859,
      "num_input_tokens_seen": 2740840448,
      "step": 9640,
      "train_runtime": 93723.6938,
      "train_tokens_per_second": 29243.837
    },
    {
      "epoch": 0.3414365544982718,
      "grad_norm": 0.5859375,
      "learning_rate": 3.566972638849445e-05,
      "loss": 0.41817731857299806,
      "num_input_tokens_seen": 2743672448,
      "step": 9650,
      "train_runtime": 93821.0805,
      "train_tokens_per_second": 29243.667
    },
    {
      "epoch": 0.3417903747620006,
      "grad_norm": 0.58203125,
      "learning_rate": 3.566065312342551e-05,
      "loss": 0.423507022857666,
      "num_input_tokens_seen": 2746535680,
      "step": 9660,
      "train_runtime": 93919.0339,
      "train_tokens_per_second": 29243.653
    },
    {
      "epoch": 0.34214419502572935,
      "grad_norm": 0.56640625,
      "learning_rate": 3.565158677870231e-05,
      "loss": 0.41780638694763184,
      "num_input_tokens_seen": 2749366528,
      "step": 9670,
      "train_runtime": 94015.7725,
      "train_tokens_per_second": 29243.673
    },
    {
      "epoch": 0.34249801528945817,
      "grad_norm": 0.5546875,
      "learning_rate": 3.564252734553221e-05,
      "loss": 0.41854009628295896,
      "num_input_tokens_seen": 2752208512,
      "step": 9680,
      "train_runtime": 94113.4013,
      "train_tokens_per_second": 29243.535
    },
    {
      "epoch": 0.34285183555318693,
      "grad_norm": 0.5859375,
      "learning_rate": 3.563347481513818e-05,
      "loss": 0.4183374881744385,
      "num_input_tokens_seen": 2755070656,
      "step": 9690,
      "train_runtime": 94209.6232,
      "train_tokens_per_second": 29244.047
    },
    {
      "epoch": 0.3432056558169157,
      "grad_norm": 0.57421875,
      "learning_rate": 3.56244291787588e-05,
      "loss": 0.4214652061462402,
      "num_input_tokens_seen": 2757961664,
      "step": 9700,
      "train_runtime": 94309.9007,
      "train_tokens_per_second": 29243.607
    },
    {
      "epoch": 0.34355947608064447,
      "grad_norm": 0.5703125,
      "learning_rate": 3.5615390427648216e-05,
      "loss": 0.4174685478210449,
      "num_input_tokens_seen": 2760823872,
      "step": 9710,
      "train_runtime": 94408.1485,
      "train_tokens_per_second": 29243.491
    },
    {
      "epoch": 0.3439132963443733,
      "grad_norm": 0.5703125,
      "learning_rate": 3.5606358553076075e-05,
      "loss": 0.4220565319061279,
      "num_input_tokens_seen": 2763666752,
      "step": 9720,
      "train_runtime": 94504.633,
      "train_tokens_per_second": 29243.717
    },
    {
      "epoch": 0.34426711660810205,
      "grad_norm": 0.58203125,
      "learning_rate": 3.5597333546327526e-05,
      "loss": 0.41991047859191893,
      "num_input_tokens_seen": 2766528320,
      "step": 9730,
      "train_runtime": 94602.6884,
      "train_tokens_per_second": 29243.654
    },
    {
      "epoch": 0.3446209368718308,
      "grad_norm": 0.5703125,
      "learning_rate": 3.5588315398703186e-05,
      "loss": 0.42338099479675295,
      "num_input_tokens_seen": 2769368640,
      "step": 9740,
      "train_runtime": 94701.3428,
      "train_tokens_per_second": 29243.182
    },
    {
      "epoch": 0.3449747571355596,
      "grad_norm": 0.58984375,
      "learning_rate": 3.557930410151907e-05,
      "loss": 0.4177273750305176,
      "num_input_tokens_seen": 2772225664,
      "step": 9750,
      "train_runtime": 94799.6095,
      "train_tokens_per_second": 29243.007
    },
    {
      "epoch": 0.3453285773992884,
      "grad_norm": 0.59375,
      "learning_rate": 3.5570299646106606e-05,
      "loss": 0.42409625053405764,
      "num_input_tokens_seen": 2775066944,
      "step": 9760,
      "train_runtime": 94895.8299,
      "train_tokens_per_second": 29243.297
    },
    {
      "epoch": 0.34568239766301717,
      "grad_norm": 0.578125,
      "learning_rate": 3.556130202381253e-05,
      "loss": 0.4192185401916504,
      "num_input_tokens_seen": 2777946752,
      "step": 9770,
      "train_runtime": 94994.7965,
      "train_tokens_per_second": 29243.147
    },
    {
      "epoch": 0.34603621792674594,
      "grad_norm": 0.546875,
      "learning_rate": 3.555231122599892e-05,
      "loss": 0.42435202598571775,
      "num_input_tokens_seen": 2780791680,
      "step": 9780,
      "train_runtime": 95092.8774,
      "train_tokens_per_second": 29242.902
    },
    {
      "epoch": 0.3463900381904747,
      "grad_norm": 0.53515625,
      "learning_rate": 3.554332724404313e-05,
      "loss": 0.4149494171142578,
      "num_input_tokens_seen": 2783628736,
      "step": 9790,
      "train_runtime": 95187.0591,
      "train_tokens_per_second": 29243.773
    },
    {
      "epoch": 0.34674385845420347,
      "grad_norm": 0.56640625,
      "learning_rate": 3.553435006933777e-05,
      "loss": 0.418074893951416,
      "num_input_tokens_seen": 2786429696,
      "step": 9800,
      "train_runtime": 95279.7489,
      "train_tokens_per_second": 29244.721
    },
    {
      "epoch": 0.3470976787179323,
      "grad_norm": 0.57421875,
      "learning_rate": 3.5525379693290626e-05,
      "loss": 0.41746997833251953,
      "num_input_tokens_seen": 2789282624,
      "step": 9810,
      "train_runtime": 95379.2067,
      "train_tokens_per_second": 29244.137
    },
    {
      "epoch": 0.34745149898166106,
      "grad_norm": 0.5625,
      "learning_rate": 3.551641610732469e-05,
      "loss": 0.41968517303466796,
      "num_input_tokens_seen": 2792052992,
      "step": 9820,
      "train_runtime": 95470.6265,
      "train_tokens_per_second": 29245.152
    },
    {
      "epoch": 0.3478053192453898,
      "grad_norm": 0.56640625,
      "learning_rate": 3.55074593028781e-05,
      "loss": 0.4130645751953125,
      "num_input_tokens_seen": 2794906496,
      "step": 9830,
      "train_runtime": 95569.0365,
      "train_tokens_per_second": 29244.896
    },
    {
      "epoch": 0.3481591395091186,
      "grad_norm": 0.55859375,
      "learning_rate": 3.5498509271404065e-05,
      "loss": 0.417772102355957,
      "num_input_tokens_seen": 2797787136,
      "step": 9840,
      "train_runtime": 95667.64,
      "train_tokens_per_second": 29244.864
    },
    {
      "epoch": 0.3485129597728474,
      "grad_norm": 0.58203125,
      "learning_rate": 3.5489566004370893e-05,
      "loss": 0.42011446952819825,
      "num_input_tokens_seen": 2800681280,
      "step": 9850,
      "train_runtime": 95768.89,
      "train_tokens_per_second": 29244.166
    },
    {
      "epoch": 0.3488667800365762,
      "grad_norm": 0.578125,
      "learning_rate": 3.548062949326194e-05,
      "loss": 0.41987051963806155,
      "num_input_tokens_seen": 2803524800,
      "step": 9860,
      "train_runtime": 95864.6644,
      "train_tokens_per_second": 29244.611
    },
    {
      "epoch": 0.34922060030030494,
      "grad_norm": 0.58203125,
      "learning_rate": 3.547169972957554e-05,
      "loss": 0.41800456047058104,
      "num_input_tokens_seen": 2806324992,
      "step": 9870,
      "train_runtime": 95960.5651,
      "train_tokens_per_second": 29244.565
    },
    {
      "epoch": 0.3495744205640337,
      "grad_norm": 0.57421875,
      "learning_rate": 3.5462776704825e-05,
      "loss": 0.4154806137084961,
      "num_input_tokens_seen": 2809198208,
      "step": 9880,
      "train_runtime": 96057.7906,
      "train_tokens_per_second": 29244.876
    },
    {
      "epoch": 0.3499282408277625,
      "grad_norm": 0.56640625,
      "learning_rate": 3.5453860410538594e-05,
      "loss": 0.41528568267822263,
      "num_input_tokens_seen": 2812001408,
      "step": 9890,
      "train_runtime": 96152.2745,
      "train_tokens_per_second": 29245.293
    },
    {
      "epoch": 0.3502820610914913,
      "grad_norm": 0.56640625,
      "learning_rate": 3.5444950838259455e-05,
      "loss": 0.4201637268066406,
      "num_input_tokens_seen": 2814878784,
      "step": 9900,
      "train_runtime": 96251.1093,
      "train_tokens_per_second": 29245.157
    },
    {
      "epoch": 0.35063588135522006,
      "grad_norm": 0.5703125,
      "learning_rate": 3.543604797954563e-05,
      "loss": 0.41726150512695315,
      "num_input_tokens_seen": 2817671616,
      "step": 9910,
      "train_runtime": 96344.644,
      "train_tokens_per_second": 29245.753
    },
    {
      "epoch": 0.3509897016189488,
      "grad_norm": 0.56640625,
      "learning_rate": 3.542715182596996e-05,
      "loss": 0.4191440105438232,
      "num_input_tokens_seen": 2820527872,
      "step": 9920,
      "train_runtime": 96442.3233,
      "train_tokens_per_second": 29245.748
    },
    {
      "epoch": 0.35134352188267765,
      "grad_norm": 0.58984375,
      "learning_rate": 3.5418262369120115e-05,
      "loss": 0.41490631103515624,
      "num_input_tokens_seen": 2823372928,
      "step": 9930,
      "train_runtime": 96538.8388,
      "train_tokens_per_second": 29245.98
    },
    {
      "epoch": 0.3516973421464064,
      "grad_norm": 0.55859375,
      "learning_rate": 3.5409379600598526e-05,
      "loss": 0.4175734519958496,
      "num_input_tokens_seen": 2826218944,
      "step": 9940,
      "train_runtime": 96635.6595,
      "train_tokens_per_second": 29246.129
    },
    {
      "epoch": 0.3520511624101352,
      "grad_norm": 0.55078125,
      "learning_rate": 3.540050351202235e-05,
      "loss": 0.41604318618774416,
      "num_input_tokens_seen": 2829055168,
      "step": 9950,
      "train_runtime": 96731.1434,
      "train_tokens_per_second": 29246.58
    },
    {
      "epoch": 0.35240498267386394,
      "grad_norm": 0.57421875,
      "learning_rate": 3.539163409502347e-05,
      "loss": 0.4203804492950439,
      "num_input_tokens_seen": 2831964224,
      "step": 9960,
      "train_runtime": 96833.5116,
      "train_tokens_per_second": 29245.704
    },
    {
      "epoch": 0.35275880293759276,
      "grad_norm": 0.6015625,
      "learning_rate": 3.5382771341248416e-05,
      "loss": 0.41888136863708497,
      "num_input_tokens_seen": 2834815616,
      "step": 9970,
      "train_runtime": 96930.2101,
      "train_tokens_per_second": 29245.945
    },
    {
      "epoch": 0.35311262320132153,
      "grad_norm": 0.578125,
      "learning_rate": 3.537391524235835e-05,
      "loss": 0.4221469879150391,
      "num_input_tokens_seen": 2837725056,
      "step": 9980,
      "train_runtime": 97031.2377,
      "train_tokens_per_second": 29245.479
    },
    {
      "epoch": 0.3534664434650503,
      "grad_norm": 0.58203125,
      "learning_rate": 3.5365065790029055e-05,
      "loss": 0.4223515510559082,
      "num_input_tokens_seen": 2840564416,
      "step": 9990,
      "train_runtime": 97129.6719,
      "train_tokens_per_second": 29245.074
    },
    {
      "epoch": 0.35382026372877906,
      "grad_norm": 0.58984375,
      "learning_rate": 3.535622297595087e-05,
      "loss": 0.4175906181335449,
      "num_input_tokens_seen": 2843412288,
      "step": 10000,
      "train_runtime": 97228.6885,
      "train_tokens_per_second": 29244.581
    },
    {
      "epoch": 0.35382026372877906,
      "eval_loss": 0.34943389892578125,
      "eval_runtime": 8.4325,
      "eval_samples_per_second": 472.932,
      "eval_steps_per_second": 3.795,
      "num_input_tokens_seen": 2843412288,
      "step": 10000
    },
    {
      "epoch": 0.3541740839925079,
      "grad_norm": 0.5546875,
      "learning_rate": 3.534738679182869e-05,
      "loss": 0.4168993949890137,
      "num_input_tokens_seen": 2846261632,
      "step": 10010,
      "train_runtime": 97352.9319,
      "train_tokens_per_second": 29236.527
    },
    {
      "epoch": 0.35452790425623665,
      "grad_norm": 0.59765625,
      "learning_rate": 3.533855722938188e-05,
      "loss": 0.4144249439239502,
      "num_input_tokens_seen": 2849122752,
      "step": 10020,
      "train_runtime": 97450.0178,
      "train_tokens_per_second": 29236.76
    },
    {
      "epoch": 0.3548817245199654,
      "grad_norm": 0.5859375,
      "learning_rate": 3.5329734280344325e-05,
      "loss": 0.4126617908477783,
      "num_input_tokens_seen": 2851970688,
      "step": 10030,
      "train_runtime": 97547.3978,
      "train_tokens_per_second": 29236.769
    },
    {
      "epoch": 0.3552355447836942,
      "grad_norm": 0.6328125,
      "learning_rate": 3.5320917936464294e-05,
      "loss": 0.42274885177612304,
      "num_input_tokens_seen": 2854831552,
      "step": 10040,
      "train_runtime": 97647.0374,
      "train_tokens_per_second": 29236.233
    },
    {
      "epoch": 0.35558936504742295,
      "grad_norm": 0.5859375,
      "learning_rate": 3.5312108189504505e-05,
      "loss": 0.41487321853637693,
      "num_input_tokens_seen": 2857663424,
      "step": 10050,
      "train_runtime": 97743.3384,
      "train_tokens_per_second": 29236.401
    },
    {
      "epoch": 0.35594318531115177,
      "grad_norm": 0.5703125,
      "learning_rate": 3.530330503124204e-05,
      "loss": 0.41655702590942384,
      "num_input_tokens_seen": 2860553920,
      "step": 10060,
      "train_runtime": 97842.8148,
      "train_tokens_per_second": 29236.219
    },
    {
      "epoch": 0.35629700557488053,
      "grad_norm": 0.5859375,
      "learning_rate": 3.5294508453468325e-05,
      "loss": 0.41977386474609374,
      "num_input_tokens_seen": 2863390528,
      "step": 10070,
      "train_runtime": 97942.9849,
      "train_tokens_per_second": 29235.279
    },
    {
      "epoch": 0.3566508258386093,
      "grad_norm": 0.61328125,
      "learning_rate": 3.528571844798908e-05,
      "loss": 0.42212538719177245,
      "num_input_tokens_seen": 2866238080,
      "step": 10080,
      "train_runtime": 98041.0984,
      "train_tokens_per_second": 29235.067
    },
    {
      "epoch": 0.35700464610233806,
      "grad_norm": 0.5859375,
      "learning_rate": 3.527693500662431e-05,
      "loss": 0.4170633316040039,
      "num_input_tokens_seen": 2869066944,
      "step": 10090,
      "train_runtime": 98136.6826,
      "train_tokens_per_second": 29235.418
    },
    {
      "epoch": 0.3573584663660669,
      "grad_norm": 0.57421875,
      "learning_rate": 3.5268158121208294e-05,
      "loss": 0.42119512557983396,
      "num_input_tokens_seen": 2871896960,
      "step": 10100,
      "train_runtime": 98234.3211,
      "train_tokens_per_second": 29235.169
    },
    {
      "epoch": 0.35771228662979565,
      "grad_norm": 0.6015625,
      "learning_rate": 3.525938778358949e-05,
      "loss": 0.4142122268676758,
      "num_input_tokens_seen": 2874711360,
      "step": 10110,
      "train_runtime": 98328.5837,
      "train_tokens_per_second": 29235.765
    },
    {
      "epoch": 0.3580661068935244,
      "grad_norm": 0.6015625,
      "learning_rate": 3.5250623985630537e-05,
      "loss": 0.41509456634521485,
      "num_input_tokens_seen": 2877534400,
      "step": 10120,
      "train_runtime": 98423.8529,
      "train_tokens_per_second": 29236.149
    },
    {
      "epoch": 0.3584199271572532,
      "grad_norm": 0.578125,
      "learning_rate": 3.524186671920826e-05,
      "loss": 0.4168394565582275,
      "num_input_tokens_seen": 2880340160,
      "step": 10130,
      "train_runtime": 98521.0819,
      "train_tokens_per_second": 29235.775
    },
    {
      "epoch": 0.358773747420982,
      "grad_norm": 0.60546875,
      "learning_rate": 3.523311597621358e-05,
      "loss": 0.42333388328552246,
      "num_input_tokens_seen": 2883145024,
      "step": 10140,
      "train_runtime": 98614.8419,
      "train_tokens_per_second": 29236.421
    },
    {
      "epoch": 0.35912756768471077,
      "grad_norm": 0.57421875,
      "learning_rate": 3.5224371748551505e-05,
      "loss": 0.42003836631774905,
      "num_input_tokens_seen": 2885943040,
      "step": 10150,
      "train_runtime": 98708.8884,
      "train_tokens_per_second": 29236.912
    },
    {
      "epoch": 0.35948138794843953,
      "grad_norm": 0.578125,
      "learning_rate": 3.521563402814109e-05,
      "loss": 0.4177816390991211,
      "num_input_tokens_seen": 2888792000,
      "step": 10160,
      "train_runtime": 98805.0323,
      "train_tokens_per_second": 29237.296
    },
    {
      "epoch": 0.3598352082121683,
      "grad_norm": 0.59375,
      "learning_rate": 3.5206902806915436e-05,
      "loss": 0.4176754951477051,
      "num_input_tokens_seen": 2891615552,
      "step": 10170,
      "train_runtime": 98901.3469,
      "train_tokens_per_second": 29237.373
    },
    {
      "epoch": 0.3601890284758971,
      "grad_norm": 0.578125,
      "learning_rate": 3.5198178076821644e-05,
      "loss": 0.41602187156677245,
      "num_input_tokens_seen": 2894430336,
      "step": 10180,
      "train_runtime": 98999.4892,
      "train_tokens_per_second": 29236.821
    },
    {
      "epoch": 0.3605428487396259,
      "grad_norm": 0.58984375,
      "learning_rate": 3.5189459829820743e-05,
      "loss": 0.41626224517822263,
      "num_input_tokens_seen": 2897246656,
      "step": 10190,
      "train_runtime": 99093.5887,
      "train_tokens_per_second": 29237.478
    },
    {
      "epoch": 0.36089666900335465,
      "grad_norm": 0.578125,
      "learning_rate": 3.5180748057887714e-05,
      "loss": 0.417949914932251,
      "num_input_tokens_seen": 2900098944,
      "step": 10200,
      "train_runtime": 99191.2792,
      "train_tokens_per_second": 29237.439
    },
    {
      "epoch": 0.3612504892670834,
      "grad_norm": 0.58203125,
      "learning_rate": 3.517204275301144e-05,
      "loss": 0.42154669761657715,
      "num_input_tokens_seen": 2902918144,
      "step": 10210,
      "train_runtime": 99288.2768,
      "train_tokens_per_second": 29237.27
    },
    {
      "epoch": 0.36160430953081224,
      "grad_norm": 0.6015625,
      "learning_rate": 3.5163343907194676e-05,
      "loss": 0.41838760375976564,
      "num_input_tokens_seen": 2905770432,
      "step": 10220,
      "train_runtime": 99384.0492,
      "train_tokens_per_second": 29237.795
    },
    {
      "epoch": 0.361958129794541,
      "grad_norm": 0.62890625,
      "learning_rate": 3.5154651512453995e-05,
      "loss": 0.41790127754211426,
      "num_input_tokens_seen": 2908571584,
      "step": 10230,
      "train_runtime": 99480.4201,
      "train_tokens_per_second": 29237.629
    },
    {
      "epoch": 0.36231195005826977,
      "grad_norm": 0.56640625,
      "learning_rate": 3.514596556081981e-05,
      "loss": 0.4139357566833496,
      "num_input_tokens_seen": 2911423872,
      "step": 10240,
      "train_runtime": 99577.2086,
      "train_tokens_per_second": 29237.854
    },
    {
      "epoch": 0.36266577032199854,
      "grad_norm": 0.5703125,
      "learning_rate": 3.513728604433628e-05,
      "loss": 0.42030978202819824,
      "num_input_tokens_seen": 2914283584,
      "step": 10250,
      "train_runtime": 99676.1521,
      "train_tokens_per_second": 29237.521
    },
    {
      "epoch": 0.36301959058572736,
      "grad_norm": 0.5625,
      "learning_rate": 3.5128612955061334e-05,
      "loss": 0.419094181060791,
      "num_input_tokens_seen": 2917172096,
      "step": 10260,
      "train_runtime": 99776.8519,
      "train_tokens_per_second": 29236.963
    },
    {
      "epoch": 0.3633734108494561,
      "grad_norm": 0.56640625,
      "learning_rate": 3.5119946285066595e-05,
      "loss": 0.4205984115600586,
      "num_input_tokens_seen": 2919902656,
      "step": 10270,
      "train_runtime": 99867.1355,
      "train_tokens_per_second": 29237.873
    },
    {
      "epoch": 0.3637272311131849,
      "grad_norm": 0.6015625,
      "learning_rate": 3.511128602643739e-05,
      "loss": 0.41779866218566897,
      "num_input_tokens_seen": 2922714496,
      "step": 10280,
      "train_runtime": 99963.6776,
      "train_tokens_per_second": 29237.765
    },
    {
      "epoch": 0.36408105137691366,
      "grad_norm": 0.5859375,
      "learning_rate": 3.510263217127269e-05,
      "loss": 0.4193439483642578,
      "num_input_tokens_seen": 2925525888,
      "step": 10290,
      "train_runtime": 100060.0874,
      "train_tokens_per_second": 29237.691
    },
    {
      "epoch": 0.3644348716406424,
      "grad_norm": 0.55859375,
      "learning_rate": 3.50939847116851e-05,
      "loss": 0.4167616367340088,
      "num_input_tokens_seen": 2928375872,
      "step": 10300,
      "train_runtime": 100156.8983,
      "train_tokens_per_second": 29237.885
    },
    {
      "epoch": 0.36478869190437124,
      "grad_norm": 0.58203125,
      "learning_rate": 3.508534363980081e-05,
      "loss": 0.4129524230957031,
      "num_input_tokens_seen": 2931220864,
      "step": 10310,
      "train_runtime": 100254.7491,
      "train_tokens_per_second": 29237.726
    },
    {
      "epoch": 0.3651425121681,
      "grad_norm": 0.57421875,
      "learning_rate": 3.507670894775958e-05,
      "loss": 0.4148233413696289,
      "num_input_tokens_seen": 2934116352,
      "step": 10320,
      "train_runtime": 100354.399,
      "train_tokens_per_second": 29237.546
    },
    {
      "epoch": 0.3654963324318288,
      "grad_norm": 0.59375,
      "learning_rate": 3.506808062771471e-05,
      "loss": 0.41765766143798827,
      "num_input_tokens_seen": 2936979328,
      "step": 10330,
      "train_runtime": 100454.047,
      "train_tokens_per_second": 29237.043
    },
    {
      "epoch": 0.36585015269555754,
      "grad_norm": 0.5625,
      "learning_rate": 3.505945867183298e-05,
      "loss": 0.41604223251342776,
      "num_input_tokens_seen": 2939864704,
      "step": 10340,
      "train_runtime": 100551.7199,
      "train_tokens_per_second": 29237.339
    },
    {
      "epoch": 0.36620397295928636,
      "grad_norm": 0.55859375,
      "learning_rate": 3.505084307229468e-05,
      "loss": 0.41513447761535643,
      "num_input_tokens_seen": 2942764864,
      "step": 10350,
      "train_runtime": 100650.5847,
      "train_tokens_per_second": 29237.434
    },
    {
      "epoch": 0.3665577932230151,
      "grad_norm": 0.58984375,
      "learning_rate": 3.5042233821293525e-05,
      "loss": 0.4170383453369141,
      "num_input_tokens_seen": 2945659584,
      "step": 10360,
      "train_runtime": 100750.4682,
      "train_tokens_per_second": 29237.18
    },
    {
      "epoch": 0.3669116134867439,
      "grad_norm": 0.5703125,
      "learning_rate": 3.503363091103664e-05,
      "loss": 0.41117086410522463,
      "num_input_tokens_seen": 2948521152,
      "step": 10370,
      "train_runtime": 100848.4551,
      "train_tokens_per_second": 29237.147
    },
    {
      "epoch": 0.36726543375047266,
      "grad_norm": 0.5546875,
      "learning_rate": 3.5025034333744545e-05,
      "loss": 0.414675235748291,
      "num_input_tokens_seen": 2951408384,
      "step": 10380,
      "train_runtime": 100949.5538,
      "train_tokens_per_second": 29236.468
    },
    {
      "epoch": 0.3676192540142015,
      "grad_norm": 0.5625,
      "learning_rate": 3.501644408165112e-05,
      "loss": 0.4157355308532715,
      "num_input_tokens_seen": 2954215808,
      "step": 10390,
      "train_runtime": 101045.5959,
      "train_tokens_per_second": 29236.463
    },
    {
      "epoch": 0.36797307427793025,
      "grad_norm": 0.5546875,
      "learning_rate": 3.500786014700357e-05,
      "loss": 0.41501941680908205,
      "num_input_tokens_seen": 2957086144,
      "step": 10400,
      "train_runtime": 101143.5927,
      "train_tokens_per_second": 29236.515
    },
    {
      "epoch": 0.368326894541659,
      "grad_norm": 0.5546875,
      "learning_rate": 3.499928252206237e-05,
      "loss": 0.41730828285217286,
      "num_input_tokens_seen": 2959937472,
      "step": 10410,
      "train_runtime": 101240.8443,
      "train_tokens_per_second": 29236.594
    },
    {
      "epoch": 0.3686807148053878,
      "grad_norm": 0.56640625,
      "learning_rate": 3.499071119910131e-05,
      "loss": 0.42185797691345217,
      "num_input_tokens_seen": 2962776128,
      "step": 10420,
      "train_runtime": 101338.0601,
      "train_tokens_per_second": 29236.559
    },
    {
      "epoch": 0.3690345350691166,
      "grad_norm": 0.57421875,
      "learning_rate": 3.498214617040739e-05,
      "loss": 0.41416540145874026,
      "num_input_tokens_seen": 2965612416,
      "step": 10430,
      "train_runtime": 101433.8272,
      "train_tokens_per_second": 29236.917
    },
    {
      "epoch": 0.36938835533284536,
      "grad_norm": 0.5703125,
      "learning_rate": 3.49735874282808e-05,
      "loss": 0.4203587532043457,
      "num_input_tokens_seen": 2968451008,
      "step": 10440,
      "train_runtime": 101530.9591,
      "train_tokens_per_second": 29236.905
    },
    {
      "epoch": 0.36974217559657413,
      "grad_norm": 0.56640625,
      "learning_rate": 3.4965034965034965e-05,
      "loss": 0.4138647079467773,
      "num_input_tokens_seen": 2971281600,
      "step": 10450,
      "train_runtime": 101624.8098,
      "train_tokens_per_second": 29237.758
    },
    {
      "epoch": 0.3700959958603029,
      "grad_norm": 0.59375,
      "learning_rate": 3.495648877299642e-05,
      "loss": 0.4175432205200195,
      "num_input_tokens_seen": 2974099712,
      "step": 10460,
      "train_runtime": 101720.1437,
      "train_tokens_per_second": 29238.06
    },
    {
      "epoch": 0.3704498161240317,
      "grad_norm": 0.6171875,
      "learning_rate": 3.494794884450483e-05,
      "loss": 0.41556100845336913,
      "num_input_tokens_seen": 2976928896,
      "step": 10470,
      "train_runtime": 101816.4647,
      "train_tokens_per_second": 29238.188
    },
    {
      "epoch": 0.3708036363877605,
      "grad_norm": 0.55859375,
      "learning_rate": 3.4939415171912954e-05,
      "loss": 0.4166508674621582,
      "num_input_tokens_seen": 2979790336,
      "step": 10480,
      "train_runtime": 101914.7493,
      "train_tokens_per_second": 29238.068
    },
    {
      "epoch": 0.37115745665148925,
      "grad_norm": 0.55859375,
      "learning_rate": 3.4930887747586616e-05,
      "loss": 0.41992530822753904,
      "num_input_tokens_seen": 2982652160,
      "step": 10490,
      "train_runtime": 102012.4065,
      "train_tokens_per_second": 29238.132
    },
    {
      "epoch": 0.371511276915218,
      "grad_norm": 0.56640625,
      "learning_rate": 3.492236656390469e-05,
      "loss": 0.420231294631958,
      "num_input_tokens_seen": 2985514176,
      "step": 10500,
      "train_runtime": 102112.612,
      "train_tokens_per_second": 29237.468
    },
    {
      "epoch": 0.37186509717894684,
      "grad_norm": 0.55859375,
      "learning_rate": 3.4913851613259034e-05,
      "loss": 0.41298365592956543,
      "num_input_tokens_seen": 2988309056,
      "step": 10510,
      "train_runtime": 102206.1399,
      "train_tokens_per_second": 29238.058
    },
    {
      "epoch": 0.3722189174426756,
      "grad_norm": 0.5703125,
      "learning_rate": 3.490534288805452e-05,
      "loss": 0.417043399810791,
      "num_input_tokens_seen": 2991174400,
      "step": 10520,
      "train_runtime": 102301.7758,
      "train_tokens_per_second": 29238.734
    },
    {
      "epoch": 0.37257273770640437,
      "grad_norm": 0.55859375,
      "learning_rate": 3.489684038070891e-05,
      "loss": 0.41370220184326173,
      "num_input_tokens_seen": 2993989824,
      "step": 10530,
      "train_runtime": 102398.3344,
      "train_tokens_per_second": 29238.657
    },
    {
      "epoch": 0.37292655797013313,
      "grad_norm": 0.6015625,
      "learning_rate": 3.488834408365296e-05,
      "loss": 0.423077392578125,
      "num_input_tokens_seen": 2996803968,
      "step": 10540,
      "train_runtime": 102493.5396,
      "train_tokens_per_second": 29238.955
    },
    {
      "epoch": 0.37328037823386195,
      "grad_norm": 0.61328125,
      "learning_rate": 3.487985398933027e-05,
      "loss": 0.416903018951416,
      "num_input_tokens_seen": 2999637952,
      "step": 10550,
      "train_runtime": 102589.4619,
      "train_tokens_per_second": 29239.241
    },
    {
      "epoch": 0.3736341984975907,
      "grad_norm": 0.58203125,
      "learning_rate": 3.4871370090197324e-05,
      "loss": 0.4210418701171875,
      "num_input_tokens_seen": 3002435456,
      "step": 10560,
      "train_runtime": 102683.9946,
      "train_tokens_per_second": 29239.566
    },
    {
      "epoch": 0.3739880187613195,
      "grad_norm": 0.5859375,
      "learning_rate": 3.486289237872343e-05,
      "loss": 0.4146622657775879,
      "num_input_tokens_seen": 3005268288,
      "step": 10570,
      "train_runtime": 102783.2083,
      "train_tokens_per_second": 29238.903
    },
    {
      "epoch": 0.37434183902504825,
      "grad_norm": 0.55859375,
      "learning_rate": 3.485442084739075e-05,
      "loss": 0.41399445533752444,
      "num_input_tokens_seen": 3008101568,
      "step": 10580,
      "train_runtime": 102881.8175,
      "train_tokens_per_second": 29238.418
    },
    {
      "epoch": 0.374695659288777,
      "grad_norm": 0.59765625,
      "learning_rate": 3.484595548869416e-05,
      "loss": 0.41747174263000486,
      "num_input_tokens_seen": 3010975040,
      "step": 10590,
      "train_runtime": 102979.1068,
      "train_tokens_per_second": 29238.698
    },
    {
      "epoch": 0.37504947955250584,
      "grad_norm": 0.55859375,
      "learning_rate": 3.4837496295141335e-05,
      "loss": 0.4180622100830078,
      "num_input_tokens_seen": 3013859904,
      "step": 10600,
      "train_runtime": 103082.2913,
      "train_tokens_per_second": 29237.417
    },
    {
      "epoch": 0.3754032998162346,
      "grad_norm": 0.57421875,
      "learning_rate": 3.482904325925266e-05,
      "loss": 0.41494026184082033,
      "num_input_tokens_seen": 3016678976,
      "step": 10610,
      "train_runtime": 103180.3196,
      "train_tokens_per_second": 29236.961
    },
    {
      "epoch": 0.37575712007996337,
      "grad_norm": 0.5625,
      "learning_rate": 3.482059637356124e-05,
      "loss": 0.41609535217285154,
      "num_input_tokens_seen": 3019452544,
      "step": 10620,
      "train_runtime": 103271.6438,
      "train_tokens_per_second": 29237.963
    },
    {
      "epoch": 0.37611094034369214,
      "grad_norm": 0.56640625,
      "learning_rate": 3.481215563061281e-05,
      "loss": 0.4127861976623535,
      "num_input_tokens_seen": 3022247808,
      "step": 10630,
      "train_runtime": 103368.0662,
      "train_tokens_per_second": 29237.732
    },
    {
      "epoch": 0.37646476060742096,
      "grad_norm": 0.57421875,
      "learning_rate": 3.4803721022965785e-05,
      "loss": 0.41522841453552245,
      "num_input_tokens_seen": 3025134272,
      "step": 10640,
      "train_runtime": 103466.6921,
      "train_tokens_per_second": 29237.76
    },
    {
      "epoch": 0.3768185808711497,
      "grad_norm": 0.578125,
      "learning_rate": 3.479529254319117e-05,
      "loss": 0.4138516426086426,
      "num_input_tokens_seen": 3027994880,
      "step": 10650,
      "train_runtime": 103564.5942,
      "train_tokens_per_second": 29237.742
    },
    {
      "epoch": 0.3771724011348785,
      "grad_norm": 0.55859375,
      "learning_rate": 3.478687018387257e-05,
      "loss": 0.42015748023986815,
      "num_input_tokens_seen": 3030848064,
      "step": 10660,
      "train_runtime": 103660.6563,
      "train_tokens_per_second": 29238.172
    },
    {
      "epoch": 0.37752622139860725,
      "grad_norm": 0.57421875,
      "learning_rate": 3.477845393760616e-05,
      "loss": 0.416491174697876,
      "num_input_tokens_seen": 3033682752,
      "step": 10670,
      "train_runtime": 103758.9775,
      "train_tokens_per_second": 29237.786
    },
    {
      "epoch": 0.3778800416623361,
      "grad_norm": 0.55078125,
      "learning_rate": 3.4770043797000614e-05,
      "loss": 0.4144142150878906,
      "num_input_tokens_seen": 3036544128,
      "step": 10680,
      "train_runtime": 103855.4374,
      "train_tokens_per_second": 29238.182
    },
    {
      "epoch": 0.37823386192606484,
      "grad_norm": 0.5625,
      "learning_rate": 3.4761639754677146e-05,
      "loss": 0.4219967365264893,
      "num_input_tokens_seen": 3039454016,
      "step": 10690,
      "train_runtime": 103958.9261,
      "train_tokens_per_second": 29237.066
    },
    {
      "epoch": 0.3785876821897936,
      "grad_norm": 0.5859375,
      "learning_rate": 3.4753241803269435e-05,
      "loss": 0.41213126182556153,
      "num_input_tokens_seen": 3042330880,
      "step": 10700,
      "train_runtime": 104057.9724,
      "train_tokens_per_second": 29236.884
    },
    {
      "epoch": 0.37894150245352237,
      "grad_norm": 0.6015625,
      "learning_rate": 3.474484993542361e-05,
      "loss": 0.4179734230041504,
      "num_input_tokens_seen": 3045096960,
      "step": 10710,
      "train_runtime": 104151.0302,
      "train_tokens_per_second": 29237.32
    },
    {
      "epoch": 0.3792953227172512,
      "grad_norm": 0.5390625,
      "learning_rate": 3.473646414379822e-05,
      "loss": 0.4129387855529785,
      "num_input_tokens_seen": 3047948800,
      "step": 10720,
      "train_runtime": 104249.9852,
      "train_tokens_per_second": 29236.923
    },
    {
      "epoch": 0.37964914298097996,
      "grad_norm": 0.55078125,
      "learning_rate": 3.472808442106422e-05,
      "loss": 0.4151453971862793,
      "num_input_tokens_seen": 3050755584,
      "step": 10730,
      "train_runtime": 104345.1266,
      "train_tokens_per_second": 29237.164
    },
    {
      "epoch": 0.3800029632447087,
      "grad_norm": 0.56640625,
      "learning_rate": 3.4719710759904936e-05,
      "loss": 0.41498217582702634,
      "num_input_tokens_seen": 3053615808,
      "step": 10740,
      "train_runtime": 104443.9086,
      "train_tokens_per_second": 29236.897
    },
    {
      "epoch": 0.3803567835084375,
      "grad_norm": 0.58203125,
      "learning_rate": 3.471134315301603e-05,
      "loss": 0.41576237678527833,
      "num_input_tokens_seen": 3056458816,
      "step": 10750,
      "train_runtime": 104541.9073,
      "train_tokens_per_second": 29236.685
    },
    {
      "epoch": 0.3807106037721663,
      "grad_norm": 0.5625,
      "learning_rate": 3.470298159310549e-05,
      "loss": 0.41095819473266604,
      "num_input_tokens_seen": 3059267264,
      "step": 10760,
      "train_runtime": 104639.1504,
      "train_tokens_per_second": 29236.354
    },
    {
      "epoch": 0.3810644240358951,
      "grad_norm": 0.5859375,
      "learning_rate": 3.4694626072893585e-05,
      "loss": 0.41770339012145996,
      "num_input_tokens_seen": 3062082304,
      "step": 10770,
      "train_runtime": 104735.0106,
      "train_tokens_per_second": 29236.473
    },
    {
      "epoch": 0.38141824429962384,
      "grad_norm": 0.55859375,
      "learning_rate": 3.468627658511285e-05,
      "loss": 0.41169066429138185,
      "num_input_tokens_seen": 3064864640,
      "step": 10780,
      "train_runtime": 104827.1761,
      "train_tokens_per_second": 29237.31
    },
    {
      "epoch": 0.3817720645633526,
      "grad_norm": 0.57421875,
      "learning_rate": 3.467793312250806e-05,
      "loss": 0.4173437118530273,
      "num_input_tokens_seen": 3067711424,
      "step": 10790,
      "train_runtime": 104923.53,
      "train_tokens_per_second": 29237.593
    },
    {
      "epoch": 0.38212588482708143,
      "grad_norm": 0.5625,
      "learning_rate": 3.466959567783619e-05,
      "loss": 0.4156506061553955,
      "num_input_tokens_seen": 3070594560,
      "step": 10800,
      "train_runtime": 105022.0814,
      "train_tokens_per_second": 29237.609
    },
    {
      "epoch": 0.3824797050908102,
      "grad_norm": 0.55859375,
      "learning_rate": 3.466126424386642e-05,
      "loss": 0.4139736175537109,
      "num_input_tokens_seen": 3073427264,
      "step": 10810,
      "train_runtime": 105118.701,
      "train_tokens_per_second": 29237.683
    },
    {
      "epoch": 0.38283352535453896,
      "grad_norm": 0.5546875,
      "learning_rate": 3.4652938813380056e-05,
      "loss": 0.414139461517334,
      "num_input_tokens_seen": 3076311168,
      "step": 10820,
      "train_runtime": 105216.2033,
      "train_tokens_per_second": 29237.998
    },
    {
      "epoch": 0.3831873456182677,
      "grad_norm": 0.57421875,
      "learning_rate": 3.464461937917057e-05,
      "loss": 0.4099103927612305,
      "num_input_tokens_seen": 3079126784,
      "step": 10830,
      "train_runtime": 105311.4041,
      "train_tokens_per_second": 29238.303
    },
    {
      "epoch": 0.3835411658819965,
      "grad_norm": 0.59375,
      "learning_rate": 3.4636305934043525e-05,
      "loss": 0.41831240653991697,
      "num_input_tokens_seen": 3082002240,
      "step": 10840,
      "train_runtime": 105412.7465,
      "train_tokens_per_second": 29237.472
    },
    {
      "epoch": 0.3838949861457253,
      "grad_norm": 0.5703125,
      "learning_rate": 3.4627998470816544e-05,
      "loss": 0.41381354331970216,
      "num_input_tokens_seen": 3084901440,
      "step": 10850,
      "train_runtime": 105509.4002,
      "train_tokens_per_second": 29238.167
    },
    {
      "epoch": 0.3842488064094541,
      "grad_norm": 0.5703125,
      "learning_rate": 3.4619696982319334e-05,
      "loss": 0.4139442443847656,
      "num_input_tokens_seen": 3087681536,
      "step": 10860,
      "train_runtime": 105603.5923,
      "train_tokens_per_second": 29238.414
    },
    {
      "epoch": 0.38460262667318285,
      "grad_norm": 0.59765625,
      "learning_rate": 3.461140146139361e-05,
      "loss": 0.41739892959594727,
      "num_input_tokens_seen": 3090515008,
      "step": 10870,
      "train_runtime": 105700.9769,
      "train_tokens_per_second": 29238.282
    },
    {
      "epoch": 0.3849564469369116,
      "grad_norm": 0.56640625,
      "learning_rate": 3.460311190089309e-05,
      "loss": 0.4170172691345215,
      "num_input_tokens_seen": 3093318016,
      "step": 10880,
      "train_runtime": 105797.094,
      "train_tokens_per_second": 29238.213
    },
    {
      "epoch": 0.38531026720064043,
      "grad_norm": 0.5625,
      "learning_rate": 3.459482829368348e-05,
      "loss": 0.41702852249145506,
      "num_input_tokens_seen": 3096163776,
      "step": 10890,
      "train_runtime": 105894.4538,
      "train_tokens_per_second": 29238.205
    },
    {
      "epoch": 0.3856640874643692,
      "grad_norm": 0.6171875,
      "learning_rate": 3.4586550632642425e-05,
      "loss": 0.41747055053710935,
      "num_input_tokens_seen": 3099002176,
      "step": 10900,
      "train_runtime": 105991.5087,
      "train_tokens_per_second": 29238.212
    },
    {
      "epoch": 0.38601790772809796,
      "grad_norm": 0.57421875,
      "learning_rate": 3.457827891065949e-05,
      "loss": 0.4186979293823242,
      "num_input_tokens_seen": 3101874688,
      "step": 10910,
      "train_runtime": 106089.9909,
      "train_tokens_per_second": 29238.146
    },
    {
      "epoch": 0.38637172799182673,
      "grad_norm": 0.57421875,
      "learning_rate": 3.457001312063614e-05,
      "loss": 0.41554436683654783,
      "num_input_tokens_seen": 3104693824,
      "step": 10920,
      "train_runtime": 106184.9611,
      "train_tokens_per_second": 29238.546
    },
    {
      "epoch": 0.38672554825555555,
      "grad_norm": 0.56640625,
      "learning_rate": 3.45617532554857e-05,
      "loss": 0.41666746139526367,
      "num_input_tokens_seen": 3107511360,
      "step": 10930,
      "train_runtime": 106282.591,
      "train_tokens_per_second": 29238.197
    },
    {
      "epoch": 0.3870793685192843,
      "grad_norm": 0.5703125,
      "learning_rate": 3.455349930813339e-05,
      "loss": 0.41462955474853513,
      "num_input_tokens_seen": 3110319744,
      "step": 10940,
      "train_runtime": 106376.9498,
      "train_tokens_per_second": 29238.663
    },
    {
      "epoch": 0.3874331887830131,
      "grad_norm": 0.56640625,
      "learning_rate": 3.45452512715162e-05,
      "loss": 0.4163802146911621,
      "num_input_tokens_seen": 3113139328,
      "step": 10950,
      "train_runtime": 106473.9351,
      "train_tokens_per_second": 29238.511
    },
    {
      "epoch": 0.38778700904674185,
      "grad_norm": 0.55859375,
      "learning_rate": 3.4537009138582935e-05,
      "loss": 0.4166153907775879,
      "num_input_tokens_seen": 3116035456,
      "step": 10960,
      "train_runtime": 106574.7197,
      "train_tokens_per_second": 29238.036
    },
    {
      "epoch": 0.38814082931047067,
      "grad_norm": 0.5625,
      "learning_rate": 3.4528772902294174e-05,
      "loss": 0.4174182891845703,
      "num_input_tokens_seen": 3118861248,
      "step": 10970,
      "train_runtime": 106668.4811,
      "train_tokens_per_second": 29238.827
    },
    {
      "epoch": 0.38849464957419944,
      "grad_norm": 0.578125,
      "learning_rate": 3.452054255562222e-05,
      "loss": 0.41211872100830077,
      "num_input_tokens_seen": 3121694144,
      "step": 10980,
      "train_runtime": 106765.0908,
      "train_tokens_per_second": 29238.903
    },
    {
      "epoch": 0.3888484698379282,
      "grad_norm": 0.56640625,
      "learning_rate": 3.451231809155115e-05,
      "loss": 0.419116735458374,
      "num_input_tokens_seen": 3124662464,
      "step": 10990,
      "train_runtime": 106875.2447,
      "train_tokens_per_second": 29236.541
    },
    {
      "epoch": 0.38920229010165697,
      "grad_norm": 0.58203125,
      "learning_rate": 3.450409950307666e-05,
      "loss": 0.4179194450378418,
      "num_input_tokens_seen": 3127486208,
      "step": 11000,
      "train_runtime": 106972.2202,
      "train_tokens_per_second": 29236.434
    },
    {
      "epoch": 0.3895561103653858,
      "grad_norm": 0.60546875,
      "learning_rate": 3.449588678320619e-05,
      "loss": 0.41327953338623047,
      "num_input_tokens_seen": 3130333248,
      "step": 11010,
      "train_runtime": 107068.9033,
      "train_tokens_per_second": 29236.624
    },
    {
      "epoch": 0.38990993062911455,
      "grad_norm": 0.578125,
      "learning_rate": 3.4487679924958767e-05,
      "loss": 0.41757965087890625,
      "num_input_tokens_seen": 3133211840,
      "step": 11020,
      "train_runtime": 107168.7551,
      "train_tokens_per_second": 29236.244
    },
    {
      "epoch": 0.3902637508928433,
      "grad_norm": 0.5859375,
      "learning_rate": 3.4479478921365076e-05,
      "loss": 0.4129302978515625,
      "num_input_tokens_seen": 3136050176,
      "step": 11030,
      "train_runtime": 107264.2321,
      "train_tokens_per_second": 29236.681
    },
    {
      "epoch": 0.3906175711565721,
      "grad_norm": 0.58203125,
      "learning_rate": 3.447128376546738e-05,
      "loss": 0.418621826171875,
      "num_input_tokens_seen": 3138931008,
      "step": 11040,
      "train_runtime": 107363.2956,
      "train_tokens_per_second": 29236.537
    },
    {
      "epoch": 0.3909713914203009,
      "grad_norm": 0.59765625,
      "learning_rate": 3.4463094450319505e-05,
      "loss": 0.41826591491699217,
      "num_input_tokens_seen": 3141796864,
      "step": 11050,
      "train_runtime": 107464.8358,
      "train_tokens_per_second": 29235.581
    },
    {
      "epoch": 0.3913252116840297,
      "grad_norm": 0.5625,
      "learning_rate": 3.4454910968986855e-05,
      "loss": 0.4192346572875977,
      "num_input_tokens_seen": 3144613824,
      "step": 11060,
      "train_runtime": 107561.4481,
      "train_tokens_per_second": 29235.51
    },
    {
      "epoch": 0.39167903194775844,
      "grad_norm": 0.55859375,
      "learning_rate": 3.4446733314546336e-05,
      "loss": 0.42135019302368165,
      "num_input_tokens_seen": 3147476032,
      "step": 11070,
      "train_runtime": 107662.3514,
      "train_tokens_per_second": 29234.695
    },
    {
      "epoch": 0.3920328522114872,
      "grad_norm": 0.5859375,
      "learning_rate": 3.443856148008633e-05,
      "loss": 0.410903263092041,
      "num_input_tokens_seen": 3150305600,
      "step": 11080,
      "train_runtime": 107758.3784,
      "train_tokens_per_second": 29234.902
    },
    {
      "epoch": 0.39238667247521597,
      "grad_norm": 0.58984375,
      "learning_rate": 3.443039545870672e-05,
      "loss": 0.419382905960083,
      "num_input_tokens_seen": 3153158784,
      "step": 11090,
      "train_runtime": 107854.9993,
      "train_tokens_per_second": 29235.166
    },
    {
      "epoch": 0.3927404927389448,
      "grad_norm": 0.56640625,
      "learning_rate": 3.442223524351883e-05,
      "loss": 0.416320276260376,
      "num_input_tokens_seen": 3156049984,
      "step": 11100,
      "train_runtime": 107957.0325,
      "train_tokens_per_second": 29234.316
    },
    {
      "epoch": 0.39309431300267356,
      "grad_norm": 0.59765625,
      "learning_rate": 3.44140808276454e-05,
      "loss": 0.41596460342407227,
      "num_input_tokens_seen": 3158898944,
      "step": 11110,
      "train_runtime": 108054.8311,
      "train_tokens_per_second": 29234.222
    },
    {
      "epoch": 0.3934481332664023,
      "grad_norm": 0.5859375,
      "learning_rate": 3.4405932204220575e-05,
      "loss": 0.41338586807250977,
      "num_input_tokens_seen": 3161758976,
      "step": 11120,
      "train_runtime": 108152.8269,
      "train_tokens_per_second": 29234.178
    },
    {
      "epoch": 0.3938019535301311,
      "grad_norm": 0.578125,
      "learning_rate": 3.4397789366389876e-05,
      "loss": 0.42003211975097654,
      "num_input_tokens_seen": 3164592256,
      "step": 11130,
      "train_runtime": 108247.4322,
      "train_tokens_per_second": 29234.802
    },
    {
      "epoch": 0.3941557737938599,
      "grad_norm": 0.59765625,
      "learning_rate": 3.438965230731016e-05,
      "loss": 0.4142338752746582,
      "num_input_tokens_seen": 3167391680,
      "step": 11140,
      "train_runtime": 108340.3606,
      "train_tokens_per_second": 29235.565
    },
    {
      "epoch": 0.3945095940575887,
      "grad_norm": 0.55859375,
      "learning_rate": 3.438152102014964e-05,
      "loss": 0.41826043128967283,
      "num_input_tokens_seen": 3170208256,
      "step": 11150,
      "train_runtime": 108436.2954,
      "train_tokens_per_second": 29235.675
    },
    {
      "epoch": 0.39486341432131744,
      "grad_norm": 0.55078125,
      "learning_rate": 3.437339549808778e-05,
      "loss": 0.41660280227661134,
      "num_input_tokens_seen": 3173073088,
      "step": 11160,
      "train_runtime": 108535.1378,
      "train_tokens_per_second": 29235.445
    },
    {
      "epoch": 0.3952172345850462,
      "grad_norm": 0.58203125,
      "learning_rate": 3.43652757343154e-05,
      "loss": 0.4146543025970459,
      "num_input_tokens_seen": 3175926528,
      "step": 11170,
      "train_runtime": 108632.8473,
      "train_tokens_per_second": 29235.416
    },
    {
      "epoch": 0.39557105484877503,
      "grad_norm": 0.58984375,
      "learning_rate": 3.435716172203449e-05,
      "loss": 0.4116074562072754,
      "num_input_tokens_seen": 3178719680,
      "step": 11180,
      "train_runtime": 108726.1964,
      "train_tokens_per_second": 29236.006
    },
    {
      "epoch": 0.3959248751125038,
      "grad_norm": 0.5859375,
      "learning_rate": 3.434905345445833e-05,
      "loss": 0.41634092330932615,
      "num_input_tokens_seen": 3181518912,
      "step": 11190,
      "train_runtime": 108822.6646,
      "train_tokens_per_second": 29235.812
    },
    {
      "epoch": 0.39627869537623256,
      "grad_norm": 0.56640625,
      "learning_rate": 3.4340950924811374e-05,
      "loss": 0.4111994743347168,
      "num_input_tokens_seen": 3184423232,
      "step": 11200,
      "train_runtime": 108920.5394,
      "train_tokens_per_second": 29236.205
    },
    {
      "epoch": 0.3966325156399613,
      "grad_norm": 0.5859375,
      "learning_rate": 3.433285412632927e-05,
      "loss": 0.41596279144287107,
      "num_input_tokens_seen": 3187193344,
      "step": 11210,
      "train_runtime": 109011.7762,
      "train_tokens_per_second": 29237.147
    },
    {
      "epoch": 0.39698633590369015,
      "grad_norm": 0.5703125,
      "learning_rate": 3.4324763052258835e-05,
      "loss": 0.4131749153137207,
      "num_input_tokens_seen": 3190018944,
      "step": 11220,
      "train_runtime": 109105.8799,
      "train_tokens_per_second": 29237.828
    },
    {
      "epoch": 0.3973401561674189,
      "grad_norm": 0.59375,
      "learning_rate": 3.4316677695858003e-05,
      "loss": 0.4115592002868652,
      "num_input_tokens_seen": 3192876288,
      "step": 11230,
      "train_runtime": 109205.9649,
      "train_tokens_per_second": 29237.197
    },
    {
      "epoch": 0.3976939764311477,
      "grad_norm": 0.5703125,
      "learning_rate": 3.430859805039583e-05,
      "loss": 0.41374759674072265,
      "num_input_tokens_seen": 3195720640,
      "step": 11240,
      "train_runtime": 109304.3384,
      "train_tokens_per_second": 29236.906
    },
    {
      "epoch": 0.39804779669487644,
      "grad_norm": 0.5703125,
      "learning_rate": 3.430052410915246e-05,
      "loss": 0.4114526271820068,
      "num_input_tokens_seen": 3198522880,
      "step": 11250,
      "train_runtime": 109397.5698,
      "train_tokens_per_second": 29237.604
    },
    {
      "epoch": 0.39840161695860526,
      "grad_norm": 0.55859375,
      "learning_rate": 3.4292455865419086e-05,
      "loss": 0.41596226692199706,
      "num_input_tokens_seen": 3201374656,
      "step": 11260,
      "train_runtime": 109495.2804,
      "train_tokens_per_second": 29237.558
    },
    {
      "epoch": 0.39875543722233403,
      "grad_norm": 0.58984375,
      "learning_rate": 3.4284393312497973e-05,
      "loss": 0.41753544807434084,
      "num_input_tokens_seen": 3204221120,
      "step": 11270,
      "train_runtime": 109597.4499,
      "train_tokens_per_second": 29236.274
    },
    {
      "epoch": 0.3991092574860628,
      "grad_norm": 0.5625,
      "learning_rate": 3.427633644370238e-05,
      "loss": 0.4173128128051758,
      "num_input_tokens_seen": 3207060992,
      "step": 11280,
      "train_runtime": 109692.2366,
      "train_tokens_per_second": 29236.9
    },
    {
      "epoch": 0.39946307774979156,
      "grad_norm": 0.5703125,
      "learning_rate": 3.4268285252356564e-05,
      "loss": 0.4158421516418457,
      "num_input_tokens_seen": 3209919680,
      "step": 11290,
      "train_runtime": 109790.1773,
      "train_tokens_per_second": 29236.857
    },
    {
      "epoch": 0.3998168980135204,
      "grad_norm": 0.56640625,
      "learning_rate": 3.426023973179575e-05,
      "loss": 0.4125174045562744,
      "num_input_tokens_seen": 3212786368,
      "step": 11300,
      "train_runtime": 109891.5353,
      "train_tokens_per_second": 29235.977
    },
    {
      "epoch": 0.40017071827724915,
      "grad_norm": 0.5703125,
      "learning_rate": 3.425219987536614e-05,
      "loss": 0.4186762809753418,
      "num_input_tokens_seen": 3215634944,
      "step": 11310,
      "train_runtime": 109988.3257,
      "train_tokens_per_second": 29236.148
    },
    {
      "epoch": 0.4005245385409779,
      "grad_norm": 0.57421875,
      "learning_rate": 3.4244165676424815e-05,
      "loss": 0.4166566848754883,
      "num_input_tokens_seen": 3218476672,
      "step": 11320,
      "train_runtime": 110083.396,
      "train_tokens_per_second": 29236.713
    },
    {
      "epoch": 0.4008783588047067,
      "grad_norm": 0.5625,
      "learning_rate": 3.423613712833979e-05,
      "loss": 0.4160191535949707,
      "num_input_tokens_seen": 3221314432,
      "step": 11330,
      "train_runtime": 110180.7442,
      "train_tokens_per_second": 29236.637
    },
    {
      "epoch": 0.40123217906843545,
      "grad_norm": 0.55859375,
      "learning_rate": 3.422811422448995e-05,
      "loss": 0.4123242378234863,
      "num_input_tokens_seen": 3224176704,
      "step": 11340,
      "train_runtime": 110279.8559,
      "train_tokens_per_second": 29236.316
    },
    {
      "epoch": 0.40158599933216427,
      "grad_norm": 0.59375,
      "learning_rate": 3.422009695826503e-05,
      "loss": 0.4157470703125,
      "num_input_tokens_seen": 3226992384,
      "step": 11350,
      "train_runtime": 110376.671,
      "train_tokens_per_second": 29236.181
    },
    {
      "epoch": 0.40193981959589303,
      "grad_norm": 0.56640625,
      "learning_rate": 3.4212085323065626e-05,
      "loss": 0.41890764236450195,
      "num_input_tokens_seen": 3229866496,
      "step": 11360,
      "train_runtime": 110476.103,
      "train_tokens_per_second": 29235.884
    },
    {
      "epoch": 0.4022936398596218,
      "grad_norm": 0.57421875,
      "learning_rate": 3.4204079312303103e-05,
      "loss": 0.4125824928283691,
      "num_input_tokens_seen": 3232696320,
      "step": 11370,
      "train_runtime": 110568.7335,
      "train_tokens_per_second": 29236.984
    },
    {
      "epoch": 0.40264746012335056,
      "grad_norm": 0.5859375,
      "learning_rate": 3.419607891939964e-05,
      "loss": 0.4136549949645996,
      "num_input_tokens_seen": 3235577024,
      "step": 11380,
      "train_runtime": 110668.3193,
      "train_tokens_per_second": 29236.705
    },
    {
      "epoch": 0.4030012803870794,
      "grad_norm": 0.58203125,
      "learning_rate": 3.4188084137788166e-05,
      "loss": 0.4150240421295166,
      "num_input_tokens_seen": 3238419904,
      "step": 11390,
      "train_runtime": 110764.7287,
      "train_tokens_per_second": 29236.924
    },
    {
      "epoch": 0.40335510065080815,
      "grad_norm": 0.59375,
      "learning_rate": 3.418009496091238e-05,
      "loss": 0.41760807037353515,
      "num_input_tokens_seen": 3241216512,
      "step": 11400,
      "train_runtime": 110857.656,
      "train_tokens_per_second": 29237.642
    },
    {
      "epoch": 0.4037089209145369,
      "grad_norm": 0.58984375,
      "learning_rate": 3.417211138222666e-05,
      "loss": 0.4124875545501709,
      "num_input_tokens_seen": 3244043904,
      "step": 11410,
      "train_runtime": 110953.3611,
      "train_tokens_per_second": 29237.906
    },
    {
      "epoch": 0.4040627411782657,
      "grad_norm": 0.58984375,
      "learning_rate": 3.416413339519612e-05,
      "loss": 0.4189607620239258,
      "num_input_tokens_seen": 3246944704,
      "step": 11420,
      "train_runtime": 111056.2458,
      "train_tokens_per_second": 29236.939
    },
    {
      "epoch": 0.4044165614419945,
      "grad_norm": 0.58984375,
      "learning_rate": 3.4156160993296524e-05,
      "loss": 0.41456027030944825,
      "num_input_tokens_seen": 3249768064,
      "step": 11430,
      "train_runtime": 111152.3585,
      "train_tokens_per_second": 29237.059
    },
    {
      "epoch": 0.40477038170572327,
      "grad_norm": 0.56640625,
      "learning_rate": 3.4148194170014295e-05,
      "loss": 0.4132650852203369,
      "num_input_tokens_seen": 3252589824,
      "step": 11440,
      "train_runtime": 111248.3799,
      "train_tokens_per_second": 29237.188
    },
    {
      "epoch": 0.40512420196945204,
      "grad_norm": 0.5703125,
      "learning_rate": 3.4140232918846484e-05,
      "loss": 0.4124772071838379,
      "num_input_tokens_seen": 3255463808,
      "step": 11450,
      "train_runtime": 111346.0604,
      "train_tokens_per_second": 29237.351
    },
    {
      "epoch": 0.4054780222331808,
      "grad_norm": 0.5859375,
      "learning_rate": 3.4132277233300753e-05,
      "loss": 0.406711483001709,
      "num_input_tokens_seen": 3258240768,
      "step": 11460,
      "train_runtime": 111437.1351,
      "train_tokens_per_second": 29238.375
    },
    {
      "epoch": 0.4058318424969096,
      "grad_norm": 0.58203125,
      "learning_rate": 3.4124327106895356e-05,
      "loss": 0.4144904136657715,
      "num_input_tokens_seen": 3261076992,
      "step": 11470,
      "train_runtime": 111533.2897,
      "train_tokens_per_second": 29238.598
    },
    {
      "epoch": 0.4061856627606384,
      "grad_norm": 0.58203125,
      "learning_rate": 3.4116382533159097e-05,
      "loss": 0.4160325050354004,
      "num_input_tokens_seen": 3263949568,
      "step": 11480,
      "train_runtime": 111631.8354,
      "train_tokens_per_second": 29238.519
    },
    {
      "epoch": 0.40653948302436715,
      "grad_norm": 0.55859375,
      "learning_rate": 3.4108443505631335e-05,
      "loss": 0.4144599914550781,
      "num_input_tokens_seen": 3266796544,
      "step": 11490,
      "train_runtime": 111728.9693,
      "train_tokens_per_second": 29238.581
    },
    {
      "epoch": 0.4068933032880959,
      "grad_norm": 0.57421875,
      "learning_rate": 3.410051001786192e-05,
      "loss": 0.4143384456634521,
      "num_input_tokens_seen": 3269649216,
      "step": 11500,
      "train_runtime": 111829.037,
      "train_tokens_per_second": 29237.927
    },
    {
      "epoch": 0.40724712355182474,
      "grad_norm": 0.56640625,
      "learning_rate": 3.409258206341124e-05,
      "loss": 0.41516814231872556,
      "num_input_tokens_seen": 3272512320,
      "step": 11510,
      "train_runtime": 111928.3504,
      "train_tokens_per_second": 29237.564
    },
    {
      "epoch": 0.4076009438155535,
      "grad_norm": 0.5703125,
      "learning_rate": 3.4084659635850134e-05,
      "loss": 0.41497111320495605,
      "num_input_tokens_seen": 3275336384,
      "step": 11520,
      "train_runtime": 112025.329,
      "train_tokens_per_second": 29237.463
    },
    {
      "epoch": 0.4079547640792823,
      "grad_norm": 0.60546875,
      "learning_rate": 3.40767427287599e-05,
      "loss": 0.4122370719909668,
      "num_input_tokens_seen": 3278275648,
      "step": 11530,
      "train_runtime": 112127.7816,
      "train_tokens_per_second": 29236.962
    },
    {
      "epoch": 0.40830858434301104,
      "grad_norm": 0.55859375,
      "learning_rate": 3.406883133573224e-05,
      "loss": 0.4149974822998047,
      "num_input_tokens_seen": 3281109184,
      "step": 11540,
      "train_runtime": 112226.1532,
      "train_tokens_per_second": 29236.582
    },
    {
      "epoch": 0.40866240460673986,
      "grad_norm": 0.578125,
      "learning_rate": 3.406092545036932e-05,
      "loss": 0.4140750885009766,
      "num_input_tokens_seen": 3283960064,
      "step": 11550,
      "train_runtime": 112322.5364,
      "train_tokens_per_second": 29236.876
    },
    {
      "epoch": 0.4090162248704686,
      "grad_norm": 0.578125,
      "learning_rate": 3.405302506628365e-05,
      "loss": 0.4131946563720703,
      "num_input_tokens_seen": 3286808128,
      "step": 11560,
      "train_runtime": 112419.7575,
      "train_tokens_per_second": 29236.926
    },
    {
      "epoch": 0.4093700451341974,
      "grad_norm": 0.55078125,
      "learning_rate": 3.404513017709813e-05,
      "loss": 0.4137443542480469,
      "num_input_tokens_seen": 3289673792,
      "step": 11570,
      "train_runtime": 112518.3897,
      "train_tokens_per_second": 29236.766
    },
    {
      "epoch": 0.40972386539792616,
      "grad_norm": 0.578125,
      "learning_rate": 3.403724077644598e-05,
      "loss": 0.4137574195861816,
      "num_input_tokens_seen": 3292509696,
      "step": 11580,
      "train_runtime": 112613.3476,
      "train_tokens_per_second": 29237.295
    },
    {
      "epoch": 0.4100776856616549,
      "grad_norm": 0.5546875,
      "learning_rate": 3.402935685797077e-05,
      "loss": 0.41542987823486327,
      "num_input_tokens_seen": 3295443840,
      "step": 11590,
      "train_runtime": 112718.0795,
      "train_tokens_per_second": 29236.16
    },
    {
      "epoch": 0.41043150592538374,
      "grad_norm": 0.60546875,
      "learning_rate": 3.4021478415326355e-05,
      "loss": 0.41658792495727537,
      "num_input_tokens_seen": 3298356224,
      "step": 11600,
      "train_runtime": 112822.1544,
      "train_tokens_per_second": 29235.005
    },
    {
      "epoch": 0.4107853261891125,
      "grad_norm": 0.5859375,
      "learning_rate": 3.401360544217687e-05,
      "loss": 0.41453099250793457,
      "num_input_tokens_seen": 3301193152,
      "step": 11610,
      "train_runtime": 112919.0471,
      "train_tokens_per_second": 29235.043
    },
    {
      "epoch": 0.4111391464528413,
      "grad_norm": 0.58984375,
      "learning_rate": 3.400573793219672e-05,
      "loss": 0.41672253608703613,
      "num_input_tokens_seen": 3304026112,
      "step": 11620,
      "train_runtime": 113016.0858,
      "train_tokens_per_second": 29235.007
    },
    {
      "epoch": 0.41149296671657004,
      "grad_norm": 0.5625,
      "learning_rate": 3.3997875879070546e-05,
      "loss": 0.41640443801879884,
      "num_input_tokens_seen": 3306898304,
      "step": 11630,
      "train_runtime": 113113.0236,
      "train_tokens_per_second": 29235.345
    },
    {
      "epoch": 0.41184678698029886,
      "grad_norm": 0.55859375,
      "learning_rate": 3.399001927649318e-05,
      "loss": 0.4163402557373047,
      "num_input_tokens_seen": 3309742208,
      "step": 11640,
      "train_runtime": 113212.4799,
      "train_tokens_per_second": 29234.782
    },
    {
      "epoch": 0.41220060724402763,
      "grad_norm": 0.57421875,
      "learning_rate": 3.398216811816968e-05,
      "loss": 0.4114383697509766,
      "num_input_tokens_seen": 3312578688,
      "step": 11650,
      "train_runtime": 113308.7347,
      "train_tokens_per_second": 29234.981
    },
    {
      "epoch": 0.4125544275077564,
      "grad_norm": 0.59765625,
      "learning_rate": 3.397432239781527e-05,
      "loss": 0.41254415512084963,
      "num_input_tokens_seen": 3315393024,
      "step": 11660,
      "train_runtime": 113405.2368,
      "train_tokens_per_second": 29234.92
    },
    {
      "epoch": 0.41290824777148516,
      "grad_norm": 0.58203125,
      "learning_rate": 3.396648210915531e-05,
      "loss": 0.4135406494140625,
      "num_input_tokens_seen": 3318191872,
      "step": 11670,
      "train_runtime": 113502.8791,
      "train_tokens_per_second": 29234.429
    },
    {
      "epoch": 0.413262068035214,
      "grad_norm": 0.58203125,
      "learning_rate": 3.3958647245925315e-05,
      "loss": 0.4148995399475098,
      "num_input_tokens_seen": 3321021888,
      "step": 11680,
      "train_runtime": 113601.2675,
      "train_tokens_per_second": 29234.021
    },
    {
      "epoch": 0.41361588829894275,
      "grad_norm": 0.56640625,
      "learning_rate": 3.3950817801870885e-05,
      "loss": 0.4123344421386719,
      "num_input_tokens_seen": 3323904960,
      "step": 11690,
      "train_runtime": 113702.991,
      "train_tokens_per_second": 29233.224
    },
    {
      "epoch": 0.4139697085626715,
      "grad_norm": 0.55078125,
      "learning_rate": 3.3942993770747735e-05,
      "loss": 0.410841703414917,
      "num_input_tokens_seen": 3326796480,
      "step": 11700,
      "train_runtime": 113801.4675,
      "train_tokens_per_second": 29233.336
    },
    {
      "epoch": 0.4143235288264003,
      "grad_norm": 0.5625,
      "learning_rate": 3.3935175146321626e-05,
      "loss": 0.416718578338623,
      "num_input_tokens_seen": 3329629056,
      "step": 11710,
      "train_runtime": 113898.1458,
      "train_tokens_per_second": 29233.391
    },
    {
      "epoch": 0.4146773490901291,
      "grad_norm": 0.578125,
      "learning_rate": 3.392736192236839e-05,
      "loss": 0.4115481376647949,
      "num_input_tokens_seen": 3332476864,
      "step": 11720,
      "train_runtime": 113993.6644,
      "train_tokens_per_second": 29233.878
    },
    {
      "epoch": 0.41503116935385787,
      "grad_norm": 0.58984375,
      "learning_rate": 3.391955409267387e-05,
      "loss": 0.4115856170654297,
      "num_input_tokens_seen": 3335266304,
      "step": 11730,
      "train_runtime": 114086.7436,
      "train_tokens_per_second": 29234.477
    },
    {
      "epoch": 0.41538498961758663,
      "grad_norm": 0.58203125,
      "learning_rate": 3.3911751651033896e-05,
      "loss": 0.41041812896728513,
      "num_input_tokens_seen": 3338176256,
      "step": 11740,
      "train_runtime": 114189.4387,
      "train_tokens_per_second": 29233.669
    },
    {
      "epoch": 0.4157388098813154,
      "grad_norm": 0.56640625,
      "learning_rate": 3.3903954591254334e-05,
      "loss": 0.41042661666870117,
      "num_input_tokens_seen": 3341021696,
      "step": 11750,
      "train_runtime": 114286.9775,
      "train_tokens_per_second": 29233.617
    },
    {
      "epoch": 0.4160926301450442,
      "grad_norm": 0.5703125,
      "learning_rate": 3.389616290715097e-05,
      "loss": 0.4141341209411621,
      "num_input_tokens_seen": 3343843200,
      "step": 11760,
      "train_runtime": 114384.4514,
      "train_tokens_per_second": 29233.372
    },
    {
      "epoch": 0.416446450408773,
      "grad_norm": 0.5703125,
      "learning_rate": 3.388837659254955e-05,
      "loss": 0.411664342880249,
      "num_input_tokens_seen": 3346653568,
      "step": 11770,
      "train_runtime": 114478.7255,
      "train_tokens_per_second": 29233.847
    },
    {
      "epoch": 0.41680027067250175,
      "grad_norm": 0.578125,
      "learning_rate": 3.3880595641285746e-05,
      "loss": 0.41227002143859864,
      "num_input_tokens_seen": 3349494080,
      "step": 11780,
      "train_runtime": 114574.7575,
      "train_tokens_per_second": 29234.136
    },
    {
      "epoch": 0.4171540909362305,
      "grad_norm": 0.61328125,
      "learning_rate": 3.387282004720513e-05,
      "loss": 0.4113319396972656,
      "num_input_tokens_seen": 3352347968,
      "step": 11790,
      "train_runtime": 114672.7997,
      "train_tokens_per_second": 29234.029
    },
    {
      "epoch": 0.41750791119995934,
      "grad_norm": 0.57421875,
      "learning_rate": 3.386504980416316e-05,
      "loss": 0.41370477676391604,
      "num_input_tokens_seen": 3355214848,
      "step": 11800,
      "train_runtime": 114770.1068,
      "train_tokens_per_second": 29234.223
    },
    {
      "epoch": 0.4178617314636881,
      "grad_norm": 0.57421875,
      "learning_rate": 3.385728490602515e-05,
      "loss": 0.40928473472595217,
      "num_input_tokens_seen": 3358055872,
      "step": 11810,
      "train_runtime": 114866.4605,
      "train_tokens_per_second": 29234.433
    },
    {
      "epoch": 0.41821555172741687,
      "grad_norm": 0.55859375,
      "learning_rate": 3.384952534666625e-05,
      "loss": 0.41118149757385253,
      "num_input_tokens_seen": 3360913728,
      "step": 11820,
      "train_runtime": 114963.7402,
      "train_tokens_per_second": 29234.554
    },
    {
      "epoch": 0.41856937199114563,
      "grad_norm": 0.5859375,
      "learning_rate": 3.3841771119971455e-05,
      "loss": 0.4141855716705322,
      "num_input_tokens_seen": 3363813568,
      "step": 11830,
      "train_runtime": 115065.9683,
      "train_tokens_per_second": 29233.783
    },
    {
      "epoch": 0.41892319225487445,
      "grad_norm": 0.57421875,
      "learning_rate": 3.383402221983554e-05,
      "loss": 0.4161659240722656,
      "num_input_tokens_seen": 3366657472,
      "step": 11840,
      "train_runtime": 115164.7511,
      "train_tokens_per_second": 29233.402
    },
    {
      "epoch": 0.4192770125186032,
      "grad_norm": 0.55859375,
      "learning_rate": 3.3826278640163064e-05,
      "loss": 0.41323318481445315,
      "num_input_tokens_seen": 3369497024,
      "step": 11850,
      "train_runtime": 115261.7681,
      "train_tokens_per_second": 29233.432
    },
    {
      "epoch": 0.419630832782332,
      "grad_norm": 0.57421875,
      "learning_rate": 3.3818540374868354e-05,
      "loss": 0.40776491165161133,
      "num_input_tokens_seen": 3372334208,
      "step": 11860,
      "train_runtime": 115357.53,
      "train_tokens_per_second": 29233.759
    },
    {
      "epoch": 0.41998465304606075,
      "grad_norm": 0.58203125,
      "learning_rate": 3.381080741787547e-05,
      "loss": 0.4171136856079102,
      "num_input_tokens_seen": 3375107712,
      "step": 11870,
      "train_runtime": 115450.9685,
      "train_tokens_per_second": 29234.122
    },
    {
      "epoch": 0.4203384733097895,
      "grad_norm": 0.578125,
      "learning_rate": 3.38030797631182e-05,
      "loss": 0.4164409637451172,
      "num_input_tokens_seen": 3377940096,
      "step": 11880,
      "train_runtime": 115546.4689,
      "train_tokens_per_second": 29234.473
    },
    {
      "epoch": 0.42069229357351834,
      "grad_norm": 0.5859375,
      "learning_rate": 3.379535740454003e-05,
      "loss": 0.4124338150024414,
      "num_input_tokens_seen": 3380774976,
      "step": 11890,
      "train_runtime": 115643.5731,
      "train_tokens_per_second": 29234.439
    },
    {
      "epoch": 0.4210461138372471,
      "grad_norm": 0.55859375,
      "learning_rate": 3.3787640336094126e-05,
      "loss": 0.4098670959472656,
      "num_input_tokens_seen": 3383618432,
      "step": 11900,
      "train_runtime": 115743.5378,
      "train_tokens_per_second": 29233.757
    },
    {
      "epoch": 0.42139993410097587,
      "grad_norm": 0.57421875,
      "learning_rate": 3.3779928551743325e-05,
      "loss": 0.4128416061401367,
      "num_input_tokens_seen": 3386487296,
      "step": 11910,
      "train_runtime": 115842.4194,
      "train_tokens_per_second": 29233.568
    },
    {
      "epoch": 0.42175375436470464,
      "grad_norm": 0.58984375,
      "learning_rate": 3.3772222045460084e-05,
      "loss": 0.4113110065460205,
      "num_input_tokens_seen": 3389325760,
      "step": 11920,
      "train_runtime": 115939.8054,
      "train_tokens_per_second": 29233.495
    },
    {
      "epoch": 0.42210757462843346,
      "grad_norm": 0.5546875,
      "learning_rate": 3.37645208112265e-05,
      "loss": 0.4120922565460205,
      "num_input_tokens_seen": 3392209920,
      "step": 11930,
      "train_runtime": 116041.1666,
      "train_tokens_per_second": 29232.815
    },
    {
      "epoch": 0.4224613948921622,
      "grad_norm": 0.56640625,
      "learning_rate": 3.3756824843034255e-05,
      "loss": 0.41185483932495115,
      "num_input_tokens_seen": 3395060672,
      "step": 11940,
      "train_runtime": 116136.7831,
      "train_tokens_per_second": 29233.294
    },
    {
      "epoch": 0.422815215155891,
      "grad_norm": 0.59765625,
      "learning_rate": 3.374913413488464e-05,
      "loss": 0.40811591148376464,
      "num_input_tokens_seen": 3397972864,
      "step": 11950,
      "train_runtime": 116238.346,
      "train_tokens_per_second": 29232.805
    },
    {
      "epoch": 0.42316903541961975,
      "grad_norm": 0.56640625,
      "learning_rate": 3.374144868078848e-05,
      "loss": 0.41322779655456543,
      "num_input_tokens_seen": 3400848320,
      "step": 11960,
      "train_runtime": 116339.4028,
      "train_tokens_per_second": 29232.128
    },
    {
      "epoch": 0.4235228556833486,
      "grad_norm": 0.59375,
      "learning_rate": 3.373376847476615e-05,
      "loss": 0.4113899230957031,
      "num_input_tokens_seen": 3403706240,
      "step": 11970,
      "train_runtime": 116438.9204,
      "train_tokens_per_second": 29231.688
    },
    {
      "epoch": 0.42387667594707734,
      "grad_norm": 0.55078125,
      "learning_rate": 3.3726093510847566e-05,
      "loss": 0.4134838104248047,
      "num_input_tokens_seen": 3406595200,
      "step": 11980,
      "train_runtime": 116536.7184,
      "train_tokens_per_second": 29231.947
    },
    {
      "epoch": 0.4242304962108061,
      "grad_norm": 0.56640625,
      "learning_rate": 3.371842378307212e-05,
      "loss": 0.40941553115844725,
      "num_input_tokens_seen": 3409425088,
      "step": 11990,
      "train_runtime": 116631.7065,
      "train_tokens_per_second": 29232.403
    },
    {
      "epoch": 0.4245843164745349,
      "grad_norm": 0.58203125,
      "learning_rate": 3.371075928548872e-05,
      "loss": 0.4122337818145752,
      "num_input_tokens_seen": 3412251712,
      "step": 12000,
      "train_runtime": 116726.7847,
      "train_tokens_per_second": 29232.808
    },
    {
      "epoch": 0.4249381367382637,
      "grad_norm": 0.5625,
      "learning_rate": 3.37031000121557e-05,
      "loss": 0.4123872756958008,
      "num_input_tokens_seen": 3415124224,
      "step": 12010,
      "train_runtime": 116827.0695,
      "train_tokens_per_second": 29232.302
    },
    {
      "epoch": 0.42529195700199246,
      "grad_norm": 0.55859375,
      "learning_rate": 3.369544595714088e-05,
      "loss": 0.4161685943603516,
      "num_input_tokens_seen": 3417996992,
      "step": 12020,
      "train_runtime": 116924.9032,
      "train_tokens_per_second": 29232.412
    },
    {
      "epoch": 0.4256457772657212,
      "grad_norm": 0.578125,
      "learning_rate": 3.368779711452148e-05,
      "loss": 0.41042294502258303,
      "num_input_tokens_seen": 3420820992,
      "step": 12030,
      "train_runtime": 117022.265,
      "train_tokens_per_second": 29232.223
    },
    {
      "epoch": 0.42599959752945,
      "grad_norm": 0.60546875,
      "learning_rate": 3.368015347838413e-05,
      "loss": 0.4107240676879883,
      "num_input_tokens_seen": 3423679168,
      "step": 12040,
      "train_runtime": 117119.4703,
      "train_tokens_per_second": 29232.366
    },
    {
      "epoch": 0.4263534177931788,
      "grad_norm": 0.59375,
      "learning_rate": 3.3672515042824855e-05,
      "loss": 0.4185525894165039,
      "num_input_tokens_seen": 3426460544,
      "step": 12050,
      "train_runtime": 117212.6937,
      "train_tokens_per_second": 29232.845
    },
    {
      "epoch": 0.4267072380569076,
      "grad_norm": 0.56640625,
      "learning_rate": 3.366488180194904e-05,
      "loss": 0.41394634246826173,
      "num_input_tokens_seen": 3429307456,
      "step": 12060,
      "train_runtime": 117312.3752,
      "train_tokens_per_second": 29232.274
    },
    {
      "epoch": 0.42706105832063634,
      "grad_norm": 0.56640625,
      "learning_rate": 3.365725374987143e-05,
      "loss": 0.4118229389190674,
      "num_input_tokens_seen": 3432160064,
      "step": 12070,
      "train_runtime": 117411.3888,
      "train_tokens_per_second": 29231.918
    },
    {
      "epoch": 0.4274148785843651,
      "grad_norm": 0.66796875,
      "learning_rate": 3.36496308807161e-05,
      "loss": 0.41736268997192383,
      "num_input_tokens_seen": 3434977472,
      "step": 12080,
      "train_runtime": 117507.87,
      "train_tokens_per_second": 29231.893
    },
    {
      "epoch": 0.42776869884809393,
      "grad_norm": 0.56640625,
      "learning_rate": 3.3642013188616426e-05,
      "loss": 0.41408915519714357,
      "num_input_tokens_seen": 3437811840,
      "step": 12090,
      "train_runtime": 117603.6562,
      "train_tokens_per_second": 29232.185
    },
    {
      "epoch": 0.4281225191118227,
      "grad_norm": 0.57421875,
      "learning_rate": 3.3634400667715074e-05,
      "loss": 0.4125615119934082,
      "num_input_tokens_seen": 3440651072,
      "step": 12100,
      "train_runtime": 117700.196,
      "train_tokens_per_second": 29232.331
    },
    {
      "epoch": 0.42847633937555146,
      "grad_norm": 0.58203125,
      "learning_rate": 3.3626793312164013e-05,
      "loss": 0.41154794692993163,
      "num_input_tokens_seen": 3443467200,
      "step": 12110,
      "train_runtime": 117796.3602,
      "train_tokens_per_second": 29232.374
    },
    {
      "epoch": 0.42883015963928023,
      "grad_norm": 0.5859375,
      "learning_rate": 3.361919111612443e-05,
      "loss": 0.4136035919189453,
      "num_input_tokens_seen": 3446304064,
      "step": 12120,
      "train_runtime": 117895.2356,
      "train_tokens_per_second": 29231.92
    },
    {
      "epoch": 0.429183979903009,
      "grad_norm": 0.58203125,
      "learning_rate": 3.361159407376678e-05,
      "loss": 0.41530570983886717,
      "num_input_tokens_seen": 3449141056,
      "step": 12130,
      "train_runtime": 117991.8594,
      "train_tokens_per_second": 29232.026
    },
    {
      "epoch": 0.4295378001667378,
      "grad_norm": 0.5625,
      "learning_rate": 3.3604002179270685e-05,
      "loss": 0.41527385711669923,
      "num_input_tokens_seen": 3451978496,
      "step": 12140,
      "train_runtime": 118089.1445,
      "train_tokens_per_second": 29231.971
    },
    {
      "epoch": 0.4298916204304666,
      "grad_norm": 0.57421875,
      "learning_rate": 3.359641542682504e-05,
      "loss": 0.41362857818603516,
      "num_input_tokens_seen": 3454860928,
      "step": 12150,
      "train_runtime": 118188.4612,
      "train_tokens_per_second": 29231.796
    },
    {
      "epoch": 0.43024544069419535,
      "grad_norm": 0.59375,
      "learning_rate": 3.3588833810627854e-05,
      "loss": 0.41366729736328123,
      "num_input_tokens_seen": 3457707904,
      "step": 12160,
      "train_runtime": 118284.5137,
      "train_tokens_per_second": 29232.127
    },
    {
      "epoch": 0.4305992609579241,
      "grad_norm": 0.55859375,
      "learning_rate": 3.358125732488632e-05,
      "loss": 0.41401143074035646,
      "num_input_tokens_seen": 3460558272,
      "step": 12170,
      "train_runtime": 118379.1147,
      "train_tokens_per_second": 29232.845
    },
    {
      "epoch": 0.43095308122165293,
      "grad_norm": 0.57421875,
      "learning_rate": 3.357368596381679e-05,
      "loss": 0.4111454010009766,
      "num_input_tokens_seen": 3463387392,
      "step": 12180,
      "train_runtime": 118477.1854,
      "train_tokens_per_second": 29232.526
    },
    {
      "epoch": 0.4313069014853817,
      "grad_norm": 0.55078125,
      "learning_rate": 3.356611972164471e-05,
      "loss": 0.41094369888305665,
      "num_input_tokens_seen": 3466219328,
      "step": 12190,
      "train_runtime": 118570.5693,
      "train_tokens_per_second": 29233.387
    },
    {
      "epoch": 0.43166072174911047,
      "grad_norm": 0.57421875,
      "learning_rate": 3.355855859260466e-05,
      "loss": 0.4134207248687744,
      "num_input_tokens_seen": 3469031360,
      "step": 12200,
      "train_runtime": 118663.8051,
      "train_tokens_per_second": 29234.115
    },
    {
      "epoch": 0.43201454201283923,
      "grad_norm": 0.59765625,
      "learning_rate": 3.3551002570940285e-05,
      "loss": 0.41342954635620116,
      "num_input_tokens_seen": 3471901376,
      "step": 12210,
      "train_runtime": 118763.4421,
      "train_tokens_per_second": 29233.755
    },
    {
      "epoch": 0.43236836227656805,
      "grad_norm": 0.546875,
      "learning_rate": 3.354345165090431e-05,
      "loss": 0.4202736377716064,
      "num_input_tokens_seen": 3474786368,
      "step": 12220,
      "train_runtime": 118862.6991,
      "train_tokens_per_second": 29233.615
    },
    {
      "epoch": 0.4327221825402968,
      "grad_norm": 0.58203125,
      "learning_rate": 3.3535905826758515e-05,
      "loss": 0.41166110038757325,
      "num_input_tokens_seen": 3477614976,
      "step": 12230,
      "train_runtime": 118956.3897,
      "train_tokens_per_second": 29234.369
    },
    {
      "epoch": 0.4330760028040256,
      "grad_norm": 0.59765625,
      "learning_rate": 3.352836509277369e-05,
      "loss": 0.4089472770690918,
      "num_input_tokens_seen": 3480460416,
      "step": 12240,
      "train_runtime": 119054.0382,
      "train_tokens_per_second": 29234.291
    },
    {
      "epoch": 0.43342982306775435,
      "grad_norm": 0.5859375,
      "learning_rate": 3.352082944322966e-05,
      "loss": 0.41441688537597654,
      "num_input_tokens_seen": 3483293248,
      "step": 12250,
      "train_runtime": 119151.4526,
      "train_tokens_per_second": 29234.165
    },
    {
      "epoch": 0.43378364333148317,
      "grad_norm": 0.5859375,
      "learning_rate": 3.351329887241524e-05,
      "loss": 0.41138763427734376,
      "num_input_tokens_seen": 3486146176,
      "step": 12260,
      "train_runtime": 119248.3364,
      "train_tokens_per_second": 29234.338
    },
    {
      "epoch": 0.43413746359521194,
      "grad_norm": 0.5703125,
      "learning_rate": 3.3505773374628225e-05,
      "loss": 0.41250147819519045,
      "num_input_tokens_seen": 3488975744,
      "step": 12270,
      "train_runtime": 119346.4506,
      "train_tokens_per_second": 29234.013
    },
    {
      "epoch": 0.4344912838589407,
      "grad_norm": 0.5703125,
      "learning_rate": 3.3498252944175354e-05,
      "loss": 0.4172475814819336,
      "num_input_tokens_seen": 3491848384,
      "step": 12280,
      "train_runtime": 119447.0021,
      "train_tokens_per_second": 29233.454
    },
    {
      "epoch": 0.43484510412266947,
      "grad_norm": 0.5859375,
      "learning_rate": 3.3490737575372326e-05,
      "loss": 0.4148406982421875,
      "num_input_tokens_seen": 3494671616,
      "step": 12290,
      "train_runtime": 119541.156,
      "train_tokens_per_second": 29234.046
    },
    {
      "epoch": 0.4351989243863983,
      "grad_norm": 0.578125,
      "learning_rate": 3.348322726254375e-05,
      "loss": 0.40831804275512695,
      "num_input_tokens_seen": 3497450944,
      "step": 12300,
      "train_runtime": 119634.6681,
      "train_tokens_per_second": 29234.427
    },
    {
      "epoch": 0.43555274465012705,
      "grad_norm": 0.5859375,
      "learning_rate": 3.347572200002315e-05,
      "loss": 0.4108287334442139,
      "num_input_tokens_seen": 3500264704,
      "step": 12310,
      "train_runtime": 119731.3464,
      "train_tokens_per_second": 29234.322
    },
    {
      "epoch": 0.4359065649138558,
      "grad_norm": 0.60546875,
      "learning_rate": 3.3468221782152924e-05,
      "loss": 0.4108120918273926,
      "num_input_tokens_seen": 3503073856,
      "step": 12320,
      "train_runtime": 119826.6025,
      "train_tokens_per_second": 29234.525
    },
    {
      "epoch": 0.4362603851775846,
      "grad_norm": 0.58984375,
      "learning_rate": 3.346072660328435e-05,
      "loss": 0.4088562488555908,
      "num_input_tokens_seen": 3505961728,
      "step": 12330,
      "train_runtime": 119926.0077,
      "train_tokens_per_second": 29234.374
    },
    {
      "epoch": 0.4366142054413134,
      "grad_norm": 0.5703125,
      "learning_rate": 3.345323645777756e-05,
      "loss": 0.4123429298400879,
      "num_input_tokens_seen": 3508781760,
      "step": 12340,
      "train_runtime": 120020.6248,
      "train_tokens_per_second": 29234.823
    },
    {
      "epoch": 0.4369680257050422,
      "grad_norm": 0.5703125,
      "learning_rate": 3.34457513400015e-05,
      "loss": 0.4115546703338623,
      "num_input_tokens_seen": 3511625344,
      "step": 12350,
      "train_runtime": 120116.7985,
      "train_tokens_per_second": 29235.089
    },
    {
      "epoch": 0.43732184596877094,
      "grad_norm": 0.6015625,
      "learning_rate": 3.343827124433396e-05,
      "loss": 0.41304798126220704,
      "num_input_tokens_seen": 3514470144,
      "step": 12360,
      "train_runtime": 120213.1841,
      "train_tokens_per_second": 29235.314
    },
    {
      "epoch": 0.4376756662324997,
      "grad_norm": 0.5859375,
      "learning_rate": 3.343079616516151e-05,
      "loss": 0.41658964157104494,
      "num_input_tokens_seen": 3517244992,
      "step": 12370,
      "train_runtime": 120306.7372,
      "train_tokens_per_second": 29235.644
    },
    {
      "epoch": 0.43802948649622847,
      "grad_norm": 0.58203125,
      "learning_rate": 3.3423326096879495e-05,
      "loss": 0.41205539703369143,
      "num_input_tokens_seen": 3520055808,
      "step": 12380,
      "train_runtime": 120402.5103,
      "train_tokens_per_second": 29235.734
    },
    {
      "epoch": 0.4383833067599573,
      "grad_norm": 0.578125,
      "learning_rate": 3.341586103389203e-05,
      "loss": 0.41498050689697263,
      "num_input_tokens_seen": 3522888384,
      "step": 12390,
      "train_runtime": 120498.0423,
      "train_tokens_per_second": 29236.063
    },
    {
      "epoch": 0.43873712702368606,
      "grad_norm": 0.578125,
      "learning_rate": 3.3408400970611995e-05,
      "loss": 0.41040878295898436,
      "num_input_tokens_seen": 3525718848,
      "step": 12400,
      "train_runtime": 120595.1917,
      "train_tokens_per_second": 29235.982
    },
    {
      "epoch": 0.4390909472874148,
      "grad_norm": 0.5703125,
      "learning_rate": 3.340094590146095e-05,
      "loss": 0.41658973693847656,
      "num_input_tokens_seen": 3528588416,
      "step": 12410,
      "train_runtime": 120691.9668,
      "train_tokens_per_second": 29236.315
    },
    {
      "epoch": 0.4394447675511436,
      "grad_norm": 0.5859375,
      "learning_rate": 3.3393495820869215e-05,
      "loss": 0.41256113052368165,
      "num_input_tokens_seen": 3531412928,
      "step": 12420,
      "train_runtime": 120790.331,
      "train_tokens_per_second": 29235.891
    },
    {
      "epoch": 0.4397985878148724,
      "grad_norm": 0.578125,
      "learning_rate": 3.338605072327576e-05,
      "loss": 0.4107187271118164,
      "num_input_tokens_seen": 3534187904,
      "step": 12430,
      "train_runtime": 120884.6733,
      "train_tokens_per_second": 29236.03
    },
    {
      "epoch": 0.4401524080786012,
      "grad_norm": 0.58984375,
      "learning_rate": 3.337861060312827e-05,
      "loss": 0.4129295825958252,
      "num_input_tokens_seen": 3537046400,
      "step": 12440,
      "train_runtime": 120980.0974,
      "train_tokens_per_second": 29236.597
    },
    {
      "epoch": 0.44050622834232994,
      "grad_norm": 0.5859375,
      "learning_rate": 3.337117545488306e-05,
      "loss": 0.4126898765563965,
      "num_input_tokens_seen": 3539862400,
      "step": 12450,
      "train_runtime": 121076.9904,
      "train_tokens_per_second": 29236.458
    },
    {
      "epoch": 0.4408600486060587,
      "grad_norm": 0.578125,
      "learning_rate": 3.33637452730051e-05,
      "loss": 0.4142162322998047,
      "num_input_tokens_seen": 3542731776,
      "step": 12460,
      "train_runtime": 121175.9922,
      "train_tokens_per_second": 29236.251
    },
    {
      "epoch": 0.44121386886978753,
      "grad_norm": 0.57421875,
      "learning_rate": 3.335632005196796e-05,
      "loss": 0.4092580795288086,
      "num_input_tokens_seen": 3545568832,
      "step": 12470,
      "train_runtime": 121273.0965,
      "train_tokens_per_second": 29236.236
    },
    {
      "epoch": 0.4415676891335163,
      "grad_norm": 0.5703125,
      "learning_rate": 3.334889978625383e-05,
      "loss": 0.4097914218902588,
      "num_input_tokens_seen": 3548391872,
      "step": 12480,
      "train_runtime": 121368.1931,
      "train_tokens_per_second": 29236.588
    },
    {
      "epoch": 0.44192150939724506,
      "grad_norm": 0.5625,
      "learning_rate": 3.3341484470353515e-05,
      "loss": 0.4096519470214844,
      "num_input_tokens_seen": 3551287488,
      "step": 12490,
      "train_runtime": 121470.7237,
      "train_tokens_per_second": 29235.748
    },
    {
      "epoch": 0.4422753296609738,
      "grad_norm": 0.59375,
      "learning_rate": 3.333407409876635e-05,
      "loss": 0.4120530128479004,
      "num_input_tokens_seen": 3554146688,
      "step": 12500,
      "train_runtime": 121567.4786,
      "train_tokens_per_second": 29235.999
    },
    {
      "epoch": 0.44262914992470265,
      "grad_norm": 0.59375,
      "learning_rate": 3.332666866600024e-05,
      "loss": 0.4091007232666016,
      "num_input_tokens_seen": 3556976192,
      "step": 12510,
      "train_runtime": 121666.4778,
      "train_tokens_per_second": 29235.466
    },
    {
      "epoch": 0.4429829701884314,
      "grad_norm": 0.5703125,
      "learning_rate": 3.331926816657162e-05,
      "loss": 0.412156867980957,
      "num_input_tokens_seen": 3559771392,
      "step": 12520,
      "train_runtime": 121760.1442,
      "train_tokens_per_second": 29235.933
    },
    {
      "epoch": 0.4433367904521602,
      "grad_norm": 0.59765625,
      "learning_rate": 3.331187259500546e-05,
      "loss": 0.41840124130249023,
      "num_input_tokens_seen": 3562673984,
      "step": 12530,
      "train_runtime": 121862.0695,
      "train_tokens_per_second": 29235.299
    },
    {
      "epoch": 0.44369061071588894,
      "grad_norm": 0.5703125,
      "learning_rate": 3.3304481945835235e-05,
      "loss": 0.4158226490020752,
      "num_input_tokens_seen": 3565495808,
      "step": 12540,
      "train_runtime": 121957.4465,
      "train_tokens_per_second": 29235.573
    },
    {
      "epoch": 0.44404443097961777,
      "grad_norm": 0.58984375,
      "learning_rate": 3.329709621360288e-05,
      "loss": 0.4107212066650391,
      "num_input_tokens_seen": 3568296192,
      "step": 12550,
      "train_runtime": 122050.3287,
      "train_tokens_per_second": 29236.269
    },
    {
      "epoch": 0.44439825124334653,
      "grad_norm": 0.59765625,
      "learning_rate": 3.328971539285882e-05,
      "loss": 0.4103407859802246,
      "num_input_tokens_seen": 3571137408,
      "step": 12560,
      "train_runtime": 122147.9686,
      "train_tokens_per_second": 29236.159
    },
    {
      "epoch": 0.4447520715070753,
      "grad_norm": 0.59375,
      "learning_rate": 3.3282339478161935e-05,
      "loss": 0.41288127899169924,
      "num_input_tokens_seen": 3573937216,
      "step": 12570,
      "train_runtime": 122240.9749,
      "train_tokens_per_second": 29236.819
    },
    {
      "epoch": 0.44510589177080406,
      "grad_norm": 0.56640625,
      "learning_rate": 3.327496846407953e-05,
      "loss": 0.4112571716308594,
      "num_input_tokens_seen": 3576755136,
      "step": 12580,
      "train_runtime": 122334.7996,
      "train_tokens_per_second": 29237.43
    },
    {
      "epoch": 0.4454597120345329,
      "grad_norm": 0.5703125,
      "learning_rate": 3.3267602345187304e-05,
      "loss": 0.416623592376709,
      "num_input_tokens_seen": 3579593728,
      "step": 12590,
      "train_runtime": 122430.3622,
      "train_tokens_per_second": 29237.794
    },
    {
      "epoch": 0.44581353229826165,
      "grad_norm": 0.5625,
      "learning_rate": 3.326024111606942e-05,
      "loss": 0.408983039855957,
      "num_input_tokens_seen": 3582422016,
      "step": 12600,
      "train_runtime": 122525.7936,
      "train_tokens_per_second": 29238.105
    },
    {
      "epoch": 0.4461673525619904,
      "grad_norm": 0.58203125,
      "learning_rate": 3.325288477131839e-05,
      "loss": 0.4170994758605957,
      "num_input_tokens_seen": 3585311552,
      "step": 12610,
      "train_runtime": 122625.6045,
      "train_tokens_per_second": 29237.871
    },
    {
      "epoch": 0.4465211728257192,
      "grad_norm": 0.5703125,
      "learning_rate": 3.324553330553507e-05,
      "loss": 0.4109673500061035,
      "num_input_tokens_seen": 3588124352,
      "step": 12620,
      "train_runtime": 122719.8513,
      "train_tokens_per_second": 29238.337
    },
    {
      "epoch": 0.44687499308944795,
      "grad_norm": 0.578125,
      "learning_rate": 3.323818671332871e-05,
      "loss": 0.41263542175292967,
      "num_input_tokens_seen": 3590967936,
      "step": 12630,
      "train_runtime": 122817.5717,
      "train_tokens_per_second": 29238.226
    },
    {
      "epoch": 0.44722881335317677,
      "grad_norm": 0.609375,
      "learning_rate": 3.323084498931687e-05,
      "loss": 0.41025333404541015,
      "num_input_tokens_seen": 3593794752,
      "step": 12640,
      "train_runtime": 122913.452,
      "train_tokens_per_second": 29238.417
    },
    {
      "epoch": 0.44758263361690553,
      "grad_norm": 0.57421875,
      "learning_rate": 3.322350812812545e-05,
      "loss": 0.4123964309692383,
      "num_input_tokens_seen": 3596652736,
      "step": 12650,
      "train_runtime": 123012.0161,
      "train_tokens_per_second": 29238.223
    },
    {
      "epoch": 0.4479364538806343,
      "grad_norm": 0.578125,
      "learning_rate": 3.321617612438862e-05,
      "loss": 0.41256985664367674,
      "num_input_tokens_seen": 3599468672,
      "step": 12660,
      "train_runtime": 123107.3889,
      "train_tokens_per_second": 29238.445
    },
    {
      "epoch": 0.44829027414436307,
      "grad_norm": 0.578125,
      "learning_rate": 3.320884897274886e-05,
      "loss": 0.4146679401397705,
      "num_input_tokens_seen": 3602314176,
      "step": 12670,
      "train_runtime": 123204.7841,
      "train_tokens_per_second": 29238.428
    },
    {
      "epoch": 0.4486440944080919,
      "grad_norm": 0.5859375,
      "learning_rate": 3.320152666785692e-05,
      "loss": 0.41847925186157225,
      "num_input_tokens_seen": 3605160320,
      "step": 12680,
      "train_runtime": 123299.835,
      "train_tokens_per_second": 29238.971
    },
    {
      "epoch": 0.44899791467182065,
      "grad_norm": 0.625,
      "learning_rate": 3.319420920437179e-05,
      "loss": 0.4097771644592285,
      "num_input_tokens_seen": 3608000448,
      "step": 12690,
      "train_runtime": 123396.0111,
      "train_tokens_per_second": 29239.198
    },
    {
      "epoch": 0.4493517349355494,
      "grad_norm": 0.578125,
      "learning_rate": 3.31868965769607e-05,
      "loss": 0.41338434219360354,
      "num_input_tokens_seen": 3610869056,
      "step": 12700,
      "train_runtime": 123495.345,
      "train_tokens_per_second": 29238.908
    },
    {
      "epoch": 0.4497055551992782,
      "grad_norm": 0.58984375,
      "learning_rate": 3.317958878029911e-05,
      "loss": 0.41217937469482424,
      "num_input_tokens_seen": 3613746304,
      "step": 12710,
      "train_runtime": 123594.8104,
      "train_tokens_per_second": 29238.657
    },
    {
      "epoch": 0.450059375463007,
      "grad_norm": 0.5546875,
      "learning_rate": 3.3172285809070665e-05,
      "loss": 0.41030683517456057,
      "num_input_tokens_seen": 3616641600,
      "step": 12720,
      "train_runtime": 123693.6921,
      "train_tokens_per_second": 29238.691
    },
    {
      "epoch": 0.45041319572673577,
      "grad_norm": 0.58203125,
      "learning_rate": 3.3164987657967214e-05,
      "loss": 0.4161098480224609,
      "num_input_tokens_seen": 3619544960,
      "step": 12730,
      "train_runtime": 123796.6688,
      "train_tokens_per_second": 29237.822
    },
    {
      "epoch": 0.45076701599046454,
      "grad_norm": 0.5625,
      "learning_rate": 3.315769432168877e-05,
      "loss": 0.4102607727050781,
      "num_input_tokens_seen": 3622380864,
      "step": 12740,
      "train_runtime": 123891.5451,
      "train_tokens_per_second": 29238.322
    },
    {
      "epoch": 0.4511208362541933,
      "grad_norm": 0.60546875,
      "learning_rate": 3.315040579494349e-05,
      "loss": 0.41478471755981444,
      "num_input_tokens_seen": 3625226368,
      "step": 12750,
      "train_runtime": 123989.0324,
      "train_tokens_per_second": 29238.283
    },
    {
      "epoch": 0.4514746565179221,
      "grad_norm": 0.56640625,
      "learning_rate": 3.31431220724477e-05,
      "loss": 0.4113792419433594,
      "num_input_tokens_seen": 3628066688,
      "step": 12760,
      "train_runtime": 124083.7917,
      "train_tokens_per_second": 29238.844
    },
    {
      "epoch": 0.4518284767816509,
      "grad_norm": 0.57421875,
      "learning_rate": 3.3135843148925834e-05,
      "loss": 0.4128225326538086,
      "num_input_tokens_seen": 3630867584,
      "step": 12770,
      "train_runtime": 124176.2962,
      "train_tokens_per_second": 29239.619
    },
    {
      "epoch": 0.45218229704537966,
      "grad_norm": 0.55859375,
      "learning_rate": 3.3128569019110414e-05,
      "loss": 0.4140755653381348,
      "num_input_tokens_seen": 3633722176,
      "step": 12780,
      "train_runtime": 124274.8195,
      "train_tokens_per_second": 29239.408
    },
    {
      "epoch": 0.4525361173091084,
      "grad_norm": 0.56640625,
      "learning_rate": 3.312129967774207e-05,
      "loss": 0.41249914169311525,
      "num_input_tokens_seen": 3636590208,
      "step": 12790,
      "train_runtime": 124374.942,
      "train_tokens_per_second": 29238.93
    },
    {
      "epoch": 0.45288993757283724,
      "grad_norm": 0.59765625,
      "learning_rate": 3.311403511956952e-05,
      "loss": 0.4087204933166504,
      "num_input_tokens_seen": 3639380736,
      "step": 12800,
      "train_runtime": 124468.6991,
      "train_tokens_per_second": 29239.325
    },
    {
      "epoch": 0.453243757836566,
      "grad_norm": 0.5703125,
      "learning_rate": 3.310677533934952e-05,
      "loss": 0.41220407485961913,
      "num_input_tokens_seen": 3642170816,
      "step": 12810,
      "train_runtime": 124563.5758,
      "train_tokens_per_second": 29239.453
    },
    {
      "epoch": 0.4535975781002948,
      "grad_norm": 0.59375,
      "learning_rate": 3.309952033184686e-05,
      "loss": 0.4085845947265625,
      "num_input_tokens_seen": 3644998592,
      "step": 12820,
      "train_runtime": 124658.3109,
      "train_tokens_per_second": 29239.916
    },
    {
      "epoch": 0.45395139836402354,
      "grad_norm": 0.5859375,
      "learning_rate": 3.309227009183439e-05,
      "loss": 0.4152225494384766,
      "num_input_tokens_seen": 3647778752,
      "step": 12830,
      "train_runtime": 124750.8097,
      "train_tokens_per_second": 29240.522
    },
    {
      "epoch": 0.45430521862775236,
      "grad_norm": 0.546875,
      "learning_rate": 3.308502461409295e-05,
      "loss": 0.4108767509460449,
      "num_input_tokens_seen": 3650677312,
      "step": 12840,
      "train_runtime": 124849.5664,
      "train_tokens_per_second": 29240.609
    },
    {
      "epoch": 0.4546590388914811,
      "grad_norm": 0.57421875,
      "learning_rate": 3.3077783893411386e-05,
      "loss": 0.4112394332885742,
      "num_input_tokens_seen": 3653550912,
      "step": 12850,
      "train_runtime": 124949.125,
      "train_tokens_per_second": 29240.308
    },
    {
      "epoch": 0.4550128591552099,
      "grad_norm": 0.5546875,
      "learning_rate": 3.30705479245865e-05,
      "loss": 0.41403961181640625,
      "num_input_tokens_seen": 3656410176,
      "step": 12860,
      "train_runtime": 125046.9119,
      "train_tokens_per_second": 29240.308
    },
    {
      "epoch": 0.45536667941893866,
      "grad_norm": 0.57421875,
      "learning_rate": 3.3063316702423094e-05,
      "loss": 0.4169203758239746,
      "num_input_tokens_seen": 3659304768,
      "step": 12870,
      "train_runtime": 125147.7847,
      "train_tokens_per_second": 29239.869
    },
    {
      "epoch": 0.4557204996826674,
      "grad_norm": 0.57421875,
      "learning_rate": 3.305609022173388e-05,
      "loss": 0.41733317375183104,
      "num_input_tokens_seen": 3662148672,
      "step": 12880,
      "train_runtime": 125244.1442,
      "train_tokens_per_second": 29240.079
    },
    {
      "epoch": 0.45607431994639624,
      "grad_norm": 0.58203125,
      "learning_rate": 3.304886847733954e-05,
      "loss": 0.4152997970581055,
      "num_input_tokens_seen": 3665006976,
      "step": 12890,
      "train_runtime": 125341.9795,
      "train_tokens_per_second": 29240.06
    },
    {
      "epoch": 0.456428140210125,
      "grad_norm": 0.578125,
      "learning_rate": 3.304165146406865e-05,
      "loss": 0.4129094123840332,
      "num_input_tokens_seen": 3667843008,
      "step": 12900,
      "train_runtime": 125439.5858,
      "train_tokens_per_second": 29239.916
    },
    {
      "epoch": 0.4567819604738538,
      "grad_norm": 0.57421875,
      "learning_rate": 3.30344391767577e-05,
      "loss": 0.40851707458496095,
      "num_input_tokens_seen": 3670698688,
      "step": 12910,
      "train_runtime": 125537.935,
      "train_tokens_per_second": 29239.757
    },
    {
      "epoch": 0.45713578073758254,
      "grad_norm": 0.59375,
      "learning_rate": 3.302723161025104e-05,
      "loss": 0.41683287620544435,
      "num_input_tokens_seen": 3673513344,
      "step": 12920,
      "train_runtime": 125631.4877,
      "train_tokens_per_second": 29240.387
    },
    {
      "epoch": 0.45748960100131136,
      "grad_norm": 0.5703125,
      "learning_rate": 3.302002875940093e-05,
      "loss": 0.41037778854370116,
      "num_input_tokens_seen": 3676334272,
      "step": 12930,
      "train_runtime": 125726.2488,
      "train_tokens_per_second": 29240.785
    },
    {
      "epoch": 0.45784342126504013,
      "grad_norm": 0.578125,
      "learning_rate": 3.3012830619067466e-05,
      "loss": 0.4098685264587402,
      "num_input_tokens_seen": 3679143232,
      "step": 12940,
      "train_runtime": 125821.3099,
      "train_tokens_per_second": 29241.018
    },
    {
      "epoch": 0.4581972415287689,
      "grad_norm": 0.57421875,
      "learning_rate": 3.300563718411857e-05,
      "loss": 0.4120169639587402,
      "num_input_tokens_seen": 3681976512,
      "step": 12950,
      "train_runtime": 125918.0031,
      "train_tokens_per_second": 29241.065
    },
    {
      "epoch": 0.45855106179249766,
      "grad_norm": 0.57421875,
      "learning_rate": 3.299844844943e-05,
      "loss": 0.40993967056274416,
      "num_input_tokens_seen": 3684819072,
      "step": 12960,
      "train_runtime": 126015.2812,
      "train_tokens_per_second": 29241.049
    },
    {
      "epoch": 0.4589048820562265,
      "grad_norm": 0.5625,
      "learning_rate": 3.299126440988535e-05,
      "loss": 0.4140603065490723,
      "num_input_tokens_seen": 3687670720,
      "step": 12970,
      "train_runtime": 126113.2231,
      "train_tokens_per_second": 29240.952
    },
    {
      "epoch": 0.45925870231995525,
      "grad_norm": 0.5859375,
      "learning_rate": 3.298408506037596e-05,
      "loss": 0.4080381393432617,
      "num_input_tokens_seen": 3690532032,
      "step": 12980,
      "train_runtime": 126212.9587,
      "train_tokens_per_second": 29240.516
    },
    {
      "epoch": 0.459612522583684,
      "grad_norm": 0.5546875,
      "learning_rate": 3.297691039580097e-05,
      "loss": 0.40605506896972654,
      "num_input_tokens_seen": 3693374976,
      "step": 12990,
      "train_runtime": 126310.4971,
      "train_tokens_per_second": 29240.444
    },
    {
      "epoch": 0.4599663428474128,
      "grad_norm": 0.5546875,
      "learning_rate": 3.29697404110673e-05,
      "loss": 0.41342811584472655,
      "num_input_tokens_seen": 3696248128,
      "step": 13000,
      "train_runtime": 126410.4816,
      "train_tokens_per_second": 29240.045
    },
    {
      "epoch": 0.4603201631111416,
      "grad_norm": 0.5859375,
      "learning_rate": 3.2962575101089594e-05,
      "loss": 0.411439323425293,
      "num_input_tokens_seen": 3699071680,
      "step": 13010,
      "train_runtime": 126505.8295,
      "train_tokens_per_second": 29240.326
    },
    {
      "epoch": 0.46067398337487037,
      "grad_norm": 0.58203125,
      "learning_rate": 3.295541446079024e-05,
      "loss": 0.4148221492767334,
      "num_input_tokens_seen": 3701891648,
      "step": 13020,
      "train_runtime": 126601.5584,
      "train_tokens_per_second": 29240.49
    },
    {
      "epoch": 0.46102780363859913,
      "grad_norm": 0.56640625,
      "learning_rate": 3.2948258485099336e-05,
      "loss": 0.41007609367370607,
      "num_input_tokens_seen": 3704756032,
      "step": 13030,
      "train_runtime": 126700.571,
      "train_tokens_per_second": 29240.247
    },
    {
      "epoch": 0.4613816239023279,
      "grad_norm": 0.58984375,
      "learning_rate": 3.29411071689547e-05,
      "loss": 0.41414527893066405,
      "num_input_tokens_seen": 3707661952,
      "step": 13040,
      "train_runtime": 126802.1024,
      "train_tokens_per_second": 29239.751
    },
    {
      "epoch": 0.4617354441660567,
      "grad_norm": 0.578125,
      "learning_rate": 3.2933960507301826e-05,
      "loss": 0.4107569694519043,
      "num_input_tokens_seen": 3710518912,
      "step": 13050,
      "train_runtime": 126900.7478,
      "train_tokens_per_second": 29239.535
    },
    {
      "epoch": 0.4620892644297855,
      "grad_norm": 0.6015625,
      "learning_rate": 3.292681849509387e-05,
      "loss": 0.4128089904785156,
      "num_input_tokens_seen": 3713319168,
      "step": 13060,
      "train_runtime": 126998.3582,
      "train_tokens_per_second": 29239.112
    },
    {
      "epoch": 0.46244308469351425,
      "grad_norm": 0.56640625,
      "learning_rate": 3.291968112729166e-05,
      "loss": 0.4093809127807617,
      "num_input_tokens_seen": 3716157248,
      "step": 13070,
      "train_runtime": 127095.2237,
      "train_tokens_per_second": 29239.157
    },
    {
      "epoch": 0.462796904957243,
      "grad_norm": 0.5703125,
      "learning_rate": 3.291254839886367e-05,
      "loss": 0.40978412628173827,
      "num_input_tokens_seen": 3718953728,
      "step": 13080,
      "train_runtime": 127192.2752,
      "train_tokens_per_second": 29238.833
    },
    {
      "epoch": 0.46315072522097184,
      "grad_norm": 0.6171875,
      "learning_rate": 3.2905420304785995e-05,
      "loss": 0.41519532203674314,
      "num_input_tokens_seen": 3721783488,
      "step": 13090,
      "train_runtime": 127287.9078,
      "train_tokens_per_second": 29239.097
    },
    {
      "epoch": 0.4635045454847006,
      "grad_norm": 0.57421875,
      "learning_rate": 3.289829684004235e-05,
      "loss": 0.41240777969360354,
      "num_input_tokens_seen": 3724641792,
      "step": 13100,
      "train_runtime": 127385.4639,
      "train_tokens_per_second": 29239.143
    },
    {
      "epoch": 0.46385836574842937,
      "grad_norm": 0.56640625,
      "learning_rate": 3.289117799962402e-05,
      "loss": 0.4131345748901367,
      "num_input_tokens_seen": 3727474368,
      "step": 13110,
      "train_runtime": 127481.9478,
      "train_tokens_per_second": 29239.233
    },
    {
      "epoch": 0.46421218601215813,
      "grad_norm": 0.5625,
      "learning_rate": 3.2884063778529914e-05,
      "loss": 0.40851125717163084,
      "num_input_tokens_seen": 3730300608,
      "step": 13120,
      "train_runtime": 127577.3435,
      "train_tokens_per_second": 29239.523
    },
    {
      "epoch": 0.4645660062758869,
      "grad_norm": 0.578125,
      "learning_rate": 3.28769541717665e-05,
      "loss": 0.41296701431274413,
      "num_input_tokens_seen": 3733139328,
      "step": 13130,
      "train_runtime": 127673.4702,
      "train_tokens_per_second": 29239.742
    },
    {
      "epoch": 0.4649198265396157,
      "grad_norm": 0.57421875,
      "learning_rate": 3.2869849174347775e-05,
      "loss": 0.40378413200378416,
      "num_input_tokens_seen": 3736004480,
      "step": 13140,
      "train_runtime": 127770.8835,
      "train_tokens_per_second": 29239.874
    },
    {
      "epoch": 0.4652736468033445,
      "grad_norm": 0.5859375,
      "learning_rate": 3.2862748781295294e-05,
      "loss": 0.4091127872467041,
      "num_input_tokens_seen": 3738816960,
      "step": 13150,
      "train_runtime": 127864.243,
      "train_tokens_per_second": 29240.52
    },
    {
      "epoch": 0.46562746706707325,
      "grad_norm": 0.5625,
      "learning_rate": 3.2855652987638146e-05,
      "loss": 0.41389904022216795,
      "num_input_tokens_seen": 3741696896,
      "step": 13160,
      "train_runtime": 127965.581,
      "train_tokens_per_second": 29239.87
    },
    {
      "epoch": 0.465981287330802,
      "grad_norm": 0.55859375,
      "learning_rate": 3.284856178841291e-05,
      "loss": 0.4076066970825195,
      "num_input_tokens_seen": 3744501888,
      "step": 13170,
      "train_runtime": 128059.821,
      "train_tokens_per_second": 29240.256
    },
    {
      "epoch": 0.46633510759453084,
      "grad_norm": 0.60546875,
      "learning_rate": 3.284147517866367e-05,
      "loss": 0.41181602478027346,
      "num_input_tokens_seen": 3747355968,
      "step": 13180,
      "train_runtime": 128158.839,
      "train_tokens_per_second": 29239.934
    },
    {
      "epoch": 0.4666889278582596,
      "grad_norm": 0.5859375,
      "learning_rate": 3.2834393153441976e-05,
      "loss": 0.4129744529724121,
      "num_input_tokens_seen": 3750236160,
      "step": 13190,
      "train_runtime": 128255.6865,
      "train_tokens_per_second": 29240.311
    },
    {
      "epoch": 0.46704274812198837,
      "grad_norm": 0.55078125,
      "learning_rate": 3.282731570780689e-05,
      "loss": 0.41068115234375,
      "num_input_tokens_seen": 3753081600,
      "step": 13200,
      "train_runtime": 128351.4484,
      "train_tokens_per_second": 29240.664
    },
    {
      "epoch": 0.46739656838571714,
      "grad_norm": 0.625,
      "learning_rate": 3.2820242836824875e-05,
      "loss": 0.41236133575439454,
      "num_input_tokens_seen": 3755909056,
      "step": 13210,
      "train_runtime": 128447.9644,
      "train_tokens_per_second": 29240.705
    },
    {
      "epoch": 0.46775038864944596,
      "grad_norm": 0.5859375,
      "learning_rate": 3.2813174535569854e-05,
      "loss": 0.41091275215148926,
      "num_input_tokens_seen": 3758709056,
      "step": 13220,
      "train_runtime": 128544.0075,
      "train_tokens_per_second": 29240.64
    },
    {
      "epoch": 0.4681042089131747,
      "grad_norm": 0.5703125,
      "learning_rate": 3.280611079912318e-05,
      "loss": 0.40787954330444337,
      "num_input_tokens_seen": 3761523904,
      "step": 13230,
      "train_runtime": 128638.6788,
      "train_tokens_per_second": 29241.002
    },
    {
      "epoch": 0.4684580291769035,
      "grad_norm": 0.57421875,
      "learning_rate": 3.279905162257358e-05,
      "loss": 0.4098822593688965,
      "num_input_tokens_seen": 3764351680,
      "step": 13240,
      "train_runtime": 128732.8753,
      "train_tokens_per_second": 29241.572
    },
    {
      "epoch": 0.46881184944063226,
      "grad_norm": 0.546875,
      "learning_rate": 3.279199700101723e-05,
      "loss": 0.4121509552001953,
      "num_input_tokens_seen": 3767180096,
      "step": 13250,
      "train_runtime": 128827.2957,
      "train_tokens_per_second": 29242.096
    },
    {
      "epoch": 0.4691656697043611,
      "grad_norm": 0.5703125,
      "learning_rate": 3.2784946929557644e-05,
      "loss": 0.40957231521606446,
      "num_input_tokens_seen": 3769964480,
      "step": 13260,
      "train_runtime": 128923.3506,
      "train_tokens_per_second": 29241.906
    },
    {
      "epoch": 0.46951948996808984,
      "grad_norm": 0.57421875,
      "learning_rate": 3.277790140330571e-05,
      "loss": 0.4122472763061523,
      "num_input_tokens_seen": 3772865984,
      "step": 13270,
      "train_runtime": 129022.4343,
      "train_tokens_per_second": 29241.938
    },
    {
      "epoch": 0.4698733102318186,
      "grad_norm": 0.578125,
      "learning_rate": 3.277086041737968e-05,
      "loss": 0.41433162689208985,
      "num_input_tokens_seen": 3775689600,
      "step": 13280,
      "train_runtime": 129116.4148,
      "train_tokens_per_second": 29242.522
    },
    {
      "epoch": 0.4702271304955474,
      "grad_norm": 0.56640625,
      "learning_rate": 3.276382396690513e-05,
      "loss": 0.4116861820220947,
      "num_input_tokens_seen": 3778486464,
      "step": 13290,
      "train_runtime": 129213.1758,
      "train_tokens_per_second": 29242.269
    },
    {
      "epoch": 0.4705809507592762,
      "grad_norm": 0.5703125,
      "learning_rate": 3.275679204701496e-05,
      "loss": 0.41191835403442384,
      "num_input_tokens_seen": 3781317632,
      "step": 13300,
      "train_runtime": 129309.344,
      "train_tokens_per_second": 29242.416
    },
    {
      "epoch": 0.47093477102300496,
      "grad_norm": 0.59765625,
      "learning_rate": 3.274976465284939e-05,
      "loss": 0.4106626510620117,
      "num_input_tokens_seen": 3784164480,
      "step": 13310,
      "train_runtime": 129405.5068,
      "train_tokens_per_second": 29242.685
    },
    {
      "epoch": 0.4712885912867337,
      "grad_norm": 0.58984375,
      "learning_rate": 3.274274177955593e-05,
      "loss": 0.41222991943359377,
      "num_input_tokens_seen": 3786995392,
      "step": 13320,
      "train_runtime": 129500.9483,
      "train_tokens_per_second": 29242.994
    },
    {
      "epoch": 0.4716424115504625,
      "grad_norm": 0.5703125,
      "learning_rate": 3.273572342228937e-05,
      "loss": 0.4107541561126709,
      "num_input_tokens_seen": 3789813440,
      "step": 13330,
      "train_runtime": 129596.7919,
      "train_tokens_per_second": 29243.112
    },
    {
      "epoch": 0.4719962318141913,
      "grad_norm": 0.578125,
      "learning_rate": 3.272870957621176e-05,
      "loss": 0.4079722881317139,
      "num_input_tokens_seen": 3792692928,
      "step": 13340,
      "train_runtime": 129694.0242,
      "train_tokens_per_second": 29243.39
    },
    {
      "epoch": 0.4723500520779201,
      "grad_norm": 0.57421875,
      "learning_rate": 3.2721700236492414e-05,
      "loss": 0.4138794898986816,
      "num_input_tokens_seen": 3795596096,
      "step": 13350,
      "train_runtime": 129796.9856,
      "train_tokens_per_second": 29242.56
    },
    {
      "epoch": 0.47270387234164885,
      "grad_norm": 0.57421875,
      "learning_rate": 3.271469539830788e-05,
      "loss": 0.4129480361938477,
      "num_input_tokens_seen": 3798444096,
      "step": 13360,
      "train_runtime": 129896.2756,
      "train_tokens_per_second": 29242.132
    },
    {
      "epoch": 0.4730576926053776,
      "grad_norm": 0.5859375,
      "learning_rate": 3.270769505684193e-05,
      "loss": 0.41733531951904296,
      "num_input_tokens_seen": 3801311488,
      "step": 13370,
      "train_runtime": 129995.6546,
      "train_tokens_per_second": 29241.835
    },
    {
      "epoch": 0.47341151286910643,
      "grad_norm": 0.6015625,
      "learning_rate": 3.2700699207285544e-05,
      "loss": 0.4079148292541504,
      "num_input_tokens_seen": 3804162368,
      "step": 13380,
      "train_runtime": 130092.3062,
      "train_tokens_per_second": 29242.024
    },
    {
      "epoch": 0.4737653331328352,
      "grad_norm": 0.58203125,
      "learning_rate": 3.269370784483691e-05,
      "loss": 0.4078085899353027,
      "num_input_tokens_seen": 3806985664,
      "step": 13390,
      "train_runtime": 130187.5312,
      "train_tokens_per_second": 29242.322
    },
    {
      "epoch": 0.47411915339656396,
      "grad_norm": 0.55859375,
      "learning_rate": 3.268672096470138e-05,
      "loss": 0.41168937683105467,
      "num_input_tokens_seen": 3809798528,
      "step": 13400,
      "train_runtime": 130282.5991,
      "train_tokens_per_second": 29242.574
    },
    {
      "epoch": 0.47447297366029273,
      "grad_norm": 0.59765625,
      "learning_rate": 3.2679738562091506e-05,
      "loss": 0.4101283073425293,
      "num_input_tokens_seen": 3812646720,
      "step": 13410,
      "train_runtime": 130379.5107,
      "train_tokens_per_second": 29242.683
    },
    {
      "epoch": 0.4748267939240215,
      "grad_norm": 0.58203125,
      "learning_rate": 3.2672760632226964e-05,
      "loss": 0.41565380096435545,
      "num_input_tokens_seen": 3815474560,
      "step": 13420,
      "train_runtime": 130478.0702,
      "train_tokens_per_second": 29242.267
    },
    {
      "epoch": 0.4751806141877503,
      "grad_norm": 0.5859375,
      "learning_rate": 3.266578717033458e-05,
      "loss": 0.4146378993988037,
      "num_input_tokens_seen": 3818355968,
      "step": 13430,
      "train_runtime": 130578.5243,
      "train_tokens_per_second": 29241.837
    },
    {
      "epoch": 0.4755344344514791,
      "grad_norm": 0.57421875,
      "learning_rate": 3.265881817164833e-05,
      "loss": 0.41215057373046876,
      "num_input_tokens_seen": 3821149376,
      "step": 13440,
      "train_runtime": 130671.1267,
      "train_tokens_per_second": 29242.492
    },
    {
      "epoch": 0.47588825471520785,
      "grad_norm": 0.59375,
      "learning_rate": 3.265185363140928e-05,
      "loss": 0.41379971504211427,
      "num_input_tokens_seen": 3824011584,
      "step": 13450,
      "train_runtime": 130768.656,
      "train_tokens_per_second": 29242.57
    },
    {
      "epoch": 0.4762420749789366,
      "grad_norm": 0.56640625,
      "learning_rate": 3.26448935448656e-05,
      "loss": 0.40944786071777345,
      "num_input_tokens_seen": 3826835904,
      "step": 13460,
      "train_runtime": 130864.3285,
      "train_tokens_per_second": 29242.773
    },
    {
      "epoch": 0.47659589524266543,
      "grad_norm": 0.5546875,
      "learning_rate": 3.263793790727256e-05,
      "loss": 0.41335458755493165,
      "num_input_tokens_seen": 3829707968,
      "step": 13470,
      "train_runtime": 130963.1124,
      "train_tokens_per_second": 29242.646
    },
    {
      "epoch": 0.4769497155063942,
      "grad_norm": 0.5859375,
      "learning_rate": 3.2630986713892495e-05,
      "loss": 0.4083230018615723,
      "num_input_tokens_seen": 3832545344,
      "step": 13480,
      "train_runtime": 131059.6954,
      "train_tokens_per_second": 29242.746
    },
    {
      "epoch": 0.47730353577012297,
      "grad_norm": 0.57421875,
      "learning_rate": 3.26240399599948e-05,
      "loss": 0.4113736152648926,
      "num_input_tokens_seen": 3835377728,
      "step": 13490,
      "train_runtime": 131156.8071,
      "train_tokens_per_second": 29242.689
    },
    {
      "epoch": 0.47765735603385173,
      "grad_norm": 0.59375,
      "learning_rate": 3.2617097640855914e-05,
      "loss": 0.4159392356872559,
      "num_input_tokens_seen": 3838192448,
      "step": 13500,
      "train_runtime": 131253.251,
      "train_tokens_per_second": 29242.647
    },
    {
      "epoch": 0.47801117629758055,
      "grad_norm": 0.58203125,
      "learning_rate": 3.2610159751759314e-05,
      "loss": 0.4103187084197998,
      "num_input_tokens_seen": 3841070848,
      "step": 13510,
      "train_runtime": 131352.8013,
      "train_tokens_per_second": 29242.398
    },
    {
      "epoch": 0.4783649965613093,
      "grad_norm": 0.62109375,
      "learning_rate": 3.26032262879955e-05,
      "loss": 0.41347618103027345,
      "num_input_tokens_seen": 3843858048,
      "step": 13520,
      "train_runtime": 131447.6004,
      "train_tokens_per_second": 29242.512
    },
    {
      "epoch": 0.4787188168250381,
      "grad_norm": 0.6015625,
      "learning_rate": 3.259629724486198e-05,
      "loss": 0.41372361183166506,
      "num_input_tokens_seen": 3846663040,
      "step": 13530,
      "train_runtime": 131542.9181,
      "train_tokens_per_second": 29242.646
    },
    {
      "epoch": 0.47907263708876685,
      "grad_norm": 0.56640625,
      "learning_rate": 3.258937261766323e-05,
      "loss": 0.4119584083557129,
      "num_input_tokens_seen": 3849578240,
      "step": 13540,
      "train_runtime": 131645.0162,
      "train_tokens_per_second": 29242.111
    },
    {
      "epoch": 0.47942645735249567,
      "grad_norm": 0.58203125,
      "learning_rate": 3.258245240171074e-05,
      "loss": 0.4128364086151123,
      "num_input_tokens_seen": 3852476672,
      "step": 13550,
      "train_runtime": 131744.1232,
      "train_tokens_per_second": 29242.114
    },
    {
      "epoch": 0.47978027761622444,
      "grad_norm": 0.5703125,
      "learning_rate": 3.2575536592322935e-05,
      "loss": 0.40924725532531736,
      "num_input_tokens_seen": 3855333952,
      "step": 13560,
      "train_runtime": 131840.7965,
      "train_tokens_per_second": 29242.344
    },
    {
      "epoch": 0.4801340978799532,
      "grad_norm": 0.5703125,
      "learning_rate": 3.256862518482523e-05,
      "loss": 0.40726399421691895,
      "num_input_tokens_seen": 3858135808,
      "step": 13570,
      "train_runtime": 131934.8294,
      "train_tokens_per_second": 29242.739
    },
    {
      "epoch": 0.48048791814368197,
      "grad_norm": 0.60546875,
      "learning_rate": 3.256171817454994e-05,
      "loss": 0.40947628021240234,
      "num_input_tokens_seen": 3860974400,
      "step": 13580,
      "train_runtime": 132031.5663,
      "train_tokens_per_second": 29242.813
    },
    {
      "epoch": 0.4808417384074108,
      "grad_norm": 0.59765625,
      "learning_rate": 3.255481555683633e-05,
      "loss": 0.41001739501953127,
      "num_input_tokens_seen": 3863829120,
      "step": 13590,
      "train_runtime": 132130.4133,
      "train_tokens_per_second": 29242.542
    },
    {
      "epoch": 0.48119555867113956,
      "grad_norm": 0.5703125,
      "learning_rate": 3.254791732703057e-05,
      "loss": 0.4125533580780029,
      "num_input_tokens_seen": 3866681408,
      "step": 13600,
      "train_runtime": 132228.978,
      "train_tokens_per_second": 29242.315
    },
    {
      "epoch": 0.4815493789348683,
      "grad_norm": 0.58984375,
      "learning_rate": 3.254102348048575e-05,
      "loss": 0.4093159675598145,
      "num_input_tokens_seen": 3869484160,
      "step": 13610,
      "train_runtime": 132323.7999,
      "train_tokens_per_second": 29242.541
    },
    {
      "epoch": 0.4819031991985971,
      "grad_norm": 0.546875,
      "learning_rate": 3.25341340125618e-05,
      "loss": 0.4039801597595215,
      "num_input_tokens_seen": 3872334656,
      "step": 13620,
      "train_runtime": 132419.413,
      "train_tokens_per_second": 29242.953
    },
    {
      "epoch": 0.4822570194623259,
      "grad_norm": 0.5703125,
      "learning_rate": 3.2527248918625575e-05,
      "loss": 0.4102001667022705,
      "num_input_tokens_seen": 3875176384,
      "step": 13630,
      "train_runtime": 132515.25,
      "train_tokens_per_second": 29243.248
    },
    {
      "epoch": 0.4826108397260547,
      "grad_norm": 0.58984375,
      "learning_rate": 3.252036819405075e-05,
      "loss": 0.4077859401702881,
      "num_input_tokens_seen": 3877983360,
      "step": 13640,
      "train_runtime": 132609.7545,
      "train_tokens_per_second": 29243.575
    },
    {
      "epoch": 0.48296465998978344,
      "grad_norm": 0.6015625,
      "learning_rate": 3.251349183421788e-05,
      "loss": 0.4132966995239258,
      "num_input_tokens_seen": 3880825280,
      "step": 13650,
      "train_runtime": 132706.497,
      "train_tokens_per_second": 29243.672
    },
    {
      "epoch": 0.4833184802535122,
      "grad_norm": 0.56640625,
      "learning_rate": 3.250661983451434e-05,
      "loss": 0.4122671127319336,
      "num_input_tokens_seen": 3883634624,
      "step": 13660,
      "train_runtime": 132800.9912,
      "train_tokens_per_second": 29244.018
    },
    {
      "epoch": 0.48367230051724097,
      "grad_norm": 0.56640625,
      "learning_rate": 3.2499752190334326e-05,
      "loss": 0.4115451812744141,
      "num_input_tokens_seen": 3886481728,
      "step": 13670,
      "train_runtime": 132901.8928,
      "train_tokens_per_second": 29243.238
    },
    {
      "epoch": 0.4840261207809698,
      "grad_norm": 0.55859375,
      "learning_rate": 3.2492888897078834e-05,
      "loss": 0.4136969089508057,
      "num_input_tokens_seen": 3889359232,
      "step": 13680,
      "train_runtime": 133000.9555,
      "train_tokens_per_second": 29243.092
    },
    {
      "epoch": 0.48437994104469856,
      "grad_norm": 0.56640625,
      "learning_rate": 3.248602995015567e-05,
      "loss": 0.4130582809448242,
      "num_input_tokens_seen": 3892185280,
      "step": 13690,
      "train_runtime": 133095.9976,
      "train_tokens_per_second": 29243.443
    },
    {
      "epoch": 0.4847337613084273,
      "grad_norm": 0.57421875,
      "learning_rate": 3.247917534497943e-05,
      "loss": 0.4113801956176758,
      "num_input_tokens_seen": 3895009856,
      "step": 13700,
      "train_runtime": 133193.6445,
      "train_tokens_per_second": 29243.211
    },
    {
      "epoch": 0.4850875815721561,
      "grad_norm": 0.578125,
      "learning_rate": 3.247232507697145e-05,
      "loss": 0.4093505859375,
      "num_input_tokens_seen": 3897859200,
      "step": 13710,
      "train_runtime": 133292.2116,
      "train_tokens_per_second": 29242.963
    },
    {
      "epoch": 0.4854414018358849,
      "grad_norm": 0.58984375,
      "learning_rate": 3.246547914155985e-05,
      "loss": 0.4139847278594971,
      "num_input_tokens_seen": 3900682816,
      "step": 13720,
      "train_runtime": 133385.5024,
      "train_tokens_per_second": 29243.679
    },
    {
      "epoch": 0.4857952220996137,
      "grad_norm": 0.578125,
      "learning_rate": 3.245863753417949e-05,
      "loss": 0.41477813720703127,
      "num_input_tokens_seen": 3903496256,
      "step": 13730,
      "train_runtime": 133479.3166,
      "train_tokens_per_second": 29244.203
    },
    {
      "epoch": 0.48614904236334244,
      "grad_norm": 0.57421875,
      "learning_rate": 3.2451800250271944e-05,
      "loss": 0.41378226280212405,
      "num_input_tokens_seen": 3906388864,
      "step": 13740,
      "train_runtime": 133580.4252,
      "train_tokens_per_second": 29243.722
    },
    {
      "epoch": 0.4865028626270712,
      "grad_norm": 0.59765625,
      "learning_rate": 3.244496728528553e-05,
      "loss": 0.41117258071899415,
      "num_input_tokens_seen": 3909218240,
      "step": 13750,
      "train_runtime": 133678.6029,
      "train_tokens_per_second": 29243.41
    },
    {
      "epoch": 0.48685668289080003,
      "grad_norm": 0.56640625,
      "learning_rate": 3.243813863467525e-05,
      "loss": 0.4135149002075195,
      "num_input_tokens_seen": 3912107584,
      "step": 13760,
      "train_runtime": 133779.1181,
      "train_tokens_per_second": 29243.036
    },
    {
      "epoch": 0.4872105031545288,
      "grad_norm": 0.56640625,
      "learning_rate": 3.243131429390281e-05,
      "loss": 0.4086174964904785,
      "num_input_tokens_seen": 3914896384,
      "step": 13770,
      "train_runtime": 133871.5528,
      "train_tokens_per_second": 29243.677
    },
    {
      "epoch": 0.48756432341825756,
      "grad_norm": 0.578125,
      "learning_rate": 3.2424494258436594e-05,
      "loss": 0.4061553955078125,
      "num_input_tokens_seen": 3917735680,
      "step": 13780,
      "train_runtime": 133970.8182,
      "train_tokens_per_second": 29243.202
    },
    {
      "epoch": 0.4879181436819863,
      "grad_norm": 0.58984375,
      "learning_rate": 3.241767852375166e-05,
      "loss": 0.4066512107849121,
      "num_input_tokens_seen": 3920528640,
      "step": 13790,
      "train_runtime": 134065.4083,
      "train_tokens_per_second": 29243.402
    },
    {
      "epoch": 0.48827196394571515,
      "grad_norm": 0.5703125,
      "learning_rate": 3.241086708532971e-05,
      "loss": 0.408186674118042,
      "num_input_tokens_seen": 3923366016,
      "step": 13800,
      "train_runtime": 134161.691,
      "train_tokens_per_second": 29243.564
    },
    {
      "epoch": 0.4886257842094439,
      "grad_norm": 0.5703125,
      "learning_rate": 3.24040599386591e-05,
      "loss": 0.41108503341674807,
      "num_input_tokens_seen": 3926291648,
      "step": 13810,
      "train_runtime": 134260.5433,
      "train_tokens_per_second": 29243.824
    },
    {
      "epoch": 0.4889796044731727,
      "grad_norm": 0.56640625,
      "learning_rate": 3.23972570792348e-05,
      "loss": 0.4115931510925293,
      "num_input_tokens_seen": 3929174336,
      "step": 13820,
      "train_runtime": 134360.2686,
      "train_tokens_per_second": 29243.573
    },
    {
      "epoch": 0.48933342473690145,
      "grad_norm": 0.625,
      "learning_rate": 3.239045850255842e-05,
      "loss": 0.4140009880065918,
      "num_input_tokens_seen": 3932069312,
      "step": 13830,
      "train_runtime": 134459.721,
      "train_tokens_per_second": 29243.474
    },
    {
      "epoch": 0.48968724500063027,
      "grad_norm": 0.5625,
      "learning_rate": 3.238366420413817e-05,
      "loss": 0.41455755233764646,
      "num_input_tokens_seen": 3934879616,
      "step": 13840,
      "train_runtime": 134553.2197,
      "train_tokens_per_second": 29244.039
    },
    {
      "epoch": 0.49004106526435903,
      "grad_norm": 0.5546875,
      "learning_rate": 3.237687417948882e-05,
      "loss": 0.411772632598877,
      "num_input_tokens_seen": 3937755456,
      "step": 13850,
      "train_runtime": 134650.9029,
      "train_tokens_per_second": 29244.182
    },
    {
      "epoch": 0.4903948855280878,
      "grad_norm": 0.55078125,
      "learning_rate": 3.2370088424131776e-05,
      "loss": 0.4148508071899414,
      "num_input_tokens_seen": 3940664064,
      "step": 13860,
      "train_runtime": 134751.9689,
      "train_tokens_per_second": 29243.833
    },
    {
      "epoch": 0.49074870579181656,
      "grad_norm": 0.58203125,
      "learning_rate": 3.236330693359497e-05,
      "loss": 0.41036348342895507,
      "num_input_tokens_seen": 3943483520,
      "step": 13870,
      "train_runtime": 134846.9012,
      "train_tokens_per_second": 29244.154
    },
    {
      "epoch": 0.4911025260555454,
      "grad_norm": 0.5625,
      "learning_rate": 3.2356529703412894e-05,
      "loss": 0.41105971336364744,
      "num_input_tokens_seen": 3946334144,
      "step": 13880,
      "train_runtime": 134944.6578,
      "train_tokens_per_second": 29244.093
    },
    {
      "epoch": 0.49145634631927415,
      "grad_norm": 0.578125,
      "learning_rate": 3.234975672912661e-05,
      "loss": 0.41156597137451173,
      "num_input_tokens_seen": 3949208896,
      "step": 13890,
      "train_runtime": 135042.804,
      "train_tokens_per_second": 29244.127
    },
    {
      "epoch": 0.4918101665830029,
      "grad_norm": 0.59765625,
      "learning_rate": 3.234298800628368e-05,
      "loss": 0.4125880241394043,
      "num_input_tokens_seen": 3952040192,
      "step": 13900,
      "train_runtime": 135139.8954,
      "train_tokens_per_second": 29244.067
    },
    {
      "epoch": 0.4921639868467317,
      "grad_norm": 0.57421875,
      "learning_rate": 3.2336223530438195e-05,
      "loss": 0.4110852241516113,
      "num_input_tokens_seen": 3954881216,
      "step": 13910,
      "train_runtime": 135237.2938,
      "train_tokens_per_second": 29244.013
    },
    {
      "epoch": 0.49251780711046045,
      "grad_norm": 0.57421875,
      "learning_rate": 3.232946329715076e-05,
      "loss": 0.4091800212860107,
      "num_input_tokens_seen": 3957707456,
      "step": 13920,
      "train_runtime": 135332.8133,
      "train_tokens_per_second": 29244.256
    },
    {
      "epoch": 0.49287162737418927,
      "grad_norm": 0.59765625,
      "learning_rate": 3.2322707301988456e-05,
      "loss": 0.4068630218505859,
      "num_input_tokens_seen": 3960528768,
      "step": 13930,
      "train_runtime": 135425.4095,
      "train_tokens_per_second": 29245.094
    },
    {
      "epoch": 0.49322544763791804,
      "grad_norm": 0.6015625,
      "learning_rate": 3.231595554052488e-05,
      "loss": 0.4096001148223877,
      "num_input_tokens_seen": 3963380224,
      "step": 13940,
      "train_runtime": 135522.603,
      "train_tokens_per_second": 29245.16
    },
    {
      "epoch": 0.4935792679016468,
      "grad_norm": 0.58984375,
      "learning_rate": 3.230920800834005e-05,
      "loss": 0.4113500118255615,
      "num_input_tokens_seen": 3966185728,
      "step": 13950,
      "train_runtime": 135618.7431,
      "train_tokens_per_second": 29245.115
    },
    {
      "epoch": 0.49393308816537557,
      "grad_norm": 0.59375,
      "learning_rate": 3.2302464701020486e-05,
      "loss": 0.41273021697998047,
      "num_input_tokens_seen": 3968997120,
      "step": 13960,
      "train_runtime": 135712.8662,
      "train_tokens_per_second": 29245.548
    },
    {
      "epoch": 0.4942869084291044,
      "grad_norm": 0.59375,
      "learning_rate": 3.2295725614159126e-05,
      "loss": 0.410627555847168,
      "num_input_tokens_seen": 3971827776,
      "step": 13970,
      "train_runtime": 135808.4213,
      "train_tokens_per_second": 29245.814
    },
    {
      "epoch": 0.49464072869283315,
      "grad_norm": 0.5703125,
      "learning_rate": 3.228899074335536e-05,
      "loss": 0.4088435173034668,
      "num_input_tokens_seen": 3974702400,
      "step": 13980,
      "train_runtime": 135910.0389,
      "train_tokens_per_second": 29245.098
    },
    {
      "epoch": 0.4949945489565619,
      "grad_norm": 0.5625,
      "learning_rate": 3.228226008421498e-05,
      "loss": 0.4095346450805664,
      "num_input_tokens_seen": 3977531136,
      "step": 13990,
      "train_runtime": 136005.1625,
      "train_tokens_per_second": 29245.442
    },
    {
      "epoch": 0.4953483692202907,
      "grad_norm": 0.5859375,
      "learning_rate": 3.2275533632350193e-05,
      "loss": 0.4095023155212402,
      "num_input_tokens_seen": 3980384000,
      "step": 14000,
      "train_runtime": 136104.1209,
      "train_tokens_per_second": 29245.139
    },
    {
      "epoch": 0.4957021894840195,
      "grad_norm": 0.62890625,
      "learning_rate": 3.226881138337963e-05,
      "loss": 0.41056323051452637,
      "num_input_tokens_seen": 3983229632,
      "step": 14010,
      "train_runtime": 136201.9369,
      "train_tokens_per_second": 29245.029
    },
    {
      "epoch": 0.49605600974774827,
      "grad_norm": 0.58203125,
      "learning_rate": 3.2262093332928256e-05,
      "loss": 0.40982680320739745,
      "num_input_tokens_seen": 3986041472,
      "step": 14020,
      "train_runtime": 136297.0031,
      "train_tokens_per_second": 29245.261
    },
    {
      "epoch": 0.49640983001147704,
      "grad_norm": 0.58203125,
      "learning_rate": 3.225537947662746e-05,
      "loss": 0.4065865516662598,
      "num_input_tokens_seen": 3988832896,
      "step": 14030,
      "train_runtime": 136387.8759,
      "train_tokens_per_second": 29246.243
    },
    {
      "epoch": 0.4967636502752058,
      "grad_norm": 0.59375,
      "learning_rate": 3.224866981011494e-05,
      "loss": 0.40512828826904296,
      "num_input_tokens_seen": 3991680640,
      "step": 14040,
      "train_runtime": 136483.1669,
      "train_tokens_per_second": 29246.688
    },
    {
      "epoch": 0.4971174705389346,
      "grad_norm": 0.5625,
      "learning_rate": 3.22419643290348e-05,
      "loss": 0.4101290225982666,
      "num_input_tokens_seen": 3994563456,
      "step": 14050,
      "train_runtime": 136581.2685,
      "train_tokens_per_second": 29246.788
    },
    {
      "epoch": 0.4974712908026634,
      "grad_norm": 0.5859375,
      "learning_rate": 3.2235263029037446e-05,
      "loss": 0.4105556488037109,
      "num_input_tokens_seen": 3997392256,
      "step": 14060,
      "train_runtime": 136680.9936,
      "train_tokens_per_second": 29246.146
    },
    {
      "epoch": 0.49782511106639216,
      "grad_norm": 0.57421875,
      "learning_rate": 3.222856590577962e-05,
      "loss": 0.411027717590332,
      "num_input_tokens_seen": 4000232832,
      "step": 14070,
      "train_runtime": 136777.8963,
      "train_tokens_per_second": 29246.194
    },
    {
      "epoch": 0.4981789313301209,
      "grad_norm": 0.56640625,
      "learning_rate": 3.222187295492436e-05,
      "loss": 0.4114194393157959,
      "num_input_tokens_seen": 4003073536,
      "step": 14080,
      "train_runtime": 136873.8234,
      "train_tokens_per_second": 29246.451
    },
    {
      "epoch": 0.49853275159384974,
      "grad_norm": 0.578125,
      "learning_rate": 3.221518417214104e-05,
      "loss": 0.41209001541137696,
      "num_input_tokens_seen": 4005809600,
      "step": 14090,
      "train_runtime": 136965.3394,
      "train_tokens_per_second": 29246.886
    },
    {
      "epoch": 0.4988865718575785,
      "grad_norm": 0.625,
      "learning_rate": 3.22084995531053e-05,
      "loss": 0.4022061824798584,
      "num_input_tokens_seen": 4008653568,
      "step": 14100,
      "train_runtime": 137063.6063,
      "train_tokens_per_second": 29246.666
    },
    {
      "epoch": 0.4992403921213073,
      "grad_norm": 0.55078125,
      "learning_rate": 3.220181909349907e-05,
      "loss": 0.40438218116760255,
      "num_input_tokens_seen": 4011502592,
      "step": 14110,
      "train_runtime": 137162.2486,
      "train_tokens_per_second": 29246.404
    },
    {
      "epoch": 0.49959421238503604,
      "grad_norm": 0.5859375,
      "learning_rate": 3.219514278901053e-05,
      "loss": 0.41073899269104003,
      "num_input_tokens_seen": 4014291200,
      "step": 14120,
      "train_runtime": 137255.7578,
      "train_tokens_per_second": 29246.796
    },
    {
      "epoch": 0.49994803264876486,
      "grad_norm": 0.5625,
      "learning_rate": 3.218847063533413e-05,
      "loss": 0.4103257656097412,
      "num_input_tokens_seen": 4017122112,
      "step": 14130,
      "train_runtime": 137350.9641,
      "train_tokens_per_second": 29247.134
    },
    {
      "epoch": 0.5003018529124936,
      "grad_norm": 0.6015625,
      "learning_rate": 3.218180262817055e-05,
      "loss": 0.41107468605041503,
      "num_input_tokens_seen": 4019970816,
      "step": 14140,
      "train_runtime": 137449.0381,
      "train_tokens_per_second": 29246.991
    },
    {
      "epoch": 0.5006556731762224,
      "grad_norm": 0.57421875,
      "learning_rate": 3.217513876322674e-05,
      "loss": 0.40564765930175783,
      "num_input_tokens_seen": 4022749120,
      "step": 14150,
      "train_runtime": 137544.9345,
      "train_tokens_per_second": 29246.799
    },
    {
      "epoch": 0.5010094934399512,
      "grad_norm": 0.59375,
      "learning_rate": 3.216847903621581e-05,
      "loss": 0.4059504508972168,
      "num_input_tokens_seen": 4025592000,
      "step": 14160,
      "train_runtime": 137641.0155,
      "train_tokens_per_second": 29247.038
    },
    {
      "epoch": 0.5013633137036799,
      "grad_norm": 0.6171875,
      "learning_rate": 3.216182344285713e-05,
      "loss": 0.4110740661621094,
      "num_input_tokens_seen": 4028470080,
      "step": 14170,
      "train_runtime": 137739.5368,
      "train_tokens_per_second": 29247.013
    },
    {
      "epoch": 0.5017171339674087,
      "grad_norm": 0.5703125,
      "learning_rate": 3.215517197887625e-05,
      "loss": 0.4101099014282227,
      "num_input_tokens_seen": 4031278400,
      "step": 14180,
      "train_runtime": 137834.806,
      "train_tokens_per_second": 29247.173
    },
    {
      "epoch": 0.5020709542311375,
      "grad_norm": 0.59765625,
      "learning_rate": 3.214852464000488e-05,
      "loss": 0.41011390686035154,
      "num_input_tokens_seen": 4034128192,
      "step": 14190,
      "train_runtime": 137929.6424,
      "train_tokens_per_second": 29247.725
    },
    {
      "epoch": 0.5024247744948663,
      "grad_norm": 0.6015625,
      "learning_rate": 3.2141881421980945e-05,
      "loss": 0.4089027404785156,
      "num_input_tokens_seen": 4036948608,
      "step": 14200,
      "train_runtime": 138025.0779,
      "train_tokens_per_second": 29247.936
    },
    {
      "epoch": 0.5027785947585951,
      "grad_norm": 0.5859375,
      "learning_rate": 3.213524232054851e-05,
      "loss": 0.4075775146484375,
      "num_input_tokens_seen": 4039743232,
      "step": 14210,
      "train_runtime": 138117.8362,
      "train_tokens_per_second": 29248.527
    },
    {
      "epoch": 0.5031324150223239,
      "grad_norm": 0.59765625,
      "learning_rate": 3.21286073314578e-05,
      "loss": 0.4120054244995117,
      "num_input_tokens_seen": 4042558016,
      "step": 14220,
      "train_runtime": 138216.1934,
      "train_tokens_per_second": 29248.078
    },
    {
      "epoch": 0.5034862352860526,
      "grad_norm": 0.6015625,
      "learning_rate": 3.2121976450465155e-05,
      "loss": 0.41195001602172854,
      "num_input_tokens_seen": 4045371264,
      "step": 14230,
      "train_runtime": 138310.5084,
      "train_tokens_per_second": 29248.474
    },
    {
      "epoch": 0.5038400555497814,
      "grad_norm": 0.57421875,
      "learning_rate": 3.211534967333308e-05,
      "loss": 0.40461182594299316,
      "num_input_tokens_seen": 4048218176,
      "step": 14240,
      "train_runtime": 138407.5156,
      "train_tokens_per_second": 29248.543
    },
    {
      "epoch": 0.5041938758135102,
      "grad_norm": 0.5703125,
      "learning_rate": 3.210872699583019e-05,
      "loss": 0.4075809955596924,
      "num_input_tokens_seen": 4051051456,
      "step": 14250,
      "train_runtime": 138503.7308,
      "train_tokens_per_second": 29248.681
    },
    {
      "epoch": 0.5045476960772389,
      "grad_norm": 0.5859375,
      "learning_rate": 3.210210841373118e-05,
      "loss": 0.41009230613708497,
      "num_input_tokens_seen": 4053911744,
      "step": 14260,
      "train_runtime": 138602.5778,
      "train_tokens_per_second": 29248.459
    },
    {
      "epoch": 0.5049015163409677,
      "grad_norm": 0.58984375,
      "learning_rate": 3.2095493922816855e-05,
      "loss": 0.4117298126220703,
      "num_input_tokens_seen": 4056795584,
      "step": 14270,
      "train_runtime": 138701.1285,
      "train_tokens_per_second": 29248.468
    },
    {
      "epoch": 0.5052553366046966,
      "grad_norm": 0.58203125,
      "learning_rate": 3.2088883518874105e-05,
      "loss": 0.40950603485107423,
      "num_input_tokens_seen": 4059580480,
      "step": 14280,
      "train_runtime": 138796.5235,
      "train_tokens_per_second": 29248.431
    },
    {
      "epoch": 0.5056091568684253,
      "grad_norm": 0.59765625,
      "learning_rate": 3.208227719769589e-05,
      "loss": 0.4050768852233887,
      "num_input_tokens_seen": 4062394432,
      "step": 14290,
      "train_runtime": 138892.5552,
      "train_tokens_per_second": 29248.468
    },
    {
      "epoch": 0.5059629771321541,
      "grad_norm": 0.5859375,
      "learning_rate": 3.207567495508124e-05,
      "loss": 0.40495052337646487,
      "num_input_tokens_seen": 4065237184,
      "step": 14300,
      "train_runtime": 138988.7476,
      "train_tokens_per_second": 29248.678
    },
    {
      "epoch": 0.5063167973958829,
      "grad_norm": 0.578125,
      "learning_rate": 3.2069076786835205e-05,
      "loss": 0.4067946434020996,
      "num_input_tokens_seen": 4068040320,
      "step": 14310,
      "train_runtime": 139085.0999,
      "train_tokens_per_second": 29248.57
    },
    {
      "epoch": 0.5066706176596116,
      "grad_norm": 0.57421875,
      "learning_rate": 3.2062482688768904e-05,
      "loss": 0.41307244300842283,
      "num_input_tokens_seen": 4070913536,
      "step": 14320,
      "train_runtime": 139185.12,
      "train_tokens_per_second": 29248.195
    },
    {
      "epoch": 0.5070244379233404,
      "grad_norm": 0.55078125,
      "learning_rate": 3.205589265669947e-05,
      "loss": 0.40561642646789553,
      "num_input_tokens_seen": 4073779840,
      "step": 14330,
      "train_runtime": 139283.7727,
      "train_tokens_per_second": 29248.058
    },
    {
      "epoch": 0.5073782581870692,
      "grad_norm": 0.5859375,
      "learning_rate": 3.204930668645005e-05,
      "loss": 0.4145057678222656,
      "num_input_tokens_seen": 4076619840,
      "step": 14340,
      "train_runtime": 139379.8003,
      "train_tokens_per_second": 29248.283
    },
    {
      "epoch": 0.5077320784507979,
      "grad_norm": 0.6015625,
      "learning_rate": 3.20427247738498e-05,
      "loss": 0.40759429931640623,
      "num_input_tokens_seen": 4079465216,
      "step": 14350,
      "train_runtime": 139478.8114,
      "train_tokens_per_second": 29247.921
    },
    {
      "epoch": 0.5080858987145268,
      "grad_norm": 0.625,
      "learning_rate": 3.2036146914733854e-05,
      "loss": 0.4121386528015137,
      "num_input_tokens_seen": 4082283904,
      "step": 14360,
      "train_runtime": 139574.8673,
      "train_tokens_per_second": 29247.987
    },
    {
      "epoch": 0.5084397189782556,
      "grad_norm": 0.578125,
      "learning_rate": 3.202957310494336e-05,
      "loss": 0.40762953758239745,
      "num_input_tokens_seen": 4085162240,
      "step": 14370,
      "train_runtime": 139673.3678,
      "train_tokens_per_second": 29247.968
    },
    {
      "epoch": 0.5087935392419843,
      "grad_norm": 0.55859375,
      "learning_rate": 3.202300334032542e-05,
      "loss": 0.399967360496521,
      "num_input_tokens_seen": 4088018432,
      "step": 14380,
      "train_runtime": 139771.0372,
      "train_tokens_per_second": 29247.965
    },
    {
      "epoch": 0.5091473595057131,
      "grad_norm": 0.5703125,
      "learning_rate": 3.201643761673308e-05,
      "loss": 0.41347856521606446,
      "num_input_tokens_seen": 4090857408,
      "step": 14390,
      "train_runtime": 139870.2281,
      "train_tokens_per_second": 29247.521
    },
    {
      "epoch": 0.5095011797694419,
      "grad_norm": 0.59765625,
      "learning_rate": 3.200987593002536e-05,
      "loss": 0.41163110733032227,
      "num_input_tokens_seen": 4093682816,
      "step": 14400,
      "train_runtime": 139966.55,
      "train_tokens_per_second": 29247.58
    },
    {
      "epoch": 0.5098550000331706,
      "grad_norm": 0.56640625,
      "learning_rate": 3.200331827606721e-05,
      "loss": 0.4086371898651123,
      "num_input_tokens_seen": 4096522432,
      "step": 14410,
      "train_runtime": 140064.191,
      "train_tokens_per_second": 29247.464
    },
    {
      "epoch": 0.5102088202968994,
      "grad_norm": 0.55078125,
      "learning_rate": 3.199676465072951e-05,
      "loss": 0.4037954330444336,
      "num_input_tokens_seen": 4099334656,
      "step": 14420,
      "train_runtime": 140159.8533,
      "train_tokens_per_second": 29247.567
    },
    {
      "epoch": 0.5105626405606282,
      "grad_norm": 0.58203125,
      "learning_rate": 3.1990215049889046e-05,
      "loss": 0.40871801376342776,
      "num_input_tokens_seen": 4102225472,
      "step": 14430,
      "train_runtime": 140259.7827,
      "train_tokens_per_second": 29247.339
    },
    {
      "epoch": 0.5109164608243569,
      "grad_norm": 0.56640625,
      "learning_rate": 3.198366946942851e-05,
      "loss": 0.41789684295654295,
      "num_input_tokens_seen": 4105131328,
      "step": 14440,
      "train_runtime": 140360.9934,
      "train_tokens_per_second": 29246.953
    },
    {
      "epoch": 0.5112702810880858,
      "grad_norm": 0.59375,
      "learning_rate": 3.1977127905236514e-05,
      "loss": 0.40943112373352053,
      "num_input_tokens_seen": 4107939968,
      "step": 14450,
      "train_runtime": 140455.4199,
      "train_tokens_per_second": 29247.287
    },
    {
      "epoch": 0.5116241013518146,
      "grad_norm": 0.5625,
      "learning_rate": 3.197059035320752e-05,
      "loss": 0.4104298114776611,
      "num_input_tokens_seen": 4110789568,
      "step": 14460,
      "train_runtime": 140554.9408,
      "train_tokens_per_second": 29246.852
    },
    {
      "epoch": 0.5119779216155433,
      "grad_norm": 0.59765625,
      "learning_rate": 3.196405680924189e-05,
      "loss": 0.40865387916564944,
      "num_input_tokens_seen": 4113665408,
      "step": 14470,
      "train_runtime": 140653.9388,
      "train_tokens_per_second": 29246.713
    },
    {
      "epoch": 0.5123317418792721,
      "grad_norm": 0.58984375,
      "learning_rate": 3.195752726924582e-05,
      "loss": 0.40996227264404295,
      "num_input_tokens_seen": 4116458560,
      "step": 14480,
      "train_runtime": 140748.8309,
      "train_tokens_per_second": 29246.84
    },
    {
      "epoch": 0.5126855621430009,
      "grad_norm": 0.578125,
      "learning_rate": 3.195100172913139e-05,
      "loss": 0.4076345443725586,
      "num_input_tokens_seen": 4119314880,
      "step": 14490,
      "train_runtime": 140848.2726,
      "train_tokens_per_second": 29246.471
    },
    {
      "epoch": 0.5130393824067296,
      "grad_norm": 0.55859375,
      "learning_rate": 3.19444801848165e-05,
      "loss": 0.41025753021240235,
      "num_input_tokens_seen": 4122162112,
      "step": 14500,
      "train_runtime": 140945.4708,
      "train_tokens_per_second": 29246.503
    },
    {
      "epoch": 0.5133932026704584,
      "grad_norm": 0.578125,
      "learning_rate": 3.1937962632224885e-05,
      "loss": 0.40854034423828123,
      "num_input_tokens_seen": 4125037696,
      "step": 14510,
      "train_runtime": 141045.013,
      "train_tokens_per_second": 29246.25
    },
    {
      "epoch": 0.5137470229341872,
      "grad_norm": 0.58984375,
      "learning_rate": 3.193144906728609e-05,
      "loss": 0.4043434143066406,
      "num_input_tokens_seen": 4127887424,
      "step": 14520,
      "train_runtime": 141140.7632,
      "train_tokens_per_second": 29246.6
    },
    {
      "epoch": 0.514100843197916,
      "grad_norm": 0.61328125,
      "learning_rate": 3.1924939485935494e-05,
      "loss": 0.4098823070526123,
      "num_input_tokens_seen": 4130725888,
      "step": 14530,
      "train_runtime": 141237.2549,
      "train_tokens_per_second": 29246.716
    },
    {
      "epoch": 0.5144546634616448,
      "grad_norm": 0.5859375,
      "learning_rate": 3.1918433884114253e-05,
      "loss": 0.40554037094116213,
      "num_input_tokens_seen": 4133562560,
      "step": 14540,
      "train_runtime": 141333.4964,
      "train_tokens_per_second": 29246.871
    },
    {
      "epoch": 0.5148084837253736,
      "grad_norm": 0.57421875,
      "learning_rate": 3.191193225776931e-05,
      "loss": 0.4158463954925537,
      "num_input_tokens_seen": 4136477056,
      "step": 14550,
      "train_runtime": 141435.577,
      "train_tokens_per_second": 29246.369
    },
    {
      "epoch": 0.5151623039891023,
      "grad_norm": 0.60546875,
      "learning_rate": 3.190543460285339e-05,
      "loss": 0.41065044403076173,
      "num_input_tokens_seen": 4139378048,
      "step": 14560,
      "train_runtime": 141538.2617,
      "train_tokens_per_second": 29245.647
    },
    {
      "epoch": 0.5155161242528311,
      "grad_norm": 0.5859375,
      "learning_rate": 3.189894091532499e-05,
      "loss": 0.41159496307373045,
      "num_input_tokens_seen": 4142189568,
      "step": 14570,
      "train_runtime": 141633.7955,
      "train_tokens_per_second": 29245.771
    },
    {
      "epoch": 0.5158699445165599,
      "grad_norm": 0.58203125,
      "learning_rate": 3.1892451191148346e-05,
      "loss": 0.4092118263244629,
      "num_input_tokens_seen": 4144996480,
      "step": 14580,
      "train_runtime": 141728.4706,
      "train_tokens_per_second": 29246.04
    },
    {
      "epoch": 0.5162237647802886,
      "grad_norm": 0.546875,
      "learning_rate": 3.1885965426293465e-05,
      "loss": 0.40863957405090334,
      "num_input_tokens_seen": 4147858752,
      "step": 14590,
      "train_runtime": 141828.8299,
      "train_tokens_per_second": 29245.526
    },
    {
      "epoch": 0.5165775850440174,
      "grad_norm": 0.58203125,
      "learning_rate": 3.187948361673606e-05,
      "loss": 0.4075927734375,
      "num_input_tokens_seen": 4150731072,
      "step": 14600,
      "train_runtime": 141929.401,
      "train_tokens_per_second": 29245.04
    },
    {
      "epoch": 0.5169314053077463,
      "grad_norm": 0.5703125,
      "learning_rate": 3.187300575845759e-05,
      "loss": 0.40454878807067873,
      "num_input_tokens_seen": 4153606400,
      "step": 14610,
      "train_runtime": 142030.9616,
      "train_tokens_per_second": 29244.373
    },
    {
      "epoch": 0.517285225571475,
      "grad_norm": 0.5859375,
      "learning_rate": 3.186653184744521e-05,
      "loss": 0.4053593635559082,
      "num_input_tokens_seen": 4156454976,
      "step": 14620,
      "train_runtime": 142132.5993,
      "train_tokens_per_second": 29243.502
    },
    {
      "epoch": 0.5176390458352038,
      "grad_norm": 0.625,
      "learning_rate": 3.18600618796918e-05,
      "loss": 0.41242570877075196,
      "num_input_tokens_seen": 4159297856,
      "step": 14630,
      "train_runtime": 142230.785,
      "train_tokens_per_second": 29243.302
    },
    {
      "epoch": 0.5179928660989326,
      "grad_norm": 0.57421875,
      "learning_rate": 3.185359585119591e-05,
      "loss": 0.4056413650512695,
      "num_input_tokens_seen": 4162070592,
      "step": 14640,
      "train_runtime": 142324.098,
      "train_tokens_per_second": 29243.611
    },
    {
      "epoch": 0.5183466863626613,
      "grad_norm": 0.58984375,
      "learning_rate": 3.184713375796178e-05,
      "loss": 0.4118755340576172,
      "num_input_tokens_seen": 4164963520,
      "step": 14650,
      "train_runtime": 142422.9052,
      "train_tokens_per_second": 29243.635
    },
    {
      "epoch": 0.5187005066263901,
      "grad_norm": 0.5390625,
      "learning_rate": 3.1840675595999344e-05,
      "loss": 0.4097257614135742,
      "num_input_tokens_seen": 4167799616,
      "step": 14660,
      "train_runtime": 142518.8997,
      "train_tokens_per_second": 29243.838
    },
    {
      "epoch": 0.5190543268901189,
      "grad_norm": 0.6015625,
      "learning_rate": 3.1834221361324155e-05,
      "loss": 0.41007165908813475,
      "num_input_tokens_seen": 4170638912,
      "step": 14670,
      "train_runtime": 142617.3679,
      "train_tokens_per_second": 29243.555
    },
    {
      "epoch": 0.5194081471538476,
      "grad_norm": 0.59375,
      "learning_rate": 3.182777104995744e-05,
      "loss": 0.4110051155090332,
      "num_input_tokens_seen": 4173507968,
      "step": 14680,
      "train_runtime": 142717.7805,
      "train_tokens_per_second": 29243.083
    },
    {
      "epoch": 0.5197619674175764,
      "grad_norm": 0.59375,
      "learning_rate": 3.182132465792609e-05,
      "loss": 0.40939817428588865,
      "num_input_tokens_seen": 4176281024,
      "step": 14690,
      "train_runtime": 142811.6736,
      "train_tokens_per_second": 29243.275
    },
    {
      "epoch": 0.5201157876813053,
      "grad_norm": 0.578125,
      "learning_rate": 3.181488218126259e-05,
      "loss": 0.4063596248626709,
      "num_input_tokens_seen": 4179121664,
      "step": 14700,
      "train_runtime": 142907.8813,
      "train_tokens_per_second": 29243.465
    },
    {
      "epoch": 0.520469607945034,
      "grad_norm": 0.57421875,
      "learning_rate": 3.180844361600506e-05,
      "loss": 0.4110249519348145,
      "num_input_tokens_seen": 4181968960,
      "step": 14710,
      "train_runtime": 143006.6258,
      "train_tokens_per_second": 29243.183
    },
    {
      "epoch": 0.5208234282087628,
      "grad_norm": 0.58984375,
      "learning_rate": 3.180200895819722e-05,
      "loss": 0.4097727298736572,
      "num_input_tokens_seen": 4184813760,
      "step": 14720,
      "train_runtime": 143106.4397,
      "train_tokens_per_second": 29242.666
    },
    {
      "epoch": 0.5211772484724916,
      "grad_norm": 0.58203125,
      "learning_rate": 3.1795578203888424e-05,
      "loss": 0.4089636325836182,
      "num_input_tokens_seen": 4187652480,
      "step": 14730,
      "train_runtime": 143201.8451,
      "train_tokens_per_second": 29243.006
    },
    {
      "epoch": 0.5215310687362203,
      "grad_norm": 0.609375,
      "learning_rate": 3.178915134913357e-05,
      "loss": 0.41102914810180663,
      "num_input_tokens_seen": 4190458368,
      "step": 14740,
      "train_runtime": 143297.8271,
      "train_tokens_per_second": 29243.0
    },
    {
      "epoch": 0.5218848889999491,
      "grad_norm": 0.5703125,
      "learning_rate": 3.178272838999316e-05,
      "loss": 0.4113306999206543,
      "num_input_tokens_seen": 4193264512,
      "step": 14750,
      "train_runtime": 143389.8787,
      "train_tokens_per_second": 29243.797
    },
    {
      "epoch": 0.5222387092636779,
      "grad_norm": 0.56640625,
      "learning_rate": 3.1776309322533274e-05,
      "loss": 0.4112546920776367,
      "num_input_tokens_seen": 4196158144,
      "step": 14760,
      "train_runtime": 143490.1834,
      "train_tokens_per_second": 29243.521
    },
    {
      "epoch": 0.5225925295274066,
      "grad_norm": 0.5859375,
      "learning_rate": 3.1769894142825536e-05,
      "loss": 0.4107497692108154,
      "num_input_tokens_seen": 4199005248,
      "step": 14770,
      "train_runtime": 143587.3436,
      "train_tokens_per_second": 29243.561
    },
    {
      "epoch": 0.5229463497911355,
      "grad_norm": 0.5859375,
      "learning_rate": 3.176348284694712e-05,
      "loss": 0.4052229881286621,
      "num_input_tokens_seen": 4201899904,
      "step": 14780,
      "train_runtime": 143685.1884,
      "train_tokens_per_second": 29243.793
    },
    {
      "epoch": 0.5233001700548643,
      "grad_norm": 0.57421875,
      "learning_rate": 3.175707543098075e-05,
      "loss": 0.40373902320861815,
      "num_input_tokens_seen": 4204755200,
      "step": 14790,
      "train_runtime": 143782.1135,
      "train_tokens_per_second": 29243.938
    },
    {
      "epoch": 0.523653990318593,
      "grad_norm": 0.57421875,
      "learning_rate": 3.1750671891014653e-05,
      "loss": 0.40833053588867185,
      "num_input_tokens_seen": 4207601856,
      "step": 14800,
      "train_runtime": 143877.7732,
      "train_tokens_per_second": 29244.28
    },
    {
      "epoch": 0.5240078105823218,
      "grad_norm": 0.57421875,
      "learning_rate": 3.174427222314262e-05,
      "loss": 0.4086447715759277,
      "num_input_tokens_seen": 4210502592,
      "step": 14810,
      "train_runtime": 143981.1024,
      "train_tokens_per_second": 29243.439
    },
    {
      "epoch": 0.5243616308460506,
      "grad_norm": 0.5546875,
      "learning_rate": 3.173787642346391e-05,
      "loss": 0.4065539360046387,
      "num_input_tokens_seen": 4213360320,
      "step": 14820,
      "train_runtime": 144078.8601,
      "train_tokens_per_second": 29243.432
    },
    {
      "epoch": 0.5247154511097794,
      "grad_norm": 0.59375,
      "learning_rate": 3.173148448808331e-05,
      "loss": 0.410731840133667,
      "num_input_tokens_seen": 4216256576,
      "step": 14830,
      "train_runtime": 144180.9939,
      "train_tokens_per_second": 29242.804
    },
    {
      "epoch": 0.5250692713735081,
      "grad_norm": 0.57421875,
      "learning_rate": 3.172509641311107e-05,
      "loss": 0.40764832496643066,
      "num_input_tokens_seen": 4219089728,
      "step": 14840,
      "train_runtime": 144277.4233,
      "train_tokens_per_second": 29242.896
    },
    {
      "epoch": 0.5254230916372369,
      "grad_norm": 0.578125,
      "learning_rate": 3.171871219466293e-05,
      "loss": 0.40964713096618655,
      "num_input_tokens_seen": 4221922816,
      "step": 14850,
      "train_runtime": 144375.7912,
      "train_tokens_per_second": 29242.595
    },
    {
      "epoch": 0.5257769119009658,
      "grad_norm": 0.5625,
      "learning_rate": 3.171233182886011e-05,
      "loss": 0.4118339061737061,
      "num_input_tokens_seen": 4224739520,
      "step": 14860,
      "train_runtime": 144470.1022,
      "train_tokens_per_second": 29243.002
    },
    {
      "epoch": 0.5261307321646945,
      "grad_norm": 0.56640625,
      "learning_rate": 3.170595531182928e-05,
      "loss": 0.407607889175415,
      "num_input_tokens_seen": 4227567616,
      "step": 14870,
      "train_runtime": 144565.6976,
      "train_tokens_per_second": 29243.228
    },
    {
      "epoch": 0.5264845524284233,
      "grad_norm": 0.59375,
      "learning_rate": 3.169958263970256e-05,
      "loss": 0.40597219467163087,
      "num_input_tokens_seen": 4230410944,
      "step": 14880,
      "train_runtime": 144662.352,
      "train_tokens_per_second": 29243.344
    },
    {
      "epoch": 0.5268383726921521,
      "grad_norm": 0.5703125,
      "learning_rate": 3.169321380861751e-05,
      "loss": 0.4079143524169922,
      "num_input_tokens_seen": 4233268416,
      "step": 14890,
      "train_runtime": 144760.6959,
      "train_tokens_per_second": 29243.217
    },
    {
      "epoch": 0.5271921929558808,
      "grad_norm": 0.5859375,
      "learning_rate": 3.168684881471711e-05,
      "loss": 0.4101587772369385,
      "num_input_tokens_seen": 4236136320,
      "step": 14900,
      "train_runtime": 144857.6424,
      "train_tokens_per_second": 29243.444
    },
    {
      "epoch": 0.5275460132196096,
      "grad_norm": 0.62109375,
      "learning_rate": 3.168048765414979e-05,
      "loss": 0.4105274200439453,
      "num_input_tokens_seen": 4238955776,
      "step": 14910,
      "train_runtime": 144954.1543,
      "train_tokens_per_second": 29243.424
    },
    {
      "epoch": 0.5278998334833384,
      "grad_norm": 0.59375,
      "learning_rate": 3.167413032306936e-05,
      "loss": 0.4076228618621826,
      "num_input_tokens_seen": 4241815808,
      "step": 14920,
      "train_runtime": 145050.2879,
      "train_tokens_per_second": 29243.76
    },
    {
      "epoch": 0.5282536537470671,
      "grad_norm": 0.578125,
      "learning_rate": 3.166777681763504e-05,
      "loss": 0.40525283813476565,
      "num_input_tokens_seen": 4244648448,
      "step": 14930,
      "train_runtime": 145146.8466,
      "train_tokens_per_second": 29243.821
    },
    {
      "epoch": 0.5286074740107959,
      "grad_norm": 0.58203125,
      "learning_rate": 3.166142713401144e-05,
      "loss": 0.4055943012237549,
      "num_input_tokens_seen": 4247458240,
      "step": 14940,
      "train_runtime": 145240.4891,
      "train_tokens_per_second": 29244.312
    },
    {
      "epoch": 0.5289612942745248,
      "grad_norm": 0.5625,
      "learning_rate": 3.165508126836857e-05,
      "loss": 0.4043453216552734,
      "num_input_tokens_seen": 4250291904,
      "step": 14950,
      "train_runtime": 145337.6913,
      "train_tokens_per_second": 29244.251
    },
    {
      "epoch": 0.5293151145382535,
      "grad_norm": 0.578125,
      "learning_rate": 3.164873921688177e-05,
      "loss": 0.41163172721862795,
      "num_input_tokens_seen": 4253113856,
      "step": 14960,
      "train_runtime": 145433.5779,
      "train_tokens_per_second": 29244.373
    },
    {
      "epoch": 0.5296689348019823,
      "grad_norm": 0.578125,
      "learning_rate": 3.164240097573178e-05,
      "loss": 0.4109938621520996,
      "num_input_tokens_seen": 4255949760,
      "step": 14970,
      "train_runtime": 145530.143,
      "train_tokens_per_second": 29244.455
    },
    {
      "epoch": 0.5300227550657111,
      "grad_norm": 0.5703125,
      "learning_rate": 3.163606654110467e-05,
      "loss": 0.40595426559448244,
      "num_input_tokens_seen": 4258822784,
      "step": 14980,
      "train_runtime": 145627.5212,
      "train_tokens_per_second": 29244.629
    },
    {
      "epoch": 0.5303765753294398,
      "grad_norm": 0.59375,
      "learning_rate": 3.162973590919187e-05,
      "loss": 0.4095478057861328,
      "num_input_tokens_seen": 4261609088,
      "step": 14990,
      "train_runtime": 145721.3922,
      "train_tokens_per_second": 29244.911
    },
    {
      "epoch": 0.5307303955931686,
      "grad_norm": 0.59375,
      "learning_rate": 3.162340907619012e-05,
      "loss": 0.40476217269897463,
      "num_input_tokens_seen": 4264439296,
      "step": 15000,
      "train_runtime": 145818.6201,
      "train_tokens_per_second": 29244.82
    },
    {
      "epoch": 0.5307303955931686,
      "eval_loss": 0.34381818771362305,
      "eval_runtime": 8.4257,
      "eval_samples_per_second": 473.311,
      "eval_steps_per_second": 3.798,
      "num_input_tokens_seen": 4264439296,
      "step": 15000
    },
    {
      "epoch": 0.5310842158568974,
      "grad_norm": 0.58984375,
      "learning_rate": 3.1617086038301516e-05,
      "loss": 0.4106022834777832,
      "num_input_tokens_seen": 4267301248,
      "step": 15010,
      "train_runtime": 145944.9471,
      "train_tokens_per_second": 29239.116
    },
    {
      "epoch": 0.5314380361206261,
      "grad_norm": 0.5703125,
      "learning_rate": 3.161076679173344e-05,
      "loss": 0.4102076530456543,
      "num_input_tokens_seen": 4270196416,
      "step": 15020,
      "train_runtime": 146044.743,
      "train_tokens_per_second": 29238.96
    },
    {
      "epoch": 0.531791856384355,
      "grad_norm": 0.59375,
      "learning_rate": 3.1604451332698575e-05,
      "loss": 0.4098836898803711,
      "num_input_tokens_seen": 4273007872,
      "step": 15030,
      "train_runtime": 146140.9001,
      "train_tokens_per_second": 29238.96
    },
    {
      "epoch": 0.5321456766480838,
      "grad_norm": 0.55859375,
      "learning_rate": 3.1598139657414923e-05,
      "loss": 0.40800275802612307,
      "num_input_tokens_seen": 4275906944,
      "step": 15040,
      "train_runtime": 146240.8114,
      "train_tokens_per_second": 29238.808
    },
    {
      "epoch": 0.5324994969118125,
      "grad_norm": 0.5625,
      "learning_rate": 3.159183176210574e-05,
      "loss": 0.4073368549346924,
      "num_input_tokens_seen": 4278740224,
      "step": 15050,
      "train_runtime": 146337.165,
      "train_tokens_per_second": 29238.917
    },
    {
      "epoch": 0.5328533171755413,
      "grad_norm": 0.6015625,
      "learning_rate": 3.1585527642999595e-05,
      "loss": 0.40842494964599607,
      "num_input_tokens_seen": 4281597824,
      "step": 15060,
      "train_runtime": 146434.478,
      "train_tokens_per_second": 29239.001
    },
    {
      "epoch": 0.5332071374392701,
      "grad_norm": 0.5703125,
      "learning_rate": 3.1579227296330294e-05,
      "loss": 0.40888800621032717,
      "num_input_tokens_seen": 4284404224,
      "step": 15070,
      "train_runtime": 146528.0021,
      "train_tokens_per_second": 29239.491
    },
    {
      "epoch": 0.5335609577029988,
      "grad_norm": 0.59765625,
      "learning_rate": 3.157293071833691e-05,
      "loss": 0.4051984786987305,
      "num_input_tokens_seen": 4287281280,
      "step": 15080,
      "train_runtime": 146626.0914,
      "train_tokens_per_second": 29239.552
    },
    {
      "epoch": 0.5339147779667276,
      "grad_norm": 0.6015625,
      "learning_rate": 3.156663790526375e-05,
      "loss": 0.413372278213501,
      "num_input_tokens_seen": 4290095232,
      "step": 15090,
      "train_runtime": 146719.6428,
      "train_tokens_per_second": 29240.088
    },
    {
      "epoch": 0.5342685982304564,
      "grad_norm": 0.5859375,
      "learning_rate": 3.156034885336039e-05,
      "loss": 0.4108406066894531,
      "num_input_tokens_seen": 4292953728,
      "step": 15100,
      "train_runtime": 146817.6716,
      "train_tokens_per_second": 29240.034
    },
    {
      "epoch": 0.5346224184941852,
      "grad_norm": 0.578125,
      "learning_rate": 3.155406355888161e-05,
      "loss": 0.40769710540771487,
      "num_input_tokens_seen": 4295809728,
      "step": 15110,
      "train_runtime": 146913.9085,
      "train_tokens_per_second": 29240.32
    },
    {
      "epoch": 0.534976238757914,
      "grad_norm": 0.59765625,
      "learning_rate": 3.1547782018087407e-05,
      "loss": 0.4042551517486572,
      "num_input_tokens_seen": 4298643584,
      "step": 15120,
      "train_runtime": 147011.0033,
      "train_tokens_per_second": 29240.285
    },
    {
      "epoch": 0.5353300590216428,
      "grad_norm": 0.55859375,
      "learning_rate": 3.154150422724299e-05,
      "loss": 0.40447430610656737,
      "num_input_tokens_seen": 4301509632,
      "step": 15130,
      "train_runtime": 147107.9939,
      "train_tokens_per_second": 29240.489
    },
    {
      "epoch": 0.5356838792853715,
      "grad_norm": 0.5703125,
      "learning_rate": 3.1535230182618783e-05,
      "loss": 0.40934176445007325,
      "num_input_tokens_seen": 4304355904,
      "step": 15140,
      "train_runtime": 147206.5555,
      "train_tokens_per_second": 29240.246
    },
    {
      "epoch": 0.5360376995491003,
      "grad_norm": 0.5546875,
      "learning_rate": 3.1528959880490387e-05,
      "loss": 0.4078420639038086,
      "num_input_tokens_seen": 4307218112,
      "step": 15150,
      "train_runtime": 147302.447,
      "train_tokens_per_second": 29240.642
    },
    {
      "epoch": 0.5363915198128291,
      "grad_norm": 0.5859375,
      "learning_rate": 3.152269331713859e-05,
      "loss": 0.4117298126220703,
      "num_input_tokens_seen": 4310026432,
      "step": 15160,
      "train_runtime": 147396.719,
      "train_tokens_per_second": 29240.993
    },
    {
      "epoch": 0.5367453400765578,
      "grad_norm": 0.5859375,
      "learning_rate": 3.151643048884935e-05,
      "loss": 0.41078572273254393,
      "num_input_tokens_seen": 4312905856,
      "step": 15170,
      "train_runtime": 147495.3044,
      "train_tokens_per_second": 29240.971
    },
    {
      "epoch": 0.5370991603402866,
      "grad_norm": 0.57421875,
      "learning_rate": 3.151017139191379e-05,
      "loss": 0.40717620849609376,
      "num_input_tokens_seen": 4315747840,
      "step": 15180,
      "train_runtime": 147595.5559,
      "train_tokens_per_second": 29240.364
    },
    {
      "epoch": 0.5374529806040154,
      "grad_norm": 0.58984375,
      "learning_rate": 3.150391602262818e-05,
      "loss": 0.4082365036010742,
      "num_input_tokens_seen": 4318558848,
      "step": 15190,
      "train_runtime": 147690.608,
      "train_tokens_per_second": 29240.579
    },
    {
      "epoch": 0.5378068008677442,
      "grad_norm": 0.58984375,
      "learning_rate": 3.149766437729394e-05,
      "loss": 0.40818538665771487,
      "num_input_tokens_seen": 4321412352,
      "step": 15200,
      "train_runtime": 147788.9902,
      "train_tokens_per_second": 29240.421
    },
    {
      "epoch": 0.538160621131473,
      "grad_norm": 0.58203125,
      "learning_rate": 3.1491416452217635e-05,
      "loss": 0.40845422744750975,
      "num_input_tokens_seen": 4324212544,
      "step": 15210,
      "train_runtime": 147883.0146,
      "train_tokens_per_second": 29240.765
    },
    {
      "epoch": 0.5385144413952018,
      "grad_norm": 0.546875,
      "learning_rate": 3.1485172243710954e-05,
      "loss": 0.40653200149536134,
      "num_input_tokens_seen": 4327046656,
      "step": 15220,
      "train_runtime": 147980.5588,
      "train_tokens_per_second": 29240.643
    },
    {
      "epoch": 0.5388682616589305,
      "grad_norm": 0.6015625,
      "learning_rate": 3.1478931748090676e-05,
      "loss": 0.4043598651885986,
      "num_input_tokens_seen": 4329853184,
      "step": 15230,
      "train_runtime": 148078.8799,
      "train_tokens_per_second": 29240.181
    },
    {
      "epoch": 0.5392220819226593,
      "grad_norm": 0.57421875,
      "learning_rate": 3.147269496167873e-05,
      "loss": 0.4014836311340332,
      "num_input_tokens_seen": 4332705792,
      "step": 15240,
      "train_runtime": 148177.291,
      "train_tokens_per_second": 29240.012
    },
    {
      "epoch": 0.5395759021863881,
      "grad_norm": 0.55859375,
      "learning_rate": 3.1466461880802105e-05,
      "loss": 0.40896382331848147,
      "num_input_tokens_seen": 4335562112,
      "step": 15250,
      "train_runtime": 148277.6878,
      "train_tokens_per_second": 29239.477
    },
    {
      "epoch": 0.5399297224501168,
      "grad_norm": 0.60546875,
      "learning_rate": 3.146023250179291e-05,
      "loss": 0.40951175689697267,
      "num_input_tokens_seen": 4338357376,
      "step": 15260,
      "train_runtime": 148372.5633,
      "train_tokens_per_second": 29239.62
    },
    {
      "epoch": 0.5402835427138456,
      "grad_norm": 0.58984375,
      "learning_rate": 3.145400682098833e-05,
      "loss": 0.4083584785461426,
      "num_input_tokens_seen": 4341228736,
      "step": 15270,
      "train_runtime": 148471.5553,
      "train_tokens_per_second": 29239.464
    },
    {
      "epoch": 0.5406373629775745,
      "grad_norm": 0.58984375,
      "learning_rate": 3.14477848347306e-05,
      "loss": 0.41114239692687987,
      "num_input_tokens_seen": 4344086976,
      "step": 15280,
      "train_runtime": 148569.1998,
      "train_tokens_per_second": 29239.486
    },
    {
      "epoch": 0.5409911832413032,
      "grad_norm": 0.59765625,
      "learning_rate": 3.144156653936703e-05,
      "loss": 0.4104591369628906,
      "num_input_tokens_seen": 4346898432,
      "step": 15290,
      "train_runtime": 148665.1077,
      "train_tokens_per_second": 29239.534
    },
    {
      "epoch": 0.541345003505032,
      "grad_norm": 0.59765625,
      "learning_rate": 3.1435351931250004e-05,
      "loss": 0.4089348316192627,
      "num_input_tokens_seen": 4349809920,
      "step": 15300,
      "train_runtime": 148768.2098,
      "train_tokens_per_second": 29238.84
    },
    {
      "epoch": 0.5416988237687608,
      "grad_norm": 0.5703125,
      "learning_rate": 3.142914100673692e-05,
      "loss": 0.4111007213592529,
      "num_input_tokens_seen": 4352621696,
      "step": 15310,
      "train_runtime": 148864.2382,
      "train_tokens_per_second": 29238.867
    },
    {
      "epoch": 0.5420526440324895,
      "grad_norm": 0.60546875,
      "learning_rate": 3.1422933762190235e-05,
      "loss": 0.40802755355834963,
      "num_input_tokens_seen": 4355470784,
      "step": 15320,
      "train_runtime": 148958.4792,
      "train_tokens_per_second": 29239.496
    },
    {
      "epoch": 0.5424064642962183,
      "grad_norm": 0.59765625,
      "learning_rate": 3.141673019397741e-05,
      "loss": 0.4092949390411377,
      "num_input_tokens_seen": 4358257216,
      "step": 15330,
      "train_runtime": 149051.4551,
      "train_tokens_per_second": 29239.951
    },
    {
      "epoch": 0.5427602845599471,
      "grad_norm": 0.578125,
      "learning_rate": 3.141053029847095e-05,
      "loss": 0.41081724166870115,
      "num_input_tokens_seen": 4361099712,
      "step": 15340,
      "train_runtime": 149148.5095,
      "train_tokens_per_second": 29239.982
    },
    {
      "epoch": 0.5431141048236758,
      "grad_norm": 0.58984375,
      "learning_rate": 3.140433407204835e-05,
      "loss": 0.40682077407836914,
      "num_input_tokens_seen": 4363925120,
      "step": 15350,
      "train_runtime": 149244.112,
      "train_tokens_per_second": 29240.183
    },
    {
      "epoch": 0.5434679250874047,
      "grad_norm": 0.578125,
      "learning_rate": 3.1398141511092124e-05,
      "loss": 0.41036109924316405,
      "num_input_tokens_seen": 4366730752,
      "step": 15360,
      "train_runtime": 149340.7947,
      "train_tokens_per_second": 29240.04
    },
    {
      "epoch": 0.5438217453511335,
      "grad_norm": 0.6015625,
      "learning_rate": 3.1391952611989736e-05,
      "loss": 0.4084867000579834,
      "num_input_tokens_seen": 4369636992,
      "step": 15370,
      "train_runtime": 149441.2993,
      "train_tokens_per_second": 29239.822
    },
    {
      "epoch": 0.5441755656148622,
      "grad_norm": 0.578125,
      "learning_rate": 3.138576737113369e-05,
      "loss": 0.4052896499633789,
      "num_input_tokens_seen": 4372542848,
      "step": 15380,
      "train_runtime": 149542.9996,
      "train_tokens_per_second": 29239.368
    },
    {
      "epoch": 0.544529385878591,
      "grad_norm": 0.59375,
      "learning_rate": 3.137958578492143e-05,
      "loss": 0.4081061363220215,
      "num_input_tokens_seen": 4375380224,
      "step": 15390,
      "train_runtime": 149640.1772,
      "train_tokens_per_second": 29239.341
    },
    {
      "epoch": 0.5448832061423198,
      "grad_norm": 0.5859375,
      "learning_rate": 3.137340784975535e-05,
      "loss": 0.407871150970459,
      "num_input_tokens_seen": 4378217408,
      "step": 15400,
      "train_runtime": 149734.0024,
      "train_tokens_per_second": 29239.968
    },
    {
      "epoch": 0.5452370264060485,
      "grad_norm": 0.61328125,
      "learning_rate": 3.136723356204285e-05,
      "loss": 0.40992279052734376,
      "num_input_tokens_seen": 4381097920,
      "step": 15410,
      "train_runtime": 149833.0925,
      "train_tokens_per_second": 29239.855
    },
    {
      "epoch": 0.5455908466697773,
      "grad_norm": 0.59375,
      "learning_rate": 3.136106291819622e-05,
      "loss": 0.40654382705688474,
      "num_input_tokens_seen": 4383983232,
      "step": 15420,
      "train_runtime": 149931.5595,
      "train_tokens_per_second": 29239.896
    },
    {
      "epoch": 0.5459446669335061,
      "grad_norm": 0.57421875,
      "learning_rate": 3.1354895914632736e-05,
      "loss": 0.4045723915100098,
      "num_input_tokens_seen": 4386808896,
      "step": 15430,
      "train_runtime": 150027.3028,
      "train_tokens_per_second": 29240.07
    },
    {
      "epoch": 0.546298487197235,
      "grad_norm": 0.58203125,
      "learning_rate": 3.1348732547774575e-05,
      "loss": 0.40632143020629885,
      "num_input_tokens_seen": 4389618368,
      "step": 15440,
      "train_runtime": 150118.6873,
      "train_tokens_per_second": 29240.986
    },
    {
      "epoch": 0.5466523074609637,
      "grad_norm": 0.5859375,
      "learning_rate": 3.134257281404883e-05,
      "loss": 0.41364545822143556,
      "num_input_tokens_seen": 4392389376,
      "step": 15450,
      "train_runtime": 150211.8433,
      "train_tokens_per_second": 29241.299
    },
    {
      "epoch": 0.5470061277246925,
      "grad_norm": 0.58984375,
      "learning_rate": 3.133641670988755e-05,
      "loss": 0.412398099899292,
      "num_input_tokens_seen": 4395211200,
      "step": 15460,
      "train_runtime": 150310.1197,
      "train_tokens_per_second": 29240.953
    },
    {
      "epoch": 0.5473599479884212,
      "grad_norm": 0.55859375,
      "learning_rate": 3.133026423172764e-05,
      "loss": 0.40074882507324217,
      "num_input_tokens_seen": 4398042880,
      "step": 15470,
      "train_runtime": 150406.5636,
      "train_tokens_per_second": 29241.03
    },
    {
      "epoch": 0.54771376825215,
      "grad_norm": 0.58203125,
      "learning_rate": 3.1324115376010897e-05,
      "loss": 0.4076112747192383,
      "num_input_tokens_seen": 4400883648,
      "step": 15480,
      "train_runtime": 150503.8433,
      "train_tokens_per_second": 29241.005
    },
    {
      "epoch": 0.5480675885158788,
      "grad_norm": 0.5859375,
      "learning_rate": 3.131797013918406e-05,
      "loss": 0.40844311714172366,
      "num_input_tokens_seen": 4403774336,
      "step": 15490,
      "train_runtime": 150603.816,
      "train_tokens_per_second": 29240.789
    },
    {
      "epoch": 0.5484214087796075,
      "grad_norm": 0.58203125,
      "learning_rate": 3.13118285176987e-05,
      "loss": 0.4066895008087158,
      "num_input_tokens_seen": 4406650432,
      "step": 15500,
      "train_runtime": 150705.3559,
      "train_tokens_per_second": 29240.171
    },
    {
      "epoch": 0.5487752290433363,
      "grad_norm": 0.578125,
      "learning_rate": 3.130569050801126e-05,
      "loss": 0.40573415756225584,
      "num_input_tokens_seen": 4409492992,
      "step": 15510,
      "train_runtime": 150807.5935,
      "train_tokens_per_second": 29239.197
    },
    {
      "epoch": 0.5491290493070651,
      "grad_norm": 0.57421875,
      "learning_rate": 3.129955610658306e-05,
      "loss": 0.4061446666717529,
      "num_input_tokens_seen": 4412328704,
      "step": 15520,
      "train_runtime": 150904.5021,
      "train_tokens_per_second": 29239.212
    },
    {
      "epoch": 0.549482869570794,
      "grad_norm": 0.6015625,
      "learning_rate": 3.129342530988025e-05,
      "loss": 0.4072696685791016,
      "num_input_tokens_seen": 4415215616,
      "step": 15530,
      "train_runtime": 151004.4551,
      "train_tokens_per_second": 29238.976
    },
    {
      "epoch": 0.5498366898345227,
      "grad_norm": 0.57421875,
      "learning_rate": 3.128729811437385e-05,
      "loss": 0.40645294189453124,
      "num_input_tokens_seen": 4418009344,
      "step": 15540,
      "train_runtime": 151097.3916,
      "train_tokens_per_second": 29239.481
    },
    {
      "epoch": 0.5501905100982515,
      "grad_norm": 0.56640625,
      "learning_rate": 3.12811745165397e-05,
      "loss": 0.4098991870880127,
      "num_input_tokens_seen": 4420809344,
      "step": 15550,
      "train_runtime": 151192.8314,
      "train_tokens_per_second": 29239.543
    },
    {
      "epoch": 0.5505443303619802,
      "grad_norm": 0.59375,
      "learning_rate": 3.127505451285846e-05,
      "loss": 0.4075554370880127,
      "num_input_tokens_seen": 4423656192,
      "step": 15560,
      "train_runtime": 151288.49,
      "train_tokens_per_second": 29239.873
    },
    {
      "epoch": 0.550898150625709,
      "grad_norm": 0.6015625,
      "learning_rate": 3.126893809981563e-05,
      "loss": 0.40610518455505373,
      "num_input_tokens_seen": 4426493120,
      "step": 15570,
      "train_runtime": 151384.0837,
      "train_tokens_per_second": 29240.149
    },
    {
      "epoch": 0.5512519708894378,
      "grad_norm": 0.5859375,
      "learning_rate": 3.126282527390149e-05,
      "loss": 0.4054409503936768,
      "num_input_tokens_seen": 4429351104,
      "step": 15580,
      "train_runtime": 151481.963,
      "train_tokens_per_second": 29240.122
    },
    {
      "epoch": 0.5516057911531665,
      "grad_norm": 0.56640625,
      "learning_rate": 3.1256716031611146e-05,
      "loss": 0.4078982353210449,
      "num_input_tokens_seen": 4432209088,
      "step": 15590,
      "train_runtime": 151580.8035,
      "train_tokens_per_second": 29239.91
    },
    {
      "epoch": 0.5519596114168953,
      "grad_norm": 0.6015625,
      "learning_rate": 3.1250610369444475e-05,
      "loss": 0.40864033699035646,
      "num_input_tokens_seen": 4435005440,
      "step": 15600,
      "train_runtime": 151675.0443,
      "train_tokens_per_second": 29240.179
    },
    {
      "epoch": 0.5523134316806242,
      "grad_norm": 0.58203125,
      "learning_rate": 3.124450828390616e-05,
      "loss": 0.4037106990814209,
      "num_input_tokens_seen": 4437808960,
      "step": 15610,
      "train_runtime": 151769.0708,
      "train_tokens_per_second": 29240.536
    },
    {
      "epoch": 0.552667251944353,
      "grad_norm": 0.59375,
      "learning_rate": 3.123840977150566e-05,
      "loss": 0.40448718070983886,
      "num_input_tokens_seen": 4440682112,
      "step": 15620,
      "train_runtime": 151867.4481,
      "train_tokens_per_second": 29240.513
    },
    {
      "epoch": 0.5530210722080817,
      "grad_norm": 0.59375,
      "learning_rate": 3.123231482875717e-05,
      "loss": 0.41243600845336914,
      "num_input_tokens_seen": 4443432064,
      "step": 15630,
      "train_runtime": 151960.5821,
      "train_tokens_per_second": 29240.689
    },
    {
      "epoch": 0.5533748924718105,
      "grad_norm": 0.58203125,
      "learning_rate": 3.122622345217967e-05,
      "loss": 0.4055476665496826,
      "num_input_tokens_seen": 4446262656,
      "step": 15640,
      "train_runtime": 152056.0163,
      "train_tokens_per_second": 29240.952
    },
    {
      "epoch": 0.5537287127355393,
      "grad_norm": 0.58203125,
      "learning_rate": 3.12201356382969e-05,
      "loss": 0.40500941276550295,
      "num_input_tokens_seen": 4449059968,
      "step": 15650,
      "train_runtime": 152148.725,
      "train_tokens_per_second": 29241.52
    },
    {
      "epoch": 0.554082532999268,
      "grad_norm": 0.5703125,
      "learning_rate": 3.1214051383637304e-05,
      "loss": 0.40358848571777345,
      "num_input_tokens_seen": 4451883072,
      "step": 15660,
      "train_runtime": 152245.8642,
      "train_tokens_per_second": 29241.406
    },
    {
      "epoch": 0.5544363532629968,
      "grad_norm": 0.6015625,
      "learning_rate": 3.1207970684734104e-05,
      "loss": 0.4064045429229736,
      "num_input_tokens_seen": 4454731648,
      "step": 15670,
      "train_runtime": 152344.2155,
      "train_tokens_per_second": 29241.226
    },
    {
      "epoch": 0.5547901735267255,
      "grad_norm": 0.60546875,
      "learning_rate": 3.120189353812521e-05,
      "loss": 0.40747699737548826,
      "num_input_tokens_seen": 4457504576,
      "step": 15680,
      "train_runtime": 152438.5674,
      "train_tokens_per_second": 29241.318
    },
    {
      "epoch": 0.5551439937904544,
      "grad_norm": 0.62890625,
      "learning_rate": 3.1195819940353294e-05,
      "loss": 0.40999584197998046,
      "num_input_tokens_seen": 4460403072,
      "step": 15690,
      "train_runtime": 152539.1046,
      "train_tokens_per_second": 29241.047
    },
    {
      "epoch": 0.5554978140541832,
      "grad_norm": 0.578125,
      "learning_rate": 3.1189749887965684e-05,
      "loss": 0.40561342239379883,
      "num_input_tokens_seen": 4463253248,
      "step": 15700,
      "train_runtime": 152636.0665,
      "train_tokens_per_second": 29241.144
    },
    {
      "epoch": 0.555851634317912,
      "grad_norm": 0.58984375,
      "learning_rate": 3.118368337751443e-05,
      "loss": 0.40368070602416994,
      "num_input_tokens_seen": 4466122112,
      "step": 15710,
      "train_runtime": 152734.2963,
      "train_tokens_per_second": 29241.121
    },
    {
      "epoch": 0.5562054545816407,
      "grad_norm": 0.59765625,
      "learning_rate": 3.1177620405556305e-05,
      "loss": 0.4062944412231445,
      "num_input_tokens_seen": 4468945472,
      "step": 15720,
      "train_runtime": 152827.5742,
      "train_tokens_per_second": 29241.748
    },
    {
      "epoch": 0.5565592748453695,
      "grad_norm": 0.58984375,
      "learning_rate": 3.117156096865272e-05,
      "loss": 0.40369558334350586,
      "num_input_tokens_seen": 4471820544,
      "step": 15730,
      "train_runtime": 152924.9119,
      "train_tokens_per_second": 29241.936
    },
    {
      "epoch": 0.5569130951090983,
      "grad_norm": 0.578125,
      "learning_rate": 3.116550506336979e-05,
      "loss": 0.4079303741455078,
      "num_input_tokens_seen": 4474630848,
      "step": 15740,
      "train_runtime": 153018.6344,
      "train_tokens_per_second": 29242.392
    },
    {
      "epoch": 0.557266915372827,
      "grad_norm": 0.5859375,
      "learning_rate": 3.11594526862783e-05,
      "loss": 0.4047048568725586,
      "num_input_tokens_seen": 4477449344,
      "step": 15750,
      "train_runtime": 153115.5054,
      "train_tokens_per_second": 29242.299
    },
    {
      "epoch": 0.5576207356365558,
      "grad_norm": 0.58203125,
      "learning_rate": 3.115340383395367e-05,
      "loss": 0.4134345054626465,
      "num_input_tokens_seen": 4480296448,
      "step": 15760,
      "train_runtime": 153212.7903,
      "train_tokens_per_second": 29242.314
    },
    {
      "epoch": 0.5579745559002846,
      "grad_norm": 0.6015625,
      "learning_rate": 3.1147358502975995e-05,
      "loss": 0.40609130859375,
      "num_input_tokens_seen": 4483132544,
      "step": 15770,
      "train_runtime": 153307.6213,
      "train_tokens_per_second": 29242.725
    },
    {
      "epoch": 0.5583283761640134,
      "grad_norm": 0.59375,
      "learning_rate": 3.114131668993e-05,
      "loss": 0.4067939281463623,
      "num_input_tokens_seen": 4486022528,
      "step": 15780,
      "train_runtime": 153409.0957,
      "train_tokens_per_second": 29242.22
    },
    {
      "epoch": 0.5586821964277422,
      "grad_norm": 0.55859375,
      "learning_rate": 3.113527839140507e-05,
      "loss": 0.40735297203063964,
      "num_input_tokens_seen": 4488831680,
      "step": 15790,
      "train_runtime": 153503.6308,
      "train_tokens_per_second": 29242.511
    },
    {
      "epoch": 0.559036016691471,
      "grad_norm": 0.5703125,
      "learning_rate": 3.112924360399517e-05,
      "loss": 0.40956878662109375,
      "num_input_tokens_seen": 4491701632,
      "step": 15800,
      "train_runtime": 153599.6774,
      "train_tokens_per_second": 29242.911
    },
    {
      "epoch": 0.5593898369551997,
      "grad_norm": 0.5703125,
      "learning_rate": 3.112321232429894e-05,
      "loss": 0.4095731735229492,
      "num_input_tokens_seen": 4494555328,
      "step": 15810,
      "train_runtime": 153698.6024,
      "train_tokens_per_second": 29242.656
    },
    {
      "epoch": 0.5597436572189285,
      "grad_norm": 0.57421875,
      "learning_rate": 3.11171845489196e-05,
      "loss": 0.4067424774169922,
      "num_input_tokens_seen": 4497396288,
      "step": 15820,
      "train_runtime": 153798.361,
      "train_tokens_per_second": 29242.16
    },
    {
      "epoch": 0.5600974774826573,
      "grad_norm": 0.58203125,
      "learning_rate": 3.1111160274464965e-05,
      "loss": 0.4107954025268555,
      "num_input_tokens_seen": 4500270336,
      "step": 15830,
      "train_runtime": 153895.5166,
      "train_tokens_per_second": 29242.375
    },
    {
      "epoch": 0.560451297746386,
      "grad_norm": 0.578125,
      "learning_rate": 3.1105139497547466e-05,
      "loss": 0.409916353225708,
      "num_input_tokens_seen": 4503136000,
      "step": 15840,
      "train_runtime": 153993.901,
      "train_tokens_per_second": 29242.301
    },
    {
      "epoch": 0.5608051180101148,
      "grad_norm": 0.57421875,
      "learning_rate": 3.109912221478411e-05,
      "loss": 0.40704822540283203,
      "num_input_tokens_seen": 4505967552,
      "step": 15850,
      "train_runtime": 154086.6973,
      "train_tokens_per_second": 29243.067
    },
    {
      "epoch": 0.5611589382738437,
      "grad_norm": 0.5859375,
      "learning_rate": 3.10931084227965e-05,
      "loss": 0.4107632637023926,
      "num_input_tokens_seen": 4508822272,
      "step": 15860,
      "train_runtime": 154185.1504,
      "train_tokens_per_second": 29242.909
    },
    {
      "epoch": 0.5615127585375724,
      "grad_norm": 0.578125,
      "learning_rate": 3.108709811821079e-05,
      "loss": 0.4064638137817383,
      "num_input_tokens_seen": 4511687872,
      "step": 15870,
      "train_runtime": 154280.9361,
      "train_tokens_per_second": 29243.327
    },
    {
      "epoch": 0.5618665788013012,
      "grad_norm": 0.59765625,
      "learning_rate": 3.1081091297657696e-05,
      "loss": 0.4065570831298828,
      "num_input_tokens_seen": 4514513728,
      "step": 15880,
      "train_runtime": 154376.883,
      "train_tokens_per_second": 29243.457
    },
    {
      "epoch": 0.56222039906503,
      "grad_norm": 0.55078125,
      "learning_rate": 3.107508795777251e-05,
      "loss": 0.40776753425598145,
      "num_input_tokens_seen": 4517418432,
      "step": 15890,
      "train_runtime": 154477.0895,
      "train_tokens_per_second": 29243.291
    },
    {
      "epoch": 0.5625742193287587,
      "grad_norm": 0.578125,
      "learning_rate": 3.1069088095195056e-05,
      "loss": 0.4102058410644531,
      "num_input_tokens_seen": 4520242240,
      "step": 15900,
      "train_runtime": 154573.3451,
      "train_tokens_per_second": 29243.349
    },
    {
      "epoch": 0.5629280395924875,
      "grad_norm": 0.6015625,
      "learning_rate": 3.1063091706569703e-05,
      "loss": 0.4087839126586914,
      "num_input_tokens_seen": 4523114112,
      "step": 15910,
      "train_runtime": 154673.2742,
      "train_tokens_per_second": 29243.023
    },
    {
      "epoch": 0.5632818598562163,
      "grad_norm": 0.60546875,
      "learning_rate": 3.105709878854535e-05,
      "loss": 0.4062658309936523,
      "num_input_tokens_seen": 4525934080,
      "step": 15920,
      "train_runtime": 154767.094,
      "train_tokens_per_second": 29243.517
    },
    {
      "epoch": 0.563635680119945,
      "grad_norm": 0.5703125,
      "learning_rate": 3.105110933777541e-05,
      "loss": 0.4077878475189209,
      "num_input_tokens_seen": 4528804608,
      "step": 15930,
      "train_runtime": 154866.661,
      "train_tokens_per_second": 29243.251
    },
    {
      "epoch": 0.5639895003836739,
      "grad_norm": 0.57421875,
      "learning_rate": 3.104512335091783e-05,
      "loss": 0.4015920162200928,
      "num_input_tokens_seen": 4531621824,
      "step": 15940,
      "train_runtime": 154962.2198,
      "train_tokens_per_second": 29243.398
    },
    {
      "epoch": 0.5643433206474027,
      "grad_norm": 0.55859375,
      "learning_rate": 3.1039140824635074e-05,
      "loss": 0.40623836517333983,
      "num_input_tokens_seen": 4534456960,
      "step": 15950,
      "train_runtime": 155057.8415,
      "train_tokens_per_second": 29243.648
    },
    {
      "epoch": 0.5646971409111314,
      "grad_norm": 0.55859375,
      "learning_rate": 3.103316175559406e-05,
      "loss": 0.4080641746520996,
      "num_input_tokens_seen": 4537251712,
      "step": 15960,
      "train_runtime": 155149.6821,
      "train_tokens_per_second": 29244.351
    },
    {
      "epoch": 0.5650509611748602,
      "grad_norm": 0.5546875,
      "learning_rate": 3.102718614046624e-05,
      "loss": 0.4091509819030762,
      "num_input_tokens_seen": 4540102528,
      "step": 15970,
      "train_runtime": 155248.8883,
      "train_tokens_per_second": 29244.026
    },
    {
      "epoch": 0.565404781438589,
      "grad_norm": 0.59375,
      "learning_rate": 3.1021213975927546e-05,
      "loss": 0.40813302993774414,
      "num_input_tokens_seen": 4542922176,
      "step": 15980,
      "train_runtime": 155344.8838,
      "train_tokens_per_second": 29244.106
    },
    {
      "epoch": 0.5657586017023177,
      "grad_norm": 0.59765625,
      "learning_rate": 3.1015245258658374e-05,
      "loss": 0.40922961235046384,
      "num_input_tokens_seen": 4545801856,
      "step": 15990,
      "train_runtime": 155444.3212,
      "train_tokens_per_second": 29243.924
    },
    {
      "epoch": 0.5661124219660465,
      "grad_norm": 0.58203125,
      "learning_rate": 3.1009279985343606e-05,
      "loss": 0.40422806739807127,
      "num_input_tokens_seen": 4548607936,
      "step": 16000,
      "train_runtime": 155538.0153,
      "train_tokens_per_second": 29244.349
    },
    {
      "epoch": 0.5664662422297753,
      "grad_norm": 0.5859375,
      "learning_rate": 3.100331815267255e-05,
      "loss": 0.4057432174682617,
      "num_input_tokens_seen": 4551433344,
      "step": 16010,
      "train_runtime": 155634.0775,
      "train_tokens_per_second": 29244.452
    },
    {
      "epoch": 0.566820062493504,
      "grad_norm": 0.62109375,
      "learning_rate": 3.099735975733902e-05,
      "loss": 0.40664067268371584,
      "num_input_tokens_seen": 4554319296,
      "step": 16020,
      "train_runtime": 155733.7403,
      "train_tokens_per_second": 29244.268
    },
    {
      "epoch": 0.5671738827572329,
      "grad_norm": 0.5859375,
      "learning_rate": 3.099140479604123e-05,
      "loss": 0.4142652988433838,
      "num_input_tokens_seen": 4557161024,
      "step": 16030,
      "train_runtime": 155830.9316,
      "train_tokens_per_second": 29244.265
    },
    {
      "epoch": 0.5675277030209617,
      "grad_norm": 0.58984375,
      "learning_rate": 3.098545326548186e-05,
      "loss": 0.4083212375640869,
      "num_input_tokens_seen": 4559967424,
      "step": 16040,
      "train_runtime": 155927.8103,
      "train_tokens_per_second": 29244.093
    },
    {
      "epoch": 0.5678815232846904,
      "grad_norm": 0.59765625,
      "learning_rate": 3.0979505162368014e-05,
      "loss": 0.406848669052124,
      "num_input_tokens_seen": 4562896704,
      "step": 16050,
      "train_runtime": 156031.1765,
      "train_tokens_per_second": 29243.494
    },
    {
      "epoch": 0.5682353435484192,
      "grad_norm": 0.56640625,
      "learning_rate": 3.097356048341121e-05,
      "loss": 0.4030649185180664,
      "num_input_tokens_seen": 4565749120,
      "step": 16060,
      "train_runtime": 156129.3172,
      "train_tokens_per_second": 29243.381
    },
    {
      "epoch": 0.568589163812148,
      "grad_norm": 0.578125,
      "learning_rate": 3.0967619225327396e-05,
      "loss": 0.4107643127441406,
      "num_input_tokens_seen": 4568584384,
      "step": 16070,
      "train_runtime": 156224.5034,
      "train_tokens_per_second": 29243.712
    },
    {
      "epoch": 0.5689429840758767,
      "grad_norm": 0.57421875,
      "learning_rate": 3.0961681384836924e-05,
      "loss": 0.40882062911987305,
      "num_input_tokens_seen": 4571418688,
      "step": 16080,
      "train_runtime": 156318.6348,
      "train_tokens_per_second": 29244.234
    },
    {
      "epoch": 0.5692968043396055,
      "grad_norm": 0.58203125,
      "learning_rate": 3.095574695866453e-05,
      "loss": 0.40875973701477053,
      "num_input_tokens_seen": 4574281664,
      "step": 16090,
      "train_runtime": 156416.8914,
      "train_tokens_per_second": 29244.167
    },
    {
      "epoch": 0.5696506246033343,
      "grad_norm": 0.578125,
      "learning_rate": 3.0949815943539355e-05,
      "loss": 0.4070535659790039,
      "num_input_tokens_seen": 4577187776,
      "step": 16100,
      "train_runtime": 156518.9301,
      "train_tokens_per_second": 29243.669
    },
    {
      "epoch": 0.5700044448670631,
      "grad_norm": 0.5625,
      "learning_rate": 3.094388833619495e-05,
      "loss": 0.40699381828308107,
      "num_input_tokens_seen": 4580032448,
      "step": 16110,
      "train_runtime": 156614.3319,
      "train_tokens_per_second": 29244.019
    },
    {
      "epoch": 0.5703582651307919,
      "grad_norm": 0.61328125,
      "learning_rate": 3.0937964133369196e-05,
      "loss": 0.40806264877319337,
      "num_input_tokens_seen": 4582878016,
      "step": 16120,
      "train_runtime": 156712.1107,
      "train_tokens_per_second": 29243.93
    },
    {
      "epoch": 0.5707120853945207,
      "grad_norm": 0.6015625,
      "learning_rate": 3.093204333180437e-05,
      "loss": 0.41038851737976073,
      "num_input_tokens_seen": 4585753088,
      "step": 16130,
      "train_runtime": 156809.9067,
      "train_tokens_per_second": 29244.027
    },
    {
      "epoch": 0.5710659056582494,
      "grad_norm": 0.58984375,
      "learning_rate": 3.0926125928247114e-05,
      "loss": 0.40847148895263674,
      "num_input_tokens_seen": 4588563328,
      "step": 16140,
      "train_runtime": 156902.3511,
      "train_tokens_per_second": 29244.707
    },
    {
      "epoch": 0.5714197259219782,
      "grad_norm": 0.58984375,
      "learning_rate": 3.092021191944842e-05,
      "loss": 0.4050747394561768,
      "num_input_tokens_seen": 4591463744,
      "step": 16150,
      "train_runtime": 157001.6358,
      "train_tokens_per_second": 29244.687
    },
    {
      "epoch": 0.571773546185707,
      "grad_norm": 0.609375,
      "learning_rate": 3.091430130216363e-05,
      "loss": 0.4123250961303711,
      "num_input_tokens_seen": 4594273024,
      "step": 16160,
      "train_runtime": 157097.8519,
      "train_tokens_per_second": 29244.658
    },
    {
      "epoch": 0.5721273664494357,
      "grad_norm": 0.578125,
      "learning_rate": 3.090839407315242e-05,
      "loss": 0.40061559677124026,
      "num_input_tokens_seen": 4597126080,
      "step": 16170,
      "train_runtime": 157196.9559,
      "train_tokens_per_second": 29244.371
    },
    {
      "epoch": 0.5724811867131645,
      "grad_norm": 0.57421875,
      "learning_rate": 3.0902490229178794e-05,
      "loss": 0.4055177688598633,
      "num_input_tokens_seen": 4600000640,
      "step": 16180,
      "train_runtime": 157294.2613,
      "train_tokens_per_second": 29244.555
    },
    {
      "epoch": 0.5728350069768934,
      "grad_norm": 0.5859375,
      "learning_rate": 3.0896589767011104e-05,
      "loss": 0.4096518039703369,
      "num_input_tokens_seen": 4602910592,
      "step": 16190,
      "train_runtime": 157396.7767,
      "train_tokens_per_second": 29243.995
    },
    {
      "epoch": 0.5731888272406221,
      "grad_norm": 0.56640625,
      "learning_rate": 3.0890692683421985e-05,
      "loss": 0.40958428382873535,
      "num_input_tokens_seen": 4605728704,
      "step": 16200,
      "train_runtime": 157493.9156,
      "train_tokens_per_second": 29243.852
    },
    {
      "epoch": 0.5735426475043509,
      "grad_norm": 0.578125,
      "learning_rate": 3.088479897518843e-05,
      "loss": 0.4066947937011719,
      "num_input_tokens_seen": 4608575232,
      "step": 16210,
      "train_runtime": 157593.0509,
      "train_tokens_per_second": 29243.518
    },
    {
      "epoch": 0.5738964677680797,
      "grad_norm": 0.58203125,
      "learning_rate": 3.087890863909168e-05,
      "loss": 0.4038949966430664,
      "num_input_tokens_seen": 4611442048,
      "step": 16220,
      "train_runtime": 157691.8594,
      "train_tokens_per_second": 29243.374
    },
    {
      "epoch": 0.5742502880318084,
      "grad_norm": 0.58203125,
      "learning_rate": 3.087302167191732e-05,
      "loss": 0.40948991775512694,
      "num_input_tokens_seen": 4614256512,
      "step": 16230,
      "train_runtime": 157786.625,
      "train_tokens_per_second": 29243.648
    },
    {
      "epoch": 0.5746041082955372,
      "grad_norm": 0.58203125,
      "learning_rate": 3.086713807045518e-05,
      "loss": 0.4068491458892822,
      "num_input_tokens_seen": 4617124736,
      "step": 16240,
      "train_runtime": 157883.9649,
      "train_tokens_per_second": 29243.785
    },
    {
      "epoch": 0.574957928559266,
      "grad_norm": 0.5703125,
      "learning_rate": 3.086125783149941e-05,
      "loss": 0.41111011505126954,
      "num_input_tokens_seen": 4620004160,
      "step": 16250,
      "train_runtime": 157985.0436,
      "train_tokens_per_second": 29243.301
    },
    {
      "epoch": 0.5753117488229947,
      "grad_norm": 0.58984375,
      "learning_rate": 3.0855380951848404e-05,
      "loss": 0.4075753688812256,
      "num_input_tokens_seen": 4622785792,
      "step": 16260,
      "train_runtime": 158077.5438,
      "train_tokens_per_second": 29243.786
    },
    {
      "epoch": 0.5756655690867235,
      "grad_norm": 0.5546875,
      "learning_rate": 3.084950742830484e-05,
      "loss": 0.40226335525512696,
      "num_input_tokens_seen": 4625604160,
      "step": 16270,
      "train_runtime": 158172.6473,
      "train_tokens_per_second": 29244.021
    },
    {
      "epoch": 0.5760193893504524,
      "grad_norm": 0.60546875,
      "learning_rate": 3.0843637257675654e-05,
      "loss": 0.40446810722351073,
      "num_input_tokens_seen": 4628437632,
      "step": 16280,
      "train_runtime": 158269.7433,
      "train_tokens_per_second": 29243.983
    },
    {
      "epoch": 0.5763732096141811,
      "grad_norm": 0.56640625,
      "learning_rate": 3.083777043677202e-05,
      "loss": 0.4045380115509033,
      "num_input_tokens_seen": 4631296000,
      "step": 16290,
      "train_runtime": 158370.1772,
      "train_tokens_per_second": 29243.486
    },
    {
      "epoch": 0.5767270298779099,
      "grad_norm": 0.5859375,
      "learning_rate": 3.083190696240936e-05,
      "loss": 0.4096792221069336,
      "num_input_tokens_seen": 4634136512,
      "step": 16300,
      "train_runtime": 158467.3551,
      "train_tokens_per_second": 29243.477
    },
    {
      "epoch": 0.5770808501416387,
      "grad_norm": 0.59765625,
      "learning_rate": 3.082604683140735e-05,
      "loss": 0.4100943565368652,
      "num_input_tokens_seen": 4636912320,
      "step": 16310,
      "train_runtime": 158560.5634,
      "train_tokens_per_second": 29243.793
    },
    {
      "epoch": 0.5774346704053674,
      "grad_norm": 0.58984375,
      "learning_rate": 3.082019004058987e-05,
      "loss": 0.4057792663574219,
      "num_input_tokens_seen": 4639789184,
      "step": 16320,
      "train_runtime": 158658.9235,
      "train_tokens_per_second": 29243.796
    },
    {
      "epoch": 0.5777884906690962,
      "grad_norm": 0.6171875,
      "learning_rate": 3.081433658678505e-05,
      "loss": 0.4112861156463623,
      "num_input_tokens_seen": 4642627456,
      "step": 16330,
      "train_runtime": 158758.137,
      "train_tokens_per_second": 29243.398
    },
    {
      "epoch": 0.578142310932825,
      "grad_norm": 0.61328125,
      "learning_rate": 3.080848646682521e-05,
      "loss": 0.40722041130065917,
      "num_input_tokens_seen": 4645507840,
      "step": 16340,
      "train_runtime": 158858.0436,
      "train_tokens_per_second": 29243.139
    },
    {
      "epoch": 0.5784961311965537,
      "grad_norm": 0.59765625,
      "learning_rate": 3.080263967754689e-05,
      "loss": 0.4033486366271973,
      "num_input_tokens_seen": 4648344960,
      "step": 16350,
      "train_runtime": 158953.8647,
      "train_tokens_per_second": 29243.359
    },
    {
      "epoch": 0.5788499514602826,
      "grad_norm": 0.59375,
      "learning_rate": 3.079679621579083e-05,
      "loss": 0.40462641716003417,
      "num_input_tokens_seen": 4651201088,
      "step": 16360,
      "train_runtime": 159049.4858,
      "train_tokens_per_second": 29243.735
    },
    {
      "epoch": 0.5792037717240114,
      "grad_norm": 0.5703125,
      "learning_rate": 3.079095607840197e-05,
      "loss": 0.41025028228759763,
      "num_input_tokens_seen": 4654067008,
      "step": 16370,
      "train_runtime": 159147.6383,
      "train_tokens_per_second": 29243.708
    },
    {
      "epoch": 0.5795575919877402,
      "grad_norm": 0.55859375,
      "learning_rate": 3.0785119262229416e-05,
      "loss": 0.4053900718688965,
      "num_input_tokens_seen": 4656871680,
      "step": 16380,
      "train_runtime": 159240.7007,
      "train_tokens_per_second": 29244.23
    },
    {
      "epoch": 0.5799114122514689,
      "grad_norm": 0.58984375,
      "learning_rate": 3.0779285764126485e-05,
      "loss": 0.4062835693359375,
      "num_input_tokens_seen": 4659745472,
      "step": 16390,
      "train_runtime": 159337.1469,
      "train_tokens_per_second": 29244.565
    },
    {
      "epoch": 0.5802652325151977,
      "grad_norm": 0.60546875,
      "learning_rate": 3.077345558095064e-05,
      "loss": 0.40163383483886717,
      "num_input_tokens_seen": 4662502656,
      "step": 16400,
      "train_runtime": 159426.5931,
      "train_tokens_per_second": 29245.451
    },
    {
      "epoch": 0.5806190527789264,
      "grad_norm": 0.5625,
      "learning_rate": 3.076762870956352e-05,
      "loss": 0.4057318687438965,
      "num_input_tokens_seen": 4665291584,
      "step": 16410,
      "train_runtime": 159522.1971,
      "train_tokens_per_second": 29245.407
    },
    {
      "epoch": 0.5809728730426552,
      "grad_norm": 0.6015625,
      "learning_rate": 3.076180514683092e-05,
      "loss": 0.4051515579223633,
      "num_input_tokens_seen": 4668126080,
      "step": 16420,
      "train_runtime": 159620.6695,
      "train_tokens_per_second": 29245.123
    },
    {
      "epoch": 0.581326693306384,
      "grad_norm": 0.60546875,
      "learning_rate": 3.075598488962278e-05,
      "loss": 0.4090378761291504,
      "num_input_tokens_seen": 4670958656,
      "step": 16430,
      "train_runtime": 159718.0407,
      "train_tokens_per_second": 29245.029
    },
    {
      "epoch": 0.5816805135701129,
      "grad_norm": 0.56640625,
      "learning_rate": 3.0750167934813207e-05,
      "loss": 0.4114226341247559,
      "num_input_tokens_seen": 4673823552,
      "step": 16440,
      "train_runtime": 159818.5283,
      "train_tokens_per_second": 29244.566
    },
    {
      "epoch": 0.5820343338338416,
      "grad_norm": 0.5859375,
      "learning_rate": 3.074435427928041e-05,
      "loss": 0.41299262046813967,
      "num_input_tokens_seen": 4676659520,
      "step": 16450,
      "train_runtime": 159915.0739,
      "train_tokens_per_second": 29244.645
    },
    {
      "epoch": 0.5823881540975704,
      "grad_norm": 0.5703125,
      "learning_rate": 3.073854391990675e-05,
      "loss": 0.40126829147338866,
      "num_input_tokens_seen": 4679561536,
      "step": 16460,
      "train_runtime": 160014.074,
      "train_tokens_per_second": 29244.687
    },
    {
      "epoch": 0.5827419743612992,
      "grad_norm": 0.55859375,
      "learning_rate": 3.073273685357871e-05,
      "loss": 0.40436415672302245,
      "num_input_tokens_seen": 4682363072,
      "step": 16470,
      "train_runtime": 160108.2698,
      "train_tokens_per_second": 29244.98
    },
    {
      "epoch": 0.5830957946250279,
      "grad_norm": 0.59765625,
      "learning_rate": 3.072693307718689e-05,
      "loss": 0.405302095413208,
      "num_input_tokens_seen": 4685215616,
      "step": 16480,
      "train_runtime": 160206.4422,
      "train_tokens_per_second": 29244.864
    },
    {
      "epoch": 0.5834496148887567,
      "grad_norm": 0.57421875,
      "learning_rate": 3.0721132587626e-05,
      "loss": 0.40462818145751955,
      "num_input_tokens_seen": 4687993344,
      "step": 16490,
      "train_runtime": 160298.7025,
      "train_tokens_per_second": 29245.361
    },
    {
      "epoch": 0.5838034351524855,
      "grad_norm": 0.59765625,
      "learning_rate": 3.071533538179482e-05,
      "loss": 0.4112078666687012,
      "num_input_tokens_seen": 4690838592,
      "step": 16500,
      "train_runtime": 160397.6336,
      "train_tokens_per_second": 29245.061
    },
    {
      "epoch": 0.5841572554162142,
      "grad_norm": 0.58984375,
      "learning_rate": 3.070954145659627e-05,
      "loss": 0.4079756259918213,
      "num_input_tokens_seen": 4693651456,
      "step": 16510,
      "train_runtime": 160492.3953,
      "train_tokens_per_second": 29245.32
    },
    {
      "epoch": 0.584511075679943,
      "grad_norm": 0.59765625,
      "learning_rate": 3.070375080893734e-05,
      "loss": 0.40964694023132325,
      "num_input_tokens_seen": 4696469440,
      "step": 16520,
      "train_runtime": 160584.6252,
      "train_tokens_per_second": 29246.072
    },
    {
      "epoch": 0.5848648959436719,
      "grad_norm": 0.56640625,
      "learning_rate": 3.06979634357291e-05,
      "loss": 0.40720019340515134,
      "num_input_tokens_seen": 4699296576,
      "step": 16530,
      "train_runtime": 160680.6172,
      "train_tokens_per_second": 29246.194
    },
    {
      "epoch": 0.5852187162074006,
      "grad_norm": 0.5546875,
      "learning_rate": 3.0692179333886686e-05,
      "loss": 0.4105440616607666,
      "num_input_tokens_seen": 4702138112,
      "step": 16540,
      "train_runtime": 160775.2258,
      "train_tokens_per_second": 29246.658
    },
    {
      "epoch": 0.5855725364711294,
      "grad_norm": 0.59765625,
      "learning_rate": 3.068639850032933e-05,
      "loss": 0.4070134162902832,
      "num_input_tokens_seen": 4705041280,
      "step": 16550,
      "train_runtime": 160876.327,
      "train_tokens_per_second": 29246.325
    },
    {
      "epoch": 0.5859263567348582,
      "grad_norm": 0.60546875,
      "learning_rate": 3.068062093198028e-05,
      "loss": 0.4061460494995117,
      "num_input_tokens_seen": 4707861184,
      "step": 16560,
      "train_runtime": 160973.0119,
      "train_tokens_per_second": 29246.276
    },
    {
      "epoch": 0.5862801769985869,
      "grad_norm": 0.578125,
      "learning_rate": 3.067484662576687e-05,
      "loss": 0.40291948318481446,
      "num_input_tokens_seen": 4710654208,
      "step": 16570,
      "train_runtime": 161066.8239,
      "train_tokens_per_second": 29246.583
    },
    {
      "epoch": 0.5866339972623157,
      "grad_norm": 0.609375,
      "learning_rate": 3.066907557862048e-05,
      "loss": 0.4098064422607422,
      "num_input_tokens_seen": 4713473728,
      "step": 16580,
      "train_runtime": 161163.9221,
      "train_tokens_per_second": 29246.457
    },
    {
      "epoch": 0.5869878175260445,
      "grad_norm": 0.57421875,
      "learning_rate": 3.066330778747651e-05,
      "loss": 0.40730891227722166,
      "num_input_tokens_seen": 4716275520,
      "step": 16590,
      "train_runtime": 161260.1269,
      "train_tokens_per_second": 29246.384
    },
    {
      "epoch": 0.5873416377897732,
      "grad_norm": 0.5859375,
      "learning_rate": 3.065754324927441e-05,
      "loss": 0.4051215171813965,
      "num_input_tokens_seen": 4719098112,
      "step": 16600,
      "train_runtime": 161357.3136,
      "train_tokens_per_second": 29246.261
    },
    {
      "epoch": 0.5876954580535021,
      "grad_norm": 0.609375,
      "learning_rate": 3.065178196095765e-05,
      "loss": 0.4059257984161377,
      "num_input_tokens_seen": 4721976000,
      "step": 16610,
      "train_runtime": 161455.859,
      "train_tokens_per_second": 29246.235
    },
    {
      "epoch": 0.5880492783172309,
      "grad_norm": 0.59765625,
      "learning_rate": 3.064602391947372e-05,
      "loss": 0.4054714679718018,
      "num_input_tokens_seen": 4724850880,
      "step": 16620,
      "train_runtime": 161554.8902,
      "train_tokens_per_second": 29246.103
    },
    {
      "epoch": 0.5884030985809596,
      "grad_norm": 0.58203125,
      "learning_rate": 3.0640269121774126e-05,
      "loss": 0.408254337310791,
      "num_input_tokens_seen": 4727732864,
      "step": 16630,
      "train_runtime": 161654.6644,
      "train_tokens_per_second": 29245.88
    },
    {
      "epoch": 0.5887569188446884,
      "grad_norm": 0.56640625,
      "learning_rate": 3.063451756481437e-05,
      "loss": 0.40446019172668457,
      "num_input_tokens_seen": 4730596480,
      "step": 16640,
      "train_runtime": 161753.4678,
      "train_tokens_per_second": 29245.719
    },
    {
      "epoch": 0.5891107391084172,
      "grad_norm": 0.6015625,
      "learning_rate": 3.062876924555396e-05,
      "loss": 0.4072012424468994,
      "num_input_tokens_seen": 4733436416,
      "step": 16650,
      "train_runtime": 161849.1676,
      "train_tokens_per_second": 29245.973
    },
    {
      "epoch": 0.5894645593721459,
      "grad_norm": 0.55859375,
      "learning_rate": 3.06230241609564e-05,
      "loss": 0.4029825210571289,
      "num_input_tokens_seen": 4736315712,
      "step": 16660,
      "train_runtime": 161947.6385,
      "train_tokens_per_second": 29245.97
    },
    {
      "epoch": 0.5898183796358747,
      "grad_norm": 0.61328125,
      "learning_rate": 3.061728230798916e-05,
      "loss": 0.40367813110351564,
      "num_input_tokens_seen": 4739207104,
      "step": 16670,
      "train_runtime": 162046.978,
      "train_tokens_per_second": 29245.884
    },
    {
      "epoch": 0.5901721998996035,
      "grad_norm": 0.60546875,
      "learning_rate": 3.061154368362373e-05,
      "loss": 0.4094395160675049,
      "num_input_tokens_seen": 4742024256,
      "step": 16680,
      "train_runtime": 162143.9247,
      "train_tokens_per_second": 29245.772
    },
    {
      "epoch": 0.5905260201633323,
      "grad_norm": 0.60546875,
      "learning_rate": 3.060580828483552e-05,
      "loss": 0.40355796813964845,
      "num_input_tokens_seen": 4744898496,
      "step": 16690,
      "train_runtime": 162243.197,
      "train_tokens_per_second": 29245.593
    },
    {
      "epoch": 0.5908798404270611,
      "grad_norm": 0.57421875,
      "learning_rate": 3.060007610860395e-05,
      "loss": 0.40567693710327146,
      "num_input_tokens_seen": 4747769280,
      "step": 16700,
      "train_runtime": 162340.5111,
      "train_tokens_per_second": 29245.746
    },
    {
      "epoch": 0.5912336606907899,
      "grad_norm": 0.59765625,
      "learning_rate": 3.059434715191236e-05,
      "loss": 0.4064697265625,
      "num_input_tokens_seen": 4750639936,
      "step": 16710,
      "train_runtime": 162439.2238,
      "train_tokens_per_second": 29245.645
    },
    {
      "epoch": 0.5915874809545186,
      "grad_norm": 0.60546875,
      "learning_rate": 3.0588621411748074e-05,
      "loss": 0.40702223777770996,
      "num_input_tokens_seen": 4753437632,
      "step": 16720,
      "train_runtime": 162531.1817,
      "train_tokens_per_second": 29246.312
    },
    {
      "epoch": 0.5919413012182474,
      "grad_norm": 0.60546875,
      "learning_rate": 3.0582898885102344e-05,
      "loss": 0.40258302688598635,
      "num_input_tokens_seen": 4756237120,
      "step": 16730,
      "train_runtime": 162622.4518,
      "train_tokens_per_second": 29247.112
    },
    {
      "epoch": 0.5922951214819762,
      "grad_norm": 0.5625,
      "learning_rate": 3.0577179568970366e-05,
      "loss": 0.4069647789001465,
      "num_input_tokens_seen": 4759045184,
      "step": 16740,
      "train_runtime": 162717.7168,
      "train_tokens_per_second": 29247.247
    },
    {
      "epoch": 0.5926489417457049,
      "grad_norm": 0.58984375,
      "learning_rate": 3.057146346035127e-05,
      "loss": 0.4019571304321289,
      "num_input_tokens_seen": 4761911616,
      "step": 16750,
      "train_runtime": 162815.3619,
      "train_tokens_per_second": 29247.312
    },
    {
      "epoch": 0.5930027620094337,
      "grad_norm": 0.59765625,
      "learning_rate": 3.056575055624811e-05,
      "loss": 0.40976428985595703,
      "num_input_tokens_seen": 4764743872,
      "step": 16760,
      "train_runtime": 162910.0742,
      "train_tokens_per_second": 29247.693
    },
    {
      "epoch": 0.5933565822731625,
      "grad_norm": 0.59375,
      "learning_rate": 3.0560040853667846e-05,
      "loss": 0.4099114418029785,
      "num_input_tokens_seen": 4767652416,
      "step": 16770,
      "train_runtime": 163014.603,
      "train_tokens_per_second": 29246.781
    },
    {
      "epoch": 0.5937104025368913,
      "grad_norm": 0.546875,
      "learning_rate": 3.0554334349621366e-05,
      "loss": 0.4050790786743164,
      "num_input_tokens_seen": 4770495296,
      "step": 16780,
      "train_runtime": 163112.1303,
      "train_tokens_per_second": 29246.723
    },
    {
      "epoch": 0.5940642228006201,
      "grad_norm": 0.5703125,
      "learning_rate": 3.054863104112347e-05,
      "loss": 0.40823917388916015,
      "num_input_tokens_seen": 4773229056,
      "step": 16790,
      "train_runtime": 163205.5212,
      "train_tokens_per_second": 29246.738
    },
    {
      "epoch": 0.5944180430643489,
      "grad_norm": 0.59375,
      "learning_rate": 3.0542930925192835e-05,
      "loss": 0.4047149658203125,
      "num_input_tokens_seen": 4776024960,
      "step": 16800,
      "train_runtime": 163298.5874,
      "train_tokens_per_second": 29247.191
    },
    {
      "epoch": 0.5947718633280776,
      "grad_norm": 0.5859375,
      "learning_rate": 3.053723399885205e-05,
      "loss": 0.4045846939086914,
      "num_input_tokens_seen": 4778855936,
      "step": 16810,
      "train_runtime": 163396.5845,
      "train_tokens_per_second": 29246.976
    },
    {
      "epoch": 0.5951256835918064,
      "grad_norm": 0.5625,
      "learning_rate": 3.0531540259127565e-05,
      "loss": 0.40632429122924807,
      "num_input_tokens_seen": 4781725888,
      "step": 16820,
      "train_runtime": 163493.8354,
      "train_tokens_per_second": 29247.133
    },
    {
      "epoch": 0.5954795038555352,
      "grad_norm": 0.58203125,
      "learning_rate": 3.052584970304976e-05,
      "loss": 0.408148193359375,
      "num_input_tokens_seen": 4784632384,
      "step": 16830,
      "train_runtime": 163595.339,
      "train_tokens_per_second": 29246.752
    },
    {
      "epoch": 0.5958333241192639,
      "grad_norm": 0.5703125,
      "learning_rate": 3.052016232765282e-05,
      "loss": 0.40592541694641116,
      "num_input_tokens_seen": 4787526400,
      "step": 16840,
      "train_runtime": 163696.1586,
      "train_tokens_per_second": 29246.419
    },
    {
      "epoch": 0.5961871443829927,
      "grad_norm": 0.59375,
      "learning_rate": 3.0514478129974845e-05,
      "loss": 0.40691165924072265,
      "num_input_tokens_seen": 4790408000,
      "step": 16850,
      "train_runtime": 163796.9337,
      "train_tokens_per_second": 29246.018
    },
    {
      "epoch": 0.5965409646467216,
      "grad_norm": 0.56640625,
      "learning_rate": 3.0508797107057785e-05,
      "loss": 0.4074373245239258,
      "num_input_tokens_seen": 4793291904,
      "step": 16860,
      "train_runtime": 163895.0921,
      "train_tokens_per_second": 29246.098
    },
    {
      "epoch": 0.5968947849104503,
      "grad_norm": 0.5859375,
      "learning_rate": 3.0503119255947432e-05,
      "loss": 0.4037750244140625,
      "num_input_tokens_seen": 4796120448,
      "step": 16870,
      "train_runtime": 163992.541,
      "train_tokens_per_second": 29245.967
    },
    {
      "epoch": 0.5972486051741791,
      "grad_norm": 0.5625,
      "learning_rate": 3.049744457369344e-05,
      "loss": 0.4029093742370605,
      "num_input_tokens_seen": 4798935680,
      "step": 16880,
      "train_runtime": 164090.1179,
      "train_tokens_per_second": 29245.732
    },
    {
      "epoch": 0.5976024254379079,
      "grad_norm": 0.58984375,
      "learning_rate": 3.049177305734929e-05,
      "loss": 0.40402708053588865,
      "num_input_tokens_seen": 4801809088,
      "step": 16890,
      "train_runtime": 164191.0366,
      "train_tokens_per_second": 29245.257
    },
    {
      "epoch": 0.5979562457016366,
      "grad_norm": 0.6015625,
      "learning_rate": 3.04861047039723e-05,
      "loss": 0.4065730571746826,
      "num_input_tokens_seen": 4804642752,
      "step": 16900,
      "train_runtime": 164287.1698,
      "train_tokens_per_second": 29245.392
    },
    {
      "epoch": 0.5983100659653654,
      "grad_norm": 0.59765625,
      "learning_rate": 3.0480439510623638e-05,
      "loss": 0.40772428512573244,
      "num_input_tokens_seen": 4807497472,
      "step": 16910,
      "train_runtime": 164383.2461,
      "train_tokens_per_second": 29245.666
    },
    {
      "epoch": 0.5986638862290942,
      "grad_norm": 0.578125,
      "learning_rate": 3.0474777474368255e-05,
      "loss": 0.4077460289001465,
      "num_input_tokens_seen": 4810281792,
      "step": 16920,
      "train_runtime": 164474.8734,
      "train_tokens_per_second": 29246.302
    },
    {
      "epoch": 0.5990177064928229,
      "grad_norm": 0.5859375,
      "learning_rate": 3.046911859227495e-05,
      "loss": 0.4049734592437744,
      "num_input_tokens_seen": 4813188032,
      "step": 16930,
      "train_runtime": 164572.3697,
      "train_tokens_per_second": 29246.635
    },
    {
      "epoch": 0.5993715267565518,
      "grad_norm": 0.62109375,
      "learning_rate": 3.0463462861416302e-05,
      "loss": 0.40674467086791993,
      "num_input_tokens_seen": 4816004416,
      "step": 16940,
      "train_runtime": 164668.1867,
      "train_tokens_per_second": 29246.72
    },
    {
      "epoch": 0.5997253470202806,
      "grad_norm": 0.58984375,
      "learning_rate": 3.045781027886873e-05,
      "loss": 0.40243988037109374,
      "num_input_tokens_seen": 4818798528,
      "step": 16950,
      "train_runtime": 164760.3024,
      "train_tokens_per_second": 29247.328
    },
    {
      "epoch": 0.6000791672840093,
      "grad_norm": 0.58203125,
      "learning_rate": 3.0452160841712408e-05,
      "loss": 0.40737123489379884,
      "num_input_tokens_seen": 4821652480,
      "step": 16960,
      "train_runtime": 164861.2112,
      "train_tokens_per_second": 29246.737
    },
    {
      "epoch": 0.6004329875477381,
      "grad_norm": 0.6015625,
      "learning_rate": 3.044651454703133e-05,
      "loss": 0.40608940124511717,
      "num_input_tokens_seen": 4824476672,
      "step": 16970,
      "train_runtime": 164959.2383,
      "train_tokens_per_second": 29246.478
    },
    {
      "epoch": 0.6007868078114669,
      "grad_norm": 0.59375,
      "learning_rate": 3.0440871391913257e-05,
      "loss": 0.40442862510681155,
      "num_input_tokens_seen": 4827343616,
      "step": 16980,
      "train_runtime": 165055.4929,
      "train_tokens_per_second": 29246.792
    },
    {
      "epoch": 0.6011406280751956,
      "grad_norm": 0.5859375,
      "learning_rate": 3.043523137344973e-05,
      "loss": 0.4066753387451172,
      "num_input_tokens_seen": 4830190848,
      "step": 16990,
      "train_runtime": 165152.3559,
      "train_tokens_per_second": 29246.878
    },
    {
      "epoch": 0.6014944483389244,
      "grad_norm": 0.58203125,
      "learning_rate": 3.0429594488736074e-05,
      "loss": 0.40218772888183596,
      "num_input_tokens_seen": 4833028416,
      "step": 17000,
      "train_runtime": 165249.3869,
      "train_tokens_per_second": 29246.877
    },
    {
      "epoch": 0.6018482686026532,
      "grad_norm": 0.5859375,
      "learning_rate": 3.0423960734871348e-05,
      "loss": 0.4041303634643555,
      "num_input_tokens_seen": 4835902656,
      "step": 17010,
      "train_runtime": 165347.6556,
      "train_tokens_per_second": 29246.878
    },
    {
      "epoch": 0.6022020888663819,
      "grad_norm": 0.5859375,
      "learning_rate": 3.0418330108958404e-05,
      "loss": 0.4078521728515625,
      "num_input_tokens_seen": 4838743168,
      "step": 17020,
      "train_runtime": 165444.1471,
      "train_tokens_per_second": 29246.989
    },
    {
      "epoch": 0.6025559091301108,
      "grad_norm": 0.609375,
      "learning_rate": 3.0412702608103828e-05,
      "loss": 0.4075908660888672,
      "num_input_tokens_seen": 4841615808,
      "step": 17030,
      "train_runtime": 165542.9198,
      "train_tokens_per_second": 29246.891
    },
    {
      "epoch": 0.6029097293938396,
      "grad_norm": 0.5859375,
      "learning_rate": 3.0407078229417944e-05,
      "loss": 0.4073158264160156,
      "num_input_tokens_seen": 4844434816,
      "step": 17040,
      "train_runtime": 165638.625,
      "train_tokens_per_second": 29247.012
    },
    {
      "epoch": 0.6032635496575683,
      "grad_norm": 0.59375,
      "learning_rate": 3.0401456970014836e-05,
      "loss": 0.40678677558898924,
      "num_input_tokens_seen": 4847282624,
      "step": 17050,
      "train_runtime": 165736.102,
      "train_tokens_per_second": 29246.993
    },
    {
      "epoch": 0.6036173699212971,
      "grad_norm": 0.609375,
      "learning_rate": 3.039583882701231e-05,
      "loss": 0.4060935974121094,
      "num_input_tokens_seen": 4850112192,
      "step": 17060,
      "train_runtime": 165833.882,
      "train_tokens_per_second": 29246.811
    },
    {
      "epoch": 0.6039711901850259,
      "grad_norm": 0.578125,
      "learning_rate": 3.0390223797531887e-05,
      "loss": 0.4084298133850098,
      "num_input_tokens_seen": 4852995648,
      "step": 17070,
      "train_runtime": 165933.5539,
      "train_tokens_per_second": 29246.62
    },
    {
      "epoch": 0.6043250104487546,
      "grad_norm": 0.60546875,
      "learning_rate": 3.0384611878698837e-05,
      "loss": 0.4082502365112305,
      "num_input_tokens_seen": 4855845952,
      "step": 17080,
      "train_runtime": 166030.9656,
      "train_tokens_per_second": 29246.628
    },
    {
      "epoch": 0.6046788307124834,
      "grad_norm": 0.5859375,
      "learning_rate": 3.0379003067642113e-05,
      "loss": 0.4072283744812012,
      "num_input_tokens_seen": 4858639744,
      "step": 17090,
      "train_runtime": 166123.6616,
      "train_tokens_per_second": 29247.126
    },
    {
      "epoch": 0.6050326509762122,
      "grad_norm": 0.57421875,
      "learning_rate": 3.037339736149441e-05,
      "loss": 0.4023797035217285,
      "num_input_tokens_seen": 4861526848,
      "step": 17100,
      "train_runtime": 166223.25,
      "train_tokens_per_second": 29246.973
    },
    {
      "epoch": 0.605386471239941,
      "grad_norm": 0.59765625,
      "learning_rate": 3.0367794757392092e-05,
      "loss": 0.4041740894317627,
      "num_input_tokens_seen": 4864316736,
      "step": 17110,
      "train_runtime": 166317.2356,
      "train_tokens_per_second": 29247.22
    },
    {
      "epoch": 0.6057402915036698,
      "grad_norm": 0.59765625,
      "learning_rate": 3.0362195252475245e-05,
      "loss": 0.4097470283508301,
      "num_input_tokens_seen": 4867139456,
      "step": 17120,
      "train_runtime": 166413.2144,
      "train_tokens_per_second": 29247.314
    },
    {
      "epoch": 0.6060941117673986,
      "grad_norm": 0.6015625,
      "learning_rate": 3.0356598843887634e-05,
      "loss": 0.4035935401916504,
      "num_input_tokens_seen": 4869951104,
      "step": 17130,
      "train_runtime": 166509.1161,
      "train_tokens_per_second": 29247.354
    },
    {
      "epoch": 0.6064479320311273,
      "grad_norm": 0.5859375,
      "learning_rate": 3.03510055287767e-05,
      "loss": 0.40491538047790526,
      "num_input_tokens_seen": 4872733760,
      "step": 17140,
      "train_runtime": 166602.3932,
      "train_tokens_per_second": 29247.682
    },
    {
      "epoch": 0.6068017522948561,
      "grad_norm": 0.57421875,
      "learning_rate": 3.0345415304293578e-05,
      "loss": 0.40308256149291993,
      "num_input_tokens_seen": 4875590400,
      "step": 17150,
      "train_runtime": 166700.8676,
      "train_tokens_per_second": 29247.541
    },
    {
      "epoch": 0.6071555725585849,
      "grad_norm": 0.57421875,
      "learning_rate": 3.0339828167593065e-05,
      "loss": 0.40570802688598634,
      "num_input_tokens_seen": 4878444224,
      "step": 17160,
      "train_runtime": 166800.6669,
      "train_tokens_per_second": 29247.151
    },
    {
      "epoch": 0.6075093928223136,
      "grad_norm": 0.56640625,
      "learning_rate": 3.033424411583363e-05,
      "loss": 0.4033414363861084,
      "num_input_tokens_seen": 4881299392,
      "step": 17170,
      "train_runtime": 166896.0823,
      "train_tokens_per_second": 29247.537
    },
    {
      "epoch": 0.6078632130860424,
      "grad_norm": 0.578125,
      "learning_rate": 3.0328663146177387e-05,
      "loss": 0.4091768264770508,
      "num_input_tokens_seen": 4884182528,
      "step": 17180,
      "train_runtime": 166996.0657,
      "train_tokens_per_second": 29247.291
    },
    {
      "epoch": 0.6082170333497713,
      "grad_norm": 0.5859375,
      "learning_rate": 3.032308525579013e-05,
      "loss": 0.40320463180541993,
      "num_input_tokens_seen": 4887034048,
      "step": 17190,
      "train_runtime": 167096.0523,
      "train_tokens_per_second": 29246.855
    },
    {
      "epoch": 0.6085708536135,
      "grad_norm": 0.56640625,
      "learning_rate": 3.031751044184127e-05,
      "loss": 0.4018968105316162,
      "num_input_tokens_seen": 4889851008,
      "step": 17200,
      "train_runtime": 167191.9173,
      "train_tokens_per_second": 29246.934
    },
    {
      "epoch": 0.6089246738772288,
      "grad_norm": 0.58984375,
      "learning_rate": 3.0311938701503872e-05,
      "loss": 0.4022430419921875,
      "num_input_tokens_seen": 4892737344,
      "step": 17210,
      "train_runtime": 167289.7295,
      "train_tokens_per_second": 29247.087
    },
    {
      "epoch": 0.6092784941409576,
      "grad_norm": 0.58203125,
      "learning_rate": 3.0306370031954646e-05,
      "loss": 0.4053038597106934,
      "num_input_tokens_seen": 4895623744,
      "step": 17220,
      "train_runtime": 167390.003,
      "train_tokens_per_second": 29246.811
    },
    {
      "epoch": 0.6096323144046863,
      "grad_norm": 0.60546875,
      "learning_rate": 3.030080443037393e-05,
      "loss": 0.4061570644378662,
      "num_input_tokens_seen": 4898499072,
      "step": 17230,
      "train_runtime": 167486.9705,
      "train_tokens_per_second": 29247.046
    },
    {
      "epoch": 0.6099861346684151,
      "grad_norm": 0.5546875,
      "learning_rate": 3.0295241893945663e-05,
      "loss": 0.4084829807281494,
      "num_input_tokens_seen": 4901417472,
      "step": 17240,
      "train_runtime": 167590.4965,
      "train_tokens_per_second": 29246.393
    },
    {
      "epoch": 0.6103399549321439,
      "grad_norm": 0.55078125,
      "learning_rate": 3.028968241985743e-05,
      "loss": 0.40891332626342775,
      "num_input_tokens_seen": 4904288448,
      "step": 17250,
      "train_runtime": 167689.8449,
      "train_tokens_per_second": 29246.186
    },
    {
      "epoch": 0.6106937751958726,
      "grad_norm": 0.5859375,
      "learning_rate": 3.0284126005300405e-05,
      "loss": 0.4050548553466797,
      "num_input_tokens_seen": 4907094080,
      "step": 17260,
      "train_runtime": 167786.8658,
      "train_tokens_per_second": 29245.996
    },
    {
      "epoch": 0.6110475954596014,
      "grad_norm": 0.55078125,
      "learning_rate": 3.0278572647469385e-05,
      "loss": 0.40810537338256836,
      "num_input_tokens_seen": 4909941952,
      "step": 17270,
      "train_runtime": 167886.6812,
      "train_tokens_per_second": 29245.572
    },
    {
      "epoch": 0.6114014157233303,
      "grad_norm": 0.6015625,
      "learning_rate": 3.027302234356275e-05,
      "loss": 0.40270819664001467,
      "num_input_tokens_seen": 4912768192,
      "step": 17280,
      "train_runtime": 167981.681,
      "train_tokens_per_second": 29245.857
    },
    {
      "epoch": 0.611755235987059,
      "grad_norm": 0.59765625,
      "learning_rate": 3.0267475090782494e-05,
      "loss": 0.4071396827697754,
      "num_input_tokens_seen": 4915623744,
      "step": 17290,
      "train_runtime": 168077.8724,
      "train_tokens_per_second": 29246.109
    },
    {
      "epoch": 0.6121090562507878,
      "grad_norm": 0.56640625,
      "learning_rate": 3.026193088633418e-05,
      "loss": 0.4104874610900879,
      "num_input_tokens_seen": 4918438208,
      "step": 17300,
      "train_runtime": 168172.8242,
      "train_tokens_per_second": 29246.332
    },
    {
      "epoch": 0.6124628765145166,
      "grad_norm": 0.59765625,
      "learning_rate": 3.0256389727426952e-05,
      "loss": 0.4036008358001709,
      "num_input_tokens_seen": 4921316352,
      "step": 17310,
      "train_runtime": 168271.8383,
      "train_tokens_per_second": 29246.227
    },
    {
      "epoch": 0.6128166967782454,
      "grad_norm": 0.58984375,
      "learning_rate": 3.0250851611273555e-05,
      "loss": 0.41010246276855467,
      "num_input_tokens_seen": 4924200896,
      "step": 17320,
      "train_runtime": 168372.4827,
      "train_tokens_per_second": 29245.877
    },
    {
      "epoch": 0.6131705170419741,
      "grad_norm": 0.63671875,
      "learning_rate": 3.0245316535090273e-05,
      "loss": 0.401582145690918,
      "num_input_tokens_seen": 4927030016,
      "step": 17330,
      "train_runtime": 168466.6279,
      "train_tokens_per_second": 29246.327
    },
    {
      "epoch": 0.6135243373057029,
      "grad_norm": 0.6015625,
      "learning_rate": 3.023978449609697e-05,
      "loss": 0.40504703521728513,
      "num_input_tokens_seen": 4929834496,
      "step": 17340,
      "train_runtime": 168561.8576,
      "train_tokens_per_second": 29246.441
    },
    {
      "epoch": 0.6138781575694316,
      "grad_norm": 0.5703125,
      "learning_rate": 3.023425549151707e-05,
      "loss": 0.40938544273376465,
      "num_input_tokens_seen": 4932647296,
      "step": 17350,
      "train_runtime": 168657.2375,
      "train_tokens_per_second": 29246.579
    },
    {
      "epoch": 0.6142319778331605,
      "grad_norm": 0.578125,
      "learning_rate": 3.022872951857754e-05,
      "loss": 0.4029240608215332,
      "num_input_tokens_seen": 4935509440,
      "step": 17360,
      "train_runtime": 168757.575,
      "train_tokens_per_second": 29246.151
    },
    {
      "epoch": 0.6145857980968893,
      "grad_norm": 0.59765625,
      "learning_rate": 3.0223206574508904e-05,
      "loss": 0.40139102935791016,
      "num_input_tokens_seen": 4938350144,
      "step": 17370,
      "train_runtime": 168854.0035,
      "train_tokens_per_second": 29246.272
    },
    {
      "epoch": 0.6149396183606181,
      "grad_norm": 0.59765625,
      "learning_rate": 3.0217686656545224e-05,
      "loss": 0.40678749084472654,
      "num_input_tokens_seen": 4941242176,
      "step": 17380,
      "train_runtime": 168954.661,
      "train_tokens_per_second": 29245.965
    },
    {
      "epoch": 0.6152934386243468,
      "grad_norm": 0.58203125,
      "learning_rate": 3.0212169761924077e-05,
      "loss": 0.40032229423522947,
      "num_input_tokens_seen": 4944175168,
      "step": 17390,
      "train_runtime": 169057.2969,
      "train_tokens_per_second": 29245.559
    },
    {
      "epoch": 0.6156472588880756,
      "grad_norm": 0.58203125,
      "learning_rate": 3.0206655887886604e-05,
      "loss": 0.4111682415008545,
      "num_input_tokens_seen": 4947047616,
      "step": 17400,
      "train_runtime": 169156.5873,
      "train_tokens_per_second": 29245.374
    },
    {
      "epoch": 0.6160010791518044,
      "grad_norm": 0.56640625,
      "learning_rate": 3.0201145031677448e-05,
      "loss": 0.40668344497680664,
      "num_input_tokens_seen": 4949891200,
      "step": 17410,
      "train_runtime": 169254.5683,
      "train_tokens_per_second": 29245.244
    },
    {
      "epoch": 0.6163548994155331,
      "grad_norm": 0.5859375,
      "learning_rate": 3.019563719054477e-05,
      "loss": 0.4065103054046631,
      "num_input_tokens_seen": 4952743552,
      "step": 17420,
      "train_runtime": 169353.1887,
      "train_tokens_per_second": 29245.056
    },
    {
      "epoch": 0.6167087196792619,
      "grad_norm": 0.5859375,
      "learning_rate": 3.019013236174025e-05,
      "loss": 0.40653672218322756,
      "num_input_tokens_seen": 4955594368,
      "step": 17430,
      "train_runtime": 169452.4934,
      "train_tokens_per_second": 29244.742
    },
    {
      "epoch": 0.6170625399429908,
      "grad_norm": 0.6015625,
      "learning_rate": 3.018463054251906e-05,
      "loss": 0.4084512710571289,
      "num_input_tokens_seen": 4958432448,
      "step": 17440,
      "train_runtime": 169546.824,
      "train_tokens_per_second": 29245.21
    },
    {
      "epoch": 0.6174163602067195,
      "grad_norm": 0.55859375,
      "learning_rate": 3.017913173013989e-05,
      "loss": 0.4048893451690674,
      "num_input_tokens_seen": 4961316288,
      "step": 17450,
      "train_runtime": 169644.6597,
      "train_tokens_per_second": 29245.343
    },
    {
      "epoch": 0.6177701804704483,
      "grad_norm": 0.58984375,
      "learning_rate": 3.017363592186491e-05,
      "loss": 0.4105884552001953,
      "num_input_tokens_seen": 4964150016,
      "step": 17460,
      "train_runtime": 169741.8624,
      "train_tokens_per_second": 29245.29
    },
    {
      "epoch": 0.6181240007341771,
      "grad_norm": 0.57421875,
      "learning_rate": 3.0168143114959797e-05,
      "loss": 0.40169200897216795,
      "num_input_tokens_seen": 4966970816,
      "step": 17470,
      "train_runtime": 169838.6365,
      "train_tokens_per_second": 29245.235
    },
    {
      "epoch": 0.6184778209979058,
      "grad_norm": 0.5625,
      "learning_rate": 3.016265330669369e-05,
      "loss": 0.4048782825469971,
      "num_input_tokens_seen": 4969811584,
      "step": 17480,
      "train_runtime": 169935.6616,
      "train_tokens_per_second": 29245.254
    },
    {
      "epoch": 0.6188316412616346,
      "grad_norm": 0.59765625,
      "learning_rate": 3.0157166494339223e-05,
      "loss": 0.40407352447509765,
      "num_input_tokens_seen": 4972673536,
      "step": 17490,
      "train_runtime": 170035.0961,
      "train_tokens_per_second": 29244.983
    },
    {
      "epoch": 0.6191854615253634,
      "grad_norm": 0.578125,
      "learning_rate": 3.0151682675172482e-05,
      "loss": 0.40251784324645995,
      "num_input_tokens_seen": 4975483200,
      "step": 17500,
      "train_runtime": 170131.8076,
      "train_tokens_per_second": 29244.874
    },
    {
      "epoch": 0.6195392817890921,
      "grad_norm": 0.640625,
      "learning_rate": 3.0146201846473034e-05,
      "loss": 0.4039877414703369,
      "num_input_tokens_seen": 4978340736,
      "step": 17510,
      "train_runtime": 170230.4562,
      "train_tokens_per_second": 29244.712
    },
    {
      "epoch": 0.6198931020528209,
      "grad_norm": 0.59765625,
      "learning_rate": 3.01407240055239e-05,
      "loss": 0.4047060966491699,
      "num_input_tokens_seen": 4981153600,
      "step": 17520,
      "train_runtime": 170325.915,
      "train_tokens_per_second": 29244.837
    },
    {
      "epoch": 0.6202469223165498,
      "grad_norm": 0.59375,
      "learning_rate": 3.013524914961157e-05,
      "loss": 0.4049750804901123,
      "num_input_tokens_seen": 4983947264,
      "step": 17530,
      "train_runtime": 170418.4329,
      "train_tokens_per_second": 29245.353
    },
    {
      "epoch": 0.6206007425802785,
      "grad_norm": 0.58203125,
      "learning_rate": 3.012977727602596e-05,
      "loss": 0.40945959091186523,
      "num_input_tokens_seen": 4986762560,
      "step": 17540,
      "train_runtime": 170512.8075,
      "train_tokens_per_second": 29245.677
    },
    {
      "epoch": 0.6209545628440073,
      "grad_norm": 0.58984375,
      "learning_rate": 3.0124308382060446e-05,
      "loss": 0.40868167877197265,
      "num_input_tokens_seen": 4989534656,
      "step": 17550,
      "train_runtime": 170608.5167,
      "train_tokens_per_second": 29245.519
    },
    {
      "epoch": 0.6213083831077361,
      "grad_norm": 0.58203125,
      "learning_rate": 3.011884246501183e-05,
      "loss": 0.4059316635131836,
      "num_input_tokens_seen": 4992401984,
      "step": 17560,
      "train_runtime": 170707.6769,
      "train_tokens_per_second": 29245.328
    },
    {
      "epoch": 0.6216622033714648,
      "grad_norm": 0.609375,
      "learning_rate": 3.0113379522180363e-05,
      "loss": 0.4088932991027832,
      "num_input_tokens_seen": 4995281408,
      "step": 17570,
      "train_runtime": 170806.7379,
      "train_tokens_per_second": 29245.225
    },
    {
      "epoch": 0.6220160236351936,
      "grad_norm": 0.5703125,
      "learning_rate": 3.0107919550869702e-05,
      "loss": 0.40520291328430175,
      "num_input_tokens_seen": 4998074432,
      "step": 17580,
      "train_runtime": 170898.1163,
      "train_tokens_per_second": 29245.931
    },
    {
      "epoch": 0.6223698438989224,
      "grad_norm": 0.61328125,
      "learning_rate": 3.0102462548386933e-05,
      "loss": 0.40500588417053224,
      "num_input_tokens_seen": 5000884480,
      "step": 17590,
      "train_runtime": 170992.7359,
      "train_tokens_per_second": 29246.181
    },
    {
      "epoch": 0.6227236641626511,
      "grad_norm": 0.6015625,
      "learning_rate": 3.0097008512042572e-05,
      "loss": 0.4085218906402588,
      "num_input_tokens_seen": 5003696768,
      "step": 17600,
      "train_runtime": 171088.4481,
      "train_tokens_per_second": 29246.257
    },
    {
      "epoch": 0.62307748442638,
      "grad_norm": 0.5859375,
      "learning_rate": 3.009155743915052e-05,
      "loss": 0.40653324127197266,
      "num_input_tokens_seen": 5006511936,
      "step": 17610,
      "train_runtime": 171184.7224,
      "train_tokens_per_second": 29246.254
    },
    {
      "epoch": 0.6234313046901088,
      "grad_norm": 0.56640625,
      "learning_rate": 3.00861093270281e-05,
      "loss": 0.40657758712768555,
      "num_input_tokens_seen": 5009304512,
      "step": 17620,
      "train_runtime": 171279.8812,
      "train_tokens_per_second": 29246.31
    },
    {
      "epoch": 0.6237851249538375,
      "grad_norm": 0.5859375,
      "learning_rate": 3.008066417299602e-05,
      "loss": 0.4066032409667969,
      "num_input_tokens_seen": 5012144896,
      "step": 17630,
      "train_runtime": 171373.7355,
      "train_tokens_per_second": 29246.867
    },
    {
      "epoch": 0.6241389452175663,
      "grad_norm": 0.60546875,
      "learning_rate": 3.00752219743784e-05,
      "loss": 0.4033350944519043,
      "num_input_tokens_seen": 5015010432,
      "step": 17640,
      "train_runtime": 171471.5681,
      "train_tokens_per_second": 29246.892
    },
    {
      "epoch": 0.6244927654812951,
      "grad_norm": 0.5703125,
      "learning_rate": 3.0069782728502737e-05,
      "loss": 0.40196542739868163,
      "num_input_tokens_seen": 5017832512,
      "step": 17650,
      "train_runtime": 171567.1023,
      "train_tokens_per_second": 29247.055
    },
    {
      "epoch": 0.6248465857450238,
      "grad_norm": 0.578125,
      "learning_rate": 3.00643464326999e-05,
      "loss": 0.404159688949585,
      "num_input_tokens_seen": 5020703808,
      "step": 17660,
      "train_runtime": 171666.1521,
      "train_tokens_per_second": 29246.906
    },
    {
      "epoch": 0.6252004060087526,
      "grad_norm": 0.6015625,
      "learning_rate": 3.0058913084304146e-05,
      "loss": 0.4051948547363281,
      "num_input_tokens_seen": 5023488640,
      "step": 17670,
      "train_runtime": 171759.9255,
      "train_tokens_per_second": 29247.152
    },
    {
      "epoch": 0.6255542262724814,
      "grad_norm": 0.609375,
      "learning_rate": 3.0053482680653112e-05,
      "loss": 0.40851216316223143,
      "num_input_tokens_seen": 5026346880,
      "step": 17680,
      "train_runtime": 171857.4776,
      "train_tokens_per_second": 29247.182
    },
    {
      "epoch": 0.6259080465362102,
      "grad_norm": 0.59375,
      "learning_rate": 3.0048055219087777e-05,
      "loss": 0.4006662368774414,
      "num_input_tokens_seen": 5029180480,
      "step": 17690,
      "train_runtime": 171954.0907,
      "train_tokens_per_second": 29247.228
    },
    {
      "epoch": 0.626261866799939,
      "grad_norm": 0.6015625,
      "learning_rate": 3.0042630696952507e-05,
      "loss": 0.404221248626709,
      "num_input_tokens_seen": 5032037952,
      "step": 17700,
      "train_runtime": 172050.627,
      "train_tokens_per_second": 29247.426
    },
    {
      "epoch": 0.6266156870636678,
      "grad_norm": 0.5859375,
      "learning_rate": 3.0037209111594994e-05,
      "loss": 0.40954113006591797,
      "num_input_tokens_seen": 5034930752,
      "step": 17710,
      "train_runtime": 172150.9555,
      "train_tokens_per_second": 29247.184
    },
    {
      "epoch": 0.6269695073273965,
      "grad_norm": 0.58984375,
      "learning_rate": 3.0031790460366305e-05,
      "loss": 0.4029120445251465,
      "num_input_tokens_seen": 5037810368,
      "step": 17720,
      "train_runtime": 172249.2074,
      "train_tokens_per_second": 29247.219
    },
    {
      "epoch": 0.6273233275911253,
      "grad_norm": 0.59375,
      "learning_rate": 3.002637474062083e-05,
      "loss": 0.4088258743286133,
      "num_input_tokens_seen": 5040589760,
      "step": 17730,
      "train_runtime": 172340.9759,
      "train_tokens_per_second": 29247.773
    },
    {
      "epoch": 0.6276771478548541,
      "grad_norm": 0.58984375,
      "learning_rate": 3.0020961949716325e-05,
      "loss": 0.40375308990478515,
      "num_input_tokens_seen": 5043435904,
      "step": 17740,
      "train_runtime": 172440.0387,
      "train_tokens_per_second": 29247.476
    },
    {
      "epoch": 0.6280309681185828,
      "grad_norm": 0.58984375,
      "learning_rate": 3.0015552085013847e-05,
      "loss": 0.4001753807067871,
      "num_input_tokens_seen": 5046189184,
      "step": 17750,
      "train_runtime": 172530.5485,
      "train_tokens_per_second": 29248.091
    },
    {
      "epoch": 0.6283847883823116,
      "grad_norm": 0.578125,
      "learning_rate": 3.0010145143877805e-05,
      "loss": 0.41054410934448243,
      "num_input_tokens_seen": 5048981696,
      "step": 17760,
      "train_runtime": 172623.9199,
      "train_tokens_per_second": 29248.448
    },
    {
      "epoch": 0.6287386086460404,
      "grad_norm": 0.59765625,
      "learning_rate": 3.0004741123675907e-05,
      "loss": 0.40427050590515134,
      "num_input_tokens_seen": 5051770688,
      "step": 17770,
      "train_runtime": 172719.1703,
      "train_tokens_per_second": 29248.465
    },
    {
      "epoch": 0.6290924289097692,
      "grad_norm": 0.58984375,
      "learning_rate": 2.9999340021779205e-05,
      "loss": 0.40260753631591795,
      "num_input_tokens_seen": 5054588800,
      "step": 17780,
      "train_runtime": 172815.2108,
      "train_tokens_per_second": 29248.518
    },
    {
      "epoch": 0.629446249173498,
      "grad_norm": 0.578125,
      "learning_rate": 2.999394183556204e-05,
      "loss": 0.40073323249816895,
      "num_input_tokens_seen": 5057500672,
      "step": 17790,
      "train_runtime": 172915.6167,
      "train_tokens_per_second": 29248.374
    },
    {
      "epoch": 0.6298000694372268,
      "grad_norm": 0.59765625,
      "learning_rate": 2.998854656240207e-05,
      "loss": 0.4047834396362305,
      "num_input_tokens_seen": 5060324160,
      "step": 17800,
      "train_runtime": 173010.135,
      "train_tokens_per_second": 29248.715
    },
    {
      "epoch": 0.6301538897009555,
      "grad_norm": 0.6015625,
      "learning_rate": 2.9983154199680257e-05,
      "loss": 0.40609073638916016,
      "num_input_tokens_seen": 5063190336,
      "step": 17810,
      "train_runtime": 173106.653,
      "train_tokens_per_second": 29248.964
    },
    {
      "epoch": 0.6305077099646843,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9977764744780845e-05,
      "loss": 0.4032859802246094,
      "num_input_tokens_seen": 5065999168,
      "step": 17820,
      "train_runtime": 173202.9847,
      "train_tokens_per_second": 29248.914
    },
    {
      "epoch": 0.6308615302284131,
      "grad_norm": 0.58984375,
      "learning_rate": 2.9972378195091384e-05,
      "loss": 0.40296058654785155,
      "num_input_tokens_seen": 5068796416,
      "step": 17830,
      "train_runtime": 173300.5102,
      "train_tokens_per_second": 29248.595
    },
    {
      "epoch": 0.6312153504921418,
      "grad_norm": 0.5625,
      "learning_rate": 2.996699454800269e-05,
      "loss": 0.4052254676818848,
      "num_input_tokens_seen": 5071733952,
      "step": 17840,
      "train_runtime": 173401.6258,
      "train_tokens_per_second": 29248.48
    },
    {
      "epoch": 0.6315691707558706,
      "grad_norm": 0.5703125,
      "learning_rate": 2.9961613800908883e-05,
      "loss": 0.4059328079223633,
      "num_input_tokens_seen": 5074574912,
      "step": 17850,
      "train_runtime": 173498.3806,
      "train_tokens_per_second": 29248.543
    },
    {
      "epoch": 0.6319229910195995,
      "grad_norm": 0.58203125,
      "learning_rate": 2.995623595120733e-05,
      "loss": 0.4021906852722168,
      "num_input_tokens_seen": 5077408448,
      "step": 17860,
      "train_runtime": 173596.1813,
      "train_tokens_per_second": 29248.388
    },
    {
      "epoch": 0.6322768112833282,
      "grad_norm": 0.578125,
      "learning_rate": 2.995086099629869e-05,
      "loss": 0.40830192565917967,
      "num_input_tokens_seen": 5080244672,
      "step": 17870,
      "train_runtime": 173692.5777,
      "train_tokens_per_second": 29248.485
    },
    {
      "epoch": 0.632630631547057,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9945488933586867e-05,
      "loss": 0.4018385887145996,
      "num_input_tokens_seen": 5083104384,
      "step": 17880,
      "train_runtime": 173792.9294,
      "train_tokens_per_second": 29248.051
    },
    {
      "epoch": 0.6329844518107858,
      "grad_norm": 0.578125,
      "learning_rate": 2.9940119760479045e-05,
      "loss": 0.4086899757385254,
      "num_input_tokens_seen": 5085932736,
      "step": 17890,
      "train_runtime": 173888.4772,
      "train_tokens_per_second": 29248.245
    },
    {
      "epoch": 0.6333382720745145,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9934753474385635e-05,
      "loss": 0.39928598403930665,
      "num_input_tokens_seen": 5088743936,
      "step": 17900,
      "train_runtime": 173983.4914,
      "train_tokens_per_second": 29248.43
    },
    {
      "epoch": 0.6336920923382433,
      "grad_norm": 0.58984375,
      "learning_rate": 2.9929390072720314e-05,
      "loss": 0.4063319206237793,
      "num_input_tokens_seen": 5091568960,
      "step": 17910,
      "train_runtime": 174080.1294,
      "train_tokens_per_second": 29248.421
    },
    {
      "epoch": 0.6340459126019721,
      "grad_norm": 0.58203125,
      "learning_rate": 2.99240295529e-05,
      "loss": 0.4067960739135742,
      "num_input_tokens_seen": 5094457088,
      "step": 17920,
      "train_runtime": 174176.846,
      "train_tokens_per_second": 29248.762
    },
    {
      "epoch": 0.6343997328657008,
      "grad_norm": 0.59375,
      "learning_rate": 2.991867191234484e-05,
      "loss": 0.4066004753112793,
      "num_input_tokens_seen": 5097324736,
      "step": 17930,
      "train_runtime": 174276.0366,
      "train_tokens_per_second": 29248.569
    },
    {
      "epoch": 0.6347535531294297,
      "grad_norm": 0.609375,
      "learning_rate": 2.991331714847822e-05,
      "loss": 0.40811758041381835,
      "num_input_tokens_seen": 5100176512,
      "step": 17940,
      "train_runtime": 174372.81,
      "train_tokens_per_second": 29248.691
    },
    {
      "epoch": 0.6351073733931585,
      "grad_norm": 0.58984375,
      "learning_rate": 2.990796525872675e-05,
      "loss": 0.4074103832244873,
      "num_input_tokens_seen": 5103029248,
      "step": 17950,
      "train_runtime": 174468.7956,
      "train_tokens_per_second": 29248.951
    },
    {
      "epoch": 0.6354611936568872,
      "grad_norm": 0.5859375,
      "learning_rate": 2.990261624052027e-05,
      "loss": 0.40820956230163574,
      "num_input_tokens_seen": 5105830400,
      "step": 17960,
      "train_runtime": 174565.1352,
      "train_tokens_per_second": 29248.855
    },
    {
      "epoch": 0.635815013920616,
      "grad_norm": 0.5703125,
      "learning_rate": 2.989727009129182e-05,
      "loss": 0.4053308963775635,
      "num_input_tokens_seen": 5108736064,
      "step": 17970,
      "train_runtime": 174661.2532,
      "train_tokens_per_second": 29249.395
    },
    {
      "epoch": 0.6361688341843448,
      "grad_norm": 0.58203125,
      "learning_rate": 2.989192680847766e-05,
      "loss": 0.40932602882385255,
      "num_input_tokens_seen": 5111624448,
      "step": 17980,
      "train_runtime": 174762.1536,
      "train_tokens_per_second": 29249.036
    },
    {
      "epoch": 0.6365226544480735,
      "grad_norm": 0.5703125,
      "learning_rate": 2.9886586389517267e-05,
      "loss": 0.4071192741394043,
      "num_input_tokens_seen": 5114494144,
      "step": 17990,
      "train_runtime": 174859.4597,
      "train_tokens_per_second": 29249.17
    },
    {
      "epoch": 0.6368764747118023,
      "grad_norm": 0.5703125,
      "learning_rate": 2.98812488318533e-05,
      "loss": 0.40278215408325196,
      "num_input_tokens_seen": 5117328640,
      "step": 18000,
      "train_runtime": 174955.7788,
      "train_tokens_per_second": 29249.269
    },
    {
      "epoch": 0.6372302949755311,
      "grad_norm": 0.58203125,
      "learning_rate": 2.987591413293162e-05,
      "loss": 0.4063523292541504,
      "num_input_tokens_seen": 5120165696,
      "step": 18010,
      "train_runtime": 175050.8834,
      "train_tokens_per_second": 29249.585
    },
    {
      "epoch": 0.6375841152392598,
      "grad_norm": 0.5703125,
      "learning_rate": 2.98705822902013e-05,
      "loss": 0.4070803165435791,
      "num_input_tokens_seen": 5123009728,
      "step": 18020,
      "train_runtime": 175149.7041,
      "train_tokens_per_second": 29249.32
    },
    {
      "epoch": 0.6379379355029887,
      "grad_norm": 0.59765625,
      "learning_rate": 2.9865253301114553e-05,
      "loss": 0.39750356674194337,
      "num_input_tokens_seen": 5125901824,
      "step": 18030,
      "train_runtime": 175250.1147,
      "train_tokens_per_second": 29249.064
    },
    {
      "epoch": 0.6382917557667175,
      "grad_norm": 0.578125,
      "learning_rate": 2.985992716312683e-05,
      "loss": 0.4061281681060791,
      "num_input_tokens_seen": 5128786752,
      "step": 18040,
      "train_runtime": 175348.1224,
      "train_tokens_per_second": 29249.168
    },
    {
      "epoch": 0.6386455760304462,
      "grad_norm": 0.578125,
      "learning_rate": 2.9854603873696703e-05,
      "loss": 0.4045412540435791,
      "num_input_tokens_seen": 5131620736,
      "step": 18050,
      "train_runtime": 175444.1673,
      "train_tokens_per_second": 29249.309
    },
    {
      "epoch": 0.638999396294175,
      "grad_norm": 0.57421875,
      "learning_rate": 2.9849283430285947e-05,
      "loss": 0.4063894271850586,
      "num_input_tokens_seen": 5134488384,
      "step": 18060,
      "train_runtime": 175544.0789,
      "train_tokens_per_second": 29248.998
    },
    {
      "epoch": 0.6393532165579038,
      "grad_norm": 0.58984375,
      "learning_rate": 2.98439658303595e-05,
      "loss": 0.4066436767578125,
      "num_input_tokens_seen": 5137327488,
      "step": 18070,
      "train_runtime": 175641.4302,
      "train_tokens_per_second": 29248.95
    },
    {
      "epoch": 0.6397070368216325,
      "grad_norm": 0.59765625,
      "learning_rate": 2.9838651071385442e-05,
      "loss": 0.40442280769348143,
      "num_input_tokens_seen": 5140219712,
      "step": 18080,
      "train_runtime": 175744.0868,
      "train_tokens_per_second": 29248.322
    },
    {
      "epoch": 0.6400608570853613,
      "grad_norm": 0.58203125,
      "learning_rate": 2.9833339150835037e-05,
      "loss": 0.4047560214996338,
      "num_input_tokens_seen": 5143062656,
      "step": 18090,
      "train_runtime": 175841.3442,
      "train_tokens_per_second": 29248.313
    },
    {
      "epoch": 0.6404146773490901,
      "grad_norm": 0.58203125,
      "learning_rate": 2.982803006618267e-05,
      "loss": 0.4041615009307861,
      "num_input_tokens_seen": 5145871104,
      "step": 18100,
      "train_runtime": 175937.1483,
      "train_tokens_per_second": 29248.349
    },
    {
      "epoch": 0.640768497612819,
      "grad_norm": 0.578125,
      "learning_rate": 2.982272381490588e-05,
      "loss": 0.4070761680603027,
      "num_input_tokens_seen": 5148714880,
      "step": 18110,
      "train_runtime": 176032.3977,
      "train_tokens_per_second": 29248.678
    },
    {
      "epoch": 0.6411223178765477,
      "grad_norm": 0.5703125,
      "learning_rate": 2.9817420394485363e-05,
      "loss": 0.4018665313720703,
      "num_input_tokens_seen": 5151576832,
      "step": 18120,
      "train_runtime": 176128.3614,
      "train_tokens_per_second": 29248.991
    },
    {
      "epoch": 0.6414761381402765,
      "grad_norm": 0.59375,
      "learning_rate": 2.9812119802404927e-05,
      "loss": 0.4062140464782715,
      "num_input_tokens_seen": 5154438400,
      "step": 18130,
      "train_runtime": 176226.6315,
      "train_tokens_per_second": 29248.919
    },
    {
      "epoch": 0.6418299584040053,
      "grad_norm": 0.58203125,
      "learning_rate": 2.9806822036151523e-05,
      "loss": 0.40354671478271487,
      "num_input_tokens_seen": 5157314816,
      "step": 18140,
      "train_runtime": 176324.9595,
      "train_tokens_per_second": 29248.921
    },
    {
      "epoch": 0.642183778667734,
      "grad_norm": 0.5859375,
      "learning_rate": 2.980152709321522e-05,
      "loss": 0.40874857902526857,
      "num_input_tokens_seen": 5160157824,
      "step": 18150,
      "train_runtime": 176420.2056,
      "train_tokens_per_second": 29249.245
    },
    {
      "epoch": 0.6425375989314628,
      "grad_norm": 0.5625,
      "learning_rate": 2.9796234971089214e-05,
      "loss": 0.4053154945373535,
      "num_input_tokens_seen": 5163078912,
      "step": 18160,
      "train_runtime": 176520.4176,
      "train_tokens_per_second": 29249.188
    },
    {
      "epoch": 0.6428914191951915,
      "grad_norm": 0.5625,
      "learning_rate": 2.979094566726981e-05,
      "loss": 0.40380334854125977,
      "num_input_tokens_seen": 5165915968,
      "step": 18170,
      "train_runtime": 176617.6219,
      "train_tokens_per_second": 29249.154
    },
    {
      "epoch": 0.6432452394589203,
      "grad_norm": 0.58984375,
      "learning_rate": 2.9785659179256427e-05,
      "loss": 0.40433244705200194,
      "num_input_tokens_seen": 5168756928,
      "step": 18180,
      "train_runtime": 176713.5416,
      "train_tokens_per_second": 29249.354
    },
    {
      "epoch": 0.6435990597226492,
      "grad_norm": 0.61328125,
      "learning_rate": 2.978037550455159e-05,
      "loss": 0.40520520210266114,
      "num_input_tokens_seen": 5171592320,
      "step": 18190,
      "train_runtime": 176811.4756,
      "train_tokens_per_second": 29249.189
    },
    {
      "epoch": 0.643952879986378,
      "grad_norm": 0.578125,
      "learning_rate": 2.977509464066091e-05,
      "loss": 0.4089812278747559,
      "num_input_tokens_seen": 5174435520,
      "step": 18200,
      "train_runtime": 176909.624,
      "train_tokens_per_second": 29249.034
    },
    {
      "epoch": 0.6443067002501067,
      "grad_norm": 0.58203125,
      "learning_rate": 2.9769816585093126e-05,
      "loss": 0.40584535598754884,
      "num_input_tokens_seen": 5177222080,
      "step": 18210,
      "train_runtime": 177005.7505,
      "train_tokens_per_second": 29248.892
    },
    {
      "epoch": 0.6446605205138355,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9764541335360024e-05,
      "loss": 0.39990782737731934,
      "num_input_tokens_seen": 5180071744,
      "step": 18220,
      "train_runtime": 177102.6002,
      "train_tokens_per_second": 29248.988
    },
    {
      "epoch": 0.6450143407775643,
      "grad_norm": 0.5625,
      "learning_rate": 2.9759268888976505e-05,
      "loss": 0.40387659072875975,
      "num_input_tokens_seen": 5182936768,
      "step": 18230,
      "train_runtime": 177203.9999,
      "train_tokens_per_second": 29248.419
    },
    {
      "epoch": 0.645368161041293,
      "grad_norm": 0.60546875,
      "learning_rate": 2.9753999243460546e-05,
      "loss": 0.4040827751159668,
      "num_input_tokens_seen": 5185806016,
      "step": 18240,
      "train_runtime": 177303.128,
      "train_tokens_per_second": 29248.249
    },
    {
      "epoch": 0.6457219813050218,
      "grad_norm": 0.6171875,
      "learning_rate": 2.974873239633319e-05,
      "loss": 0.4052712917327881,
      "num_input_tokens_seen": 5188599488,
      "step": 18250,
      "train_runtime": 177397.8722,
      "train_tokens_per_second": 29248.375
    },
    {
      "epoch": 0.6460758015687506,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9743468345118563e-05,
      "loss": 0.40115108489990237,
      "num_input_tokens_seen": 5191436992,
      "step": 18260,
      "train_runtime": 177496.5386,
      "train_tokens_per_second": 29248.103
    },
    {
      "epoch": 0.6464296218324794,
      "grad_norm": 0.62109375,
      "learning_rate": 2.9738207087343844e-05,
      "loss": 0.4042972564697266,
      "num_input_tokens_seen": 5194292288,
      "step": 18270,
      "train_runtime": 177593.7099,
      "train_tokens_per_second": 29248.177
    },
    {
      "epoch": 0.6467834420962082,
      "grad_norm": 0.55078125,
      "learning_rate": 2.9732948620539287e-05,
      "loss": 0.4030327320098877,
      "num_input_tokens_seen": 5197122944,
      "step": 18280,
      "train_runtime": 177688.6268,
      "train_tokens_per_second": 29248.484
    },
    {
      "epoch": 0.647137262359937,
      "grad_norm": 0.59765625,
      "learning_rate": 2.9727692942238188e-05,
      "loss": 0.39977207183837893,
      "num_input_tokens_seen": 5199970304,
      "step": 18290,
      "train_runtime": 177781.784,
      "train_tokens_per_second": 29249.174
    },
    {
      "epoch": 0.6474910826236657,
      "grad_norm": 0.5703125,
      "learning_rate": 2.97224400499769e-05,
      "loss": 0.4066838264465332,
      "num_input_tokens_seen": 5202848960,
      "step": 18300,
      "train_runtime": 177882.1591,
      "train_tokens_per_second": 29248.852
    },
    {
      "epoch": 0.6478449028873945,
      "grad_norm": 0.58203125,
      "learning_rate": 2.9717189941294826e-05,
      "loss": 0.4029226303100586,
      "num_input_tokens_seen": 5205752704,
      "step": 18310,
      "train_runtime": 177982.2667,
      "train_tokens_per_second": 29248.716
    },
    {
      "epoch": 0.6481987231511233,
      "grad_norm": 0.6015625,
      "learning_rate": 2.97119426137344e-05,
      "loss": 0.4060703754425049,
      "num_input_tokens_seen": 5208582144,
      "step": 18320,
      "train_runtime": 178080.4096,
      "train_tokens_per_second": 29248.485
    },
    {
      "epoch": 0.648552543414852,
      "grad_norm": 0.6015625,
      "learning_rate": 2.9706698064841103e-05,
      "loss": 0.40630078315734863,
      "num_input_tokens_seen": 5211359552,
      "step": 18330,
      "train_runtime": 178172.6739,
      "train_tokens_per_second": 29248.927
    },
    {
      "epoch": 0.6489063636785808,
      "grad_norm": 0.5703125,
      "learning_rate": 2.970145629216345e-05,
      "loss": 0.4037792205810547,
      "num_input_tokens_seen": 5214195008,
      "step": 18340,
      "train_runtime": 178268.9692,
      "train_tokens_per_second": 29249.033
    },
    {
      "epoch": 0.6492601839423096,
      "grad_norm": 0.57421875,
      "learning_rate": 2.969621729325296e-05,
      "loss": 0.412615966796875,
      "num_input_tokens_seen": 5217021248,
      "step": 18350,
      "train_runtime": 178365.9207,
      "train_tokens_per_second": 29248.98
    },
    {
      "epoch": 0.6496140042060384,
      "grad_norm": 0.59765625,
      "learning_rate": 2.969098106566421e-05,
      "loss": 0.40331106185913085,
      "num_input_tokens_seen": 5219914240,
      "step": 18360,
      "train_runtime": 178467.6098,
      "train_tokens_per_second": 29248.524
    },
    {
      "epoch": 0.6499678244697672,
      "grad_norm": 0.625,
      "learning_rate": 2.968574760695476e-05,
      "loss": 0.40604658126831056,
      "num_input_tokens_seen": 5222842560,
      "step": 18370,
      "train_runtime": 178570.8671,
      "train_tokens_per_second": 29248.01
    },
    {
      "epoch": 0.650321644733496,
      "grad_norm": 0.57421875,
      "learning_rate": 2.9680516914685206e-05,
      "loss": 0.40002927780151365,
      "num_input_tokens_seen": 5225691136,
      "step": 18380,
      "train_runtime": 178667.3796,
      "train_tokens_per_second": 29248.155
    },
    {
      "epoch": 0.6506754649972247,
      "grad_norm": 0.56640625,
      "learning_rate": 2.9675288986419148e-05,
      "loss": 0.40667295455932617,
      "num_input_tokens_seen": 5228528256,
      "step": 18390,
      "train_runtime": 178764.8603,
      "train_tokens_per_second": 29248.076
    },
    {
      "epoch": 0.6510292852609535,
      "grad_norm": 0.5859375,
      "learning_rate": 2.967006381972317e-05,
      "loss": 0.40378704071044924,
      "num_input_tokens_seen": 5231359040,
      "step": 18400,
      "train_runtime": 178863.0201,
      "train_tokens_per_second": 29247.851
    },
    {
      "epoch": 0.6513831055246823,
      "grad_norm": 0.60546875,
      "learning_rate": 2.9664841412166882e-05,
      "loss": 0.405350399017334,
      "num_input_tokens_seen": 5234193408,
      "step": 18410,
      "train_runtime": 178961.058,
      "train_tokens_per_second": 29247.667
    },
    {
      "epoch": 0.651736925788411,
      "grad_norm": 0.5859375,
      "learning_rate": 2.965962176132287e-05,
      "loss": 0.40405893325805664,
      "num_input_tokens_seen": 5237039872,
      "step": 18420,
      "train_runtime": 179058.2821,
      "train_tokens_per_second": 29247.683
    },
    {
      "epoch": 0.6520907460521398,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9654404864766706e-05,
      "loss": 0.3987222671508789,
      "num_input_tokens_seen": 5239866176,
      "step": 18430,
      "train_runtime": 179153.1633,
      "train_tokens_per_second": 29247.969
    },
    {
      "epoch": 0.6524445663158687,
      "grad_norm": 0.68359375,
      "learning_rate": 2.9649190720076962e-05,
      "loss": 0.40628809928894044,
      "num_input_tokens_seen": 5242772352,
      "step": 18440,
      "train_runtime": 179254.7858,
      "train_tokens_per_second": 29247.6
    },
    {
      "epoch": 0.6527983865795974,
      "grad_norm": 0.56640625,
      "learning_rate": 2.9643979324835176e-05,
      "loss": 0.40418100357055664,
      "num_input_tokens_seen": 5245648320,
      "step": 18450,
      "train_runtime": 179355.7168,
      "train_tokens_per_second": 29247.177
    },
    {
      "epoch": 0.6531522068433262,
      "grad_norm": 0.59765625,
      "learning_rate": 2.9638770676625866e-05,
      "loss": 0.4036346435546875,
      "num_input_tokens_seen": 5248480512,
      "step": 18460,
      "train_runtime": 179455.5824,
      "train_tokens_per_second": 29246.683
    },
    {
      "epoch": 0.653506027107055,
      "grad_norm": 0.57421875,
      "learning_rate": 2.9633564773036517e-05,
      "loss": 0.4022249221801758,
      "num_input_tokens_seen": 5251268608,
      "step": 18470,
      "train_runtime": 179548.7016,
      "train_tokens_per_second": 29247.043
    },
    {
      "epoch": 0.6538598473707837,
      "grad_norm": 0.59765625,
      "learning_rate": 2.9628361611657578e-05,
      "loss": 0.40443782806396483,
      "num_input_tokens_seen": 5254129984,
      "step": 18480,
      "train_runtime": 179649.4192,
      "train_tokens_per_second": 29246.574
    },
    {
      "epoch": 0.6542136676345125,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9623161190082472e-05,
      "loss": 0.4029820919036865,
      "num_input_tokens_seen": 5256959360,
      "step": 18490,
      "train_runtime": 179749.2926,
      "train_tokens_per_second": 29246.064
    },
    {
      "epoch": 0.6545674878982413,
      "grad_norm": 0.58984375,
      "learning_rate": 2.9617963505907554e-05,
      "loss": 0.4015929698944092,
      "num_input_tokens_seen": 5259849216,
      "step": 18500,
      "train_runtime": 179852.3165,
      "train_tokens_per_second": 29245.379
    },
    {
      "epoch": 0.65492130816197,
      "grad_norm": 0.56640625,
      "learning_rate": 2.9612768556732145e-05,
      "loss": 0.40746254920959474,
      "num_input_tokens_seen": 5262726208,
      "step": 18510,
      "train_runtime": 179950.3076,
      "train_tokens_per_second": 29245.442
    },
    {
      "epoch": 0.6552751284256989,
      "grad_norm": 0.5703125,
      "learning_rate": 2.960757634015852e-05,
      "loss": 0.40365753173828123,
      "num_input_tokens_seen": 5265541184,
      "step": 18520,
      "train_runtime": 180046.651,
      "train_tokens_per_second": 29245.427
    },
    {
      "epoch": 0.6556289486894277,
      "grad_norm": 0.58203125,
      "learning_rate": 2.9602386853791885e-05,
      "loss": 0.4015194892883301,
      "num_input_tokens_seen": 5268363008,
      "step": 18530,
      "train_runtime": 180139.0506,
      "train_tokens_per_second": 29246.091
    },
    {
      "epoch": 0.6559827689531564,
      "grad_norm": 0.6015625,
      "learning_rate": 2.959720009524038e-05,
      "loss": 0.40960206985473635,
      "num_input_tokens_seen": 5271242496,
      "step": 18540,
      "train_runtime": 180239.942,
      "train_tokens_per_second": 29245.696
    },
    {
      "epoch": 0.6563365892168852,
      "grad_norm": 0.62890625,
      "learning_rate": 2.9592016062115085e-05,
      "loss": 0.4019481658935547,
      "num_input_tokens_seen": 5274129024,
      "step": 18550,
      "train_runtime": 180337.0544,
      "train_tokens_per_second": 29245.953
    },
    {
      "epoch": 0.656690409480614,
      "grad_norm": 0.6171875,
      "learning_rate": 2.9586834752030002e-05,
      "loss": 0.40447111129760743,
      "num_input_tokens_seen": 5277002112,
      "step": 18560,
      "train_runtime": 180434.7461,
      "train_tokens_per_second": 29246.042
    },
    {
      "epoch": 0.6570442297443427,
      "grad_norm": 0.57421875,
      "learning_rate": 2.958165616260206e-05,
      "loss": 0.40413599014282225,
      "num_input_tokens_seen": 5279807488,
      "step": 18570,
      "train_runtime": 180527.5096,
      "train_tokens_per_second": 29246.554
    },
    {
      "epoch": 0.6573980500080715,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9576480291451114e-05,
      "loss": 0.4024855613708496,
      "num_input_tokens_seen": 5282667136,
      "step": 18580,
      "train_runtime": 180624.931,
      "train_tokens_per_second": 29246.611
    },
    {
      "epoch": 0.6577518702718003,
      "grad_norm": 0.56640625,
      "learning_rate": 2.957130713619991e-05,
      "loss": 0.4043850898742676,
      "num_input_tokens_seen": 5285451392,
      "step": 18590,
      "train_runtime": 180718.729,
      "train_tokens_per_second": 29246.838
    },
    {
      "epoch": 0.658105690535529,
      "grad_norm": 0.57421875,
      "learning_rate": 2.9566136694474138e-05,
      "loss": 0.40281333923339846,
      "num_input_tokens_seen": 5288344704,
      "step": 18600,
      "train_runtime": 180816.6277,
      "train_tokens_per_second": 29247.004
    },
    {
      "epoch": 0.6584595107992579,
      "grad_norm": 0.6015625,
      "learning_rate": 2.9560968963902365e-05,
      "loss": 0.4039157867431641,
      "num_input_tokens_seen": 5291221376,
      "step": 18610,
      "train_runtime": 180914.1267,
      "train_tokens_per_second": 29247.143
    },
    {
      "epoch": 0.6588133310629867,
      "grad_norm": 0.5703125,
      "learning_rate": 2.9555803942116062e-05,
      "loss": 0.40124988555908203,
      "num_input_tokens_seen": 5294061632,
      "step": 18620,
      "train_runtime": 181008.1448,
      "train_tokens_per_second": 29247.643
    },
    {
      "epoch": 0.6591671513267154,
      "grad_norm": 0.578125,
      "learning_rate": 2.955064162674961e-05,
      "loss": 0.4066600799560547,
      "num_input_tokens_seen": 5296885440,
      "step": 18630,
      "train_runtime": 181103.6768,
      "train_tokens_per_second": 29247.807
    },
    {
      "epoch": 0.6595209715904442,
      "grad_norm": 0.60546875,
      "learning_rate": 2.9545482015440266e-05,
      "loss": 0.4041609764099121,
      "num_input_tokens_seen": 5299739584,
      "step": 18640,
      "train_runtime": 181201.6057,
      "train_tokens_per_second": 29247.752
    },
    {
      "epoch": 0.659874791854173,
      "grad_norm": 0.58203125,
      "learning_rate": 2.954032510582819e-05,
      "loss": 0.40394792556762693,
      "num_input_tokens_seen": 5302608896,
      "step": 18650,
      "train_runtime": 181302.4812,
      "train_tokens_per_second": 29247.305
    },
    {
      "epoch": 0.6602286121179017,
      "grad_norm": 0.5625,
      "learning_rate": 2.95351708955564e-05,
      "loss": 0.4041898727416992,
      "num_input_tokens_seen": 5305451584,
      "step": 18660,
      "train_runtime": 181397.7855,
      "train_tokens_per_second": 29247.609
    },
    {
      "epoch": 0.6605824323816305,
      "grad_norm": 0.58203125,
      "learning_rate": 2.9530019382270823e-05,
      "loss": 0.40572643280029297,
      "num_input_tokens_seen": 5308258240,
      "step": 18670,
      "train_runtime": 181493.0896,
      "train_tokens_per_second": 29247.715
    },
    {
      "epoch": 0.6609362526453593,
      "grad_norm": 0.609375,
      "learning_rate": 2.9524870563620233e-05,
      "loss": 0.40117921829223635,
      "num_input_tokens_seen": 5311137216,
      "step": 18680,
      "train_runtime": 181594.5034,
      "train_tokens_per_second": 29247.236
    },
    {
      "epoch": 0.6612900729090881,
      "grad_norm": 0.5625,
      "learning_rate": 2.9519724437256287e-05,
      "loss": 0.4087681770324707,
      "num_input_tokens_seen": 5314011520,
      "step": 18690,
      "train_runtime": 181693.8391,
      "train_tokens_per_second": 29247.065
    },
    {
      "epoch": 0.6616438931728169,
      "grad_norm": 0.60546875,
      "learning_rate": 2.9514581000833496e-05,
      "loss": 0.4028594970703125,
      "num_input_tokens_seen": 5316814848,
      "step": 18700,
      "train_runtime": 181787.3546,
      "train_tokens_per_second": 29247.441
    },
    {
      "epoch": 0.6619977134365457,
      "grad_norm": 0.58203125,
      "learning_rate": 2.950944025200924e-05,
      "loss": 0.40320191383361814,
      "num_input_tokens_seen": 5319664256,
      "step": 18710,
      "train_runtime": 181883.3731,
      "train_tokens_per_second": 29247.667
    },
    {
      "epoch": 0.6623515337002744,
      "grad_norm": 0.58984375,
      "learning_rate": 2.950430218844375e-05,
      "loss": 0.400652551651001,
      "num_input_tokens_seen": 5322458624,
      "step": 18720,
      "train_runtime": 181976.3269,
      "train_tokens_per_second": 29248.083
    },
    {
      "epoch": 0.6627053539640032,
      "grad_norm": 0.578125,
      "learning_rate": 2.9499166807800106e-05,
      "loss": 0.40508260726928713,
      "num_input_tokens_seen": 5325399872,
      "step": 18730,
      "train_runtime": 182078.5306,
      "train_tokens_per_second": 29247.819
    },
    {
      "epoch": 0.663059174227732,
      "grad_norm": 0.59375,
      "learning_rate": 2.949403410774424e-05,
      "loss": 0.4024787425994873,
      "num_input_tokens_seen": 5328314432,
      "step": 18740,
      "train_runtime": 182180.7125,
      "train_tokens_per_second": 29247.412
    },
    {
      "epoch": 0.6634129944914607,
      "grad_norm": 0.57421875,
      "learning_rate": 2.948890408594492e-05,
      "loss": 0.4029059410095215,
      "num_input_tokens_seen": 5331233920,
      "step": 18750,
      "train_runtime": 182282.7735,
      "train_tokens_per_second": 29247.053
    },
    {
      "epoch": 0.6637668147551895,
      "grad_norm": 0.56640625,
      "learning_rate": 2.9483776740073748e-05,
      "loss": 0.40323166847229003,
      "num_input_tokens_seen": 5334099904,
      "step": 18760,
      "train_runtime": 182381.0214,
      "train_tokens_per_second": 29247.012
    },
    {
      "epoch": 0.6641206350189184,
      "grad_norm": 0.59375,
      "learning_rate": 2.947865206780518e-05,
      "loss": 0.4053342819213867,
      "num_input_tokens_seen": 5336928320,
      "step": 18770,
      "train_runtime": 182479.2312,
      "train_tokens_per_second": 29246.771
    },
    {
      "epoch": 0.6644744552826471,
      "grad_norm": 0.5859375,
      "learning_rate": 2.947353006681647e-05,
      "loss": 0.4063398361206055,
      "num_input_tokens_seen": 5339757504,
      "step": 18780,
      "train_runtime": 182577.407,
      "train_tokens_per_second": 29246.54
    },
    {
      "epoch": 0.6648282755463759,
      "grad_norm": 0.62890625,
      "learning_rate": 2.9468410734787717e-05,
      "loss": 0.4052834987640381,
      "num_input_tokens_seen": 5342539648,
      "step": 18790,
      "train_runtime": 182670.8025,
      "train_tokens_per_second": 29246.818
    },
    {
      "epoch": 0.6651820958101047,
      "grad_norm": 0.58984375,
      "learning_rate": 2.9463294069401838e-05,
      "loss": 0.4044794082641602,
      "num_input_tokens_seen": 5345349760,
      "step": 18800,
      "train_runtime": 182767.8813,
      "train_tokens_per_second": 29246.658
    },
    {
      "epoch": 0.6655359160738334,
      "grad_norm": 0.59375,
      "learning_rate": 2.9458180068344555e-05,
      "loss": 0.40369515419006347,
      "num_input_tokens_seen": 5348219840,
      "step": 18810,
      "train_runtime": 182864.9127,
      "train_tokens_per_second": 29246.835
    },
    {
      "epoch": 0.6658897363375622,
      "grad_norm": 0.5703125,
      "learning_rate": 2.9453068729304407e-05,
      "loss": 0.40727624893188474,
      "num_input_tokens_seen": 5351087040,
      "step": 18820,
      "train_runtime": 182963.0546,
      "train_tokens_per_second": 29246.817
    },
    {
      "epoch": 0.666243556601291,
      "grad_norm": 0.61328125,
      "learning_rate": 2.944796004997274e-05,
      "loss": 0.40417704582214353,
      "num_input_tokens_seen": 5353951744,
      "step": 18830,
      "train_runtime": 183062.4412,
      "train_tokens_per_second": 29246.588
    },
    {
      "epoch": 0.6665973768650197,
      "grad_norm": 0.59375,
      "learning_rate": 2.94428540280437e-05,
      "loss": 0.40285549163818357,
      "num_input_tokens_seen": 5356777280,
      "step": 18840,
      "train_runtime": 183159.0265,
      "train_tokens_per_second": 29246.592
    },
    {
      "epoch": 0.6669511971287485,
      "grad_norm": 0.59375,
      "learning_rate": 2.9437750661214242e-05,
      "loss": 0.4091742992401123,
      "num_input_tokens_seen": 5359628992,
      "step": 18850,
      "train_runtime": 183257.4138,
      "train_tokens_per_second": 29246.451
    },
    {
      "epoch": 0.6673050173924774,
      "grad_norm": 0.60546875,
      "learning_rate": 2.9432649947184094e-05,
      "loss": 0.40563039779663085,
      "num_input_tokens_seen": 5362466496,
      "step": 18860,
      "train_runtime": 183354.4268,
      "train_tokens_per_second": 29246.452
    },
    {
      "epoch": 0.6676588376562061,
      "grad_norm": 0.58203125,
      "learning_rate": 2.942755188365579e-05,
      "loss": 0.39965798854827883,
      "num_input_tokens_seen": 5365309760,
      "step": 18870,
      "train_runtime": 183448.7223,
      "train_tokens_per_second": 29246.918
    },
    {
      "epoch": 0.6680126579199349,
      "grad_norm": 0.58203125,
      "learning_rate": 2.9422456468334635e-05,
      "loss": 0.4059948444366455,
      "num_input_tokens_seen": 5368185728,
      "step": 18880,
      "train_runtime": 183547.5694,
      "train_tokens_per_second": 29246.836
    },
    {
      "epoch": 0.6683664781836637,
      "grad_norm": 0.57421875,
      "learning_rate": 2.9417363698928734e-05,
      "loss": 0.4070162296295166,
      "num_input_tokens_seen": 5371068544,
      "step": 18890,
      "train_runtime": 183646.1575,
      "train_tokens_per_second": 29246.833
    },
    {
      "epoch": 0.6687202984473924,
      "grad_norm": 0.58203125,
      "learning_rate": 2.9412273573148946e-05,
      "loss": 0.406981897354126,
      "num_input_tokens_seen": 5373968000,
      "step": 18900,
      "train_runtime": 183744.5213,
      "train_tokens_per_second": 29246.956
    },
    {
      "epoch": 0.6690741187111212,
      "grad_norm": 0.5859375,
      "learning_rate": 2.940718608870891e-05,
      "loss": 0.40624003410339354,
      "num_input_tokens_seen": 5376826432,
      "step": 18910,
      "train_runtime": 183840.8011,
      "train_tokens_per_second": 29247.188
    },
    {
      "epoch": 0.66942793897485,
      "grad_norm": 0.5703125,
      "learning_rate": 2.940210124332504e-05,
      "loss": 0.4022181510925293,
      "num_input_tokens_seen": 5379688192,
      "step": 18920,
      "train_runtime": 183938.5568,
      "train_tokens_per_second": 29247.202
    },
    {
      "epoch": 0.6697817592385787,
      "grad_norm": 0.62109375,
      "learning_rate": 2.9397019034716504e-05,
      "loss": 0.3991296052932739,
      "num_input_tokens_seen": 5382474368,
      "step": 18930,
      "train_runtime": 184031.302,
      "train_tokens_per_second": 29247.602
    },
    {
      "epoch": 0.6701355795023076,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9391939460605233e-05,
      "loss": 0.4051826953887939,
      "num_input_tokens_seen": 5385377792,
      "step": 18940,
      "train_runtime": 184132.8964,
      "train_tokens_per_second": 29247.233
    },
    {
      "epoch": 0.6704893997660364,
      "grad_norm": 0.578125,
      "learning_rate": 2.9386862518715914e-05,
      "loss": 0.4025622844696045,
      "num_input_tokens_seen": 5388224832,
      "step": 18950,
      "train_runtime": 184230.0332,
      "train_tokens_per_second": 29247.266
    },
    {
      "epoch": 0.6708432200297652,
      "grad_norm": 0.5625,
      "learning_rate": 2.9381788206775966e-05,
      "loss": 0.4046022415161133,
      "num_input_tokens_seen": 5391128576,
      "step": 18960,
      "train_runtime": 184329.5082,
      "train_tokens_per_second": 29247.236
    },
    {
      "epoch": 0.6711970402934939,
      "grad_norm": 0.58984375,
      "learning_rate": 2.937671652251559e-05,
      "loss": 0.40441365242004396,
      "num_input_tokens_seen": 5393952960,
      "step": 18970,
      "train_runtime": 184424.9768,
      "train_tokens_per_second": 29247.41
    },
    {
      "epoch": 0.6715508605572227,
      "grad_norm": 0.58203125,
      "learning_rate": 2.9371647463667696e-05,
      "loss": 0.4034428119659424,
      "num_input_tokens_seen": 5396779968,
      "step": 18980,
      "train_runtime": 184520.3678,
      "train_tokens_per_second": 29247.611
    },
    {
      "epoch": 0.6719046808209514,
      "grad_norm": 0.57421875,
      "learning_rate": 2.9366581027967943e-05,
      "loss": 0.40221443176269533,
      "num_input_tokens_seen": 5399650560,
      "step": 18990,
      "train_runtime": 184618.8784,
      "train_tokens_per_second": 29247.554
    },
    {
      "epoch": 0.6722585010846802,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9361517213154726e-05,
      "loss": 0.39937705993652345,
      "num_input_tokens_seen": 5402479168,
      "step": 19000,
      "train_runtime": 184716.1811,
      "train_tokens_per_second": 29247.46
    },
    {
      "epoch": 0.672612321348409,
      "grad_norm": 0.60546875,
      "learning_rate": 2.935645601696917e-05,
      "loss": 0.4040325164794922,
      "num_input_tokens_seen": 5405312896,
      "step": 19010,
      "train_runtime": 184811.7598,
      "train_tokens_per_second": 29247.667
    },
    {
      "epoch": 0.6729661416121379,
      "grad_norm": 0.59765625,
      "learning_rate": 2.9351397437155114e-05,
      "loss": 0.4082945346832275,
      "num_input_tokens_seen": 5408159488,
      "step": 19020,
      "train_runtime": 184910.5219,
      "train_tokens_per_second": 29247.441
    },
    {
      "epoch": 0.6733199618758666,
      "grad_norm": 0.58984375,
      "learning_rate": 2.934634147145913e-05,
      "loss": 0.4019640922546387,
      "num_input_tokens_seen": 5411064832,
      "step": 19030,
      "train_runtime": 185014.0706,
      "train_tokens_per_second": 29246.775
    },
    {
      "epoch": 0.6736737821395954,
      "grad_norm": 0.609375,
      "learning_rate": 2.9341288117630495e-05,
      "loss": 0.4040031909942627,
      "num_input_tokens_seen": 5413963968,
      "step": 19040,
      "train_runtime": 185116.6782,
      "train_tokens_per_second": 29246.225
    },
    {
      "epoch": 0.6740276024033242,
      "grad_norm": 0.59375,
      "learning_rate": 2.933623737342121e-05,
      "loss": 0.40644545555114747,
      "num_input_tokens_seen": 5416799744,
      "step": 19050,
      "train_runtime": 185213.9087,
      "train_tokens_per_second": 29246.182
    },
    {
      "epoch": 0.6743814226670529,
      "grad_norm": 0.58203125,
      "learning_rate": 2.9331189236585977e-05,
      "loss": 0.4047755241394043,
      "num_input_tokens_seen": 5419656768,
      "step": 19060,
      "train_runtime": 185311.5507,
      "train_tokens_per_second": 29246.19
    },
    {
      "epoch": 0.6747352429307817,
      "grad_norm": 0.578125,
      "learning_rate": 2.9326143704882192e-05,
      "loss": 0.403855037689209,
      "num_input_tokens_seen": 5422565120,
      "step": 19070,
      "train_runtime": 185412.0806,
      "train_tokens_per_second": 29246.018
    },
    {
      "epoch": 0.6750890631945105,
      "grad_norm": 0.58203125,
      "learning_rate": 2.932110077606997e-05,
      "loss": 0.4038975238800049,
      "num_input_tokens_seen": 5425444800,
      "step": 19080,
      "train_runtime": 185508.9213,
      "train_tokens_per_second": 29246.274
    },
    {
      "epoch": 0.6754428834582392,
      "grad_norm": 0.59375,
      "learning_rate": 2.931606044791212e-05,
      "loss": 0.406279182434082,
      "num_input_tokens_seen": 5428307008,
      "step": 19090,
      "train_runtime": 185605.7343,
      "train_tokens_per_second": 29246.44
    },
    {
      "epoch": 0.675796703721968,
      "grad_norm": 0.62109375,
      "learning_rate": 2.9311022718174114e-05,
      "loss": 0.4090275764465332,
      "num_input_tokens_seen": 5431148672,
      "step": 19100,
      "train_runtime": 185701.6331,
      "train_tokens_per_second": 29246.639
    },
    {
      "epoch": 0.6761505239856969,
      "grad_norm": 0.58984375,
      "learning_rate": 2.930598758462415e-05,
      "loss": 0.4032048225402832,
      "num_input_tokens_seen": 5433984128,
      "step": 19110,
      "train_runtime": 185797.9786,
      "train_tokens_per_second": 29246.734
    },
    {
      "epoch": 0.6765043442494256,
      "grad_norm": 0.578125,
      "learning_rate": 2.9300955045033072e-05,
      "loss": 0.40287294387817385,
      "num_input_tokens_seen": 5436845760,
      "step": 19120,
      "train_runtime": 185896.0779,
      "train_tokens_per_second": 29246.694
    },
    {
      "epoch": 0.6768581645131544,
      "grad_norm": 0.609375,
      "learning_rate": 2.9295925097174435e-05,
      "loss": 0.40950851440429686,
      "num_input_tokens_seen": 5439687872,
      "step": 19130,
      "train_runtime": 185992.9622,
      "train_tokens_per_second": 29246.74
    },
    {
      "epoch": 0.6772119847768832,
      "grad_norm": 0.58984375,
      "learning_rate": 2.929089773882445e-05,
      "loss": 0.39822540283203123,
      "num_input_tokens_seen": 5442598016,
      "step": 19140,
      "train_runtime": 186092.7309,
      "train_tokens_per_second": 29246.699
    },
    {
      "epoch": 0.6775658050406119,
      "grad_norm": 0.58984375,
      "learning_rate": 2.9285872967762007e-05,
      "loss": 0.40432300567626955,
      "num_input_tokens_seen": 5445430784,
      "step": 19150,
      "train_runtime": 186190.2248,
      "train_tokens_per_second": 29246.599
    },
    {
      "epoch": 0.6779196253043407,
      "grad_norm": 0.56640625,
      "learning_rate": 2.9280850781768638e-05,
      "loss": 0.40349321365356444,
      "num_input_tokens_seen": 5448254016,
      "step": 19160,
      "train_runtime": 186285.4888,
      "train_tokens_per_second": 29246.798
    },
    {
      "epoch": 0.6782734455680695,
      "grad_norm": 0.58203125,
      "learning_rate": 2.9275831178628576e-05,
      "loss": 0.4038459300994873,
      "num_input_tokens_seen": 5451079232,
      "step": 19170,
      "train_runtime": 186383.2017,
      "train_tokens_per_second": 29246.623
    },
    {
      "epoch": 0.6786272658317982,
      "grad_norm": 0.5546875,
      "learning_rate": 2.927081415612869e-05,
      "loss": 0.40450258255004884,
      "num_input_tokens_seen": 5453943552,
      "step": 19180,
      "train_runtime": 186479.595,
      "train_tokens_per_second": 29246.865
    },
    {
      "epoch": 0.6789810860955271,
      "grad_norm": 0.55859375,
      "learning_rate": 2.9265799712058505e-05,
      "loss": 0.4013828277587891,
      "num_input_tokens_seen": 5456769984,
      "step": 19190,
      "train_runtime": 186577.021,
      "train_tokens_per_second": 29246.742
    },
    {
      "epoch": 0.6793349063592559,
      "grad_norm": 0.57421875,
      "learning_rate": 2.926078784421019e-05,
      "loss": 0.40457735061645506,
      "num_input_tokens_seen": 5459565120,
      "step": 19200,
      "train_runtime": 186673.2695,
      "train_tokens_per_second": 29246.636
    },
    {
      "epoch": 0.6796887266229846,
      "grad_norm": 0.58203125,
      "learning_rate": 2.9255778550378575e-05,
      "loss": 0.40732393264770506,
      "num_input_tokens_seen": 5462365376,
      "step": 19210,
      "train_runtime": 186767.7108,
      "train_tokens_per_second": 29246.84
    },
    {
      "epoch": 0.6800425468867134,
      "grad_norm": 0.6015625,
      "learning_rate": 2.9250771828361118e-05,
      "loss": 0.40843663215637205,
      "num_input_tokens_seen": 5465167104,
      "step": 19220,
      "train_runtime": 186862.1113,
      "train_tokens_per_second": 29247.059
    },
    {
      "epoch": 0.6803963671504422,
      "grad_norm": 0.56640625,
      "learning_rate": 2.9245767675957923e-05,
      "loss": 0.40323553085327146,
      "num_input_tokens_seen": 5468045760,
      "step": 19230,
      "train_runtime": 186963.0588,
      "train_tokens_per_second": 29246.664
    },
    {
      "epoch": 0.6807501874141709,
      "grad_norm": 0.59765625,
      "learning_rate": 2.924076609097172e-05,
      "loss": 0.4037144184112549,
      "num_input_tokens_seen": 5470898304,
      "step": 19240,
      "train_runtime": 187059.6483,
      "train_tokens_per_second": 29246.812
    },
    {
      "epoch": 0.6811040076778997,
      "grad_norm": 0.59765625,
      "learning_rate": 2.9235767071207876e-05,
      "loss": 0.4081141471862793,
      "num_input_tokens_seen": 5473720128,
      "step": 19250,
      "train_runtime": 187157.8261,
      "train_tokens_per_second": 29246.547
    },
    {
      "epoch": 0.6814578279416285,
      "grad_norm": 0.59375,
      "learning_rate": 2.9230770614474383e-05,
      "loss": 0.4026987075805664,
      "num_input_tokens_seen": 5476543296,
      "step": 19260,
      "train_runtime": 187251.8523,
      "train_tokens_per_second": 29246.938
    },
    {
      "epoch": 0.6818116482053573,
      "grad_norm": 0.55859375,
      "learning_rate": 2.9225776718581842e-05,
      "loss": 0.4051219940185547,
      "num_input_tokens_seen": 5479396800,
      "step": 19270,
      "train_runtime": 187351.6803,
      "train_tokens_per_second": 29246.585
    },
    {
      "epoch": 0.6821654684690861,
      "grad_norm": 0.62890625,
      "learning_rate": 2.9220785381343495e-05,
      "loss": 0.40059242248535154,
      "num_input_tokens_seen": 5482251968,
      "step": 19280,
      "train_runtime": 187447.6358,
      "train_tokens_per_second": 29246.845
    },
    {
      "epoch": 0.6825192887328149,
      "grad_norm": 0.58203125,
      "learning_rate": 2.9215796600575174e-05,
      "loss": 0.4052286624908447,
      "num_input_tokens_seen": 5485147520,
      "step": 19290,
      "train_runtime": 187546.0509,
      "train_tokens_per_second": 29246.937
    },
    {
      "epoch": 0.6828731089965436,
      "grad_norm": 0.6015625,
      "learning_rate": 2.9210810374095332e-05,
      "loss": 0.40407209396362304,
      "num_input_tokens_seen": 5488081408,
      "step": 19300,
      "train_runtime": 187649.9708,
      "train_tokens_per_second": 29246.375
    },
    {
      "epoch": 0.6832269292602724,
      "grad_norm": 0.6015625,
      "learning_rate": 2.9205826699725026e-05,
      "loss": 0.403245210647583,
      "num_input_tokens_seen": 5490922368,
      "step": 19310,
      "train_runtime": 187746.8513,
      "train_tokens_per_second": 29246.415
    },
    {
      "epoch": 0.6835807495240012,
      "grad_norm": 0.58203125,
      "learning_rate": 2.920084557528791e-05,
      "loss": 0.40039286613464353,
      "num_input_tokens_seen": 5493779520,
      "step": 19320,
      "train_runtime": 187844.489,
      "train_tokens_per_second": 29246.424
    },
    {
      "epoch": 0.6839345697877299,
      "grad_norm": 0.58984375,
      "learning_rate": 2.9195866998610245e-05,
      "loss": 0.40297584533691405,
      "num_input_tokens_seen": 5496683584,
      "step": 19330,
      "train_runtime": 187944.5477,
      "train_tokens_per_second": 29246.305
    },
    {
      "epoch": 0.6842883900514587,
      "grad_norm": 0.5703125,
      "learning_rate": 2.9190890967520878e-05,
      "loss": 0.4039573669433594,
      "num_input_tokens_seen": 5499504832,
      "step": 19340,
      "train_runtime": 188040.7147,
      "train_tokens_per_second": 29246.351
    },
    {
      "epoch": 0.6846422103151875,
      "grad_norm": 0.58984375,
      "learning_rate": 2.9185917479851243e-05,
      "loss": 0.40180506706237795,
      "num_input_tokens_seen": 5502320448,
      "step": 19350,
      "train_runtime": 188134.4847,
      "train_tokens_per_second": 29246.74
    },
    {
      "epoch": 0.6849960305789163,
      "grad_norm": 0.59375,
      "learning_rate": 2.9180946533435366e-05,
      "loss": 0.40492935180664064,
      "num_input_tokens_seen": 5505159232,
      "step": 19360,
      "train_runtime": 188232.8225,
      "train_tokens_per_second": 29246.542
    },
    {
      "epoch": 0.6853498508426451,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9175978126109853e-05,
      "loss": 0.40566396713256836,
      "num_input_tokens_seen": 5508060096,
      "step": 19370,
      "train_runtime": 188331.6847,
      "train_tokens_per_second": 29246.593
    },
    {
      "epoch": 0.6857036711063739,
      "grad_norm": 0.58984375,
      "learning_rate": 2.9171012255713885e-05,
      "loss": 0.40088791847229005,
      "num_input_tokens_seen": 5510877184,
      "step": 19380,
      "train_runtime": 188426.1362,
      "train_tokens_per_second": 29246.883
    },
    {
      "epoch": 0.6860574913701026,
      "grad_norm": 0.57421875,
      "learning_rate": 2.9166048920089217e-05,
      "loss": 0.40096755027770997,
      "num_input_tokens_seen": 5513754944,
      "step": 19390,
      "train_runtime": 188525.269,
      "train_tokens_per_second": 29246.769
    },
    {
      "epoch": 0.6864113116338314,
      "grad_norm": 0.59375,
      "learning_rate": 2.9161088117080175e-05,
      "loss": 0.4008651256561279,
      "num_input_tokens_seen": 5516608448,
      "step": 19400,
      "train_runtime": 188622.2714,
      "train_tokens_per_second": 29246.856
    },
    {
      "epoch": 0.6867651318975602,
      "grad_norm": 0.6015625,
      "learning_rate": 2.9156129844533658e-05,
      "loss": 0.4028066635131836,
      "num_input_tokens_seen": 5519425216,
      "step": 19410,
      "train_runtime": 188718.2727,
      "train_tokens_per_second": 29246.904
    },
    {
      "epoch": 0.6871189521612889,
      "grad_norm": 0.59765625,
      "learning_rate": 2.9151174100299104e-05,
      "loss": 0.40504870414733884,
      "num_input_tokens_seen": 5522282560,
      "step": 19420,
      "train_runtime": 188815.3237,
      "train_tokens_per_second": 29247.004
    },
    {
      "epoch": 0.6874727724250177,
      "grad_norm": 0.6015625,
      "learning_rate": 2.9146220882228535e-05,
      "loss": 0.40459461212158204,
      "num_input_tokens_seen": 5525120704,
      "step": 19430,
      "train_runtime": 188911.3949,
      "train_tokens_per_second": 29247.154
    },
    {
      "epoch": 0.6878265926887466,
      "grad_norm": 0.5703125,
      "learning_rate": 2.914127018817651e-05,
      "loss": 0.404727840423584,
      "num_input_tokens_seen": 5527968768,
      "step": 19440,
      "train_runtime": 189007.6364,
      "train_tokens_per_second": 29247.33
    },
    {
      "epoch": 0.6881804129524753,
      "grad_norm": 0.578125,
      "learning_rate": 2.9136322016000152e-05,
      "loss": 0.40401601791381836,
      "num_input_tokens_seen": 5530816640,
      "step": 19450,
      "train_runtime": 189104.0575,
      "train_tokens_per_second": 29247.477
    },
    {
      "epoch": 0.6885342332162041,
      "grad_norm": 0.6015625,
      "learning_rate": 2.9131376363559116e-05,
      "loss": 0.4041296005249023,
      "num_input_tokens_seen": 5533670784,
      "step": 19460,
      "train_runtime": 189202.7079,
      "train_tokens_per_second": 29247.313
    },
    {
      "epoch": 0.6888880534799329,
      "grad_norm": 0.5703125,
      "learning_rate": 2.9126433228715606e-05,
      "loss": 0.40099411010742186,
      "num_input_tokens_seen": 5536528192,
      "step": 19470,
      "train_runtime": 189300.3092,
      "train_tokens_per_second": 29247.328
    },
    {
      "epoch": 0.6892418737436616,
      "grad_norm": 0.61328125,
      "learning_rate": 2.9121492609334354e-05,
      "loss": 0.3998518466949463,
      "num_input_tokens_seen": 5539371584,
      "step": 19480,
      "train_runtime": 189398.7008,
      "train_tokens_per_second": 29247.147
    },
    {
      "epoch": 0.6895956940073904,
      "grad_norm": 0.578125,
      "learning_rate": 2.9116554503282656e-05,
      "loss": 0.40839290618896484,
      "num_input_tokens_seen": 5542197632,
      "step": 19490,
      "train_runtime": 189496.4358,
      "train_tokens_per_second": 29246.976
    },
    {
      "epoch": 0.6899495142711192,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9111618908430303e-05,
      "loss": 0.40813555717468264,
      "num_input_tokens_seen": 5545063616,
      "step": 19500,
      "train_runtime": 189593.6999,
      "train_tokens_per_second": 29247.088
    },
    {
      "epoch": 0.6903033345348479,
      "grad_norm": 0.5703125,
      "learning_rate": 2.9106685822649643e-05,
      "loss": 0.4031411647796631,
      "num_input_tokens_seen": 5547930752,
      "step": 19510,
      "train_runtime": 189693.8308,
      "train_tokens_per_second": 29246.764
    },
    {
      "epoch": 0.6906571547985768,
      "grad_norm": 0.58203125,
      "learning_rate": 2.910175524381552e-05,
      "loss": 0.4062369346618652,
      "num_input_tokens_seen": 5550838656,
      "step": 19520,
      "train_runtime": 189795.7178,
      "train_tokens_per_second": 29246.385
    },
    {
      "epoch": 0.6910109750623056,
      "grad_norm": 0.58984375,
      "learning_rate": 2.9096827169805318e-05,
      "loss": 0.40814809799194335,
      "num_input_tokens_seen": 5553703552,
      "step": 19530,
      "train_runtime": 189892.8483,
      "train_tokens_per_second": 29246.512
    },
    {
      "epoch": 0.6913647953260343,
      "grad_norm": 0.61328125,
      "learning_rate": 2.9091901598498923e-05,
      "loss": 0.3996427536010742,
      "num_input_tokens_seen": 5556544960,
      "step": 19540,
      "train_runtime": 189988.2257,
      "train_tokens_per_second": 29246.786
    },
    {
      "epoch": 0.6917186155897631,
      "grad_norm": 0.55078125,
      "learning_rate": 2.9086978527778736e-05,
      "loss": 0.4041744709014893,
      "num_input_tokens_seen": 5559399936,
      "step": 19550,
      "train_runtime": 190088.2654,
      "train_tokens_per_second": 29246.413
    },
    {
      "epoch": 0.6920724358534919,
      "grad_norm": 0.59375,
      "learning_rate": 2.9082057955529668e-05,
      "loss": 0.4075783252716064,
      "num_input_tokens_seen": 5562200832,
      "step": 19560,
      "train_runtime": 190182.9895,
      "train_tokens_per_second": 29246.574
    },
    {
      "epoch": 0.6924262561172206,
      "grad_norm": 0.5859375,
      "learning_rate": 2.907713987963914e-05,
      "loss": 0.40821161270141604,
      "num_input_tokens_seen": 5565021952,
      "step": 19570,
      "train_runtime": 190278.3861,
      "train_tokens_per_second": 29246.737
    },
    {
      "epoch": 0.6927800763809494,
      "grad_norm": 0.5703125,
      "learning_rate": 2.9072224297997058e-05,
      "loss": 0.4013996601104736,
      "num_input_tokens_seen": 5567839552,
      "step": 19580,
      "train_runtime": 190373.5433,
      "train_tokens_per_second": 29246.919
    },
    {
      "epoch": 0.6931338966446782,
      "grad_norm": 0.6015625,
      "learning_rate": 2.9067311208495834e-05,
      "loss": 0.40244307518005373,
      "num_input_tokens_seen": 5570708480,
      "step": 19590,
      "train_runtime": 190471.9239,
      "train_tokens_per_second": 29246.875
    },
    {
      "epoch": 0.6934877169084069,
      "grad_norm": 0.59765625,
      "learning_rate": 2.9062400609030372e-05,
      "loss": 0.4038554191589355,
      "num_input_tokens_seen": 5573613440,
      "step": 19600,
      "train_runtime": 190576.8293,
      "train_tokens_per_second": 29246.018
    },
    {
      "epoch": 0.6938415371721358,
      "grad_norm": 0.58984375,
      "learning_rate": 2.905749249749806e-05,
      "loss": 0.40879669189453127,
      "num_input_tokens_seen": 5576448704,
      "step": 19610,
      "train_runtime": 190672.7253,
      "train_tokens_per_second": 29246.179
    },
    {
      "epoch": 0.6941953574358646,
      "grad_norm": 0.59375,
      "learning_rate": 2.905258687179878e-05,
      "loss": 0.40604748725891116,
      "num_input_tokens_seen": 5579299264,
      "step": 19620,
      "train_runtime": 190771.4989,
      "train_tokens_per_second": 29245.979
    },
    {
      "epoch": 0.6945491776995933,
      "grad_norm": 0.55859375,
      "learning_rate": 2.904768372983488e-05,
      "loss": 0.3974184989929199,
      "num_input_tokens_seen": 5582173632,
      "step": 19630,
      "train_runtime": 190870.5376,
      "train_tokens_per_second": 29245.863
    },
    {
      "epoch": 0.6949029979633221,
      "grad_norm": 0.60546875,
      "learning_rate": 2.90427830695112e-05,
      "loss": 0.4015629768371582,
      "num_input_tokens_seen": 5585031104,
      "step": 19640,
      "train_runtime": 190967.3811,
      "train_tokens_per_second": 29245.995
    },
    {
      "epoch": 0.6952568182270509,
      "grad_norm": 0.578125,
      "learning_rate": 2.9037884888735044e-05,
      "loss": 0.40587544441223145,
      "num_input_tokens_seen": 5587897024,
      "step": 19650,
      "train_runtime": 191064.6038,
      "train_tokens_per_second": 29246.113
    },
    {
      "epoch": 0.6956106384907796,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9032989185416194e-05,
      "loss": 0.40116868019104,
      "num_input_tokens_seen": 5590733440,
      "step": 19660,
      "train_runtime": 191159.7141,
      "train_tokens_per_second": 29246.4
    },
    {
      "epoch": 0.6959644587545084,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9028095957466888e-05,
      "loss": 0.4036966323852539,
      "num_input_tokens_seen": 5593578688,
      "step": 19670,
      "train_runtime": 191259.5813,
      "train_tokens_per_second": 29246.005
    },
    {
      "epoch": 0.6963182790182372,
      "grad_norm": 0.57421875,
      "learning_rate": 2.9023205202801833e-05,
      "loss": 0.4065293312072754,
      "num_input_tokens_seen": 5596428224,
      "step": 19680,
      "train_runtime": 191359.8402,
      "train_tokens_per_second": 29245.573
    },
    {
      "epoch": 0.696672099281966,
      "grad_norm": 0.5625,
      "learning_rate": 2.90183169193382e-05,
      "loss": 0.40302457809448244,
      "num_input_tokens_seen": 5599307520,
      "step": 19690,
      "train_runtime": 191458.7617,
      "train_tokens_per_second": 29245.502
    },
    {
      "epoch": 0.6970259195456948,
      "grad_norm": 0.58984375,
      "learning_rate": 2.90134311049956e-05,
      "loss": 0.4052546501159668,
      "num_input_tokens_seen": 5602154304,
      "step": 19700,
      "train_runtime": 191557.3991,
      "train_tokens_per_second": 29245.304
    },
    {
      "epoch": 0.6973797398094236,
      "grad_norm": 0.58984375,
      "learning_rate": 2.90085477576961e-05,
      "loss": 0.4053903579711914,
      "num_input_tokens_seen": 5605031168,
      "step": 19710,
      "train_runtime": 191657.1076,
      "train_tokens_per_second": 29245.099
    },
    {
      "epoch": 0.6977335600731523,
      "grad_norm": 0.5859375,
      "learning_rate": 2.9003666875364226e-05,
      "loss": 0.40281357765197756,
      "num_input_tokens_seen": 5607879552,
      "step": 19720,
      "train_runtime": 191757.3348,
      "train_tokens_per_second": 29244.668
    },
    {
      "epoch": 0.6980873803368811,
      "grad_norm": 0.5859375,
      "learning_rate": 2.899878845592694e-05,
      "loss": 0.4050135612487793,
      "num_input_tokens_seen": 5610736960,
      "step": 19730,
      "train_runtime": 191854.1822,
      "train_tokens_per_second": 29244.799
    },
    {
      "epoch": 0.6984412006006099,
      "grad_norm": 0.59765625,
      "learning_rate": 2.8993912497313636e-05,
      "loss": 0.40939970016479493,
      "num_input_tokens_seen": 5613547520,
      "step": 19740,
      "train_runtime": 191949.555,
      "train_tokens_per_second": 29244.91
    },
    {
      "epoch": 0.6987950208643386,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8989038997456153e-05,
      "loss": 0.4023500919342041,
      "num_input_tokens_seen": 5616413824,
      "step": 19750,
      "train_runtime": 192049.239,
      "train_tokens_per_second": 29244.655
    },
    {
      "epoch": 0.6991488411280674,
      "grad_norm": 0.609375,
      "learning_rate": 2.898416795428876e-05,
      "loss": 0.39735095500946044,
      "num_input_tokens_seen": 5619273088,
      "step": 19760,
      "train_runtime": 192147.72,
      "train_tokens_per_second": 29244.547
    },
    {
      "epoch": 0.6995026613917963,
      "grad_norm": 0.57421875,
      "learning_rate": 2.897929936574816e-05,
      "loss": 0.40691452026367186,
      "num_input_tokens_seen": 5622143360,
      "step": 19770,
      "train_runtime": 192246.3783,
      "train_tokens_per_second": 29244.47
    },
    {
      "epoch": 0.699856481655525,
      "grad_norm": 0.5703125,
      "learning_rate": 2.8974433229773478e-05,
      "loss": 0.4022841930389404,
      "num_input_tokens_seen": 5624978304,
      "step": 19780,
      "train_runtime": 192343.5262,
      "train_tokens_per_second": 29244.438
    },
    {
      "epoch": 0.7002103019192538,
      "grad_norm": 0.578125,
      "learning_rate": 2.896956954430625e-05,
      "loss": 0.4064777374267578,
      "num_input_tokens_seen": 5627826944,
      "step": 19790,
      "train_runtime": 192439.5079,
      "train_tokens_per_second": 29244.655
    },
    {
      "epoch": 0.7005641221829826,
      "grad_norm": 0.57421875,
      "learning_rate": 2.896470830729045e-05,
      "loss": 0.40154685974121096,
      "num_input_tokens_seen": 5630693376,
      "step": 19800,
      "train_runtime": 192538.2174,
      "train_tokens_per_second": 29244.549
    },
    {
      "epoch": 0.7009179424467114,
      "grad_norm": 0.6015625,
      "learning_rate": 2.895984951667245e-05,
      "loss": 0.3982991218566895,
      "num_input_tokens_seen": 5633505024,
      "step": 19810,
      "train_runtime": 192634.6196,
      "train_tokens_per_second": 29244.51
    },
    {
      "epoch": 0.7012717627104401,
      "grad_norm": 0.58203125,
      "learning_rate": 2.8954993170401047e-05,
      "loss": 0.40444517135620117,
      "num_input_tokens_seen": 5636371072,
      "step": 19820,
      "train_runtime": 192733.9249,
      "train_tokens_per_second": 29244.312
    },
    {
      "epoch": 0.7016255829741689,
      "grad_norm": 0.6171875,
      "learning_rate": 2.8950139266427423e-05,
      "loss": 0.4031824111938477,
      "num_input_tokens_seen": 5639171712,
      "step": 19830,
      "train_runtime": 192833.5845,
      "train_tokens_per_second": 29243.722
    },
    {
      "epoch": 0.7019794032378976,
      "grad_norm": 0.5703125,
      "learning_rate": 2.8945287802705186e-05,
      "loss": 0.40310916900634763,
      "num_input_tokens_seen": 5642090560,
      "step": 19840,
      "train_runtime": 192937.0257,
      "train_tokens_per_second": 29243.172
    },
    {
      "epoch": 0.7023332235016264,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8940438777190336e-05,
      "loss": 0.4033676624298096,
      "num_input_tokens_seen": 5644966464,
      "step": 19850,
      "train_runtime": 193034.7286,
      "train_tokens_per_second": 29243.269
    },
    {
      "epoch": 0.7026870437653553,
      "grad_norm": 0.58984375,
      "learning_rate": 2.8935592187841265e-05,
      "loss": 0.4016928195953369,
      "num_input_tokens_seen": 5647789632,
      "step": 19860,
      "train_runtime": 193131.0233,
      "train_tokens_per_second": 29243.306
    },
    {
      "epoch": 0.7030408640290841,
      "grad_norm": 0.59375,
      "learning_rate": 2.893074803261876e-05,
      "loss": 0.40922183990478517,
      "num_input_tokens_seen": 5650654400,
      "step": 19870,
      "train_runtime": 193229.4359,
      "train_tokens_per_second": 29243.238
    },
    {
      "epoch": 0.7033946842928128,
      "grad_norm": 0.5703125,
      "learning_rate": 2.8925906309485995e-05,
      "loss": 0.4039417266845703,
      "num_input_tokens_seen": 5653547328,
      "step": 19880,
      "train_runtime": 193328.4804,
      "train_tokens_per_second": 29243.22
    },
    {
      "epoch": 0.7037485045565416,
      "grad_norm": 0.578125,
      "learning_rate": 2.8921067016408532e-05,
      "loss": 0.3992153644561768,
      "num_input_tokens_seen": 5656377152,
      "step": 19890,
      "train_runtime": 193422.4224,
      "train_tokens_per_second": 29243.648
    },
    {
      "epoch": 0.7041023248202704,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8916230151354316e-05,
      "loss": 0.40674762725830077,
      "num_input_tokens_seen": 5659159808,
      "step": 19900,
      "train_runtime": 193517.1736,
      "train_tokens_per_second": 29243.708
    },
    {
      "epoch": 0.7044561450839991,
      "grad_norm": 0.58984375,
      "learning_rate": 2.8911395712293677e-05,
      "loss": 0.4048741817474365,
      "num_input_tokens_seen": 5661991040,
      "step": 19910,
      "train_runtime": 193613.7713,
      "train_tokens_per_second": 29243.741
    },
    {
      "epoch": 0.7048099653477279,
      "grad_norm": 0.59375,
      "learning_rate": 2.8906563697199294e-05,
      "loss": 0.40169076919555663,
      "num_input_tokens_seen": 5664812992,
      "step": 19920,
      "train_runtime": 193708.9337,
      "train_tokens_per_second": 29243.943
    },
    {
      "epoch": 0.7051637856114567,
      "grad_norm": 0.5703125,
      "learning_rate": 2.8901734104046245e-05,
      "loss": 0.400034236907959,
      "num_input_tokens_seen": 5667594496,
      "step": 19930,
      "train_runtime": 193803.5676,
      "train_tokens_per_second": 29244.015
    },
    {
      "epoch": 0.7055176058751855,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8896906930811964e-05,
      "loss": 0.40499048233032225,
      "num_input_tokens_seen": 5670416896,
      "step": 19940,
      "train_runtime": 193897.6874,
      "train_tokens_per_second": 29244.376
    },
    {
      "epoch": 0.7058714261389143,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8892082175476243e-05,
      "loss": 0.40171356201171876,
      "num_input_tokens_seen": 5673184896,
      "step": 19950,
      "train_runtime": 193988.9811,
      "train_tokens_per_second": 29244.882
    },
    {
      "epoch": 0.7062252464026431,
      "grad_norm": 0.59765625,
      "learning_rate": 2.8887259836021246e-05,
      "loss": 0.4029547214508057,
      "num_input_tokens_seen": 5676008320,
      "step": 19960,
      "train_runtime": 194084.197,
      "train_tokens_per_second": 29245.082
    },
    {
      "epoch": 0.7065790666663718,
      "grad_norm": 0.59765625,
      "learning_rate": 2.8882439910431493e-05,
      "loss": 0.4039130210876465,
      "num_input_tokens_seen": 5678878592,
      "step": 19970,
      "train_runtime": 194184.391,
      "train_tokens_per_second": 29244.774
    },
    {
      "epoch": 0.7069328869301006,
      "grad_norm": 0.578125,
      "learning_rate": 2.8877622396693843e-05,
      "loss": 0.4053036689758301,
      "num_input_tokens_seen": 5681686336,
      "step": 19980,
      "train_runtime": 194278.5254,
      "train_tokens_per_second": 29245.056
    },
    {
      "epoch": 0.7072867071938294,
      "grad_norm": 0.578125,
      "learning_rate": 2.8872807292797516e-05,
      "loss": 0.4042822360992432,
      "num_input_tokens_seen": 5684484224,
      "step": 19990,
      "train_runtime": 194372.274,
      "train_tokens_per_second": 29245.345
    },
    {
      "epoch": 0.7076405274575581,
      "grad_norm": 0.546875,
      "learning_rate": 2.8867994596734084e-05,
      "loss": 0.399719762802124,
      "num_input_tokens_seen": 5687337600,
      "step": 20000,
      "train_runtime": 194469.5424,
      "train_tokens_per_second": 29245.39
    },
    {
      "epoch": 0.7076405274575581,
      "eval_loss": 0.3405376076698303,
      "eval_runtime": 8.4222,
      "eval_samples_per_second": 473.509,
      "eval_steps_per_second": 3.799,
      "num_input_tokens_seen": 5687337600,
      "step": 20000
    },
    {
      "epoch": 0.7079943477212869,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8863184306497437e-05,
      "loss": 0.40539979934692383,
      "num_input_tokens_seen": 5690213376,
      "step": 20010,
      "train_runtime": 194599.4527,
      "train_tokens_per_second": 29240.644
    },
    {
      "epoch": 0.7083481679850158,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8858376420083848e-05,
      "loss": 0.4044796466827393,
      "num_input_tokens_seen": 5693051008,
      "step": 20020,
      "train_runtime": 194694.6999,
      "train_tokens_per_second": 29240.914
    },
    {
      "epoch": 0.7087019882487445,
      "grad_norm": 0.58203125,
      "learning_rate": 2.8853570935491873e-05,
      "loss": 0.40385894775390624,
      "num_input_tokens_seen": 5695909696,
      "step": 20030,
      "train_runtime": 194791.9793,
      "train_tokens_per_second": 29240.987
    },
    {
      "epoch": 0.7090558085124733,
      "grad_norm": 0.578125,
      "learning_rate": 2.8848767850722447e-05,
      "loss": 0.40356879234313964,
      "num_input_tokens_seen": 5698707008,
      "step": 20040,
      "train_runtime": 194884.5988,
      "train_tokens_per_second": 29241.444
    },
    {
      "epoch": 0.7094096287762021,
      "grad_norm": 0.61328125,
      "learning_rate": 2.88439671637788e-05,
      "loss": 0.4030400276184082,
      "num_input_tokens_seen": 5701546880,
      "step": 20050,
      "train_runtime": 194978.3596,
      "train_tokens_per_second": 29241.947
    },
    {
      "epoch": 0.7097634490399308,
      "grad_norm": 0.578125,
      "learning_rate": 2.8839168872666505e-05,
      "loss": 0.39932711124420167,
      "num_input_tokens_seen": 5704427968,
      "step": 20060,
      "train_runtime": 195077.5007,
      "train_tokens_per_second": 29241.855
    },
    {
      "epoch": 0.7101172693036596,
      "grad_norm": 0.59375,
      "learning_rate": 2.8834372975393448e-05,
      "loss": 0.40181674957275393,
      "num_input_tokens_seen": 5707261696,
      "step": 20070,
      "train_runtime": 195176.1073,
      "train_tokens_per_second": 29241.6
    },
    {
      "epoch": 0.7104710895673884,
      "grad_norm": 0.59765625,
      "learning_rate": 2.8829579469969843e-05,
      "loss": 0.4052821159362793,
      "num_input_tokens_seen": 5710143360,
      "step": 20080,
      "train_runtime": 195276.995,
      "train_tokens_per_second": 29241.25
    },
    {
      "epoch": 0.7108249098311171,
      "grad_norm": 0.59375,
      "learning_rate": 2.88247883544082e-05,
      "loss": 0.39843552112579345,
      "num_input_tokens_seen": 5712987712,
      "step": 20090,
      "train_runtime": 195372.1321,
      "train_tokens_per_second": 29241.569
    },
    {
      "epoch": 0.7111787300948459,
      "grad_norm": 0.609375,
      "learning_rate": 2.8819999626723366e-05,
      "loss": 0.39776394367218015,
      "num_input_tokens_seen": 5715797888,
      "step": 20100,
      "train_runtime": 195469.6609,
      "train_tokens_per_second": 29241.356
    },
    {
      "epoch": 0.7115325503585748,
      "grad_norm": 0.58984375,
      "learning_rate": 2.881521328493248e-05,
      "loss": 0.4054411888122559,
      "num_input_tokens_seen": 5718698624,
      "step": 20110,
      "train_runtime": 195572.3946,
      "train_tokens_per_second": 29240.827
    },
    {
      "epoch": 0.7118863706223035,
      "grad_norm": 0.58203125,
      "learning_rate": 2.881042932705499e-05,
      "loss": 0.4002941608428955,
      "num_input_tokens_seen": 5721542016,
      "step": 20120,
      "train_runtime": 195670.7542,
      "train_tokens_per_second": 29240.66
    },
    {
      "epoch": 0.7122401908860323,
      "grad_norm": 0.5546875,
      "learning_rate": 2.8805647751112635e-05,
      "loss": 0.40181598663330076,
      "num_input_tokens_seen": 5724366592,
      "step": 20130,
      "train_runtime": 195766.4846,
      "train_tokens_per_second": 29240.79
    },
    {
      "epoch": 0.7125940111497611,
      "grad_norm": 0.60546875,
      "learning_rate": 2.8800868555129462e-05,
      "loss": 0.4026060104370117,
      "num_input_tokens_seen": 5727242496,
      "step": 20140,
      "train_runtime": 195865.8114,
      "train_tokens_per_second": 29240.644
    },
    {
      "epoch": 0.7129478314134898,
      "grad_norm": 0.609375,
      "learning_rate": 2.8796091737131814e-05,
      "loss": 0.40285167694091795,
      "num_input_tokens_seen": 5730071552,
      "step": 20150,
      "train_runtime": 195966.3893,
      "train_tokens_per_second": 29240.073
    },
    {
      "epoch": 0.7133016516772186,
      "grad_norm": 0.55078125,
      "learning_rate": 2.8791317295148322e-05,
      "loss": 0.4056499481201172,
      "num_input_tokens_seen": 5732937792,
      "step": 20160,
      "train_runtime": 196064.4829,
      "train_tokens_per_second": 29240.063
    },
    {
      "epoch": 0.7136554719409474,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8786545227209893e-05,
      "loss": 0.39931330680847166,
      "num_input_tokens_seen": 5735873920,
      "step": 20170,
      "train_runtime": 196167.0414,
      "train_tokens_per_second": 29239.743
    },
    {
      "epoch": 0.7140092922046761,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8781775531349742e-05,
      "loss": 0.40167598724365233,
      "num_input_tokens_seen": 5738691328,
      "step": 20180,
      "train_runtime": 196263.5948,
      "train_tokens_per_second": 29239.714
    },
    {
      "epoch": 0.714363112468405,
      "grad_norm": 0.5625,
      "learning_rate": 2.8777008205603334e-05,
      "loss": 0.4030892372131348,
      "num_input_tokens_seen": 5741575872,
      "step": 20190,
      "train_runtime": 196362.2094,
      "train_tokens_per_second": 29239.719
    },
    {
      "epoch": 0.7147169327321338,
      "grad_norm": 0.609375,
      "learning_rate": 2.877224324800844e-05,
      "loss": 0.4055344581604004,
      "num_input_tokens_seen": 5744369344,
      "step": 20200,
      "train_runtime": 196457.1184,
      "train_tokens_per_second": 29239.813
    },
    {
      "epoch": 0.7150707529958625,
      "grad_norm": 0.609375,
      "learning_rate": 2.8767480656605078e-05,
      "loss": 0.40584325790405273,
      "num_input_tokens_seen": 5747209984,
      "step": 20210,
      "train_runtime": 196552.4578,
      "train_tokens_per_second": 29240.082
    },
    {
      "epoch": 0.7154245732595913,
      "grad_norm": 0.60546875,
      "learning_rate": 2.8762720429435556e-05,
      "loss": 0.4035944938659668,
      "num_input_tokens_seen": 5750020864,
      "step": 20220,
      "train_runtime": 196647.9429,
      "train_tokens_per_second": 29240.178
    },
    {
      "epoch": 0.7157783935233201,
      "grad_norm": 0.61328125,
      "learning_rate": 2.8757962564544448e-05,
      "loss": 0.40030131340026853,
      "num_input_tokens_seen": 5752873536,
      "step": 20230,
      "train_runtime": 196745.7264,
      "train_tokens_per_second": 29240.145
    },
    {
      "epoch": 0.7161322137870488,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8753207059978586e-05,
      "loss": 0.40159144401550295,
      "num_input_tokens_seen": 5755677440,
      "step": 20240,
      "train_runtime": 196840.2367,
      "train_tokens_per_second": 29240.35
    },
    {
      "epoch": 0.7164860340507776,
      "grad_norm": 0.56640625,
      "learning_rate": 2.874845391378706e-05,
      "loss": 0.40175318717956543,
      "num_input_tokens_seen": 5758570688,
      "step": 20250,
      "train_runtime": 196940.1201,
      "train_tokens_per_second": 29240.211
    },
    {
      "epoch": 0.7168398543145064,
      "grad_norm": 0.640625,
      "learning_rate": 2.874370312402121e-05,
      "loss": 0.40808658599853515,
      "num_input_tokens_seen": 5761387904,
      "step": 20260,
      "train_runtime": 197034.8403,
      "train_tokens_per_second": 29240.453
    },
    {
      "epoch": 0.7171936745782352,
      "grad_norm": 0.60546875,
      "learning_rate": 2.8738954688734647e-05,
      "loss": 0.4072543144226074,
      "num_input_tokens_seen": 5764247104,
      "step": 20270,
      "train_runtime": 197135.1515,
      "train_tokens_per_second": 29240.077
    },
    {
      "epoch": 0.717547494841964,
      "grad_norm": 0.56640625,
      "learning_rate": 2.8734208605983226e-05,
      "loss": 0.40230717658996584,
      "num_input_tokens_seen": 5767154432,
      "step": 20280,
      "train_runtime": 197237.9037,
      "train_tokens_per_second": 29239.585
    },
    {
      "epoch": 0.7179013151056928,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8729464873825035e-05,
      "loss": 0.3948399305343628,
      "num_input_tokens_seen": 5769997696,
      "step": 20290,
      "train_runtime": 197338.0873,
      "train_tokens_per_second": 29239.149
    },
    {
      "epoch": 0.7182551353694215,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8724723490320427e-05,
      "loss": 0.401629638671875,
      "num_input_tokens_seen": 5772786240,
      "step": 20300,
      "train_runtime": 197432.4498,
      "train_tokens_per_second": 29239.298
    },
    {
      "epoch": 0.7186089556331503,
      "grad_norm": 0.578125,
      "learning_rate": 2.8719984453531983e-05,
      "loss": 0.4044965267181396,
      "num_input_tokens_seen": 5775593920,
      "step": 20310,
      "train_runtime": 197526.0038,
      "train_tokens_per_second": 29239.664
    },
    {
      "epoch": 0.7189627758968791,
      "grad_norm": 0.5859375,
      "learning_rate": 2.871524776152452e-05,
      "loss": 0.40209026336669923,
      "num_input_tokens_seen": 5778514880,
      "step": 20320,
      "train_runtime": 197629.6449,
      "train_tokens_per_second": 29239.11
    },
    {
      "epoch": 0.7193165961606078,
      "grad_norm": 0.60546875,
      "learning_rate": 2.8710513412365093e-05,
      "loss": 0.4041339874267578,
      "num_input_tokens_seen": 5781344768,
      "step": 20330,
      "train_runtime": 197726.8976,
      "train_tokens_per_second": 29239.041
    },
    {
      "epoch": 0.7196704164243366,
      "grad_norm": 0.58984375,
      "learning_rate": 2.8705781404122977e-05,
      "loss": 0.40406408309936526,
      "num_input_tokens_seen": 5784193728,
      "step": 20340,
      "train_runtime": 197825.6613,
      "train_tokens_per_second": 29238.844
    },
    {
      "epoch": 0.7200242366880654,
      "grad_norm": 0.60546875,
      "learning_rate": 2.87010517348697e-05,
      "loss": 0.402649974822998,
      "num_input_tokens_seen": 5787004928,
      "step": 20350,
      "train_runtime": 197918.4293,
      "train_tokens_per_second": 29239.343
    },
    {
      "epoch": 0.7203780569517942,
      "grad_norm": 0.5546875,
      "learning_rate": 2.8696324402678976e-05,
      "loss": 0.402455472946167,
      "num_input_tokens_seen": 5789859136,
      "step": 20360,
      "train_runtime": 198017.0073,
      "train_tokens_per_second": 29239.201
    },
    {
      "epoch": 0.720731877215523,
      "grad_norm": 0.58984375,
      "learning_rate": 2.8691599405626764e-05,
      "loss": 0.3999408006668091,
      "num_input_tokens_seen": 5792650304,
      "step": 20370,
      "train_runtime": 198109.327,
      "train_tokens_per_second": 29239.665
    },
    {
      "epoch": 0.7210856974792518,
      "grad_norm": 0.56640625,
      "learning_rate": 2.8686876741791234e-05,
      "loss": 0.4015204429626465,
      "num_input_tokens_seen": 5795493184,
      "step": 20380,
      "train_runtime": 198204.4984,
      "train_tokens_per_second": 29239.968
    },
    {
      "epoch": 0.7214395177429805,
      "grad_norm": 0.609375,
      "learning_rate": 2.8682156409252775e-05,
      "loss": 0.4027289867401123,
      "num_input_tokens_seen": 5798369408,
      "step": 20390,
      "train_runtime": 198305.9255,
      "train_tokens_per_second": 29239.517
    },
    {
      "epoch": 0.7217933380067093,
      "grad_norm": 0.58203125,
      "learning_rate": 2.8677438406093976e-05,
      "loss": 0.40537252426147463,
      "num_input_tokens_seen": 5801295872,
      "step": 20400,
      "train_runtime": 198407.8089,
      "train_tokens_per_second": 29239.252
    },
    {
      "epoch": 0.7221471582704381,
      "grad_norm": 0.59375,
      "learning_rate": 2.867272273039964e-05,
      "loss": 0.4008030414581299,
      "num_input_tokens_seen": 5804118912,
      "step": 20410,
      "train_runtime": 198506.0513,
      "train_tokens_per_second": 29239.002
    },
    {
      "epoch": 0.7225009785341668,
      "grad_norm": 0.59375,
      "learning_rate": 2.8668009380256765e-05,
      "loss": 0.40213832855224607,
      "num_input_tokens_seen": 5806953088,
      "step": 20420,
      "train_runtime": 198603.0817,
      "train_tokens_per_second": 29238.988
    },
    {
      "epoch": 0.7228547987978956,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8663298353754574e-05,
      "loss": 0.40180349349975586,
      "num_input_tokens_seen": 5809839424,
      "step": 20430,
      "train_runtime": 198700.4143,
      "train_tokens_per_second": 29239.191
    },
    {
      "epoch": 0.7232086190616245,
      "grad_norm": 0.60546875,
      "learning_rate": 2.865858964898445e-05,
      "loss": 0.3995216369628906,
      "num_input_tokens_seen": 5812671488,
      "step": 20440,
      "train_runtime": 198798.6111,
      "train_tokens_per_second": 29238.994
    },
    {
      "epoch": 0.7235624393253532,
      "grad_norm": 0.578125,
      "learning_rate": 2.8653883264040004e-05,
      "loss": 0.40815205574035646,
      "num_input_tokens_seen": 5815518528,
      "step": 20450,
      "train_runtime": 198897.4057,
      "train_tokens_per_second": 29238.785
    },
    {
      "epoch": 0.723916259589082,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8649179197017017e-05,
      "loss": 0.39934215545654295,
      "num_input_tokens_seen": 5818345600,
      "step": 20460,
      "train_runtime": 198991.8652,
      "train_tokens_per_second": 29239.113
    },
    {
      "epoch": 0.7242700798528108,
      "grad_norm": 0.62109375,
      "learning_rate": 2.864447744601347e-05,
      "loss": 0.4011113166809082,
      "num_input_tokens_seen": 5821212608,
      "step": 20470,
      "train_runtime": 199092.5066,
      "train_tokens_per_second": 29238.733
    },
    {
      "epoch": 0.7246239001165395,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8639778009129507e-05,
      "loss": 0.40370659828186034,
      "num_input_tokens_seen": 5824051264,
      "step": 20480,
      "train_runtime": 199188.8059,
      "train_tokens_per_second": 29238.848
    },
    {
      "epoch": 0.7249777203802683,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8635080884467492e-05,
      "loss": 0.40807380676269533,
      "num_input_tokens_seen": 5826942080,
      "step": 20490,
      "train_runtime": 199289.2425,
      "train_tokens_per_second": 29238.618
    },
    {
      "epoch": 0.7253315406439971,
      "grad_norm": 0.609375,
      "learning_rate": 2.8630386070131927e-05,
      "loss": 0.4022520542144775,
      "num_input_tokens_seen": 5829822528,
      "step": 20500,
      "train_runtime": 199386.9549,
      "train_tokens_per_second": 29238.736
    },
    {
      "epoch": 0.7256853609077258,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8625693564229507e-05,
      "loss": 0.4057596206665039,
      "num_input_tokens_seen": 5832641408,
      "step": 20510,
      "train_runtime": 199483.4091,
      "train_tokens_per_second": 29238.729
    },
    {
      "epoch": 0.7260391811714547,
      "grad_norm": 0.5625,
      "learning_rate": 2.8621003364869102e-05,
      "loss": 0.4062325477600098,
      "num_input_tokens_seen": 5835486976,
      "step": 20520,
      "train_runtime": 199581.8938,
      "train_tokens_per_second": 29238.559
    },
    {
      "epoch": 0.7263930014351835,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8616315470161738e-05,
      "loss": 0.4033400535583496,
      "num_input_tokens_seen": 5838337664,
      "step": 20530,
      "train_runtime": 199676.2128,
      "train_tokens_per_second": 29239.024
    },
    {
      "epoch": 0.7267468216989122,
      "grad_norm": 0.58203125,
      "learning_rate": 2.8611629878220624e-05,
      "loss": 0.4002431869506836,
      "num_input_tokens_seen": 5841176896,
      "step": 20540,
      "train_runtime": 199770.084,
      "train_tokens_per_second": 29239.498
    },
    {
      "epoch": 0.727100641962641,
      "grad_norm": 0.59375,
      "learning_rate": 2.8606946587161116e-05,
      "loss": 0.40137500762939454,
      "num_input_tokens_seen": 5844045184,
      "step": 20550,
      "train_runtime": 199865.8017,
      "train_tokens_per_second": 29239.846
    },
    {
      "epoch": 0.7274544622263698,
      "grad_norm": 0.5859375,
      "learning_rate": 2.860226559510072e-05,
      "loss": 0.4063122749328613,
      "num_input_tokens_seen": 5846909568,
      "step": 20560,
      "train_runtime": 199962.8617,
      "train_tokens_per_second": 29239.977
    },
    {
      "epoch": 0.7278082824900985,
      "grad_norm": 0.58984375,
      "learning_rate": 2.859758690015913e-05,
      "loss": 0.401547908782959,
      "num_input_tokens_seen": 5849704832,
      "step": 20570,
      "train_runtime": 200059.2629,
      "train_tokens_per_second": 29239.86
    },
    {
      "epoch": 0.7281621027538273,
      "grad_norm": 0.59765625,
      "learning_rate": 2.8592910500458154e-05,
      "loss": 0.40147933959960935,
      "num_input_tokens_seen": 5852570432,
      "step": 20580,
      "train_runtime": 200154.6351,
      "train_tokens_per_second": 29240.244
    },
    {
      "epoch": 0.7285159230175561,
      "grad_norm": 0.5546875,
      "learning_rate": 2.858823639412178e-05,
      "loss": 0.40390429496765134,
      "num_input_tokens_seen": 5855476864,
      "step": 20590,
      "train_runtime": 200256.7957,
      "train_tokens_per_second": 29239.841
    },
    {
      "epoch": 0.7288697432812848,
      "grad_norm": 0.59375,
      "learning_rate": 2.858356457927613e-05,
      "loss": 0.40361733436584474,
      "num_input_tokens_seen": 5858295936,
      "step": 20600,
      "train_runtime": 200353.234,
      "train_tokens_per_second": 29239.837
    },
    {
      "epoch": 0.7292235635450137,
      "grad_norm": 0.56640625,
      "learning_rate": 2.857889505404946e-05,
      "loss": 0.4030409812927246,
      "num_input_tokens_seen": 5861133632,
      "step": 20610,
      "train_runtime": 200453.0145,
      "train_tokens_per_second": 29239.439
    },
    {
      "epoch": 0.7295773838087425,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8574227816572184e-05,
      "loss": 0.4005866050720215,
      "num_input_tokens_seen": 5863989568,
      "step": 20620,
      "train_runtime": 200552.1593,
      "train_tokens_per_second": 29239.224
    },
    {
      "epoch": 0.7299312040724713,
      "grad_norm": 0.6015625,
      "learning_rate": 2.856956286497684e-05,
      "loss": 0.4035504341125488,
      "num_input_tokens_seen": 5866803392,
      "step": 20630,
      "train_runtime": 200646.3106,
      "train_tokens_per_second": 29239.528
    },
    {
      "epoch": 0.7302850243362,
      "grad_norm": 0.578125,
      "learning_rate": 2.85649001973981e-05,
      "loss": 0.40082406997680664,
      "num_input_tokens_seen": 5869583616,
      "step": 20640,
      "train_runtime": 200741.3622,
      "train_tokens_per_second": 29239.533
    },
    {
      "epoch": 0.7306388445999288,
      "grad_norm": 0.5703125,
      "learning_rate": 2.8560239811972754e-05,
      "loss": 0.39939250946044924,
      "num_input_tokens_seen": 5872449152,
      "step": 20650,
      "train_runtime": 200838.8262,
      "train_tokens_per_second": 29239.611
    },
    {
      "epoch": 0.7309926648636575,
      "grad_norm": 0.59375,
      "learning_rate": 2.855558170683976e-05,
      "loss": 0.4068933963775635,
      "num_input_tokens_seen": 5875205056,
      "step": 20660,
      "train_runtime": 200929.6156,
      "train_tokens_per_second": 29240.115
    },
    {
      "epoch": 0.7313464851273863,
      "grad_norm": 0.5859375,
      "learning_rate": 2.855092588014017e-05,
      "loss": 0.4007720470428467,
      "num_input_tokens_seen": 5878038464,
      "step": 20670,
      "train_runtime": 201025.0608,
      "train_tokens_per_second": 29240.327
    },
    {
      "epoch": 0.7317003053911151,
      "grad_norm": 0.5859375,
      "learning_rate": 2.854627233001715e-05,
      "loss": 0.4012059211730957,
      "num_input_tokens_seen": 5880880256,
      "step": 20680,
      "train_runtime": 201122.0765,
      "train_tokens_per_second": 29240.352
    },
    {
      "epoch": 0.732054125654844,
      "grad_norm": 0.58984375,
      "learning_rate": 2.8541621054615992e-05,
      "loss": 0.40305156707763673,
      "num_input_tokens_seen": 5883739392,
      "step": 20690,
      "train_runtime": 201220.5033,
      "train_tokens_per_second": 29240.258
    },
    {
      "epoch": 0.7324079459185727,
      "grad_norm": 0.59375,
      "learning_rate": 2.8536972052084122e-05,
      "loss": 0.4060030460357666,
      "num_input_tokens_seen": 5886572544,
      "step": 20700,
      "train_runtime": 201316.2142,
      "train_tokens_per_second": 29240.429
    },
    {
      "epoch": 0.7327617661823015,
      "grad_norm": 0.57421875,
      "learning_rate": 2.853232532057104e-05,
      "loss": 0.4038423538208008,
      "num_input_tokens_seen": 5889413760,
      "step": 20710,
      "train_runtime": 201415.8851,
      "train_tokens_per_second": 29240.066
    },
    {
      "epoch": 0.7331155864460303,
      "grad_norm": 0.58984375,
      "learning_rate": 2.8527680858228395e-05,
      "loss": 0.39737353324890134,
      "num_input_tokens_seen": 5892251136,
      "step": 20720,
      "train_runtime": 201513.2588,
      "train_tokens_per_second": 29240.017
    },
    {
      "epoch": 0.733469406709759,
      "grad_norm": 0.578125,
      "learning_rate": 2.8523038663209906e-05,
      "loss": 0.40528287887573244,
      "num_input_tokens_seen": 5895028096,
      "step": 20730,
      "train_runtime": 201605.9507,
      "train_tokens_per_second": 29240.348
    },
    {
      "epoch": 0.7338232269734878,
      "grad_norm": 0.58984375,
      "learning_rate": 2.8518398733671427e-05,
      "loss": 0.4017350196838379,
      "num_input_tokens_seen": 5897904192,
      "step": 20740,
      "train_runtime": 201704.795,
      "train_tokens_per_second": 29240.278
    },
    {
      "epoch": 0.7341770472372166,
      "grad_norm": 0.578125,
      "learning_rate": 2.8513761067770877e-05,
      "loss": 0.40599832534790037,
      "num_input_tokens_seen": 5900780160,
      "step": 20750,
      "train_runtime": 201802.2408,
      "train_tokens_per_second": 29240.41
    },
    {
      "epoch": 0.7345308675009453,
      "grad_norm": 0.56640625,
      "learning_rate": 2.85091256636683e-05,
      "loss": 0.40285730361938477,
      "num_input_tokens_seen": 5903594944,
      "step": 20760,
      "train_runtime": 201897.7366,
      "train_tokens_per_second": 29240.521
    },
    {
      "epoch": 0.7348846877646742,
      "grad_norm": 0.5859375,
      "learning_rate": 2.850449251952582e-05,
      "loss": 0.402994441986084,
      "num_input_tokens_seen": 5906506368,
      "step": 20770,
      "train_runtime": 201998.167,
      "train_tokens_per_second": 29240.396
    },
    {
      "epoch": 0.735238508028403,
      "grad_norm": 0.59375,
      "learning_rate": 2.849986163350766e-05,
      "loss": 0.40232067108154296,
      "num_input_tokens_seen": 5909345856,
      "step": 20780,
      "train_runtime": 202096.4309,
      "train_tokens_per_second": 29240.229
    },
    {
      "epoch": 0.7355923282921317,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8495233003780103e-05,
      "loss": 0.40167899131774903,
      "num_input_tokens_seen": 5912204480,
      "step": 20790,
      "train_runtime": 202193.146,
      "train_tokens_per_second": 29240.38
    },
    {
      "epoch": 0.7359461485558605,
      "grad_norm": 0.58203125,
      "learning_rate": 2.8490606628511557e-05,
      "loss": 0.3991173267364502,
      "num_input_tokens_seen": 5915059072,
      "step": 20800,
      "train_runtime": 202291.8092,
      "train_tokens_per_second": 29240.23
    },
    {
      "epoch": 0.7362999688195893,
      "grad_norm": 0.5625,
      "learning_rate": 2.8485982505872476e-05,
      "loss": 0.4016124725341797,
      "num_input_tokens_seen": 5917924608,
      "step": 20810,
      "train_runtime": 202391.1153,
      "train_tokens_per_second": 29240.041
    },
    {
      "epoch": 0.736653789083318,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8481360634035415e-05,
      "loss": 0.4056361198425293,
      "num_input_tokens_seen": 5920770944,
      "step": 20820,
      "train_runtime": 202487.5751,
      "train_tokens_per_second": 29240.169
    },
    {
      "epoch": 0.7370076093470468,
      "grad_norm": 0.5859375,
      "learning_rate": 2.847674101117499e-05,
      "loss": 0.403035831451416,
      "num_input_tokens_seen": 5923643264,
      "step": 20830,
      "train_runtime": 202583.6837,
      "train_tokens_per_second": 29240.476
    },
    {
      "epoch": 0.7373614296107756,
      "grad_norm": 0.578125,
      "learning_rate": 2.8472123635467896e-05,
      "loss": 0.4044186592102051,
      "num_input_tokens_seen": 5926487104,
      "step": 20840,
      "train_runtime": 202681.0483,
      "train_tokens_per_second": 29240.46
    },
    {
      "epoch": 0.7377152498745044,
      "grad_norm": 0.5859375,
      "learning_rate": 2.846750850509289e-05,
      "loss": 0.40087318420410156,
      "num_input_tokens_seen": 5929334592,
      "step": 20850,
      "train_runtime": 202775.8194,
      "train_tokens_per_second": 29240.837
    },
    {
      "epoch": 0.7380690701382332,
      "grad_norm": 0.578125,
      "learning_rate": 2.8462895618230813e-05,
      "loss": 0.4043607711791992,
      "num_input_tokens_seen": 5932241536,
      "step": 20860,
      "train_runtime": 202875.6707,
      "train_tokens_per_second": 29240.774
    },
    {
      "epoch": 0.738422890401962,
      "grad_norm": 0.61328125,
      "learning_rate": 2.845828497306453e-05,
      "loss": 0.39979393482208253,
      "num_input_tokens_seen": 5935053632,
      "step": 20870,
      "train_runtime": 202972.5017,
      "train_tokens_per_second": 29240.678
    },
    {
      "epoch": 0.7387767106656907,
      "grad_norm": 0.58203125,
      "learning_rate": 2.8453676567779018e-05,
      "loss": 0.40284051895141604,
      "num_input_tokens_seen": 5937934784,
      "step": 20880,
      "train_runtime": 203069.6837,
      "train_tokens_per_second": 29240.873
    },
    {
      "epoch": 0.7391305309294195,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8449070400561267e-05,
      "loss": 0.40182785987854003,
      "num_input_tokens_seen": 5940800320,
      "step": 20890,
      "train_runtime": 203169.2245,
      "train_tokens_per_second": 29240.651
    },
    {
      "epoch": 0.7394843511931483,
      "grad_norm": 0.5703125,
      "learning_rate": 2.8444466469600343e-05,
      "loss": 0.40529537200927734,
      "num_input_tokens_seen": 5943635072,
      "step": 20900,
      "train_runtime": 203266.1073,
      "train_tokens_per_second": 29240.66
    },
    {
      "epoch": 0.739838171456877,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8439864773087354e-05,
      "loss": 0.40120744705200195,
      "num_input_tokens_seen": 5946503040,
      "step": 20910,
      "train_runtime": 203365.7783,
      "train_tokens_per_second": 29240.431
    },
    {
      "epoch": 0.7401919917206058,
      "grad_norm": 0.59375,
      "learning_rate": 2.8435265309215465e-05,
      "loss": 0.4001636505126953,
      "num_input_tokens_seen": 5949331712,
      "step": 20920,
      "train_runtime": 203462.3254,
      "train_tokens_per_second": 29240.459
    },
    {
      "epoch": 0.7405458119843346,
      "grad_norm": 0.54296875,
      "learning_rate": 2.843066807617987e-05,
      "loss": 0.4073075294494629,
      "num_input_tokens_seen": 5952235008,
      "step": 20930,
      "train_runtime": 203562.527,
      "train_tokens_per_second": 29240.328
    },
    {
      "epoch": 0.7408996322480634,
      "grad_norm": 0.59375,
      "learning_rate": 2.8426073072177827e-05,
      "loss": 0.4028350830078125,
      "num_input_tokens_seen": 5955103680,
      "step": 20940,
      "train_runtime": 203658.8896,
      "train_tokens_per_second": 29240.578
    },
    {
      "epoch": 0.7412534525117922,
      "grad_norm": 0.58203125,
      "learning_rate": 2.8421480295408616e-05,
      "loss": 0.40394368171691897,
      "num_input_tokens_seen": 5957967168,
      "step": 20950,
      "train_runtime": 203758.4437,
      "train_tokens_per_second": 29240.345
    },
    {
      "epoch": 0.741607272775521,
      "grad_norm": 0.59765625,
      "learning_rate": 2.841688974407355e-05,
      "loss": 0.403903865814209,
      "num_input_tokens_seen": 5960793536,
      "step": 20960,
      "train_runtime": 203855.9122,
      "train_tokens_per_second": 29240.229
    },
    {
      "epoch": 0.7419610930392497,
      "grad_norm": 0.59375,
      "learning_rate": 2.8412301416375984e-05,
      "loss": 0.4063393592834473,
      "num_input_tokens_seen": 5963631616,
      "step": 20970,
      "train_runtime": 203952.8107,
      "train_tokens_per_second": 29240.252
    },
    {
      "epoch": 0.7423149133029785,
      "grad_norm": 0.58984375,
      "learning_rate": 2.840771531052131e-05,
      "loss": 0.40227136611938474,
      "num_input_tokens_seen": 5966481920,
      "step": 20980,
      "train_runtime": 204053.3711,
      "train_tokens_per_second": 29239.811
    },
    {
      "epoch": 0.7426687335667073,
      "grad_norm": 0.54296875,
      "learning_rate": 2.840313142471694e-05,
      "loss": 0.39959182739257815,
      "num_input_tokens_seen": 5969371072,
      "step": 20990,
      "train_runtime": 204156.5419,
      "train_tokens_per_second": 29239.186
    },
    {
      "epoch": 0.743022553830436,
      "grad_norm": 0.5625,
      "learning_rate": 2.83985497571723e-05,
      "loss": 0.4067965030670166,
      "num_input_tokens_seen": 5972228032,
      "step": 21000,
      "train_runtime": 204252.475,
      "train_tokens_per_second": 29239.44
    },
    {
      "epoch": 0.7433763740941648,
      "grad_norm": 0.60546875,
      "learning_rate": 2.8393970306098843e-05,
      "loss": 0.4042391777038574,
      "num_input_tokens_seen": 5975085504,
      "step": 21010,
      "train_runtime": 204349.7381,
      "train_tokens_per_second": 29239.507
    },
    {
      "epoch": 0.7437301943578937,
      "grad_norm": 0.5625,
      "learning_rate": 2.8389393069710053e-05,
      "loss": 0.4100473403930664,
      "num_input_tokens_seen": 5977922240,
      "step": 21020,
      "train_runtime": 204448.1148,
      "train_tokens_per_second": 29239.312
    },
    {
      "epoch": 0.7440840146216224,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8384818046221418e-05,
      "loss": 0.40555534362792967,
      "num_input_tokens_seen": 5980802240,
      "step": 21030,
      "train_runtime": 204551.5903,
      "train_tokens_per_second": 29238.601
    },
    {
      "epoch": 0.7444378348853512,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8380245233850434e-05,
      "loss": 0.4050168037414551,
      "num_input_tokens_seen": 5983692864,
      "step": 21040,
      "train_runtime": 204650.7446,
      "train_tokens_per_second": 29238.559
    },
    {
      "epoch": 0.74479165514908,
      "grad_norm": 0.6171875,
      "learning_rate": 2.837567463081662e-05,
      "loss": 0.40543422698974607,
      "num_input_tokens_seen": 5986513920,
      "step": 21050,
      "train_runtime": 204747.3409,
      "train_tokens_per_second": 29238.543
    },
    {
      "epoch": 0.7451454754128087,
      "grad_norm": 0.5625,
      "learning_rate": 2.8371106235341477e-05,
      "loss": 0.3963713884353638,
      "num_input_tokens_seen": 5989338944,
      "step": 21060,
      "train_runtime": 204841.2029,
      "train_tokens_per_second": 29238.937
    },
    {
      "epoch": 0.7454992956765375,
      "grad_norm": 0.59765625,
      "learning_rate": 2.836654004564855e-05,
      "loss": 0.4016937255859375,
      "num_input_tokens_seen": 5992180160,
      "step": 21070,
      "train_runtime": 204938.0576,
      "train_tokens_per_second": 29238.982
    },
    {
      "epoch": 0.7458531159402663,
      "grad_norm": 0.6171875,
      "learning_rate": 2.8361976059963348e-05,
      "loss": 0.40276317596435546,
      "num_input_tokens_seen": 5995003712,
      "step": 21080,
      "train_runtime": 205032.5377,
      "train_tokens_per_second": 29239.28
    },
    {
      "epoch": 0.746206936203995,
      "grad_norm": 0.640625,
      "learning_rate": 2.835741427651339e-05,
      "loss": 0.4051965713500977,
      "num_input_tokens_seen": 5997859648,
      "step": 21090,
      "train_runtime": 205132.8705,
      "train_tokens_per_second": 29238.901
    },
    {
      "epoch": 0.7465607564677239,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8352854693528197e-05,
      "loss": 0.4022191047668457,
      "num_input_tokens_seen": 6000715712,
      "step": 21100,
      "train_runtime": 205231.0828,
      "train_tokens_per_second": 29238.825
    },
    {
      "epoch": 0.7469145767314527,
      "grad_norm": 0.59765625,
      "learning_rate": 2.8348297309239274e-05,
      "loss": 0.40061092376708984,
      "num_input_tokens_seen": 6003573504,
      "step": 21110,
      "train_runtime": 205326.9624,
      "train_tokens_per_second": 29239.09
    },
    {
      "epoch": 0.7472683969951814,
      "grad_norm": 0.58984375,
      "learning_rate": 2.834374212188012e-05,
      "loss": 0.40214147567749026,
      "num_input_tokens_seen": 6006483328,
      "step": 21120,
      "train_runtime": 205427.131,
      "train_tokens_per_second": 29238.997
    },
    {
      "epoch": 0.7476222172589102,
      "grad_norm": 0.578125,
      "learning_rate": 2.833918912968621e-05,
      "loss": 0.4019974708557129,
      "num_input_tokens_seen": 6009366272,
      "step": 21130,
      "train_runtime": 205527.2053,
      "train_tokens_per_second": 29238.787
    },
    {
      "epoch": 0.747976037522639,
      "grad_norm": 0.609375,
      "learning_rate": 2.833463833089502e-05,
      "loss": 0.4017859935760498,
      "num_input_tokens_seen": 6012217152,
      "step": 21140,
      "train_runtime": 205625.3154,
      "train_tokens_per_second": 29238.701
    },
    {
      "epoch": 0.7483298577863677,
      "grad_norm": 0.60546875,
      "learning_rate": 2.8330089723745985e-05,
      "loss": 0.4010505676269531,
      "num_input_tokens_seen": 6015078784,
      "step": 21150,
      "train_runtime": 205725.2724,
      "train_tokens_per_second": 29238.405
    },
    {
      "epoch": 0.7486836780500965,
      "grad_norm": 0.578125,
      "learning_rate": 2.8325543306480538e-05,
      "loss": 0.3991249084472656,
      "num_input_tokens_seen": 6017903040,
      "step": 21160,
      "train_runtime": 205822.6474,
      "train_tokens_per_second": 29238.294
    },
    {
      "epoch": 0.7490374983138253,
      "grad_norm": 0.55859375,
      "learning_rate": 2.832099907734208e-05,
      "loss": 0.4030435562133789,
      "num_input_tokens_seen": 6020737728,
      "step": 21170,
      "train_runtime": 205917.2135,
      "train_tokens_per_second": 29238.632
    },
    {
      "epoch": 0.749391318577554,
      "grad_norm": 0.57421875,
      "learning_rate": 2.831645703457598e-05,
      "loss": 0.4036706924438477,
      "num_input_tokens_seen": 6023592192,
      "step": 21180,
      "train_runtime": 206016.2433,
      "train_tokens_per_second": 29238.433
    },
    {
      "epoch": 0.7497451388412829,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8311917176429582e-05,
      "loss": 0.40347042083740237,
      "num_input_tokens_seen": 6026465792,
      "step": 21190,
      "train_runtime": 206114.2207,
      "train_tokens_per_second": 29238.476
    },
    {
      "epoch": 0.7500989591050117,
      "grad_norm": 0.578125,
      "learning_rate": 2.830737950115218e-05,
      "loss": 0.4009533405303955,
      "num_input_tokens_seen": 6029296384,
      "step": 21200,
      "train_runtime": 206208.1955,
      "train_tokens_per_second": 29238.879
    },
    {
      "epoch": 0.7504527793687404,
      "grad_norm": 0.5625,
      "learning_rate": 2.8302844006995062e-05,
      "loss": 0.40364627838134765,
      "num_input_tokens_seen": 6032204928,
      "step": 21210,
      "train_runtime": 206310.0667,
      "train_tokens_per_second": 29238.539
    },
    {
      "epoch": 0.7508065996324692,
      "grad_norm": 0.57421875,
      "learning_rate": 2.829831069221145e-05,
      "loss": 0.4018427848815918,
      "num_input_tokens_seen": 6035024960,
      "step": 21220,
      "train_runtime": 206403.1199,
      "train_tokens_per_second": 29239.02
    },
    {
      "epoch": 0.751160419896198,
      "grad_norm": 0.59765625,
      "learning_rate": 2.8293779555056532e-05,
      "loss": 0.40345458984375,
      "num_input_tokens_seen": 6037872512,
      "step": 21230,
      "train_runtime": 206503.7871,
      "train_tokens_per_second": 29238.556
    },
    {
      "epoch": 0.7515142401599267,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8289250593787454e-05,
      "loss": 0.40208120346069337,
      "num_input_tokens_seen": 6040666880,
      "step": 21240,
      "train_runtime": 206596.1275,
      "train_tokens_per_second": 29239.013
    },
    {
      "epoch": 0.7518680604236555,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8284723806663316e-05,
      "loss": 0.39830024242401124,
      "num_input_tokens_seen": 6043495232,
      "step": 21250,
      "train_runtime": 206693.3955,
      "train_tokens_per_second": 29238.937
    },
    {
      "epoch": 0.7522218806873843,
      "grad_norm": 0.578125,
      "learning_rate": 2.828019919194515e-05,
      "loss": 0.40095744132995603,
      "num_input_tokens_seen": 6046314368,
      "step": 21260,
      "train_runtime": 206791.1022,
      "train_tokens_per_second": 29238.755
    },
    {
      "epoch": 0.7525757009511131,
      "grad_norm": 0.58984375,
      "learning_rate": 2.827567674789597e-05,
      "loss": 0.40511088371276854,
      "num_input_tokens_seen": 6049228096,
      "step": 21270,
      "train_runtime": 206890.2838,
      "train_tokens_per_second": 29238.822
    },
    {
      "epoch": 0.7529295212148419,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8271156472780697e-05,
      "loss": 0.3994009017944336,
      "num_input_tokens_seen": 6052079168,
      "step": 21280,
      "train_runtime": 206991.0537,
      "train_tokens_per_second": 29238.361
    },
    {
      "epoch": 0.7532833414785707,
      "grad_norm": 0.58203125,
      "learning_rate": 2.826663836486621e-05,
      "loss": 0.4032484531402588,
      "num_input_tokens_seen": 6054953216,
      "step": 21290,
      "train_runtime": 207090.0406,
      "train_tokens_per_second": 29238.264
    },
    {
      "epoch": 0.7536371617422994,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8262122422421326e-05,
      "loss": 0.40679135322570803,
      "num_input_tokens_seen": 6057801984,
      "step": 21300,
      "train_runtime": 207185.9263,
      "train_tokens_per_second": 29238.482
    },
    {
      "epoch": 0.7539909820060282,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8257608643716797e-05,
      "loss": 0.40395288467407225,
      "num_input_tokens_seen": 6060640320,
      "step": 21310,
      "train_runtime": 207283.7265,
      "train_tokens_per_second": 29238.38
    },
    {
      "epoch": 0.754344802269757,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8253097027025312e-05,
      "loss": 0.4087357521057129,
      "num_input_tokens_seen": 6063462656,
      "step": 21320,
      "train_runtime": 207378.5905,
      "train_tokens_per_second": 29238.614
    },
    {
      "epoch": 0.7546986225334857,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8248587570621472e-05,
      "loss": 0.3998574256896973,
      "num_input_tokens_seen": 6066305152,
      "step": 21330,
      "train_runtime": 207475.2812,
      "train_tokens_per_second": 29238.689
    },
    {
      "epoch": 0.7550524427972145,
      "grad_norm": 0.58203125,
      "learning_rate": 2.824408027278182e-05,
      "loss": 0.4021791458129883,
      "num_input_tokens_seen": 6069158208,
      "step": 21340,
      "train_runtime": 207571.7768,
      "train_tokens_per_second": 29238.841
    },
    {
      "epoch": 0.7554062630609434,
      "grad_norm": 0.61328125,
      "learning_rate": 2.823957513178483e-05,
      "loss": 0.4011404037475586,
      "num_input_tokens_seen": 6071980864,
      "step": 21350,
      "train_runtime": 207665.9035,
      "train_tokens_per_second": 29239.181
    },
    {
      "epoch": 0.7557600833246721,
      "grad_norm": 0.5703125,
      "learning_rate": 2.8235072145910878e-05,
      "loss": 0.4065885543823242,
      "num_input_tokens_seen": 6074852992,
      "step": 21360,
      "train_runtime": 207765.9766,
      "train_tokens_per_second": 29238.921
    },
    {
      "epoch": 0.7561139035884009,
      "grad_norm": 0.58203125,
      "learning_rate": 2.8230571313442273e-05,
      "loss": 0.397721004486084,
      "num_input_tokens_seen": 6077675520,
      "step": 21370,
      "train_runtime": 207861.9349,
      "train_tokens_per_second": 29239.002
    },
    {
      "epoch": 0.7564677238521297,
      "grad_norm": 0.6015625,
      "learning_rate": 2.822607263266324e-05,
      "loss": 0.40192575454711915,
      "num_input_tokens_seen": 6080512448,
      "step": 21380,
      "train_runtime": 207956.6422,
      "train_tokens_per_second": 29239.328
    },
    {
      "epoch": 0.7568215441158584,
      "grad_norm": 0.5546875,
      "learning_rate": 2.822157610185991e-05,
      "loss": 0.3952006816864014,
      "num_input_tokens_seen": 6083380992,
      "step": 21390,
      "train_runtime": 208052.5695,
      "train_tokens_per_second": 29239.634
    },
    {
      "epoch": 0.7571753643795872,
      "grad_norm": 0.61328125,
      "learning_rate": 2.821708171932032e-05,
      "loss": 0.40285372734069824,
      "num_input_tokens_seen": 6086255040,
      "step": 21400,
      "train_runtime": 208151.89,
      "train_tokens_per_second": 29239.49
    },
    {
      "epoch": 0.757529184643316,
      "grad_norm": 0.56640625,
      "learning_rate": 2.821258948333444e-05,
      "loss": 0.4052440643310547,
      "num_input_tokens_seen": 6089119296,
      "step": 21410,
      "train_runtime": 208250.642,
      "train_tokens_per_second": 29239.378
    },
    {
      "epoch": 0.7578830049070447,
      "grad_norm": 0.60546875,
      "learning_rate": 2.8208099392194116e-05,
      "loss": 0.4063704967498779,
      "num_input_tokens_seen": 6091926080,
      "step": 21420,
      "train_runtime": 208346.0056,
      "train_tokens_per_second": 29239.467
    },
    {
      "epoch": 0.7582368251707735,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8203611444193113e-05,
      "loss": 0.40077948570251465,
      "num_input_tokens_seen": 6094779584,
      "step": 21430,
      "train_runtime": 208441.7447,
      "train_tokens_per_second": 29239.726
    },
    {
      "epoch": 0.7585906454345024,
      "grad_norm": 0.5703125,
      "learning_rate": 2.8199125637627084e-05,
      "loss": 0.3984208106994629,
      "num_input_tokens_seen": 6097571456,
      "step": 21440,
      "train_runtime": 208535.5667,
      "train_tokens_per_second": 29239.959
    },
    {
      "epoch": 0.7589444656982312,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8194641970793593e-05,
      "loss": 0.40079245567321775,
      "num_input_tokens_seen": 6100396672,
      "step": 21450,
      "train_runtime": 208630.3426,
      "train_tokens_per_second": 29240.218
    },
    {
      "epoch": 0.7592982859619599,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8190160441992092e-05,
      "loss": 0.40223073959350586,
      "num_input_tokens_seen": 6103225152,
      "step": 21460,
      "train_runtime": 208728.2877,
      "train_tokens_per_second": 29240.048
    },
    {
      "epoch": 0.7596521062256887,
      "grad_norm": 0.609375,
      "learning_rate": 2.818568104952392e-05,
      "loss": 0.40602335929870603,
      "num_input_tokens_seen": 6106011520,
      "step": 21470,
      "train_runtime": 208819.8816,
      "train_tokens_per_second": 29240.566
    },
    {
      "epoch": 0.7600059264894174,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8181203791692306e-05,
      "loss": 0.4030517578125,
      "num_input_tokens_seen": 6108848960,
      "step": 21480,
      "train_runtime": 208914.7528,
      "train_tokens_per_second": 29240.869
    },
    {
      "epoch": 0.7603597467531462,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8176728666802375e-05,
      "loss": 0.4021156311035156,
      "num_input_tokens_seen": 6111702336,
      "step": 21490,
      "train_runtime": 209012.9843,
      "train_tokens_per_second": 29240.778
    },
    {
      "epoch": 0.760713567016875,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8172255673161124e-05,
      "loss": 0.39834303855895997,
      "num_input_tokens_seen": 6114556160,
      "step": 21500,
      "train_runtime": 209110.1639,
      "train_tokens_per_second": 29240.837
    },
    {
      "epoch": 0.7610673872806037,
      "grad_norm": 0.578125,
      "learning_rate": 2.8167784809077436e-05,
      "loss": 0.4024876594543457,
      "num_input_tokens_seen": 6117416192,
      "step": 21510,
      "train_runtime": 209206.3248,
      "train_tokens_per_second": 29241.067
    },
    {
      "epoch": 0.7614212075443326,
      "grad_norm": 0.62109375,
      "learning_rate": 2.816331607286207e-05,
      "loss": 0.3980637788772583,
      "num_input_tokens_seen": 6120260224,
      "step": 21520,
      "train_runtime": 209303.9946,
      "train_tokens_per_second": 29241.01
    },
    {
      "epoch": 0.7617750278080614,
      "grad_norm": 0.5703125,
      "learning_rate": 2.815884946282766e-05,
      "loss": 0.4038565635681152,
      "num_input_tokens_seen": 6123067136,
      "step": 21530,
      "train_runtime": 209399.925,
      "train_tokens_per_second": 29241.019
    },
    {
      "epoch": 0.7621288480717902,
      "grad_norm": 0.62109375,
      "learning_rate": 2.815438497728871e-05,
      "loss": 0.40122556686401367,
      "num_input_tokens_seen": 6125931520,
      "step": 21540,
      "train_runtime": 209500.3182,
      "train_tokens_per_second": 29240.679
    },
    {
      "epoch": 0.7624826683355189,
      "grad_norm": 0.5859375,
      "learning_rate": 2.8149922614561607e-05,
      "loss": 0.40387563705444335,
      "num_input_tokens_seen": 6128745536,
      "step": 21550,
      "train_runtime": 209598.6607,
      "train_tokens_per_second": 29240.385
    },
    {
      "epoch": 0.7628364885992477,
      "grad_norm": 0.60546875,
      "learning_rate": 2.8145462372964588e-05,
      "loss": 0.40125017166137694,
      "num_input_tokens_seen": 6131573312,
      "step": 21560,
      "train_runtime": 209695.8369,
      "train_tokens_per_second": 29240.32
    },
    {
      "epoch": 0.7631903088629765,
      "grad_norm": 0.58984375,
      "learning_rate": 2.814100425081777e-05,
      "loss": 0.4010325908660889,
      "num_input_tokens_seen": 6134375040,
      "step": 21570,
      "train_runtime": 209791.1187,
      "train_tokens_per_second": 29240.394
    },
    {
      "epoch": 0.7635441291267052,
      "grad_norm": 0.59765625,
      "learning_rate": 2.8136548246443117e-05,
      "loss": 0.39951772689819337,
      "num_input_tokens_seen": 6137264960,
      "step": 21580,
      "train_runtime": 209892.3549,
      "train_tokens_per_second": 29240.06
    },
    {
      "epoch": 0.763897949390434,
      "grad_norm": 0.63671875,
      "learning_rate": 2.8132094358164468e-05,
      "loss": 0.40428409576416013,
      "num_input_tokens_seen": 6140138816,
      "step": 21590,
      "train_runtime": 209989.922,
      "train_tokens_per_second": 29240.16
    },
    {
      "epoch": 0.7642517696541629,
      "grad_norm": 0.5546875,
      "learning_rate": 2.8127642584307508e-05,
      "loss": 0.4010955810546875,
      "num_input_tokens_seen": 6142984064,
      "step": 21600,
      "train_runtime": 210088.3028,
      "train_tokens_per_second": 29240.01
    },
    {
      "epoch": 0.7646055899178916,
      "grad_norm": 0.59765625,
      "learning_rate": 2.8123192923199776e-05,
      "loss": 0.4032616138458252,
      "num_input_tokens_seen": 6145847488,
      "step": 21610,
      "train_runtime": 210185.2461,
      "train_tokens_per_second": 29240.147
    },
    {
      "epoch": 0.7649594101816204,
      "grad_norm": 0.60546875,
      "learning_rate": 2.8118745373170667e-05,
      "loss": 0.40500125885009763,
      "num_input_tokens_seen": 6148713792,
      "step": 21620,
      "train_runtime": 210283.9607,
      "train_tokens_per_second": 29240.051
    },
    {
      "epoch": 0.7653132304453492,
      "grad_norm": 0.58984375,
      "learning_rate": 2.8114299932551423e-05,
      "loss": 0.4008623600006104,
      "num_input_tokens_seen": 6151501696,
      "step": 21630,
      "train_runtime": 210376.6638,
      "train_tokens_per_second": 29240.419
    },
    {
      "epoch": 0.7656670507090779,
      "grad_norm": 0.6015625,
      "learning_rate": 2.8109856599675145e-05,
      "loss": 0.40099372863769533,
      "num_input_tokens_seen": 6154360064,
      "step": 21640,
      "train_runtime": 210476.1105,
      "train_tokens_per_second": 29240.183
    },
    {
      "epoch": 0.7660208709728067,
      "grad_norm": 0.58203125,
      "learning_rate": 2.8105415372876742e-05,
      "loss": 0.4013160228729248,
      "num_input_tokens_seen": 6157238080,
      "step": 21650,
      "train_runtime": 210575.7179,
      "train_tokens_per_second": 29240.019
    },
    {
      "epoch": 0.7663746912365355,
      "grad_norm": 0.578125,
      "learning_rate": 2.8100976250493e-05,
      "loss": 0.4019191741943359,
      "num_input_tokens_seen": 6160097728,
      "step": 21660,
      "train_runtime": 210676.1204,
      "train_tokens_per_second": 29239.658
    },
    {
      "epoch": 0.7667285115002642,
      "grad_norm": 0.59375,
      "learning_rate": 2.8096539230862527e-05,
      "loss": 0.4069089889526367,
      "num_input_tokens_seen": 6162970048,
      "step": 21670,
      "train_runtime": 210775.8123,
      "train_tokens_per_second": 29239.456
    },
    {
      "epoch": 0.767082331763993,
      "grad_norm": 0.58203125,
      "learning_rate": 2.809210431232578e-05,
      "loss": 0.40045604705810545,
      "num_input_tokens_seen": 6165827072,
      "step": 21680,
      "train_runtime": 210874.1769,
      "train_tokens_per_second": 29239.365
    },
    {
      "epoch": 0.7674361520277219,
      "grad_norm": 0.60546875,
      "learning_rate": 2.8087671493225025e-05,
      "loss": 0.4019759178161621,
      "num_input_tokens_seen": 6168667520,
      "step": 21690,
      "train_runtime": 210970.5362,
      "train_tokens_per_second": 29239.474
    },
    {
      "epoch": 0.7677899722914506,
      "grad_norm": 0.5859375,
      "learning_rate": 2.808324077190438e-05,
      "loss": 0.4014717102050781,
      "num_input_tokens_seen": 6171544384,
      "step": 21700,
      "train_runtime": 211070.563,
      "train_tokens_per_second": 29239.247
    },
    {
      "epoch": 0.7681437925551794,
      "grad_norm": 0.58203125,
      "learning_rate": 2.807881214670978e-05,
      "loss": 0.4000603675842285,
      "num_input_tokens_seen": 6174348288,
      "step": 21710,
      "train_runtime": 211164.9174,
      "train_tokens_per_second": 29239.461
    },
    {
      "epoch": 0.7684976128189082,
      "grad_norm": 0.609375,
      "learning_rate": 2.8074385615988996e-05,
      "loss": 0.403455924987793,
      "num_input_tokens_seen": 6177181760,
      "step": 21720,
      "train_runtime": 211261.3836,
      "train_tokens_per_second": 29239.521
    },
    {
      "epoch": 0.7688514330826369,
      "grad_norm": 0.578125,
      "learning_rate": 2.8069961178091602e-05,
      "loss": 0.4034730911254883,
      "num_input_tokens_seen": 6180022592,
      "step": 21730,
      "train_runtime": 211358.0672,
      "train_tokens_per_second": 29239.587
    },
    {
      "epoch": 0.7692052533463657,
      "grad_norm": 0.58984375,
      "learning_rate": 2.806553883136901e-05,
      "loss": 0.40419206619262693,
      "num_input_tokens_seen": 6182888576,
      "step": 21740,
      "train_runtime": 211457.1505,
      "train_tokens_per_second": 29239.44
    },
    {
      "epoch": 0.7695590736100945,
      "grad_norm": 0.58203125,
      "learning_rate": 2.806111857417445e-05,
      "loss": 0.4033034801483154,
      "num_input_tokens_seen": 6185731200,
      "step": 21750,
      "train_runtime": 211554.4256,
      "train_tokens_per_second": 29239.432
    },
    {
      "epoch": 0.7699128938738232,
      "grad_norm": 0.58984375,
      "learning_rate": 2.8056700404862957e-05,
      "loss": 0.40367727279663085,
      "num_input_tokens_seen": 6188538752,
      "step": 21760,
      "train_runtime": 211649.2799,
      "train_tokens_per_second": 29239.593
    },
    {
      "epoch": 0.7702667141375521,
      "grad_norm": 0.59375,
      "learning_rate": 2.8052284321791383e-05,
      "loss": 0.4004530429840088,
      "num_input_tokens_seen": 6191344448,
      "step": 21770,
      "train_runtime": 211742.7244,
      "train_tokens_per_second": 29239.939
    },
    {
      "epoch": 0.7706205344012809,
      "grad_norm": 0.58984375,
      "learning_rate": 2.8047870323318382e-05,
      "loss": 0.40040907859802244,
      "num_input_tokens_seen": 6194210752,
      "step": 21780,
      "train_runtime": 211841.9374,
      "train_tokens_per_second": 29239.776
    },
    {
      "epoch": 0.7709743546650096,
      "grad_norm": 0.5703125,
      "learning_rate": 2.8043458407804428e-05,
      "loss": 0.40522065162658694,
      "num_input_tokens_seen": 6197046656,
      "step": 21790,
      "train_runtime": 211939.2523,
      "train_tokens_per_second": 29239.731
    },
    {
      "epoch": 0.7713281749287384,
      "grad_norm": 0.59765625,
      "learning_rate": 2.8039048573611805e-05,
      "loss": 0.4014880180358887,
      "num_input_tokens_seen": 6199852800,
      "step": 21800,
      "train_runtime": 212034.7268,
      "train_tokens_per_second": 29239.799
    },
    {
      "epoch": 0.7716819951924672,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8034640819104574e-05,
      "loss": 0.400075626373291,
      "num_input_tokens_seen": 6202663808,
      "step": 21810,
      "train_runtime": 212129.1713,
      "train_tokens_per_second": 29240.032
    },
    {
      "epoch": 0.7720358154561959,
      "grad_norm": 0.59375,
      "learning_rate": 2.803023514264861e-05,
      "loss": 0.40038352012634276,
      "num_input_tokens_seen": 6205479296,
      "step": 21820,
      "train_runtime": 212227.6669,
      "train_tokens_per_second": 29239.728
    },
    {
      "epoch": 0.7723896357199247,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8025831542611596e-05,
      "loss": 0.40074548721313474,
      "num_input_tokens_seen": 6208357056,
      "step": 21830,
      "train_runtime": 212326.1394,
      "train_tokens_per_second": 29239.721
    },
    {
      "epoch": 0.7727434559836535,
      "grad_norm": 0.6171875,
      "learning_rate": 2.8021430017362994e-05,
      "loss": 0.40484304428100587,
      "num_input_tokens_seen": 6211215936,
      "step": 21840,
      "train_runtime": 212422.6916,
      "train_tokens_per_second": 29239.889
    },
    {
      "epoch": 0.7730972762473823,
      "grad_norm": 0.58203125,
      "learning_rate": 2.801703056527406e-05,
      "loss": 0.40180673599243166,
      "num_input_tokens_seen": 6214038720,
      "step": 21850,
      "train_runtime": 212517.8797,
      "train_tokens_per_second": 29240.075
    },
    {
      "epoch": 0.7734510965111111,
      "grad_norm": 0.58203125,
      "learning_rate": 2.801263318471784e-05,
      "loss": 0.40531578063964846,
      "num_input_tokens_seen": 6216856064,
      "step": 21860,
      "train_runtime": 212611.0702,
      "train_tokens_per_second": 29240.51
    },
    {
      "epoch": 0.7738049167748399,
      "grad_norm": 0.61328125,
      "learning_rate": 2.800823787406917e-05,
      "loss": 0.402863883972168,
      "num_input_tokens_seen": 6219774528,
      "step": 21870,
      "train_runtime": 212712.9938,
      "train_tokens_per_second": 29240.219
    },
    {
      "epoch": 0.7741587370385686,
      "grad_norm": 0.57421875,
      "learning_rate": 2.8003844631704678e-05,
      "loss": 0.4025244235992432,
      "num_input_tokens_seen": 6222680512,
      "step": 21880,
      "train_runtime": 212813.8948,
      "train_tokens_per_second": 29240.01
    },
    {
      "epoch": 0.7745125573022974,
      "grad_norm": 0.58203125,
      "learning_rate": 2.7999453456002755e-05,
      "loss": 0.40511693954467776,
      "num_input_tokens_seen": 6225528896,
      "step": 21890,
      "train_runtime": 212912.072,
      "train_tokens_per_second": 29239.906
    },
    {
      "epoch": 0.7748663775660262,
      "grad_norm": 0.6015625,
      "learning_rate": 2.799506434534359e-05,
      "loss": 0.40181965827941896,
      "num_input_tokens_seen": 6228382528,
      "step": 21900,
      "train_runtime": 213011.1105,
      "train_tokens_per_second": 29239.707
    },
    {
      "epoch": 0.7752201978297549,
      "grad_norm": 0.625,
      "learning_rate": 2.7990677298109136e-05,
      "loss": 0.40729293823242185,
      "num_input_tokens_seen": 6231212864,
      "step": 21910,
      "train_runtime": 213107.5718,
      "train_tokens_per_second": 29239.753
    },
    {
      "epoch": 0.7755740180934837,
      "grad_norm": 0.609375,
      "learning_rate": 2.7986292312683125e-05,
      "loss": 0.4004612922668457,
      "num_input_tokens_seen": 6233996096,
      "step": 21920,
      "train_runtime": 213199.4556,
      "train_tokens_per_second": 29240.206
    },
    {
      "epoch": 0.7759278383572125,
      "grad_norm": 0.59375,
      "learning_rate": 2.798190938745106e-05,
      "loss": 0.40383257865905764,
      "num_input_tokens_seen": 6236845568,
      "step": 21930,
      "train_runtime": 213294.8793,
      "train_tokens_per_second": 29240.484
    },
    {
      "epoch": 0.7762816586209413,
      "grad_norm": 0.578125,
      "learning_rate": 2.7977528520800218e-05,
      "loss": 0.4037925243377686,
      "num_input_tokens_seen": 6239686912,
      "step": 21940,
      "train_runtime": 213389.3728,
      "train_tokens_per_second": 29240.851
    },
    {
      "epoch": 0.7766354788846701,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7973149711119635e-05,
      "loss": 0.403082275390625,
      "num_input_tokens_seen": 6242501824,
      "step": 21950,
      "train_runtime": 213484.7204,
      "train_tokens_per_second": 29240.977
    },
    {
      "epoch": 0.7769892991483989,
      "grad_norm": 0.59765625,
      "learning_rate": 2.7968772956800115e-05,
      "loss": 0.3998307466506958,
      "num_input_tokens_seen": 6245323904,
      "step": 21960,
      "train_runtime": 213580.5955,
      "train_tokens_per_second": 29241.064
    },
    {
      "epoch": 0.7773431194121276,
      "grad_norm": 0.58203125,
      "learning_rate": 2.796439825623423e-05,
      "loss": 0.3972943305969238,
      "num_input_tokens_seen": 6248202496,
      "step": 21970,
      "train_runtime": 213682.8308,
      "train_tokens_per_second": 29240.545
    },
    {
      "epoch": 0.7776969396758564,
      "grad_norm": 0.5859375,
      "learning_rate": 2.79600256078163e-05,
      "loss": 0.40175352096557615,
      "num_input_tokens_seen": 6251015872,
      "step": 21980,
      "train_runtime": 213780.1149,
      "train_tokens_per_second": 29240.399
    },
    {
      "epoch": 0.7780507599395852,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7955655009942416e-05,
      "loss": 0.398640775680542,
      "num_input_tokens_seen": 6253897536,
      "step": 21990,
      "train_runtime": 213879.2381,
      "train_tokens_per_second": 29240.321
    },
    {
      "epoch": 0.7784045802033139,
      "grad_norm": 0.5625,
      "learning_rate": 2.79512864610104e-05,
      "loss": 0.3993380069732666,
      "num_input_tokens_seen": 6256718016,
      "step": 22000,
      "train_runtime": 213975.8102,
      "train_tokens_per_second": 29240.305
    },
    {
      "epoch": 0.7787584004670427,
      "grad_norm": 0.5859375,
      "learning_rate": 2.794691995941986e-05,
      "loss": 0.3973658800125122,
      "num_input_tokens_seen": 6259563648,
      "step": 22010,
      "train_runtime": 214072.9752,
      "train_tokens_per_second": 29240.326
    },
    {
      "epoch": 0.7791122207307716,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7942555503572115e-05,
      "loss": 0.40152807235717775,
      "num_input_tokens_seen": 6262393408,
      "step": 22020,
      "train_runtime": 214170.3637,
      "train_tokens_per_second": 29240.243
    },
    {
      "epoch": 0.7794660409945003,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7938193091870268e-05,
      "loss": 0.40468440055847166,
      "num_input_tokens_seen": 6265231040,
      "step": 22030,
      "train_runtime": 214267.0567,
      "train_tokens_per_second": 29240.291
    },
    {
      "epoch": 0.7798198612582291,
      "grad_norm": 0.5703125,
      "learning_rate": 2.793383272271914e-05,
      "loss": 0.40171175003051757,
      "num_input_tokens_seen": 6268078272,
      "step": 22040,
      "train_runtime": 214362.9292,
      "train_tokens_per_second": 29240.496
    },
    {
      "epoch": 0.7801736815219579,
      "grad_norm": 0.59375,
      "learning_rate": 2.7929474394525307e-05,
      "loss": 0.40299282073974607,
      "num_input_tokens_seen": 6270969152,
      "step": 22050,
      "train_runtime": 214463.4114,
      "train_tokens_per_second": 29240.275
    },
    {
      "epoch": 0.7805275017856866,
      "grad_norm": 0.56640625,
      "learning_rate": 2.7925118105697078e-05,
      "loss": 0.40417985916137694,
      "num_input_tokens_seen": 6273804032,
      "step": 22060,
      "train_runtime": 214559.7929,
      "train_tokens_per_second": 29240.353
    },
    {
      "epoch": 0.7808813220494154,
      "grad_norm": 0.58984375,
      "learning_rate": 2.792076385464451e-05,
      "loss": 0.4009377479553223,
      "num_input_tokens_seen": 6276699200,
      "step": 22070,
      "train_runtime": 214659.8802,
      "train_tokens_per_second": 29240.206
    },
    {
      "epoch": 0.7812351423131442,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7916411639779383e-05,
      "loss": 0.4046659469604492,
      "num_input_tokens_seen": 6279558528,
      "step": 22080,
      "train_runtime": 214760.4374,
      "train_tokens_per_second": 29239.829
    },
    {
      "epoch": 0.7815889625768729,
      "grad_norm": 0.6171875,
      "learning_rate": 2.7912061459515215e-05,
      "loss": 0.4031815052032471,
      "num_input_tokens_seen": 6282408320,
      "step": 22090,
      "train_runtime": 214860.6102,
      "train_tokens_per_second": 29239.46
    },
    {
      "epoch": 0.7819427828406018,
      "grad_norm": 0.6015625,
      "learning_rate": 2.7907713312267247e-05,
      "loss": 0.40269718170166013,
      "num_input_tokens_seen": 6285253888,
      "step": 22100,
      "train_runtime": 214958.1226,
      "train_tokens_per_second": 29239.434
    },
    {
      "epoch": 0.7822966031043306,
      "grad_norm": 0.578125,
      "learning_rate": 2.7903367196452457e-05,
      "loss": 0.40325264930725097,
      "num_input_tokens_seen": 6288175424,
      "step": 22110,
      "train_runtime": 215058.0007,
      "train_tokens_per_second": 29239.44
    },
    {
      "epoch": 0.7826504233680593,
      "grad_norm": 0.578125,
      "learning_rate": 2.7899023110489554e-05,
      "loss": 0.4025906562805176,
      "num_input_tokens_seen": 6291064832,
      "step": 22120,
      "train_runtime": 215156.3358,
      "train_tokens_per_second": 29239.505
    },
    {
      "epoch": 0.7830042436317881,
      "grad_norm": 0.59375,
      "learning_rate": 2.789468105279895e-05,
      "loss": 0.39742507934570315,
      "num_input_tokens_seen": 6293907776,
      "step": 22130,
      "train_runtime": 215252.8934,
      "train_tokens_per_second": 29239.597
    },
    {
      "epoch": 0.7833580638955169,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7890341021802786e-05,
      "loss": 0.4017388343811035,
      "num_input_tokens_seen": 6296731904,
      "step": 22140,
      "train_runtime": 215349.3687,
      "train_tokens_per_second": 29239.612
    },
    {
      "epoch": 0.7837118841592456,
      "grad_norm": 0.578125,
      "learning_rate": 2.7886003015924933e-05,
      "loss": 0.40320391654968263,
      "num_input_tokens_seen": 6299597952,
      "step": 22150,
      "train_runtime": 215447.1514,
      "train_tokens_per_second": 29239.644
    },
    {
      "epoch": 0.7840657044229744,
      "grad_norm": 0.5703125,
      "learning_rate": 2.788166703359096e-05,
      "loss": 0.40161876678466796,
      "num_input_tokens_seen": 6302420160,
      "step": 22160,
      "train_runtime": 215541.2025,
      "train_tokens_per_second": 29239.979
    },
    {
      "epoch": 0.7844195246867032,
      "grad_norm": 0.57421875,
      "learning_rate": 2.787733307322816e-05,
      "loss": 0.4034140586853027,
      "num_input_tokens_seen": 6305252352,
      "step": 22170,
      "train_runtime": 215636.4358,
      "train_tokens_per_second": 29240.199
    },
    {
      "epoch": 0.7847733449504319,
      "grad_norm": 0.57421875,
      "learning_rate": 2.7873001133265518e-05,
      "loss": 0.4045859813690186,
      "num_input_tokens_seen": 6308094464,
      "step": 22180,
      "train_runtime": 215733.5725,
      "train_tokens_per_second": 29240.208
    },
    {
      "epoch": 0.7851271652141608,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7868671212133768e-05,
      "loss": 0.4012325286865234,
      "num_input_tokens_seen": 6311009408,
      "step": 22190,
      "train_runtime": 215835.8427,
      "train_tokens_per_second": 29239.858
    },
    {
      "epoch": 0.7854809854778896,
      "grad_norm": 0.546875,
      "learning_rate": 2.7864343308265313e-05,
      "loss": 0.3999070882797241,
      "num_input_tokens_seen": 6313870208,
      "step": 22200,
      "train_runtime": 215932.385,
      "train_tokens_per_second": 29240.034
    },
    {
      "epoch": 0.7858348057416183,
      "grad_norm": 0.5625,
      "learning_rate": 2.7860017420094258e-05,
      "loss": 0.40146970748901367,
      "num_input_tokens_seen": 6316758400,
      "step": 22210,
      "train_runtime": 216031.218,
      "train_tokens_per_second": 29240.026
    },
    {
      "epoch": 0.7861886260053471,
      "grad_norm": 0.59765625,
      "learning_rate": 2.7855693546056445e-05,
      "loss": 0.3986906766891479,
      "num_input_tokens_seen": 6319635136,
      "step": 22220,
      "train_runtime": 216130.7995,
      "train_tokens_per_second": 29239.864
    },
    {
      "epoch": 0.7865424462690759,
      "grad_norm": 0.59765625,
      "learning_rate": 2.785137168458937e-05,
      "loss": 0.4019612312316895,
      "num_input_tokens_seen": 6322477056,
      "step": 22230,
      "train_runtime": 216225.1908,
      "train_tokens_per_second": 29240.243
    },
    {
      "epoch": 0.7868962665328046,
      "grad_norm": 0.57421875,
      "learning_rate": 2.7847051834132265e-05,
      "loss": 0.4017789840698242,
      "num_input_tokens_seen": 6325317824,
      "step": 22240,
      "train_runtime": 216322.4894,
      "train_tokens_per_second": 29240.223
    },
    {
      "epoch": 0.7872500867965334,
      "grad_norm": 0.57421875,
      "learning_rate": 2.784273399312603e-05,
      "loss": 0.4004362106323242,
      "num_input_tokens_seen": 6328111424,
      "step": 22250,
      "train_runtime": 216415.653,
      "train_tokens_per_second": 29240.544
    },
    {
      "epoch": 0.7876039070602622,
      "grad_norm": 0.57421875,
      "learning_rate": 2.783841816001326e-05,
      "loss": 0.40133142471313477,
      "num_input_tokens_seen": 6330975616,
      "step": 22260,
      "train_runtime": 216514.8248,
      "train_tokens_per_second": 29240.379
    },
    {
      "epoch": 0.787957727323991,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7834104333238255e-05,
      "loss": 0.40468730926513674,
      "num_input_tokens_seen": 6333824768,
      "step": 22270,
      "train_runtime": 216610.8411,
      "train_tokens_per_second": 29240.571
    },
    {
      "epoch": 0.7883115475877198,
      "grad_norm": 0.5625,
      "learning_rate": 2.7829792511246987e-05,
      "loss": 0.3991037368774414,
      "num_input_tokens_seen": 6336694208,
      "step": 22280,
      "train_runtime": 216710.6825,
      "train_tokens_per_second": 29240.341
    },
    {
      "epoch": 0.7886653678514486,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7825482692487114e-05,
      "loss": 0.40187225341796873,
      "num_input_tokens_seen": 6339521664,
      "step": 22290,
      "train_runtime": 216806.6155,
      "train_tokens_per_second": 29240.444
    },
    {
      "epoch": 0.7890191881151774,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7821174875407985e-05,
      "loss": 0.3983335494995117,
      "num_input_tokens_seen": 6342405632,
      "step": 22300,
      "train_runtime": 216908.0163,
      "train_tokens_per_second": 29240.07
    },
    {
      "epoch": 0.7893730083789061,
      "grad_norm": 0.6015625,
      "learning_rate": 2.781686905846062e-05,
      "loss": 0.397584056854248,
      "num_input_tokens_seen": 6345213056,
      "step": 22310,
      "train_runtime": 217003.4208,
      "train_tokens_per_second": 29240.152
    },
    {
      "epoch": 0.7897268286426349,
      "grad_norm": 0.57421875,
      "learning_rate": 2.7812565240097716e-05,
      "loss": 0.3988661289215088,
      "num_input_tokens_seen": 6348006400,
      "step": 22320,
      "train_runtime": 217097.2848,
      "train_tokens_per_second": 29240.377
    },
    {
      "epoch": 0.7900806489063636,
      "grad_norm": 0.5625,
      "learning_rate": 2.7808263418773656e-05,
      "loss": 0.4023624897003174,
      "num_input_tokens_seen": 6350875264,
      "step": 22330,
      "train_runtime": 217194.5227,
      "train_tokens_per_second": 29240.495
    },
    {
      "epoch": 0.7904344691700924,
      "grad_norm": 0.59375,
      "learning_rate": 2.7803963592944482e-05,
      "loss": 0.3968529462814331,
      "num_input_tokens_seen": 6353698624,
      "step": 22340,
      "train_runtime": 217289.9557,
      "train_tokens_per_second": 29240.646
    },
    {
      "epoch": 0.7907882894338213,
      "grad_norm": 0.59765625,
      "learning_rate": 2.779966576106792e-05,
      "loss": 0.40396957397460936,
      "num_input_tokens_seen": 6356481472,
      "step": 22350,
      "train_runtime": 217382.8834,
      "train_tokens_per_second": 29240.947
    },
    {
      "epoch": 0.7911421096975501,
      "grad_norm": 0.58984375,
      "learning_rate": 2.779536992160336e-05,
      "loss": 0.401324987411499,
      "num_input_tokens_seen": 6359324096,
      "step": 22360,
      "train_runtime": 217480.7217,
      "train_tokens_per_second": 29240.863
    },
    {
      "epoch": 0.7914959299612788,
      "grad_norm": 0.609375,
      "learning_rate": 2.7791076073011845e-05,
      "loss": 0.4039212703704834,
      "num_input_tokens_seen": 6362175808,
      "step": 22370,
      "train_runtime": 217579.1175,
      "train_tokens_per_second": 29240.746
    },
    {
      "epoch": 0.7918497502250076,
      "grad_norm": 0.609375,
      "learning_rate": 2.7786784213756102e-05,
      "loss": 0.4011545181274414,
      "num_input_tokens_seen": 6365002496,
      "step": 22380,
      "train_runtime": 217676.1372,
      "train_tokens_per_second": 29240.699
    },
    {
      "epoch": 0.7922035704887364,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7782494342300502e-05,
      "loss": 0.4010001659393311,
      "num_input_tokens_seen": 6367800384,
      "step": 22390,
      "train_runtime": 217766.7573,
      "train_tokens_per_second": 29241.379
    },
    {
      "epoch": 0.7925573907524651,
      "grad_norm": 0.5859375,
      "learning_rate": 2.777820645711109e-05,
      "loss": 0.4051644325256348,
      "num_input_tokens_seen": 6370698624,
      "step": 22400,
      "train_runtime": 217866.8912,
      "train_tokens_per_second": 29241.243
    },
    {
      "epoch": 0.7929112110161939,
      "grad_norm": 0.5859375,
      "learning_rate": 2.777392055665556e-05,
      "loss": 0.40039501190185545,
      "num_input_tokens_seen": 6373549824,
      "step": 22410,
      "train_runtime": 217964.9791,
      "train_tokens_per_second": 29241.165
    },
    {
      "epoch": 0.7932650312799226,
      "grad_norm": 0.58203125,
      "learning_rate": 2.7769636639403252e-05,
      "loss": 0.400991153717041,
      "num_input_tokens_seen": 6376342400,
      "step": 22420,
      "train_runtime": 218059.1645,
      "train_tokens_per_second": 29241.341
    },
    {
      "epoch": 0.7936188515436514,
      "grad_norm": 0.59375,
      "learning_rate": 2.776535470382518e-05,
      "loss": 0.40067543983459475,
      "num_input_tokens_seen": 6379187776,
      "step": 22430,
      "train_runtime": 218157.2092,
      "train_tokens_per_second": 29241.242
    },
    {
      "epoch": 0.7939726718073803,
      "grad_norm": 0.57421875,
      "learning_rate": 2.776107474839399e-05,
      "loss": 0.40239601135253905,
      "num_input_tokens_seen": 6382054144,
      "step": 22440,
      "train_runtime": 218257.6177,
      "train_tokens_per_second": 29240.923
    },
    {
      "epoch": 0.7943264920711091,
      "grad_norm": 0.5703125,
      "learning_rate": 2.7756796771583976e-05,
      "loss": 0.39692723751068115,
      "num_input_tokens_seen": 6384900416,
      "step": 22450,
      "train_runtime": 218353.4945,
      "train_tokens_per_second": 29241.119
    },
    {
      "epoch": 0.7946803123348378,
      "grad_norm": 0.59765625,
      "learning_rate": 2.775252077187109e-05,
      "loss": 0.39790935516357423,
      "num_input_tokens_seen": 6387747456,
      "step": 22460,
      "train_runtime": 218450.358,
      "train_tokens_per_second": 29241.186
    },
    {
      "epoch": 0.7950341325985666,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7748246747732902e-05,
      "loss": 0.4012287616729736,
      "num_input_tokens_seen": 6390513600,
      "step": 22470,
      "train_runtime": 218544.5121,
      "train_tokens_per_second": 29241.245
    },
    {
      "epoch": 0.7953879528622954,
      "grad_norm": 0.59765625,
      "learning_rate": 2.774397469764866e-05,
      "loss": 0.4026956558227539,
      "num_input_tokens_seen": 6393333760,
      "step": 22480,
      "train_runtime": 218639.1148,
      "train_tokens_per_second": 29241.491
    },
    {
      "epoch": 0.7957417731260241,
      "grad_norm": 0.58203125,
      "learning_rate": 2.7739704620099226e-05,
      "loss": 0.40224671363830566,
      "num_input_tokens_seen": 6396174656,
      "step": 22490,
      "train_runtime": 218738.023,
      "train_tokens_per_second": 29241.257
    },
    {
      "epoch": 0.7960955933897529,
      "grad_norm": 0.57421875,
      "learning_rate": 2.7735436513567098e-05,
      "loss": 0.3987330436706543,
      "num_input_tokens_seen": 6399009600,
      "step": 22500,
      "train_runtime": 218836.2403,
      "train_tokens_per_second": 29241.087
    },
    {
      "epoch": 0.7964494136534817,
      "grad_norm": 0.6015625,
      "learning_rate": 2.7731170376536402e-05,
      "loss": 0.40046215057373047,
      "num_input_tokens_seen": 6401880448,
      "step": 22510,
      "train_runtime": 218934.1744,
      "train_tokens_per_second": 29241.12
    },
    {
      "epoch": 0.7968032339172105,
      "grad_norm": 0.5625,
      "learning_rate": 2.7726906207492925e-05,
      "loss": 0.40197367668151857,
      "num_input_tokens_seen": 6404782144,
      "step": 22520,
      "train_runtime": 219035.2656,
      "train_tokens_per_second": 29240.872
    },
    {
      "epoch": 0.7971570541809393,
      "grad_norm": 0.59375,
      "learning_rate": 2.7722644004924048e-05,
      "loss": 0.4049616813659668,
      "num_input_tokens_seen": 6407612544,
      "step": 22530,
      "train_runtime": 219133.1562,
      "train_tokens_per_second": 29240.726
    },
    {
      "epoch": 0.7975108744446681,
      "grad_norm": 0.578125,
      "learning_rate": 2.77183837673188e-05,
      "loss": 0.3956486225128174,
      "num_input_tokens_seen": 6410450176,
      "step": 22540,
      "train_runtime": 219231.7534,
      "train_tokens_per_second": 29240.519
    },
    {
      "epoch": 0.7978646947083968,
      "grad_norm": 0.65625,
      "learning_rate": 2.7714125493167835e-05,
      "loss": 0.39937601089477537,
      "num_input_tokens_seen": 6413332416,
      "step": 22550,
      "train_runtime": 219333.9944,
      "train_tokens_per_second": 29240.029
    },
    {
      "epoch": 0.7982185149721256,
      "grad_norm": 0.59375,
      "learning_rate": 2.770986918096342e-05,
      "loss": 0.3993282079696655,
      "num_input_tokens_seen": 6416186432,
      "step": 22560,
      "train_runtime": 219432.4533,
      "train_tokens_per_second": 29239.916
    },
    {
      "epoch": 0.7985723352358544,
      "grad_norm": 0.61328125,
      "learning_rate": 2.7705614829199454e-05,
      "loss": 0.40130019187927246,
      "num_input_tokens_seen": 6418982848,
      "step": 22570,
      "train_runtime": 219527.2396,
      "train_tokens_per_second": 29240.029
    },
    {
      "epoch": 0.7989261554995831,
      "grad_norm": 0.5703125,
      "learning_rate": 2.7701362436371437e-05,
      "loss": 0.4002102851867676,
      "num_input_tokens_seen": 6421845056,
      "step": 22580,
      "train_runtime": 219626.3015,
      "train_tokens_per_second": 29239.873
    },
    {
      "epoch": 0.7992799757633119,
      "grad_norm": 0.59375,
      "learning_rate": 2.7697112000976492e-05,
      "loss": 0.3958927869796753,
      "num_input_tokens_seen": 6424694080,
      "step": 22590,
      "train_runtime": 219722.4644,
      "train_tokens_per_second": 29240.042
    },
    {
      "epoch": 0.7996337960270408,
      "grad_norm": 0.56640625,
      "learning_rate": 2.769286352151338e-05,
      "loss": 0.4024338722229004,
      "num_input_tokens_seen": 6427556096,
      "step": 22600,
      "train_runtime": 219819.9892,
      "train_tokens_per_second": 29240.089
    },
    {
      "epoch": 0.7999876162907695,
      "grad_norm": 0.5859375,
      "learning_rate": 2.768861699648243e-05,
      "loss": 0.39873199462890624,
      "num_input_tokens_seen": 6430488064,
      "step": 22610,
      "train_runtime": 219919.6272,
      "train_tokens_per_second": 29240.174
    },
    {
      "epoch": 0.8003414365544983,
      "grad_norm": 0.61328125,
      "learning_rate": 2.7684372424385613e-05,
      "loss": 0.4021897315979004,
      "num_input_tokens_seen": 6433337280,
      "step": 22620,
      "train_runtime": 220016.2226,
      "train_tokens_per_second": 29240.286
    },
    {
      "epoch": 0.8006952568182271,
      "grad_norm": 0.57421875,
      "learning_rate": 2.7680129803726496e-05,
      "loss": 0.397288179397583,
      "num_input_tokens_seen": 6436162176,
      "step": 22630,
      "train_runtime": 220113.6745,
      "train_tokens_per_second": 29240.174
    },
    {
      "epoch": 0.8010490770819558,
      "grad_norm": 0.5703125,
      "learning_rate": 2.767588913301025e-05,
      "loss": 0.40419631004333495,
      "num_input_tokens_seen": 6439001344,
      "step": 22640,
      "train_runtime": 220209.7097,
      "train_tokens_per_second": 29240.315
    },
    {
      "epoch": 0.8014028973456846,
      "grad_norm": 0.62109375,
      "learning_rate": 2.767165041074365e-05,
      "loss": 0.40021524429321287,
      "num_input_tokens_seen": 6441841536,
      "step": 22650,
      "train_runtime": 220305.0654,
      "train_tokens_per_second": 29240.551
    },
    {
      "epoch": 0.8017567176094134,
      "grad_norm": 0.5625,
      "learning_rate": 2.766741363543507e-05,
      "loss": 0.4055091381072998,
      "num_input_tokens_seen": 6444707584,
      "step": 22660,
      "train_runtime": 220403.9741,
      "train_tokens_per_second": 29240.433
    },
    {
      "epoch": 0.8021105378731421,
      "grad_norm": 0.60546875,
      "learning_rate": 2.7663178805594482e-05,
      "loss": 0.3965758800506592,
      "num_input_tokens_seen": 6447578112,
      "step": 22670,
      "train_runtime": 220501.3667,
      "train_tokens_per_second": 29240.536
    },
    {
      "epoch": 0.8024643581368709,
      "grad_norm": 0.56640625,
      "learning_rate": 2.7658945919733463e-05,
      "loss": 0.4039391040802002,
      "num_input_tokens_seen": 6450483520,
      "step": 22680,
      "train_runtime": 220604.4382,
      "train_tokens_per_second": 29240.044
    },
    {
      "epoch": 0.8028181784005998,
      "grad_norm": 0.57421875,
      "learning_rate": 2.7654714976365164e-05,
      "loss": 0.39691615104675293,
      "num_input_tokens_seen": 6453366400,
      "step": 22690,
      "train_runtime": 220701.4209,
      "train_tokens_per_second": 29240.258
    },
    {
      "epoch": 0.8031719986643285,
      "grad_norm": 0.5703125,
      "learning_rate": 2.765048597400436e-05,
      "loss": 0.4009274482727051,
      "num_input_tokens_seen": 6456204928,
      "step": 22700,
      "train_runtime": 220797.4659,
      "train_tokens_per_second": 29240.394
    },
    {
      "epoch": 0.8035258189280573,
      "grad_norm": 0.56640625,
      "learning_rate": 2.7646258911167373e-05,
      "loss": 0.4031529426574707,
      "num_input_tokens_seen": 6459031936,
      "step": 22710,
      "train_runtime": 220894.3986,
      "train_tokens_per_second": 29240.361
    },
    {
      "epoch": 0.8038796391917861,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7642033786372145e-05,
      "loss": 0.40304059982299806,
      "num_input_tokens_seen": 6461922624,
      "step": 22720,
      "train_runtime": 220991.3738,
      "train_tokens_per_second": 29240.61
    },
    {
      "epoch": 0.8042334594555148,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7637810598138193e-05,
      "loss": 0.3979023456573486,
      "num_input_tokens_seen": 6464810112,
      "step": 22730,
      "train_runtime": 221087.8941,
      "train_tokens_per_second": 29240.905
    },
    {
      "epoch": 0.8045872797192436,
      "grad_norm": 0.60546875,
      "learning_rate": 2.7633589344986615e-05,
      "loss": 0.4018411636352539,
      "num_input_tokens_seen": 6467718464,
      "step": 22740,
      "train_runtime": 221189.9903,
      "train_tokens_per_second": 29240.557
    },
    {
      "epoch": 0.8049410999829724,
      "grad_norm": 0.60546875,
      "learning_rate": 2.7629370025440087e-05,
      "loss": 0.40100917816162107,
      "num_input_tokens_seen": 6470567360,
      "step": 22750,
      "train_runtime": 221289.593,
      "train_tokens_per_second": 29240.27
    },
    {
      "epoch": 0.8052949202467011,
      "grad_norm": 0.5859375,
      "learning_rate": 2.762515263802287e-05,
      "loss": 0.40035085678100585,
      "num_input_tokens_seen": 6473497920,
      "step": 22760,
      "train_runtime": 221389.8537,
      "train_tokens_per_second": 29240.265
    },
    {
      "epoch": 0.80564874051043,
      "grad_norm": 0.578125,
      "learning_rate": 2.762093718126081e-05,
      "loss": 0.40250182151794434,
      "num_input_tokens_seen": 6476327488,
      "step": 22770,
      "train_runtime": 221482.7278,
      "train_tokens_per_second": 29240.779
    },
    {
      "epoch": 0.8060025607741588,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7616723653681298e-05,
      "loss": 0.4035053253173828,
      "num_input_tokens_seen": 6479164608,
      "step": 22780,
      "train_runtime": 221578.8002,
      "train_tokens_per_second": 29240.905
    },
    {
      "epoch": 0.8063563810378875,
      "grad_norm": 0.60546875,
      "learning_rate": 2.761251205381332e-05,
      "loss": 0.39922728538513186,
      "num_input_tokens_seen": 6482004800,
      "step": 22790,
      "train_runtime": 221675.5232,
      "train_tokens_per_second": 29240.959
    },
    {
      "epoch": 0.8067102013016163,
      "grad_norm": 0.5859375,
      "learning_rate": 2.760830238018744e-05,
      "loss": 0.4007260322570801,
      "num_input_tokens_seen": 6484838592,
      "step": 22800,
      "train_runtime": 221773.2235,
      "train_tokens_per_second": 29240.855
    },
    {
      "epoch": 0.8070640215653451,
      "grad_norm": 0.59375,
      "learning_rate": 2.760409463133576e-05,
      "loss": 0.39955615997314453,
      "num_input_tokens_seen": 6487655488,
      "step": 22810,
      "train_runtime": 221866.7595,
      "train_tokens_per_second": 29241.223
    },
    {
      "epoch": 0.8074178418290738,
      "grad_norm": 0.61328125,
      "learning_rate": 2.7599888805791967e-05,
      "loss": 0.40296120643615724,
      "num_input_tokens_seen": 6490529792,
      "step": 22820,
      "train_runtime": 221968.1833,
      "train_tokens_per_second": 29240.811
    },
    {
      "epoch": 0.8077716620928026,
      "grad_norm": 0.5703125,
      "learning_rate": 2.759568490209132e-05,
      "loss": 0.3974020004272461,
      "num_input_tokens_seen": 6493384512,
      "step": 22830,
      "train_runtime": 222065.6348,
      "train_tokens_per_second": 29240.835
    },
    {
      "epoch": 0.8081254823565314,
      "grad_norm": 0.6015625,
      "learning_rate": 2.7591482918770613e-05,
      "loss": 0.39903085231781005,
      "num_input_tokens_seen": 6496235776,
      "step": 22840,
      "train_runtime": 222162.364,
      "train_tokens_per_second": 29240.937
    },
    {
      "epoch": 0.8084793026202602,
      "grad_norm": 0.5625,
      "learning_rate": 2.7587282854368222e-05,
      "loss": 0.39942545890808107,
      "num_input_tokens_seen": 6499074240,
      "step": 22850,
      "train_runtime": 222260.5674,
      "train_tokens_per_second": 29240.788
    },
    {
      "epoch": 0.808833122883989,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7583084707424072e-05,
      "loss": 0.3966909170150757,
      "num_input_tokens_seen": 6501894272,
      "step": 22860,
      "train_runtime": 222356.4561,
      "train_tokens_per_second": 29240.861
    },
    {
      "epoch": 0.8091869431477178,
      "grad_norm": 0.58203125,
      "learning_rate": 2.7578888476479643e-05,
      "loss": 0.402421760559082,
      "num_input_tokens_seen": 6504719232,
      "step": 22870,
      "train_runtime": 222453.212,
      "train_tokens_per_second": 29240.842
    },
    {
      "epoch": 0.8095407634114465,
      "grad_norm": 0.60546875,
      "learning_rate": 2.7574694160077964e-05,
      "loss": 0.3987089157104492,
      "num_input_tokens_seen": 6507529088,
      "step": 22880,
      "train_runtime": 222548.3018,
      "train_tokens_per_second": 29240.974
    },
    {
      "epoch": 0.8098945836751753,
      "grad_norm": 0.578125,
      "learning_rate": 2.757050175676362e-05,
      "loss": 0.4025400161743164,
      "num_input_tokens_seen": 6510378048,
      "step": 22890,
      "train_runtime": 222647.7113,
      "train_tokens_per_second": 29240.714
    },
    {
      "epoch": 0.8102484039389041,
      "grad_norm": 0.54296875,
      "learning_rate": 2.756631126508274e-05,
      "loss": 0.401155948638916,
      "num_input_tokens_seen": 6513212288,
      "step": 22900,
      "train_runtime": 222741.6124,
      "train_tokens_per_second": 29241.111
    },
    {
      "epoch": 0.8106022242026328,
      "grad_norm": 0.58203125,
      "learning_rate": 2.7562122683583004e-05,
      "loss": 0.4013826847076416,
      "num_input_tokens_seen": 6516022976,
      "step": 22910,
      "train_runtime": 222840.1361,
      "train_tokens_per_second": 29240.796
    },
    {
      "epoch": 0.8109560444663616,
      "grad_norm": 0.5703125,
      "learning_rate": 2.7557936010813634e-05,
      "loss": 0.40588741302490233,
      "num_input_tokens_seen": 6518820352,
      "step": 22920,
      "train_runtime": 222934.9214,
      "train_tokens_per_second": 29240.912
    },
    {
      "epoch": 0.8113098647300904,
      "grad_norm": 0.58203125,
      "learning_rate": 2.75537512453254e-05,
      "loss": 0.4023984432220459,
      "num_input_tokens_seen": 6521648768,
      "step": 22930,
      "train_runtime": 223034.0193,
      "train_tokens_per_second": 29240.601
    },
    {
      "epoch": 0.8116636849938192,
      "grad_norm": 0.61328125,
      "learning_rate": 2.7549568385670595e-05,
      "loss": 0.40431985855102537,
      "num_input_tokens_seen": 6524491072,
      "step": 22940,
      "train_runtime": 223131.9726,
      "train_tokens_per_second": 29240.503
    },
    {
      "epoch": 0.812017505257548,
      "grad_norm": 0.609375,
      "learning_rate": 2.754538743040307e-05,
      "loss": 0.4044541835784912,
      "num_input_tokens_seen": 6527338176,
      "step": 22950,
      "train_runtime": 223229.1498,
      "train_tokens_per_second": 29240.528
    },
    {
      "epoch": 0.8123713255212768,
      "grad_norm": 0.59765625,
      "learning_rate": 2.75412083780782e-05,
      "loss": 0.3979846239089966,
      "num_input_tokens_seen": 6530195456,
      "step": 22960,
      "train_runtime": 223326.5544,
      "train_tokens_per_second": 29240.569
    },
    {
      "epoch": 0.8127251457850055,
      "grad_norm": 0.5703125,
      "learning_rate": 2.7537031227252898e-05,
      "loss": 0.4007590293884277,
      "num_input_tokens_seen": 6533060800,
      "step": 22970,
      "train_runtime": 223426.6871,
      "train_tokens_per_second": 29240.289
    },
    {
      "epoch": 0.8130789660487343,
      "grad_norm": 0.578125,
      "learning_rate": 2.75328559764856e-05,
      "loss": 0.405216121673584,
      "num_input_tokens_seen": 6535907968,
      "step": 22980,
      "train_runtime": 223522.5217,
      "train_tokens_per_second": 29240.49
    },
    {
      "epoch": 0.8134327863124631,
      "grad_norm": 0.59765625,
      "learning_rate": 2.752868262433629e-05,
      "loss": 0.40304250717163087,
      "num_input_tokens_seen": 6538782976,
      "step": 22990,
      "train_runtime": 223622.3276,
      "train_tokens_per_second": 29240.296
    },
    {
      "epoch": 0.8137866065761918,
      "grad_norm": 0.58984375,
      "learning_rate": 2.752451116936646e-05,
      "loss": 0.4003291606903076,
      "num_input_tokens_seen": 6541612096,
      "step": 23000,
      "train_runtime": 223717.8049,
      "train_tokens_per_second": 29240.463
    },
    {
      "epoch": 0.8141404268399206,
      "grad_norm": 0.609375,
      "learning_rate": 2.752034161013915e-05,
      "loss": 0.398400092124939,
      "num_input_tokens_seen": 6544474112,
      "step": 23010,
      "train_runtime": 223818.1918,
      "train_tokens_per_second": 29240.135
    },
    {
      "epoch": 0.8144942471036495,
      "grad_norm": 0.5703125,
      "learning_rate": 2.7516173945218888e-05,
      "loss": 0.39873456954956055,
      "num_input_tokens_seen": 6547316608,
      "step": 23020,
      "train_runtime": 223912.2222,
      "train_tokens_per_second": 29240.55
    },
    {
      "epoch": 0.8148480673673782,
      "grad_norm": 0.5859375,
      "learning_rate": 2.751200817317176e-05,
      "loss": 0.4035696029663086,
      "num_input_tokens_seen": 6550139776,
      "step": 23030,
      "train_runtime": 224009.0095,
      "train_tokens_per_second": 29240.519
    },
    {
      "epoch": 0.815201887631107,
      "grad_norm": 0.578125,
      "learning_rate": 2.7507844292565354e-05,
      "loss": 0.39484448432922364,
      "num_input_tokens_seen": 6552959808,
      "step": 23040,
      "train_runtime": 224105.9147,
      "train_tokens_per_second": 29240.459
    },
    {
      "epoch": 0.8155557078948358,
      "grad_norm": 0.6171875,
      "learning_rate": 2.7503682301968763e-05,
      "loss": 0.39975433349609374,
      "num_input_tokens_seen": 6555845696,
      "step": 23050,
      "train_runtime": 224206.8054,
      "train_tokens_per_second": 29240.173
    },
    {
      "epoch": 0.8159095281585645,
      "grad_norm": 0.59375,
      "learning_rate": 2.7499522199952627e-05,
      "loss": 0.4039126396179199,
      "num_input_tokens_seen": 6558741824,
      "step": 23060,
      "train_runtime": 224309.5896,
      "train_tokens_per_second": 29239.685
    },
    {
      "epoch": 0.8162633484222933,
      "grad_norm": 0.609375,
      "learning_rate": 2.7495363985089072e-05,
      "loss": 0.398949408531189,
      "num_input_tokens_seen": 6561577408,
      "step": 23070,
      "train_runtime": 224407.1797,
      "train_tokens_per_second": 29239.606
    },
    {
      "epoch": 0.8166171686860221,
      "grad_norm": 0.578125,
      "learning_rate": 2.7491207655951745e-05,
      "loss": 0.4011551856994629,
      "num_input_tokens_seen": 6564394368,
      "step": 23080,
      "train_runtime": 224502.6535,
      "train_tokens_per_second": 29239.718
    },
    {
      "epoch": 0.8169709889497508,
      "grad_norm": 0.58203125,
      "learning_rate": 2.7487053211115797e-05,
      "loss": 0.3990474224090576,
      "num_input_tokens_seen": 6567258048,
      "step": 23090,
      "train_runtime": 224603.8309,
      "train_tokens_per_second": 29239.297
    },
    {
      "epoch": 0.8173248092134797,
      "grad_norm": 0.6171875,
      "learning_rate": 2.748290064915789e-05,
      "loss": 0.4018836975097656,
      "num_input_tokens_seen": 6570109760,
      "step": 23100,
      "train_runtime": 224701.3323,
      "train_tokens_per_second": 29239.3
    },
    {
      "epoch": 0.8176786294772085,
      "grad_norm": 0.58984375,
      "learning_rate": 2.747874996865619e-05,
      "loss": 0.4037341117858887,
      "num_input_tokens_seen": 6572910400,
      "step": 23110,
      "train_runtime": 224797.346,
      "train_tokens_per_second": 29239.27
    },
    {
      "epoch": 0.8180324497409373,
      "grad_norm": 0.60546875,
      "learning_rate": 2.747460116819037e-05,
      "loss": 0.40352768898010255,
      "num_input_tokens_seen": 6575717312,
      "step": 23120,
      "train_runtime": 224890.6347,
      "train_tokens_per_second": 29239.623
    },
    {
      "epoch": 0.818386270004666,
      "grad_norm": 0.609375,
      "learning_rate": 2.7470454246341597e-05,
      "loss": 0.40348501205444337,
      "num_input_tokens_seen": 6578460416,
      "step": 23130,
      "train_runtime": 224982.1247,
      "train_tokens_per_second": 29239.925
    },
    {
      "epoch": 0.8187400902683948,
      "grad_norm": 0.58203125,
      "learning_rate": 2.746630920169254e-05,
      "loss": 0.40070600509643556,
      "num_input_tokens_seen": 6581277248,
      "step": 23140,
      "train_runtime": 225080.2647,
      "train_tokens_per_second": 29239.69
    },
    {
      "epoch": 0.8190939105321235,
      "grad_norm": 0.59765625,
      "learning_rate": 2.7462166032827358e-05,
      "loss": 0.40209321975708007,
      "num_input_tokens_seen": 6584125312,
      "step": 23150,
      "train_runtime": 225178.987,
      "train_tokens_per_second": 29239.519
    },
    {
      "epoch": 0.8194477307958523,
      "grad_norm": 0.578125,
      "learning_rate": 2.745802473833171e-05,
      "loss": 0.4015172004699707,
      "num_input_tokens_seen": 6586948032,
      "step": 23160,
      "train_runtime": 225274.9314,
      "train_tokens_per_second": 29239.596
    },
    {
      "epoch": 0.8198015510595811,
      "grad_norm": 0.578125,
      "learning_rate": 2.7453885316792755e-05,
      "loss": 0.39780478477478026,
      "num_input_tokens_seen": 6589785408,
      "step": 23170,
      "train_runtime": 225371.6977,
      "train_tokens_per_second": 29239.632
    },
    {
      "epoch": 0.8201553713233098,
      "grad_norm": 0.57421875,
      "learning_rate": 2.7449747766799135e-05,
      "loss": 0.3977693557739258,
      "num_input_tokens_seen": 6592632320,
      "step": 23180,
      "train_runtime": 225471.259,
      "train_tokens_per_second": 29239.347
    },
    {
      "epoch": 0.8205091915870387,
      "grad_norm": 0.59375,
      "learning_rate": 2.744561208694097e-05,
      "loss": 0.40313167572021485,
      "num_input_tokens_seen": 6595490112,
      "step": 23190,
      "train_runtime": 225568.1201,
      "train_tokens_per_second": 29239.46
    },
    {
      "epoch": 0.8208630118507675,
      "grad_norm": 0.578125,
      "learning_rate": 2.7441478275809884e-05,
      "loss": 0.40102329254150393,
      "num_input_tokens_seen": 6598388416,
      "step": 23200,
      "train_runtime": 225669.2044,
      "train_tokens_per_second": 29239.206
    },
    {
      "epoch": 0.8212168321144963,
      "grad_norm": 0.62109375,
      "learning_rate": 2.743734633199898e-05,
      "loss": 0.4089224338531494,
      "num_input_tokens_seen": 6601239296,
      "step": 23210,
      "train_runtime": 225767.9258,
      "train_tokens_per_second": 29239.048
    },
    {
      "epoch": 0.821570652378225,
      "grad_norm": 0.58203125,
      "learning_rate": 2.743321625410284e-05,
      "loss": 0.4025871276855469,
      "num_input_tokens_seen": 6604035840,
      "step": 23220,
      "train_runtime": 225860.6639,
      "train_tokens_per_second": 29239.425
    },
    {
      "epoch": 0.8219244726419538,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7429088040717526e-05,
      "loss": 0.40401582717895507,
      "num_input_tokens_seen": 6606898432,
      "step": 23230,
      "train_runtime": 225957.8124,
      "train_tokens_per_second": 29239.522
    },
    {
      "epoch": 0.8222782929056826,
      "grad_norm": 0.59375,
      "learning_rate": 2.7424961690440586e-05,
      "loss": 0.39657657146453856,
      "num_input_tokens_seen": 6609739008,
      "step": 23240,
      "train_runtime": 226051.7867,
      "train_tokens_per_second": 29239.933
    },
    {
      "epoch": 0.8226321131694113,
      "grad_norm": 0.62109375,
      "learning_rate": 2.742083720187103e-05,
      "loss": 0.39868438243865967,
      "num_input_tokens_seen": 6612523008,
      "step": 23250,
      "train_runtime": 226143.6327,
      "train_tokens_per_second": 29240.368
    },
    {
      "epoch": 0.8229859334331401,
      "grad_norm": 0.61328125,
      "learning_rate": 2.7416714573609363e-05,
      "loss": 0.40042638778686523,
      "num_input_tokens_seen": 6615339584,
      "step": 23260,
      "train_runtime": 226240.2902,
      "train_tokens_per_second": 29240.325
    },
    {
      "epoch": 0.823339753696869,
      "grad_norm": 0.578125,
      "learning_rate": 2.7412593804257543e-05,
      "loss": 0.40003418922424316,
      "num_input_tokens_seen": 6618179392,
      "step": 23270,
      "train_runtime": 226334.7984,
      "train_tokens_per_second": 29240.662
    },
    {
      "epoch": 0.8236935739605977,
      "grad_norm": 0.59765625,
      "learning_rate": 2.740847489241901e-05,
      "loss": 0.40078253746032716,
      "num_input_tokens_seen": 6621007104,
      "step": 23280,
      "train_runtime": 226432.7942,
      "train_tokens_per_second": 29240.496
    },
    {
      "epoch": 0.8240473942243265,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7404357836698668e-05,
      "loss": 0.40749382972717285,
      "num_input_tokens_seen": 6623822912,
      "step": 23290,
      "train_runtime": 226527.7947,
      "train_tokens_per_second": 29240.663
    },
    {
      "epoch": 0.8244012144880553,
      "grad_norm": 0.58203125,
      "learning_rate": 2.740024263570289e-05,
      "loss": 0.40078063011169435,
      "num_input_tokens_seen": 6626688576,
      "step": 23300,
      "train_runtime": 226625.6531,
      "train_tokens_per_second": 29240.682
    },
    {
      "epoch": 0.824755034751784,
      "grad_norm": 0.60546875,
      "learning_rate": 2.7396129288039503e-05,
      "loss": 0.3996004104614258,
      "num_input_tokens_seen": 6629458304,
      "step": 23310,
      "train_runtime": 226717.9459,
      "train_tokens_per_second": 29240.995
    },
    {
      "epoch": 0.8251088550155128,
      "grad_norm": 0.6171875,
      "learning_rate": 2.739201779231782e-05,
      "loss": 0.40511474609375,
      "num_input_tokens_seen": 6632267136,
      "step": 23320,
      "train_runtime": 226811.3615,
      "train_tokens_per_second": 29241.336
    },
    {
      "epoch": 0.8254626752792416,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7387908147148587e-05,
      "loss": 0.4056240081787109,
      "num_input_tokens_seen": 6635163584,
      "step": 23330,
      "train_runtime": 226911.4497,
      "train_tokens_per_second": 29241.202
    },
    {
      "epoch": 0.8258164955429703,
      "grad_norm": 0.6171875,
      "learning_rate": 2.7383800351144027e-05,
      "loss": 0.40821280479431155,
      "num_input_tokens_seen": 6638039808,
      "step": 23340,
      "train_runtime": 227007.081,
      "train_tokens_per_second": 29241.554
    },
    {
      "epoch": 0.8261703158066992,
      "grad_norm": 0.56640625,
      "learning_rate": 2.737969440291782e-05,
      "loss": 0.40048990249633787,
      "num_input_tokens_seen": 6640829056,
      "step": 23350,
      "train_runtime": 227101.362,
      "train_tokens_per_second": 29241.696
    },
    {
      "epoch": 0.826524136070428,
      "grad_norm": 0.58203125,
      "learning_rate": 2.7375590301085087e-05,
      "loss": 0.4001479625701904,
      "num_input_tokens_seen": 6643602688,
      "step": 23360,
      "train_runtime": 227195.9009,
      "train_tokens_per_second": 29241.737
    },
    {
      "epoch": 0.8268779563341567,
      "grad_norm": 0.61328125,
      "learning_rate": 2.7371488044262406e-05,
      "loss": 0.40163044929504393,
      "num_input_tokens_seen": 6646403584,
      "step": 23370,
      "train_runtime": 227288.1138,
      "train_tokens_per_second": 29242.196
    },
    {
      "epoch": 0.8272317765978855,
      "grad_norm": 0.5703125,
      "learning_rate": 2.7367387631067816e-05,
      "loss": 0.397481107711792,
      "num_input_tokens_seen": 6649190208,
      "step": 23380,
      "train_runtime": 227382.4682,
      "train_tokens_per_second": 29242.317
    },
    {
      "epoch": 0.8275855968616143,
      "grad_norm": 0.59375,
      "learning_rate": 2.73632890601208e-05,
      "loss": 0.40126667022705076,
      "num_input_tokens_seen": 6651995392,
      "step": 23390,
      "train_runtime": 227477.2003,
      "train_tokens_per_second": 29242.471
    },
    {
      "epoch": 0.827939417125343,
      "grad_norm": 0.58203125,
      "learning_rate": 2.7359192330042277e-05,
      "loss": 0.3992583990097046,
      "num_input_tokens_seen": 6654835136,
      "step": 23400,
      "train_runtime": 227574.5842,
      "train_tokens_per_second": 29242.436
    },
    {
      "epoch": 0.8282932373890718,
      "grad_norm": 0.578125,
      "learning_rate": 2.7355097439454625e-05,
      "loss": 0.40599637031555175,
      "num_input_tokens_seen": 6657648832,
      "step": 23410,
      "train_runtime": 227668.8874,
      "train_tokens_per_second": 29242.682
    },
    {
      "epoch": 0.8286470576528006,
      "grad_norm": 0.59375,
      "learning_rate": 2.735100438698166e-05,
      "loss": 0.4006871223449707,
      "num_input_tokens_seen": 6660510272,
      "step": 23420,
      "train_runtime": 227767.5554,
      "train_tokens_per_second": 29242.577
    },
    {
      "epoch": 0.8290008779165293,
      "grad_norm": 0.59375,
      "learning_rate": 2.7346913171248635e-05,
      "loss": 0.4014723300933838,
      "num_input_tokens_seen": 6663379200,
      "step": 23430,
      "train_runtime": 227870.0338,
      "train_tokens_per_second": 29242.016
    },
    {
      "epoch": 0.8293546981802582,
      "grad_norm": 0.6015625,
      "learning_rate": 2.734282379088225e-05,
      "loss": 0.3993005037307739,
      "num_input_tokens_seen": 6666237120,
      "step": 23440,
      "train_runtime": 227970.6983,
      "train_tokens_per_second": 29241.64
    },
    {
      "epoch": 0.829708518443987,
      "grad_norm": 0.609375,
      "learning_rate": 2.733873624451063e-05,
      "loss": 0.40107412338256837,
      "num_input_tokens_seen": 6669092544,
      "step": 23450,
      "train_runtime": 228069.286,
      "train_tokens_per_second": 29241.52
    },
    {
      "epoch": 0.8300623387077157,
      "grad_norm": 0.609375,
      "learning_rate": 2.733465053076335e-05,
      "loss": 0.40411982536315916,
      "num_input_tokens_seen": 6671894848,
      "step": 23460,
      "train_runtime": 228164.9418,
      "train_tokens_per_second": 29241.543
    },
    {
      "epoch": 0.8304161589714445,
      "grad_norm": 0.5625,
      "learning_rate": 2.733056664827141e-05,
      "loss": 0.3983922004699707,
      "num_input_tokens_seen": 6674719872,
      "step": 23470,
      "train_runtime": 228259.0279,
      "train_tokens_per_second": 29241.866
    },
    {
      "epoch": 0.8307699792351733,
      "grad_norm": 0.60546875,
      "learning_rate": 2.732648459566724e-05,
      "loss": 0.4013786792755127,
      "num_input_tokens_seen": 6677592960,
      "step": 23480,
      "train_runtime": 228358.8753,
      "train_tokens_per_second": 29241.662
    },
    {
      "epoch": 0.831123799498902,
      "grad_norm": 0.58984375,
      "learning_rate": 2.73224043715847e-05,
      "loss": 0.40276732444763186,
      "num_input_tokens_seen": 6680380736,
      "step": 23490,
      "train_runtime": 228452.0276,
      "train_tokens_per_second": 29241.941
    },
    {
      "epoch": 0.8314776197626308,
      "grad_norm": 0.59765625,
      "learning_rate": 2.731832597465908e-05,
      "loss": 0.40181670188903806,
      "num_input_tokens_seen": 6683200000,
      "step": 23500,
      "train_runtime": 228548.9566,
      "train_tokens_per_second": 29241.875
    },
    {
      "epoch": 0.8318314400263596,
      "grad_norm": 0.625,
      "learning_rate": 2.7314249403527097e-05,
      "loss": 0.3983565330505371,
      "num_input_tokens_seen": 6686038400,
      "step": 23510,
      "train_runtime": 228648.2155,
      "train_tokens_per_second": 29241.594
    },
    {
      "epoch": 0.8321852602900884,
      "grad_norm": 0.6015625,
      "learning_rate": 2.731017465682689e-05,
      "loss": 0.4006089210510254,
      "num_input_tokens_seen": 6688851328,
      "step": 23520,
      "train_runtime": 228743.8829,
      "train_tokens_per_second": 29241.662
    },
    {
      "epoch": 0.8325390805538172,
      "grad_norm": 0.57421875,
      "learning_rate": 2.730610173319802e-05,
      "loss": 0.4024630546569824,
      "num_input_tokens_seen": 6691664960,
      "step": 23530,
      "train_runtime": 228838.0542,
      "train_tokens_per_second": 29241.924
    },
    {
      "epoch": 0.832892900817546,
      "grad_norm": 0.59765625,
      "learning_rate": 2.7302030631281454e-05,
      "loss": 0.40010366439819334,
      "num_input_tokens_seen": 6694520128,
      "step": 23540,
      "train_runtime": 228935.3453,
      "train_tokens_per_second": 29241.968
    },
    {
      "epoch": 0.8332467210812747,
      "grad_norm": 0.6015625,
      "learning_rate": 2.7297961349719608e-05,
      "loss": 0.4005733489990234,
      "num_input_tokens_seen": 6697419392,
      "step": 23550,
      "train_runtime": 229037.5235,
      "train_tokens_per_second": 29241.581
    },
    {
      "epoch": 0.8336005413450035,
      "grad_norm": 0.61328125,
      "learning_rate": 2.729389388715628e-05,
      "loss": 0.403204345703125,
      "num_input_tokens_seen": 6700298176,
      "step": 23560,
      "train_runtime": 229138.2028,
      "train_tokens_per_second": 29241.297
    },
    {
      "epoch": 0.8339543616087323,
      "grad_norm": 0.60546875,
      "learning_rate": 2.728982824223671e-05,
      "loss": 0.40502614974975587,
      "num_input_tokens_seen": 6703155264,
      "step": 23570,
      "train_runtime": 229234.0075,
      "train_tokens_per_second": 29241.539
    },
    {
      "epoch": 0.834308181872461,
      "grad_norm": 0.60546875,
      "learning_rate": 2.7285764413607527e-05,
      "loss": 0.4047130584716797,
      "num_input_tokens_seen": 6706059072,
      "step": 23580,
      "train_runtime": 229332.7372,
      "train_tokens_per_second": 29241.613
    },
    {
      "epoch": 0.8346620021361898,
      "grad_norm": 0.59765625,
      "learning_rate": 2.7281702399916786e-05,
      "loss": 0.4016727924346924,
      "num_input_tokens_seen": 6708845440,
      "step": 23590,
      "train_runtime": 229426.6707,
      "train_tokens_per_second": 29241.785
    },
    {
      "epoch": 0.8350158223999187,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7277642199813954e-05,
      "loss": 0.4033057689666748,
      "num_input_tokens_seen": 6711675456,
      "step": 23600,
      "train_runtime": 229523.5159,
      "train_tokens_per_second": 29241.777
    },
    {
      "epoch": 0.8353696426636474,
      "grad_norm": 0.59375,
      "learning_rate": 2.7273583811949882e-05,
      "loss": 0.4035799026489258,
      "num_input_tokens_seen": 6714522112,
      "step": 23610,
      "train_runtime": 229618.9217,
      "train_tokens_per_second": 29242.024
    },
    {
      "epoch": 0.8357234629273762,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7269527234976844e-05,
      "loss": 0.4005948543548584,
      "num_input_tokens_seen": 6717364800,
      "step": 23620,
      "train_runtime": 229714.8264,
      "train_tokens_per_second": 29242.191
    },
    {
      "epoch": 0.836077283191105,
      "grad_norm": 0.5859375,
      "learning_rate": 2.726547246754852e-05,
      "loss": 0.40173816680908203,
      "num_input_tokens_seen": 6720234496,
      "step": 23630,
      "train_runtime": 229815.0756,
      "train_tokens_per_second": 29241.922
    },
    {
      "epoch": 0.8364311034548337,
      "grad_norm": 0.59765625,
      "learning_rate": 2.726141950831998e-05,
      "loss": 0.3997680425643921,
      "num_input_tokens_seen": 6723101184,
      "step": 23640,
      "train_runtime": 229909.6251,
      "train_tokens_per_second": 29242.365
    },
    {
      "epoch": 0.8367849237185625,
      "grad_norm": 0.60546875,
      "learning_rate": 2.7257368355947686e-05,
      "loss": 0.40159969329833983,
      "num_input_tokens_seen": 6725948544,
      "step": 23650,
      "train_runtime": 230007.4266,
      "train_tokens_per_second": 29242.31
    },
    {
      "epoch": 0.8371387439822913,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7253319009089528e-05,
      "loss": 0.4024930953979492,
      "num_input_tokens_seen": 6728794496,
      "step": 23660,
      "train_runtime": 230102.5708,
      "train_tokens_per_second": 29242.587
    },
    {
      "epoch": 0.83749256424602,
      "grad_norm": 0.56640625,
      "learning_rate": 2.7249271466404758e-05,
      "loss": 0.39936907291412355,
      "num_input_tokens_seen": 6731619392,
      "step": 23670,
      "train_runtime": 230199.6247,
      "train_tokens_per_second": 29242.53
    },
    {
      "epoch": 0.8378463845097489,
      "grad_norm": 0.59375,
      "learning_rate": 2.724522572655403e-05,
      "loss": 0.4012170314788818,
      "num_input_tokens_seen": 6734455360,
      "step": 23680,
      "train_runtime": 230298.2275,
      "train_tokens_per_second": 29242.324
    },
    {
      "epoch": 0.8382002047734777,
      "grad_norm": 0.6015625,
      "learning_rate": 2.7241181788199405e-05,
      "loss": 0.3996647596359253,
      "num_input_tokens_seen": 6737257792,
      "step": 23690,
      "train_runtime": 230391.5221,
      "train_tokens_per_second": 29242.646
    },
    {
      "epoch": 0.8385540250372064,
      "grad_norm": 0.60546875,
      "learning_rate": 2.7237139650004316e-05,
      "loss": 0.4030764579772949,
      "num_input_tokens_seen": 6740072256,
      "step": 23700,
      "train_runtime": 230486.0785,
      "train_tokens_per_second": 29242.861
    },
    {
      "epoch": 0.8389078453009352,
      "grad_norm": 0.58203125,
      "learning_rate": 2.7233099310633598e-05,
      "loss": 0.40282392501831055,
      "num_input_tokens_seen": 6742937472,
      "step": 23710,
      "train_runtime": 230584.2223,
      "train_tokens_per_second": 29242.84
    },
    {
      "epoch": 0.839261665564664,
      "grad_norm": 0.59765625,
      "learning_rate": 2.722906076875345e-05,
      "loss": 0.40260043144226076,
      "num_input_tokens_seen": 6745802560,
      "step": 23720,
      "train_runtime": 230680.812,
      "train_tokens_per_second": 29243.015
    },
    {
      "epoch": 0.8396154858283927,
      "grad_norm": 0.62109375,
      "learning_rate": 2.7225024023031487e-05,
      "loss": 0.40151286125183105,
      "num_input_tokens_seen": 6748624448,
      "step": 23730,
      "train_runtime": 230776.9835,
      "train_tokens_per_second": 29243.057
    },
    {
      "epoch": 0.8399693060921215,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7220989072136676e-05,
      "loss": 0.40032100677490234,
      "num_input_tokens_seen": 6751487104,
      "step": 23740,
      "train_runtime": 230874.4852,
      "train_tokens_per_second": 29243.106
    },
    {
      "epoch": 0.8403231263558503,
      "grad_norm": 0.6171875,
      "learning_rate": 2.7216955914739385e-05,
      "loss": 0.40537681579589846,
      "num_input_tokens_seen": 6754324864,
      "step": 23750,
      "train_runtime": 230971.7696,
      "train_tokens_per_second": 29243.075
    },
    {
      "epoch": 0.840676946619579,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7212924549511348e-05,
      "loss": 0.3997394323348999,
      "num_input_tokens_seen": 6757160064,
      "step": 23760,
      "train_runtime": 231069.1339,
      "train_tokens_per_second": 29243.023
    },
    {
      "epoch": 0.8410307668833079,
      "grad_norm": 0.58203125,
      "learning_rate": 2.7208894975125688e-05,
      "loss": 0.4048455238342285,
      "num_input_tokens_seen": 6759985152,
      "step": 23770,
      "train_runtime": 231164.95,
      "train_tokens_per_second": 29243.123
    },
    {
      "epoch": 0.8413845871470367,
      "grad_norm": 0.5859375,
      "learning_rate": 2.720486719025689e-05,
      "loss": 0.3983281135559082,
      "num_input_tokens_seen": 6762899520,
      "step": 23780,
      "train_runtime": 231267.742,
      "train_tokens_per_second": 29242.727
    },
    {
      "epoch": 0.8417384074107654,
      "grad_norm": 0.61328125,
      "learning_rate": 2.7200841193580816e-05,
      "loss": 0.4020063400268555,
      "num_input_tokens_seen": 6765737088,
      "step": 23790,
      "train_runtime": 231363.193,
      "train_tokens_per_second": 29242.928
    },
    {
      "epoch": 0.8420922276744942,
      "grad_norm": 0.578125,
      "learning_rate": 2.7196816983774708e-05,
      "loss": 0.402175760269165,
      "num_input_tokens_seen": 6768535104,
      "step": 23800,
      "train_runtime": 231458.8809,
      "train_tokens_per_second": 29242.927
    },
    {
      "epoch": 0.842446047938223,
      "grad_norm": 0.60546875,
      "learning_rate": 2.7192794559517166e-05,
      "loss": 0.39998593330383303,
      "num_input_tokens_seen": 6771360384,
      "step": 23810,
      "train_runtime": 231554.3962,
      "train_tokens_per_second": 29243.066
    },
    {
      "epoch": 0.8427998682019517,
      "grad_norm": 0.578125,
      "learning_rate": 2.7188773919488165e-05,
      "loss": 0.4000765323638916,
      "num_input_tokens_seen": 6774220352,
      "step": 23820,
      "train_runtime": 231654.2219,
      "train_tokens_per_second": 29242.81
    },
    {
      "epoch": 0.8431536884656805,
      "grad_norm": 0.59765625,
      "learning_rate": 2.7184755062369043e-05,
      "loss": 0.4026946544647217,
      "num_input_tokens_seen": 6777075840,
      "step": 23830,
      "train_runtime": 231753.8958,
      "train_tokens_per_second": 29242.554
    },
    {
      "epoch": 0.8435075087294093,
      "grad_norm": 0.640625,
      "learning_rate": 2.7180737986842498e-05,
      "loss": 0.3993961334228516,
      "num_input_tokens_seen": 6779957440,
      "step": 23840,
      "train_runtime": 231852.0896,
      "train_tokens_per_second": 29242.598
    },
    {
      "epoch": 0.8438613289931381,
      "grad_norm": 0.5859375,
      "learning_rate": 2.71767226915926e-05,
      "loss": 0.4022849082946777,
      "num_input_tokens_seen": 6782824192,
      "step": 23850,
      "train_runtime": 231950.8175,
      "train_tokens_per_second": 29242.51
    },
    {
      "epoch": 0.8442151492568669,
      "grad_norm": 0.5625,
      "learning_rate": 2.7172709175304767e-05,
      "loss": 0.39954342842102053,
      "num_input_tokens_seen": 6785715776,
      "step": 23860,
      "train_runtime": 232048.5861,
      "train_tokens_per_second": 29242.651
    },
    {
      "epoch": 0.8445689695205957,
      "grad_norm": 0.60546875,
      "learning_rate": 2.716869743666579e-05,
      "loss": 0.40131635665893556,
      "num_input_tokens_seen": 6788554176,
      "step": 23870,
      "train_runtime": 232142.7891,
      "train_tokens_per_second": 29243.011
    },
    {
      "epoch": 0.8449227897843244,
      "grad_norm": 0.58203125,
      "learning_rate": 2.7164687474363803e-05,
      "loss": 0.3996366262435913,
      "num_input_tokens_seen": 6791387200,
      "step": 23880,
      "train_runtime": 232240.5321,
      "train_tokens_per_second": 29242.902
    },
    {
      "epoch": 0.8452766100480532,
      "grad_norm": 0.61328125,
      "learning_rate": 2.7160679287088307e-05,
      "loss": 0.39946019649505615,
      "num_input_tokens_seen": 6794182400,
      "step": 23890,
      "train_runtime": 232335.9508,
      "train_tokens_per_second": 29242.923
    },
    {
      "epoch": 0.845630430311782,
      "grad_norm": 0.60546875,
      "learning_rate": 2.7156672873530147e-05,
      "loss": 0.40001659393310546,
      "num_input_tokens_seen": 6797066368,
      "step": 23900,
      "train_runtime": 232436.8081,
      "train_tokens_per_second": 29242.642
    },
    {
      "epoch": 0.8459842505755107,
      "grad_norm": 0.578125,
      "learning_rate": 2.715266823238152e-05,
      "loss": 0.40672922134399414,
      "num_input_tokens_seen": 6799884352,
      "step": 23910,
      "train_runtime": 232533.3231,
      "train_tokens_per_second": 29242.623
    },
    {
      "epoch": 0.8463380708392395,
      "grad_norm": 0.58984375,
      "learning_rate": 2.714866536233598e-05,
      "loss": 0.3905314922332764,
      "num_input_tokens_seen": 6802746560,
      "step": 23920,
      "train_runtime": 232631.7704,
      "train_tokens_per_second": 29242.552
    },
    {
      "epoch": 0.8466918911029684,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7144664262088422e-05,
      "loss": 0.402878475189209,
      "num_input_tokens_seen": 6805556800,
      "step": 23930,
      "train_runtime": 232726.2261,
      "train_tokens_per_second": 29242.758
    },
    {
      "epoch": 0.8470457113666972,
      "grad_norm": 0.578125,
      "learning_rate": 2.7140664930335084e-05,
      "loss": 0.40422534942626953,
      "num_input_tokens_seen": 6808367872,
      "step": 23940,
      "train_runtime": 232821.2996,
      "train_tokens_per_second": 29242.891
    },
    {
      "epoch": 0.8473995316304259,
      "grad_norm": 0.58203125,
      "learning_rate": 2.713666736577356e-05,
      "loss": 0.40209207534790037,
      "num_input_tokens_seen": 6811223296,
      "step": 23950,
      "train_runtime": 232919.0216,
      "train_tokens_per_second": 29242.881
    },
    {
      "epoch": 0.8477533518941547,
      "grad_norm": 0.59765625,
      "learning_rate": 2.713267156710278e-05,
      "loss": 0.39720659255981444,
      "num_input_tokens_seen": 6814095104,
      "step": 23960,
      "train_runtime": 233016.6498,
      "train_tokens_per_second": 29242.954
    },
    {
      "epoch": 0.8481071721578834,
      "grad_norm": 0.578125,
      "learning_rate": 2.712867753302301e-05,
      "loss": 0.4034297466278076,
      "num_input_tokens_seen": 6816930432,
      "step": 23970,
      "train_runtime": 233113.8682,
      "train_tokens_per_second": 29242.921
    },
    {
      "epoch": 0.8484609924216122,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7124685262235865e-05,
      "loss": 0.400280237197876,
      "num_input_tokens_seen": 6819820480,
      "step": 23980,
      "train_runtime": 233214.4467,
      "train_tokens_per_second": 29242.702
    },
    {
      "epoch": 0.848814812685341,
      "grad_norm": 0.6015625,
      "learning_rate": 2.7120694753444287e-05,
      "loss": 0.3994969606399536,
      "num_input_tokens_seen": 6822655680,
      "step": 23990,
      "train_runtime": 233312.7123,
      "train_tokens_per_second": 29242.537
    },
    {
      "epoch": 0.8491686329490697,
      "grad_norm": 0.6015625,
      "learning_rate": 2.7116706005352547e-05,
      "loss": 0.4037921905517578,
      "num_input_tokens_seen": 6825531264,
      "step": 24000,
      "train_runtime": 233411.416,
      "train_tokens_per_second": 29242.491
    },
    {
      "epoch": 0.8495224532127985,
      "grad_norm": 0.5859375,
      "learning_rate": 2.7112719016666282e-05,
      "loss": 0.3995975494384766,
      "num_input_tokens_seen": 6828340608,
      "step": 24010,
      "train_runtime": 233507.6233,
      "train_tokens_per_second": 29242.474
    },
    {
      "epoch": 0.8498762734765274,
      "grad_norm": 0.609375,
      "learning_rate": 2.7108733786092427e-05,
      "loss": 0.39865851402282715,
      "num_input_tokens_seen": 6831119424,
      "step": 24020,
      "train_runtime": 233602.7315,
      "train_tokens_per_second": 29242.464
    },
    {
      "epoch": 0.8502300937402562,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7104750312339254e-05,
      "loss": 0.39898841381072997,
      "num_input_tokens_seen": 6833910272,
      "step": 24030,
      "train_runtime": 233693.4329,
      "train_tokens_per_second": 29243.057
    },
    {
      "epoch": 0.8505839140039849,
      "grad_norm": 0.60546875,
      "learning_rate": 2.7100768594116382e-05,
      "loss": 0.40321874618530273,
      "num_input_tokens_seen": 6836786432,
      "step": 24040,
      "train_runtime": 233790.5278,
      "train_tokens_per_second": 29243.214
    },
    {
      "epoch": 0.8509377342677137,
      "grad_norm": 0.59375,
      "learning_rate": 2.709678863013473e-05,
      "loss": 0.40262203216552733,
      "num_input_tokens_seen": 6839663232,
      "step": 24050,
      "train_runtime": 233888.8888,
      "train_tokens_per_second": 29243.216
    },
    {
      "epoch": 0.8512915545314425,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7092810419106562e-05,
      "loss": 0.4003917217254639,
      "num_input_tokens_seen": 6842474688,
      "step": 24060,
      "train_runtime": 233985.006,
      "train_tokens_per_second": 29243.219
    },
    {
      "epoch": 0.8516453747951712,
      "grad_norm": 0.61328125,
      "learning_rate": 2.7088833959745457e-05,
      "loss": 0.39735984802246094,
      "num_input_tokens_seen": 6845340864,
      "step": 24070,
      "train_runtime": 234084.7902,
      "train_tokens_per_second": 29242.997
    },
    {
      "epoch": 0.8519991950589,
      "grad_norm": 0.5859375,
      "learning_rate": 2.708485925076631e-05,
      "loss": 0.3959228754043579,
      "num_input_tokens_seen": 6848187072,
      "step": 24080,
      "train_runtime": 234182.0074,
      "train_tokens_per_second": 29243.011
    },
    {
      "epoch": 0.8523530153226287,
      "grad_norm": 0.62109375,
      "learning_rate": 2.708088629088535e-05,
      "loss": 0.4006394386291504,
      "num_input_tokens_seen": 6851030336,
      "step": 24090,
      "train_runtime": 234277.1921,
      "train_tokens_per_second": 29243.266
    },
    {
      "epoch": 0.8527068355863576,
      "grad_norm": 0.5703125,
      "learning_rate": 2.7076915078820115e-05,
      "loss": 0.4071969985961914,
      "num_input_tokens_seen": 6853932416,
      "step": 24100,
      "train_runtime": 234375.7507,
      "train_tokens_per_second": 29243.351
    },
    {
      "epoch": 0.8530606558500864,
      "grad_norm": 0.578125,
      "learning_rate": 2.707294561328945e-05,
      "loss": 0.4035952568054199,
      "num_input_tokens_seen": 6856750656,
      "step": 24110,
      "train_runtime": 234471.5787,
      "train_tokens_per_second": 29243.419
    },
    {
      "epoch": 0.8534144761138152,
      "grad_norm": 0.609375,
      "learning_rate": 2.706897789301353e-05,
      "loss": 0.39940571784973145,
      "num_input_tokens_seen": 6859575616,
      "step": 24120,
      "train_runtime": 234569.1009,
      "train_tokens_per_second": 29243.304
    },
    {
      "epoch": 0.8537682963775439,
      "grad_norm": 0.578125,
      "learning_rate": 2.706501191671384e-05,
      "loss": 0.3971100807189941,
      "num_input_tokens_seen": 6862452608,
      "step": 24130,
      "train_runtime": 234671.6616,
      "train_tokens_per_second": 29242.784
    },
    {
      "epoch": 0.8541221166412727,
      "grad_norm": 0.6015625,
      "learning_rate": 2.7061047683113165e-05,
      "loss": 0.4009715557098389,
      "num_input_tokens_seen": 6865240064,
      "step": 24140,
      "train_runtime": 234765.4482,
      "train_tokens_per_second": 29242.975
    },
    {
      "epoch": 0.8544759369050015,
      "grad_norm": 0.609375,
      "learning_rate": 2.705708519093561e-05,
      "loss": 0.39855546951293946,
      "num_input_tokens_seen": 6868096128,
      "step": 24150,
      "train_runtime": 234860.3994,
      "train_tokens_per_second": 29243.313
    },
    {
      "epoch": 0.8548297571687302,
      "grad_norm": 0.61328125,
      "learning_rate": 2.705312443890658e-05,
      "loss": 0.4019068717956543,
      "num_input_tokens_seen": 6870921280,
      "step": 24160,
      "train_runtime": 234956.2379,
      "train_tokens_per_second": 29243.409
    },
    {
      "epoch": 0.855183577432459,
      "grad_norm": 0.58984375,
      "learning_rate": 2.70491654257528e-05,
      "loss": 0.4043871402740479,
      "num_input_tokens_seen": 6873722496,
      "step": 24170,
      "train_runtime": 235049.7705,
      "train_tokens_per_second": 29243.689
    },
    {
      "epoch": 0.8555373976961879,
      "grad_norm": 0.57421875,
      "learning_rate": 2.7045208150202274e-05,
      "loss": 0.4042680740356445,
      "num_input_tokens_seen": 6876630144,
      "step": 24180,
      "train_runtime": 235149.5498,
      "train_tokens_per_second": 29243.646
    },
    {
      "epoch": 0.8558912179599166,
      "grad_norm": 0.578125,
      "learning_rate": 2.704125261098433e-05,
      "loss": 0.39805164337158205,
      "num_input_tokens_seen": 6879503168,
      "step": 24190,
      "train_runtime": 235247.475,
      "train_tokens_per_second": 29243.685
    },
    {
      "epoch": 0.8562450382236454,
      "grad_norm": 0.6171875,
      "learning_rate": 2.7037298806829586e-05,
      "loss": 0.39963650703430176,
      "num_input_tokens_seen": 6882324928,
      "step": 24200,
      "train_runtime": 235344.347,
      "train_tokens_per_second": 29243.638
    },
    {
      "epoch": 0.8565988584873742,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7033346736469967e-05,
      "loss": 0.4035895347595215,
      "num_input_tokens_seen": 6885142784,
      "step": 24210,
      "train_runtime": 235440.7271,
      "train_tokens_per_second": 29243.635
    },
    {
      "epoch": 0.8569526787511029,
      "grad_norm": 0.5859375,
      "learning_rate": 2.702939639863869e-05,
      "loss": 0.3984901189804077,
      "num_input_tokens_seen": 6888005184,
      "step": 24220,
      "train_runtime": 235538.8077,
      "train_tokens_per_second": 29243.611
    },
    {
      "epoch": 0.8573064990148317,
      "grad_norm": 0.59375,
      "learning_rate": 2.7025447792070262e-05,
      "loss": 0.40219988822937014,
      "num_input_tokens_seen": 6890802816,
      "step": 24230,
      "train_runtime": 235633.1091,
      "train_tokens_per_second": 29243.78
    },
    {
      "epoch": 0.8576603192785605,
      "grad_norm": 0.61328125,
      "learning_rate": 2.7021500915500493e-05,
      "loss": 0.40181446075439453,
      "num_input_tokens_seen": 6893650624,
      "step": 24240,
      "train_runtime": 235728.7996,
      "train_tokens_per_second": 29243.99
    },
    {
      "epoch": 0.8580141395422892,
      "grad_norm": 0.59765625,
      "learning_rate": 2.701755576766648e-05,
      "loss": 0.39760627746582033,
      "num_input_tokens_seen": 6896463104,
      "step": 24250,
      "train_runtime": 235822.8686,
      "train_tokens_per_second": 29244.251
    },
    {
      "epoch": 0.858367959806018,
      "grad_norm": 0.5703125,
      "learning_rate": 2.701361234730661e-05,
      "loss": 0.398471736907959,
      "num_input_tokens_seen": 6899282112,
      "step": 24260,
      "train_runtime": 235919.7098,
      "train_tokens_per_second": 29244.195
    },
    {
      "epoch": 0.8587217800697469,
      "grad_norm": 0.57421875,
      "learning_rate": 2.700967065316056e-05,
      "loss": 0.4050884246826172,
      "num_input_tokens_seen": 6902066688,
      "step": 24270,
      "train_runtime": 236012.4411,
      "train_tokens_per_second": 29244.504
    },
    {
      "epoch": 0.8590756003334756,
      "grad_norm": 0.6015625,
      "learning_rate": 2.70057306839693e-05,
      "loss": 0.39977211952209474,
      "num_input_tokens_seen": 6904942080,
      "step": 24280,
      "train_runtime": 236109.1139,
      "train_tokens_per_second": 29244.708
    },
    {
      "epoch": 0.8594294205972044,
      "grad_norm": 0.58984375,
      "learning_rate": 2.7001792438475076e-05,
      "loss": 0.4012551784515381,
      "num_input_tokens_seen": 6907759808,
      "step": 24290,
      "train_runtime": 236206.8632,
      "train_tokens_per_second": 29244.535
    },
    {
      "epoch": 0.8597832408609332,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6997855915421416e-05,
      "loss": 0.401644229888916,
      "num_input_tokens_seen": 6910643456,
      "step": 24300,
      "train_runtime": 236306.3071,
      "train_tokens_per_second": 29244.431
    },
    {
      "epoch": 0.8601370611246619,
      "grad_norm": 0.609375,
      "learning_rate": 2.6993921113553145e-05,
      "loss": 0.4017834186553955,
      "num_input_tokens_seen": 6913510976,
      "step": 24310,
      "train_runtime": 236406.522,
      "train_tokens_per_second": 29244.163
    },
    {
      "epoch": 0.8604908813883907,
      "grad_norm": 0.59375,
      "learning_rate": 2.6989988031616343e-05,
      "loss": 0.4023585796356201,
      "num_input_tokens_seen": 6916365952,
      "step": 24320,
      "train_runtime": 236505.0915,
      "train_tokens_per_second": 29244.047
    },
    {
      "epoch": 0.8608447016521195,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6986056668358396e-05,
      "loss": 0.3998257160186768,
      "num_input_tokens_seen": 6919160512,
      "step": 24330,
      "train_runtime": 236598.5858,
      "train_tokens_per_second": 29244.302
    },
    {
      "epoch": 0.8611985219158482,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6982127022527947e-05,
      "loss": 0.3963127613067627,
      "num_input_tokens_seen": 6921967232,
      "step": 24340,
      "train_runtime": 236691.6587,
      "train_tokens_per_second": 29244.661
    },
    {
      "epoch": 0.8615523421795771,
      "grad_norm": 0.578125,
      "learning_rate": 2.697819909287493e-05,
      "loss": 0.40001840591430665,
      "num_input_tokens_seen": 6924779200,
      "step": 24350,
      "train_runtime": 236786.8895,
      "train_tokens_per_second": 29244.775
    },
    {
      "epoch": 0.8619061624433059,
      "grad_norm": 0.640625,
      "learning_rate": 2.697427287815053e-05,
      "loss": 0.4025301933288574,
      "num_input_tokens_seen": 6927630272,
      "step": 24360,
      "train_runtime": 236884.3528,
      "train_tokens_per_second": 29244.778
    },
    {
      "epoch": 0.8622599827070346,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6970348377107236e-05,
      "loss": 0.3963738441467285,
      "num_input_tokens_seen": 6930432384,
      "step": 24370,
      "train_runtime": 236979.1943,
      "train_tokens_per_second": 29244.898
    },
    {
      "epoch": 0.8626138029707634,
      "grad_norm": 0.5546875,
      "learning_rate": 2.696642558849877e-05,
      "loss": 0.3980403900146484,
      "num_input_tokens_seen": 6933254400,
      "step": 24380,
      "train_runtime": 237076.1712,
      "train_tokens_per_second": 29244.839
    },
    {
      "epoch": 0.8629676232344922,
      "grad_norm": 0.625,
      "learning_rate": 2.696250451108016e-05,
      "loss": 0.3990329265594482,
      "num_input_tokens_seen": 6936082688,
      "step": 24390,
      "train_runtime": 237171.9176,
      "train_tokens_per_second": 29244.958
    },
    {
      "epoch": 0.8633214434982209,
      "grad_norm": 0.6171875,
      "learning_rate": 2.6958585143607662e-05,
      "loss": 0.4002392292022705,
      "num_input_tokens_seen": 6938952640,
      "step": 24400,
      "train_runtime": 237271.7665,
      "train_tokens_per_second": 29244.746
    },
    {
      "epoch": 0.8636752637619497,
      "grad_norm": 0.6015625,
      "learning_rate": 2.695466748483883e-05,
      "loss": 0.39701406955718993,
      "num_input_tokens_seen": 6941816064,
      "step": 24410,
      "train_runtime": 237370.5961,
      "train_tokens_per_second": 29244.633
    },
    {
      "epoch": 0.8640290840256785,
      "grad_norm": 0.59765625,
      "learning_rate": 2.695075153353247e-05,
      "loss": 0.3982418060302734,
      "num_input_tokens_seen": 6944680640,
      "step": 24420,
      "train_runtime": 237469.2819,
      "train_tokens_per_second": 29244.543
    },
    {
      "epoch": 0.8643829042894073,
      "grad_norm": 0.55859375,
      "learning_rate": 2.6946837288448646e-05,
      "loss": 0.40138587951660154,
      "num_input_tokens_seen": 6947518400,
      "step": 24430,
      "train_runtime": 237566.6297,
      "train_tokens_per_second": 29244.505
    },
    {
      "epoch": 0.8647367245531361,
      "grad_norm": 0.5625,
      "learning_rate": 2.6942924748348684e-05,
      "loss": 0.40033588409423826,
      "num_input_tokens_seen": 6950360640,
      "step": 24440,
      "train_runtime": 237665.1867,
      "train_tokens_per_second": 29244.336
    },
    {
      "epoch": 0.8650905448168649,
      "grad_norm": 0.609375,
      "learning_rate": 2.693901391199517e-05,
      "loss": 0.40167908668518065,
      "num_input_tokens_seen": 6953204544,
      "step": 24450,
      "train_runtime": 237764.675,
      "train_tokens_per_second": 29244.061
    },
    {
      "epoch": 0.8654443650805936,
      "grad_norm": 0.625,
      "learning_rate": 2.6935104778151943e-05,
      "loss": 0.40256080627441404,
      "num_input_tokens_seen": 6956102976,
      "step": 24460,
      "train_runtime": 237863.9386,
      "train_tokens_per_second": 29244.042
    },
    {
      "epoch": 0.8657981853443224,
      "grad_norm": 0.61328125,
      "learning_rate": 2.6931197345584102e-05,
      "loss": 0.397590184211731,
      "num_input_tokens_seen": 6958895232,
      "step": 24470,
      "train_runtime": 237958.5446,
      "train_tokens_per_second": 29244.149
    },
    {
      "epoch": 0.8661520056080512,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6927291613058003e-05,
      "loss": 0.3997004270553589,
      "num_input_tokens_seen": 6961713088,
      "step": 24480,
      "train_runtime": 238053.9018,
      "train_tokens_per_second": 29244.272
    },
    {
      "epoch": 0.8665058258717799,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6923387579341254e-05,
      "loss": 0.39673662185668945,
      "num_input_tokens_seen": 6964551872,
      "step": 24490,
      "train_runtime": 238147.4014,
      "train_tokens_per_second": 29244.711
    },
    {
      "epoch": 0.8668596461355087,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6919485243202693e-05,
      "loss": 0.39530434608459475,
      "num_input_tokens_seen": 6967388480,
      "step": 24500,
      "train_runtime": 238244.6588,
      "train_tokens_per_second": 29244.679
    },
    {
      "epoch": 0.8672134663992375,
      "grad_norm": 0.578125,
      "learning_rate": 2.6915584603412434e-05,
      "loss": 0.39744112491607664,
      "num_input_tokens_seen": 6970218048,
      "step": 24510,
      "train_runtime": 238340.1691,
      "train_tokens_per_second": 29244.831
    },
    {
      "epoch": 0.8675672866629663,
      "grad_norm": 0.6015625,
      "learning_rate": 2.6911685658741827e-05,
      "loss": 0.40381155014038084,
      "num_input_tokens_seen": 6973067072,
      "step": 24520,
      "train_runtime": 238438.8011,
      "train_tokens_per_second": 29244.683
    },
    {
      "epoch": 0.8679211069266951,
      "grad_norm": 0.5859375,
      "learning_rate": 2.690778840796346e-05,
      "loss": 0.3999814987182617,
      "num_input_tokens_seen": 6975971712,
      "step": 24530,
      "train_runtime": 238537.6247,
      "train_tokens_per_second": 29244.744
    },
    {
      "epoch": 0.8682749271904239,
      "grad_norm": 0.578125,
      "learning_rate": 2.6903892849851175e-05,
      "loss": 0.40328664779663087,
      "num_input_tokens_seen": 6978857408,
      "step": 24540,
      "train_runtime": 238636.612,
      "train_tokens_per_second": 29244.705
    },
    {
      "epoch": 0.8686287474541526,
      "grad_norm": 0.578125,
      "learning_rate": 2.6899998983180057e-05,
      "loss": 0.40062580108642576,
      "num_input_tokens_seen": 6981693952,
      "step": 24550,
      "train_runtime": 238733.2755,
      "train_tokens_per_second": 29244.746
    },
    {
      "epoch": 0.8689825677178814,
      "grad_norm": 0.58203125,
      "learning_rate": 2.689610680672642e-05,
      "loss": 0.39992246627807615,
      "num_input_tokens_seen": 6984531072,
      "step": 24560,
      "train_runtime": 238829.546,
      "train_tokens_per_second": 29244.837
    },
    {
      "epoch": 0.8693363879816102,
      "grad_norm": 0.61328125,
      "learning_rate": 2.6892216319267843e-05,
      "loss": 0.39810681343078613,
      "num_input_tokens_seen": 6987390016,
      "step": 24570,
      "train_runtime": 238927.9982,
      "train_tokens_per_second": 29244.752
    },
    {
      "epoch": 0.8696902082453389,
      "grad_norm": 0.6015625,
      "learning_rate": 2.6888327519583107e-05,
      "loss": 0.4005607604980469,
      "num_input_tokens_seen": 6990256832,
      "step": 24580,
      "train_runtime": 239027.352,
      "train_tokens_per_second": 29244.59
    },
    {
      "epoch": 0.8700440285090677,
      "grad_norm": 0.5859375,
      "learning_rate": 2.688444040645225e-05,
      "loss": 0.3984896421432495,
      "num_input_tokens_seen": 6993125056,
      "step": 24590,
      "train_runtime": 239127.0516,
      "train_tokens_per_second": 29244.391
    },
    {
      "epoch": 0.8703978487727966,
      "grad_norm": 0.58203125,
      "learning_rate": 2.688055497865654e-05,
      "loss": 0.39867496490478516,
      "num_input_tokens_seen": 6995943936,
      "step": 24600,
      "train_runtime": 239222.0633,
      "train_tokens_per_second": 29244.56
    },
    {
      "epoch": 0.8707516690365253,
      "grad_norm": 0.57421875,
      "learning_rate": 2.6876671234978483e-05,
      "loss": 0.4010897636413574,
      "num_input_tokens_seen": 6998836736,
      "step": 24610,
      "train_runtime": 239324.5679,
      "train_tokens_per_second": 29244.121
    },
    {
      "epoch": 0.8711054893002541,
      "grad_norm": 0.57421875,
      "learning_rate": 2.6872789174201807e-05,
      "loss": 0.39871530532836913,
      "num_input_tokens_seen": 7001703360,
      "step": 24620,
      "train_runtime": 239424.464,
      "train_tokens_per_second": 29243.893
    },
    {
      "epoch": 0.8714593095639829,
      "grad_norm": 0.609375,
      "learning_rate": 2.686890879511147e-05,
      "loss": 0.4025277137756348,
      "num_input_tokens_seen": 7004562624,
      "step": 24630,
      "train_runtime": 239525.0201,
      "train_tokens_per_second": 29243.553
    },
    {
      "epoch": 0.8718131298277116,
      "grad_norm": 0.62890625,
      "learning_rate": 2.6865030096493665e-05,
      "loss": 0.4040116310119629,
      "num_input_tokens_seen": 7007431616,
      "step": 24640,
      "train_runtime": 239624.2932,
      "train_tokens_per_second": 29243.411
    },
    {
      "epoch": 0.8721669500914404,
      "grad_norm": 0.59375,
      "learning_rate": 2.6861153077135815e-05,
      "loss": 0.3997276544570923,
      "num_input_tokens_seen": 7010275584,
      "step": 24650,
      "train_runtime": 239721.3124,
      "train_tokens_per_second": 29243.439
    },
    {
      "epoch": 0.8725207703551692,
      "grad_norm": 0.57421875,
      "learning_rate": 2.6857277735826545e-05,
      "loss": 0.4015812873840332,
      "num_input_tokens_seen": 7013106304,
      "step": 24660,
      "train_runtime": 239817.2248,
      "train_tokens_per_second": 29243.547
    },
    {
      "epoch": 0.8728745906188979,
      "grad_norm": 0.578125,
      "learning_rate": 2.685340407135573e-05,
      "loss": 0.3989898681640625,
      "num_input_tokens_seen": 7016033600,
      "step": 24670,
      "train_runtime": 239919.9575,
      "train_tokens_per_second": 29243.226
    },
    {
      "epoch": 0.8732284108826268,
      "grad_norm": 0.59375,
      "learning_rate": 2.6849532082514445e-05,
      "loss": 0.39873347282409666,
      "num_input_tokens_seen": 7018875776,
      "step": 24680,
      "train_runtime": 240017.5015,
      "train_tokens_per_second": 29243.183
    },
    {
      "epoch": 0.8735822311463556,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6845661768095005e-05,
      "loss": 0.3964445352554321,
      "num_input_tokens_seen": 7021816512,
      "step": 24690,
      "train_runtime": 240120.5197,
      "train_tokens_per_second": 29242.884
    },
    {
      "epoch": 0.8739360514100843,
      "grad_norm": 0.61328125,
      "learning_rate": 2.6841793126890925e-05,
      "loss": 0.4021657943725586,
      "num_input_tokens_seen": 7024625344,
      "step": 24700,
      "train_runtime": 240213.2829,
      "train_tokens_per_second": 29243.284
    },
    {
      "epoch": 0.8742898716738131,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6837926157696946e-05,
      "loss": 0.4011106014251709,
      "num_input_tokens_seen": 7027444544,
      "step": 24710,
      "train_runtime": 240312.21,
      "train_tokens_per_second": 29242.977
    },
    {
      "epoch": 0.8746436919375419,
      "grad_norm": 0.609375,
      "learning_rate": 2.6834060859309018e-05,
      "loss": 0.40004878044128417,
      "num_input_tokens_seen": 7030291968,
      "step": 24720,
      "train_runtime": 240408.3043,
      "train_tokens_per_second": 29243.133
    },
    {
      "epoch": 0.8749975122012706,
      "grad_norm": 0.609375,
      "learning_rate": 2.6830197230524317e-05,
      "loss": 0.3969268321990967,
      "num_input_tokens_seen": 7033122624,
      "step": 24730,
      "train_runtime": 240501.6399,
      "train_tokens_per_second": 29243.554
    },
    {
      "epoch": 0.8753513324649994,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6826335270141216e-05,
      "loss": 0.39848220348358154,
      "num_input_tokens_seen": 7036029120,
      "step": 24740,
      "train_runtime": 240604.1567,
      "train_tokens_per_second": 29243.174
    },
    {
      "epoch": 0.8757051527287282,
      "grad_norm": 0.59375,
      "learning_rate": 2.6822474976959312e-05,
      "loss": 0.39956305027008054,
      "num_input_tokens_seen": 7038888256,
      "step": 24750,
      "train_runtime": 240703.4345,
      "train_tokens_per_second": 29242.991
    },
    {
      "epoch": 0.8760589729924569,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6818616349779397e-05,
      "loss": 0.40245985984802246,
      "num_input_tokens_seen": 7041737792,
      "step": 24760,
      "train_runtime": 240800.5963,
      "train_tokens_per_second": 29243.025
    },
    {
      "epoch": 0.8764127932561858,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6814759387403482e-05,
      "loss": 0.4003581523895264,
      "num_input_tokens_seen": 7044557952,
      "step": 24770,
      "train_runtime": 240899.2249,
      "train_tokens_per_second": 29242.759
    },
    {
      "epoch": 0.8767666135199146,
      "grad_norm": 0.6015625,
      "learning_rate": 2.681090408863477e-05,
      "loss": 0.4022761344909668,
      "num_input_tokens_seen": 7047406528,
      "step": 24780,
      "train_runtime": 240999.2404,
      "train_tokens_per_second": 29242.443
    },
    {
      "epoch": 0.8771204337836433,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6807050452277694e-05,
      "loss": 0.39827733039855956,
      "num_input_tokens_seen": 7050232384,
      "step": 24790,
      "train_runtime": 241093.8931,
      "train_tokens_per_second": 29242.683
    },
    {
      "epoch": 0.8774742540473721,
      "grad_norm": 0.62890625,
      "learning_rate": 2.6803198477137853e-05,
      "loss": 0.4002978324890137,
      "num_input_tokens_seen": 7053055232,
      "step": 24800,
      "train_runtime": 241188.5029,
      "train_tokens_per_second": 29242.916
    },
    {
      "epoch": 0.8778280743111009,
      "grad_norm": 0.61328125,
      "learning_rate": 2.6799348162022082e-05,
      "loss": 0.39804773330688475,
      "num_input_tokens_seen": 7055884096,
      "step": 24810,
      "train_runtime": 241284.7615,
      "train_tokens_per_second": 29242.974
    },
    {
      "epoch": 0.8781818945748296,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6795499505738397e-05,
      "loss": 0.404567289352417,
      "num_input_tokens_seen": 7058721152,
      "step": 24820,
      "train_runtime": 241381.3902,
      "train_tokens_per_second": 29243.021
    },
    {
      "epoch": 0.8785357148385584,
      "grad_norm": 0.59765625,
      "learning_rate": 2.679165250709601e-05,
      "loss": 0.4042099952697754,
      "num_input_tokens_seen": 7061523712,
      "step": 24830,
      "train_runtime": 241477.7887,
      "train_tokens_per_second": 29242.953
    },
    {
      "epoch": 0.8788895351022872,
      "grad_norm": 0.57421875,
      "learning_rate": 2.678780716490533e-05,
      "loss": 0.3966546058654785,
      "num_input_tokens_seen": 7064400064,
      "step": 24840,
      "train_runtime": 241578.7272,
      "train_tokens_per_second": 29242.641
    },
    {
      "epoch": 0.879243355366016,
      "grad_norm": 0.58984375,
      "learning_rate": 2.678396347797798e-05,
      "loss": 0.3962850570678711,
      "num_input_tokens_seen": 7067241536,
      "step": 24850,
      "train_runtime": 241673.9546,
      "train_tokens_per_second": 29242.876
    },
    {
      "epoch": 0.8795971756297448,
      "grad_norm": 0.5625,
      "learning_rate": 2.6780121445126756e-05,
      "loss": 0.3981971263885498,
      "num_input_tokens_seen": 7070086080,
      "step": 24860,
      "train_runtime": 241771.045,
      "train_tokens_per_second": 29242.898
    },
    {
      "epoch": 0.8799509958934736,
      "grad_norm": 0.5703125,
      "learning_rate": 2.6776281065165644e-05,
      "loss": 0.3988051414489746,
      "num_input_tokens_seen": 7072959040,
      "step": 24870,
      "train_runtime": 241869.7212,
      "train_tokens_per_second": 29242.846
    },
    {
      "epoch": 0.8803048161572024,
      "grad_norm": 0.5859375,
      "learning_rate": 2.677244233690983e-05,
      "loss": 0.3995032787322998,
      "num_input_tokens_seen": 7075815808,
      "step": 24880,
      "train_runtime": 241968.6112,
      "train_tokens_per_second": 29242.701
    },
    {
      "epoch": 0.8806586364209311,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6768605259175694e-05,
      "loss": 0.4009195327758789,
      "num_input_tokens_seen": 7078660800,
      "step": 24890,
      "train_runtime": 242068.0372,
      "train_tokens_per_second": 29242.443
    },
    {
      "epoch": 0.8810124566846599,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6764769830780784e-05,
      "loss": 0.4026834964752197,
      "num_input_tokens_seen": 7081475520,
      "step": 24900,
      "train_runtime": 242161.042,
      "train_tokens_per_second": 29242.836
    },
    {
      "epoch": 0.8813662769483886,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6760936050543857e-05,
      "loss": 0.399591326713562,
      "num_input_tokens_seen": 7084299776,
      "step": 24910,
      "train_runtime": 242256.131,
      "train_tokens_per_second": 29243.015
    },
    {
      "epoch": 0.8817200972121174,
      "grad_norm": 0.60546875,
      "learning_rate": 2.675710391728483e-05,
      "loss": 0.4019454002380371,
      "num_input_tokens_seen": 7087195264,
      "step": 24920,
      "train_runtime": 242356.2848,
      "train_tokens_per_second": 29242.878
    },
    {
      "epoch": 0.8820739174758463,
      "grad_norm": 0.5625,
      "learning_rate": 2.6753273429824822e-05,
      "loss": 0.40071897506713866,
      "num_input_tokens_seen": 7090094592,
      "step": 24930,
      "train_runtime": 242454.7721,
      "train_tokens_per_second": 29242.957
    },
    {
      "epoch": 0.8824277377395751,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6749444586986127e-05,
      "loss": 0.39962120056152345,
      "num_input_tokens_seen": 7092936384,
      "step": 24940,
      "train_runtime": 242551.4582,
      "train_tokens_per_second": 29243.017
    },
    {
      "epoch": 0.8827815580033038,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6745617387592214e-05,
      "loss": 0.39986209869384765,
      "num_input_tokens_seen": 7095776448,
      "step": 24950,
      "train_runtime": 242646.492,
      "train_tokens_per_second": 29243.268
    },
    {
      "epoch": 0.8831353782670326,
      "grad_norm": 0.578125,
      "learning_rate": 2.6741791830467728e-05,
      "loss": 0.39818081855773924,
      "num_input_tokens_seen": 7098612160,
      "step": 24960,
      "train_runtime": 242742.7557,
      "train_tokens_per_second": 29243.353
    },
    {
      "epoch": 0.8834891985307614,
      "grad_norm": 0.58203125,
      "learning_rate": 2.67379679144385e-05,
      "loss": 0.3946235179901123,
      "num_input_tokens_seen": 7101494464,
      "step": 24970,
      "train_runtime": 242841.5357,
      "train_tokens_per_second": 29243.327
    },
    {
      "epoch": 0.8838430187944901,
      "grad_norm": 0.609375,
      "learning_rate": 2.6734145638331536e-05,
      "loss": 0.3985111951828003,
      "num_input_tokens_seen": 7104353216,
      "step": 24980,
      "train_runtime": 242940.834,
      "train_tokens_per_second": 29243.142
    },
    {
      "epoch": 0.8841968390582189,
      "grad_norm": 0.5859375,
      "learning_rate": 2.673032500097501e-05,
      "loss": 0.40055456161499026,
      "num_input_tokens_seen": 7107186624,
      "step": 24990,
      "train_runtime": 243036.3765,
      "train_tokens_per_second": 29243.304
    },
    {
      "epoch": 0.8845506593219477,
      "grad_norm": 0.58203125,
      "learning_rate": 2.672650600119826e-05,
      "loss": 0.39823911190032957,
      "num_input_tokens_seen": 7110093696,
      "step": 25000,
      "train_runtime": 243137.6937,
      "train_tokens_per_second": 29243.075
    },
    {
      "epoch": 0.8845506593219477,
      "eval_loss": 0.3392401933670044,
      "eval_runtime": 8.4326,
      "eval_samples_per_second": 472.929,
      "eval_steps_per_second": 3.795,
      "num_input_tokens_seen": 7110093696,
      "step": 25000
    },
    {
      "epoch": 0.8849044795856764,
      "grad_norm": 0.6015625,
      "learning_rate": 2.672268863783181e-05,
      "loss": 0.4022231578826904,
      "num_input_tokens_seen": 7112954752,
      "step": 25010,
      "train_runtime": 243266.2006,
      "train_tokens_per_second": 29239.388
    },
    {
      "epoch": 0.8852582998494053,
      "grad_norm": 0.5703125,
      "learning_rate": 2.6718872909707347e-05,
      "loss": 0.40130186080932617,
      "num_input_tokens_seen": 7115726848,
      "step": 25020,
      "train_runtime": 243357.0304,
      "train_tokens_per_second": 29239.866
    },
    {
      "epoch": 0.8856121201131341,
      "grad_norm": 0.58984375,
      "learning_rate": 2.671505881565772e-05,
      "loss": 0.3980755567550659,
      "num_input_tokens_seen": 7118565376,
      "step": 25030,
      "train_runtime": 243455.1707,
      "train_tokens_per_second": 29239.738
    },
    {
      "epoch": 0.8859659403768628,
      "grad_norm": 0.58203125,
      "learning_rate": 2.671124635451696e-05,
      "loss": 0.3961543798446655,
      "num_input_tokens_seen": 7121407552,
      "step": 25040,
      "train_runtime": 243551.3333,
      "train_tokens_per_second": 29239.863
    },
    {
      "epoch": 0.8863197606405916,
      "grad_norm": 0.6328125,
      "learning_rate": 2.6707435525120237e-05,
      "loss": 0.3972560167312622,
      "num_input_tokens_seen": 7124235968,
      "step": 25050,
      "train_runtime": 243648.9175,
      "train_tokens_per_second": 29239.76
    },
    {
      "epoch": 0.8866735809043204,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6703626326303906e-05,
      "loss": 0.39791016578674315,
      "num_input_tokens_seen": 7127074624,
      "step": 25060,
      "train_runtime": 243747.0575,
      "train_tokens_per_second": 29239.633
    },
    {
      "epoch": 0.8870274011680491,
      "grad_norm": 0.6171875,
      "learning_rate": 2.669981875690547e-05,
      "loss": 0.40055389404296876,
      "num_input_tokens_seen": 7129976704,
      "step": 25070,
      "train_runtime": 243846.1321,
      "train_tokens_per_second": 29239.655
    },
    {
      "epoch": 0.8873812214317779,
      "grad_norm": 0.56640625,
      "learning_rate": 2.6696012815763604e-05,
      "loss": 0.4002669811248779,
      "num_input_tokens_seen": 7132845120,
      "step": 25080,
      "train_runtime": 243947.6133,
      "train_tokens_per_second": 29239.249
    },
    {
      "epoch": 0.8877350416955067,
      "grad_norm": 0.5703125,
      "learning_rate": 2.669220850171813e-05,
      "loss": 0.40085816383361816,
      "num_input_tokens_seen": 7135742464,
      "step": 25090,
      "train_runtime": 244046.1847,
      "train_tokens_per_second": 29239.312
    },
    {
      "epoch": 0.8880888619592355,
      "grad_norm": 0.5859375,
      "learning_rate": 2.668840581361003e-05,
      "loss": 0.40192832946777346,
      "num_input_tokens_seen": 7138578432,
      "step": 25100,
      "train_runtime": 244141.7837,
      "train_tokens_per_second": 29239.478
    },
    {
      "epoch": 0.8884426822229643,
      "grad_norm": 0.59765625,
      "learning_rate": 2.668460475028145e-05,
      "loss": 0.40016489028930663,
      "num_input_tokens_seen": 7141411008,
      "step": 25110,
      "train_runtime": 244238.6767,
      "train_tokens_per_second": 29239.476
    },
    {
      "epoch": 0.8887965024866931,
      "grad_norm": 0.62890625,
      "learning_rate": 2.6680805310575686e-05,
      "loss": 0.40253238677978515,
      "num_input_tokens_seen": 7144279872,
      "step": 25120,
      "train_runtime": 244338.9,
      "train_tokens_per_second": 29239.224
    },
    {
      "epoch": 0.8891503227504218,
      "grad_norm": 0.59375,
      "learning_rate": 2.6677007493337164e-05,
      "loss": 0.4012650489807129,
      "num_input_tokens_seen": 7147075776,
      "step": 25130,
      "train_runtime": 244432.7077,
      "train_tokens_per_second": 29239.441
    },
    {
      "epoch": 0.8895041430141506,
      "grad_norm": 0.578125,
      "learning_rate": 2.6673211297411503e-05,
      "loss": 0.39514741897583006,
      "num_input_tokens_seen": 7149911360,
      "step": 25140,
      "train_runtime": 244529.86,
      "train_tokens_per_second": 29239.42
    },
    {
      "epoch": 0.8898579632778794,
      "grad_norm": 0.61328125,
      "learning_rate": 2.6669416721645447e-05,
      "loss": 0.40055103302001954,
      "num_input_tokens_seen": 7152746304,
      "step": 25150,
      "train_runtime": 244625.3545,
      "train_tokens_per_second": 29239.595
    },
    {
      "epoch": 0.8902117835416081,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6665623764886876e-05,
      "loss": 0.3993120908737183,
      "num_input_tokens_seen": 7155609408,
      "step": 25160,
      "train_runtime": 244724.8088,
      "train_tokens_per_second": 29239.412
    },
    {
      "epoch": 0.8905656038053369,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6661832425984847e-05,
      "loss": 0.39555697441101073,
      "num_input_tokens_seen": 7158470784,
      "step": 25170,
      "train_runtime": 244824.7674,
      "train_tokens_per_second": 29239.161
    },
    {
      "epoch": 0.8909194240690658,
      "grad_norm": 0.58203125,
      "learning_rate": 2.665804270378953e-05,
      "loss": 0.4029559135437012,
      "num_input_tokens_seen": 7161297344,
      "step": 25180,
      "train_runtime": 244922.7688,
      "train_tokens_per_second": 29239.002
    },
    {
      "epoch": 0.8912732443327945,
      "grad_norm": 0.57421875,
      "learning_rate": 2.665425459715228e-05,
      "loss": 0.3993260383605957,
      "num_input_tokens_seen": 7164144192,
      "step": 25190,
      "train_runtime": 245021.0608,
      "train_tokens_per_second": 29238.891
    },
    {
      "epoch": 0.8916270645965233,
      "grad_norm": 0.609375,
      "learning_rate": 2.6650468104925547e-05,
      "loss": 0.3969135761260986,
      "num_input_tokens_seen": 7167005376,
      "step": 25200,
      "train_runtime": 245118.7175,
      "train_tokens_per_second": 29238.915
    },
    {
      "epoch": 0.8919808848602521,
      "grad_norm": 0.59765625,
      "learning_rate": 2.664668322596295e-05,
      "loss": 0.39873499870300294,
      "num_input_tokens_seen": 7169872000,
      "step": 25210,
      "train_runtime": 245219.1725,
      "train_tokens_per_second": 29238.627
    },
    {
      "epoch": 0.8923347051239808,
      "grad_norm": 0.62109375,
      "learning_rate": 2.664289995911925e-05,
      "loss": 0.3989361047744751,
      "num_input_tokens_seen": 7172743232,
      "step": 25220,
      "train_runtime": 245317.432,
      "train_tokens_per_second": 29238.62
    },
    {
      "epoch": 0.8926885253877096,
      "grad_norm": 0.60546875,
      "learning_rate": 2.6639118303250328e-05,
      "loss": 0.4051660537719727,
      "num_input_tokens_seen": 7175549440,
      "step": 25230,
      "train_runtime": 245413.0108,
      "train_tokens_per_second": 29238.668
    },
    {
      "epoch": 0.8930423456514384,
      "grad_norm": 0.6015625,
      "learning_rate": 2.663533825721321e-05,
      "loss": 0.400730037689209,
      "num_input_tokens_seen": 7178427264,
      "step": 25240,
      "train_runtime": 245515.5353,
      "train_tokens_per_second": 29238.179
    },
    {
      "epoch": 0.8933961659151671,
      "grad_norm": 0.6015625,
      "learning_rate": 2.6631559819866065e-05,
      "loss": 0.39889414310455323,
      "num_input_tokens_seen": 7181239232,
      "step": 25250,
      "train_runtime": 245610.5753,
      "train_tokens_per_second": 29238.314
    },
    {
      "epoch": 0.8937499861788959,
      "grad_norm": 0.578125,
      "learning_rate": 2.662778299006819e-05,
      "loss": 0.4035154342651367,
      "num_input_tokens_seen": 7184113536,
      "step": 25260,
      "train_runtime": 245708.786,
      "train_tokens_per_second": 29238.326
    },
    {
      "epoch": 0.8941038064426248,
      "grad_norm": 0.60546875,
      "learning_rate": 2.6624007766680004e-05,
      "loss": 0.4019301891326904,
      "num_input_tokens_seen": 7186963200,
      "step": 25270,
      "train_runtime": 245804.6284,
      "train_tokens_per_second": 29238.519
    },
    {
      "epoch": 0.8944576267063535,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6620234148563066e-05,
      "loss": 0.3962251663208008,
      "num_input_tokens_seen": 7189763328,
      "step": 25280,
      "train_runtime": 245899.5989,
      "train_tokens_per_second": 29238.613
    },
    {
      "epoch": 0.8948114469700823,
      "grad_norm": 0.6015625,
      "learning_rate": 2.6616462134580072e-05,
      "loss": 0.3975823402404785,
      "num_input_tokens_seen": 7192593280,
      "step": 25290,
      "train_runtime": 245997.0826,
      "train_tokens_per_second": 29238.531
    },
    {
      "epoch": 0.8951652672338111,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6612691723594835e-05,
      "loss": 0.397947883605957,
      "num_input_tokens_seen": 7195397888,
      "step": 25300,
      "train_runtime": 246092.1813,
      "train_tokens_per_second": 29238.629
    },
    {
      "epoch": 0.8955190874975398,
      "grad_norm": 0.60546875,
      "learning_rate": 2.6608922914472283e-05,
      "loss": 0.3953337907791138,
      "num_input_tokens_seen": 7198229056,
      "step": 25310,
      "train_runtime": 246188.5198,
      "train_tokens_per_second": 29238.687
    },
    {
      "epoch": 0.8958729077612686,
      "grad_norm": 0.5859375,
      "learning_rate": 2.66051557060785e-05,
      "loss": 0.4060483455657959,
      "num_input_tokens_seen": 7200989056,
      "step": 25320,
      "train_runtime": 246279.8988,
      "train_tokens_per_second": 29239.045
    },
    {
      "epoch": 0.8962267280249974,
      "grad_norm": 0.6015625,
      "learning_rate": 2.6601390097280665e-05,
      "loss": 0.3961021423339844,
      "num_input_tokens_seen": 7203796608,
      "step": 25330,
      "train_runtime": 246376.2437,
      "train_tokens_per_second": 29239.007
    },
    {
      "epoch": 0.8965805482887261,
      "grad_norm": 0.60546875,
      "learning_rate": 2.6597626086947097e-05,
      "loss": 0.40268239974975584,
      "num_input_tokens_seen": 7206658048,
      "step": 25340,
      "train_runtime": 246474.6913,
      "train_tokens_per_second": 29238.937
    },
    {
      "epoch": 0.896934368552455,
      "grad_norm": 0.59765625,
      "learning_rate": 2.659386367394722e-05,
      "loss": 0.4050790786743164,
      "num_input_tokens_seen": 7209520704,
      "step": 25350,
      "train_runtime": 246572.0756,
      "train_tokens_per_second": 29238.999
    },
    {
      "epoch": 0.8972881888161838,
      "grad_norm": 0.5625,
      "learning_rate": 2.65901028571516e-05,
      "loss": 0.39623663425445554,
      "num_input_tokens_seen": 7212382976,
      "step": 25360,
      "train_runtime": 246670.3547,
      "train_tokens_per_second": 29238.953
    },
    {
      "epoch": 0.8976420090799125,
      "grad_norm": 0.59375,
      "learning_rate": 2.6586343635431887e-05,
      "loss": 0.4008064270019531,
      "num_input_tokens_seen": 7215223936,
      "step": 25370,
      "train_runtime": 246766.3552,
      "train_tokens_per_second": 29239.091
    },
    {
      "epoch": 0.8979958293436413,
      "grad_norm": 0.6171875,
      "learning_rate": 2.658258600766088e-05,
      "loss": 0.40004734992980956,
      "num_input_tokens_seen": 7218056128,
      "step": 25380,
      "train_runtime": 246861.552,
      "train_tokens_per_second": 29239.288
    },
    {
      "epoch": 0.8983496496073701,
      "grad_norm": 0.61328125,
      "learning_rate": 2.657882997271247e-05,
      "loss": 0.4022518157958984,
      "num_input_tokens_seen": 7220825728,
      "step": 25390,
      "train_runtime": 246952.8938,
      "train_tokens_per_second": 29239.689
    },
    {
      "epoch": 0.8987034698710988,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6575075529461685e-05,
      "loss": 0.3985184669494629,
      "num_input_tokens_seen": 7223679360,
      "step": 25400,
      "train_runtime": 247052.6012,
      "train_tokens_per_second": 29239.439
    },
    {
      "epoch": 0.8990572901348276,
      "grad_norm": 0.578125,
      "learning_rate": 2.657132267678464e-05,
      "loss": 0.3990170001983643,
      "num_input_tokens_seen": 7226557696,
      "step": 25410,
      "train_runtime": 247152.0079,
      "train_tokens_per_second": 29239.324
    },
    {
      "epoch": 0.8994111103985564,
      "grad_norm": 0.6015625,
      "learning_rate": 2.656757141355857e-05,
      "loss": 0.4003419876098633,
      "num_input_tokens_seen": 7229419072,
      "step": 25420,
      "train_runtime": 247252.9209,
      "train_tokens_per_second": 29238.963
    },
    {
      "epoch": 0.8997649306622852,
      "grad_norm": 0.578125,
      "learning_rate": 2.6563821738661833e-05,
      "loss": 0.3981818675994873,
      "num_input_tokens_seen": 7232323200,
      "step": 25430,
      "train_runtime": 247351.7281,
      "train_tokens_per_second": 29239.024
    },
    {
      "epoch": 0.900118750926014,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6560073650973873e-05,
      "loss": 0.3981840372085571,
      "num_input_tokens_seen": 7235181440,
      "step": 25440,
      "train_runtime": 247448.0445,
      "train_tokens_per_second": 29239.194
    },
    {
      "epoch": 0.9004725711897428,
      "grad_norm": 0.609375,
      "learning_rate": 2.655632714937525e-05,
      "loss": 0.40260777473449705,
      "num_input_tokens_seen": 7238069568,
      "step": 25450,
      "train_runtime": 247547.4707,
      "train_tokens_per_second": 29239.117
    },
    {
      "epoch": 0.9008263914534715,
      "grad_norm": 0.60546875,
      "learning_rate": 2.6552582232747637e-05,
      "loss": 0.39693174362182615,
      "num_input_tokens_seen": 7240984192,
      "step": 25460,
      "train_runtime": 247648.7242,
      "train_tokens_per_second": 29238.932
    },
    {
      "epoch": 0.9011802117172003,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6548838899973794e-05,
      "loss": 0.3956117630004883,
      "num_input_tokens_seen": 7243801856,
      "step": 25470,
      "train_runtime": 247744.5052,
      "train_tokens_per_second": 29239.001
    },
    {
      "epoch": 0.9015340319809291,
      "grad_norm": 0.62109375,
      "learning_rate": 2.6545097149937598e-05,
      "loss": 0.3974839925765991,
      "num_input_tokens_seen": 7246685504,
      "step": 25480,
      "train_runtime": 247846.3743,
      "train_tokens_per_second": 29238.618
    },
    {
      "epoch": 0.9018878522446578,
      "grad_norm": 0.61328125,
      "learning_rate": 2.6541356981524018e-05,
      "loss": 0.39944212436676024,
      "num_input_tokens_seen": 7249484288,
      "step": 25490,
      "train_runtime": 247941.7978,
      "train_tokens_per_second": 29238.653
    },
    {
      "epoch": 0.9022416725083866,
      "grad_norm": 0.578125,
      "learning_rate": 2.653761839361913e-05,
      "loss": 0.39823575019836427,
      "num_input_tokens_seen": 7252360000,
      "step": 25500,
      "train_runtime": 248041.0483,
      "train_tokens_per_second": 29238.548
    },
    {
      "epoch": 0.9025954927721154,
      "grad_norm": 0.59375,
      "learning_rate": 2.6533881385110097e-05,
      "loss": 0.40156192779541017,
      "num_input_tokens_seen": 7255188736,
      "step": 25510,
      "train_runtime": 248136.859,
      "train_tokens_per_second": 29238.658
    },
    {
      "epoch": 0.9029493130358442,
      "grad_norm": 0.62109375,
      "learning_rate": 2.6530145954885187e-05,
      "loss": 0.40303540229797363,
      "num_input_tokens_seen": 7258065280,
      "step": 25520,
      "train_runtime": 248235.3909,
      "train_tokens_per_second": 29238.64
    },
    {
      "epoch": 0.903303133299573,
      "grad_norm": 0.57421875,
      "learning_rate": 2.6526412101833764e-05,
      "loss": 0.39902620315551757,
      "num_input_tokens_seen": 7260926016,
      "step": 25530,
      "train_runtime": 248332.3966,
      "train_tokens_per_second": 29238.739
    },
    {
      "epoch": 0.9036569535633018,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6522679824846287e-05,
      "loss": 0.39557886123657227,
      "num_input_tokens_seen": 7263808640,
      "step": 25540,
      "train_runtime": 248430.5092,
      "train_tokens_per_second": 29238.795
    },
    {
      "epoch": 0.9040107738270305,
      "grad_norm": 0.56640625,
      "learning_rate": 2.6518949122814297e-05,
      "loss": 0.39488649368286133,
      "num_input_tokens_seen": 7266746560,
      "step": 25550,
      "train_runtime": 248531.237,
      "train_tokens_per_second": 29238.766
    },
    {
      "epoch": 0.9043645940907593,
      "grad_norm": 0.60546875,
      "learning_rate": 2.651521999463043e-05,
      "loss": 0.3974877595901489,
      "num_input_tokens_seen": 7269593280,
      "step": 25560,
      "train_runtime": 248630.7839,
      "train_tokens_per_second": 29238.508
    },
    {
      "epoch": 0.9047184143544881,
      "grad_norm": 0.61328125,
      "learning_rate": 2.651149243918842e-05,
      "loss": 0.40156941413879393,
      "num_input_tokens_seen": 7272358272,
      "step": 25570,
      "train_runtime": 248721.3389,
      "train_tokens_per_second": 29238.98
    },
    {
      "epoch": 0.9050722346182168,
      "grad_norm": 0.57421875,
      "learning_rate": 2.650776645538308e-05,
      "loss": 0.39846274852752683,
      "num_input_tokens_seen": 7275191680,
      "step": 25580,
      "train_runtime": 248817.4997,
      "train_tokens_per_second": 29239.068
    },
    {
      "epoch": 0.9054260548819456,
      "grad_norm": 0.59375,
      "learning_rate": 2.650404204211032e-05,
      "loss": 0.40518627166748045,
      "num_input_tokens_seen": 7278065728,
      "step": 25590,
      "train_runtime": 248915.292,
      "train_tokens_per_second": 29239.127
    },
    {
      "epoch": 0.9057798751456745,
      "grad_norm": 0.6015625,
      "learning_rate": 2.6500319198267126e-05,
      "loss": 0.40259642601013185,
      "num_input_tokens_seen": 7280872960,
      "step": 25600,
      "train_runtime": 249011.8018,
      "train_tokens_per_second": 29239.068
    },
    {
      "epoch": 0.9061336954094033,
      "grad_norm": 0.62890625,
      "learning_rate": 2.6496597922751572e-05,
      "loss": 0.4010054588317871,
      "num_input_tokens_seen": 7283647360,
      "step": 25610,
      "train_runtime": 249104.7891,
      "train_tokens_per_second": 29239.291
    },
    {
      "epoch": 0.906487515673132,
      "grad_norm": 0.609375,
      "learning_rate": 2.6492878214462818e-05,
      "loss": 0.4013101100921631,
      "num_input_tokens_seen": 7286489344,
      "step": 25620,
      "train_runtime": 249202.7918,
      "train_tokens_per_second": 29239.196
    },
    {
      "epoch": 0.9068413359368608,
      "grad_norm": 0.57421875,
      "learning_rate": 2.648916007230109e-05,
      "loss": 0.39694030284881593,
      "num_input_tokens_seen": 7289338816,
      "step": 25630,
      "train_runtime": 249296.3978,
      "train_tokens_per_second": 29239.648
    },
    {
      "epoch": 0.9071951562005895,
      "grad_norm": 0.609375,
      "learning_rate": 2.6485443495167722e-05,
      "loss": 0.39302549362182615,
      "num_input_tokens_seen": 7292164416,
      "step": 25640,
      "train_runtime": 249391.2613,
      "train_tokens_per_second": 29239.855
    },
    {
      "epoch": 0.9075489764643183,
      "grad_norm": 0.578125,
      "learning_rate": 2.6481728481965102e-05,
      "loss": 0.3992034435272217,
      "num_input_tokens_seen": 7295053952,
      "step": 25650,
      "train_runtime": 249493.0045,
      "train_tokens_per_second": 29239.513
    },
    {
      "epoch": 0.9079027967280471,
      "grad_norm": 0.59375,
      "learning_rate": 2.6478015031596705e-05,
      "loss": 0.39673738479614257,
      "num_input_tokens_seen": 7297896512,
      "step": 25660,
      "train_runtime": 249589.8775,
      "train_tokens_per_second": 29239.553
    },
    {
      "epoch": 0.9082566169917758,
      "grad_norm": 0.6171875,
      "learning_rate": 2.6474303142967076e-05,
      "loss": 0.39556074142456055,
      "num_input_tokens_seen": 7300728320,
      "step": 25670,
      "train_runtime": 249686.3369,
      "train_tokens_per_second": 29239.599
    },
    {
      "epoch": 0.9086104372555047,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6470592814981852e-05,
      "loss": 0.40029754638671877,
      "num_input_tokens_seen": 7303594368,
      "step": 25680,
      "train_runtime": 249785.9909,
      "train_tokens_per_second": 29239.407
    },
    {
      "epoch": 0.9089642575192335,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6466884046547725e-05,
      "loss": 0.4009224891662598,
      "num_input_tokens_seen": 7306414336,
      "step": 25690,
      "train_runtime": 249882.4836,
      "train_tokens_per_second": 29239.402
    },
    {
      "epoch": 0.9093180777829623,
      "grad_norm": 0.58984375,
      "learning_rate": 2.646317683657247e-05,
      "loss": 0.4027262687683105,
      "num_input_tokens_seen": 7309275328,
      "step": 25700,
      "train_runtime": 249981.0619,
      "train_tokens_per_second": 29239.316
    },
    {
      "epoch": 0.909671898046691,
      "grad_norm": 0.59375,
      "learning_rate": 2.645947118396491e-05,
      "loss": 0.39997520446777346,
      "num_input_tokens_seen": 7312084864,
      "step": 25710,
      "train_runtime": 250075.4069,
      "train_tokens_per_second": 29239.52
    },
    {
      "epoch": 0.9100257183104198,
      "grad_norm": 0.59375,
      "learning_rate": 2.6455767087634982e-05,
      "loss": 0.4024055957794189,
      "num_input_tokens_seen": 7314917504,
      "step": 25720,
      "train_runtime": 250173.7783,
      "train_tokens_per_second": 29239.345
    },
    {
      "epoch": 0.9103795385741486,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6452064546493642e-05,
      "loss": 0.4042919635772705,
      "num_input_tokens_seen": 7317741568,
      "step": 25730,
      "train_runtime": 250270.8729,
      "train_tokens_per_second": 29239.286
    },
    {
      "epoch": 0.9107333588378773,
      "grad_norm": 0.5859375,
      "learning_rate": 2.644836355945295e-05,
      "loss": 0.4048019886016846,
      "num_input_tokens_seen": 7320575360,
      "step": 25740,
      "train_runtime": 250366.6792,
      "train_tokens_per_second": 29239.415
    },
    {
      "epoch": 0.9110871791016061,
      "grad_norm": 0.5703125,
      "learning_rate": 2.6444664125426012e-05,
      "loss": 0.4019166469573975,
      "num_input_tokens_seen": 7323435072,
      "step": 25750,
      "train_runtime": 250464.3066,
      "train_tokens_per_second": 29239.436
    },
    {
      "epoch": 0.9114409993653348,
      "grad_norm": 0.59375,
      "learning_rate": 2.6440966243326985e-05,
      "loss": 0.39493412971496583,
      "num_input_tokens_seen": 7326340928,
      "step": 25760,
      "train_runtime": 250563.9251,
      "train_tokens_per_second": 29239.408
    },
    {
      "epoch": 0.9117948196290637,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6437269912071132e-05,
      "loss": 0.40195207595825194,
      "num_input_tokens_seen": 7329211904,
      "step": 25770,
      "train_runtime": 250664.9738,
      "train_tokens_per_second": 29239.075
    },
    {
      "epoch": 0.9121486398927925,
      "grad_norm": 0.59375,
      "learning_rate": 2.6433575130574732e-05,
      "loss": 0.3957659721374512,
      "num_input_tokens_seen": 7332064704,
      "step": 25780,
      "train_runtime": 250764.9594,
      "train_tokens_per_second": 29238.793
    },
    {
      "epoch": 0.9125024601565213,
      "grad_norm": 0.6015625,
      "learning_rate": 2.6429881897755154e-05,
      "loss": 0.39867801666259767,
      "num_input_tokens_seen": 7334932160,
      "step": 25790,
      "train_runtime": 250861.5924,
      "train_tokens_per_second": 29238.96
    },
    {
      "epoch": 0.91285628042025,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6426190212530806e-05,
      "loss": 0.4010892391204834,
      "num_input_tokens_seen": 7337747072,
      "step": 25800,
      "train_runtime": 250956.3016,
      "train_tokens_per_second": 29239.143
    },
    {
      "epoch": 0.9132101006839788,
      "grad_norm": 0.5703125,
      "learning_rate": 2.6422500073821168e-05,
      "loss": 0.39749746322631835,
      "num_input_tokens_seen": 7340579904,
      "step": 25810,
      "train_runtime": 251052.6994,
      "train_tokens_per_second": 29239.199
    },
    {
      "epoch": 0.9135639209477076,
      "grad_norm": 0.61328125,
      "learning_rate": 2.6418811480546757e-05,
      "loss": 0.4014753341674805,
      "num_input_tokens_seen": 7343382400,
      "step": 25820,
      "train_runtime": 251149.0877,
      "train_tokens_per_second": 29239.136
    },
    {
      "epoch": 0.9139177412114363,
      "grad_norm": 0.60546875,
      "learning_rate": 2.641512443162917e-05,
      "loss": 0.4016080856323242,
      "num_input_tokens_seen": 7346285056,
      "step": 25830,
      "train_runtime": 251250.4477,
      "train_tokens_per_second": 29238.893
    },
    {
      "epoch": 0.9142715614751651,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6411438925991034e-05,
      "loss": 0.40257940292358396,
      "num_input_tokens_seen": 7349125248,
      "step": 25840,
      "train_runtime": 251349.1903,
      "train_tokens_per_second": 29238.707
    },
    {
      "epoch": 0.914625381738894,
      "grad_norm": 0.59375,
      "learning_rate": 2.6407754962556046e-05,
      "loss": 0.4022639274597168,
      "num_input_tokens_seen": 7351978560,
      "step": 25850,
      "train_runtime": 251444.1783,
      "train_tokens_per_second": 29239.009
    },
    {
      "epoch": 0.9149792020026227,
      "grad_norm": 0.59375,
      "learning_rate": 2.640407254024894e-05,
      "loss": 0.39742555618286135,
      "num_input_tokens_seen": 7354887424,
      "step": 25860,
      "train_runtime": 251544.593,
      "train_tokens_per_second": 29238.901
    },
    {
      "epoch": 0.9153330222663515,
      "grad_norm": 0.6171875,
      "learning_rate": 2.640039165799551e-05,
      "loss": 0.39743528366088865,
      "num_input_tokens_seen": 7357754112,
      "step": 25870,
      "train_runtime": 251642.4702,
      "train_tokens_per_second": 29238.92
    },
    {
      "epoch": 0.9156868425300803,
      "grad_norm": 0.609375,
      "learning_rate": 2.6396712314722586e-05,
      "loss": 0.4021452903747559,
      "num_input_tokens_seen": 7360567936,
      "step": 25880,
      "train_runtime": 251739.4888,
      "train_tokens_per_second": 29238.829
    },
    {
      "epoch": 0.916040662793809,
      "grad_norm": 0.55859375,
      "learning_rate": 2.6393034509358054e-05,
      "loss": 0.39592957496643066,
      "num_input_tokens_seen": 7363417344,
      "step": 25890,
      "train_runtime": 251838.2718,
      "train_tokens_per_second": 29238.675
    },
    {
      "epoch": 0.9163944830575378,
      "grad_norm": 0.6015625,
      "learning_rate": 2.6389358240830854e-05,
      "loss": 0.39899747371673583,
      "num_input_tokens_seen": 7366247552,
      "step": 25900,
      "train_runtime": 251932.9911,
      "train_tokens_per_second": 29238.916
    },
    {
      "epoch": 0.9167483033212666,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6385683508070942e-05,
      "loss": 0.4007877826690674,
      "num_input_tokens_seen": 7369084736,
      "step": 25910,
      "train_runtime": 252029.5419,
      "train_tokens_per_second": 29238.972
    },
    {
      "epoch": 0.9171021235849953,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6382010310009335e-05,
      "loss": 0.39722001552581787,
      "num_input_tokens_seen": 7371948800,
      "step": 25920,
      "train_runtime": 252127.8505,
      "train_tokens_per_second": 29238.931
    },
    {
      "epoch": 0.9174559438487242,
      "grad_norm": 0.5859375,
      "learning_rate": 2.63783386455781e-05,
      "loss": 0.3985212087631226,
      "num_input_tokens_seen": 7374769472,
      "step": 25930,
      "train_runtime": 252222.4581,
      "train_tokens_per_second": 29239.147
    },
    {
      "epoch": 0.917809764112453,
      "grad_norm": 0.59375,
      "learning_rate": 2.637466851371032e-05,
      "loss": 0.40396957397460936,
      "num_input_tokens_seen": 7377641152,
      "step": 25940,
      "train_runtime": 252322.3132,
      "train_tokens_per_second": 29238.957
    },
    {
      "epoch": 0.9181635843761817,
      "grad_norm": 0.58984375,
      "learning_rate": 2.637099991334015e-05,
      "loss": 0.39706361293792725,
      "num_input_tokens_seen": 7380539456,
      "step": 25950,
      "train_runtime": 252423.4736,
      "train_tokens_per_second": 29238.721
    },
    {
      "epoch": 0.9185174046399105,
      "grad_norm": 0.62109375,
      "learning_rate": 2.636733284340275e-05,
      "loss": 0.3972465038299561,
      "num_input_tokens_seen": 7383350272,
      "step": 25960,
      "train_runtime": 252518.1685,
      "train_tokens_per_second": 29238.887
    },
    {
      "epoch": 0.9188712249036393,
      "grad_norm": 0.5625,
      "learning_rate": 2.6363667302834326e-05,
      "loss": 0.3986637115478516,
      "num_input_tokens_seen": 7386183232,
      "step": 25970,
      "train_runtime": 252614.8926,
      "train_tokens_per_second": 29238.907
    },
    {
      "epoch": 0.919225045167368,
      "grad_norm": 0.60546875,
      "learning_rate": 2.636000329057214e-05,
      "loss": 0.39251768589019775,
      "num_input_tokens_seen": 7389062656,
      "step": 25980,
      "train_runtime": 252715.0361,
      "train_tokens_per_second": 29238.714
    },
    {
      "epoch": 0.9195788654310968,
      "grad_norm": 0.60546875,
      "learning_rate": 2.6356340805554452e-05,
      "loss": 0.4001704216003418,
      "num_input_tokens_seen": 7391905792,
      "step": 25990,
      "train_runtime": 252813.0801,
      "train_tokens_per_second": 29238.621
    },
    {
      "epoch": 0.9199326856948256,
      "grad_norm": 0.6015625,
      "learning_rate": 2.635267984672058e-05,
      "loss": 0.39476704597473145,
      "num_input_tokens_seen": 7394704704,
      "step": 26000,
      "train_runtime": 252907.1595,
      "train_tokens_per_second": 29238.811
    },
    {
      "epoch": 0.9202865059585543,
      "grad_norm": 0.5703125,
      "learning_rate": 2.6349020413010862e-05,
      "loss": 0.4021094799041748,
      "num_input_tokens_seen": 7397520192,
      "step": 26010,
      "train_runtime": 253003.469,
      "train_tokens_per_second": 29238.809
    },
    {
      "epoch": 0.9206403262222832,
      "grad_norm": 0.5625,
      "learning_rate": 2.6345362503366676e-05,
      "loss": 0.39967756271362304,
      "num_input_tokens_seen": 7400370176,
      "step": 26020,
      "train_runtime": 253103.9036,
      "train_tokens_per_second": 29238.467
    },
    {
      "epoch": 0.920994146486012,
      "grad_norm": 0.609375,
      "learning_rate": 2.634170611673042e-05,
      "loss": 0.40238361358642577,
      "num_input_tokens_seen": 7403118144,
      "step": 26030,
      "train_runtime": 253196.691,
      "train_tokens_per_second": 29238.605
    },
    {
      "epoch": 0.9213479667497407,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6338051252045516e-05,
      "loss": 0.3964683055877686,
      "num_input_tokens_seen": 7405969792,
      "step": 26040,
      "train_runtime": 253296.1415,
      "train_tokens_per_second": 29238.384
    },
    {
      "epoch": 0.9217017870134695,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6334397908256412e-05,
      "loss": 0.39841771125793457,
      "num_input_tokens_seen": 7408837952,
      "step": 26050,
      "train_runtime": 253392.2897,
      "train_tokens_per_second": 29238.608
    },
    {
      "epoch": 0.9220556072771983,
      "grad_norm": 0.62890625,
      "learning_rate": 2.63307460843086e-05,
      "loss": 0.3997972011566162,
      "num_input_tokens_seen": 7411620864,
      "step": 26060,
      "train_runtime": 253485.8687,
      "train_tokens_per_second": 29238.793
    },
    {
      "epoch": 0.922409427540927,
      "grad_norm": 0.578125,
      "learning_rate": 2.6327095779148576e-05,
      "loss": 0.40070438385009766,
      "num_input_tokens_seen": 7414539200,
      "step": 26070,
      "train_runtime": 253588.2792,
      "train_tokens_per_second": 29238.493
    },
    {
      "epoch": 0.9227632478046558,
      "grad_norm": 0.578125,
      "learning_rate": 2.6323446991723856e-05,
      "loss": 0.3976911544799805,
      "num_input_tokens_seen": 7417369856,
      "step": 26080,
      "train_runtime": 253681.817,
      "train_tokens_per_second": 29238.871
    },
    {
      "epoch": 0.9231170680683846,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6319799720982985e-05,
      "loss": 0.4010918617248535,
      "num_input_tokens_seen": 7420172992,
      "step": 26090,
      "train_runtime": 253777.73,
      "train_tokens_per_second": 29238.866
    },
    {
      "epoch": 0.9234708883321134,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6316153965875527e-05,
      "loss": 0.39762158393859864,
      "num_input_tokens_seen": 7422991104,
      "step": 26100,
      "train_runtime": 253873.1305,
      "train_tokens_per_second": 29238.979
    },
    {
      "epoch": 0.9238247085958422,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6312509725352074e-05,
      "loss": 0.40204601287841796,
      "num_input_tokens_seen": 7425821120,
      "step": 26110,
      "train_runtime": 253971.1555,
      "train_tokens_per_second": 29238.837
    },
    {
      "epoch": 0.924178528859571,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6308866998364208e-05,
      "loss": 0.396649694442749,
      "num_input_tokens_seen": 7428651392,
      "step": 26120,
      "train_runtime": 254066.4012,
      "train_tokens_per_second": 29239.015
    },
    {
      "epoch": 0.9245323491232997,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6305225783864556e-05,
      "loss": 0.39858369827270507,
      "num_input_tokens_seen": 7431515456,
      "step": 26130,
      "train_runtime": 254164.2058,
      "train_tokens_per_second": 29239.032
    },
    {
      "epoch": 0.9248861693870285,
      "grad_norm": 0.60546875,
      "learning_rate": 2.6301586080806734e-05,
      "loss": 0.3980096340179443,
      "num_input_tokens_seen": 7434345792,
      "step": 26140,
      "train_runtime": 254263.7872,
      "train_tokens_per_second": 29238.713
    },
    {
      "epoch": 0.9252399896507573,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6297947888145395e-05,
      "loss": 0.3948065280914307,
      "num_input_tokens_seen": 7437112960,
      "step": 26150,
      "train_runtime": 254356.4414,
      "train_tokens_per_second": 29238.941
    },
    {
      "epoch": 0.925593809914486,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6294311204836185e-05,
      "loss": 0.3992250919342041,
      "num_input_tokens_seen": 7440015104,
      "step": 26160,
      "train_runtime": 254456.0999,
      "train_tokens_per_second": 29238.895
    },
    {
      "epoch": 0.9259476301782148,
      "grad_norm": 0.6171875,
      "learning_rate": 2.6290676029835777e-05,
      "loss": 0.4019043445587158,
      "num_input_tokens_seen": 7442824320,
      "step": 26170,
      "train_runtime": 254551.2084,
      "train_tokens_per_second": 29239.006
    },
    {
      "epoch": 0.9263014504419437,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6287042362101834e-05,
      "loss": 0.40125131607055664,
      "num_input_tokens_seen": 7445649280,
      "step": 26180,
      "train_runtime": 254645.1906,
      "train_tokens_per_second": 29239.308
    },
    {
      "epoch": 0.9266552707056724,
      "grad_norm": 0.578125,
      "learning_rate": 2.628341020059304e-05,
      "loss": 0.3943807363510132,
      "num_input_tokens_seen": 7448544256,
      "step": 26190,
      "train_runtime": 254747.2541,
      "train_tokens_per_second": 29238.958
    },
    {
      "epoch": 0.9270090909694012,
      "grad_norm": 0.59375,
      "learning_rate": 2.627977954426909e-05,
      "loss": 0.39886474609375,
      "num_input_tokens_seen": 7451411648,
      "step": 26200,
      "train_runtime": 254846.7685,
      "train_tokens_per_second": 29238.792
    },
    {
      "epoch": 0.92736291123313,
      "grad_norm": 0.6171875,
      "learning_rate": 2.627615039209067e-05,
      "loss": 0.40076217651367185,
      "num_input_tokens_seen": 7454233792,
      "step": 26210,
      "train_runtime": 254942.4137,
      "train_tokens_per_second": 29238.892
    },
    {
      "epoch": 0.9277167314968587,
      "grad_norm": 0.6015625,
      "learning_rate": 2.6272522743019483e-05,
      "loss": 0.39693379402160645,
      "num_input_tokens_seen": 7457067968,
      "step": 26220,
      "train_runtime": 255041.4614,
      "train_tokens_per_second": 29238.65
    },
    {
      "epoch": 0.9280705517605875,
      "grad_norm": 0.609375,
      "learning_rate": 2.6268896596018234e-05,
      "loss": 0.3957505226135254,
      "num_input_tokens_seen": 7459882048,
      "step": 26230,
      "train_runtime": 255134.9414,
      "train_tokens_per_second": 29238.967
    },
    {
      "epoch": 0.9284243720243163,
      "grad_norm": 0.62109375,
      "learning_rate": 2.6265271950050624e-05,
      "loss": 0.3964117765426636,
      "num_input_tokens_seen": 7462726464,
      "step": 26240,
      "train_runtime": 255232.2901,
      "train_tokens_per_second": 29238.959
    },
    {
      "epoch": 0.928778192288045,
      "grad_norm": 0.57421875,
      "learning_rate": 2.626164880408136e-05,
      "loss": 0.3972675085067749,
      "num_input_tokens_seen": 7465592512,
      "step": 26250,
      "train_runtime": 255330.8272,
      "train_tokens_per_second": 29238.9
    },
    {
      "epoch": 0.9291320125517738,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6258027157076137e-05,
      "loss": 0.3954314231872559,
      "num_input_tokens_seen": 7468436288,
      "step": 26260,
      "train_runtime": 255429.3498,
      "train_tokens_per_second": 29238.755
    },
    {
      "epoch": 0.9294858328155027,
      "grad_norm": 0.62890625,
      "learning_rate": 2.625440700800167e-05,
      "loss": 0.3998711109161377,
      "num_input_tokens_seen": 7471270400,
      "step": 26270,
      "train_runtime": 255527.3188,
      "train_tokens_per_second": 29238.637
    },
    {
      "epoch": 0.9298396530792314,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6250788355825646e-05,
      "loss": 0.3990406274795532,
      "num_input_tokens_seen": 7474175040,
      "step": 26280,
      "train_runtime": 255628.7445,
      "train_tokens_per_second": 29238.398
    },
    {
      "epoch": 0.9301934733429602,
      "grad_norm": 0.59375,
      "learning_rate": 2.6247171199516767e-05,
      "loss": 0.39998886585235593,
      "num_input_tokens_seen": 7477086272,
      "step": 26290,
      "train_runtime": 255729.8865,
      "train_tokens_per_second": 29238.218
    },
    {
      "epoch": 0.930547293606689,
      "grad_norm": 0.56640625,
      "learning_rate": 2.6243555538044717e-05,
      "loss": 0.3982580900192261,
      "num_input_tokens_seen": 7479947072,
      "step": 26300,
      "train_runtime": 255828.4484,
      "train_tokens_per_second": 29238.136
    },
    {
      "epoch": 0.9309011138704177,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6239941370380184e-05,
      "loss": 0.39694933891296386,
      "num_input_tokens_seen": 7482773952,
      "step": 26310,
      "train_runtime": 255925.0286,
      "train_tokens_per_second": 29238.148
    },
    {
      "epoch": 0.9312549341341465,
      "grad_norm": 0.60546875,
      "learning_rate": 2.6236328695494834e-05,
      "loss": 0.39450922012329104,
      "num_input_tokens_seen": 7485591360,
      "step": 26320,
      "train_runtime": 256019.371,
      "train_tokens_per_second": 29238.379
    },
    {
      "epoch": 0.9316087543978753,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6232717512361343e-05,
      "loss": 0.39752721786499023,
      "num_input_tokens_seen": 7488479360,
      "step": 26330,
      "train_runtime": 256119.0977,
      "train_tokens_per_second": 29238.27
    },
    {
      "epoch": 0.931962574661604,
      "grad_norm": 0.59375,
      "learning_rate": 2.6229107819953357e-05,
      "loss": 0.3971027612686157,
      "num_input_tokens_seen": 7491326208,
      "step": 26340,
      "train_runtime": 256215.9879,
      "train_tokens_per_second": 29238.325
    },
    {
      "epoch": 0.9323163949253329,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6225499617245525e-05,
      "loss": 0.401854133605957,
      "num_input_tokens_seen": 7494171328,
      "step": 26350,
      "train_runtime": 256310.7447,
      "train_tokens_per_second": 29238.616
    },
    {
      "epoch": 0.9326702151890617,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6221892903213473e-05,
      "loss": 0.4006021976470947,
      "num_input_tokens_seen": 7496991168,
      "step": 26360,
      "train_runtime": 256406.7733,
      "train_tokens_per_second": 29238.663
    },
    {
      "epoch": 0.9330240354527904,
      "grad_norm": 0.6328125,
      "learning_rate": 2.621828767683382e-05,
      "loss": 0.3983322620391846,
      "num_input_tokens_seen": 7499831552,
      "step": 26370,
      "train_runtime": 256505.1867,
      "train_tokens_per_second": 29238.518
    },
    {
      "epoch": 0.9333778557165192,
      "grad_norm": 0.60546875,
      "learning_rate": 2.6214683937084162e-05,
      "loss": 0.40112690925598143,
      "num_input_tokens_seen": 7502663936,
      "step": 26380,
      "train_runtime": 256600.5239,
      "train_tokens_per_second": 29238.693
    },
    {
      "epoch": 0.933731675980248,
      "grad_norm": 0.61328125,
      "learning_rate": 2.621108168294309e-05,
      "loss": 0.40163602828979494,
      "num_input_tokens_seen": 7505471744,
      "step": 26390,
      "train_runtime": 256694.7296,
      "train_tokens_per_second": 29238.901
    },
    {
      "epoch": 0.9340854962439767,
      "grad_norm": 0.58984375,
      "learning_rate": 2.620748091339016e-05,
      "loss": 0.3998089790344238,
      "num_input_tokens_seen": 7508353152,
      "step": 26400,
      "train_runtime": 256789.0575,
      "train_tokens_per_second": 29239.381
    },
    {
      "epoch": 0.9344393165077055,
      "grad_norm": 0.61328125,
      "learning_rate": 2.620388162740593e-05,
      "loss": 0.4036846160888672,
      "num_input_tokens_seen": 7511148928,
      "step": 26410,
      "train_runtime": 256883.6016,
      "train_tokens_per_second": 29239.503
    },
    {
      "epoch": 0.9347931367714343,
      "grad_norm": 0.58984375,
      "learning_rate": 2.620028382397192e-05,
      "loss": 0.3989966154098511,
      "num_input_tokens_seen": 7514030912,
      "step": 26420,
      "train_runtime": 256982.5912,
      "train_tokens_per_second": 29239.455
    },
    {
      "epoch": 0.9351469570351632,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6196687502070645e-05,
      "loss": 0.40350589752197263,
      "num_input_tokens_seen": 7516908224,
      "step": 26430,
      "train_runtime": 257080.4288,
      "train_tokens_per_second": 29239.52
    },
    {
      "epoch": 0.9355007772988919,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6193092660685577e-05,
      "loss": 0.397203254699707,
      "num_input_tokens_seen": 7519726528,
      "step": 26440,
      "train_runtime": 257172.7595,
      "train_tokens_per_second": 29239.981
    },
    {
      "epoch": 0.9358545975626207,
      "grad_norm": 0.6015625,
      "learning_rate": 2.618949929880118e-05,
      "loss": 0.4003242015838623,
      "num_input_tokens_seen": 7522534016,
      "step": 26450,
      "train_runtime": 257268.5171,
      "train_tokens_per_second": 29240.01
    },
    {
      "epoch": 0.9362084178263494,
      "grad_norm": 0.60546875,
      "learning_rate": 2.618590741540289e-05,
      "loss": 0.3972476959228516,
      "num_input_tokens_seen": 7525328512,
      "step": 26460,
      "train_runtime": 257360.7375,
      "train_tokens_per_second": 29240.391
    },
    {
      "epoch": 0.9365622380900782,
      "grad_norm": 0.59375,
      "learning_rate": 2.6182317009477104e-05,
      "loss": 0.3991569519042969,
      "num_input_tokens_seen": 7528202048,
      "step": 26470,
      "train_runtime": 257459.2788,
      "train_tokens_per_second": 29240.36
    },
    {
      "epoch": 0.936916058353807,
      "grad_norm": 0.609375,
      "learning_rate": 2.617872808001122e-05,
      "loss": 0.3979495525360107,
      "num_input_tokens_seen": 7531066304,
      "step": 26480,
      "train_runtime": 257557.1297,
      "train_tokens_per_second": 29240.372
    },
    {
      "epoch": 0.9372698786175357,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6175140625993583e-05,
      "loss": 0.3976877689361572,
      "num_input_tokens_seen": 7533873984,
      "step": 26490,
      "train_runtime": 257652.221,
      "train_tokens_per_second": 29240.478
    },
    {
      "epoch": 0.9376236988812645,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6171554646413508e-05,
      "loss": 0.39845259189605714,
      "num_input_tokens_seen": 7536732416,
      "step": 26500,
      "train_runtime": 257750.4679,
      "train_tokens_per_second": 29240.422
    },
    {
      "epoch": 0.9379775191449934,
      "grad_norm": 0.578125,
      "learning_rate": 2.61679701402613e-05,
      "loss": 0.3973482847213745,
      "num_input_tokens_seen": 7539570496,
      "step": 26510,
      "train_runtime": 257849.0575,
      "train_tokens_per_second": 29240.248
    },
    {
      "epoch": 0.9383313394087222,
      "grad_norm": 0.59375,
      "learning_rate": 2.6164387106528197e-05,
      "loss": 0.40004758834838866,
      "num_input_tokens_seen": 7542419776,
      "step": 26520,
      "train_runtime": 257948.9355,
      "train_tokens_per_second": 29239.972
    },
    {
      "epoch": 0.9386851596724509,
      "grad_norm": 0.6015625,
      "learning_rate": 2.6160805544206447e-05,
      "loss": 0.3993862152099609,
      "num_input_tokens_seen": 7545247616,
      "step": 26530,
      "train_runtime": 258046.3643,
      "train_tokens_per_second": 29239.891
    },
    {
      "epoch": 0.9390389799361797,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6157225452289223e-05,
      "loss": 0.39635534286499025,
      "num_input_tokens_seen": 7548103680,
      "step": 26540,
      "train_runtime": 258141.6549,
      "train_tokens_per_second": 29240.162
    },
    {
      "epoch": 0.9393928001999085,
      "grad_norm": 0.6015625,
      "learning_rate": 2.6153646829770696e-05,
      "loss": 0.39853811264038086,
      "num_input_tokens_seen": 7550937152,
      "step": 26550,
      "train_runtime": 258239.1635,
      "train_tokens_per_second": 29240.093
    },
    {
      "epoch": 0.9397466204636372,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6150069675645972e-05,
      "loss": 0.39964966773986815,
      "num_input_tokens_seen": 7553826304,
      "step": 26560,
      "train_runtime": 258339.811,
      "train_tokens_per_second": 29239.885
    },
    {
      "epoch": 0.940100440727366,
      "grad_norm": 0.625,
      "learning_rate": 2.6146493988911137e-05,
      "loss": 0.4021864891052246,
      "num_input_tokens_seen": 7556638656,
      "step": 26570,
      "train_runtime": 258434.1137,
      "train_tokens_per_second": 29240.097
    },
    {
      "epoch": 0.9404542609910947,
      "grad_norm": 0.6328125,
      "learning_rate": 2.6142919768563224e-05,
      "loss": 0.4010316848754883,
      "num_input_tokens_seen": 7559477248,
      "step": 26580,
      "train_runtime": 258530.2279,
      "train_tokens_per_second": 29240.206
    },
    {
      "epoch": 0.9408080812548235,
      "grad_norm": 0.58984375,
      "learning_rate": 2.613934701360024e-05,
      "loss": 0.40176992416381835,
      "num_input_tokens_seen": 7562282368,
      "step": 26590,
      "train_runtime": 258623.3783,
      "train_tokens_per_second": 29240.521
    },
    {
      "epoch": 0.9411619015185524,
      "grad_norm": 0.609375,
      "learning_rate": 2.6135775723021143e-05,
      "loss": 0.396843409538269,
      "num_input_tokens_seen": 7565127424,
      "step": 26600,
      "train_runtime": 258719.4161,
      "train_tokens_per_second": 29240.664
    },
    {
      "epoch": 0.9415157217822812,
      "grad_norm": 0.59375,
      "learning_rate": 2.613220589582585e-05,
      "loss": 0.4043242931365967,
      "num_input_tokens_seen": 7568018496,
      "step": 26610,
      "train_runtime": 258819.3856,
      "train_tokens_per_second": 29240.54
    },
    {
      "epoch": 0.9418695420460099,
      "grad_norm": 0.6171875,
      "learning_rate": 2.612863753101522e-05,
      "loss": 0.39943852424621584,
      "num_input_tokens_seen": 7570843008,
      "step": 26620,
      "train_runtime": 258913.5032,
      "train_tokens_per_second": 29240.819
    },
    {
      "epoch": 0.9422233623097387,
      "grad_norm": 0.5703125,
      "learning_rate": 2.612507062759109e-05,
      "loss": 0.4017926216125488,
      "num_input_tokens_seen": 7573606592,
      "step": 26630,
      "train_runtime": 259004.5437,
      "train_tokens_per_second": 29241.211
    },
    {
      "epoch": 0.9425771825734675,
      "grad_norm": 0.59375,
      "learning_rate": 2.612150518455625e-05,
      "loss": 0.3972273349761963,
      "num_input_tokens_seen": 7576424576,
      "step": 26640,
      "train_runtime": 259102.0225,
      "train_tokens_per_second": 29241.086
    },
    {
      "epoch": 0.9429310028371962,
      "grad_norm": 0.60546875,
      "learning_rate": 2.6117941200914407e-05,
      "loss": 0.39843101501464845,
      "num_input_tokens_seen": 7579285376,
      "step": 26650,
      "train_runtime": 259197.1733,
      "train_tokens_per_second": 29241.389
    },
    {
      "epoch": 0.943284823100925,
      "grad_norm": 0.57421875,
      "learning_rate": 2.6114378675670266e-05,
      "loss": 0.3988903045654297,
      "num_input_tokens_seen": 7582086912,
      "step": 26660,
      "train_runtime": 259290.5894,
      "train_tokens_per_second": 29241.659
    },
    {
      "epoch": 0.9436386433646538,
      "grad_norm": 0.58203125,
      "learning_rate": 2.611081760782944e-05,
      "loss": 0.39679226875305174,
      "num_input_tokens_seen": 7584917376,
      "step": 26670,
      "train_runtime": 259386.5904,
      "train_tokens_per_second": 29241.748
    },
    {
      "epoch": 0.9439924636283826,
      "grad_norm": 0.56640625,
      "learning_rate": 2.6107257996398533e-05,
      "loss": 0.3978052854537964,
      "num_input_tokens_seen": 7587733568,
      "step": 26680,
      "train_runtime": 259482.9664,
      "train_tokens_per_second": 29241.74
    },
    {
      "epoch": 0.9443462838921114,
      "grad_norm": 0.609375,
      "learning_rate": 2.610369984038506e-05,
      "loss": 0.4018243312835693,
      "num_input_tokens_seen": 7590547328,
      "step": 26690,
      "train_runtime": 259576.8924,
      "train_tokens_per_second": 29241.999
    },
    {
      "epoch": 0.9447001041558402,
      "grad_norm": 0.6015625,
      "learning_rate": 2.6100143138797508e-05,
      "loss": 0.3984825134277344,
      "num_input_tokens_seen": 7593391104,
      "step": 26700,
      "train_runtime": 259673.2033,
      "train_tokens_per_second": 29242.105
    },
    {
      "epoch": 0.9450539244195689,
      "grad_norm": 0.58203125,
      "learning_rate": 2.6096587890645286e-05,
      "loss": 0.4009200096130371,
      "num_input_tokens_seen": 7596243136,
      "step": 26710,
      "train_runtime": 259771.0683,
      "train_tokens_per_second": 29242.068
    },
    {
      "epoch": 0.9454077446832977,
      "grad_norm": 0.578125,
      "learning_rate": 2.6093034094938772e-05,
      "loss": 0.4034112453460693,
      "num_input_tokens_seen": 7599072192,
      "step": 26720,
      "train_runtime": 259867.6947,
      "train_tokens_per_second": 29242.081
    },
    {
      "epoch": 0.9457615649470265,
      "grad_norm": 0.6015625,
      "learning_rate": 2.608948175068927e-05,
      "loss": 0.3975165843963623,
      "num_input_tokens_seen": 7601949120,
      "step": 26730,
      "train_runtime": 259964.2155,
      "train_tokens_per_second": 29242.291
    },
    {
      "epoch": 0.9461153852107552,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6085930856909035e-05,
      "loss": 0.39958236217498777,
      "num_input_tokens_seen": 7604805952,
      "step": 26740,
      "train_runtime": 260060.0049,
      "train_tokens_per_second": 29242.505
    },
    {
      "epoch": 0.946469205474484,
      "grad_norm": 0.5859375,
      "learning_rate": 2.608238141261126e-05,
      "loss": 0.39940547943115234,
      "num_input_tokens_seen": 7607641280,
      "step": 26750,
      "train_runtime": 260155.1379,
      "train_tokens_per_second": 29242.71
    },
    {
      "epoch": 0.9468230257382129,
      "grad_norm": 0.609375,
      "learning_rate": 2.6078833416810072e-05,
      "loss": 0.39827785491943357,
      "num_input_tokens_seen": 7610425536,
      "step": 26760,
      "train_runtime": 260248.7823,
      "train_tokens_per_second": 29242.886
    },
    {
      "epoch": 0.9471768460019416,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6075286868520543e-05,
      "loss": 0.3989081382751465,
      "num_input_tokens_seen": 7613237632,
      "step": 26770,
      "train_runtime": 260342.3794,
      "train_tokens_per_second": 29243.175
    },
    {
      "epoch": 0.9475306662656704,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6071741766758678e-05,
      "loss": 0.4026822090148926,
      "num_input_tokens_seen": 7616021312,
      "step": 26780,
      "train_runtime": 260437.1807,
      "train_tokens_per_second": 29243.218
    },
    {
      "epoch": 0.9478844865293992,
      "grad_norm": 0.5703125,
      "learning_rate": 2.6068198110541432e-05,
      "loss": 0.4010054111480713,
      "num_input_tokens_seen": 7618848512,
      "step": 26790,
      "train_runtime": 260534.1842,
      "train_tokens_per_second": 29243.182
    },
    {
      "epoch": 0.9482383067931279,
      "grad_norm": 0.59765625,
      "learning_rate": 2.606465589888667e-05,
      "loss": 0.401930570602417,
      "num_input_tokens_seen": 7621735040,
      "step": 26800,
      "train_runtime": 260633.4887,
      "train_tokens_per_second": 29243.115
    },
    {
      "epoch": 0.9485921270568567,
      "grad_norm": 0.57421875,
      "learning_rate": 2.606111513081321e-05,
      "loss": 0.3994897365570068,
      "num_input_tokens_seen": 7624568320,
      "step": 26810,
      "train_runtime": 260729.2195,
      "train_tokens_per_second": 29243.245
    },
    {
      "epoch": 0.9489459473205855,
      "grad_norm": 0.6171875,
      "learning_rate": 2.60575758053408e-05,
      "loss": 0.39697816371917727,
      "num_input_tokens_seen": 7627436160,
      "step": 26820,
      "train_runtime": 260828.2178,
      "train_tokens_per_second": 29243.14
    },
    {
      "epoch": 0.9492997675843142,
      "grad_norm": 0.58984375,
      "learning_rate": 2.605403792149012e-05,
      "loss": 0.396331787109375,
      "num_input_tokens_seen": 7630342400,
      "step": 26830,
      "train_runtime": 260928.4661,
      "train_tokens_per_second": 29243.043
    },
    {
      "epoch": 0.949653587848043,
      "grad_norm": 0.609375,
      "learning_rate": 2.6050501478282775e-05,
      "loss": 0.3953064203262329,
      "num_input_tokens_seen": 7633210560,
      "step": 26840,
      "train_runtime": 261028.1649,
      "train_tokens_per_second": 29242.862
    },
    {
      "epoch": 0.9500074081117719,
      "grad_norm": 0.59375,
      "learning_rate": 2.6046966474741293e-05,
      "loss": 0.39481778144836427,
      "num_input_tokens_seen": 7636083072,
      "step": 26850,
      "train_runtime": 261124.3796,
      "train_tokens_per_second": 29243.087
    },
    {
      "epoch": 0.9503612283755006,
      "grad_norm": 0.59375,
      "learning_rate": 2.6043432909889148e-05,
      "loss": 0.397759222984314,
      "num_input_tokens_seen": 7638949888,
      "step": 26860,
      "train_runtime": 261220.6694,
      "train_tokens_per_second": 29243.283
    },
    {
      "epoch": 0.9507150486392294,
      "grad_norm": 0.5703125,
      "learning_rate": 2.6039900782750737e-05,
      "loss": 0.39381046295166017,
      "num_input_tokens_seen": 7641783168,
      "step": 26870,
      "train_runtime": 261316.8062,
      "train_tokens_per_second": 29243.367
    },
    {
      "epoch": 0.9510688689029582,
      "grad_norm": 0.61328125,
      "learning_rate": 2.6036370092351377e-05,
      "loss": 0.39339656829833985,
      "num_input_tokens_seen": 7644615424,
      "step": 26880,
      "train_runtime": 261414.7572,
      "train_tokens_per_second": 29243.244
    },
    {
      "epoch": 0.9514226891666869,
      "grad_norm": 0.59765625,
      "learning_rate": 2.6032840837717298e-05,
      "loss": 0.4019001007080078,
      "num_input_tokens_seen": 7647469376,
      "step": 26890,
      "train_runtime": 261513.0631,
      "train_tokens_per_second": 29243.164
    },
    {
      "epoch": 0.9517765094304157,
      "grad_norm": 0.57421875,
      "learning_rate": 2.6029313017875682e-05,
      "loss": 0.400600004196167,
      "num_input_tokens_seen": 7650284608,
      "step": 26900,
      "train_runtime": 261607.3265,
      "train_tokens_per_second": 29243.388
    },
    {
      "epoch": 0.9521303296941445,
      "grad_norm": 0.58984375,
      "learning_rate": 2.602578663185461e-05,
      "loss": 0.40262789726257325,
      "num_input_tokens_seen": 7653107456,
      "step": 26910,
      "train_runtime": 261706.5654,
      "train_tokens_per_second": 29243.085
    },
    {
      "epoch": 0.9524841499578732,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6022261678683102e-05,
      "loss": 0.3949984073638916,
      "num_input_tokens_seen": 7655948864,
      "step": 26920,
      "train_runtime": 261804.2575,
      "train_tokens_per_second": 29243.027
    },
    {
      "epoch": 0.9528379702216021,
      "grad_norm": 0.60546875,
      "learning_rate": 2.601873815739108e-05,
      "loss": 0.3995465040206909,
      "num_input_tokens_seen": 7658779392,
      "step": 26930,
      "train_runtime": 261901.3699,
      "train_tokens_per_second": 29242.991
    },
    {
      "epoch": 0.9531917904853309,
      "grad_norm": 0.6015625,
      "learning_rate": 2.6015216067009395e-05,
      "loss": 0.3973482847213745,
      "num_input_tokens_seen": 7661614784,
      "step": 26940,
      "train_runtime": 261996.1394,
      "train_tokens_per_second": 29243.235
    },
    {
      "epoch": 0.9535456107490596,
      "grad_norm": 0.56640625,
      "learning_rate": 2.6011695406569824e-05,
      "loss": 0.39684038162231444,
      "num_input_tokens_seen": 7664483008,
      "step": 26950,
      "train_runtime": 262094.8116,
      "train_tokens_per_second": 29243.17
    },
    {
      "epoch": 0.9538994310127884,
      "grad_norm": 0.5859375,
      "learning_rate": 2.6008176175105048e-05,
      "loss": 0.3989076614379883,
      "num_input_tokens_seen": 7667332416,
      "step": 26960,
      "train_runtime": 262189.4958,
      "train_tokens_per_second": 29243.477
    },
    {
      "epoch": 0.9542532512765172,
      "grad_norm": 0.609375,
      "learning_rate": 2.6004658371648672e-05,
      "loss": 0.4005279541015625,
      "num_input_tokens_seen": 7670159680,
      "step": 26970,
      "train_runtime": 262285.7287,
      "train_tokens_per_second": 29243.527
    },
    {
      "epoch": 0.9546070715402459,
      "grad_norm": 0.58984375,
      "learning_rate": 2.6001141995235195e-05,
      "loss": 0.39890542030334475,
      "num_input_tokens_seen": 7673056704,
      "step": 26980,
      "train_runtime": 262384.7522,
      "train_tokens_per_second": 29243.531
    },
    {
      "epoch": 0.9549608918039747,
      "grad_norm": 0.5703125,
      "learning_rate": 2.599762704490007e-05,
      "loss": 0.40028767585754393,
      "num_input_tokens_seen": 7675956352,
      "step": 26990,
      "train_runtime": 262486.2876,
      "train_tokens_per_second": 29243.266
    },
    {
      "epoch": 0.9553147120677035,
      "grad_norm": 0.59765625,
      "learning_rate": 2.599411351967963e-05,
      "loss": 0.39893522262573244,
      "num_input_tokens_seen": 7678770112,
      "step": 27000,
      "train_runtime": 262582.6036,
      "train_tokens_per_second": 29243.255
    },
    {
      "epoch": 0.9556685323314323,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5990601418611127e-05,
      "loss": 0.3983405590057373,
      "num_input_tokens_seen": 7681633920,
      "step": 27010,
      "train_runtime": 262682.5763,
      "train_tokens_per_second": 29243.028
    },
    {
      "epoch": 0.9560223525951611,
      "grad_norm": 0.62109375,
      "learning_rate": 2.598709074073272e-05,
      "loss": 0.39721479415893557,
      "num_input_tokens_seen": 7684447104,
      "step": 27020,
      "train_runtime": 262776.9718,
      "train_tokens_per_second": 29243.229
    },
    {
      "epoch": 0.9563761728588899,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5983581485083498e-05,
      "loss": 0.4035499095916748,
      "num_input_tokens_seen": 7687253632,
      "step": 27030,
      "train_runtime": 262873.4507,
      "train_tokens_per_second": 29243.172
    },
    {
      "epoch": 0.9567299931226186,
      "grad_norm": 0.58984375,
      "learning_rate": 2.598007365070343e-05,
      "loss": 0.40035438537597656,
      "num_input_tokens_seen": 7690112640,
      "step": 27040,
      "train_runtime": 262973.2914,
      "train_tokens_per_second": 29242.942
    },
    {
      "epoch": 0.9570838133863474,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5976567236633403e-05,
      "loss": 0.40668430328369143,
      "num_input_tokens_seen": 7692974400,
      "step": 27050,
      "train_runtime": 263072.4014,
      "train_tokens_per_second": 29242.803
    },
    {
      "epoch": 0.9574376336500762,
      "grad_norm": 0.6328125,
      "learning_rate": 2.5973062241915225e-05,
      "loss": 0.39524426460266116,
      "num_input_tokens_seen": 7695779648,
      "step": 27060,
      "train_runtime": 263169.327,
      "train_tokens_per_second": 29242.692
    },
    {
      "epoch": 0.9577914539138049,
      "grad_norm": 0.61328125,
      "learning_rate": 2.5969558665591576e-05,
      "loss": 0.4012258529663086,
      "num_input_tokens_seen": 7698627264,
      "step": 27070,
      "train_runtime": 263267.095,
      "train_tokens_per_second": 29242.649
    },
    {
      "epoch": 0.9581452741775337,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5966056506706078e-05,
      "loss": 0.40055179595947266,
      "num_input_tokens_seen": 7701469120,
      "step": 27080,
      "train_runtime": 263365.9532,
      "train_tokens_per_second": 29242.463
    },
    {
      "epoch": 0.9584990944412625,
      "grad_norm": 0.61328125,
      "learning_rate": 2.596255576430322e-05,
      "loss": 0.40211167335510256,
      "num_input_tokens_seen": 7704293632,
      "step": 27090,
      "train_runtime": 263460.8687,
      "train_tokens_per_second": 29242.649
    },
    {
      "epoch": 0.9588529147049913,
      "grad_norm": 0.59375,
      "learning_rate": 2.5959056437428414e-05,
      "loss": 0.4005706787109375,
      "num_input_tokens_seen": 7707106816,
      "step": 27100,
      "train_runtime": 263556.7867,
      "train_tokens_per_second": 29242.68
    },
    {
      "epoch": 0.9592067349687201,
      "grad_norm": 0.61328125,
      "learning_rate": 2.5955558525127977e-05,
      "loss": 0.3992138385772705,
      "num_input_tokens_seen": 7709932096,
      "step": 27110,
      "train_runtime": 263651.5026,
      "train_tokens_per_second": 29242.891
    },
    {
      "epoch": 0.9595605552324489,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5952062026449096e-05,
      "loss": 0.4039638519287109,
      "num_input_tokens_seen": 7712776768,
      "step": 27120,
      "train_runtime": 263749.3231,
      "train_tokens_per_second": 29242.831
    },
    {
      "epoch": 0.9599143754961776,
      "grad_norm": 0.640625,
      "learning_rate": 2.59485669404399e-05,
      "loss": 0.39955973625183105,
      "num_input_tokens_seen": 7715607616,
      "step": 27130,
      "train_runtime": 263844.754,
      "train_tokens_per_second": 29242.983
    },
    {
      "epoch": 0.9602681957599064,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5945073266149366e-05,
      "loss": 0.39974355697631836,
      "num_input_tokens_seen": 7718424512,
      "step": 27140,
      "train_runtime": 263939.8514,
      "train_tokens_per_second": 29243.119
    },
    {
      "epoch": 0.9606220160236352,
      "grad_norm": 0.59765625,
      "learning_rate": 2.594158100262741e-05,
      "loss": 0.39653887748718264,
      "num_input_tokens_seen": 7721271360,
      "step": 27150,
      "train_runtime": 264035.8988,
      "train_tokens_per_second": 29243.263
    },
    {
      "epoch": 0.9609758362873639,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5938090148924815e-05,
      "loss": 0.40317444801330565,
      "num_input_tokens_seen": 7724100992,
      "step": 27160,
      "train_runtime": 264132.0322,
      "train_tokens_per_second": 29243.333
    },
    {
      "epoch": 0.9613296565510927,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5934600704093282e-05,
      "loss": 0.40064101219177245,
      "num_input_tokens_seen": 7726932160,
      "step": 27170,
      "train_runtime": 264228.039,
      "train_tokens_per_second": 29243.422
    },
    {
      "epoch": 0.9616834768148216,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5931112667185365e-05,
      "loss": 0.39939146041870116,
      "num_input_tokens_seen": 7729738432,
      "step": 27180,
      "train_runtime": 264323.5664,
      "train_tokens_per_second": 29243.471
    },
    {
      "epoch": 0.9620372970785503,
      "grad_norm": 0.58984375,
      "learning_rate": 2.592762603725456e-05,
      "loss": 0.3980076313018799,
      "num_input_tokens_seen": 7732588608,
      "step": 27190,
      "train_runtime": 264419.3244,
      "train_tokens_per_second": 29243.659
    },
    {
      "epoch": 0.9623911173422791,
      "grad_norm": 0.62890625,
      "learning_rate": 2.592414081335522e-05,
      "loss": 0.39870409965515136,
      "num_input_tokens_seen": 7735442176,
      "step": 27200,
      "train_runtime": 264514.4786,
      "train_tokens_per_second": 29243.927
    },
    {
      "epoch": 0.9627449376060079,
      "grad_norm": 0.58984375,
      "learning_rate": 2.592065699454259e-05,
      "loss": 0.3983691930770874,
      "num_input_tokens_seen": 7738321792,
      "step": 27210,
      "train_runtime": 264614.3441,
      "train_tokens_per_second": 29243.773
    },
    {
      "epoch": 0.9630987578697366,
      "grad_norm": 0.5703125,
      "learning_rate": 2.5917174579872815e-05,
      "loss": 0.3958759784698486,
      "num_input_tokens_seen": 7741176000,
      "step": 27220,
      "train_runtime": 264710.7654,
      "train_tokens_per_second": 29243.903
    },
    {
      "epoch": 0.9634525781334654,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5913693568402926e-05,
      "loss": 0.3938683032989502,
      "num_input_tokens_seen": 7744038848,
      "step": 27230,
      "train_runtime": 264808.8046,
      "train_tokens_per_second": 29243.887
    },
    {
      "epoch": 0.9638063983971942,
      "grad_norm": 0.59375,
      "learning_rate": 2.5910213959190833e-05,
      "loss": 0.40143728256225586,
      "num_input_tokens_seen": 7746888704,
      "step": 27240,
      "train_runtime": 264907.4816,
      "train_tokens_per_second": 29243.752
    },
    {
      "epoch": 0.9641602186609229,
      "grad_norm": 0.609375,
      "learning_rate": 2.590673575129534e-05,
      "loss": 0.40088181495666503,
      "num_input_tokens_seen": 7749687296,
      "step": 27250,
      "train_runtime": 265001.2782,
      "train_tokens_per_second": 29243.962
    },
    {
      "epoch": 0.9645140389246518,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5903258943776116e-05,
      "loss": 0.39911527633666993,
      "num_input_tokens_seen": 7752568768,
      "step": 27260,
      "train_runtime": 265102.9423,
      "train_tokens_per_second": 29243.616
    },
    {
      "epoch": 0.9648678591883806,
      "grad_norm": 0.6171875,
      "learning_rate": 2.5899783535693745e-05,
      "loss": 0.3927340030670166,
      "num_input_tokens_seen": 7755390336,
      "step": 27270,
      "train_runtime": 265198.5543,
      "train_tokens_per_second": 29243.713
    },
    {
      "epoch": 0.9652216794521093,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5896309526109663e-05,
      "loss": 0.3992013454437256,
      "num_input_tokens_seen": 7758272320,
      "step": 27280,
      "train_runtime": 265298.2601,
      "train_tokens_per_second": 29243.585
    },
    {
      "epoch": 0.9655754997158381,
      "grad_norm": 0.6171875,
      "learning_rate": 2.589283691408621e-05,
      "loss": 0.3973858118057251,
      "num_input_tokens_seen": 7761131648,
      "step": 27290,
      "train_runtime": 265396.3767,
      "train_tokens_per_second": 29243.548
    },
    {
      "epoch": 0.9659293199795669,
      "grad_norm": 0.58984375,
      "learning_rate": 2.588936569868658e-05,
      "loss": 0.40364990234375,
      "num_input_tokens_seen": 7763973760,
      "step": 27300,
      "train_runtime": 265494.3461,
      "train_tokens_per_second": 29243.462
    },
    {
      "epoch": 0.9662831402432956,
      "grad_norm": 0.56640625,
      "learning_rate": 2.5885895878974877e-05,
      "loss": 0.400461483001709,
      "num_input_tokens_seen": 7766862464,
      "step": 27310,
      "train_runtime": 265595.0834,
      "train_tokens_per_second": 29243.246
    },
    {
      "epoch": 0.9666369605070244,
      "grad_norm": 0.578125,
      "learning_rate": 2.5882427454016055e-05,
      "loss": 0.4009994983673096,
      "num_input_tokens_seen": 7769708480,
      "step": 27320,
      "train_runtime": 265693.6931,
      "train_tokens_per_second": 29243.105
    },
    {
      "epoch": 0.9669907807707532,
      "grad_norm": 0.59765625,
      "learning_rate": 2.587896042287596e-05,
      "loss": 0.3973134279251099,
      "num_input_tokens_seen": 7772516544,
      "step": 27330,
      "train_runtime": 265790.1556,
      "train_tokens_per_second": 29243.057
    },
    {
      "epoch": 0.9673446010344819,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5875494784621317e-05,
      "loss": 0.3973072052001953,
      "num_input_tokens_seen": 7775338240,
      "step": 27340,
      "train_runtime": 265886.1814,
      "train_tokens_per_second": 29243.108
    },
    {
      "epoch": 0.9676984212982108,
      "grad_norm": 0.58984375,
      "learning_rate": 2.587203053831971e-05,
      "loss": 0.3967430591583252,
      "num_input_tokens_seen": 7778188160,
      "step": 27350,
      "train_runtime": 265985.7147,
      "train_tokens_per_second": 29242.879
    },
    {
      "epoch": 0.9680522415619396,
      "grad_norm": 0.59765625,
      "learning_rate": 2.586856768303961e-05,
      "loss": 0.39706850051879883,
      "num_input_tokens_seen": 7781007616,
      "step": 27360,
      "train_runtime": 266082.6795,
      "train_tokens_per_second": 29242.819
    },
    {
      "epoch": 0.9684060618256684,
      "grad_norm": 0.6171875,
      "learning_rate": 2.586510621785036e-05,
      "loss": 0.39858102798461914,
      "num_input_tokens_seen": 7783851840,
      "step": 27370,
      "train_runtime": 266177.0195,
      "train_tokens_per_second": 29243.14
    },
    {
      "epoch": 0.9687598820893971,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5861646141822164e-05,
      "loss": 0.40108633041381836,
      "num_input_tokens_seen": 7786691392,
      "step": 27380,
      "train_runtime": 266277.0223,
      "train_tokens_per_second": 29242.821
    },
    {
      "epoch": 0.9691137023531259,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5858187454026106e-05,
      "loss": 0.39533224105834963,
      "num_input_tokens_seen": 7789517312,
      "step": 27390,
      "train_runtime": 266371.5938,
      "train_tokens_per_second": 29243.048
    },
    {
      "epoch": 0.9694675226168546,
      "grad_norm": 0.578125,
      "learning_rate": 2.5854730153534134e-05,
      "loss": 0.3947713851928711,
      "num_input_tokens_seen": 7792367616,
      "step": 27400,
      "train_runtime": 266470.3422,
      "train_tokens_per_second": 29242.908
    },
    {
      "epoch": 0.9698213428805834,
      "grad_norm": 0.57421875,
      "learning_rate": 2.585127423941907e-05,
      "loss": 0.4002467155456543,
      "num_input_tokens_seen": 7795244736,
      "step": 27410,
      "train_runtime": 266571.1649,
      "train_tokens_per_second": 29242.64
    },
    {
      "epoch": 0.9701751631443122,
      "grad_norm": 0.59765625,
      "learning_rate": 2.58478197107546e-05,
      "loss": 0.3985337972640991,
      "num_input_tokens_seen": 7798101888,
      "step": 27420,
      "train_runtime": 266667.7506,
      "train_tokens_per_second": 29242.763
    },
    {
      "epoch": 0.9705289834080411,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5844366566615273e-05,
      "loss": 0.39818401336669923,
      "num_input_tokens_seen": 7801014976,
      "step": 27430,
      "train_runtime": 266768.1858,
      "train_tokens_per_second": 29242.674
    },
    {
      "epoch": 0.9708828036717698,
      "grad_norm": 0.59375,
      "learning_rate": 2.584091480607651e-05,
      "loss": 0.40388174057006837,
      "num_input_tokens_seen": 7803891904,
      "step": 27440,
      "train_runtime": 266867.9715,
      "train_tokens_per_second": 29242.52
    },
    {
      "epoch": 0.9712366239354986,
      "grad_norm": 0.59375,
      "learning_rate": 2.5837464428214592e-05,
      "loss": 0.3994109869003296,
      "num_input_tokens_seen": 7806784704,
      "step": 27450,
      "train_runtime": 266968.4823,
      "train_tokens_per_second": 29242.346
    },
    {
      "epoch": 0.9715904441992274,
      "grad_norm": 0.61328125,
      "learning_rate": 2.5834015432106666e-05,
      "loss": 0.39685707092285155,
      "num_input_tokens_seen": 7809631744,
      "step": 27460,
      "train_runtime": 267064.7881,
      "train_tokens_per_second": 29242.461
    },
    {
      "epoch": 0.9719442644629561,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5830567816830737e-05,
      "loss": 0.39645776748657224,
      "num_input_tokens_seen": 7812457856,
      "step": 27470,
      "train_runtime": 267161.074,
      "train_tokens_per_second": 29242.501
    },
    {
      "epoch": 0.9722980847266849,
      "grad_norm": 0.58203125,
      "learning_rate": 2.5827121581465668e-05,
      "loss": 0.3955065727233887,
      "num_input_tokens_seen": 7815289536,
      "step": 27480,
      "train_runtime": 267259.7847,
      "train_tokens_per_second": 29242.295
    },
    {
      "epoch": 0.9726519049904137,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5823676725091193e-05,
      "loss": 0.4005618095397949,
      "num_input_tokens_seen": 7818122176,
      "step": 27490,
      "train_runtime": 267357.1218,
      "train_tokens_per_second": 29242.244
    },
    {
      "epoch": 0.9730057252541424,
      "grad_norm": 0.5625,
      "learning_rate": 2.5820233246787902e-05,
      "loss": 0.40039587020874023,
      "num_input_tokens_seen": 7820980032,
      "step": 27500,
      "train_runtime": 267453.5882,
      "train_tokens_per_second": 29242.382
    },
    {
      "epoch": 0.9733595455178713,
      "grad_norm": 0.59375,
      "learning_rate": 2.581679114563723e-05,
      "loss": 0.4029428005218506,
      "num_input_tokens_seen": 7823928128,
      "step": 27510,
      "train_runtime": 267557.7607,
      "train_tokens_per_second": 29242.015
    },
    {
      "epoch": 0.9737133657816001,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5813350420721488e-05,
      "loss": 0.4005100250244141,
      "num_input_tokens_seen": 7826748928,
      "step": 27520,
      "train_runtime": 267652.1895,
      "train_tokens_per_second": 29242.238
    },
    {
      "epoch": 0.9740671860453288,
      "grad_norm": 0.5703125,
      "learning_rate": 2.5809911071123834e-05,
      "loss": 0.3956948757171631,
      "num_input_tokens_seen": 7829627584,
      "step": 27530,
      "train_runtime": 267751.987,
      "train_tokens_per_second": 29242.09
    },
    {
      "epoch": 0.9744210063090576,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5806473095928274e-05,
      "loss": 0.3960397720336914,
      "num_input_tokens_seen": 7832470336,
      "step": 27540,
      "train_runtime": 267846.8999,
      "train_tokens_per_second": 29242.341
    },
    {
      "epoch": 0.9747748265727864,
      "grad_norm": 0.6171875,
      "learning_rate": 2.5803036494219678e-05,
      "loss": 0.3994396209716797,
      "num_input_tokens_seen": 7835371520,
      "step": 27550,
      "train_runtime": 267948.1227,
      "train_tokens_per_second": 29242.121
    },
    {
      "epoch": 0.9751286468365151,
      "grad_norm": 0.58203125,
      "learning_rate": 2.5799601265083767e-05,
      "loss": 0.400991153717041,
      "num_input_tokens_seen": 7838242752,
      "step": 27560,
      "train_runtime": 268046.1473,
      "train_tokens_per_second": 29242.139
    },
    {
      "epoch": 0.9754824671002439,
      "grad_norm": 0.578125,
      "learning_rate": 2.5796167407607113e-05,
      "loss": 0.3959287881851196,
      "num_input_tokens_seen": 7841110848,
      "step": 27570,
      "train_runtime": 268143.133,
      "train_tokens_per_second": 29242.259
    },
    {
      "epoch": 0.9758362873639727,
      "grad_norm": 0.609375,
      "learning_rate": 2.5792734920877138e-05,
      "loss": 0.3995499134063721,
      "num_input_tokens_seen": 7843947392,
      "step": 27580,
      "train_runtime": 268239.4221,
      "train_tokens_per_second": 29242.336
    },
    {
      "epoch": 0.9761901076277014,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5789303803982114e-05,
      "loss": 0.39961650371551516,
      "num_input_tokens_seen": 7846760832,
      "step": 27590,
      "train_runtime": 268334.4578,
      "train_tokens_per_second": 29242.464
    },
    {
      "epoch": 0.9765439278914303,
      "grad_norm": 0.5703125,
      "learning_rate": 2.578587405601115e-05,
      "loss": 0.40409126281738283,
      "num_input_tokens_seen": 7849578048,
      "step": 27600,
      "train_runtime": 268430.4744,
      "train_tokens_per_second": 29242.5
    },
    {
      "epoch": 0.9768977481551591,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5782445676054233e-05,
      "loss": 0.39781885147094725,
      "num_input_tokens_seen": 7852493312,
      "step": 27610,
      "train_runtime": 268530.5108,
      "train_tokens_per_second": 29242.462
    },
    {
      "epoch": 0.9772515684188878,
      "grad_norm": 0.58203125,
      "learning_rate": 2.577901866320217e-05,
      "loss": 0.4021693229675293,
      "num_input_tokens_seen": 7855302656,
      "step": 27620,
      "train_runtime": 268625.3663,
      "train_tokens_per_second": 29242.594
    },
    {
      "epoch": 0.9776053886826166,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5775593016546618e-05,
      "loss": 0.3960843563079834,
      "num_input_tokens_seen": 7858197760,
      "step": 27630,
      "train_runtime": 268726.3392,
      "train_tokens_per_second": 29242.38
    },
    {
      "epoch": 0.9779592089463454,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5772168735180084e-05,
      "loss": 0.40301008224487306,
      "num_input_tokens_seen": 7861040896,
      "step": 27640,
      "train_runtime": 268824.2922,
      "train_tokens_per_second": 29242.301
    },
    {
      "epoch": 0.9783130292100741,
      "grad_norm": 0.5703125,
      "learning_rate": 2.5768745818195927e-05,
      "loss": 0.3999293565750122,
      "num_input_tokens_seen": 7863913856,
      "step": 27650,
      "train_runtime": 268922.9966,
      "train_tokens_per_second": 29242.251
    },
    {
      "epoch": 0.9786668494738029,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5765324264688327e-05,
      "loss": 0.3959509372711182,
      "num_input_tokens_seen": 7866820160,
      "step": 27660,
      "train_runtime": 269022.7867,
      "train_tokens_per_second": 29242.208
    },
    {
      "epoch": 0.9790206697375317,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5761904073752323e-05,
      "loss": 0.39629387855529785,
      "num_input_tokens_seen": 7869606464,
      "step": 27670,
      "train_runtime": 269114.9455,
      "train_tokens_per_second": 29242.547
    },
    {
      "epoch": 0.9793744900012605,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5758485244483788e-05,
      "loss": 0.40226335525512696,
      "num_input_tokens_seen": 7872438336,
      "step": 27680,
      "train_runtime": 269212.2273,
      "train_tokens_per_second": 29242.499
    },
    {
      "epoch": 0.9797283102649893,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5755067775979443e-05,
      "loss": 0.3981515645980835,
      "num_input_tokens_seen": 7875197248,
      "step": 27690,
      "train_runtime": 269303.2712,
      "train_tokens_per_second": 29242.858
    },
    {
      "epoch": 0.9800821305287181,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5751651667336824e-05,
      "loss": 0.3977331638336182,
      "num_input_tokens_seen": 7878105984,
      "step": 27700,
      "train_runtime": 269404.3895,
      "train_tokens_per_second": 29242.679
    },
    {
      "epoch": 0.9804359507924468,
      "grad_norm": 0.61328125,
      "learning_rate": 2.5748236917654333e-05,
      "loss": 0.4015091896057129,
      "num_input_tokens_seen": 7880935232,
      "step": 27710,
      "train_runtime": 269497.8693,
      "train_tokens_per_second": 29243.034
    },
    {
      "epoch": 0.9807897710561756,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5744823526031202e-05,
      "loss": 0.40073838233947756,
      "num_input_tokens_seen": 7883814720,
      "step": 27720,
      "train_runtime": 269595.6837,
      "train_tokens_per_second": 29243.104
    },
    {
      "epoch": 0.9811435913199044,
      "grad_norm": 0.61328125,
      "learning_rate": 2.5741411491567484e-05,
      "loss": 0.3967472553253174,
      "num_input_tokens_seen": 7886692032,
      "step": 27730,
      "train_runtime": 269694.2669,
      "train_tokens_per_second": 29243.084
    },
    {
      "epoch": 0.9814974115836331,
      "grad_norm": 0.59765625,
      "learning_rate": 2.573800081336408e-05,
      "loss": 0.39780030250549314,
      "num_input_tokens_seen": 7889550912,
      "step": 27740,
      "train_runtime": 269792.0549,
      "train_tokens_per_second": 29243.081
    },
    {
      "epoch": 0.9818512318473619,
      "grad_norm": 0.625,
      "learning_rate": 2.573459149052272e-05,
      "loss": 0.3954323768615723,
      "num_input_tokens_seen": 7892358528,
      "step": 27750,
      "train_runtime": 269885.2353,
      "train_tokens_per_second": 29243.388
    },
    {
      "epoch": 0.9822050521110908,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5731183522145968e-05,
      "loss": 0.4019524574279785,
      "num_input_tokens_seen": 7895227200,
      "step": 27760,
      "train_runtime": 269986.1562,
      "train_tokens_per_second": 29243.082
    },
    {
      "epoch": 0.9825588723748195,
      "grad_norm": 0.56640625,
      "learning_rate": 2.5727776907337226e-05,
      "loss": 0.39240384101867676,
      "num_input_tokens_seen": 7898030208,
      "step": 27770,
      "train_runtime": 270079.829,
      "train_tokens_per_second": 29243.318
    },
    {
      "epoch": 0.9829126926385483,
      "grad_norm": 0.609375,
      "learning_rate": 2.5724371645200708e-05,
      "loss": 0.39812841415405276,
      "num_input_tokens_seen": 7900816960,
      "step": 27780,
      "train_runtime": 270176.9517,
      "train_tokens_per_second": 29243.12
    },
    {
      "epoch": 0.9832665129022771,
      "grad_norm": 0.609375,
      "learning_rate": 2.572096773484147e-05,
      "loss": 0.4045284271240234,
      "num_input_tokens_seen": 7903646848,
      "step": 27790,
      "train_runtime": 270272.9178,
      "train_tokens_per_second": 29243.207
    },
    {
      "epoch": 0.9836203331660058,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5717565175365416e-05,
      "loss": 0.39740467071533203,
      "num_input_tokens_seen": 7906475520,
      "step": 27800,
      "train_runtime": 270367.5579,
      "train_tokens_per_second": 29243.433
    },
    {
      "epoch": 0.9839741534297346,
      "grad_norm": 0.59765625,
      "learning_rate": 2.571416396587924e-05,
      "loss": 0.3975717067718506,
      "num_input_tokens_seen": 7909321216,
      "step": 27810,
      "train_runtime": 270465.8021,
      "train_tokens_per_second": 29243.332
    },
    {
      "epoch": 0.9843279736934634,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5710764105490486e-05,
      "loss": 0.3997790336608887,
      "num_input_tokens_seen": 7912120320,
      "step": 27820,
      "train_runtime": 270559.6413,
      "train_tokens_per_second": 29243.535
    },
    {
      "epoch": 0.9846817939571921,
      "grad_norm": 0.58984375,
      "learning_rate": 2.570736559330752e-05,
      "loss": 0.39911911487579343,
      "num_input_tokens_seen": 7914969536,
      "step": 27830,
      "train_runtime": 270656.305,
      "train_tokens_per_second": 29243.618
    },
    {
      "epoch": 0.9850356142209209,
      "grad_norm": 0.64453125,
      "learning_rate": 2.570396842843953e-05,
      "loss": 0.39789586067199706,
      "num_input_tokens_seen": 7917792064,
      "step": 27840,
      "train_runtime": 270751.9984,
      "train_tokens_per_second": 29243.707
    },
    {
      "epoch": 0.9853894344846498,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5700572609996526e-05,
      "loss": 0.3981212615966797,
      "num_input_tokens_seen": 7920633152,
      "step": 27850,
      "train_runtime": 270850.3418,
      "train_tokens_per_second": 29243.578
    },
    {
      "epoch": 0.9857432547483785,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5697178137089357e-05,
      "loss": 0.3947503089904785,
      "num_input_tokens_seen": 7923523520,
      "step": 27860,
      "train_runtime": 270947.7642,
      "train_tokens_per_second": 29243.731
    },
    {
      "epoch": 0.9860970750121073,
      "grad_norm": 0.58203125,
      "learning_rate": 2.569378500882966e-05,
      "loss": 0.3959986686706543,
      "num_input_tokens_seen": 7926298176,
      "step": 27870,
      "train_runtime": 271043.6654,
      "train_tokens_per_second": 29243.621
    },
    {
      "epoch": 0.9864508952758361,
      "grad_norm": 0.5859375,
      "learning_rate": 2.569039322432994e-05,
      "loss": 0.40237884521484374,
      "num_input_tokens_seen": 7929133824,
      "step": 27880,
      "train_runtime": 271139.3219,
      "train_tokens_per_second": 29243.762
    },
    {
      "epoch": 0.9868047155395648,
      "grad_norm": 0.625,
      "learning_rate": 2.5687002782703474e-05,
      "loss": 0.4016533851623535,
      "num_input_tokens_seen": 7931983168,
      "step": 27890,
      "train_runtime": 271237.5189,
      "train_tokens_per_second": 29243.68
    },
    {
      "epoch": 0.9871585358032936,
      "grad_norm": 0.6015625,
      "learning_rate": 2.568361368306439e-05,
      "loss": 0.39665908813476564,
      "num_input_tokens_seen": 7934843328,
      "step": 27900,
      "train_runtime": 271337.0613,
      "train_tokens_per_second": 29243.493
    },
    {
      "epoch": 0.9875123560670224,
      "grad_norm": 0.59375,
      "learning_rate": 2.5680225924527623e-05,
      "loss": 0.39907145500183105,
      "num_input_tokens_seen": 7937700352,
      "step": 27910,
      "train_runtime": 271434.8647,
      "train_tokens_per_second": 29243.481
    },
    {
      "epoch": 0.9878661763307511,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5676839506208927e-05,
      "loss": 0.3972276210784912,
      "num_input_tokens_seen": 7940547648,
      "step": 27920,
      "train_runtime": 271531.7043,
      "train_tokens_per_second": 29243.538
    },
    {
      "epoch": 0.98821999659448,
      "grad_norm": 0.56640625,
      "learning_rate": 2.567345442722487e-05,
      "loss": 0.39655632972717286,
      "num_input_tokens_seen": 7943389504,
      "step": 27930,
      "train_runtime": 271627.0183,
      "train_tokens_per_second": 29243.739
    },
    {
      "epoch": 0.9885738168582088,
      "grad_norm": 0.58984375,
      "learning_rate": 2.567007068669283e-05,
      "loss": 0.3965986967086792,
      "num_input_tokens_seen": 7946204800,
      "step": 27940,
      "train_runtime": 271720.5643,
      "train_tokens_per_second": 29244.032
    },
    {
      "epoch": 0.9889276371219375,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5666688283731016e-05,
      "loss": 0.3946433782577515,
      "num_input_tokens_seen": 7948974080,
      "step": 27950,
      "train_runtime": 271811.0232,
      "train_tokens_per_second": 29244.488
    },
    {
      "epoch": 0.9892814573856663,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5663307217458428e-05,
      "loss": 0.3974971532821655,
      "num_input_tokens_seen": 7951878912,
      "step": 27960,
      "train_runtime": 271910.2389,
      "train_tokens_per_second": 29244.5
    },
    {
      "epoch": 0.9896352776493951,
      "grad_norm": 0.58203125,
      "learning_rate": 2.5659927486994893e-05,
      "loss": 0.39725556373596194,
      "num_input_tokens_seen": 7954701824,
      "step": 27970,
      "train_runtime": 272005.9134,
      "train_tokens_per_second": 29244.592
    },
    {
      "epoch": 0.9899890979131238,
      "grad_norm": 0.578125,
      "learning_rate": 2.565654909146105e-05,
      "loss": 0.39673457145690916,
      "num_input_tokens_seen": 7957524096,
      "step": 27980,
      "train_runtime": 272102.9322,
      "train_tokens_per_second": 29244.536
    },
    {
      "epoch": 0.9903429181768526,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5653172029978346e-05,
      "loss": 0.40122127532958984,
      "num_input_tokens_seen": 7960404800,
      "step": 27990,
      "train_runtime": 272199.7308,
      "train_tokens_per_second": 29244.72
    },
    {
      "epoch": 0.9906967384405814,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5649796301669014e-05,
      "loss": 0.3940152883529663,
      "num_input_tokens_seen": 7963204224,
      "step": 28000,
      "train_runtime": 272295.2892,
      "train_tokens_per_second": 29244.737
    },
    {
      "epoch": 0.9910505587043102,
      "grad_norm": 0.61328125,
      "learning_rate": 2.5646421905656148e-05,
      "loss": 0.3974417209625244,
      "num_input_tokens_seen": 7966060928,
      "step": 28010,
      "train_runtime": 272391.7257,
      "train_tokens_per_second": 29244.871
    },
    {
      "epoch": 0.991404378968039,
      "grad_norm": 0.62109375,
      "learning_rate": 2.5643048841063587e-05,
      "loss": 0.3976799964904785,
      "num_input_tokens_seen": 7968937280,
      "step": 28020,
      "train_runtime": 272489.8236,
      "train_tokens_per_second": 29244.899
    },
    {
      "epoch": 0.9917581992317678,
      "grad_norm": 0.62109375,
      "learning_rate": 2.563967710701603e-05,
      "loss": 0.4041726112365723,
      "num_input_tokens_seen": 7971751104,
      "step": 28030,
      "train_runtime": 272585.5933,
      "train_tokens_per_second": 29244.947
    },
    {
      "epoch": 0.9921120194954965,
      "grad_norm": 0.62109375,
      "learning_rate": 2.5636306702638946e-05,
      "loss": 0.40075006484985354,
      "num_input_tokens_seen": 7974506944,
      "step": 28040,
      "train_runtime": 272677.9714,
      "train_tokens_per_second": 29245.145
    },
    {
      "epoch": 0.9924658397592253,
      "grad_norm": 0.59375,
      "learning_rate": 2.5632937627058624e-05,
      "loss": 0.40079808235168457,
      "num_input_tokens_seen": 7977336320,
      "step": 28050,
      "train_runtime": 272773.5431,
      "train_tokens_per_second": 29245.271
    },
    {
      "epoch": 0.9928196600229541,
      "grad_norm": 0.58203125,
      "learning_rate": 2.562956987940216e-05,
      "loss": 0.3951700210571289,
      "num_input_tokens_seen": 7980157952,
      "step": 28060,
      "train_runtime": 272868.5018,
      "train_tokens_per_second": 29245.435
    },
    {
      "epoch": 0.9931734802866828,
      "grad_norm": 0.609375,
      "learning_rate": 2.5626203458797432e-05,
      "loss": 0.39908695220947266,
      "num_input_tokens_seen": 7983067456,
      "step": 28070,
      "train_runtime": 272969.7001,
      "train_tokens_per_second": 29245.251
    },
    {
      "epoch": 0.9935273005504116,
      "grad_norm": 0.578125,
      "learning_rate": 2.562283836437315e-05,
      "loss": 0.3960444450378418,
      "num_input_tokens_seen": 7985969856,
      "step": 28080,
      "train_runtime": 273069.6747,
      "train_tokens_per_second": 29245.173
    },
    {
      "epoch": 0.9938811208141404,
      "grad_norm": 0.62109375,
      "learning_rate": 2.561947459525879e-05,
      "loss": 0.39877910614013673,
      "num_input_tokens_seen": 7988810048,
      "step": 28090,
      "train_runtime": 273165.3761,
      "train_tokens_per_second": 29245.324
    },
    {
      "epoch": 0.9942349410778692,
      "grad_norm": 0.58203125,
      "learning_rate": 2.561611215058466e-05,
      "loss": 0.39794418811798093,
      "num_input_tokens_seen": 7991675072,
      "step": 28100,
      "train_runtime": 273263.9158,
      "train_tokens_per_second": 29245.263
    },
    {
      "epoch": 0.994588761341598,
      "grad_norm": 0.59375,
      "learning_rate": 2.5612751029481845e-05,
      "loss": 0.3959181785583496,
      "num_input_tokens_seen": 7994518144,
      "step": 28110,
      "train_runtime": 273359.6536,
      "train_tokens_per_second": 29245.421
    },
    {
      "epoch": 0.9949425816053268,
      "grad_norm": 0.58203125,
      "learning_rate": 2.560939123108225e-05,
      "loss": 0.3949803590774536,
      "num_input_tokens_seen": 7997373312,
      "step": 28120,
      "train_runtime": 273460.239,
      "train_tokens_per_second": 29245.105
    },
    {
      "epoch": 0.9952964018690555,
      "grad_norm": 0.6015625,
      "learning_rate": 2.560603275451855e-05,
      "loss": 0.4009438991546631,
      "num_input_tokens_seen": 8000276352,
      "step": 28130,
      "train_runtime": 273562.1617,
      "train_tokens_per_second": 29244.821
    },
    {
      "epoch": 0.9956502221327843,
      "grad_norm": 0.609375,
      "learning_rate": 2.5602675598924227e-05,
      "loss": 0.4008786201477051,
      "num_input_tokens_seen": 8003102528,
      "step": 28140,
      "train_runtime": 273655.0942,
      "train_tokens_per_second": 29245.217
    },
    {
      "epoch": 0.9960040423965131,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5599319763433566e-05,
      "loss": 0.39966888427734376,
      "num_input_tokens_seen": 8005923840,
      "step": 28150,
      "train_runtime": 273752.4196,
      "train_tokens_per_second": 29245.125
    },
    {
      "epoch": 0.9963578626602418,
      "grad_norm": 0.5703125,
      "learning_rate": 2.559596524718164e-05,
      "loss": 0.39789068698883057,
      "num_input_tokens_seen": 8008785664,
      "step": 28160,
      "train_runtime": 273849.4826,
      "train_tokens_per_second": 29245.21
    },
    {
      "epoch": 0.9967116829239706,
      "grad_norm": 0.58203125,
      "learning_rate": 2.559261204930431e-05,
      "loss": 0.39697449207305907,
      "num_input_tokens_seen": 8011625856,
      "step": 28170,
      "train_runtime": 273946.75,
      "train_tokens_per_second": 29245.194
    },
    {
      "epoch": 0.9970655031876995,
      "grad_norm": 0.59765625,
      "learning_rate": 2.558926016893824e-05,
      "loss": 0.39841408729553224,
      "num_input_tokens_seen": 8014461440,
      "step": 28180,
      "train_runtime": 274042.2682,
      "train_tokens_per_second": 29245.348
    },
    {
      "epoch": 0.9974193234514283,
      "grad_norm": 0.57421875,
      "learning_rate": 2.5585909605220877e-05,
      "loss": 0.3999458312988281,
      "num_input_tokens_seen": 8017315840,
      "step": 28190,
      "train_runtime": 274140.9999,
      "train_tokens_per_second": 29245.227
    },
    {
      "epoch": 0.997773143715157,
      "grad_norm": 0.57421875,
      "learning_rate": 2.558256035729046e-05,
      "loss": 0.39957804679870607,
      "num_input_tokens_seen": 8020148032,
      "step": 28200,
      "train_runtime": 274235.8306,
      "train_tokens_per_second": 29245.442
    },
    {
      "epoch": 0.9981269639788858,
      "grad_norm": 0.58203125,
      "learning_rate": 2.557921242428602e-05,
      "loss": 0.4022542953491211,
      "num_input_tokens_seen": 8022949056,
      "step": 28210,
      "train_runtime": 274330.5863,
      "train_tokens_per_second": 29245.551
    },
    {
      "epoch": 0.9984807842426145,
      "grad_norm": 0.6015625,
      "learning_rate": 2.557586580534737e-05,
      "loss": 0.3926943302154541,
      "num_input_tokens_seen": 8025749568,
      "step": 28220,
      "train_runtime": 274424.8234,
      "train_tokens_per_second": 29245.713
    },
    {
      "epoch": 0.9988346045063433,
      "grad_norm": 0.609375,
      "learning_rate": 2.5572520499615127e-05,
      "loss": 0.3961955070495605,
      "num_input_tokens_seen": 8028631680,
      "step": 28230,
      "train_runtime": 274524.7178,
      "train_tokens_per_second": 29245.569
    },
    {
      "epoch": 0.9991884247700721,
      "grad_norm": 0.58203125,
      "learning_rate": 2.5569176506230667e-05,
      "loss": 0.4020658016204834,
      "num_input_tokens_seen": 8031430592,
      "step": 28240,
      "train_runtime": 274620.7369,
      "train_tokens_per_second": 29245.536
    },
    {
      "epoch": 0.9995422450338008,
      "grad_norm": 0.5703125,
      "learning_rate": 2.5565833824336183e-05,
      "loss": 0.39832544326782227,
      "num_input_tokens_seen": 8034253696,
      "step": 28250,
      "train_runtime": 274714.2057,
      "train_tokens_per_second": 29245.862
    },
    {
      "epoch": 0.9998960652975297,
      "grad_norm": 0.56640625,
      "learning_rate": 2.5562492453074633e-05,
      "loss": 0.39608800411224365,
      "num_input_tokens_seen": 8037056512,
      "step": 28260,
      "train_runtime": 274806.3584,
      "train_tokens_per_second": 29246.254
    },
    {
      "epoch": 1.0002476741846102,
      "grad_norm": 0.60546875,
      "learning_rate": 2.555915239158976e-05,
      "loss": 0.3948505878448486,
      "num_input_tokens_seen": 8039909376,
      "step": 28270,
      "train_runtime": 274916.5534,
      "train_tokens_per_second": 29244.908
    },
    {
      "epoch": 1.000601494448339,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5555813639026093e-05,
      "loss": 0.38819246292114257,
      "num_input_tokens_seen": 8042756800,
      "step": 28280,
      "train_runtime": 275017.0202,
      "train_tokens_per_second": 29244.578
    },
    {
      "epoch": 1.0009553147120678,
      "grad_norm": 0.59375,
      "learning_rate": 2.5552476194528947e-05,
      "loss": 0.3980375289916992,
      "num_input_tokens_seen": 8045619328,
      "step": 28290,
      "train_runtime": 275116.1622,
      "train_tokens_per_second": 29244.444
    },
    {
      "epoch": 1.0013091349757965,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5549140057244413e-05,
      "loss": 0.38842923641204835,
      "num_input_tokens_seen": 8048455744,
      "step": 28300,
      "train_runtime": 275215.1701,
      "train_tokens_per_second": 29244.23
    },
    {
      "epoch": 1.0016629552395253,
      "grad_norm": 0.59375,
      "learning_rate": 2.5545805226319362e-05,
      "loss": 0.39246401786804197,
      "num_input_tokens_seen": 8051281152,
      "step": 28310,
      "train_runtime": 275311.9869,
      "train_tokens_per_second": 29244.209
    },
    {
      "epoch": 1.002016775503254,
      "grad_norm": 0.578125,
      "learning_rate": 2.5542471700901453e-05,
      "loss": 0.394031023979187,
      "num_input_tokens_seen": 8054102272,
      "step": 28320,
      "train_runtime": 275406.79,
      "train_tokens_per_second": 29244.385
    },
    {
      "epoch": 1.0023705957669828,
      "grad_norm": 0.6328125,
      "learning_rate": 2.5539139480139113e-05,
      "loss": 0.39044389724731443,
      "num_input_tokens_seen": 8056987648,
      "step": 28330,
      "train_runtime": 275507.1303,
      "train_tokens_per_second": 29244.207
    },
    {
      "epoch": 1.0027244160307116,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5535808563181546e-05,
      "loss": 0.390697717666626,
      "num_input_tokens_seen": 8059853504,
      "step": 28340,
      "train_runtime": 275603.6438,
      "train_tokens_per_second": 29244.365
    },
    {
      "epoch": 1.0030782362944404,
      "grad_norm": 0.59375,
      "learning_rate": 2.553247894917874e-05,
      "loss": 0.3869602680206299,
      "num_input_tokens_seen": 8062724288,
      "step": 28350,
      "train_runtime": 275701.4999,
      "train_tokens_per_second": 29244.398
    },
    {
      "epoch": 1.0034320565581691,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5529150637281453e-05,
      "loss": 0.3930094242095947,
      "num_input_tokens_seen": 8065563840,
      "step": 28360,
      "train_runtime": 275800.9129,
      "train_tokens_per_second": 29244.152
    },
    {
      "epoch": 1.003785876821898,
      "grad_norm": 0.56640625,
      "learning_rate": 2.552582362664122e-05,
      "loss": 0.3876837491989136,
      "num_input_tokens_seen": 8068325568,
      "step": 28370,
      "train_runtime": 275893.2306,
      "train_tokens_per_second": 29244.377
    },
    {
      "epoch": 1.0041396970856267,
      "grad_norm": 0.55859375,
      "learning_rate": 2.5522497916410353e-05,
      "loss": 0.39084525108337403,
      "num_input_tokens_seen": 8071166848,
      "step": 28380,
      "train_runtime": 275989.7705,
      "train_tokens_per_second": 29244.442
    },
    {
      "epoch": 1.0044935173493554,
      "grad_norm": 0.59375,
      "learning_rate": 2.5519173505741933e-05,
      "loss": 0.3946462869644165,
      "num_input_tokens_seen": 8074037760,
      "step": 28390,
      "train_runtime": 276090.4518,
      "train_tokens_per_second": 29244.176
    },
    {
      "epoch": 1.0048473376130842,
      "grad_norm": 0.61328125,
      "learning_rate": 2.551585039378981e-05,
      "loss": 0.3910089015960693,
      "num_input_tokens_seen": 8076849536,
      "step": 28400,
      "train_runtime": 276186.128,
      "train_tokens_per_second": 29244.226
    },
    {
      "epoch": 1.005201157876813,
      "grad_norm": 0.6171875,
      "learning_rate": 2.551252857970861e-05,
      "loss": 0.3901904821395874,
      "num_input_tokens_seen": 8079745472,
      "step": 28410,
      "train_runtime": 276285.7455,
      "train_tokens_per_second": 29244.163
    },
    {
      "epoch": 1.0055549781405417,
      "grad_norm": 0.609375,
      "learning_rate": 2.5509208062653734e-05,
      "loss": 0.39221858978271484,
      "num_input_tokens_seen": 8082548672,
      "step": 28420,
      "train_runtime": 276380.8867,
      "train_tokens_per_second": 29244.239
    },
    {
      "epoch": 1.0059087984042707,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5505888841781333e-05,
      "loss": 0.3890850067138672,
      "num_input_tokens_seen": 8085400384,
      "step": 28430,
      "train_runtime": 276479.0026,
      "train_tokens_per_second": 29244.175
    },
    {
      "epoch": 1.0062626186679995,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5502570916248353e-05,
      "loss": 0.39583353996276854,
      "num_input_tokens_seen": 8088302272,
      "step": 28440,
      "train_runtime": 276580.6365,
      "train_tokens_per_second": 29243.921
    },
    {
      "epoch": 1.0066164389317283,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5499254285212486e-05,
      "loss": 0.39234309196472167,
      "num_input_tokens_seen": 8091105088,
      "step": 28450,
      "train_runtime": 276674.0766,
      "train_tokens_per_second": 29244.175
    },
    {
      "epoch": 1.006970259195457,
      "grad_norm": 0.6171875,
      "learning_rate": 2.5495938947832204e-05,
      "loss": 0.3897040605545044,
      "num_input_tokens_seen": 8093894528,
      "step": 28460,
      "train_runtime": 276770.773,
      "train_tokens_per_second": 29244.036
    },
    {
      "epoch": 1.0073240794591858,
      "grad_norm": 0.62890625,
      "learning_rate": 2.5492624903266732e-05,
      "loss": 0.39056398868560793,
      "num_input_tokens_seen": 8096779008,
      "step": 28470,
      "train_runtime": 276869.1037,
      "train_tokens_per_second": 29244.068
    },
    {
      "epoch": 1.0076778997229145,
      "grad_norm": 0.6015625,
      "learning_rate": 2.548931215067607e-05,
      "loss": 0.39957690238952637,
      "num_input_tokens_seen": 8099626048,
      "step": 28480,
      "train_runtime": 276966.8954,
      "train_tokens_per_second": 29244.022
    },
    {
      "epoch": 1.0080317199866433,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5486000689220985e-05,
      "loss": 0.38661961555480956,
      "num_input_tokens_seen": 8102472448,
      "step": 28490,
      "train_runtime": 277062.7847,
      "train_tokens_per_second": 29244.175
    },
    {
      "epoch": 1.008385540250372,
      "grad_norm": 0.57421875,
      "learning_rate": 2.5482690518062994e-05,
      "loss": 0.39012315273284914,
      "num_input_tokens_seen": 8105298176,
      "step": 28500,
      "train_runtime": 277160.4093,
      "train_tokens_per_second": 29244.069
    },
    {
      "epoch": 1.0087393605141008,
      "grad_norm": 0.6015625,
      "learning_rate": 2.547938163636439e-05,
      "loss": 0.388866662979126,
      "num_input_tokens_seen": 8108137024,
      "step": 28510,
      "train_runtime": 277256.926,
      "train_tokens_per_second": 29244.128
    },
    {
      "epoch": 1.0090931807778296,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5476074043288212e-05,
      "loss": 0.39122862815856935,
      "num_input_tokens_seen": 8110957248,
      "step": 28520,
      "train_runtime": 277351.5339,
      "train_tokens_per_second": 29244.321
    },
    {
      "epoch": 1.0094470010415584,
      "grad_norm": 0.5859375,
      "learning_rate": 2.547276773799827e-05,
      "loss": 0.3896785736083984,
      "num_input_tokens_seen": 8113789056,
      "step": 28530,
      "train_runtime": 277445.8992,
      "train_tokens_per_second": 29244.581
    },
    {
      "epoch": 1.0098008213052871,
      "grad_norm": 0.5703125,
      "learning_rate": 2.5469462719659138e-05,
      "loss": 0.388866925239563,
      "num_input_tokens_seen": 8116633408,
      "step": 28540,
      "train_runtime": 277545.8829,
      "train_tokens_per_second": 29244.294
    },
    {
      "epoch": 1.010154641569016,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5466158987436135e-05,
      "loss": 0.3922267913818359,
      "num_input_tokens_seen": 8119549888,
      "step": 28550,
      "train_runtime": 277647.306,
      "train_tokens_per_second": 29244.116
    },
    {
      "epoch": 1.0105084618327447,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5462856540495345e-05,
      "loss": 0.3909679174423218,
      "num_input_tokens_seen": 8122393088,
      "step": 28560,
      "train_runtime": 277744.8676,
      "train_tokens_per_second": 29244.08
    },
    {
      "epoch": 1.0108622820964734,
      "grad_norm": 0.59375,
      "learning_rate": 2.5459555378003607e-05,
      "loss": 0.39356722831726076,
      "num_input_tokens_seen": 8125268992,
      "step": 28570,
      "train_runtime": 277847.0507,
      "train_tokens_per_second": 29243.676
    },
    {
      "epoch": 1.0112161023602022,
      "grad_norm": 0.58984375,
      "learning_rate": 2.545625549912853e-05,
      "loss": 0.39147493839263914,
      "num_input_tokens_seen": 8128035648,
      "step": 28580,
      "train_runtime": 277940.762,
      "train_tokens_per_second": 29243.77
    },
    {
      "epoch": 1.011569922623931,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5452956903038445e-05,
      "loss": 0.3919403076171875,
      "num_input_tokens_seen": 8130815040,
      "step": 28590,
      "train_runtime": 278033.6585,
      "train_tokens_per_second": 29243.995
    },
    {
      "epoch": 1.01192374288766,
      "grad_norm": 0.58203125,
      "learning_rate": 2.5449659588902468e-05,
      "loss": 0.38667960166931153,
      "num_input_tokens_seen": 8133671872,
      "step": 28600,
      "train_runtime": 278132.4553,
      "train_tokens_per_second": 29243.879
    },
    {
      "epoch": 1.0122775631513887,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5446363555890464e-05,
      "loss": 0.39066686630249026,
      "num_input_tokens_seen": 8136553088,
      "step": 28610,
      "train_runtime": 278230.0244,
      "train_tokens_per_second": 29243.979
    },
    {
      "epoch": 1.0126313834151175,
      "grad_norm": 0.59375,
      "learning_rate": 2.5443068803173025e-05,
      "loss": 0.3922752857208252,
      "num_input_tokens_seen": 8139360896,
      "step": 28620,
      "train_runtime": 278325.0413,
      "train_tokens_per_second": 29244.084
    },
    {
      "epoch": 1.0129852036788463,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5439775329921533e-05,
      "loss": 0.3946134090423584,
      "num_input_tokens_seen": 8142155072,
      "step": 28630,
      "train_runtime": 278418.2579,
      "train_tokens_per_second": 29244.329
    },
    {
      "epoch": 1.013339023942575,
      "grad_norm": 0.6171875,
      "learning_rate": 2.5436483135308092e-05,
      "loss": 0.3948752641677856,
      "num_input_tokens_seen": 8144966400,
      "step": 28640,
      "train_runtime": 278513.889,
      "train_tokens_per_second": 29244.381
    },
    {
      "epoch": 1.0136928442063038,
      "grad_norm": 0.5703125,
      "learning_rate": 2.543319221850557e-05,
      "loss": 0.3923147916793823,
      "num_input_tokens_seen": 8147819264,
      "step": 28650,
      "train_runtime": 278610.3917,
      "train_tokens_per_second": 29244.492
    },
    {
      "epoch": 1.0140466644700326,
      "grad_norm": 0.60546875,
      "learning_rate": 2.542990257868757e-05,
      "loss": 0.39341349601745607,
      "num_input_tokens_seen": 8150625984,
      "step": 28660,
      "train_runtime": 278705.637,
      "train_tokens_per_second": 29244.568
    },
    {
      "epoch": 1.0144004847337613,
      "grad_norm": 0.62890625,
      "learning_rate": 2.542661421502846e-05,
      "loss": 0.39618620872497556,
      "num_input_tokens_seen": 8153499264,
      "step": 28670,
      "train_runtime": 278803.3415,
      "train_tokens_per_second": 29244.625
    },
    {
      "epoch": 1.01475430499749,
      "grad_norm": 0.61328125,
      "learning_rate": 2.5423327126703344e-05,
      "loss": 0.392274808883667,
      "num_input_tokens_seen": 8156353152,
      "step": 28680,
      "train_runtime": 278900.3714,
      "train_tokens_per_second": 29244.684
    },
    {
      "epoch": 1.0151081252612189,
      "grad_norm": 0.5703125,
      "learning_rate": 2.5420041312888078e-05,
      "loss": 0.3968669414520264,
      "num_input_tokens_seen": 8159211072,
      "step": 28690,
      "train_runtime": 278997.2729,
      "train_tokens_per_second": 29244.77
    },
    {
      "epoch": 1.0154619455249476,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5416756772759254e-05,
      "loss": 0.38858857154846194,
      "num_input_tokens_seen": 8162032448,
      "step": 28700,
      "train_runtime": 279090.5133,
      "train_tokens_per_second": 29245.109
    },
    {
      "epoch": 1.0158157657886764,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5413473505494222e-05,
      "loss": 0.3933609962463379,
      "num_input_tokens_seen": 8164826816,
      "step": 28710,
      "train_runtime": 279187.8334,
      "train_tokens_per_second": 29244.923
    },
    {
      "epoch": 1.0161695860524051,
      "grad_norm": 0.59375,
      "learning_rate": 2.5410191510271063e-05,
      "loss": 0.3880609512329102,
      "num_input_tokens_seen": 8167666688,
      "step": 28720,
      "train_runtime": 279283.8765,
      "train_tokens_per_second": 29245.035
    },
    {
      "epoch": 1.016523406316134,
      "grad_norm": 0.5703125,
      "learning_rate": 2.5406910786268613e-05,
      "loss": 0.39503769874572753,
      "num_input_tokens_seen": 8170509952,
      "step": 28730,
      "train_runtime": 279382.4255,
      "train_tokens_per_second": 29244.896
    },
    {
      "epoch": 1.0168772265798627,
      "grad_norm": 0.61328125,
      "learning_rate": 2.5403631332666444e-05,
      "loss": 0.39555599689483645,
      "num_input_tokens_seen": 8173393088,
      "step": 28740,
      "train_runtime": 279481.2334,
      "train_tokens_per_second": 29244.873
    },
    {
      "epoch": 1.0172310468435914,
      "grad_norm": 0.6328125,
      "learning_rate": 2.5400353148644863e-05,
      "loss": 0.389239764213562,
      "num_input_tokens_seen": 8176212416,
      "step": 28750,
      "train_runtime": 279578.2223,
      "train_tokens_per_second": 29244.812
    },
    {
      "epoch": 1.0175848671073202,
      "grad_norm": 0.57421875,
      "learning_rate": 2.5397076233384925e-05,
      "loss": 0.3915641069412231,
      "num_input_tokens_seen": 8179049152,
      "step": 28760,
      "train_runtime": 279674.1033,
      "train_tokens_per_second": 29244.928
    },
    {
      "epoch": 1.0179386873710492,
      "grad_norm": 0.6015625,
      "learning_rate": 2.539380058606843e-05,
      "loss": 0.39095616340637207,
      "num_input_tokens_seen": 8181930752,
      "step": 28770,
      "train_runtime": 279772.2446,
      "train_tokens_per_second": 29244.969
    },
    {
      "epoch": 1.018292507634778,
      "grad_norm": 0.6171875,
      "learning_rate": 2.5390526205877895e-05,
      "loss": 0.3957646369934082,
      "num_input_tokens_seen": 8184747072,
      "step": 28780,
      "train_runtime": 279867.1103,
      "train_tokens_per_second": 29245.119
    },
    {
      "epoch": 1.0186463278985067,
      "grad_norm": 0.60546875,
      "learning_rate": 2.53872530919966e-05,
      "loss": 0.3865208625793457,
      "num_input_tokens_seen": 8187612480,
      "step": 28790,
      "train_runtime": 279967.3498,
      "train_tokens_per_second": 29244.883
    },
    {
      "epoch": 1.0190001481622355,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5383981243608547e-05,
      "loss": 0.3902866840362549,
      "num_input_tokens_seen": 8190449536,
      "step": 28800,
      "train_runtime": 280064.0818,
      "train_tokens_per_second": 29244.912
    },
    {
      "epoch": 1.0193539684259643,
      "grad_norm": 0.59375,
      "learning_rate": 2.5380710659898476e-05,
      "loss": 0.39726145267486573,
      "num_input_tokens_seen": 8193309440,
      "step": 28810,
      "train_runtime": 280161.9844,
      "train_tokens_per_second": 29244.901
    },
    {
      "epoch": 1.019707788689693,
      "grad_norm": 0.6015625,
      "learning_rate": 2.537744134005187e-05,
      "loss": 0.3934477806091309,
      "num_input_tokens_seen": 8196130880,
      "step": 28820,
      "train_runtime": 280259.0353,
      "train_tokens_per_second": 29244.841
    },
    {
      "epoch": 1.0200616089534218,
      "grad_norm": 0.6015625,
      "learning_rate": 2.537417328325493e-05,
      "loss": 0.39667491912841796,
      "num_input_tokens_seen": 8198935744,
      "step": 28830,
      "train_runtime": 280352.8095,
      "train_tokens_per_second": 29245.064
    },
    {
      "epoch": 1.0204154292171506,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5370906488694606e-05,
      "loss": 0.3903061389923096,
      "num_input_tokens_seen": 8201811456,
      "step": 28840,
      "train_runtime": 280452.0619,
      "train_tokens_per_second": 29244.968
    },
    {
      "epoch": 1.0207692494808793,
      "grad_norm": 0.5703125,
      "learning_rate": 2.536764095555857e-05,
      "loss": 0.3888975143432617,
      "num_input_tokens_seen": 8204692032,
      "step": 28850,
      "train_runtime": 280550.8779,
      "train_tokens_per_second": 29244.934
    },
    {
      "epoch": 1.021123069744608,
      "grad_norm": 0.59375,
      "learning_rate": 2.536437668303524e-05,
      "loss": 0.3873569488525391,
      "num_input_tokens_seen": 8207528256,
      "step": 28860,
      "train_runtime": 280649.1832,
      "train_tokens_per_second": 29244.797
    },
    {
      "epoch": 1.0214768900083369,
      "grad_norm": 0.58203125,
      "learning_rate": 2.5361113670313745e-05,
      "loss": 0.3949462890625,
      "num_input_tokens_seen": 8210386176,
      "step": 28870,
      "train_runtime": 280748.3918,
      "train_tokens_per_second": 29244.642
    },
    {
      "epoch": 1.0218307102720656,
      "grad_norm": 0.625,
      "learning_rate": 2.5357851916583964e-05,
      "loss": 0.3935708045959473,
      "num_input_tokens_seen": 8213194304,
      "step": 28880,
      "train_runtime": 280843.3946,
      "train_tokens_per_second": 29244.748
    },
    {
      "epoch": 1.0221845305357944,
      "grad_norm": 0.6171875,
      "learning_rate": 2.5354591421036485e-05,
      "loss": 0.3913459300994873,
      "num_input_tokens_seen": 8216071488,
      "step": 28890,
      "train_runtime": 280942.0698,
      "train_tokens_per_second": 29244.718
    },
    {
      "epoch": 1.0225383507995232,
      "grad_norm": 0.609375,
      "learning_rate": 2.5351332182862643e-05,
      "loss": 0.3918105125427246,
      "num_input_tokens_seen": 8218899968,
      "step": 28900,
      "train_runtime": 281037.8746,
      "train_tokens_per_second": 29244.813
    },
    {
      "epoch": 1.022892171063252,
      "grad_norm": 0.59375,
      "learning_rate": 2.5348074201254484e-05,
      "loss": 0.3912977933883667,
      "num_input_tokens_seen": 8221649344,
      "step": 28910,
      "train_runtime": 281129.0699,
      "train_tokens_per_second": 29245.106
    },
    {
      "epoch": 1.0232459913269807,
      "grad_norm": 0.61328125,
      "learning_rate": 2.53448174754048e-05,
      "loss": 0.3941735982894897,
      "num_input_tokens_seen": 8224446976,
      "step": 28920,
      "train_runtime": 281222.128,
      "train_tokens_per_second": 29245.376
    },
    {
      "epoch": 1.0235998115907097,
      "grad_norm": 0.61328125,
      "learning_rate": 2.5341562004507087e-05,
      "loss": 0.39432220458984374,
      "num_input_tokens_seen": 8227288704,
      "step": 28930,
      "train_runtime": 281316.8823,
      "train_tokens_per_second": 29245.627
    },
    {
      "epoch": 1.0239536318544384,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5338307787755583e-05,
      "loss": 0.3874164581298828,
      "num_input_tokens_seen": 8230196608,
      "step": 28940,
      "train_runtime": 281417.625,
      "train_tokens_per_second": 29245.491
    },
    {
      "epoch": 1.0243074521181672,
      "grad_norm": 0.59375,
      "learning_rate": 2.533505482434524e-05,
      "loss": 0.390718412399292,
      "num_input_tokens_seen": 8233026688,
      "step": 28950,
      "train_runtime": 281514.7085,
      "train_tokens_per_second": 29245.458
    },
    {
      "epoch": 1.024661272381896,
      "grad_norm": 0.59375,
      "learning_rate": 2.533180311347174e-05,
      "loss": 0.391047739982605,
      "num_input_tokens_seen": 8235850176,
      "step": 28960,
      "train_runtime": 281608.7161,
      "train_tokens_per_second": 29245.722
    },
    {
      "epoch": 1.0250150926456247,
      "grad_norm": 0.60546875,
      "learning_rate": 2.532855265433149e-05,
      "loss": 0.3936030387878418,
      "num_input_tokens_seen": 8238641280,
      "step": 28970,
      "train_runtime": 281705.395,
      "train_tokens_per_second": 29245.593
    },
    {
      "epoch": 1.0253689129093535,
      "grad_norm": 0.61328125,
      "learning_rate": 2.5325303446121602e-05,
      "loss": 0.3960809469223022,
      "num_input_tokens_seen": 8241520448,
      "step": 28980,
      "train_runtime": 281807.0434,
      "train_tokens_per_second": 29245.261
    },
    {
      "epoch": 1.0257227331730823,
      "grad_norm": 0.578125,
      "learning_rate": 2.532205548803992e-05,
      "loss": 0.391988205909729,
      "num_input_tokens_seen": 8244379776,
      "step": 28990,
      "train_runtime": 281905.4003,
      "train_tokens_per_second": 29245.2
    },
    {
      "epoch": 1.026076553436811,
      "grad_norm": 0.59765625,
      "learning_rate": 2.531880877928502e-05,
      "loss": 0.3893109321594238,
      "num_input_tokens_seen": 8247211264,
      "step": 29000,
      "train_runtime": 282000.9201,
      "train_tokens_per_second": 29245.335
    },
    {
      "epoch": 1.0264303737005398,
      "grad_norm": 0.5703125,
      "learning_rate": 2.5315563319056173e-05,
      "loss": 0.3934291124343872,
      "num_input_tokens_seen": 8250072320,
      "step": 29010,
      "train_runtime": 282100.5032,
      "train_tokens_per_second": 29245.153
    },
    {
      "epoch": 1.0267841939642686,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5312319106553384e-05,
      "loss": 0.3905926704406738,
      "num_input_tokens_seen": 8252970432,
      "step": 29020,
      "train_runtime": 282198.289,
      "train_tokens_per_second": 29245.289
    },
    {
      "epoch": 1.0271380142279973,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5309076140977378e-05,
      "loss": 0.3938968896865845,
      "num_input_tokens_seen": 8255750656,
      "step": 29030,
      "train_runtime": 282291.0442,
      "train_tokens_per_second": 29245.528
    },
    {
      "epoch": 1.027491834491726,
      "grad_norm": 0.5859375,
      "learning_rate": 2.530583442152958e-05,
      "loss": 0.3917114734649658,
      "num_input_tokens_seen": 8258627008,
      "step": 29040,
      "train_runtime": 282390.9767,
      "train_tokens_per_second": 29245.364
    },
    {
      "epoch": 1.0278456547554549,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5302593947412147e-05,
      "loss": 0.3924980640411377,
      "num_input_tokens_seen": 8261513408,
      "step": 29050,
      "train_runtime": 282489.3091,
      "train_tokens_per_second": 29245.402
    },
    {
      "epoch": 1.0281994750191836,
      "grad_norm": 0.60546875,
      "learning_rate": 2.529935471782794e-05,
      "loss": 0.3921475887298584,
      "num_input_tokens_seen": 8264319104,
      "step": 29060,
      "train_runtime": 282585.9333,
      "train_tokens_per_second": 29245.331
    },
    {
      "epoch": 1.0285532952829124,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5296116731980547e-05,
      "loss": 0.38964266777038575,
      "num_input_tokens_seen": 8267183424,
      "step": 29070,
      "train_runtime": 282684.9284,
      "train_tokens_per_second": 29245.222
    },
    {
      "epoch": 1.0289071155466412,
      "grad_norm": 0.62890625,
      "learning_rate": 2.529287998907426e-05,
      "loss": 0.38799319267272947,
      "num_input_tokens_seen": 8270033920,
      "step": 29080,
      "train_runtime": 282782.373,
      "train_tokens_per_second": 29245.224
    },
    {
      "epoch": 1.02926093581037,
      "grad_norm": 0.59375,
      "learning_rate": 2.528964448831408e-05,
      "loss": 0.3917055606842041,
      "num_input_tokens_seen": 8272901376,
      "step": 29090,
      "train_runtime": 282877.3973,
      "train_tokens_per_second": 29245.537
    },
    {
      "epoch": 1.029614756074099,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5286410228905727e-05,
      "loss": 0.3961790084838867,
      "num_input_tokens_seen": 8275804544,
      "step": 29100,
      "train_runtime": 282979.0001,
      "train_tokens_per_second": 29245.296
    },
    {
      "epoch": 1.0299685763378277,
      "grad_norm": 0.5703125,
      "learning_rate": 2.5283177210055632e-05,
      "loss": 0.39419353008270264,
      "num_input_tokens_seen": 8278685440,
      "step": 29110,
      "train_runtime": 283076.32,
      "train_tokens_per_second": 29245.418
    },
    {
      "epoch": 1.0303223966015564,
      "grad_norm": 0.609375,
      "learning_rate": 2.527994543097093e-05,
      "loss": 0.39393742084503175,
      "num_input_tokens_seen": 8281519168,
      "step": 29120,
      "train_runtime": 283170.7197,
      "train_tokens_per_second": 29245.676
    },
    {
      "epoch": 1.0306762168652852,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5276714890859472e-05,
      "loss": 0.3897049903869629,
      "num_input_tokens_seen": 8284398976,
      "step": 29130,
      "train_runtime": 283269.3824,
      "train_tokens_per_second": 29245.656
    },
    {
      "epoch": 1.031030037129014,
      "grad_norm": 0.5703125,
      "learning_rate": 2.527348558892981e-05,
      "loss": 0.388124418258667,
      "num_input_tokens_seen": 8287274432,
      "step": 29140,
      "train_runtime": 283368.6999,
      "train_tokens_per_second": 29245.553
    },
    {
      "epoch": 1.0313838573927427,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5270257524391223e-05,
      "loss": 0.39238662719726564,
      "num_input_tokens_seen": 8290079808,
      "step": 29150,
      "train_runtime": 283465.155,
      "train_tokens_per_second": 29245.499
    },
    {
      "epoch": 1.0317376776564715,
      "grad_norm": 0.58984375,
      "learning_rate": 2.526703069645366e-05,
      "loss": 0.38860769271850587,
      "num_input_tokens_seen": 8292949888,
      "step": 29160,
      "train_runtime": 283564.6928,
      "train_tokens_per_second": 29245.354
    },
    {
      "epoch": 1.0320914979202003,
      "grad_norm": 0.578125,
      "learning_rate": 2.5263805104327815e-05,
      "loss": 0.3866438865661621,
      "num_input_tokens_seen": 8295789568,
      "step": 29170,
      "train_runtime": 283661.722,
      "train_tokens_per_second": 29245.361
    },
    {
      "epoch": 1.032445318183929,
      "grad_norm": 0.58984375,
      "learning_rate": 2.526058074722506e-05,
      "loss": 0.3949902534484863,
      "num_input_tokens_seen": 8298608576,
      "step": 29180,
      "train_runtime": 283755.695,
      "train_tokens_per_second": 29245.611
    },
    {
      "epoch": 1.0327991384476578,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5257357624357485e-05,
      "loss": 0.39396042823791505,
      "num_input_tokens_seen": 8301422400,
      "step": 29190,
      "train_runtime": 283851.9774,
      "train_tokens_per_second": 29245.604
    },
    {
      "epoch": 1.0331529587113866,
      "grad_norm": 0.58203125,
      "learning_rate": 2.525413573493788e-05,
      "loss": 0.3916712522506714,
      "num_input_tokens_seen": 8304241792,
      "step": 29200,
      "train_runtime": 283947.9157,
      "train_tokens_per_second": 29245.652
    },
    {
      "epoch": 1.0335067789751153,
      "grad_norm": 0.58203125,
      "learning_rate": 2.525091507817974e-05,
      "loss": 0.39437236785888674,
      "num_input_tokens_seen": 8307077248,
      "step": 29210,
      "train_runtime": 284044.0184,
      "train_tokens_per_second": 29245.739
    },
    {
      "epoch": 1.033860599238844,
      "grad_norm": 0.59375,
      "learning_rate": 2.524769565329725e-05,
      "loss": 0.3935762166976929,
      "num_input_tokens_seen": 8309927040,
      "step": 29220,
      "train_runtime": 284141.5342,
      "train_tokens_per_second": 29245.732
    },
    {
      "epoch": 1.0342144195025729,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5244477459505317e-05,
      "loss": 0.38967466354370117,
      "num_input_tokens_seen": 8312778944,
      "step": 29230,
      "train_runtime": 284241.9251,
      "train_tokens_per_second": 29245.436
    },
    {
      "epoch": 1.0345682397663016,
      "grad_norm": 0.58984375,
      "learning_rate": 2.524126049601953e-05,
      "loss": 0.3948280572891235,
      "num_input_tokens_seen": 8315668160,
      "step": 29240,
      "train_runtime": 284341.5166,
      "train_tokens_per_second": 29245.353
    },
    {
      "epoch": 1.0349220600300304,
      "grad_norm": 0.65625,
      "learning_rate": 2.523804476205619e-05,
      "loss": 0.3941455841064453,
      "num_input_tokens_seen": 8318459328,
      "step": 29250,
      "train_runtime": 284434.9622,
      "train_tokens_per_second": 29245.559
    },
    {
      "epoch": 1.0352758802937592,
      "grad_norm": 0.609375,
      "learning_rate": 2.5234830256832292e-05,
      "loss": 0.3910728931427002,
      "num_input_tokens_seen": 8321299904,
      "step": 29260,
      "train_runtime": 284534.4454,
      "train_tokens_per_second": 29245.316
    },
    {
      "epoch": 1.0356297005574882,
      "grad_norm": 0.59765625,
      "learning_rate": 2.523161697956552e-05,
      "loss": 0.39293575286865234,
      "num_input_tokens_seen": 8324117120,
      "step": 29270,
      "train_runtime": 284629.3624,
      "train_tokens_per_second": 29245.462
    },
    {
      "epoch": 1.035983520821217,
      "grad_norm": 0.59765625,
      "learning_rate": 2.522840492947427e-05,
      "loss": 0.3858278751373291,
      "num_input_tokens_seen": 8326990528,
      "step": 29280,
      "train_runtime": 284728.6879,
      "train_tokens_per_second": 29245.351
    },
    {
      "epoch": 1.0363373410849457,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5225194105777628e-05,
      "loss": 0.3904426574707031,
      "num_input_tokens_seen": 8329845760,
      "step": 29290,
      "train_runtime": 284827.7666,
      "train_tokens_per_second": 29245.203
    },
    {
      "epoch": 1.0366911613486745,
      "grad_norm": 0.58203125,
      "learning_rate": 2.5221984507695367e-05,
      "loss": 0.39294967651367185,
      "num_input_tokens_seen": 8332707136,
      "step": 29300,
      "train_runtime": 284923.6179,
      "train_tokens_per_second": 29245.407
    },
    {
      "epoch": 1.0370449816124032,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5218776134447975e-05,
      "loss": 0.39381191730499265,
      "num_input_tokens_seen": 8335544064,
      "step": 29310,
      "train_runtime": 285021.1171,
      "train_tokens_per_second": 29245.356
    },
    {
      "epoch": 1.037398801876132,
      "grad_norm": 0.58203125,
      "learning_rate": 2.5215568985256617e-05,
      "loss": 0.39267659187316895,
      "num_input_tokens_seen": 8338327616,
      "step": 29320,
      "train_runtime": 285113.7349,
      "train_tokens_per_second": 29245.619
    },
    {
      "epoch": 1.0377526221398607,
      "grad_norm": 0.609375,
      "learning_rate": 2.5212363059343154e-05,
      "loss": 0.3902092456817627,
      "num_input_tokens_seen": 8341245440,
      "step": 29330,
      "train_runtime": 285214.8337,
      "train_tokens_per_second": 29245.483
    },
    {
      "epoch": 1.0381064424035895,
      "grad_norm": 0.609375,
      "learning_rate": 2.520915835593014e-05,
      "loss": 0.3982834339141846,
      "num_input_tokens_seen": 8344038784,
      "step": 29340,
      "train_runtime": 285308.3448,
      "train_tokens_per_second": 29245.688
    },
    {
      "epoch": 1.0384602626673183,
      "grad_norm": 0.61328125,
      "learning_rate": 2.5205954874240828e-05,
      "loss": 0.3928164005279541,
      "num_input_tokens_seen": 8346882368,
      "step": 29350,
      "train_runtime": 285404.1704,
      "train_tokens_per_second": 29245.832
    },
    {
      "epoch": 1.038814082931047,
      "grad_norm": 0.58203125,
      "learning_rate": 2.5202752613499154e-05,
      "loss": 0.3947235107421875,
      "num_input_tokens_seen": 8349779264,
      "step": 29360,
      "train_runtime": 285503.65,
      "train_tokens_per_second": 29245.788
    },
    {
      "epoch": 1.0391679031947758,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5199551572929743e-05,
      "loss": 0.39569816589355467,
      "num_input_tokens_seen": 8352670208,
      "step": 29370,
      "train_runtime": 285604.8208,
      "train_tokens_per_second": 29245.55
    },
    {
      "epoch": 1.0395217234585046,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5196351751757913e-05,
      "loss": 0.39073193073272705,
      "num_input_tokens_seen": 8355529344,
      "step": 29380,
      "train_runtime": 285702.3099,
      "train_tokens_per_second": 29245.579
    },
    {
      "epoch": 1.0398755437222333,
      "grad_norm": 0.62890625,
      "learning_rate": 2.5193153149209664e-05,
      "loss": 0.3907917976379395,
      "num_input_tokens_seen": 8358456000,
      "step": 29390,
      "train_runtime": 285806.7424,
      "train_tokens_per_second": 29245.132
    },
    {
      "epoch": 1.040229363985962,
      "grad_norm": 0.625,
      "learning_rate": 2.5189955764511702e-05,
      "loss": 0.39262776374816893,
      "num_input_tokens_seen": 8361291264,
      "step": 29400,
      "train_runtime": 285904.0918,
      "train_tokens_per_second": 29245.091
    },
    {
      "epoch": 1.0405831842496909,
      "grad_norm": 0.6171875,
      "learning_rate": 2.5186759596891398e-05,
      "loss": 0.38845105171203614,
      "num_input_tokens_seen": 8364111616,
      "step": 29410,
      "train_runtime": 286000.2778,
      "train_tokens_per_second": 29245.117
    },
    {
      "epoch": 1.0409370045134196,
      "grad_norm": 0.6015625,
      "learning_rate": 2.518356464557682e-05,
      "loss": 0.38805246353149414,
      "num_input_tokens_seen": 8366962112,
      "step": 29420,
      "train_runtime": 286097.6714,
      "train_tokens_per_second": 29245.125
    },
    {
      "epoch": 1.0412908247771484,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5180370909796718e-05,
      "loss": 0.39016094207763674,
      "num_input_tokens_seen": 8369782848,
      "step": 29430,
      "train_runtime": 286190.9234,
      "train_tokens_per_second": 29245.452
    },
    {
      "epoch": 1.0416446450408774,
      "grad_norm": 0.59375,
      "learning_rate": 2.5177178388780527e-05,
      "loss": 0.38773393630981445,
      "num_input_tokens_seen": 8372579840,
      "step": 29440,
      "train_runtime": 286283.5771,
      "train_tokens_per_second": 29245.757
    },
    {
      "epoch": 1.0419984653046062,
      "grad_norm": 0.609375,
      "learning_rate": 2.5173987081758372e-05,
      "loss": 0.3923208713531494,
      "num_input_tokens_seen": 8375406144,
      "step": 29450,
      "train_runtime": 286378.8227,
      "train_tokens_per_second": 29245.899
    },
    {
      "epoch": 1.042352285568335,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5170796987961044e-05,
      "loss": 0.39222288131713867,
      "num_input_tokens_seen": 8378244928,
      "step": 29460,
      "train_runtime": 286473.6574,
      "train_tokens_per_second": 29246.127
    },
    {
      "epoch": 1.0427061058320637,
      "grad_norm": 0.58984375,
      "learning_rate": 2.516760810662004e-05,
      "loss": 0.3883936405181885,
      "num_input_tokens_seen": 8381078592,
      "step": 29470,
      "train_runtime": 286570.4073,
      "train_tokens_per_second": 29246.141
    },
    {
      "epoch": 1.0430599260957925,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5164420436967517e-05,
      "loss": 0.38690016269683836,
      "num_input_tokens_seen": 8383939776,
      "step": 29480,
      "train_runtime": 286668.4832,
      "train_tokens_per_second": 29246.116
    },
    {
      "epoch": 1.0434137463595212,
      "grad_norm": 0.59375,
      "learning_rate": 2.516123397823632e-05,
      "loss": 0.38701939582824707,
      "num_input_tokens_seen": 8386789760,
      "step": 29490,
      "train_runtime": 286765.5134,
      "train_tokens_per_second": 29246.159
    },
    {
      "epoch": 1.04376756662325,
      "grad_norm": 0.6171875,
      "learning_rate": 2.5158048729659988e-05,
      "loss": 0.3961660146713257,
      "num_input_tokens_seen": 8389647168,
      "step": 29500,
      "train_runtime": 286863.4687,
      "train_tokens_per_second": 29246.133
    },
    {
      "epoch": 1.0441213868869788,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5154864690472714e-05,
      "loss": 0.3914198398590088,
      "num_input_tokens_seen": 8392480128,
      "step": 29510,
      "train_runtime": 286960.4671,
      "train_tokens_per_second": 29246.12
    },
    {
      "epoch": 1.0444752071507075,
      "grad_norm": 0.62890625,
      "learning_rate": 2.5151681859909383e-05,
      "loss": 0.3919799566268921,
      "num_input_tokens_seen": 8395303808,
      "step": 29520,
      "train_runtime": 287057.03,
      "train_tokens_per_second": 29246.118
    },
    {
      "epoch": 1.0448290274144363,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5148500237205564e-05,
      "loss": 0.3953500986099243,
      "num_input_tokens_seen": 8398207104,
      "step": 29530,
      "train_runtime": 287159.2559,
      "train_tokens_per_second": 29245.817
    },
    {
      "epoch": 1.045182847678165,
      "grad_norm": 0.59765625,
      "learning_rate": 2.514531982159748e-05,
      "loss": 0.39100074768066406,
      "num_input_tokens_seen": 8401092352,
      "step": 29540,
      "train_runtime": 287258.6658,
      "train_tokens_per_second": 29245.74
    },
    {
      "epoch": 1.0455366679418938,
      "grad_norm": 0.61328125,
      "learning_rate": 2.5142140612322056e-05,
      "loss": 0.392481255531311,
      "num_input_tokens_seen": 8403973376,
      "step": 29550,
      "train_runtime": 287358.0226,
      "train_tokens_per_second": 29245.654
    },
    {
      "epoch": 1.0458904882056226,
      "grad_norm": 0.61328125,
      "learning_rate": 2.5138962608616883e-05,
      "loss": 0.39355695247650146,
      "num_input_tokens_seen": 8406783168,
      "step": 29560,
      "train_runtime": 287451.1813,
      "train_tokens_per_second": 29245.951
    },
    {
      "epoch": 1.0462443084693513,
      "grad_norm": 0.58984375,
      "learning_rate": 2.513578580972022e-05,
      "loss": 0.38819451332092286,
      "num_input_tokens_seen": 8409632128,
      "step": 29570,
      "train_runtime": 287548.4739,
      "train_tokens_per_second": 29245.963
    },
    {
      "epoch": 1.0465981287330801,
      "grad_norm": 0.6171875,
      "learning_rate": 2.5132610214871e-05,
      "loss": 0.3934628486633301,
      "num_input_tokens_seen": 8412488576,
      "step": 29580,
      "train_runtime": 287646.1058,
      "train_tokens_per_second": 29245.967
    },
    {
      "epoch": 1.0469519489968089,
      "grad_norm": 0.6171875,
      "learning_rate": 2.5129435823308835e-05,
      "loss": 0.38758509159088134,
      "num_input_tokens_seen": 8415376256,
      "step": 29590,
      "train_runtime": 287744.4316,
      "train_tokens_per_second": 29246.009
    },
    {
      "epoch": 1.0473057692605379,
      "grad_norm": 0.58203125,
      "learning_rate": 2.5126262634274012e-05,
      "loss": 0.3935369968414307,
      "num_input_tokens_seen": 8418174016,
      "step": 29600,
      "train_runtime": 287840.7621,
      "train_tokens_per_second": 29245.941
    },
    {
      "epoch": 1.0476595895242666,
      "grad_norm": 0.60546875,
      "learning_rate": 2.512309064700748e-05,
      "loss": 0.39109344482421876,
      "num_input_tokens_seen": 8420998912,
      "step": 29610,
      "train_runtime": 287937.6684,
      "train_tokens_per_second": 29245.909
    },
    {
      "epoch": 1.0480134097879954,
      "grad_norm": 0.59765625,
      "learning_rate": 2.511991986075086e-05,
      "loss": 0.3876230239868164,
      "num_input_tokens_seen": 8423861184,
      "step": 29620,
      "train_runtime": 288034.9024,
      "train_tokens_per_second": 29245.974
    },
    {
      "epoch": 1.0483672300517242,
      "grad_norm": 0.6015625,
      "learning_rate": 2.511675027474645e-05,
      "loss": 0.39133505821228026,
      "num_input_tokens_seen": 8426761344,
      "step": 29630,
      "train_runtime": 288133.4812,
      "train_tokens_per_second": 29246.033
    },
    {
      "epoch": 1.048721050315453,
      "grad_norm": 0.6171875,
      "learning_rate": 2.5113581888237215e-05,
      "loss": 0.39363951683044435,
      "num_input_tokens_seen": 8429587008,
      "step": 29640,
      "train_runtime": 288229.0876,
      "train_tokens_per_second": 29246.136
    },
    {
      "epoch": 1.0490748705791817,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5110414700466782e-05,
      "loss": 0.38858904838562014,
      "num_input_tokens_seen": 8432394048,
      "step": 29650,
      "train_runtime": 288325.6057,
      "train_tokens_per_second": 29246.081
    },
    {
      "epoch": 1.0494286908429105,
      "grad_norm": 0.59375,
      "learning_rate": 2.5107248710679447e-05,
      "loss": 0.3939572095870972,
      "num_input_tokens_seen": 8435300096,
      "step": 29660,
      "train_runtime": 288424.8863,
      "train_tokens_per_second": 29246.09
    },
    {
      "epoch": 1.0497825111066392,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5104083918120187e-05,
      "loss": 0.3967156887054443,
      "num_input_tokens_seen": 8438190336,
      "step": 29670,
      "train_runtime": 288525.1346,
      "train_tokens_per_second": 29245.945
    },
    {
      "epoch": 1.050136331370368,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5100920322034615e-05,
      "loss": 0.39824814796447755,
      "num_input_tokens_seen": 8441038080,
      "step": 29680,
      "train_runtime": 288623.869,
      "train_tokens_per_second": 29245.807
    },
    {
      "epoch": 1.0504901516340968,
      "grad_norm": 0.62109375,
      "learning_rate": 2.5097757921669047e-05,
      "loss": 0.3882847785949707,
      "num_input_tokens_seen": 8443878784,
      "step": 29690,
      "train_runtime": 288721.6497,
      "train_tokens_per_second": 29245.742
    },
    {
      "epoch": 1.0508439718978255,
      "grad_norm": 0.6015625,
      "learning_rate": 2.509459671627043e-05,
      "loss": 0.39636979103088377,
      "num_input_tokens_seen": 8446693824,
      "step": 29700,
      "train_runtime": 288816.1708,
      "train_tokens_per_second": 29245.917
    },
    {
      "epoch": 1.0511977921615543,
      "grad_norm": 0.59375,
      "learning_rate": 2.50914367050864e-05,
      "loss": 0.39093470573425293,
      "num_input_tokens_seen": 8449562112,
      "step": 29710,
      "train_runtime": 288914.1834,
      "train_tokens_per_second": 29245.924
    },
    {
      "epoch": 1.051551612425283,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5088277887365236e-05,
      "loss": 0.3928231239318848,
      "num_input_tokens_seen": 8452375104,
      "step": 29720,
      "train_runtime": 289009.2335,
      "train_tokens_per_second": 29246.038
    },
    {
      "epoch": 1.0519054326890118,
      "grad_norm": 0.578125,
      "learning_rate": 2.5085120262355895e-05,
      "loss": 0.3927906513214111,
      "num_input_tokens_seen": 8455174848,
      "step": 29730,
      "train_runtime": 289102.6452,
      "train_tokens_per_second": 29246.273
    },
    {
      "epoch": 1.0522592529527406,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5081963829307987e-05,
      "loss": 0.3902306079864502,
      "num_input_tokens_seen": 8458032256,
      "step": 29740,
      "train_runtime": 289202.9291,
      "train_tokens_per_second": 29246.012
    },
    {
      "epoch": 1.0526130732164694,
      "grad_norm": 0.60546875,
      "learning_rate": 2.507880858747178e-05,
      "loss": 0.3920453548431396,
      "num_input_tokens_seen": 8460895680,
      "step": 29750,
      "train_runtime": 289299.0348,
      "train_tokens_per_second": 29246.194
    },
    {
      "epoch": 1.0529668934801983,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5075654536098213e-05,
      "loss": 0.3894463300704956,
      "num_input_tokens_seen": 8463737536,
      "step": 29760,
      "train_runtime": 289395.1597,
      "train_tokens_per_second": 29246.3
    },
    {
      "epoch": 1.053320713743927,
      "grad_norm": 0.62890625,
      "learning_rate": 2.5072501674438875e-05,
      "loss": 0.39032104015350344,
      "num_input_tokens_seen": 8466601408,
      "step": 29770,
      "train_runtime": 289494.8925,
      "train_tokens_per_second": 29246.117
    },
    {
      "epoch": 1.0536745340076559,
      "grad_norm": 0.5703125,
      "learning_rate": 2.5069350001746017e-05,
      "loss": 0.3964667320251465,
      "num_input_tokens_seen": 8469489536,
      "step": 29780,
      "train_runtime": 289597.6525,
      "train_tokens_per_second": 29245.712
    },
    {
      "epoch": 1.0540283542713846,
      "grad_norm": 0.61328125,
      "learning_rate": 2.506619951727255e-05,
      "loss": 0.392992639541626,
      "num_input_tokens_seen": 8472355136,
      "step": 29790,
      "train_runtime": 289697.756,
      "train_tokens_per_second": 29245.498
    },
    {
      "epoch": 1.0543821745351134,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5063050220272033e-05,
      "loss": 0.3889392137527466,
      "num_input_tokens_seen": 8475185664,
      "step": 29800,
      "train_runtime": 289793.0504,
      "train_tokens_per_second": 29245.648
    },
    {
      "epoch": 1.0547359947988422,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5059902109998696e-05,
      "loss": 0.3884976863861084,
      "num_input_tokens_seen": 8478013504,
      "step": 29810,
      "train_runtime": 289887.7784,
      "train_tokens_per_second": 29245.847
    },
    {
      "epoch": 1.055089815062571,
      "grad_norm": 0.5703125,
      "learning_rate": 2.505675518570741e-05,
      "loss": 0.38781890869140623,
      "num_input_tokens_seen": 8480837376,
      "step": 29820,
      "train_runtime": 289984.0125,
      "train_tokens_per_second": 29245.879
    },
    {
      "epoch": 1.0554436353262997,
      "grad_norm": 0.61328125,
      "learning_rate": 2.505360944665371e-05,
      "loss": 0.38853533267974855,
      "num_input_tokens_seen": 8483714496,
      "step": 29830,
      "train_runtime": 290084.4879,
      "train_tokens_per_second": 29245.668
    },
    {
      "epoch": 1.0557974555900285,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5050464892093777e-05,
      "loss": 0.389754056930542,
      "num_input_tokens_seen": 8486601152,
      "step": 29840,
      "train_runtime": 290185.1493,
      "train_tokens_per_second": 29245.47
    },
    {
      "epoch": 1.0561512758537572,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5047321521284463e-05,
      "loss": 0.39377455711364745,
      "num_input_tokens_seen": 8489445376,
      "step": 29850,
      "train_runtime": 290283.3663,
      "train_tokens_per_second": 29245.373
    },
    {
      "epoch": 1.056505096117486,
      "grad_norm": 0.59375,
      "learning_rate": 2.5044179333483243e-05,
      "loss": 0.39593095779418946,
      "num_input_tokens_seen": 8492256192,
      "step": 29860,
      "train_runtime": 290378.1811,
      "train_tokens_per_second": 29245.504
    },
    {
      "epoch": 1.0568589163812148,
      "grad_norm": 0.62109375,
      "learning_rate": 2.5041038327948274e-05,
      "loss": 0.39040825366973875,
      "num_input_tokens_seen": 8495081984,
      "step": 29870,
      "train_runtime": 290474.3289,
      "train_tokens_per_second": 29245.552
    },
    {
      "epoch": 1.0572127366449435,
      "grad_norm": 0.60546875,
      "learning_rate": 2.5037898503938346e-05,
      "loss": 0.38826441764831543,
      "num_input_tokens_seen": 8497885824,
      "step": 29880,
      "train_runtime": 290569.0143,
      "train_tokens_per_second": 29245.671
    },
    {
      "epoch": 1.0575665569086723,
      "grad_norm": 0.59765625,
      "learning_rate": 2.5034759860712903e-05,
      "loss": 0.3920740604400635,
      "num_input_tokens_seen": 8500759808,
      "step": 29890,
      "train_runtime": 290667.3682,
      "train_tokens_per_second": 29245.663
    },
    {
      "epoch": 1.057920377172401,
      "grad_norm": 0.58984375,
      "learning_rate": 2.5031622397532047e-05,
      "loss": 0.3881415367126465,
      "num_input_tokens_seen": 8503626176,
      "step": 29900,
      "train_runtime": 290762.8635,
      "train_tokens_per_second": 29245.916
    },
    {
      "epoch": 1.0582741974361298,
      "grad_norm": 0.58203125,
      "learning_rate": 2.502848611365651e-05,
      "loss": 0.3909889221191406,
      "num_input_tokens_seen": 8506471488,
      "step": 29910,
      "train_runtime": 290862.676,
      "train_tokens_per_second": 29245.662
    },
    {
      "epoch": 1.0586280176998586,
      "grad_norm": 0.57421875,
      "learning_rate": 2.50253510083477e-05,
      "loss": 0.391344690322876,
      "num_input_tokens_seen": 8509321664,
      "step": 29920,
      "train_runtime": 290957.3531,
      "train_tokens_per_second": 29245.941
    },
    {
      "epoch": 1.0589818379635876,
      "grad_norm": 0.6171875,
      "learning_rate": 2.502221708086765e-05,
      "loss": 0.39687111377716067,
      "num_input_tokens_seen": 8512201600,
      "step": 29930,
      "train_runtime": 291055.5012,
      "train_tokens_per_second": 29245.974
    },
    {
      "epoch": 1.0593356582273163,
      "grad_norm": 0.59375,
      "learning_rate": 2.501908433047904e-05,
      "loss": 0.39103198051452637,
      "num_input_tokens_seen": 8515051712,
      "step": 29940,
      "train_runtime": 291153.0473,
      "train_tokens_per_second": 29245.965
    },
    {
      "epoch": 1.0596894784910451,
      "grad_norm": 0.6015625,
      "learning_rate": 2.501595275644521e-05,
      "loss": 0.39113216400146483,
      "num_input_tokens_seen": 8517857344,
      "step": 29950,
      "train_runtime": 291246.9731,
      "train_tokens_per_second": 29246.166
    },
    {
      "epoch": 1.0600432987547739,
      "grad_norm": 0.5859375,
      "learning_rate": 2.5012822358030136e-05,
      "loss": 0.39150357246398926,
      "num_input_tokens_seen": 8520674944,
      "step": 29960,
      "train_runtime": 291341.0129,
      "train_tokens_per_second": 29246.397
    },
    {
      "epoch": 1.0603971190185026,
      "grad_norm": 0.6015625,
      "learning_rate": 2.500969313449844e-05,
      "loss": 0.3969014406204224,
      "num_input_tokens_seen": 8523534848,
      "step": 29970,
      "train_runtime": 291439.628,
      "train_tokens_per_second": 29246.314
    },
    {
      "epoch": 1.0607509392822314,
      "grad_norm": 0.61328125,
      "learning_rate": 2.500656508511539e-05,
      "loss": 0.3912001609802246,
      "num_input_tokens_seen": 8526372224,
      "step": 29980,
      "train_runtime": 291535.8571,
      "train_tokens_per_second": 29246.393
    },
    {
      "epoch": 1.0611047595459602,
      "grad_norm": 0.59765625,
      "learning_rate": 2.500343820914689e-05,
      "loss": 0.3907221078872681,
      "num_input_tokens_seen": 8529185856,
      "step": 29990,
      "train_runtime": 291632.5782,
      "train_tokens_per_second": 29246.341
    },
    {
      "epoch": 1.061458579809689,
      "grad_norm": 0.6015625,
      "learning_rate": 2.5000312505859502e-05,
      "loss": 0.39152894020080564,
      "num_input_tokens_seen": 8532056576,
      "step": 30000,
      "train_runtime": 291732.4245,
      "train_tokens_per_second": 29246.172
    },
    {
      "epoch": 1.061458579809689,
      "eval_loss": 0.33860355615615845,
      "eval_runtime": 8.4123,
      "eval_samples_per_second": 474.068,
      "eval_steps_per_second": 3.804,
      "num_input_tokens_seen": 8532056576,
      "step": 30000
    },
    {
      "epoch": 1.0618124000734177,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4997187974520408e-05,
      "loss": 0.38997437953948977,
      "num_input_tokens_seen": 8534891904,
      "step": 30010,
      "train_runtime": 291861.1082,
      "train_tokens_per_second": 29242.991
    },
    {
      "epoch": 1.0621662203371465,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4994064614397442e-05,
      "loss": 0.3899205684661865,
      "num_input_tokens_seen": 8537737984,
      "step": 30020,
      "train_runtime": 291958.4356,
      "train_tokens_per_second": 29242.991
    },
    {
      "epoch": 1.0625200406008752,
      "grad_norm": 0.59765625,
      "learning_rate": 2.499094242475909e-05,
      "loss": 0.39099855422973634,
      "num_input_tokens_seen": 8540617600,
      "step": 30030,
      "train_runtime": 292059.5355,
      "train_tokens_per_second": 29242.728
    },
    {
      "epoch": 1.062873860864604,
      "grad_norm": 0.59375,
      "learning_rate": 2.4987821404874463e-05,
      "loss": 0.3915839672088623,
      "num_input_tokens_seen": 8543440512,
      "step": 30040,
      "train_runtime": 292156.2345,
      "train_tokens_per_second": 29242.712
    },
    {
      "epoch": 1.0632276811283328,
      "grad_norm": 0.57421875,
      "learning_rate": 2.49847015540133e-05,
      "loss": 0.39154677391052245,
      "num_input_tokens_seen": 8546294080,
      "step": 30050,
      "train_runtime": 292253.4059,
      "train_tokens_per_second": 29242.753
    },
    {
      "epoch": 1.0635815013920615,
      "grad_norm": 0.625,
      "learning_rate": 2.498158287144601e-05,
      "loss": 0.39322285652160643,
      "num_input_tokens_seen": 8549108480,
      "step": 30060,
      "train_runtime": 292348.6232,
      "train_tokens_per_second": 29242.855
    },
    {
      "epoch": 1.0639353216557903,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4978465356443613e-05,
      "loss": 0.3957035541534424,
      "num_input_tokens_seen": 8551931200,
      "step": 30070,
      "train_runtime": 292445.378,
      "train_tokens_per_second": 29242.832
    },
    {
      "epoch": 1.064289141919519,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4975349008277774e-05,
      "loss": 0.3882534265518188,
      "num_input_tokens_seen": 8554742592,
      "step": 30080,
      "train_runtime": 292542.3738,
      "train_tokens_per_second": 29242.747
    },
    {
      "epoch": 1.0646429621832478,
      "grad_norm": 0.59375,
      "learning_rate": 2.4972233826220794e-05,
      "loss": 0.3938943386077881,
      "num_input_tokens_seen": 8557563520,
      "step": 30090,
      "train_runtime": 292640.0618,
      "train_tokens_per_second": 29242.625
    },
    {
      "epoch": 1.0649967824469768,
      "grad_norm": 0.62109375,
      "learning_rate": 2.496911980954561e-05,
      "loss": 0.3884228467941284,
      "num_input_tokens_seen": 8560374208,
      "step": 30100,
      "train_runtime": 292735.4993,
      "train_tokens_per_second": 29242.693
    },
    {
      "epoch": 1.0653506027107056,
      "grad_norm": 0.60546875,
      "learning_rate": 2.496600695752579e-05,
      "loss": 0.39510948657989503,
      "num_input_tokens_seen": 8563199168,
      "step": 30110,
      "train_runtime": 292832.422,
      "train_tokens_per_second": 29242.661
    },
    {
      "epoch": 1.0657044229744344,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4962895269435543e-05,
      "loss": 0.38712444305419924,
      "num_input_tokens_seen": 8566012160,
      "step": 30120,
      "train_runtime": 292928.7818,
      "train_tokens_per_second": 29242.644
    },
    {
      "epoch": 1.0660582432381631,
      "grad_norm": 0.58984375,
      "learning_rate": 2.49597847445497e-05,
      "loss": 0.38688387870788576,
      "num_input_tokens_seen": 8568842368,
      "step": 30130,
      "train_runtime": 293026.3627,
      "train_tokens_per_second": 29242.565
    },
    {
      "epoch": 1.0664120635018919,
      "grad_norm": 0.58203125,
      "learning_rate": 2.495667538214374e-05,
      "loss": 0.3950577974319458,
      "num_input_tokens_seen": 8571620096,
      "step": 30140,
      "train_runtime": 293118.5252,
      "train_tokens_per_second": 29242.847
    },
    {
      "epoch": 1.0667658837656206,
      "grad_norm": 0.59375,
      "learning_rate": 2.4953567181493754e-05,
      "loss": 0.39014949798583987,
      "num_input_tokens_seen": 8574470656,
      "step": 30150,
      "train_runtime": 293217.5208,
      "train_tokens_per_second": 29242.695
    },
    {
      "epoch": 1.0671197040293494,
      "grad_norm": 0.61328125,
      "learning_rate": 2.495046014187648e-05,
      "loss": 0.3923321723937988,
      "num_input_tokens_seen": 8577330752,
      "step": 30160,
      "train_runtime": 293318.587,
      "train_tokens_per_second": 29242.37
    },
    {
      "epoch": 1.0674735242930782,
      "grad_norm": 0.6015625,
      "learning_rate": 2.494735426256928e-05,
      "loss": 0.3899068355560303,
      "num_input_tokens_seen": 8580153856,
      "step": 30170,
      "train_runtime": 293413.7994,
      "train_tokens_per_second": 29242.503
    },
    {
      "epoch": 1.067827344556807,
      "grad_norm": 0.62109375,
      "learning_rate": 2.4944249542850143e-05,
      "loss": 0.385516619682312,
      "num_input_tokens_seen": 8582978368,
      "step": 30180,
      "train_runtime": 293510.5507,
      "train_tokens_per_second": 29242.487
    },
    {
      "epoch": 1.0681811648205357,
      "grad_norm": 0.58984375,
      "learning_rate": 2.494114598199769e-05,
      "loss": 0.39456958770751954,
      "num_input_tokens_seen": 8585800064,
      "step": 30190,
      "train_runtime": 293603.5382,
      "train_tokens_per_second": 29242.836
    },
    {
      "epoch": 1.0685349850842645,
      "grad_norm": 0.578125,
      "learning_rate": 2.493804357929118e-05,
      "loss": 0.3906743049621582,
      "num_input_tokens_seen": 8588638464,
      "step": 30200,
      "train_runtime": 293700.7107,
      "train_tokens_per_second": 29242.825
    },
    {
      "epoch": 1.0688888053479932,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4934942334010474e-05,
      "loss": 0.3913243770599365,
      "num_input_tokens_seen": 8591515776,
      "step": 30210,
      "train_runtime": 293797.821,
      "train_tokens_per_second": 29242.953
    },
    {
      "epoch": 1.069242625611722,
      "grad_norm": 0.57421875,
      "learning_rate": 2.493184224543609e-05,
      "loss": 0.3911145210266113,
      "num_input_tokens_seen": 8594383872,
      "step": 30220,
      "train_runtime": 293897.5497,
      "train_tokens_per_second": 29242.788
    },
    {
      "epoch": 1.0695964458754508,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4928743312849144e-05,
      "loss": 0.39372868537902833,
      "num_input_tokens_seen": 8597279040,
      "step": 30230,
      "train_runtime": 293996.6376,
      "train_tokens_per_second": 29242.78
    },
    {
      "epoch": 1.0699502661391795,
      "grad_norm": 0.56640625,
      "learning_rate": 2.4925645535531404e-05,
      "loss": 0.3938077688217163,
      "num_input_tokens_seen": 8600124544,
      "step": 30240,
      "train_runtime": 294094.6164,
      "train_tokens_per_second": 29242.713
    },
    {
      "epoch": 1.0703040864029083,
      "grad_norm": 0.59375,
      "learning_rate": 2.4922548912765234e-05,
      "loss": 0.38884749412536623,
      "num_input_tokens_seen": 8603017472,
      "step": 30250,
      "train_runtime": 294193.5933,
      "train_tokens_per_second": 29242.708
    },
    {
      "epoch": 1.070657906666637,
      "grad_norm": 0.58203125,
      "learning_rate": 2.4919453443833642e-05,
      "loss": 0.39093847274780275,
      "num_input_tokens_seen": 8605882240,
      "step": 30260,
      "train_runtime": 294290.5123,
      "train_tokens_per_second": 29242.812
    },
    {
      "epoch": 1.071011726930366,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4916359128020264e-05,
      "loss": 0.3939260482788086,
      "num_input_tokens_seen": 8608732864,
      "step": 30270,
      "train_runtime": 294390.7041,
      "train_tokens_per_second": 29242.543
    },
    {
      "epoch": 1.0713655471940948,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4913265964609342e-05,
      "loss": 0.3872772216796875,
      "num_input_tokens_seen": 8611570944,
      "step": 30280,
      "train_runtime": 294487.3008,
      "train_tokens_per_second": 29242.588
    },
    {
      "epoch": 1.0717193674578236,
      "grad_norm": 0.62890625,
      "learning_rate": 2.4910173952885737e-05,
      "loss": 0.38878345489501953,
      "num_input_tokens_seen": 8614446272,
      "step": 30290,
      "train_runtime": 294584.4167,
      "train_tokens_per_second": 29242.709
    },
    {
      "epoch": 1.0720731877215524,
      "grad_norm": 0.58984375,
      "learning_rate": 2.490708309213496e-05,
      "loss": 0.39459388256072997,
      "num_input_tokens_seen": 8617270976,
      "step": 30300,
      "train_runtime": 294680.9229,
      "train_tokens_per_second": 29242.717
    },
    {
      "epoch": 1.0724270079852811,
      "grad_norm": 0.62890625,
      "learning_rate": 2.4903993381643107e-05,
      "loss": 0.3869975805282593,
      "num_input_tokens_seen": 8620095744,
      "step": 30310,
      "train_runtime": 294776.3496,
      "train_tokens_per_second": 29242.834
    },
    {
      "epoch": 1.0727808282490099,
      "grad_norm": 0.609375,
      "learning_rate": 2.4900904820696916e-05,
      "loss": 0.39280712604522705,
      "num_input_tokens_seen": 8622953920,
      "step": 30320,
      "train_runtime": 294875.9985,
      "train_tokens_per_second": 29242.644
    },
    {
      "epoch": 1.0731346485127387,
      "grad_norm": 0.59765625,
      "learning_rate": 2.489781740858374e-05,
      "loss": 0.39142396450042727,
      "num_input_tokens_seen": 8625767808,
      "step": 30330,
      "train_runtime": 294972.992,
      "train_tokens_per_second": 29242.568
    },
    {
      "epoch": 1.0734884687764674,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4894731144591545e-05,
      "loss": 0.39249932765960693,
      "num_input_tokens_seen": 8628583552,
      "step": 30340,
      "train_runtime": 295068.6476,
      "train_tokens_per_second": 29242.631
    },
    {
      "epoch": 1.0738422890401962,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4891646028008923e-05,
      "loss": 0.39223670959472656,
      "num_input_tokens_seen": 8631424640,
      "step": 30350,
      "train_runtime": 295166.9822,
      "train_tokens_per_second": 29242.514
    },
    {
      "epoch": 1.074196109303925,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4888562058125068e-05,
      "loss": 0.3923863410949707,
      "num_input_tokens_seen": 8634254720,
      "step": 30360,
      "train_runtime": 295263.0897,
      "train_tokens_per_second": 29242.581
    },
    {
      "epoch": 1.0745499295676537,
      "grad_norm": 0.58984375,
      "learning_rate": 2.488547923422981e-05,
      "loss": 0.39220428466796875,
      "num_input_tokens_seen": 8637137792,
      "step": 30370,
      "train_runtime": 295362.7215,
      "train_tokens_per_second": 29242.478
    },
    {
      "epoch": 1.0749037498313825,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4882397555613582e-05,
      "loss": 0.3963631629943848,
      "num_input_tokens_seen": 8639975552,
      "step": 30380,
      "train_runtime": 295458.3872,
      "train_tokens_per_second": 29242.614
    },
    {
      "epoch": 1.0752575700951112,
      "grad_norm": 0.578125,
      "learning_rate": 2.487931702156743e-05,
      "loss": 0.3911131381988525,
      "num_input_tokens_seen": 8642767680,
      "step": 30390,
      "train_runtime": 295554.0562,
      "train_tokens_per_second": 29242.595
    },
    {
      "epoch": 1.07561139035884,
      "grad_norm": 0.59375,
      "learning_rate": 2.4876237631383026e-05,
      "loss": 0.3917921781539917,
      "num_input_tokens_seen": 8645595200,
      "step": 30400,
      "train_runtime": 295648.8771,
      "train_tokens_per_second": 29242.78
    },
    {
      "epoch": 1.0759652106225688,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4873159384352647e-05,
      "loss": 0.3936903953552246,
      "num_input_tokens_seen": 8648428928,
      "step": 30410,
      "train_runtime": 295744.4223,
      "train_tokens_per_second": 29242.915
    },
    {
      "epoch": 1.0763190308862975,
      "grad_norm": 0.609375,
      "learning_rate": 2.487008227976918e-05,
      "loss": 0.3960005283355713,
      "num_input_tokens_seen": 8651207872,
      "step": 30420,
      "train_runtime": 295836.6888,
      "train_tokens_per_second": 29243.188
    },
    {
      "epoch": 1.0766728511500263,
      "grad_norm": 0.6015625,
      "learning_rate": 2.486700631692613e-05,
      "loss": 0.3870538711547852,
      "num_input_tokens_seen": 8654055616,
      "step": 30430,
      "train_runtime": 295933.39,
      "train_tokens_per_second": 29243.255
    },
    {
      "epoch": 1.0770266714137553,
      "grad_norm": 0.61328125,
      "learning_rate": 2.486393149511762e-05,
      "loss": 0.39339210987091067,
      "num_input_tokens_seen": 8656875840,
      "step": 30440,
      "train_runtime": 296029.6617,
      "train_tokens_per_second": 29243.272
    },
    {
      "epoch": 1.077380491677484,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4860857813638365e-05,
      "loss": 0.3922523021697998,
      "num_input_tokens_seen": 8659703808,
      "step": 30450,
      "train_runtime": 296126.5514,
      "train_tokens_per_second": 29243.253
    },
    {
      "epoch": 1.0777343119412128,
      "grad_norm": 0.58984375,
      "learning_rate": 2.485778527178371e-05,
      "loss": 0.3936240911483765,
      "num_input_tokens_seen": 8662570112,
      "step": 30460,
      "train_runtime": 296224.2327,
      "train_tokens_per_second": 29243.287
    },
    {
      "epoch": 1.0780881322049416,
      "grad_norm": 0.578125,
      "learning_rate": 2.48547138688496e-05,
      "loss": 0.39096481800079347,
      "num_input_tokens_seen": 8665451456,
      "step": 30470,
      "train_runtime": 296326.1169,
      "train_tokens_per_second": 29242.956
    },
    {
      "epoch": 1.0784419524686704,
      "grad_norm": 0.61328125,
      "learning_rate": 2.485164360413258e-05,
      "loss": 0.39667930603027346,
      "num_input_tokens_seen": 8668232768,
      "step": 30480,
      "train_runtime": 296422.0746,
      "train_tokens_per_second": 29242.872
    },
    {
      "epoch": 1.0787957727323991,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4848574476929823e-05,
      "loss": 0.3877444267272949,
      "num_input_tokens_seen": 8671091072,
      "step": 30490,
      "train_runtime": 296522.9176,
      "train_tokens_per_second": 29242.566
    },
    {
      "epoch": 1.079149592996128,
      "grad_norm": 0.59375,
      "learning_rate": 2.484550648653909e-05,
      "loss": 0.38915562629699707,
      "num_input_tokens_seen": 8673882240,
      "step": 30500,
      "train_runtime": 296615.8692,
      "train_tokens_per_second": 29242.812
    },
    {
      "epoch": 1.0795034132598567,
      "grad_norm": 0.578125,
      "learning_rate": 2.484243963225877e-05,
      "loss": 0.3889951229095459,
      "num_input_tokens_seen": 8676791552,
      "step": 30510,
      "train_runtime": 296715.2644,
      "train_tokens_per_second": 29242.822
    },
    {
      "epoch": 1.0798572335235854,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4839373913387835e-05,
      "loss": 0.3849780082702637,
      "num_input_tokens_seen": 8679678208,
      "step": 30520,
      "train_runtime": 296812.6414,
      "train_tokens_per_second": 29242.953
    },
    {
      "epoch": 1.0802110537873142,
      "grad_norm": 0.609375,
      "learning_rate": 2.4836309329225873e-05,
      "loss": 0.3888519763946533,
      "num_input_tokens_seen": 8682504064,
      "step": 30530,
      "train_runtime": 296911.5965,
      "train_tokens_per_second": 29242.725
    },
    {
      "epoch": 1.080564874051043,
      "grad_norm": 0.58984375,
      "learning_rate": 2.483324587907309e-05,
      "loss": 0.3945186138153076,
      "num_input_tokens_seen": 8685337408,
      "step": 30540,
      "train_runtime": 297009.4306,
      "train_tokens_per_second": 29242.632
    },
    {
      "epoch": 1.0809186943147717,
      "grad_norm": 0.60546875,
      "learning_rate": 2.483018356223027e-05,
      "loss": 0.3936718463897705,
      "num_input_tokens_seen": 8688214592,
      "step": 30550,
      "train_runtime": 297107.9761,
      "train_tokens_per_second": 29242.616
    },
    {
      "epoch": 1.0812725145785005,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4827122377998815e-05,
      "loss": 0.38601295948028563,
      "num_input_tokens_seen": 8691076224,
      "step": 30560,
      "train_runtime": 297204.5527,
      "train_tokens_per_second": 29242.743
    },
    {
      "epoch": 1.0816263348422293,
      "grad_norm": 0.609375,
      "learning_rate": 2.4824062325680737e-05,
      "loss": 0.38950190544128416,
      "num_input_tokens_seen": 8693898432,
      "step": 30570,
      "train_runtime": 297301.4989,
      "train_tokens_per_second": 29242.7
    },
    {
      "epoch": 1.081980155105958,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4821003404578623e-05,
      "loss": 0.3913022518157959,
      "num_input_tokens_seen": 8696720704,
      "step": 30580,
      "train_runtime": 297396.1197,
      "train_tokens_per_second": 29242.886
    },
    {
      "epoch": 1.082333975369687,
      "grad_norm": 0.609375,
      "learning_rate": 2.4817945613995703e-05,
      "loss": 0.3862746238708496,
      "num_input_tokens_seen": 8699545408,
      "step": 30590,
      "train_runtime": 297488.9721,
      "train_tokens_per_second": 29243.253
    },
    {
      "epoch": 1.0826877956334158,
      "grad_norm": 0.58203125,
      "learning_rate": 2.481488895323577e-05,
      "loss": 0.3895613670349121,
      "num_input_tokens_seen": 8702389504,
      "step": 30600,
      "train_runtime": 297584.367,
      "train_tokens_per_second": 29243.436
    },
    {
      "epoch": 1.0830416158971445,
      "grad_norm": 0.609375,
      "learning_rate": 2.481183342160324e-05,
      "loss": 0.3940150737762451,
      "num_input_tokens_seen": 8705240000,
      "step": 30610,
      "train_runtime": 297682.6145,
      "train_tokens_per_second": 29243.36
    },
    {
      "epoch": 1.0833954361608733,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4808779018403107e-05,
      "loss": 0.38983974456787107,
      "num_input_tokens_seen": 8708054656,
      "step": 30620,
      "train_runtime": 297779.6245,
      "train_tokens_per_second": 29243.286
    },
    {
      "epoch": 1.083749256424602,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4805725742940987e-05,
      "loss": 0.3891287326812744,
      "num_input_tokens_seen": 8710902016,
      "step": 30630,
      "train_runtime": 297877.6803,
      "train_tokens_per_second": 29243.218
    },
    {
      "epoch": 1.0841030766883308,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4802673594523086e-05,
      "loss": 0.3929009437561035,
      "num_input_tokens_seen": 8713767680,
      "step": 30640,
      "train_runtime": 297979.0452,
      "train_tokens_per_second": 29242.887
    },
    {
      "epoch": 1.0844568969520596,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4799622572456198e-05,
      "loss": 0.38962838649749754,
      "num_input_tokens_seen": 8716583552,
      "step": 30650,
      "train_runtime": 298072.9508,
      "train_tokens_per_second": 29243.122
    },
    {
      "epoch": 1.0848107172157884,
      "grad_norm": 0.58984375,
      "learning_rate": 2.479657267604772e-05,
      "loss": 0.3892711639404297,
      "num_input_tokens_seen": 8719445056,
      "step": 30660,
      "train_runtime": 298169.2462,
      "train_tokens_per_second": 29243.274
    },
    {
      "epoch": 1.0851645374795171,
      "grad_norm": 0.5703125,
      "learning_rate": 2.4793523904605657e-05,
      "loss": 0.389400053024292,
      "num_input_tokens_seen": 8722262400,
      "step": 30670,
      "train_runtime": 298264.3982,
      "train_tokens_per_second": 29243.391
    },
    {
      "epoch": 1.085518357743246,
      "grad_norm": 0.5859375,
      "learning_rate": 2.479047625743859e-05,
      "loss": 0.39056341648101806,
      "num_input_tokens_seen": 8725083008,
      "step": 30680,
      "train_runtime": 298360.6569,
      "train_tokens_per_second": 29243.41
    },
    {
      "epoch": 1.0858721780069747,
      "grad_norm": 0.59375,
      "learning_rate": 2.4787429733855697e-05,
      "loss": 0.39096994400024415,
      "num_input_tokens_seen": 8727941952,
      "step": 30690,
      "train_runtime": 298459.0667,
      "train_tokens_per_second": 29243.347
    },
    {
      "epoch": 1.0862259982707034,
      "grad_norm": 0.6015625,
      "learning_rate": 2.478438433316677e-05,
      "loss": 0.3891505241394043,
      "num_input_tokens_seen": 8730815872,
      "step": 30700,
      "train_runtime": 298558.4874,
      "train_tokens_per_second": 29243.235
    },
    {
      "epoch": 1.0865798185344322,
      "grad_norm": 0.6484375,
      "learning_rate": 2.478134005468217e-05,
      "loss": 0.3924570083618164,
      "num_input_tokens_seen": 8733630720,
      "step": 30710,
      "train_runtime": 298654.5416,
      "train_tokens_per_second": 29243.254
    },
    {
      "epoch": 1.086933638798161,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4778296897712865e-05,
      "loss": 0.3891456604003906,
      "num_input_tokens_seen": 8736459968,
      "step": 30720,
      "train_runtime": 298749.7811,
      "train_tokens_per_second": 29243.402
    },
    {
      "epoch": 1.0872874590618897,
      "grad_norm": 0.59375,
      "learning_rate": 2.4775254861570422e-05,
      "loss": 0.39409651756286623,
      "num_input_tokens_seen": 8739337600,
      "step": 30730,
      "train_runtime": 298847.9506,
      "train_tokens_per_second": 29243.425
    },
    {
      "epoch": 1.0876412793256185,
      "grad_norm": 0.63671875,
      "learning_rate": 2.4772213945566974e-05,
      "loss": 0.39090328216552733,
      "num_input_tokens_seen": 8742197120,
      "step": 30740,
      "train_runtime": 298946.2306,
      "train_tokens_per_second": 29243.376
    },
    {
      "epoch": 1.0879950995893473,
      "grad_norm": 0.59375,
      "learning_rate": 2.4769174149015266e-05,
      "loss": 0.39293441772460935,
      "num_input_tokens_seen": 8745038464,
      "step": 30750,
      "train_runtime": 299042.2485,
      "train_tokens_per_second": 29243.488
    },
    {
      "epoch": 1.0883489198530762,
      "grad_norm": 0.61328125,
      "learning_rate": 2.476613547122863e-05,
      "loss": 0.38801114559173583,
      "num_input_tokens_seen": 8747859200,
      "step": 30760,
      "train_runtime": 299135.2603,
      "train_tokens_per_second": 29243.825
    },
    {
      "epoch": 1.088702740116805,
      "grad_norm": 0.58203125,
      "learning_rate": 2.476309791152098e-05,
      "loss": 0.39524762630462645,
      "num_input_tokens_seen": 8750696064,
      "step": 30770,
      "train_runtime": 299234.0534,
      "train_tokens_per_second": 29243.65
    },
    {
      "epoch": 1.0890565603805338,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4760061469206827e-05,
      "loss": 0.3922133445739746,
      "num_input_tokens_seen": 8753561088,
      "step": 30780,
      "train_runtime": 299333.4186,
      "train_tokens_per_second": 29243.514
    },
    {
      "epoch": 1.0894103806442625,
      "grad_norm": 0.609375,
      "learning_rate": 2.4757026143601268e-05,
      "loss": 0.39108834266662595,
      "num_input_tokens_seen": 8756371328,
      "step": 30790,
      "train_runtime": 299425.136,
      "train_tokens_per_second": 29243.942
    },
    {
      "epoch": 1.0897642009079913,
      "grad_norm": 0.5703125,
      "learning_rate": 2.475399193401998e-05,
      "loss": 0.3922947883605957,
      "num_input_tokens_seen": 8759267776,
      "step": 30800,
      "train_runtime": 299525.7874,
      "train_tokens_per_second": 29243.785
    },
    {
      "epoch": 1.09011802117172,
      "grad_norm": 0.62109375,
      "learning_rate": 2.475095883977924e-05,
      "loss": 0.3926777601242065,
      "num_input_tokens_seen": 8762098944,
      "step": 30810,
      "train_runtime": 299623.3458,
      "train_tokens_per_second": 29243.712
    },
    {
      "epoch": 1.0904718414354488,
      "grad_norm": 0.58984375,
      "learning_rate": 2.474792686019591e-05,
      "loss": 0.3913769245147705,
      "num_input_tokens_seen": 8764981312,
      "step": 30820,
      "train_runtime": 299724.0777,
      "train_tokens_per_second": 29243.501
    },
    {
      "epoch": 1.0908256616991776,
      "grad_norm": 0.6015625,
      "learning_rate": 2.474489599458742e-05,
      "loss": 0.39033284187316897,
      "num_input_tokens_seen": 8767852096,
      "step": 30830,
      "train_runtime": 299821.3553,
      "train_tokens_per_second": 29243.588
    },
    {
      "epoch": 1.0911794819629064,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4741866242271804e-05,
      "loss": 0.3910961627960205,
      "num_input_tokens_seen": 8770695360,
      "step": 30840,
      "train_runtime": 299918.2345,
      "train_tokens_per_second": 29243.622
    },
    {
      "epoch": 1.0915333022266351,
      "grad_norm": 0.578125,
      "learning_rate": 2.473883760256767e-05,
      "loss": 0.39322586059570314,
      "num_input_tokens_seen": 8773544896,
      "step": 30850,
      "train_runtime": 300016.6479,
      "train_tokens_per_second": 29243.527
    },
    {
      "epoch": 1.091887122490364,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4735810074794217e-05,
      "loss": 0.39442267417907717,
      "num_input_tokens_seen": 8776427072,
      "step": 30860,
      "train_runtime": 300117.4523,
      "train_tokens_per_second": 29243.308
    },
    {
      "epoch": 1.0922409427540927,
      "grad_norm": 0.59765625,
      "learning_rate": 2.473278365827123e-05,
      "loss": 0.39067783355712893,
      "num_input_tokens_seen": 8779285632,
      "step": 30870,
      "train_runtime": 300215.7366,
      "train_tokens_per_second": 29243.256
    },
    {
      "epoch": 1.0925947630178214,
      "grad_norm": 0.60546875,
      "learning_rate": 2.472975835231906e-05,
      "loss": 0.3886127471923828,
      "num_input_tokens_seen": 8782105792,
      "step": 30880,
      "train_runtime": 300311.6188,
      "train_tokens_per_second": 29243.31
    },
    {
      "epoch": 1.0929485832815502,
      "grad_norm": 0.609375,
      "learning_rate": 2.4726734156258647e-05,
      "loss": 0.39494221210479735,
      "num_input_tokens_seen": 8784940160,
      "step": 30890,
      "train_runtime": 300408.7423,
      "train_tokens_per_second": 29243.291
    },
    {
      "epoch": 1.093302403545279,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4723711069411526e-05,
      "loss": 0.3886883735656738,
      "num_input_tokens_seen": 8787793280,
      "step": 30900,
      "train_runtime": 300506.1262,
      "train_tokens_per_second": 29243.308
    },
    {
      "epoch": 1.0936562238090077,
      "grad_norm": 0.6171875,
      "learning_rate": 2.472068909109979e-05,
      "loss": 0.3853020668029785,
      "num_input_tokens_seen": 8790602368,
      "step": 30910,
      "train_runtime": 300603.7881,
      "train_tokens_per_second": 29243.152
    },
    {
      "epoch": 1.0940100440727365,
      "grad_norm": 0.578125,
      "learning_rate": 2.471766822064614e-05,
      "loss": 0.3937170743942261,
      "num_input_tokens_seen": 8793475968,
      "step": 30920,
      "train_runtime": 300703.3074,
      "train_tokens_per_second": 29243.03
    },
    {
      "epoch": 1.0943638643364655,
      "grad_norm": 0.625,
      "learning_rate": 2.4714648457373824e-05,
      "loss": 0.3961985111236572,
      "num_input_tokens_seen": 8796281344,
      "step": 30930,
      "train_runtime": 300798.3225,
      "train_tokens_per_second": 29243.12
    },
    {
      "epoch": 1.0947176846001943,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4711629800606694e-05,
      "loss": 0.39090895652770996,
      "num_input_tokens_seen": 8799133248,
      "step": 30940,
      "train_runtime": 300896.7396,
      "train_tokens_per_second": 29243.033
    },
    {
      "epoch": 1.095071504863923,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4708612249669157e-05,
      "loss": 0.38562180995941164,
      "num_input_tokens_seen": 8801989376,
      "step": 30950,
      "train_runtime": 300995.2973,
      "train_tokens_per_second": 29242.946
    },
    {
      "epoch": 1.0954253251276518,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4705595803886223e-05,
      "loss": 0.39196100234985354,
      "num_input_tokens_seen": 8804858112,
      "step": 30960,
      "train_runtime": 301095.1445,
      "train_tokens_per_second": 29242.777
    },
    {
      "epoch": 1.0957791453913805,
      "grad_norm": 0.59375,
      "learning_rate": 2.4702580462583468e-05,
      "loss": 0.3861811399459839,
      "num_input_tokens_seen": 8807736384,
      "step": 30970,
      "train_runtime": 301193.0945,
      "train_tokens_per_second": 29242.823
    },
    {
      "epoch": 1.0961329656551093,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4699566225087032e-05,
      "loss": 0.39705095291137693,
      "num_input_tokens_seen": 8810558976,
      "step": 30980,
      "train_runtime": 301286.9113,
      "train_tokens_per_second": 29243.086
    },
    {
      "epoch": 1.096486785918838,
      "grad_norm": 0.578125,
      "learning_rate": 2.4696553090723644e-05,
      "loss": 0.3898041248321533,
      "num_input_tokens_seen": 8813443392,
      "step": 30990,
      "train_runtime": 301385.8839,
      "train_tokens_per_second": 29243.053
    },
    {
      "epoch": 1.0968406061825668,
      "grad_norm": 0.59375,
      "learning_rate": 2.4693541058820606e-05,
      "loss": 0.39616055488586427,
      "num_input_tokens_seen": 8816337728,
      "step": 31000,
      "train_runtime": 301485.4949,
      "train_tokens_per_second": 29242.991
    },
    {
      "epoch": 1.0971944264462956,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4690530128705798e-05,
      "loss": 0.39245381355285647,
      "num_input_tokens_seen": 8819114624,
      "step": 31010,
      "train_runtime": 301576.6941,
      "train_tokens_per_second": 29243.356
    },
    {
      "epoch": 1.0975482467100244,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4687520299707655e-05,
      "loss": 0.3932775020599365,
      "num_input_tokens_seen": 8821949056,
      "step": 31020,
      "train_runtime": 301674.1395,
      "train_tokens_per_second": 29243.306
    },
    {
      "epoch": 1.0979020669737531,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4684511571155208e-05,
      "loss": 0.39643881320953367,
      "num_input_tokens_seen": 8824721216,
      "step": 31030,
      "train_runtime": 301769.5844,
      "train_tokens_per_second": 29243.243
    },
    {
      "epoch": 1.098255887237482,
      "grad_norm": 0.59375,
      "learning_rate": 2.4681503942378048e-05,
      "loss": 0.3923227787017822,
      "num_input_tokens_seen": 8827554688,
      "step": 31040,
      "train_runtime": 301866.2976,
      "train_tokens_per_second": 29243.26
    },
    {
      "epoch": 1.0986097075012107,
      "grad_norm": 0.59375,
      "learning_rate": 2.4678497412706344e-05,
      "loss": 0.3933370590209961,
      "num_input_tokens_seen": 8830366976,
      "step": 31050,
      "train_runtime": 301964.2767,
      "train_tokens_per_second": 29243.085
    },
    {
      "epoch": 1.0989635277649394,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4675491981470825e-05,
      "loss": 0.39458398818969725,
      "num_input_tokens_seen": 8833177664,
      "step": 31060,
      "train_runtime": 302059.1236,
      "train_tokens_per_second": 29243.208
    },
    {
      "epoch": 1.0993173480286682,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4672487648002803e-05,
      "loss": 0.3981754779815674,
      "num_input_tokens_seen": 8835960448,
      "step": 31070,
      "train_runtime": 302155.6452,
      "train_tokens_per_second": 29243.076
    },
    {
      "epoch": 1.099671168292397,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4669484411634147e-05,
      "loss": 0.3921816349029541,
      "num_input_tokens_seen": 8838765440,
      "step": 31080,
      "train_runtime": 302252.7105,
      "train_tokens_per_second": 29242.965
    },
    {
      "epoch": 1.1000249885561257,
      "grad_norm": 0.578125,
      "learning_rate": 2.466648227169731e-05,
      "loss": 0.38764967918396,
      "num_input_tokens_seen": 8841659136,
      "step": 31090,
      "train_runtime": 302352.4358,
      "train_tokens_per_second": 29242.89
    },
    {
      "epoch": 1.1003788088198547,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4663481227525304e-05,
      "loss": 0.3933694362640381,
      "num_input_tokens_seen": 8844499712,
      "step": 31100,
      "train_runtime": 302450.1423,
      "train_tokens_per_second": 29242.835
    },
    {
      "epoch": 1.1007326290835835,
      "grad_norm": 0.56640625,
      "learning_rate": 2.4660481278451704e-05,
      "loss": 0.3893763542175293,
      "num_input_tokens_seen": 8847438720,
      "step": 31110,
      "train_runtime": 302553.9717,
      "train_tokens_per_second": 29242.514
    },
    {
      "epoch": 1.1010864493473123,
      "grad_norm": 0.6015625,
      "learning_rate": 2.465748242381067e-05,
      "loss": 0.39050793647766113,
      "num_input_tokens_seen": 8850241344,
      "step": 31120,
      "train_runtime": 302648.4457,
      "train_tokens_per_second": 29242.646
    },
    {
      "epoch": 1.101440269611041,
      "grad_norm": 0.58203125,
      "learning_rate": 2.4654484662936915e-05,
      "loss": 0.3951855659484863,
      "num_input_tokens_seen": 8853025600,
      "step": 31130,
      "train_runtime": 302743.1624,
      "train_tokens_per_second": 29242.694
    },
    {
      "epoch": 1.1017940898747698,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4651487995165712e-05,
      "loss": 0.38728985786437986,
      "num_input_tokens_seen": 8855847872,
      "step": 31140,
      "train_runtime": 302837.9759,
      "train_tokens_per_second": 29242.858
    },
    {
      "epoch": 1.1021479101384986,
      "grad_norm": 0.578125,
      "learning_rate": 2.4648492419832915e-05,
      "loss": 0.39473934173583985,
      "num_input_tokens_seen": 8858661760,
      "step": 31150,
      "train_runtime": 302933.4312,
      "train_tokens_per_second": 29242.932
    },
    {
      "epoch": 1.1025017304022273,
      "grad_norm": 0.59375,
      "learning_rate": 2.4645497936274937e-05,
      "loss": 0.3893791675567627,
      "num_input_tokens_seen": 8861430016,
      "step": 31160,
      "train_runtime": 303025.6612,
      "train_tokens_per_second": 29243.167
    },
    {
      "epoch": 1.102855550665956,
      "grad_norm": 0.609375,
      "learning_rate": 2.4642504543828753e-05,
      "loss": 0.39426631927490235,
      "num_input_tokens_seen": 8864291968,
      "step": 31170,
      "train_runtime": 303122.6259,
      "train_tokens_per_second": 29243.254
    },
    {
      "epoch": 1.1032093709296849,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4639512241831904e-05,
      "loss": 0.3910731315612793,
      "num_input_tokens_seen": 8867147392,
      "step": 31180,
      "train_runtime": 303222.8837,
      "train_tokens_per_second": 29243.002
    },
    {
      "epoch": 1.1035631911934136,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4636521029622485e-05,
      "loss": 0.3913859844207764,
      "num_input_tokens_seen": 8869901376,
      "step": 31190,
      "train_runtime": 303313.6855,
      "train_tokens_per_second": 29243.327
    },
    {
      "epoch": 1.1039170114571424,
      "grad_norm": 0.59375,
      "learning_rate": 2.463353090653917e-05,
      "loss": 0.39261603355407715,
      "num_input_tokens_seen": 8872737920,
      "step": 31200,
      "train_runtime": 303410.8262,
      "train_tokens_per_second": 29243.314
    },
    {
      "epoch": 1.1042708317208711,
      "grad_norm": 0.62109375,
      "learning_rate": 2.4630541871921187e-05,
      "loss": 0.3920186758041382,
      "num_input_tokens_seen": 8875596736,
      "step": 31210,
      "train_runtime": 303509.819,
      "train_tokens_per_second": 29243.195
    },
    {
      "epoch": 1.1046246519846,
      "grad_norm": 0.58203125,
      "learning_rate": 2.4627553925108313e-05,
      "loss": 0.3933091163635254,
      "num_input_tokens_seen": 8878420480,
      "step": 31220,
      "train_runtime": 303608.1885,
      "train_tokens_per_second": 29243.021
    },
    {
      "epoch": 1.1049784722483287,
      "grad_norm": 0.59375,
      "learning_rate": 2.4624567065440906e-05,
      "loss": 0.38926043510437014,
      "num_input_tokens_seen": 8881234816,
      "step": 31230,
      "train_runtime": 303703.4122,
      "train_tokens_per_second": 29243.118
    },
    {
      "epoch": 1.1053322925120574,
      "grad_norm": 0.62890625,
      "learning_rate": 2.4621581292259868e-05,
      "loss": 0.393221378326416,
      "num_input_tokens_seen": 8883989888,
      "step": 31240,
      "train_runtime": 303797.0095,
      "train_tokens_per_second": 29243.178
    },
    {
      "epoch": 1.1056861127757862,
      "grad_norm": 0.58203125,
      "learning_rate": 2.4618596604906673e-05,
      "loss": 0.389432954788208,
      "num_input_tokens_seen": 8886794368,
      "step": 31250,
      "train_runtime": 303891.7611,
      "train_tokens_per_second": 29243.288
    },
    {
      "epoch": 1.106039933039515,
      "grad_norm": 0.63671875,
      "learning_rate": 2.4615613002723347e-05,
      "loss": 0.3893982648849487,
      "num_input_tokens_seen": 8889625600,
      "step": 31260,
      "train_runtime": 303984.9766,
      "train_tokens_per_second": 29243.635
    },
    {
      "epoch": 1.106393753303244,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4612630485052466e-05,
      "loss": 0.3953744888305664,
      "num_input_tokens_seen": 8892476480,
      "step": 31270,
      "train_runtime": 304083.0955,
      "train_tokens_per_second": 29243.574
    },
    {
      "epoch": 1.1067475735669727,
      "grad_norm": 0.59375,
      "learning_rate": 2.4609649051237175e-05,
      "loss": 0.3910569667816162,
      "num_input_tokens_seen": 8895300608,
      "step": 31280,
      "train_runtime": 304176.9587,
      "train_tokens_per_second": 29243.834
    },
    {
      "epoch": 1.1071013938307015,
      "grad_norm": 0.62109375,
      "learning_rate": 2.460666870062118e-05,
      "loss": 0.39440701007843015,
      "num_input_tokens_seen": 8898108736,
      "step": 31290,
      "train_runtime": 304272.1462,
      "train_tokens_per_second": 29243.915
    },
    {
      "epoch": 1.1074552140944303,
      "grad_norm": 0.59375,
      "learning_rate": 2.4603689432548728e-05,
      "loss": 0.38909599781036375,
      "num_input_tokens_seen": 8900946240,
      "step": 31300,
      "train_runtime": 304369.5293,
      "train_tokens_per_second": 29243.881
    },
    {
      "epoch": 1.107809034358159,
      "grad_norm": 0.640625,
      "learning_rate": 2.460071124636463e-05,
      "loss": 0.39116530418395995,
      "num_input_tokens_seen": 8903786752,
      "step": 31310,
      "train_runtime": 304467.0999,
      "train_tokens_per_second": 29243.839
    },
    {
      "epoch": 1.1081628546218878,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4597734141414254e-05,
      "loss": 0.3890060901641846,
      "num_input_tokens_seen": 8906625472,
      "step": 31320,
      "train_runtime": 304565.2535,
      "train_tokens_per_second": 29243.735
    },
    {
      "epoch": 1.1085166748856166,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4594758117043516e-05,
      "loss": 0.38748846054077146,
      "num_input_tokens_seen": 8909447808,
      "step": 31330,
      "train_runtime": 304660.9074,
      "train_tokens_per_second": 29243.817
    },
    {
      "epoch": 1.1088704951493453,
      "grad_norm": 0.609375,
      "learning_rate": 2.4591783172598898e-05,
      "loss": 0.39062840938568116,
      "num_input_tokens_seen": 8912378432,
      "step": 31340,
      "train_runtime": 304765.4424,
      "train_tokens_per_second": 29243.402
    },
    {
      "epoch": 1.109224315413074,
      "grad_norm": 0.57421875,
      "learning_rate": 2.458880930742742e-05,
      "loss": 0.39377598762512206,
      "num_input_tokens_seen": 8915198144,
      "step": 31350,
      "train_runtime": 304860.6907,
      "train_tokens_per_second": 29243.515
    },
    {
      "epoch": 1.1095781356768029,
      "grad_norm": 0.58984375,
      "learning_rate": 2.458583652087666e-05,
      "loss": 0.39347522258758544,
      "num_input_tokens_seen": 8918044992,
      "step": 31360,
      "train_runtime": 304956.3497,
      "train_tokens_per_second": 29243.677
    },
    {
      "epoch": 1.1099319559405316,
      "grad_norm": 0.609375,
      "learning_rate": 2.4582864812294757e-05,
      "loss": 0.3891420364379883,
      "num_input_tokens_seen": 8920859072,
      "step": 31370,
      "train_runtime": 305052.0724,
      "train_tokens_per_second": 29243.726
    },
    {
      "epoch": 1.1102857762042604,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4579894181030387e-05,
      "loss": 0.39308948516845704,
      "num_input_tokens_seen": 8923675840,
      "step": 31380,
      "train_runtime": 305146.9005,
      "train_tokens_per_second": 29243.869
    },
    {
      "epoch": 1.1106395964679892,
      "grad_norm": 0.59765625,
      "learning_rate": 2.457692462643279e-05,
      "loss": 0.38944506645202637,
      "num_input_tokens_seen": 8926519936,
      "step": 31390,
      "train_runtime": 305243.7572,
      "train_tokens_per_second": 29243.907
    },
    {
      "epoch": 1.110993416731718,
      "grad_norm": 0.609375,
      "learning_rate": 2.4573956147851734e-05,
      "loss": 0.3937228679656982,
      "num_input_tokens_seen": 8929360768,
      "step": 31400,
      "train_runtime": 305340.8982,
      "train_tokens_per_second": 29243.907
    },
    {
      "epoch": 1.1113472369954467,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4570988744637573e-05,
      "loss": 0.38926851749420166,
      "num_input_tokens_seen": 8932176064,
      "step": 31410,
      "train_runtime": 305435.1044,
      "train_tokens_per_second": 29244.104
    },
    {
      "epoch": 1.1117010572591755,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4568022416141178e-05,
      "loss": 0.3905486106872559,
      "num_input_tokens_seen": 8935036928,
      "step": 31420,
      "train_runtime": 305530.6639,
      "train_tokens_per_second": 29244.321
    },
    {
      "epoch": 1.1120548775229042,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4565057161713987e-05,
      "loss": 0.39151678085327146,
      "num_input_tokens_seen": 8937931264,
      "step": 31430,
      "train_runtime": 305630.9228,
      "train_tokens_per_second": 29244.198
    },
    {
      "epoch": 1.1124086977866332,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4562092980707975e-05,
      "loss": 0.39062275886535647,
      "num_input_tokens_seen": 8940802240,
      "step": 31440,
      "train_runtime": 305727.964,
      "train_tokens_per_second": 29244.306
    },
    {
      "epoch": 1.112762518050362,
      "grad_norm": 0.6171875,
      "learning_rate": 2.455912987247566e-05,
      "loss": 0.38751611709594724,
      "num_input_tokens_seen": 8943665088,
      "step": 31450,
      "train_runtime": 305825.8198,
      "train_tokens_per_second": 29244.31
    },
    {
      "epoch": 1.1131163383140907,
      "grad_norm": 0.62890625,
      "learning_rate": 2.455616783637013e-05,
      "loss": 0.3899212837219238,
      "num_input_tokens_seen": 8946493184,
      "step": 31460,
      "train_runtime": 305924.4112,
      "train_tokens_per_second": 29244.13
    },
    {
      "epoch": 1.1134701585778195,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4553206871745e-05,
      "loss": 0.39320101737976076,
      "num_input_tokens_seen": 8949372224,
      "step": 31470,
      "train_runtime": 306023.8287,
      "train_tokens_per_second": 29244.037
    },
    {
      "epoch": 1.1138239788415483,
      "grad_norm": 0.62109375,
      "learning_rate": 2.4550246977954422e-05,
      "loss": 0.39439618587493896,
      "num_input_tokens_seen": 8952211392,
      "step": 31480,
      "train_runtime": 306120.495,
      "train_tokens_per_second": 29244.077
    },
    {
      "epoch": 1.114177799105277,
      "grad_norm": 0.6171875,
      "learning_rate": 2.454728815435312e-05,
      "loss": 0.39246206283569335,
      "num_input_tokens_seen": 8955083968,
      "step": 31490,
      "train_runtime": 306220.7326,
      "train_tokens_per_second": 29243.885
    },
    {
      "epoch": 1.1145316193690058,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4544330400296333e-05,
      "loss": 0.3932910919189453,
      "num_input_tokens_seen": 8957897152,
      "step": 31500,
      "train_runtime": 306315.9622,
      "train_tokens_per_second": 29243.978
    },
    {
      "epoch": 1.1148854396327346,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4541373715139876e-05,
      "loss": 0.394961953163147,
      "num_input_tokens_seen": 8960745408,
      "step": 31510,
      "train_runtime": 306416.1858,
      "train_tokens_per_second": 29243.708
    },
    {
      "epoch": 1.1152392598964633,
      "grad_norm": 0.58203125,
      "learning_rate": 2.453841809824007e-05,
      "loss": 0.3874105930328369,
      "num_input_tokens_seen": 8963585856,
      "step": 31520,
      "train_runtime": 306514.2837,
      "train_tokens_per_second": 29243.615
    },
    {
      "epoch": 1.115593080160192,
      "grad_norm": 0.62109375,
      "learning_rate": 2.4535463548953808e-05,
      "loss": 0.38852055072784425,
      "num_input_tokens_seen": 8966380032,
      "step": 31530,
      "train_runtime": 306607.3951,
      "train_tokens_per_second": 29243.848
    },
    {
      "epoch": 1.1159469004239209,
      "grad_norm": 0.56640625,
      "learning_rate": 2.4532510066638515e-05,
      "loss": 0.3929260730743408,
      "num_input_tokens_seen": 8969227840,
      "step": 31540,
      "train_runtime": 306707.6512,
      "train_tokens_per_second": 29243.574
    },
    {
      "epoch": 1.1163007206876496,
      "grad_norm": 0.59375,
      "learning_rate": 2.452955765065216e-05,
      "loss": 0.39009239673614504,
      "num_input_tokens_seen": 8972045184,
      "step": 31550,
      "train_runtime": 306803.2046,
      "train_tokens_per_second": 29243.649
    },
    {
      "epoch": 1.1166545409513784,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4526606300353234e-05,
      "loss": 0.3912105321884155,
      "num_input_tokens_seen": 8974774208,
      "step": 31560,
      "train_runtime": 306892.0353,
      "train_tokens_per_second": 29244.077
    },
    {
      "epoch": 1.1170083612151072,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4523656015100795e-05,
      "loss": 0.3898076772689819,
      "num_input_tokens_seen": 8977614272,
      "step": 31570,
      "train_runtime": 306990.7283,
      "train_tokens_per_second": 29243.926
    },
    {
      "epoch": 1.117362181478836,
      "grad_norm": 0.57421875,
      "learning_rate": 2.4520706794254433e-05,
      "loss": 0.3923133373260498,
      "num_input_tokens_seen": 8980489344,
      "step": 31580,
      "train_runtime": 307090.6597,
      "train_tokens_per_second": 29243.772
    },
    {
      "epoch": 1.117716001742565,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4517758637174266e-05,
      "loss": 0.3897665500640869,
      "num_input_tokens_seen": 8983301248,
      "step": 31590,
      "train_runtime": 307183.1044,
      "train_tokens_per_second": 29244.125
    },
    {
      "epoch": 1.1180698220062937,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4514811543220957e-05,
      "loss": 0.3889149188995361,
      "num_input_tokens_seen": 8986166976,
      "step": 31600,
      "train_runtime": 307280.6148,
      "train_tokens_per_second": 29244.171
    },
    {
      "epoch": 1.1184236422700224,
      "grad_norm": 0.62109375,
      "learning_rate": 2.4511865511755717e-05,
      "loss": 0.3887796878814697,
      "num_input_tokens_seen": 8989008256,
      "step": 31610,
      "train_runtime": 307376.2393,
      "train_tokens_per_second": 29244.317
    },
    {
      "epoch": 1.1187774625337512,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4508920542140266e-05,
      "loss": 0.3897322416305542,
      "num_input_tokens_seen": 8991880064,
      "step": 31620,
      "train_runtime": 307473.3479,
      "train_tokens_per_second": 29244.421
    },
    {
      "epoch": 1.11913128279748,
      "grad_norm": 0.62109375,
      "learning_rate": 2.4505976633736905e-05,
      "loss": 0.39379243850708007,
      "num_input_tokens_seen": 8994712896,
      "step": 31630,
      "train_runtime": 307569.3225,
      "train_tokens_per_second": 29244.506
    },
    {
      "epoch": 1.1194851030612087,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4503033785908423e-05,
      "loss": 0.3944609880447388,
      "num_input_tokens_seen": 8997562816,
      "step": 31640,
      "train_runtime": 307667.1114,
      "train_tokens_per_second": 29244.474
    },
    {
      "epoch": 1.1198389233249375,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4500091998018178e-05,
      "loss": 0.3878974437713623,
      "num_input_tokens_seen": 9000375744,
      "step": 31650,
      "train_runtime": 307761.4758,
      "train_tokens_per_second": 29244.647
    },
    {
      "epoch": 1.1201927435886663,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4497151269430056e-05,
      "loss": 0.390798020362854,
      "num_input_tokens_seen": 9003194944,
      "step": 31660,
      "train_runtime": 307855.3584,
      "train_tokens_per_second": 29244.886
    },
    {
      "epoch": 1.120546563852395,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4494211599508456e-05,
      "loss": 0.38975703716278076,
      "num_input_tokens_seen": 9006043008,
      "step": 31670,
      "train_runtime": 307952.3983,
      "train_tokens_per_second": 29244.919
    },
    {
      "epoch": 1.1209003841161238,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4491272987618343e-05,
      "loss": 0.38620786666870116,
      "num_input_tokens_seen": 9008929856,
      "step": 31680,
      "train_runtime": 308054.1393,
      "train_tokens_per_second": 29244.632
    },
    {
      "epoch": 1.1212542043798526,
      "grad_norm": 0.59375,
      "learning_rate": 2.44883354331252e-05,
      "loss": 0.3955803871154785,
      "num_input_tokens_seen": 9011756992,
      "step": 31690,
      "train_runtime": 308149.6609,
      "train_tokens_per_second": 29244.741
    },
    {
      "epoch": 1.1216080246435813,
      "grad_norm": 0.65625,
      "learning_rate": 2.4485398935395037e-05,
      "loss": 0.39271974563598633,
      "num_input_tokens_seen": 9014561344,
      "step": 31700,
      "train_runtime": 308246.9085,
      "train_tokens_per_second": 29244.612
    },
    {
      "epoch": 1.12196184490731,
      "grad_norm": 0.5703125,
      "learning_rate": 2.44824634937944e-05,
      "loss": 0.39099879264831544,
      "num_input_tokens_seen": 9017416320,
      "step": 31710,
      "train_runtime": 308343.2466,
      "train_tokens_per_second": 29244.734
    },
    {
      "epoch": 1.1223156651710389,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4479529107690374e-05,
      "loss": 0.3882007122039795,
      "num_input_tokens_seen": 9020226688,
      "step": 31720,
      "train_runtime": 308436.7245,
      "train_tokens_per_second": 29244.983
    },
    {
      "epoch": 1.1226694854347676,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4476595776450564e-05,
      "loss": 0.389832067489624,
      "num_input_tokens_seen": 9023073728,
      "step": 31730,
      "train_runtime": 308535.1464,
      "train_tokens_per_second": 29244.881
    },
    {
      "epoch": 1.1230233056984964,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4473663499443112e-05,
      "loss": 0.38889412879943847,
      "num_input_tokens_seen": 9025939264,
      "step": 31740,
      "train_runtime": 308633.2035,
      "train_tokens_per_second": 29244.874
    },
    {
      "epoch": 1.1233771259622252,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4470732276036698e-05,
      "loss": 0.39253294467926025,
      "num_input_tokens_seen": 9028742144,
      "step": 31750,
      "train_runtime": 308727.6061,
      "train_tokens_per_second": 29245.011
    },
    {
      "epoch": 1.1237309462259542,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4467802105600504e-05,
      "loss": 0.39611001014709474,
      "num_input_tokens_seen": 9031582592,
      "step": 31760,
      "train_runtime": 308825.3862,
      "train_tokens_per_second": 29244.949
    },
    {
      "epoch": 1.124084766489683,
      "grad_norm": 0.578125,
      "learning_rate": 2.4464872987504275e-05,
      "loss": 0.39229393005371094,
      "num_input_tokens_seen": 9034511360,
      "step": 31770,
      "train_runtime": 308928.8806,
      "train_tokens_per_second": 29244.632
    },
    {
      "epoch": 1.1244385867534117,
      "grad_norm": 0.60546875,
      "learning_rate": 2.446194492111825e-05,
      "loss": 0.3919085502624512,
      "num_input_tokens_seen": 9037327936,
      "step": 31780,
      "train_runtime": 309025.1635,
      "train_tokens_per_second": 29244.634
    },
    {
      "epoch": 1.1247924070171404,
      "grad_norm": 0.5859375,
      "learning_rate": 2.445901790581323e-05,
      "loss": 0.3887171268463135,
      "num_input_tokens_seen": 9040213888,
      "step": 31790,
      "train_runtime": 309122.2702,
      "train_tokens_per_second": 29244.784
    },
    {
      "epoch": 1.1251462272808692,
      "grad_norm": 0.5859375,
      "learning_rate": 2.445609194096051e-05,
      "loss": 0.39116415977478025,
      "num_input_tokens_seen": 9043037632,
      "step": 31800,
      "train_runtime": 309216.7238,
      "train_tokens_per_second": 29244.982
    },
    {
      "epoch": 1.125500047544598,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4453167025931932e-05,
      "loss": 0.3889565706253052,
      "num_input_tokens_seen": 9045867008,
      "step": 31810,
      "train_runtime": 309311.2437,
      "train_tokens_per_second": 29245.193
    },
    {
      "epoch": 1.1258538678083267,
      "grad_norm": 0.578125,
      "learning_rate": 2.445024316009987e-05,
      "loss": 0.39447908401489257,
      "num_input_tokens_seen": 9048692928,
      "step": 31820,
      "train_runtime": 309408.0861,
      "train_tokens_per_second": 29245.173
    },
    {
      "epoch": 1.1262076880720555,
      "grad_norm": 0.58203125,
      "learning_rate": 2.4447320342837192e-05,
      "loss": 0.39153826236724854,
      "num_input_tokens_seen": 9051539264,
      "step": 31830,
      "train_runtime": 309504.2724,
      "train_tokens_per_second": 29245.281
    },
    {
      "epoch": 1.1265615083357843,
      "grad_norm": 0.6171875,
      "learning_rate": 2.444439857351732e-05,
      "loss": 0.39404993057250975,
      "num_input_tokens_seen": 9054397632,
      "step": 31840,
      "train_runtime": 309603.0445,
      "train_tokens_per_second": 29245.183
    },
    {
      "epoch": 1.126915328599513,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4441477851514195e-05,
      "loss": 0.3889914512634277,
      "num_input_tokens_seen": 9057182848,
      "step": 31850,
      "train_runtime": 309698.2147,
      "train_tokens_per_second": 29245.189
    },
    {
      "epoch": 1.1272691488632418,
      "grad_norm": 0.609375,
      "learning_rate": 2.443855817620227e-05,
      "loss": 0.389780592918396,
      "num_input_tokens_seen": 9060000320,
      "step": 31860,
      "train_runtime": 309796.5989,
      "train_tokens_per_second": 29244.996
    },
    {
      "epoch": 1.1276229691269706,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4435639546956534e-05,
      "loss": 0.39208064079284666,
      "num_input_tokens_seen": 9062860096,
      "step": 31870,
      "train_runtime": 309895.1214,
      "train_tokens_per_second": 29244.927
    },
    {
      "epoch": 1.1279767893906993,
      "grad_norm": 0.62890625,
      "learning_rate": 2.4432721963152482e-05,
      "loss": 0.3877956628799438,
      "num_input_tokens_seen": 9065692160,
      "step": 31880,
      "train_runtime": 309989.995,
      "train_tokens_per_second": 29245.112
    },
    {
      "epoch": 1.128330609654428,
      "grad_norm": 0.6328125,
      "learning_rate": 2.442980542416615e-05,
      "loss": 0.3879082679748535,
      "num_input_tokens_seen": 9068529280,
      "step": 31890,
      "train_runtime": 310083.0174,
      "train_tokens_per_second": 29245.488
    },
    {
      "epoch": 1.1286844299181569,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4426889929374085e-05,
      "loss": 0.3877959966659546,
      "num_input_tokens_seen": 9071363904,
      "step": 31900,
      "train_runtime": 310176.5124,
      "train_tokens_per_second": 29245.812
    },
    {
      "epoch": 1.1290382501818856,
      "grad_norm": 0.5859375,
      "learning_rate": 2.442397547815336e-05,
      "loss": 0.388504695892334,
      "num_input_tokens_seen": 9074169152,
      "step": 31910,
      "train_runtime": 310272.4356,
      "train_tokens_per_second": 29245.811
    },
    {
      "epoch": 1.1293920704456144,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4421062069881545e-05,
      "loss": 0.39296574592590333,
      "num_input_tokens_seen": 9077015808,
      "step": 31920,
      "train_runtime": 310371.6294,
      "train_tokens_per_second": 29245.636
    },
    {
      "epoch": 1.1297458907093434,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4418149703936773e-05,
      "loss": 0.394504189491272,
      "num_input_tokens_seen": 9079837120,
      "step": 31930,
      "train_runtime": 310469.283,
      "train_tokens_per_second": 29245.525
    },
    {
      "epoch": 1.1300997109730722,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4415238379697657e-05,
      "loss": 0.39421706199645995,
      "num_input_tokens_seen": 9082718400,
      "step": 31940,
      "train_runtime": 310567.7098,
      "train_tokens_per_second": 29245.534
    },
    {
      "epoch": 1.130453531236801,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4412328096543355e-05,
      "loss": 0.3945643901824951,
      "num_input_tokens_seen": 9085524160,
      "step": 31950,
      "train_runtime": 310663.5349,
      "train_tokens_per_second": 29245.544
    },
    {
      "epoch": 1.1308073515005297,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4409418853853517e-05,
      "loss": 0.39127635955810547,
      "num_input_tokens_seen": 9088371776,
      "step": 31960,
      "train_runtime": 310761.9664,
      "train_tokens_per_second": 29245.444
    },
    {
      "epoch": 1.1311611717642585,
      "grad_norm": 0.609375,
      "learning_rate": 2.440651065100833e-05,
      "loss": 0.3958575963973999,
      "num_input_tokens_seen": 9091224704,
      "step": 31970,
      "train_runtime": 310860.1038,
      "train_tokens_per_second": 29245.389
    },
    {
      "epoch": 1.1315149920279872,
      "grad_norm": 0.59765625,
      "learning_rate": 2.44036034873885e-05,
      "loss": 0.3935561180114746,
      "num_input_tokens_seen": 9094091584,
      "step": 31980,
      "train_runtime": 310958.2393,
      "train_tokens_per_second": 29245.379
    },
    {
      "epoch": 1.131868812291716,
      "grad_norm": 0.609375,
      "learning_rate": 2.440069736237523e-05,
      "loss": 0.38922598361968996,
      "num_input_tokens_seen": 9096930752,
      "step": 31990,
      "train_runtime": 311051.8854,
      "train_tokens_per_second": 29245.702
    },
    {
      "epoch": 1.1322226325554448,
      "grad_norm": 0.609375,
      "learning_rate": 2.439779227535026e-05,
      "loss": 0.3947909832000732,
      "num_input_tokens_seen": 9099779968,
      "step": 32000,
      "train_runtime": 311148.7996,
      "train_tokens_per_second": 29245.75
    },
    {
      "epoch": 1.1325764528191735,
      "grad_norm": 0.6328125,
      "learning_rate": 2.439488822569583e-05,
      "loss": 0.39146995544433594,
      "num_input_tokens_seen": 9102639104,
      "step": 32010,
      "train_runtime": 311244.7328,
      "train_tokens_per_second": 29245.922
    },
    {
      "epoch": 1.1329302730829023,
      "grad_norm": 0.63671875,
      "learning_rate": 2.4391985212794695e-05,
      "loss": 0.38922438621520994,
      "num_input_tokens_seen": 9105475008,
      "step": 32020,
      "train_runtime": 311343.203,
      "train_tokens_per_second": 29245.781
    },
    {
      "epoch": 1.133284093346631,
      "grad_norm": 0.640625,
      "learning_rate": 2.4389083236030146e-05,
      "loss": 0.39569735527038574,
      "num_input_tokens_seen": 9108276032,
      "step": 32030,
      "train_runtime": 311438.609,
      "train_tokens_per_second": 29245.815
    },
    {
      "epoch": 1.1336379136103598,
      "grad_norm": 0.578125,
      "learning_rate": 2.4386182294785952e-05,
      "loss": 0.3915308713912964,
      "num_input_tokens_seen": 9111154176,
      "step": 32040,
      "train_runtime": 311538.3799,
      "train_tokens_per_second": 29245.688
    },
    {
      "epoch": 1.1339917338740886,
      "grad_norm": 0.6015625,
      "learning_rate": 2.438328238844642e-05,
      "loss": 0.39095933437347413,
      "num_input_tokens_seen": 9113959488,
      "step": 32050,
      "train_runtime": 311631.6651,
      "train_tokens_per_second": 29245.935
    },
    {
      "epoch": 1.1343455541378173,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4380383516396372e-05,
      "loss": 0.39179201126098634,
      "num_input_tokens_seen": 9116817472,
      "step": 32060,
      "train_runtime": 311727.0025,
      "train_tokens_per_second": 29246.159
    },
    {
      "epoch": 1.1346993744015461,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4377485678021113e-05,
      "loss": 0.3900890350341797,
      "num_input_tokens_seen": 9119695808,
      "step": 32070,
      "train_runtime": 311824.6078,
      "train_tokens_per_second": 29246.235
    },
    {
      "epoch": 1.1350531946652749,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4374588872706498e-05,
      "loss": 0.38930773735046387,
      "num_input_tokens_seen": 9122551872,
      "step": 32080,
      "train_runtime": 311923.1444,
      "train_tokens_per_second": 29246.153
    },
    {
      "epoch": 1.1354070149290036,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4371693099838865e-05,
      "loss": 0.39177489280700684,
      "num_input_tokens_seen": 9125410176,
      "step": 32090,
      "train_runtime": 312024.172,
      "train_tokens_per_second": 29245.844
    },
    {
      "epoch": 1.1357608351927326,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4368798358805075e-05,
      "loss": 0.3898944616317749,
      "num_input_tokens_seen": 9128252480,
      "step": 32100,
      "train_runtime": 312121.1576,
      "train_tokens_per_second": 29245.863
    },
    {
      "epoch": 1.1361146554564614,
      "grad_norm": 0.59375,
      "learning_rate": 2.4365904648992495e-05,
      "loss": 0.39085969924926756,
      "num_input_tokens_seen": 9131125248,
      "step": 32110,
      "train_runtime": 312216.5121,
      "train_tokens_per_second": 29246.132
    },
    {
      "epoch": 1.1364684757201902,
      "grad_norm": 0.61328125,
      "learning_rate": 2.436301196978899e-05,
      "loss": 0.3918031692504883,
      "num_input_tokens_seen": 9133973888,
      "step": 32120,
      "train_runtime": 312314.9469,
      "train_tokens_per_second": 29246.035
    },
    {
      "epoch": 1.136822295983919,
      "grad_norm": 0.59765625,
      "learning_rate": 2.436012032058296e-05,
      "loss": 0.38938679695129397,
      "num_input_tokens_seen": 9136808640,
      "step": 32130,
      "train_runtime": 312410.8143,
      "train_tokens_per_second": 29246.134
    },
    {
      "epoch": 1.1371761162476477,
      "grad_norm": 0.58203125,
      "learning_rate": 2.4357229700763283e-05,
      "loss": 0.3918055534362793,
      "num_input_tokens_seen": 9139679488,
      "step": 32140,
      "train_runtime": 312506.9487,
      "train_tokens_per_second": 29246.324
    },
    {
      "epoch": 1.1375299365113765,
      "grad_norm": 0.59765625,
      "learning_rate": 2.435434010971937e-05,
      "loss": 0.3938034772872925,
      "num_input_tokens_seen": 9142476992,
      "step": 32150,
      "train_runtime": 312600.8925,
      "train_tokens_per_second": 29246.484
    },
    {
      "epoch": 1.1378837567751052,
      "grad_norm": 0.609375,
      "learning_rate": 2.4351451546841118e-05,
      "loss": 0.3926145315170288,
      "num_input_tokens_seen": 9145374272,
      "step": 32160,
      "train_runtime": 312701.5992,
      "train_tokens_per_second": 29246.33
    },
    {
      "epoch": 1.138237577038834,
      "grad_norm": 0.59765625,
      "learning_rate": 2.434856401151895e-05,
      "loss": 0.39398512840270994,
      "num_input_tokens_seen": 9148207488,
      "step": 32170,
      "train_runtime": 312799.9376,
      "train_tokens_per_second": 29246.193
    },
    {
      "epoch": 1.1385913973025628,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4345677503143778e-05,
      "loss": 0.39225115776062014,
      "num_input_tokens_seen": 9151062912,
      "step": 32180,
      "train_runtime": 312898.1305,
      "train_tokens_per_second": 29246.141
    },
    {
      "epoch": 1.1389452175662915,
      "grad_norm": 0.578125,
      "learning_rate": 2.4342792021107033e-05,
      "loss": 0.39346837997436523,
      "num_input_tokens_seen": 9153943872,
      "step": 32190,
      "train_runtime": 312995.9529,
      "train_tokens_per_second": 29246.205
    },
    {
      "epoch": 1.1392990378300203,
      "grad_norm": 0.625,
      "learning_rate": 2.433990756480064e-05,
      "loss": 0.3894240379333496,
      "num_input_tokens_seen": 9156787648,
      "step": 32200,
      "train_runtime": 313092.8912,
      "train_tokens_per_second": 29246.233
    },
    {
      "epoch": 1.139652858093749,
      "grad_norm": 0.62109375,
      "learning_rate": 2.433702413361703e-05,
      "loss": 0.394959020614624,
      "num_input_tokens_seen": 9159608000,
      "step": 32210,
      "train_runtime": 313188.7594,
      "train_tokens_per_second": 29246.286
    },
    {
      "epoch": 1.1400066783574778,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4334141726949147e-05,
      "loss": 0.38838396072387693,
      "num_input_tokens_seen": 9162429568,
      "step": 32220,
      "train_runtime": 313284.2726,
      "train_tokens_per_second": 29246.376
    },
    {
      "epoch": 1.1403604986212066,
      "grad_norm": 0.59375,
      "learning_rate": 2.433126034419042e-05,
      "loss": 0.3915014982223511,
      "num_input_tokens_seen": 9165374848,
      "step": 32230,
      "train_runtime": 313386.5248,
      "train_tokens_per_second": 29246.231
    },
    {
      "epoch": 1.1407143188849354,
      "grad_norm": 0.62890625,
      "learning_rate": 2.4328379984734804e-05,
      "loss": 0.3841994762420654,
      "num_input_tokens_seen": 9168251840,
      "step": 32240,
      "train_runtime": 313485.7653,
      "train_tokens_per_second": 29246.15
    },
    {
      "epoch": 1.1410681391486643,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4325500647976738e-05,
      "loss": 0.39467051029205324,
      "num_input_tokens_seen": 9171082368,
      "step": 32250,
      "train_runtime": 313581.2743,
      "train_tokens_per_second": 29246.269
    },
    {
      "epoch": 1.1414219594123929,
      "grad_norm": 0.59375,
      "learning_rate": 2.4322622333311165e-05,
      "loss": 0.38826684951782225,
      "num_input_tokens_seen": 9173994176,
      "step": 32260,
      "train_runtime": 313679.8139,
      "train_tokens_per_second": 29246.365
    },
    {
      "epoch": 1.1417757796761219,
      "grad_norm": 0.578125,
      "learning_rate": 2.4319745040133543e-05,
      "loss": 0.39078197479248045,
      "num_input_tokens_seen": 9176779456,
      "step": 32270,
      "train_runtime": 313775.6903,
      "train_tokens_per_second": 29246.305
    },
    {
      "epoch": 1.1421295999398506,
      "grad_norm": 0.60546875,
      "learning_rate": 2.431686876783981e-05,
      "loss": 0.39177799224853516,
      "num_input_tokens_seen": 9179582464,
      "step": 32280,
      "train_runtime": 313872.58,
      "train_tokens_per_second": 29246.207
    },
    {
      "epoch": 1.1424834202035794,
      "grad_norm": 0.58984375,
      "learning_rate": 2.431399351582642e-05,
      "loss": 0.3909552574157715,
      "num_input_tokens_seen": 9182380544,
      "step": 32290,
      "train_runtime": 313968.3097,
      "train_tokens_per_second": 29246.202
    },
    {
      "epoch": 1.1428372404673082,
      "grad_norm": 0.57421875,
      "learning_rate": 2.431111928349032e-05,
      "loss": 0.38819835186004636,
      "num_input_tokens_seen": 9185219008,
      "step": 32300,
      "train_runtime": 314066.3769,
      "train_tokens_per_second": 29246.107
    },
    {
      "epoch": 1.143191060731037,
      "grad_norm": 0.62109375,
      "learning_rate": 2.4308246070228956e-05,
      "loss": 0.3869027614593506,
      "num_input_tokens_seen": 9188072448,
      "step": 32310,
      "train_runtime": 314164.1727,
      "train_tokens_per_second": 29246.086
    },
    {
      "epoch": 1.1435448809947657,
      "grad_norm": 0.59765625,
      "learning_rate": 2.430537387544027e-05,
      "loss": 0.3925086259841919,
      "num_input_tokens_seen": 9190898560,
      "step": 32320,
      "train_runtime": 314261.412,
      "train_tokens_per_second": 29246.03
    },
    {
      "epoch": 1.1438987012584945,
      "grad_norm": 0.62109375,
      "learning_rate": 2.4302502698522707e-05,
      "loss": 0.3948864221572876,
      "num_input_tokens_seen": 9193738752,
      "step": 32330,
      "train_runtime": 314359.4429,
      "train_tokens_per_second": 29245.944
    },
    {
      "epoch": 1.1442525215222232,
      "grad_norm": 0.61328125,
      "learning_rate": 2.429963253887521e-05,
      "loss": 0.39093406200408937,
      "num_input_tokens_seen": 9196551616,
      "step": 32340,
      "train_runtime": 314455.2591,
      "train_tokens_per_second": 29245.978
    },
    {
      "epoch": 1.144606341785952,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4296763395897212e-05,
      "loss": 0.38929646015167235,
      "num_input_tokens_seen": 9199440448,
      "step": 32350,
      "train_runtime": 314554.2006,
      "train_tokens_per_second": 29245.963
    },
    {
      "epoch": 1.1449601620496808,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4293895268988646e-05,
      "loss": 0.3914044380187988,
      "num_input_tokens_seen": 9202239616,
      "step": 32360,
      "train_runtime": 314648.7416,
      "train_tokens_per_second": 29246.072
    },
    {
      "epoch": 1.1453139823134095,
      "grad_norm": 0.57421875,
      "learning_rate": 2.429102815754994e-05,
      "loss": 0.3884486198425293,
      "num_input_tokens_seen": 9205112192,
      "step": 32370,
      "train_runtime": 314749.3006,
      "train_tokens_per_second": 29245.854
    },
    {
      "epoch": 1.1456678025771383,
      "grad_norm": 0.59765625,
      "learning_rate": 2.428816206098202e-05,
      "loss": 0.3874303579330444,
      "num_input_tokens_seen": 9208002176,
      "step": 32380,
      "train_runtime": 314848.5331,
      "train_tokens_per_second": 29245.816
    },
    {
      "epoch": 1.146021622840867,
      "grad_norm": 0.58984375,
      "learning_rate": 2.42852969786863e-05,
      "loss": 0.39446771144866943,
      "num_input_tokens_seen": 9210931712,
      "step": 32390,
      "train_runtime": 314949.878,
      "train_tokens_per_second": 29245.707
    },
    {
      "epoch": 1.1463754431045958,
      "grad_norm": 0.59375,
      "learning_rate": 2.42824329100647e-05,
      "loss": 0.39263575077056884,
      "num_input_tokens_seen": 9213806912,
      "step": 32400,
      "train_runtime": 315047.3592,
      "train_tokens_per_second": 29245.784
    },
    {
      "epoch": 1.1467292633683246,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4279569854519623e-05,
      "loss": 0.3886662483215332,
      "num_input_tokens_seen": 9216752256,
      "step": 32410,
      "train_runtime": 315155.1344,
      "train_tokens_per_second": 29245.128
    },
    {
      "epoch": 1.1470830836320536,
      "grad_norm": 0.609375,
      "learning_rate": 2.427670781145397e-05,
      "loss": 0.39008941650390627,
      "num_input_tokens_seen": 9219542528,
      "step": 32420,
      "train_runtime": 315248.5001,
      "train_tokens_per_second": 29245.318
    },
    {
      "epoch": 1.1474369038957821,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4273846780271132e-05,
      "loss": 0.39446125030517576,
      "num_input_tokens_seen": 9222449664,
      "step": 32430,
      "train_runtime": 315350.6589,
      "train_tokens_per_second": 29245.062
    },
    {
      "epoch": 1.147790724159511,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4270986760374985e-05,
      "loss": 0.38886094093322754,
      "num_input_tokens_seen": 9225259136,
      "step": 32440,
      "train_runtime": 315447.2086,
      "train_tokens_per_second": 29245.018
    },
    {
      "epoch": 1.1481445444232399,
      "grad_norm": 0.625,
      "learning_rate": 2.4268127751169922e-05,
      "loss": 0.3905381202697754,
      "num_input_tokens_seen": 9228095808,
      "step": 32450,
      "train_runtime": 315544.1472,
      "train_tokens_per_second": 29245.023
    },
    {
      "epoch": 1.1484983646869686,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4265269752060792e-05,
      "loss": 0.3935687065124512,
      "num_input_tokens_seen": 9230960960,
      "step": 32460,
      "train_runtime": 315642.7926,
      "train_tokens_per_second": 29244.96
    },
    {
      "epoch": 1.1488521849506974,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4262412762452966e-05,
      "loss": 0.38972697257995603,
      "num_input_tokens_seen": 9233762240,
      "step": 32470,
      "train_runtime": 315736.3797,
      "train_tokens_per_second": 29245.164
    },
    {
      "epoch": 1.1492060052144262,
      "grad_norm": 0.62890625,
      "learning_rate": 2.4259556781752288e-05,
      "loss": 0.385425329208374,
      "num_input_tokens_seen": 9236616256,
      "step": 32480,
      "train_runtime": 315832.7851,
      "train_tokens_per_second": 29245.274
    },
    {
      "epoch": 1.149559825478155,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4256701809365084e-05,
      "loss": 0.39088973999023435,
      "num_input_tokens_seen": 9239468160,
      "step": 32490,
      "train_runtime": 315931.4306,
      "train_tokens_per_second": 29245.169
    },
    {
      "epoch": 1.1499136457418837,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4253847844698192e-05,
      "loss": 0.3911791563034058,
      "num_input_tokens_seen": 9242305792,
      "step": 32500,
      "train_runtime": 316025.0078,
      "train_tokens_per_second": 29245.489
    },
    {
      "epoch": 1.1502674660056125,
      "grad_norm": 0.578125,
      "learning_rate": 2.4250994887158923e-05,
      "loss": 0.3907617568969727,
      "num_input_tokens_seen": 9245208512,
      "step": 32510,
      "train_runtime": 316125.8865,
      "train_tokens_per_second": 29245.338
    },
    {
      "epoch": 1.1506212862693412,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4248142936155078e-05,
      "loss": 0.391890811920166,
      "num_input_tokens_seen": 9248095040,
      "step": 32520,
      "train_runtime": 316225.4804,
      "train_tokens_per_second": 29245.256
    },
    {
      "epoch": 1.15097510653307,
      "grad_norm": 0.60546875,
      "learning_rate": 2.424529199109494e-05,
      "loss": 0.3876257181167603,
      "num_input_tokens_seen": 9250908736,
      "step": 32530,
      "train_runtime": 316319.4834,
      "train_tokens_per_second": 29245.46
    },
    {
      "epoch": 1.1513289267967988,
      "grad_norm": 0.578125,
      "learning_rate": 2.42424420513873e-05,
      "loss": 0.38611156940460206,
      "num_input_tokens_seen": 9253798272,
      "step": 32540,
      "train_runtime": 316418.6449,
      "train_tokens_per_second": 29245.427
    },
    {
      "epoch": 1.1516827470605275,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4239593116441407e-05,
      "loss": 0.394504976272583,
      "num_input_tokens_seen": 9256601152,
      "step": 32550,
      "train_runtime": 316514.7657,
      "train_tokens_per_second": 29245.401
    },
    {
      "epoch": 1.1520365673242563,
      "grad_norm": 0.5859375,
      "learning_rate": 2.423674518566701e-05,
      "loss": 0.38872513771057127,
      "num_input_tokens_seen": 9259409984,
      "step": 32560,
      "train_runtime": 316609.1938,
      "train_tokens_per_second": 29245.55
    },
    {
      "epoch": 1.152390387587985,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4233898258474347e-05,
      "loss": 0.39398970603942873,
      "num_input_tokens_seen": 9262257600,
      "step": 32570,
      "train_runtime": 316708.237,
      "train_tokens_per_second": 29245.395
    },
    {
      "epoch": 1.1527442078517138,
      "grad_norm": 0.640625,
      "learning_rate": 2.4231052334274144e-05,
      "loss": 0.3939439058303833,
      "num_input_tokens_seen": 9265102848,
      "step": 32580,
      "train_runtime": 316803.5251,
      "train_tokens_per_second": 29245.58
    },
    {
      "epoch": 1.1530980281154428,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4228207412477587e-05,
      "loss": 0.3905972957611084,
      "num_input_tokens_seen": 9267959744,
      "step": 32590,
      "train_runtime": 316900.841,
      "train_tokens_per_second": 29245.614
    },
    {
      "epoch": 1.1534518483791714,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4225363492496376e-05,
      "loss": 0.3900750637054443,
      "num_input_tokens_seen": 9270835776,
      "step": 32600,
      "train_runtime": 317001.3905,
      "train_tokens_per_second": 29245.41
    },
    {
      "epoch": 1.1538056686429004,
      "grad_norm": 0.578125,
      "learning_rate": 2.422252057374267e-05,
      "loss": 0.38943114280700686,
      "num_input_tokens_seen": 9273656704,
      "step": 32610,
      "train_runtime": 317096.3531,
      "train_tokens_per_second": 29245.548
    },
    {
      "epoch": 1.1541594889066291,
      "grad_norm": 0.58984375,
      "learning_rate": 2.421967865562914e-05,
      "loss": 0.38686678409576414,
      "num_input_tokens_seen": 9276506176,
      "step": 32620,
      "train_runtime": 317195.089,
      "train_tokens_per_second": 29245.428
    },
    {
      "epoch": 1.1545133091703579,
      "grad_norm": 0.62109375,
      "learning_rate": 2.42168377375689e-05,
      "loss": 0.3897801399230957,
      "num_input_tokens_seen": 9279303040,
      "step": 32630,
      "train_runtime": 317290.2676,
      "train_tokens_per_second": 29245.47
    },
    {
      "epoch": 1.1548671294340866,
      "grad_norm": 0.6015625,
      "learning_rate": 2.421399781897558e-05,
      "loss": 0.3919291734695435,
      "num_input_tokens_seen": 9282128768,
      "step": 32640,
      "train_runtime": 317385.6821,
      "train_tokens_per_second": 29245.581
    },
    {
      "epoch": 1.1552209496978154,
      "grad_norm": 0.625,
      "learning_rate": 2.421115889926327e-05,
      "loss": 0.3873903751373291,
      "num_input_tokens_seen": 9284994944,
      "step": 32650,
      "train_runtime": 317481.1753,
      "train_tokens_per_second": 29245.813
    },
    {
      "epoch": 1.1555747699615442,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4208320977846555e-05,
      "loss": 0.3920799732208252,
      "num_input_tokens_seen": 9287864128,
      "step": 32660,
      "train_runtime": 317583.2458,
      "train_tokens_per_second": 29245.447
    },
    {
      "epoch": 1.155928590225273,
      "grad_norm": 0.62109375,
      "learning_rate": 2.4205484054140493e-05,
      "loss": 0.3898750305175781,
      "num_input_tokens_seen": 9290722688,
      "step": 32670,
      "train_runtime": 317677.5102,
      "train_tokens_per_second": 29245.768
    },
    {
      "epoch": 1.1562824104890017,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4202648127560616e-05,
      "loss": 0.3876809597015381,
      "num_input_tokens_seen": 9293619328,
      "step": 32680,
      "train_runtime": 317777.1099,
      "train_tokens_per_second": 29245.717
    },
    {
      "epoch": 1.1566362307527305,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4199813197522947e-05,
      "loss": 0.3931845426559448,
      "num_input_tokens_seen": 9296439552,
      "step": 32690,
      "train_runtime": 317871.6796,
      "train_tokens_per_second": 29245.888
    },
    {
      "epoch": 1.1569900510164592,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4196979263443985e-05,
      "loss": 0.39022073745727537,
      "num_input_tokens_seen": 9299260096,
      "step": 32700,
      "train_runtime": 317966.7076,
      "train_tokens_per_second": 29246.018
    },
    {
      "epoch": 1.157343871280188,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4194146324740697e-05,
      "loss": 0.3913739681243896,
      "num_input_tokens_seen": 9302155392,
      "step": 32710,
      "train_runtime": 318065.7709,
      "train_tokens_per_second": 29246.012
    },
    {
      "epoch": 1.1576976915439168,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4191314380830545e-05,
      "loss": 0.39194812774658205,
      "num_input_tokens_seen": 9305032704,
      "step": 32720,
      "train_runtime": 318165.6922,
      "train_tokens_per_second": 29245.871
    },
    {
      "epoch": 1.1580515118076455,
      "grad_norm": 0.59375,
      "learning_rate": 2.4188483431131453e-05,
      "loss": 0.3865678310394287,
      "num_input_tokens_seen": 9307922176,
      "step": 32730,
      "train_runtime": 318265.3656,
      "train_tokens_per_second": 29245.79
    },
    {
      "epoch": 1.1584053320713743,
      "grad_norm": 0.609375,
      "learning_rate": 2.4185653475061826e-05,
      "loss": 0.3931853771209717,
      "num_input_tokens_seen": 9310772928,
      "step": 32740,
      "train_runtime": 318363.1105,
      "train_tokens_per_second": 29245.766
    },
    {
      "epoch": 1.158759152335103,
      "grad_norm": 0.59375,
      "learning_rate": 2.418282451204055e-05,
      "loss": 0.3928253650665283,
      "num_input_tokens_seen": 9313617920,
      "step": 32750,
      "train_runtime": 318458.9538,
      "train_tokens_per_second": 29245.897
    },
    {
      "epoch": 1.159112972598832,
      "grad_norm": 0.609375,
      "learning_rate": 2.417999654148698e-05,
      "loss": 0.38882534503936766,
      "num_input_tokens_seen": 9316380864,
      "step": 32760,
      "train_runtime": 318552.7872,
      "train_tokens_per_second": 29245.956
    },
    {
      "epoch": 1.1594667928625608,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4177169562820957e-05,
      "loss": 0.39058284759521483,
      "num_input_tokens_seen": 9319236608,
      "step": 32770,
      "train_runtime": 318648.7624,
      "train_tokens_per_second": 29246.11
    },
    {
      "epoch": 1.1598206131262896,
      "grad_norm": 0.60546875,
      "learning_rate": 2.417434357546278e-05,
      "loss": 0.3878857851028442,
      "num_input_tokens_seen": 9322090112,
      "step": 32780,
      "train_runtime": 318747.3846,
      "train_tokens_per_second": 29246.013
    },
    {
      "epoch": 1.1601744333900184,
      "grad_norm": 0.61328125,
      "learning_rate": 2.417151857883324e-05,
      "loss": 0.39049334526062013,
      "num_input_tokens_seen": 9324975552,
      "step": 32790,
      "train_runtime": 318847.8388,
      "train_tokens_per_second": 29245.848
    },
    {
      "epoch": 1.1605282536537471,
      "grad_norm": 0.6015625,
      "learning_rate": 2.416869457235358e-05,
      "loss": 0.3914969444274902,
      "num_input_tokens_seen": 9327805440,
      "step": 32800,
      "train_runtime": 318945.9341,
      "train_tokens_per_second": 29245.726
    },
    {
      "epoch": 1.1608820739174759,
      "grad_norm": 0.59375,
      "learning_rate": 2.416587155544554e-05,
      "loss": 0.3889345169067383,
      "num_input_tokens_seen": 9330690368,
      "step": 32810,
      "train_runtime": 319048.2211,
      "train_tokens_per_second": 29245.392
    },
    {
      "epoch": 1.1612358941812047,
      "grad_norm": 0.578125,
      "learning_rate": 2.416304952753132e-05,
      "loss": 0.38731536865234373,
      "num_input_tokens_seen": 9333535360,
      "step": 32820,
      "train_runtime": 319147.6692,
      "train_tokens_per_second": 29245.194
    },
    {
      "epoch": 1.1615897144449334,
      "grad_norm": 0.578125,
      "learning_rate": 2.4160228488033597e-05,
      "loss": 0.3884768486022949,
      "num_input_tokens_seen": 9336407552,
      "step": 32830,
      "train_runtime": 319245.1271,
      "train_tokens_per_second": 29245.263
    },
    {
      "epoch": 1.1619435347086622,
      "grad_norm": 0.59375,
      "learning_rate": 2.415740843637551e-05,
      "loss": 0.3937070846557617,
      "num_input_tokens_seen": 9339243968,
      "step": 32840,
      "train_runtime": 319341.3287,
      "train_tokens_per_second": 29245.334
    },
    {
      "epoch": 1.162297354972391,
      "grad_norm": 0.59375,
      "learning_rate": 2.4154589371980672e-05,
      "loss": 0.3886354923248291,
      "num_input_tokens_seen": 9342061376,
      "step": 32850,
      "train_runtime": 319437.9461,
      "train_tokens_per_second": 29245.309
    },
    {
      "epoch": 1.1626511752361197,
      "grad_norm": 0.59765625,
      "learning_rate": 2.415177129427319e-05,
      "loss": 0.3893561840057373,
      "num_input_tokens_seen": 9344861312,
      "step": 32860,
      "train_runtime": 319531.9223,
      "train_tokens_per_second": 29245.47
    },
    {
      "epoch": 1.1630049954998485,
      "grad_norm": 0.625,
      "learning_rate": 2.414895420267761e-05,
      "loss": 0.39281692504882815,
      "num_input_tokens_seen": 9347685952,
      "step": 32870,
      "train_runtime": 319628.6971,
      "train_tokens_per_second": 29245.453
    },
    {
      "epoch": 1.1633588157635772,
      "grad_norm": 0.59765625,
      "learning_rate": 2.414613809661895e-05,
      "loss": 0.38995981216430664,
      "num_input_tokens_seen": 9350535936,
      "step": 32880,
      "train_runtime": 319723.3539,
      "train_tokens_per_second": 29245.708
    },
    {
      "epoch": 1.163712636027306,
      "grad_norm": 0.609375,
      "learning_rate": 2.4143322975522726e-05,
      "loss": 0.3920666933059692,
      "num_input_tokens_seen": 9353344832,
      "step": 32890,
      "train_runtime": 319819.2069,
      "train_tokens_per_second": 29245.726
    },
    {
      "epoch": 1.1640664562910348,
      "grad_norm": 0.578125,
      "learning_rate": 2.414050883881489e-05,
      "loss": 0.39069352149963377,
      "num_input_tokens_seen": 9356229248,
      "step": 32900,
      "train_runtime": 319918.5376,
      "train_tokens_per_second": 29245.661
    },
    {
      "epoch": 1.1644202765547635,
      "grad_norm": 0.59375,
      "learning_rate": 2.413769568592188e-05,
      "loss": 0.39227004051208497,
      "num_input_tokens_seen": 9359088832,
      "step": 32910,
      "train_runtime": 320015.7986,
      "train_tokens_per_second": 29245.709
    },
    {
      "epoch": 1.1647740968184923,
      "grad_norm": 0.625,
      "learning_rate": 2.41348835162706e-05,
      "loss": 0.3868955850601196,
      "num_input_tokens_seen": 9361896832,
      "step": 32920,
      "train_runtime": 320107.2464,
      "train_tokens_per_second": 29246.126
    },
    {
      "epoch": 1.1651279170822213,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4132072329288414e-05,
      "loss": 0.390920090675354,
      "num_input_tokens_seen": 9364762048,
      "step": 32930,
      "train_runtime": 320205.0195,
      "train_tokens_per_second": 29246.144
    },
    {
      "epoch": 1.16548173734595,
      "grad_norm": 0.625,
      "learning_rate": 2.4129262124403165e-05,
      "loss": 0.38875927925109866,
      "num_input_tokens_seen": 9367664576,
      "step": 32940,
      "train_runtime": 320307.9694,
      "train_tokens_per_second": 29245.806
    },
    {
      "epoch": 1.1658355576096788,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4126452901043155e-05,
      "loss": 0.3936948299407959,
      "num_input_tokens_seen": 9370459584,
      "step": 32950,
      "train_runtime": 320404.2293,
      "train_tokens_per_second": 29245.742
    },
    {
      "epoch": 1.1661893778734076,
      "grad_norm": 0.62109375,
      "learning_rate": 2.412364465863714e-05,
      "loss": 0.38907370567321775,
      "num_input_tokens_seen": 9373308352,
      "step": 32960,
      "train_runtime": 320503.1836,
      "train_tokens_per_second": 29245.601
    },
    {
      "epoch": 1.1665431981371364,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4120837396614367e-05,
      "loss": 0.38681161403656006,
      "num_input_tokens_seen": 9376113856,
      "step": 32970,
      "train_runtime": 320597.0282,
      "train_tokens_per_second": 29245.792
    },
    {
      "epoch": 1.1668970184008651,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4118031114404525e-05,
      "loss": 0.3915836811065674,
      "num_input_tokens_seen": 9378983360,
      "step": 32980,
      "train_runtime": 320694.8308,
      "train_tokens_per_second": 29245.82
    },
    {
      "epoch": 1.167250838664594,
      "grad_norm": 0.6015625,
      "learning_rate": 2.411522581143778e-05,
      "loss": 0.39067468643188474,
      "num_input_tokens_seen": 9381851328,
      "step": 32990,
      "train_runtime": 320792.8586,
      "train_tokens_per_second": 29245.824
    },
    {
      "epoch": 1.1676046589283227,
      "grad_norm": 0.5859375,
      "learning_rate": 2.4112421487144762e-05,
      "loss": 0.3922233581542969,
      "num_input_tokens_seen": 9384704960,
      "step": 33000,
      "train_runtime": 320890.7299,
      "train_tokens_per_second": 29245.796
    },
    {
      "epoch": 1.1679584791920514,
      "grad_norm": 0.61328125,
      "learning_rate": 2.410961814095655e-05,
      "loss": 0.3909944534301758,
      "num_input_tokens_seen": 9387591232,
      "step": 33010,
      "train_runtime": 320989.9865,
      "train_tokens_per_second": 29245.745
    },
    {
      "epoch": 1.1683122994557802,
      "grad_norm": 0.61328125,
      "learning_rate": 2.410681577230471e-05,
      "loss": 0.390471076965332,
      "num_input_tokens_seen": 9390438848,
      "step": 33020,
      "train_runtime": 321089.079,
      "train_tokens_per_second": 29245.588
    },
    {
      "epoch": 1.168666119719509,
      "grad_norm": 0.59375,
      "learning_rate": 2.410401438062125e-05,
      "loss": 0.390967583656311,
      "num_input_tokens_seen": 9393268608,
      "step": 33030,
      "train_runtime": 321185.8725,
      "train_tokens_per_second": 29245.585
    },
    {
      "epoch": 1.1690199399832377,
      "grad_norm": 0.6171875,
      "learning_rate": 2.4101213965338647e-05,
      "loss": 0.3953163385391235,
      "num_input_tokens_seen": 9396069120,
      "step": 33040,
      "train_runtime": 321279.5651,
      "train_tokens_per_second": 29245.773
    },
    {
      "epoch": 1.1693737602469665,
      "grad_norm": 0.58984375,
      "learning_rate": 2.4098414525889836e-05,
      "loss": 0.38777520656585696,
      "num_input_tokens_seen": 9398898688,
      "step": 33050,
      "train_runtime": 321376.7717,
      "train_tokens_per_second": 29245.731
    },
    {
      "epoch": 1.1697275805106953,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4095616061708218e-05,
      "loss": 0.3901984214782715,
      "num_input_tokens_seen": 9401752448,
      "step": 33060,
      "train_runtime": 321472.9812,
      "train_tokens_per_second": 29245.856
    },
    {
      "epoch": 1.170081400774424,
      "grad_norm": 0.62109375,
      "learning_rate": 2.4092818572227663e-05,
      "loss": 0.3925659656524658,
      "num_input_tokens_seen": 9404559936,
      "step": 33070,
      "train_runtime": 321570.0961,
      "train_tokens_per_second": 29245.754
    },
    {
      "epoch": 1.1704352210381528,
      "grad_norm": 0.578125,
      "learning_rate": 2.4090022056882475e-05,
      "loss": 0.3923038959503174,
      "num_input_tokens_seen": 9407359360,
      "step": 33080,
      "train_runtime": 321666.0551,
      "train_tokens_per_second": 29245.732
    },
    {
      "epoch": 1.1707890413018816,
      "grad_norm": 0.609375,
      "learning_rate": 2.408722651510744e-05,
      "loss": 0.39384050369262696,
      "num_input_tokens_seen": 9410227264,
      "step": 33090,
      "train_runtime": 321767.0328,
      "train_tokens_per_second": 29245.467
    },
    {
      "epoch": 1.1711428615656105,
      "grad_norm": 0.6015625,
      "learning_rate": 2.40844319463378e-05,
      "loss": 0.3934753179550171,
      "num_input_tokens_seen": 9413071936,
      "step": 33100,
      "train_runtime": 321865.6523,
      "train_tokens_per_second": 29245.345
    },
    {
      "epoch": 1.1714966818293393,
      "grad_norm": 0.609375,
      "learning_rate": 2.4081638350009247e-05,
      "loss": 0.3884658098220825,
      "num_input_tokens_seen": 9415914944,
      "step": 33110,
      "train_runtime": 321961.2524,
      "train_tokens_per_second": 29245.491
    },
    {
      "epoch": 1.171850502093068,
      "grad_norm": 0.5859375,
      "learning_rate": 2.407884572555794e-05,
      "loss": 0.39093215465545655,
      "num_input_tokens_seen": 9418758016,
      "step": 33120,
      "train_runtime": 322059.8125,
      "train_tokens_per_second": 29245.369
    },
    {
      "epoch": 1.1722043223567968,
      "grad_norm": 0.59765625,
      "learning_rate": 2.4076054072420486e-05,
      "loss": 0.39130733013153074,
      "num_input_tokens_seen": 9421605568,
      "step": 33130,
      "train_runtime": 322156.6887,
      "train_tokens_per_second": 29245.413
    },
    {
      "epoch": 1.1725581426205256,
      "grad_norm": 0.58203125,
      "learning_rate": 2.4073263390033957e-05,
      "loss": 0.39181196689605713,
      "num_input_tokens_seen": 9424470656,
      "step": 33140,
      "train_runtime": 322253.452,
      "train_tokens_per_second": 29245.523
    },
    {
      "epoch": 1.1729119628842544,
      "grad_norm": 0.6171875,
      "learning_rate": 2.407047367783588e-05,
      "loss": 0.3926100730895996,
      "num_input_tokens_seen": 9427328064,
      "step": 33150,
      "train_runtime": 322351.3235,
      "train_tokens_per_second": 29245.508
    },
    {
      "epoch": 1.1732657831479831,
      "grad_norm": 0.61328125,
      "learning_rate": 2.406768493526424e-05,
      "loss": 0.39453699588775637,
      "num_input_tokens_seen": 9430267776,
      "step": 33160,
      "train_runtime": 322453.8805,
      "train_tokens_per_second": 29245.323
    },
    {
      "epoch": 1.173619603411712,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4064897161757468e-05,
      "loss": 0.39337542057037356,
      "num_input_tokens_seen": 9433076864,
      "step": 33170,
      "train_runtime": 322546.3481,
      "train_tokens_per_second": 29245.648
    },
    {
      "epoch": 1.1739734236754407,
      "grad_norm": 0.61328125,
      "learning_rate": 2.4062110356754464e-05,
      "loss": 0.39212806224823,
      "num_input_tokens_seen": 9435879104,
      "step": 33180,
      "train_runtime": 322640.9313,
      "train_tokens_per_second": 29245.76
    },
    {
      "epoch": 1.1743272439391694,
      "grad_norm": 0.625,
      "learning_rate": 2.4059324519694572e-05,
      "loss": 0.38807973861694334,
      "num_input_tokens_seen": 9438735936,
      "step": 33190,
      "train_runtime": 322737.4517,
      "train_tokens_per_second": 29245.865
    },
    {
      "epoch": 1.1746810642028982,
      "grad_norm": 0.61328125,
      "learning_rate": 2.40565396500176e-05,
      "loss": 0.3934789180755615,
      "num_input_tokens_seen": 9441674880,
      "step": 33200,
      "train_runtime": 322840.2738,
      "train_tokens_per_second": 29245.654
    },
    {
      "epoch": 1.175034884466627,
      "grad_norm": 0.5859375,
      "learning_rate": 2.40537557471638e-05,
      "loss": 0.3919606924057007,
      "num_input_tokens_seen": 9444507392,
      "step": 33210,
      "train_runtime": 322937.084,
      "train_tokens_per_second": 29245.658
    },
    {
      "epoch": 1.1753887047303557,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4050972810573875e-05,
      "loss": 0.3910083770751953,
      "num_input_tokens_seen": 9447403584,
      "step": 33220,
      "train_runtime": 323038.894,
      "train_tokens_per_second": 29245.406
    },
    {
      "epoch": 1.1757425249940845,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4048190839689003e-05,
      "loss": 0.39489221572875977,
      "num_input_tokens_seen": 9450276032,
      "step": 33230,
      "train_runtime": 323136.2946,
      "train_tokens_per_second": 29245.48
    },
    {
      "epoch": 1.1760963452578133,
      "grad_norm": 0.58203125,
      "learning_rate": 2.404540983395078e-05,
      "loss": 0.39011573791503906,
      "num_input_tokens_seen": 9453128704,
      "step": 33240,
      "train_runtime": 323233.1523,
      "train_tokens_per_second": 29245.542
    },
    {
      "epoch": 1.1764501655215422,
      "grad_norm": 0.5859375,
      "learning_rate": 2.404262979280129e-05,
      "loss": 0.38842086791992186,
      "num_input_tokens_seen": 9455938816,
      "step": 33250,
      "train_runtime": 323329.3394,
      "train_tokens_per_second": 29245.533
    },
    {
      "epoch": 1.1768039857852708,
      "grad_norm": 0.61328125,
      "learning_rate": 2.403985071568304e-05,
      "loss": 0.39597139358520506,
      "num_input_tokens_seen": 9458846912,
      "step": 33260,
      "train_runtime": 323431.0297,
      "train_tokens_per_second": 29245.329
    },
    {
      "epoch": 1.1771578060489998,
      "grad_norm": 0.6015625,
      "learning_rate": 2.4037072602039002e-05,
      "loss": 0.3884121894836426,
      "num_input_tokens_seen": 9461655808,
      "step": 33270,
      "train_runtime": 323527.0981,
      "train_tokens_per_second": 29245.327
    },
    {
      "epoch": 1.1775116263127285,
      "grad_norm": 0.62109375,
      "learning_rate": 2.4034295451312596e-05,
      "loss": 0.39089303016662597,
      "num_input_tokens_seen": 9464510336,
      "step": 33280,
      "train_runtime": 323624.5172,
      "train_tokens_per_second": 29245.344
    },
    {
      "epoch": 1.1778654465764573,
      "grad_norm": 0.578125,
      "learning_rate": 2.403151926294769e-05,
      "loss": 0.38807382583618166,
      "num_input_tokens_seen": 9467379904,
      "step": 33290,
      "train_runtime": 323721.2511,
      "train_tokens_per_second": 29245.469
    },
    {
      "epoch": 1.178219266840186,
      "grad_norm": 0.62890625,
      "learning_rate": 2.4028744036388606e-05,
      "loss": 0.39189863204956055,
      "num_input_tokens_seen": 9470173248,
      "step": 33300,
      "train_runtime": 323814.7964,
      "train_tokens_per_second": 29245.647
    },
    {
      "epoch": 1.1785730871039148,
      "grad_norm": 0.60546875,
      "learning_rate": 2.4025969771080115e-05,
      "loss": 0.39383530616760254,
      "num_input_tokens_seen": 9473054848,
      "step": 33310,
      "train_runtime": 323915.1414,
      "train_tokens_per_second": 29245.483
    },
    {
      "epoch": 1.1789269073676436,
      "grad_norm": 0.609375,
      "learning_rate": 2.402319646646742e-05,
      "loss": 0.3898460388183594,
      "num_input_tokens_seen": 9475877696,
      "step": 33320,
      "train_runtime": 324010.8769,
      "train_tokens_per_second": 29245.554
    },
    {
      "epoch": 1.1792807276313724,
      "grad_norm": 0.60546875,
      "learning_rate": 2.40204241219962e-05,
      "loss": 0.387637996673584,
      "num_input_tokens_seen": 9478680000,
      "step": 33330,
      "train_runtime": 324104.6018,
      "train_tokens_per_second": 29245.743
    },
    {
      "epoch": 1.1796345478951011,
      "grad_norm": 0.58203125,
      "learning_rate": 2.4017652737112568e-05,
      "loss": 0.38987891674041747,
      "num_input_tokens_seen": 9481559616,
      "step": 33340,
      "train_runtime": 324204.3136,
      "train_tokens_per_second": 29245.631
    },
    {
      "epoch": 1.17998836815883,
      "grad_norm": 0.578125,
      "learning_rate": 2.4014882311263073e-05,
      "loss": 0.38961448669433596,
      "num_input_tokens_seen": 9484425536,
      "step": 33350,
      "train_runtime": 324302.8261,
      "train_tokens_per_second": 29245.584
    },
    {
      "epoch": 1.1803421884225587,
      "grad_norm": 0.578125,
      "learning_rate": 2.4012112843894734e-05,
      "loss": 0.39429607391357424,
      "num_input_tokens_seen": 9487289024,
      "step": 33360,
      "train_runtime": 324400.6646,
      "train_tokens_per_second": 29245.591
    },
    {
      "epoch": 1.1806960086862874,
      "grad_norm": 0.6015625,
      "learning_rate": 2.400934433445499e-05,
      "loss": 0.39247570037841795,
      "num_input_tokens_seen": 9490138112,
      "step": 33370,
      "train_runtime": 324497.3767,
      "train_tokens_per_second": 29245.654
    },
    {
      "epoch": 1.1810498289500162,
      "grad_norm": 0.62890625,
      "learning_rate": 2.4006576782391752e-05,
      "loss": 0.38724794387817385,
      "num_input_tokens_seen": 9492945728,
      "step": 33380,
      "train_runtime": 324590.6473,
      "train_tokens_per_second": 29245.9
    },
    {
      "epoch": 1.181403649213745,
      "grad_norm": 0.61328125,
      "learning_rate": 2.400381018715336e-05,
      "loss": 0.3899810791015625,
      "num_input_tokens_seen": 9495778368,
      "step": 33390,
      "train_runtime": 324686.0745,
      "train_tokens_per_second": 29246.029
    },
    {
      "epoch": 1.1817574694774737,
      "grad_norm": 0.609375,
      "learning_rate": 2.40010445481886e-05,
      "loss": 0.3902117252349854,
      "num_input_tokens_seen": 9498635840,
      "step": 33400,
      "train_runtime": 324783.6944,
      "train_tokens_per_second": 29246.037
    },
    {
      "epoch": 1.1821112897412025,
      "grad_norm": 0.59375,
      "learning_rate": 2.399827986494671e-05,
      "loss": 0.38477678298950196,
      "num_input_tokens_seen": 9501464128,
      "step": 33410,
      "train_runtime": 324878.1467,
      "train_tokens_per_second": 29246.24
    },
    {
      "epoch": 1.1824651100049315,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3995516136877368e-05,
      "loss": 0.3882612943649292,
      "num_input_tokens_seen": 9504287872,
      "step": 33420,
      "train_runtime": 324972.9569,
      "train_tokens_per_second": 29246.396
    },
    {
      "epoch": 1.18281893026866,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3992753363430694e-05,
      "loss": 0.39034693241119384,
      "num_input_tokens_seen": 9507071424,
      "step": 33430,
      "train_runtime": 325065.6531,
      "train_tokens_per_second": 29246.619
    },
    {
      "epoch": 1.183172750532389,
      "grad_norm": 0.609375,
      "learning_rate": 2.3989991544057252e-05,
      "loss": 0.3888209342956543,
      "num_input_tokens_seen": 9509908096,
      "step": 33440,
      "train_runtime": 325160.0972,
      "train_tokens_per_second": 29246.848
    },
    {
      "epoch": 1.1835265707961178,
      "grad_norm": 0.59375,
      "learning_rate": 2.3987230678208044e-05,
      "loss": 0.39576475620269774,
      "num_input_tokens_seen": 9512749568,
      "step": 33450,
      "train_runtime": 325257.0272,
      "train_tokens_per_second": 29246.869
    },
    {
      "epoch": 1.1838803910598465,
      "grad_norm": 0.59375,
      "learning_rate": 2.3984470765334527e-05,
      "loss": 0.38622972965240476,
      "num_input_tokens_seen": 9515630272,
      "step": 33460,
      "train_runtime": 325356.6755,
      "train_tokens_per_second": 29246.765
    },
    {
      "epoch": 1.1842342113235753,
      "grad_norm": 0.59765625,
      "learning_rate": 2.398171180488859e-05,
      "loss": 0.39551146030426027,
      "num_input_tokens_seen": 9518488256,
      "step": 33470,
      "train_runtime": 325455.8428,
      "train_tokens_per_second": 29246.635
    },
    {
      "epoch": 1.184588031587304,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3978953796322565e-05,
      "loss": 0.3946110248565674,
      "num_input_tokens_seen": 9521336768,
      "step": 33480,
      "train_runtime": 325553.9345,
      "train_tokens_per_second": 29246.573
    },
    {
      "epoch": 1.1849418518510328,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3976196739089218e-05,
      "loss": 0.39379899501800536,
      "num_input_tokens_seen": 9524181696,
      "step": 33490,
      "train_runtime": 325652.0931,
      "train_tokens_per_second": 29246.493
    },
    {
      "epoch": 1.1852956721147616,
      "grad_norm": 0.60546875,
      "learning_rate": 2.397344063264177e-05,
      "loss": 0.3925729513168335,
      "num_input_tokens_seen": 9527027968,
      "step": 33500,
      "train_runtime": 325749.8174,
      "train_tokens_per_second": 29246.457
    },
    {
      "epoch": 1.1856494923784904,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3970685476433867e-05,
      "loss": 0.38857645988464357,
      "num_input_tokens_seen": 9529800960,
      "step": 33510,
      "train_runtime": 325842.2382,
      "train_tokens_per_second": 29246.672
    },
    {
      "epoch": 1.1860033126422191,
      "grad_norm": 0.59375,
      "learning_rate": 2.396793126991961e-05,
      "loss": 0.39438643455505373,
      "num_input_tokens_seen": 9532695360,
      "step": 33520,
      "train_runtime": 325942.5265,
      "train_tokens_per_second": 29246.553
    },
    {
      "epoch": 1.186357132905948,
      "grad_norm": 0.61328125,
      "learning_rate": 2.396517801255352e-05,
      "loss": 0.3892963409423828,
      "num_input_tokens_seen": 9535536512,
      "step": 33530,
      "train_runtime": 326038.781,
      "train_tokens_per_second": 29246.633
    },
    {
      "epoch": 1.1867109531696767,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3962425703790577e-05,
      "loss": 0.3927242994308472,
      "num_input_tokens_seen": 9538409280,
      "step": 33540,
      "train_runtime": 326135.9822,
      "train_tokens_per_second": 29246.725
    },
    {
      "epoch": 1.1870647734334054,
      "grad_norm": 0.61328125,
      "learning_rate": 2.395967434308618e-05,
      "loss": 0.388751220703125,
      "num_input_tokens_seen": 9541272320,
      "step": 33550,
      "train_runtime": 326232.7178,
      "train_tokens_per_second": 29246.828
    },
    {
      "epoch": 1.1874185936971342,
      "grad_norm": 0.60546875,
      "learning_rate": 2.395692392989619e-05,
      "loss": 0.3861258506774902,
      "num_input_tokens_seen": 9544064320,
      "step": 33560,
      "train_runtime": 326327.8187,
      "train_tokens_per_second": 29246.861
    },
    {
      "epoch": 1.187772413960863,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3954174463676868e-05,
      "loss": 0.3948571443557739,
      "num_input_tokens_seen": 9546886080,
      "step": 33570,
      "train_runtime": 326425.6548,
      "train_tokens_per_second": 29246.739
    },
    {
      "epoch": 1.1881262342245917,
      "grad_norm": 0.59375,
      "learning_rate": 2.395142594388495e-05,
      "loss": 0.3900721788406372,
      "num_input_tokens_seen": 9549726336,
      "step": 33580,
      "train_runtime": 326520.4389,
      "train_tokens_per_second": 29246.948
    },
    {
      "epoch": 1.1884800544883207,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3948678369977577e-05,
      "loss": 0.3959944725036621,
      "num_input_tokens_seen": 9552575936,
      "step": 33590,
      "train_runtime": 326618.5543,
      "train_tokens_per_second": 29246.887
    },
    {
      "epoch": 1.1888338747520493,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3945931741412362e-05,
      "loss": 0.3917521953582764,
      "num_input_tokens_seen": 9555412928,
      "step": 33600,
      "train_runtime": 326716.4835,
      "train_tokens_per_second": 29246.804
    },
    {
      "epoch": 1.1891876950157783,
      "grad_norm": 0.5859375,
      "learning_rate": 2.394318605764731e-05,
      "loss": 0.3932295322418213,
      "num_input_tokens_seen": 9558267328,
      "step": 33610,
      "train_runtime": 326814.7158,
      "train_tokens_per_second": 29246.747
    },
    {
      "epoch": 1.189541515279507,
      "grad_norm": 0.609375,
      "learning_rate": 2.3940441318140893e-05,
      "loss": 0.38986716270446775,
      "num_input_tokens_seen": 9561136192,
      "step": 33620,
      "train_runtime": 326913.7845,
      "train_tokens_per_second": 29246.66
    },
    {
      "epoch": 1.1898953355432358,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3937697522352013e-05,
      "loss": 0.3913094520568848,
      "num_input_tokens_seen": 9564021504,
      "step": 33630,
      "train_runtime": 327014.8582,
      "train_tokens_per_second": 29246.443
    },
    {
      "epoch": 1.1902491558069646,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3934954669739984e-05,
      "loss": 0.3905174970626831,
      "num_input_tokens_seen": 9566898944,
      "step": 33640,
      "train_runtime": 327114.9105,
      "train_tokens_per_second": 29246.294
    },
    {
      "epoch": 1.1906029760706933,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3932212759764587e-05,
      "loss": 0.3906787157058716,
      "num_input_tokens_seen": 9569715968,
      "step": 33650,
      "train_runtime": 327209.5814,
      "train_tokens_per_second": 29246.442
    },
    {
      "epoch": 1.190956796334422,
      "grad_norm": 0.640625,
      "learning_rate": 2.3929471791886006e-05,
      "loss": 0.3931232452392578,
      "num_input_tokens_seen": 9572561984,
      "step": 33660,
      "train_runtime": 327307.4466,
      "train_tokens_per_second": 29246.392
    },
    {
      "epoch": 1.1913106165981509,
      "grad_norm": 0.59375,
      "learning_rate": 2.3926731765564872e-05,
      "loss": 0.3944676399230957,
      "num_input_tokens_seen": 9575418432,
      "step": 33670,
      "train_runtime": 327406.388,
      "train_tokens_per_second": 29246.279
    },
    {
      "epoch": 1.1916644368618796,
      "grad_norm": 0.609375,
      "learning_rate": 2.3923992680262258e-05,
      "loss": 0.3941665172576904,
      "num_input_tokens_seen": 9578277056,
      "step": 33680,
      "train_runtime": 327504.9892,
      "train_tokens_per_second": 29246.202
    },
    {
      "epoch": 1.1920182571256084,
      "grad_norm": 0.57421875,
      "learning_rate": 2.3921254535439645e-05,
      "loss": 0.3895529270172119,
      "num_input_tokens_seen": 9581097408,
      "step": 33690,
      "train_runtime": 327599.2772,
      "train_tokens_per_second": 29246.394
    },
    {
      "epoch": 1.1923720773893371,
      "grad_norm": 0.59375,
      "learning_rate": 2.391851733055896e-05,
      "loss": 0.3941351890563965,
      "num_input_tokens_seen": 9583940608,
      "step": 33700,
      "train_runtime": 327697.4113,
      "train_tokens_per_second": 29246.312
    },
    {
      "epoch": 1.192725897653066,
      "grad_norm": 0.63671875,
      "learning_rate": 2.3915781065082566e-05,
      "loss": 0.3885958194732666,
      "num_input_tokens_seen": 9586741440,
      "step": 33710,
      "train_runtime": 327790.943,
      "train_tokens_per_second": 29246.511
    },
    {
      "epoch": 1.1930797179167947,
      "grad_norm": 0.6015625,
      "learning_rate": 2.391304573847324e-05,
      "loss": 0.39181630611419677,
      "num_input_tokens_seen": 9589594176,
      "step": 33720,
      "train_runtime": 327888.5916,
      "train_tokens_per_second": 29246.501
    },
    {
      "epoch": 1.1934335381805234,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3910311350194205e-05,
      "loss": 0.3927597999572754,
      "num_input_tokens_seen": 9592433600,
      "step": 33730,
      "train_runtime": 327986.9084,
      "train_tokens_per_second": 29246.392
    },
    {
      "epoch": 1.1937873584442522,
      "grad_norm": 0.5859375,
      "learning_rate": 2.39075778997091e-05,
      "loss": 0.3883208751678467,
      "num_input_tokens_seen": 9595292032,
      "step": 33740,
      "train_runtime": 328083.1119,
      "train_tokens_per_second": 29246.528
    },
    {
      "epoch": 1.194141178707981,
      "grad_norm": 0.59765625,
      "learning_rate": 2.390484538648201e-05,
      "loss": 0.3834574222564697,
      "num_input_tokens_seen": 9598131072,
      "step": 33750,
      "train_runtime": 328180.4448,
      "train_tokens_per_second": 29246.505
    },
    {
      "epoch": 1.19449499897171,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3902113809977433e-05,
      "loss": 0.3930532455444336,
      "num_input_tokens_seen": 9601000896,
      "step": 33760,
      "train_runtime": 328279.463,
      "train_tokens_per_second": 29246.426
    },
    {
      "epoch": 1.1948488192354387,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3899383169660297e-05,
      "loss": 0.38910708427429197,
      "num_input_tokens_seen": 9603877632,
      "step": 33770,
      "train_runtime": 328377.339,
      "train_tokens_per_second": 29246.469
    },
    {
      "epoch": 1.1952026394991675,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3896653464995968e-05,
      "loss": 0.39278712272644045,
      "num_input_tokens_seen": 9606686784,
      "step": 33780,
      "train_runtime": 328469.3258,
      "train_tokens_per_second": 29246.831
    },
    {
      "epoch": 1.1955564597628963,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3893924695450232e-05,
      "loss": 0.39176130294799805,
      "num_input_tokens_seen": 9609472128,
      "step": 33790,
      "train_runtime": 328564.3086,
      "train_tokens_per_second": 29246.853
    },
    {
      "epoch": 1.195910280026625,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3891196860489297e-05,
      "loss": 0.3888782739639282,
      "num_input_tokens_seen": 9612316480,
      "step": 33800,
      "train_runtime": 328658.962,
      "train_tokens_per_second": 29247.085
    },
    {
      "epoch": 1.1962641002903538,
      "grad_norm": 0.609375,
      "learning_rate": 2.3888469959579813e-05,
      "loss": 0.3965553045272827,
      "num_input_tokens_seen": 9615149248,
      "step": 33810,
      "train_runtime": 328754.7708,
      "train_tokens_per_second": 29247.178
    },
    {
      "epoch": 1.1966179205540826,
      "grad_norm": 0.61328125,
      "learning_rate": 2.388574399218883e-05,
      "loss": 0.3913938045501709,
      "num_input_tokens_seen": 9618028096,
      "step": 33820,
      "train_runtime": 328853.1235,
      "train_tokens_per_second": 29247.185
    },
    {
      "epoch": 1.1969717408178113,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3883018957783854e-05,
      "loss": 0.390216588973999,
      "num_input_tokens_seen": 9620882880,
      "step": 33830,
      "train_runtime": 328952.5224,
      "train_tokens_per_second": 29247.026
    },
    {
      "epoch": 1.19732556108154,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3880294855832802e-05,
      "loss": 0.38720552921295165,
      "num_input_tokens_seen": 9623748352,
      "step": 33840,
      "train_runtime": 329051.7335,
      "train_tokens_per_second": 29246.916
    },
    {
      "epoch": 1.1976793813452689,
      "grad_norm": 0.57421875,
      "learning_rate": 2.387757168580401e-05,
      "loss": 0.39399967193603513,
      "num_input_tokens_seen": 9626657536,
      "step": 33850,
      "train_runtime": 329154.6699,
      "train_tokens_per_second": 29246.608
    },
    {
      "epoch": 1.1980332016089976,
      "grad_norm": 0.609375,
      "learning_rate": 2.387484944716625e-05,
      "loss": 0.3881422519683838,
      "num_input_tokens_seen": 9629524736,
      "step": 33860,
      "train_runtime": 329252.0356,
      "train_tokens_per_second": 29246.667
    },
    {
      "epoch": 1.1983870218727264,
      "grad_norm": 0.58984375,
      "learning_rate": 2.387212813938871e-05,
      "loss": 0.39118475914001466,
      "num_input_tokens_seen": 9632425664,
      "step": 33870,
      "train_runtime": 329351.5655,
      "train_tokens_per_second": 29246.637
    },
    {
      "epoch": 1.1987408421364552,
      "grad_norm": 0.578125,
      "learning_rate": 2.3869407761940998e-05,
      "loss": 0.3896019697189331,
      "num_input_tokens_seen": 9635266304,
      "step": 33880,
      "train_runtime": 329447.0366,
      "train_tokens_per_second": 29246.784
    },
    {
      "epoch": 1.199094662400184,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3866688314293158e-05,
      "loss": 0.3908462762832642,
      "num_input_tokens_seen": 9638069824,
      "step": 33890,
      "train_runtime": 329540.353,
      "train_tokens_per_second": 29247.009
    },
    {
      "epoch": 1.1994484826639127,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3863969795915643e-05,
      "loss": 0.3864020347595215,
      "num_input_tokens_seen": 9640934016,
      "step": 33900,
      "train_runtime": 329638.5296,
      "train_tokens_per_second": 29246.988
    },
    {
      "epoch": 1.1998023029276415,
      "grad_norm": 0.609375,
      "learning_rate": 2.3861252206279336e-05,
      "loss": 0.3906244277954102,
      "num_input_tokens_seen": 9643792320,
      "step": 33910,
      "train_runtime": 329736.0231,
      "train_tokens_per_second": 29247.009
    },
    {
      "epoch": 1.2001561231913702,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3858535544855544e-05,
      "loss": 0.38703858852386475,
      "num_input_tokens_seen": 9646623424,
      "step": 33920,
      "train_runtime": 329833.3356,
      "train_tokens_per_second": 29246.963
    },
    {
      "epoch": 1.2005099434550992,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3855819811115987e-05,
      "loss": 0.39233222007751467,
      "num_input_tokens_seen": 9649468544,
      "step": 33930,
      "train_runtime": 329928.0479,
      "train_tokens_per_second": 29247.191
    },
    {
      "epoch": 1.200863763718828,
      "grad_norm": 0.58984375,
      "learning_rate": 2.385310500453281e-05,
      "loss": 0.3908323049545288,
      "num_input_tokens_seen": 9652265792,
      "step": 33940,
      "train_runtime": 330025.0956,
      "train_tokens_per_second": 29247.066
    },
    {
      "epoch": 1.2012175839825567,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3850391124578573e-05,
      "loss": 0.3900197982788086,
      "num_input_tokens_seen": 9655174464,
      "step": 33950,
      "train_runtime": 330123.6009,
      "train_tokens_per_second": 29247.15
    },
    {
      "epoch": 1.2015714042462855,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3847678170726275e-05,
      "loss": 0.3944376468658447,
      "num_input_tokens_seen": 9658019648,
      "step": 33960,
      "train_runtime": 330219.7724,
      "train_tokens_per_second": 29247.248
    },
    {
      "epoch": 1.2019252245100143,
      "grad_norm": 0.609375,
      "learning_rate": 2.3844966142449312e-05,
      "loss": 0.3907630205154419,
      "num_input_tokens_seen": 9660785984,
      "step": 33970,
      "train_runtime": 330312.3133,
      "train_tokens_per_second": 29247.429
    },
    {
      "epoch": 1.202279044773743,
      "grad_norm": 0.59375,
      "learning_rate": 2.384225503922151e-05,
      "loss": 0.3928015470504761,
      "num_input_tokens_seen": 9663646464,
      "step": 33980,
      "train_runtime": 330410.0502,
      "train_tokens_per_second": 29247.435
    },
    {
      "epoch": 1.2026328650374718,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3839544860517103e-05,
      "loss": 0.3893132209777832,
      "num_input_tokens_seen": 9666481280,
      "step": 33990,
      "train_runtime": 330506.9991,
      "train_tokens_per_second": 29247.433
    },
    {
      "epoch": 1.2029866853012006,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3836835605810765e-05,
      "loss": 0.39322385787963865,
      "num_input_tokens_seen": 9669289216,
      "step": 34000,
      "train_runtime": 330603.2874,
      "train_tokens_per_second": 29247.408
    },
    {
      "epoch": 1.2033405055649293,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3834127274577564e-05,
      "loss": 0.3904114723205566,
      "num_input_tokens_seen": 9672089728,
      "step": 34010,
      "train_runtime": 330696.22,
      "train_tokens_per_second": 29247.657
    },
    {
      "epoch": 1.203694325828658,
      "grad_norm": 0.5703125,
      "learning_rate": 2.3831419866293005e-05,
      "loss": 0.3903738260269165,
      "num_input_tokens_seen": 9674950272,
      "step": 34020,
      "train_runtime": 330793.7231,
      "train_tokens_per_second": 29247.684
    },
    {
      "epoch": 1.2040481460923869,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3828713380432997e-05,
      "loss": 0.39186806678771974,
      "num_input_tokens_seen": 9677808448,
      "step": 34030,
      "train_runtime": 330892.5336,
      "train_tokens_per_second": 29247.588
    },
    {
      "epoch": 1.2044019663561156,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3826007816473866e-05,
      "loss": 0.3914055824279785,
      "num_input_tokens_seen": 9680670464,
      "step": 34040,
      "train_runtime": 330992.3933,
      "train_tokens_per_second": 29247.411
    },
    {
      "epoch": 1.2047557866198444,
      "grad_norm": 0.609375,
      "learning_rate": 2.3823303173892365e-05,
      "loss": 0.3886967897415161,
      "num_input_tokens_seen": 9683566656,
      "step": 34050,
      "train_runtime": 331092.4404,
      "train_tokens_per_second": 29247.32
    },
    {
      "epoch": 1.2051096068835732,
      "grad_norm": 0.60546875,
      "learning_rate": 2.382059945216565e-05,
      "loss": 0.3942198514938354,
      "num_input_tokens_seen": 9686376832,
      "step": 34060,
      "train_runtime": 331187.8427,
      "train_tokens_per_second": 29247.38
    },
    {
      "epoch": 1.205463427147302,
      "grad_norm": 0.59375,
      "learning_rate": 2.3817896650771297e-05,
      "loss": 0.3941674709320068,
      "num_input_tokens_seen": 9689171392,
      "step": 34070,
      "train_runtime": 331282.3607,
      "train_tokens_per_second": 29247.471
    },
    {
      "epoch": 1.205817247411031,
      "grad_norm": 0.62109375,
      "learning_rate": 2.38151947691873e-05,
      "loss": 0.389211368560791,
      "num_input_tokens_seen": 9692054784,
      "step": 34080,
      "train_runtime": 331379.7936,
      "train_tokens_per_second": 29247.573
    },
    {
      "epoch": 1.2061710676747595,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3812493806892066e-05,
      "loss": 0.3886995315551758,
      "num_input_tokens_seen": 9694916736,
      "step": 34090,
      "train_runtime": 331477.8089,
      "train_tokens_per_second": 29247.559
    },
    {
      "epoch": 1.2065248879384884,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3809793763364406e-05,
      "loss": 0.3905021190643311,
      "num_input_tokens_seen": 9697683648,
      "step": 34100,
      "train_runtime": 331568.8774,
      "train_tokens_per_second": 29247.871
    },
    {
      "epoch": 1.2068787082022172,
      "grad_norm": 0.59375,
      "learning_rate": 2.3807094638083575e-05,
      "loss": 0.39222333431243894,
      "num_input_tokens_seen": 9700529856,
      "step": 34110,
      "train_runtime": 331667.8596,
      "train_tokens_per_second": 29247.724
    },
    {
      "epoch": 1.207232528465946,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3804396430529196e-05,
      "loss": 0.38743300437927247,
      "num_input_tokens_seen": 9703314112,
      "step": 34120,
      "train_runtime": 331757.5566,
      "train_tokens_per_second": 29248.208
    },
    {
      "epoch": 1.2075863487296747,
      "grad_norm": 0.60546875,
      "learning_rate": 2.380169914018134e-05,
      "loss": 0.3886707305908203,
      "num_input_tokens_seen": 9706138560,
      "step": 34130,
      "train_runtime": 331852.6993,
      "train_tokens_per_second": 29248.334
    },
    {
      "epoch": 1.2079401689934035,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3799002766520473e-05,
      "loss": 0.39063329696655275,
      "num_input_tokens_seen": 9708987008,
      "step": 34140,
      "train_runtime": 331948.238,
      "train_tokens_per_second": 29248.497
    },
    {
      "epoch": 1.2082939892571323,
      "grad_norm": 0.58984375,
      "learning_rate": 2.379630730902749e-05,
      "loss": 0.3893106698989868,
      "num_input_tokens_seen": 9711806784,
      "step": 34150,
      "train_runtime": 332045.9647,
      "train_tokens_per_second": 29248.381
    },
    {
      "epoch": 1.208647809520861,
      "grad_norm": 0.58203125,
      "learning_rate": 2.3793612767183668e-05,
      "loss": 0.38825669288635256,
      "num_input_tokens_seen": 9714642624,
      "step": 34160,
      "train_runtime": 332141.4053,
      "train_tokens_per_second": 29248.514
    },
    {
      "epoch": 1.2090016297845898,
      "grad_norm": 0.625,
      "learning_rate": 2.3790919140470726e-05,
      "loss": 0.38932573795318604,
      "num_input_tokens_seen": 9717448704,
      "step": 34170,
      "train_runtime": 332235.293,
      "train_tokens_per_second": 29248.695
    },
    {
      "epoch": 1.2093554500483186,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3788226428370774e-05,
      "loss": 0.3852090835571289,
      "num_input_tokens_seen": 9720349056,
      "step": 34180,
      "train_runtime": 332334.6336,
      "train_tokens_per_second": 29248.679
    },
    {
      "epoch": 1.2097092703120473,
      "grad_norm": 0.59375,
      "learning_rate": 2.3785534630366355e-05,
      "loss": 0.39013381004333497,
      "num_input_tokens_seen": 9723184512,
      "step": 34190,
      "train_runtime": 332429.7361,
      "train_tokens_per_second": 29248.841
    },
    {
      "epoch": 1.210063090575776,
      "grad_norm": 0.58984375,
      "learning_rate": 2.378284374594038e-05,
      "loss": 0.3916423559188843,
      "num_input_tokens_seen": 9726001856,
      "step": 34200,
      "train_runtime": 332524.4548,
      "train_tokens_per_second": 29248.982
    },
    {
      "epoch": 1.2104169108395049,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3780153774576215e-05,
      "loss": 0.39029865264892577,
      "num_input_tokens_seen": 9728830720,
      "step": 34210,
      "train_runtime": 332616.7762,
      "train_tokens_per_second": 29249.369
    },
    {
      "epoch": 1.2107707311032336,
      "grad_norm": 0.609375,
      "learning_rate": 2.3777464715757608e-05,
      "loss": 0.39426584243774415,
      "num_input_tokens_seen": 9731613632,
      "step": 34220,
      "train_runtime": 332710.034,
      "train_tokens_per_second": 29249.535
    },
    {
      "epoch": 1.2111245513669624,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3774776568968716e-05,
      "loss": 0.3904907703399658,
      "num_input_tokens_seen": 9734413504,
      "step": 34230,
      "train_runtime": 332805.1423,
      "train_tokens_per_second": 29249.589
    },
    {
      "epoch": 1.2114783716306912,
      "grad_norm": 0.6015625,
      "learning_rate": 2.377208933369413e-05,
      "loss": 0.39004790782928467,
      "num_input_tokens_seen": 9737197568,
      "step": 34240,
      "train_runtime": 332899.1789,
      "train_tokens_per_second": 29249.689
    },
    {
      "epoch": 1.2118321918944202,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3769403009418808e-05,
      "loss": 0.38975651264190675,
      "num_input_tokens_seen": 9740070400,
      "step": 34250,
      "train_runtime": 332998.5262,
      "train_tokens_per_second": 29249.59
    },
    {
      "epoch": 1.2121860121581487,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3766717595628145e-05,
      "loss": 0.38875398635864256,
      "num_input_tokens_seen": 9742856832,
      "step": 34260,
      "train_runtime": 333094.1005,
      "train_tokens_per_second": 29249.563
    },
    {
      "epoch": 1.2125398324218777,
      "grad_norm": 0.64453125,
      "learning_rate": 2.376403309180794e-05,
      "loss": 0.39031352996826174,
      "num_input_tokens_seen": 9745677376,
      "step": 34270,
      "train_runtime": 333189.1765,
      "train_tokens_per_second": 29249.682
    },
    {
      "epoch": 1.2128936526856064,
      "grad_norm": 0.609375,
      "learning_rate": 2.3761349497444387e-05,
      "loss": 0.3901947021484375,
      "num_input_tokens_seen": 9748482816,
      "step": 34280,
      "train_runtime": 333283.9658,
      "train_tokens_per_second": 29249.78
    },
    {
      "epoch": 1.2132474729493352,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3758666812024093e-05,
      "loss": 0.38884131908416747,
      "num_input_tokens_seen": 9751270656,
      "step": 34290,
      "train_runtime": 333375.7528,
      "train_tokens_per_second": 29250.09
    },
    {
      "epoch": 1.213601293213064,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3755985035034072e-05,
      "loss": 0.3917670726776123,
      "num_input_tokens_seen": 9754067584,
      "step": 34300,
      "train_runtime": 333467.5621,
      "train_tokens_per_second": 29250.424
    },
    {
      "epoch": 1.2139551134767927,
      "grad_norm": 0.6328125,
      "learning_rate": 2.3753304165961733e-05,
      "loss": 0.39243981838226316,
      "num_input_tokens_seen": 9756888192,
      "step": 34310,
      "train_runtime": 333565.1188,
      "train_tokens_per_second": 29250.325
    },
    {
      "epoch": 1.2143089337405215,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3750624204294913e-05,
      "loss": 0.39138786792755126,
      "num_input_tokens_seen": 9759714240,
      "step": 34320,
      "train_runtime": 333662.4836,
      "train_tokens_per_second": 29250.26
    },
    {
      "epoch": 1.2146627540042503,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3747945149521823e-05,
      "loss": 0.3886219024658203,
      "num_input_tokens_seen": 9762568192,
      "step": 34330,
      "train_runtime": 333759.2047,
      "train_tokens_per_second": 29250.334
    },
    {
      "epoch": 1.215016574267979,
      "grad_norm": 0.6015625,
      "learning_rate": 2.374526700113111e-05,
      "loss": 0.3916738986968994,
      "num_input_tokens_seen": 9765409216,
      "step": 34340,
      "train_runtime": 333856.5116,
      "train_tokens_per_second": 29250.318
    },
    {
      "epoch": 1.2153703945317078,
      "grad_norm": 0.59765625,
      "learning_rate": 2.374258975861179e-05,
      "loss": 0.3902529239654541,
      "num_input_tokens_seen": 9768232192,
      "step": 34350,
      "train_runtime": 333949.5065,
      "train_tokens_per_second": 29250.626
    },
    {
      "epoch": 1.2157242147954366,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3739913421453303e-05,
      "loss": 0.39206860065460203,
      "num_input_tokens_seen": 9771099072,
      "step": 34360,
      "train_runtime": 334046.4062,
      "train_tokens_per_second": 29250.724
    },
    {
      "epoch": 1.2160780350591653,
      "grad_norm": 0.6015625,
      "learning_rate": 2.37372379891455e-05,
      "loss": 0.3900717258453369,
      "num_input_tokens_seen": 9773976512,
      "step": 34370,
      "train_runtime": 334148.7894,
      "train_tokens_per_second": 29250.372
    },
    {
      "epoch": 1.216431855322894,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3734563461178615e-05,
      "loss": 0.38710927963256836,
      "num_input_tokens_seen": 9776848576,
      "step": 34380,
      "train_runtime": 334251.7521,
      "train_tokens_per_second": 29249.955
    },
    {
      "epoch": 1.2167856755866229,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3731889837043294e-05,
      "loss": 0.39729976654052734,
      "num_input_tokens_seen": 9779679808,
      "step": 34390,
      "train_runtime": 334349.2568,
      "train_tokens_per_second": 29249.892
    },
    {
      "epoch": 1.2171394958503516,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3729217116230586e-05,
      "loss": 0.3876819133758545,
      "num_input_tokens_seen": 9782507968,
      "step": 34400,
      "train_runtime": 334444.0391,
      "train_tokens_per_second": 29250.059
    },
    {
      "epoch": 1.2174933161140804,
      "grad_norm": 0.609375,
      "learning_rate": 2.3726545298231932e-05,
      "loss": 0.38928420543670655,
      "num_input_tokens_seen": 9785377728,
      "step": 34410,
      "train_runtime": 334544.0338,
      "train_tokens_per_second": 29249.895
    },
    {
      "epoch": 1.2178471363778094,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3723874382539182e-05,
      "loss": 0.39150271415710447,
      "num_input_tokens_seen": 9788234496,
      "step": 34420,
      "train_runtime": 334640.2833,
      "train_tokens_per_second": 29250.019
    },
    {
      "epoch": 1.218200956641538,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3721204368644585e-05,
      "loss": 0.3904736042022705,
      "num_input_tokens_seen": 9791061312,
      "step": 34430,
      "train_runtime": 334738.5852,
      "train_tokens_per_second": 29249.874
    },
    {
      "epoch": 1.218554776905267,
      "grad_norm": 0.6171875,
      "learning_rate": 2.371853525604079e-05,
      "loss": 0.393033504486084,
      "num_input_tokens_seen": 9793960000,
      "step": 34440,
      "train_runtime": 334839.5084,
      "train_tokens_per_second": 29249.714
    },
    {
      "epoch": 1.2189085971689957,
      "grad_norm": 0.58203125,
      "learning_rate": 2.371586704422084e-05,
      "loss": 0.3932281255722046,
      "num_input_tokens_seen": 9796792448,
      "step": 34450,
      "train_runtime": 334937.395,
      "train_tokens_per_second": 29249.623
    },
    {
      "epoch": 1.2192624174327245,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3713199732678183e-05,
      "loss": 0.3898531436920166,
      "num_input_tokens_seen": 9799579008,
      "step": 34460,
      "train_runtime": 335030.9717,
      "train_tokens_per_second": 29249.77
    },
    {
      "epoch": 1.2196162376964532,
      "grad_norm": 0.58203125,
      "learning_rate": 2.371053332090666e-05,
      "loss": 0.3909220457077026,
      "num_input_tokens_seen": 9802470912,
      "step": 34470,
      "train_runtime": 335131.0845,
      "train_tokens_per_second": 29249.662
    },
    {
      "epoch": 1.219970057960182,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3707867808400527e-05,
      "loss": 0.391217565536499,
      "num_input_tokens_seen": 9805286080,
      "step": 34480,
      "train_runtime": 335225.352,
      "train_tokens_per_second": 29249.835
    },
    {
      "epoch": 1.2203238782239108,
      "grad_norm": 0.62890625,
      "learning_rate": 2.370520319465442e-05,
      "loss": 0.39152050018310547,
      "num_input_tokens_seen": 9808145920,
      "step": 34490,
      "train_runtime": 335323.4415,
      "train_tokens_per_second": 29249.807
    },
    {
      "epoch": 1.2206776984876395,
      "grad_norm": 0.609375,
      "learning_rate": 2.370253947916337e-05,
      "loss": 0.38847715854644777,
      "num_input_tokens_seen": 9810961536,
      "step": 34500,
      "train_runtime": 335423.0923,
      "train_tokens_per_second": 29249.511
    },
    {
      "epoch": 1.2210315187513683,
      "grad_norm": 0.56640625,
      "learning_rate": 2.369987666142282e-05,
      "loss": 0.38788018226623533,
      "num_input_tokens_seen": 9813835456,
      "step": 34510,
      "train_runtime": 335522.217,
      "train_tokens_per_second": 29249.436
    },
    {
      "epoch": 1.221385339015097,
      "grad_norm": 0.609375,
      "learning_rate": 2.36972147409286e-05,
      "loss": 0.38880462646484376,
      "num_input_tokens_seen": 9816663680,
      "step": 34520,
      "train_runtime": 335618.0042,
      "train_tokens_per_second": 29249.515
    },
    {
      "epoch": 1.2217391592788258,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3694553717176945e-05,
      "loss": 0.38937208652496336,
      "num_input_tokens_seen": 9819550080,
      "step": 34530,
      "train_runtime": 335719.9094,
      "train_tokens_per_second": 29249.234
    },
    {
      "epoch": 1.2220929795425546,
      "grad_norm": 0.6015625,
      "learning_rate": 2.369189358966447e-05,
      "loss": 0.391521692276001,
      "num_input_tokens_seen": 9822366336,
      "step": 34540,
      "train_runtime": 335815.4327,
      "train_tokens_per_second": 29249.3
    },
    {
      "epoch": 1.2224467998062833,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3689234357888205e-05,
      "loss": 0.3923304557800293,
      "num_input_tokens_seen": 9825223488,
      "step": 34550,
      "train_runtime": 335913.1046,
      "train_tokens_per_second": 29249.301
    },
    {
      "epoch": 1.2228006200700121,
      "grad_norm": 0.5703125,
      "learning_rate": 2.368657602134556e-05,
      "loss": 0.38765337467193606,
      "num_input_tokens_seen": 9828064064,
      "step": 34560,
      "train_runtime": 336008.0003,
      "train_tokens_per_second": 29249.494
    },
    {
      "epoch": 1.2231544403337409,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3683918579534347e-05,
      "loss": 0.39066851139068604,
      "num_input_tokens_seen": 9830890048,
      "step": 34570,
      "train_runtime": 336105.5677,
      "train_tokens_per_second": 29249.411
    },
    {
      "epoch": 1.2235082605974696,
      "grad_norm": 0.59375,
      "learning_rate": 2.368126203195277e-05,
      "loss": 0.38983473777770994,
      "num_input_tokens_seen": 9833731520,
      "step": 34580,
      "train_runtime": 336202.7445,
      "train_tokens_per_second": 29249.409
    },
    {
      "epoch": 1.2238620808611986,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3678606378099428e-05,
      "loss": 0.38753561973571776,
      "num_input_tokens_seen": 9836584960,
      "step": 34590,
      "train_runtime": 336303.4783,
      "train_tokens_per_second": 29249.132
    },
    {
      "epoch": 1.2242159011249274,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3675951617473315e-05,
      "loss": 0.3899605989456177,
      "num_input_tokens_seen": 9839436032,
      "step": 34600,
      "train_runtime": 336400.8276,
      "train_tokens_per_second": 29249.143
    },
    {
      "epoch": 1.2245697213886562,
      "grad_norm": 0.58203125,
      "learning_rate": 2.3673297749573805e-05,
      "loss": 0.39180185794830324,
      "num_input_tokens_seen": 9842257152,
      "step": 34610,
      "train_runtime": 336496.2138,
      "train_tokens_per_second": 29249.236
    },
    {
      "epoch": 1.224923541652385,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3670644773900692e-05,
      "loss": 0.3895064353942871,
      "num_input_tokens_seen": 9845072128,
      "step": 34620,
      "train_runtime": 336592.198,
      "train_tokens_per_second": 29249.258
    },
    {
      "epoch": 1.2252773619161137,
      "grad_norm": 0.625,
      "learning_rate": 2.3667992689954135e-05,
      "loss": 0.39354586601257324,
      "num_input_tokens_seen": 9847911808,
      "step": 34630,
      "train_runtime": 336688.1621,
      "train_tokens_per_second": 29249.356
    },
    {
      "epoch": 1.2256311821798425,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3665341497234704e-05,
      "loss": 0.38932151794433595,
      "num_input_tokens_seen": 9850730624,
      "step": 34640,
      "train_runtime": 336785.3759,
      "train_tokens_per_second": 29249.283
    },
    {
      "epoch": 1.2259850024435712,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3662691195243346e-05,
      "loss": 0.39045109748840334,
      "num_input_tokens_seen": 9853537408,
      "step": 34650,
      "train_runtime": 336879.5923,
      "train_tokens_per_second": 29249.434
    },
    {
      "epoch": 1.2263388227073,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3660041783481403e-05,
      "loss": 0.3947850227355957,
      "num_input_tokens_seen": 9856290560,
      "step": 34660,
      "train_runtime": 336968.9291,
      "train_tokens_per_second": 29249.85
    },
    {
      "epoch": 1.2266926429710288,
      "grad_norm": 0.59375,
      "learning_rate": 2.365739326145062e-05,
      "loss": 0.39142184257507323,
      "num_input_tokens_seen": 9859116416,
      "step": 34670,
      "train_runtime": 337065.8149,
      "train_tokens_per_second": 29249.826
    },
    {
      "epoch": 1.2270464632347575,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3654745628653117e-05,
      "loss": 0.3880918025970459,
      "num_input_tokens_seen": 9861939776,
      "step": 34680,
      "train_runtime": 337162.894,
      "train_tokens_per_second": 29249.778
    },
    {
      "epoch": 1.2274002834984863,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3652098884591413e-05,
      "loss": 0.38976516723632815,
      "num_input_tokens_seen": 9864776704,
      "step": 34690,
      "train_runtime": 337259.5001,
      "train_tokens_per_second": 29249.811
    },
    {
      "epoch": 1.227754103762215,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3649453028768406e-05,
      "loss": 0.3898470401763916,
      "num_input_tokens_seen": 9867605056,
      "step": 34700,
      "train_runtime": 337353.4724,
      "train_tokens_per_second": 29250.047
    },
    {
      "epoch": 1.2281079240259438,
      "grad_norm": 0.609375,
      "learning_rate": 2.3646808060687397e-05,
      "loss": 0.38961086273193357,
      "num_input_tokens_seen": 9870411776,
      "step": 34710,
      "train_runtime": 337446.9877,
      "train_tokens_per_second": 29250.259
    },
    {
      "epoch": 1.2284617442896726,
      "grad_norm": 0.609375,
      "learning_rate": 2.364416397985207e-05,
      "loss": 0.39345560073852537,
      "num_input_tokens_seen": 9873246208,
      "step": 34720,
      "train_runtime": 337542.4438,
      "train_tokens_per_second": 29250.384
    },
    {
      "epoch": 1.2288155645534014,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3641520785766487e-05,
      "loss": 0.3879849433898926,
      "num_input_tokens_seen": 9876161920,
      "step": 34730,
      "train_runtime": 337644.6231,
      "train_tokens_per_second": 29250.168
    },
    {
      "epoch": 1.2291693848171301,
      "grad_norm": 0.6015625,
      "learning_rate": 2.363887847793512e-05,
      "loss": 0.3916689157485962,
      "num_input_tokens_seen": 9878980416,
      "step": 34740,
      "train_runtime": 337740.7509,
      "train_tokens_per_second": 29250.188
    },
    {
      "epoch": 1.2295232050808589,
      "grad_norm": 0.61328125,
      "learning_rate": 2.363623705586281e-05,
      "loss": 0.39472362995147703,
      "num_input_tokens_seen": 9881828992,
      "step": 34750,
      "train_runtime": 337839.8374,
      "train_tokens_per_second": 29250.041
    },
    {
      "epoch": 1.2298770253445879,
      "grad_norm": 0.60546875,
      "learning_rate": 2.363359651905479e-05,
      "loss": 0.3958576679229736,
      "num_input_tokens_seen": 9884673792,
      "step": 34760,
      "train_runtime": 337936.5957,
      "train_tokens_per_second": 29250.084
    },
    {
      "epoch": 1.2302308456083166,
      "grad_norm": 0.609375,
      "learning_rate": 2.3630956867016688e-05,
      "loss": 0.38894104957580566,
      "num_input_tokens_seen": 9887490688,
      "step": 34770,
      "train_runtime": 338031.4344,
      "train_tokens_per_second": 29250.211
    },
    {
      "epoch": 1.2305846658720454,
      "grad_norm": 0.609375,
      "learning_rate": 2.3628318099254506e-05,
      "loss": 0.38692293167114256,
      "num_input_tokens_seen": 9890319040,
      "step": 34780,
      "train_runtime": 338125.6299,
      "train_tokens_per_second": 29250.427
    },
    {
      "epoch": 1.2309384861357742,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3625680215274635e-05,
      "loss": 0.38958425521850587,
      "num_input_tokens_seen": 9893143104,
      "step": 34790,
      "train_runtime": 338223.8932,
      "train_tokens_per_second": 29250.279
    },
    {
      "epoch": 1.231292306399503,
      "grad_norm": 0.60546875,
      "learning_rate": 2.362304321458386e-05,
      "loss": 0.3958865165710449,
      "num_input_tokens_seen": 9896034752,
      "step": 34800,
      "train_runtime": 338325.475,
      "train_tokens_per_second": 29250.043
    },
    {
      "epoch": 1.2316461266632317,
      "grad_norm": 0.578125,
      "learning_rate": 2.362040709668934e-05,
      "loss": 0.3878112554550171,
      "num_input_tokens_seen": 9898863040,
      "step": 34810,
      "train_runtime": 338422.8142,
      "train_tokens_per_second": 29249.987
    },
    {
      "epoch": 1.2319999469269605,
      "grad_norm": 0.578125,
      "learning_rate": 2.361777186109863e-05,
      "loss": 0.3871556997299194,
      "num_input_tokens_seen": 9901685696,
      "step": 34820,
      "train_runtime": 338517.6135,
      "train_tokens_per_second": 29250.134
    },
    {
      "epoch": 1.2323537671906892,
      "grad_norm": 0.59375,
      "learning_rate": 2.3615137507319672e-05,
      "loss": 0.39195685386657714,
      "num_input_tokens_seen": 9904598208,
      "step": 34830,
      "train_runtime": 338619.4897,
      "train_tokens_per_second": 29249.935
    },
    {
      "epoch": 1.232707587454418,
      "grad_norm": 0.6171875,
      "learning_rate": 2.361250403486076e-05,
      "loss": 0.3917252540588379,
      "num_input_tokens_seen": 9907452864,
      "step": 34840,
      "train_runtime": 338717.0262,
      "train_tokens_per_second": 29249.941
    },
    {
      "epoch": 1.2330614077181468,
      "grad_norm": 0.609375,
      "learning_rate": 2.360987144323062e-05,
      "loss": 0.38925833702087403,
      "num_input_tokens_seen": 9910414720,
      "step": 34850,
      "train_runtime": 338819.5251,
      "train_tokens_per_second": 29249.834
    },
    {
      "epoch": 1.2334152279818755,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3607239731938324e-05,
      "loss": 0.3893608570098877,
      "num_input_tokens_seen": 9913257856,
      "step": 34860,
      "train_runtime": 338917.6403,
      "train_tokens_per_second": 29249.755
    },
    {
      "epoch": 1.2337690482456043,
      "grad_norm": 0.65234375,
      "learning_rate": 2.3604608900493343e-05,
      "loss": 0.3903757095336914,
      "num_input_tokens_seen": 9916116224,
      "step": 34870,
      "train_runtime": 339016.155,
      "train_tokens_per_second": 29249.686
    },
    {
      "epoch": 1.234122868509333,
      "grad_norm": 0.62109375,
      "learning_rate": 2.360197894840553e-05,
      "loss": 0.39066381454467775,
      "num_input_tokens_seen": 9918965312,
      "step": 34880,
      "train_runtime": 339113.0984,
      "train_tokens_per_second": 29249.726
    },
    {
      "epoch": 1.2344766887730618,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3599349875185117e-05,
      "loss": 0.3920109272003174,
      "num_input_tokens_seen": 9921834944,
      "step": 34890,
      "train_runtime": 339211.2045,
      "train_tokens_per_second": 29249.726
    },
    {
      "epoch": 1.2348305090367906,
      "grad_norm": 0.58203125,
      "learning_rate": 2.359672168034272e-05,
      "loss": 0.391188383102417,
      "num_input_tokens_seen": 9924674240,
      "step": 34900,
      "train_runtime": 339310.2829,
      "train_tokens_per_second": 29249.553
    },
    {
      "epoch": 1.2351843293005194,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3594094363389327e-05,
      "loss": 0.3887295722961426,
      "num_input_tokens_seen": 9927486080,
      "step": 34910,
      "train_runtime": 339404.4318,
      "train_tokens_per_second": 29249.724
    },
    {
      "epoch": 1.2355381495642481,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3591467923836324e-05,
      "loss": 0.39623570442199707,
      "num_input_tokens_seen": 9930325312,
      "step": 34920,
      "train_runtime": 339504.7869,
      "train_tokens_per_second": 29249.441
    },
    {
      "epoch": 1.235891969827977,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3588842361195467e-05,
      "loss": 0.3905511379241943,
      "num_input_tokens_seen": 9933175360,
      "step": 34930,
      "train_runtime": 339601.7731,
      "train_tokens_per_second": 29249.48
    },
    {
      "epoch": 1.2362457900917059,
      "grad_norm": 0.578125,
      "learning_rate": 2.3586217674978893e-05,
      "loss": 0.39021623134613037,
      "num_input_tokens_seen": 9936098560,
      "step": 34940,
      "train_runtime": 339703.6504,
      "train_tokens_per_second": 29249.313
    },
    {
      "epoch": 1.2365996103554346,
      "grad_norm": 0.60546875,
      "learning_rate": 2.358359386469912e-05,
      "loss": 0.38912899494171144,
      "num_input_tokens_seen": 9938961792,
      "step": 34950,
      "train_runtime": 339801.4642,
      "train_tokens_per_second": 29249.32
    },
    {
      "epoch": 1.2369534306191634,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3580970929869047e-05,
      "loss": 0.3933791875839233,
      "num_input_tokens_seen": 9941761600,
      "step": 34960,
      "train_runtime": 339894.0911,
      "train_tokens_per_second": 29249.586
    },
    {
      "epoch": 1.2373072508828922,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3578348870001952e-05,
      "loss": 0.3955677032470703,
      "num_input_tokens_seen": 9944624256,
      "step": 34970,
      "train_runtime": 339993.3756,
      "train_tokens_per_second": 29249.465
    },
    {
      "epoch": 1.237661071146621,
      "grad_norm": 0.60546875,
      "learning_rate": 2.357572768461149e-05,
      "loss": 0.38793082237243653,
      "num_input_tokens_seen": 9947495936,
      "step": 34980,
      "train_runtime": 340093.9304,
      "train_tokens_per_second": 29249.26
    },
    {
      "epoch": 1.2380148914103497,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3573107373211687e-05,
      "loss": 0.39339404106140136,
      "num_input_tokens_seen": 9950286144,
      "step": 34990,
      "train_runtime": 340184.6625,
      "train_tokens_per_second": 29249.661
    },
    {
      "epoch": 1.2383687116740785,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3570487935316964e-05,
      "loss": 0.3933719158172607,
      "num_input_tokens_seen": 9953103872,
      "step": 35000,
      "train_runtime": 340281.5165,
      "train_tokens_per_second": 29249.617
    },
    {
      "epoch": 1.2383687116740785,
      "eval_loss": 0.33749818801879883,
      "eval_runtime": 8.4167,
      "eval_samples_per_second": 473.818,
      "eval_steps_per_second": 3.802,
      "num_input_tokens_seen": 9953103872,
      "step": 35000
    },
    {
      "epoch": 1.2387225319378072,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3567869370442104e-05,
      "loss": 0.38792977333068845,
      "num_input_tokens_seen": 9955961216,
      "step": 35010,
      "train_runtime": 340411.1001,
      "train_tokens_per_second": 29246.876
    },
    {
      "epoch": 1.239076352201536,
      "grad_norm": 0.5859375,
      "learning_rate": 2.356525167810228e-05,
      "loss": 0.39495229721069336,
      "num_input_tokens_seen": 9958810304,
      "step": 35020,
      "train_runtime": 340511.7608,
      "train_tokens_per_second": 29246.597
    },
    {
      "epoch": 1.2394301724652648,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3562634857813034e-05,
      "loss": 0.39720590114593507,
      "num_input_tokens_seen": 9961596352,
      "step": 35030,
      "train_runtime": 340606.3562,
      "train_tokens_per_second": 29246.654
    },
    {
      "epoch": 1.2397839927289935,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3560018909090288e-05,
      "loss": 0.3956779956817627,
      "num_input_tokens_seen": 9964467392,
      "step": 35040,
      "train_runtime": 340705.0098,
      "train_tokens_per_second": 29246.612
    },
    {
      "epoch": 1.2401378129927223,
      "grad_norm": 0.6171875,
      "learning_rate": 2.355740383145033e-05,
      "loss": 0.391188383102417,
      "num_input_tokens_seen": 9967367488,
      "step": 35050,
      "train_runtime": 340805.7111,
      "train_tokens_per_second": 29246.48
    },
    {
      "epoch": 1.240491633256451,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3554789624409836e-05,
      "loss": 0.3957855701446533,
      "num_input_tokens_seen": 9970211264,
      "step": 35060,
      "train_runtime": 340901.7307,
      "train_tokens_per_second": 29246.584
    },
    {
      "epoch": 1.2408454535201798,
      "grad_norm": 0.58984375,
      "learning_rate": 2.355217628748586e-05,
      "loss": 0.38939361572265624,
      "num_input_tokens_seen": 9973072000,
      "step": 35070,
      "train_runtime": 340997.8008,
      "train_tokens_per_second": 29246.734
    },
    {
      "epoch": 1.2411992737839088,
      "grad_norm": 0.61328125,
      "learning_rate": 2.354956382019581e-05,
      "loss": 0.39279704093933104,
      "num_input_tokens_seen": 9975927488,
      "step": 35080,
      "train_runtime": 341094.3148,
      "train_tokens_per_second": 29246.83
    },
    {
      "epoch": 1.2415530940476374,
      "grad_norm": 0.59765625,
      "learning_rate": 2.35469522220575e-05,
      "loss": 0.3957968235015869,
      "num_input_tokens_seen": 9978769408,
      "step": 35090,
      "train_runtime": 341188.578,
      "train_tokens_per_second": 29247.079
    },
    {
      "epoch": 1.2419069143113664,
      "grad_norm": 0.625,
      "learning_rate": 2.3544341492589096e-05,
      "loss": 0.395880389213562,
      "num_input_tokens_seen": 9981646080,
      "step": 35100,
      "train_runtime": 341288.0979,
      "train_tokens_per_second": 29246.98
    },
    {
      "epoch": 1.2422607345750951,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3541731631309126e-05,
      "loss": 0.38951592445373534,
      "num_input_tokens_seen": 9984491392,
      "step": 35110,
      "train_runtime": 341386.188,
      "train_tokens_per_second": 29246.911
    },
    {
      "epoch": 1.2426145548388239,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3539122637736537e-05,
      "loss": 0.38961329460144045,
      "num_input_tokens_seen": 9987369216,
      "step": 35120,
      "train_runtime": 341484.2428,
      "train_tokens_per_second": 29246.94
    },
    {
      "epoch": 1.2429683751025526,
      "grad_norm": 0.58984375,
      "learning_rate": 2.35365145113906e-05,
      "loss": 0.3884893417358398,
      "num_input_tokens_seen": 9990212224,
      "step": 35130,
      "train_runtime": 341580.9817,
      "train_tokens_per_second": 29246.98
    },
    {
      "epoch": 1.2433221953662814,
      "grad_norm": 0.609375,
      "learning_rate": 2.3533907251790983e-05,
      "loss": 0.3930696964263916,
      "num_input_tokens_seen": 9993065216,
      "step": 35140,
      "train_runtime": 341679.9127,
      "train_tokens_per_second": 29246.862
    },
    {
      "epoch": 1.2436760156300102,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3531300858457727e-05,
      "loss": 0.3917020082473755,
      "num_input_tokens_seen": 9995957056,
      "step": 35150,
      "train_runtime": 341782.3056,
      "train_tokens_per_second": 29246.561
    },
    {
      "epoch": 1.244029835893739,
      "grad_norm": 0.609375,
      "learning_rate": 2.352869533091124e-05,
      "loss": 0.3884159564971924,
      "num_input_tokens_seen": 9998854080,
      "step": 35160,
      "train_runtime": 341883.0793,
      "train_tokens_per_second": 29246.414
    },
    {
      "epoch": 1.2443836561574677,
      "grad_norm": 0.609375,
      "learning_rate": 2.3526090668672306e-05,
      "loss": 0.39250574111938474,
      "num_input_tokens_seen": 10001740096,
      "step": 35170,
      "train_runtime": 341984.2574,
      "train_tokens_per_second": 29246.2
    },
    {
      "epoch": 1.2447374764211965,
      "grad_norm": 0.59375,
      "learning_rate": 2.3523486871262068e-05,
      "loss": 0.3898512363433838,
      "num_input_tokens_seen": 10004669440,
      "step": 35180,
      "train_runtime": 342084.8514,
      "train_tokens_per_second": 29246.163
    },
    {
      "epoch": 1.2450912966849252,
      "grad_norm": 0.59375,
      "learning_rate": 2.3520883938202048e-05,
      "loss": 0.3902533292770386,
      "num_input_tokens_seen": 10007498176,
      "step": 35190,
      "train_runtime": 342181.3145,
      "train_tokens_per_second": 29246.185
    },
    {
      "epoch": 1.245445116948654,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3518281869014153e-05,
      "loss": 0.3891111373901367,
      "num_input_tokens_seen": 10010338560,
      "step": 35200,
      "train_runtime": 342276.5217,
      "train_tokens_per_second": 29246.349
    },
    {
      "epoch": 1.2457989372123828,
      "grad_norm": 0.609375,
      "learning_rate": 2.3515680663220632e-05,
      "loss": 0.39291832447052,
      "num_input_tokens_seen": 10013225984,
      "step": 35210,
      "train_runtime": 342378.8767,
      "train_tokens_per_second": 29246.039
    },
    {
      "epoch": 1.2461527574761115,
      "grad_norm": 0.58984375,
      "learning_rate": 2.351308032034413e-05,
      "loss": 0.3892541408538818,
      "num_input_tokens_seen": 10016053440,
      "step": 35220,
      "train_runtime": 342474.8532,
      "train_tokens_per_second": 29246.099
    },
    {
      "epoch": 1.2465065777398403,
      "grad_norm": 0.609375,
      "learning_rate": 2.3510480839907635e-05,
      "loss": 0.39425790309906006,
      "num_input_tokens_seen": 10018930112,
      "step": 35230,
      "train_runtime": 342576.108,
      "train_tokens_per_second": 29245.852
    },
    {
      "epoch": 1.246860398003569,
      "grad_norm": 0.59375,
      "learning_rate": 2.350788222143453e-05,
      "loss": 0.39088053703308107,
      "num_input_tokens_seen": 10021760448,
      "step": 35240,
      "train_runtime": 342673.0866,
      "train_tokens_per_second": 29245.835
    },
    {
      "epoch": 1.247214218267298,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3505284464448558e-05,
      "loss": 0.3901395797729492,
      "num_input_tokens_seen": 10024582720,
      "step": 35250,
      "train_runtime": 342769.6239,
      "train_tokens_per_second": 29245.832
    },
    {
      "epoch": 1.2475680385310266,
      "grad_norm": 0.59375,
      "learning_rate": 2.3502687568473815e-05,
      "loss": 0.393476676940918,
      "num_input_tokens_seen": 10027407872,
      "step": 35260,
      "train_runtime": 342866.3879,
      "train_tokens_per_second": 29245.818
    },
    {
      "epoch": 1.2479218587947556,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3500091533034785e-05,
      "loss": 0.39229159355163573,
      "num_input_tokens_seen": 10030243520,
      "step": 35270,
      "train_runtime": 342963.2879,
      "train_tokens_per_second": 29245.823
    },
    {
      "epoch": 1.2482756790584844,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3497496357656313e-05,
      "loss": 0.3912361621856689,
      "num_input_tokens_seen": 10033080960,
      "step": 35280,
      "train_runtime": 343060.9269,
      "train_tokens_per_second": 29245.77
    },
    {
      "epoch": 1.2486294993222131,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3494902041863607e-05,
      "loss": 0.39301586151123047,
      "num_input_tokens_seen": 10035942016,
      "step": 35290,
      "train_runtime": 343157.4138,
      "train_tokens_per_second": 29245.884
    },
    {
      "epoch": 1.2489833195859419,
      "grad_norm": 0.60546875,
      "learning_rate": 2.349230858518225e-05,
      "loss": 0.3922626495361328,
      "num_input_tokens_seen": 10038792768,
      "step": 35300,
      "train_runtime": 343254.2584,
      "train_tokens_per_second": 29245.938
    },
    {
      "epoch": 1.2493371398496707,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3489715987138178e-05,
      "loss": 0.38910613059997556,
      "num_input_tokens_seen": 10041574528,
      "step": 35310,
      "train_runtime": 343348.9419,
      "train_tokens_per_second": 29245.975
    },
    {
      "epoch": 1.2496909601133994,
      "grad_norm": 0.62109375,
      "learning_rate": 2.348712424725771e-05,
      "loss": 0.3897295713424683,
      "num_input_tokens_seen": 10044415808,
      "step": 35320,
      "train_runtime": 343447.6273,
      "train_tokens_per_second": 29245.844
    },
    {
      "epoch": 1.2500447803771282,
      "grad_norm": 0.59375,
      "learning_rate": 2.3484533365067526e-05,
      "loss": 0.39483957290649413,
      "num_input_tokens_seen": 10047284864,
      "step": 35330,
      "train_runtime": 343549.4455,
      "train_tokens_per_second": 29245.528
    },
    {
      "epoch": 1.250398600640857,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3481943340094656e-05,
      "loss": 0.38970365524291994,
      "num_input_tokens_seen": 10050130944,
      "step": 35340,
      "train_runtime": 343645.337,
      "train_tokens_per_second": 29245.649
    },
    {
      "epoch": 1.2507524209045857,
      "grad_norm": 0.609375,
      "learning_rate": 2.3479354171866517e-05,
      "loss": 0.3914132833480835,
      "num_input_tokens_seen": 10052974080,
      "step": 35350,
      "train_runtime": 343741.368,
      "train_tokens_per_second": 29245.75
    },
    {
      "epoch": 1.2511062411683145,
      "grad_norm": 0.58203125,
      "learning_rate": 2.3476765859910877e-05,
      "loss": 0.39165487289428713,
      "num_input_tokens_seen": 10055788288,
      "step": 35360,
      "train_runtime": 343835.5828,
      "train_tokens_per_second": 29245.921
    },
    {
      "epoch": 1.2514600614320432,
      "grad_norm": 0.609375,
      "learning_rate": 2.347417840375587e-05,
      "loss": 0.3855934143066406,
      "num_input_tokens_seen": 10058621248,
      "step": 35370,
      "train_runtime": 343933.5432,
      "train_tokens_per_second": 29245.828
    },
    {
      "epoch": 1.251813881695772,
      "grad_norm": 0.6015625,
      "learning_rate": 2.347159180293e-05,
      "loss": 0.3871392965316772,
      "num_input_tokens_seen": 10061456640,
      "step": 35380,
      "train_runtime": 344028.9755,
      "train_tokens_per_second": 29245.957
    },
    {
      "epoch": 1.2521677019595008,
      "grad_norm": 0.58984375,
      "learning_rate": 2.346900605696213e-05,
      "loss": 0.3917234897613525,
      "num_input_tokens_seen": 10064292736,
      "step": 35390,
      "train_runtime": 344123.7692,
      "train_tokens_per_second": 29246.142
    },
    {
      "epoch": 1.2525215222232295,
      "grad_norm": 0.58203125,
      "learning_rate": 2.3466421165381486e-05,
      "loss": 0.38956403732299805,
      "num_input_tokens_seen": 10067125568,
      "step": 35400,
      "train_runtime": 344219.9752,
      "train_tokens_per_second": 29246.198
    },
    {
      "epoch": 1.2528753424869583,
      "grad_norm": 0.6015625,
      "learning_rate": 2.346383712771766e-05,
      "loss": 0.39185781478881837,
      "num_input_tokens_seen": 10069973440,
      "step": 35410,
      "train_runtime": 344316.2755,
      "train_tokens_per_second": 29246.289
    },
    {
      "epoch": 1.2532291627506873,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3461253943500598e-05,
      "loss": 0.392056941986084,
      "num_input_tokens_seen": 10072860544,
      "step": 35420,
      "train_runtime": 344415.4029,
      "train_tokens_per_second": 29246.255
    },
    {
      "epoch": 1.2535829830144158,
      "grad_norm": 0.60546875,
      "learning_rate": 2.345867161226062e-05,
      "loss": 0.39003934860229494,
      "num_input_tokens_seen": 10075716096,
      "step": 35430,
      "train_runtime": 344513.0706,
      "train_tokens_per_second": 29246.252
    },
    {
      "epoch": 1.2539368032781448,
      "grad_norm": 0.5859375,
      "learning_rate": 2.34560901335284e-05,
      "loss": 0.38649041652679444,
      "num_input_tokens_seen": 10078532096,
      "step": 35440,
      "train_runtime": 344608.808,
      "train_tokens_per_second": 29246.299
    },
    {
      "epoch": 1.2542906235418736,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3453509506834984e-05,
      "loss": 0.3915275812149048,
      "num_input_tokens_seen": 10081347968,
      "step": 35450,
      "train_runtime": 344707.7571,
      "train_tokens_per_second": 29246.072
    },
    {
      "epoch": 1.2546444438056024,
      "grad_norm": 0.609375,
      "learning_rate": 2.345092973171176e-05,
      "loss": 0.38986341953277587,
      "num_input_tokens_seen": 10084213888,
      "step": 35460,
      "train_runtime": 344807.3865,
      "train_tokens_per_second": 29245.933
    },
    {
      "epoch": 1.2549982640693311,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3448350807690487e-05,
      "loss": 0.3932187080383301,
      "num_input_tokens_seen": 10087038336,
      "step": 35470,
      "train_runtime": 344903.9498,
      "train_tokens_per_second": 29245.935
    },
    {
      "epoch": 1.25535208433306,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3445772734303292e-05,
      "loss": 0.3870553016662598,
      "num_input_tokens_seen": 10089841024,
      "step": 35480,
      "train_runtime": 344999.2428,
      "train_tokens_per_second": 29245.98
    },
    {
      "epoch": 1.2557059045967887,
      "grad_norm": 0.609375,
      "learning_rate": 2.344319551108265e-05,
      "loss": 0.38880760669708253,
      "num_input_tokens_seen": 10092678848,
      "step": 35490,
      "train_runtime": 345091.8232,
      "train_tokens_per_second": 29246.358
    },
    {
      "epoch": 1.2560597248605174,
      "grad_norm": 0.578125,
      "learning_rate": 2.3440619137561402e-05,
      "loss": 0.38543813228607177,
      "num_input_tokens_seen": 10095558208,
      "step": 35500,
      "train_runtime": 345194.9715,
      "train_tokens_per_second": 29245.96
    },
    {
      "epoch": 1.2564135451242462,
      "grad_norm": 0.60546875,
      "learning_rate": 2.343804361327275e-05,
      "loss": 0.38905441761016846,
      "num_input_tokens_seen": 10098391104,
      "step": 35510,
      "train_runtime": 345289.871,
      "train_tokens_per_second": 29246.126
    },
    {
      "epoch": 1.256767365387975,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3435468937750252e-05,
      "loss": 0.3927861452102661,
      "num_input_tokens_seen": 10101243008,
      "step": 35520,
      "train_runtime": 345385.7238,
      "train_tokens_per_second": 29246.267
    },
    {
      "epoch": 1.2571211856517037,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3432895110527815e-05,
      "loss": 0.392937970161438,
      "num_input_tokens_seen": 10104065792,
      "step": 35530,
      "train_runtime": 345481.266,
      "train_tokens_per_second": 29246.349
    },
    {
      "epoch": 1.2574750059154325,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3430322131139726e-05,
      "loss": 0.38762712478637695,
      "num_input_tokens_seen": 10106898368,
      "step": 35540,
      "train_runtime": 345579.5951,
      "train_tokens_per_second": 29246.224
    },
    {
      "epoch": 1.2578288261791613,
      "grad_norm": 0.6015625,
      "learning_rate": 2.342774999912061e-05,
      "loss": 0.39256060123443604,
      "num_input_tokens_seen": 10109664512,
      "step": 35550,
      "train_runtime": 345670.996,
      "train_tokens_per_second": 29246.493
    },
    {
      "epoch": 1.25818264644289,
      "grad_norm": 0.578125,
      "learning_rate": 2.342517871400546e-05,
      "loss": 0.39326491355896,
      "num_input_tokens_seen": 10112520512,
      "step": 35560,
      "train_runtime": 345768.1898,
      "train_tokens_per_second": 29246.532
    },
    {
      "epoch": 1.258536466706619,
      "grad_norm": 0.5859375,
      "learning_rate": 2.342260827532962e-05,
      "loss": 0.3897913932800293,
      "num_input_tokens_seen": 10115328000,
      "step": 35570,
      "train_runtime": 345865.7365,
      "train_tokens_per_second": 29246.401
    },
    {
      "epoch": 1.2588902869703475,
      "grad_norm": 0.6015625,
      "learning_rate": 2.34200386826288e-05,
      "loss": 0.389065146446228,
      "num_input_tokens_seen": 10118209472,
      "step": 35580,
      "train_runtime": 345964.5616,
      "train_tokens_per_second": 29246.375
    },
    {
      "epoch": 1.2592441072340765,
      "grad_norm": 0.6328125,
      "learning_rate": 2.3417469935439052e-05,
      "loss": 0.3881157398223877,
      "num_input_tokens_seen": 10121055488,
      "step": 35590,
      "train_runtime": 346062.4692,
      "train_tokens_per_second": 29246.325
    },
    {
      "epoch": 1.259597927497805,
      "grad_norm": 0.59765625,
      "learning_rate": 2.34149020332968e-05,
      "loss": 0.3917097568511963,
      "num_input_tokens_seen": 10123917824,
      "step": 35600,
      "train_runtime": 346161.0508,
      "train_tokens_per_second": 29246.265
    },
    {
      "epoch": 1.259951747761534,
      "grad_norm": 0.59375,
      "learning_rate": 2.3412334975738813e-05,
      "loss": 0.394685959815979,
      "num_input_tokens_seen": 10126750016,
      "step": 35610,
      "train_runtime": 346258.6902,
      "train_tokens_per_second": 29246.197
    },
    {
      "epoch": 1.2603055680252628,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3409768762302222e-05,
      "loss": 0.3866231918334961,
      "num_input_tokens_seen": 10129580608,
      "step": 35620,
      "train_runtime": 346356.9092,
      "train_tokens_per_second": 29246.076
    },
    {
      "epoch": 1.2606593882889916,
      "grad_norm": 0.609375,
      "learning_rate": 2.340720339252451e-05,
      "loss": 0.3914416074752808,
      "num_input_tokens_seen": 10132420992,
      "step": 35630,
      "train_runtime": 346454.3603,
      "train_tokens_per_second": 29246.048
    },
    {
      "epoch": 1.2610132085527204,
      "grad_norm": 0.58203125,
      "learning_rate": 2.3404638865943507e-05,
      "loss": 0.3888051509857178,
      "num_input_tokens_seen": 10135264512,
      "step": 35640,
      "train_runtime": 346549.7353,
      "train_tokens_per_second": 29246.205
    },
    {
      "epoch": 1.2613670288164491,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3402075182097412e-05,
      "loss": 0.3925264596939087,
      "num_input_tokens_seen": 10138134912,
      "step": 35650,
      "train_runtime": 346649.717,
      "train_tokens_per_second": 29246.05
    },
    {
      "epoch": 1.261720849080178,
      "grad_norm": 0.6015625,
      "learning_rate": 2.339951234052477e-05,
      "loss": 0.38530669212341306,
      "num_input_tokens_seen": 10140988224,
      "step": 35660,
      "train_runtime": 346749.2634,
      "train_tokens_per_second": 29245.883
    },
    {
      "epoch": 1.2620746693439067,
      "grad_norm": 0.62890625,
      "learning_rate": 2.3396950340764478e-05,
      "loss": 0.3853438377380371,
      "num_input_tokens_seen": 10143858304,
      "step": 35670,
      "train_runtime": 346850.5364,
      "train_tokens_per_second": 29245.618
    },
    {
      "epoch": 1.2624284896076354,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3394389182355793e-05,
      "loss": 0.39094090461730957,
      "num_input_tokens_seen": 10146706752,
      "step": 35680,
      "train_runtime": 346949.634,
      "train_tokens_per_second": 29245.475
    },
    {
      "epoch": 1.2627823098713642,
      "grad_norm": 0.6171875,
      "learning_rate": 2.339182886483832e-05,
      "loss": 0.391913628578186,
      "num_input_tokens_seen": 10149506048,
      "step": 35690,
      "train_runtime": 347043.6888,
      "train_tokens_per_second": 29245.615
    },
    {
      "epoch": 1.263136130135093,
      "grad_norm": 0.58203125,
      "learning_rate": 2.3389269387752017e-05,
      "loss": 0.38967850208282473,
      "num_input_tokens_seen": 10152397888,
      "step": 35700,
      "train_runtime": 347143.1268,
      "train_tokens_per_second": 29245.568
    },
    {
      "epoch": 1.2634899503988217,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3386710750637194e-05,
      "loss": 0.39403934478759767,
      "num_input_tokens_seen": 10155230208,
      "step": 35710,
      "train_runtime": 347240.178,
      "train_tokens_per_second": 29245.551
    },
    {
      "epoch": 1.2638437706625505,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3384152953034514e-05,
      "loss": 0.3918935298919678,
      "num_input_tokens_seen": 10158070912,
      "step": 35720,
      "train_runtime": 347338.8769,
      "train_tokens_per_second": 29245.419
    },
    {
      "epoch": 1.2641975909262793,
      "grad_norm": 0.6015625,
      "learning_rate": 2.338159599448499e-05,
      "loss": 0.3861706733703613,
      "num_input_tokens_seen": 10160917056,
      "step": 35730,
      "train_runtime": 347436.0195,
      "train_tokens_per_second": 29245.434
    },
    {
      "epoch": 1.2645514111900082,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3379039874529996e-05,
      "loss": 0.3889310359954834,
      "num_input_tokens_seen": 10163797952,
      "step": 35740,
      "train_runtime": 347534.1027,
      "train_tokens_per_second": 29245.469
    },
    {
      "epoch": 1.2649052314537368,
      "grad_norm": 0.625,
      "learning_rate": 2.337648459271124e-05,
      "loss": 0.39087882041931155,
      "num_input_tokens_seen": 10166626048,
      "step": 35750,
      "train_runtime": 347631.708,
      "train_tokens_per_second": 29245.393
    },
    {
      "epoch": 1.2652590517174658,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3373930148570787e-05,
      "loss": 0.3912909984588623,
      "num_input_tokens_seen": 10169436416,
      "step": 35760,
      "train_runtime": 347728.0864,
      "train_tokens_per_second": 29245.37
    },
    {
      "epoch": 1.2656128719811943,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3371376541651063e-05,
      "loss": 0.3960239887237549,
      "num_input_tokens_seen": 10172264768,
      "step": 35770,
      "train_runtime": 347823.1062,
      "train_tokens_per_second": 29245.512
    },
    {
      "epoch": 1.2659666922449233,
      "grad_norm": 0.6484375,
      "learning_rate": 2.336882377149483e-05,
      "loss": 0.3915431499481201,
      "num_input_tokens_seen": 10175120896,
      "step": 35780,
      "train_runtime": 347921.1709,
      "train_tokens_per_second": 29245.478
    },
    {
      "epoch": 1.266320512508652,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3366271837645207e-05,
      "loss": 0.39311091899871825,
      "num_input_tokens_seen": 10178012672,
      "step": 35790,
      "train_runtime": 348019.6358,
      "train_tokens_per_second": 29245.513
    },
    {
      "epoch": 1.2666743327723808,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3363720739645652e-05,
      "loss": 0.39456839561462403,
      "num_input_tokens_seen": 10180854720,
      "step": 35800,
      "train_runtime": 348116.6634,
      "train_tokens_per_second": 29245.525
    },
    {
      "epoch": 1.2670281530361096,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3361170477039995e-05,
      "loss": 0.3854986190795898,
      "num_input_tokens_seen": 10183678464,
      "step": 35810,
      "train_runtime": 348211.9287,
      "train_tokens_per_second": 29245.634
    },
    {
      "epoch": 1.2673819732998384,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3358621049372383e-05,
      "loss": 0.3889935970306396,
      "num_input_tokens_seen": 10186474624,
      "step": 35820,
      "train_runtime": 348308.5498,
      "train_tokens_per_second": 29245.549
    },
    {
      "epoch": 1.2677357935635671,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3356072456187336e-05,
      "loss": 0.3913357019424438,
      "num_input_tokens_seen": 10189268608,
      "step": 35830,
      "train_runtime": 348401.8643,
      "train_tokens_per_second": 29245.735
    },
    {
      "epoch": 1.268089613827296,
      "grad_norm": 0.59765625,
      "learning_rate": 2.335352469702972e-05,
      "loss": 0.3917271137237549,
      "num_input_tokens_seen": 10192053568,
      "step": 35840,
      "train_runtime": 348496.9661,
      "train_tokens_per_second": 29245.745
    },
    {
      "epoch": 1.2684434340910247,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3350977771444724e-05,
      "loss": 0.38952884674072263,
      "num_input_tokens_seen": 10194905920,
      "step": 35850,
      "train_runtime": 348597.0177,
      "train_tokens_per_second": 29245.534
    },
    {
      "epoch": 1.2687972543547534,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3348431678977914e-05,
      "loss": 0.3941945552825928,
      "num_input_tokens_seen": 10197752832,
      "step": 35860,
      "train_runtime": 348693.8026,
      "train_tokens_per_second": 29245.581
    },
    {
      "epoch": 1.2691510746184822,
      "grad_norm": 0.58984375,
      "learning_rate": 2.334588641917519e-05,
      "loss": 0.3887597322463989,
      "num_input_tokens_seen": 10200570048,
      "step": 35870,
      "train_runtime": 348787.3616,
      "train_tokens_per_second": 29245.813
    },
    {
      "epoch": 1.269504894882211,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3343341991582792e-05,
      "loss": 0.3919421672821045,
      "num_input_tokens_seen": 10203438656,
      "step": 35880,
      "train_runtime": 348887.7515,
      "train_tokens_per_second": 29245.62
    },
    {
      "epoch": 1.2698587151459397,
      "grad_norm": 0.6015625,
      "learning_rate": 2.334079839574732e-05,
      "loss": 0.38967156410217285,
      "num_input_tokens_seen": 10206278592,
      "step": 35890,
      "train_runtime": 348982.7272,
      "train_tokens_per_second": 29245.799
    },
    {
      "epoch": 1.2702125354096685,
      "grad_norm": 0.58984375,
      "learning_rate": 2.333825563121571e-05,
      "loss": 0.39358320236206057,
      "num_input_tokens_seen": 10209164416,
      "step": 35900,
      "train_runtime": 349081.2782,
      "train_tokens_per_second": 29245.809
    },
    {
      "epoch": 1.2705663556733975,
      "grad_norm": 0.59375,
      "learning_rate": 2.333571369753525e-05,
      "loss": 0.3916492938995361,
      "num_input_tokens_seen": 10212010368,
      "step": 35910,
      "train_runtime": 349179.6844,
      "train_tokens_per_second": 29245.717
    },
    {
      "epoch": 1.270920175937126,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3333172594253564e-05,
      "loss": 0.39238781929016114,
      "num_input_tokens_seen": 10214850176,
      "step": 35920,
      "train_runtime": 349277.039,
      "train_tokens_per_second": 29245.696
    },
    {
      "epoch": 1.271273996200855,
      "grad_norm": 0.578125,
      "learning_rate": 2.3330632320918626e-05,
      "loss": 0.3926822662353516,
      "num_input_tokens_seen": 10217684864,
      "step": 35930,
      "train_runtime": 349374.2686,
      "train_tokens_per_second": 29245.671
    },
    {
      "epoch": 1.2716278164645836,
      "grad_norm": 0.62890625,
      "learning_rate": 2.3328092877078757e-05,
      "loss": 0.38837418556213377,
      "num_input_tokens_seen": 10220489984,
      "step": 35940,
      "train_runtime": 349469.6099,
      "train_tokens_per_second": 29245.719
    },
    {
      "epoch": 1.2719816367283125,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3325554262282617e-05,
      "loss": 0.39333467483520507,
      "num_input_tokens_seen": 10223374272,
      "step": 35950,
      "train_runtime": 349567.8213,
      "train_tokens_per_second": 29245.753
    },
    {
      "epoch": 1.2723354569920413,
      "grad_norm": 0.65234375,
      "learning_rate": 2.332301647607922e-05,
      "loss": 0.38719959259033204,
      "num_input_tokens_seen": 10226270016,
      "step": 35960,
      "train_runtime": 349667.1947,
      "train_tokens_per_second": 29245.723
    },
    {
      "epoch": 1.27268927725577,
      "grad_norm": 0.58203125,
      "learning_rate": 2.3320479518017907e-05,
      "loss": 0.38618924617767336,
      "num_input_tokens_seen": 10229135040,
      "step": 35970,
      "train_runtime": 349763.3216,
      "train_tokens_per_second": 29245.877
    },
    {
      "epoch": 1.2730430975194988,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3317943387648372e-05,
      "loss": 0.39114928245544434,
      "num_input_tokens_seen": 10232022784,
      "step": 35980,
      "train_runtime": 349866.146,
      "train_tokens_per_second": 29245.535
    },
    {
      "epoch": 1.2733969177832276,
      "grad_norm": 0.59375,
      "learning_rate": 2.3315408084520653e-05,
      "loss": 0.3932966232299805,
      "num_input_tokens_seen": 10234852992,
      "step": 35990,
      "train_runtime": 349963.5043,
      "train_tokens_per_second": 29245.487
    },
    {
      "epoch": 1.2737507380469564,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3312873608185128e-05,
      "loss": 0.38701791763305665,
      "num_input_tokens_seen": 10237701504,
      "step": 36000,
      "train_runtime": 350061.0874,
      "train_tokens_per_second": 29245.471
    },
    {
      "epoch": 1.2741045583106851,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3310339958192516e-05,
      "loss": 0.3945901393890381,
      "num_input_tokens_seen": 10240542208,
      "step": 36010,
      "train_runtime": 350158.2051,
      "train_tokens_per_second": 29245.473
    },
    {
      "epoch": 1.274458378574414,
      "grad_norm": 0.5625,
      "learning_rate": 2.3307807134093877e-05,
      "loss": 0.3894914388656616,
      "num_input_tokens_seen": 10243427712,
      "step": 36020,
      "train_runtime": 350258.1559,
      "train_tokens_per_second": 29245.365
    },
    {
      "epoch": 1.2748121988381427,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3305275135440616e-05,
      "loss": 0.3927293300628662,
      "num_input_tokens_seen": 10246277184,
      "step": 36030,
      "train_runtime": 350355.3845,
      "train_tokens_per_second": 29245.382
    },
    {
      "epoch": 1.2751660191018714,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3302743961784472e-05,
      "loss": 0.39076340198516846,
      "num_input_tokens_seen": 10249089600,
      "step": 36040,
      "train_runtime": 350450.8584,
      "train_tokens_per_second": 29245.44
    },
    {
      "epoch": 1.2755198393656002,
      "grad_norm": 0.59375,
      "learning_rate": 2.3300213612677545e-05,
      "loss": 0.38801989555358884,
      "num_input_tokens_seen": 10252022528,
      "step": 36050,
      "train_runtime": 350551.8324,
      "train_tokens_per_second": 29245.383
    },
    {
      "epoch": 1.275873659629329,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3297684087672242e-05,
      "loss": 0.39103810787200927,
      "num_input_tokens_seen": 10254893120,
      "step": 36060,
      "train_runtime": 350647.6531,
      "train_tokens_per_second": 29245.578
    },
    {
      "epoch": 1.2762274798930577,
      "grad_norm": 0.625,
      "learning_rate": 2.3295155386321336e-05,
      "loss": 0.39308369159698486,
      "num_input_tokens_seen": 10257762624,
      "step": 36070,
      "train_runtime": 350745.5495,
      "train_tokens_per_second": 29245.596
    },
    {
      "epoch": 1.2765813001567867,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3292627508177933e-05,
      "loss": 0.3874519824981689,
      "num_input_tokens_seen": 10260627968,
      "step": 36080,
      "train_runtime": 350841.7539,
      "train_tokens_per_second": 29245.744
    },
    {
      "epoch": 1.2769351204205153,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3290100452795474e-05,
      "loss": 0.39218294620513916,
      "num_input_tokens_seen": 10263424576,
      "step": 36090,
      "train_runtime": 350937.0391,
      "train_tokens_per_second": 29245.772
    },
    {
      "epoch": 1.2772889406842443,
      "grad_norm": 0.6328125,
      "learning_rate": 2.3287574219727752e-05,
      "loss": 0.39055254459381106,
      "num_input_tokens_seen": 10266299072,
      "step": 36100,
      "train_runtime": 351038.9557,
      "train_tokens_per_second": 29245.47
    },
    {
      "epoch": 1.277642760947973,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3285048808528872e-05,
      "loss": 0.38941469192504885,
      "num_input_tokens_seen": 10269100928,
      "step": 36110,
      "train_runtime": 351134.099,
      "train_tokens_per_second": 29245.525
    },
    {
      "epoch": 1.2779965812117018,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3282524218753308e-05,
      "loss": 0.39226882457733153,
      "num_input_tokens_seen": 10271921792,
      "step": 36120,
      "train_runtime": 351229.719,
      "train_tokens_per_second": 29245.594
    },
    {
      "epoch": 1.2783504014754306,
      "grad_norm": 0.59765625,
      "learning_rate": 2.328000044995586e-05,
      "loss": 0.3881704330444336,
      "num_input_tokens_seen": 10274797312,
      "step": 36130,
      "train_runtime": 351326.457,
      "train_tokens_per_second": 29245.726
    },
    {
      "epoch": 1.2787042217391593,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3277477501691647e-05,
      "loss": 0.3907644271850586,
      "num_input_tokens_seen": 10277620416,
      "step": 36140,
      "train_runtime": 351422.6468,
      "train_tokens_per_second": 29245.754
    },
    {
      "epoch": 1.279058042002888,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3274955373516163e-05,
      "loss": 0.391316556930542,
      "num_input_tokens_seen": 10280484224,
      "step": 36150,
      "train_runtime": 351519.7564,
      "train_tokens_per_second": 29245.822
    },
    {
      "epoch": 1.2794118622666169,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3272434064985205e-05,
      "loss": 0.3923471450805664,
      "num_input_tokens_seen": 10283366336,
      "step": 36160,
      "train_runtime": 351617.6747,
      "train_tokens_per_second": 29245.874
    },
    {
      "epoch": 1.2797656825303456,
      "grad_norm": 0.58203125,
      "learning_rate": 2.3269913575654932e-05,
      "loss": 0.38865563869476316,
      "num_input_tokens_seen": 10286268288,
      "step": 36170,
      "train_runtime": 351717.6281,
      "train_tokens_per_second": 29245.814
    },
    {
      "epoch": 1.2801195027940744,
      "grad_norm": 0.59765625,
      "learning_rate": 2.326739390508182e-05,
      "loss": 0.39372360706329346,
      "num_input_tokens_seen": 10289162304,
      "step": 36180,
      "train_runtime": 351819.8586,
      "train_tokens_per_second": 29245.542
    },
    {
      "epoch": 1.2804733230578031,
      "grad_norm": 0.609375,
      "learning_rate": 2.326487505282269e-05,
      "loss": 0.3919198751449585,
      "num_input_tokens_seen": 10291963072,
      "step": 36190,
      "train_runtime": 351913.2904,
      "train_tokens_per_second": 29245.736
    },
    {
      "epoch": 1.280827143321532,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3262357018434698e-05,
      "loss": 0.39367635250091554,
      "num_input_tokens_seen": 10294793920,
      "step": 36200,
      "train_runtime": 352009.7273,
      "train_tokens_per_second": 29245.765
    },
    {
      "epoch": 1.2811809635852607,
      "grad_norm": 0.625,
      "learning_rate": 2.3259839801475338e-05,
      "loss": 0.39380860328674316,
      "num_input_tokens_seen": 10297646528,
      "step": 36210,
      "train_runtime": 352107.488,
      "train_tokens_per_second": 29245.747
    },
    {
      "epoch": 1.2815347838489894,
      "grad_norm": 0.6171875,
      "learning_rate": 2.325732340150243e-05,
      "loss": 0.3922597646713257,
      "num_input_tokens_seen": 10300463360,
      "step": 36220,
      "train_runtime": 352203.6652,
      "train_tokens_per_second": 29245.759
    },
    {
      "epoch": 1.2818886041127182,
      "grad_norm": 0.62109375,
      "learning_rate": 2.325480781807415e-05,
      "loss": 0.3932282209396362,
      "num_input_tokens_seen": 10303317952,
      "step": 36230,
      "train_runtime": 352304.2576,
      "train_tokens_per_second": 29245.511
    },
    {
      "epoch": 1.282242424376447,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3252293050748975e-05,
      "loss": 0.3873154640197754,
      "num_input_tokens_seen": 10306168768,
      "step": 36240,
      "train_runtime": 352400.8422,
      "train_tokens_per_second": 29245.585
    },
    {
      "epoch": 1.282596244640176,
      "grad_norm": 0.59765625,
      "learning_rate": 2.324977909908575e-05,
      "loss": 0.39307422637939454,
      "num_input_tokens_seen": 10308941952,
      "step": 36250,
      "train_runtime": 352496.0846,
      "train_tokens_per_second": 29245.55
    },
    {
      "epoch": 1.2829500649039045,
      "grad_norm": 0.625,
      "learning_rate": 2.3247265962643632e-05,
      "loss": 0.39191014766693116,
      "num_input_tokens_seen": 10311799040,
      "step": 36260,
      "train_runtime": 352595.1545,
      "train_tokens_per_second": 29245.436
    },
    {
      "epoch": 1.2833038851676335,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3244753640982106e-05,
      "loss": 0.3940455436706543,
      "num_input_tokens_seen": 10314670976,
      "step": 36270,
      "train_runtime": 352695.7194,
      "train_tokens_per_second": 29245.24
    },
    {
      "epoch": 1.2836577054313623,
      "grad_norm": 0.62890625,
      "learning_rate": 2.324224213366102e-05,
      "loss": 0.3956727981567383,
      "num_input_tokens_seen": 10317556096,
      "step": 36280,
      "train_runtime": 352795.6582,
      "train_tokens_per_second": 29245.133
    },
    {
      "epoch": 1.284011525695091,
      "grad_norm": 0.609375,
      "learning_rate": 2.3239731440240538e-05,
      "loss": 0.3826421022415161,
      "num_input_tokens_seen": 10320370240,
      "step": 36290,
      "train_runtime": 352891.3146,
      "train_tokens_per_second": 29245.181
    },
    {
      "epoch": 1.2843653459588198,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3237221560281143e-05,
      "loss": 0.39660933017730715,
      "num_input_tokens_seen": 10323213696,
      "step": 36300,
      "train_runtime": 352991.1144,
      "train_tokens_per_second": 29244.968
    },
    {
      "epoch": 1.2847191662225486,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3234712493343673e-05,
      "loss": 0.39238438606262205,
      "num_input_tokens_seen": 10326142016,
      "step": 36310,
      "train_runtime": 353091.1406,
      "train_tokens_per_second": 29244.976
    },
    {
      "epoch": 1.2850729864862773,
      "grad_norm": 0.625,
      "learning_rate": 2.323220423898928e-05,
      "loss": 0.39208695888519285,
      "num_input_tokens_seen": 10329028416,
      "step": 36320,
      "train_runtime": 353189.9915,
      "train_tokens_per_second": 29244.964
    },
    {
      "epoch": 1.285426806750006,
      "grad_norm": 0.58203125,
      "learning_rate": 2.3229696796779456e-05,
      "loss": 0.39110984802246096,
      "num_input_tokens_seen": 10331877632,
      "step": 36330,
      "train_runtime": 353286.6885,
      "train_tokens_per_second": 29245.024
    },
    {
      "epoch": 1.2857806270137349,
      "grad_norm": 0.55859375,
      "learning_rate": 2.3227190166276033e-05,
      "loss": 0.3885376453399658,
      "num_input_tokens_seen": 10334730560,
      "step": 36340,
      "train_runtime": 353382.0732,
      "train_tokens_per_second": 29245.203
    },
    {
      "epoch": 1.2861344472774636,
      "grad_norm": 0.59375,
      "learning_rate": 2.3224684347041153e-05,
      "loss": 0.3953958511352539,
      "num_input_tokens_seen": 10337580928,
      "step": 36350,
      "train_runtime": 353479.2301,
      "train_tokens_per_second": 29245.229
    },
    {
      "epoch": 1.2864882675411924,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3222179338637306e-05,
      "loss": 0.3921929597854614,
      "num_input_tokens_seen": 10340425472,
      "step": 36360,
      "train_runtime": 353576.0464,
      "train_tokens_per_second": 29245.266
    },
    {
      "epoch": 1.2868420878049212,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3219675140627302e-05,
      "loss": 0.38900277614593504,
      "num_input_tokens_seen": 10343287040,
      "step": 36370,
      "train_runtime": 353672.324,
      "train_tokens_per_second": 29245.396
    },
    {
      "epoch": 1.28719590806865,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3217171752574292e-05,
      "loss": 0.39088029861450196,
      "num_input_tokens_seen": 10346137920,
      "step": 36380,
      "train_runtime": 353769.0309,
      "train_tokens_per_second": 29245.46
    },
    {
      "epoch": 1.2875497283323787,
      "grad_norm": 0.6640625,
      "learning_rate": 2.3214669174041745e-05,
      "loss": 0.3967722415924072,
      "num_input_tokens_seen": 10348913664,
      "step": 36390,
      "train_runtime": 353861.0586,
      "train_tokens_per_second": 29245.698
    },
    {
      "epoch": 1.2879035485961077,
      "grad_norm": 0.62890625,
      "learning_rate": 2.321216740459346e-05,
      "loss": 0.3920655012130737,
      "num_input_tokens_seen": 10351740544,
      "step": 36400,
      "train_runtime": 353956.1517,
      "train_tokens_per_second": 29245.827
    },
    {
      "epoch": 1.2882573688598362,
      "grad_norm": 0.609375,
      "learning_rate": 2.3209666443793578e-05,
      "loss": 0.39097981452941893,
      "num_input_tokens_seen": 10354564224,
      "step": 36410,
      "train_runtime": 354054.1439,
      "train_tokens_per_second": 29245.708
    },
    {
      "epoch": 1.2886111891235652,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3207166291206553e-05,
      "loss": 0.3885443449020386,
      "num_input_tokens_seen": 10357410304,
      "step": 36420,
      "train_runtime": 354151.5021,
      "train_tokens_per_second": 29245.705
    },
    {
      "epoch": 1.2889650093872937,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3204666946397175e-05,
      "loss": 0.3927842378616333,
      "num_input_tokens_seen": 10360279296,
      "step": 36430,
      "train_runtime": 354249.9769,
      "train_tokens_per_second": 29245.674
    },
    {
      "epoch": 1.2893188296510227,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3202168408930565e-05,
      "loss": 0.39349889755249023,
      "num_input_tokens_seen": 10363111616,
      "step": 36440,
      "train_runtime": 354346.4421,
      "train_tokens_per_second": 29245.705
    },
    {
      "epoch": 1.2896726499147515,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3199670678372166e-05,
      "loss": 0.3919225692749023,
      "num_input_tokens_seen": 10365997312,
      "step": 36450,
      "train_runtime": 354445.3674,
      "train_tokens_per_second": 29245.684
    },
    {
      "epoch": 1.2900264701784803,
      "grad_norm": 0.578125,
      "learning_rate": 2.3197173754287747e-05,
      "loss": 0.388189959526062,
      "num_input_tokens_seen": 10368827776,
      "step": 36460,
      "train_runtime": 354538.5494,
      "train_tokens_per_second": 29245.981
    },
    {
      "epoch": 1.290380290442209,
      "grad_norm": 0.6015625,
      "learning_rate": 2.319467763624341e-05,
      "loss": 0.3922209978103638,
      "num_input_tokens_seen": 10371667456,
      "step": 36470,
      "train_runtime": 354634.5094,
      "train_tokens_per_second": 29246.075
    },
    {
      "epoch": 1.2907341107059378,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3192182323805577e-05,
      "loss": 0.38714075088500977,
      "num_input_tokens_seen": 10374532864,
      "step": 36480,
      "train_runtime": 354734.4045,
      "train_tokens_per_second": 29245.917
    },
    {
      "epoch": 1.2910879309696666,
      "grad_norm": 0.6171875,
      "learning_rate": 2.318968781654101e-05,
      "loss": 0.39092111587524414,
      "num_input_tokens_seen": 10377353600,
      "step": 36490,
      "train_runtime": 354828.4586,
      "train_tokens_per_second": 29246.114
    },
    {
      "epoch": 1.2914417512333953,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3187194114016778e-05,
      "loss": 0.39103260040283205,
      "num_input_tokens_seen": 10380203328,
      "step": 36500,
      "train_runtime": 354924.7475,
      "train_tokens_per_second": 29246.209
    },
    {
      "epoch": 1.291795571497124,
      "grad_norm": 0.60546875,
      "learning_rate": 2.318470121580029e-05,
      "loss": 0.39037678241729734,
      "num_input_tokens_seen": 10383002688,
      "step": 36510,
      "train_runtime": 355020.5526,
      "train_tokens_per_second": 29246.202
    },
    {
      "epoch": 1.2921493917608529,
      "grad_norm": 0.5859375,
      "learning_rate": 2.318220912145927e-05,
      "loss": 0.39055805206298827,
      "num_input_tokens_seen": 10385819840,
      "step": 36520,
      "train_runtime": 355117.4685,
      "train_tokens_per_second": 29246.153
    },
    {
      "epoch": 1.2925032120245816,
      "grad_norm": 0.59375,
      "learning_rate": 2.3179717830561783e-05,
      "loss": 0.3888389110565186,
      "num_input_tokens_seen": 10388652800,
      "step": 36530,
      "train_runtime": 355212.3957,
      "train_tokens_per_second": 29246.313
    },
    {
      "epoch": 1.2928570322883104,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3177227342676204e-05,
      "loss": 0.3895245552062988,
      "num_input_tokens_seen": 10391493696,
      "step": 36540,
      "train_runtime": 355310.1496,
      "train_tokens_per_second": 29246.262
    },
    {
      "epoch": 1.2932108525520392,
      "grad_norm": 0.5625,
      "learning_rate": 2.317473765737124e-05,
      "loss": 0.38902356624603274,
      "num_input_tokens_seen": 10394314816,
      "step": 36550,
      "train_runtime": 355406.287,
      "train_tokens_per_second": 29246.289
    },
    {
      "epoch": 1.293564672815768,
      "grad_norm": 0.58203125,
      "learning_rate": 2.317224877421591e-05,
      "loss": 0.3910688877105713,
      "num_input_tokens_seen": 10397156672,
      "step": 36560,
      "train_runtime": 355502.9464,
      "train_tokens_per_second": 29246.331
    },
    {
      "epoch": 1.293918493079497,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3169760692779578e-05,
      "loss": 0.39177327156066893,
      "num_input_tokens_seen": 10400008960,
      "step": 36570,
      "train_runtime": 355601.1802,
      "train_tokens_per_second": 29246.272
    },
    {
      "epoch": 1.2942723133432255,
      "grad_norm": 0.6171875,
      "learning_rate": 2.316727341263191e-05,
      "loss": 0.3879788875579834,
      "num_input_tokens_seen": 10402812224,
      "step": 36580,
      "train_runtime": 355698.4023,
      "train_tokens_per_second": 29246.16
    },
    {
      "epoch": 1.2946261336069544,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3164786933342915e-05,
      "loss": 0.39709744453430174,
      "num_input_tokens_seen": 10405665600,
      "step": 36590,
      "train_runtime": 355796.0953,
      "train_tokens_per_second": 29246.149
    },
    {
      "epoch": 1.294979953870683,
      "grad_norm": 0.609375,
      "learning_rate": 2.3162301254482914e-05,
      "loss": 0.39253203868865966,
      "num_input_tokens_seen": 10408514944,
      "step": 36600,
      "train_runtime": 355893.6424,
      "train_tokens_per_second": 29246.139
    },
    {
      "epoch": 1.295333774134412,
      "grad_norm": 0.58203125,
      "learning_rate": 2.3159816375622546e-05,
      "loss": 0.38924694061279297,
      "num_input_tokens_seen": 10411365760,
      "step": 36610,
      "train_runtime": 355991.4874,
      "train_tokens_per_second": 29246.109
    },
    {
      "epoch": 1.2956875943981407,
      "grad_norm": 0.5703125,
      "learning_rate": 2.3157332296332783e-05,
      "loss": 0.3876067638397217,
      "num_input_tokens_seen": 10414225920,
      "step": 36620,
      "train_runtime": 356089.5858,
      "train_tokens_per_second": 29246.084
    },
    {
      "epoch": 1.2960414146618695,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3154849016184914e-05,
      "loss": 0.3923532485961914,
      "num_input_tokens_seen": 10417155072,
      "step": 36630,
      "train_runtime": 356192.2649,
      "train_tokens_per_second": 29245.877
    },
    {
      "epoch": 1.2963952349255983,
      "grad_norm": 0.625,
      "learning_rate": 2.3152366534750554e-05,
      "loss": 0.38785197734832766,
      "num_input_tokens_seen": 10420005760,
      "step": 36640,
      "train_runtime": 356289.3908,
      "train_tokens_per_second": 29245.905
    },
    {
      "epoch": 1.296749055189327,
      "grad_norm": 0.625,
      "learning_rate": 2.3149884851601633e-05,
      "loss": 0.38750295639038085,
      "num_input_tokens_seen": 10422820672,
      "step": 36650,
      "train_runtime": 356386.0458,
      "train_tokens_per_second": 29245.872
    },
    {
      "epoch": 1.2971028754530558,
      "grad_norm": 0.58984375,
      "learning_rate": 2.31474039663104e-05,
      "loss": 0.3918515920639038,
      "num_input_tokens_seen": 10425622016,
      "step": 36660,
      "train_runtime": 356479.0735,
      "train_tokens_per_second": 29246.098
    },
    {
      "epoch": 1.2974566957167846,
      "grad_norm": 0.6171875,
      "learning_rate": 2.314492387844944e-05,
      "loss": 0.39035797119140625,
      "num_input_tokens_seen": 10428388224,
      "step": 36670,
      "train_runtime": 356576.1579,
      "train_tokens_per_second": 29245.893
    },
    {
      "epoch": 1.2978105159805133,
      "grad_norm": 0.61328125,
      "learning_rate": 2.314244458759163e-05,
      "loss": 0.3924812078475952,
      "num_input_tokens_seen": 10431217152,
      "step": 36680,
      "train_runtime": 356674.9117,
      "train_tokens_per_second": 29245.727
    },
    {
      "epoch": 1.298164336244242,
      "grad_norm": 0.6328125,
      "learning_rate": 2.3139966093310202e-05,
      "loss": 0.38553104400634763,
      "num_input_tokens_seen": 10434115456,
      "step": 36690,
      "train_runtime": 356771.6916,
      "train_tokens_per_second": 29245.917
    },
    {
      "epoch": 1.2985181565079709,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3137488395178695e-05,
      "loss": 0.389219331741333,
      "num_input_tokens_seen": 10436962176,
      "step": 36700,
      "train_runtime": 356868.3728,
      "train_tokens_per_second": 29245.971
    },
    {
      "epoch": 1.2988719767716996,
      "grad_norm": 0.58984375,
      "learning_rate": 2.313501149277095e-05,
      "loss": 0.38736035823822024,
      "num_input_tokens_seen": 10439813568,
      "step": 36710,
      "train_runtime": 356964.1742,
      "train_tokens_per_second": 29246.11
    },
    {
      "epoch": 1.2992257970354284,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3132535385661143e-05,
      "loss": 0.39359226226806643,
      "num_input_tokens_seen": 10442624128,
      "step": 36720,
      "train_runtime": 357060.1122,
      "train_tokens_per_second": 29246.123
    },
    {
      "epoch": 1.2995796172991572,
      "grad_norm": 0.6015625,
      "learning_rate": 2.313006007342377e-05,
      "loss": 0.39149534702301025,
      "num_input_tokens_seen": 10445464384,
      "step": 36730,
      "train_runtime": 357155.96,
      "train_tokens_per_second": 29246.227
    },
    {
      "epoch": 1.2999334375628862,
      "grad_norm": 0.578125,
      "learning_rate": 2.312758555563365e-05,
      "loss": 0.3922810316085815,
      "num_input_tokens_seen": 10448279744,
      "step": 36740,
      "train_runtime": 357251.4893,
      "train_tokens_per_second": 29246.287
    },
    {
      "epoch": 1.3002872578266147,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3125111831865908e-05,
      "loss": 0.3928556442260742,
      "num_input_tokens_seen": 10451154112,
      "step": 36750,
      "train_runtime": 357351.5242,
      "train_tokens_per_second": 29246.144
    },
    {
      "epoch": 1.3006410780903437,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3122638901695983e-05,
      "loss": 0.3899819374084473,
      "num_input_tokens_seen": 10454018944,
      "step": 36760,
      "train_runtime": 357452.2415,
      "train_tokens_per_second": 29245.918
    },
    {
      "epoch": 1.3009948983540722,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3120166764699654e-05,
      "loss": 0.39181995391845703,
      "num_input_tokens_seen": 10456844672,
      "step": 36770,
      "train_runtime": 357549.7234,
      "train_tokens_per_second": 29245.847
    },
    {
      "epoch": 1.3013487186178012,
      "grad_norm": 0.63671875,
      "learning_rate": 2.3117695420453e-05,
      "loss": 0.39190337657928465,
      "num_input_tokens_seen": 10459667520,
      "step": 36780,
      "train_runtime": 357645.8494,
      "train_tokens_per_second": 29245.88
    },
    {
      "epoch": 1.30170253888153,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3115224868532416e-05,
      "loss": 0.39562134742736815,
      "num_input_tokens_seen": 10462504000,
      "step": 36790,
      "train_runtime": 357741.0773,
      "train_tokens_per_second": 29246.024
    },
    {
      "epoch": 1.3020563591452587,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3112755108514628e-05,
      "loss": 0.3894166946411133,
      "num_input_tokens_seen": 10465319296,
      "step": 36800,
      "train_runtime": 357835.7976,
      "train_tokens_per_second": 29246.15
    },
    {
      "epoch": 1.3024101794089875,
      "grad_norm": 0.62890625,
      "learning_rate": 2.311028613997666e-05,
      "loss": 0.39465794563293455,
      "num_input_tokens_seen": 10468055424,
      "step": 36810,
      "train_runtime": 357927.6291,
      "train_tokens_per_second": 29246.291
    },
    {
      "epoch": 1.3027639996727163,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3107817962495873e-05,
      "loss": 0.38817002773284914,
      "num_input_tokens_seen": 10470895104,
      "step": 36820,
      "train_runtime": 358023.7671,
      "train_tokens_per_second": 29246.369
    },
    {
      "epoch": 1.303117819936445,
      "grad_norm": 0.59375,
      "learning_rate": 2.3105350575649924e-05,
      "loss": 0.3915217399597168,
      "num_input_tokens_seen": 10473714944,
      "step": 36830,
      "train_runtime": 358119.1685,
      "train_tokens_per_second": 29246.452
    },
    {
      "epoch": 1.3034716402001738,
      "grad_norm": 0.6015625,
      "learning_rate": 2.31028839790168e-05,
      "loss": 0.38786485195159914,
      "num_input_tokens_seen": 10476587264,
      "step": 36840,
      "train_runtime": 358214.9748,
      "train_tokens_per_second": 29246.648
    },
    {
      "epoch": 1.3038254604639026,
      "grad_norm": 0.6015625,
      "learning_rate": 2.310041817217479e-05,
      "loss": 0.3925775051116943,
      "num_input_tokens_seen": 10479458496,
      "step": 36850,
      "train_runtime": 358312.5866,
      "train_tokens_per_second": 29246.694
    },
    {
      "epoch": 1.3041792807276313,
      "grad_norm": 0.58203125,
      "learning_rate": 2.3097953154702514e-05,
      "loss": 0.3880384683609009,
      "num_input_tokens_seen": 10482322368,
      "step": 36860,
      "train_runtime": 358409.6312,
      "train_tokens_per_second": 29246.765
    },
    {
      "epoch": 1.30453310099136,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3095488926178896e-05,
      "loss": 0.3910287618637085,
      "num_input_tokens_seen": 10485198720,
      "step": 36870,
      "train_runtime": 358507.8019,
      "train_tokens_per_second": 29246.78
    },
    {
      "epoch": 1.3048869212550889,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3093025486183175e-05,
      "loss": 0.39112653732299807,
      "num_input_tokens_seen": 10488079168,
      "step": 36880,
      "train_runtime": 358605.28,
      "train_tokens_per_second": 29246.862
    },
    {
      "epoch": 1.3052407415188176,
      "grad_norm": 0.6171875,
      "learning_rate": 2.3090562834294908e-05,
      "loss": 0.38792574405670166,
      "num_input_tokens_seen": 10490917376,
      "step": 36890,
      "train_runtime": 358703.1434,
      "train_tokens_per_second": 29246.795
    },
    {
      "epoch": 1.3055945617825464,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3088100970093966e-05,
      "loss": 0.38968286514282224,
      "num_input_tokens_seen": 10493685440,
      "step": 36900,
      "train_runtime": 358797.598,
      "train_tokens_per_second": 29246.811
    },
    {
      "epoch": 1.3059483820462754,
      "grad_norm": 0.609375,
      "learning_rate": 2.3085639893160527e-05,
      "loss": 0.39174542427062986,
      "num_input_tokens_seen": 10496578880,
      "step": 36910,
      "train_runtime": 358899.126,
      "train_tokens_per_second": 29246.599
    },
    {
      "epoch": 1.306302202310004,
      "grad_norm": 0.63671875,
      "learning_rate": 2.3083179603075084e-05,
      "loss": 0.3913600444793701,
      "num_input_tokens_seen": 10499381056,
      "step": 36920,
      "train_runtime": 358995.3682,
      "train_tokens_per_second": 29246.564
    },
    {
      "epoch": 1.306656022573733,
      "grad_norm": 0.62109375,
      "learning_rate": 2.308072009941845e-05,
      "loss": 0.3898455619812012,
      "num_input_tokens_seen": 10502223168,
      "step": 36930,
      "train_runtime": 359091.1348,
      "train_tokens_per_second": 29246.679
    },
    {
      "epoch": 1.3070098428374615,
      "grad_norm": 0.609375,
      "learning_rate": 2.3078261381771748e-05,
      "loss": 0.39620747566223147,
      "num_input_tokens_seen": 10505056384,
      "step": 36940,
      "train_runtime": 359190.0985,
      "train_tokens_per_second": 29246.509
    },
    {
      "epoch": 1.3073636631011905,
      "grad_norm": 0.609375,
      "learning_rate": 2.3075803449716407e-05,
      "loss": 0.388875150680542,
      "num_input_tokens_seen": 10507953088,
      "step": 36950,
      "train_runtime": 359290.4044,
      "train_tokens_per_second": 29246.406
    },
    {
      "epoch": 1.3077174833649192,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3073346302834174e-05,
      "loss": 0.38850765228271483,
      "num_input_tokens_seen": 10510797760,
      "step": 36960,
      "train_runtime": 359386.994,
      "train_tokens_per_second": 29246.461
    },
    {
      "epoch": 1.308071303628648,
      "grad_norm": 0.59375,
      "learning_rate": 2.3070889940707103e-05,
      "loss": 0.3870737552642822,
      "num_input_tokens_seen": 10513629440,
      "step": 36970,
      "train_runtime": 359481.5065,
      "train_tokens_per_second": 29246.649
    },
    {
      "epoch": 1.3084251238923768,
      "grad_norm": 0.61328125,
      "learning_rate": 2.3068434362917562e-05,
      "loss": 0.3883809566497803,
      "num_input_tokens_seen": 10516463040,
      "step": 36980,
      "train_runtime": 359579.0445,
      "train_tokens_per_second": 29246.596
    },
    {
      "epoch": 1.3087789441561055,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3065979569048235e-05,
      "loss": 0.3883203983306885,
      "num_input_tokens_seen": 10519315840,
      "step": 36990,
      "train_runtime": 359677.9472,
      "train_tokens_per_second": 29246.485
    },
    {
      "epoch": 1.3091327644198343,
      "grad_norm": 0.609375,
      "learning_rate": 2.306352555868211e-05,
      "loss": 0.39303610324859617,
      "num_input_tokens_seen": 10522161280,
      "step": 37000,
      "train_runtime": 359779.6993,
      "train_tokens_per_second": 29246.123
    },
    {
      "epoch": 1.309486584683563,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3061072331402486e-05,
      "loss": 0.38878188133239744,
      "num_input_tokens_seen": 10525001856,
      "step": 37010,
      "train_runtime": 359877.1936,
      "train_tokens_per_second": 29246.093
    },
    {
      "epoch": 1.3098404049472918,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3058619886792977e-05,
      "loss": 0.3925739288330078,
      "num_input_tokens_seen": 10527894912,
      "step": 37020,
      "train_runtime": 359978.4156,
      "train_tokens_per_second": 29245.906
    },
    {
      "epoch": 1.3101942252110206,
      "grad_norm": 0.6015625,
      "learning_rate": 2.30561682244375e-05,
      "loss": 0.391937255859375,
      "num_input_tokens_seen": 10530746304,
      "step": 37030,
      "train_runtime": 360076.1454,
      "train_tokens_per_second": 29245.887
    },
    {
      "epoch": 1.3105480454747493,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3053717343920285e-05,
      "loss": 0.3854072093963623,
      "num_input_tokens_seen": 10533536064,
      "step": 37040,
      "train_runtime": 360167.7451,
      "train_tokens_per_second": 29246.195
    },
    {
      "epoch": 1.3109018657384781,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3051267244825882e-05,
      "loss": 0.38942029476165774,
      "num_input_tokens_seen": 10536402112,
      "step": 37050,
      "train_runtime": 360267.4987,
      "train_tokens_per_second": 29246.052
    },
    {
      "epoch": 1.3112556860022069,
      "grad_norm": 0.59765625,
      "learning_rate": 2.3048817926739122e-05,
      "loss": 0.38984384536743166,
      "num_input_tokens_seen": 10539198208,
      "step": 37060,
      "train_runtime": 360362.529,
      "train_tokens_per_second": 29246.099
    },
    {
      "epoch": 1.3116095062659356,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3046369389245177e-05,
      "loss": 0.39116687774658204,
      "num_input_tokens_seen": 10542040704,
      "step": 37070,
      "train_runtime": 360459.2542,
      "train_tokens_per_second": 29246.137
    },
    {
      "epoch": 1.3119633265296646,
      "grad_norm": 0.609375,
      "learning_rate": 2.3043921631929502e-05,
      "loss": 0.3927873134613037,
      "num_input_tokens_seen": 10544857600,
      "step": 37080,
      "train_runtime": 360554.3536,
      "train_tokens_per_second": 29246.236
    },
    {
      "epoch": 1.3123171467933932,
      "grad_norm": 0.60546875,
      "learning_rate": 2.304147465437788e-05,
      "loss": 0.39102873802185056,
      "num_input_tokens_seen": 10547648128,
      "step": 37090,
      "train_runtime": 360651.1462,
      "train_tokens_per_second": 29246.124
    },
    {
      "epoch": 1.3126709670571222,
      "grad_norm": 0.609375,
      "learning_rate": 2.3039028456176388e-05,
      "loss": 0.3926230907440186,
      "num_input_tokens_seen": 10550496320,
      "step": 37100,
      "train_runtime": 360749.5896,
      "train_tokens_per_second": 29246.038
    },
    {
      "epoch": 1.313024787320851,
      "grad_norm": 0.5859375,
      "learning_rate": 2.3036583036911418e-05,
      "loss": 0.39110236167907714,
      "num_input_tokens_seen": 10553359488,
      "step": 37110,
      "train_runtime": 360848.9736,
      "train_tokens_per_second": 29245.918
    },
    {
      "epoch": 1.3133786075845797,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3034138396169667e-05,
      "loss": 0.3842885971069336,
      "num_input_tokens_seen": 10556238080,
      "step": 37120,
      "train_runtime": 360949.0415,
      "train_tokens_per_second": 29245.785
    },
    {
      "epoch": 1.3137324278483085,
      "grad_norm": 0.609375,
      "learning_rate": 2.3031694533538127e-05,
      "loss": 0.3885584115982056,
      "num_input_tokens_seen": 10559099584,
      "step": 37130,
      "train_runtime": 361045.5472,
      "train_tokens_per_second": 29245.893
    },
    {
      "epoch": 1.3140862481120372,
      "grad_norm": 0.6015625,
      "learning_rate": 2.302925144860413e-05,
      "loss": 0.3924171686172485,
      "num_input_tokens_seen": 10561887616,
      "step": 37140,
      "train_runtime": 361137.2381,
      "train_tokens_per_second": 29246.188
    },
    {
      "epoch": 1.314440068375766,
      "grad_norm": 0.625,
      "learning_rate": 2.302680914095527e-05,
      "loss": 0.38936653137207033,
      "num_input_tokens_seen": 10564708288,
      "step": 37150,
      "train_runtime": 361233.1305,
      "train_tokens_per_second": 29246.233
    },
    {
      "epoch": 1.3147938886394948,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3024367610179485e-05,
      "loss": 0.39456074237823485,
      "num_input_tokens_seen": 10567546368,
      "step": 37160,
      "train_runtime": 361329.6712,
      "train_tokens_per_second": 29246.273
    },
    {
      "epoch": 1.3151477089032235,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3021926855864995e-05,
      "loss": 0.3941436767578125,
      "num_input_tokens_seen": 10570346432,
      "step": 37170,
      "train_runtime": 361423.3379,
      "train_tokens_per_second": 29246.441
    },
    {
      "epoch": 1.3155015291669523,
      "grad_norm": 0.6015625,
      "learning_rate": 2.3019486877600337e-05,
      "loss": 0.3912742376327515,
      "num_input_tokens_seen": 10573251008,
      "step": 37180,
      "train_runtime": 361525.5141,
      "train_tokens_per_second": 29246.21
    },
    {
      "epoch": 1.315855349430681,
      "grad_norm": 0.62109375,
      "learning_rate": 2.3017047674974354e-05,
      "loss": 0.3891000270843506,
      "num_input_tokens_seen": 10576018752,
      "step": 37190,
      "train_runtime": 361620.103,
      "train_tokens_per_second": 29246.214
    },
    {
      "epoch": 1.3162091696944098,
      "grad_norm": 0.60546875,
      "learning_rate": 2.301460924757618e-05,
      "loss": 0.38751907348632814,
      "num_input_tokens_seen": 10578802304,
      "step": 37200,
      "train_runtime": 361712.2268,
      "train_tokens_per_second": 29246.46
    },
    {
      "epoch": 1.3165629899581386,
      "grad_norm": 0.625,
      "learning_rate": 2.3012171594995272e-05,
      "loss": 0.39267170429229736,
      "num_input_tokens_seen": 10581651136,
      "step": 37210,
      "train_runtime": 361809.9441,
      "train_tokens_per_second": 29246.435
    },
    {
      "epoch": 1.3169168102218674,
      "grad_norm": 0.62890625,
      "learning_rate": 2.3009734716821385e-05,
      "loss": 0.3916393995285034,
      "num_input_tokens_seen": 10584571264,
      "step": 37220,
      "train_runtime": 361911.3705,
      "train_tokens_per_second": 29246.308
    },
    {
      "epoch": 1.3172706304855961,
      "grad_norm": 0.60546875,
      "learning_rate": 2.3007298612644562e-05,
      "loss": 0.3918531656265259,
      "num_input_tokens_seen": 10587413376,
      "step": 37230,
      "train_runtime": 362007.2937,
      "train_tokens_per_second": 29246.409
    },
    {
      "epoch": 1.3176244507493249,
      "grad_norm": 0.62890625,
      "learning_rate": 2.3004863282055178e-05,
      "loss": 0.3942422866821289,
      "num_input_tokens_seen": 10590196160,
      "step": 37240,
      "train_runtime": 362101.6099,
      "train_tokens_per_second": 29246.476
    },
    {
      "epoch": 1.3179782710130539,
      "grad_norm": 0.625,
      "learning_rate": 2.300242872464389e-05,
      "loss": 0.3919389247894287,
      "num_input_tokens_seen": 10593082240,
      "step": 37250,
      "train_runtime": 362200.9103,
      "train_tokens_per_second": 29246.426
    },
    {
      "epoch": 1.3183320912767824,
      "grad_norm": 0.62109375,
      "learning_rate": 2.299999494000167e-05,
      "loss": 0.3921942949295044,
      "num_input_tokens_seen": 10595914944,
      "step": 37260,
      "train_runtime": 362298.88,
      "train_tokens_per_second": 29246.336
    },
    {
      "epoch": 1.3186859115405114,
      "grad_norm": 0.60546875,
      "learning_rate": 2.299756192771979e-05,
      "loss": 0.3891026973724365,
      "num_input_tokens_seen": 10598733952,
      "step": 37270,
      "train_runtime": 362393.0457,
      "train_tokens_per_second": 29246.516
    },
    {
      "epoch": 1.3190397318042402,
      "grad_norm": 0.61328125,
      "learning_rate": 2.299512968738981e-05,
      "loss": 0.3897572040557861,
      "num_input_tokens_seen": 10601590592,
      "step": 37280,
      "train_runtime": 362492.5509,
      "train_tokens_per_second": 29246.368
    },
    {
      "epoch": 1.319393552067969,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2992698218603617e-05,
      "loss": 0.39164884090423585,
      "num_input_tokens_seen": 10604454208,
      "step": 37290,
      "train_runtime": 362589.9875,
      "train_tokens_per_second": 29246.407
    },
    {
      "epoch": 1.3197473723316977,
      "grad_norm": 0.59375,
      "learning_rate": 2.2990267520953383e-05,
      "loss": 0.3887930870056152,
      "num_input_tokens_seen": 10607281792,
      "step": 37300,
      "train_runtime": 362687.7187,
      "train_tokens_per_second": 29246.322
    },
    {
      "epoch": 1.3201011925954265,
      "grad_norm": 0.625,
      "learning_rate": 2.2987837594031587e-05,
      "loss": 0.3866581916809082,
      "num_input_tokens_seen": 10610142208,
      "step": 37310,
      "train_runtime": 362787.1685,
      "train_tokens_per_second": 29246.189
    },
    {
      "epoch": 1.3204550128591552,
      "grad_norm": 0.59375,
      "learning_rate": 2.2985408437431016e-05,
      "loss": 0.38734891414642336,
      "num_input_tokens_seen": 10613000000,
      "step": 37320,
      "train_runtime": 362883.2676,
      "train_tokens_per_second": 29246.32
    },
    {
      "epoch": 1.320808833122884,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2982980050744743e-05,
      "loss": 0.39428086280822755,
      "num_input_tokens_seen": 10615812480,
      "step": 37330,
      "train_runtime": 362979.1906,
      "train_tokens_per_second": 29246.339
    },
    {
      "epoch": 1.3211626533866128,
      "grad_norm": 0.578125,
      "learning_rate": 2.2980552433566154e-05,
      "loss": 0.3905577421188354,
      "num_input_tokens_seen": 10618701824,
      "step": 37340,
      "train_runtime": 363080.3312,
      "train_tokens_per_second": 29246.15
    },
    {
      "epoch": 1.3215164736503415,
      "grad_norm": 0.60546875,
      "learning_rate": 2.297812558548893e-05,
      "loss": 0.3894331932067871,
      "num_input_tokens_seen": 10621494208,
      "step": 37350,
      "train_runtime": 363172.9061,
      "train_tokens_per_second": 29246.384
    },
    {
      "epoch": 1.3218702939140703,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2975699506107056e-05,
      "loss": 0.39195420742034914,
      "num_input_tokens_seen": 10624335168,
      "step": 37360,
      "train_runtime": 363272.1554,
      "train_tokens_per_second": 29246.214
    },
    {
      "epoch": 1.322224114177799,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2973274195014816e-05,
      "loss": 0.3894282102584839,
      "num_input_tokens_seen": 10627284608,
      "step": 37370,
      "train_runtime": 363376.8585,
      "train_tokens_per_second": 29245.904
    },
    {
      "epoch": 1.3225779344415278,
      "grad_norm": 0.58203125,
      "learning_rate": 2.2970849651806796e-05,
      "loss": 0.3907696962356567,
      "num_input_tokens_seen": 10630163712,
      "step": 37380,
      "train_runtime": 363474.0165,
      "train_tokens_per_second": 29246.007
    },
    {
      "epoch": 1.3229317547052566,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2968425876077866e-05,
      "loss": 0.39137604236602785,
      "num_input_tokens_seen": 10633005760,
      "step": 37390,
      "train_runtime": 363570.2619,
      "train_tokens_per_second": 29246.082
    },
    {
      "epoch": 1.3232855749689856,
      "grad_norm": 0.57421875,
      "learning_rate": 2.2966002867423225e-05,
      "loss": 0.391874885559082,
      "num_input_tokens_seen": 10635808064,
      "step": 37400,
      "train_runtime": 363667.7049,
      "train_tokens_per_second": 29245.952
    },
    {
      "epoch": 1.3236393952327141,
      "grad_norm": 0.6015625,
      "learning_rate": 2.296358062543834e-05,
      "loss": 0.3922929525375366,
      "num_input_tokens_seen": 10638664576,
      "step": 37410,
      "train_runtime": 363767.4395,
      "train_tokens_per_second": 29245.786
    },
    {
      "epoch": 1.323993215496443,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2961159149718993e-05,
      "loss": 0.3879826545715332,
      "num_input_tokens_seen": 10641406976,
      "step": 37420,
      "train_runtime": 363859.8428,
      "train_tokens_per_second": 29245.896
    },
    {
      "epoch": 1.3243470357601717,
      "grad_norm": 0.60546875,
      "learning_rate": 2.295873843986126e-05,
      "loss": 0.39168686866760255,
      "num_input_tokens_seen": 10644213824,
      "step": 37430,
      "train_runtime": 363956.1724,
      "train_tokens_per_second": 29245.867
    },
    {
      "epoch": 1.3247008560239006,
      "grad_norm": 0.5703125,
      "learning_rate": 2.2956318495461526e-05,
      "loss": 0.3887831687927246,
      "num_input_tokens_seen": 10647067776,
      "step": 37440,
      "train_runtime": 364055.0182,
      "train_tokens_per_second": 29245.766
    },
    {
      "epoch": 1.3250546762876294,
      "grad_norm": 0.625,
      "learning_rate": 2.295389931611645e-05,
      "loss": 0.39048161506652834,
      "num_input_tokens_seen": 10649914880,
      "step": 37450,
      "train_runtime": 364150.4294,
      "train_tokens_per_second": 29245.922
    },
    {
      "epoch": 1.3254084965513582,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2951480901423017e-05,
      "loss": 0.3884414196014404,
      "num_input_tokens_seen": 10652834368,
      "step": 37460,
      "train_runtime": 364252.4646,
      "train_tokens_per_second": 29245.744
    },
    {
      "epoch": 1.325762316815087,
      "grad_norm": 0.59375,
      "learning_rate": 2.294906325097848e-05,
      "loss": 0.39214463233947755,
      "num_input_tokens_seen": 10655695232,
      "step": 37470,
      "train_runtime": 364349.9782,
      "train_tokens_per_second": 29245.769
    },
    {
      "epoch": 1.3261161370788157,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2946646364380415e-05,
      "loss": 0.39381585121154783,
      "num_input_tokens_seen": 10658500800,
      "step": 37480,
      "train_runtime": 364445.9941,
      "train_tokens_per_second": 29245.762
    },
    {
      "epoch": 1.3264699573425445,
      "grad_norm": 0.6015625,
      "learning_rate": 2.294423024122668e-05,
      "loss": 0.38974153995513916,
      "num_input_tokens_seen": 10661347136,
      "step": 37490,
      "train_runtime": 364543.7907,
      "train_tokens_per_second": 29245.724
    },
    {
      "epoch": 1.3268237776062732,
      "grad_norm": 0.6328125,
      "learning_rate": 2.2941814881115428e-05,
      "loss": 0.3915058374404907,
      "num_input_tokens_seen": 10664161344,
      "step": 37500,
      "train_runtime": 364640.5279,
      "train_tokens_per_second": 29245.683
    },
    {
      "epoch": 1.327177597870002,
      "grad_norm": 0.6015625,
      "learning_rate": 2.293940028364512e-05,
      "loss": 0.3865213871002197,
      "num_input_tokens_seen": 10666978752,
      "step": 37510,
      "train_runtime": 364738.2333,
      "train_tokens_per_second": 29245.573
    },
    {
      "epoch": 1.3275314181337308,
      "grad_norm": 0.57421875,
      "learning_rate": 2.2936986448414498e-05,
      "loss": 0.38599205017089844,
      "num_input_tokens_seen": 10669854080,
      "step": 37520,
      "train_runtime": 364835.1879,
      "train_tokens_per_second": 29245.683
    },
    {
      "epoch": 1.3278852383974595,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2934573375022613e-05,
      "loss": 0.39447388648986814,
      "num_input_tokens_seen": 10672654272,
      "step": 37530,
      "train_runtime": 364931.9291,
      "train_tokens_per_second": 29245.603
    },
    {
      "epoch": 1.3282390586611883,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2932161063068803e-05,
      "loss": 0.3906829357147217,
      "num_input_tokens_seen": 10675512896,
      "step": 37540,
      "train_runtime": 365030.3218,
      "train_tokens_per_second": 29245.551
    },
    {
      "epoch": 1.328592878924917,
      "grad_norm": 0.60546875,
      "learning_rate": 2.29297495121527e-05,
      "loss": 0.3903957366943359,
      "num_input_tokens_seen": 10678393792,
      "step": 37550,
      "train_runtime": 365129.2312,
      "train_tokens_per_second": 29245.519
    },
    {
      "epoch": 1.3289466991886458,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2927338721874245e-05,
      "loss": 0.39175052642822267,
      "num_input_tokens_seen": 10681257408,
      "step": 37560,
      "train_runtime": 365228.3417,
      "train_tokens_per_second": 29245.423
    },
    {
      "epoch": 1.3293005194523748,
      "grad_norm": 0.6328125,
      "learning_rate": 2.2924928691833646e-05,
      "loss": 0.3884971857070923,
      "num_input_tokens_seen": 10684109248,
      "step": 37570,
      "train_runtime": 365324.8351,
      "train_tokens_per_second": 29245.505
    },
    {
      "epoch": 1.3296543397161034,
      "grad_norm": 0.6328125,
      "learning_rate": 2.292251942163143e-05,
      "loss": 0.39147922992706297,
      "num_input_tokens_seen": 10686949504,
      "step": 37580,
      "train_runtime": 365425.8234,
      "train_tokens_per_second": 29245.195
    },
    {
      "epoch": 1.3300081599798323,
      "grad_norm": 0.6171875,
      "learning_rate": 2.292011091086841e-05,
      "loss": 0.38724594116210936,
      "num_input_tokens_seen": 10689757504,
      "step": 37590,
      "train_runtime": 365519.8107,
      "train_tokens_per_second": 29245.357
    },
    {
      "epoch": 1.330361980243561,
      "grad_norm": 0.6015625,
      "learning_rate": 2.291770315914569e-05,
      "loss": 0.3886714458465576,
      "num_input_tokens_seen": 10692563456,
      "step": 37600,
      "train_runtime": 365615.9653,
      "train_tokens_per_second": 29245.341
    },
    {
      "epoch": 1.3307158005072899,
      "grad_norm": 0.578125,
      "learning_rate": 2.291529616606466e-05,
      "loss": 0.3949223518371582,
      "num_input_tokens_seen": 10695440576,
      "step": 37610,
      "train_runtime": 365714.4005,
      "train_tokens_per_second": 29245.336
    },
    {
      "epoch": 1.3310696207710186,
      "grad_norm": 0.5859375,
      "learning_rate": 2.291288993122703e-05,
      "loss": 0.3912301540374756,
      "num_input_tokens_seen": 10698309440,
      "step": 37620,
      "train_runtime": 365810.5334,
      "train_tokens_per_second": 29245.493
    },
    {
      "epoch": 1.3314234410347474,
      "grad_norm": 0.60546875,
      "learning_rate": 2.291048445423477e-05,
      "loss": 0.39014320373535155,
      "num_input_tokens_seen": 10701123648,
      "step": 37630,
      "train_runtime": 365906.3739,
      "train_tokens_per_second": 29245.524
    },
    {
      "epoch": 1.3317772612984762,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2908079734690165e-05,
      "loss": 0.3915010929107666,
      "num_input_tokens_seen": 10704008128,
      "step": 37640,
      "train_runtime": 366007.6815,
      "train_tokens_per_second": 29245.31
    },
    {
      "epoch": 1.332131081562205,
      "grad_norm": 0.60546875,
      "learning_rate": 2.290567577219578e-05,
      "loss": 0.3960436820983887,
      "num_input_tokens_seen": 10706837312,
      "step": 37650,
      "train_runtime": 366106.6337,
      "train_tokens_per_second": 29245.133
    },
    {
      "epoch": 1.3324849018259337,
      "grad_norm": 0.6328125,
      "learning_rate": 2.2903272566354474e-05,
      "loss": 0.39026584625244143,
      "num_input_tokens_seen": 10709667584,
      "step": 37660,
      "train_runtime": 366202.3384,
      "train_tokens_per_second": 29245.219
    },
    {
      "epoch": 1.3328387220896625,
      "grad_norm": 0.57421875,
      "learning_rate": 2.2900870116769398e-05,
      "loss": 0.3882603168487549,
      "num_input_tokens_seen": 10712546048,
      "step": 37670,
      "train_runtime": 366302.1771,
      "train_tokens_per_second": 29245.106
    },
    {
      "epoch": 1.3331925423533912,
      "grad_norm": 0.609375,
      "learning_rate": 2.289846842304401e-05,
      "loss": 0.39290494918823243,
      "num_input_tokens_seen": 10715385920,
      "step": 37680,
      "train_runtime": 366394.953,
      "train_tokens_per_second": 29245.452
    },
    {
      "epoch": 1.33354636261712,
      "grad_norm": 0.6328125,
      "learning_rate": 2.2896067484782027e-05,
      "loss": 0.38860411643981935,
      "num_input_tokens_seen": 10718208256,
      "step": 37690,
      "train_runtime": 366491.0424,
      "train_tokens_per_second": 29245.485
    },
    {
      "epoch": 1.3339001828808488,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2893667301587484e-05,
      "loss": 0.39140305519104,
      "num_input_tokens_seen": 10721053248,
      "step": 37700,
      "train_runtime": 366586.5835,
      "train_tokens_per_second": 29245.624
    },
    {
      "epoch": 1.3342540031445775,
      "grad_norm": 0.60546875,
      "learning_rate": 2.28912678730647e-05,
      "loss": 0.38918509483337405,
      "num_input_tokens_seen": 10723918528,
      "step": 37710,
      "train_runtime": 366684.5357,
      "train_tokens_per_second": 29245.625
    },
    {
      "epoch": 1.3346078234083063,
      "grad_norm": 0.625,
      "learning_rate": 2.2888869198818276e-05,
      "loss": 0.3890595197677612,
      "num_input_tokens_seen": 10726782400,
      "step": 37720,
      "train_runtime": 366783.3785,
      "train_tokens_per_second": 29245.552
    },
    {
      "epoch": 1.334961643672035,
      "grad_norm": 0.62109375,
      "learning_rate": 2.288647127845311e-05,
      "loss": 0.38911287784576415,
      "num_input_tokens_seen": 10729592704,
      "step": 37730,
      "train_runtime": 366878.2245,
      "train_tokens_per_second": 29245.652
    },
    {
      "epoch": 1.335315463935764,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2884074111574384e-05,
      "loss": 0.3911198616027832,
      "num_input_tokens_seen": 10732409408,
      "step": 37740,
      "train_runtime": 366972.7822,
      "train_tokens_per_second": 29245.791
    },
    {
      "epoch": 1.3356692841994926,
      "grad_norm": 0.640625,
      "learning_rate": 2.2881677697787584e-05,
      "loss": 0.38768208026885986,
      "num_input_tokens_seen": 10735275520,
      "step": 37750,
      "train_runtime": 367072.5985,
      "train_tokens_per_second": 29245.647
    },
    {
      "epoch": 1.3360231044632216,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2879282036698465e-05,
      "loss": 0.38977029323577883,
      "num_input_tokens_seen": 10738120000,
      "step": 37760,
      "train_runtime": 367168.2674,
      "train_tokens_per_second": 29245.774
    },
    {
      "epoch": 1.3363769247269501,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2876887127913084e-05,
      "loss": 0.39130654335021975,
      "num_input_tokens_seen": 10740943936,
      "step": 37770,
      "train_runtime": 367266.4072,
      "train_tokens_per_second": 29245.648
    },
    {
      "epoch": 1.3367307449906791,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2874492971037785e-05,
      "loss": 0.38621296882629397,
      "num_input_tokens_seen": 10743787392,
      "step": 37780,
      "train_runtime": 367362.5428,
      "train_tokens_per_second": 29245.735
    },
    {
      "epoch": 1.3370845652544079,
      "grad_norm": 0.609375,
      "learning_rate": 2.2872099565679195e-05,
      "loss": 0.39300858974456787,
      "num_input_tokens_seen": 10746590016,
      "step": 37790,
      "train_runtime": 367456.0712,
      "train_tokens_per_second": 29245.918
    },
    {
      "epoch": 1.3374383855181367,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2869706911444235e-05,
      "loss": 0.3949438095092773,
      "num_input_tokens_seen": 10749435136,
      "step": 37800,
      "train_runtime": 367551.7817,
      "train_tokens_per_second": 29246.043
    },
    {
      "epoch": 1.3377922057818654,
      "grad_norm": 0.68359375,
      "learning_rate": 2.286731500794011e-05,
      "loss": 0.389025616645813,
      "num_input_tokens_seen": 10752276416,
      "step": 37810,
      "train_runtime": 367652.5325,
      "train_tokens_per_second": 29245.756
    },
    {
      "epoch": 1.3381460260455942,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2864923854774315e-05,
      "loss": 0.3898641347885132,
      "num_input_tokens_seen": 10755136384,
      "step": 37820,
      "train_runtime": 367748.2194,
      "train_tokens_per_second": 29245.924
    },
    {
      "epoch": 1.338499846309323,
      "grad_norm": 0.59765625,
      "learning_rate": 2.286253345155463e-05,
      "loss": 0.39429464340209963,
      "num_input_tokens_seen": 10758000256,
      "step": 37830,
      "train_runtime": 367844.7631,
      "train_tokens_per_second": 29246.033
    },
    {
      "epoch": 1.3388536665730517,
      "grad_norm": 0.6015625,
      "learning_rate": 2.286014379788912e-05,
      "loss": 0.391705322265625,
      "num_input_tokens_seen": 10760865920,
      "step": 37840,
      "train_runtime": 367943.1908,
      "train_tokens_per_second": 29245.998
    },
    {
      "epoch": 1.3392074868367805,
      "grad_norm": 0.6171875,
      "learning_rate": 2.285775489338615e-05,
      "loss": 0.3914210557937622,
      "num_input_tokens_seen": 10763731520,
      "step": 37850,
      "train_runtime": 368043.1102,
      "train_tokens_per_second": 29245.844
    },
    {
      "epoch": 1.3395613071005092,
      "grad_norm": 0.60546875,
      "learning_rate": 2.285536673765435e-05,
      "loss": 0.38653855323791503,
      "num_input_tokens_seen": 10766653312,
      "step": 37860,
      "train_runtime": 368145.8487,
      "train_tokens_per_second": 29245.619
    },
    {
      "epoch": 1.339915127364238,
      "grad_norm": 0.58984375,
      "learning_rate": 2.285297933030265e-05,
      "loss": 0.3879828929901123,
      "num_input_tokens_seen": 10769527616,
      "step": 37870,
      "train_runtime": 368244.8072,
      "train_tokens_per_second": 29245.565
    },
    {
      "epoch": 1.3402689476279668,
      "grad_norm": 0.640625,
      "learning_rate": 2.2850592670940272e-05,
      "loss": 0.38802452087402345,
      "num_input_tokens_seen": 10772364544,
      "step": 37880,
      "train_runtime": 368341.1486,
      "train_tokens_per_second": 29245.618
    },
    {
      "epoch": 1.3406227678916955,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2848206759176703e-05,
      "loss": 0.38828909397125244,
      "num_input_tokens_seen": 10775186496,
      "step": 37890,
      "train_runtime": 368434.1479,
      "train_tokens_per_second": 29245.895
    },
    {
      "epoch": 1.3409765881554243,
      "grad_norm": 0.6015625,
      "learning_rate": 2.284582159462173e-05,
      "loss": 0.3904193162918091,
      "num_input_tokens_seen": 10778001152,
      "step": 37900,
      "train_runtime": 368528.8872,
      "train_tokens_per_second": 29246.014
    },
    {
      "epoch": 1.3413304084191533,
      "grad_norm": 0.5859375,
      "learning_rate": 2.284343717688543e-05,
      "loss": 0.3872607469558716,
      "num_input_tokens_seen": 10780771584,
      "step": 37910,
      "train_runtime": 368620.7312,
      "train_tokens_per_second": 29246.243
    },
    {
      "epoch": 1.3416842286828818,
      "grad_norm": 0.59765625,
      "learning_rate": 2.284105350557815e-05,
      "loss": 0.3839366912841797,
      "num_input_tokens_seen": 10783595904,
      "step": 37920,
      "train_runtime": 368716.4703,
      "train_tokens_per_second": 29246.309
    },
    {
      "epoch": 1.3420380489466108,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2838670580310526e-05,
      "loss": 0.3887828826904297,
      "num_input_tokens_seen": 10786434432,
      "step": 37930,
      "train_runtime": 368814.158,
      "train_tokens_per_second": 29246.259
    },
    {
      "epoch": 1.3423918692103396,
      "grad_norm": 0.58984375,
      "learning_rate": 2.283628840069349e-05,
      "loss": 0.3892253875732422,
      "num_input_tokens_seen": 10789229440,
      "step": 37940,
      "train_runtime": 368911.9678,
      "train_tokens_per_second": 29246.081
    },
    {
      "epoch": 1.3427456894740684,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2833906966338237e-05,
      "loss": 0.3865851402282715,
      "num_input_tokens_seen": 10792035136,
      "step": 37950,
      "train_runtime": 369005.6976,
      "train_tokens_per_second": 29246.256
    },
    {
      "epoch": 1.3430995097377971,
      "grad_norm": 0.625,
      "learning_rate": 2.2831526276856274e-05,
      "loss": 0.38839163780212405,
      "num_input_tokens_seen": 10794873920,
      "step": 37960,
      "train_runtime": 369102.7428,
      "train_tokens_per_second": 29246.258
    },
    {
      "epoch": 1.343453330001526,
      "grad_norm": 0.60546875,
      "learning_rate": 2.282914633185936e-05,
      "loss": 0.39214296340942384,
      "num_input_tokens_seen": 10797723776,
      "step": 37970,
      "train_runtime": 369199.4787,
      "train_tokens_per_second": 29246.314
    },
    {
      "epoch": 1.3438071502652547,
      "grad_norm": 0.6328125,
      "learning_rate": 2.282676713095956e-05,
      "loss": 0.3928501129150391,
      "num_input_tokens_seen": 10800582464,
      "step": 37980,
      "train_runtime": 369296.7543,
      "train_tokens_per_second": 29246.351
    },
    {
      "epoch": 1.3441609705289834,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2824388673769212e-05,
      "loss": 0.38679208755493166,
      "num_input_tokens_seen": 10803405056,
      "step": 37990,
      "train_runtime": 369393.6148,
      "train_tokens_per_second": 29246.323
    },
    {
      "epoch": 1.3445147907927122,
      "grad_norm": 0.59765625,
      "learning_rate": 2.282201095990094e-05,
      "loss": 0.3891578435897827,
      "num_input_tokens_seen": 10806262080,
      "step": 38000,
      "train_runtime": 369490.0114,
      "train_tokens_per_second": 29246.425
    },
    {
      "epoch": 1.344868611056441,
      "grad_norm": 0.61328125,
      "learning_rate": 2.281963398896765e-05,
      "loss": 0.38672671318054197,
      "num_input_tokens_seen": 10809123968,
      "step": 38010,
      "train_runtime": 369589.1446,
      "train_tokens_per_second": 29246.324
    },
    {
      "epoch": 1.3452224313201697,
      "grad_norm": 0.6328125,
      "learning_rate": 2.281725776058253e-05,
      "loss": 0.38184943199157717,
      "num_input_tokens_seen": 10812005120,
      "step": 38020,
      "train_runtime": 369689.4453,
      "train_tokens_per_second": 29246.183
    },
    {
      "epoch": 1.3455762515838985,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2814882274359045e-05,
      "loss": 0.38944220542907715,
      "num_input_tokens_seen": 10814812608,
      "step": 38030,
      "train_runtime": 369784.863,
      "train_tokens_per_second": 29246.229
    },
    {
      "epoch": 1.3459300718476273,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2812507529910955e-05,
      "loss": 0.3935706377029419,
      "num_input_tokens_seen": 10817676160,
      "step": 38040,
      "train_runtime": 369886.1042,
      "train_tokens_per_second": 29245.965
    },
    {
      "epoch": 1.346283892111356,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2810133526852288e-05,
      "loss": 0.39360015392303466,
      "num_input_tokens_seen": 10820531584,
      "step": 38050,
      "train_runtime": 369981.3543,
      "train_tokens_per_second": 29246.154
    },
    {
      "epoch": 1.3466377123750848,
      "grad_norm": 0.59375,
      "learning_rate": 2.2807760264797352e-05,
      "loss": 0.3902523279190063,
      "num_input_tokens_seen": 10823374656,
      "step": 38060,
      "train_runtime": 370076.2582,
      "train_tokens_per_second": 29246.336
    },
    {
      "epoch": 1.3469915326388135,
      "grad_norm": 0.59375,
      "learning_rate": 2.280538774336075e-05,
      "loss": 0.3879945516586304,
      "num_input_tokens_seen": 10826221568,
      "step": 38070,
      "train_runtime": 370172.4738,
      "train_tokens_per_second": 29246.425
    },
    {
      "epoch": 1.3473453529025425,
      "grad_norm": 0.60546875,
      "learning_rate": 2.280301596215735e-05,
      "loss": 0.3895974636077881,
      "num_input_tokens_seen": 10829106240,
      "step": 38080,
      "train_runtime": 370271.2613,
      "train_tokens_per_second": 29246.413
    },
    {
      "epoch": 1.347699173166271,
      "grad_norm": 0.5859375,
      "learning_rate": 2.280064492080231e-05,
      "loss": 0.39406661987304686,
      "num_input_tokens_seen": 10831973376,
      "step": 38090,
      "train_runtime": 370372.5527,
      "train_tokens_per_second": 29246.156
    },
    {
      "epoch": 1.34805299343,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2798274618911075e-05,
      "loss": 0.3909964323043823,
      "num_input_tokens_seen": 10834827968,
      "step": 38100,
      "train_runtime": 370471.3694,
      "train_tokens_per_second": 29246.06
    },
    {
      "epoch": 1.3484068136937288,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2795905056099345e-05,
      "loss": 0.38561186790466306,
      "num_input_tokens_seen": 10837663360,
      "step": 38110,
      "train_runtime": 370564.8954,
      "train_tokens_per_second": 29246.33
    },
    {
      "epoch": 1.3487606339574576,
      "grad_norm": 0.609375,
      "learning_rate": 2.2793536231983123e-05,
      "loss": 0.3913773536682129,
      "num_input_tokens_seen": 10840504384,
      "step": 38120,
      "train_runtime": 370662.5758,
      "train_tokens_per_second": 29246.288
    },
    {
      "epoch": 1.3491144542211864,
      "grad_norm": 0.6171875,
      "learning_rate": 2.279116814617868e-05,
      "loss": 0.39183220863342283,
      "num_input_tokens_seen": 10843375552,
      "step": 38130,
      "train_runtime": 370761.463,
      "train_tokens_per_second": 29246.231
    },
    {
      "epoch": 1.3494682744849151,
      "grad_norm": 0.625,
      "learning_rate": 2.278880079830257e-05,
      "loss": 0.3905796527862549,
      "num_input_tokens_seen": 10846235072,
      "step": 38140,
      "train_runtime": 370856.124,
      "train_tokens_per_second": 29246.477
    },
    {
      "epoch": 1.349822094748644,
      "grad_norm": 0.6171875,
      "learning_rate": 2.278643418797163e-05,
      "loss": 0.3909287691116333,
      "num_input_tokens_seen": 10849056384,
      "step": 38150,
      "train_runtime": 370950.4726,
      "train_tokens_per_second": 29246.644
    },
    {
      "epoch": 1.3501759150123727,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2784068314802963e-05,
      "loss": 0.390028977394104,
      "num_input_tokens_seen": 10851937984,
      "step": 38160,
      "train_runtime": 371051.7917,
      "train_tokens_per_second": 29246.424
    },
    {
      "epoch": 1.3505297352761014,
      "grad_norm": 0.6015625,
      "learning_rate": 2.278170317841396e-05,
      "loss": 0.39056925773620604,
      "num_input_tokens_seen": 10854759168,
      "step": 38170,
      "train_runtime": 371146.6208,
      "train_tokens_per_second": 29246.553
    },
    {
      "epoch": 1.3508835555398302,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2779338778422287e-05,
      "loss": 0.39166011810302737,
      "num_input_tokens_seen": 10857565568,
      "step": 38180,
      "train_runtime": 371243.4216,
      "train_tokens_per_second": 29246.486
    },
    {
      "epoch": 1.351237375803559,
      "grad_norm": 0.59375,
      "learning_rate": 2.2776975114445885e-05,
      "loss": 0.39177675247192384,
      "num_input_tokens_seen": 10860431872,
      "step": 38190,
      "train_runtime": 371343.3564,
      "train_tokens_per_second": 29246.334
    },
    {
      "epoch": 1.3515911960672877,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2774612186102988e-05,
      "loss": 0.39413065910339357,
      "num_input_tokens_seen": 10863265536,
      "step": 38200,
      "train_runtime": 371439.2387,
      "train_tokens_per_second": 29246.413
    },
    {
      "epoch": 1.3519450163310165,
      "grad_norm": 0.625,
      "learning_rate": 2.277224999301208e-05,
      "loss": 0.3913625001907349,
      "num_input_tokens_seen": 10866100608,
      "step": 38210,
      "train_runtime": 371537.5405,
      "train_tokens_per_second": 29246.306
    },
    {
      "epoch": 1.3522988365947453,
      "grad_norm": 0.625,
      "learning_rate": 2.276988853479194e-05,
      "loss": 0.3887608528137207,
      "num_input_tokens_seen": 10868931328,
      "step": 38220,
      "train_runtime": 371637.8878,
      "train_tokens_per_second": 29246.026
    },
    {
      "epoch": 1.352652656858474,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2767527811061636e-05,
      "loss": 0.3916268587112427,
      "num_input_tokens_seen": 10871714048,
      "step": 38230,
      "train_runtime": 371731.4685,
      "train_tokens_per_second": 29246.149
    },
    {
      "epoch": 1.3530064771222028,
      "grad_norm": 0.609375,
      "learning_rate": 2.2765167821440473e-05,
      "loss": 0.3936856746673584,
      "num_input_tokens_seen": 10874593664,
      "step": 38240,
      "train_runtime": 371830.4005,
      "train_tokens_per_second": 29246.112
    },
    {
      "epoch": 1.3533602973859318,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2762808565548063e-05,
      "loss": 0.38674130439758303,
      "num_input_tokens_seen": 10877424960,
      "step": 38250,
      "train_runtime": 371927.4276,
      "train_tokens_per_second": 29246.095
    },
    {
      "epoch": 1.3537141176496603,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2760450043004303e-05,
      "loss": 0.3925855398178101,
      "num_input_tokens_seen": 10880254464,
      "step": 38260,
      "train_runtime": 372024.7213,
      "train_tokens_per_second": 29246.052
    },
    {
      "epoch": 1.3540679379133893,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2758092253429327e-05,
      "loss": 0.38815057277679443,
      "num_input_tokens_seen": 10883046016,
      "step": 38270,
      "train_runtime": 372117.903,
      "train_tokens_per_second": 29246.231
    },
    {
      "epoch": 1.354421758177118,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2755735196443585e-05,
      "loss": 0.39426164627075194,
      "num_input_tokens_seen": 10885894336,
      "step": 38280,
      "train_runtime": 372215.8005,
      "train_tokens_per_second": 29246.191
    },
    {
      "epoch": 1.3547755784408468,
      "grad_norm": 0.609375,
      "learning_rate": 2.2753378871667773e-05,
      "loss": 0.3930710792541504,
      "num_input_tokens_seen": 10888731584,
      "step": 38290,
      "train_runtime": 372312.7758,
      "train_tokens_per_second": 29246.194
    },
    {
      "epoch": 1.3551293987045756,
      "grad_norm": 0.62109375,
      "learning_rate": 2.275102327872288e-05,
      "loss": 0.3935257911682129,
      "num_input_tokens_seen": 10891600512,
      "step": 38300,
      "train_runtime": 372410.3619,
      "train_tokens_per_second": 29246.234
    },
    {
      "epoch": 1.3554832189683044,
      "grad_norm": 0.62109375,
      "learning_rate": 2.274866841723015e-05,
      "loss": 0.39020767211914065,
      "num_input_tokens_seen": 10894444864,
      "step": 38310,
      "train_runtime": 372504.6177,
      "train_tokens_per_second": 29246.469
    },
    {
      "epoch": 1.3558370392320331,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2746314286811134e-05,
      "loss": 0.38763995170593263,
      "num_input_tokens_seen": 10897268608,
      "step": 38320,
      "train_runtime": 372602.1022,
      "train_tokens_per_second": 29246.396
    },
    {
      "epoch": 1.356190859495762,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2743960887087616e-05,
      "loss": 0.3916309833526611,
      "num_input_tokens_seen": 10900054720,
      "step": 38330,
      "train_runtime": 372696.1836,
      "train_tokens_per_second": 29246.489
    },
    {
      "epoch": 1.3565446797594907,
      "grad_norm": 0.6171875,
      "learning_rate": 2.274160821768169e-05,
      "loss": 0.39238176345825193,
      "num_input_tokens_seen": 10902898240,
      "step": 38340,
      "train_runtime": 372794.7408,
      "train_tokens_per_second": 29246.384
    },
    {
      "epoch": 1.3568985000232194,
      "grad_norm": 0.609375,
      "learning_rate": 2.2739256278215697e-05,
      "loss": 0.3912178039550781,
      "num_input_tokens_seen": 10905727360,
      "step": 38350,
      "train_runtime": 372892.5124,
      "train_tokens_per_second": 29246.303
    },
    {
      "epoch": 1.3572523202869482,
      "grad_norm": 0.60546875,
      "learning_rate": 2.273690506831227e-05,
      "loss": 0.39323103427886963,
      "num_input_tokens_seen": 10908560768,
      "step": 38360,
      "train_runtime": 372990.561,
      "train_tokens_per_second": 29246.211
    },
    {
      "epoch": 1.357606140550677,
      "grad_norm": 0.59765625,
      "learning_rate": 2.273455458759431e-05,
      "loss": 0.3869377851486206,
      "num_input_tokens_seen": 10911420800,
      "step": 38370,
      "train_runtime": 373087.0696,
      "train_tokens_per_second": 29246.312
    },
    {
      "epoch": 1.3579599608144057,
      "grad_norm": 0.609375,
      "learning_rate": 2.2732204835684982e-05,
      "loss": 0.3878058671951294,
      "num_input_tokens_seen": 10914297472,
      "step": 38380,
      "train_runtime": 373186.1111,
      "train_tokens_per_second": 29246.258
    },
    {
      "epoch": 1.3583137810781345,
      "grad_norm": 0.609375,
      "learning_rate": 2.2729855812207737e-05,
      "loss": 0.3878204107284546,
      "num_input_tokens_seen": 10917116480,
      "step": 38390,
      "train_runtime": 373282.4818,
      "train_tokens_per_second": 29246.26
    },
    {
      "epoch": 1.3586676013418635,
      "grad_norm": 0.63671875,
      "learning_rate": 2.272750751678628e-05,
      "loss": 0.38982443809509276,
      "num_input_tokens_seen": 10919937344,
      "step": 38400,
      "train_runtime": 373378.6009,
      "train_tokens_per_second": 29246.286
    },
    {
      "epoch": 1.359021421605592,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2725159949044613e-05,
      "loss": 0.39234676361083987,
      "num_input_tokens_seen": 10922721856,
      "step": 38410,
      "train_runtime": 373472.5134,
      "train_tokens_per_second": 29246.388
    },
    {
      "epoch": 1.359375241869321,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2722813108606992e-05,
      "loss": 0.38744635581970216,
      "num_input_tokens_seen": 10925587264,
      "step": 38420,
      "train_runtime": 373569.6247,
      "train_tokens_per_second": 29246.455
    },
    {
      "epoch": 1.3597290621330496,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2720466995097943e-05,
      "loss": 0.39301509857177735,
      "num_input_tokens_seen": 10928521728,
      "step": 38430,
      "train_runtime": 373674.4998,
      "train_tokens_per_second": 29246.1
    },
    {
      "epoch": 1.3600828823967785,
      "grad_norm": 0.6015625,
      "learning_rate": 2.271812160814228e-05,
      "loss": 0.39125211238861085,
      "num_input_tokens_seen": 10931379456,
      "step": 38440,
      "train_runtime": 373775.0728,
      "train_tokens_per_second": 29245.876
    },
    {
      "epoch": 1.3604367026605073,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2715776947365065e-05,
      "loss": 0.3927396535873413,
      "num_input_tokens_seen": 10934192128,
      "step": 38450,
      "train_runtime": 373866.461,
      "train_tokens_per_second": 29246.25
    },
    {
      "epoch": 1.360790522924236,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2713433012391654e-05,
      "loss": 0.3895401954650879,
      "num_input_tokens_seen": 10936995136,
      "step": 38460,
      "train_runtime": 373960.6791,
      "train_tokens_per_second": 29246.377
    },
    {
      "epoch": 1.3611443431879648,
      "grad_norm": 0.609375,
      "learning_rate": 2.2711089802847657e-05,
      "loss": 0.3885706901550293,
      "num_input_tokens_seen": 10939837696,
      "step": 38470,
      "train_runtime": 374059.2763,
      "train_tokens_per_second": 29246.268
    },
    {
      "epoch": 1.3614981634516936,
      "grad_norm": 0.625,
      "learning_rate": 2.270874731835896e-05,
      "loss": 0.38973586559295653,
      "num_input_tokens_seen": 10942637888,
      "step": 38480,
      "train_runtime": 374155.5545,
      "train_tokens_per_second": 29246.226
    },
    {
      "epoch": 1.3618519837154224,
      "grad_norm": 0.61328125,
      "learning_rate": 2.270640555855172e-05,
      "loss": 0.3894874095916748,
      "num_input_tokens_seen": 10945469248,
      "step": 38490,
      "train_runtime": 374252.4342,
      "train_tokens_per_second": 29246.221
    },
    {
      "epoch": 1.3622058039791511,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2704064523052372e-05,
      "loss": 0.39103376865386963,
      "num_input_tokens_seen": 10948361600,
      "step": 38500,
      "train_runtime": 374355.9209,
      "train_tokens_per_second": 29245.862
    },
    {
      "epoch": 1.36255962424288,
      "grad_norm": 0.6640625,
      "learning_rate": 2.270172421148759e-05,
      "loss": 0.392916202545166,
      "num_input_tokens_seen": 10951189440,
      "step": 38510,
      "train_runtime": 374452.5684,
      "train_tokens_per_second": 29245.865
    },
    {
      "epoch": 1.3629134445066087,
      "grad_norm": 0.6171875,
      "learning_rate": 2.269938462348436e-05,
      "loss": 0.3911635398864746,
      "num_input_tokens_seen": 10954091712,
      "step": 38520,
      "train_runtime": 374550.9593,
      "train_tokens_per_second": 29245.932
    },
    {
      "epoch": 1.3632672647703374,
      "grad_norm": 0.60546875,
      "learning_rate": 2.26970457586699e-05,
      "loss": 0.3889798402786255,
      "num_input_tokens_seen": 10956958592,
      "step": 38530,
      "train_runtime": 374647.7384,
      "train_tokens_per_second": 29246.029
    },
    {
      "epoch": 1.3636210850340662,
      "grad_norm": 0.609375,
      "learning_rate": 2.2694707616671726e-05,
      "loss": 0.39226632118225097,
      "num_input_tokens_seen": 10959828992,
      "step": 38540,
      "train_runtime": 374745.2825,
      "train_tokens_per_second": 29246.076
    },
    {
      "epoch": 1.363974905297795,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2692370197117605e-05,
      "loss": 0.38927111625671384,
      "num_input_tokens_seen": 10962646848,
      "step": 38550,
      "train_runtime": 374841.2973,
      "train_tokens_per_second": 29246.102
    },
    {
      "epoch": 1.3643287255615237,
      "grad_norm": 0.62890625,
      "learning_rate": 2.269003349963557e-05,
      "loss": 0.386190128326416,
      "num_input_tokens_seen": 10965488256,
      "step": 38560,
      "train_runtime": 374935.5501,
      "train_tokens_per_second": 29246.328
    },
    {
      "epoch": 1.3646825458252527,
      "grad_norm": 0.609375,
      "learning_rate": 2.268769752385393e-05,
      "loss": 0.38511104583740235,
      "num_input_tokens_seen": 10968309696,
      "step": 38570,
      "train_runtime": 375030.3355,
      "train_tokens_per_second": 29246.46
    },
    {
      "epoch": 1.3650363660889813,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2685362269401272e-05,
      "loss": 0.3887251615524292,
      "num_input_tokens_seen": 10971198272,
      "step": 38580,
      "train_runtime": 375127.6258,
      "train_tokens_per_second": 29246.575
    },
    {
      "epoch": 1.3653901863527103,
      "grad_norm": 0.609375,
      "learning_rate": 2.268302773590643e-05,
      "loss": 0.3868979215621948,
      "num_input_tokens_seen": 10974057664,
      "step": 38590,
      "train_runtime": 375225.2979,
      "train_tokens_per_second": 29246.583
    },
    {
      "epoch": 1.3657440066164388,
      "grad_norm": 0.578125,
      "learning_rate": 2.268069392299851e-05,
      "loss": 0.3841234683990479,
      "num_input_tokens_seen": 10976914496,
      "step": 38600,
      "train_runtime": 375323.7025,
      "train_tokens_per_second": 29246.526
    },
    {
      "epoch": 1.3660978268801678,
      "grad_norm": 0.65234375,
      "learning_rate": 2.267836083030689e-05,
      "loss": 0.39037380218505857,
      "num_input_tokens_seen": 10979768960,
      "step": 38610,
      "train_runtime": 375422.6566,
      "train_tokens_per_second": 29246.421
    },
    {
      "epoch": 1.3664516471438966,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2676028457461225e-05,
      "loss": 0.38668346405029297,
      "num_input_tokens_seen": 10982613952,
      "step": 38620,
      "train_runtime": 375519.797,
      "train_tokens_per_second": 29246.431
    },
    {
      "epoch": 1.3668054674076253,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2673696804091418e-05,
      "loss": 0.38927206993103025,
      "num_input_tokens_seen": 10985410048,
      "step": 38630,
      "train_runtime": 375612.8179,
      "train_tokens_per_second": 29246.633
    },
    {
      "epoch": 1.367159287671354,
      "grad_norm": 0.59375,
      "learning_rate": 2.267136586982764e-05,
      "loss": 0.38597822189331055,
      "num_input_tokens_seen": 10988300032,
      "step": 38640,
      "train_runtime": 375711.3313,
      "train_tokens_per_second": 29246.656
    },
    {
      "epoch": 1.3675131079350829,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2669035654300346e-05,
      "loss": 0.38661832809448243,
      "num_input_tokens_seen": 10991144128,
      "step": 38650,
      "train_runtime": 375810.4459,
      "train_tokens_per_second": 29246.51
    },
    {
      "epoch": 1.3678669281988116,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2666706157140242e-05,
      "loss": 0.3878067970275879,
      "num_input_tokens_seen": 10994020800,
      "step": 38660,
      "train_runtime": 375909.6378,
      "train_tokens_per_second": 29246.446
    },
    {
      "epoch": 1.3682207484625404,
      "grad_norm": 0.59375,
      "learning_rate": 2.2664377377978295e-05,
      "loss": 0.38732447624206545,
      "num_input_tokens_seen": 10996851072,
      "step": 38670,
      "train_runtime": 376006.2827,
      "train_tokens_per_second": 29246.456
    },
    {
      "epoch": 1.3685745687262691,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2662049316445754e-05,
      "loss": 0.39151811599731445,
      "num_input_tokens_seen": 10999660864,
      "step": 38680,
      "train_runtime": 376100.6271,
      "train_tokens_per_second": 29246.59
    },
    {
      "epoch": 1.368928388989998,
      "grad_norm": 0.59375,
      "learning_rate": 2.2659721972174117e-05,
      "loss": 0.3887642383575439,
      "num_input_tokens_seen": 11002482368,
      "step": 38690,
      "train_runtime": 376198.4601,
      "train_tokens_per_second": 29246.484
    },
    {
      "epoch": 1.3692822092537267,
      "grad_norm": 0.609375,
      "learning_rate": 2.2657395344795153e-05,
      "loss": 0.39536571502685547,
      "num_input_tokens_seen": 11005339264,
      "step": 38700,
      "train_runtime": 376296.1956,
      "train_tokens_per_second": 29246.48
    },
    {
      "epoch": 1.3696360295174554,
      "grad_norm": 0.59765625,
      "learning_rate": 2.265506943394091e-05,
      "loss": 0.3898845911026001,
      "num_input_tokens_seen": 11008131072,
      "step": 38710,
      "train_runtime": 376387.4887,
      "train_tokens_per_second": 29246.804
    },
    {
      "epoch": 1.3699898497811842,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2652744239243668e-05,
      "loss": 0.38378276824951174,
      "num_input_tokens_seen": 11010988160,
      "step": 38720,
      "train_runtime": 376484.812,
      "train_tokens_per_second": 29246.832
    },
    {
      "epoch": 1.370343670044913,
      "grad_norm": 0.640625,
      "learning_rate": 2.2650419760336006e-05,
      "loss": 0.3887014389038086,
      "num_input_tokens_seen": 11013866176,
      "step": 38730,
      "train_runtime": 376584.8316,
      "train_tokens_per_second": 29246.707
    },
    {
      "epoch": 1.370697490308642,
      "grad_norm": 0.6328125,
      "learning_rate": 2.2648095996850735e-05,
      "loss": 0.3874682903289795,
      "num_input_tokens_seen": 11016722496,
      "step": 38740,
      "train_runtime": 376682.2946,
      "train_tokens_per_second": 29246.722
    },
    {
      "epoch": 1.3710513105723705,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2645772948420957e-05,
      "loss": 0.3867891073226929,
      "num_input_tokens_seen": 11019531904,
      "step": 38750,
      "train_runtime": 376776.9646,
      "train_tokens_per_second": 29246.83
    },
    {
      "epoch": 1.3714051308360995,
      "grad_norm": 0.58203125,
      "learning_rate": 2.264345061468003e-05,
      "loss": 0.38934223651885985,
      "num_input_tokens_seen": 11022393792,
      "step": 38760,
      "train_runtime": 376876.3004,
      "train_tokens_per_second": 29246.715
    },
    {
      "epoch": 1.371758951099828,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2641128995261554e-05,
      "loss": 0.3914985656738281,
      "num_input_tokens_seen": 11025283008,
      "step": 38770,
      "train_runtime": 376976.6337,
      "train_tokens_per_second": 29246.595
    },
    {
      "epoch": 1.372112771363557,
      "grad_norm": 0.640625,
      "learning_rate": 2.2638808089799417e-05,
      "loss": 0.38730630874633787,
      "num_input_tokens_seen": 11028088128,
      "step": 38780,
      "train_runtime": 377072.1014,
      "train_tokens_per_second": 29246.63
    },
    {
      "epoch": 1.3724665916272858,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2636487897927773e-05,
      "loss": 0.3903202056884766,
      "num_input_tokens_seen": 11030944000,
      "step": 38790,
      "train_runtime": 377172.8151,
      "train_tokens_per_second": 29246.392
    },
    {
      "epoch": 1.3728204118910146,
      "grad_norm": 0.609375,
      "learning_rate": 2.2634168419281007e-05,
      "loss": 0.3899411678314209,
      "num_input_tokens_seen": 11033770048,
      "step": 38800,
      "train_runtime": 377268.3092,
      "train_tokens_per_second": 29246.48
    },
    {
      "epoch": 1.3731742321547433,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2631849653493803e-05,
      "loss": 0.3898441791534424,
      "num_input_tokens_seen": 11036662784,
      "step": 38810,
      "train_runtime": 377364.9128,
      "train_tokens_per_second": 29246.659
    },
    {
      "epoch": 1.373528052418472,
      "grad_norm": 0.609375,
      "learning_rate": 2.2629531600201077e-05,
      "loss": 0.3927628517150879,
      "num_input_tokens_seen": 11039528064,
      "step": 38820,
      "train_runtime": 377462.1501,
      "train_tokens_per_second": 29246.715
    },
    {
      "epoch": 1.3738818726822009,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2627214259038028e-05,
      "loss": 0.3915828227996826,
      "num_input_tokens_seen": 11042412992,
      "step": 38830,
      "train_runtime": 377561.3635,
      "train_tokens_per_second": 29246.671
    },
    {
      "epoch": 1.3742356929459296,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2624897629640108e-05,
      "loss": 0.39042153358459475,
      "num_input_tokens_seen": 11045297216,
      "step": 38840,
      "train_runtime": 377660.8228,
      "train_tokens_per_second": 29246.606
    },
    {
      "epoch": 1.3745895132096584,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2622581711643022e-05,
      "loss": 0.39333410263061525,
      "num_input_tokens_seen": 11048189056,
      "step": 38850,
      "train_runtime": 377761.0502,
      "train_tokens_per_second": 29246.501
    },
    {
      "epoch": 1.3749433334733872,
      "grad_norm": 0.6171875,
      "learning_rate": 2.262026650468275e-05,
      "loss": 0.3897092819213867,
      "num_input_tokens_seen": 11051043840,
      "step": 38860,
      "train_runtime": 377858.3855,
      "train_tokens_per_second": 29246.523
    },
    {
      "epoch": 1.375297153737116,
      "grad_norm": 0.61328125,
      "learning_rate": 2.261795200839553e-05,
      "loss": 0.39409472942352297,
      "num_input_tokens_seen": 11053920512,
      "step": 38870,
      "train_runtime": 377958.2752,
      "train_tokens_per_second": 29246.404
    },
    {
      "epoch": 1.3756509740008447,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2615638222417856e-05,
      "loss": 0.39360010623931885,
      "num_input_tokens_seen": 11056773632,
      "step": 38880,
      "train_runtime": 378059.8846,
      "train_tokens_per_second": 29246.091
    },
    {
      "epoch": 1.3760047942645734,
      "grad_norm": 0.59375,
      "learning_rate": 2.261332514638648e-05,
      "loss": 0.38776426315307616,
      "num_input_tokens_seen": 11059602944,
      "step": 38890,
      "train_runtime": 378156.459,
      "train_tokens_per_second": 29246.104
    },
    {
      "epoch": 1.3763586145283022,
      "grad_norm": 0.5859375,
      "learning_rate": 2.261101277993842e-05,
      "loss": 0.3868075370788574,
      "num_input_tokens_seen": 11062481856,
      "step": 38900,
      "train_runtime": 378255.3864,
      "train_tokens_per_second": 29246.066
    },
    {
      "epoch": 1.3767124347920312,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2608701122710953e-05,
      "loss": 0.39182395935058595,
      "num_input_tokens_seen": 11065335296,
      "step": 38910,
      "train_runtime": 378351.5243,
      "train_tokens_per_second": 29246.176
    },
    {
      "epoch": 1.3770662550557597,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2606390174341617e-05,
      "loss": 0.39125618934631345,
      "num_input_tokens_seen": 11068160768,
      "step": 38920,
      "train_runtime": 378447.1497,
      "train_tokens_per_second": 29246.252
    },
    {
      "epoch": 1.3774200753194887,
      "grad_norm": 0.63671875,
      "learning_rate": 2.2604079934468193e-05,
      "loss": 0.3885795116424561,
      "num_input_tokens_seen": 11070971520,
      "step": 38930,
      "train_runtime": 378542.2916,
      "train_tokens_per_second": 29246.327
    },
    {
      "epoch": 1.3777738955832175,
      "grad_norm": 0.5859375,
      "learning_rate": 2.260177040272875e-05,
      "loss": 0.3865159511566162,
      "num_input_tokens_seen": 11073798144,
      "step": 38940,
      "train_runtime": 378637.9301,
      "train_tokens_per_second": 29246.405
    },
    {
      "epoch": 1.3781277158469463,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2599461578761593e-05,
      "loss": 0.3925990104675293,
      "num_input_tokens_seen": 11076636224,
      "step": 38950,
      "train_runtime": 378735.2566,
      "train_tokens_per_second": 29246.383
    },
    {
      "epoch": 1.378481536110675,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2597153462205292e-05,
      "loss": 0.3931520938873291,
      "num_input_tokens_seen": 11079494464,
      "step": 38960,
      "train_runtime": 378833.8381,
      "train_tokens_per_second": 29246.317
    },
    {
      "epoch": 1.3788353563744038,
      "grad_norm": 0.609375,
      "learning_rate": 2.259484605269868e-05,
      "loss": 0.3876009941101074,
      "num_input_tokens_seen": 11082340352,
      "step": 38970,
      "train_runtime": 378932.6962,
      "train_tokens_per_second": 29246.197
    },
    {
      "epoch": 1.3791891766381326,
      "grad_norm": 0.61328125,
      "learning_rate": 2.259253934988084e-05,
      "loss": 0.393019962310791,
      "num_input_tokens_seen": 11085170816,
      "step": 38980,
      "train_runtime": 379030.247,
      "train_tokens_per_second": 29246.138
    },
    {
      "epoch": 1.3795429969018613,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2590233353391124e-05,
      "loss": 0.387861442565918,
      "num_input_tokens_seen": 11087992192,
      "step": 38990,
      "train_runtime": 379125.7156,
      "train_tokens_per_second": 29246.215
    },
    {
      "epoch": 1.37989681716559,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2587928062869124e-05,
      "loss": 0.39212684631347655,
      "num_input_tokens_seen": 11090834624,
      "step": 39000,
      "train_runtime": 379222.6804,
      "train_tokens_per_second": 29246.232
    },
    {
      "epoch": 1.3802506374293189,
      "grad_norm": 0.59765625,
      "learning_rate": 2.258562347795471e-05,
      "loss": 0.3850613832473755,
      "num_input_tokens_seen": 11093671168,
      "step": 39010,
      "train_runtime": 379320.2688,
      "train_tokens_per_second": 29246.186
    },
    {
      "epoch": 1.3806044576930476,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2583319598287993e-05,
      "loss": 0.38744335174560546,
      "num_input_tokens_seen": 11096512000,
      "step": 39020,
      "train_runtime": 379421.3887,
      "train_tokens_per_second": 29245.879
    },
    {
      "epoch": 1.3809582779567764,
      "grad_norm": 0.59375,
      "learning_rate": 2.258101642350935e-05,
      "loss": 0.3917128324508667,
      "num_input_tokens_seen": 11099313280,
      "step": 39030,
      "train_runtime": 379514.7702,
      "train_tokens_per_second": 29246.064
    },
    {
      "epoch": 1.3813120982205052,
      "grad_norm": 0.58203125,
      "learning_rate": 2.2578713953259414e-05,
      "loss": 0.38610689640045165,
      "num_input_tokens_seen": 11102135744,
      "step": 39040,
      "train_runtime": 379611.9395,
      "train_tokens_per_second": 29246.013
    },
    {
      "epoch": 1.381665918484234,
      "grad_norm": 0.64453125,
      "learning_rate": 2.2576412187179074e-05,
      "loss": 0.3876628398895264,
      "num_input_tokens_seen": 11104966592,
      "step": 39050,
      "train_runtime": 379710.1414,
      "train_tokens_per_second": 29245.905
    },
    {
      "epoch": 1.3820197387479627,
      "grad_norm": 0.60546875,
      "learning_rate": 2.257411112490946e-05,
      "loss": 0.3869986057281494,
      "num_input_tokens_seen": 11107825536,
      "step": 39060,
      "train_runtime": 379809.8551,
      "train_tokens_per_second": 29245.754
    },
    {
      "epoch": 1.3823735590116915,
      "grad_norm": 0.58984375,
      "learning_rate": 2.257181076609199e-05,
      "loss": 0.39197072982788084,
      "num_input_tokens_seen": 11110701056,
      "step": 39070,
      "train_runtime": 379908.152,
      "train_tokens_per_second": 29245.756
    },
    {
      "epoch": 1.3827273792754204,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2569511110368306e-05,
      "loss": 0.3927911758422852,
      "num_input_tokens_seen": 11113572480,
      "step": 39080,
      "train_runtime": 380004.3382,
      "train_tokens_per_second": 29245.909
    },
    {
      "epoch": 1.383081199539149,
      "grad_norm": 0.59765625,
      "learning_rate": 2.256721215738032e-05,
      "loss": 0.3901776075363159,
      "num_input_tokens_seen": 11116431744,
      "step": 39090,
      "train_runtime": 380104.2094,
      "train_tokens_per_second": 29245.748
    },
    {
      "epoch": 1.383435019802878,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2564913906770207e-05,
      "loss": 0.38931324481964114,
      "num_input_tokens_seen": 11119239296,
      "step": 39100,
      "train_runtime": 380197.4414,
      "train_tokens_per_second": 29245.96
    },
    {
      "epoch": 1.3837888400666067,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2562616358180384e-05,
      "loss": 0.38906543254852294,
      "num_input_tokens_seen": 11122072192,
      "step": 39110,
      "train_runtime": 380291.4675,
      "train_tokens_per_second": 29246.179
    },
    {
      "epoch": 1.3841426603303355,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2560319511253522e-05,
      "loss": 0.388606595993042,
      "num_input_tokens_seen": 11124964864,
      "step": 39120,
      "train_runtime": 380390.9688,
      "train_tokens_per_second": 29246.133
    },
    {
      "epoch": 1.3844964805940643,
      "grad_norm": 0.58203125,
      "learning_rate": 2.255802336563256e-05,
      "loss": 0.389971661567688,
      "num_input_tokens_seen": 11127832384,
      "step": 39130,
      "train_runtime": 380488.4181,
      "train_tokens_per_second": 29246.179
    },
    {
      "epoch": 1.384850300857793,
      "grad_norm": 0.625,
      "learning_rate": 2.255572792096067e-05,
      "loss": 0.3933126926422119,
      "num_input_tokens_seen": 11130690944,
      "step": 39140,
      "train_runtime": 380589.6429,
      "train_tokens_per_second": 29245.911
    },
    {
      "epoch": 1.3852041211215218,
      "grad_norm": 0.609375,
      "learning_rate": 2.2553433176881303e-05,
      "loss": 0.39327635765075686,
      "num_input_tokens_seen": 11133547840,
      "step": 39150,
      "train_runtime": 380688.7151,
      "train_tokens_per_second": 29245.805
    },
    {
      "epoch": 1.3855579413852506,
      "grad_norm": 0.625,
      "learning_rate": 2.2551139133038147e-05,
      "loss": 0.3896777153015137,
      "num_input_tokens_seen": 11136403328,
      "step": 39160,
      "train_runtime": 380787.163,
      "train_tokens_per_second": 29245.743
    },
    {
      "epoch": 1.3859117616489793,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2548845789075145e-05,
      "loss": 0.3917239189147949,
      "num_input_tokens_seen": 11139264832,
      "step": 39170,
      "train_runtime": 380885.4899,
      "train_tokens_per_second": 29245.705
    },
    {
      "epoch": 1.386265581912708,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2546553144636503e-05,
      "loss": 0.3911376714706421,
      "num_input_tokens_seen": 11142194368,
      "step": 39180,
      "train_runtime": 380990.5183,
      "train_tokens_per_second": 29245.332
    },
    {
      "epoch": 1.3866194021764369,
      "grad_norm": 0.61328125,
      "learning_rate": 2.254426119936668e-05,
      "loss": 0.3860174179077148,
      "num_input_tokens_seen": 11145013632,
      "step": 39190,
      "train_runtime": 381086.0626,
      "train_tokens_per_second": 29245.398
    },
    {
      "epoch": 1.3869732224401656,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2541969952910365e-05,
      "loss": 0.3878906726837158,
      "num_input_tokens_seen": 11147898752,
      "step": 39200,
      "train_runtime": 381184.88,
      "train_tokens_per_second": 29245.385
    },
    {
      "epoch": 1.3873270427038944,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2539679404912533e-05,
      "loss": 0.3896648645401001,
      "num_input_tokens_seen": 11150699776,
      "step": 39210,
      "train_runtime": 381280.9754,
      "train_tokens_per_second": 29245.361
    },
    {
      "epoch": 1.3876808629676232,
      "grad_norm": 0.59375,
      "learning_rate": 2.253738955501838e-05,
      "loss": 0.38965606689453125,
      "num_input_tokens_seen": 11153533632,
      "step": 39220,
      "train_runtime": 381377.9565,
      "train_tokens_per_second": 29245.355
    },
    {
      "epoch": 1.3880346832313522,
      "grad_norm": 0.609375,
      "learning_rate": 2.2535100402873383e-05,
      "loss": 0.3914762496948242,
      "num_input_tokens_seen": 11156396864,
      "step": 39230,
      "train_runtime": 381474.7135,
      "train_tokens_per_second": 29245.443
    },
    {
      "epoch": 1.3883885034950807,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2532811948123252e-05,
      "loss": 0.3892830371856689,
      "num_input_tokens_seen": 11159236160,
      "step": 39240,
      "train_runtime": 381574.2574,
      "train_tokens_per_second": 29245.254
    },
    {
      "epoch": 1.3887423237588097,
      "grad_norm": 0.609375,
      "learning_rate": 2.2530524190413955e-05,
      "loss": 0.3849909782409668,
      "num_input_tokens_seen": 11162052224,
      "step": 39250,
      "train_runtime": 381668.3629,
      "train_tokens_per_second": 29245.422
    },
    {
      "epoch": 1.3890961440225382,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2528237129391706e-05,
      "loss": 0.39268150329589846,
      "num_input_tokens_seen": 11164892736,
      "step": 39260,
      "train_runtime": 381767.2125,
      "train_tokens_per_second": 29245.29
    },
    {
      "epoch": 1.3894499642862672,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2525950764702985e-05,
      "loss": 0.39043426513671875,
      "num_input_tokens_seen": 11167730112,
      "step": 39270,
      "train_runtime": 381863.5923,
      "train_tokens_per_second": 29245.339
    },
    {
      "epoch": 1.389803784549996,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2523665095994505e-05,
      "loss": 0.38786580562591555,
      "num_input_tokens_seen": 11170539264,
      "step": 39280,
      "train_runtime": 381957.7934,
      "train_tokens_per_second": 29245.481
    },
    {
      "epoch": 1.3901576048137247,
      "grad_norm": 0.6328125,
      "learning_rate": 2.252138012291324e-05,
      "loss": 0.3902053594589233,
      "num_input_tokens_seen": 11173406528,
      "step": 39290,
      "train_runtime": 382058.2143,
      "train_tokens_per_second": 29245.299
    },
    {
      "epoch": 1.3905114250774535,
      "grad_norm": 0.58203125,
      "learning_rate": 2.2519095845106414e-05,
      "loss": 0.39381895065307615,
      "num_input_tokens_seen": 11176267392,
      "step": 39300,
      "train_runtime": 382155.9994,
      "train_tokens_per_second": 29245.301
    },
    {
      "epoch": 1.3908652453411823,
      "grad_norm": 0.59765625,
      "learning_rate": 2.25168122622215e-05,
      "loss": 0.38964109420776366,
      "num_input_tokens_seen": 11179142976,
      "step": 39310,
      "train_runtime": 382254.918,
      "train_tokens_per_second": 29245.256
    },
    {
      "epoch": 1.391219065604911,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2514529373906217e-05,
      "loss": 0.3899622678756714,
      "num_input_tokens_seen": 11181973504,
      "step": 39320,
      "train_runtime": 382349.174,
      "train_tokens_per_second": 29245.45
    },
    {
      "epoch": 1.3915728858686398,
      "grad_norm": 0.62109375,
      "learning_rate": 2.251224717980855e-05,
      "loss": 0.3894974708557129,
      "num_input_tokens_seen": 11184771456,
      "step": 39330,
      "train_runtime": 382441.6519,
      "train_tokens_per_second": 29245.694
    },
    {
      "epoch": 1.3919267061323686,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2509965679576713e-05,
      "loss": 0.39198126792907717,
      "num_input_tokens_seen": 11187658048,
      "step": 39340,
      "train_runtime": 382541.81,
      "train_tokens_per_second": 29245.582
    },
    {
      "epoch": 1.3922805263960973,
      "grad_norm": 0.59765625,
      "learning_rate": 2.250768487285918e-05,
      "loss": 0.38835816383361815,
      "num_input_tokens_seen": 11190528448,
      "step": 39350,
      "train_runtime": 382640.6157,
      "train_tokens_per_second": 29245.532
    },
    {
      "epoch": 1.392634346659826,
      "grad_norm": 0.60546875,
      "learning_rate": 2.250540475930467e-05,
      "loss": 0.3910823345184326,
      "num_input_tokens_seen": 11193331648,
      "step": 39360,
      "train_runtime": 382735.9042,
      "train_tokens_per_second": 29245.575
    },
    {
      "epoch": 1.3929881669235549,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2503125338562166e-05,
      "loss": 0.3896183013916016,
      "num_input_tokens_seen": 11196209024,
      "step": 39370,
      "train_runtime": 382834.6952,
      "train_tokens_per_second": 29245.544
    },
    {
      "epoch": 1.3933419871872836,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2500846610280868e-05,
      "loss": 0.38942005634307864,
      "num_input_tokens_seen": 11199065536,
      "step": 39380,
      "train_runtime": 382934.9725,
      "train_tokens_per_second": 29245.345
    },
    {
      "epoch": 1.3936958074510124,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2498568574110262e-05,
      "loss": 0.38760254383087156,
      "num_input_tokens_seen": 11201837056,
      "step": 39390,
      "train_runtime": 383027.5662,
      "train_tokens_per_second": 29245.511
    },
    {
      "epoch": 1.3940496277147414,
      "grad_norm": 0.609375,
      "learning_rate": 2.249629122970006e-05,
      "loss": 0.3866061449050903,
      "num_input_tokens_seen": 11204617408,
      "step": 39400,
      "train_runtime": 383120.8743,
      "train_tokens_per_second": 29245.646
    },
    {
      "epoch": 1.39440344797847,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2494014576700217e-05,
      "loss": 0.38678503036499023,
      "num_input_tokens_seen": 11207412032,
      "step": 39410,
      "train_runtime": 383215.2734,
      "train_tokens_per_second": 29245.734
    },
    {
      "epoch": 1.394757268242199,
      "grad_norm": 0.625,
      "learning_rate": 2.2491738614760956e-05,
      "loss": 0.3901569128036499,
      "num_input_tokens_seen": 11210267648,
      "step": 39420,
      "train_runtime": 383313.1594,
      "train_tokens_per_second": 29245.716
    },
    {
      "epoch": 1.3951110885059275,
      "grad_norm": 0.625,
      "learning_rate": 2.2489463343532742e-05,
      "loss": 0.39046335220336914,
      "num_input_tokens_seen": 11213072064,
      "step": 39430,
      "train_runtime": 383408.2707,
      "train_tokens_per_second": 29245.775
    },
    {
      "epoch": 1.3954649087696565,
      "grad_norm": 0.62890625,
      "learning_rate": 2.2487188762666268e-05,
      "loss": 0.3915881156921387,
      "num_input_tokens_seen": 11215926144,
      "step": 39440,
      "train_runtime": 383507.0138,
      "train_tokens_per_second": 29245.687
    },
    {
      "epoch": 1.3958187290333852,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2484914871812504e-05,
      "loss": 0.3882216215133667,
      "num_input_tokens_seen": 11218763200,
      "step": 39450,
      "train_runtime": 383604.3437,
      "train_tokens_per_second": 29245.663
    },
    {
      "epoch": 1.396172549297114,
      "grad_norm": 0.578125,
      "learning_rate": 2.248264167062264e-05,
      "loss": 0.3908745288848877,
      "num_input_tokens_seen": 11221601472,
      "step": 39460,
      "train_runtime": 383698.7409,
      "train_tokens_per_second": 29245.865
    },
    {
      "epoch": 1.3965263695608428,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2480369158748138e-05,
      "loss": 0.3922338724136353,
      "num_input_tokens_seen": 11224429952,
      "step": 39470,
      "train_runtime": 383796.9044,
      "train_tokens_per_second": 29245.754
    },
    {
      "epoch": 1.3968801898245715,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2478097335840682e-05,
      "loss": 0.39245095252990725,
      "num_input_tokens_seen": 11227295360,
      "step": 39480,
      "train_runtime": 383897.2996,
      "train_tokens_per_second": 29245.57
    },
    {
      "epoch": 1.3972340100883003,
      "grad_norm": 0.58203125,
      "learning_rate": 2.2475826201552215e-05,
      "loss": 0.39073567390441893,
      "num_input_tokens_seen": 11230140864,
      "step": 39490,
      "train_runtime": 383993.99,
      "train_tokens_per_second": 29245.616
    },
    {
      "epoch": 1.397587830352029,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2473555755534935e-05,
      "loss": 0.39198329448699953,
      "num_input_tokens_seen": 11232961856,
      "step": 39500,
      "train_runtime": 384090.3125,
      "train_tokens_per_second": 29245.627
    },
    {
      "epoch": 1.3979416506157578,
      "grad_norm": 0.609375,
      "learning_rate": 2.2471285997441267e-05,
      "loss": 0.3873871326446533,
      "num_input_tokens_seen": 11235836672,
      "step": 39510,
      "train_runtime": 384189.067,
      "train_tokens_per_second": 29245.592
    },
    {
      "epoch": 1.3982954708794866,
      "grad_norm": 0.6171875,
      "learning_rate": 2.246901692692389e-05,
      "loss": 0.39047234058380126,
      "num_input_tokens_seen": 11238604864,
      "step": 39520,
      "train_runtime": 384281.7974,
      "train_tokens_per_second": 29245.738
    },
    {
      "epoch": 1.3986492911432153,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2466748543635735e-05,
      "loss": 0.38958890438079835,
      "num_input_tokens_seen": 11241440704,
      "step": 39530,
      "train_runtime": 384376.1892,
      "train_tokens_per_second": 29245.934
    },
    {
      "epoch": 1.399003111406944,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2464480847229967e-05,
      "loss": 0.38456265926361083,
      "num_input_tokens_seen": 11244246336,
      "step": 39540,
      "train_runtime": 384472.2667,
      "train_tokens_per_second": 29245.923
    },
    {
      "epoch": 1.3993569316706729,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2462213837360003e-05,
      "loss": 0.38970279693603516,
      "num_input_tokens_seen": 11247100608,
      "step": 39550,
      "train_runtime": 384566.7038,
      "train_tokens_per_second": 29246.163
    },
    {
      "epoch": 1.3997107519344016,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2459947513679502e-05,
      "loss": 0.3863519191741943,
      "num_input_tokens_seen": 11249952448,
      "step": 39560,
      "train_runtime": 384664.4729,
      "train_tokens_per_second": 29246.144
    },
    {
      "epoch": 1.4000645721981306,
      "grad_norm": 0.58203125,
      "learning_rate": 2.2457681875842372e-05,
      "loss": 0.3867347717285156,
      "num_input_tokens_seen": 11252786880,
      "step": 39570,
      "train_runtime": 384759.4287,
      "train_tokens_per_second": 29246.293
    },
    {
      "epoch": 1.4004183924618592,
      "grad_norm": 0.58203125,
      "learning_rate": 2.245541692350276e-05,
      "loss": 0.39089341163635255,
      "num_input_tokens_seen": 11255565824,
      "step": 39580,
      "train_runtime": 384851.6433,
      "train_tokens_per_second": 29246.506
    },
    {
      "epoch": 1.4007722127255882,
      "grad_norm": 0.62890625,
      "learning_rate": 2.245315265631506e-05,
      "loss": 0.3921431303024292,
      "num_input_tokens_seen": 11258436608,
      "step": 39590,
      "train_runtime": 384951.7604,
      "train_tokens_per_second": 29246.357
    },
    {
      "epoch": 1.4011260329893167,
      "grad_norm": 0.609375,
      "learning_rate": 2.2450889073933906e-05,
      "loss": 0.3916007518768311,
      "num_input_tokens_seen": 11261232640,
      "step": 39600,
      "train_runtime": 385044.439,
      "train_tokens_per_second": 29246.579
    },
    {
      "epoch": 1.4014798532530457,
      "grad_norm": 0.59375,
      "learning_rate": 2.2448626176014187e-05,
      "loss": 0.3927267551422119,
      "num_input_tokens_seen": 11264059072,
      "step": 39610,
      "train_runtime": 385141.9845,
      "train_tokens_per_second": 29246.51
    },
    {
      "epoch": 1.4018336735167745,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2446363962211015e-05,
      "loss": 0.39011099338531496,
      "num_input_tokens_seen": 11266859392,
      "step": 39620,
      "train_runtime": 385238.1925,
      "train_tokens_per_second": 29246.476
    },
    {
      "epoch": 1.4021874937805032,
      "grad_norm": 0.59375,
      "learning_rate": 2.244410243217976e-05,
      "loss": 0.3889866828918457,
      "num_input_tokens_seen": 11269693120,
      "step": 39630,
      "train_runtime": 385334.2888,
      "train_tokens_per_second": 29246.536
    },
    {
      "epoch": 1.402541314044232,
      "grad_norm": 0.58203125,
      "learning_rate": 2.2441841585576043e-05,
      "loss": 0.389142370223999,
      "num_input_tokens_seen": 11272529792,
      "step": 39640,
      "train_runtime": 385431.9442,
      "train_tokens_per_second": 29246.486
    },
    {
      "epoch": 1.4028951343079608,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2439581422055703e-05,
      "loss": 0.394963812828064,
      "num_input_tokens_seen": 11275426048,
      "step": 39650,
      "train_runtime": 385532.5329,
      "train_tokens_per_second": 29246.367
    },
    {
      "epoch": 1.4032489545716895,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2437321941274843e-05,
      "loss": 0.39005801677703855,
      "num_input_tokens_seen": 11278256192,
      "step": 39660,
      "train_runtime": 385630.7891,
      "train_tokens_per_second": 29246.254
    },
    {
      "epoch": 1.4036027748354183,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2435063142889803e-05,
      "loss": 0.39021968841552734,
      "num_input_tokens_seen": 11281119616,
      "step": 39670,
      "train_runtime": 385729.2435,
      "train_tokens_per_second": 29246.213
    },
    {
      "epoch": 1.403956595099147,
      "grad_norm": 0.62890625,
      "learning_rate": 2.2432805026557163e-05,
      "loss": 0.38866124153137205,
      "num_input_tokens_seen": 11284013440,
      "step": 39680,
      "train_runtime": 385830.2866,
      "train_tokens_per_second": 29246.054
    },
    {
      "epoch": 1.4043104153628758,
      "grad_norm": 0.6171875,
      "learning_rate": 2.243054759193374e-05,
      "loss": 0.39093899726867676,
      "num_input_tokens_seen": 11286918848,
      "step": 39690,
      "train_runtime": 385932.5904,
      "train_tokens_per_second": 29245.83
    },
    {
      "epoch": 1.4046642356266046,
      "grad_norm": 0.609375,
      "learning_rate": 2.2428290838676597e-05,
      "loss": 0.39354887008666994,
      "num_input_tokens_seen": 11289822400,
      "step": 39700,
      "train_runtime": 386030.8565,
      "train_tokens_per_second": 29245.907
    },
    {
      "epoch": 1.4050180558903334,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2426034766443042e-05,
      "loss": 0.388654899597168,
      "num_input_tokens_seen": 11292672064,
      "step": 39710,
      "train_runtime": 386127.8401,
      "train_tokens_per_second": 29245.941
    },
    {
      "epoch": 1.4053718761540621,
      "grad_norm": 0.625,
      "learning_rate": 2.2423779374890626e-05,
      "loss": 0.388590669631958,
      "num_input_tokens_seen": 11295548544,
      "step": 39720,
      "train_runtime": 386229.5597,
      "train_tokens_per_second": 29245.686
    },
    {
      "epoch": 1.4057256964177909,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2421524663677132e-05,
      "loss": 0.3916128396987915,
      "num_input_tokens_seen": 11298352256,
      "step": 39730,
      "train_runtime": 386324.3274,
      "train_tokens_per_second": 29245.77
    },
    {
      "epoch": 1.4060795166815199,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2419270632460588e-05,
      "loss": 0.3905921936035156,
      "num_input_tokens_seen": 11301168896,
      "step": 39740,
      "train_runtime": 386419.523,
      "train_tokens_per_second": 29245.854
    },
    {
      "epoch": 1.4064333369452484,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2417017280899265e-05,
      "loss": 0.39098193645477297,
      "num_input_tokens_seen": 11303974528,
      "step": 39750,
      "train_runtime": 386516.1704,
      "train_tokens_per_second": 29245.8
    },
    {
      "epoch": 1.4067871572089774,
      "grad_norm": 0.59375,
      "learning_rate": 2.2414764608651668e-05,
      "loss": 0.3909113883972168,
      "num_input_tokens_seen": 11306792128,
      "step": 39760,
      "train_runtime": 386614.3664,
      "train_tokens_per_second": 29245.66
    },
    {
      "epoch": 1.407140977472706,
      "grad_norm": 0.6015625,
      "learning_rate": 2.241251261537655e-05,
      "loss": 0.3887632369995117,
      "num_input_tokens_seen": 11309587840,
      "step": 39770,
      "train_runtime": 386708.711,
      "train_tokens_per_second": 29245.754
    },
    {
      "epoch": 1.407494797736435,
      "grad_norm": 0.62890625,
      "learning_rate": 2.2410261300732903e-05,
      "loss": 0.3937871217727661,
      "num_input_tokens_seen": 11312487232,
      "step": 39780,
      "train_runtime": 386811.6242,
      "train_tokens_per_second": 29245.469
    },
    {
      "epoch": 1.4078486180001637,
      "grad_norm": 0.58203125,
      "learning_rate": 2.240801066437995e-05,
      "loss": 0.3941979169845581,
      "num_input_tokens_seen": 11315314880,
      "step": 39790,
      "train_runtime": 386906.2005,
      "train_tokens_per_second": 29245.628
    },
    {
      "epoch": 1.4082024382638925,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2405760705977166e-05,
      "loss": 0.3887176513671875,
      "num_input_tokens_seen": 11318206912,
      "step": 39800,
      "train_runtime": 387005.9583,
      "train_tokens_per_second": 29245.562
    },
    {
      "epoch": 1.4085562585276212,
      "grad_norm": 0.62109375,
      "learning_rate": 2.240351142518425e-05,
      "loss": 0.3900728225708008,
      "num_input_tokens_seen": 11321046528,
      "step": 39810,
      "train_runtime": 387102.7036,
      "train_tokens_per_second": 29245.589
    },
    {
      "epoch": 1.40891007879135,
      "grad_norm": 0.578125,
      "learning_rate": 2.240126282166116e-05,
      "loss": 0.38937814235687257,
      "num_input_tokens_seen": 11323874880,
      "step": 39820,
      "train_runtime": 387200.5701,
      "train_tokens_per_second": 29245.502
    },
    {
      "epoch": 1.4092638990550788,
      "grad_norm": 0.625,
      "learning_rate": 2.239901489506808e-05,
      "loss": 0.3895012378692627,
      "num_input_tokens_seen": 11326753280,
      "step": 39830,
      "train_runtime": 387301.2846,
      "train_tokens_per_second": 29245.328
    },
    {
      "epoch": 1.4096177193188075,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2396767645065424e-05,
      "loss": 0.39227278232574464,
      "num_input_tokens_seen": 11329605440,
      "step": 39840,
      "train_runtime": 387398.4679,
      "train_tokens_per_second": 29245.354
    },
    {
      "epoch": 1.4099715395825363,
      "grad_norm": 0.609375,
      "learning_rate": 2.239452107131387e-05,
      "loss": 0.39105401039123533,
      "num_input_tokens_seen": 11332447360,
      "step": 39850,
      "train_runtime": 387495.2068,
      "train_tokens_per_second": 29245.387
    },
    {
      "epoch": 1.410325359846265,
      "grad_norm": 0.6015625,
      "learning_rate": 2.239227517347431e-05,
      "loss": 0.3898653984069824,
      "num_input_tokens_seen": 11335302336,
      "step": 39860,
      "train_runtime": 387593.0355,
      "train_tokens_per_second": 29245.372
    },
    {
      "epoch": 1.4106791801099938,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2390029951207876e-05,
      "loss": 0.3870972156524658,
      "num_input_tokens_seen": 11338120000,
      "step": 39870,
      "train_runtime": 387686.425,
      "train_tokens_per_second": 29245.595
    },
    {
      "epoch": 1.4110330003737226,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2387785404175965e-05,
      "loss": 0.3906590938568115,
      "num_input_tokens_seen": 11340983232,
      "step": 39880,
      "train_runtime": 387785.0483,
      "train_tokens_per_second": 29245.54
    },
    {
      "epoch": 1.4113868206374514,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2385541532040178e-05,
      "loss": 0.39721705913543703,
      "num_input_tokens_seen": 11343874368,
      "step": 39890,
      "train_runtime": 387885.3249,
      "train_tokens_per_second": 29245.433
    },
    {
      "epoch": 1.4117406409011801,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2383298334462368e-05,
      "loss": 0.39101126194000246,
      "num_input_tokens_seen": 11346669056,
      "step": 39900,
      "train_runtime": 387979.7176,
      "train_tokens_per_second": 29245.521
    },
    {
      "epoch": 1.412094461164909,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2381055811104627e-05,
      "loss": 0.3916982650756836,
      "num_input_tokens_seen": 11349581440,
      "step": 39910,
      "train_runtime": 388081.7648,
      "train_tokens_per_second": 29245.336
    },
    {
      "epoch": 1.4124482814286377,
      "grad_norm": 0.64453125,
      "learning_rate": 2.2378813961629274e-05,
      "loss": 0.392187762260437,
      "num_input_tokens_seen": 11352445824,
      "step": 39920,
      "train_runtime": 388181.5717,
      "train_tokens_per_second": 29245.195
    },
    {
      "epoch": 1.4128021016923666,
      "grad_norm": 0.62890625,
      "learning_rate": 2.2376572785698884e-05,
      "loss": 0.3917304515838623,
      "num_input_tokens_seen": 11355256192,
      "step": 39930,
      "train_runtime": 388276.3498,
      "train_tokens_per_second": 29245.294
    },
    {
      "epoch": 1.4131559219560954,
      "grad_norm": 0.6171875,
      "learning_rate": 2.237433228297625e-05,
      "loss": 0.3887251138687134,
      "num_input_tokens_seen": 11358158912,
      "step": 39940,
      "train_runtime": 388375.2664,
      "train_tokens_per_second": 29245.32
    },
    {
      "epoch": 1.4135097422198242,
      "grad_norm": 0.59375,
      "learning_rate": 2.2372092453124402e-05,
      "loss": 0.3879792928695679,
      "num_input_tokens_seen": 11360970368,
      "step": 39950,
      "train_runtime": 388470.9916,
      "train_tokens_per_second": 29245.351
    },
    {
      "epoch": 1.413863562483553,
      "grad_norm": 0.59765625,
      "learning_rate": 2.236985329580662e-05,
      "loss": 0.38984549045562744,
      "num_input_tokens_seen": 11363792768,
      "step": 39960,
      "train_runtime": 388564.4927,
      "train_tokens_per_second": 29245.577
    },
    {
      "epoch": 1.4142173827472817,
      "grad_norm": 0.63671875,
      "learning_rate": 2.236761481068641e-05,
      "loss": 0.3871484279632568,
      "num_input_tokens_seen": 11366586624,
      "step": 39970,
      "train_runtime": 388659.9159,
      "train_tokens_per_second": 29245.585
    },
    {
      "epoch": 1.4145712030110105,
      "grad_norm": 0.71875,
      "learning_rate": 2.2365376997427508e-05,
      "loss": 0.3879386901855469,
      "num_input_tokens_seen": 11369383424,
      "step": 39980,
      "train_runtime": 388753.0564,
      "train_tokens_per_second": 29245.772
    },
    {
      "epoch": 1.4149250232747392,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2363139855693906e-05,
      "loss": 0.38856356143951415,
      "num_input_tokens_seen": 11372213504,
      "step": 39990,
      "train_runtime": 388851.2578,
      "train_tokens_per_second": 29245.665
    },
    {
      "epoch": 1.415278843538468,
      "grad_norm": 0.59375,
      "learning_rate": 2.2360903385149807e-05,
      "loss": 0.38887643814086914,
      "num_input_tokens_seen": 11374996992,
      "step": 40000,
      "train_runtime": 388944.4839,
      "train_tokens_per_second": 29245.811
    },
    {
      "epoch": 1.415278843538468,
      "eval_loss": 0.33797845244407654,
      "eval_runtime": 8.4274,
      "eval_samples_per_second": 473.217,
      "eval_steps_per_second": 3.797,
      "num_input_tokens_seen": 11374996992,
      "step": 40000
    },
    {
      "epoch": 1.4156326638021968,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2358667585459662e-05,
      "loss": 0.38672637939453125,
      "num_input_tokens_seen": 11377803456,
      "step": 40010,
      "train_runtime": 389069.8515,
      "train_tokens_per_second": 29243.601
    },
    {
      "epoch": 1.4159864840659255,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2356432456288154e-05,
      "loss": 0.38859717845916747,
      "num_input_tokens_seen": 11380581760,
      "step": 40020,
      "train_runtime": 389161.6461,
      "train_tokens_per_second": 29243.842
    },
    {
      "epoch": 1.4163403043296543,
      "grad_norm": 0.59375,
      "learning_rate": 2.235419799730021e-05,
      "loss": 0.3892946243286133,
      "num_input_tokens_seen": 11383398080,
      "step": 40030,
      "train_runtime": 389256.8893,
      "train_tokens_per_second": 29243.922
    },
    {
      "epoch": 1.416694124593383,
      "grad_norm": 0.63671875,
      "learning_rate": 2.2351964208160966e-05,
      "loss": 0.393123722076416,
      "num_input_tokens_seen": 11386258048,
      "step": 40040,
      "train_runtime": 389353.9447,
      "train_tokens_per_second": 29243.978
    },
    {
      "epoch": 1.4170479448571118,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2349731088535826e-05,
      "loss": 0.3896328926086426,
      "num_input_tokens_seen": 11389134464,
      "step": 40050,
      "train_runtime": 389451.9993,
      "train_tokens_per_second": 29244.001
    },
    {
      "epoch": 1.4174017651208406,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2347498638090404e-05,
      "loss": 0.3914656639099121,
      "num_input_tokens_seen": 11391962304,
      "step": 40060,
      "train_runtime": 389547.4846,
      "train_tokens_per_second": 29244.092
    },
    {
      "epoch": 1.4177555853845694,
      "grad_norm": 0.60546875,
      "learning_rate": 2.234526685649055e-05,
      "loss": 0.389146614074707,
      "num_input_tokens_seen": 11394846720,
      "step": 40070,
      "train_runtime": 389646.0307,
      "train_tokens_per_second": 29244.098
    },
    {
      "epoch": 1.4181094056482983,
      "grad_norm": 0.609375,
      "learning_rate": 2.234303574340235e-05,
      "loss": 0.3896371364593506,
      "num_input_tokens_seen": 11397669696,
      "step": 40080,
      "train_runtime": 389741.2251,
      "train_tokens_per_second": 29244.198
    },
    {
      "epoch": 1.418463225912027,
      "grad_norm": 0.6015625,
      "learning_rate": 2.234080529849214e-05,
      "loss": 0.3892864942550659,
      "num_input_tokens_seen": 11400518080,
      "step": 40090,
      "train_runtime": 389840.6003,
      "train_tokens_per_second": 29244.05
    },
    {
      "epoch": 1.4188170461757559,
      "grad_norm": 0.59375,
      "learning_rate": 2.2338575521426467e-05,
      "loss": 0.3886976718902588,
      "num_input_tokens_seen": 11403364352,
      "step": 40100,
      "train_runtime": 389939.6015,
      "train_tokens_per_second": 29243.925
    },
    {
      "epoch": 1.4191708664394846,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2336346411872113e-05,
      "loss": 0.3863805294036865,
      "num_input_tokens_seen": 11406277248,
      "step": 40110,
      "train_runtime": 390039.1615,
      "train_tokens_per_second": 29243.928
    },
    {
      "epoch": 1.4195246867032134,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2334117969496103e-05,
      "loss": 0.39302668571472166,
      "num_input_tokens_seen": 11409132672,
      "step": 40120,
      "train_runtime": 390133.9945,
      "train_tokens_per_second": 29244.139
    },
    {
      "epoch": 1.4198785069669422,
      "grad_norm": 0.62890625,
      "learning_rate": 2.2331890193965688e-05,
      "loss": 0.39189538955688474,
      "num_input_tokens_seen": 11411977280,
      "step": 40130,
      "train_runtime": 390232.8637,
      "train_tokens_per_second": 29244.019
    },
    {
      "epoch": 1.420232327230671,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2329663084948358e-05,
      "loss": 0.3895774602890015,
      "num_input_tokens_seen": 11414848768,
      "step": 40140,
      "train_runtime": 390330.0745,
      "train_tokens_per_second": 29244.092
    },
    {
      "epoch": 1.4205861474943997,
      "grad_norm": 0.609375,
      "learning_rate": 2.2327436642111822e-05,
      "loss": 0.39326364994049073,
      "num_input_tokens_seen": 11417682944,
      "step": 40150,
      "train_runtime": 390425.1589,
      "train_tokens_per_second": 29244.229
    },
    {
      "epoch": 1.4209399677581285,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2325210865124034e-05,
      "loss": 0.39289882183074953,
      "num_input_tokens_seen": 11420514240,
      "step": 40160,
      "train_runtime": 390522.7983,
      "train_tokens_per_second": 29244.168
    },
    {
      "epoch": 1.4212937880218572,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2322985753653177e-05,
      "loss": 0.3914470195770264,
      "num_input_tokens_seen": 11423375296,
      "step": 40170,
      "train_runtime": 390617.9384,
      "train_tokens_per_second": 29244.369
    },
    {
      "epoch": 1.421647608285586,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2320761307367658e-05,
      "loss": 0.3858524799346924,
      "num_input_tokens_seen": 11426219264,
      "step": 40180,
      "train_runtime": 390715.8501,
      "train_tokens_per_second": 29244.32
    },
    {
      "epoch": 1.4220014285493148,
      "grad_norm": 0.609375,
      "learning_rate": 2.2318537525936124e-05,
      "loss": 0.39272003173828124,
      "num_input_tokens_seen": 11429033600,
      "step": 40190,
      "train_runtime": 390812.3121,
      "train_tokens_per_second": 29244.303
    },
    {
      "epoch": 1.4223552488130435,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2316314409027443e-05,
      "loss": 0.3908493757247925,
      "num_input_tokens_seen": 11431858624,
      "step": 40200,
      "train_runtime": 390910.306,
      "train_tokens_per_second": 29244.199
    },
    {
      "epoch": 1.4227090690767723,
      "grad_norm": 0.625,
      "learning_rate": 2.2314091956310723e-05,
      "loss": 0.39589080810546873,
      "num_input_tokens_seen": 11434702080,
      "step": 40210,
      "train_runtime": 391007.777,
      "train_tokens_per_second": 29244.181
    },
    {
      "epoch": 1.423062889340501,
      "grad_norm": 0.59375,
      "learning_rate": 2.231187016745531e-05,
      "loss": 0.38591485023498534,
      "num_input_tokens_seen": 11437572544,
      "step": 40220,
      "train_runtime": 391108.7964,
      "train_tokens_per_second": 29243.967
    },
    {
      "epoch": 1.42341670960423,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2309649042130753e-05,
      "loss": 0.3878044128417969,
      "num_input_tokens_seen": 11440431808,
      "step": 40230,
      "train_runtime": 391208.1155,
      "train_tokens_per_second": 29243.851
    },
    {
      "epoch": 1.4237705298679586,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2307428580006863e-05,
      "loss": 0.3874910831451416,
      "num_input_tokens_seen": 11443291968,
      "step": 40240,
      "train_runtime": 391305.5926,
      "train_tokens_per_second": 29243.875
    },
    {
      "epoch": 1.4241243501316876,
      "grad_norm": 0.59765625,
      "learning_rate": 2.230520878075366e-05,
      "loss": 0.3884744167327881,
      "num_input_tokens_seen": 11446056000,
      "step": 40250,
      "train_runtime": 391397.1458,
      "train_tokens_per_second": 29244.097
    },
    {
      "epoch": 1.4244781703954161,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2302989644041396e-05,
      "loss": 0.3924331426620483,
      "num_input_tokens_seen": 11448894656,
      "step": 40260,
      "train_runtime": 391492.6087,
      "train_tokens_per_second": 29244.217
    },
    {
      "epoch": 1.4248319906591451,
      "grad_norm": 0.578125,
      "learning_rate": 2.2300771169540562e-05,
      "loss": 0.3892298698425293,
      "num_input_tokens_seen": 11451740096,
      "step": 40270,
      "train_runtime": 391587.7858,
      "train_tokens_per_second": 29244.375
    },
    {
      "epoch": 1.4251858109228739,
      "grad_norm": 0.59375,
      "learning_rate": 2.2298553356921874e-05,
      "loss": 0.38946716785430907,
      "num_input_tokens_seen": 11454567040,
      "step": 40280,
      "train_runtime": 391682.0183,
      "train_tokens_per_second": 29244.557
    },
    {
      "epoch": 1.4255396311866027,
      "grad_norm": 0.625,
      "learning_rate": 2.2296336205856278e-05,
      "loss": 0.3936901569366455,
      "num_input_tokens_seen": 11457382976,
      "step": 40290,
      "train_runtime": 391776.4845,
      "train_tokens_per_second": 29244.693
    },
    {
      "epoch": 1.4258934514503314,
      "grad_norm": 0.58203125,
      "learning_rate": 2.2294119716014944e-05,
      "loss": 0.38647236824035647,
      "num_input_tokens_seen": 11460279872,
      "step": 40300,
      "train_runtime": 391879.8901,
      "train_tokens_per_second": 29244.368
    },
    {
      "epoch": 1.4262472717140602,
      "grad_norm": 0.609375,
      "learning_rate": 2.2291903887069275e-05,
      "loss": 0.38946921825408937,
      "num_input_tokens_seen": 11463110400,
      "step": 40310,
      "train_runtime": 391975.3813,
      "train_tokens_per_second": 29244.465
    },
    {
      "epoch": 1.426601091977789,
      "grad_norm": 0.59765625,
      "learning_rate": 2.22896887186909e-05,
      "loss": 0.38969926834106444,
      "num_input_tokens_seen": 11465984384,
      "step": 40320,
      "train_runtime": 392075.605,
      "train_tokens_per_second": 29244.32
    },
    {
      "epoch": 1.4269549122415177,
      "grad_norm": 0.609375,
      "learning_rate": 2.2287474210551678e-05,
      "loss": 0.3908782243728638,
      "num_input_tokens_seen": 11468849600,
      "step": 40330,
      "train_runtime": 392175.8523,
      "train_tokens_per_second": 29244.15
    },
    {
      "epoch": 1.4273087325052465,
      "grad_norm": 0.6015625,
      "learning_rate": 2.22852603623237e-05,
      "loss": 0.38668394088745117,
      "num_input_tokens_seen": 11471715584,
      "step": 40340,
      "train_runtime": 392273.4114,
      "train_tokens_per_second": 29244.183
    },
    {
      "epoch": 1.4276625527689752,
      "grad_norm": 0.58203125,
      "learning_rate": 2.2283047173679276e-05,
      "loss": 0.3864229679107666,
      "num_input_tokens_seen": 11474555008,
      "step": 40350,
      "train_runtime": 392370.7603,
      "train_tokens_per_second": 29244.164
    },
    {
      "epoch": 1.428016373032704,
      "grad_norm": 0.609375,
      "learning_rate": 2.2280834644290958e-05,
      "loss": 0.3852095365524292,
      "num_input_tokens_seen": 11477404096,
      "step": 40360,
      "train_runtime": 392468.9914,
      "train_tokens_per_second": 29244.104
    },
    {
      "epoch": 1.4283701932964328,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2278622773831507e-05,
      "loss": 0.3873852252960205,
      "num_input_tokens_seen": 11480238592,
      "step": 40370,
      "train_runtime": 392564.3631,
      "train_tokens_per_second": 29244.22
    },
    {
      "epoch": 1.4287240135601615,
      "grad_norm": 0.62890625,
      "learning_rate": 2.227641156197392e-05,
      "loss": 0.3899340867996216,
      "num_input_tokens_seen": 11483001536,
      "step": 40380,
      "train_runtime": 392655.7495,
      "train_tokens_per_second": 29244.45
    },
    {
      "epoch": 1.4290778338238903,
      "grad_norm": 0.609375,
      "learning_rate": 2.227420100839143e-05,
      "loss": 0.3916586399078369,
      "num_input_tokens_seen": 11485833152,
      "step": 40390,
      "train_runtime": 392751.5872,
      "train_tokens_per_second": 29244.524
    },
    {
      "epoch": 1.4294316540876193,
      "grad_norm": 0.59375,
      "learning_rate": 2.227199111275749e-05,
      "loss": 0.3894233226776123,
      "num_input_tokens_seen": 11488681728,
      "step": 40400,
      "train_runtime": 392849.262,
      "train_tokens_per_second": 29244.504
    },
    {
      "epoch": 1.4297854743513478,
      "grad_norm": 0.6015625,
      "learning_rate": 2.226978187474577e-05,
      "loss": 0.3920260429382324,
      "num_input_tokens_seen": 11491614080,
      "step": 40410,
      "train_runtime": 392951.2174,
      "train_tokens_per_second": 29244.378
    },
    {
      "epoch": 1.4301392946150768,
      "grad_norm": 0.609375,
      "learning_rate": 2.226757329403018e-05,
      "loss": 0.3896705389022827,
      "num_input_tokens_seen": 11494459264,
      "step": 40420,
      "train_runtime": 393047.4887,
      "train_tokens_per_second": 29244.454
    },
    {
      "epoch": 1.4304931148788054,
      "grad_norm": 0.6015625,
      "learning_rate": 2.226536537028485e-05,
      "loss": 0.3868753433227539,
      "num_input_tokens_seen": 11497338880,
      "step": 40430,
      "train_runtime": 393146.7239,
      "train_tokens_per_second": 29244.397
    },
    {
      "epoch": 1.4308469351425344,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2263158103184143e-05,
      "loss": 0.38915555477142333,
      "num_input_tokens_seen": 11500204352,
      "step": 40440,
      "train_runtime": 393244.1577,
      "train_tokens_per_second": 29244.438
    },
    {
      "epoch": 1.4312007554062631,
      "grad_norm": 0.609375,
      "learning_rate": 2.2260951492402636e-05,
      "loss": 0.3887815475463867,
      "num_input_tokens_seen": 11502991680,
      "step": 40450,
      "train_runtime": 393337.5303,
      "train_tokens_per_second": 29244.582
    },
    {
      "epoch": 1.431554575669992,
      "grad_norm": 0.58203125,
      "learning_rate": 2.225874553761514e-05,
      "loss": 0.3865049362182617,
      "num_input_tokens_seen": 11505850112,
      "step": 40460,
      "train_runtime": 393432.8216,
      "train_tokens_per_second": 29244.764
    },
    {
      "epoch": 1.4319083959337207,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2256540238496687e-05,
      "loss": 0.387589693069458,
      "num_input_tokens_seen": 11508674048,
      "step": 40470,
      "train_runtime": 393528.6376,
      "train_tokens_per_second": 29244.82
    },
    {
      "epoch": 1.4322622161974494,
      "grad_norm": 0.609375,
      "learning_rate": 2.225433559472255e-05,
      "loss": 0.3893230199813843,
      "num_input_tokens_seen": 11511497024,
      "step": 40480,
      "train_runtime": 393626.089,
      "train_tokens_per_second": 29244.751
    },
    {
      "epoch": 1.4326160364611782,
      "grad_norm": 0.6171875,
      "learning_rate": 2.22521316059682e-05,
      "loss": 0.3911158800125122,
      "num_input_tokens_seen": 11514339264,
      "step": 40490,
      "train_runtime": 393722.019,
      "train_tokens_per_second": 29244.845
    },
    {
      "epoch": 1.432969856724907,
      "grad_norm": 0.62109375,
      "learning_rate": 2.224992827190935e-05,
      "loss": 0.3901453733444214,
      "num_input_tokens_seen": 11517171136,
      "step": 40500,
      "train_runtime": 393818.7668,
      "train_tokens_per_second": 29244.851
    },
    {
      "epoch": 1.4333236769886357,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2247725592221942e-05,
      "loss": 0.38905446529388427,
      "num_input_tokens_seen": 11520055104,
      "step": 40510,
      "train_runtime": 393918.7668,
      "train_tokens_per_second": 29244.748
    },
    {
      "epoch": 1.4336774972523645,
      "grad_norm": 0.58984375,
      "learning_rate": 2.224552356658213e-05,
      "loss": 0.3883650779724121,
      "num_input_tokens_seen": 11522938176,
      "step": 40520,
      "train_runtime": 394019.906,
      "train_tokens_per_second": 29244.558
    },
    {
      "epoch": 1.4340313175160933,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2243322194666296e-05,
      "loss": 0.3913144111633301,
      "num_input_tokens_seen": 11525780544,
      "step": 40530,
      "train_runtime": 394117.0546,
      "train_tokens_per_second": 29244.562
    },
    {
      "epoch": 1.434385137779822,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2241121476151057e-05,
      "loss": 0.3855109691619873,
      "num_input_tokens_seen": 11528624000,
      "step": 40540,
      "train_runtime": 394212.708,
      "train_tokens_per_second": 29244.679
    },
    {
      "epoch": 1.4347389580435508,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2238921410713234e-05,
      "loss": 0.38794527053833006,
      "num_input_tokens_seen": 11531470016,
      "step": 40550,
      "train_runtime": 394309.3802,
      "train_tokens_per_second": 29244.727
    },
    {
      "epoch": 1.4350927783072795,
      "grad_norm": 0.62109375,
      "learning_rate": 2.223672199802989e-05,
      "loss": 0.39396731853485106,
      "num_input_tokens_seen": 11534291712,
      "step": 40560,
      "train_runtime": 394406.6627,
      "train_tokens_per_second": 29244.668
    },
    {
      "epoch": 1.4354465985710085,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2234523237778308e-05,
      "loss": 0.3863222122192383,
      "num_input_tokens_seen": 11537152192,
      "step": 40570,
      "train_runtime": 394507.4858,
      "train_tokens_per_second": 29244.444
    },
    {
      "epoch": 1.435800418834737,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2232325129635972e-05,
      "loss": 0.38979282379150393,
      "num_input_tokens_seen": 11539980224,
      "step": 40580,
      "train_runtime": 394604.2177,
      "train_tokens_per_second": 29244.442
    },
    {
      "epoch": 1.436154239098466,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2230127673280628e-05,
      "loss": 0.38915553092956545,
      "num_input_tokens_seen": 11542885888,
      "step": 40590,
      "train_runtime": 394706.5566,
      "train_tokens_per_second": 29244.221
    },
    {
      "epoch": 1.4365080593621946,
      "grad_norm": 0.58984375,
      "learning_rate": 2.222793086839022e-05,
      "loss": 0.38700313568115235,
      "num_input_tokens_seen": 11545725440,
      "step": 40600,
      "train_runtime": 394806.9318,
      "train_tokens_per_second": 29243.979
    },
    {
      "epoch": 1.4368618796259236,
      "grad_norm": 0.63671875,
      "learning_rate": 2.222573471464291e-05,
      "loss": 0.3864602565765381,
      "num_input_tokens_seen": 11548548416,
      "step": 40610,
      "train_runtime": 394905.1277,
      "train_tokens_per_second": 29243.855
    },
    {
      "epoch": 1.4372156998896524,
      "grad_norm": 0.59765625,
      "learning_rate": 2.22235392117171e-05,
      "loss": 0.3960279941558838,
      "num_input_tokens_seen": 11551403648,
      "step": 40620,
      "train_runtime": 395003.0354,
      "train_tokens_per_second": 29243.835
    },
    {
      "epoch": 1.4375695201533811,
      "grad_norm": 0.609375,
      "learning_rate": 2.22213443592914e-05,
      "loss": 0.3902994155883789,
      "num_input_tokens_seen": 11554256640,
      "step": 40630,
      "train_runtime": 395101.2051,
      "train_tokens_per_second": 29243.79
    },
    {
      "epoch": 1.43792334041711,
      "grad_norm": 0.61328125,
      "learning_rate": 2.221915015704465e-05,
      "loss": 0.3881173610687256,
      "num_input_tokens_seen": 11557072000,
      "step": 40640,
      "train_runtime": 395195.2645,
      "train_tokens_per_second": 29243.954
    },
    {
      "epoch": 1.4382771606808387,
      "grad_norm": 0.58203125,
      "learning_rate": 2.221695660465592e-05,
      "loss": 0.38553333282470703,
      "num_input_tokens_seen": 11559902336,
      "step": 40650,
      "train_runtime": 395294.6429,
      "train_tokens_per_second": 29243.762
    },
    {
      "epoch": 1.4386309809445674,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2214763701804485e-05,
      "loss": 0.3853966236114502,
      "num_input_tokens_seen": 11562712000,
      "step": 40660,
      "train_runtime": 395389.7903,
      "train_tokens_per_second": 29243.83
    },
    {
      "epoch": 1.4389848012082962,
      "grad_norm": 0.58203125,
      "learning_rate": 2.221257144816984e-05,
      "loss": 0.3895285606384277,
      "num_input_tokens_seen": 11565586112,
      "step": 40670,
      "train_runtime": 395487.0615,
      "train_tokens_per_second": 29243.905
    },
    {
      "epoch": 1.439338621472025,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2210379843431728e-05,
      "loss": 0.3875574588775635,
      "num_input_tokens_seen": 11568444672,
      "step": 40680,
      "train_runtime": 395583.8155,
      "train_tokens_per_second": 29243.979
    },
    {
      "epoch": 1.4396924417357537,
      "grad_norm": 0.61328125,
      "learning_rate": 2.220818888727008e-05,
      "loss": 0.39140543937683103,
      "num_input_tokens_seen": 11571286464,
      "step": 40690,
      "train_runtime": 395679.8891,
      "train_tokens_per_second": 29244.06
    },
    {
      "epoch": 1.4400462619994825,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2205998579365066e-05,
      "loss": 0.38942081928253175,
      "num_input_tokens_seen": 11574163904,
      "step": 40700,
      "train_runtime": 395777.0879,
      "train_tokens_per_second": 29244.148
    },
    {
      "epoch": 1.4404000822632113,
      "grad_norm": 0.6015625,
      "learning_rate": 2.220380891939708e-05,
      "loss": 0.39285242557525635,
      "num_input_tokens_seen": 11577004992,
      "step": 40710,
      "train_runtime": 395871.9878,
      "train_tokens_per_second": 29244.315
    },
    {
      "epoch": 1.44075390252694,
      "grad_norm": 0.61328125,
      "learning_rate": 2.220161990704673e-05,
      "loss": 0.3845607995986938,
      "num_input_tokens_seen": 11579829248,
      "step": 40720,
      "train_runtime": 395967.9229,
      "train_tokens_per_second": 29244.362
    },
    {
      "epoch": 1.4411077227906688,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2199431541994836e-05,
      "loss": 0.3867394208908081,
      "num_input_tokens_seen": 11582653632,
      "step": 40730,
      "train_runtime": 396065.5919,
      "train_tokens_per_second": 29244.281
    },
    {
      "epoch": 1.4414615430543978,
      "grad_norm": 0.609375,
      "learning_rate": 2.2197243823922452e-05,
      "loss": 0.38645544052124026,
      "num_input_tokens_seen": 11585477120,
      "step": 40740,
      "train_runtime": 396162.0728,
      "train_tokens_per_second": 29244.286
    },
    {
      "epoch": 1.4418153633181263,
      "grad_norm": 0.63671875,
      "learning_rate": 2.2195056752510847e-05,
      "loss": 0.3891130924224854,
      "num_input_tokens_seen": 11588314752,
      "step": 40750,
      "train_runtime": 396258.9957,
      "train_tokens_per_second": 29244.294
    },
    {
      "epoch": 1.4421691835818553,
      "grad_norm": 0.625,
      "learning_rate": 2.2192870327441508e-05,
      "loss": 0.3863529682159424,
      "num_input_tokens_seen": 11591223168,
      "step": 40760,
      "train_runtime": 396359.3425,
      "train_tokens_per_second": 29244.228
    },
    {
      "epoch": 1.442523003845584,
      "grad_norm": 0.58984375,
      "learning_rate": 2.219068454839615e-05,
      "loss": 0.3863476276397705,
      "num_input_tokens_seen": 11594067584,
      "step": 40770,
      "train_runtime": 396454.1755,
      "train_tokens_per_second": 29244.408
    },
    {
      "epoch": 1.4428768241093128,
      "grad_norm": 0.609375,
      "learning_rate": 2.218849941505669e-05,
      "loss": 0.3878518342971802,
      "num_input_tokens_seen": 11596892352,
      "step": 40780,
      "train_runtime": 396550.635,
      "train_tokens_per_second": 29244.418
    },
    {
      "epoch": 1.4432306443730416,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2186314927105285e-05,
      "loss": 0.38821682929992674,
      "num_input_tokens_seen": 11599698048,
      "step": 40790,
      "train_runtime": 396644.0766,
      "train_tokens_per_second": 29244.602
    },
    {
      "epoch": 1.4435844646367704,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2184131084224292e-05,
      "loss": 0.3887503147125244,
      "num_input_tokens_seen": 11602548928,
      "step": 40800,
      "train_runtime": 396742.2958,
      "train_tokens_per_second": 29244.548
    },
    {
      "epoch": 1.4439382849004991,
      "grad_norm": 0.65625,
      "learning_rate": 2.2181947886096305e-05,
      "loss": 0.3889749526977539,
      "num_input_tokens_seen": 11605396096,
      "step": 40810,
      "train_runtime": 396838.6051,
      "train_tokens_per_second": 29244.625
    },
    {
      "epoch": 1.444292105164228,
      "grad_norm": 0.60546875,
      "learning_rate": 2.217976533240412e-05,
      "loss": 0.39027340412139894,
      "num_input_tokens_seen": 11608202496,
      "step": 40820,
      "train_runtime": 396933.7398,
      "train_tokens_per_second": 29244.686
    },
    {
      "epoch": 1.4446459254279567,
      "grad_norm": 0.609375,
      "learning_rate": 2.217758342283076e-05,
      "loss": 0.39053912162780763,
      "num_input_tokens_seen": 11611019392,
      "step": 40830,
      "train_runtime": 397028.8914,
      "train_tokens_per_second": 29244.772
    },
    {
      "epoch": 1.4449997456916854,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2175402157059465e-05,
      "loss": 0.3942587852478027,
      "num_input_tokens_seen": 11613865792,
      "step": 40840,
      "train_runtime": 397126.1768,
      "train_tokens_per_second": 29244.775
    },
    {
      "epoch": 1.4453535659554142,
      "grad_norm": 0.59375,
      "learning_rate": 2.2173221534773696e-05,
      "loss": 0.3930384635925293,
      "num_input_tokens_seen": 11616688384,
      "step": 40850,
      "train_runtime": 397220.0521,
      "train_tokens_per_second": 29244.97
    },
    {
      "epoch": 1.445707386219143,
      "grad_norm": 0.609375,
      "learning_rate": 2.217104155565713e-05,
      "loss": 0.3854065418243408,
      "num_input_tokens_seen": 11619584704,
      "step": 40860,
      "train_runtime": 397322.9183,
      "train_tokens_per_second": 29244.688
    },
    {
      "epoch": 1.4460612064828717,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2168862219393656e-05,
      "loss": 0.3892513275146484,
      "num_input_tokens_seen": 11622397504,
      "step": 40870,
      "train_runtime": 397416.9661,
      "train_tokens_per_second": 29244.845
    },
    {
      "epoch": 1.4464150267466005,
      "grad_norm": 0.59375,
      "learning_rate": 2.2166683525667382e-05,
      "loss": 0.38978431224822996,
      "num_input_tokens_seen": 11625274112,
      "step": 40880,
      "train_runtime": 397516.0864,
      "train_tokens_per_second": 29244.789
    },
    {
      "epoch": 1.4467688470103293,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2164505474162645e-05,
      "loss": 0.38984863758087157,
      "num_input_tokens_seen": 11628109952,
      "step": 40890,
      "train_runtime": 397611.2174,
      "train_tokens_per_second": 29244.924
    },
    {
      "epoch": 1.447122667274058,
      "grad_norm": 0.609375,
      "learning_rate": 2.216232806456398e-05,
      "loss": 0.391741943359375,
      "num_input_tokens_seen": 11630886720,
      "step": 40900,
      "train_runtime": 397703.835,
      "train_tokens_per_second": 29245.096
    },
    {
      "epoch": 1.447476487537787,
      "grad_norm": 0.6328125,
      "learning_rate": 2.2160151296556163e-05,
      "loss": 0.3904669761657715,
      "num_input_tokens_seen": 11633691392,
      "step": 40910,
      "train_runtime": 397798.411,
      "train_tokens_per_second": 29245.193
    },
    {
      "epoch": 1.4478303078015156,
      "grad_norm": 0.609375,
      "learning_rate": 2.2157975169824165e-05,
      "loss": 0.39179158210754395,
      "num_input_tokens_seen": 11636556800,
      "step": 40920,
      "train_runtime": 397896.6563,
      "train_tokens_per_second": 29245.174
    },
    {
      "epoch": 1.4481841280652445,
      "grad_norm": 0.63671875,
      "learning_rate": 2.2155799684053176e-05,
      "loss": 0.3884803295135498,
      "num_input_tokens_seen": 11639413632,
      "step": 40930,
      "train_runtime": 397994.2706,
      "train_tokens_per_second": 29245.179
    },
    {
      "epoch": 1.4485379483289733,
      "grad_norm": 0.61328125,
      "learning_rate": 2.215362483892862e-05,
      "loss": 0.3948474168777466,
      "num_input_tokens_seen": 11642220480,
      "step": 40940,
      "train_runtime": 398088.3793,
      "train_tokens_per_second": 29245.316
    },
    {
      "epoch": 1.448891768592702,
      "grad_norm": 0.59375,
      "learning_rate": 2.2151450634136108e-05,
      "loss": 0.38910617828369143,
      "num_input_tokens_seen": 11645031872,
      "step": 40950,
      "train_runtime": 398184.17,
      "train_tokens_per_second": 29245.341
    },
    {
      "epoch": 1.4492455888564308,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2149277069361504e-05,
      "loss": 0.38561482429504396,
      "num_input_tokens_seen": 11647837376,
      "step": 40960,
      "train_runtime": 398279.122,
      "train_tokens_per_second": 29245.413
    },
    {
      "epoch": 1.4495994091201596,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2147104144290857e-05,
      "loss": 0.39092767238616943,
      "num_input_tokens_seen": 11650645696,
      "step": 40970,
      "train_runtime": 398375.1057,
      "train_tokens_per_second": 29245.416
    },
    {
      "epoch": 1.4499532293838884,
      "grad_norm": 0.609375,
      "learning_rate": 2.2144931858610438e-05,
      "loss": 0.3938960790634155,
      "num_input_tokens_seen": 11653494912,
      "step": 40980,
      "train_runtime": 398474.9027,
      "train_tokens_per_second": 29245.242
    },
    {
      "epoch": 1.4503070496476171,
      "grad_norm": 0.58984375,
      "learning_rate": 2.214276021200674e-05,
      "loss": 0.38536319732666013,
      "num_input_tokens_seen": 11656305984,
      "step": 40990,
      "train_runtime": 398571.0246,
      "train_tokens_per_second": 29245.242
    },
    {
      "epoch": 1.450660869911346,
      "grad_norm": 0.6640625,
      "learning_rate": 2.2140589204166476e-05,
      "loss": 0.38586225509643557,
      "num_input_tokens_seen": 11659153792,
      "step": 41000,
      "train_runtime": 398667.6044,
      "train_tokens_per_second": 29245.3
    },
    {
      "epoch": 1.4510146901750747,
      "grad_norm": 0.625,
      "learning_rate": 2.2138418834776557e-05,
      "loss": 0.3900140762329102,
      "num_input_tokens_seen": 11661997248,
      "step": 41010,
      "train_runtime": 398766.339,
      "train_tokens_per_second": 29245.19
    },
    {
      "epoch": 1.4513685104388034,
      "grad_norm": 0.61328125,
      "learning_rate": 2.213624910352413e-05,
      "loss": 0.38683252334594725,
      "num_input_tokens_seen": 11664893056,
      "step": 41020,
      "train_runtime": 398867.2381,
      "train_tokens_per_second": 29245.052
    },
    {
      "epoch": 1.4517223307025322,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2134080010096532e-05,
      "loss": 0.38812670707702634,
      "num_input_tokens_seen": 11667749184,
      "step": 41030,
      "train_runtime": 398968.3211,
      "train_tokens_per_second": 29244.801
    },
    {
      "epoch": 1.452076150966261,
      "grad_norm": 0.609375,
      "learning_rate": 2.213191155418133e-05,
      "loss": 0.39428157806396485,
      "num_input_tokens_seen": 11670609856,
      "step": 41040,
      "train_runtime": 399066.7455,
      "train_tokens_per_second": 29244.757
    },
    {
      "epoch": 1.4524299712299897,
      "grad_norm": 0.61328125,
      "learning_rate": 2.212974373546631e-05,
      "loss": 0.39076895713806153,
      "num_input_tokens_seen": 11673447424,
      "step": 41050,
      "train_runtime": 399164.6891,
      "train_tokens_per_second": 29244.69
    },
    {
      "epoch": 1.4527837914937185,
      "grad_norm": 0.63671875,
      "learning_rate": 2.212757655363946e-05,
      "loss": 0.38721442222595215,
      "num_input_tokens_seen": 11676262784,
      "step": 41060,
      "train_runtime": 399259.2753,
      "train_tokens_per_second": 29244.813
    },
    {
      "epoch": 1.4531376117574473,
      "grad_norm": 0.63671875,
      "learning_rate": 2.2125410008388984e-05,
      "loss": 0.3874940872192383,
      "num_input_tokens_seen": 11679105472,
      "step": 41070,
      "train_runtime": 399357.5361,
      "train_tokens_per_second": 29244.735
    },
    {
      "epoch": 1.4534914320211763,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2123244099403306e-05,
      "loss": 0.38756141662597654,
      "num_input_tokens_seen": 11681939456,
      "step": 41080,
      "train_runtime": 399453.2044,
      "train_tokens_per_second": 29244.826
    },
    {
      "epoch": 1.4538452522849048,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2121078826371056e-05,
      "loss": 0.3879857063293457,
      "num_input_tokens_seen": 11684796032,
      "step": 41090,
      "train_runtime": 399549.6661,
      "train_tokens_per_second": 29244.915
    },
    {
      "epoch": 1.4541990725486338,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2118914188981088e-05,
      "loss": 0.38819236755371095,
      "num_input_tokens_seen": 11687639744,
      "step": 41100,
      "train_runtime": 399647.8629,
      "train_tokens_per_second": 29244.845
    },
    {
      "epoch": 1.4545528928123626,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2116750186922453e-05,
      "loss": 0.39173412322998047,
      "num_input_tokens_seen": 11690419520,
      "step": 41110,
      "train_runtime": 399741.0738,
      "train_tokens_per_second": 29244.98
    },
    {
      "epoch": 1.4549067130760913,
      "grad_norm": 0.6171875,
      "learning_rate": 2.2114586819884432e-05,
      "loss": 0.3884713649749756,
      "num_input_tokens_seen": 11693284544,
      "step": 41120,
      "train_runtime": 399841.5348,
      "train_tokens_per_second": 29244.797
    },
    {
      "epoch": 1.45526053333982,
      "grad_norm": 0.60546875,
      "learning_rate": 2.21124240875565e-05,
      "loss": 0.3936263084411621,
      "num_input_tokens_seen": 11696187968,
      "step": 41130,
      "train_runtime": 399942.7547,
      "train_tokens_per_second": 29244.655
    },
    {
      "epoch": 1.4556143536035488,
      "grad_norm": 0.609375,
      "learning_rate": 2.2110261989628366e-05,
      "loss": 0.3882664680480957,
      "num_input_tokens_seen": 11699114240,
      "step": 41140,
      "train_runtime": 400047.8965,
      "train_tokens_per_second": 29244.284
    },
    {
      "epoch": 1.4559681738672776,
      "grad_norm": 0.59375,
      "learning_rate": 2.210810052578994e-05,
      "loss": 0.3900323390960693,
      "num_input_tokens_seen": 11701960832,
      "step": 41150,
      "train_runtime": 400145.8045,
      "train_tokens_per_second": 29244.242
    },
    {
      "epoch": 1.4563219941310064,
      "grad_norm": 0.59375,
      "learning_rate": 2.2105939695731333e-05,
      "loss": 0.389192533493042,
      "num_input_tokens_seen": 11704807872,
      "step": 41160,
      "train_runtime": 400242.7667,
      "train_tokens_per_second": 29244.271
    },
    {
      "epoch": 1.4566758143947351,
      "grad_norm": 0.609375,
      "learning_rate": 2.210377949914289e-05,
      "loss": 0.3876333713531494,
      "num_input_tokens_seen": 11707594944,
      "step": 41170,
      "train_runtime": 400336.2974,
      "train_tokens_per_second": 29244.4
    },
    {
      "epoch": 1.457029634658464,
      "grad_norm": 0.58203125,
      "learning_rate": 2.2101619935715153e-05,
      "loss": 0.39046659469604494,
      "num_input_tokens_seen": 11710382912,
      "step": 41180,
      "train_runtime": 400429.4773,
      "train_tokens_per_second": 29244.558
    },
    {
      "epoch": 1.4573834549221927,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2099461005138885e-05,
      "loss": 0.38741326332092285,
      "num_input_tokens_seen": 11713193856,
      "step": 41190,
      "train_runtime": 400525.2661,
      "train_tokens_per_second": 29244.582
    },
    {
      "epoch": 1.4577372751859214,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2097302707105054e-05,
      "loss": 0.38676726818084717,
      "num_input_tokens_seen": 11715963584,
      "step": 41200,
      "train_runtime": 400616.7709,
      "train_tokens_per_second": 29244.816
    },
    {
      "epoch": 1.4580910954496502,
      "grad_norm": 0.6171875,
      "learning_rate": 2.209514504130483e-05,
      "loss": 0.38543100357055665,
      "num_input_tokens_seen": 11718841600,
      "step": 41210,
      "train_runtime": 400717.7699,
      "train_tokens_per_second": 29244.627
    },
    {
      "epoch": 1.458444915713379,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2092988007429625e-05,
      "loss": 0.3888906478881836,
      "num_input_tokens_seen": 11721698304,
      "step": 41220,
      "train_runtime": 400816.5829,
      "train_tokens_per_second": 29244.544
    },
    {
      "epoch": 1.458798735977108,
      "grad_norm": 0.61328125,
      "learning_rate": 2.209083160517102e-05,
      "loss": 0.39154150485992434,
      "num_input_tokens_seen": 11724616320,
      "step": 41230,
      "train_runtime": 400920.3522,
      "train_tokens_per_second": 29244.253
    },
    {
      "epoch": 1.4591525562408365,
      "grad_norm": 0.62109375,
      "learning_rate": 2.208867583422084e-05,
      "loss": 0.38861167430877686,
      "num_input_tokens_seen": 11727516416,
      "step": 41240,
      "train_runtime": 401021.1499,
      "train_tokens_per_second": 29244.134
    },
    {
      "epoch": 1.4595063765045655,
      "grad_norm": 0.625,
      "learning_rate": 2.2086520694271108e-05,
      "loss": 0.3876093864440918,
      "num_input_tokens_seen": 11730358784,
      "step": 41250,
      "train_runtime": 401120.1848,
      "train_tokens_per_second": 29244.0
    },
    {
      "epoch": 1.459860196768294,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2084366185014058e-05,
      "loss": 0.3886312484741211,
      "num_input_tokens_seen": 11733238336,
      "step": 41260,
      "train_runtime": 401219.768,
      "train_tokens_per_second": 29243.919
    },
    {
      "epoch": 1.460214017032023,
      "grad_norm": 0.60546875,
      "learning_rate": 2.208221230614213e-05,
      "loss": 0.3850902318954468,
      "num_input_tokens_seen": 11736064640,
      "step": 41270,
      "train_runtime": 401316.0489,
      "train_tokens_per_second": 29243.945
    },
    {
      "epoch": 1.4605678372957518,
      "grad_norm": 0.60546875,
      "learning_rate": 2.208005905734798e-05,
      "loss": 0.3880853414535522,
      "num_input_tokens_seen": 11738981952,
      "step": 41280,
      "train_runtime": 401415.5316,
      "train_tokens_per_second": 29243.965
    },
    {
      "epoch": 1.4609216575594806,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2077906438324474e-05,
      "loss": 0.3935344457626343,
      "num_input_tokens_seen": 11741814272,
      "step": 41290,
      "train_runtime": 401511.0538,
      "train_tokens_per_second": 29244.062
    },
    {
      "epoch": 1.4612754778232093,
      "grad_norm": 0.609375,
      "learning_rate": 2.2075754448764683e-05,
      "loss": 0.3901233196258545,
      "num_input_tokens_seen": 11744600640,
      "step": 41300,
      "train_runtime": 401604.7705,
      "train_tokens_per_second": 29244.176
    },
    {
      "epoch": 1.461629298086938,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2073603088361887e-05,
      "loss": 0.3872868061065674,
      "num_input_tokens_seen": 11747479040,
      "step": 41310,
      "train_runtime": 401704.5716,
      "train_tokens_per_second": 29244.076
    },
    {
      "epoch": 1.4619831183506669,
      "grad_norm": 0.60546875,
      "learning_rate": 2.2071452356809586e-05,
      "loss": 0.38918461799621584,
      "num_input_tokens_seen": 11750318720,
      "step": 41320,
      "train_runtime": 401800.4639,
      "train_tokens_per_second": 29244.164
    },
    {
      "epoch": 1.4623369386143956,
      "grad_norm": 0.58984375,
      "learning_rate": 2.2069302253801474e-05,
      "loss": 0.3885365009307861,
      "num_input_tokens_seen": 11753198720,
      "step": 41330,
      "train_runtime": 401897.5005,
      "train_tokens_per_second": 29244.269
    },
    {
      "epoch": 1.4626907588781244,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2067152779031465e-05,
      "loss": 0.3934452056884766,
      "num_input_tokens_seen": 11756035008,
      "step": 41340,
      "train_runtime": 401991.9097,
      "train_tokens_per_second": 29244.457
    },
    {
      "epoch": 1.4630445791418532,
      "grad_norm": 0.62109375,
      "learning_rate": 2.206500393219367e-05,
      "loss": 0.3892578840255737,
      "num_input_tokens_seen": 11758838976,
      "step": 41350,
      "train_runtime": 402086.2275,
      "train_tokens_per_second": 29244.57
    },
    {
      "epoch": 1.463398399405582,
      "grad_norm": 0.58984375,
      "learning_rate": 2.206285571298242e-05,
      "loss": 0.3952140808105469,
      "num_input_tokens_seen": 11761715328,
      "step": 41360,
      "train_runtime": 402184.6644,
      "train_tokens_per_second": 29244.564
    },
    {
      "epoch": 1.4637522196693107,
      "grad_norm": 0.60546875,
      "learning_rate": 2.206070812109226e-05,
      "loss": 0.38689894676208497,
      "num_input_tokens_seen": 11764582592,
      "step": 41370,
      "train_runtime": 402282.2075,
      "train_tokens_per_second": 29244.601
    },
    {
      "epoch": 1.4641060399330394,
      "grad_norm": 0.6484375,
      "learning_rate": 2.2058561156217913e-05,
      "loss": 0.3905513763427734,
      "num_input_tokens_seen": 11767450752,
      "step": 41380,
      "train_runtime": 402379.7553,
      "train_tokens_per_second": 29244.639
    },
    {
      "epoch": 1.4644598601967682,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2056414818054347e-05,
      "loss": 0.38667137622833253,
      "num_input_tokens_seen": 11770238464,
      "step": 41390,
      "train_runtime": 402471.5769,
      "train_tokens_per_second": 29244.894
    },
    {
      "epoch": 1.4648136804604972,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2054269106296714e-05,
      "loss": 0.3852600336074829,
      "num_input_tokens_seen": 11773071424,
      "step": 41400,
      "train_runtime": 402567.8338,
      "train_tokens_per_second": 29244.938
    },
    {
      "epoch": 1.4651675007242257,
      "grad_norm": 0.62890625,
      "learning_rate": 2.205212402064038e-05,
      "loss": 0.3888318300247192,
      "num_input_tokens_seen": 11775928576,
      "step": 41410,
      "train_runtime": 402662.9419,
      "train_tokens_per_second": 29245.126
    },
    {
      "epoch": 1.4655213209879547,
      "grad_norm": 0.58203125,
      "learning_rate": 2.204997956078092e-05,
      "loss": 0.3925034999847412,
      "num_input_tokens_seen": 11778787136,
      "step": 41420,
      "train_runtime": 402760.0548,
      "train_tokens_per_second": 29245.172
    },
    {
      "epoch": 1.4658751412516833,
      "grad_norm": 0.609375,
      "learning_rate": 2.2047835726414116e-05,
      "loss": 0.3913259506225586,
      "num_input_tokens_seen": 11781597376,
      "step": 41430,
      "train_runtime": 402855.1457,
      "train_tokens_per_second": 29245.245
    },
    {
      "epoch": 1.4662289615154123,
      "grad_norm": 0.6015625,
      "learning_rate": 2.2045692517235952e-05,
      "loss": 0.3940769672393799,
      "num_input_tokens_seen": 11784486528,
      "step": 41440,
      "train_runtime": 402957.0544,
      "train_tokens_per_second": 29245.019
    },
    {
      "epoch": 1.466582781779141,
      "grad_norm": 0.625,
      "learning_rate": 2.204354993294263e-05,
      "loss": 0.388521409034729,
      "num_input_tokens_seen": 11787246976,
      "step": 41450,
      "train_runtime": 403050.9107,
      "train_tokens_per_second": 29245.057
    },
    {
      "epoch": 1.4669366020428698,
      "grad_norm": 0.62109375,
      "learning_rate": 2.2041407973230548e-05,
      "loss": 0.3861745119094849,
      "num_input_tokens_seen": 11790027328,
      "step": 41460,
      "train_runtime": 403144.7412,
      "train_tokens_per_second": 29245.147
    },
    {
      "epoch": 1.4672904223065986,
      "grad_norm": 0.60546875,
      "learning_rate": 2.203926663779631e-05,
      "loss": 0.39077186584472656,
      "num_input_tokens_seen": 11792876288,
      "step": 41470,
      "train_runtime": 403241.5868,
      "train_tokens_per_second": 29245.189
    },
    {
      "epoch": 1.4676442425703273,
      "grad_norm": 0.57421875,
      "learning_rate": 2.2037125926336733e-05,
      "loss": 0.38942904472351075,
      "num_input_tokens_seen": 11795772096,
      "step": 41480,
      "train_runtime": 403343.6798,
      "train_tokens_per_second": 29244.966
    },
    {
      "epoch": 1.467998062834056,
      "grad_norm": 0.58203125,
      "learning_rate": 2.2034985838548835e-05,
      "loss": 0.3886574745178223,
      "num_input_tokens_seen": 11798590080,
      "step": 41490,
      "train_runtime": 403440.9354,
      "train_tokens_per_second": 29244.901
    },
    {
      "epoch": 1.4683518830977849,
      "grad_norm": 0.609375,
      "learning_rate": 2.2032846374129848e-05,
      "loss": 0.38815042972564695,
      "num_input_tokens_seen": 11801416064,
      "step": 41500,
      "train_runtime": 403538.3232,
      "train_tokens_per_second": 29244.846
    },
    {
      "epoch": 1.4687057033615136,
      "grad_norm": 0.62890625,
      "learning_rate": 2.2030707532777196e-05,
      "loss": 0.39616124629974364,
      "num_input_tokens_seen": 11804203456,
      "step": 41510,
      "train_runtime": 403632.2213,
      "train_tokens_per_second": 29244.948
    },
    {
      "epoch": 1.4690595236252424,
      "grad_norm": 0.625,
      "learning_rate": 2.2028569314188528e-05,
      "loss": 0.3893098831176758,
      "num_input_tokens_seen": 11807072768,
      "step": 41520,
      "train_runtime": 403730.4441,
      "train_tokens_per_second": 29244.94
    },
    {
      "epoch": 1.4694133438889712,
      "grad_norm": 0.57421875,
      "learning_rate": 2.2026431718061676e-05,
      "loss": 0.3874368667602539,
      "num_input_tokens_seen": 11809909504,
      "step": 41530,
      "train_runtime": 403827.9398,
      "train_tokens_per_second": 29244.904
    },
    {
      "epoch": 1.4697671641527,
      "grad_norm": 0.61328125,
      "learning_rate": 2.202429474409469e-05,
      "loss": 0.39329066276550295,
      "num_input_tokens_seen": 11812725696,
      "step": 41540,
      "train_runtime": 403923.1493,
      "train_tokens_per_second": 29244.983
    },
    {
      "epoch": 1.4701209844164287,
      "grad_norm": 0.58203125,
      "learning_rate": 2.202215839198583e-05,
      "loss": 0.38404526710510256,
      "num_input_tokens_seen": 11815580224,
      "step": 41550,
      "train_runtime": 404020.6708,
      "train_tokens_per_second": 29244.989
    },
    {
      "epoch": 1.4704748046801575,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2020022661433544e-05,
      "loss": 0.389002251625061,
      "num_input_tokens_seen": 11818447168,
      "step": 41560,
      "train_runtime": 404118.0932,
      "train_tokens_per_second": 29245.033
    },
    {
      "epoch": 1.4708286249438864,
      "grad_norm": 0.63671875,
      "learning_rate": 2.20178875521365e-05,
      "loss": 0.3907487869262695,
      "num_input_tokens_seen": 11821257664,
      "step": 41570,
      "train_runtime": 404214.8997,
      "train_tokens_per_second": 29244.982
    },
    {
      "epoch": 1.471182445207615,
      "grad_norm": 0.59765625,
      "learning_rate": 2.2015753063793566e-05,
      "loss": 0.39040122032165525,
      "num_input_tokens_seen": 11824069248,
      "step": 41580,
      "train_runtime": 404307.0474,
      "train_tokens_per_second": 29245.271
    },
    {
      "epoch": 1.471536265471344,
      "grad_norm": 0.609375,
      "learning_rate": 2.2013619196103812e-05,
      "loss": 0.3934812068939209,
      "num_input_tokens_seen": 11826916096,
      "step": 41590,
      "train_runtime": 404405.3064,
      "train_tokens_per_second": 29245.205
    },
    {
      "epoch": 1.4718900857350725,
      "grad_norm": 0.5703125,
      "learning_rate": 2.2011485948766514e-05,
      "loss": 0.38592445850372314,
      "num_input_tokens_seen": 11829774080,
      "step": 41600,
      "train_runtime": 404502.8361,
      "train_tokens_per_second": 29245.219
    },
    {
      "epoch": 1.4722439059988015,
      "grad_norm": 0.625,
      "learning_rate": 2.2009353321481148e-05,
      "loss": 0.3861634016036987,
      "num_input_tokens_seen": 11832622592,
      "step": 41610,
      "train_runtime": 404598.1632,
      "train_tokens_per_second": 29245.369
    },
    {
      "epoch": 1.4725977262625303,
      "grad_norm": 0.609375,
      "learning_rate": 2.2007221313947404e-05,
      "loss": 0.3925305366516113,
      "num_input_tokens_seen": 11835515584,
      "step": 41620,
      "train_runtime": 404697.2593,
      "train_tokens_per_second": 29245.356
    },
    {
      "epoch": 1.472951546526259,
      "grad_norm": 0.62890625,
      "learning_rate": 2.2005089925865162e-05,
      "loss": 0.39200472831726074,
      "num_input_tokens_seen": 11838331520,
      "step": 41630,
      "train_runtime": 404793.9421,
      "train_tokens_per_second": 29245.328
    },
    {
      "epoch": 1.4733053667899878,
      "grad_norm": 0.5859375,
      "learning_rate": 2.2002959156934518e-05,
      "loss": 0.38784592151641845,
      "num_input_tokens_seen": 11841169152,
      "step": 41640,
      "train_runtime": 404887.8089,
      "train_tokens_per_second": 29245.556
    },
    {
      "epoch": 1.4736591870537166,
      "grad_norm": 0.62890625,
      "learning_rate": 2.2000829006855763e-05,
      "loss": 0.39212713241577146,
      "num_input_tokens_seen": 11844018688,
      "step": 41650,
      "train_runtime": 404984.5536,
      "train_tokens_per_second": 29245.606
    },
    {
      "epoch": 1.4740130073174453,
      "grad_norm": 0.58984375,
      "learning_rate": 2.199869947532939e-05,
      "loss": 0.3888967037200928,
      "num_input_tokens_seen": 11846870592,
      "step": 41660,
      "train_runtime": 405080.0081,
      "train_tokens_per_second": 29245.755
    },
    {
      "epoch": 1.474366827581174,
      "grad_norm": 0.578125,
      "learning_rate": 2.199657056205611e-05,
      "loss": 0.38810622692108154,
      "num_input_tokens_seen": 11849697280,
      "step": 41670,
      "train_runtime": 405175.5263,
      "train_tokens_per_second": 29245.837
    },
    {
      "epoch": 1.4747206478449029,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1994442266736807e-05,
      "loss": 0.3954521656036377,
      "num_input_tokens_seen": 11852599104,
      "step": 41680,
      "train_runtime": 405273.8043,
      "train_tokens_per_second": 29245.905
    },
    {
      "epoch": 1.4750744681086316,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1992314589072605e-05,
      "loss": 0.38598241806030276,
      "num_input_tokens_seen": 11855496576,
      "step": 41690,
      "train_runtime": 405375.459,
      "train_tokens_per_second": 29245.719
    },
    {
      "epoch": 1.4754282883723604,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1990187528764796e-05,
      "loss": 0.3855717182159424,
      "num_input_tokens_seen": 11858326784,
      "step": 41700,
      "train_runtime": 405470.619,
      "train_tokens_per_second": 29245.835
    },
    {
      "epoch": 1.4757821086360892,
      "grad_norm": 0.62890625,
      "learning_rate": 2.19880610855149e-05,
      "loss": 0.38348994255065916,
      "num_input_tokens_seen": 11861230528,
      "step": 41710,
      "train_runtime": 405567.4784,
      "train_tokens_per_second": 29246.01
    },
    {
      "epoch": 1.476135928899818,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1985935259024618e-05,
      "loss": 0.38772623538970946,
      "num_input_tokens_seen": 11864121472,
      "step": 41720,
      "train_runtime": 405667.0795,
      "train_tokens_per_second": 29245.956
    },
    {
      "epoch": 1.4764897491635467,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1983810048995874e-05,
      "loss": 0.38765480518341067,
      "num_input_tokens_seen": 11866918208,
      "step": 41730,
      "train_runtime": 405762.2506,
      "train_tokens_per_second": 29245.989
    },
    {
      "epoch": 1.4768435694272757,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1981685455130778e-05,
      "loss": 0.3916560173034668,
      "num_input_tokens_seen": 11869772224,
      "step": 41740,
      "train_runtime": 405862.2649,
      "train_tokens_per_second": 29245.814
    },
    {
      "epoch": 1.4771973896910042,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1979561477131642e-05,
      "loss": 0.39403786659240725,
      "num_input_tokens_seen": 11872650752,
      "step": 41750,
      "train_runtime": 405962.4937,
      "train_tokens_per_second": 29245.684
    },
    {
      "epoch": 1.4775512099547332,
      "grad_norm": 0.59375,
      "learning_rate": 2.197743811470099e-05,
      "loss": 0.3879851341247559,
      "num_input_tokens_seen": 11875491328,
      "step": 41760,
      "train_runtime": 406058.9136,
      "train_tokens_per_second": 29245.735
    },
    {
      "epoch": 1.477905030218462,
      "grad_norm": 0.62109375,
      "learning_rate": 2.197531536754154e-05,
      "loss": 0.39210987091064453,
      "num_input_tokens_seen": 11878366272,
      "step": 41770,
      "train_runtime": 406158.8204,
      "train_tokens_per_second": 29245.619
    },
    {
      "epoch": 1.4782588504821907,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1973193235356205e-05,
      "loss": 0.3847451686859131,
      "num_input_tokens_seen": 11881177856,
      "step": 41780,
      "train_runtime": 406255.8606,
      "train_tokens_per_second": 29245.554
    },
    {
      "epoch": 1.4786126707459195,
      "grad_norm": 0.625,
      "learning_rate": 2.197107171784811e-05,
      "loss": 0.38843379020690916,
      "num_input_tokens_seen": 11884000640,
      "step": 41790,
      "train_runtime": 406347.8649,
      "train_tokens_per_second": 29245.879
    },
    {
      "epoch": 1.4789664910096483,
      "grad_norm": 0.62890625,
      "learning_rate": 2.196895081472058e-05,
      "loss": 0.38926947116851807,
      "num_input_tokens_seen": 11886802112,
      "step": 41800,
      "train_runtime": 406445.1468,
      "train_tokens_per_second": 29245.772
    },
    {
      "epoch": 1.479320311273377,
      "grad_norm": 0.62109375,
      "learning_rate": 2.196683052567713e-05,
      "loss": 0.3918159008026123,
      "num_input_tokens_seen": 11889605760,
      "step": 41810,
      "train_runtime": 406542.3383,
      "train_tokens_per_second": 29245.677
    },
    {
      "epoch": 1.4796741315371058,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1964710850421486e-05,
      "loss": 0.3893465995788574,
      "num_input_tokens_seen": 11892479424,
      "step": 41820,
      "train_runtime": 406638.8454,
      "train_tokens_per_second": 29245.803
    },
    {
      "epoch": 1.4800279518008346,
      "grad_norm": 0.6171875,
      "learning_rate": 2.196259178865757e-05,
      "loss": 0.3939209461212158,
      "num_input_tokens_seen": 11895332288,
      "step": 41830,
      "train_runtime": 406735.9528,
      "train_tokens_per_second": 29245.834
    },
    {
      "epoch": 1.4803817720645633,
      "grad_norm": 0.63671875,
      "learning_rate": 2.1960473340089502e-05,
      "loss": 0.3899192810058594,
      "num_input_tokens_seen": 11898161408,
      "step": 41840,
      "train_runtime": 406833.3941,
      "train_tokens_per_second": 29245.784
    },
    {
      "epoch": 1.480735592328292,
      "grad_norm": 0.61328125,
      "learning_rate": 2.19583555044216e-05,
      "loss": 0.3907729148864746,
      "num_input_tokens_seen": 11900976064,
      "step": 41850,
      "train_runtime": 406929.0064,
      "train_tokens_per_second": 29245.829
    },
    {
      "epoch": 1.4810894125920209,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1956238281358394e-05,
      "loss": 0.38986475467681886,
      "num_input_tokens_seen": 11903796608,
      "step": 41860,
      "train_runtime": 407022.4065,
      "train_tokens_per_second": 29246.047
    },
    {
      "epoch": 1.4814432328557496,
      "grad_norm": 0.640625,
      "learning_rate": 2.1954121670604595e-05,
      "loss": 0.3915048122406006,
      "num_input_tokens_seen": 11906668864,
      "step": 41870,
      "train_runtime": 407121.282,
      "train_tokens_per_second": 29246.0
    },
    {
      "epoch": 1.4817970531194784,
      "grad_norm": 0.6328125,
      "learning_rate": 2.1952005671865124e-05,
      "loss": 0.38891217708587644,
      "num_input_tokens_seen": 11909509120,
      "step": 41880,
      "train_runtime": 407217.1733,
      "train_tokens_per_second": 29246.088
    },
    {
      "epoch": 1.4821508733832072,
      "grad_norm": 0.609375,
      "learning_rate": 2.194989028484511e-05,
      "loss": 0.38334877490997316,
      "num_input_tokens_seen": 11912281088,
      "step": 41890,
      "train_runtime": 407311.0434,
      "train_tokens_per_second": 29246.153
    },
    {
      "epoch": 1.482504693646936,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1947775509249857e-05,
      "loss": 0.38658182621002196,
      "num_input_tokens_seen": 11915101440,
      "step": 41900,
      "train_runtime": 407407.1951,
      "train_tokens_per_second": 29246.173
    },
    {
      "epoch": 1.482858513910665,
      "grad_norm": 0.59765625,
      "learning_rate": 2.194566134478489e-05,
      "loss": 0.39027931690216067,
      "num_input_tokens_seen": 11917887168,
      "step": 41910,
      "train_runtime": 407498.8048,
      "train_tokens_per_second": 29246.435
    },
    {
      "epoch": 1.4832123341743935,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1943547791155918e-05,
      "loss": 0.3932937145233154,
      "num_input_tokens_seen": 11920766592,
      "step": 41920,
      "train_runtime": 407595.7417,
      "train_tokens_per_second": 29246.544
    },
    {
      "epoch": 1.4835661544381225,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1941434848068865e-05,
      "loss": 0.3913945198059082,
      "num_input_tokens_seen": 11923626560,
      "step": 41930,
      "train_runtime": 407691.4506,
      "train_tokens_per_second": 29246.693
    },
    {
      "epoch": 1.4839199747018512,
      "grad_norm": 0.609375,
      "learning_rate": 2.193932251522983e-05,
      "loss": 0.38409278392791746,
      "num_input_tokens_seen": 11926494080,
      "step": 41940,
      "train_runtime": 407792.0428,
      "train_tokens_per_second": 29246.51
    },
    {
      "epoch": 1.48427379496558,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1937210792345127e-05,
      "loss": 0.3881162166595459,
      "num_input_tokens_seen": 11929369280,
      "step": 41950,
      "train_runtime": 407892.2239,
      "train_tokens_per_second": 29246.376
    },
    {
      "epoch": 1.4846276152293088,
      "grad_norm": 0.6171875,
      "learning_rate": 2.193509967912126e-05,
      "loss": 0.39225139617919924,
      "num_input_tokens_seen": 11932191616,
      "step": 41960,
      "train_runtime": 407989.1695,
      "train_tokens_per_second": 29246.344
    },
    {
      "epoch": 1.4849814354930375,
      "grad_norm": 0.609375,
      "learning_rate": 2.193298917526494e-05,
      "loss": 0.3883184909820557,
      "num_input_tokens_seen": 11935069632,
      "step": 41970,
      "train_runtime": 408087.9126,
      "train_tokens_per_second": 29246.32
    },
    {
      "epoch": 1.4853352557567663,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1930879280483068e-05,
      "loss": 0.3937432289123535,
      "num_input_tokens_seen": 11937905984,
      "step": 41980,
      "train_runtime": 408184.6134,
      "train_tokens_per_second": 29246.34
    },
    {
      "epoch": 1.485689076020495,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1928769994482743e-05,
      "loss": 0.39079980850219725,
      "num_input_tokens_seen": 11940753536,
      "step": 41990,
      "train_runtime": 408282.423,
      "train_tokens_per_second": 29246.308
    },
    {
      "epoch": 1.4860428962842238,
      "grad_norm": 0.625,
      "learning_rate": 2.1926661316971257e-05,
      "loss": 0.39066352844238283,
      "num_input_tokens_seen": 11943568640,
      "step": 42000,
      "train_runtime": 408375.7694,
      "train_tokens_per_second": 29246.516
    },
    {
      "epoch": 1.4863967165479526,
      "grad_norm": 0.59765625,
      "learning_rate": 2.192455324765611e-05,
      "loss": 0.39258360862731934,
      "num_input_tokens_seen": 11946471296,
      "step": 42010,
      "train_runtime": 408480.7985,
      "train_tokens_per_second": 29246.102
    },
    {
      "epoch": 1.4867505368116813,
      "grad_norm": 0.60546875,
      "learning_rate": 2.192244578624499e-05,
      "loss": 0.388304615020752,
      "num_input_tokens_seen": 11949288896,
      "step": 42020,
      "train_runtime": 408576.321,
      "train_tokens_per_second": 29246.161
    },
    {
      "epoch": 1.48710435707541,
      "grad_norm": 0.59375,
      "learning_rate": 2.1920338932445783e-05,
      "loss": 0.3867853879928589,
      "num_input_tokens_seen": 11952160768,
      "step": 42030,
      "train_runtime": 408675.1986,
      "train_tokens_per_second": 29246.112
    },
    {
      "epoch": 1.4874581773391389,
      "grad_norm": 0.6015625,
      "learning_rate": 2.191823268596658e-05,
      "loss": 0.39093039035797117,
      "num_input_tokens_seen": 11955001600,
      "step": 42040,
      "train_runtime": 408773.2994,
      "train_tokens_per_second": 29246.043
    },
    {
      "epoch": 1.4878119976028676,
      "grad_norm": 0.6015625,
      "learning_rate": 2.191612704651565e-05,
      "loss": 0.3905934810638428,
      "num_input_tokens_seen": 11957859520,
      "step": 42050,
      "train_runtime": 408872.0825,
      "train_tokens_per_second": 29245.967
    },
    {
      "epoch": 1.4881658178665966,
      "grad_norm": 0.62109375,
      "learning_rate": 2.191402201380148e-05,
      "loss": 0.38708198070526123,
      "num_input_tokens_seen": 11960717312,
      "step": 42060,
      "train_runtime": 408968.2515,
      "train_tokens_per_second": 29246.078
    },
    {
      "epoch": 1.4885196381303252,
      "grad_norm": 0.6328125,
      "learning_rate": 2.191191758753273e-05,
      "loss": 0.3918651819229126,
      "num_input_tokens_seen": 11963604416,
      "step": 42070,
      "train_runtime": 409065.7062,
      "train_tokens_per_second": 29246.168
    },
    {
      "epoch": 1.4888734583940542,
      "grad_norm": 0.61328125,
      "learning_rate": 2.190981376741828e-05,
      "loss": 0.3869763374328613,
      "num_input_tokens_seen": 11966445696,
      "step": 42080,
      "train_runtime": 409163.0258,
      "train_tokens_per_second": 29246.156
    },
    {
      "epoch": 1.4892272786577827,
      "grad_norm": 0.625,
      "learning_rate": 2.190771055316719e-05,
      "loss": 0.38770365715026855,
      "num_input_tokens_seen": 11969239296,
      "step": 42090,
      "train_runtime": 409257.3887,
      "train_tokens_per_second": 29246.239
    },
    {
      "epoch": 1.4895810989215117,
      "grad_norm": 0.6328125,
      "learning_rate": 2.1905607944488718e-05,
      "loss": 0.38915128707885743,
      "num_input_tokens_seen": 11972024704,
      "step": 42100,
      "train_runtime": 409351.5883,
      "train_tokens_per_second": 29246.313
    },
    {
      "epoch": 1.4899349191852405,
      "grad_norm": 0.6328125,
      "learning_rate": 2.1903505941092315e-05,
      "loss": 0.39298579692840574,
      "num_input_tokens_seen": 11974864512,
      "step": 42110,
      "train_runtime": 409452.4827,
      "train_tokens_per_second": 29246.042
    },
    {
      "epoch": 1.4902887394489692,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1901404542687634e-05,
      "loss": 0.3821543216705322,
      "num_input_tokens_seen": 11977747904,
      "step": 42120,
      "train_runtime": 409551.0011,
      "train_tokens_per_second": 29246.047
    },
    {
      "epoch": 1.490642559712698,
      "grad_norm": 0.62109375,
      "learning_rate": 2.189930374898452e-05,
      "loss": 0.38375015258789064,
      "num_input_tokens_seen": 11980582336,
      "step": 42130,
      "train_runtime": 409646.78,
      "train_tokens_per_second": 29246.128
    },
    {
      "epoch": 1.4909963799764268,
      "grad_norm": 0.609375,
      "learning_rate": 2.1897203559693018e-05,
      "loss": 0.38978650569915774,
      "num_input_tokens_seen": 11983450432,
      "step": 42140,
      "train_runtime": 409748.0475,
      "train_tokens_per_second": 29245.9
    },
    {
      "epoch": 1.4913502002401555,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1895103974523347e-05,
      "loss": 0.38601322174072267,
      "num_input_tokens_seen": 11986294720,
      "step": 42150,
      "train_runtime": 409845.928,
      "train_tokens_per_second": 29245.855
    },
    {
      "epoch": 1.4917040205038843,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1893004993185944e-05,
      "loss": 0.3908984184265137,
      "num_input_tokens_seen": 11989119424,
      "step": 42160,
      "train_runtime": 409940.0123,
      "train_tokens_per_second": 29246.034
    },
    {
      "epoch": 1.492057840767613,
      "grad_norm": 0.60546875,
      "learning_rate": 2.189090661539144e-05,
      "loss": 0.39123926162719724,
      "num_input_tokens_seen": 11991940736,
      "step": 42170,
      "train_runtime": 410035.6492,
      "train_tokens_per_second": 29246.093
    },
    {
      "epoch": 1.4924116610313418,
      "grad_norm": 0.609375,
      "learning_rate": 2.188880884085063e-05,
      "loss": 0.38629870414733886,
      "num_input_tokens_seen": 11994823424,
      "step": 42180,
      "train_runtime": 410133.1666,
      "train_tokens_per_second": 29246.168
    },
    {
      "epoch": 1.4927654812950706,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1886711669274547e-05,
      "loss": 0.3888279914855957,
      "num_input_tokens_seen": 11997691712,
      "step": 42190,
      "train_runtime": 410231.6169,
      "train_tokens_per_second": 29246.141
    },
    {
      "epoch": 1.4931193015587994,
      "grad_norm": 0.609375,
      "learning_rate": 2.188461510037438e-05,
      "loss": 0.3864429473876953,
      "num_input_tokens_seen": 12000527232,
      "step": 42200,
      "train_runtime": 410326.9727,
      "train_tokens_per_second": 29246.255
    },
    {
      "epoch": 1.4934731218225281,
      "grad_norm": 0.640625,
      "learning_rate": 2.1882519133861538e-05,
      "loss": 0.38834528923034667,
      "num_input_tokens_seen": 12003381312,
      "step": 42210,
      "train_runtime": 410424.9339,
      "train_tokens_per_second": 29246.228
    },
    {
      "epoch": 1.4938269420862569,
      "grad_norm": 0.57421875,
      "learning_rate": 2.1880423769447607e-05,
      "loss": 0.39205136299133303,
      "num_input_tokens_seen": 12006237440,
      "step": 42220,
      "train_runtime": 410522.1044,
      "train_tokens_per_second": 29246.263
    },
    {
      "epoch": 1.4941807623499859,
      "grad_norm": 0.59375,
      "learning_rate": 2.1878329006844368e-05,
      "loss": 0.38584532737731936,
      "num_input_tokens_seen": 12009064128,
      "step": 42230,
      "train_runtime": 410620.6729,
      "train_tokens_per_second": 29246.126
    },
    {
      "epoch": 1.4945345826137144,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1876234845763806e-05,
      "loss": 0.3882946491241455,
      "num_input_tokens_seen": 12011854592,
      "step": 42240,
      "train_runtime": 410716.0073,
      "train_tokens_per_second": 29246.132
    },
    {
      "epoch": 1.4948884028774434,
      "grad_norm": 0.625,
      "learning_rate": 2.1874141285918086e-05,
      "loss": 0.385976505279541,
      "num_input_tokens_seen": 12014750016,
      "step": 42250,
      "train_runtime": 410818.3783,
      "train_tokens_per_second": 29245.892
    },
    {
      "epoch": 1.495242223141172,
      "grad_norm": 0.609375,
      "learning_rate": 2.1872048327019575e-05,
      "loss": 0.39275107383728025,
      "num_input_tokens_seen": 12017579008,
      "step": 42260,
      "train_runtime": 410913.728,
      "train_tokens_per_second": 29245.991
    },
    {
      "epoch": 1.495596043404901,
      "grad_norm": 0.59375,
      "learning_rate": 2.1869955968780834e-05,
      "loss": 0.3885845184326172,
      "num_input_tokens_seen": 12020409536,
      "step": 42270,
      "train_runtime": 411008.0546,
      "train_tokens_per_second": 29246.165
    },
    {
      "epoch": 1.4959498636686297,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1867864210914605e-05,
      "loss": 0.3897651195526123,
      "num_input_tokens_seen": 12023295872,
      "step": 42280,
      "train_runtime": 411109.5461,
      "train_tokens_per_second": 29245.966
    },
    {
      "epoch": 1.4963036839323585,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1865773053133827e-05,
      "loss": 0.3897970914840698,
      "num_input_tokens_seen": 12026124864,
      "step": 42290,
      "train_runtime": 411205.7304,
      "train_tokens_per_second": 29246.005
    },
    {
      "epoch": 1.4966575041960872,
      "grad_norm": 0.62890625,
      "learning_rate": 2.186368249515164e-05,
      "loss": 0.38759841918945315,
      "num_input_tokens_seen": 12028913472,
      "step": 42300,
      "train_runtime": 411299.913,
      "train_tokens_per_second": 29246.088
    },
    {
      "epoch": 1.497011324459816,
      "grad_norm": 0.625,
      "learning_rate": 2.1861592536681362e-05,
      "loss": 0.38748767375946047,
      "num_input_tokens_seen": 12031675968,
      "step": 42310,
      "train_runtime": 411394.025,
      "train_tokens_per_second": 29246.113
    },
    {
      "epoch": 1.4973651447235448,
      "grad_norm": 0.59375,
      "learning_rate": 2.1859503177436517e-05,
      "loss": 0.3915513038635254,
      "num_input_tokens_seen": 12034532864,
      "step": 42320,
      "train_runtime": 411493.8369,
      "train_tokens_per_second": 29245.961
    },
    {
      "epoch": 1.4977189649872735,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1857414417130808e-05,
      "loss": 0.38543593883514404,
      "num_input_tokens_seen": 12037375104,
      "step": 42330,
      "train_runtime": 411591.5308,
      "train_tokens_per_second": 29245.925
    },
    {
      "epoch": 1.4980727852510023,
      "grad_norm": 0.63671875,
      "learning_rate": 2.1855326255478136e-05,
      "loss": 0.38600897789001465,
      "num_input_tokens_seen": 12040229696,
      "step": 42340,
      "train_runtime": 411690.2796,
      "train_tokens_per_second": 29245.844
    },
    {
      "epoch": 1.498426605514731,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1853238692192597e-05,
      "loss": 0.3888068914413452,
      "num_input_tokens_seen": 12043131968,
      "step": 42350,
      "train_runtime": 411792.2459,
      "train_tokens_per_second": 29245.65
    },
    {
      "epoch": 1.4987804257784598,
      "grad_norm": 0.609375,
      "learning_rate": 2.185115172698846e-05,
      "loss": 0.39337098598480225,
      "num_input_tokens_seen": 12045978368,
      "step": 42360,
      "train_runtime": 411889.3799,
      "train_tokens_per_second": 29245.664
    },
    {
      "epoch": 1.4991342460421886,
      "grad_norm": 0.63671875,
      "learning_rate": 2.1849065359580216e-05,
      "loss": 0.38784613609313967,
      "num_input_tokens_seen": 12048829120,
      "step": 42370,
      "train_runtime": 411986.9061,
      "train_tokens_per_second": 29245.66
    },
    {
      "epoch": 1.4994880663059174,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1846979589682517e-05,
      "loss": 0.38994128704071046,
      "num_input_tokens_seen": 12051688192,
      "step": 42380,
      "train_runtime": 412084.554,
      "train_tokens_per_second": 29245.668
    },
    {
      "epoch": 1.4998418865696461,
      "grad_norm": 0.60546875,
      "learning_rate": 2.184489441701022e-05,
      "loss": 0.38850257396697996,
      "num_input_tokens_seen": 12054501504,
      "step": 42390,
      "train_runtime": 412180.2209,
      "train_tokens_per_second": 29245.706
    },
    {
      "epoch": 1.500195706833375,
      "grad_norm": 0.609375,
      "learning_rate": 2.1842809841278376e-05,
      "loss": 0.3886709451675415,
      "num_input_tokens_seen": 12057332352,
      "step": 42400,
      "train_runtime": 412276.3762,
      "train_tokens_per_second": 29245.751
    },
    {
      "epoch": 1.5005495270971037,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1840725862202214e-05,
      "loss": 0.388775110244751,
      "num_input_tokens_seen": 12060198464,
      "step": 42410,
      "train_runtime": 412375.602,
      "train_tokens_per_second": 29245.664
    },
    {
      "epoch": 1.5009033473608326,
      "grad_norm": 0.59765625,
      "learning_rate": 2.183864247949716e-05,
      "loss": 0.38613133430480956,
      "num_input_tokens_seen": 12063074496,
      "step": 42420,
      "train_runtime": 412477.185,
      "train_tokens_per_second": 29245.434
    },
    {
      "epoch": 1.5012571676245612,
      "grad_norm": 0.609375,
      "learning_rate": 2.183655969287883e-05,
      "loss": 0.3919047832489014,
      "num_input_tokens_seen": 12065890304,
      "step": 42430,
      "train_runtime": 412571.6475,
      "train_tokens_per_second": 29245.563
    },
    {
      "epoch": 1.5016109878882902,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1834477502063034e-05,
      "loss": 0.3853446960449219,
      "num_input_tokens_seen": 12068698816,
      "step": 42440,
      "train_runtime": 412667.7823,
      "train_tokens_per_second": 29245.556
    },
    {
      "epoch": 1.5019648081520187,
      "grad_norm": 0.60546875,
      "learning_rate": 2.183239590676576e-05,
      "loss": 0.38905317783355714,
      "num_input_tokens_seen": 12071541632,
      "step": 42450,
      "train_runtime": 412766.2836,
      "train_tokens_per_second": 29245.464
    },
    {
      "epoch": 1.5023186284157477,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1830314906703198e-05,
      "loss": 0.3881096839904785,
      "num_input_tokens_seen": 12074359232,
      "step": 42460,
      "train_runtime": 412860.7051,
      "train_tokens_per_second": 29245.6
    },
    {
      "epoch": 1.5026724486794765,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1828234501591717e-05,
      "loss": 0.3910414218902588,
      "num_input_tokens_seen": 12077214400,
      "step": 42470,
      "train_runtime": 412959.1654,
      "train_tokens_per_second": 29245.541
    },
    {
      "epoch": 1.5030262689432052,
      "grad_norm": 0.61328125,
      "learning_rate": 2.182615469114789e-05,
      "loss": 0.39313597679138185,
      "num_input_tokens_seen": 12079984256,
      "step": 42480,
      "train_runtime": 413052.2198,
      "train_tokens_per_second": 29245.659
    },
    {
      "epoch": 1.503380089206934,
      "grad_norm": 0.59375,
      "learning_rate": 2.182407547508846e-05,
      "loss": 0.3910232067108154,
      "num_input_tokens_seen": 12082777152,
      "step": 42490,
      "train_runtime": 413145.249,
      "train_tokens_per_second": 29245.833
    },
    {
      "epoch": 1.5037339094706628,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1821996853130363e-05,
      "loss": 0.39245460033416746,
      "num_input_tokens_seen": 12085594880,
      "step": 42500,
      "train_runtime": 413239.6262,
      "train_tokens_per_second": 29245.973
    },
    {
      "epoch": 1.5040877297343915,
      "grad_norm": 0.640625,
      "learning_rate": 2.1819918824990745e-05,
      "loss": 0.391314697265625,
      "num_input_tokens_seen": 12088394176,
      "step": 42510,
      "train_runtime": 413335.1861,
      "train_tokens_per_second": 29245.984
    },
    {
      "epoch": 1.5044415499981203,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1817841390386913e-05,
      "loss": 0.3880293607711792,
      "num_input_tokens_seen": 12091286464,
      "step": 42520,
      "train_runtime": 413437.4819,
      "train_tokens_per_second": 29245.743
    },
    {
      "epoch": 1.504795370261849,
      "grad_norm": 0.578125,
      "learning_rate": 2.1815764549036372e-05,
      "loss": 0.38989307880401614,
      "num_input_tokens_seen": 12094099776,
      "step": 42530,
      "train_runtime": 413532.075,
      "train_tokens_per_second": 29245.857
    },
    {
      "epoch": 1.5051491905255778,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1813688300656823e-05,
      "loss": 0.3885371446609497,
      "num_input_tokens_seen": 12096892544,
      "step": 42540,
      "train_runtime": 413625.6383,
      "train_tokens_per_second": 29245.993
    },
    {
      "epoch": 1.5055030107893068,
      "grad_norm": 0.59375,
      "learning_rate": 2.181161264496615e-05,
      "loss": 0.3847520351409912,
      "num_input_tokens_seen": 12099722304,
      "step": 42550,
      "train_runtime": 413724.8359,
      "train_tokens_per_second": 29245.821
    },
    {
      "epoch": 1.5058568310530354,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1809537581682414e-05,
      "loss": 0.38471245765686035,
      "num_input_tokens_seen": 12102568576,
      "step": 42560,
      "train_runtime": 413825.3597,
      "train_tokens_per_second": 29245.594
    },
    {
      "epoch": 1.5062106513167643,
      "grad_norm": 0.609375,
      "learning_rate": 2.1807463110523877e-05,
      "loss": 0.39519548416137695,
      "num_input_tokens_seen": 12105425984,
      "step": 42570,
      "train_runtime": 413924.7316,
      "train_tokens_per_second": 29245.476
    },
    {
      "epoch": 1.506564471580493,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1805389231208993e-05,
      "loss": 0.387813925743103,
      "num_input_tokens_seen": 12108240576,
      "step": 42580,
      "train_runtime": 414019.6556,
      "train_tokens_per_second": 29245.569
    },
    {
      "epoch": 1.5069182918442219,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1803315943456386e-05,
      "loss": 0.3898014545440674,
      "num_input_tokens_seen": 12111070208,
      "step": 42590,
      "train_runtime": 414118.1279,
      "train_tokens_per_second": 29245.448
    },
    {
      "epoch": 1.5072721121079504,
      "grad_norm": 0.62890625,
      "learning_rate": 2.180124324698488e-05,
      "loss": 0.39206585884094236,
      "num_input_tokens_seen": 12113892736,
      "step": 42600,
      "train_runtime": 414213.2032,
      "train_tokens_per_second": 29245.549
    },
    {
      "epoch": 1.5076259323716794,
      "grad_norm": 0.59375,
      "learning_rate": 2.1799171141513476e-05,
      "loss": 0.38569135665893556,
      "num_input_tokens_seen": 12116740160,
      "step": 42610,
      "train_runtime": 414311.518,
      "train_tokens_per_second": 29245.482
    },
    {
      "epoch": 1.507979752635408,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1797099626761385e-05,
      "loss": 0.388590669631958,
      "num_input_tokens_seen": 12119576064,
      "step": 42620,
      "train_runtime": 414408.8388,
      "train_tokens_per_second": 29245.457
    },
    {
      "epoch": 1.508333572899137,
      "grad_norm": 0.59765625,
      "learning_rate": 2.179502870244797e-05,
      "loss": 0.3887790679931641,
      "num_input_tokens_seen": 12122361472,
      "step": 42630,
      "train_runtime": 414504.0187,
      "train_tokens_per_second": 29245.462
    },
    {
      "epoch": 1.5086873931628657,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1792958368292814e-05,
      "loss": 0.3878232479095459,
      "num_input_tokens_seen": 12125185728,
      "step": 42640,
      "train_runtime": 414598.1624,
      "train_tokens_per_second": 29245.633
    },
    {
      "epoch": 1.5090412134265945,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1790888624015656e-05,
      "loss": 0.38990106582641604,
      "num_input_tokens_seen": 12128054336,
      "step": 42650,
      "train_runtime": 414693.8636,
      "train_tokens_per_second": 29245.801
    },
    {
      "epoch": 1.5093950336903232,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1788819469336452e-05,
      "loss": 0.38659181594848635,
      "num_input_tokens_seen": 12130873984,
      "step": 42660,
      "train_runtime": 414788.646,
      "train_tokens_per_second": 29245.916
    },
    {
      "epoch": 1.509748853954052,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1786750903975316e-05,
      "loss": 0.3870596647262573,
      "num_input_tokens_seen": 12133724032,
      "step": 42670,
      "train_runtime": 414886.7789,
      "train_tokens_per_second": 29245.868
    },
    {
      "epoch": 1.5101026742177808,
      "grad_norm": 0.60546875,
      "learning_rate": 2.178468292765257e-05,
      "loss": 0.3873318672180176,
      "num_input_tokens_seen": 12136596352,
      "step": 42680,
      "train_runtime": 414985.2269,
      "train_tokens_per_second": 29245.852
    },
    {
      "epoch": 1.5104564944815095,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1782615540088703e-05,
      "loss": 0.3861170053482056,
      "num_input_tokens_seen": 12139420544,
      "step": 42690,
      "train_runtime": 415083.1337,
      "train_tokens_per_second": 29245.757
    },
    {
      "epoch": 1.5108103147452383,
      "grad_norm": 0.58203125,
      "learning_rate": 2.178054874100441e-05,
      "loss": 0.38941497802734376,
      "num_input_tokens_seen": 12142253632,
      "step": 42700,
      "train_runtime": 415180.3735,
      "train_tokens_per_second": 29245.731
    },
    {
      "epoch": 1.511164135008967,
      "grad_norm": 0.59375,
      "learning_rate": 2.177848253012055e-05,
      "loss": 0.3882740020751953,
      "num_input_tokens_seen": 12145089920,
      "step": 42710,
      "train_runtime": 415276.7605,
      "train_tokens_per_second": 29245.773
    },
    {
      "epoch": 1.511517955272696,
      "grad_norm": 0.6171875,
      "learning_rate": 2.177641690715819e-05,
      "loss": 0.3938926696777344,
      "num_input_tokens_seen": 12148039232,
      "step": 42720,
      "train_runtime": 415380.1582,
      "train_tokens_per_second": 29245.593
    },
    {
      "epoch": 1.5118717755364246,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1774351871838557e-05,
      "loss": 0.3898476123809814,
      "num_input_tokens_seen": 12150827264,
      "step": 42730,
      "train_runtime": 415473.5174,
      "train_tokens_per_second": 29245.732
    },
    {
      "epoch": 1.5122255958001536,
      "grad_norm": 0.58984375,
      "learning_rate": 2.177228742388308e-05,
      "loss": 0.38460755348205566,
      "num_input_tokens_seen": 12153618240,
      "step": 42740,
      "train_runtime": 415568.7935,
      "train_tokens_per_second": 29245.743
    },
    {
      "epoch": 1.5125794160638821,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1770223563013374e-05,
      "loss": 0.3908083915710449,
      "num_input_tokens_seen": 12156494272,
      "step": 42750,
      "train_runtime": 415667.4388,
      "train_tokens_per_second": 29245.722
    },
    {
      "epoch": 1.5129332363276111,
      "grad_norm": 0.6015625,
      "learning_rate": 2.176816028895123e-05,
      "loss": 0.39047675132751464,
      "num_input_tokens_seen": 12159317760,
      "step": 42760,
      "train_runtime": 415761.6049,
      "train_tokens_per_second": 29245.889
    },
    {
      "epoch": 1.5132870565913397,
      "grad_norm": 0.609375,
      "learning_rate": 2.1766097601418624e-05,
      "loss": 0.38875522613525393,
      "num_input_tokens_seen": 12162160128,
      "step": 42770,
      "train_runtime": 415857.3658,
      "train_tokens_per_second": 29245.989
    },
    {
      "epoch": 1.5136408768550687,
      "grad_norm": 0.6328125,
      "learning_rate": 2.176403550013773e-05,
      "loss": 0.38666815757751466,
      "num_input_tokens_seen": 12164983872,
      "step": 42780,
      "train_runtime": 415954.08,
      "train_tokens_per_second": 29245.978
    },
    {
      "epoch": 1.5139946971187974,
      "grad_norm": 0.609375,
      "learning_rate": 2.176197398483088e-05,
      "loss": 0.3883677005767822,
      "num_input_tokens_seen": 12167836288,
      "step": 42790,
      "train_runtime": 416050.914,
      "train_tokens_per_second": 29246.027
    },
    {
      "epoch": 1.5143485173825262,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1759913055220623e-05,
      "loss": 0.3917588233947754,
      "num_input_tokens_seen": 12170711296,
      "step": 42800,
      "train_runtime": 416147.5851,
      "train_tokens_per_second": 29246.142
    },
    {
      "epoch": 1.514702337646255,
      "grad_norm": 0.609375,
      "learning_rate": 2.1757852711029664e-05,
      "loss": 0.3863863468170166,
      "num_input_tokens_seen": 12173560000,
      "step": 42810,
      "train_runtime": 416245.8584,
      "train_tokens_per_second": 29246.081
    },
    {
      "epoch": 1.5150561579099837,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1755792951980896e-05,
      "loss": 0.3919061660766602,
      "num_input_tokens_seen": 12176367744,
      "step": 42820,
      "train_runtime": 416342.7478,
      "train_tokens_per_second": 29246.019
    },
    {
      "epoch": 1.5154099781737125,
      "grad_norm": 0.59765625,
      "learning_rate": 2.175373377779742e-05,
      "loss": 0.3897319078445435,
      "num_input_tokens_seen": 12179276928,
      "step": 42830,
      "train_runtime": 416442.7031,
      "train_tokens_per_second": 29245.985
    },
    {
      "epoch": 1.5157637984374412,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1751675188202493e-05,
      "loss": 0.3912724494934082,
      "num_input_tokens_seen": 12182130112,
      "step": 42840,
      "train_runtime": 416538.2024,
      "train_tokens_per_second": 29246.129
    },
    {
      "epoch": 1.51611761870117,
      "grad_norm": 0.6171875,
      "learning_rate": 2.174961718291956e-05,
      "loss": 0.38510124683380126,
      "num_input_tokens_seen": 12184963072,
      "step": 42850,
      "train_runtime": 416633.4323,
      "train_tokens_per_second": 29246.244
    },
    {
      "epoch": 1.5164714389648988,
      "grad_norm": 0.58984375,
      "learning_rate": 2.174755976167226e-05,
      "loss": 0.3861616373062134,
      "num_input_tokens_seen": 12187821696,
      "step": 42860,
      "train_runtime": 416730.2961,
      "train_tokens_per_second": 29246.306
    },
    {
      "epoch": 1.5168252592286275,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1745502924184407e-05,
      "loss": 0.38733668327331544,
      "num_input_tokens_seen": 12190711616,
      "step": 42870,
      "train_runtime": 416826.4984,
      "train_tokens_per_second": 29246.489
    },
    {
      "epoch": 1.5171790794923563,
      "grad_norm": 0.609375,
      "learning_rate": 2.1743446670180005e-05,
      "loss": 0.3877874374389648,
      "num_input_tokens_seen": 12193600256,
      "step": 42880,
      "train_runtime": 416922.3548,
      "train_tokens_per_second": 29246.693
    },
    {
      "epoch": 1.5175328997560853,
      "grad_norm": 0.609375,
      "learning_rate": 2.1741390999383225e-05,
      "loss": 0.3919093370437622,
      "num_input_tokens_seen": 12196479168,
      "step": 42890,
      "train_runtime": 417020.8529,
      "train_tokens_per_second": 29246.689
    },
    {
      "epoch": 1.5178867200198138,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1739335911518432e-05,
      "loss": 0.38647260665893557,
      "num_input_tokens_seen": 12199325632,
      "step": 42900,
      "train_runtime": 417115.739,
      "train_tokens_per_second": 29246.86
    },
    {
      "epoch": 1.5182405402835428,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1737281406310186e-05,
      "loss": 0.38724236488342284,
      "num_input_tokens_seen": 12202148992,
      "step": 42910,
      "train_runtime": 417213.8127,
      "train_tokens_per_second": 29246.752
    },
    {
      "epoch": 1.5185943605472714,
      "grad_norm": 0.6015625,
      "learning_rate": 2.17352274834832e-05,
      "loss": 0.39097347259521487,
      "num_input_tokens_seen": 12204924544,
      "step": 42920,
      "train_runtime": 417307.722,
      "train_tokens_per_second": 29246.822
    },
    {
      "epoch": 1.5189481808110004,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1733174142762388e-05,
      "loss": 0.38936848640441896,
      "num_input_tokens_seen": 12207786432,
      "step": 42930,
      "train_runtime": 417405.3406,
      "train_tokens_per_second": 29246.838
    },
    {
      "epoch": 1.519302001074729,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1731121383872852e-05,
      "loss": 0.38815531730651853,
      "num_input_tokens_seen": 12210541632,
      "step": 42940,
      "train_runtime": 417496.5612,
      "train_tokens_per_second": 29247.047
    },
    {
      "epoch": 1.519655821338458,
      "grad_norm": 0.5859375,
      "learning_rate": 2.1729069206539847e-05,
      "loss": 0.3922411441802979,
      "num_input_tokens_seen": 12213365440,
      "step": 42950,
      "train_runtime": 417593.9831,
      "train_tokens_per_second": 29246.986
    },
    {
      "epoch": 1.5200096416021867,
      "grad_norm": 0.625,
      "learning_rate": 2.172701761048885e-05,
      "loss": 0.3850921630859375,
      "num_input_tokens_seen": 12216205056,
      "step": 42960,
      "train_runtime": 417692.3333,
      "train_tokens_per_second": 29246.898
    },
    {
      "epoch": 1.5203634618659154,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1724966595445483e-05,
      "loss": 0.38896331787109373,
      "num_input_tokens_seen": 12219023168,
      "step": 42970,
      "train_runtime": 417787.0532,
      "train_tokens_per_second": 29247.013
    },
    {
      "epoch": 1.5207172821296442,
      "grad_norm": 0.59765625,
      "learning_rate": 2.172291616113557e-05,
      "loss": 0.39084429740905763,
      "num_input_tokens_seen": 12221892480,
      "step": 42980,
      "train_runtime": 417887.3888,
      "train_tokens_per_second": 29246.856
    },
    {
      "epoch": 1.521071102393373,
      "grad_norm": 0.59765625,
      "learning_rate": 2.172086630728511e-05,
      "loss": 0.3847005367279053,
      "num_input_tokens_seen": 12224784640,
      "step": 42990,
      "train_runtime": 417987.2754,
      "train_tokens_per_second": 29246.787
    },
    {
      "epoch": 1.5214249226571017,
      "grad_norm": 0.58984375,
      "learning_rate": 2.171881703362028e-05,
      "loss": 0.3871195316314697,
      "num_input_tokens_seen": 12227680960,
      "step": 43000,
      "train_runtime": 418087.3327,
      "train_tokens_per_second": 29246.715
    },
    {
      "epoch": 1.5217787429208305,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1716768339867445e-05,
      "loss": 0.3897268533706665,
      "num_input_tokens_seen": 12230492544,
      "step": 43010,
      "train_runtime": 418181.6483,
      "train_tokens_per_second": 29246.842
    },
    {
      "epoch": 1.5221325631845593,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1714720225753148e-05,
      "loss": 0.3881224155426025,
      "num_input_tokens_seen": 12233338944,
      "step": 43020,
      "train_runtime": 418276.2105,
      "train_tokens_per_second": 29247.035
    },
    {
      "epoch": 1.522486383448288,
      "grad_norm": 0.5859375,
      "learning_rate": 2.1712672691004098e-05,
      "loss": 0.38787715435028075,
      "num_input_tokens_seen": 12236178880,
      "step": 43030,
      "train_runtime": 418375.055,
      "train_tokens_per_second": 29246.913
    },
    {
      "epoch": 1.522840203712017,
      "grad_norm": 0.5859375,
      "learning_rate": 2.171062573534722e-05,
      "loss": 0.39142684936523436,
      "num_input_tokens_seen": 12238977280,
      "step": 43040,
      "train_runtime": 418470.0868,
      "train_tokens_per_second": 29246.959
    },
    {
      "epoch": 1.5231940239757455,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1708579358509578e-05,
      "loss": 0.3917617082595825,
      "num_input_tokens_seen": 12241750400,
      "step": 43050,
      "train_runtime": 418564.536,
      "train_tokens_per_second": 29246.984
    },
    {
      "epoch": 1.5235478442394745,
      "grad_norm": 0.59375,
      "learning_rate": 2.1706533560218447e-05,
      "loss": 0.3920008182525635,
      "num_input_tokens_seen": 12244605504,
      "step": 43060,
      "train_runtime": 418661.8291,
      "train_tokens_per_second": 29247.007
    },
    {
      "epoch": 1.523901664503203,
      "grad_norm": 0.625,
      "learning_rate": 2.1704488340201262e-05,
      "loss": 0.39245123863220216,
      "num_input_tokens_seen": 12247415680,
      "step": 43070,
      "train_runtime": 418758.0275,
      "train_tokens_per_second": 29246.999
    },
    {
      "epoch": 1.524255484766932,
      "grad_norm": 0.5859375,
      "learning_rate": 2.170244369818564e-05,
      "loss": 0.38855547904968263,
      "num_input_tokens_seen": 12250276096,
      "step": 43080,
      "train_runtime": 418853.7655,
      "train_tokens_per_second": 29247.143
    },
    {
      "epoch": 1.5246093050306606,
      "grad_norm": 0.63671875,
      "learning_rate": 2.17003996338994e-05,
      "loss": 0.384521484375,
      "num_input_tokens_seen": 12253115968,
      "step": 43090,
      "train_runtime": 418951.1559,
      "train_tokens_per_second": 29247.123
    },
    {
      "epoch": 1.5249631252943896,
      "grad_norm": 0.578125,
      "learning_rate": 2.1698356147070515e-05,
      "loss": 0.38777961730957033,
      "num_input_tokens_seen": 12255938880,
      "step": 43100,
      "train_runtime": 419044.361,
      "train_tokens_per_second": 29247.354
    },
    {
      "epoch": 1.5253169455581181,
      "grad_norm": 0.6015625,
      "learning_rate": 2.169631323742714e-05,
      "loss": 0.3913033246994019,
      "num_input_tokens_seen": 12258805568,
      "step": 43110,
      "train_runtime": 419140.5134,
      "train_tokens_per_second": 29247.484
    },
    {
      "epoch": 1.5256707658218471,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1694270904697623e-05,
      "loss": 0.38835670948028567,
      "num_input_tokens_seen": 12261658944,
      "step": 43120,
      "train_runtime": 419236.7749,
      "train_tokens_per_second": 29247.575
    },
    {
      "epoch": 1.526024586085576,
      "grad_norm": 0.63671875,
      "learning_rate": 2.169222914861048e-05,
      "loss": 0.39002649784088134,
      "num_input_tokens_seen": 12264533888,
      "step": 43130,
      "train_runtime": 419334.8963,
      "train_tokens_per_second": 29247.587
    },
    {
      "epoch": 1.5263784063493047,
      "grad_norm": 0.609375,
      "learning_rate": 2.1690187968894404e-05,
      "loss": 0.3865140438079834,
      "num_input_tokens_seen": 12267309504,
      "step": 43140,
      "train_runtime": 419427.1286,
      "train_tokens_per_second": 29247.773
    },
    {
      "epoch": 1.5267322266130334,
      "grad_norm": 0.6015625,
      "learning_rate": 2.168814736527828e-05,
      "loss": 0.3910662651062012,
      "num_input_tokens_seen": 12270153472,
      "step": 43150,
      "train_runtime": 419525.1251,
      "train_tokens_per_second": 29247.72
    },
    {
      "epoch": 1.5270860468767622,
      "grad_norm": 0.59375,
      "learning_rate": 2.1686107337491155e-05,
      "loss": 0.39094431400299073,
      "num_input_tokens_seen": 12272988096,
      "step": 43160,
      "train_runtime": 419621.0618,
      "train_tokens_per_second": 29247.789
    },
    {
      "epoch": 1.527439867140491,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1684067885262266e-05,
      "loss": 0.39046144485473633,
      "num_input_tokens_seen": 12275819520,
      "step": 43170,
      "train_runtime": 419717.1906,
      "train_tokens_per_second": 29247.836
    },
    {
      "epoch": 1.5277936874042197,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1682029008321024e-05,
      "loss": 0.39098682403564455,
      "num_input_tokens_seen": 12278695552,
      "step": 43180,
      "train_runtime": 419816.8514,
      "train_tokens_per_second": 29247.743
    },
    {
      "epoch": 1.5281475076679485,
      "grad_norm": 0.60546875,
      "learning_rate": 2.167999070639702e-05,
      "loss": 0.38949458599090575,
      "num_input_tokens_seen": 12281526208,
      "step": 43190,
      "train_runtime": 419915.4518,
      "train_tokens_per_second": 29247.617
    },
    {
      "epoch": 1.5285013279316773,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1677952979220015e-05,
      "loss": 0.3916356563568115,
      "num_input_tokens_seen": 12284343616,
      "step": 43200,
      "train_runtime": 420011.7067,
      "train_tokens_per_second": 29247.622
    },
    {
      "epoch": 1.5288551481954062,
      "grad_norm": 0.609375,
      "learning_rate": 2.167591582651996e-05,
      "loss": 0.39000668525695803,
      "num_input_tokens_seen": 12287138176,
      "step": 43210,
      "train_runtime": 420107.5473,
      "train_tokens_per_second": 29247.602
    },
    {
      "epoch": 1.5292089684591348,
      "grad_norm": 0.609375,
      "learning_rate": 2.1673879248026977e-05,
      "loss": 0.3884908437728882,
      "num_input_tokens_seen": 12289980992,
      "step": 43220,
      "train_runtime": 420204.4627,
      "train_tokens_per_second": 29247.621
    },
    {
      "epoch": 1.5295627887228638,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1671843243471366e-05,
      "loss": 0.39139065742492674,
      "num_input_tokens_seen": 12292877440,
      "step": 43230,
      "train_runtime": 420303.3254,
      "train_tokens_per_second": 29247.633
    },
    {
      "epoch": 1.5299166089865923,
      "grad_norm": 0.609375,
      "learning_rate": 2.1669807812583602e-05,
      "loss": 0.3898471355438232,
      "num_input_tokens_seen": 12295747712,
      "step": 43240,
      "train_runtime": 420402.7648,
      "train_tokens_per_second": 29247.542
    },
    {
      "epoch": 1.5302704292503213,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1667772955094347e-05,
      "loss": 0.387610125541687,
      "num_input_tokens_seen": 12298618496,
      "step": 43250,
      "train_runtime": 420499.9824,
      "train_tokens_per_second": 29247.608
    },
    {
      "epoch": 1.5306242495140499,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1665738670734422e-05,
      "loss": 0.3915060043334961,
      "num_input_tokens_seen": 12301503552,
      "step": 43260,
      "train_runtime": 420601.8545,
      "train_tokens_per_second": 29247.383
    },
    {
      "epoch": 1.5309780697777788,
      "grad_norm": 0.57421875,
      "learning_rate": 2.166370495923485e-05,
      "loss": 0.396150803565979,
      "num_input_tokens_seen": 12304304192,
      "step": 43270,
      "train_runtime": 420695.1807,
      "train_tokens_per_second": 29247.552
    },
    {
      "epoch": 1.5313318900415074,
      "grad_norm": 0.57421875,
      "learning_rate": 2.16616718203268e-05,
      "loss": 0.39034366607666016,
      "num_input_tokens_seen": 12307178752,
      "step": 43280,
      "train_runtime": 420792.5669,
      "train_tokens_per_second": 29247.614
    },
    {
      "epoch": 1.5316857103052364,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1659639253741637e-05,
      "loss": 0.3915281295776367,
      "num_input_tokens_seen": 12310115008,
      "step": 43290,
      "train_runtime": 420897.1642,
      "train_tokens_per_second": 29247.322
    },
    {
      "epoch": 1.5320395305689651,
      "grad_norm": 0.59375,
      "learning_rate": 2.1657607259210906e-05,
      "loss": 0.3863854885101318,
      "num_input_tokens_seen": 12312915904,
      "step": 43300,
      "train_runtime": 420990.8421,
      "train_tokens_per_second": 29247.467
    },
    {
      "epoch": 1.532393350832694,
      "grad_norm": 0.59375,
      "learning_rate": 2.1655575836466328e-05,
      "loss": 0.38784518241882326,
      "num_input_tokens_seen": 12315801152,
      "step": 43310,
      "train_runtime": 421087.3088,
      "train_tokens_per_second": 29247.619
    },
    {
      "epoch": 1.5327471710964227,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1653544985239775e-05,
      "loss": 0.3899970769882202,
      "num_input_tokens_seen": 12318612608,
      "step": 43320,
      "train_runtime": 421182.8268,
      "train_tokens_per_second": 29247.661
    },
    {
      "epoch": 1.5331009913601514,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1651514705263324e-05,
      "loss": 0.38704278469085696,
      "num_input_tokens_seen": 12321439360,
      "step": 43330,
      "train_runtime": 421279.7469,
      "train_tokens_per_second": 29247.642
    },
    {
      "epoch": 1.5334548116238802,
      "grad_norm": 0.59375,
      "learning_rate": 2.164948499626922e-05,
      "loss": 0.3893830060958862,
      "num_input_tokens_seen": 12324267840,
      "step": 43340,
      "train_runtime": 421375.5103,
      "train_tokens_per_second": 29247.708
    },
    {
      "epoch": 1.533808631887609,
      "grad_norm": 0.609375,
      "learning_rate": 2.1647455857989867e-05,
      "loss": 0.3891900539398193,
      "num_input_tokens_seen": 12327083776,
      "step": 43350,
      "train_runtime": 421473.195,
      "train_tokens_per_second": 29247.61
    },
    {
      "epoch": 1.5341624521513377,
      "grad_norm": 0.62890625,
      "learning_rate": 2.164542729015788e-05,
      "loss": 0.3932191848754883,
      "num_input_tokens_seen": 12329918656,
      "step": 43360,
      "train_runtime": 421568.3761,
      "train_tokens_per_second": 29247.731
    },
    {
      "epoch": 1.5345162724150665,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1643399292506014e-05,
      "loss": 0.3885686159133911,
      "num_input_tokens_seen": 12332732800,
      "step": 43370,
      "train_runtime": 421662.9073,
      "train_tokens_per_second": 29247.848
    },
    {
      "epoch": 1.5348700926787955,
      "grad_norm": 0.60546875,
      "learning_rate": 2.164137186476722e-05,
      "loss": 0.39277896881103513,
      "num_input_tokens_seen": 12335606464,
      "step": 43380,
      "train_runtime": 421759.9218,
      "train_tokens_per_second": 29247.934
    },
    {
      "epoch": 1.535223912942524,
      "grad_norm": 0.63671875,
      "learning_rate": 2.1639345006674607e-05,
      "loss": 0.38790106773376465,
      "num_input_tokens_seen": 12338514496,
      "step": 43390,
      "train_runtime": 421859.7552,
      "train_tokens_per_second": 29247.906
    },
    {
      "epoch": 1.535577733206253,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1637318717961477e-05,
      "loss": 0.3909615993499756,
      "num_input_tokens_seen": 12341371840,
      "step": 43400,
      "train_runtime": 421957.0989,
      "train_tokens_per_second": 29247.93
    },
    {
      "epoch": 1.5359315534699816,
      "grad_norm": 0.6015625,
      "learning_rate": 2.16352929983613e-05,
      "loss": 0.3921290397644043,
      "num_input_tokens_seen": 12344204800,
      "step": 43410,
      "train_runtime": 422052.037,
      "train_tokens_per_second": 29248.064
    },
    {
      "epoch": 1.5362853737337105,
      "grad_norm": 0.59375,
      "learning_rate": 2.1633267847607718e-05,
      "loss": 0.3894049882888794,
      "num_input_tokens_seen": 12347099648,
      "step": 43420,
      "train_runtime": 422154.1794,
      "train_tokens_per_second": 29247.844
    },
    {
      "epoch": 1.536639193997439,
      "grad_norm": 0.5859375,
      "learning_rate": 2.1631243265434543e-05,
      "loss": 0.39225380420684813,
      "num_input_tokens_seen": 12349994944,
      "step": 43430,
      "train_runtime": 422251.2379,
      "train_tokens_per_second": 29247.978
    },
    {
      "epoch": 1.536993014261168,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1629219251575783e-05,
      "loss": 0.3890273094177246,
      "num_input_tokens_seen": 12352848384,
      "step": 43440,
      "train_runtime": 422347.2348,
      "train_tokens_per_second": 29248.086
    },
    {
      "epoch": 1.5373468345248966,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1627195805765592e-05,
      "loss": 0.3877848148345947,
      "num_input_tokens_seen": 12355717632,
      "step": 43450,
      "train_runtime": 422446.7528,
      "train_tokens_per_second": 29247.988
    },
    {
      "epoch": 1.5377006547886256,
      "grad_norm": 0.609375,
      "learning_rate": 2.162517292773831e-05,
      "loss": 0.3912092685699463,
      "num_input_tokens_seen": 12358574016,
      "step": 43460,
      "train_runtime": 422543.7612,
      "train_tokens_per_second": 29248.033
    },
    {
      "epoch": 1.5380544750523544,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1623150617228456e-05,
      "loss": 0.384053897857666,
      "num_input_tokens_seen": 12361382976,
      "step": 43470,
      "train_runtime": 422639.1415,
      "train_tokens_per_second": 29248.079
    },
    {
      "epoch": 1.5384082953160831,
      "grad_norm": 0.63671875,
      "learning_rate": 2.1621128873970715e-05,
      "loss": 0.3927616119384766,
      "num_input_tokens_seen": 12364179456,
      "step": 43480,
      "train_runtime": 422733.6381,
      "train_tokens_per_second": 29248.156
    },
    {
      "epoch": 1.538762115579812,
      "grad_norm": 0.57421875,
      "learning_rate": 2.1619107697699955e-05,
      "loss": 0.382613205909729,
      "num_input_tokens_seen": 12367064192,
      "step": 43490,
      "train_runtime": 422831.9757,
      "train_tokens_per_second": 29248.176
    },
    {
      "epoch": 1.5391159358435407,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1617087088151208e-05,
      "loss": 0.38992834091186523,
      "num_input_tokens_seen": 12369846336,
      "step": 43500,
      "train_runtime": 422923.1623,
      "train_tokens_per_second": 29248.449
    },
    {
      "epoch": 1.5394697561072694,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1615067045059685e-05,
      "loss": 0.3945052146911621,
      "num_input_tokens_seen": 12372748672,
      "step": 43510,
      "train_runtime": 423026.0475,
      "train_tokens_per_second": 29248.196
    },
    {
      "epoch": 1.5398235763709982,
      "grad_norm": 0.59375,
      "learning_rate": 2.161304756816076e-05,
      "loss": 0.38994674682617186,
      "num_input_tokens_seen": 12375564032,
      "step": 43520,
      "train_runtime": 423120.2176,
      "train_tokens_per_second": 29248.34
    },
    {
      "epoch": 1.540177396634727,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1611028657189997e-05,
      "loss": 0.3891166687011719,
      "num_input_tokens_seen": 12378426496,
      "step": 43530,
      "train_runtime": 423217.9715,
      "train_tokens_per_second": 29248.348
    },
    {
      "epoch": 1.5405312168984557,
      "grad_norm": 0.6484375,
      "learning_rate": 2.160901031188312e-05,
      "loss": 0.3958103179931641,
      "num_input_tokens_seen": 12381221568,
      "step": 43540,
      "train_runtime": 423311.8014,
      "train_tokens_per_second": 29248.468
    },
    {
      "epoch": 1.5408850371621847,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1606992531976027e-05,
      "loss": 0.38954644203186034,
      "num_input_tokens_seen": 12384109184,
      "step": 43550,
      "train_runtime": 423411.4636,
      "train_tokens_per_second": 29248.403
    },
    {
      "epoch": 1.5412388574259133,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1604975317204805e-05,
      "loss": 0.38824265003204345,
      "num_input_tokens_seen": 12386943936,
      "step": 43560,
      "train_runtime": 423507.6904,
      "train_tokens_per_second": 29248.451
    },
    {
      "epoch": 1.5415926776896423,
      "grad_norm": 0.6328125,
      "learning_rate": 2.160295866730568e-05,
      "loss": 0.3835118293762207,
      "num_input_tokens_seen": 12389767488,
      "step": 43570,
      "train_runtime": 423604.7275,
      "train_tokens_per_second": 29248.416
    },
    {
      "epoch": 1.5419464979533708,
      "grad_norm": 0.609375,
      "learning_rate": 2.160094258201508e-05,
      "loss": 0.39267544746398925,
      "num_input_tokens_seen": 12392641280,
      "step": 43580,
      "train_runtime": 423704.1664,
      "train_tokens_per_second": 29248.335
    },
    {
      "epoch": 1.5423003182170998,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1598927061069596e-05,
      "loss": 0.3896996736526489,
      "num_input_tokens_seen": 12395480384,
      "step": 43590,
      "train_runtime": 423800.3095,
      "train_tokens_per_second": 29248.399
    },
    {
      "epoch": 1.5426541384808283,
      "grad_norm": 0.61328125,
      "learning_rate": 2.159691210420599e-05,
      "loss": 0.3902219533920288,
      "num_input_tokens_seen": 12398313600,
      "step": 43600,
      "train_runtime": 423897.716,
      "train_tokens_per_second": 29248.361
    },
    {
      "epoch": 1.5430079587445573,
      "grad_norm": 0.609375,
      "learning_rate": 2.1594897711161196e-05,
      "loss": 0.386974573135376,
      "num_input_tokens_seen": 12401137984,
      "step": 43610,
      "train_runtime": 423995.171,
      "train_tokens_per_second": 29248.3
    },
    {
      "epoch": 1.543361779008286,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1592883881672317e-05,
      "loss": 0.38967456817626955,
      "num_input_tokens_seen": 12403944384,
      "step": 43620,
      "train_runtime": 424090.8996,
      "train_tokens_per_second": 29248.315
    },
    {
      "epoch": 1.5437155992720148,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1590870615476628e-05,
      "loss": 0.388608455657959,
      "num_input_tokens_seen": 12406708288,
      "step": 43630,
      "train_runtime": 424182.7507,
      "train_tokens_per_second": 29248.498
    },
    {
      "epoch": 1.5440694195357436,
      "grad_norm": 0.578125,
      "learning_rate": 2.158885791231158e-05,
      "loss": 0.38902058601379397,
      "num_input_tokens_seen": 12409582400,
      "step": 43640,
      "train_runtime": 424283.7626,
      "train_tokens_per_second": 29248.309
    },
    {
      "epoch": 1.5444232397994724,
      "grad_norm": 0.578125,
      "learning_rate": 2.15868457719148e-05,
      "loss": 0.38901457786560056,
      "num_input_tokens_seen": 12412413120,
      "step": 43650,
      "train_runtime": 424380.7537,
      "train_tokens_per_second": 29248.294
    },
    {
      "epoch": 1.5447770600632011,
      "grad_norm": 0.61328125,
      "learning_rate": 2.158483419402407e-05,
      "loss": 0.3894495964050293,
      "num_input_tokens_seen": 12415259136,
      "step": 43660,
      "train_runtime": 424476.3555,
      "train_tokens_per_second": 29248.412
    },
    {
      "epoch": 1.54513088032693,
      "grad_norm": 0.6171875,
      "learning_rate": 2.158282317837736e-05,
      "loss": 0.3907548189163208,
      "num_input_tokens_seen": 12418113728,
      "step": 43670,
      "train_runtime": 424575.742,
      "train_tokens_per_second": 29248.288
    },
    {
      "epoch": 1.5454847005906587,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1580812724712792e-05,
      "loss": 0.385688853263855,
      "num_input_tokens_seen": 12420983232,
      "step": 43680,
      "train_runtime": 424673.6897,
      "train_tokens_per_second": 29248.299
    },
    {
      "epoch": 1.5458385208543874,
      "grad_norm": 0.60546875,
      "learning_rate": 2.157880283276868e-05,
      "loss": 0.3868332624435425,
      "num_input_tokens_seen": 12423910528,
      "step": 43690,
      "train_runtime": 424778.5156,
      "train_tokens_per_second": 29247.973
    },
    {
      "epoch": 1.5461923411181162,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1576793502283494e-05,
      "loss": 0.38932070732116697,
      "num_input_tokens_seen": 12426757504,
      "step": 43700,
      "train_runtime": 424877.2188,
      "train_tokens_per_second": 29247.879
    },
    {
      "epoch": 1.546546161381845,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1574784732995884e-05,
      "loss": 0.3905416250228882,
      "num_input_tokens_seen": 12429555392,
      "step": 43710,
      "train_runtime": 424970.9095,
      "train_tokens_per_second": 29248.015
    },
    {
      "epoch": 1.546899981645574,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1572776524644658e-05,
      "loss": 0.38866915702819826,
      "num_input_tokens_seen": 12432411840,
      "step": 43720,
      "train_runtime": 425071.9054,
      "train_tokens_per_second": 29247.785
    },
    {
      "epoch": 1.5472538019093025,
      "grad_norm": 0.609375,
      "learning_rate": 2.1570768876968802e-05,
      "loss": 0.38974111080169677,
      "num_input_tokens_seen": 12435229312,
      "step": 43730,
      "train_runtime": 425166.919,
      "train_tokens_per_second": 29247.876
    },
    {
      "epoch": 1.5476076221730315,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1568761789707474e-05,
      "loss": 0.3852193355560303,
      "num_input_tokens_seen": 12438057088,
      "step": 43740,
      "train_runtime": 425262.7791,
      "train_tokens_per_second": 29247.933
    },
    {
      "epoch": 1.54796144243676,
      "grad_norm": 0.609375,
      "learning_rate": 2.15667552626e-05,
      "loss": 0.3881957530975342,
      "num_input_tokens_seen": 12440882112,
      "step": 43750,
      "train_runtime": 425359.9747,
      "train_tokens_per_second": 29247.891
    },
    {
      "epoch": 1.548315262700489,
      "grad_norm": 0.58203125,
      "learning_rate": 2.1564749295385872e-05,
      "loss": 0.390362286567688,
      "num_input_tokens_seen": 12443744384,
      "step": 43760,
      "train_runtime": 425458.2629,
      "train_tokens_per_second": 29247.862
    },
    {
      "epoch": 1.5486690829642176,
      "grad_norm": 0.625,
      "learning_rate": 2.1562743887804757e-05,
      "loss": 0.3880608081817627,
      "num_input_tokens_seen": 12446624256,
      "step": 43770,
      "train_runtime": 425558.5004,
      "train_tokens_per_second": 29247.74
    },
    {
      "epoch": 1.5490229032279466,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1560739039596485e-05,
      "loss": 0.3888096332550049,
      "num_input_tokens_seen": 12449464896,
      "step": 43780,
      "train_runtime": 425656.4557,
      "train_tokens_per_second": 29247.683
    },
    {
      "epoch": 1.5493767234916753,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1558734750501068e-05,
      "loss": 0.3873999834060669,
      "num_input_tokens_seen": 12452267776,
      "step": 43790,
      "train_runtime": 425751.7086,
      "train_tokens_per_second": 29247.722
    },
    {
      "epoch": 1.549730543755404,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1556731020258664e-05,
      "loss": 0.39227685928344724,
      "num_input_tokens_seen": 12455164288,
      "step": 43800,
      "train_runtime": 425851.5039,
      "train_tokens_per_second": 29247.67
    },
    {
      "epoch": 1.5500843640191329,
      "grad_norm": 0.609375,
      "learning_rate": 2.1554727848609627e-05,
      "loss": 0.3890377998352051,
      "num_input_tokens_seen": 12458012544,
      "step": 43810,
      "train_runtime": 425947.3468,
      "train_tokens_per_second": 29247.776
    },
    {
      "epoch": 1.5504381842828616,
      "grad_norm": 0.61328125,
      "learning_rate": 2.155272523529446e-05,
      "loss": 0.3918537378311157,
      "num_input_tokens_seen": 12460797376,
      "step": 43820,
      "train_runtime": 426043.9057,
      "train_tokens_per_second": 29247.684
    },
    {
      "epoch": 1.5507920045465904,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1550723180053844e-05,
      "loss": 0.3858006954193115,
      "num_input_tokens_seen": 12463622784,
      "step": 43830,
      "train_runtime": 426141.5019,
      "train_tokens_per_second": 29247.615
    },
    {
      "epoch": 1.5511458248103192,
      "grad_norm": 0.609375,
      "learning_rate": 2.154872168262863e-05,
      "loss": 0.3911353588104248,
      "num_input_tokens_seen": 12466459840,
      "step": 43840,
      "train_runtime": 426238.7432,
      "train_tokens_per_second": 29247.599
    },
    {
      "epoch": 1.551499645074048,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1546720742759827e-05,
      "loss": 0.3917374610900879,
      "num_input_tokens_seen": 12469283840,
      "step": 43850,
      "train_runtime": 426337.9126,
      "train_tokens_per_second": 29247.42
    },
    {
      "epoch": 1.5518534653377767,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1544720360188627e-05,
      "loss": 0.3883843421936035,
      "num_input_tokens_seen": 12472106944,
      "step": 43860,
      "train_runtime": 426434.3304,
      "train_tokens_per_second": 29247.427
    },
    {
      "epoch": 1.5522072856015057,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1542720534656376e-05,
      "loss": 0.39139976501464846,
      "num_input_tokens_seen": 12474971072,
      "step": 43870,
      "train_runtime": 426534.5777,
      "train_tokens_per_second": 29247.268
    },
    {
      "epoch": 1.5525611058652342,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1540721265904594e-05,
      "loss": 0.39008517265319825,
      "num_input_tokens_seen": 12477812096,
      "step": 43880,
      "train_runtime": 426630.7019,
      "train_tokens_per_second": 29247.337
    },
    {
      "epoch": 1.5529149261289632,
      "grad_norm": 0.61328125,
      "learning_rate": 2.153872255367498e-05,
      "loss": 0.3899892807006836,
      "num_input_tokens_seen": 12480686400,
      "step": 43890,
      "train_runtime": 426729.1607,
      "train_tokens_per_second": 29247.325
    },
    {
      "epoch": 1.5532687463926917,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1536724397709374e-05,
      "loss": 0.3917407512664795,
      "num_input_tokens_seen": 12483482560,
      "step": 43900,
      "train_runtime": 426823.1444,
      "train_tokens_per_second": 29247.436
    },
    {
      "epoch": 1.5536225666564207,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1534726797749813e-05,
      "loss": 0.39223179817199705,
      "num_input_tokens_seen": 12486327744,
      "step": 43910,
      "train_runtime": 426918.045,
      "train_tokens_per_second": 29247.599
    },
    {
      "epoch": 1.5539763869201493,
      "grad_norm": 0.60546875,
      "learning_rate": 2.153272975353848e-05,
      "loss": 0.3840668201446533,
      "num_input_tokens_seen": 12489117760,
      "step": 43920,
      "train_runtime": 427013.3095,
      "train_tokens_per_second": 29247.608
    },
    {
      "epoch": 1.5543302071838783,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1530733264817737e-05,
      "loss": 0.3913280010223389,
      "num_input_tokens_seen": 12491962432,
      "step": 43930,
      "train_runtime": 427110.43,
      "train_tokens_per_second": 29247.617
    },
    {
      "epoch": 1.5546840274476068,
      "grad_norm": 0.58203125,
      "learning_rate": 2.152873733133011e-05,
      "loss": 0.3897716522216797,
      "num_input_tokens_seen": 12494818688,
      "step": 43940,
      "train_runtime": 427209.6269,
      "train_tokens_per_second": 29247.512
    },
    {
      "epoch": 1.5550378477113358,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1526741952818296e-05,
      "loss": 0.39046201705932615,
      "num_input_tokens_seen": 12497623616,
      "step": 43950,
      "train_runtime": 427305.5774,
      "train_tokens_per_second": 29247.509
    },
    {
      "epoch": 1.5553916679750646,
      "grad_norm": 0.609375,
      "learning_rate": 2.152474712902514e-05,
      "loss": 0.38759191036224366,
      "num_input_tokens_seen": 12500495488,
      "step": 43960,
      "train_runtime": 427403.6832,
      "train_tokens_per_second": 29247.515
    },
    {
      "epoch": 1.5557454882387933,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1522752859693686e-05,
      "loss": 0.39222135543823244,
      "num_input_tokens_seen": 12503396544,
      "step": 43970,
      "train_runtime": 427501.8889,
      "train_tokens_per_second": 29247.582
    },
    {
      "epoch": 1.556099308502522,
      "grad_norm": 0.578125,
      "learning_rate": 2.152075914456711e-05,
      "loss": 0.38758115768432616,
      "num_input_tokens_seen": 12506165056,
      "step": 43980,
      "train_runtime": 427595.4729,
      "train_tokens_per_second": 29247.655
    },
    {
      "epoch": 1.5564531287662509,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1518765983388788e-05,
      "loss": 0.39447574615478515,
      "num_input_tokens_seen": 12508950272,
      "step": 43990,
      "train_runtime": 427689.8732,
      "train_tokens_per_second": 29247.712
    },
    {
      "epoch": 1.5568069490299796,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1516773375902238e-05,
      "loss": 0.3862939357757568,
      "num_input_tokens_seen": 12511763648,
      "step": 44000,
      "train_runtime": 427785.0979,
      "train_tokens_per_second": 29247.778
    },
    {
      "epoch": 1.5571607692937084,
      "grad_norm": 0.66796875,
      "learning_rate": 2.1514781321851152e-05,
      "loss": 0.387599515914917,
      "num_input_tokens_seen": 12514606592,
      "step": 44010,
      "train_runtime": 427879.6165,
      "train_tokens_per_second": 29247.962
    },
    {
      "epoch": 1.5575145895574372,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1512789820979384e-05,
      "loss": 0.3897282600402832,
      "num_input_tokens_seen": 12517426304,
      "step": 44020,
      "train_runtime": 427974.4037,
      "train_tokens_per_second": 29248.072
    },
    {
      "epoch": 1.557868409821166,
      "grad_norm": 0.5703125,
      "learning_rate": 2.1510798873030965e-05,
      "loss": 0.3901822090148926,
      "num_input_tokens_seen": 12520287168,
      "step": 44030,
      "train_runtime": 428071.0461,
      "train_tokens_per_second": 29248.152
    },
    {
      "epoch": 1.558222230084895,
      "grad_norm": 0.609375,
      "learning_rate": 2.1508808477750084e-05,
      "loss": 0.3936939716339111,
      "num_input_tokens_seen": 12523135232,
      "step": 44040,
      "train_runtime": 428166.5373,
      "train_tokens_per_second": 29248.281
    },
    {
      "epoch": 1.5585760503486235,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1506818634881096e-05,
      "loss": 0.384035062789917,
      "num_input_tokens_seen": 12525946816,
      "step": 44050,
      "train_runtime": 428260.64,
      "train_tokens_per_second": 29248.419
    },
    {
      "epoch": 1.5589298706123524,
      "grad_norm": 0.609375,
      "learning_rate": 2.1504829344168518e-05,
      "loss": 0.38800787925720215,
      "num_input_tokens_seen": 12528793280,
      "step": 44060,
      "train_runtime": 428359.5153,
      "train_tokens_per_second": 29248.313
    },
    {
      "epoch": 1.559283690876081,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1502840605357042e-05,
      "loss": 0.3905655384063721,
      "num_input_tokens_seen": 12531591424,
      "step": 44070,
      "train_runtime": 428455.0368,
      "train_tokens_per_second": 29248.323
    },
    {
      "epoch": 1.55963751113981,
      "grad_norm": 0.609375,
      "learning_rate": 2.1500852418191518e-05,
      "loss": 0.3879871845245361,
      "num_input_tokens_seen": 12534477184,
      "step": 44080,
      "train_runtime": 428557.426,
      "train_tokens_per_second": 29248.069
    },
    {
      "epoch": 1.5599913314035385,
      "grad_norm": 0.59375,
      "learning_rate": 2.1498864782416958e-05,
      "loss": 0.389221715927124,
      "num_input_tokens_seen": 12537293888,
      "step": 44090,
      "train_runtime": 428651.5733,
      "train_tokens_per_second": 29248.216
    },
    {
      "epoch": 1.5603451516672675,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1496877697778547e-05,
      "loss": 0.38833768367767335,
      "num_input_tokens_seen": 12540124032,
      "step": 44100,
      "train_runtime": 428746.2996,
      "train_tokens_per_second": 29248.355
    },
    {
      "epoch": 1.560698971930996,
      "grad_norm": 0.5859375,
      "learning_rate": 2.1494891164021637e-05,
      "loss": 0.38802669048309324,
      "num_input_tokens_seen": 12542967296,
      "step": 44110,
      "train_runtime": 428843.677,
      "train_tokens_per_second": 29248.344
    },
    {
      "epoch": 1.561052792194725,
      "grad_norm": 0.61328125,
      "learning_rate": 2.149290518089173e-05,
      "loss": 0.3914804220199585,
      "num_input_tokens_seen": 12545837312,
      "step": 44120,
      "train_runtime": 428942.6341,
      "train_tokens_per_second": 29248.287
    },
    {
      "epoch": 1.5614066124584538,
      "grad_norm": 0.6328125,
      "learning_rate": 2.1490919748134512e-05,
      "loss": 0.38834688663482664,
      "num_input_tokens_seen": 12548620288,
      "step": 44130,
      "train_runtime": 429037.5439,
      "train_tokens_per_second": 29248.303
    },
    {
      "epoch": 1.5617604327221826,
      "grad_norm": 0.609375,
      "learning_rate": 2.1488934865495812e-05,
      "loss": 0.3912999629974365,
      "num_input_tokens_seen": 12551454720,
      "step": 44140,
      "train_runtime": 429131.673,
      "train_tokens_per_second": 29248.493
    },
    {
      "epoch": 1.5621142529859113,
      "grad_norm": 0.609375,
      "learning_rate": 2.1486950532721642e-05,
      "loss": 0.38909385204315183,
      "num_input_tokens_seen": 12554275392,
      "step": 44150,
      "train_runtime": 429227.0067,
      "train_tokens_per_second": 29248.568
    },
    {
      "epoch": 1.56246807324964,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1484966749558173e-05,
      "loss": 0.39074001312255857,
      "num_input_tokens_seen": 12557107904,
      "step": 44160,
      "train_runtime": 429321.2483,
      "train_tokens_per_second": 29248.745
    },
    {
      "epoch": 1.5628218935133689,
      "grad_norm": 0.59375,
      "learning_rate": 2.1482983515751732e-05,
      "loss": 0.3922168731689453,
      "num_input_tokens_seen": 12559961920,
      "step": 44170,
      "train_runtime": 429417.3568,
      "train_tokens_per_second": 29248.845
    },
    {
      "epoch": 1.5631757137770976,
      "grad_norm": 0.625,
      "learning_rate": 2.1481000831048812e-05,
      "loss": 0.39490609169006347,
      "num_input_tokens_seen": 12562849856,
      "step": 44180,
      "train_runtime": 429512.4419,
      "train_tokens_per_second": 29249.094
    },
    {
      "epoch": 1.5635295340408264,
      "grad_norm": 0.62109375,
      "learning_rate": 2.147901869519609e-05,
      "loss": 0.38771209716796873,
      "num_input_tokens_seen": 12565730240,
      "step": 44190,
      "train_runtime": 429613.2651,
      "train_tokens_per_second": 29248.934
    },
    {
      "epoch": 1.5638833543045552,
      "grad_norm": 0.578125,
      "learning_rate": 2.147703710794037e-05,
      "loss": 0.3866820096969604,
      "num_input_tokens_seen": 12568560064,
      "step": 44200,
      "train_runtime": 429707.3922,
      "train_tokens_per_second": 29249.113
    },
    {
      "epoch": 1.5642371745682841,
      "grad_norm": 0.59765625,
      "learning_rate": 2.147505606902865e-05,
      "loss": 0.3912502288818359,
      "num_input_tokens_seen": 12571387584,
      "step": 44210,
      "train_runtime": 429802.7927,
      "train_tokens_per_second": 29249.199
    },
    {
      "epoch": 1.5645909948320127,
      "grad_norm": 0.58984375,
      "learning_rate": 2.147307557820808e-05,
      "loss": 0.38916351795196535,
      "num_input_tokens_seen": 12574206720,
      "step": 44220,
      "train_runtime": 429899.3628,
      "train_tokens_per_second": 29249.187
    },
    {
      "epoch": 1.5649448150957417,
      "grad_norm": 0.5859375,
      "learning_rate": 2.147109563522597e-05,
      "loss": 0.3939924716949463,
      "num_input_tokens_seen": 12577028288,
      "step": 44230,
      "train_runtime": 429992.1132,
      "train_tokens_per_second": 29249.439
    },
    {
      "epoch": 1.5652986353594702,
      "grad_norm": 0.609375,
      "learning_rate": 2.1469116239829804e-05,
      "loss": 0.3916449069976807,
      "num_input_tokens_seen": 12579834816,
      "step": 44240,
      "train_runtime": 430085.2853,
      "train_tokens_per_second": 29249.628
    },
    {
      "epoch": 1.5656524556231992,
      "grad_norm": 0.625,
      "learning_rate": 2.1467137391767213e-05,
      "loss": 0.39118366241455077,
      "num_input_tokens_seen": 12582689984,
      "step": 44250,
      "train_runtime": 430182.2923,
      "train_tokens_per_second": 29249.67
    },
    {
      "epoch": 1.5660062758869278,
      "grad_norm": 0.59375,
      "learning_rate": 2.146515909078601e-05,
      "loss": 0.3892606019973755,
      "num_input_tokens_seen": 12585563968,
      "step": 44260,
      "train_runtime": 430283.4127,
      "train_tokens_per_second": 29249.475
    },
    {
      "epoch": 1.5663600961506567,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1463181336634156e-05,
      "loss": 0.39255824089050295,
      "num_input_tokens_seen": 12588414528,
      "step": 44270,
      "train_runtime": 430380.6113,
      "train_tokens_per_second": 29249.493
    },
    {
      "epoch": 1.5667139164143853,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1461204129059776e-05,
      "loss": 0.39202563762664794,
      "num_input_tokens_seen": 12591275648,
      "step": 44280,
      "train_runtime": 430478.8885,
      "train_tokens_per_second": 29249.461
    },
    {
      "epoch": 1.5670677366781143,
      "grad_norm": 0.59375,
      "learning_rate": 2.145922746781116e-05,
      "loss": 0.38499107360839846,
      "num_input_tokens_seen": 12594127680,
      "step": 44290,
      "train_runtime": 430574.1974,
      "train_tokens_per_second": 29249.611
    },
    {
      "epoch": 1.567421556941843,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1457251352636765e-05,
      "loss": 0.38577277660369874,
      "num_input_tokens_seen": 12596940032,
      "step": 44300,
      "train_runtime": 430666.7557,
      "train_tokens_per_second": 29249.855
    },
    {
      "epoch": 1.5677753772055718,
      "grad_norm": 0.609375,
      "learning_rate": 2.1455275783285203e-05,
      "loss": 0.3907612800598145,
      "num_input_tokens_seen": 12599772992,
      "step": 44310,
      "train_runtime": 430764.1901,
      "train_tokens_per_second": 29249.815
    },
    {
      "epoch": 1.5681291974693006,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1453300759505255e-05,
      "loss": 0.38887863159179686,
      "num_input_tokens_seen": 12602666496,
      "step": 44320,
      "train_runtime": 430862.7641,
      "train_tokens_per_second": 29249.839
    },
    {
      "epoch": 1.5684830177330293,
      "grad_norm": 0.6171875,
      "learning_rate": 2.145132628104585e-05,
      "loss": 0.39121160507202146,
      "num_input_tokens_seen": 12605518912,
      "step": 44330,
      "train_runtime": 430960.3884,
      "train_tokens_per_second": 29249.832
    },
    {
      "epoch": 1.568836837996758,
      "grad_norm": 0.64453125,
      "learning_rate": 2.1449352347656098e-05,
      "loss": 0.3926108360290527,
      "num_input_tokens_seen": 12608373696,
      "step": 44340,
      "train_runtime": 431059.8077,
      "train_tokens_per_second": 29249.708
    },
    {
      "epoch": 1.5691906582604869,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1447378959085256e-05,
      "loss": 0.3865676403045654,
      "num_input_tokens_seen": 12611231296,
      "step": 44350,
      "train_runtime": 431156.9887,
      "train_tokens_per_second": 29249.743
    },
    {
      "epoch": 1.5695444785242156,
      "grad_norm": 0.625,
      "learning_rate": 2.1445406115082747e-05,
      "loss": 0.38885016441345216,
      "num_input_tokens_seen": 12614064960,
      "step": 44360,
      "train_runtime": 431254.7606,
      "train_tokens_per_second": 29249.683
    },
    {
      "epoch": 1.5698982987879444,
      "grad_norm": 0.59375,
      "learning_rate": 2.1443433815398163e-05,
      "loss": 0.3882179260253906,
      "num_input_tokens_seen": 12616857728,
      "step": 44370,
      "train_runtime": 431350.3679,
      "train_tokens_per_second": 29249.674
    },
    {
      "epoch": 1.5702521190516734,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1441462059781238e-05,
      "loss": 0.3887664794921875,
      "num_input_tokens_seen": 12619720000,
      "step": 44380,
      "train_runtime": 431448.5504,
      "train_tokens_per_second": 29249.652
    },
    {
      "epoch": 1.570605939315402,
      "grad_norm": 0.65625,
      "learning_rate": 2.1439490847981885e-05,
      "loss": 0.3904621124267578,
      "num_input_tokens_seen": 12622584512,
      "step": 44390,
      "train_runtime": 431547.9387,
      "train_tokens_per_second": 29249.553
    },
    {
      "epoch": 1.570959759579131,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1437520179750174e-05,
      "loss": 0.3902522087097168,
      "num_input_tokens_seen": 12625504576,
      "step": 44400,
      "train_runtime": 431650.6297,
      "train_tokens_per_second": 29249.36
    },
    {
      "epoch": 1.5713135798428595,
      "grad_norm": 0.609375,
      "learning_rate": 2.1435550054836332e-05,
      "loss": 0.38826086521148684,
      "num_input_tokens_seen": 12628376768,
      "step": 44410,
      "train_runtime": 431751.2738,
      "train_tokens_per_second": 29249.194
    },
    {
      "epoch": 1.5716674001065885,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1433580472990743e-05,
      "loss": 0.38236465454101565,
      "num_input_tokens_seen": 12631232192,
      "step": 44420,
      "train_runtime": 431848.5327,
      "train_tokens_per_second": 29249.219
    },
    {
      "epoch": 1.572021220370317,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1431611433963964e-05,
      "loss": 0.3914022922515869,
      "num_input_tokens_seen": 12634029568,
      "step": 44430,
      "train_runtime": 431944.1389,
      "train_tokens_per_second": 29249.221
    },
    {
      "epoch": 1.572375040634046,
      "grad_norm": 0.58984375,
      "learning_rate": 2.14296429375067e-05,
      "loss": 0.3884912014007568,
      "num_input_tokens_seen": 12636909120,
      "step": 44440,
      "train_runtime": 432044.6001,
      "train_tokens_per_second": 29249.085
    },
    {
      "epoch": 1.5727288608977745,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1427674983369826e-05,
      "loss": 0.393666934967041,
      "num_input_tokens_seen": 12639779136,
      "step": 44450,
      "train_runtime": 432145.4449,
      "train_tokens_per_second": 29248.901
    },
    {
      "epoch": 1.5730826811615035,
      "grad_norm": 0.609375,
      "learning_rate": 2.1425707571304366e-05,
      "loss": 0.3901156663894653,
      "num_input_tokens_seen": 12642660672,
      "step": 44460,
      "train_runtime": 432246.2274,
      "train_tokens_per_second": 29248.747
    },
    {
      "epoch": 1.5734365014252323,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1423740701061512e-05,
      "loss": 0.3896439552307129,
      "num_input_tokens_seen": 12645494208,
      "step": 44470,
      "train_runtime": 432344.2171,
      "train_tokens_per_second": 29248.672
    },
    {
      "epoch": 1.573790321688961,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1421774372392622e-05,
      "loss": 0.3929807424545288,
      "num_input_tokens_seen": 12648281280,
      "step": 44480,
      "train_runtime": 432436.8241,
      "train_tokens_per_second": 29248.853
    },
    {
      "epoch": 1.5741441419526898,
      "grad_norm": 0.59375,
      "learning_rate": 2.1419808585049194e-05,
      "loss": 0.39260187149047854,
      "num_input_tokens_seen": 12651189952,
      "step": 44490,
      "train_runtime": 432536.3823,
      "train_tokens_per_second": 29248.846
    },
    {
      "epoch": 1.5744979622164186,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1417843338782907e-05,
      "loss": 0.387771463394165,
      "num_input_tokens_seen": 12654037952,
      "step": 44500,
      "train_runtime": 432633.9228,
      "train_tokens_per_second": 29248.834
    },
    {
      "epoch": 1.5748517824801473,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1415878633345592e-05,
      "loss": 0.391402530670166,
      "num_input_tokens_seen": 12656910592,
      "step": 44510,
      "train_runtime": 432734.6176,
      "train_tokens_per_second": 29248.667
    },
    {
      "epoch": 1.575205602743876,
      "grad_norm": 0.5859375,
      "learning_rate": 2.1413914468489227e-05,
      "loss": 0.38649921417236327,
      "num_input_tokens_seen": 12659733376,
      "step": 44520,
      "train_runtime": 432832.1401,
      "train_tokens_per_second": 29248.598
    },
    {
      "epoch": 1.5755594230076049,
      "grad_norm": 0.578125,
      "learning_rate": 2.141195084396597e-05,
      "loss": 0.38875837326049806,
      "num_input_tokens_seen": 12662600768,
      "step": 44530,
      "train_runtime": 432928.0458,
      "train_tokens_per_second": 29248.742
    },
    {
      "epoch": 1.5759132432713336,
      "grad_norm": 0.609375,
      "learning_rate": 2.140998775952812e-05,
      "loss": 0.3887928009033203,
      "num_input_tokens_seen": 12665407296,
      "step": 44540,
      "train_runtime": 433024.383,
      "train_tokens_per_second": 29248.716
    },
    {
      "epoch": 1.5762670635350626,
      "grad_norm": 0.59765625,
      "learning_rate": 2.140802521492814e-05,
      "loss": 0.38804116249084475,
      "num_input_tokens_seen": 12668250752,
      "step": 44550,
      "train_runtime": 433120.6943,
      "train_tokens_per_second": 29248.777
    },
    {
      "epoch": 1.5766208837987912,
      "grad_norm": 0.60546875,
      "learning_rate": 2.140606320991867e-05,
      "loss": 0.39308843612670896,
      "num_input_tokens_seen": 12671056320,
      "step": 44560,
      "train_runtime": 433218.6944,
      "train_tokens_per_second": 29248.637
    },
    {
      "epoch": 1.5769747040625202,
      "grad_norm": 0.61328125,
      "learning_rate": 2.140410174425248e-05,
      "loss": 0.3906198024749756,
      "num_input_tokens_seen": 12673875200,
      "step": 44570,
      "train_runtime": 433314.9591,
      "train_tokens_per_second": 29248.645
    },
    {
      "epoch": 1.5773285243262487,
      "grad_norm": 0.6328125,
      "learning_rate": 2.140214081768251e-05,
      "loss": 0.38810865879058837,
      "num_input_tokens_seen": 12676647872,
      "step": 44580,
      "train_runtime": 433410.6798,
      "train_tokens_per_second": 29248.582
    },
    {
      "epoch": 1.5776823445899777,
      "grad_norm": 0.60546875,
      "learning_rate": 2.140018042996187e-05,
      "loss": 0.3846935987472534,
      "num_input_tokens_seen": 12679409024,
      "step": 44590,
      "train_runtime": 433502.3061,
      "train_tokens_per_second": 29248.769
    },
    {
      "epoch": 1.5780361648537062,
      "grad_norm": 0.609375,
      "learning_rate": 2.1398220580843807e-05,
      "loss": 0.3881331443786621,
      "num_input_tokens_seen": 12682261504,
      "step": 44600,
      "train_runtime": 433599.7376,
      "train_tokens_per_second": 29248.776
    },
    {
      "epoch": 1.5783899851174352,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1396261270081748e-05,
      "loss": 0.3894944190979004,
      "num_input_tokens_seen": 12685153856,
      "step": 44610,
      "train_runtime": 433700.4043,
      "train_tokens_per_second": 29248.656
    },
    {
      "epoch": 1.578743805381164,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1394302497429258e-05,
      "loss": 0.389498233795166,
      "num_input_tokens_seen": 12688015296,
      "step": 44620,
      "train_runtime": 433797.4886,
      "train_tokens_per_second": 29248.706
    },
    {
      "epoch": 1.5790976256448928,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1392344262640077e-05,
      "loss": 0.38886559009552,
      "num_input_tokens_seen": 12690933824,
      "step": 44630,
      "train_runtime": 433897.5332,
      "train_tokens_per_second": 29248.689
    },
    {
      "epoch": 1.5794514459086215,
      "grad_norm": 0.5859375,
      "learning_rate": 2.139038656546809e-05,
      "loss": 0.38655881881713866,
      "num_input_tokens_seen": 12693750144,
      "step": 44640,
      "train_runtime": 433990.4033,
      "train_tokens_per_second": 29248.919
    },
    {
      "epoch": 1.5798052661723503,
      "grad_norm": 0.61328125,
      "learning_rate": 2.138842940566735e-05,
      "loss": 0.39056284427642823,
      "num_input_tokens_seen": 12696573056,
      "step": 44650,
      "train_runtime": 434085.0021,
      "train_tokens_per_second": 29249.048
    },
    {
      "epoch": 1.580159086436079,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1386472782992055e-05,
      "loss": 0.38646736145019533,
      "num_input_tokens_seen": 12699389568,
      "step": 44660,
      "train_runtime": 434180.2714,
      "train_tokens_per_second": 29249.117
    },
    {
      "epoch": 1.5805129066998078,
      "grad_norm": 0.6171875,
      "learning_rate": 2.138451669719658e-05,
      "loss": 0.39318141937255857,
      "num_input_tokens_seen": 12702281792,
      "step": 44670,
      "train_runtime": 434281.4869,
      "train_tokens_per_second": 29248.96
    },
    {
      "epoch": 1.5808667269635366,
      "grad_norm": 0.59375,
      "learning_rate": 2.138256114803543e-05,
      "loss": 0.38870036602020264,
      "num_input_tokens_seen": 12705152832,
      "step": 44680,
      "train_runtime": 434378.9183,
      "train_tokens_per_second": 29249.009
    },
    {
      "epoch": 1.5812205472272653,
      "grad_norm": 0.59375,
      "learning_rate": 2.138060613526329e-05,
      "loss": 0.3939640045166016,
      "num_input_tokens_seen": 12707983872,
      "step": 44690,
      "train_runtime": 434476.3898,
      "train_tokens_per_second": 29248.963
    },
    {
      "epoch": 1.5815743674909941,
      "grad_norm": 0.66015625,
      "learning_rate": 2.1378651658634993e-05,
      "loss": 0.3891609191894531,
      "num_input_tokens_seen": 12710842048,
      "step": 44700,
      "train_runtime": 434574.9814,
      "train_tokens_per_second": 29248.904
    },
    {
      "epoch": 1.5819281877547229,
      "grad_norm": 0.60546875,
      "learning_rate": 2.137669771790553e-05,
      "loss": 0.38781180381774905,
      "num_input_tokens_seen": 12713751936,
      "step": 44710,
      "train_runtime": 434675.7516,
      "train_tokens_per_second": 29248.818
    },
    {
      "epoch": 1.5822820080184519,
      "grad_norm": 0.6328125,
      "learning_rate": 2.1374744312830048e-05,
      "loss": 0.38753108978271483,
      "num_input_tokens_seen": 12716627776,
      "step": 44720,
      "train_runtime": 434775.6943,
      "train_tokens_per_second": 29248.709
    },
    {
      "epoch": 1.5826358282821804,
      "grad_norm": 0.63671875,
      "learning_rate": 2.137279144316385e-05,
      "loss": 0.39143168926239014,
      "num_input_tokens_seen": 12719462592,
      "step": 44730,
      "train_runtime": 434868.5703,
      "train_tokens_per_second": 29248.981
    },
    {
      "epoch": 1.5829896485459094,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1370839108662393e-05,
      "loss": 0.3870562553405762,
      "num_input_tokens_seen": 12722289728,
      "step": 44740,
      "train_runtime": 434964.6183,
      "train_tokens_per_second": 29249.022
    },
    {
      "epoch": 1.583343468809638,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1368887309081305e-05,
      "loss": 0.3888025999069214,
      "num_input_tokens_seen": 12725142528,
      "step": 44750,
      "train_runtime": 435062.3952,
      "train_tokens_per_second": 29249.006
    },
    {
      "epoch": 1.583697289073367,
      "grad_norm": 0.6171875,
      "learning_rate": 2.136693604417635e-05,
      "loss": 0.38753998279571533,
      "num_input_tokens_seen": 12727975808,
      "step": 44760,
      "train_runtime": 435159.9033,
      "train_tokens_per_second": 29248.963
    },
    {
      "epoch": 1.5840511093370955,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1364985313703465e-05,
      "loss": 0.3935102462768555,
      "num_input_tokens_seen": 12730808960,
      "step": 44770,
      "train_runtime": 435255.8914,
      "train_tokens_per_second": 29249.022
    },
    {
      "epoch": 1.5844049296008245,
      "grad_norm": 0.59375,
      "learning_rate": 2.1363035117418724e-05,
      "loss": 0.3910824298858643,
      "num_input_tokens_seen": 12733634816,
      "step": 44780,
      "train_runtime": 435351.7661,
      "train_tokens_per_second": 29249.071
    },
    {
      "epoch": 1.5847587498645532,
      "grad_norm": 0.625,
      "learning_rate": 2.1361085455078375e-05,
      "loss": 0.3940113544464111,
      "num_input_tokens_seen": 12736443904,
      "step": 44790,
      "train_runtime": 435448.0749,
      "train_tokens_per_second": 29249.053
    },
    {
      "epoch": 1.585112570128282,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1359136326438812e-05,
      "loss": 0.39459786415100095,
      "num_input_tokens_seen": 12739237568,
      "step": 44800,
      "train_runtime": 435541.8822,
      "train_tokens_per_second": 29249.168
    },
    {
      "epoch": 1.5854663903920108,
      "grad_norm": 0.59375,
      "learning_rate": 2.1357187731256588e-05,
      "loss": 0.384906005859375,
      "num_input_tokens_seen": 12742100800,
      "step": 44810,
      "train_runtime": 435640.6836,
      "train_tokens_per_second": 29249.107
    },
    {
      "epoch": 1.5858202106557395,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1355239669288418e-05,
      "loss": 0.3929803133010864,
      "num_input_tokens_seen": 12745032448,
      "step": 44820,
      "train_runtime": 435740.5623,
      "train_tokens_per_second": 29249.13
    },
    {
      "epoch": 1.5861740309194683,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1353292140291152e-05,
      "loss": 0.39290447235107423,
      "num_input_tokens_seen": 12747848768,
      "step": 44830,
      "train_runtime": 435838.1182,
      "train_tokens_per_second": 29249.045
    },
    {
      "epoch": 1.586527851183197,
      "grad_norm": 0.5859375,
      "learning_rate": 2.1351345144021818e-05,
      "loss": 0.3892096519470215,
      "num_input_tokens_seen": 12750732672,
      "step": 44840,
      "train_runtime": 435940.0591,
      "train_tokens_per_second": 29248.821
    },
    {
      "epoch": 1.5868816714469258,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1349398680237585e-05,
      "loss": 0.3912916660308838,
      "num_input_tokens_seen": 12753564352,
      "step": 44850,
      "train_runtime": 436036.6689,
      "train_tokens_per_second": 29248.834
    },
    {
      "epoch": 1.5872354917106546,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1347452748695784e-05,
      "loss": 0.3866436243057251,
      "num_input_tokens_seen": 12756367616,
      "step": 44860,
      "train_runtime": 436131.8071,
      "train_tokens_per_second": 29248.882
    },
    {
      "epoch": 1.5875893119743836,
      "grad_norm": 0.609375,
      "learning_rate": 2.1345507349153896e-05,
      "loss": 0.38961215019226075,
      "num_input_tokens_seen": 12759225088,
      "step": 44870,
      "train_runtime": 436226.7441,
      "train_tokens_per_second": 29249.067
    },
    {
      "epoch": 1.5879431322381121,
      "grad_norm": 0.5859375,
      "learning_rate": 2.1343562481369555e-05,
      "loss": 0.38921217918395995,
      "num_input_tokens_seen": 12762026688,
      "step": 44880,
      "train_runtime": 436321.0909,
      "train_tokens_per_second": 29249.163
    },
    {
      "epoch": 1.588296952501841,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1341618145100557e-05,
      "loss": 0.3892656326293945,
      "num_input_tokens_seen": 12764949504,
      "step": 44890,
      "train_runtime": 436422.795,
      "train_tokens_per_second": 29249.044
    },
    {
      "epoch": 1.5886507727655697,
      "grad_norm": 0.65234375,
      "learning_rate": 2.133967434010485e-05,
      "loss": 0.3880063533782959,
      "num_input_tokens_seen": 12767798144,
      "step": 44900,
      "train_runtime": 436522.1908,
      "train_tokens_per_second": 29248.91
    },
    {
      "epoch": 1.5890045930292986,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1337731066140536e-05,
      "loss": 0.38448147773742675,
      "num_input_tokens_seen": 12770677440,
      "step": 44910,
      "train_runtime": 436618.5099,
      "train_tokens_per_second": 29249.052
    },
    {
      "epoch": 1.5893584132930272,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1335788322965864e-05,
      "loss": 0.39184203147888186,
      "num_input_tokens_seen": 12773505792,
      "step": 44920,
      "train_runtime": 436710.9335,
      "train_tokens_per_second": 29249.338
    },
    {
      "epoch": 1.5897122335567562,
      "grad_norm": 0.5859375,
      "learning_rate": 2.1333846110339246e-05,
      "loss": 0.384472131729126,
      "num_input_tokens_seen": 12776366080,
      "step": 44930,
      "train_runtime": 436807.0916,
      "train_tokens_per_second": 29249.447
    },
    {
      "epoch": 1.5900660538204847,
      "grad_norm": 0.6015625,
      "learning_rate": 2.133190442801925e-05,
      "loss": 0.39309532642364503,
      "num_input_tokens_seen": 12779222592,
      "step": 44940,
      "train_runtime": 436906.3582,
      "train_tokens_per_second": 29249.34
    },
    {
      "epoch": 1.5904198740842137,
      "grad_norm": 0.5859375,
      "learning_rate": 2.1329963275764586e-05,
      "loss": 0.3858497142791748,
      "num_input_tokens_seen": 12782075776,
      "step": 44950,
      "train_runtime": 437002.5757,
      "train_tokens_per_second": 29249.429
    },
    {
      "epoch": 1.5907736943479425,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1328022653334127e-05,
      "loss": 0.39380998611450196,
      "num_input_tokens_seen": 12784884160,
      "step": 44960,
      "train_runtime": 437094.4987,
      "train_tokens_per_second": 29249.703
    },
    {
      "epoch": 1.5911275146116712,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1326082560486895e-05,
      "loss": 0.38759121894836424,
      "num_input_tokens_seen": 12787754624,
      "step": 44970,
      "train_runtime": 437189.1406,
      "train_tokens_per_second": 29249.937
    },
    {
      "epoch": 1.5914813348754,
      "grad_norm": 0.625,
      "learning_rate": 2.132414299698207e-05,
      "loss": 0.38825697898864747,
      "num_input_tokens_seen": 12790593216,
      "step": 44980,
      "train_runtime": 437284.6757,
      "train_tokens_per_second": 29250.038
    },
    {
      "epoch": 1.5918351551391288,
      "grad_norm": 0.59375,
      "learning_rate": 2.1322203962578984e-05,
      "loss": 0.38610210418701174,
      "num_input_tokens_seen": 12793465920,
      "step": 44990,
      "train_runtime": 437382.2998,
      "train_tokens_per_second": 29250.077
    },
    {
      "epoch": 1.5921889754028575,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1320265457037115e-05,
      "loss": 0.39192681312561034,
      "num_input_tokens_seen": 12796284416,
      "step": 45000,
      "train_runtime": 437480.7379,
      "train_tokens_per_second": 29249.938
    },
    {
      "epoch": 1.5921889754028575,
      "eval_loss": 0.33652210235595703,
      "eval_runtime": 8.4198,
      "eval_samples_per_second": 473.647,
      "eval_steps_per_second": 3.801,
      "num_input_tokens_seen": 12796284416,
      "step": 45000
    },
    {
      "epoch": 1.5925427956665863,
      "grad_norm": 0.609375,
      "learning_rate": 2.1318327480116112e-05,
      "loss": 0.3895860195159912,
      "num_input_tokens_seen": 12799080832,
      "step": 45010,
      "train_runtime": 437604.1809,
      "train_tokens_per_second": 29248.077
    },
    {
      "epoch": 1.592896615930315,
      "grad_norm": 0.60546875,
      "learning_rate": 2.131639003157575e-05,
      "loss": 0.3914334297180176,
      "num_input_tokens_seen": 12801917888,
      "step": 45020,
      "train_runtime": 437702.9683,
      "train_tokens_per_second": 29247.958
    },
    {
      "epoch": 1.5932504361940438,
      "grad_norm": 0.609375,
      "learning_rate": 2.1314453111175983e-05,
      "loss": 0.3910184383392334,
      "num_input_tokens_seen": 12804813248,
      "step": 45030,
      "train_runtime": 437801.331,
      "train_tokens_per_second": 29248.0
    },
    {
      "epoch": 1.5936042564577728,
      "grad_norm": 0.5625,
      "learning_rate": 2.13125167186769e-05,
      "loss": 0.39088263511657717,
      "num_input_tokens_seen": 12807639168,
      "step": 45040,
      "train_runtime": 437895.9329,
      "train_tokens_per_second": 29248.135
    },
    {
      "epoch": 1.5939580767215014,
      "grad_norm": 0.625,
      "learning_rate": 2.131058085383875e-05,
      "loss": 0.39203720092773436,
      "num_input_tokens_seen": 12810487744,
      "step": 45050,
      "train_runtime": 437995.3044,
      "train_tokens_per_second": 29248.002
    },
    {
      "epoch": 1.5943118969852303,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1308645516421935e-05,
      "loss": 0.3851248502731323,
      "num_input_tokens_seen": 12813344448,
      "step": 45060,
      "train_runtime": 438091.3401,
      "train_tokens_per_second": 29248.112
    },
    {
      "epoch": 1.594665717248959,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1306710706187006e-05,
      "loss": 0.38852543830871583,
      "num_input_tokens_seen": 12816212480,
      "step": 45070,
      "train_runtime": 438189.2873,
      "train_tokens_per_second": 29248.119
    },
    {
      "epoch": 1.5950195375126879,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1304776422894667e-05,
      "loss": 0.38821632862091066,
      "num_input_tokens_seen": 12819068672,
      "step": 45080,
      "train_runtime": 438284.7482,
      "train_tokens_per_second": 29248.265
    },
    {
      "epoch": 1.5953733577764164,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1302842666305776e-05,
      "loss": 0.3872946500778198,
      "num_input_tokens_seen": 12821913664,
      "step": 45090,
      "train_runtime": 438379.9603,
      "train_tokens_per_second": 29248.403
    },
    {
      "epoch": 1.5957271780401454,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1300909436181343e-05,
      "loss": 0.39091858863830564,
      "num_input_tokens_seen": 12824738816,
      "step": 45100,
      "train_runtime": 438477.1787,
      "train_tokens_per_second": 29248.361
    },
    {
      "epoch": 1.596080998303874,
      "grad_norm": 0.58984375,
      "learning_rate": 2.129897673228253e-05,
      "loss": 0.3875157833099365,
      "num_input_tokens_seen": 12827594304,
      "step": 45110,
      "train_runtime": 438576.9868,
      "train_tokens_per_second": 29248.216
    },
    {
      "epoch": 1.596434818567603,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1297044554370643e-05,
      "loss": 0.38906288146972656,
      "num_input_tokens_seen": 12830426944,
      "step": 45120,
      "train_runtime": 438673.163,
      "train_tokens_per_second": 29248.26
    },
    {
      "epoch": 1.5967886388313317,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1295112902207147e-05,
      "loss": 0.39132840633392335,
      "num_input_tokens_seen": 12833234624,
      "step": 45130,
      "train_runtime": 438767.2574,
      "train_tokens_per_second": 29248.387
    },
    {
      "epoch": 1.5971424590950605,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1293181775553662e-05,
      "loss": 0.3892688274383545,
      "num_input_tokens_seen": 12836122624,
      "step": 45140,
      "train_runtime": 438867.1395,
      "train_tokens_per_second": 29248.311
    },
    {
      "epoch": 1.5974962793587892,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1291251174171946e-05,
      "loss": 0.3857736587524414,
      "num_input_tokens_seen": 12838963712,
      "step": 45150,
      "train_runtime": 438963.9186,
      "train_tokens_per_second": 29248.335
    },
    {
      "epoch": 1.597850099622518,
      "grad_norm": 0.59765625,
      "learning_rate": 2.128932109782393e-05,
      "loss": 0.38685131072998047,
      "num_input_tokens_seen": 12841800960,
      "step": 45160,
      "train_runtime": 439057.2823,
      "train_tokens_per_second": 29248.578
    },
    {
      "epoch": 1.5982039198862468,
      "grad_norm": 0.6171875,
      "learning_rate": 2.128739154627167e-05,
      "loss": 0.3905055999755859,
      "num_input_tokens_seen": 12844650624,
      "step": 45170,
      "train_runtime": 439155.7038,
      "train_tokens_per_second": 29248.511
    },
    {
      "epoch": 1.5985577401499755,
      "grad_norm": 0.66015625,
      "learning_rate": 2.128546251927739e-05,
      "loss": 0.3858360528945923,
      "num_input_tokens_seen": 12847525312,
      "step": 45180,
      "train_runtime": 439252.4011,
      "train_tokens_per_second": 29248.617
    },
    {
      "epoch": 1.5989115604137043,
      "grad_norm": 0.61328125,
      "learning_rate": 2.128353401660346e-05,
      "loss": 0.39108691215515134,
      "num_input_tokens_seen": 12850308864,
      "step": 45190,
      "train_runtime": 439345.6058,
      "train_tokens_per_second": 29248.748
    },
    {
      "epoch": 1.599265380677433,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1281606038012402e-05,
      "loss": 0.3894501209259033,
      "num_input_tokens_seen": 12853207808,
      "step": 45200,
      "train_runtime": 439446.3421,
      "train_tokens_per_second": 29248.64
    },
    {
      "epoch": 1.599619200941162,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1279678583266886e-05,
      "loss": 0.39052419662475585,
      "num_input_tokens_seen": 12856019520,
      "step": 45210,
      "train_runtime": 439543.9966,
      "train_tokens_per_second": 29248.539
    },
    {
      "epoch": 1.5999730212048906,
      "grad_norm": 0.6171875,
      "learning_rate": 2.127775165212974e-05,
      "loss": 0.39018614292144777,
      "num_input_tokens_seen": 12858788544,
      "step": 45220,
      "train_runtime": 439635.1388,
      "train_tokens_per_second": 29248.773
    },
    {
      "epoch": 1.6003268414686196,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1275825244363922e-05,
      "loss": 0.3904735803604126,
      "num_input_tokens_seen": 12861605056,
      "step": 45230,
      "train_runtime": 439732.5926,
      "train_tokens_per_second": 29248.696
    },
    {
      "epoch": 1.6006806617323481,
      "grad_norm": 0.609375,
      "learning_rate": 2.127389935973257e-05,
      "loss": 0.3859109401702881,
      "num_input_tokens_seen": 12864459136,
      "step": 45240,
      "train_runtime": 439830.607,
      "train_tokens_per_second": 29248.667
    },
    {
      "epoch": 1.6010344819960771,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1271973997998947e-05,
      "loss": 0.38423891067504884,
      "num_input_tokens_seen": 12867307072,
      "step": 45250,
      "train_runtime": 439926.8426,
      "train_tokens_per_second": 29248.743
    },
    {
      "epoch": 1.6013883022598057,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1270049158926483e-05,
      "loss": 0.3884052515029907,
      "num_input_tokens_seen": 12870160896,
      "step": 45260,
      "train_runtime": 440022.2624,
      "train_tokens_per_second": 29248.886
    },
    {
      "epoch": 1.6017421225235347,
      "grad_norm": 0.58203125,
      "learning_rate": 2.1268124842278744e-05,
      "loss": 0.3884448528289795,
      "num_input_tokens_seen": 12873036864,
      "step": 45270,
      "train_runtime": 440121.0662,
      "train_tokens_per_second": 29248.854
    },
    {
      "epoch": 1.6020959427872632,
      "grad_norm": 0.609375,
      "learning_rate": 2.1266201047819458e-05,
      "loss": 0.39068496227264404,
      "num_input_tokens_seen": 12875860800,
      "step": 45280,
      "train_runtime": 440216.6337,
      "train_tokens_per_second": 29248.919
    },
    {
      "epoch": 1.6024497630509922,
      "grad_norm": 0.61328125,
      "learning_rate": 2.126427777531249e-05,
      "loss": 0.38841488361358645,
      "num_input_tokens_seen": 12878672064,
      "step": 45290,
      "train_runtime": 440313.4151,
      "train_tokens_per_second": 29248.875
    },
    {
      "epoch": 1.602803583314721,
      "grad_norm": 0.6171875,
      "learning_rate": 2.126235502452186e-05,
      "loss": 0.3897926092147827,
      "num_input_tokens_seen": 12881511680,
      "step": 45300,
      "train_runtime": 440411.8428,
      "train_tokens_per_second": 29248.786
    },
    {
      "epoch": 1.6031574035784497,
      "grad_norm": 0.59375,
      "learning_rate": 2.1260432795211747e-05,
      "loss": 0.39048631191253663,
      "num_input_tokens_seen": 12884358464,
      "step": 45310,
      "train_runtime": 440507.1425,
      "train_tokens_per_second": 29248.921
    },
    {
      "epoch": 1.6035112238421785,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1258511087146463e-05,
      "loss": 0.38648035526275637,
      "num_input_tokens_seen": 12887202048,
      "step": 45320,
      "train_runtime": 440605.1579,
      "train_tokens_per_second": 29248.868
    },
    {
      "epoch": 1.6038650441059072,
      "grad_norm": 0.6015625,
      "learning_rate": 2.125658990009048e-05,
      "loss": 0.39255719184875487,
      "num_input_tokens_seen": 12890055360,
      "step": 45330,
      "train_runtime": 440702.3899,
      "train_tokens_per_second": 29248.889
    },
    {
      "epoch": 1.604218864369636,
      "grad_norm": 0.625,
      "learning_rate": 2.125466923380842e-05,
      "loss": 0.39010910987854003,
      "num_input_tokens_seen": 12892913984,
      "step": 45340,
      "train_runtime": 440796.0499,
      "train_tokens_per_second": 29249.16
    },
    {
      "epoch": 1.6045726846333648,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1252749088065038e-05,
      "loss": 0.38816065788269044,
      "num_input_tokens_seen": 12895766784,
      "step": 45350,
      "train_runtime": 440892.5637,
      "train_tokens_per_second": 29249.227
    },
    {
      "epoch": 1.6049265048970935,
      "grad_norm": 0.60546875,
      "learning_rate": 2.125082946262526e-05,
      "loss": 0.38496904373168944,
      "num_input_tokens_seen": 12898648448,
      "step": 45360,
      "train_runtime": 440992.4429,
      "train_tokens_per_second": 29249.137
    },
    {
      "epoch": 1.6052803251608223,
      "grad_norm": 0.6328125,
      "learning_rate": 2.1248910357254146e-05,
      "loss": 0.38748290538787844,
      "num_input_tokens_seen": 12901506176,
      "step": 45370,
      "train_runtime": 441089.2587,
      "train_tokens_per_second": 29249.196
    },
    {
      "epoch": 1.6056341454245513,
      "grad_norm": 0.5859375,
      "learning_rate": 2.1246991771716913e-05,
      "loss": 0.3928998470306396,
      "num_input_tokens_seen": 12904395392,
      "step": 45380,
      "train_runtime": 441189.34,
      "train_tokens_per_second": 29249.11
    },
    {
      "epoch": 1.6059879656882798,
      "grad_norm": 0.61328125,
      "learning_rate": 2.124507370577891e-05,
      "loss": 0.385508394241333,
      "num_input_tokens_seen": 12907297152,
      "step": 45390,
      "train_runtime": 441289.8673,
      "train_tokens_per_second": 29249.022
    },
    {
      "epoch": 1.6063417859520088,
      "grad_norm": 0.5859375,
      "learning_rate": 2.1243156159205653e-05,
      "loss": 0.3894524574279785,
      "num_input_tokens_seen": 12910122304,
      "step": 45400,
      "train_runtime": 441385.4849,
      "train_tokens_per_second": 29249.087
    },
    {
      "epoch": 1.6066956062157374,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1241239131762808e-05,
      "loss": 0.3883928060531616,
      "num_input_tokens_seen": 12912981696,
      "step": 45410,
      "train_runtime": 441480.7367,
      "train_tokens_per_second": 29249.253
    },
    {
      "epoch": 1.6070494264794664,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1239322623216165e-05,
      "loss": 0.38733625411987305,
      "num_input_tokens_seen": 12915846144,
      "step": 45420,
      "train_runtime": 441579.0986,
      "train_tokens_per_second": 29249.224
    },
    {
      "epoch": 1.607403246743195,
      "grad_norm": 0.6328125,
      "learning_rate": 2.1237406633331682e-05,
      "loss": 0.39173614978790283,
      "num_input_tokens_seen": 12918590592,
      "step": 45430,
      "train_runtime": 441672.402,
      "train_tokens_per_second": 29249.259
    },
    {
      "epoch": 1.607757067006924,
      "grad_norm": 0.64453125,
      "learning_rate": 2.1235491161875466e-05,
      "loss": 0.3900144577026367,
      "num_input_tokens_seen": 12921430848,
      "step": 45440,
      "train_runtime": 441769.4881,
      "train_tokens_per_second": 29249.261
    },
    {
      "epoch": 1.6081108872706524,
      "grad_norm": 0.62890625,
      "learning_rate": 2.123357620861376e-05,
      "loss": 0.3882885932922363,
      "num_input_tokens_seen": 12924311040,
      "step": 45450,
      "train_runtime": 441868.9054,
      "train_tokens_per_second": 29249.198
    },
    {
      "epoch": 1.6084647075343814,
      "grad_norm": 0.59375,
      "learning_rate": 2.1231661773312958e-05,
      "loss": 0.38944187164306643,
      "num_input_tokens_seen": 12927140416,
      "step": 45460,
      "train_runtime": 441964.2725,
      "train_tokens_per_second": 29249.288
    },
    {
      "epoch": 1.6088185277981102,
      "grad_norm": 0.60546875,
      "learning_rate": 2.122974785573961e-05,
      "loss": 0.3875762939453125,
      "num_input_tokens_seen": 12929985536,
      "step": 45470,
      "train_runtime": 442061.0217,
      "train_tokens_per_second": 29249.323
    },
    {
      "epoch": 1.609172348061839,
      "grad_norm": 0.6328125,
      "learning_rate": 2.12278344556604e-05,
      "loss": 0.38895392417907715,
      "num_input_tokens_seen": 12932874688,
      "step": 45480,
      "train_runtime": 442161.6975,
      "train_tokens_per_second": 29249.197
    },
    {
      "epoch": 1.6095261683255677,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1225921572842165e-05,
      "loss": 0.3914012432098389,
      "num_input_tokens_seen": 12935733760,
      "step": 45490,
      "train_runtime": 442259.7988,
      "train_tokens_per_second": 29249.174
    },
    {
      "epoch": 1.6098799885892965,
      "grad_norm": 0.6015625,
      "learning_rate": 2.12240092070519e-05,
      "loss": 0.38959786891937254,
      "num_input_tokens_seen": 12938590528,
      "step": 45500,
      "train_runtime": 442356.5536,
      "train_tokens_per_second": 29249.234
    },
    {
      "epoch": 1.6102338088530253,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1222097358056727e-05,
      "loss": 0.387194299697876,
      "num_input_tokens_seen": 12941407296,
      "step": 45510,
      "train_runtime": 442452.0517,
      "train_tokens_per_second": 29249.288
    },
    {
      "epoch": 1.610587629116754,
      "grad_norm": 0.59765625,
      "learning_rate": 2.122018602562393e-05,
      "loss": 0.3864673376083374,
      "num_input_tokens_seen": 12944279360,
      "step": 45520,
      "train_runtime": 442547.8466,
      "train_tokens_per_second": 29249.446
    },
    {
      "epoch": 1.6109414493804828,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1218275209520926e-05,
      "loss": 0.3891880989074707,
      "num_input_tokens_seen": 12947076288,
      "step": 45530,
      "train_runtime": 442642.5694,
      "train_tokens_per_second": 29249.506
    },
    {
      "epoch": 1.6112952696442115,
      "grad_norm": 0.58203125,
      "learning_rate": 2.1216364909515306e-05,
      "loss": 0.3909165382385254,
      "num_input_tokens_seen": 12949908672,
      "step": 45540,
      "train_runtime": 442738.512,
      "train_tokens_per_second": 29249.565
    },
    {
      "epoch": 1.6116490899079405,
      "grad_norm": 0.625,
      "learning_rate": 2.1214455125374763e-05,
      "loss": 0.3852527618408203,
      "num_input_tokens_seen": 12952833728,
      "step": 45550,
      "train_runtime": 442841.4899,
      "train_tokens_per_second": 29249.368
    },
    {
      "epoch": 1.612002910171669,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1212545856867176e-05,
      "loss": 0.3862185478210449,
      "num_input_tokens_seen": 12955658368,
      "step": 45560,
      "train_runtime": 442936.5167,
      "train_tokens_per_second": 29249.47
    },
    {
      "epoch": 1.612356730435398,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1210637103760558e-05,
      "loss": 0.3919167995452881,
      "num_input_tokens_seen": 12958541632,
      "step": 45570,
      "train_runtime": 443037.5259,
      "train_tokens_per_second": 29249.309
    },
    {
      "epoch": 1.6127105506991266,
      "grad_norm": 0.59375,
      "learning_rate": 2.1208728865823054e-05,
      "loss": 0.389019775390625,
      "num_input_tokens_seen": 12961378560,
      "step": 45580,
      "train_runtime": 443131.7045,
      "train_tokens_per_second": 29249.495
    },
    {
      "epoch": 1.6130643709628556,
      "grad_norm": 0.60546875,
      "learning_rate": 2.120682114282298e-05,
      "loss": 0.38559646606445314,
      "num_input_tokens_seen": 12964219904,
      "step": 45590,
      "train_runtime": 443227.6238,
      "train_tokens_per_second": 29249.576
    },
    {
      "epoch": 1.6134181912265841,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1204913934528784e-05,
      "loss": 0.38884451389312746,
      "num_input_tokens_seen": 12967115520,
      "step": 45600,
      "train_runtime": 443328.553,
      "train_tokens_per_second": 29249.448
    },
    {
      "epoch": 1.6137720114903131,
      "grad_norm": 0.59765625,
      "learning_rate": 2.120300724070905e-05,
      "loss": 0.3895893573760986,
      "num_input_tokens_seen": 12969981824,
      "step": 45610,
      "train_runtime": 443424.6611,
      "train_tokens_per_second": 29249.573
    },
    {
      "epoch": 1.614125831754042,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1201101061132525e-05,
      "loss": 0.39345195293426516,
      "num_input_tokens_seen": 12972808768,
      "step": 45620,
      "train_runtime": 443518.7004,
      "train_tokens_per_second": 29249.745
    },
    {
      "epoch": 1.6144796520177707,
      "grad_norm": 0.6328125,
      "learning_rate": 2.1199195395568095e-05,
      "loss": 0.38239092826843263,
      "num_input_tokens_seen": 12975654144,
      "step": 45630,
      "train_runtime": 443613.8372,
      "train_tokens_per_second": 29249.886
    },
    {
      "epoch": 1.6148334722814994,
      "grad_norm": 0.61328125,
      "learning_rate": 2.119729024378479e-05,
      "loss": 0.39530048370361326,
      "num_input_tokens_seen": 12978517376,
      "step": 45640,
      "train_runtime": 443710.8838,
      "train_tokens_per_second": 29249.941
    },
    {
      "epoch": 1.6151872925452282,
      "grad_norm": 0.5859375,
      "learning_rate": 2.119538560555179e-05,
      "loss": 0.3886423587799072,
      "num_input_tokens_seen": 12981345408,
      "step": 45650,
      "train_runtime": 443806.7277,
      "train_tokens_per_second": 29249.997
    },
    {
      "epoch": 1.615541112808957,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1193481480638405e-05,
      "loss": 0.3916743516921997,
      "num_input_tokens_seen": 12984199744,
      "step": 45660,
      "train_runtime": 443904.6842,
      "train_tokens_per_second": 29249.972
    },
    {
      "epoch": 1.6158949330726857,
      "grad_norm": 0.6328125,
      "learning_rate": 2.119157786881411e-05,
      "loss": 0.38446879386901855,
      "num_input_tokens_seen": 12986996160,
      "step": 45670,
      "train_runtime": 443999.7416,
      "train_tokens_per_second": 29250.008
    },
    {
      "epoch": 1.6162487533364145,
      "grad_norm": 0.6484375,
      "learning_rate": 2.118967476984852e-05,
      "loss": 0.3916933536529541,
      "num_input_tokens_seen": 12989805504,
      "step": 45680,
      "train_runtime": 444097.1134,
      "train_tokens_per_second": 29249.921
    },
    {
      "epoch": 1.6166025736001433,
      "grad_norm": 0.6015625,
      "learning_rate": 2.118777218351139e-05,
      "loss": 0.38768706321716306,
      "num_input_tokens_seen": 12992623296,
      "step": 45690,
      "train_runtime": 444194.4075,
      "train_tokens_per_second": 29249.858
    },
    {
      "epoch": 1.616956393863872,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1185870109572614e-05,
      "loss": 0.3936758041381836,
      "num_input_tokens_seen": 12995437504,
      "step": 45700,
      "train_runtime": 444289.8534,
      "train_tokens_per_second": 29249.908
    },
    {
      "epoch": 1.6173102141276008,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1183968547802243e-05,
      "loss": 0.3878281593322754,
      "num_input_tokens_seen": 12998207936,
      "step": 45710,
      "train_runtime": 444386.8101,
      "train_tokens_per_second": 29249.761
    },
    {
      "epoch": 1.6176640343913298,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1182067497970464e-05,
      "loss": 0.3898247957229614,
      "num_input_tokens_seen": 13001048000,
      "step": 45720,
      "train_runtime": 444484.811,
      "train_tokens_per_second": 29249.701
    },
    {
      "epoch": 1.6180178546550583,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1180166959847613e-05,
      "loss": 0.3905017614364624,
      "num_input_tokens_seen": 13003955456,
      "step": 45730,
      "train_runtime": 444588.9167,
      "train_tokens_per_second": 29249.392
    },
    {
      "epoch": 1.6183716749187873,
      "grad_norm": 0.609375,
      "learning_rate": 2.1178266933204168e-05,
      "loss": 0.38474531173706056,
      "num_input_tokens_seen": 13006771584,
      "step": 45740,
      "train_runtime": 444685.5909,
      "train_tokens_per_second": 29249.366
    },
    {
      "epoch": 1.6187254951825158,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1176367417810754e-05,
      "loss": 0.38822102546691895,
      "num_input_tokens_seen": 13009608512,
      "step": 45750,
      "train_runtime": 444782.9026,
      "train_tokens_per_second": 29249.345
    },
    {
      "epoch": 1.6190793154462448,
      "grad_norm": 0.6328125,
      "learning_rate": 2.1174468413438138e-05,
      "loss": 0.39193115234375,
      "num_input_tokens_seen": 13012532928,
      "step": 45760,
      "train_runtime": 444885.5319,
      "train_tokens_per_second": 29249.171
    },
    {
      "epoch": 1.6194331357099734,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1172569919857225e-05,
      "loss": 0.38861267566680907,
      "num_input_tokens_seen": 13015375424,
      "step": 45770,
      "train_runtime": 444982.4495,
      "train_tokens_per_second": 29249.188
    },
    {
      "epoch": 1.6197869559737024,
      "grad_norm": 0.640625,
      "learning_rate": 2.117067193683907e-05,
      "loss": 0.3913105487823486,
      "num_input_tokens_seen": 13018198976,
      "step": 45780,
      "train_runtime": 445077.6808,
      "train_tokens_per_second": 29249.274
    },
    {
      "epoch": 1.6201407762374311,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1168774464154877e-05,
      "loss": 0.3899451732635498,
      "num_input_tokens_seen": 13021067456,
      "step": 45790,
      "train_runtime": 445175.8421,
      "train_tokens_per_second": 29249.268
    },
    {
      "epoch": 1.62049459650116,
      "grad_norm": 0.578125,
      "learning_rate": 2.116687750157598e-05,
      "loss": 0.38467795848846437,
      "num_input_tokens_seen": 13023915008,
      "step": 45800,
      "train_runtime": 445275.6805,
      "train_tokens_per_second": 29249.105
    },
    {
      "epoch": 1.6208484167648887,
      "grad_norm": 0.625,
      "learning_rate": 2.116498104887387e-05,
      "loss": 0.38592872619628904,
      "num_input_tokens_seen": 13026771648,
      "step": 45810,
      "train_runtime": 445374.3353,
      "train_tokens_per_second": 29249.04
    },
    {
      "epoch": 1.6212022370286174,
      "grad_norm": 0.59375,
      "learning_rate": 2.1163085105820175e-05,
      "loss": 0.3860755681991577,
      "num_input_tokens_seen": 13029592256,
      "step": 45820,
      "train_runtime": 445469.7779,
      "train_tokens_per_second": 29249.105
    },
    {
      "epoch": 1.6215560572923462,
      "grad_norm": 0.625,
      "learning_rate": 2.116118967218666e-05,
      "loss": 0.38577728271484374,
      "num_input_tokens_seen": 13032440768,
      "step": 45830,
      "train_runtime": 445566.9765,
      "train_tokens_per_second": 29249.117
    },
    {
      "epoch": 1.621909877556075,
      "grad_norm": 0.62109375,
      "learning_rate": 2.115929474774525e-05,
      "loss": 0.38961329460144045,
      "num_input_tokens_seen": 13035298816,
      "step": 45840,
      "train_runtime": 445662.474,
      "train_tokens_per_second": 29249.263
    },
    {
      "epoch": 1.6222636978198037,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1157400332267988e-05,
      "loss": 0.3902440071105957,
      "num_input_tokens_seen": 13038126784,
      "step": 45850,
      "train_runtime": 445758.477,
      "train_tokens_per_second": 29249.308
    },
    {
      "epoch": 1.6226175180835325,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1155506425527083e-05,
      "loss": 0.3902561187744141,
      "num_input_tokens_seen": 13040989568,
      "step": 45860,
      "train_runtime": 445854.7956,
      "train_tokens_per_second": 29249.41
    },
    {
      "epoch": 1.6229713383472615,
      "grad_norm": 0.625,
      "learning_rate": 2.115361302729488e-05,
      "loss": 0.3927320003509521,
      "num_input_tokens_seen": 13043850432,
      "step": 45870,
      "train_runtime": 445953.749,
      "train_tokens_per_second": 29249.335
    },
    {
      "epoch": 1.62332515861099,
      "grad_norm": 0.6171875,
      "learning_rate": 2.115172013734386e-05,
      "loss": 0.38945329189300537,
      "num_input_tokens_seen": 13046672512,
      "step": 45880,
      "train_runtime": 446049.2243,
      "train_tokens_per_second": 29249.401
    },
    {
      "epoch": 1.623678978874719,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1149827755446652e-05,
      "loss": 0.3876248836517334,
      "num_input_tokens_seen": 13049484224,
      "step": 45890,
      "train_runtime": 446147.7794,
      "train_tokens_per_second": 29249.242
    },
    {
      "epoch": 1.6240327991384476,
      "grad_norm": 0.59375,
      "learning_rate": 2.1147935881376024e-05,
      "loss": 0.3892193794250488,
      "num_input_tokens_seen": 13052268352,
      "step": 45900,
      "train_runtime": 446238.3939,
      "train_tokens_per_second": 29249.541
    },
    {
      "epoch": 1.6243866194021765,
      "grad_norm": 0.60546875,
      "learning_rate": 2.114604451490489e-05,
      "loss": 0.3882743835449219,
      "num_input_tokens_seen": 13055154752,
      "step": 45910,
      "train_runtime": 446336.3084,
      "train_tokens_per_second": 29249.592
    },
    {
      "epoch": 1.624740439665905,
      "grad_norm": 0.58203125,
      "learning_rate": 2.1144153655806313e-05,
      "loss": 0.3892354011535645,
      "num_input_tokens_seen": 13058002880,
      "step": 45920,
      "train_runtime": 446432.8786,
      "train_tokens_per_second": 29249.644
    },
    {
      "epoch": 1.625094259929634,
      "grad_norm": 0.59375,
      "learning_rate": 2.114226330385347e-05,
      "loss": 0.38631439208984375,
      "num_input_tokens_seen": 13060896256,
      "step": 45930,
      "train_runtime": 446534.7978,
      "train_tokens_per_second": 29249.448
    },
    {
      "epoch": 1.6254480801933626,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1140373458819724e-05,
      "loss": 0.3903128862380981,
      "num_input_tokens_seen": 13063714752,
      "step": 45940,
      "train_runtime": 446634.0336,
      "train_tokens_per_second": 29249.26
    },
    {
      "epoch": 1.6258019004570916,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1138484120478535e-05,
      "loss": 0.38848493099212644,
      "num_input_tokens_seen": 13066550336,
      "step": 45950,
      "train_runtime": 446728.5558,
      "train_tokens_per_second": 29249.418
    },
    {
      "epoch": 1.6261557207208204,
      "grad_norm": 0.59375,
      "learning_rate": 2.1136595288603532e-05,
      "loss": 0.3935427188873291,
      "num_input_tokens_seen": 13069404032,
      "step": 45960,
      "train_runtime": 446830.6119,
      "train_tokens_per_second": 29249.124
    },
    {
      "epoch": 1.6265095409845491,
      "grad_norm": 0.62890625,
      "learning_rate": 2.113470696296849e-05,
      "loss": 0.3888894319534302,
      "num_input_tokens_seen": 13072281472,
      "step": 45970,
      "train_runtime": 446928.6995,
      "train_tokens_per_second": 29249.143
    },
    {
      "epoch": 1.626863361248278,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1132819143347296e-05,
      "loss": 0.39215869903564454,
      "num_input_tokens_seen": 13075117376,
      "step": 45980,
      "train_runtime": 447027.062,
      "train_tokens_per_second": 29249.051
    },
    {
      "epoch": 1.6272171815120067,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1130931829514005e-05,
      "loss": 0.38678665161132814,
      "num_input_tokens_seen": 13077915200,
      "step": 45990,
      "train_runtime": 447120.9937,
      "train_tokens_per_second": 29249.164
    },
    {
      "epoch": 1.6275710017757354,
      "grad_norm": 0.61328125,
      "learning_rate": 2.112904502124281e-05,
      "loss": 0.39141292572021485,
      "num_input_tokens_seen": 13080718592,
      "step": 46000,
      "train_runtime": 447215.0943,
      "train_tokens_per_second": 29249.278
    },
    {
      "epoch": 1.6279248220394642,
      "grad_norm": 0.60546875,
      "learning_rate": 2.112715871830803e-05,
      "loss": 0.38442325592041016,
      "num_input_tokens_seen": 13083587968,
      "step": 46010,
      "train_runtime": 447313.7075,
      "train_tokens_per_second": 29249.244
    },
    {
      "epoch": 1.628278642303193,
      "grad_norm": 0.62890625,
      "learning_rate": 2.112527292048414e-05,
      "loss": 0.39490082263946535,
      "num_input_tokens_seen": 13086399744,
      "step": 46020,
      "train_runtime": 447409.6308,
      "train_tokens_per_second": 29249.258
    },
    {
      "epoch": 1.6286324625669217,
      "grad_norm": 0.609375,
      "learning_rate": 2.1123387627545748e-05,
      "loss": 0.3847757339477539,
      "num_input_tokens_seen": 13089255104,
      "step": 46030,
      "train_runtime": 447506.9711,
      "train_tokens_per_second": 29249.276
    },
    {
      "epoch": 1.6289862828306507,
      "grad_norm": 0.62109375,
      "learning_rate": 2.112150283926761e-05,
      "loss": 0.3872460603713989,
      "num_input_tokens_seen": 13092094592,
      "step": 46040,
      "train_runtime": 447605.0378,
      "train_tokens_per_second": 29249.212
    },
    {
      "epoch": 1.6293401030943793,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1119618555424612e-05,
      "loss": 0.3925561666488647,
      "num_input_tokens_seen": 13094937920,
      "step": 46050,
      "train_runtime": 447701.5626,
      "train_tokens_per_second": 29249.257
    },
    {
      "epoch": 1.6296939233581083,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1117734775791794e-05,
      "loss": 0.39198276996612547,
      "num_input_tokens_seen": 13097787008,
      "step": 46060,
      "train_runtime": 447801.6987,
      "train_tokens_per_second": 29249.078
    },
    {
      "epoch": 1.6300477436218368,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1115851500144324e-05,
      "loss": 0.3884898662567139,
      "num_input_tokens_seen": 13100669632,
      "step": 46070,
      "train_runtime": 447903.4452,
      "train_tokens_per_second": 29248.87
    },
    {
      "epoch": 1.6304015638855658,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1113968728257516e-05,
      "loss": 0.3887759208679199,
      "num_input_tokens_seen": 13103485440,
      "step": 46080,
      "train_runtime": 447996.9217,
      "train_tokens_per_second": 29249.052
    },
    {
      "epoch": 1.6307553841492943,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1112086459906828e-05,
      "loss": 0.3887644290924072,
      "num_input_tokens_seen": 13106362880,
      "step": 46090,
      "train_runtime": 448094.9702,
      "train_tokens_per_second": 29249.074
    },
    {
      "epoch": 1.6311092044130233,
      "grad_norm": 0.625,
      "learning_rate": 2.1110204694867846e-05,
      "loss": 0.38816585540771487,
      "num_input_tokens_seen": 13109184320,
      "step": 46100,
      "train_runtime": 448191.558,
      "train_tokens_per_second": 29249.066
    },
    {
      "epoch": 1.6314630246767519,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1108323432916305e-05,
      "loss": 0.3907257080078125,
      "num_input_tokens_seen": 13111998528,
      "step": 46110,
      "train_runtime": 448286.0098,
      "train_tokens_per_second": 29249.181
    },
    {
      "epoch": 1.6318168449404808,
      "grad_norm": 0.609375,
      "learning_rate": 2.1106442673828086e-05,
      "loss": 0.38648648262023927,
      "num_input_tokens_seen": 13114796928,
      "step": 46120,
      "train_runtime": 448381.8918,
      "train_tokens_per_second": 29249.167
    },
    {
      "epoch": 1.6321706652042096,
      "grad_norm": 0.58203125,
      "learning_rate": 2.1104562417379202e-05,
      "loss": 0.38654747009277346,
      "num_input_tokens_seen": 13117667904,
      "step": 46130,
      "train_runtime": 448479.9792,
      "train_tokens_per_second": 29249.172
    },
    {
      "epoch": 1.6325244854679384,
      "grad_norm": 0.609375,
      "learning_rate": 2.1102682663345797e-05,
      "loss": 0.39447746276855467,
      "num_input_tokens_seen": 13120485824,
      "step": 46140,
      "train_runtime": 448578.4906,
      "train_tokens_per_second": 29249.03
    },
    {
      "epoch": 1.6328783057316671,
      "grad_norm": 0.625,
      "learning_rate": 2.110080341150417e-05,
      "loss": 0.3891158580780029,
      "num_input_tokens_seen": 13123296384,
      "step": 46150,
      "train_runtime": 448675.9349,
      "train_tokens_per_second": 29248.942
    },
    {
      "epoch": 1.633232125995396,
      "grad_norm": 0.64453125,
      "learning_rate": 2.1098924661630748e-05,
      "loss": 0.3895108222961426,
      "num_input_tokens_seen": 13126158592,
      "step": 46160,
      "train_runtime": 448773.8159,
      "train_tokens_per_second": 29248.94
    },
    {
      "epoch": 1.6335859462591247,
      "grad_norm": 0.59375,
      "learning_rate": 2.109704641350211e-05,
      "loss": 0.3934329032897949,
      "num_input_tokens_seen": 13128961920,
      "step": 46170,
      "train_runtime": 448868.9857,
      "train_tokens_per_second": 29248.984
    },
    {
      "epoch": 1.6339397665228534,
      "grad_norm": 0.58203125,
      "learning_rate": 2.1095168666894962e-05,
      "loss": 0.3871035099029541,
      "num_input_tokens_seen": 13131829312,
      "step": 46180,
      "train_runtime": 448964.9138,
      "train_tokens_per_second": 29249.121
    },
    {
      "epoch": 1.6342935867865822,
      "grad_norm": 0.59375,
      "learning_rate": 2.1093291421586156e-05,
      "loss": 0.3858203411102295,
      "num_input_tokens_seen": 13134680704,
      "step": 46190,
      "train_runtime": 449064.2398,
      "train_tokens_per_second": 29249.002
    },
    {
      "epoch": 1.634647407050311,
      "grad_norm": 0.6015625,
      "learning_rate": 2.1091414677352676e-05,
      "loss": 0.38942317962646483,
      "num_input_tokens_seen": 13137536192,
      "step": 46200,
      "train_runtime": 449163.3913,
      "train_tokens_per_second": 29248.902
    },
    {
      "epoch": 1.63500122731404,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1089538433971655e-05,
      "loss": 0.38621108531951903,
      "num_input_tokens_seen": 13140289920,
      "step": 46210,
      "train_runtime": 449252.792,
      "train_tokens_per_second": 29249.211
    },
    {
      "epoch": 1.6353550475777685,
      "grad_norm": 0.58203125,
      "learning_rate": 2.1087662691220356e-05,
      "loss": 0.38844685554504393,
      "num_input_tokens_seen": 13143133696,
      "step": 46220,
      "train_runtime": 449352.7402,
      "train_tokens_per_second": 29249.034
    },
    {
      "epoch": 1.6357088678414975,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1085787448876184e-05,
      "loss": 0.38851022720336914,
      "num_input_tokens_seen": 13145959488,
      "step": 46230,
      "train_runtime": 449447.8473,
      "train_tokens_per_second": 29249.132
    },
    {
      "epoch": 1.636062688105226,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1083912706716684e-05,
      "loss": 0.3890007734298706,
      "num_input_tokens_seen": 13148823552,
      "step": 46240,
      "train_runtime": 449546.7192,
      "train_tokens_per_second": 29249.07
    },
    {
      "epoch": 1.636416508368955,
      "grad_norm": 0.61328125,
      "learning_rate": 2.108203846451954e-05,
      "loss": 0.3935135841369629,
      "num_input_tokens_seen": 13151654272,
      "step": 46250,
      "train_runtime": 449641.8274,
      "train_tokens_per_second": 29249.179
    },
    {
      "epoch": 1.6367703286326836,
      "grad_norm": 0.609375,
      "learning_rate": 2.1080164722062563e-05,
      "loss": 0.3888871192932129,
      "num_input_tokens_seen": 13154499264,
      "step": 46260,
      "train_runtime": 449740.3228,
      "train_tokens_per_second": 29249.099
    },
    {
      "epoch": 1.6371241488964126,
      "grad_norm": 0.59765625,
      "learning_rate": 2.107829147912372e-05,
      "loss": 0.38654944896697996,
      "num_input_tokens_seen": 13157342336,
      "step": 46270,
      "train_runtime": 449834.7353,
      "train_tokens_per_second": 29249.28
    },
    {
      "epoch": 1.637477969160141,
      "grad_norm": 0.59375,
      "learning_rate": 2.1076418735481103e-05,
      "loss": 0.3874382019042969,
      "num_input_tokens_seen": 13160202240,
      "step": 46280,
      "train_runtime": 449931.8959,
      "train_tokens_per_second": 29249.321
    },
    {
      "epoch": 1.63783178942387,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1074546490912953e-05,
      "loss": 0.39109230041503906,
      "num_input_tokens_seen": 13163059008,
      "step": 46290,
      "train_runtime": 450026.4284,
      "train_tokens_per_second": 29249.524
    },
    {
      "epoch": 1.6381856096875989,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1072674745197633e-05,
      "loss": 0.3839420318603516,
      "num_input_tokens_seen": 13165887680,
      "step": 46300,
      "train_runtime": 450122.0673,
      "train_tokens_per_second": 29249.594
    },
    {
      "epoch": 1.6385394299513276,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1070803498113666e-05,
      "loss": 0.38668949604034425,
      "num_input_tokens_seen": 13168763136,
      "step": 46310,
      "train_runtime": 450218.0087,
      "train_tokens_per_second": 29249.748
    },
    {
      "epoch": 1.6388932502150564,
      "grad_norm": 0.640625,
      "learning_rate": 2.1068932749439683e-05,
      "loss": 0.38903012275695803,
      "num_input_tokens_seen": 13171664704,
      "step": 46320,
      "train_runtime": 450317.5456,
      "train_tokens_per_second": 29249.726
    },
    {
      "epoch": 1.6392470704787852,
      "grad_norm": 0.61328125,
      "learning_rate": 2.106706249895448e-05,
      "loss": 0.3871425151824951,
      "num_input_tokens_seen": 13174498368,
      "step": 46330,
      "train_runtime": 450411.0583,
      "train_tokens_per_second": 29249.944
    },
    {
      "epoch": 1.639600890742514,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1065192746436978e-05,
      "loss": 0.39148578643798826,
      "num_input_tokens_seen": 13177369472,
      "step": 46340,
      "train_runtime": 450509.897,
      "train_tokens_per_second": 29249.9
    },
    {
      "epoch": 1.6399547110062427,
      "grad_norm": 0.609375,
      "learning_rate": 2.1063323491666236e-05,
      "loss": 0.3871280670166016,
      "num_input_tokens_seen": 13180198144,
      "step": 46350,
      "train_runtime": 450605.6525,
      "train_tokens_per_second": 29249.962
    },
    {
      "epoch": 1.6403085312699714,
      "grad_norm": 0.65234375,
      "learning_rate": 2.1061454734421454e-05,
      "loss": 0.3891870975494385,
      "num_input_tokens_seen": 13183083840,
      "step": 46360,
      "train_runtime": 450706.9556,
      "train_tokens_per_second": 29249.79
    },
    {
      "epoch": 1.6406623515337002,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1059586474481964e-05,
      "loss": 0.3866166353225708,
      "num_input_tokens_seen": 13185916928,
      "step": 46370,
      "train_runtime": 450805.1468,
      "train_tokens_per_second": 29249.704
    },
    {
      "epoch": 1.6410161717974292,
      "grad_norm": 0.5859375,
      "learning_rate": 2.1057718711627242e-05,
      "loss": 0.3884653329849243,
      "num_input_tokens_seen": 13188786176,
      "step": 46380,
      "train_runtime": 450903.0062,
      "train_tokens_per_second": 29249.719
    },
    {
      "epoch": 1.6413699920611577,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1055851445636883e-05,
      "loss": 0.3931423187255859,
      "num_input_tokens_seen": 13191628032,
      "step": 46390,
      "train_runtime": 451005.7145,
      "train_tokens_per_second": 29249.359
    },
    {
      "epoch": 1.6417238123248867,
      "grad_norm": 0.5859375,
      "learning_rate": 2.105398467629065e-05,
      "loss": 0.3949535846710205,
      "num_input_tokens_seen": 13194503104,
      "step": 46400,
      "train_runtime": 451104.3716,
      "train_tokens_per_second": 29249.335
    },
    {
      "epoch": 1.6420776325886153,
      "grad_norm": 0.65625,
      "learning_rate": 2.105211840336841e-05,
      "loss": 0.38668758869171144,
      "num_input_tokens_seen": 13197430976,
      "step": 46410,
      "train_runtime": 451209.5355,
      "train_tokens_per_second": 29249.007
    },
    {
      "epoch": 1.6424314528523443,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1050252626650186e-05,
      "loss": 0.3920552730560303,
      "num_input_tokens_seen": 13200295744,
      "step": 46420,
      "train_runtime": 451308.4595,
      "train_tokens_per_second": 29248.944
    },
    {
      "epoch": 1.6427852731160728,
      "grad_norm": 0.59375,
      "learning_rate": 2.104838734591614e-05,
      "loss": 0.3914467811584473,
      "num_input_tokens_seen": 13203172032,
      "step": 46430,
      "train_runtime": 451407.9919,
      "train_tokens_per_second": 29248.866
    },
    {
      "epoch": 1.6431390933798018,
      "grad_norm": 0.625,
      "learning_rate": 2.1046522560946552e-05,
      "loss": 0.3910368919372559,
      "num_input_tokens_seen": 13206040256,
      "step": 46440,
      "train_runtime": 451507.5691,
      "train_tokens_per_second": 29248.768
    },
    {
      "epoch": 1.6434929136435306,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1044658271521856e-05,
      "loss": 0.3880784034729004,
      "num_input_tokens_seen": 13208871616,
      "step": 46450,
      "train_runtime": 451603.4092,
      "train_tokens_per_second": 29248.831
    },
    {
      "epoch": 1.6438467339072593,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1042794477422607e-05,
      "loss": 0.388734245300293,
      "num_input_tokens_seen": 13211724224,
      "step": 46460,
      "train_runtime": 451701.49,
      "train_tokens_per_second": 29248.795
    },
    {
      "epoch": 1.644200554170988,
      "grad_norm": 0.62890625,
      "learning_rate": 2.104093117842952e-05,
      "loss": 0.3912392854690552,
      "num_input_tokens_seen": 13214587840,
      "step": 46470,
      "train_runtime": 451800.011,
      "train_tokens_per_second": 29248.755
    },
    {
      "epoch": 1.6445543744347169,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1039068374323418e-05,
      "loss": 0.3883832931518555,
      "num_input_tokens_seen": 13217446528,
      "step": 46480,
      "train_runtime": 451897.7304,
      "train_tokens_per_second": 29248.756
    },
    {
      "epoch": 1.6449081946984456,
      "grad_norm": 0.59765625,
      "learning_rate": 2.1037206064885274e-05,
      "loss": 0.3877273082733154,
      "num_input_tokens_seen": 13220312960,
      "step": 46490,
      "train_runtime": 451997.8253,
      "train_tokens_per_second": 29248.621
    },
    {
      "epoch": 1.6452620149621744,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1035344249896195e-05,
      "loss": 0.38761367797851565,
      "num_input_tokens_seen": 13223205056,
      "step": 46500,
      "train_runtime": 452097.2946,
      "train_tokens_per_second": 29248.583
    },
    {
      "epoch": 1.6456158352259032,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1033482929137428e-05,
      "loss": 0.38792369365692136,
      "num_input_tokens_seen": 13226083584,
      "step": 46510,
      "train_runtime": 452198.3255,
      "train_tokens_per_second": 29248.413
    },
    {
      "epoch": 1.645969655489632,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1031622102390343e-05,
      "loss": 0.38866071701049804,
      "num_input_tokens_seen": 13228925952,
      "step": 46520,
      "train_runtime": 452294.3403,
      "train_tokens_per_second": 29248.489
    },
    {
      "epoch": 1.6463234757533607,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1029761769436462e-05,
      "loss": 0.39198248386383056,
      "num_input_tokens_seen": 13231716608,
      "step": 46530,
      "train_runtime": 452389.4176,
      "train_tokens_per_second": 29248.51
    },
    {
      "epoch": 1.6466772960170895,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1027901930057424e-05,
      "loss": 0.38909015655517576,
      "num_input_tokens_seen": 13234541632,
      "step": 46540,
      "train_runtime": 452485.4925,
      "train_tokens_per_second": 29248.544
    },
    {
      "epoch": 1.6470311162808184,
      "grad_norm": 0.59375,
      "learning_rate": 2.1026042584035024e-05,
      "loss": 0.3909201145172119,
      "num_input_tokens_seen": 13237397568,
      "step": 46550,
      "train_runtime": 452583.3906,
      "train_tokens_per_second": 29248.527
    },
    {
      "epoch": 1.647384936544547,
      "grad_norm": 0.62890625,
      "learning_rate": 2.102418373115117e-05,
      "loss": 0.38832712173461914,
      "num_input_tokens_seen": 13240272896,
      "step": 46560,
      "train_runtime": 452682.235,
      "train_tokens_per_second": 29248.492
    },
    {
      "epoch": 1.647738756808276,
      "grad_norm": 0.6328125,
      "learning_rate": 2.1022325371187922e-05,
      "loss": 0.386026930809021,
      "num_input_tokens_seen": 13243127168,
      "step": 46570,
      "train_runtime": 452779.1908,
      "train_tokens_per_second": 29248.533
    },
    {
      "epoch": 1.6480925770720045,
      "grad_norm": 0.58984375,
      "learning_rate": 2.102046750392747e-05,
      "loss": 0.3893928527832031,
      "num_input_tokens_seen": 13245971712,
      "step": 46580,
      "train_runtime": 452875.5863,
      "train_tokens_per_second": 29248.589
    },
    {
      "epoch": 1.6484463973357335,
      "grad_norm": 0.60546875,
      "learning_rate": 2.1018610129152124e-05,
      "loss": 0.3930018901824951,
      "num_input_tokens_seen": 13248825600,
      "step": 46590,
      "train_runtime": 452975.5923,
      "train_tokens_per_second": 29248.432
    },
    {
      "epoch": 1.648800217599462,
      "grad_norm": 0.59765625,
      "learning_rate": 2.101675324664436e-05,
      "loss": 0.3875734806060791,
      "num_input_tokens_seen": 13251676864,
      "step": 46600,
      "train_runtime": 453070.9867,
      "train_tokens_per_second": 29248.566
    },
    {
      "epoch": 1.649154037863191,
      "grad_norm": 0.59765625,
      "learning_rate": 2.101489685618676e-05,
      "loss": 0.38723509311676024,
      "num_input_tokens_seen": 13254531008,
      "step": 46610,
      "train_runtime": 453168.4579,
      "train_tokens_per_second": 29248.574
    },
    {
      "epoch": 1.6495078581269198,
      "grad_norm": 0.6328125,
      "learning_rate": 2.1013040957562053e-05,
      "loss": 0.3871634006500244,
      "num_input_tokens_seen": 13257378880,
      "step": 46620,
      "train_runtime": 453265.9474,
      "train_tokens_per_second": 29248.566
    },
    {
      "epoch": 1.6498616783906486,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1011185550553098e-05,
      "loss": 0.38614802360534667,
      "num_input_tokens_seen": 13260264704,
      "step": 46630,
      "train_runtime": 453363.3088,
      "train_tokens_per_second": 29248.65
    },
    {
      "epoch": 1.6502154986543773,
      "grad_norm": 0.61328125,
      "learning_rate": 2.1009330634942888e-05,
      "loss": 0.3922013759613037,
      "num_input_tokens_seen": 13263092864,
      "step": 46640,
      "train_runtime": 453458.5091,
      "train_tokens_per_second": 29248.746
    },
    {
      "epoch": 1.650569318918106,
      "grad_norm": 0.59375,
      "learning_rate": 2.1007476210514558e-05,
      "loss": 0.3882549524307251,
      "num_input_tokens_seen": 13265986880,
      "step": 46650,
      "train_runtime": 453557.7782,
      "train_tokens_per_second": 29248.725
    },
    {
      "epoch": 1.6509231391818349,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1005622277051373e-05,
      "loss": 0.38733229637145994,
      "num_input_tokens_seen": 13268831936,
      "step": 46660,
      "train_runtime": 453654.8428,
      "train_tokens_per_second": 29248.739
    },
    {
      "epoch": 1.6512769594455636,
      "grad_norm": 0.6171875,
      "learning_rate": 2.1003768834336722e-05,
      "loss": 0.38726320266723635,
      "num_input_tokens_seen": 13271678208,
      "step": 46670,
      "train_runtime": 453753.74,
      "train_tokens_per_second": 29248.637
    },
    {
      "epoch": 1.6516307797092924,
      "grad_norm": 0.609375,
      "learning_rate": 2.100191588215414e-05,
      "loss": 0.3897682189941406,
      "num_input_tokens_seen": 13274531392,
      "step": 46680,
      "train_runtime": 453852.5892,
      "train_tokens_per_second": 29248.553
    },
    {
      "epoch": 1.6519845999730212,
      "grad_norm": 0.59375,
      "learning_rate": 2.1000063420287294e-05,
      "loss": 0.390595817565918,
      "num_input_tokens_seen": 13277452096,
      "step": 46690,
      "train_runtime": 453953.6452,
      "train_tokens_per_second": 29248.476
    },
    {
      "epoch": 1.6523384202367501,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0998211448519977e-05,
      "loss": 0.38735337257385255,
      "num_input_tokens_seen": 13280262400,
      "step": 46700,
      "train_runtime": 454048.711,
      "train_tokens_per_second": 29248.541
    },
    {
      "epoch": 1.6526922405004787,
      "grad_norm": 0.5859375,
      "learning_rate": 2.0996359966636126e-05,
      "loss": 0.38974733352661134,
      "num_input_tokens_seen": 13283095104,
      "step": 46710,
      "train_runtime": 454145.5845,
      "train_tokens_per_second": 29248.54
    },
    {
      "epoch": 1.6530460607642077,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0994508974419803e-05,
      "loss": 0.38871474266052247,
      "num_input_tokens_seen": 13285955648,
      "step": 46720,
      "train_runtime": 454245.3277,
      "train_tokens_per_second": 29248.415
    },
    {
      "epoch": 1.6533998810279362,
      "grad_norm": 0.625,
      "learning_rate": 2.0992658471655204e-05,
      "loss": 0.38708810806274413,
      "num_input_tokens_seen": 13288854080,
      "step": 46730,
      "train_runtime": 454344.7538,
      "train_tokens_per_second": 29248.393
    },
    {
      "epoch": 1.6537537012916652,
      "grad_norm": 0.60546875,
      "learning_rate": 2.099080845812666e-05,
      "loss": 0.38606536388397217,
      "num_input_tokens_seen": 13291678272,
      "step": 46740,
      "train_runtime": 454441.144,
      "train_tokens_per_second": 29248.404
    },
    {
      "epoch": 1.6541075215553938,
      "grad_norm": 0.640625,
      "learning_rate": 2.0988958933618644e-05,
      "loss": 0.38857297897338866,
      "num_input_tokens_seen": 13294490624,
      "step": 46750,
      "train_runtime": 454534.9157,
      "train_tokens_per_second": 29248.558
    },
    {
      "epoch": 1.6544613418191227,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0987109897915743e-05,
      "loss": 0.3870939493179321,
      "num_input_tokens_seen": 13297313344,
      "step": 46760,
      "train_runtime": 454628.9877,
      "train_tokens_per_second": 29248.714
    },
    {
      "epoch": 1.6548151620828513,
      "grad_norm": 0.62109375,
      "learning_rate": 2.098526135080269e-05,
      "loss": 0.38361668586730957,
      "num_input_tokens_seen": 13300200448,
      "step": 46770,
      "train_runtime": 454731.5799,
      "train_tokens_per_second": 29248.464
    },
    {
      "epoch": 1.6551689823465803,
      "grad_norm": 0.65234375,
      "learning_rate": 2.098341329206435e-05,
      "loss": 0.3892812728881836,
      "num_input_tokens_seen": 13303008704,
      "step": 46780,
      "train_runtime": 454828.0272,
      "train_tokens_per_second": 29248.437
    },
    {
      "epoch": 1.655522802610309,
      "grad_norm": 0.625,
      "learning_rate": 2.0981565721485717e-05,
      "loss": 0.38861937522888185,
      "num_input_tokens_seen": 13305876544,
      "step": 46790,
      "train_runtime": 454926.9153,
      "train_tokens_per_second": 29248.383
    },
    {
      "epoch": 1.6558766228740378,
      "grad_norm": 0.58203125,
      "learning_rate": 2.097971863885191e-05,
      "loss": 0.38390097618103025,
      "num_input_tokens_seen": 13308730368,
      "step": 46800,
      "train_runtime": 455024.6788,
      "train_tokens_per_second": 29248.37
    },
    {
      "epoch": 1.6562304431377666,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0977872043948204e-05,
      "loss": 0.3919650077819824,
      "num_input_tokens_seen": 13311532672,
      "step": 46810,
      "train_runtime": 455119.1797,
      "train_tokens_per_second": 29248.455
    },
    {
      "epoch": 1.6565842634014953,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0976025936559985e-05,
      "loss": 0.3910619497299194,
      "num_input_tokens_seen": 13314393216,
      "step": 46820,
      "train_runtime": 455217.2855,
      "train_tokens_per_second": 29248.435
    },
    {
      "epoch": 1.656938083665224,
      "grad_norm": 0.6640625,
      "learning_rate": 2.0974180316472775e-05,
      "loss": 0.38962488174438475,
      "num_input_tokens_seen": 13317245120,
      "step": 46830,
      "train_runtime": 455315.2829,
      "train_tokens_per_second": 29248.404
    },
    {
      "epoch": 1.6572919039289529,
      "grad_norm": 0.59765625,
      "learning_rate": 2.097233518347223e-05,
      "loss": 0.3907878875732422,
      "num_input_tokens_seen": 13320087360,
      "step": 46840,
      "train_runtime": 455409.8375,
      "train_tokens_per_second": 29248.572
    },
    {
      "epoch": 1.6576457241926816,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0970490537344147e-05,
      "loss": 0.3881929874420166,
      "num_input_tokens_seen": 13322951360,
      "step": 46850,
      "train_runtime": 455508.4169,
      "train_tokens_per_second": 29248.529
    },
    {
      "epoch": 1.6579995444564104,
      "grad_norm": 0.6015625,
      "learning_rate": 2.096864637787444e-05,
      "loss": 0.38553380966186523,
      "num_input_tokens_seen": 13325792384,
      "step": 46860,
      "train_runtime": 455607.6401,
      "train_tokens_per_second": 29248.395
    },
    {
      "epoch": 1.6583533647201394,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0966802704849163e-05,
      "loss": 0.3898475170135498,
      "num_input_tokens_seen": 13328666816,
      "step": 46870,
      "train_runtime": 455707.071,
      "train_tokens_per_second": 29248.321
    },
    {
      "epoch": 1.658707184983868,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0964959518054493e-05,
      "loss": 0.39019615650177003,
      "num_input_tokens_seen": 13331439104,
      "step": 46880,
      "train_runtime": 455799.4746,
      "train_tokens_per_second": 29248.474
    },
    {
      "epoch": 1.659061005247597,
      "grad_norm": 0.59375,
      "learning_rate": 2.0963116817276757e-05,
      "loss": 0.3866535425186157,
      "num_input_tokens_seen": 13334280384,
      "step": 46890,
      "train_runtime": 455898.4874,
      "train_tokens_per_second": 29248.354
    },
    {
      "epoch": 1.6594148255113255,
      "grad_norm": 0.6171875,
      "learning_rate": 2.09612746023024e-05,
      "loss": 0.3936209201812744,
      "num_input_tokens_seen": 13337033728,
      "step": 46900,
      "train_runtime": 455989.5888,
      "train_tokens_per_second": 29248.549
    },
    {
      "epoch": 1.6597686457750545,
      "grad_norm": 0.59765625,
      "learning_rate": 2.095943287291799e-05,
      "loss": 0.38361918926239014,
      "num_input_tokens_seen": 13339911232,
      "step": 46910,
      "train_runtime": 456089.4049,
      "train_tokens_per_second": 29248.457
    },
    {
      "epoch": 1.660122466038783,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0957591628910243e-05,
      "loss": 0.3843294382095337,
      "num_input_tokens_seen": 13342779200,
      "step": 46920,
      "train_runtime": 456187.6947,
      "train_tokens_per_second": 29248.442
    },
    {
      "epoch": 1.660476286302512,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0955750870066006e-05,
      "loss": 0.3919475555419922,
      "num_input_tokens_seen": 13345669440,
      "step": 46930,
      "train_runtime": 456286.4408,
      "train_tokens_per_second": 29248.446
    },
    {
      "epoch": 1.6608301065662405,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0953910596172238e-05,
      "loss": 0.38295903205871584,
      "num_input_tokens_seen": 13348459328,
      "step": 46940,
      "train_runtime": 456381.235,
      "train_tokens_per_second": 29248.484
    },
    {
      "epoch": 1.6611839268299695,
      "grad_norm": 0.61328125,
      "learning_rate": 2.095207080701605e-05,
      "loss": 0.388715934753418,
      "num_input_tokens_seen": 13351299712,
      "step": 46950,
      "train_runtime": 456481.3803,
      "train_tokens_per_second": 29248.29
    },
    {
      "epoch": 1.6615377470936983,
      "grad_norm": 0.6171875,
      "learning_rate": 2.095023150238468e-05,
      "loss": 0.3895021200180054,
      "num_input_tokens_seen": 13354214464,
      "step": 46960,
      "train_runtime": 456580.4923,
      "train_tokens_per_second": 29248.325
    },
    {
      "epoch": 1.661891567357427,
      "grad_norm": 0.64453125,
      "learning_rate": 2.094839268206548e-05,
      "loss": 0.39182310104370116,
      "num_input_tokens_seen": 13357048896,
      "step": 46970,
      "train_runtime": 456675.9918,
      "train_tokens_per_second": 29248.415
    },
    {
      "epoch": 1.6622453876211558,
      "grad_norm": 0.6171875,
      "learning_rate": 2.094655434584595e-05,
      "loss": 0.3866722583770752,
      "num_input_tokens_seen": 13359905216,
      "step": 46980,
      "train_runtime": 456772.4087,
      "train_tokens_per_second": 29248.494
    },
    {
      "epoch": 1.6625992078848846,
      "grad_norm": 0.64453125,
      "learning_rate": 2.0944716493513714e-05,
      "loss": 0.38572053909301757,
      "num_input_tokens_seen": 13362752320,
      "step": 46990,
      "train_runtime": 456872.7681,
      "train_tokens_per_second": 29248.301
    },
    {
      "epoch": 1.6629530281486133,
      "grad_norm": 0.625,
      "learning_rate": 2.0942879124856534e-05,
      "loss": 0.3886034727096558,
      "num_input_tokens_seen": 13365596352,
      "step": 47000,
      "train_runtime": 456970.1335,
      "train_tokens_per_second": 29248.293
    },
    {
      "epoch": 1.663306848412342,
      "grad_norm": 0.609375,
      "learning_rate": 2.094104223966229e-05,
      "loss": 0.39440786838531494,
      "num_input_tokens_seen": 13368435264,
      "step": 47010,
      "train_runtime": 457070.1147,
      "train_tokens_per_second": 29248.106
    },
    {
      "epoch": 1.6636606686760709,
      "grad_norm": 0.59375,
      "learning_rate": 2.0939205837719e-05,
      "loss": 0.3892618179321289,
      "num_input_tokens_seen": 13371336128,
      "step": 47020,
      "train_runtime": 457170.7487,
      "train_tokens_per_second": 29248.013
    },
    {
      "epoch": 1.6640144889397996,
      "grad_norm": 0.609375,
      "learning_rate": 2.093736991881481e-05,
      "loss": 0.3866738796234131,
      "num_input_tokens_seen": 13374194624,
      "step": 47030,
      "train_runtime": 457268.1129,
      "train_tokens_per_second": 29248.037
    },
    {
      "epoch": 1.6643683092035286,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0935534482737994e-05,
      "loss": 0.3849113941192627,
      "num_input_tokens_seen": 13377049408,
      "step": 47040,
      "train_runtime": 457367.2466,
      "train_tokens_per_second": 29247.939
    },
    {
      "epoch": 1.6647221294672572,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0933699529276963e-05,
      "loss": 0.3889256238937378,
      "num_input_tokens_seen": 13379879360,
      "step": 47050,
      "train_runtime": 457466.1342,
      "train_tokens_per_second": 29247.803
    },
    {
      "epoch": 1.6650759497309862,
      "grad_norm": 0.60546875,
      "learning_rate": 2.093186505822025e-05,
      "loss": 0.38730587959289553,
      "num_input_tokens_seen": 13382698880,
      "step": 47060,
      "train_runtime": 457560.5806,
      "train_tokens_per_second": 29247.928
    },
    {
      "epoch": 1.6654297699947147,
      "grad_norm": 0.61328125,
      "learning_rate": 2.093003106935652e-05,
      "loss": 0.3923827648162842,
      "num_input_tokens_seen": 13385529024,
      "step": 47070,
      "train_runtime": 457654.9817,
      "train_tokens_per_second": 29248.079
    },
    {
      "epoch": 1.6657835902584437,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0928197562474576e-05,
      "loss": 0.39537787437438965,
      "num_input_tokens_seen": 13388344192,
      "step": 47080,
      "train_runtime": 457750.8909,
      "train_tokens_per_second": 29248.101
    },
    {
      "epoch": 1.6661374105221722,
      "grad_norm": 0.625,
      "learning_rate": 2.0926364537363328e-05,
      "loss": 0.3837417125701904,
      "num_input_tokens_seen": 13391186560,
      "step": 47090,
      "train_runtime": 457846.9027,
      "train_tokens_per_second": 29248.175
    },
    {
      "epoch": 1.6664912307859012,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0924531993811842e-05,
      "loss": 0.39069561958312987,
      "num_input_tokens_seen": 13393996416,
      "step": 47100,
      "train_runtime": 457940.1759,
      "train_tokens_per_second": 29248.354
    },
    {
      "epoch": 1.6668450510496298,
      "grad_norm": 0.59375,
      "learning_rate": 2.0922699931609298e-05,
      "loss": 0.390627384185791,
      "num_input_tokens_seen": 13396853760,
      "step": 47110,
      "train_runtime": 458039.7808,
      "train_tokens_per_second": 29248.232
    },
    {
      "epoch": 1.6671988713133588,
      "grad_norm": 0.61328125,
      "learning_rate": 2.092086835054501e-05,
      "loss": 0.3899845123291016,
      "num_input_tokens_seen": 13399712064,
      "step": 47120,
      "train_runtime": 458141.0149,
      "train_tokens_per_second": 29248.008
    },
    {
      "epoch": 1.6675526915770875,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0919037250408417e-05,
      "loss": 0.39252281188964844,
      "num_input_tokens_seen": 13402605056,
      "step": 47130,
      "train_runtime": 458243.2845,
      "train_tokens_per_second": 29247.794
    },
    {
      "epoch": 1.6679065118408163,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0917206630989093e-05,
      "loss": 0.3857570171356201,
      "num_input_tokens_seen": 13405434112,
      "step": 47140,
      "train_runtime": 458338.0847,
      "train_tokens_per_second": 29247.917
    },
    {
      "epoch": 1.668260332104545,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0915376492076734e-05,
      "loss": 0.3930667400360107,
      "num_input_tokens_seen": 13408317248,
      "step": 47150,
      "train_runtime": 458437.5575,
      "train_tokens_per_second": 29247.859
    },
    {
      "epoch": 1.6686141523682738,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0913546833461167e-05,
      "loss": 0.38636794090271,
      "num_input_tokens_seen": 13411214528,
      "step": 47160,
      "train_runtime": 458536.2124,
      "train_tokens_per_second": 29247.885
    },
    {
      "epoch": 1.6689679726320026,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0911717654932358e-05,
      "loss": 0.39260072708129884,
      "num_input_tokens_seen": 13414142848,
      "step": 47170,
      "train_runtime": 458637.8854,
      "train_tokens_per_second": 29247.786
    },
    {
      "epoch": 1.6693217928957313,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0909888956280382e-05,
      "loss": 0.3901290655136108,
      "num_input_tokens_seen": 13416974656,
      "step": 47180,
      "train_runtime": 458736.3746,
      "train_tokens_per_second": 29247.68
    },
    {
      "epoch": 1.6696756131594601,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0908060737295463e-05,
      "loss": 0.388273024559021,
      "num_input_tokens_seen": 13419807360,
      "step": 47190,
      "train_runtime": 458834.9417,
      "train_tokens_per_second": 29247.571
    },
    {
      "epoch": 1.6700294334231889,
      "grad_norm": 0.6328125,
      "learning_rate": 2.0906232997767933e-05,
      "loss": 0.3891755104064941,
      "num_input_tokens_seen": 13422652160,
      "step": 47200,
      "train_runtime": 458933.3821,
      "train_tokens_per_second": 29247.496
    },
    {
      "epoch": 1.6703832536869179,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0904405737488273e-05,
      "loss": 0.3939460515975952,
      "num_input_tokens_seen": 13425501952,
      "step": 47210,
      "train_runtime": 459031.8822,
      "train_tokens_per_second": 29247.428
    },
    {
      "epoch": 1.6707370739506464,
      "grad_norm": 0.59375,
      "learning_rate": 2.0902578956247073e-05,
      "loss": 0.3910306453704834,
      "num_input_tokens_seen": 13428351808,
      "step": 47220,
      "train_runtime": 459127.969,
      "train_tokens_per_second": 29247.514
    },
    {
      "epoch": 1.6710908942143754,
      "grad_norm": 0.6328125,
      "learning_rate": 2.090075265383507e-05,
      "loss": 0.38813965320587157,
      "num_input_tokens_seen": 13431173376,
      "step": 47230,
      "train_runtime": 459222.3928,
      "train_tokens_per_second": 29247.645
    },
    {
      "epoch": 1.671444714478104,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0898926830043105e-05,
      "loss": 0.38821945190429685,
      "num_input_tokens_seen": 13434019200,
      "step": 47240,
      "train_runtime": 459322.2859,
      "train_tokens_per_second": 29247.48
    },
    {
      "epoch": 1.671798534741833,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0897101484662172e-05,
      "loss": 0.38429558277130127,
      "num_input_tokens_seen": 13436852224,
      "step": 47250,
      "train_runtime": 459418.4656,
      "train_tokens_per_second": 29247.523
    },
    {
      "epoch": 1.6721523550055615,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0895276617483384e-05,
      "loss": 0.3929542779922485,
      "num_input_tokens_seen": 13439749248,
      "step": 47260,
      "train_runtime": 459520.6449,
      "train_tokens_per_second": 29247.324
    },
    {
      "epoch": 1.6725061752692905,
      "grad_norm": 0.640625,
      "learning_rate": 2.089345222829797e-05,
      "loss": 0.3935800552368164,
      "num_input_tokens_seen": 13442612416,
      "step": 47270,
      "train_runtime": 459623.1406,
      "train_tokens_per_second": 29247.031
    },
    {
      "epoch": 1.672859995533019,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0891628316897297e-05,
      "loss": 0.3896228075027466,
      "num_input_tokens_seen": 13445467136,
      "step": 47280,
      "train_runtime": 459720.5324,
      "train_tokens_per_second": 29247.045
    },
    {
      "epoch": 1.673213815796748,
      "grad_norm": 0.6328125,
      "learning_rate": 2.0889804883072863e-05,
      "loss": 0.38893582820892336,
      "num_input_tokens_seen": 13448285184,
      "step": 47290,
      "train_runtime": 459813.4449,
      "train_tokens_per_second": 29247.264
    },
    {
      "epoch": 1.6735676360604768,
      "grad_norm": 0.6171875,
      "learning_rate": 2.088798192661629e-05,
      "loss": 0.3861669063568115,
      "num_input_tokens_seen": 13451164352,
      "step": 47300,
      "train_runtime": 459914.6584,
      "train_tokens_per_second": 29247.088
    },
    {
      "epoch": 1.6739214563242055,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0886159447319314e-05,
      "loss": 0.3862429618835449,
      "num_input_tokens_seen": 13453996416,
      "step": 47310,
      "train_runtime": 460007.9611,
      "train_tokens_per_second": 29247.312
    },
    {
      "epoch": 1.6742752765879343,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0884337444973824e-05,
      "loss": 0.3916969060897827,
      "num_input_tokens_seen": 13456834240,
      "step": 47320,
      "train_runtime": 460107.4836,
      "train_tokens_per_second": 29247.154
    },
    {
      "epoch": 1.674629096851663,
      "grad_norm": 0.60546875,
      "learning_rate": 2.088251591937182e-05,
      "loss": 0.3888957977294922,
      "num_input_tokens_seen": 13459677824,
      "step": 47330,
      "train_runtime": 460202.2009,
      "train_tokens_per_second": 29247.313
    },
    {
      "epoch": 1.6749829171153918,
      "grad_norm": 0.59375,
      "learning_rate": 2.0880694870305427e-05,
      "loss": 0.3945812702178955,
      "num_input_tokens_seen": 13462523200,
      "step": 47340,
      "train_runtime": 460297.2586,
      "train_tokens_per_second": 29247.455
    },
    {
      "epoch": 1.6753367373791206,
      "grad_norm": 0.61328125,
      "learning_rate": 2.08788742975669e-05,
      "loss": 0.3929840087890625,
      "num_input_tokens_seen": 13465363840,
      "step": 47350,
      "train_runtime": 460393.4131,
      "train_tokens_per_second": 29247.516
    },
    {
      "epoch": 1.6756905576428494,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0877054200948624e-05,
      "loss": 0.38920021057128906,
      "num_input_tokens_seen": 13468191488,
      "step": 47360,
      "train_runtime": 460488.8508,
      "train_tokens_per_second": 29247.595
    },
    {
      "epoch": 1.6760443779065781,
      "grad_norm": 0.60546875,
      "learning_rate": 2.08752345802431e-05,
      "loss": 0.3889219522476196,
      "num_input_tokens_seen": 13471042048,
      "step": 47370,
      "train_runtime": 460586.2864,
      "train_tokens_per_second": 29247.597
    },
    {
      "epoch": 1.676398198170307,
      "grad_norm": 0.58203125,
      "learning_rate": 2.0873415435242988e-05,
      "loss": 0.3904151201248169,
      "num_input_tokens_seen": 13473888448,
      "step": 47380,
      "train_runtime": 460683.149,
      "train_tokens_per_second": 29247.626
    },
    {
      "epoch": 1.6767520184340357,
      "grad_norm": 0.60546875,
      "learning_rate": 2.087159676574102e-05,
      "loss": 0.39035844802856445,
      "num_input_tokens_seen": 13476737984,
      "step": 47390,
      "train_runtime": 460780.1048,
      "train_tokens_per_second": 29247.656
    },
    {
      "epoch": 1.6771058386977646,
      "grad_norm": 0.6171875,
      "learning_rate": 2.086977857153011e-05,
      "loss": 0.388238525390625,
      "num_input_tokens_seen": 13479566400,
      "step": 47400,
      "train_runtime": 460876.8322,
      "train_tokens_per_second": 29247.655
    },
    {
      "epoch": 1.6774596589614932,
      "grad_norm": 0.609375,
      "learning_rate": 2.086796085240326e-05,
      "loss": 0.38954696655273435,
      "num_input_tokens_seen": 13482357696,
      "step": 47410,
      "train_runtime": 460972.0755,
      "train_tokens_per_second": 29247.667
    },
    {
      "epoch": 1.6778134792252222,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0866143608153614e-05,
      "loss": 0.387572717666626,
      "num_input_tokens_seen": 13485198336,
      "step": 47420,
      "train_runtime": 461070.4873,
      "train_tokens_per_second": 29247.585
    },
    {
      "epoch": 1.6781672994889507,
      "grad_norm": 0.625,
      "learning_rate": 2.086432683857444e-05,
      "loss": 0.39342925548553465,
      "num_input_tokens_seen": 13488037952,
      "step": 47430,
      "train_runtime": 461167.6907,
      "train_tokens_per_second": 29247.578
    },
    {
      "epoch": 1.6785211197526797,
      "grad_norm": 0.625,
      "learning_rate": 2.0862510543459127e-05,
      "loss": 0.389902925491333,
      "num_input_tokens_seen": 13490893952,
      "step": 47440,
      "train_runtime": 461263.9646,
      "train_tokens_per_second": 29247.665
    },
    {
      "epoch": 1.6788749400164085,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0860694722601197e-05,
      "loss": 0.38869054317474366,
      "num_input_tokens_seen": 13493688384,
      "step": 47450,
      "train_runtime": 461358.8109,
      "train_tokens_per_second": 29247.709
    },
    {
      "epoch": 1.6792287602801372,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0858879375794306e-05,
      "loss": 0.38864936828613283,
      "num_input_tokens_seen": 13496522752,
      "step": 47460,
      "train_runtime": 461455.8821,
      "train_tokens_per_second": 29247.699
    },
    {
      "epoch": 1.679582580543866,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0857064502832207e-05,
      "loss": 0.392928409576416,
      "num_input_tokens_seen": 13499400384,
      "step": 47470,
      "train_runtime": 461557.7966,
      "train_tokens_per_second": 29247.476
    },
    {
      "epoch": 1.6799364008075948,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0855250103508805e-05,
      "loss": 0.38923797607421873,
      "num_input_tokens_seen": 13502198976,
      "step": 47480,
      "train_runtime": 461651.2542,
      "train_tokens_per_second": 29247.617
    },
    {
      "epoch": 1.6802902210713235,
      "grad_norm": 0.59375,
      "learning_rate": 2.085343617761812e-05,
      "loss": 0.38673949241638184,
      "num_input_tokens_seen": 13505095424,
      "step": 47490,
      "train_runtime": 461752.7296,
      "train_tokens_per_second": 29247.462
    },
    {
      "epoch": 1.6806440413350523,
      "grad_norm": 0.609375,
      "learning_rate": 2.08516227249543e-05,
      "loss": 0.3887539625167847,
      "num_input_tokens_seen": 13507979072,
      "step": 47500,
      "train_runtime": 461851.8082,
      "train_tokens_per_second": 29247.431
    },
    {
      "epoch": 1.680997861598781,
      "grad_norm": 0.61328125,
      "learning_rate": 2.084980974531162e-05,
      "loss": 0.39057395458221433,
      "num_input_tokens_seen": 13510811264,
      "step": 47510,
      "train_runtime": 461950.3962,
      "train_tokens_per_second": 29247.32
    },
    {
      "epoch": 1.6813516818625098,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0847997238484477e-05,
      "loss": 0.3878222227096558,
      "num_input_tokens_seen": 13513655808,
      "step": 47520,
      "train_runtime": 462044.0756,
      "train_tokens_per_second": 29247.547
    },
    {
      "epoch": 1.6817055021262386,
      "grad_norm": 0.58984375,
      "learning_rate": 2.084618520426739e-05,
      "loss": 0.3899877071380615,
      "num_input_tokens_seen": 13516457088,
      "step": 47530,
      "train_runtime": 462139.674,
      "train_tokens_per_second": 29247.558
    },
    {
      "epoch": 1.6820593223899674,
      "grad_norm": 0.6328125,
      "learning_rate": 2.0844373642455005e-05,
      "loss": 0.38895201683044434,
      "num_input_tokens_seen": 13519269760,
      "step": 47540,
      "train_runtime": 462238.1319,
      "train_tokens_per_second": 29247.413
    },
    {
      "epoch": 1.6824131426536963,
      "grad_norm": 0.58203125,
      "learning_rate": 2.0842562552842103e-05,
      "loss": 0.3893259525299072,
      "num_input_tokens_seen": 13522149376,
      "step": 47550,
      "train_runtime": 462335.3976,
      "train_tokens_per_second": 29247.489
    },
    {
      "epoch": 1.682766962917425,
      "grad_norm": 0.66015625,
      "learning_rate": 2.0840751935223573e-05,
      "loss": 0.3897470235824585,
      "num_input_tokens_seen": 13525025600,
      "step": 47560,
      "train_runtime": 462432.3599,
      "train_tokens_per_second": 29247.576
    },
    {
      "epoch": 1.6831207831811539,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0838941789394442e-05,
      "loss": 0.3892627000808716,
      "num_input_tokens_seen": 13527870528,
      "step": 47570,
      "train_runtime": 462528.3003,
      "train_tokens_per_second": 29247.66
    },
    {
      "epoch": 1.6834746034448824,
      "grad_norm": 0.63671875,
      "learning_rate": 2.083713211514986e-05,
      "loss": 0.3881516933441162,
      "num_input_tokens_seen": 13530731712,
      "step": 47580,
      "train_runtime": 462628.8423,
      "train_tokens_per_second": 29247.488
    },
    {
      "epoch": 1.6838284237086114,
      "grad_norm": 0.61328125,
      "learning_rate": 2.083532291228509e-05,
      "loss": 0.3877270698547363,
      "num_input_tokens_seen": 13533541440,
      "step": 47590,
      "train_runtime": 462723.4868,
      "train_tokens_per_second": 29247.578
    },
    {
      "epoch": 1.68418224397234,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0833514180595527e-05,
      "loss": 0.38462152481079104,
      "num_input_tokens_seen": 13536369920,
      "step": 47600,
      "train_runtime": 462821.7333,
      "train_tokens_per_second": 29247.481
    },
    {
      "epoch": 1.684536064236069,
      "grad_norm": 0.61328125,
      "learning_rate": 2.08317059198767e-05,
      "loss": 0.38976967334747314,
      "num_input_tokens_seen": 13539229248,
      "step": 47610,
      "train_runtime": 462919.077,
      "train_tokens_per_second": 29247.508
    },
    {
      "epoch": 1.6848898844997977,
      "grad_norm": 0.5859375,
      "learning_rate": 2.0829898129924246e-05,
      "loss": 0.38527483940124513,
      "num_input_tokens_seen": 13542116416,
      "step": 47620,
      "train_runtime": 463019.6292,
      "train_tokens_per_second": 29247.392
    },
    {
      "epoch": 1.6852437047635265,
      "grad_norm": 0.609375,
      "learning_rate": 2.0828090810533936e-05,
      "loss": 0.3903650760650635,
      "num_input_tokens_seen": 13544968064,
      "step": 47630,
      "train_runtime": 463118.1215,
      "train_tokens_per_second": 29247.329
    },
    {
      "epoch": 1.6855975250272552,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0826283961501665e-05,
      "loss": 0.38622426986694336,
      "num_input_tokens_seen": 13547780288,
      "step": 47640,
      "train_runtime": 463214.1933,
      "train_tokens_per_second": 29247.334
    },
    {
      "epoch": 1.685951345290984,
      "grad_norm": 0.59375,
      "learning_rate": 2.0824477582623445e-05,
      "loss": 0.3926991939544678,
      "num_input_tokens_seen": 13550642496,
      "step": 47650,
      "train_runtime": 463311.4677,
      "train_tokens_per_second": 29247.371
    },
    {
      "epoch": 1.6863051655547128,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0822671673695414e-05,
      "loss": 0.3876538038253784,
      "num_input_tokens_seen": 13553570048,
      "step": 47660,
      "train_runtime": 463411.6836,
      "train_tokens_per_second": 29247.364
    },
    {
      "epoch": 1.6866589858184415,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0820866234513844e-05,
      "loss": 0.39308881759643555,
      "num_input_tokens_seen": 13556407616,
      "step": 47670,
      "train_runtime": 463510.247,
      "train_tokens_per_second": 29247.266
    },
    {
      "epoch": 1.6870128060821703,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0819061264875116e-05,
      "loss": 0.39126613140106203,
      "num_input_tokens_seen": 13559210688,
      "step": 47680,
      "train_runtime": 463602.5828,
      "train_tokens_per_second": 29247.487
    },
    {
      "epoch": 1.687366626345899,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0817256764575738e-05,
      "loss": 0.39210147857666017,
      "num_input_tokens_seen": 13561989696,
      "step": 47690,
      "train_runtime": 463697.3214,
      "train_tokens_per_second": 29247.505
    },
    {
      "epoch": 1.687720446609628,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0815452733412354e-05,
      "loss": 0.3905423641204834,
      "num_input_tokens_seen": 13564833792,
      "step": 47700,
      "train_runtime": 463792.5491,
      "train_tokens_per_second": 29247.632
    },
    {
      "epoch": 1.6880742668733566,
      "grad_norm": 0.59765625,
      "learning_rate": 2.081364917118171e-05,
      "loss": 0.3887256383895874,
      "num_input_tokens_seen": 13567649856,
      "step": 47710,
      "train_runtime": 463886.8042,
      "train_tokens_per_second": 29247.76
    },
    {
      "epoch": 1.6884280871370856,
      "grad_norm": 0.60546875,
      "learning_rate": 2.08118460776807e-05,
      "loss": 0.38867313861846925,
      "num_input_tokens_seen": 13570492544,
      "step": 47720,
      "train_runtime": 463982.3738,
      "train_tokens_per_second": 29247.862
    },
    {
      "epoch": 1.6887819074008141,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0810043452706317e-05,
      "loss": 0.39161770343780516,
      "num_input_tokens_seen": 13573312896,
      "step": 47730,
      "train_runtime": 464077.5377,
      "train_tokens_per_second": 29247.942
    },
    {
      "epoch": 1.6891357276645431,
      "grad_norm": 0.578125,
      "learning_rate": 2.08082412960557e-05,
      "loss": 0.3880011558532715,
      "num_input_tokens_seen": 13576127424,
      "step": 47740,
      "train_runtime": 464171.6178,
      "train_tokens_per_second": 29248.077
    },
    {
      "epoch": 1.6894895479282717,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0806439607526085e-05,
      "loss": 0.39341139793395996,
      "num_input_tokens_seen": 13578962368,
      "step": 47750,
      "train_runtime": 464266.3035,
      "train_tokens_per_second": 29248.219
    },
    {
      "epoch": 1.6898433681920006,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0804638386914853e-05,
      "loss": 0.39020657539367676,
      "num_input_tokens_seen": 13581805696,
      "step": 47760,
      "train_runtime": 464363.7679,
      "train_tokens_per_second": 29248.203
    },
    {
      "epoch": 1.6901971884557292,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0802837634019497e-05,
      "loss": 0.39258518218994143,
      "num_input_tokens_seen": 13584596864,
      "step": 47770,
      "train_runtime": 464458.5562,
      "train_tokens_per_second": 29248.243
    },
    {
      "epoch": 1.6905510087194582,
      "grad_norm": 0.61328125,
      "learning_rate": 2.080103734863764e-05,
      "loss": 0.38879125118255614,
      "num_input_tokens_seen": 13587456064,
      "step": 47780,
      "train_runtime": 464557.3039,
      "train_tokens_per_second": 29248.181
    },
    {
      "epoch": 1.690904828983187,
      "grad_norm": 0.609375,
      "learning_rate": 2.0799237530567022e-05,
      "loss": 0.39292140007019044,
      "num_input_tokens_seen": 13590344512,
      "step": 47790,
      "train_runtime": 464654.547,
      "train_tokens_per_second": 29248.276
    },
    {
      "epoch": 1.6912586492469157,
      "grad_norm": 0.6015625,
      "learning_rate": 2.079743817960551e-05,
      "loss": 0.3895714282989502,
      "num_input_tokens_seen": 13593122368,
      "step": 47800,
      "train_runtime": 464750.6976,
      "train_tokens_per_second": 29248.202
    },
    {
      "epoch": 1.6916124695106445,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0795639295551074e-05,
      "loss": 0.3920772552490234,
      "num_input_tokens_seen": 13596015680,
      "step": 47810,
      "train_runtime": 464849.825,
      "train_tokens_per_second": 29248.189
    },
    {
      "epoch": 1.6919662897743732,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0793840878201835e-05,
      "loss": 0.3850001335144043,
      "num_input_tokens_seen": 13598832384,
      "step": 47820,
      "train_runtime": 464944.5073,
      "train_tokens_per_second": 29248.291
    },
    {
      "epoch": 1.692320110038102,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0792042927356023e-05,
      "loss": 0.3832183837890625,
      "num_input_tokens_seen": 13601646720,
      "step": 47830,
      "train_runtime": 465040.5788,
      "train_tokens_per_second": 29248.301
    },
    {
      "epoch": 1.6926739303018308,
      "grad_norm": 0.609375,
      "learning_rate": 2.079024544281199e-05,
      "loss": 0.39319667816162107,
      "num_input_tokens_seen": 13604438848,
      "step": 47840,
      "train_runtime": 465135.6664,
      "train_tokens_per_second": 29248.324
    },
    {
      "epoch": 1.6930277505655595,
      "grad_norm": 0.625,
      "learning_rate": 2.078844842436821e-05,
      "loss": 0.3922778606414795,
      "num_input_tokens_seen": 13607228928,
      "step": 47850,
      "train_runtime": 465229.8747,
      "train_tokens_per_second": 29248.399
    },
    {
      "epoch": 1.6933815708292883,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0786651871823272e-05,
      "loss": 0.38624837398529055,
      "num_input_tokens_seen": 13610089280,
      "step": 47860,
      "train_runtime": 465329.7429,
      "train_tokens_per_second": 29248.269
    },
    {
      "epoch": 1.6937353910930173,
      "grad_norm": 0.640625,
      "learning_rate": 2.0784855784975907e-05,
      "loss": 0.3850614070892334,
      "num_input_tokens_seen": 13612907328,
      "step": 47870,
      "train_runtime": 465424.1986,
      "train_tokens_per_second": 29248.388
    },
    {
      "epoch": 1.6940892113567458,
      "grad_norm": 0.59375,
      "learning_rate": 2.078306016362495e-05,
      "loss": 0.3858610153198242,
      "num_input_tokens_seen": 13615737856,
      "step": 47880,
      "train_runtime": 465520.9037,
      "train_tokens_per_second": 29248.392
    },
    {
      "epoch": 1.6944430316204748,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0781265007569355e-05,
      "loss": 0.3895831823348999,
      "num_input_tokens_seen": 13618588352,
      "step": 47890,
      "train_runtime": 465617.6169,
      "train_tokens_per_second": 29248.439
    },
    {
      "epoch": 1.6947968518842034,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0779470316608218e-05,
      "loss": 0.3867624521255493,
      "num_input_tokens_seen": 13621466304,
      "step": 47900,
      "train_runtime": 465715.334,
      "train_tokens_per_second": 29248.481
    },
    {
      "epoch": 1.6951506721479324,
      "grad_norm": 0.59375,
      "learning_rate": 2.0777676090540738e-05,
      "loss": 0.3871964931488037,
      "num_input_tokens_seen": 13624274496,
      "step": 47910,
      "train_runtime": 465809.163,
      "train_tokens_per_second": 29248.618
    },
    {
      "epoch": 1.695504492411661,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0775882329166235e-05,
      "loss": 0.38729941844940186,
      "num_input_tokens_seen": 13627144192,
      "step": 47920,
      "train_runtime": 465904.7439,
      "train_tokens_per_second": 29248.778
    },
    {
      "epoch": 1.69585831267539,
      "grad_norm": 0.625,
      "learning_rate": 2.0774089032284164e-05,
      "loss": 0.39241800308227537,
      "num_input_tokens_seen": 13629971648,
      "step": 47930,
      "train_runtime": 465997.9116,
      "train_tokens_per_second": 29248.997
    },
    {
      "epoch": 1.6962121329391184,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0772296199694092e-05,
      "loss": 0.3910749197006226,
      "num_input_tokens_seen": 13632793792,
      "step": 47940,
      "train_runtime": 466093.9172,
      "train_tokens_per_second": 29249.027
    },
    {
      "epoch": 1.6965659532028474,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0770503831195703e-05,
      "loss": 0.3873137474060059,
      "num_input_tokens_seen": 13635600512,
      "step": 47950,
      "train_runtime": 466189.6038,
      "train_tokens_per_second": 29249.045
    },
    {
      "epoch": 1.6969197734665762,
      "grad_norm": 0.625,
      "learning_rate": 2.0768711926588813e-05,
      "loss": 0.3928596258163452,
      "num_input_tokens_seen": 13638457728,
      "step": 47960,
      "train_runtime": 466289.8642,
      "train_tokens_per_second": 29248.883
    },
    {
      "epoch": 1.697273593730305,
      "grad_norm": 0.58984375,
      "learning_rate": 2.076692048567335e-05,
      "loss": 0.38697829246521,
      "num_input_tokens_seen": 13641345664,
      "step": 47970,
      "train_runtime": 466386.8521,
      "train_tokens_per_second": 29248.993
    },
    {
      "epoch": 1.6976274139940337,
      "grad_norm": 0.59765625,
      "learning_rate": 2.076512950824937e-05,
      "loss": 0.3890304327011108,
      "num_input_tokens_seen": 13644170176,
      "step": 47980,
      "train_runtime": 466481.2059,
      "train_tokens_per_second": 29249.132
    },
    {
      "epoch": 1.6979812342577625,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0763338994117043e-05,
      "loss": 0.3881545066833496,
      "num_input_tokens_seen": 13647034688,
      "step": 47990,
      "train_runtime": 466577.0557,
      "train_tokens_per_second": 29249.262
    },
    {
      "epoch": 1.6983350545214912,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0761548943076657e-05,
      "loss": 0.38659062385559084,
      "num_input_tokens_seen": 13649907456,
      "step": 48000,
      "train_runtime": 466678.2925,
      "train_tokens_per_second": 29249.073
    },
    {
      "epoch": 1.69868887478522,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0759759354928637e-05,
      "loss": 0.3865540027618408,
      "num_input_tokens_seen": 13652747200,
      "step": 48010,
      "train_runtime": 466776.2689,
      "train_tokens_per_second": 29249.017
    },
    {
      "epoch": 1.6990426950489488,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0757970229473512e-05,
      "loss": 0.39294488430023194,
      "num_input_tokens_seen": 13655570176,
      "step": 48020,
      "train_runtime": 466873.1559,
      "train_tokens_per_second": 29248.994
    },
    {
      "epoch": 1.6993965153126775,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0756181566511933e-05,
      "loss": 0.3923866987228394,
      "num_input_tokens_seen": 13658403008,
      "step": 48030,
      "train_runtime": 466968.8544,
      "train_tokens_per_second": 29249.066
    },
    {
      "epoch": 1.6997503355764065,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0754393365844677e-05,
      "loss": 0.3864671945571899,
      "num_input_tokens_seen": 13661246656,
      "step": 48040,
      "train_runtime": 467066.846,
      "train_tokens_per_second": 29249.018
    },
    {
      "epoch": 1.700104155840135,
      "grad_norm": 0.62890625,
      "learning_rate": 2.075260562727264e-05,
      "loss": 0.3852688789367676,
      "num_input_tokens_seen": 13664068160,
      "step": 48050,
      "train_runtime": 467162.0993,
      "train_tokens_per_second": 29249.094
    },
    {
      "epoch": 1.700457976103864,
      "grad_norm": 0.625,
      "learning_rate": 2.0750818350596838e-05,
      "loss": 0.3894263982772827,
      "num_input_tokens_seen": 13666918592,
      "step": 48060,
      "train_runtime": 467261.8188,
      "train_tokens_per_second": 29248.952
    },
    {
      "epoch": 1.7008117963675926,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0749031535618403e-05,
      "loss": 0.38974287509918215,
      "num_input_tokens_seen": 13669768384,
      "step": 48070,
      "train_runtime": 467356.0299,
      "train_tokens_per_second": 29249.154
    },
    {
      "epoch": 1.7011656166313216,
      "grad_norm": 0.5859375,
      "learning_rate": 2.074724518213859e-05,
      "loss": 0.3871318817138672,
      "num_input_tokens_seen": 13672655936,
      "step": 48080,
      "train_runtime": 467457.3216,
      "train_tokens_per_second": 29248.993
    },
    {
      "epoch": 1.7015194368950501,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0745459289958773e-05,
      "loss": 0.38607001304626465,
      "num_input_tokens_seen": 13675506176,
      "step": 48090,
      "train_runtime": 467555.6855,
      "train_tokens_per_second": 29248.936
    },
    {
      "epoch": 1.7018732571587791,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0743673858880447e-05,
      "loss": 0.3865602731704712,
      "num_input_tokens_seen": 13678392192,
      "step": 48100,
      "train_runtime": 467656.405,
      "train_tokens_per_second": 29248.808
    },
    {
      "epoch": 1.7022270774225077,
      "grad_norm": 0.625,
      "learning_rate": 2.074188888870523e-05,
      "loss": 0.3908292055130005,
      "num_input_tokens_seen": 13681223040,
      "step": 48110,
      "train_runtime": 467754.0079,
      "train_tokens_per_second": 29248.756
    },
    {
      "epoch": 1.7025808976862367,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0740104379234847e-05,
      "loss": 0.3867603302001953,
      "num_input_tokens_seen": 13684064896,
      "step": 48120,
      "train_runtime": 467851.1323,
      "train_tokens_per_second": 29248.759
    },
    {
      "epoch": 1.7029347179499654,
      "grad_norm": 0.59375,
      "learning_rate": 2.073832033027116e-05,
      "loss": 0.38911683559417726,
      "num_input_tokens_seen": 13686884928,
      "step": 48130,
      "train_runtime": 467945.4125,
      "train_tokens_per_second": 29248.892
    },
    {
      "epoch": 1.7032885382136942,
      "grad_norm": 0.609375,
      "learning_rate": 2.0736536741616128e-05,
      "loss": 0.38747320175170896,
      "num_input_tokens_seen": 13689754880,
      "step": 48140,
      "train_runtime": 468041.4136,
      "train_tokens_per_second": 29249.025
    },
    {
      "epoch": 1.703642358477423,
      "grad_norm": 0.6015625,
      "learning_rate": 2.073475361307185e-05,
      "loss": 0.38791484832763673,
      "num_input_tokens_seen": 13692595840,
      "step": 48150,
      "train_runtime": 468137.1248,
      "train_tokens_per_second": 29249.113
    },
    {
      "epoch": 1.7039961787411517,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0732970944440534e-05,
      "loss": 0.38873212337493895,
      "num_input_tokens_seen": 13695499712,
      "step": 48160,
      "train_runtime": 468235.9639,
      "train_tokens_per_second": 29249.141
    },
    {
      "epoch": 1.7043499990048805,
      "grad_norm": 0.6171875,
      "learning_rate": 2.073118873552452e-05,
      "loss": 0.3841092586517334,
      "num_input_tokens_seen": 13698357888,
      "step": 48170,
      "train_runtime": 468334.7198,
      "train_tokens_per_second": 29249.076
    },
    {
      "epoch": 1.7047038192686093,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0729406986126238e-05,
      "loss": 0.38967161178588866,
      "num_input_tokens_seen": 13701224064,
      "step": 48180,
      "train_runtime": 468433.5833,
      "train_tokens_per_second": 29249.022
    },
    {
      "epoch": 1.705057639532338,
      "grad_norm": 0.63671875,
      "learning_rate": 2.0727625696048264e-05,
      "loss": 0.39243450164794924,
      "num_input_tokens_seen": 13704029888,
      "step": 48190,
      "train_runtime": 468529.842,
      "train_tokens_per_second": 29249.001
    },
    {
      "epoch": 1.7054114597960668,
      "grad_norm": 0.59375,
      "learning_rate": 2.0725844865093287e-05,
      "loss": 0.39022054672241213,
      "num_input_tokens_seen": 13706867200,
      "step": 48200,
      "train_runtime": 468624.7872,
      "train_tokens_per_second": 29249.13
    },
    {
      "epoch": 1.7057652800597958,
      "grad_norm": 0.6015625,
      "learning_rate": 2.07240644930641e-05,
      "loss": 0.38955373764038087,
      "num_input_tokens_seen": 13709739264,
      "step": 48210,
      "train_runtime": 468723.729,
      "train_tokens_per_second": 29249.083
    },
    {
      "epoch": 1.7061191003235243,
      "grad_norm": 0.609375,
      "learning_rate": 2.0722284579763633e-05,
      "loss": 0.3858583927154541,
      "num_input_tokens_seen": 13712579456,
      "step": 48220,
      "train_runtime": 468821.056,
      "train_tokens_per_second": 29249.069
    },
    {
      "epoch": 1.7064729205872533,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0720505124994934e-05,
      "loss": 0.389461612701416,
      "num_input_tokens_seen": 13715411648,
      "step": 48230,
      "train_runtime": 468917.6904,
      "train_tokens_per_second": 29249.081
    },
    {
      "epoch": 1.7068267408509818,
      "grad_norm": 0.609375,
      "learning_rate": 2.0718726128561156e-05,
      "loss": 0.3888113021850586,
      "num_input_tokens_seen": 13718270848,
      "step": 48240,
      "train_runtime": 469013.8477,
      "train_tokens_per_second": 29249.181
    },
    {
      "epoch": 1.7071805611147108,
      "grad_norm": 0.5859375,
      "learning_rate": 2.0716947590265572e-05,
      "loss": 0.39095044136047363,
      "num_input_tokens_seen": 13721091776,
      "step": 48250,
      "train_runtime": 469109.6785,
      "train_tokens_per_second": 29249.219
    },
    {
      "epoch": 1.7075343813784394,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0715169509911588e-05,
      "loss": 0.3898854970932007,
      "num_input_tokens_seen": 13723950912,
      "step": 48260,
      "train_runtime": 469207.281,
      "train_tokens_per_second": 29249.228
    },
    {
      "epoch": 1.7078882016421684,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0713391887302706e-05,
      "loss": 0.3899320125579834,
      "num_input_tokens_seen": 13726796736,
      "step": 48270,
      "train_runtime": 469305.9885,
      "train_tokens_per_second": 29249.14
    },
    {
      "epoch": 1.708242021905897,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0711614722242566e-05,
      "loss": 0.3884573459625244,
      "num_input_tokens_seen": 13729610496,
      "step": 48280,
      "train_runtime": 469402.8319,
      "train_tokens_per_second": 29249.1
    },
    {
      "epoch": 1.708595842169626,
      "grad_norm": 0.59375,
      "learning_rate": 2.0709838014534923e-05,
      "loss": 0.38820972442626955,
      "num_input_tokens_seen": 13732433536,
      "step": 48290,
      "train_runtime": 469499.4594,
      "train_tokens_per_second": 29249.093
    },
    {
      "epoch": 1.7089496624333547,
      "grad_norm": 0.609375,
      "learning_rate": 2.0708061763983633e-05,
      "loss": 0.3893574714660645,
      "num_input_tokens_seen": 13735299264,
      "step": 48300,
      "train_runtime": 469597.641,
      "train_tokens_per_second": 29249.081
    },
    {
      "epoch": 1.7093034826970834,
      "grad_norm": 0.62109375,
      "learning_rate": 2.070628597039269e-05,
      "loss": 0.38619108200073243,
      "num_input_tokens_seen": 13738148736,
      "step": 48310,
      "train_runtime": 469690.9236,
      "train_tokens_per_second": 29249.338
    },
    {
      "epoch": 1.7096573029608122,
      "grad_norm": 0.62109375,
      "learning_rate": 2.07045106335662e-05,
      "loss": 0.3881126165390015,
      "num_input_tokens_seen": 13741024576,
      "step": 48320,
      "train_runtime": 469789.7111,
      "train_tokens_per_second": 29249.309
    },
    {
      "epoch": 1.710011123224541,
      "grad_norm": 0.625,
      "learning_rate": 2.0702735753308372e-05,
      "loss": 0.3866866111755371,
      "num_input_tokens_seen": 13743871680,
      "step": 48330,
      "train_runtime": 469885.2731,
      "train_tokens_per_second": 29249.42
    },
    {
      "epoch": 1.7103649434882697,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0700961329423558e-05,
      "loss": 0.38760910034179685,
      "num_input_tokens_seen": 13746730624,
      "step": 48340,
      "train_runtime": 469983.7262,
      "train_tokens_per_second": 29249.376
    },
    {
      "epoch": 1.7107187637519985,
      "grad_norm": 0.60546875,
      "learning_rate": 2.06991873617162e-05,
      "loss": 0.390235161781311,
      "num_input_tokens_seen": 13749591872,
      "step": 48350,
      "train_runtime": 470083.279,
      "train_tokens_per_second": 29249.268
    },
    {
      "epoch": 1.7110725840157273,
      "grad_norm": 0.609375,
      "learning_rate": 2.0697413849990883e-05,
      "loss": 0.3880808115005493,
      "num_input_tokens_seen": 13752402496,
      "step": 48360,
      "train_runtime": 470178.898,
      "train_tokens_per_second": 29249.298
    },
    {
      "epoch": 1.711426404279456,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0695640794052288e-05,
      "loss": 0.38985333442687986,
      "num_input_tokens_seen": 13755270592,
      "step": 48370,
      "train_runtime": 470281.2909,
      "train_tokens_per_second": 29249.028
    },
    {
      "epoch": 1.711780224543185,
      "grad_norm": 0.625,
      "learning_rate": 2.0693868193705228e-05,
      "loss": 0.39145784378051757,
      "num_input_tokens_seen": 13758136064,
      "step": 48380,
      "train_runtime": 470381.9039,
      "train_tokens_per_second": 29248.863
    },
    {
      "epoch": 1.7121340448069136,
      "grad_norm": 0.60546875,
      "learning_rate": 2.069209604875463e-05,
      "loss": 0.386849045753479,
      "num_input_tokens_seen": 13761048000,
      "step": 48390,
      "train_runtime": 470483.2625,
      "train_tokens_per_second": 29248.751
    },
    {
      "epoch": 1.7124878650706425,
      "grad_norm": 0.58984375,
      "learning_rate": 2.069032435900553e-05,
      "loss": 0.39316654205322266,
      "num_input_tokens_seen": 13763955520,
      "step": 48400,
      "train_runtime": 470586.4994,
      "train_tokens_per_second": 29248.513
    },
    {
      "epoch": 1.712841685334371,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0688553124263086e-05,
      "loss": 0.39151811599731445,
      "num_input_tokens_seen": 13766769280,
      "step": 48410,
      "train_runtime": 470682.3274,
      "train_tokens_per_second": 29248.537
    },
    {
      "epoch": 1.7131955055981,
      "grad_norm": 0.5859375,
      "learning_rate": 2.068678234433257e-05,
      "loss": 0.3873628616333008,
      "num_input_tokens_seen": 13769620032,
      "step": 48420,
      "train_runtime": 470779.6541,
      "train_tokens_per_second": 29248.545
    },
    {
      "epoch": 1.7135493258618286,
      "grad_norm": 0.59375,
      "learning_rate": 2.0685012019019385e-05,
      "loss": 0.3883244037628174,
      "num_input_tokens_seen": 13772479168,
      "step": 48430,
      "train_runtime": 470876.1222,
      "train_tokens_per_second": 29248.625
    },
    {
      "epoch": 1.7139031461255576,
      "grad_norm": 0.59375,
      "learning_rate": 2.0683242148129027e-05,
      "loss": 0.39118180274963377,
      "num_input_tokens_seen": 13775317824,
      "step": 48440,
      "train_runtime": 470971.4519,
      "train_tokens_per_second": 29248.732
    },
    {
      "epoch": 1.7142569663892864,
      "grad_norm": 0.60546875,
      "learning_rate": 2.068147273146712e-05,
      "loss": 0.38933897018432617,
      "num_input_tokens_seen": 13778182400,
      "step": 48450,
      "train_runtime": 471069.309,
      "train_tokens_per_second": 29248.737
    },
    {
      "epoch": 1.7146107866530151,
      "grad_norm": 0.609375,
      "learning_rate": 2.067970376883942e-05,
      "loss": 0.3872515201568604,
      "num_input_tokens_seen": 13781024832,
      "step": 48460,
      "train_runtime": 471167.1516,
      "train_tokens_per_second": 29248.696
    },
    {
      "epoch": 1.714964606916744,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0677935260051766e-05,
      "loss": 0.38962535858154296,
      "num_input_tokens_seen": 13783832896,
      "step": 48470,
      "train_runtime": 471261.0837,
      "train_tokens_per_second": 29248.825
    },
    {
      "epoch": 1.7153184271804727,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0676167204910145e-05,
      "loss": 0.3942216634750366,
      "num_input_tokens_seen": 13786695424,
      "step": 48480,
      "train_runtime": 471357.0438,
      "train_tokens_per_second": 29248.943
    },
    {
      "epoch": 1.7156722474442014,
      "grad_norm": 0.61328125,
      "learning_rate": 2.067439960322063e-05,
      "loss": 0.38683304786682127,
      "num_input_tokens_seen": 13789540672,
      "step": 48490,
      "train_runtime": 471454.8238,
      "train_tokens_per_second": 29248.912
    },
    {
      "epoch": 1.7160260677079302,
      "grad_norm": 0.61328125,
      "learning_rate": 2.067263245478945e-05,
      "loss": 0.38875441551208495,
      "num_input_tokens_seen": 13792359680,
      "step": 48500,
      "train_runtime": 471551.9756,
      "train_tokens_per_second": 29248.864
    },
    {
      "epoch": 1.716379887971659,
      "grad_norm": 0.62890625,
      "learning_rate": 2.06708657594229e-05,
      "loss": 0.3962899684906006,
      "num_input_tokens_seen": 13795189312,
      "step": 48510,
      "train_runtime": 471647.285,
      "train_tokens_per_second": 29248.953
    },
    {
      "epoch": 1.7167337082353877,
      "grad_norm": 0.6171875,
      "learning_rate": 2.066909951692744e-05,
      "loss": 0.38721909523010256,
      "num_input_tokens_seen": 13798056320,
      "step": 48520,
      "train_runtime": 471746.6991,
      "train_tokens_per_second": 29248.867
    },
    {
      "epoch": 1.7170875284991165,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0667333727109607e-05,
      "loss": 0.38817930221557617,
      "num_input_tokens_seen": 13800873792,
      "step": 48530,
      "train_runtime": 471842.123,
      "train_tokens_per_second": 29248.923
    },
    {
      "epoch": 1.7174413487628453,
      "grad_norm": 0.61328125,
      "learning_rate": 2.066556838977608e-05,
      "loss": 0.3903515338897705,
      "num_input_tokens_seen": 13803753728,
      "step": 48540,
      "train_runtime": 471942.606,
      "train_tokens_per_second": 29248.798
    },
    {
      "epoch": 1.7177951690265743,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0663803504733643e-05,
      "loss": 0.38574464321136476,
      "num_input_tokens_seen": 13806596224,
      "step": 48550,
      "train_runtime": 472038.1316,
      "train_tokens_per_second": 29248.9
    },
    {
      "epoch": 1.7181489892903028,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0662039071789186e-05,
      "loss": 0.3923600196838379,
      "num_input_tokens_seen": 13809494848,
      "step": 48560,
      "train_runtime": 472139.5578,
      "train_tokens_per_second": 29248.756
    },
    {
      "epoch": 1.7185028095540318,
      "grad_norm": 0.61328125,
      "learning_rate": 2.066027509074973e-05,
      "loss": 0.3877371311187744,
      "num_input_tokens_seen": 13812316224,
      "step": 48570,
      "train_runtime": 472233.1679,
      "train_tokens_per_second": 29248.933
    },
    {
      "epoch": 1.7188566298177603,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0658511561422408e-05,
      "loss": 0.38576905727386473,
      "num_input_tokens_seen": 13815134080,
      "step": 48580,
      "train_runtime": 472329.3796,
      "train_tokens_per_second": 29248.941
    },
    {
      "epoch": 1.7192104500814893,
      "grad_norm": 0.63671875,
      "learning_rate": 2.0656748483614464e-05,
      "loss": 0.3853257656097412,
      "num_input_tokens_seen": 13818018304,
      "step": 48590,
      "train_runtime": 472428.0954,
      "train_tokens_per_second": 29248.934
    },
    {
      "epoch": 1.7195642703452179,
      "grad_norm": 0.609375,
      "learning_rate": 2.0654985857133256e-05,
      "loss": 0.38887615203857423,
      "num_input_tokens_seen": 13820798656,
      "step": 48600,
      "train_runtime": 472521.6071,
      "train_tokens_per_second": 29249.03
    },
    {
      "epoch": 1.7199180906089468,
      "grad_norm": 0.6328125,
      "learning_rate": 2.065322368178626e-05,
      "loss": 0.3885476589202881,
      "num_input_tokens_seen": 13823666880,
      "step": 48610,
      "train_runtime": 472618.1759,
      "train_tokens_per_second": 29249.122
    },
    {
      "epoch": 1.7202719108726756,
      "grad_norm": 0.578125,
      "learning_rate": 2.0651461957381073e-05,
      "loss": 0.3862586498260498,
      "num_input_tokens_seen": 13826486272,
      "step": 48620,
      "train_runtime": 472713.7131,
      "train_tokens_per_second": 29249.175
    },
    {
      "epoch": 1.7206257311364044,
      "grad_norm": 0.625,
      "learning_rate": 2.064970068372539e-05,
      "loss": 0.39878201484680176,
      "num_input_tokens_seen": 13829337600,
      "step": 48630,
      "train_runtime": 472811.3425,
      "train_tokens_per_second": 29249.166
    },
    {
      "epoch": 1.7209795514001331,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0647939860627043e-05,
      "loss": 0.3905741453170776,
      "num_input_tokens_seen": 13832139200,
      "step": 48640,
      "train_runtime": 472907.059,
      "train_tokens_per_second": 29249.17
    },
    {
      "epoch": 1.721333371663862,
      "grad_norm": 0.59375,
      "learning_rate": 2.0646179487893964e-05,
      "loss": 0.38819422721862795,
      "num_input_tokens_seen": 13835023680,
      "step": 48650,
      "train_runtime": 473007.0874,
      "train_tokens_per_second": 29249.083
    },
    {
      "epoch": 1.7216871919275907,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0644419565334194e-05,
      "loss": 0.39127411842346194,
      "num_input_tokens_seen": 13837870400,
      "step": 48660,
      "train_runtime": 473106.0407,
      "train_tokens_per_second": 29248.983
    },
    {
      "epoch": 1.7220410121913194,
      "grad_norm": 0.59375,
      "learning_rate": 2.0642660092755907e-05,
      "loss": 0.38996338844299316,
      "num_input_tokens_seen": 13840713216,
      "step": 48670,
      "train_runtime": 473202.1329,
      "train_tokens_per_second": 29249.051
    },
    {
      "epoch": 1.7223948324550482,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0640901069967375e-05,
      "loss": 0.38815195560455323,
      "num_input_tokens_seen": 13843563264,
      "step": 48680,
      "train_runtime": 473297.6681,
      "train_tokens_per_second": 29249.169
    },
    {
      "epoch": 1.722748652718777,
      "grad_norm": 0.65625,
      "learning_rate": 2.0639142496777e-05,
      "loss": 0.38807406425476076,
      "num_input_tokens_seen": 13846371072,
      "step": 48690,
      "train_runtime": 473393.6627,
      "train_tokens_per_second": 29249.169
    },
    {
      "epoch": 1.723102472982506,
      "grad_norm": 0.59375,
      "learning_rate": 2.0637384372993275e-05,
      "loss": 0.39033467769622804,
      "num_input_tokens_seen": 13849165696,
      "step": 48700,
      "train_runtime": 473486.5234,
      "train_tokens_per_second": 29249.334
    },
    {
      "epoch": 1.7234562932462345,
      "grad_norm": 0.5859375,
      "learning_rate": 2.0635626698424836e-05,
      "loss": 0.387131404876709,
      "num_input_tokens_seen": 13851974848,
      "step": 48710,
      "train_runtime": 473581.8964,
      "train_tokens_per_second": 29249.376
    },
    {
      "epoch": 1.7238101135099635,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0633869472880413e-05,
      "loss": 0.3892030715942383,
      "num_input_tokens_seen": 13854800512,
      "step": 48720,
      "train_runtime": 473675.5049,
      "train_tokens_per_second": 29249.561
    },
    {
      "epoch": 1.724163933773692,
      "grad_norm": 0.59375,
      "learning_rate": 2.0632112696168856e-05,
      "loss": 0.3873451709747314,
      "num_input_tokens_seen": 13857649536,
      "step": 48730,
      "train_runtime": 473772.3268,
      "train_tokens_per_second": 29249.597
    },
    {
      "epoch": 1.724517754037421,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0630356368099127e-05,
      "loss": 0.392289924621582,
      "num_input_tokens_seen": 13860484224,
      "step": 48740,
      "train_runtime": 473866.4495,
      "train_tokens_per_second": 29249.769
    },
    {
      "epoch": 1.7248715743011496,
      "grad_norm": 0.625,
      "learning_rate": 2.06286004884803e-05,
      "loss": 0.3891010284423828,
      "num_input_tokens_seen": 13863294144,
      "step": 48750,
      "train_runtime": 473960.3885,
      "train_tokens_per_second": 29249.9
    },
    {
      "epoch": 1.7252253945648786,
      "grad_norm": 0.6171875,
      "learning_rate": 2.062684505712157e-05,
      "loss": 0.3863518238067627,
      "num_input_tokens_seen": 13866109632,
      "step": 48760,
      "train_runtime": 474053.8263,
      "train_tokens_per_second": 29250.074
    },
    {
      "epoch": 1.725579214828607,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0625090073832243e-05,
      "loss": 0.3932070016860962,
      "num_input_tokens_seen": 13868989440,
      "step": 48770,
      "train_runtime": 474153.7189,
      "train_tokens_per_second": 29249.986
    },
    {
      "epoch": 1.725933035092336,
      "grad_norm": 0.59375,
      "learning_rate": 2.0623335538421733e-05,
      "loss": 0.3887565851211548,
      "num_input_tokens_seen": 13871900992,
      "step": 48780,
      "train_runtime": 474254.4155,
      "train_tokens_per_second": 29249.914
    },
    {
      "epoch": 1.7262868553560649,
      "grad_norm": 0.5859375,
      "learning_rate": 2.0621581450699574e-05,
      "loss": 0.38957719802856444,
      "num_input_tokens_seen": 13874715264,
      "step": 48790,
      "train_runtime": 474350.9089,
      "train_tokens_per_second": 29249.897
    },
    {
      "epoch": 1.7266406756197936,
      "grad_norm": 0.6484375,
      "learning_rate": 2.0619827810475418e-05,
      "loss": 0.38720059394836426,
      "num_input_tokens_seen": 13877607872,
      "step": 48800,
      "train_runtime": 474449.114,
      "train_tokens_per_second": 29249.939
    },
    {
      "epoch": 1.7269944958835224,
      "grad_norm": 0.63671875,
      "learning_rate": 2.0618074617559007e-05,
      "loss": 0.38745574951171874,
      "num_input_tokens_seen": 13880471104,
      "step": 48810,
      "train_runtime": 474547.5285,
      "train_tokens_per_second": 29249.907
    },
    {
      "epoch": 1.7273483161472512,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0616321871760226e-05,
      "loss": 0.3931020736694336,
      "num_input_tokens_seen": 13883314176,
      "step": 48820,
      "train_runtime": 474643.2385,
      "train_tokens_per_second": 29249.999
    },
    {
      "epoch": 1.72770213641098,
      "grad_norm": 0.59375,
      "learning_rate": 2.0614569572889056e-05,
      "loss": 0.3858001232147217,
      "num_input_tokens_seen": 13886184000,
      "step": 48830,
      "train_runtime": 474740.6214,
      "train_tokens_per_second": 29250.044
    },
    {
      "epoch": 1.7280559566747087,
      "grad_norm": 0.60546875,
      "learning_rate": 2.061281772075559e-05,
      "loss": 0.3869338512420654,
      "num_input_tokens_seen": 13888998912,
      "step": 48840,
      "train_runtime": 474836.4905,
      "train_tokens_per_second": 29250.066
    },
    {
      "epoch": 1.7284097769384374,
      "grad_norm": 0.609375,
      "learning_rate": 2.0611066315170047e-05,
      "loss": 0.3855573654174805,
      "num_input_tokens_seen": 13891844032,
      "step": 48850,
      "train_runtime": 474933.3176,
      "train_tokens_per_second": 29250.094
    },
    {
      "epoch": 1.7287635972021662,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0609315355942744e-05,
      "loss": 0.3895872592926025,
      "num_input_tokens_seen": 13894653056,
      "step": 48860,
      "train_runtime": 475027.446,
      "train_tokens_per_second": 29250.211
    },
    {
      "epoch": 1.7291174174658952,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0607564842884116e-05,
      "loss": 0.38496174812316897,
      "num_input_tokens_seen": 13897499200,
      "step": 48870,
      "train_runtime": 475124.3129,
      "train_tokens_per_second": 29250.238
    },
    {
      "epoch": 1.7294712377296237,
      "grad_norm": 0.59375,
      "learning_rate": 2.0605814775804717e-05,
      "loss": 0.3888188600540161,
      "num_input_tokens_seen": 13900315904,
      "step": 48880,
      "train_runtime": 475221.6379,
      "train_tokens_per_second": 29250.175
    },
    {
      "epoch": 1.7298250579933527,
      "grad_norm": 0.59375,
      "learning_rate": 2.0604065154515207e-05,
      "loss": 0.3855093479156494,
      "num_input_tokens_seen": 13903188096,
      "step": 48890,
      "train_runtime": 475324.7147,
      "train_tokens_per_second": 29249.874
    },
    {
      "epoch": 1.7301788782570813,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0602315978826363e-05,
      "loss": 0.38578956127166747,
      "num_input_tokens_seen": 13906031168,
      "step": 48900,
      "train_runtime": 475420.1049,
      "train_tokens_per_second": 29249.985
    },
    {
      "epoch": 1.7305326985208103,
      "grad_norm": 0.59375,
      "learning_rate": 2.0600567248549062e-05,
      "loss": 0.3835385799407959,
      "num_input_tokens_seen": 13908900416,
      "step": 48910,
      "train_runtime": 475519.2943,
      "train_tokens_per_second": 29249.918
    },
    {
      "epoch": 1.7308865187845388,
      "grad_norm": 0.609375,
      "learning_rate": 2.0598818963494312e-05,
      "loss": 0.3878213882446289,
      "num_input_tokens_seen": 13911761984,
      "step": 48920,
      "train_runtime": 475617.5046,
      "train_tokens_per_second": 29249.895
    },
    {
      "epoch": 1.7312403390482678,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0597071123473223e-05,
      "loss": 0.39088256359100343,
      "num_input_tokens_seen": 13914586048,
      "step": 48930,
      "train_runtime": 475714.3398,
      "train_tokens_per_second": 29249.877
    },
    {
      "epoch": 1.7315941593119963,
      "grad_norm": 0.58203125,
      "learning_rate": 2.0595323728297013e-05,
      "loss": 0.3886402130126953,
      "num_input_tokens_seen": 13917437120,
      "step": 48940,
      "train_runtime": 475812.2584,
      "train_tokens_per_second": 29249.85
    },
    {
      "epoch": 1.7319479795757253,
      "grad_norm": 0.62109375,
      "learning_rate": 2.059357677777702e-05,
      "loss": 0.3934603691101074,
      "num_input_tokens_seen": 13920332864,
      "step": 48950,
      "train_runtime": 475913.5294,
      "train_tokens_per_second": 29249.71
    },
    {
      "epoch": 1.732301799839454,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0591830271724687e-05,
      "loss": 0.3913571357727051,
      "num_input_tokens_seen": 13923163072,
      "step": 48960,
      "train_runtime": 476008.2784,
      "train_tokens_per_second": 29249.834
    },
    {
      "epoch": 1.7326556201031829,
      "grad_norm": 0.5859375,
      "learning_rate": 2.0590084209951588e-05,
      "loss": 0.3880268096923828,
      "num_input_tokens_seen": 13926025600,
      "step": 48970,
      "train_runtime": 476104.2661,
      "train_tokens_per_second": 29249.949
    },
    {
      "epoch": 1.7330094403669116,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0588338592269375e-05,
      "loss": 0.3864912986755371,
      "num_input_tokens_seen": 13928862016,
      "step": 48980,
      "train_runtime": 476200.6625,
      "train_tokens_per_second": 29249.985
    },
    {
      "epoch": 1.7333632606306404,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0586593418489846e-05,
      "loss": 0.3858344554901123,
      "num_input_tokens_seen": 13931706304,
      "step": 48990,
      "train_runtime": 476296.892,
      "train_tokens_per_second": 29250.047
    },
    {
      "epoch": 1.7337170808943692,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0584848688424894e-05,
      "loss": 0.38674952983856203,
      "num_input_tokens_seen": 13934492032,
      "step": 49000,
      "train_runtime": 476390.1694,
      "train_tokens_per_second": 29250.167
    },
    {
      "epoch": 1.734070901158098,
      "grad_norm": 0.57421875,
      "learning_rate": 2.0583104401886512e-05,
      "loss": 0.38558807373046877,
      "num_input_tokens_seen": 13937352640,
      "step": 49010,
      "train_runtime": 476487.3211,
      "train_tokens_per_second": 29250.207
    },
    {
      "epoch": 1.7344247214218267,
      "grad_norm": 0.625,
      "learning_rate": 2.0581360558686828e-05,
      "loss": 0.389147424697876,
      "num_input_tokens_seen": 13940169664,
      "step": 49020,
      "train_runtime": 476584.5349,
      "train_tokens_per_second": 29250.151
    },
    {
      "epoch": 1.7347785416855555,
      "grad_norm": 0.62890625,
      "learning_rate": 2.057961715863807e-05,
      "loss": 0.39339375495910645,
      "num_input_tokens_seen": 13942974080,
      "step": 49030,
      "train_runtime": 476681.9763,
      "train_tokens_per_second": 29250.055
    },
    {
      "epoch": 1.7351323619492844,
      "grad_norm": 0.59375,
      "learning_rate": 2.0577874201552578e-05,
      "loss": 0.38551983833312986,
      "num_input_tokens_seen": 13945876096,
      "step": 49040,
      "train_runtime": 476782.1259,
      "train_tokens_per_second": 29249.998
    },
    {
      "epoch": 1.735486182213013,
      "grad_norm": 0.625,
      "learning_rate": 2.0576131687242803e-05,
      "loss": 0.38854217529296875,
      "num_input_tokens_seen": 13948727552,
      "step": 49050,
      "train_runtime": 476878.5845,
      "train_tokens_per_second": 29250.061
    },
    {
      "epoch": 1.735840002476742,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0574389615521303e-05,
      "loss": 0.38984825611114504,
      "num_input_tokens_seen": 13951510016,
      "step": 49060,
      "train_runtime": 476972.6515,
      "train_tokens_per_second": 29250.126
    },
    {
      "epoch": 1.7361938227404705,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0572647986200757e-05,
      "loss": 0.38666157722473143,
      "num_input_tokens_seen": 13954370496,
      "step": 49070,
      "train_runtime": 477069.8123,
      "train_tokens_per_second": 29250.165
    },
    {
      "epoch": 1.7365476430041995,
      "grad_norm": 0.61328125,
      "learning_rate": 2.057090679909395e-05,
      "loss": 0.3921320915222168,
      "num_input_tokens_seen": 13957226688,
      "step": 49080,
      "train_runtime": 477166.6995,
      "train_tokens_per_second": 29250.211
    },
    {
      "epoch": 1.736901463267928,
      "grad_norm": 0.640625,
      "learning_rate": 2.056916605401378e-05,
      "loss": 0.3907826662063599,
      "num_input_tokens_seen": 13960103168,
      "step": 49090,
      "train_runtime": 477263.1189,
      "train_tokens_per_second": 29250.329
    },
    {
      "epoch": 1.737255283531657,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0567425750773247e-05,
      "loss": 0.38638982772827146,
      "num_input_tokens_seen": 13962869504,
      "step": 49100,
      "train_runtime": 477356.8745,
      "train_tokens_per_second": 29250.379
    },
    {
      "epoch": 1.7376091037953856,
      "grad_norm": 0.5859375,
      "learning_rate": 2.056568588918547e-05,
      "loss": 0.38939127922058103,
      "num_input_tokens_seen": 13965735744,
      "step": 49110,
      "train_runtime": 477453.2158,
      "train_tokens_per_second": 29250.48
    },
    {
      "epoch": 1.7379629240591146,
      "grad_norm": 0.578125,
      "learning_rate": 2.056394646906368e-05,
      "loss": 0.3872859001159668,
      "num_input_tokens_seen": 13968618560,
      "step": 49120,
      "train_runtime": 477553.4373,
      "train_tokens_per_second": 29250.378
    },
    {
      "epoch": 1.7383167443228433,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0562207490221216e-05,
      "loss": 0.3951389789581299,
      "num_input_tokens_seen": 13971409984,
      "step": 49130,
      "train_runtime": 477649.583,
      "train_tokens_per_second": 29250.334
    },
    {
      "epoch": 1.738670564586572,
      "grad_norm": 0.58203125,
      "learning_rate": 2.056046895247152e-05,
      "loss": 0.38391273021697997,
      "num_input_tokens_seen": 13974325824,
      "step": 49140,
      "train_runtime": 477751.9794,
      "train_tokens_per_second": 29250.168
    },
    {
      "epoch": 1.7390243848503009,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0558730855628155e-05,
      "loss": 0.3855523347854614,
      "num_input_tokens_seen": 13977164992,
      "step": 49150,
      "train_runtime": 477849.3774,
      "train_tokens_per_second": 29250.148
    },
    {
      "epoch": 1.7393782051140296,
      "grad_norm": 0.578125,
      "learning_rate": 2.055699319950479e-05,
      "loss": 0.3909287929534912,
      "num_input_tokens_seen": 13980006592,
      "step": 49160,
      "train_runtime": 477947.0516,
      "train_tokens_per_second": 29250.116
    },
    {
      "epoch": 1.7397320253777584,
      "grad_norm": 0.59765625,
      "learning_rate": 2.055525598391521e-05,
      "loss": 0.38961195945739746,
      "num_input_tokens_seen": 13982844032,
      "step": 49170,
      "train_runtime": 478041.199,
      "train_tokens_per_second": 29250.291
    },
    {
      "epoch": 1.7400858456414872,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0553519208673293e-05,
      "loss": 0.3888551712036133,
      "num_input_tokens_seen": 13985742656,
      "step": 49180,
      "train_runtime": 478141.1087,
      "train_tokens_per_second": 29250.241
    },
    {
      "epoch": 1.740439665905216,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0551782873593057e-05,
      "loss": 0.39024658203125,
      "num_input_tokens_seen": 13988579712,
      "step": 49190,
      "train_runtime": 478237.363,
      "train_tokens_per_second": 29250.286
    },
    {
      "epoch": 1.7407934861689447,
      "grad_norm": 0.609375,
      "learning_rate": 2.0550046978488592e-05,
      "loss": 0.3875107288360596,
      "num_input_tokens_seen": 13991383168,
      "step": 49200,
      "train_runtime": 478334.1302,
      "train_tokens_per_second": 29250.23
    },
    {
      "epoch": 1.7411473064326737,
      "grad_norm": 0.609375,
      "learning_rate": 2.0548311523174133e-05,
      "loss": 0.3859492301940918,
      "num_input_tokens_seen": 13994221952,
      "step": 49210,
      "train_runtime": 478432.8478,
      "train_tokens_per_second": 29250.128
    },
    {
      "epoch": 1.7415011266964022,
      "grad_norm": 0.59375,
      "learning_rate": 2.0546576507464003e-05,
      "loss": 0.38080430030822754,
      "num_input_tokens_seen": 13997047232,
      "step": 49220,
      "train_runtime": 478527.5028,
      "train_tokens_per_second": 29250.246
    },
    {
      "epoch": 1.7418549469601312,
      "grad_norm": 0.62109375,
      "learning_rate": 2.054484193117264e-05,
      "loss": 0.38934381008148194,
      "num_input_tokens_seen": 13999908032,
      "step": 49230,
      "train_runtime": 478627.0181,
      "train_tokens_per_second": 29250.142
    },
    {
      "epoch": 1.7422087672238598,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0543107794114597e-05,
      "loss": 0.39241857528686525,
      "num_input_tokens_seen": 14002766016,
      "step": 49240,
      "train_runtime": 478725.1502,
      "train_tokens_per_second": 29250.116
    },
    {
      "epoch": 1.7425625874875887,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0541374096104523e-05,
      "loss": 0.39133267402648925,
      "num_input_tokens_seen": 14005568576,
      "step": 49250,
      "train_runtime": 478819.4384,
      "train_tokens_per_second": 29250.209
    },
    {
      "epoch": 1.7429164077513173,
      "grad_norm": 0.6171875,
      "learning_rate": 2.05396408369572e-05,
      "loss": 0.3865822315216064,
      "num_input_tokens_seen": 14008406912,
      "step": 49260,
      "train_runtime": 478916.8275,
      "train_tokens_per_second": 29250.187
    },
    {
      "epoch": 1.7432702280150463,
      "grad_norm": 0.62109375,
      "learning_rate": 2.05379080164875e-05,
      "loss": 0.3883687973022461,
      "num_input_tokens_seen": 14011195456,
      "step": 49270,
      "train_runtime": 479011.4427,
      "train_tokens_per_second": 29250.231
    },
    {
      "epoch": 1.743624048278775,
      "grad_norm": 0.6015625,
      "learning_rate": 2.053617563451041e-05,
      "loss": 0.394056510925293,
      "num_input_tokens_seen": 14014009280,
      "step": 49280,
      "train_runtime": 479106.6861,
      "train_tokens_per_second": 29250.29
    },
    {
      "epoch": 1.7439778685425038,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0534443690841026e-05,
      "loss": 0.39061501026153567,
      "num_input_tokens_seen": 14016871680,
      "step": 49290,
      "train_runtime": 479202.8156,
      "train_tokens_per_second": 29250.395
    },
    {
      "epoch": 1.7443316888062326,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0532712185294546e-05,
      "loss": 0.39222581386566163,
      "num_input_tokens_seen": 14019706496,
      "step": 49300,
      "train_runtime": 479299.0917,
      "train_tokens_per_second": 29250.434
    },
    {
      "epoch": 1.7446855090699613,
      "grad_norm": 0.59375,
      "learning_rate": 2.0530981117686292e-05,
      "loss": 0.3871941566467285,
      "num_input_tokens_seen": 14022544512,
      "step": 49310,
      "train_runtime": 479395.8669,
      "train_tokens_per_second": 29250.449
    },
    {
      "epoch": 1.74503932933369,
      "grad_norm": 0.609375,
      "learning_rate": 2.0529250487831685e-05,
      "loss": 0.38560240268707274,
      "num_input_tokens_seen": 14025402624,
      "step": 49320,
      "train_runtime": 479493.1182,
      "train_tokens_per_second": 29250.477
    },
    {
      "epoch": 1.7453931495974189,
      "grad_norm": 0.609375,
      "learning_rate": 2.0527520295546256e-05,
      "loss": 0.3916832447052002,
      "num_input_tokens_seen": 14028195008,
      "step": 49330,
      "train_runtime": 479587.8461,
      "train_tokens_per_second": 29250.522
    },
    {
      "epoch": 1.7457469698611476,
      "grad_norm": 0.58203125,
      "learning_rate": 2.052579054064565e-05,
      "loss": 0.3858347415924072,
      "num_input_tokens_seen": 14031070592,
      "step": 49340,
      "train_runtime": 479686.2644,
      "train_tokens_per_second": 29250.516
    },
    {
      "epoch": 1.7461007901248764,
      "grad_norm": 0.59765625,
      "learning_rate": 2.052406122294561e-05,
      "loss": 0.3910540580749512,
      "num_input_tokens_seen": 14033917056,
      "step": 49350,
      "train_runtime": 479784.6796,
      "train_tokens_per_second": 29250.448
    },
    {
      "epoch": 1.7464546103886052,
      "grad_norm": 0.5859375,
      "learning_rate": 2.0522332342261998e-05,
      "loss": 0.38623507022857667,
      "num_input_tokens_seen": 14036744256,
      "step": 49360,
      "train_runtime": 479880.8659,
      "train_tokens_per_second": 29250.477
    },
    {
      "epoch": 1.746808430652334,
      "grad_norm": 0.6015625,
      "learning_rate": 2.052060389841078e-05,
      "loss": 0.3902060747146606,
      "num_input_tokens_seen": 14039614784,
      "step": 49370,
      "train_runtime": 479979.2337,
      "train_tokens_per_second": 29250.463
    },
    {
      "epoch": 1.747162250916063,
      "grad_norm": 0.64453125,
      "learning_rate": 2.0518875891208032e-05,
      "loss": 0.38612380027771,
      "num_input_tokens_seen": 14042505024,
      "step": 49380,
      "train_runtime": 480076.6135,
      "train_tokens_per_second": 29250.55
    },
    {
      "epoch": 1.7475160711797915,
      "grad_norm": 0.625,
      "learning_rate": 2.051714832046994e-05,
      "loss": 0.3875981092453003,
      "num_input_tokens_seen": 14045327680,
      "step": 49390,
      "train_runtime": 480171.0525,
      "train_tokens_per_second": 29250.676
    },
    {
      "epoch": 1.7478698914435205,
      "grad_norm": 0.59375,
      "learning_rate": 2.051542118601279e-05,
      "loss": 0.3948683500289917,
      "num_input_tokens_seen": 14048168320,
      "step": 49400,
      "train_runtime": 480268.5161,
      "train_tokens_per_second": 29250.654
    },
    {
      "epoch": 1.748223711707249,
      "grad_norm": 0.609375,
      "learning_rate": 2.0513694487652985e-05,
      "loss": 0.38892192840576173,
      "num_input_tokens_seen": 14050945984,
      "step": 49410,
      "train_runtime": 480361.8669,
      "train_tokens_per_second": 29250.752
    },
    {
      "epoch": 1.748577531970978,
      "grad_norm": 0.63671875,
      "learning_rate": 2.051196822520703e-05,
      "loss": 0.39353418350219727,
      "num_input_tokens_seen": 14053775424,
      "step": 49420,
      "train_runtime": 480455.7062,
      "train_tokens_per_second": 29250.928
    },
    {
      "epoch": 1.7489313522347065,
      "grad_norm": 0.59375,
      "learning_rate": 2.0510242398491553e-05,
      "loss": 0.38719897270202636,
      "num_input_tokens_seen": 14056553600,
      "step": 49430,
      "train_runtime": 480550.6452,
      "train_tokens_per_second": 29250.931
    },
    {
      "epoch": 1.7492851724984355,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0508517007323266e-05,
      "loss": 0.3912373065948486,
      "num_input_tokens_seen": 14059430400,
      "step": 49440,
      "train_runtime": 480648.6139,
      "train_tokens_per_second": 29250.954
    },
    {
      "epoch": 1.7496389927621643,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0506792051519007e-05,
      "loss": 0.38600988388061525,
      "num_input_tokens_seen": 14062304448,
      "step": 49450,
      "train_runtime": 480748.0998,
      "train_tokens_per_second": 29250.879
    },
    {
      "epoch": 1.749992813025893,
      "grad_norm": 0.6015625,
      "learning_rate": 2.050506753089571e-05,
      "loss": 0.39232099056243896,
      "num_input_tokens_seen": 14065174720,
      "step": 49460,
      "train_runtime": 480847.976,
      "train_tokens_per_second": 29250.772
    },
    {
      "epoch": 1.7503466332896218,
      "grad_norm": 0.6015625,
      "learning_rate": 2.050334344527043e-05,
      "loss": 0.38351552486419677,
      "num_input_tokens_seen": 14067993536,
      "step": 49470,
      "train_runtime": 480943.036,
      "train_tokens_per_second": 29250.852
    },
    {
      "epoch": 1.7507004535533506,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0501619794460322e-05,
      "loss": 0.38890819549560546,
      "num_input_tokens_seen": 14070812992,
      "step": 49480,
      "train_runtime": 481037.6774,
      "train_tokens_per_second": 29250.958
    },
    {
      "epoch": 1.7510542738170793,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0499896578282647e-05,
      "loss": 0.38720974922180174,
      "num_input_tokens_seen": 14073621120,
      "step": 49490,
      "train_runtime": 481132.7232,
      "train_tokens_per_second": 29251.016
    },
    {
      "epoch": 1.751408094080808,
      "grad_norm": 0.6328125,
      "learning_rate": 2.049817379655477e-05,
      "loss": 0.3879827976226807,
      "num_input_tokens_seen": 14076485120,
      "step": 49500,
      "train_runtime": 481230.7111,
      "train_tokens_per_second": 29251.012
    },
    {
      "epoch": 1.7517619143445369,
      "grad_norm": 0.625,
      "learning_rate": 2.049645144909418e-05,
      "loss": 0.3924839019775391,
      "num_input_tokens_seen": 14079344448,
      "step": 49510,
      "train_runtime": 481328.2738,
      "train_tokens_per_second": 29251.023
    },
    {
      "epoch": 1.7521157346082656,
      "grad_norm": 0.578125,
      "learning_rate": 2.0494729535718454e-05,
      "loss": 0.38801000118255613,
      "num_input_tokens_seen": 14082116544,
      "step": 49520,
      "train_runtime": 481423.1651,
      "train_tokens_per_second": 29251.016
    },
    {
      "epoch": 1.7524695548719946,
      "grad_norm": 0.6328125,
      "learning_rate": 2.0493008056245292e-05,
      "loss": 0.3944086074829102,
      "num_input_tokens_seen": 14085002496,
      "step": 49530,
      "train_runtime": 481521.9948,
      "train_tokens_per_second": 29251.005
    },
    {
      "epoch": 1.7528233751357232,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0491287010492484e-05,
      "loss": 0.3899362564086914,
      "num_input_tokens_seen": 14087823168,
      "step": 49540,
      "train_runtime": 481618.526,
      "train_tokens_per_second": 29250.999
    },
    {
      "epoch": 1.7531771953994522,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0489566398277943e-05,
      "loss": 0.38977460861206054,
      "num_input_tokens_seen": 14090702016,
      "step": 49550,
      "train_runtime": 481715.9026,
      "train_tokens_per_second": 29251.063
    },
    {
      "epoch": 1.7535310156631807,
      "grad_norm": 0.59375,
      "learning_rate": 2.0487846219419683e-05,
      "loss": 0.3842364549636841,
      "num_input_tokens_seen": 14093561536,
      "step": 49560,
      "train_runtime": 481812.2191,
      "train_tokens_per_second": 29251.15
    },
    {
      "epoch": 1.7538848359269097,
      "grad_norm": 0.640625,
      "learning_rate": 2.048612647373583e-05,
      "loss": 0.38803696632385254,
      "num_input_tokens_seen": 14096403200,
      "step": 49570,
      "train_runtime": 481908.0421,
      "train_tokens_per_second": 29251.23
    },
    {
      "epoch": 1.7542386561906382,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0484407161044598e-05,
      "loss": 0.38816018104553224,
      "num_input_tokens_seen": 14099277248,
      "step": 49580,
      "train_runtime": 482011.921,
      "train_tokens_per_second": 29250.889
    },
    {
      "epoch": 1.7545924764543672,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0482688281164333e-05,
      "loss": 0.38619720935821533,
      "num_input_tokens_seen": 14102163968,
      "step": 49590,
      "train_runtime": 482109.9581,
      "train_tokens_per_second": 29250.929
    },
    {
      "epoch": 1.7549462967180958,
      "grad_norm": 0.6015625,
      "learning_rate": 2.048096983391347e-05,
      "loss": 0.3884623050689697,
      "num_input_tokens_seen": 14104997632,
      "step": 49600,
      "train_runtime": 482204.5257,
      "train_tokens_per_second": 29251.069
    },
    {
      "epoch": 1.7553001169818248,
      "grad_norm": 0.6015625,
      "learning_rate": 2.047925181911056e-05,
      "loss": 0.39087040424346925,
      "num_input_tokens_seen": 14107825920,
      "step": 49610,
      "train_runtime": 482300.8649,
      "train_tokens_per_second": 29251.09
    },
    {
      "epoch": 1.7556539372455535,
      "grad_norm": 0.625,
      "learning_rate": 2.0477534236574254e-05,
      "loss": 0.3876131534576416,
      "num_input_tokens_seen": 14110684672,
      "step": 49620,
      "train_runtime": 482397.3718,
      "train_tokens_per_second": 29251.164
    },
    {
      "epoch": 1.7560077575092823,
      "grad_norm": 0.625,
      "learning_rate": 2.0475817086123313e-05,
      "loss": 0.38890039920806885,
      "num_input_tokens_seen": 14113532736,
      "step": 49630,
      "train_runtime": 482496.6899,
      "train_tokens_per_second": 29251.046
    },
    {
      "epoch": 1.756361577773011,
      "grad_norm": 0.59765625,
      "learning_rate": 2.047410036757661e-05,
      "loss": 0.3884609222412109,
      "num_input_tokens_seen": 14116398912,
      "step": 49640,
      "train_runtime": 482594.4803,
      "train_tokens_per_second": 29251.058
    },
    {
      "epoch": 1.7567153980367398,
      "grad_norm": 0.59375,
      "learning_rate": 2.0472384080753113e-05,
      "loss": 0.3850518226623535,
      "num_input_tokens_seen": 14119203904,
      "step": 49650,
      "train_runtime": 482688.6484,
      "train_tokens_per_second": 29251.162
    },
    {
      "epoch": 1.7570692183004686,
      "grad_norm": 0.6328125,
      "learning_rate": 2.04706682254719e-05,
      "loss": 0.38535640239715574,
      "num_input_tokens_seen": 14122062720,
      "step": 49660,
      "train_runtime": 482787.2379,
      "train_tokens_per_second": 29251.11
    },
    {
      "epoch": 1.7574230385641973,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0468952801552166e-05,
      "loss": 0.38881592750549315,
      "num_input_tokens_seen": 14124929664,
      "step": 49670,
      "train_runtime": 482886.9309,
      "train_tokens_per_second": 29251.008
    },
    {
      "epoch": 1.7577768588279261,
      "grad_norm": 0.6171875,
      "learning_rate": 2.046723780881319e-05,
      "loss": 0.3919193744659424,
      "num_input_tokens_seen": 14127732224,
      "step": 49680,
      "train_runtime": 482984.0892,
      "train_tokens_per_second": 29250.927
    },
    {
      "epoch": 1.7581306790916549,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0465523247074376e-05,
      "loss": 0.38100900650024416,
      "num_input_tokens_seen": 14130582720,
      "step": 49690,
      "train_runtime": 483079.7414,
      "train_tokens_per_second": 29251.036
    },
    {
      "epoch": 1.7584844993553839,
      "grad_norm": 0.58984375,
      "learning_rate": 2.046380911615523e-05,
      "loss": 0.38820900917053225,
      "num_input_tokens_seen": 14133457536,
      "step": 49700,
      "train_runtime": 483177.5975,
      "train_tokens_per_second": 29251.061
    },
    {
      "epoch": 1.7588383196191124,
      "grad_norm": 0.63671875,
      "learning_rate": 2.0462095415875357e-05,
      "loss": 0.38717851638793943,
      "num_input_tokens_seen": 14136300352,
      "step": 49710,
      "train_runtime": 483273.7354,
      "train_tokens_per_second": 29251.125
    },
    {
      "epoch": 1.7591921398828414,
      "grad_norm": 0.58203125,
      "learning_rate": 2.046038214605448e-05,
      "loss": 0.38627569675445556,
      "num_input_tokens_seen": 14139115072,
      "step": 49720,
      "train_runtime": 483367.6709,
      "train_tokens_per_second": 29251.263
    },
    {
      "epoch": 1.75954596014657,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0458669306512404e-05,
      "loss": 0.3881821155548096,
      "num_input_tokens_seen": 14141986880,
      "step": 49730,
      "train_runtime": 483467.0762,
      "train_tokens_per_second": 29251.189
    },
    {
      "epoch": 1.759899780410299,
      "grad_norm": 0.61328125,
      "learning_rate": 2.045695689706907e-05,
      "loss": 0.38967266082763674,
      "num_input_tokens_seen": 14144775616,
      "step": 49740,
      "train_runtime": 483559.8751,
      "train_tokens_per_second": 29251.343
    },
    {
      "epoch": 1.7602536006740275,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0455244917544507e-05,
      "loss": 0.3913793325424194,
      "num_input_tokens_seen": 14147649088,
      "step": 49750,
      "train_runtime": 483657.1635,
      "train_tokens_per_second": 29251.4
    },
    {
      "epoch": 1.7606074209377565,
      "grad_norm": 0.58203125,
      "learning_rate": 2.0453533367758846e-05,
      "loss": 0.3884756326675415,
      "num_input_tokens_seen": 14150497472,
      "step": 49760,
      "train_runtime": 483755.5038,
      "train_tokens_per_second": 29251.342
    },
    {
      "epoch": 1.760961241201485,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0451822247532338e-05,
      "loss": 0.3908435821533203,
      "num_input_tokens_seen": 14153397504,
      "step": 49770,
      "train_runtime": 483855.3439,
      "train_tokens_per_second": 29251.299
    },
    {
      "epoch": 1.761315061465214,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0450111556685323e-05,
      "loss": 0.385576868057251,
      "num_input_tokens_seen": 14156242496,
      "step": 49780,
      "train_runtime": 483953.2473,
      "train_tokens_per_second": 29251.26
    },
    {
      "epoch": 1.7616688817289428,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0448401295038257e-05,
      "loss": 0.3850072145462036,
      "num_input_tokens_seen": 14159148352,
      "step": 49790,
      "train_runtime": 484050.2214,
      "train_tokens_per_second": 29251.404
    },
    {
      "epoch": 1.7620227019926715,
      "grad_norm": 0.60546875,
      "learning_rate": 2.04466914624117e-05,
      "loss": 0.39039804935455324,
      "num_input_tokens_seen": 14162018816,
      "step": 49800,
      "train_runtime": 484146.162,
      "train_tokens_per_second": 29251.536
    },
    {
      "epoch": 1.7623765222564003,
      "grad_norm": 0.609375,
      "learning_rate": 2.0444982058626314e-05,
      "loss": 0.3904542922973633,
      "num_input_tokens_seen": 14164775744,
      "step": 49810,
      "train_runtime": 484236.7633,
      "train_tokens_per_second": 29251.756
    },
    {
      "epoch": 1.762730342520129,
      "grad_norm": 0.609375,
      "learning_rate": 2.0443273083502863e-05,
      "loss": 0.3892939567565918,
      "num_input_tokens_seen": 14167685312,
      "step": 49820,
      "train_runtime": 484335.816,
      "train_tokens_per_second": 29251.781
    },
    {
      "epoch": 1.7630841627838578,
      "grad_norm": 0.59375,
      "learning_rate": 2.044156453686223e-05,
      "loss": 0.39123556613922117,
      "num_input_tokens_seen": 14170503936,
      "step": 49830,
      "train_runtime": 484429.7172,
      "train_tokens_per_second": 29251.93
    },
    {
      "epoch": 1.7634379830475866,
      "grad_norm": 0.63671875,
      "learning_rate": 2.0439856418525386e-05,
      "loss": 0.38967945575714114,
      "num_input_tokens_seen": 14173361536,
      "step": 49840,
      "train_runtime": 484524.8368,
      "train_tokens_per_second": 29252.085
    },
    {
      "epoch": 1.7637918033113154,
      "grad_norm": 0.60546875,
      "learning_rate": 2.043814872831341e-05,
      "loss": 0.3876481056213379,
      "num_input_tokens_seen": 14176290624,
      "step": 49850,
      "train_runtime": 484624.3168,
      "train_tokens_per_second": 29252.124
    },
    {
      "epoch": 1.7641456235750441,
      "grad_norm": 0.609375,
      "learning_rate": 2.0436441466047494e-05,
      "loss": 0.3952652454376221,
      "num_input_tokens_seen": 14179099648,
      "step": 49860,
      "train_runtime": 484717.9606,
      "train_tokens_per_second": 29252.268
    },
    {
      "epoch": 1.764499443838773,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0434734631548927e-05,
      "loss": 0.3850527048110962,
      "num_input_tokens_seen": 14181980544,
      "step": 49870,
      "train_runtime": 484820.7565,
      "train_tokens_per_second": 29252.008
    },
    {
      "epoch": 1.7648532641025017,
      "grad_norm": 0.66796875,
      "learning_rate": 2.0433028224639106e-05,
      "loss": 0.39000668525695803,
      "num_input_tokens_seen": 14184771008,
      "step": 49880,
      "train_runtime": 484911.9545,
      "train_tokens_per_second": 29252.261
    },
    {
      "epoch": 1.7652070843662306,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0431322245139533e-05,
      "loss": 0.3878610610961914,
      "num_input_tokens_seen": 14187629440,
      "step": 49890,
      "train_runtime": 485009.7506,
      "train_tokens_per_second": 29252.256
    },
    {
      "epoch": 1.7655609046299592,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0429616692871808e-05,
      "loss": 0.38987021446228026,
      "num_input_tokens_seen": 14190438336,
      "step": 49900,
      "train_runtime": 485105.7331,
      "train_tokens_per_second": 29252.259
    },
    {
      "epoch": 1.7659147248936882,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0427911567657644e-05,
      "loss": 0.3873212575912476,
      "num_input_tokens_seen": 14193285184,
      "step": 49910,
      "train_runtime": 485204.9153,
      "train_tokens_per_second": 29252.146
    },
    {
      "epoch": 1.7662685451574167,
      "grad_norm": 0.625,
      "learning_rate": 2.0426206869318852e-05,
      "loss": 0.3871925354003906,
      "num_input_tokens_seen": 14196087936,
      "step": 49920,
      "train_runtime": 485298.3057,
      "train_tokens_per_second": 29252.292
    },
    {
      "epoch": 1.7666223654211457,
      "grad_norm": 0.58203125,
      "learning_rate": 2.0424502597677357e-05,
      "loss": 0.3897804021835327,
      "num_input_tokens_seen": 14198965376,
      "step": 49930,
      "train_runtime": 485399.9089,
      "train_tokens_per_second": 29252.097
    },
    {
      "epoch": 1.7669761856848742,
      "grad_norm": 0.5859375,
      "learning_rate": 2.042279875255517e-05,
      "loss": 0.3890186309814453,
      "num_input_tokens_seen": 14201795712,
      "step": 49940,
      "train_runtime": 485497.8847,
      "train_tokens_per_second": 29252.024
    },
    {
      "epoch": 1.7673300059486032,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0421095333774413e-05,
      "loss": 0.3899000883102417,
      "num_input_tokens_seen": 14204625984,
      "step": 49950,
      "train_runtime": 485593.7096,
      "train_tokens_per_second": 29252.08
    },
    {
      "epoch": 1.767683826212332,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0419392341157327e-05,
      "loss": 0.3857257843017578,
      "num_input_tokens_seen": 14207496960,
      "step": 49960,
      "train_runtime": 485694.9456,
      "train_tokens_per_second": 29251.894
    },
    {
      "epoch": 1.7680376464760608,
      "grad_norm": 0.56640625,
      "learning_rate": 2.0417689774526235e-05,
      "loss": 0.388786768913269,
      "num_input_tokens_seen": 14210322496,
      "step": 49970,
      "train_runtime": 485792.1481,
      "train_tokens_per_second": 29251.857
    },
    {
      "epoch": 1.7683914667397895,
      "grad_norm": 0.625,
      "learning_rate": 2.0415987633703583e-05,
      "loss": 0.3869042873382568,
      "num_input_tokens_seen": 14213167616,
      "step": 49980,
      "train_runtime": 485890.2902,
      "train_tokens_per_second": 29251.804
    },
    {
      "epoch": 1.7687452870035183,
      "grad_norm": 0.625,
      "learning_rate": 2.04142859185119e-05,
      "loss": 0.3896949768066406,
      "num_input_tokens_seen": 14216049408,
      "step": 49990,
      "train_runtime": 485989.8037,
      "train_tokens_per_second": 29251.744
    },
    {
      "epoch": 1.769099107267247,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0412584628773835e-05,
      "loss": 0.39403295516967773,
      "num_input_tokens_seen": 14218881984,
      "step": 50000,
      "train_runtime": 486088.2502,
      "train_tokens_per_second": 29251.647
    },
    {
      "epoch": 1.769099107267247,
      "eval_loss": 0.33621031045913696,
      "eval_runtime": 8.4143,
      "eval_samples_per_second": 473.954,
      "eval_steps_per_second": 3.803,
      "num_input_tokens_seen": 14218881984,
      "step": 50000
    },
    {
      "epoch": 1.7694529275309758,
      "grad_norm": 0.58203125,
      "learning_rate": 2.0410883764312135e-05,
      "loss": 0.38748767375946047,
      "num_input_tokens_seen": 14221693248,
      "step": 50010,
      "train_runtime": 486213.8011,
      "train_tokens_per_second": 29249.876
    },
    {
      "epoch": 1.7698067477947046,
      "grad_norm": 0.625,
      "learning_rate": 2.0409183324949646e-05,
      "loss": 0.389849328994751,
      "num_input_tokens_seen": 14224536320,
      "step": 50020,
      "train_runtime": 486314.1335,
      "train_tokens_per_second": 29249.687
    },
    {
      "epoch": 1.7701605680584334,
      "grad_norm": 0.625,
      "learning_rate": 2.0407483310509333e-05,
      "loss": 0.38871333599090574,
      "num_input_tokens_seen": 14227418304,
      "step": 50030,
      "train_runtime": 486415.575,
      "train_tokens_per_second": 29249.512
    },
    {
      "epoch": 1.7705143883221623,
      "grad_norm": 0.6171875,
      "learning_rate": 2.040578372081424e-05,
      "loss": 0.389327073097229,
      "num_input_tokens_seen": 14230233536,
      "step": 50040,
      "train_runtime": 486510.9739,
      "train_tokens_per_second": 29249.563
    },
    {
      "epoch": 1.770868208585891,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0404084555687525e-05,
      "loss": 0.3867353439331055,
      "num_input_tokens_seen": 14233023552,
      "step": 50050,
      "train_runtime": 486603.7164,
      "train_tokens_per_second": 29249.722
    },
    {
      "epoch": 1.7712220288496199,
      "grad_norm": 0.609375,
      "learning_rate": 2.0402385814952464e-05,
      "loss": 0.3870378494262695,
      "num_input_tokens_seen": 14235863680,
      "step": 50060,
      "train_runtime": 486699.7038,
      "train_tokens_per_second": 29249.789
    },
    {
      "epoch": 1.7715758491133484,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0400687498432415e-05,
      "loss": 0.38392367362976076,
      "num_input_tokens_seen": 14238697344,
      "step": 50070,
      "train_runtime": 486795.689,
      "train_tokens_per_second": 29249.843
    },
    {
      "epoch": 1.7719296693770774,
      "grad_norm": 0.62109375,
      "learning_rate": 2.039898960595085e-05,
      "loss": 0.3898477077484131,
      "num_input_tokens_seen": 14241492992,
      "step": 50080,
      "train_runtime": 486891.6743,
      "train_tokens_per_second": 29249.818
    },
    {
      "epoch": 1.772283489640806,
      "grad_norm": 0.59765625,
      "learning_rate": 2.039729213733134e-05,
      "loss": 0.3867949962615967,
      "num_input_tokens_seen": 14244341760,
      "step": 50090,
      "train_runtime": 486990.4964,
      "train_tokens_per_second": 29249.733
    },
    {
      "epoch": 1.772637309904535,
      "grad_norm": 0.671875,
      "learning_rate": 2.0395595092397554e-05,
      "loss": 0.3867375373840332,
      "num_input_tokens_seen": 14247171968,
      "step": 50100,
      "train_runtime": 487089.4145,
      "train_tokens_per_second": 29249.603
    },
    {
      "epoch": 1.7729911301682635,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0393898470973276e-05,
      "loss": 0.3887359380722046,
      "num_input_tokens_seen": 14249991808,
      "step": 50110,
      "train_runtime": 487187.819,
      "train_tokens_per_second": 29249.483
    },
    {
      "epoch": 1.7733449504319925,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0392202272882384e-05,
      "loss": 0.39030065536499026,
      "num_input_tokens_seen": 14252898176,
      "step": 50120,
      "train_runtime": 487288.2392,
      "train_tokens_per_second": 29249.42
    },
    {
      "epoch": 1.7736987706957212,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0390506497948854e-05,
      "loss": 0.39316370487213137,
      "num_input_tokens_seen": 14255738752,
      "step": 50130,
      "train_runtime": 487386.6149,
      "train_tokens_per_second": 29249.344
    },
    {
      "epoch": 1.77405259095945,
      "grad_norm": 0.6171875,
      "learning_rate": 2.038881114599678e-05,
      "loss": 0.39061784744262695,
      "num_input_tokens_seen": 14258539840,
      "step": 50140,
      "train_runtime": 487480.1763,
      "train_tokens_per_second": 29249.476
    },
    {
      "epoch": 1.7744064112231788,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0387116216850345e-05,
      "loss": 0.38534982204437257,
      "num_input_tokens_seen": 14261400064,
      "step": 50150,
      "train_runtime": 487575.3235,
      "train_tokens_per_second": 29249.635
    },
    {
      "epoch": 1.7747602314869075,
      "grad_norm": 0.609375,
      "learning_rate": 2.0385421710333838e-05,
      "loss": 0.3840333938598633,
      "num_input_tokens_seen": 14264280128,
      "step": 50160,
      "train_runtime": 487672.4498,
      "train_tokens_per_second": 29249.715
    },
    {
      "epoch": 1.7751140517506363,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0383727626271648e-05,
      "loss": 0.39063568115234376,
      "num_input_tokens_seen": 14267157120,
      "step": 50170,
      "train_runtime": 487770.4123,
      "train_tokens_per_second": 29249.739
    },
    {
      "epoch": 1.775467872014365,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0382033964488274e-05,
      "loss": 0.3923508167266846,
      "num_input_tokens_seen": 14270035136,
      "step": 50180,
      "train_runtime": 487868.3916,
      "train_tokens_per_second": 29249.764
    },
    {
      "epoch": 1.7758216922780938,
      "grad_norm": 0.625,
      "learning_rate": 2.0380340724808307e-05,
      "loss": 0.3892721652984619,
      "num_input_tokens_seen": 14272912768,
      "step": 50190,
      "train_runtime": 487968.5315,
      "train_tokens_per_second": 29249.658
    },
    {
      "epoch": 1.7761755125418226,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0378647907056447e-05,
      "loss": 0.39028334617614746,
      "num_input_tokens_seen": 14275817984,
      "step": 50200,
      "train_runtime": 488068.5493,
      "train_tokens_per_second": 29249.617
    },
    {
      "epoch": 1.7765293328055516,
      "grad_norm": 0.609375,
      "learning_rate": 2.0376955511057493e-05,
      "loss": 0.38695802688598635,
      "num_input_tokens_seen": 14278631360,
      "step": 50210,
      "train_runtime": 488167.1444,
      "train_tokens_per_second": 29249.472
    },
    {
      "epoch": 1.7768831530692801,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0375263536636348e-05,
      "loss": 0.38867812156677245,
      "num_input_tokens_seen": 14281497792,
      "step": 50220,
      "train_runtime": 488264.505,
      "train_tokens_per_second": 29249.511
    },
    {
      "epoch": 1.7772369733330091,
      "grad_norm": 0.6328125,
      "learning_rate": 2.0373571983618014e-05,
      "loss": 0.39181008338928225,
      "num_input_tokens_seen": 14284357120,
      "step": 50230,
      "train_runtime": 488363.8163,
      "train_tokens_per_second": 29249.417
    },
    {
      "epoch": 1.7775907935967377,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0371880851827597e-05,
      "loss": 0.38538293838500975,
      "num_input_tokens_seen": 14287188160,
      "step": 50240,
      "train_runtime": 488459.4548,
      "train_tokens_per_second": 29249.486
    },
    {
      "epoch": 1.7779446138604666,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0370190141090298e-05,
      "loss": 0.38669528961181643,
      "num_input_tokens_seen": 14290047232,
      "step": 50250,
      "train_runtime": 488555.5429,
      "train_tokens_per_second": 29249.586
    },
    {
      "epoch": 1.7782984341241952,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0368499851231433e-05,
      "loss": 0.3892032623291016,
      "num_input_tokens_seen": 14292963264,
      "step": 50260,
      "train_runtime": 488655.9293,
      "train_tokens_per_second": 29249.544
    },
    {
      "epoch": 1.7786522543879242,
      "grad_norm": 0.63671875,
      "learning_rate": 2.0366809982076404e-05,
      "loss": 0.38855705261230467,
      "num_input_tokens_seen": 14295811392,
      "step": 50270,
      "train_runtime": 488755.0564,
      "train_tokens_per_second": 29249.439
    },
    {
      "epoch": 1.779006074651653,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0365120533450732e-05,
      "loss": 0.39193086624145507,
      "num_input_tokens_seen": 14298666560,
      "step": 50280,
      "train_runtime": 488851.2157,
      "train_tokens_per_second": 29249.526
    },
    {
      "epoch": 1.7793598949153817,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0363431505180018e-05,
      "loss": 0.38649711608886717,
      "num_input_tokens_seen": 14301523264,
      "step": 50290,
      "train_runtime": 488952.5037,
      "train_tokens_per_second": 29249.31
    },
    {
      "epoch": 1.7797137151791105,
      "grad_norm": 0.61328125,
      "learning_rate": 2.036174289708998e-05,
      "loss": 0.3914536714553833,
      "num_input_tokens_seen": 14304374016,
      "step": 50300,
      "train_runtime": 489051.6167,
      "train_tokens_per_second": 29249.211
    },
    {
      "epoch": 1.7800675354428392,
      "grad_norm": 0.59765625,
      "learning_rate": 2.036005470900643e-05,
      "loss": 0.38511826992034914,
      "num_input_tokens_seen": 14307237504,
      "step": 50310,
      "train_runtime": 489149.4595,
      "train_tokens_per_second": 29249.215
    },
    {
      "epoch": 1.780421355706568,
      "grad_norm": 0.6171875,
      "learning_rate": 2.035836694075529e-05,
      "loss": 0.3879786491394043,
      "num_input_tokens_seen": 14310040896,
      "step": 50320,
      "train_runtime": 489241.3717,
      "train_tokens_per_second": 29249.45
    },
    {
      "epoch": 1.7807751759702968,
      "grad_norm": 0.63671875,
      "learning_rate": 2.0356679592162574e-05,
      "loss": 0.38781094551086426,
      "num_input_tokens_seen": 14312854592,
      "step": 50330,
      "train_runtime": 489336.0319,
      "train_tokens_per_second": 29249.542
    },
    {
      "epoch": 1.7811289962340255,
      "grad_norm": 0.625,
      "learning_rate": 2.0354992663054395e-05,
      "loss": 0.38694162368774415,
      "num_input_tokens_seen": 14315676736,
      "step": 50340,
      "train_runtime": 489432.7855,
      "train_tokens_per_second": 29249.525
    },
    {
      "epoch": 1.7814828164977543,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0353306153256975e-05,
      "loss": 0.3925336837768555,
      "num_input_tokens_seen": 14318516544,
      "step": 50350,
      "train_runtime": 489535.1432,
      "train_tokens_per_second": 29249.211
    },
    {
      "epoch": 1.781836636761483,
      "grad_norm": 0.58203125,
      "learning_rate": 2.0351620062596634e-05,
      "loss": 0.3856451749801636,
      "num_input_tokens_seen": 14321402944,
      "step": 50360,
      "train_runtime": 489635.4111,
      "train_tokens_per_second": 29249.116
    },
    {
      "epoch": 1.7821904570252118,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0349934390899796e-05,
      "loss": 0.38912713527679443,
      "num_input_tokens_seen": 14324217408,
      "step": 50370,
      "train_runtime": 489729.8469,
      "train_tokens_per_second": 29249.223
    },
    {
      "epoch": 1.7825442772889408,
      "grad_norm": 0.609375,
      "learning_rate": 2.0348249137992967e-05,
      "loss": 0.38780841827392576,
      "num_input_tokens_seen": 14327077504,
      "step": 50380,
      "train_runtime": 489826.7211,
      "train_tokens_per_second": 29249.277
    },
    {
      "epoch": 1.7828980975526694,
      "grad_norm": 0.609375,
      "learning_rate": 2.034656430370278e-05,
      "loss": 0.3941533088684082,
      "num_input_tokens_seen": 14329920000,
      "step": 50390,
      "train_runtime": 489924.0967,
      "train_tokens_per_second": 29249.266
    },
    {
      "epoch": 1.7832519178163984,
      "grad_norm": 0.625,
      "learning_rate": 2.0344879887855952e-05,
      "loss": 0.3906111717224121,
      "num_input_tokens_seen": 14332805952,
      "step": 50400,
      "train_runtime": 490023.3647,
      "train_tokens_per_second": 29249.23
    },
    {
      "epoch": 1.783605738080127,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0343195890279314e-05,
      "loss": 0.39227895736694335,
      "num_input_tokens_seen": 14335667776,
      "step": 50410,
      "train_runtime": 490121.0244,
      "train_tokens_per_second": 29249.241
    },
    {
      "epoch": 1.783959558343856,
      "grad_norm": 0.58203125,
      "learning_rate": 2.034151231079977e-05,
      "loss": 0.3909771919250488,
      "num_input_tokens_seen": 14338499456,
      "step": 50420,
      "train_runtime": 490217.6106,
      "train_tokens_per_second": 29249.254
    },
    {
      "epoch": 1.7843133786075844,
      "grad_norm": 0.62109375,
      "learning_rate": 2.033982914924436e-05,
      "loss": 0.39262781143188474,
      "num_input_tokens_seen": 14341356736,
      "step": 50430,
      "train_runtime": 490317.5493,
      "train_tokens_per_second": 29249.12
    },
    {
      "epoch": 1.7846671988713134,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0338146405440198e-05,
      "loss": 0.3838437557220459,
      "num_input_tokens_seen": 14344204160,
      "step": 50440,
      "train_runtime": 490414.1884,
      "train_tokens_per_second": 29249.162
    },
    {
      "epoch": 1.7850210191350422,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0336464079214507e-05,
      "loss": 0.38554673194885253,
      "num_input_tokens_seen": 14347045632,
      "step": 50450,
      "train_runtime": 490511.2604,
      "train_tokens_per_second": 29249.167
    },
    {
      "epoch": 1.785374839398771,
      "grad_norm": 0.62109375,
      "learning_rate": 2.033478217039461e-05,
      "loss": 0.3890625,
      "num_input_tokens_seen": 14349957440,
      "step": 50460,
      "train_runtime": 490613.135,
      "train_tokens_per_second": 29249.028
    },
    {
      "epoch": 1.7857286596624997,
      "grad_norm": 0.6015625,
      "learning_rate": 2.033310067880794e-05,
      "loss": 0.3881007194519043,
      "num_input_tokens_seen": 14352851520,
      "step": 50470,
      "train_runtime": 490711.7109,
      "train_tokens_per_second": 29249.05
    },
    {
      "epoch": 1.7860824799262285,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0331419604282e-05,
      "loss": 0.38754591941833494,
      "num_input_tokens_seen": 14355658304,
      "step": 50480,
      "train_runtime": 490806.4584,
      "train_tokens_per_second": 29249.123
    },
    {
      "epoch": 1.7864363001899572,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0329738946644425e-05,
      "loss": 0.38557004928588867,
      "num_input_tokens_seen": 14358490496,
      "step": 50490,
      "train_runtime": 490904.2026,
      "train_tokens_per_second": 29249.068
    },
    {
      "epoch": 1.786790120453686,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0328058705722937e-05,
      "loss": 0.3874515056610107,
      "num_input_tokens_seen": 14361369600,
      "step": 50500,
      "train_runtime": 491003.4368,
      "train_tokens_per_second": 29249.021
    },
    {
      "epoch": 1.7871439407174148,
      "grad_norm": 0.64453125,
      "learning_rate": 2.0326378881345356e-05,
      "loss": 0.388558554649353,
      "num_input_tokens_seen": 14364233472,
      "step": 50510,
      "train_runtime": 491099.3591,
      "train_tokens_per_second": 29249.139
    },
    {
      "epoch": 1.7874977609811435,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0324699473339605e-05,
      "loss": 0.3865520477294922,
      "num_input_tokens_seen": 14367140736,
      "step": 50520,
      "train_runtime": 491199.9063,
      "train_tokens_per_second": 29249.071
    },
    {
      "epoch": 1.7878515812448725,
      "grad_norm": 0.63671875,
      "learning_rate": 2.0323020481533697e-05,
      "loss": 0.3855280876159668,
      "num_input_tokens_seen": 14369975680,
      "step": 50530,
      "train_runtime": 491301.4099,
      "train_tokens_per_second": 29248.798
    },
    {
      "epoch": 1.788205401508601,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0321341905755763e-05,
      "loss": 0.3844597578048706,
      "num_input_tokens_seen": 14372866816,
      "step": 50540,
      "train_runtime": 491400.7593,
      "train_tokens_per_second": 29248.768
    },
    {
      "epoch": 1.78855922177233,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0319663745834018e-05,
      "loss": 0.39006433486938474,
      "num_input_tokens_seen": 14375748352,
      "step": 50550,
      "train_runtime": 491501.2781,
      "train_tokens_per_second": 29248.649
    },
    {
      "epoch": 1.7889130420360586,
      "grad_norm": 0.62109375,
      "learning_rate": 2.031798600159678e-05,
      "loss": 0.3903589487075806,
      "num_input_tokens_seen": 14378604800,
      "step": 50560,
      "train_runtime": 491599.3989,
      "train_tokens_per_second": 29248.622
    },
    {
      "epoch": 1.7892668622997876,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0316308672872466e-05,
      "loss": 0.38635687828063964,
      "num_input_tokens_seen": 14381446144,
      "step": 50570,
      "train_runtime": 491694.3128,
      "train_tokens_per_second": 29248.754
    },
    {
      "epoch": 1.7896206825635161,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0314631759489597e-05,
      "loss": 0.3876587629318237,
      "num_input_tokens_seen": 14384275968,
      "step": 50580,
      "train_runtime": 491791.4679,
      "train_tokens_per_second": 29248.73
    },
    {
      "epoch": 1.7899745028272451,
      "grad_norm": 0.59375,
      "learning_rate": 2.0312955261276783e-05,
      "loss": 0.38613548278808596,
      "num_input_tokens_seen": 14387086464,
      "step": 50590,
      "train_runtime": 491886.9151,
      "train_tokens_per_second": 29248.768
    },
    {
      "epoch": 1.7903283230909737,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0311279178062754e-05,
      "loss": 0.3904703617095947,
      "num_input_tokens_seen": 14389862144,
      "step": 50600,
      "train_runtime": 491978.3965,
      "train_tokens_per_second": 29248.972
    },
    {
      "epoch": 1.7906821433547027,
      "grad_norm": 0.63671875,
      "learning_rate": 2.0309603509676307e-05,
      "loss": 0.39000988006591797,
      "num_input_tokens_seen": 14392652352,
      "step": 50610,
      "train_runtime": 492070.7514,
      "train_tokens_per_second": 29249.152
    },
    {
      "epoch": 1.7910359636184314,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0307928255946362e-05,
      "loss": 0.3903473377227783,
      "num_input_tokens_seen": 14395532992,
      "step": 50620,
      "train_runtime": 492168.4055,
      "train_tokens_per_second": 29249.202
    },
    {
      "epoch": 1.7913897838821602,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0306253416701935e-05,
      "loss": 0.38553466796875,
      "num_input_tokens_seen": 14398337088,
      "step": 50630,
      "train_runtime": 492262.4211,
      "train_tokens_per_second": 29249.312
    },
    {
      "epoch": 1.791743604145889,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0304578991772127e-05,
      "loss": 0.38522820472717284,
      "num_input_tokens_seen": 14401125888,
      "step": 50640,
      "train_runtime": 492355.2867,
      "train_tokens_per_second": 29249.459
    },
    {
      "epoch": 1.7920974244096177,
      "grad_norm": 0.5859375,
      "learning_rate": 2.030290498098616e-05,
      "loss": 0.3891852855682373,
      "num_input_tokens_seen": 14403993088,
      "step": 50650,
      "train_runtime": 492454.1209,
      "train_tokens_per_second": 29249.411
    },
    {
      "epoch": 1.7924512446733465,
      "grad_norm": 0.609375,
      "learning_rate": 2.0301231384173324e-05,
      "loss": 0.3905333995819092,
      "num_input_tokens_seen": 14406857472,
      "step": 50660,
      "train_runtime": 492551.175,
      "train_tokens_per_second": 29249.463
    },
    {
      "epoch": 1.7928050649370753,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0299558201163047e-05,
      "loss": 0.387405252456665,
      "num_input_tokens_seen": 14409669824,
      "step": 50670,
      "train_runtime": 492646.5498,
      "train_tokens_per_second": 29249.509
    },
    {
      "epoch": 1.793158885200804,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0297885431784816e-05,
      "loss": 0.391724967956543,
      "num_input_tokens_seen": 14412567104,
      "step": 50680,
      "train_runtime": 492745.1343,
      "train_tokens_per_second": 29249.537
    },
    {
      "epoch": 1.7935127054645328,
      "grad_norm": 0.625,
      "learning_rate": 2.0296213075868243e-05,
      "loss": 0.38786237239837645,
      "num_input_tokens_seen": 14415413184,
      "step": 50690,
      "train_runtime": 492842.4688,
      "train_tokens_per_second": 29249.535
    },
    {
      "epoch": 1.7938665257282618,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0294541133243025e-05,
      "loss": 0.38762688636779785,
      "num_input_tokens_seen": 14418257472,
      "step": 50700,
      "train_runtime": 492938.2576,
      "train_tokens_per_second": 29249.622
    },
    {
      "epoch": 1.7942203459919903,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0292869603738964e-05,
      "loss": 0.3917722702026367,
      "num_input_tokens_seen": 14421087296,
      "step": 50710,
      "train_runtime": 493035.0339,
      "train_tokens_per_second": 29249.62
    },
    {
      "epoch": 1.7945741662557193,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0291198487185956e-05,
      "loss": 0.3887815237045288,
      "num_input_tokens_seen": 14423951360,
      "step": 50720,
      "train_runtime": 493132.4879,
      "train_tokens_per_second": 29249.647
    },
    {
      "epoch": 1.7949279865194478,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0289527783413993e-05,
      "loss": 0.3890718460083008,
      "num_input_tokens_seen": 14426803840,
      "step": 50730,
      "train_runtime": 493226.9739,
      "train_tokens_per_second": 29249.827
    },
    {
      "epoch": 1.7952818067831768,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0287857492253182e-05,
      "loss": 0.39197487831115724,
      "num_input_tokens_seen": 14429652032,
      "step": 50740,
      "train_runtime": 493322.6958,
      "train_tokens_per_second": 29249.925
    },
    {
      "epoch": 1.7956356270469054,
      "grad_norm": 0.578125,
      "learning_rate": 2.0286187613533696e-05,
      "loss": 0.388737154006958,
      "num_input_tokens_seen": 14432463744,
      "step": 50750,
      "train_runtime": 493417.379,
      "train_tokens_per_second": 29250.011
    },
    {
      "epoch": 1.7959894473106344,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0284518147085836e-05,
      "loss": 0.3835196256637573,
      "num_input_tokens_seen": 14435292544,
      "step": 50760,
      "train_runtime": 493514.6831,
      "train_tokens_per_second": 29249.976
    },
    {
      "epoch": 1.796343267574363,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0282849092739986e-05,
      "loss": 0.39655866622924807,
      "num_input_tokens_seen": 14438149312,
      "step": 50770,
      "train_runtime": 493613.1606,
      "train_tokens_per_second": 29249.928
    },
    {
      "epoch": 1.796697087838092,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0281180450326632e-05,
      "loss": 0.38693928718566895,
      "num_input_tokens_seen": 14441028352,
      "step": 50780,
      "train_runtime": 493714.0791,
      "train_tokens_per_second": 29249.78
    },
    {
      "epoch": 1.7970509081018207,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0279512219676352e-05,
      "loss": 0.3872836589813232,
      "num_input_tokens_seen": 14443823424,
      "step": 50790,
      "train_runtime": 493807.194,
      "train_tokens_per_second": 29249.925
    },
    {
      "epoch": 1.7974047283655494,
      "grad_norm": 0.58984375,
      "learning_rate": 2.027784440061983e-05,
      "loss": 0.39070749282836914,
      "num_input_tokens_seen": 14446639488,
      "step": 50800,
      "train_runtime": 493904.4922,
      "train_tokens_per_second": 29249.865
    },
    {
      "epoch": 1.7977585486292782,
      "grad_norm": 0.6328125,
      "learning_rate": 2.0276176992987846e-05,
      "loss": 0.38802294731140136,
      "num_input_tokens_seen": 14449491200,
      "step": 50810,
      "train_runtime": 494003.0615,
      "train_tokens_per_second": 29249.801
    },
    {
      "epoch": 1.798112368893007,
      "grad_norm": 0.62109375,
      "learning_rate": 2.027450999661126e-05,
      "loss": 0.39028615951538087,
      "num_input_tokens_seen": 14452334592,
      "step": 50820,
      "train_runtime": 494098.4464,
      "train_tokens_per_second": 29249.909
    },
    {
      "epoch": 1.7984661891567357,
      "grad_norm": 0.58984375,
      "learning_rate": 2.027284341132106e-05,
      "loss": 0.39107933044433596,
      "num_input_tokens_seen": 14455210944,
      "step": 50830,
      "train_runtime": 494198.7975,
      "train_tokens_per_second": 29249.79
    },
    {
      "epoch": 1.7988200094204645,
      "grad_norm": 0.609375,
      "learning_rate": 2.0271177236948313e-05,
      "loss": 0.3896492958068848,
      "num_input_tokens_seen": 14458058048,
      "step": 50840,
      "train_runtime": 494295.6551,
      "train_tokens_per_second": 29249.818
    },
    {
      "epoch": 1.7991738296841933,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0269511473324176e-05,
      "loss": 0.3889066934585571,
      "num_input_tokens_seen": 14460942528,
      "step": 50850,
      "train_runtime": 494396.2546,
      "train_tokens_per_second": 29249.701
    },
    {
      "epoch": 1.799527649947922,
      "grad_norm": 0.578125,
      "learning_rate": 2.0267846120279927e-05,
      "loss": 0.3871904373168945,
      "num_input_tokens_seen": 14463786432,
      "step": 50860,
      "train_runtime": 494491.7351,
      "train_tokens_per_second": 29249.804
    },
    {
      "epoch": 1.799881470211651,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0266181177646913e-05,
      "loss": 0.3901465177536011,
      "num_input_tokens_seen": 14466568000,
      "step": 50870,
      "train_runtime": 494584.0623,
      "train_tokens_per_second": 29249.968
    },
    {
      "epoch": 1.8002352904753796,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0264516645256593e-05,
      "loss": 0.3877096652984619,
      "num_input_tokens_seen": 14469439616,
      "step": 50880,
      "train_runtime": 494682.7995,
      "train_tokens_per_second": 29249.935
    },
    {
      "epoch": 1.8005891107391085,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0262852522940528e-05,
      "loss": 0.3882959604263306,
      "num_input_tokens_seen": 14472290816,
      "step": 50890,
      "train_runtime": 494775.3336,
      "train_tokens_per_second": 29250.227
    },
    {
      "epoch": 1.800942931002837,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0261188810530366e-05,
      "loss": 0.39019896984100344,
      "num_input_tokens_seen": 14475117760,
      "step": 50900,
      "train_runtime": 494872.4894,
      "train_tokens_per_second": 29250.197
    },
    {
      "epoch": 1.801296751266566,
      "grad_norm": 0.59375,
      "learning_rate": 2.025952550785785e-05,
      "loss": 0.3897997379302979,
      "num_input_tokens_seen": 14478041856,
      "step": 50910,
      "train_runtime": 494976.8509,
      "train_tokens_per_second": 29249.937
    },
    {
      "epoch": 1.8016505715302946,
      "grad_norm": 0.609375,
      "learning_rate": 2.0257862614754837e-05,
      "loss": 0.38805365562438965,
      "num_input_tokens_seen": 14480872128,
      "step": 50920,
      "train_runtime": 495075.6908,
      "train_tokens_per_second": 29249.815
    },
    {
      "epoch": 1.8020043917940236,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0256200131053257e-05,
      "loss": 0.3874636173248291,
      "num_input_tokens_seen": 14483709056,
      "step": 50930,
      "train_runtime": 495171.2177,
      "train_tokens_per_second": 29249.901
    },
    {
      "epoch": 1.8023582120577522,
      "grad_norm": 0.59765625,
      "learning_rate": 2.025453805658515e-05,
      "loss": 0.3914686679840088,
      "num_input_tokens_seen": 14486550784,
      "step": 50940,
      "train_runtime": 495268.0834,
      "train_tokens_per_second": 29249.918
    },
    {
      "epoch": 1.8027120323214811,
      "grad_norm": 0.609375,
      "learning_rate": 2.0252876391182653e-05,
      "loss": 0.38584802150726316,
      "num_input_tokens_seen": 14489419584,
      "step": 50950,
      "train_runtime": 495364.9799,
      "train_tokens_per_second": 29249.988
    },
    {
      "epoch": 1.80306585258521,
      "grad_norm": 0.625,
      "learning_rate": 2.0251215134677996e-05,
      "loss": 0.38669064044952395,
      "num_input_tokens_seen": 14492228352,
      "step": 50960,
      "train_runtime": 495458.7484,
      "train_tokens_per_second": 29250.121
    },
    {
      "epoch": 1.8034196728489387,
      "grad_norm": 0.6015625,
      "learning_rate": 2.02495542869035e-05,
      "loss": 0.3868887901306152,
      "num_input_tokens_seen": 14495119616,
      "step": 50970,
      "train_runtime": 495557.0897,
      "train_tokens_per_second": 29250.151
    },
    {
      "epoch": 1.8037734931126674,
      "grad_norm": 0.6328125,
      "learning_rate": 2.0247893847691597e-05,
      "loss": 0.38996105194091796,
      "num_input_tokens_seen": 14497977664,
      "step": 50980,
      "train_runtime": 495655.7573,
      "train_tokens_per_second": 29250.094
    },
    {
      "epoch": 1.8041273133763962,
      "grad_norm": 0.59765625,
      "learning_rate": 2.02462338168748e-05,
      "loss": 0.38741815090179443,
      "num_input_tokens_seen": 14500845632,
      "step": 50990,
      "train_runtime": 495753.7292,
      "train_tokens_per_second": 29250.099
    },
    {
      "epoch": 1.804481133640125,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0244574194285724e-05,
      "loss": 0.387392520904541,
      "num_input_tokens_seen": 14503726720,
      "step": 51000,
      "train_runtime": 495851.7451,
      "train_tokens_per_second": 29250.127
    },
    {
      "epoch": 1.8048349539038537,
      "grad_norm": 0.59375,
      "learning_rate": 2.0242914979757085e-05,
      "loss": 0.3885485649108887,
      "num_input_tokens_seen": 14506607104,
      "step": 51010,
      "train_runtime": 495951.4604,
      "train_tokens_per_second": 29250.054
    },
    {
      "epoch": 1.8051887741675825,
      "grad_norm": 0.609375,
      "learning_rate": 2.024125617312169e-05,
      "loss": 0.38977670669555664,
      "num_input_tokens_seen": 14509458560,
      "step": 51020,
      "train_runtime": 496048.1631,
      "train_tokens_per_second": 29250.1
    },
    {
      "epoch": 1.8055425944313113,
      "grad_norm": 0.609375,
      "learning_rate": 2.0239597774212434e-05,
      "loss": 0.3883575201034546,
      "num_input_tokens_seen": 14512300352,
      "step": 51030,
      "train_runtime": 496145.2953,
      "train_tokens_per_second": 29250.102
    },
    {
      "epoch": 1.8058964146950403,
      "grad_norm": 0.62890625,
      "learning_rate": 2.023793978286233e-05,
      "loss": 0.39366955757141114,
      "num_input_tokens_seen": 14515180032,
      "step": 51040,
      "train_runtime": 496244.5307,
      "train_tokens_per_second": 29250.055
    },
    {
      "epoch": 1.8062502349587688,
      "grad_norm": 0.609375,
      "learning_rate": 2.023628219890446e-05,
      "loss": 0.38989949226379395,
      "num_input_tokens_seen": 14518073408,
      "step": 51050,
      "train_runtime": 496344.502,
      "train_tokens_per_second": 29249.993
    },
    {
      "epoch": 1.8066040552224978,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0234625022172022e-05,
      "loss": 0.39272136688232423,
      "num_input_tokens_seen": 14520882176,
      "step": 51060,
      "train_runtime": 496440.6228,
      "train_tokens_per_second": 29249.988
    },
    {
      "epoch": 1.8069578754862263,
      "grad_norm": 0.59375,
      "learning_rate": 2.0232968252498296e-05,
      "loss": 0.38747172355651854,
      "num_input_tokens_seen": 14523785792,
      "step": 51070,
      "train_runtime": 496541.3672,
      "train_tokens_per_second": 29249.901
    },
    {
      "epoch": 1.8073116957499553,
      "grad_norm": 0.5859375,
      "learning_rate": 2.0231311889716673e-05,
      "loss": 0.38672194480895994,
      "num_input_tokens_seen": 14526678976,
      "step": 51080,
      "train_runtime": 496642.4783,
      "train_tokens_per_second": 29249.771
    },
    {
      "epoch": 1.8076655160136839,
      "grad_norm": 0.59765625,
      "learning_rate": 2.022965593366062e-05,
      "loss": 0.3924574851989746,
      "num_input_tokens_seen": 14529515904,
      "step": 51090,
      "train_runtime": 496739.2706,
      "train_tokens_per_second": 29249.783
    },
    {
      "epoch": 1.8080193362774128,
      "grad_norm": 0.625,
      "learning_rate": 2.0228000384163716e-05,
      "loss": 0.38303341865539553,
      "num_input_tokens_seen": 14532425728,
      "step": 51100,
      "train_runtime": 496838.0788,
      "train_tokens_per_second": 29249.823
    },
    {
      "epoch": 1.8083731565411414,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0226345241059623e-05,
      "loss": 0.38330740928649903,
      "num_input_tokens_seen": 14535256000,
      "step": 51110,
      "train_runtime": 496933.1687,
      "train_tokens_per_second": 29249.921
    },
    {
      "epoch": 1.8087269768048704,
      "grad_norm": 0.6015625,
      "learning_rate": 2.022469050418211e-05,
      "loss": 0.38721954822540283,
      "num_input_tokens_seen": 14538068928,
      "step": 51120,
      "train_runtime": 497028.3479,
      "train_tokens_per_second": 29249.979
    },
    {
      "epoch": 1.8090807970685991,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0223036173365033e-05,
      "loss": 0.3908969879150391,
      "num_input_tokens_seen": 14540915776,
      "step": 51130,
      "train_runtime": 497125.9455,
      "train_tokens_per_second": 29249.964
    },
    {
      "epoch": 1.809434617332328,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0221382248442346e-05,
      "loss": 0.39102463722229003,
      "num_input_tokens_seen": 14543712704,
      "step": 51140,
      "train_runtime": 497220.2099,
      "train_tokens_per_second": 29250.043
    },
    {
      "epoch": 1.8097884375960567,
      "grad_norm": 0.6328125,
      "learning_rate": 2.0219728729248095e-05,
      "loss": 0.3879990577697754,
      "num_input_tokens_seen": 14546570112,
      "step": 51150,
      "train_runtime": 497318.0571,
      "train_tokens_per_second": 29250.034
    },
    {
      "epoch": 1.8101422578597854,
      "grad_norm": 0.59375,
      "learning_rate": 2.0218075615616425e-05,
      "loss": 0.3864192724227905,
      "num_input_tokens_seen": 14549390400,
      "step": 51160,
      "train_runtime": 497413.9602,
      "train_tokens_per_second": 29250.064
    },
    {
      "epoch": 1.8104960781235142,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0216422907381577e-05,
      "loss": 0.3912806034088135,
      "num_input_tokens_seen": 14552233984,
      "step": 51170,
      "train_runtime": 497510.5844,
      "train_tokens_per_second": 29250.099
    },
    {
      "epoch": 1.810849898387243,
      "grad_norm": 0.6015625,
      "learning_rate": 2.021477060437788e-05,
      "loss": 0.3887171268463135,
      "num_input_tokens_seen": 14555083712,
      "step": 51180,
      "train_runtime": 497607.9912,
      "train_tokens_per_second": 29250.1
    },
    {
      "epoch": 1.8112037186509717,
      "grad_norm": 0.63671875,
      "learning_rate": 2.021311870643976e-05,
      "loss": 0.38593878746032717,
      "num_input_tokens_seen": 14557916544,
      "step": 51190,
      "train_runtime": 497703.4896,
      "train_tokens_per_second": 29250.18
    },
    {
      "epoch": 1.8115575389147005,
      "grad_norm": 0.609375,
      "learning_rate": 2.0211467213401743e-05,
      "loss": 0.39083452224731446,
      "num_input_tokens_seen": 14560795200,
      "step": 51200,
      "train_runtime": 497804.4282,
      "train_tokens_per_second": 29250.031
    },
    {
      "epoch": 1.8119113591784295,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0209816125098453e-05,
      "loss": 0.3883590459823608,
      "num_input_tokens_seen": 14563599168,
      "step": 51210,
      "train_runtime": 497899.0444,
      "train_tokens_per_second": 29250.105
    },
    {
      "epoch": 1.812265179442158,
      "grad_norm": 0.62890625,
      "learning_rate": 2.020816544136459e-05,
      "loss": 0.38519399166107177,
      "num_input_tokens_seen": 14566429632,
      "step": 51220,
      "train_runtime": 497994.5208,
      "train_tokens_per_second": 29250.181
    },
    {
      "epoch": 1.812618999705887,
      "grad_norm": 0.625,
      "learning_rate": 2.0206515162034965e-05,
      "loss": 0.38746793270111085,
      "num_input_tokens_seen": 14569262720,
      "step": 51230,
      "train_runtime": 498090.7423,
      "train_tokens_per_second": 29250.218
    },
    {
      "epoch": 1.8129728199696156,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0204865286944477e-05,
      "loss": 0.38526105880737305,
      "num_input_tokens_seen": 14572097408,
      "step": 51240,
      "train_runtime": 498189.7699,
      "train_tokens_per_second": 29250.094
    },
    {
      "epoch": 1.8133266402333446,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0203215815928123e-05,
      "loss": 0.38857054710388184,
      "num_input_tokens_seen": 14574926208,
      "step": 51250,
      "train_runtime": 498286.82,
      "train_tokens_per_second": 29250.074
    },
    {
      "epoch": 1.813680460497073,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0201566748820995e-05,
      "loss": 0.38366098403930665,
      "num_input_tokens_seen": 14577741888,
      "step": 51260,
      "train_runtime": 498381.9565,
      "train_tokens_per_second": 29250.14
    },
    {
      "epoch": 1.814034280760802,
      "grad_norm": 0.625,
      "learning_rate": 2.019991808545827e-05,
      "loss": 0.3872786521911621,
      "num_input_tokens_seen": 14580565376,
      "step": 51270,
      "train_runtime": 498478.4629,
      "train_tokens_per_second": 29250.141
    },
    {
      "epoch": 1.8143881010245309,
      "grad_norm": 0.59765625,
      "learning_rate": 2.019826982567523e-05,
      "loss": 0.3879929304122925,
      "num_input_tokens_seen": 14583456704,
      "step": 51280,
      "train_runtime": 498579.5898,
      "train_tokens_per_second": 29250.007
    },
    {
      "epoch": 1.8147419212882596,
      "grad_norm": 0.6328125,
      "learning_rate": 2.019662196930725e-05,
      "loss": 0.39245221614837644,
      "num_input_tokens_seen": 14586282368,
      "step": 51290,
      "train_runtime": 498678.8841,
      "train_tokens_per_second": 29249.85
    },
    {
      "epoch": 1.8150957415519884,
      "grad_norm": 0.609375,
      "learning_rate": 2.0194974516189784e-05,
      "loss": 0.3849053382873535,
      "num_input_tokens_seen": 14589166144,
      "step": 51300,
      "train_runtime": 498778.6819,
      "train_tokens_per_second": 29249.779
    },
    {
      "epoch": 1.8154495618157171,
      "grad_norm": 0.6171875,
      "learning_rate": 2.019332746615841e-05,
      "loss": 0.38883171081542967,
      "num_input_tokens_seen": 14592027904,
      "step": 51310,
      "train_runtime": 498876.9275,
      "train_tokens_per_second": 29249.755
    },
    {
      "epoch": 1.815803382079446,
      "grad_norm": 0.6328125,
      "learning_rate": 2.0191680819048758e-05,
      "loss": 0.38864936828613283,
      "num_input_tokens_seen": 14594885632,
      "step": 51320,
      "train_runtime": 498974.4771,
      "train_tokens_per_second": 29249.764
    },
    {
      "epoch": 1.8161572023431747,
      "grad_norm": 0.625,
      "learning_rate": 2.0190034574696592e-05,
      "loss": 0.3953663110733032,
      "num_input_tokens_seen": 14597692416,
      "step": 51330,
      "train_runtime": 499069.0467,
      "train_tokens_per_second": 29249.845
    },
    {
      "epoch": 1.8165110226069034,
      "grad_norm": 0.609375,
      "learning_rate": 2.0188388732937753e-05,
      "loss": 0.3909308433532715,
      "num_input_tokens_seen": 14600551936,
      "step": 51340,
      "train_runtime": 499168.0665,
      "train_tokens_per_second": 29249.772
    },
    {
      "epoch": 1.8168648428706322,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0186743293608164e-05,
      "loss": 0.3911138534545898,
      "num_input_tokens_seen": 14603393216,
      "step": 51350,
      "train_runtime": 499263.8772,
      "train_tokens_per_second": 29249.849
    },
    {
      "epoch": 1.817218663134361,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0185098256543866e-05,
      "loss": 0.39126052856445315,
      "num_input_tokens_seen": 14606230912,
      "step": 51360,
      "train_runtime": 499362.5553,
      "train_tokens_per_second": 29249.752
    },
    {
      "epoch": 1.8175724833980897,
      "grad_norm": 0.609375,
      "learning_rate": 2.018345362158098e-05,
      "loss": 0.3843747854232788,
      "num_input_tokens_seen": 14609116672,
      "step": 51370,
      "train_runtime": 499460.4393,
      "train_tokens_per_second": 29249.797
    },
    {
      "epoch": 1.8179263036618187,
      "grad_norm": 0.58203125,
      "learning_rate": 2.018180938855571e-05,
      "loss": 0.38645114898681643,
      "num_input_tokens_seen": 14611964800,
      "step": 51380,
      "train_runtime": 499558.0418,
      "train_tokens_per_second": 29249.784
    },
    {
      "epoch": 1.8182801239255473,
      "grad_norm": 0.65234375,
      "learning_rate": 2.0180165557304377e-05,
      "loss": 0.38966965675354004,
      "num_input_tokens_seen": 14614870656,
      "step": 51390,
      "train_runtime": 499656.6361,
      "train_tokens_per_second": 29249.828
    },
    {
      "epoch": 1.8186339441892763,
      "grad_norm": 0.63671875,
      "learning_rate": 2.0178522127663373e-05,
      "loss": 0.38861324787139895,
      "num_input_tokens_seen": 14617749760,
      "step": 51400,
      "train_runtime": 499754.2409,
      "train_tokens_per_second": 29249.876
    },
    {
      "epoch": 1.8189877644530048,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0176879099469196e-05,
      "loss": 0.3853743553161621,
      "num_input_tokens_seen": 14620598784,
      "step": 51410,
      "train_runtime": 499852.5015,
      "train_tokens_per_second": 29249.826
    },
    {
      "epoch": 1.8193415847167338,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0175236472558443e-05,
      "loss": 0.3880643129348755,
      "num_input_tokens_seen": 14623430720,
      "step": 51420,
      "train_runtime": 499948.6648,
      "train_tokens_per_second": 29249.865
    },
    {
      "epoch": 1.8196954049804623,
      "grad_norm": 0.6328125,
      "learning_rate": 2.017359424676778e-05,
      "loss": 0.39051432609558107,
      "num_input_tokens_seen": 14626291456,
      "step": 51430,
      "train_runtime": 500046.9283,
      "train_tokens_per_second": 29249.838
    },
    {
      "epoch": 1.8200492252441913,
      "grad_norm": 0.63671875,
      "learning_rate": 2.0171952421933997e-05,
      "loss": 0.3882394552230835,
      "num_input_tokens_seen": 14629185536,
      "step": 51440,
      "train_runtime": 500147.3461,
      "train_tokens_per_second": 29249.751
    },
    {
      "epoch": 1.82040304550792,
      "grad_norm": 0.59375,
      "learning_rate": 2.017031099789395e-05,
      "loss": 0.3860489368438721,
      "num_input_tokens_seen": 14631982272,
      "step": 51450,
      "train_runtime": 500244.0866,
      "train_tokens_per_second": 29249.686
    },
    {
      "epoch": 1.8207568657716489,
      "grad_norm": 0.59375,
      "learning_rate": 2.0168669974484606e-05,
      "loss": 0.3885204792022705,
      "num_input_tokens_seen": 14634767872,
      "step": 51460,
      "train_runtime": 500335.4841,
      "train_tokens_per_second": 29249.91
    },
    {
      "epoch": 1.8211106860353776,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0167029351543015e-05,
      "loss": 0.39079971313476564,
      "num_input_tokens_seen": 14637595008,
      "step": 51470,
      "train_runtime": 500432.4641,
      "train_tokens_per_second": 29249.891
    },
    {
      "epoch": 1.8214645062991064,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0165389128906324e-05,
      "loss": 0.3891436576843262,
      "num_input_tokens_seen": 14640451712,
      "step": 51480,
      "train_runtime": 500529.9603,
      "train_tokens_per_second": 29249.901
    },
    {
      "epoch": 1.8218183265628352,
      "grad_norm": 0.625,
      "learning_rate": 2.0163749306411775e-05,
      "loss": 0.3888471364974976,
      "num_input_tokens_seen": 14643309056,
      "step": 51490,
      "train_runtime": 500631.0391,
      "train_tokens_per_second": 29249.703
    },
    {
      "epoch": 1.822172146826564,
      "grad_norm": 0.59375,
      "learning_rate": 2.0162109883896694e-05,
      "loss": 0.3852571725845337,
      "num_input_tokens_seen": 14646130432,
      "step": 51500,
      "train_runtime": 500728.5953,
      "train_tokens_per_second": 29249.639
    },
    {
      "epoch": 1.8225259670902927,
      "grad_norm": 0.5703125,
      "learning_rate": 2.016047086119851e-05,
      "loss": 0.3879885196685791,
      "num_input_tokens_seen": 14649022720,
      "step": 51510,
      "train_runtime": 500828.2873,
      "train_tokens_per_second": 29249.591
    },
    {
      "epoch": 1.8228797873540215,
      "grad_norm": 0.63671875,
      "learning_rate": 2.015883223815473e-05,
      "loss": 0.38724889755249026,
      "num_input_tokens_seen": 14651866240,
      "step": 51520,
      "train_runtime": 500924.9569,
      "train_tokens_per_second": 29249.623
    },
    {
      "epoch": 1.8232336076177504,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0157194014602974e-05,
      "loss": 0.3896310329437256,
      "num_input_tokens_seen": 14654656192,
      "step": 51530,
      "train_runtime": 501020.631,
      "train_tokens_per_second": 29249.606
    },
    {
      "epoch": 1.823587427881479,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0155556190380938e-05,
      "loss": 0.39223451614379884,
      "num_input_tokens_seen": 14657508992,
      "step": 51540,
      "train_runtime": 501118.8798,
      "train_tokens_per_second": 29249.564
    },
    {
      "epoch": 1.823941248145208,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0153918765326414e-05,
      "loss": 0.3866464853286743,
      "num_input_tokens_seen": 14660358720,
      "step": 51550,
      "train_runtime": 501218.5585,
      "train_tokens_per_second": 29249.433
    },
    {
      "epoch": 1.8242950684089365,
      "grad_norm": 0.6328125,
      "learning_rate": 2.0152281739277286e-05,
      "loss": 0.38678860664367676,
      "num_input_tokens_seen": 14663172096,
      "step": 51560,
      "train_runtime": 501313.8766,
      "train_tokens_per_second": 29249.484
    },
    {
      "epoch": 1.8246488886726655,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0150645112071542e-05,
      "loss": 0.38818650245666503,
      "num_input_tokens_seen": 14666016192,
      "step": 51570,
      "train_runtime": 501411.9656,
      "train_tokens_per_second": 29249.434
    },
    {
      "epoch": 1.825002708936394,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0149008883547245e-05,
      "loss": 0.38543081283569336,
      "num_input_tokens_seen": 14668849408,
      "step": 51580,
      "train_runtime": 501509.2997,
      "train_tokens_per_second": 29249.407
    },
    {
      "epoch": 1.825356529200123,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0147373053542558e-05,
      "loss": 0.3867804527282715,
      "num_input_tokens_seen": 14671720320,
      "step": 51590,
      "train_runtime": 501608.2592,
      "train_tokens_per_second": 29249.36
    },
    {
      "epoch": 1.8257103494638516,
      "grad_norm": 0.62890625,
      "learning_rate": 2.014573762189573e-05,
      "loss": 0.39119749069213866,
      "num_input_tokens_seen": 14674547072,
      "step": 51600,
      "train_runtime": 501703.147,
      "train_tokens_per_second": 29249.462
    },
    {
      "epoch": 1.8260641697275806,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0144102588445116e-05,
      "loss": 0.39220762252807617,
      "num_input_tokens_seen": 14677419392,
      "step": 51610,
      "train_runtime": 501803.2585,
      "train_tokens_per_second": 29249.351
    },
    {
      "epoch": 1.8264179899913093,
      "grad_norm": 0.62109375,
      "learning_rate": 2.014246795302915e-05,
      "loss": 0.38856210708618166,
      "num_input_tokens_seen": 14680263424,
      "step": 51620,
      "train_runtime": 501902.3775,
      "train_tokens_per_second": 29249.241
    },
    {
      "epoch": 1.826771810255038,
      "grad_norm": 0.6328125,
      "learning_rate": 2.0140833715486358e-05,
      "loss": 0.39266233444213866,
      "num_input_tokens_seen": 14683072576,
      "step": 51630,
      "train_runtime": 501996.335,
      "train_tokens_per_second": 29249.362
    },
    {
      "epoch": 1.8271256305187669,
      "grad_norm": 0.61328125,
      "learning_rate": 2.013919987565536e-05,
      "loss": 0.3925272703170776,
      "num_input_tokens_seen": 14685962816,
      "step": 51640,
      "train_runtime": 502097.4776,
      "train_tokens_per_second": 29249.226
    },
    {
      "epoch": 1.8274794507824956,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0137566433374875e-05,
      "loss": 0.3927711009979248,
      "num_input_tokens_seen": 14688814784,
      "step": 51650,
      "train_runtime": 502193.8079,
      "train_tokens_per_second": 29249.295
    },
    {
      "epoch": 1.8278332710462244,
      "grad_norm": 0.6328125,
      "learning_rate": 2.0135933388483708e-05,
      "loss": 0.3837876796722412,
      "num_input_tokens_seen": 14691624512,
      "step": 51660,
      "train_runtime": 502288.6677,
      "train_tokens_per_second": 29249.365
    },
    {
      "epoch": 1.8281870913099532,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0134300740820746e-05,
      "loss": 0.3872742414474487,
      "num_input_tokens_seen": 14694446016,
      "step": 51670,
      "train_runtime": 502384.053,
      "train_tokens_per_second": 29249.428
    },
    {
      "epoch": 1.828540911573682,
      "grad_norm": 0.59375,
      "learning_rate": 2.0132668490224984e-05,
      "loss": 0.3876617431640625,
      "num_input_tokens_seen": 14697304640,
      "step": 51680,
      "train_runtime": 502481.7891,
      "train_tokens_per_second": 29249.427
    },
    {
      "epoch": 1.8288947318374107,
      "grad_norm": 0.5703125,
      "learning_rate": 2.0131036636535498e-05,
      "loss": 0.38526296615600586,
      "num_input_tokens_seen": 14700168192,
      "step": 51690,
      "train_runtime": 502580.5036,
      "train_tokens_per_second": 29249.38
    },
    {
      "epoch": 1.8292485521011397,
      "grad_norm": 0.6171875,
      "learning_rate": 2.012940517959146e-05,
      "loss": 0.38771839141845704,
      "num_input_tokens_seen": 14703018816,
      "step": 51700,
      "train_runtime": 502675.857,
      "train_tokens_per_second": 29249.503
    },
    {
      "epoch": 1.8296023723648682,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0127774119232128e-05,
      "loss": 0.3868736743927002,
      "num_input_tokens_seen": 14705870720,
      "step": 51710,
      "train_runtime": 502772.4787,
      "train_tokens_per_second": 29249.554
    },
    {
      "epoch": 1.8299561926285972,
      "grad_norm": 0.609375,
      "learning_rate": 2.0126143455296853e-05,
      "loss": 0.388594388961792,
      "num_input_tokens_seen": 14708772032,
      "step": 51720,
      "train_runtime": 502872.9558,
      "train_tokens_per_second": 29249.479
    },
    {
      "epoch": 1.8303100128923258,
      "grad_norm": 0.60546875,
      "learning_rate": 2.012451318762508e-05,
      "loss": 0.3878726243972778,
      "num_input_tokens_seen": 14711634496,
      "step": 51730,
      "train_runtime": 502970.4788,
      "train_tokens_per_second": 29249.499
    },
    {
      "epoch": 1.8306638331560547,
      "grad_norm": 0.609375,
      "learning_rate": 2.0122883316056348e-05,
      "loss": 0.38696799278259275,
      "num_input_tokens_seen": 14714460160,
      "step": 51740,
      "train_runtime": 503066.7496,
      "train_tokens_per_second": 29249.518
    },
    {
      "epoch": 1.8310176534197833,
      "grad_norm": 0.578125,
      "learning_rate": 2.0121253840430276e-05,
      "loss": 0.3870852470397949,
      "num_input_tokens_seen": 14717380800,
      "step": 51750,
      "train_runtime": 503169.045,
      "train_tokens_per_second": 29249.376
    },
    {
      "epoch": 1.8313714736835123,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0119624760586583e-05,
      "loss": 0.3874931335449219,
      "num_input_tokens_seen": 14720253952,
      "step": 51760,
      "train_runtime": 503267.78,
      "train_tokens_per_second": 29249.347
    },
    {
      "epoch": 1.8317252939472408,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0117996076365077e-05,
      "loss": 0.38912889957427976,
      "num_input_tokens_seen": 14723105536,
      "step": 51770,
      "train_runtime": 503365.1076,
      "train_tokens_per_second": 29249.357
    },
    {
      "epoch": 1.8320791142109698,
      "grad_norm": 0.5859375,
      "learning_rate": 2.0116367787605653e-05,
      "loss": 0.38758020401000975,
      "num_input_tokens_seen": 14725951360,
      "step": 51780,
      "train_runtime": 503462.9393,
      "train_tokens_per_second": 29249.325
    },
    {
      "epoch": 1.8324329344746986,
      "grad_norm": 0.625,
      "learning_rate": 2.01147398941483e-05,
      "loss": 0.38751816749572754,
      "num_input_tokens_seen": 14728826624,
      "step": 51790,
      "train_runtime": 503561.3377,
      "train_tokens_per_second": 29249.32
    },
    {
      "epoch": 1.8327867547384273,
      "grad_norm": 0.6015625,
      "learning_rate": 2.01131123958331e-05,
      "loss": 0.3913458824157715,
      "num_input_tokens_seen": 14731645760,
      "step": 51800,
      "train_runtime": 503653.669,
      "train_tokens_per_second": 29249.555
    },
    {
      "epoch": 1.833140575002156,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0111485292500217e-05,
      "loss": 0.38903346061706545,
      "num_input_tokens_seen": 14734529472,
      "step": 51810,
      "train_runtime": 503752.0737,
      "train_tokens_per_second": 29249.566
    },
    {
      "epoch": 1.8334943952658849,
      "grad_norm": 0.61328125,
      "learning_rate": 2.010985858398992e-05,
      "loss": 0.38600361347198486,
      "num_input_tokens_seen": 14737382208,
      "step": 51820,
      "train_runtime": 503848.3798,
      "train_tokens_per_second": 29249.637
    },
    {
      "epoch": 1.8338482155296136,
      "grad_norm": 0.63671875,
      "learning_rate": 2.0108232270142547e-05,
      "loss": 0.38599767684936526,
      "num_input_tokens_seen": 14740214656,
      "step": 51830,
      "train_runtime": 503945.8543,
      "train_tokens_per_second": 29249.6
    },
    {
      "epoch": 1.8342020357933424,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0106606350798547e-05,
      "loss": 0.39305527210235597,
      "num_input_tokens_seen": 14743062976,
      "step": 51840,
      "train_runtime": 504044.6211,
      "train_tokens_per_second": 29249.52
    },
    {
      "epoch": 1.8345558560570712,
      "grad_norm": 0.58203125,
      "learning_rate": 2.0104980825798458e-05,
      "loss": 0.383164644241333,
      "num_input_tokens_seen": 14745890688,
      "step": 51850,
      "train_runtime": 504137.9903,
      "train_tokens_per_second": 29249.711
    },
    {
      "epoch": 1.8349096763208,
      "grad_norm": 0.640625,
      "learning_rate": 2.0103355694982888e-05,
      "loss": 0.3885310649871826,
      "num_input_tokens_seen": 14748708992,
      "step": 51860,
      "train_runtime": 504232.4301,
      "train_tokens_per_second": 29249.822
    },
    {
      "epoch": 1.835263496584529,
      "grad_norm": 0.6171875,
      "learning_rate": 2.010173095819256e-05,
      "loss": 0.3892142057418823,
      "num_input_tokens_seen": 14751567872,
      "step": 51870,
      "train_runtime": 504332.7875,
      "train_tokens_per_second": 29249.671
    },
    {
      "epoch": 1.8356173168482575,
      "grad_norm": 0.59765625,
      "learning_rate": 2.0100106615268267e-05,
      "loss": 0.3875137805938721,
      "num_input_tokens_seen": 14754377344,
      "step": 51880,
      "train_runtime": 504429.2492,
      "train_tokens_per_second": 29249.647
    },
    {
      "epoch": 1.8359711371119865,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0098482666050902e-05,
      "loss": 0.3896777868270874,
      "num_input_tokens_seen": 14757212480,
      "step": 51890,
      "train_runtime": 504529.0345,
      "train_tokens_per_second": 29249.481
    },
    {
      "epoch": 1.836324957375715,
      "grad_norm": 0.5859375,
      "learning_rate": 2.009685911038145e-05,
      "loss": 0.3893177270889282,
      "num_input_tokens_seen": 14760058624,
      "step": 51900,
      "train_runtime": 504625.389,
      "train_tokens_per_second": 29249.536
    },
    {
      "epoch": 1.836678777639444,
      "grad_norm": 0.609375,
      "learning_rate": 2.0095235948100982e-05,
      "loss": 0.3884300708770752,
      "num_input_tokens_seen": 14762878976,
      "step": 51910,
      "train_runtime": 504719.8282,
      "train_tokens_per_second": 29249.651
    },
    {
      "epoch": 1.8370325979031725,
      "grad_norm": 0.640625,
      "learning_rate": 2.0093613179050662e-05,
      "loss": 0.3878193378448486,
      "num_input_tokens_seen": 14765716864,
      "step": 51920,
      "train_runtime": 504816.5655,
      "train_tokens_per_second": 29249.668
    },
    {
      "epoch": 1.8373864181669015,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0091990803071733e-05,
      "loss": 0.3849036693572998,
      "num_input_tokens_seen": 14768661952,
      "step": 51930,
      "train_runtime": 504920.4873,
      "train_tokens_per_second": 29249.481
    },
    {
      "epoch": 1.83774023843063,
      "grad_norm": 0.61328125,
      "learning_rate": 2.009036882000554e-05,
      "loss": 0.38760464191436766,
      "num_input_tokens_seen": 14771459328,
      "step": 51940,
      "train_runtime": 505013.5867,
      "train_tokens_per_second": 29249.628
    },
    {
      "epoch": 1.838094058694359,
      "grad_norm": 0.61328125,
      "learning_rate": 2.008874722969352e-05,
      "loss": 0.38884477615356444,
      "num_input_tokens_seen": 14774306688,
      "step": 51950,
      "train_runtime": 505113.5556,
      "train_tokens_per_second": 29249.476
    },
    {
      "epoch": 1.8384478789580878,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0087126031977192e-05,
      "loss": 0.38777933120727537,
      "num_input_tokens_seen": 14777162176,
      "step": 51960,
      "train_runtime": 505210.7378,
      "train_tokens_per_second": 29249.501
    },
    {
      "epoch": 1.8388016992218166,
      "grad_norm": 0.60546875,
      "learning_rate": 2.008550522669815e-05,
      "loss": 0.3837061166763306,
      "num_input_tokens_seen": 14779979264,
      "step": 51970,
      "train_runtime": 505307.3182,
      "train_tokens_per_second": 29249.486
    },
    {
      "epoch": 1.8391555194855453,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0083884813698114e-05,
      "loss": 0.38832640647888184,
      "num_input_tokens_seen": 14782824768,
      "step": 51980,
      "train_runtime": 505403.8279,
      "train_tokens_per_second": 29249.531
    },
    {
      "epoch": 1.839509339749274,
      "grad_norm": 0.62890625,
      "learning_rate": 2.008226479281886e-05,
      "loss": 0.3870819568634033,
      "num_input_tokens_seen": 14785633472,
      "step": 51990,
      "train_runtime": 505499.5477,
      "train_tokens_per_second": 29249.548
    },
    {
      "epoch": 1.8398631600130029,
      "grad_norm": 0.6171875,
      "learning_rate": 2.008064516390227e-05,
      "loss": 0.38431243896484374,
      "num_input_tokens_seen": 14788486272,
      "step": 52000,
      "train_runtime": 505596.1193,
      "train_tokens_per_second": 29249.604
    },
    {
      "epoch": 1.8402169802767316,
      "grad_norm": 0.59375,
      "learning_rate": 2.0079025926790312e-05,
      "loss": 0.38648500442504885,
      "num_input_tokens_seen": 14791354112,
      "step": 52010,
      "train_runtime": 505693.7442,
      "train_tokens_per_second": 29249.628
    },
    {
      "epoch": 1.8405708005404604,
      "grad_norm": 0.609375,
      "learning_rate": 2.007740708132504e-05,
      "loss": 0.3847785472869873,
      "num_input_tokens_seen": 14794183552,
      "step": 52020,
      "train_runtime": 505793.0866,
      "train_tokens_per_second": 29249.478
    },
    {
      "epoch": 1.8409246208041892,
      "grad_norm": 0.640625,
      "learning_rate": 2.0075788627348595e-05,
      "loss": 0.3903635025024414,
      "num_input_tokens_seen": 14797066048,
      "step": 52030,
      "train_runtime": 505891.9421,
      "train_tokens_per_second": 29249.46
    },
    {
      "epoch": 1.8412784410679182,
      "grad_norm": 0.6015625,
      "learning_rate": 2.007417056470322e-05,
      "loss": 0.38654141426086425,
      "num_input_tokens_seen": 14799977664,
      "step": 52040,
      "train_runtime": 505992.337,
      "train_tokens_per_second": 29249.411
    },
    {
      "epoch": 1.8416322613316467,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0072552893231236e-05,
      "loss": 0.3915912389755249,
      "num_input_tokens_seen": 14802843776,
      "step": 52050,
      "train_runtime": 506090.6381,
      "train_tokens_per_second": 29249.393
    },
    {
      "epoch": 1.8419860815953757,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0070935612775056e-05,
      "loss": 0.3833825349807739,
      "num_input_tokens_seen": 14805666496,
      "step": 52060,
      "train_runtime": 506186.0841,
      "train_tokens_per_second": 29249.454
    },
    {
      "epoch": 1.8423399018591042,
      "grad_norm": 0.609375,
      "learning_rate": 2.006931872317718e-05,
      "loss": 0.3913029670715332,
      "num_input_tokens_seen": 14808541888,
      "step": 52070,
      "train_runtime": 506287.0803,
      "train_tokens_per_second": 29249.298
    },
    {
      "epoch": 1.8426937221228332,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0067702224280192e-05,
      "loss": 0.3902284622192383,
      "num_input_tokens_seen": 14811345728,
      "step": 52080,
      "train_runtime": 506380.27,
      "train_tokens_per_second": 29249.453
    },
    {
      "epoch": 1.8430475423865618,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0066086115926782e-05,
      "loss": 0.39210846424102785,
      "num_input_tokens_seen": 14814217152,
      "step": 52090,
      "train_runtime": 506477.9481,
      "train_tokens_per_second": 29249.481
    },
    {
      "epoch": 1.8434013626502908,
      "grad_norm": 0.625,
      "learning_rate": 2.0064470397959713e-05,
      "loss": 0.39340524673461913,
      "num_input_tokens_seen": 14817025984,
      "step": 52100,
      "train_runtime": 506572.8461,
      "train_tokens_per_second": 29249.546
    },
    {
      "epoch": 1.8437551829140195,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0062855070221834e-05,
      "loss": 0.3835534334182739,
      "num_input_tokens_seen": 14819859648,
      "step": 52110,
      "train_runtime": 506669.8526,
      "train_tokens_per_second": 29249.539
    },
    {
      "epoch": 1.8441090031777483,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0061240132556102e-05,
      "loss": 0.3896360158920288,
      "num_input_tokens_seen": 14822672128,
      "step": 52120,
      "train_runtime": 506766.0747,
      "train_tokens_per_second": 29249.535
    },
    {
      "epoch": 1.844462823441477,
      "grad_norm": 0.59375,
      "learning_rate": 2.0059625584805542e-05,
      "loss": 0.38804266452789304,
      "num_input_tokens_seen": 14825601920,
      "step": 52130,
      "train_runtime": 506869.6919,
      "train_tokens_per_second": 29249.336
    },
    {
      "epoch": 1.8448166437052058,
      "grad_norm": 0.578125,
      "learning_rate": 2.0058011426813275e-05,
      "loss": 0.39243288040161134,
      "num_input_tokens_seen": 14828473088,
      "step": 52140,
      "train_runtime": 506970.8204,
      "train_tokens_per_second": 29249.165
    },
    {
      "epoch": 1.8451704639689346,
      "grad_norm": 0.609375,
      "learning_rate": 2.0056397658422517e-05,
      "loss": 0.39058969020843504,
      "num_input_tokens_seen": 14831337536,
      "step": 52150,
      "train_runtime": 507069.6306,
      "train_tokens_per_second": 29249.114
    },
    {
      "epoch": 1.8455242842326633,
      "grad_norm": 0.609375,
      "learning_rate": 2.005478427947656e-05,
      "loss": 0.39193899631500245,
      "num_input_tokens_seen": 14834190336,
      "step": 52160,
      "train_runtime": 507166.6087,
      "train_tokens_per_second": 29249.146
    },
    {
      "epoch": 1.8458781044963921,
      "grad_norm": 0.60546875,
      "learning_rate": 2.005317128981879e-05,
      "loss": 0.3927168846130371,
      "num_input_tokens_seen": 14837054976,
      "step": 52170,
      "train_runtime": 507267.6291,
      "train_tokens_per_second": 29248.969
    },
    {
      "epoch": 1.8462319247601209,
      "grad_norm": 0.640625,
      "learning_rate": 2.005155868929269e-05,
      "loss": 0.3869420051574707,
      "num_input_tokens_seen": 14839866816,
      "step": 52180,
      "train_runtime": 507363.1584,
      "train_tokens_per_second": 29249.004
    },
    {
      "epoch": 1.8465857450238496,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0049946477741817e-05,
      "loss": 0.38881487846374513,
      "num_input_tokens_seen": 14842770944,
      "step": 52190,
      "train_runtime": 507463.65,
      "train_tokens_per_second": 29248.934
    },
    {
      "epoch": 1.8469395652875784,
      "grad_norm": 0.578125,
      "learning_rate": 2.0048334655009818e-05,
      "loss": 0.38770217895507814,
      "num_input_tokens_seen": 14845628288,
      "step": 52200,
      "train_runtime": 507558.5093,
      "train_tokens_per_second": 29249.097
    },
    {
      "epoch": 1.8472933855513074,
      "grad_norm": 0.6015625,
      "learning_rate": 2.004672322094043e-05,
      "loss": 0.38989987373352053,
      "num_input_tokens_seen": 14848459520,
      "step": 52210,
      "train_runtime": 507652.5757,
      "train_tokens_per_second": 29249.255
    },
    {
      "epoch": 1.847647205815036,
      "grad_norm": 0.62109375,
      "learning_rate": 2.00451121753775e-05,
      "loss": 0.3895394802093506,
      "num_input_tokens_seen": 14851264192,
      "step": 52220,
      "train_runtime": 507747.0476,
      "train_tokens_per_second": 29249.336
    },
    {
      "epoch": 1.848001026078765,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0043501518164916e-05,
      "loss": 0.3878207206726074,
      "num_input_tokens_seen": 14854066368,
      "step": 52230,
      "train_runtime": 507840.7596,
      "train_tokens_per_second": 29249.457
    },
    {
      "epoch": 1.8483548463424935,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0041891249146692e-05,
      "loss": 0.3886967658996582,
      "num_input_tokens_seen": 14856942720,
      "step": 52240,
      "train_runtime": 507940.7347,
      "train_tokens_per_second": 29249.363
    },
    {
      "epoch": 1.8487086666062225,
      "grad_norm": 0.63671875,
      "learning_rate": 2.0040281368166923e-05,
      "loss": 0.3910306692123413,
      "num_input_tokens_seen": 14859740992,
      "step": 52250,
      "train_runtime": 508035.4659,
      "train_tokens_per_second": 29249.417
    },
    {
      "epoch": 1.849062486869951,
      "grad_norm": 0.640625,
      "learning_rate": 2.0038671875069774e-05,
      "loss": 0.38906378746032716,
      "num_input_tokens_seen": 14862599808,
      "step": 52260,
      "train_runtime": 508135.6772,
      "train_tokens_per_second": 29249.274
    },
    {
      "epoch": 1.84941630713368,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0037062769699525e-05,
      "loss": 0.3852107524871826,
      "num_input_tokens_seen": 14865442304,
      "step": 52270,
      "train_runtime": 508233.7452,
      "train_tokens_per_second": 29249.223
    },
    {
      "epoch": 1.8497701273974088,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0035454051900516e-05,
      "loss": 0.3884259223937988,
      "num_input_tokens_seen": 14868314304,
      "step": 52280,
      "train_runtime": 508332.1903,
      "train_tokens_per_second": 29249.209
    },
    {
      "epoch": 1.8501239476611375,
      "grad_norm": 0.59765625,
      "learning_rate": 2.003384572151719e-05,
      "loss": 0.38514723777771,
      "num_input_tokens_seen": 14871239232,
      "step": 52290,
      "train_runtime": 508432.5702,
      "train_tokens_per_second": 29249.187
    },
    {
      "epoch": 1.8504777679248663,
      "grad_norm": 0.578125,
      "learning_rate": 2.0032237778394082e-05,
      "loss": 0.38457412719726564,
      "num_input_tokens_seen": 14874111104,
      "step": 52300,
      "train_runtime": 508529.5118,
      "train_tokens_per_second": 29249.258
    },
    {
      "epoch": 1.850831588188595,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0030630222375794e-05,
      "loss": 0.3913117408752441,
      "num_input_tokens_seen": 14876953088,
      "step": 52310,
      "train_runtime": 508627.3621,
      "train_tokens_per_second": 29249.219
    },
    {
      "epoch": 1.8511854084523238,
      "grad_norm": 0.59375,
      "learning_rate": 2.0029023053307043e-05,
      "loss": 0.3888120174407959,
      "num_input_tokens_seen": 14879821632,
      "step": 52320,
      "train_runtime": 508726.9111,
      "train_tokens_per_second": 29249.134
    },
    {
      "epoch": 1.8515392287160526,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0027416271032604e-05,
      "loss": 0.38582258224487304,
      "num_input_tokens_seen": 14882681024,
      "step": 52330,
      "train_runtime": 508824.0878,
      "train_tokens_per_second": 29249.168
    },
    {
      "epoch": 1.8518930489797814,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0025809875397362e-05,
      "loss": 0.3930107593536377,
      "num_input_tokens_seen": 14885498688,
      "step": 52340,
      "train_runtime": 508921.175,
      "train_tokens_per_second": 29249.124
    },
    {
      "epoch": 1.8522468692435101,
      "grad_norm": 0.61328125,
      "learning_rate": 2.002420386624628e-05,
      "loss": 0.39712607860565186,
      "num_input_tokens_seen": 14888316288,
      "step": 52350,
      "train_runtime": 509016.525,
      "train_tokens_per_second": 29249.181
    },
    {
      "epoch": 1.852600689507239,
      "grad_norm": 0.60546875,
      "learning_rate": 2.0022598243424406e-05,
      "loss": 0.39173412322998047,
      "num_input_tokens_seen": 14891132864,
      "step": 52360,
      "train_runtime": 509111.5333,
      "train_tokens_per_second": 29249.255
    },
    {
      "epoch": 1.8529545097709677,
      "grad_norm": 0.625,
      "learning_rate": 2.0020993006776878e-05,
      "loss": 0.3871928691864014,
      "num_input_tokens_seen": 14893995456,
      "step": 52370,
      "train_runtime": 509210.0886,
      "train_tokens_per_second": 29249.215
    },
    {
      "epoch": 1.8533083300346966,
      "grad_norm": 0.58984375,
      "learning_rate": 2.0019388156148922e-05,
      "loss": 0.390073299407959,
      "num_input_tokens_seen": 14896844864,
      "step": 52380,
      "train_runtime": 509305.57,
      "train_tokens_per_second": 29249.326
    },
    {
      "epoch": 1.8536621502984252,
      "grad_norm": 0.60546875,
      "learning_rate": 2.001778369138585e-05,
      "loss": 0.3903347969055176,
      "num_input_tokens_seen": 14899698176,
      "step": 52390,
      "train_runtime": 509402.5361,
      "train_tokens_per_second": 29249.36
    },
    {
      "epoch": 1.8540159705621542,
      "grad_norm": 0.59375,
      "learning_rate": 2.001617961233306e-05,
      "loss": 0.39125995635986327,
      "num_input_tokens_seen": 14902509760,
      "step": 52400,
      "train_runtime": 509500.6138,
      "train_tokens_per_second": 29249.248
    },
    {
      "epoch": 1.8543697908258827,
      "grad_norm": 0.61328125,
      "learning_rate": 2.0014575918836026e-05,
      "loss": 0.38801701068878175,
      "num_input_tokens_seen": 14905346752,
      "step": 52410,
      "train_runtime": 509595.0325,
      "train_tokens_per_second": 29249.396
    },
    {
      "epoch": 1.8547236110896117,
      "grad_norm": 0.62890625,
      "learning_rate": 2.0012972610740336e-05,
      "loss": 0.3901921272277832,
      "num_input_tokens_seen": 14908187008,
      "step": 52420,
      "train_runtime": 509691.1106,
      "train_tokens_per_second": 29249.455
    },
    {
      "epoch": 1.8550774313533402,
      "grad_norm": 0.61328125,
      "learning_rate": 2.001136968789164e-05,
      "loss": 0.3896796464920044,
      "num_input_tokens_seen": 14911065536,
      "step": 52430,
      "train_runtime": 509789.3757,
      "train_tokens_per_second": 29249.463
    },
    {
      "epoch": 1.8554312516170692,
      "grad_norm": 0.58984375,
      "learning_rate": 2.000976715013568e-05,
      "loss": 0.38902010917663576,
      "num_input_tokens_seen": 14913905792,
      "step": 52440,
      "train_runtime": 509885.2164,
      "train_tokens_per_second": 29249.536
    },
    {
      "epoch": 1.855785071880798,
      "grad_norm": 0.609375,
      "learning_rate": 2.0008164997318294e-05,
      "loss": 0.39191834926605223,
      "num_input_tokens_seen": 14916731520,
      "step": 52450,
      "train_runtime": 509981.4176,
      "train_tokens_per_second": 29249.559
    },
    {
      "epoch": 1.8561388921445268,
      "grad_norm": 0.58984375,
      "learning_rate": 2.000656322928539e-05,
      "loss": 0.3873609066009521,
      "num_input_tokens_seen": 14919531776,
      "step": 52460,
      "train_runtime": 510076.5417,
      "train_tokens_per_second": 29249.594
    },
    {
      "epoch": 1.8564927124082555,
      "grad_norm": 0.609375,
      "learning_rate": 2.000496184588298e-05,
      "loss": 0.3859537601470947,
      "num_input_tokens_seen": 14922358592,
      "step": 52470,
      "train_runtime": 510175.3012,
      "train_tokens_per_second": 29249.473
    },
    {
      "epoch": 1.8568465326719843,
      "grad_norm": 0.62109375,
      "learning_rate": 2.0003360846957154e-05,
      "loss": 0.38540210723876955,
      "num_input_tokens_seen": 14925212928,
      "step": 52480,
      "train_runtime": 510273.4346,
      "train_tokens_per_second": 29249.441
    },
    {
      "epoch": 1.857200352935713,
      "grad_norm": 0.6171875,
      "learning_rate": 2.0001760232354076e-05,
      "loss": 0.3895920991897583,
      "num_input_tokens_seen": 14928026944,
      "step": 52490,
      "train_runtime": 510366.6321,
      "train_tokens_per_second": 29249.614
    },
    {
      "epoch": 1.8575541731994418,
      "grad_norm": 0.6015625,
      "learning_rate": 2.0000160001920026e-05,
      "loss": 0.38834171295166015,
      "num_input_tokens_seen": 14930867008,
      "step": 52500,
      "train_runtime": 510463.152,
      "train_tokens_per_second": 29249.647
    },
    {
      "epoch": 1.8579079934631706,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9998560155501343e-05,
      "loss": 0.3922633171081543,
      "num_input_tokens_seen": 14933773568,
      "step": 52510,
      "train_runtime": 510566.8691,
      "train_tokens_per_second": 29249.398
    },
    {
      "epoch": 1.8582618137268994,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9996960692944456e-05,
      "loss": 0.3900697469711304,
      "num_input_tokens_seen": 14936665216,
      "step": 52520,
      "train_runtime": 510667.5934,
      "train_tokens_per_second": 29249.291
    },
    {
      "epoch": 1.8586156339906283,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9995361614095893e-05,
      "loss": 0.39076385498046873,
      "num_input_tokens_seen": 14939450240,
      "step": 52530,
      "train_runtime": 510763.0833,
      "train_tokens_per_second": 29249.276
    },
    {
      "epoch": 1.858969454254357,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9993762918802263e-05,
      "loss": 0.3901973247528076,
      "num_input_tokens_seen": 14942286464,
      "step": 52540,
      "train_runtime": 510859.3885,
      "train_tokens_per_second": 29249.314
    },
    {
      "epoch": 1.8593232745180859,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9992164606910253e-05,
      "loss": 0.3884864807128906,
      "num_input_tokens_seen": 14945110144,
      "step": 52550,
      "train_runtime": 510953.3283,
      "train_tokens_per_second": 29249.462
    },
    {
      "epoch": 1.8596770947818144,
      "grad_norm": 0.62890625,
      "learning_rate": 1.999056667826664e-05,
      "loss": 0.38579819202423093,
      "num_input_tokens_seen": 14947939904,
      "step": 52560,
      "train_runtime": 511048.6292,
      "train_tokens_per_second": 29249.545
    },
    {
      "epoch": 1.8600309150455434,
      "grad_norm": 0.609375,
      "learning_rate": 1.9988969132718285e-05,
      "loss": 0.3899259567260742,
      "num_input_tokens_seen": 14950793920,
      "step": 52570,
      "train_runtime": 511144.9867,
      "train_tokens_per_second": 29249.615
    },
    {
      "epoch": 1.860384735309272,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9987371970112145e-05,
      "loss": 0.3879676342010498,
      "num_input_tokens_seen": 14953684864,
      "step": 52580,
      "train_runtime": 511244.4133,
      "train_tokens_per_second": 29249.581
    },
    {
      "epoch": 1.860738555573001,
      "grad_norm": 0.6171875,
      "learning_rate": 1.998577519029525e-05,
      "loss": 0.3869981288909912,
      "num_input_tokens_seen": 14956571904,
      "step": 52590,
      "train_runtime": 511344.5647,
      "train_tokens_per_second": 29249.498
    },
    {
      "epoch": 1.8610923758367295,
      "grad_norm": 0.625,
      "learning_rate": 1.9984178793114727e-05,
      "loss": 0.38635365962982177,
      "num_input_tokens_seen": 14959440576,
      "step": 52600,
      "train_runtime": 511443.1942,
      "train_tokens_per_second": 29249.467
    },
    {
      "epoch": 1.8614461961004585,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9982582778417768e-05,
      "loss": 0.38664755821228025,
      "num_input_tokens_seen": 14962240768,
      "step": 52610,
      "train_runtime": 511538.3098,
      "train_tokens_per_second": 29249.502
    },
    {
      "epoch": 1.8618000163641872,
      "grad_norm": 0.625,
      "learning_rate": 1.9980987146051677e-05,
      "loss": 0.38433499336242677,
      "num_input_tokens_seen": 14965126656,
      "step": 52620,
      "train_runtime": 511639.8379,
      "train_tokens_per_second": 29249.338
    },
    {
      "epoch": 1.862153836627916,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9979391895863827e-05,
      "loss": 0.3902010679244995,
      "num_input_tokens_seen": 14967964416,
      "step": 52630,
      "train_runtime": 511734.8524,
      "train_tokens_per_second": 29249.453
    },
    {
      "epoch": 1.8625076568916448,
      "grad_norm": 0.62109375,
      "learning_rate": 1.997779702770168e-05,
      "loss": 0.3881723880767822,
      "num_input_tokens_seen": 14970794816,
      "step": 52640,
      "train_runtime": 511830.6282,
      "train_tokens_per_second": 29249.509
    },
    {
      "epoch": 1.8628614771553735,
      "grad_norm": 0.62890625,
      "learning_rate": 1.997620254141278e-05,
      "loss": 0.38869125843048097,
      "num_input_tokens_seen": 14973588160,
      "step": 52650,
      "train_runtime": 511923.8461,
      "train_tokens_per_second": 29249.64
    },
    {
      "epoch": 1.8632152974191023,
      "grad_norm": 0.59765625,
      "learning_rate": 1.997460843684476e-05,
      "loss": 0.3851171493530273,
      "num_input_tokens_seen": 14976479104,
      "step": 52660,
      "train_runtime": 512024.1628,
      "train_tokens_per_second": 29249.555
    },
    {
      "epoch": 1.863569117682831,
      "grad_norm": 0.61328125,
      "learning_rate": 1.997301471384534e-05,
      "loss": 0.38960976600646974,
      "num_input_tokens_seen": 14979315392,
      "step": 52670,
      "train_runtime": 512121.102,
      "train_tokens_per_second": 29249.557
    },
    {
      "epoch": 1.8639229379465598,
      "grad_norm": 0.625,
      "learning_rate": 1.997142137226232e-05,
      "loss": 0.38511836528778076,
      "num_input_tokens_seen": 14982132672,
      "step": 52680,
      "train_runtime": 512214.3851,
      "train_tokens_per_second": 29249.73
    },
    {
      "epoch": 1.8642767582102886,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9969828411943588e-05,
      "loss": 0.3878340244293213,
      "num_input_tokens_seen": 14984967808,
      "step": 52690,
      "train_runtime": 512309.5573,
      "train_tokens_per_second": 29249.831
    },
    {
      "epoch": 1.8646305784740176,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9968235832737117e-05,
      "loss": 0.3881404161453247,
      "num_input_tokens_seen": 14987848256,
      "step": 52700,
      "train_runtime": 512408.4854,
      "train_tokens_per_second": 29249.805
    },
    {
      "epoch": 1.8649843987377461,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9966643634490965e-05,
      "loss": 0.38811697959899905,
      "num_input_tokens_seen": 14990697344,
      "step": 52710,
      "train_runtime": 512505.8956,
      "train_tokens_per_second": 29249.805
    },
    {
      "epoch": 1.8653382190014751,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9965051817053272e-05,
      "loss": 0.38676812648773196,
      "num_input_tokens_seen": 14993497344,
      "step": 52720,
      "train_runtime": 512603.546,
      "train_tokens_per_second": 29249.695
    },
    {
      "epoch": 1.8656920392652037,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9963460380272264e-05,
      "loss": 0.38706045150756835,
      "num_input_tokens_seen": 14996344832,
      "step": 52730,
      "train_runtime": 512700.6838,
      "train_tokens_per_second": 29249.707
    },
    {
      "epoch": 1.8660458595289326,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9961869323996253e-05,
      "loss": 0.38314013481140136,
      "num_input_tokens_seen": 14999178816,
      "step": 52740,
      "train_runtime": 512795.7838,
      "train_tokens_per_second": 29249.809
    },
    {
      "epoch": 1.8663996797926612,
      "grad_norm": 0.625,
      "learning_rate": 1.996027864807364e-05,
      "loss": 0.3890821933746338,
      "num_input_tokens_seen": 15002017408,
      "step": 52750,
      "train_runtime": 512891.9241,
      "train_tokens_per_second": 29249.861
    },
    {
      "epoch": 1.8667535000563902,
      "grad_norm": 0.625,
      "learning_rate": 1.9958688352352895e-05,
      "loss": 0.39221336841583254,
      "num_input_tokens_seen": 15004853824,
      "step": 52760,
      "train_runtime": 512989.8352,
      "train_tokens_per_second": 29249.807
    },
    {
      "epoch": 1.8671073203201187,
      "grad_norm": 0.609375,
      "learning_rate": 1.995709843668259e-05,
      "loss": 0.3884114265441895,
      "num_input_tokens_seen": 15007682048,
      "step": 52770,
      "train_runtime": 513086.3345,
      "train_tokens_per_second": 29249.818
    },
    {
      "epoch": 1.8674611405838477,
      "grad_norm": 0.59375,
      "learning_rate": 1.995550890091138e-05,
      "loss": 0.38825714588165283,
      "num_input_tokens_seen": 15010514304,
      "step": 52780,
      "train_runtime": 513181.9569,
      "train_tokens_per_second": 29249.887
    },
    {
      "epoch": 1.8678149608475765,
      "grad_norm": 0.609375,
      "learning_rate": 1.9953919744887985e-05,
      "loss": 0.39096574783325194,
      "num_input_tokens_seen": 15013325632,
      "step": 52790,
      "train_runtime": 513278.8824,
      "train_tokens_per_second": 29249.841
    },
    {
      "epoch": 1.8681687811113052,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9952330968461233e-05,
      "loss": 0.38544411659240724,
      "num_input_tokens_seen": 15016185792,
      "step": 52800,
      "train_runtime": 513376.2024,
      "train_tokens_per_second": 29249.867
    },
    {
      "epoch": 1.868522601375034,
      "grad_norm": 0.64453125,
      "learning_rate": 1.995074257148002e-05,
      "loss": 0.38976938724517823,
      "num_input_tokens_seen": 15019031040,
      "step": 52810,
      "train_runtime": 513472.5359,
      "train_tokens_per_second": 29249.921
    },
    {
      "epoch": 1.8688764216387628,
      "grad_norm": 0.63671875,
      "learning_rate": 1.9949154553793342e-05,
      "loss": 0.3901804447174072,
      "num_input_tokens_seen": 15021916736,
      "step": 52820,
      "train_runtime": 513572.9633,
      "train_tokens_per_second": 29249.82
    },
    {
      "epoch": 1.8692302419024915,
      "grad_norm": 0.59375,
      "learning_rate": 1.994756691525026e-05,
      "loss": 0.38826560974121094,
      "num_input_tokens_seen": 15024798208,
      "step": 52830,
      "train_runtime": 513673.7044,
      "train_tokens_per_second": 29249.693
    },
    {
      "epoch": 1.8695840621662203,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9945979655699932e-05,
      "loss": 0.38898324966430664,
      "num_input_tokens_seen": 15027600960,
      "step": 52840,
      "train_runtime": 513765.1444,
      "train_tokens_per_second": 29249.943
    },
    {
      "epoch": 1.869937882429949,
      "grad_norm": 0.6171875,
      "learning_rate": 1.99443927749916e-05,
      "loss": 0.39112677574157717,
      "num_input_tokens_seen": 15030380544,
      "step": 52850,
      "train_runtime": 513861.6642,
      "train_tokens_per_second": 29249.858
    },
    {
      "epoch": 1.8702917026936778,
      "grad_norm": 0.59375,
      "learning_rate": 1.994280627297458e-05,
      "loss": 0.3858879566192627,
      "num_input_tokens_seen": 15033159424,
      "step": 52860,
      "train_runtime": 513954.0953,
      "train_tokens_per_second": 29250.004
    },
    {
      "epoch": 1.8706455229574068,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9941220149498282e-05,
      "loss": 0.3906075954437256,
      "num_input_tokens_seen": 15035982464,
      "step": 52870,
      "train_runtime": 514051.9026,
      "train_tokens_per_second": 29249.931
    },
    {
      "epoch": 1.8709993432211354,
      "grad_norm": 0.59765625,
      "learning_rate": 1.99396344044122e-05,
      "loss": 0.39344029426574706,
      "num_input_tokens_seen": 15038833600,
      "step": 52880,
      "train_runtime": 514148.8733,
      "train_tokens_per_second": 29249.959
    },
    {
      "epoch": 1.8713531634848644,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9938049037565903e-05,
      "loss": 0.38948564529418944,
      "num_input_tokens_seen": 15041661824,
      "step": 52890,
      "train_runtime": 514245.6409,
      "train_tokens_per_second": 29249.955
    },
    {
      "epoch": 1.871706983748593,
      "grad_norm": 0.609375,
      "learning_rate": 1.993646404880905e-05,
      "loss": 0.39232783317565917,
      "num_input_tokens_seen": 15044500672,
      "step": 52900,
      "train_runtime": 514343.6075,
      "train_tokens_per_second": 29249.903
    },
    {
      "epoch": 1.8720608040123219,
      "grad_norm": 0.6171875,
      "learning_rate": 1.993487943799139e-05,
      "loss": 0.3886898040771484,
      "num_input_tokens_seen": 15047343552,
      "step": 52910,
      "train_runtime": 514438.3456,
      "train_tokens_per_second": 29250.043
    },
    {
      "epoch": 1.8724146242760504,
      "grad_norm": 0.609375,
      "learning_rate": 1.9933295204962733e-05,
      "loss": 0.38711795806884763,
      "num_input_tokens_seen": 15050183808,
      "step": 52920,
      "train_runtime": 514535.842,
      "train_tokens_per_second": 29250.02
    },
    {
      "epoch": 1.8727684445397794,
      "grad_norm": 0.625,
      "learning_rate": 1.9931711349572995e-05,
      "loss": 0.38999531269073484,
      "num_input_tokens_seen": 15053017280,
      "step": 52930,
      "train_runtime": 514633.5178,
      "train_tokens_per_second": 29249.975
    },
    {
      "epoch": 1.873122264803508,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9930127871672175e-05,
      "loss": 0.3885352611541748,
      "num_input_tokens_seen": 15055880320,
      "step": 52940,
      "train_runtime": 514732.5528,
      "train_tokens_per_second": 29249.909
    },
    {
      "epoch": 1.873476085067237,
      "grad_norm": 0.625,
      "learning_rate": 1.992854477111034e-05,
      "loss": 0.39130687713623047,
      "num_input_tokens_seen": 15058738560,
      "step": 52950,
      "train_runtime": 514829.2295,
      "train_tokens_per_second": 29249.968
    },
    {
      "epoch": 1.8738299053309657,
      "grad_norm": 0.625,
      "learning_rate": 1.9926962047737652e-05,
      "loss": 0.38676133155822756,
      "num_input_tokens_seen": 15061586240,
      "step": 52960,
      "train_runtime": 514927.0767,
      "train_tokens_per_second": 29249.94
    },
    {
      "epoch": 1.8741837255946945,
      "grad_norm": 0.6171875,
      "learning_rate": 1.992537970140435e-05,
      "loss": 0.3870056629180908,
      "num_input_tokens_seen": 15064406464,
      "step": 52970,
      "train_runtime": 515020.2596,
      "train_tokens_per_second": 29250.124
    },
    {
      "epoch": 1.8745375458584232,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9923797731960762e-05,
      "loss": 0.3875922679901123,
      "num_input_tokens_seen": 15067219712,
      "step": 52980,
      "train_runtime": 515114.4994,
      "train_tokens_per_second": 29250.234
    },
    {
      "epoch": 1.874891366122152,
      "grad_norm": 0.609375,
      "learning_rate": 1.9922216139257303e-05,
      "loss": 0.3881427049636841,
      "num_input_tokens_seen": 15070048576,
      "step": 52990,
      "train_runtime": 515211.6136,
      "train_tokens_per_second": 29250.211
    },
    {
      "epoch": 1.8752451863858808,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9920634923144448e-05,
      "loss": 0.38758556842803954,
      "num_input_tokens_seen": 15072914944,
      "step": 53000,
      "train_runtime": 515307.7614,
      "train_tokens_per_second": 29250.316
    },
    {
      "epoch": 1.8755990066496095,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9919054083472788e-05,
      "loss": 0.3880557298660278,
      "num_input_tokens_seen": 15075765440,
      "step": 53010,
      "train_runtime": 515404.2855,
      "train_tokens_per_second": 29250.369
    },
    {
      "epoch": 1.8759528269133383,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9917473620092976e-05,
      "loss": 0.3893942594528198,
      "num_input_tokens_seen": 15078580992,
      "step": 53020,
      "train_runtime": 515497.3784,
      "train_tokens_per_second": 29250.548
    },
    {
      "epoch": 1.876306647177067,
      "grad_norm": 0.59375,
      "learning_rate": 1.9915893532855746e-05,
      "loss": 0.39071145057678225,
      "num_input_tokens_seen": 15081389120,
      "step": 53030,
      "train_runtime": 515591.3678,
      "train_tokens_per_second": 29250.663
    },
    {
      "epoch": 1.876660467440796,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9914313821611926e-05,
      "loss": 0.3903194904327393,
      "num_input_tokens_seen": 15084214336,
      "step": 53040,
      "train_runtime": 515684.9541,
      "train_tokens_per_second": 29250.833
    },
    {
      "epoch": 1.8770142877045246,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9912734486212426e-05,
      "loss": 0.38924849033355713,
      "num_input_tokens_seen": 15087038464,
      "step": 53050,
      "train_runtime": 515782.5988,
      "train_tokens_per_second": 29250.771
    },
    {
      "epoch": 1.8773681079682536,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9911155526508233e-05,
      "loss": 0.3888861179351807,
      "num_input_tokens_seen": 15089893760,
      "step": 53060,
      "train_runtime": 515879.7605,
      "train_tokens_per_second": 29250.796
    },
    {
      "epoch": 1.8777219282319821,
      "grad_norm": 0.59375,
      "learning_rate": 1.9909576942350416e-05,
      "loss": 0.3890857219696045,
      "num_input_tokens_seen": 15092741632,
      "step": 53070,
      "train_runtime": 515979.9138,
      "train_tokens_per_second": 29250.638
    },
    {
      "epoch": 1.8780757484957111,
      "grad_norm": 0.625,
      "learning_rate": 1.9907998733590127e-05,
      "loss": 0.3897523880004883,
      "num_input_tokens_seen": 15095558784,
      "step": 53080,
      "train_runtime": 516075.1505,
      "train_tokens_per_second": 29250.699
    },
    {
      "epoch": 1.8784295687594397,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9906420900078614e-05,
      "loss": 0.3894433736801147,
      "num_input_tokens_seen": 15098389696,
      "step": 53090,
      "train_runtime": 516169.9889,
      "train_tokens_per_second": 29250.809
    },
    {
      "epoch": 1.8787833890231687,
      "grad_norm": 0.671875,
      "learning_rate": 1.9904843441667184e-05,
      "loss": 0.38627848625183103,
      "num_input_tokens_seen": 15101230144,
      "step": 53100,
      "train_runtime": 516269.9061,
      "train_tokens_per_second": 29250.65
    },
    {
      "epoch": 1.8791372092868974,
      "grad_norm": 0.61328125,
      "learning_rate": 1.990326635820725e-05,
      "loss": 0.38816375732421876,
      "num_input_tokens_seen": 15104124416,
      "step": 53110,
      "train_runtime": 516368.0914,
      "train_tokens_per_second": 29250.693
    },
    {
      "epoch": 1.8794910295506262,
      "grad_norm": 0.625,
      "learning_rate": 1.9901689649550285e-05,
      "loss": 0.3869650840759277,
      "num_input_tokens_seen": 15106953216,
      "step": 53120,
      "train_runtime": 516464.4259,
      "train_tokens_per_second": 29250.714
    },
    {
      "epoch": 1.879844849814355,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9900113315547862e-05,
      "loss": 0.389239764213562,
      "num_input_tokens_seen": 15109785728,
      "step": 53130,
      "train_runtime": 516560.4768,
      "train_tokens_per_second": 29250.758
    },
    {
      "epoch": 1.8801986700780837,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9898537356051635e-05,
      "loss": 0.38719141483306885,
      "num_input_tokens_seen": 15112611712,
      "step": 53140,
      "train_runtime": 516659.3849,
      "train_tokens_per_second": 29250.628
    },
    {
      "epoch": 1.8805524903418125,
      "grad_norm": 0.6484375,
      "learning_rate": 1.9896961770913325e-05,
      "loss": 0.39161033630371095,
      "num_input_tokens_seen": 15115498368,
      "step": 53150,
      "train_runtime": 516761.9015,
      "train_tokens_per_second": 29250.412
    },
    {
      "epoch": 1.8809063106055413,
      "grad_norm": 0.59765625,
      "learning_rate": 1.989538655998475e-05,
      "loss": 0.3917238235473633,
      "num_input_tokens_seen": 15118345792,
      "step": 53160,
      "train_runtime": 516860.7018,
      "train_tokens_per_second": 29250.329
    },
    {
      "epoch": 1.88126013086927,
      "grad_norm": 0.62109375,
      "learning_rate": 1.989381172311781e-05,
      "loss": 0.38785481452941895,
      "num_input_tokens_seen": 15121195328,
      "step": 53170,
      "train_runtime": 516955.3983,
      "train_tokens_per_second": 29250.483
    },
    {
      "epoch": 1.8816139511329988,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9892237260164473e-05,
      "loss": 0.3875454902648926,
      "num_input_tokens_seen": 15123988544,
      "step": 53180,
      "train_runtime": 517049.8868,
      "train_tokens_per_second": 29250.54
    },
    {
      "epoch": 1.8819677713967276,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9890663170976807e-05,
      "loss": 0.38889455795288086,
      "num_input_tokens_seen": 15126827200,
      "step": 53190,
      "train_runtime": 517145.7013,
      "train_tokens_per_second": 29250.61
    },
    {
      "epoch": 1.8823215916604563,
      "grad_norm": 0.625,
      "learning_rate": 1.9889089455406948e-05,
      "loss": 0.3917581081390381,
      "num_input_tokens_seen": 15129682240,
      "step": 53200,
      "train_runtime": 517241.1588,
      "train_tokens_per_second": 29250.731
    },
    {
      "epoch": 1.8826754119241853,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9887516113307126e-05,
      "loss": 0.3889960527420044,
      "num_input_tokens_seen": 15132541760,
      "step": 53210,
      "train_runtime": 517339.2663,
      "train_tokens_per_second": 29250.712
    },
    {
      "epoch": 1.8830292321879138,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9885943144529635e-05,
      "loss": 0.3879654884338379,
      "num_input_tokens_seen": 15135366592,
      "step": 53220,
      "train_runtime": 517434.2953,
      "train_tokens_per_second": 29250.799
    },
    {
      "epoch": 1.8833830524516428,
      "grad_norm": 0.625,
      "learning_rate": 1.9884370548926874e-05,
      "loss": 0.38990857601165774,
      "num_input_tokens_seen": 15138289152,
      "step": 53230,
      "train_runtime": 517534.567,
      "train_tokens_per_second": 29250.779
    },
    {
      "epoch": 1.8837368727153714,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9882798326351305e-05,
      "loss": 0.38763885498046874,
      "num_input_tokens_seen": 15141172672,
      "step": 53240,
      "train_runtime": 517635.5553,
      "train_tokens_per_second": 29250.643
    },
    {
      "epoch": 1.8840906929791004,
      "grad_norm": 0.609375,
      "learning_rate": 1.9881226476655475e-05,
      "loss": 0.38727731704711915,
      "num_input_tokens_seen": 15144026176,
      "step": 53250,
      "train_runtime": 517732.476,
      "train_tokens_per_second": 29250.678
    },
    {
      "epoch": 1.884444513242829,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9879654999692028e-05,
      "loss": 0.38786969184875486,
      "num_input_tokens_seen": 15146872384,
      "step": 53260,
      "train_runtime": 517833.151,
      "train_tokens_per_second": 29250.488
    },
    {
      "epoch": 1.884798333506558,
      "grad_norm": 0.62109375,
      "learning_rate": 1.987808389531366e-05,
      "loss": 0.3893772840499878,
      "num_input_tokens_seen": 15149702272,
      "step": 53270,
      "train_runtime": 517930.3972,
      "train_tokens_per_second": 29250.46
    },
    {
      "epoch": 1.8851521537702867,
      "grad_norm": 0.59375,
      "learning_rate": 1.987651316337318e-05,
      "loss": 0.3891500234603882,
      "num_input_tokens_seen": 15152548288,
      "step": 53280,
      "train_runtime": 518027.8306,
      "train_tokens_per_second": 29250.452
    },
    {
      "epoch": 1.8855059740340154,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9874942803723465e-05,
      "loss": 0.3888816833496094,
      "num_input_tokens_seen": 15155368576,
      "step": 53290,
      "train_runtime": 518120.5242,
      "train_tokens_per_second": 29250.662
    },
    {
      "epoch": 1.8858597942977442,
      "grad_norm": 0.6328125,
      "learning_rate": 1.987337281621746e-05,
      "loss": 0.3868882656097412,
      "num_input_tokens_seen": 15158285760,
      "step": 53300,
      "train_runtime": 518218.0611,
      "train_tokens_per_second": 29250.786
    },
    {
      "epoch": 1.886213614561473,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9871803200708214e-05,
      "loss": 0.3888881206512451,
      "num_input_tokens_seen": 15161112320,
      "step": 53310,
      "train_runtime": 518311.1369,
      "train_tokens_per_second": 29250.987
    },
    {
      "epoch": 1.8865674348252017,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9870233957048844e-05,
      "loss": 0.3868952751159668,
      "num_input_tokens_seen": 15163967936,
      "step": 53320,
      "train_runtime": 518408.5207,
      "train_tokens_per_second": 29251.001
    },
    {
      "epoch": 1.8869212550889305,
      "grad_norm": 0.609375,
      "learning_rate": 1.9868665085092556e-05,
      "loss": 0.3902482032775879,
      "num_input_tokens_seen": 15166826496,
      "step": 53330,
      "train_runtime": 518506.9476,
      "train_tokens_per_second": 29250.961
    },
    {
      "epoch": 1.8872750753526593,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9867096584692624e-05,
      "loss": 0.3900275230407715,
      "num_input_tokens_seen": 15169662656,
      "step": 53340,
      "train_runtime": 518603.1434,
      "train_tokens_per_second": 29251.004
    },
    {
      "epoch": 1.887628895616388,
      "grad_norm": 0.609375,
      "learning_rate": 1.9865528455702416e-05,
      "loss": 0.3940988302230835,
      "num_input_tokens_seen": 15172499712,
      "step": 53350,
      "train_runtime": 518700.8679,
      "train_tokens_per_second": 29250.963
    },
    {
      "epoch": 1.887982715880117,
      "grad_norm": 0.65234375,
      "learning_rate": 1.9863960697975378e-05,
      "loss": 0.38994994163513186,
      "num_input_tokens_seen": 15175370112,
      "step": 53360,
      "train_runtime": 518800.3126,
      "train_tokens_per_second": 29250.889
    },
    {
      "epoch": 1.8883365361438456,
      "grad_norm": 0.578125,
      "learning_rate": 1.986239331136503e-05,
      "loss": 0.3894001007080078,
      "num_input_tokens_seen": 15178147008,
      "step": 53370,
      "train_runtime": 518892.6364,
      "train_tokens_per_second": 29251.036
    },
    {
      "epoch": 1.8886903564075745,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9860826295724985e-05,
      "loss": 0.39093894958496095,
      "num_input_tokens_seen": 15181014080,
      "step": 53380,
      "train_runtime": 518994.4726,
      "train_tokens_per_second": 29250.82
    },
    {
      "epoch": 1.889044176671303,
      "grad_norm": 0.61328125,
      "learning_rate": 1.985925965090893e-05,
      "loss": 0.3889622688293457,
      "num_input_tokens_seen": 15183882944,
      "step": 53390,
      "train_runtime": 519092.1551,
      "train_tokens_per_second": 29250.843
    },
    {
      "epoch": 1.889397996935032,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9857693376770625e-05,
      "loss": 0.39083285331726075,
      "num_input_tokens_seen": 15186685824,
      "step": 53400,
      "train_runtime": 519187.7971,
      "train_tokens_per_second": 29250.853
    },
    {
      "epoch": 1.8897518171987606,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9856127473163922e-05,
      "loss": 0.38647093772888186,
      "num_input_tokens_seen": 15189505536,
      "step": 53410,
      "train_runtime": 519282.5959,
      "train_tokens_per_second": 29250.943
    },
    {
      "epoch": 1.8901056374624896,
      "grad_norm": 0.59375,
      "learning_rate": 1.9854561939942756e-05,
      "loss": 0.38783090114593505,
      "num_input_tokens_seen": 15192307968,
      "step": 53420,
      "train_runtime": 519379.6171,
      "train_tokens_per_second": 29250.874
    },
    {
      "epoch": 1.8904594577262182,
      "grad_norm": 0.60546875,
      "learning_rate": 1.985299677696113e-05,
      "loss": 0.3890634536743164,
      "num_input_tokens_seen": 15195184448,
      "step": 53430,
      "train_runtime": 519477.2443,
      "train_tokens_per_second": 29250.914
    },
    {
      "epoch": 1.8908132779899471,
      "grad_norm": 0.59375,
      "learning_rate": 1.9851431984073143e-05,
      "loss": 0.3879124164581299,
      "num_input_tokens_seen": 15198006784,
      "step": 53440,
      "train_runtime": 519571.6041,
      "train_tokens_per_second": 29251.034
    },
    {
      "epoch": 1.891167098253676,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9849867561132955e-05,
      "loss": 0.3906916379928589,
      "num_input_tokens_seen": 15200875904,
      "step": 53450,
      "train_runtime": 519668.583,
      "train_tokens_per_second": 29251.097
    },
    {
      "epoch": 1.8915209185174047,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9848303507994826e-05,
      "loss": 0.3866857051849365,
      "num_input_tokens_seen": 15203784192,
      "step": 53460,
      "train_runtime": 519770.6749,
      "train_tokens_per_second": 29250.946
    },
    {
      "epoch": 1.8918747387811334,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9846739824513087e-05,
      "loss": 0.3907691478729248,
      "num_input_tokens_seen": 15206600512,
      "step": 53470,
      "train_runtime": 519865.0141,
      "train_tokens_per_second": 29251.056
    },
    {
      "epoch": 1.8922285590448622,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9845176510542145e-05,
      "loss": 0.3860243082046509,
      "num_input_tokens_seen": 15209420352,
      "step": 53480,
      "train_runtime": 519959.256,
      "train_tokens_per_second": 29251.177
    },
    {
      "epoch": 1.892582379308591,
      "grad_norm": 0.5859375,
      "learning_rate": 1.9843613565936495e-05,
      "loss": 0.3895238161087036,
      "num_input_tokens_seen": 15212237376,
      "step": 53490,
      "train_runtime": 520054.5685,
      "train_tokens_per_second": 29251.233
    },
    {
      "epoch": 1.8929361995723197,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9842050990550707e-05,
      "loss": 0.3885918617248535,
      "num_input_tokens_seen": 15215105664,
      "step": 53500,
      "train_runtime": 520153.2804,
      "train_tokens_per_second": 29251.196
    },
    {
      "epoch": 1.8932900198360485,
      "grad_norm": 0.609375,
      "learning_rate": 1.9840488784239442e-05,
      "loss": 0.3880393266677856,
      "num_input_tokens_seen": 15217921088,
      "step": 53510,
      "train_runtime": 520247.3298,
      "train_tokens_per_second": 29251.32
    },
    {
      "epoch": 1.8936438400997773,
      "grad_norm": 0.640625,
      "learning_rate": 1.983892694685743e-05,
      "loss": 0.3908080101013184,
      "num_input_tokens_seen": 15220754240,
      "step": 53520,
      "train_runtime": 520343.9307,
      "train_tokens_per_second": 29251.334
    },
    {
      "epoch": 1.8939976603635063,
      "grad_norm": 0.625,
      "learning_rate": 1.983736547825947e-05,
      "loss": 0.3856344699859619,
      "num_input_tokens_seen": 15223586240,
      "step": 53530,
      "train_runtime": 520440.2466,
      "train_tokens_per_second": 29251.362
    },
    {
      "epoch": 1.8943514806272348,
      "grad_norm": 0.609375,
      "learning_rate": 1.983580437830047e-05,
      "loss": 0.3907046318054199,
      "num_input_tokens_seen": 15226450176,
      "step": 53540,
      "train_runtime": 520538.1517,
      "train_tokens_per_second": 29251.363
    },
    {
      "epoch": 1.8947053008909638,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9834243646835407e-05,
      "loss": 0.38485419750213623,
      "num_input_tokens_seen": 15229264320,
      "step": 53550,
      "train_runtime": 520633.7646,
      "train_tokens_per_second": 29251.396
    },
    {
      "epoch": 1.8950591211546923,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9832683283719318e-05,
      "loss": 0.38997886180877683,
      "num_input_tokens_seen": 15232128128,
      "step": 53560,
      "train_runtime": 520734.4648,
      "train_tokens_per_second": 29251.239
    },
    {
      "epoch": 1.8954129414184213,
      "grad_norm": 0.609375,
      "learning_rate": 1.9831123288807342e-05,
      "loss": 0.38659234046936036,
      "num_input_tokens_seen": 15234941632,
      "step": 53570,
      "train_runtime": 520832.0475,
      "train_tokens_per_second": 29251.16
    },
    {
      "epoch": 1.8957667616821499,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9829563661954696e-05,
      "loss": 0.3892052412033081,
      "num_input_tokens_seen": 15237821568,
      "step": 53580,
      "train_runtime": 520933.1464,
      "train_tokens_per_second": 29251.012
    },
    {
      "epoch": 1.8961205819458788,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9828004403016665e-05,
      "loss": 0.3880744457244873,
      "num_input_tokens_seen": 15240696384,
      "step": 53590,
      "train_runtime": 521032.0135,
      "train_tokens_per_second": 29250.979
    },
    {
      "epoch": 1.8964744022096074,
      "grad_norm": 0.640625,
      "learning_rate": 1.9826445511848627e-05,
      "loss": 0.3953718900680542,
      "num_input_tokens_seen": 15243539328,
      "step": 53600,
      "train_runtime": 521126.9955,
      "train_tokens_per_second": 29251.103
    },
    {
      "epoch": 1.8968282224733364,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9824886988306024e-05,
      "loss": 0.3863143682479858,
      "num_input_tokens_seen": 15246413888,
      "step": 53610,
      "train_runtime": 521224.8278,
      "train_tokens_per_second": 29251.127
    },
    {
      "epoch": 1.8971820427370651,
      "grad_norm": 0.609375,
      "learning_rate": 1.9823328832244396e-05,
      "loss": 0.3872296571731567,
      "num_input_tokens_seen": 15249245376,
      "step": 53620,
      "train_runtime": 521323.1341,
      "train_tokens_per_second": 29251.043
    },
    {
      "epoch": 1.897535863000794,
      "grad_norm": 0.578125,
      "learning_rate": 1.9821771043519348e-05,
      "loss": 0.3867331981658936,
      "num_input_tokens_seen": 15252124224,
      "step": 53630,
      "train_runtime": 521422.567,
      "train_tokens_per_second": 29250.986
    },
    {
      "epoch": 1.8978896832645227,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9820213621986572e-05,
      "loss": 0.3917325496673584,
      "num_input_tokens_seen": 15254925440,
      "step": 53640,
      "train_runtime": 521516.3369,
      "train_tokens_per_second": 29251.098
    },
    {
      "epoch": 1.8982435035282514,
      "grad_norm": 0.6484375,
      "learning_rate": 1.9818656567501832e-05,
      "loss": 0.38895363807678224,
      "num_input_tokens_seen": 15257742528,
      "step": 53650,
      "train_runtime": 521613.1825,
      "train_tokens_per_second": 29251.068
    },
    {
      "epoch": 1.8985973237919802,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9817099879920984e-05,
      "loss": 0.3855937957763672,
      "num_input_tokens_seen": 15260600320,
      "step": 53660,
      "train_runtime": 521710.7091,
      "train_tokens_per_second": 29251.077
    },
    {
      "epoch": 1.898951144055709,
      "grad_norm": 0.59765625,
      "learning_rate": 1.981554355909995e-05,
      "loss": 0.38933491706848145,
      "num_input_tokens_seen": 15263387648,
      "step": 53670,
      "train_runtime": 521805.3432,
      "train_tokens_per_second": 29251.114
    },
    {
      "epoch": 1.8993049643194377,
      "grad_norm": 0.609375,
      "learning_rate": 1.9813987604894733e-05,
      "loss": 0.38747310638427734,
      "num_input_tokens_seen": 15266260160,
      "step": 53680,
      "train_runtime": 521905.618,
      "train_tokens_per_second": 29250.998
    },
    {
      "epoch": 1.8996587845831665,
      "grad_norm": 0.6328125,
      "learning_rate": 1.981243201716143e-05,
      "loss": 0.3825028657913208,
      "num_input_tokens_seen": 15269102656,
      "step": 53690,
      "train_runtime": 522003.5787,
      "train_tokens_per_second": 29250.954
    },
    {
      "epoch": 1.9000126048468955,
      "grad_norm": 0.5859375,
      "learning_rate": 1.98108767957562e-05,
      "loss": 0.39011249542236326,
      "num_input_tokens_seen": 15271902464,
      "step": 53700,
      "train_runtime": 522095.9074,
      "train_tokens_per_second": 29251.144
    },
    {
      "epoch": 1.900366425110624,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9809321940535282e-05,
      "loss": 0.3835816144943237,
      "num_input_tokens_seen": 15274761856,
      "step": 53710,
      "train_runtime": 522194.221,
      "train_tokens_per_second": 29251.112
    },
    {
      "epoch": 1.900720245374353,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9807767451355004e-05,
      "loss": 0.38843545913696287,
      "num_input_tokens_seen": 15277596800,
      "step": 53720,
      "train_runtime": 522290.1021,
      "train_tokens_per_second": 29251.17
    },
    {
      "epoch": 1.9010740656380816,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9806213328071767e-05,
      "loss": 0.38711061477661135,
      "num_input_tokens_seen": 15280457344,
      "step": 53730,
      "train_runtime": 522386.7806,
      "train_tokens_per_second": 29251.233
    },
    {
      "epoch": 1.9014278859018106,
      "grad_norm": 0.625,
      "learning_rate": 1.980465957054205e-05,
      "loss": 0.38663573265075685,
      "num_input_tokens_seen": 15283255808,
      "step": 53740,
      "train_runtime": 522482.6489,
      "train_tokens_per_second": 29251.222
    },
    {
      "epoch": 1.901781706165539,
      "grad_norm": 0.62109375,
      "learning_rate": 1.980310617862242e-05,
      "loss": 0.391534423828125,
      "num_input_tokens_seen": 15286074816,
      "step": 53750,
      "train_runtime": 522580.4353,
      "train_tokens_per_second": 29251.143
    },
    {
      "epoch": 1.902135526429268,
      "grad_norm": 0.62890625,
      "learning_rate": 1.98015531521695e-05,
      "loss": 0.3836332321166992,
      "num_input_tokens_seen": 15288933504,
      "step": 53760,
      "train_runtime": 522676.9832,
      "train_tokens_per_second": 29251.209
    },
    {
      "epoch": 1.9024893466929966,
      "grad_norm": 0.60546875,
      "learning_rate": 1.980000049104002e-05,
      "loss": 0.39131228923797606,
      "num_input_tokens_seen": 15291791680,
      "step": 53770,
      "train_runtime": 522775.5348,
      "train_tokens_per_second": 29251.162
    },
    {
      "epoch": 1.9028431669567256,
      "grad_norm": 0.60546875,
      "learning_rate": 1.979844819509077e-05,
      "loss": 0.3852677345275879,
      "num_input_tokens_seen": 15294617856,
      "step": 53780,
      "train_runtime": 522872.9132,
      "train_tokens_per_second": 29251.119
    },
    {
      "epoch": 1.9031969872204544,
      "grad_norm": 0.62890625,
      "learning_rate": 1.979689626417863e-05,
      "loss": 0.3901498556137085,
      "num_input_tokens_seen": 15297405056,
      "step": 53790,
      "train_runtime": 522966.9226,
      "train_tokens_per_second": 29251.19
    },
    {
      "epoch": 1.9035508074841831,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9795344698160537e-05,
      "loss": 0.3894065856933594,
      "num_input_tokens_seen": 15300277568,
      "step": 53800,
      "train_runtime": 523064.7036,
      "train_tokens_per_second": 29251.214
    },
    {
      "epoch": 1.903904627747912,
      "grad_norm": 0.6015625,
      "learning_rate": 1.979379349689354e-05,
      "loss": 0.3846890926361084,
      "num_input_tokens_seen": 15303062272,
      "step": 53810,
      "train_runtime": 523158.1422,
      "train_tokens_per_second": 29251.312
    },
    {
      "epoch": 1.9042584480116407,
      "grad_norm": 0.609375,
      "learning_rate": 1.979224266023474e-05,
      "loss": 0.39230320453643797,
      "num_input_tokens_seen": 15305852416,
      "step": 53820,
      "train_runtime": 523254.4533,
      "train_tokens_per_second": 29251.261
    },
    {
      "epoch": 1.9046122682753694,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9790692188041323e-05,
      "loss": 0.3900423526763916,
      "num_input_tokens_seen": 15308687872,
      "step": 53830,
      "train_runtime": 523350.118,
      "train_tokens_per_second": 29251.332
    },
    {
      "epoch": 1.9049660885390982,
      "grad_norm": 0.625,
      "learning_rate": 1.9789142080170562e-05,
      "loss": 0.3864589691162109,
      "num_input_tokens_seen": 15311520832,
      "step": 53840,
      "train_runtime": 523445.8033,
      "train_tokens_per_second": 29251.397
    },
    {
      "epoch": 1.905319908802827,
      "grad_norm": 0.5859375,
      "learning_rate": 1.9787592336479794e-05,
      "loss": 0.386370587348938,
      "num_input_tokens_seen": 15314368448,
      "step": 53850,
      "train_runtime": 523542.0465,
      "train_tokens_per_second": 29251.459
    },
    {
      "epoch": 1.9056737290665557,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9786042956826444e-05,
      "loss": 0.38846635818481445,
      "num_input_tokens_seen": 15317277248,
      "step": 53860,
      "train_runtime": 523643.7517,
      "train_tokens_per_second": 29251.332
    },
    {
      "epoch": 1.9060275493302847,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9784493941068015e-05,
      "loss": 0.3874437093734741,
      "num_input_tokens_seen": 15320099968,
      "step": 53870,
      "train_runtime": 523737.3723,
      "train_tokens_per_second": 29251.493
    },
    {
      "epoch": 1.9063813695940133,
      "grad_norm": 0.59375,
      "learning_rate": 1.9782945289062085e-05,
      "loss": 0.3873290538787842,
      "num_input_tokens_seen": 15322939968,
      "step": 53880,
      "train_runtime": 523834.1951,
      "train_tokens_per_second": 29251.508
    },
    {
      "epoch": 1.9067351898577423,
      "grad_norm": 0.6171875,
      "learning_rate": 1.978139700066631e-05,
      "loss": 0.38614840507507325,
      "num_input_tokens_seen": 15325730816,
      "step": 53890,
      "train_runtime": 523926.7606,
      "train_tokens_per_second": 29251.666
    },
    {
      "epoch": 1.9070890101214708,
      "grad_norm": 0.63671875,
      "learning_rate": 1.977984907573842e-05,
      "loss": 0.385538125038147,
      "num_input_tokens_seen": 15328558592,
      "step": 53900,
      "train_runtime": 524022.7305,
      "train_tokens_per_second": 29251.706
    },
    {
      "epoch": 1.9074428303851998,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9778301514136232e-05,
      "loss": 0.3856320858001709,
      "num_input_tokens_seen": 15331444736,
      "step": 53910,
      "train_runtime": 524123.3013,
      "train_tokens_per_second": 29251.599
    },
    {
      "epoch": 1.9077966506489283,
      "grad_norm": 0.60546875,
      "learning_rate": 1.977675431571764e-05,
      "loss": 0.38629310131072997,
      "num_input_tokens_seen": 15334280256,
      "step": 53920,
      "train_runtime": 524221.6893,
      "train_tokens_per_second": 29251.518
    },
    {
      "epoch": 1.9081504709126573,
      "grad_norm": 0.625,
      "learning_rate": 1.977520748034061e-05,
      "loss": 0.38968839645385744,
      "num_input_tokens_seen": 15337088064,
      "step": 53930,
      "train_runtime": 524317.2451,
      "train_tokens_per_second": 29251.542
    },
    {
      "epoch": 1.9085042911763859,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9773661007863183e-05,
      "loss": 0.3931389808654785,
      "num_input_tokens_seen": 15339925120,
      "step": 53940,
      "train_runtime": 524416.8215,
      "train_tokens_per_second": 29251.398
    },
    {
      "epoch": 1.9088581114401149,
      "grad_norm": 0.60546875,
      "learning_rate": 1.977211489814349e-05,
      "loss": 0.3909188508987427,
      "num_input_tokens_seen": 15342791616,
      "step": 53950,
      "train_runtime": 524516.9827,
      "train_tokens_per_second": 29251.277
    },
    {
      "epoch": 1.9092119317038436,
      "grad_norm": 0.62890625,
      "learning_rate": 1.977056915103973e-05,
      "loss": 0.38784518241882326,
      "num_input_tokens_seen": 15345632192,
      "step": 53960,
      "train_runtime": 524612.927,
      "train_tokens_per_second": 29251.342
    },
    {
      "epoch": 1.9095657519675724,
      "grad_norm": 0.625,
      "learning_rate": 1.9769023766410176e-05,
      "loss": 0.3866534471511841,
      "num_input_tokens_seen": 15348445056,
      "step": 53970,
      "train_runtime": 524710.1247,
      "train_tokens_per_second": 29251.284
    },
    {
      "epoch": 1.9099195722313012,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9767478744113192e-05,
      "loss": 0.3944977045059204,
      "num_input_tokens_seen": 15351364864,
      "step": 53980,
      "train_runtime": 524810.9224,
      "train_tokens_per_second": 29251.23
    },
    {
      "epoch": 1.91027339249503,
      "grad_norm": 0.62109375,
      "learning_rate": 1.976593408400721e-05,
      "loss": 0.3887478828430176,
      "num_input_tokens_seen": 15354176640,
      "step": 53990,
      "train_runtime": 524906.03,
      "train_tokens_per_second": 29251.286
    },
    {
      "epoch": 1.9106272127587587,
      "grad_norm": 0.6015625,
      "learning_rate": 1.976438978595074e-05,
      "loss": 0.38941264152526855,
      "num_input_tokens_seen": 15357006080,
      "step": 54000,
      "train_runtime": 525001.8646,
      "train_tokens_per_second": 29251.336
    },
    {
      "epoch": 1.9109810330224875,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9762845849802375e-05,
      "loss": 0.3852963924407959,
      "num_input_tokens_seen": 15359838528,
      "step": 54010,
      "train_runtime": 525096.894,
      "train_tokens_per_second": 29251.437
    },
    {
      "epoch": 1.9113348532862162,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9761302275420775e-05,
      "loss": 0.3875159502029419,
      "num_input_tokens_seen": 15362698240,
      "step": 54020,
      "train_runtime": 525194.2941,
      "train_tokens_per_second": 29251.457
    },
    {
      "epoch": 1.911688673549945,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9759759062664685e-05,
      "loss": 0.3879960775375366,
      "num_input_tokens_seen": 15365653056,
      "step": 54030,
      "train_runtime": 525299.0726,
      "train_tokens_per_second": 29251.247
    },
    {
      "epoch": 1.912042493813674,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9758216211392925e-05,
      "loss": 0.3892364501953125,
      "num_input_tokens_seen": 15368471552,
      "step": 54040,
      "train_runtime": 525396.1679,
      "train_tokens_per_second": 29251.206
    },
    {
      "epoch": 1.9123963140774025,
      "grad_norm": 0.609375,
      "learning_rate": 1.97566737214644e-05,
      "loss": 0.38932032585144044,
      "num_input_tokens_seen": 15371287040,
      "step": 54050,
      "train_runtime": 525489.7998,
      "train_tokens_per_second": 29251.352
    },
    {
      "epoch": 1.9127501343411315,
      "grad_norm": 0.625,
      "learning_rate": 1.9755131592738073e-05,
      "loss": 0.39127469062805176,
      "num_input_tokens_seen": 15374138880,
      "step": 54060,
      "train_runtime": 525587.2948,
      "train_tokens_per_second": 29251.352
    },
    {
      "epoch": 1.91310395460486,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9753589825073005e-05,
      "loss": 0.3866180181503296,
      "num_input_tokens_seen": 15376968384,
      "step": 54070,
      "train_runtime": 525680.9897,
      "train_tokens_per_second": 29251.521
    },
    {
      "epoch": 1.913457774868589,
      "grad_norm": 0.58984375,
      "learning_rate": 1.975204841832832e-05,
      "loss": 0.3887892723083496,
      "num_input_tokens_seen": 15379848192,
      "step": 54080,
      "train_runtime": 525779.5947,
      "train_tokens_per_second": 29251.512
    },
    {
      "epoch": 1.9138115951323176,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9750507372363228e-05,
      "loss": 0.3888261318206787,
      "num_input_tokens_seen": 15382733440,
      "step": 54090,
      "train_runtime": 525877.4496,
      "train_tokens_per_second": 29251.556
    },
    {
      "epoch": 1.9141654153960466,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9748966687037003e-05,
      "loss": 0.38535704612731936,
      "num_input_tokens_seen": 15385608064,
      "step": 54100,
      "train_runtime": 525977.8039,
      "train_tokens_per_second": 29251.44
    },
    {
      "epoch": 1.9145192356597753,
      "grad_norm": 0.59375,
      "learning_rate": 1.974742636220902e-05,
      "loss": 0.38651299476623535,
      "num_input_tokens_seen": 15388443712,
      "step": 54110,
      "train_runtime": 526074.9859,
      "train_tokens_per_second": 29251.426
    },
    {
      "epoch": 1.914873055923504,
      "grad_norm": 0.58984375,
      "learning_rate": 1.97458863977387e-05,
      "loss": 0.3876023530960083,
      "num_input_tokens_seen": 15391288576,
      "step": 54120,
      "train_runtime": 526169.9775,
      "train_tokens_per_second": 29251.552
    },
    {
      "epoch": 1.9152268761872329,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9744346793485567e-05,
      "loss": 0.38829994201660156,
      "num_input_tokens_seen": 15394125504,
      "step": 54130,
      "train_runtime": 526265.5389,
      "train_tokens_per_second": 29251.631
    },
    {
      "epoch": 1.9155806964509616,
      "grad_norm": 0.60546875,
      "learning_rate": 1.97428075493092e-05,
      "loss": 0.38388900756835936,
      "num_input_tokens_seen": 15396968256,
      "step": 54140,
      "train_runtime": 526364.1278,
      "train_tokens_per_second": 29251.553
    },
    {
      "epoch": 1.9159345167146904,
      "grad_norm": 0.59375,
      "learning_rate": 1.9741268665069273e-05,
      "loss": 0.3890402793884277,
      "num_input_tokens_seen": 15399818176,
      "step": 54150,
      "train_runtime": 526459.7926,
      "train_tokens_per_second": 29251.651
    },
    {
      "epoch": 1.9162883369784192,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9739730140625527e-05,
      "loss": 0.3863030195236206,
      "num_input_tokens_seen": 15402693504,
      "step": 54160,
      "train_runtime": 526557.9933,
      "train_tokens_per_second": 29251.656
    },
    {
      "epoch": 1.916642157242148,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9738191975837777e-05,
      "loss": 0.38698320388793944,
      "num_input_tokens_seen": 15405510592,
      "step": 54170,
      "train_runtime": 526653.4313,
      "train_tokens_per_second": 29251.705
    },
    {
      "epoch": 1.9169959775058767,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9736654170565928e-05,
      "loss": 0.3885756015777588,
      "num_input_tokens_seen": 15408342016,
      "step": 54180,
      "train_runtime": 526749.6089,
      "train_tokens_per_second": 29251.739
    },
    {
      "epoch": 1.9173497977696055,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9735116724669948e-05,
      "loss": 0.38958680629730225,
      "num_input_tokens_seen": 15411184448,
      "step": 54190,
      "train_runtime": 526849.3708,
      "train_tokens_per_second": 29251.595
    },
    {
      "epoch": 1.9177036180333342,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9733579638009878e-05,
      "loss": 0.3934484958648682,
      "num_input_tokens_seen": 15414026560,
      "step": 54200,
      "train_runtime": 526945.343,
      "train_tokens_per_second": 29251.661
    },
    {
      "epoch": 1.9180574382970632,
      "grad_norm": 0.6171875,
      "learning_rate": 1.973204291044585e-05,
      "loss": 0.38809478282928467,
      "num_input_tokens_seen": 15416833152,
      "step": 54210,
      "train_runtime": 527039.8845,
      "train_tokens_per_second": 29251.739
    },
    {
      "epoch": 1.9184112585607918,
      "grad_norm": 0.625,
      "learning_rate": 1.973050654183807e-05,
      "loss": 0.3903543472290039,
      "num_input_tokens_seen": 15419706752,
      "step": 54220,
      "train_runtime": 527138.8681,
      "train_tokens_per_second": 29251.698
    },
    {
      "epoch": 1.9187650788245207,
      "grad_norm": 0.62890625,
      "learning_rate": 1.97289705320468e-05,
      "loss": 0.3884843587875366,
      "num_input_tokens_seen": 15422575872,
      "step": 54230,
      "train_runtime": 527236.1031,
      "train_tokens_per_second": 29251.745
    },
    {
      "epoch": 1.9191188990882493,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9727434880932412e-05,
      "loss": 0.39126129150390626,
      "num_input_tokens_seen": 15425396160,
      "step": 54240,
      "train_runtime": 527334.4476,
      "train_tokens_per_second": 29251.638
    },
    {
      "epoch": 1.9194727193519783,
      "grad_norm": 0.625,
      "learning_rate": 1.972589958835532e-05,
      "loss": 0.3899538993835449,
      "num_input_tokens_seen": 15428272832,
      "step": 54250,
      "train_runtime": 527434.0875,
      "train_tokens_per_second": 29251.566
    },
    {
      "epoch": 1.9198265396157068,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9724364654176036e-05,
      "loss": 0.39056031703948973,
      "num_input_tokens_seen": 15431049664,
      "step": 54260,
      "train_runtime": 527527.3331,
      "train_tokens_per_second": 29251.659
    },
    {
      "epoch": 1.9201803598794358,
      "grad_norm": 0.6484375,
      "learning_rate": 1.9722830078255144e-05,
      "loss": 0.3884063720703125,
      "num_input_tokens_seen": 15433842304,
      "step": 54270,
      "train_runtime": 527622.3744,
      "train_tokens_per_second": 29251.683
    },
    {
      "epoch": 1.9205341801431646,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9721295860453297e-05,
      "loss": 0.3904250621795654,
      "num_input_tokens_seen": 15436725376,
      "step": 54280,
      "train_runtime": 527723.882,
      "train_tokens_per_second": 29251.519
    },
    {
      "epoch": 1.9208880004068933,
      "grad_norm": 0.609375,
      "learning_rate": 1.971976200063123e-05,
      "loss": 0.38887896537780764,
      "num_input_tokens_seen": 15439570304,
      "step": 54290,
      "train_runtime": 527819.9619,
      "train_tokens_per_second": 29251.585
    },
    {
      "epoch": 1.921241820670622,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9718228498649746e-05,
      "loss": 0.3917547225952148,
      "num_input_tokens_seen": 15442452736,
      "step": 54300,
      "train_runtime": 527922.1731,
      "train_tokens_per_second": 29251.381
    },
    {
      "epoch": 1.9215956409343509,
      "grad_norm": 0.625,
      "learning_rate": 1.9716695354369742e-05,
      "loss": 0.39081730842590334,
      "num_input_tokens_seen": 15445309888,
      "step": 54310,
      "train_runtime": 528019.7,
      "train_tokens_per_second": 29251.389
    },
    {
      "epoch": 1.9219494611980796,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9715162567652167e-05,
      "loss": 0.38950605392456056,
      "num_input_tokens_seen": 15448118336,
      "step": 54320,
      "train_runtime": 528116.2842,
      "train_tokens_per_second": 29251.358
    },
    {
      "epoch": 1.9223032814618084,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9713630138358067e-05,
      "loss": 0.3867368221282959,
      "num_input_tokens_seen": 15450951168,
      "step": 54330,
      "train_runtime": 528215.0257,
      "train_tokens_per_second": 29251.253
    },
    {
      "epoch": 1.9226571017255372,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9712098066348543e-05,
      "loss": 0.390903902053833,
      "num_input_tokens_seen": 15453823552,
      "step": 54340,
      "train_runtime": 528313.835,
      "train_tokens_per_second": 29251.219
    },
    {
      "epoch": 1.923010921989266,
      "grad_norm": 0.62109375,
      "learning_rate": 1.971056635148479e-05,
      "loss": 0.38699500560760497,
      "num_input_tokens_seen": 15456641280,
      "step": 54350,
      "train_runtime": 528410.2386,
      "train_tokens_per_second": 29251.215
    },
    {
      "epoch": 1.923364742252995,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9709034993628065e-05,
      "loss": 0.38905603885650636,
      "num_input_tokens_seen": 15459451392,
      "step": 54360,
      "train_runtime": 528505.8363,
      "train_tokens_per_second": 29251.241
    },
    {
      "epoch": 1.9237185625167235,
      "grad_norm": 0.609375,
      "learning_rate": 1.9707503992639716e-05,
      "loss": 0.38847765922546384,
      "num_input_tokens_seen": 15462280512,
      "step": 54370,
      "train_runtime": 528602.5439,
      "train_tokens_per_second": 29251.241
    },
    {
      "epoch": 1.9240723827804525,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9705973348381145e-05,
      "loss": 0.3889357328414917,
      "num_input_tokens_seen": 15465139904,
      "step": 54380,
      "train_runtime": 528699.0955,
      "train_tokens_per_second": 29251.308
    },
    {
      "epoch": 1.924426203044181,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9704443060713846e-05,
      "loss": 0.3833740234375,
      "num_input_tokens_seen": 15467953216,
      "step": 54390,
      "train_runtime": 528794.5454,
      "train_tokens_per_second": 29251.348
    },
    {
      "epoch": 1.92478002330791,
      "grad_norm": 0.625,
      "learning_rate": 1.9702913129499384e-05,
      "loss": 0.3912911653518677,
      "num_input_tokens_seen": 15470823872,
      "step": 54400,
      "train_runtime": 528889.608,
      "train_tokens_per_second": 29251.518
    },
    {
      "epoch": 1.9251338435716385,
      "grad_norm": 0.59375,
      "learning_rate": 1.9701383554599397e-05,
      "loss": 0.3886124610900879,
      "num_input_tokens_seen": 15473629376,
      "step": 54410,
      "train_runtime": 528988.0865,
      "train_tokens_per_second": 29251.376
    },
    {
      "epoch": 1.9254876638353675,
      "grad_norm": 0.640625,
      "learning_rate": 1.9699854335875603e-05,
      "loss": 0.3872699499130249,
      "num_input_tokens_seen": 15476444864,
      "step": 54420,
      "train_runtime": 529083.0722,
      "train_tokens_per_second": 29251.446
    },
    {
      "epoch": 1.925841484099096,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9698325473189788e-05,
      "loss": 0.38746678829193115,
      "num_input_tokens_seen": 15479297408,
      "step": 54430,
      "train_runtime": 529180.0971,
      "train_tokens_per_second": 29251.473
    },
    {
      "epoch": 1.926195304362825,
      "grad_norm": 0.62109375,
      "learning_rate": 1.969679696640382e-05,
      "loss": 0.3851868391036987,
      "num_input_tokens_seen": 15482150528,
      "step": 54440,
      "train_runtime": 529276.4678,
      "train_tokens_per_second": 29251.538
    },
    {
      "epoch": 1.9265491246265538,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9695268815379634e-05,
      "loss": 0.39093623161315916,
      "num_input_tokens_seen": 15485021312,
      "step": 54450,
      "train_runtime": 529377.4337,
      "train_tokens_per_second": 29251.382
    },
    {
      "epoch": 1.9269029448902826,
      "grad_norm": 0.5859375,
      "learning_rate": 1.9693741019979244e-05,
      "loss": 0.3886695384979248,
      "num_input_tokens_seen": 15487830080,
      "step": 54460,
      "train_runtime": 529471.2129,
      "train_tokens_per_second": 29251.505
    },
    {
      "epoch": 1.9272567651540113,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9692213580064744e-05,
      "loss": 0.3880319595336914,
      "num_input_tokens_seen": 15490663424,
      "step": 54470,
      "train_runtime": 529571.1838,
      "train_tokens_per_second": 29251.334
    },
    {
      "epoch": 1.92761058541774,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9690686495498302e-05,
      "loss": 0.38658292293548585,
      "num_input_tokens_seen": 15493534464,
      "step": 54480,
      "train_runtime": 529670.4976,
      "train_tokens_per_second": 29251.269
    },
    {
      "epoch": 1.9279644056814689,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9689159766142147e-05,
      "loss": 0.38579521179199217,
      "num_input_tokens_seen": 15496378368,
      "step": 54490,
      "train_runtime": 529767.244,
      "train_tokens_per_second": 29251.296
    },
    {
      "epoch": 1.9283182259451976,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9687633391858605e-05,
      "loss": 0.3875744104385376,
      "num_input_tokens_seen": 15499240064,
      "step": 54500,
      "train_runtime": 529866.2531,
      "train_tokens_per_second": 29251.231
    },
    {
      "epoch": 1.9286720462089264,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9686107372510057e-05,
      "loss": 0.390608549118042,
      "num_input_tokens_seen": 15502129664,
      "step": 54510,
      "train_runtime": 529963.7982,
      "train_tokens_per_second": 29251.299
    },
    {
      "epoch": 1.9290258664726552,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9684581707958966e-05,
      "loss": 0.38645851612091064,
      "num_input_tokens_seen": 15504955584,
      "step": 54520,
      "train_runtime": 530059.3261,
      "train_tokens_per_second": 29251.359
    },
    {
      "epoch": 1.9293796867363842,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9683056398067875e-05,
      "loss": 0.39103331565856936,
      "num_input_tokens_seen": 15507822912,
      "step": 54530,
      "train_runtime": 530159.4858,
      "train_tokens_per_second": 29251.241
    },
    {
      "epoch": 1.9297335070001127,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9681531442699394e-05,
      "loss": 0.39015212059021,
      "num_input_tokens_seen": 15510617088,
      "step": 54540,
      "train_runtime": 530256.33,
      "train_tokens_per_second": 29251.168
    },
    {
      "epoch": 1.9300873272638417,
      "grad_norm": 0.640625,
      "learning_rate": 1.9680006841716207e-05,
      "loss": 0.38644578456878664,
      "num_input_tokens_seen": 15513424640,
      "step": 54550,
      "train_runtime": 530354.1682,
      "train_tokens_per_second": 29251.066
    },
    {
      "epoch": 1.9304411475275702,
      "grad_norm": 0.6015625,
      "learning_rate": 1.967848259498108e-05,
      "loss": 0.385676646232605,
      "num_input_tokens_seen": 15516254272,
      "step": 54560,
      "train_runtime": 530449.7798,
      "train_tokens_per_second": 29251.128
    },
    {
      "epoch": 1.9307949677912992,
      "grad_norm": 0.625,
      "learning_rate": 1.967695870235685e-05,
      "loss": 0.38791708946228026,
      "num_input_tokens_seen": 15519045440,
      "step": 54570,
      "train_runtime": 530546.9817,
      "train_tokens_per_second": 29251.03
    },
    {
      "epoch": 1.9311487880550278,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9675435163706427e-05,
      "loss": 0.3904177904129028,
      "num_input_tokens_seen": 15521874368,
      "step": 54580,
      "train_runtime": 530641.2351,
      "train_tokens_per_second": 29251.165
    },
    {
      "epoch": 1.9315026083187568,
      "grad_norm": 0.62890625,
      "learning_rate": 1.967391197889279e-05,
      "loss": 0.39180641174316405,
      "num_input_tokens_seen": 15524670592,
      "step": 54590,
      "train_runtime": 530734.8551,
      "train_tokens_per_second": 29251.274
    },
    {
      "epoch": 1.9318564285824853,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9672389147779e-05,
      "loss": 0.3862171173095703,
      "num_input_tokens_seen": 15527475904,
      "step": 54600,
      "train_runtime": 530828.7138,
      "train_tokens_per_second": 29251.387
    },
    {
      "epoch": 1.9322102488462143,
      "grad_norm": 0.63671875,
      "learning_rate": 1.9670866670228194e-05,
      "loss": 0.3870074272155762,
      "num_input_tokens_seen": 15530364736,
      "step": 54610,
      "train_runtime": 530926.2054,
      "train_tokens_per_second": 29251.456
    },
    {
      "epoch": 1.932564069109943,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9669344546103576e-05,
      "loss": 0.3859997749328613,
      "num_input_tokens_seen": 15533226624,
      "step": 54620,
      "train_runtime": 531025.9391,
      "train_tokens_per_second": 29251.352
    },
    {
      "epoch": 1.9329178893736718,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9667822775268428e-05,
      "loss": 0.3879380226135254,
      "num_input_tokens_seen": 15536062208,
      "step": 54630,
      "train_runtime": 531119.9328,
      "train_tokens_per_second": 29251.514
    },
    {
      "epoch": 1.9332717096374006,
      "grad_norm": 0.625,
      "learning_rate": 1.9666301357586107e-05,
      "loss": 0.3917035341262817,
      "num_input_tokens_seen": 15538880256,
      "step": 54640,
      "train_runtime": 531215.0583,
      "train_tokens_per_second": 29251.581
    },
    {
      "epoch": 1.9336255299011293,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9664780292920036e-05,
      "loss": 0.3878628730773926,
      "num_input_tokens_seen": 15541767296,
      "step": 54650,
      "train_runtime": 531315.2997,
      "train_tokens_per_second": 29251.496
    },
    {
      "epoch": 1.9339793501648581,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9663259581133726e-05,
      "loss": 0.3857172966003418,
      "num_input_tokens_seen": 15544632128,
      "step": 54660,
      "train_runtime": 531412.7246,
      "train_tokens_per_second": 29251.524
    },
    {
      "epoch": 1.9343331704285869,
      "grad_norm": 0.59375,
      "learning_rate": 1.9661739222090753e-05,
      "loss": 0.3852613210678101,
      "num_input_tokens_seen": 15547487360,
      "step": 54670,
      "train_runtime": 531510.6582,
      "train_tokens_per_second": 29251.506
    },
    {
      "epoch": 1.9346869906923156,
      "grad_norm": 0.62890625,
      "learning_rate": 1.966021921565476e-05,
      "loss": 0.387258243560791,
      "num_input_tokens_seen": 15550336384,
      "step": 54680,
      "train_runtime": 531608.558,
      "train_tokens_per_second": 29251.479
    },
    {
      "epoch": 1.9350408109560444,
      "grad_norm": 0.60546875,
      "learning_rate": 1.965869956168948e-05,
      "loss": 0.3893598556518555,
      "num_input_tokens_seen": 15553125504,
      "step": 54690,
      "train_runtime": 531701.0293,
      "train_tokens_per_second": 29251.637
    },
    {
      "epoch": 1.9353946312197734,
      "grad_norm": 0.6328125,
      "learning_rate": 1.965718026005871e-05,
      "loss": 0.38754892349243164,
      "num_input_tokens_seen": 15555997760,
      "step": 54700,
      "train_runtime": 531798.7089,
      "train_tokens_per_second": 29251.665
    },
    {
      "epoch": 1.935748451483502,
      "grad_norm": 0.59375,
      "learning_rate": 1.9655661310626317e-05,
      "loss": 0.39072866439819337,
      "num_input_tokens_seen": 15558850560,
      "step": 54710,
      "train_runtime": 531896.8391,
      "train_tokens_per_second": 29251.632
    },
    {
      "epoch": 1.936102271747231,
      "grad_norm": 0.5859375,
      "learning_rate": 1.965414271325625e-05,
      "loss": 0.38925814628601074,
      "num_input_tokens_seen": 15561740672,
      "step": 54720,
      "train_runtime": 531995.1334,
      "train_tokens_per_second": 29251.66
    },
    {
      "epoch": 1.9364560920109595,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9652624467812534e-05,
      "loss": 0.3912940740585327,
      "num_input_tokens_seen": 15564547840,
      "step": 54730,
      "train_runtime": 532090.7253,
      "train_tokens_per_second": 29251.68
    },
    {
      "epoch": 1.9368099122746885,
      "grad_norm": 0.59765625,
      "learning_rate": 1.965110657415925e-05,
      "loss": 0.3876797199249268,
      "num_input_tokens_seen": 15567381760,
      "step": 54740,
      "train_runtime": 532185.5498,
      "train_tokens_per_second": 29251.793
    },
    {
      "epoch": 1.937163732538417,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9649589032160576e-05,
      "loss": 0.38842966556549074,
      "num_input_tokens_seen": 15570193216,
      "step": 54750,
      "train_runtime": 532280.3788,
      "train_tokens_per_second": 29251.864
    },
    {
      "epoch": 1.937517552802146,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9648071841680746e-05,
      "loss": 0.38835141658782957,
      "num_input_tokens_seen": 15573023872,
      "step": 54760,
      "train_runtime": 532375.9339,
      "train_tokens_per_second": 29251.931
    },
    {
      "epoch": 1.9378713730658745,
      "grad_norm": 0.61328125,
      "learning_rate": 1.964655500258407e-05,
      "loss": 0.3879629850387573,
      "num_input_tokens_seen": 15575857152,
      "step": 54770,
      "train_runtime": 532471.4762,
      "train_tokens_per_second": 29252.003
    },
    {
      "epoch": 1.9382251933296035,
      "grad_norm": 0.625,
      "learning_rate": 1.964503851473494e-05,
      "loss": 0.3855945587158203,
      "num_input_tokens_seen": 15578714112,
      "step": 54780,
      "train_runtime": 532572.447,
      "train_tokens_per_second": 29251.821
    },
    {
      "epoch": 1.9385790135933323,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9643522377997807e-05,
      "loss": 0.3841076374053955,
      "num_input_tokens_seen": 15581578688,
      "step": 54790,
      "train_runtime": 532673.4553,
      "train_tokens_per_second": 29251.652
    },
    {
      "epoch": 1.938932833857061,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9642006592237216e-05,
      "loss": 0.3880176305770874,
      "num_input_tokens_seen": 15584446336,
      "step": 54800,
      "train_runtime": 532777.2403,
      "train_tokens_per_second": 29251.337
    },
    {
      "epoch": 1.9392866541207898,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9640491157317767e-05,
      "loss": 0.38619556427001955,
      "num_input_tokens_seen": 15587297472,
      "step": 54810,
      "train_runtime": 532874.7419,
      "train_tokens_per_second": 29251.335
    },
    {
      "epoch": 1.9396404743845186,
      "grad_norm": 0.5859375,
      "learning_rate": 1.9638976073104138e-05,
      "loss": 0.38499722480773924,
      "num_input_tokens_seen": 15590176064,
      "step": 54820,
      "train_runtime": 532976.3237,
      "train_tokens_per_second": 29251.161
    },
    {
      "epoch": 1.9399942946482474,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9637461339461084e-05,
      "loss": 0.3904792308807373,
      "num_input_tokens_seen": 15593021568,
      "step": 54830,
      "train_runtime": 533074.5882,
      "train_tokens_per_second": 29251.107
    },
    {
      "epoch": 1.9403481149119761,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9635946956253426e-05,
      "loss": 0.3888221740722656,
      "num_input_tokens_seen": 15595863488,
      "step": 54840,
      "train_runtime": 533171.6126,
      "train_tokens_per_second": 29251.114
    },
    {
      "epoch": 1.9407019351757049,
      "grad_norm": 0.625,
      "learning_rate": 1.9634432923346063e-05,
      "loss": 0.3841134548187256,
      "num_input_tokens_seen": 15598663552,
      "step": 54850,
      "train_runtime": 533266.4843,
      "train_tokens_per_second": 29251.161
    },
    {
      "epoch": 1.9410557554394336,
      "grad_norm": 0.6484375,
      "learning_rate": 1.963291924060397e-05,
      "loss": 0.3886404514312744,
      "num_input_tokens_seen": 15601518336,
      "step": 54860,
      "train_runtime": 533364.9072,
      "train_tokens_per_second": 29251.115
    },
    {
      "epoch": 1.9414095757031626,
      "grad_norm": 0.60546875,
      "learning_rate": 1.963140590789219e-05,
      "loss": 0.3887904644012451,
      "num_input_tokens_seen": 15604354688,
      "step": 54870,
      "train_runtime": 533460.479,
      "train_tokens_per_second": 29251.192
    },
    {
      "epoch": 1.9417633959668912,
      "grad_norm": 0.63671875,
      "learning_rate": 1.9629892925075834e-05,
      "loss": 0.389663028717041,
      "num_input_tokens_seen": 15607185728,
      "step": 54880,
      "train_runtime": 533557.067,
      "train_tokens_per_second": 29251.202
    },
    {
      "epoch": 1.9421172162306202,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9628380292020093e-05,
      "loss": 0.38409695625305174,
      "num_input_tokens_seen": 15609983744,
      "step": 54890,
      "train_runtime": 533650.4948,
      "train_tokens_per_second": 29251.324
    },
    {
      "epoch": 1.9424710364943487,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9626868008590238e-05,
      "loss": 0.3847958564758301,
      "num_input_tokens_seen": 15612857152,
      "step": 54900,
      "train_runtime": 533748.9016,
      "train_tokens_per_second": 29251.315
    },
    {
      "epoch": 1.9428248567580777,
      "grad_norm": 0.625,
      "learning_rate": 1.9625356074651595e-05,
      "loss": 0.3890111207962036,
      "num_input_tokens_seen": 15615759808,
      "step": 54910,
      "train_runtime": 533849.0432,
      "train_tokens_per_second": 29251.265
    },
    {
      "epoch": 1.9431786770218062,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9623844490069577e-05,
      "loss": 0.388214111328125,
      "num_input_tokens_seen": 15618584640,
      "step": 54920,
      "train_runtime": 533944.4144,
      "train_tokens_per_second": 29251.331
    },
    {
      "epoch": 1.9435324972855352,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9622333254709655e-05,
      "loss": 0.38620638847351074,
      "num_input_tokens_seen": 15621465664,
      "step": 54930,
      "train_runtime": 534043.6179,
      "train_tokens_per_second": 29251.292
    },
    {
      "epoch": 1.943886317549264,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9620822368437388e-05,
      "loss": 0.3910114288330078,
      "num_input_tokens_seen": 15624342592,
      "step": 54940,
      "train_runtime": 534142.347,
      "train_tokens_per_second": 29251.271
    },
    {
      "epoch": 1.9442401378129928,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9619311831118405e-05,
      "loss": 0.3886824607849121,
      "num_input_tokens_seen": 15627171520,
      "step": 54950,
      "train_runtime": 534239.2919,
      "train_tokens_per_second": 29251.258
    },
    {
      "epoch": 1.9445939580767215,
      "grad_norm": 0.60546875,
      "learning_rate": 1.961780164261839e-05,
      "loss": 0.3930304288864136,
      "num_input_tokens_seen": 15629987008,
      "step": 54960,
      "train_runtime": 534333.8572,
      "train_tokens_per_second": 29251.351
    },
    {
      "epoch": 1.9449477783404503,
      "grad_norm": 0.61328125,
      "learning_rate": 1.961629180280313e-05,
      "loss": 0.39078459739685056,
      "num_input_tokens_seen": 15632884096,
      "step": 54970,
      "train_runtime": 534436.5523,
      "train_tokens_per_second": 29251.151
    },
    {
      "epoch": 1.945301598604179,
      "grad_norm": 0.64453125,
      "learning_rate": 1.9614782311538455e-05,
      "loss": 0.38828024864196775,
      "num_input_tokens_seen": 15635758080,
      "step": 54980,
      "train_runtime": 534533.7213,
      "train_tokens_per_second": 29251.21
    },
    {
      "epoch": 1.9456554188679078,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9613273168690278e-05,
      "loss": 0.3859945058822632,
      "num_input_tokens_seen": 15638642944,
      "step": 54990,
      "train_runtime": 534631.5177,
      "train_tokens_per_second": 29251.255
    },
    {
      "epoch": 1.9460092391316366,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9611764374124592e-05,
      "loss": 0.39281420707702636,
      "num_input_tokens_seen": 15641534336,
      "step": 55000,
      "train_runtime": 534733.4733,
      "train_tokens_per_second": 29251.085
    },
    {
      "epoch": 1.9460092391316366,
      "eval_loss": 0.33642661571502686,
      "eval_runtime": 8.4306,
      "eval_samples_per_second": 473.041,
      "eval_steps_per_second": 3.796,
      "num_input_tokens_seen": 15641534336,
      "step": 55000
    },
    {
      "epoch": 1.9463630593953654,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9610255927707456e-05,
      "loss": 0.39242579936981203,
      "num_input_tokens_seen": 15644404544,
      "step": 55010,
      "train_runtime": 534859.6357,
      "train_tokens_per_second": 29249.552
    },
    {
      "epoch": 1.9467168796590941,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9608747829304998e-05,
      "loss": 0.38872780799865725,
      "num_input_tokens_seen": 15647263296,
      "step": 55020,
      "train_runtime": 534959.4949,
      "train_tokens_per_second": 29249.436
    },
    {
      "epoch": 1.947070699922823,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9607240078783416e-05,
      "loss": 0.38651509284973146,
      "num_input_tokens_seen": 15650120320,
      "step": 55030,
      "train_runtime": 535056.7762,
      "train_tokens_per_second": 29249.457
    },
    {
      "epoch": 1.9474245201865519,
      "grad_norm": 0.609375,
      "learning_rate": 1.960573267600899e-05,
      "loss": 0.3863197326660156,
      "num_input_tokens_seen": 15652989696,
      "step": 55040,
      "train_runtime": 535155.7273,
      "train_tokens_per_second": 29249.411
    },
    {
      "epoch": 1.9477783404502804,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9604225620848072e-05,
      "loss": 0.3943453788757324,
      "num_input_tokens_seen": 15655835264,
      "step": 55050,
      "train_runtime": 535254.4405,
      "train_tokens_per_second": 29249.333
    },
    {
      "epoch": 1.9481321607140094,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9602718913167075e-05,
      "loss": 0.3903839349746704,
      "num_input_tokens_seen": 15658645696,
      "step": 55060,
      "train_runtime": 535349.621,
      "train_tokens_per_second": 29249.382
    },
    {
      "epoch": 1.948485980977738,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9601212552832483e-05,
      "loss": 0.3878039360046387,
      "num_input_tokens_seen": 15661494656,
      "step": 55070,
      "train_runtime": 535448.9371,
      "train_tokens_per_second": 29249.278
    },
    {
      "epoch": 1.948839801241467,
      "grad_norm": 0.61328125,
      "learning_rate": 1.959970653971087e-05,
      "loss": 0.3857141017913818,
      "num_input_tokens_seen": 15664384128,
      "step": 55080,
      "train_runtime": 535549.4843,
      "train_tokens_per_second": 29249.182
    },
    {
      "epoch": 1.9491936215051955,
      "grad_norm": 0.65234375,
      "learning_rate": 1.9598200873668862e-05,
      "loss": 0.38744616508483887,
      "num_input_tokens_seen": 15667203200,
      "step": 55090,
      "train_runtime": 535644.6271,
      "train_tokens_per_second": 29249.249
    },
    {
      "epoch": 1.9495474417689245,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9596695554573167e-05,
      "loss": 0.388596248626709,
      "num_input_tokens_seen": 15670068224,
      "step": 55100,
      "train_runtime": 535746.8419,
      "train_tokens_per_second": 29249.017
    },
    {
      "epoch": 1.9499012620326532,
      "grad_norm": 0.625,
      "learning_rate": 1.959519058229056e-05,
      "loss": 0.3864539861679077,
      "num_input_tokens_seen": 15672869056,
      "step": 55110,
      "train_runtime": 535842.5524,
      "train_tokens_per_second": 29249.019
    },
    {
      "epoch": 1.950255082296382,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9593685956687895e-05,
      "loss": 0.3895367860794067,
      "num_input_tokens_seen": 15675753984,
      "step": 55120,
      "train_runtime": 535940.4164,
      "train_tokens_per_second": 29249.061
    },
    {
      "epoch": 1.9506089025601108,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9592181677632094e-05,
      "loss": 0.3872324705123901,
      "num_input_tokens_seen": 15678597760,
      "step": 55130,
      "train_runtime": 536038.4367,
      "train_tokens_per_second": 29249.018
    },
    {
      "epoch": 1.9509627228238395,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9590677744990145e-05,
      "loss": 0.3908626794815063,
      "num_input_tokens_seen": 15681442496,
      "step": 55140,
      "train_runtime": 536136.6538,
      "train_tokens_per_second": 29248.966
    },
    {
      "epoch": 1.9513165430875683,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9589174158629106e-05,
      "loss": 0.3930805206298828,
      "num_input_tokens_seen": 15684275456,
      "step": 55150,
      "train_runtime": 536233.9829,
      "train_tokens_per_second": 29248.94
    },
    {
      "epoch": 1.951670363351297,
      "grad_norm": 0.58984375,
      "learning_rate": 1.958767091841612e-05,
      "loss": 0.38980922698974607,
      "num_input_tokens_seen": 15687186944,
      "step": 55160,
      "train_runtime": 536332.8837,
      "train_tokens_per_second": 29248.975
    },
    {
      "epoch": 1.9520241836150258,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9586168024218395e-05,
      "loss": 0.3865967273712158,
      "num_input_tokens_seen": 15690080128,
      "step": 55170,
      "train_runtime": 536433.4917,
      "train_tokens_per_second": 29248.882
    },
    {
      "epoch": 1.9523780038787546,
      "grad_norm": 0.578125,
      "learning_rate": 1.9584665475903198e-05,
      "loss": 0.38882744312286377,
      "num_input_tokens_seen": 15692963840,
      "step": 55180,
      "train_runtime": 536533.9838,
      "train_tokens_per_second": 29248.779
    },
    {
      "epoch": 1.9527318241424836,
      "grad_norm": 0.59375,
      "learning_rate": 1.9583163273337892e-05,
      "loss": 0.39128913879394533,
      "num_input_tokens_seen": 15695856960,
      "step": 55190,
      "train_runtime": 536634.8755,
      "train_tokens_per_second": 29248.671
    },
    {
      "epoch": 1.9530856444062121,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9581661416389887e-05,
      "loss": 0.3890889883041382,
      "num_input_tokens_seen": 15698729088,
      "step": 55200,
      "train_runtime": 536735.4092,
      "train_tokens_per_second": 29248.544
    },
    {
      "epoch": 1.9534394646699411,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9580159904926674e-05,
      "loss": 0.38544507026672364,
      "num_input_tokens_seen": 15701555520,
      "step": 55210,
      "train_runtime": 536829.7376,
      "train_tokens_per_second": 29248.669
    },
    {
      "epoch": 1.9537932849336697,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9578658738815823e-05,
      "loss": 0.39132275581359866,
      "num_input_tokens_seen": 15704386112,
      "step": 55220,
      "train_runtime": 536928.25,
      "train_tokens_per_second": 29248.575
    },
    {
      "epoch": 1.9541471051973986,
      "grad_norm": 0.6171875,
      "learning_rate": 1.957715791792496e-05,
      "loss": 0.3887802600860596,
      "num_input_tokens_seen": 15707238720,
      "step": 55230,
      "train_runtime": 537027.4782,
      "train_tokens_per_second": 29248.482
    },
    {
      "epoch": 1.9545009254611272,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9575657442121788e-05,
      "loss": 0.38661653995513917,
      "num_input_tokens_seen": 15710040640,
      "step": 55240,
      "train_runtime": 537122.3265,
      "train_tokens_per_second": 29248.534
    },
    {
      "epoch": 1.9548547457248562,
      "grad_norm": 0.58984375,
      "learning_rate": 1.957415731127409e-05,
      "loss": 0.390841007232666,
      "num_input_tokens_seen": 15712885248,
      "step": 55250,
      "train_runtime": 537220.5205,
      "train_tokens_per_second": 29248.483
    },
    {
      "epoch": 1.9552085659885847,
      "grad_norm": 0.62890625,
      "learning_rate": 1.957265752524971e-05,
      "loss": 0.38548755645751953,
      "num_input_tokens_seen": 15715707648,
      "step": 55260,
      "train_runtime": 537315.4663,
      "train_tokens_per_second": 29248.567
    },
    {
      "epoch": 1.9555623862523137,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9571158083916562e-05,
      "loss": 0.3947559118270874,
      "num_input_tokens_seen": 15718585024,
      "step": 55270,
      "train_runtime": 537415.503,
      "train_tokens_per_second": 29248.477
    },
    {
      "epoch": 1.9559162065160425,
      "grad_norm": 0.625,
      "learning_rate": 1.9569658987142637e-05,
      "loss": 0.39704484939575196,
      "num_input_tokens_seen": 15721424576,
      "step": 55280,
      "train_runtime": 537511.4409,
      "train_tokens_per_second": 29248.539
    },
    {
      "epoch": 1.9562700267797712,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9568160234795992e-05,
      "loss": 0.39054360389709475,
      "num_input_tokens_seen": 15724310208,
      "step": 55290,
      "train_runtime": 537611.4371,
      "train_tokens_per_second": 29248.467
    },
    {
      "epoch": 1.9566238470435,
      "grad_norm": 0.61328125,
      "learning_rate": 1.956666182674476e-05,
      "loss": 0.38492422103881835,
      "num_input_tokens_seen": 15727165312,
      "step": 55300,
      "train_runtime": 537710.8509,
      "train_tokens_per_second": 29248.369
    },
    {
      "epoch": 1.9569776673072288,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9565163762857133e-05,
      "loss": 0.3898935079574585,
      "num_input_tokens_seen": 15730059968,
      "step": 55310,
      "train_runtime": 537812.0899,
      "train_tokens_per_second": 29248.245
    },
    {
      "epoch": 1.9573314875709575,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9563666043001394e-05,
      "loss": 0.387939977645874,
      "num_input_tokens_seen": 15732864960,
      "step": 55320,
      "train_runtime": 537908.2302,
      "train_tokens_per_second": 29248.232
    },
    {
      "epoch": 1.9576853078346863,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9562168667045876e-05,
      "loss": 0.388346004486084,
      "num_input_tokens_seen": 15735718464,
      "step": 55330,
      "train_runtime": 538005.2773,
      "train_tokens_per_second": 29248.26
    },
    {
      "epoch": 1.958039128098415,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9560671634858988e-05,
      "loss": 0.3887749910354614,
      "num_input_tokens_seen": 15738593088,
      "step": 55340,
      "train_runtime": 538103.4229,
      "train_tokens_per_second": 29248.268
    },
    {
      "epoch": 1.9583929483621438,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9559174946309216e-05,
      "loss": 0.38655991554260255,
      "num_input_tokens_seen": 15741446784,
      "step": 55350,
      "train_runtime": 538198.8122,
      "train_tokens_per_second": 29248.386
    },
    {
      "epoch": 1.9587467686258728,
      "grad_norm": 0.6328125,
      "learning_rate": 1.955767860126512e-05,
      "loss": 0.390248966217041,
      "num_input_tokens_seen": 15744336704,
      "step": 55360,
      "train_runtime": 538297.5492,
      "train_tokens_per_second": 29248.39
    },
    {
      "epoch": 1.9591005888896014,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9556182599595313e-05,
      "loss": 0.3863346576690674,
      "num_input_tokens_seen": 15747229056,
      "step": 55370,
      "train_runtime": 538395.4857,
      "train_tokens_per_second": 29248.442
    },
    {
      "epoch": 1.9594544091533304,
      "grad_norm": 0.59765625,
      "learning_rate": 1.95546869411685e-05,
      "loss": 0.3862210035324097,
      "num_input_tokens_seen": 15750091136,
      "step": 55380,
      "train_runtime": 538491.7963,
      "train_tokens_per_second": 29248.526
    },
    {
      "epoch": 1.959808229417059,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9553191625853427e-05,
      "loss": 0.3949150085449219,
      "num_input_tokens_seen": 15752914624,
      "step": 55390,
      "train_runtime": 538588.7906,
      "train_tokens_per_second": 29248.501
    },
    {
      "epoch": 1.9601620496807879,
      "grad_norm": 0.64453125,
      "learning_rate": 1.9551696653518943e-05,
      "loss": 0.3891268491744995,
      "num_input_tokens_seen": 15755769664,
      "step": 55400,
      "train_runtime": 538687.1132,
      "train_tokens_per_second": 29248.462
    },
    {
      "epoch": 1.9605158699445164,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9550202024033947e-05,
      "loss": 0.3915924310684204,
      "num_input_tokens_seen": 15758620224,
      "step": 55410,
      "train_runtime": 538782.619,
      "train_tokens_per_second": 29248.568
    },
    {
      "epoch": 1.9608696902082454,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9548707737267412e-05,
      "loss": 0.38542590141296384,
      "num_input_tokens_seen": 15761456640,
      "step": 55420,
      "train_runtime": 538878.9925,
      "train_tokens_per_second": 29248.601
    },
    {
      "epoch": 1.961223510471974,
      "grad_norm": 0.64453125,
      "learning_rate": 1.9547213793088384e-05,
      "loss": 0.39220314025878905,
      "num_input_tokens_seen": 15764266944,
      "step": 55430,
      "train_runtime": 538974.1982,
      "train_tokens_per_second": 29248.649
    },
    {
      "epoch": 1.961577330735703,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9545720191365983e-05,
      "loss": 0.38616414070129396,
      "num_input_tokens_seen": 15767081344,
      "step": 55440,
      "train_runtime": 539069.0127,
      "train_tokens_per_second": 29248.725
    },
    {
      "epoch": 1.9619311509994317,
      "grad_norm": 0.59375,
      "learning_rate": 1.9544226931969382e-05,
      "loss": 0.3876357078552246,
      "num_input_tokens_seen": 15769919872,
      "step": 55450,
      "train_runtime": 539166.2276,
      "train_tokens_per_second": 29248.716
    },
    {
      "epoch": 1.9622849712631605,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9542734014767848e-05,
      "loss": 0.39226813316345216,
      "num_input_tokens_seen": 15772834752,
      "step": 55460,
      "train_runtime": 539267.1381,
      "train_tokens_per_second": 29248.648
    },
    {
      "epoch": 1.9626387915268892,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9541241439630693e-05,
      "loss": 0.38219451904296875,
      "num_input_tokens_seen": 15775712128,
      "step": 55470,
      "train_runtime": 539368.8579,
      "train_tokens_per_second": 29248.467
    },
    {
      "epoch": 1.962992611790618,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9539749206427322e-05,
      "loss": 0.3890096664428711,
      "num_input_tokens_seen": 15778560000,
      "step": 55480,
      "train_runtime": 539466.8098,
      "train_tokens_per_second": 29248.435
    },
    {
      "epoch": 1.9633464320543468,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9538257315027194e-05,
      "loss": 0.38710412979125974,
      "num_input_tokens_seen": 15781436992,
      "step": 55490,
      "train_runtime": 539567.1331,
      "train_tokens_per_second": 29248.329
    },
    {
      "epoch": 1.9637002523180755,
      "grad_norm": 0.62109375,
      "learning_rate": 1.953676576529984e-05,
      "loss": 0.387100887298584,
      "num_input_tokens_seen": 15784325056,
      "step": 55500,
      "train_runtime": 539666.7467,
      "train_tokens_per_second": 29248.282
    },
    {
      "epoch": 1.9640540725818043,
      "grad_norm": 0.61328125,
      "learning_rate": 1.953527455711487e-05,
      "loss": 0.38877167701721194,
      "num_input_tokens_seen": 15787178496,
      "step": 55510,
      "train_runtime": 539763.6015,
      "train_tokens_per_second": 29248.32
    },
    {
      "epoch": 1.964407892845533,
      "grad_norm": 0.59375,
      "learning_rate": 1.9533783690341955e-05,
      "loss": 0.389346981048584,
      "num_input_tokens_seen": 15790024832,
      "step": 55520,
      "train_runtime": 539860.1604,
      "train_tokens_per_second": 29248.361
    },
    {
      "epoch": 1.964761713109262,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9532293164850833e-05,
      "loss": 0.38842344284057617,
      "num_input_tokens_seen": 15792909888,
      "step": 55530,
      "train_runtime": 539959.3154,
      "train_tokens_per_second": 29248.333
    },
    {
      "epoch": 1.9651155333729906,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9530802980511325e-05,
      "loss": 0.3873558044433594,
      "num_input_tokens_seen": 15795838272,
      "step": 55540,
      "train_runtime": 540060.8468,
      "train_tokens_per_second": 29248.257
    },
    {
      "epoch": 1.9654693536367196,
      "grad_norm": 0.609375,
      "learning_rate": 1.95293131371933e-05,
      "loss": 0.38912575244903563,
      "num_input_tokens_seen": 15798662848,
      "step": 55550,
      "train_runtime": 540156.1108,
      "train_tokens_per_second": 29248.328
    },
    {
      "epoch": 1.9658231739004481,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9527823634766724e-05,
      "loss": 0.3886443853378296,
      "num_input_tokens_seen": 15801563648,
      "step": 55560,
      "train_runtime": 540256.3085,
      "train_tokens_per_second": 29248.272
    },
    {
      "epoch": 1.9661769941641771,
      "grad_norm": 0.6015625,
      "learning_rate": 1.952633447310161e-05,
      "loss": 0.3855548858642578,
      "num_input_tokens_seen": 15804382336,
      "step": 55570,
      "train_runtime": 540352.1454,
      "train_tokens_per_second": 29248.301
    },
    {
      "epoch": 1.9665308144279057,
      "grad_norm": 0.64453125,
      "learning_rate": 1.9524845652068046e-05,
      "loss": 0.39172892570495604,
      "num_input_tokens_seen": 15807244096,
      "step": 55580,
      "train_runtime": 540448.7253,
      "train_tokens_per_second": 29248.37
    },
    {
      "epoch": 1.9668846346916347,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9523357171536195e-05,
      "loss": 0.395935583114624,
      "num_input_tokens_seen": 15810162048,
      "step": 55590,
      "train_runtime": 540551.5516,
      "train_tokens_per_second": 29248.204
    },
    {
      "epoch": 1.9672384549553632,
      "grad_norm": 0.625,
      "learning_rate": 1.9521869031376288e-05,
      "loss": 0.39039177894592286,
      "num_input_tokens_seen": 15813060352,
      "step": 55600,
      "train_runtime": 540650.1283,
      "train_tokens_per_second": 29248.232
    },
    {
      "epoch": 1.9675922752190922,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9520381231458616e-05,
      "loss": 0.3864325284957886,
      "num_input_tokens_seen": 15815886656,
      "step": 55610,
      "train_runtime": 540745.1056,
      "train_tokens_per_second": 29248.321
    },
    {
      "epoch": 1.967946095482821,
      "grad_norm": 0.65234375,
      "learning_rate": 1.9518893771653555e-05,
      "loss": 0.3902462959289551,
      "num_input_tokens_seen": 15818718272,
      "step": 55620,
      "train_runtime": 540839.5121,
      "train_tokens_per_second": 29248.452
    },
    {
      "epoch": 1.9682999157465497,
      "grad_norm": 0.59765625,
      "learning_rate": 1.951740665183153e-05,
      "loss": 0.38970136642456055,
      "num_input_tokens_seen": 15821595008,
      "step": 55630,
      "train_runtime": 540939.8841,
      "train_tokens_per_second": 29248.343
    },
    {
      "epoch": 1.9686537360102785,
      "grad_norm": 0.609375,
      "learning_rate": 1.9515919871863057e-05,
      "loss": 0.3938803195953369,
      "num_input_tokens_seen": 15824438784,
      "step": 55640,
      "train_runtime": 541038.8022,
      "train_tokens_per_second": 29248.251
    },
    {
      "epoch": 1.9690075562740073,
      "grad_norm": 0.609375,
      "learning_rate": 1.9514433431618707e-05,
      "loss": 0.3875724792480469,
      "num_input_tokens_seen": 15827281600,
      "step": 55650,
      "train_runtime": 541134.8987,
      "train_tokens_per_second": 29248.311
    },
    {
      "epoch": 1.969361376537736,
      "grad_norm": 0.59375,
      "learning_rate": 1.951294733096912e-05,
      "loss": 0.3896567106246948,
      "num_input_tokens_seen": 15830160896,
      "step": 55660,
      "train_runtime": 541235.4517,
      "train_tokens_per_second": 29248.197
    },
    {
      "epoch": 1.9697151968014648,
      "grad_norm": 0.60546875,
      "learning_rate": 1.951146156978501e-05,
      "loss": 0.39091081619262696,
      "num_input_tokens_seen": 15833006400,
      "step": 55670,
      "train_runtime": 541334.0766,
      "train_tokens_per_second": 29248.124
    },
    {
      "epoch": 1.9700690170651936,
      "grad_norm": 0.57421875,
      "learning_rate": 1.9509976147937163e-05,
      "loss": 0.3877918243408203,
      "num_input_tokens_seen": 15835852608,
      "step": 55680,
      "train_runtime": 541429.3107,
      "train_tokens_per_second": 29248.237
    },
    {
      "epoch": 1.9704228373289223,
      "grad_norm": 0.65234375,
      "learning_rate": 1.950849106529642e-05,
      "loss": 0.3898444652557373,
      "num_input_tokens_seen": 15838724736,
      "step": 55690,
      "train_runtime": 541528.9814,
      "train_tokens_per_second": 29248.157
    },
    {
      "epoch": 1.9707766575926513,
      "grad_norm": 0.6015625,
      "learning_rate": 1.950700632173371e-05,
      "loss": 0.3855905532836914,
      "num_input_tokens_seen": 15841599232,
      "step": 55700,
      "train_runtime": 541629.4067,
      "train_tokens_per_second": 29248.041
    },
    {
      "epoch": 1.9711304778563798,
      "grad_norm": 0.640625,
      "learning_rate": 1.9505521917120013e-05,
      "loss": 0.38884222507476807,
      "num_input_tokens_seen": 15844439424,
      "step": 55710,
      "train_runtime": 541727.5996,
      "train_tokens_per_second": 29247.983
    },
    {
      "epoch": 1.9714842981201088,
      "grad_norm": 0.65234375,
      "learning_rate": 1.9504037851326387e-05,
      "loss": 0.3871743679046631,
      "num_input_tokens_seen": 15847310976,
      "step": 55720,
      "train_runtime": 541828.6158,
      "train_tokens_per_second": 29247.83
    },
    {
      "epoch": 1.9718381183838374,
      "grad_norm": 0.59375,
      "learning_rate": 1.950255412422396e-05,
      "loss": 0.3910228729248047,
      "num_input_tokens_seen": 15850160768,
      "step": 55730,
      "train_runtime": 541924.5694,
      "train_tokens_per_second": 29247.91
    },
    {
      "epoch": 1.9721919386475664,
      "grad_norm": 0.56640625,
      "learning_rate": 1.9501070735683924e-05,
      "loss": 0.38800630569458006,
      "num_input_tokens_seen": 15853018368,
      "step": 55740,
      "train_runtime": 542020.6958,
      "train_tokens_per_second": 29247.995
    },
    {
      "epoch": 1.972545758911295,
      "grad_norm": 0.5859375,
      "learning_rate": 1.9499587685577537e-05,
      "loss": 0.3861918687820435,
      "num_input_tokens_seen": 15855846528,
      "step": 55750,
      "train_runtime": 542115.6907,
      "train_tokens_per_second": 29248.086
    },
    {
      "epoch": 1.972899579175024,
      "grad_norm": 0.59375,
      "learning_rate": 1.9498104973776137e-05,
      "loss": 0.3848401069641113,
      "num_input_tokens_seen": 15858727552,
      "step": 55760,
      "train_runtime": 542215.3893,
      "train_tokens_per_second": 29248.022
    },
    {
      "epoch": 1.9732533994387524,
      "grad_norm": 0.61328125,
      "learning_rate": 1.949662260015112e-05,
      "loss": 0.3858680009841919,
      "num_input_tokens_seen": 15861522944,
      "step": 55770,
      "train_runtime": 542310.0314,
      "train_tokens_per_second": 29248.072
    },
    {
      "epoch": 1.9736072197024814,
      "grad_norm": 0.59375,
      "learning_rate": 1.9495140564573953e-05,
      "loss": 0.38861143589019775,
      "num_input_tokens_seen": 15864311616,
      "step": 55780,
      "train_runtime": 542403.9228,
      "train_tokens_per_second": 29248.151
    },
    {
      "epoch": 1.9739610399662102,
      "grad_norm": 0.609375,
      "learning_rate": 1.9493658866916166e-05,
      "loss": 0.3888195514678955,
      "num_input_tokens_seen": 15867128704,
      "step": 55790,
      "train_runtime": 542499.9532,
      "train_tokens_per_second": 29248.166
    },
    {
      "epoch": 1.974314860229939,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9492177507049375e-05,
      "loss": 0.3892224788665771,
      "num_input_tokens_seen": 15869987776,
      "step": 55800,
      "train_runtime": 542595.498,
      "train_tokens_per_second": 29248.285
    },
    {
      "epoch": 1.9746686804936677,
      "grad_norm": 0.609375,
      "learning_rate": 1.9490696484845252e-05,
      "loss": 0.389974308013916,
      "num_input_tokens_seen": 15872832768,
      "step": 55810,
      "train_runtime": 542691.2321,
      "train_tokens_per_second": 29248.368
    },
    {
      "epoch": 1.9750225007573965,
      "grad_norm": 0.640625,
      "learning_rate": 1.948921580017553e-05,
      "loss": 0.39040675163269045,
      "num_input_tokens_seen": 15875630080,
      "step": 55820,
      "train_runtime": 542787.8396,
      "train_tokens_per_second": 29248.316
    },
    {
      "epoch": 1.9753763210211253,
      "grad_norm": 0.60546875,
      "learning_rate": 1.948773545291202e-05,
      "loss": 0.3840803623199463,
      "num_input_tokens_seen": 15878497024,
      "step": 55830,
      "train_runtime": 542886.1844,
      "train_tokens_per_second": 29248.298
    },
    {
      "epoch": 1.975730141284854,
      "grad_norm": 0.62109375,
      "learning_rate": 1.94862554429266e-05,
      "loss": 0.38790407180786135,
      "num_input_tokens_seen": 15881281920,
      "step": 55840,
      "train_runtime": 542980.2494,
      "train_tokens_per_second": 29248.36
    },
    {
      "epoch": 1.9760839615485828,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9484775770091214e-05,
      "loss": 0.38820159435272217,
      "num_input_tokens_seen": 15884091584,
      "step": 55850,
      "train_runtime": 543075.7874,
      "train_tokens_per_second": 29248.388
    },
    {
      "epoch": 1.9764377818123116,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9483296434277877e-05,
      "loss": 0.3876596450805664,
      "num_input_tokens_seen": 15886903616,
      "step": 55860,
      "train_runtime": 543171.9104,
      "train_tokens_per_second": 29248.39
    },
    {
      "epoch": 1.9767916020760405,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9481817435358675e-05,
      "loss": 0.38994905948638914,
      "num_input_tokens_seen": 15889770048,
      "step": 55870,
      "train_runtime": 543269.8322,
      "train_tokens_per_second": 29248.394
    },
    {
      "epoch": 1.977145422339769,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9480338773205753e-05,
      "loss": 0.3916220188140869,
      "num_input_tokens_seen": 15892611776,
      "step": 55880,
      "train_runtime": 543365.8853,
      "train_tokens_per_second": 29248.453
    },
    {
      "epoch": 1.977499242603498,
      "grad_norm": 0.609375,
      "learning_rate": 1.9478860447691327e-05,
      "loss": 0.3893189191818237,
      "num_input_tokens_seen": 15895480896,
      "step": 55890,
      "train_runtime": 543463.0586,
      "train_tokens_per_second": 29248.503
    },
    {
      "epoch": 1.9778530628672266,
      "grad_norm": 0.68359375,
      "learning_rate": 1.9477382458687682e-05,
      "loss": 0.3878438472747803,
      "num_input_tokens_seen": 15898312192,
      "step": 55900,
      "train_runtime": 543557.1126,
      "train_tokens_per_second": 29248.651
    },
    {
      "epoch": 1.9782068831309556,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9475904806067173e-05,
      "loss": 0.38989160060882566,
      "num_input_tokens_seen": 15901135168,
      "step": 55910,
      "train_runtime": 543652.7197,
      "train_tokens_per_second": 29248.7
    },
    {
      "epoch": 1.9785607033946842,
      "grad_norm": 0.625,
      "learning_rate": 1.947442748970222e-05,
      "loss": 0.38657007217407224,
      "num_input_tokens_seen": 15904008320,
      "step": 55920,
      "train_runtime": 543753.1734,
      "train_tokens_per_second": 29248.58
    },
    {
      "epoch": 1.9789145236584131,
      "grad_norm": 0.6171875,
      "learning_rate": 1.947295050946531e-05,
      "loss": 0.39104328155517576,
      "num_input_tokens_seen": 15906858176,
      "step": 55930,
      "train_runtime": 543851.6428,
      "train_tokens_per_second": 29248.525
    },
    {
      "epoch": 1.979268343922142,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9471473865229005e-05,
      "loss": 0.3895110607147217,
      "num_input_tokens_seen": 15909717376,
      "step": 55940,
      "train_runtime": 543948.9459,
      "train_tokens_per_second": 29248.549
    },
    {
      "epoch": 1.9796221641858707,
      "grad_norm": 0.625,
      "learning_rate": 1.946999755686592e-05,
      "loss": 0.3883556365966797,
      "num_input_tokens_seen": 15912560960,
      "step": 55950,
      "train_runtime": 544045.2682,
      "train_tokens_per_second": 29248.597
    },
    {
      "epoch": 1.9799759844495994,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9468521584248754e-05,
      "loss": 0.3900402069091797,
      "num_input_tokens_seen": 15915391488,
      "step": 55960,
      "train_runtime": 544140.7013,
      "train_tokens_per_second": 29248.669
    },
    {
      "epoch": 1.9803298047133282,
      "grad_norm": 0.57421875,
      "learning_rate": 1.9467045947250262e-05,
      "loss": 0.3871407985687256,
      "num_input_tokens_seen": 15918301120,
      "step": 55970,
      "train_runtime": 544244.2884,
      "train_tokens_per_second": 29248.449
    },
    {
      "epoch": 1.980683624977057,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9465570645743265e-05,
      "loss": 0.387880802154541,
      "num_input_tokens_seen": 15921138688,
      "step": 55980,
      "train_runtime": 544339.5857,
      "train_tokens_per_second": 29248.541
    },
    {
      "epoch": 1.9810374452407857,
      "grad_norm": 0.6328125,
      "learning_rate": 1.946409567960067e-05,
      "loss": 0.39166951179504395,
      "num_input_tokens_seen": 15923974144,
      "step": 55990,
      "train_runtime": 544435.1966,
      "train_tokens_per_second": 29248.613
    },
    {
      "epoch": 1.9813912655045145,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9462621048695433e-05,
      "loss": 0.3900148868560791,
      "num_input_tokens_seen": 15926836800,
      "step": 56000,
      "train_runtime": 544533.6879,
      "train_tokens_per_second": 29248.579
    },
    {
      "epoch": 1.9817450857682433,
      "grad_norm": 0.65234375,
      "learning_rate": 1.9461146752900578e-05,
      "loss": 0.3865171432495117,
      "num_input_tokens_seen": 15929623360,
      "step": 56010,
      "train_runtime": 544628.1686,
      "train_tokens_per_second": 29248.622
    },
    {
      "epoch": 1.982098906031972,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9459672792089206e-05,
      "loss": 0.38361918926239014,
      "num_input_tokens_seen": 15932404736,
      "step": 56020,
      "train_runtime": 544721.5726,
      "train_tokens_per_second": 29248.713
    },
    {
      "epoch": 1.9824527262957008,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9458199166134474e-05,
      "loss": 0.38777186870574953,
      "num_input_tokens_seen": 15935241664,
      "step": 56030,
      "train_runtime": 544819.1346,
      "train_tokens_per_second": 29248.682
    },
    {
      "epoch": 1.9828065465594298,
      "grad_norm": 0.58984375,
      "learning_rate": 1.945672587490962e-05,
      "loss": 0.3864882469177246,
      "num_input_tokens_seen": 15938158144,
      "step": 56040,
      "train_runtime": 544919.7016,
      "train_tokens_per_second": 29248.636
    },
    {
      "epoch": 1.9831603668231583,
      "grad_norm": 0.62890625,
      "learning_rate": 1.945525291828794e-05,
      "loss": 0.3837608814239502,
      "num_input_tokens_seen": 15941030912,
      "step": 56050,
      "train_runtime": 545018.1996,
      "train_tokens_per_second": 29248.621
    },
    {
      "epoch": 1.9835141870868873,
      "grad_norm": 0.6171875,
      "learning_rate": 1.945378029614279e-05,
      "loss": 0.392111611366272,
      "num_input_tokens_seen": 15943890688,
      "step": 56060,
      "train_runtime": 545116.1005,
      "train_tokens_per_second": 29248.615
    },
    {
      "epoch": 1.9838680073506159,
      "grad_norm": 0.640625,
      "learning_rate": 1.9452308008347617e-05,
      "loss": 0.387205982208252,
      "num_input_tokens_seen": 15946700800,
      "step": 56070,
      "train_runtime": 545209.0562,
      "train_tokens_per_second": 29248.782
    },
    {
      "epoch": 1.9842218276143448,
      "grad_norm": 0.62109375,
      "learning_rate": 1.945083605477591e-05,
      "loss": 0.39066190719604493,
      "num_input_tokens_seen": 15949568640,
      "step": 56080,
      "train_runtime": 545307.5681,
      "train_tokens_per_second": 29248.757
    },
    {
      "epoch": 1.9845756478780734,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9449364435301235e-05,
      "loss": 0.3899407863616943,
      "num_input_tokens_seen": 15952373440,
      "step": 56090,
      "train_runtime": 545402.9177,
      "train_tokens_per_second": 29248.786
    },
    {
      "epoch": 1.9849294681418024,
      "grad_norm": 0.58203125,
      "learning_rate": 1.944789314979723e-05,
      "loss": 0.38867807388305664,
      "num_input_tokens_seen": 15955251264,
      "step": 56100,
      "train_runtime": 545500.1708,
      "train_tokens_per_second": 29248.847
    },
    {
      "epoch": 1.9852832884055311,
      "grad_norm": 0.60546875,
      "learning_rate": 1.944642219813759e-05,
      "loss": 0.38737475872039795,
      "num_input_tokens_seen": 15958147520,
      "step": 56110,
      "train_runtime": 545599.7706,
      "train_tokens_per_second": 29248.816
    },
    {
      "epoch": 1.98563710866926,
      "grad_norm": 0.6171875,
      "learning_rate": 1.944495158019608e-05,
      "loss": 0.39335243701934813,
      "num_input_tokens_seen": 15960998592,
      "step": 56120,
      "train_runtime": 545698.4075,
      "train_tokens_per_second": 29248.754
    },
    {
      "epoch": 1.9859909289329887,
      "grad_norm": 0.609375,
      "learning_rate": 1.9443481295846537e-05,
      "loss": 0.3889739990234375,
      "num_input_tokens_seen": 15963850880,
      "step": 56130,
      "train_runtime": 545795.3756,
      "train_tokens_per_second": 29248.784
    },
    {
      "epoch": 1.9863447491967174,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9442011344962863e-05,
      "loss": 0.3894783020019531,
      "num_input_tokens_seen": 15966651520,
      "step": 56140,
      "train_runtime": 545888.3274,
      "train_tokens_per_second": 29248.934
    },
    {
      "epoch": 1.9866985694604462,
      "grad_norm": 0.578125,
      "learning_rate": 1.9440541727419024e-05,
      "loss": 0.38727807998657227,
      "num_input_tokens_seen": 15969476416,
      "step": 56150,
      "train_runtime": 545984.6154,
      "train_tokens_per_second": 29248.949
    },
    {
      "epoch": 1.987052389724175,
      "grad_norm": 0.625,
      "learning_rate": 1.9439072443089045e-05,
      "loss": 0.3868678569793701,
      "num_input_tokens_seen": 15972270208,
      "step": 56160,
      "train_runtime": 546080.1405,
      "train_tokens_per_second": 29248.949
    },
    {
      "epoch": 1.9874062099879037,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9437603491847038e-05,
      "loss": 0.38737854957580564,
      "num_input_tokens_seen": 15975082880,
      "step": 56170,
      "train_runtime": 546173.4227,
      "train_tokens_per_second": 29249.103
    },
    {
      "epoch": 1.9877600302516325,
      "grad_norm": 0.765625,
      "learning_rate": 1.943613487356717e-05,
      "loss": 0.3907593250274658,
      "num_input_tokens_seen": 15977881472,
      "step": 56180,
      "train_runtime": 546268.4217,
      "train_tokens_per_second": 29249.14
    },
    {
      "epoch": 1.9881138505153615,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9434666588123665e-05,
      "loss": 0.38759555816650393,
      "num_input_tokens_seen": 15980732160,
      "step": 56190,
      "train_runtime": 546367.0793,
      "train_tokens_per_second": 29249.076
    },
    {
      "epoch": 1.98846767077909,
      "grad_norm": 0.59765625,
      "learning_rate": 1.943319863539083e-05,
      "loss": 0.38571479320526125,
      "num_input_tokens_seen": 15983606720,
      "step": 56200,
      "train_runtime": 546466.9789,
      "train_tokens_per_second": 29248.989
    },
    {
      "epoch": 1.988821491042819,
      "grad_norm": 0.609375,
      "learning_rate": 1.9431731015243026e-05,
      "loss": 0.38887112140655516,
      "num_input_tokens_seen": 15986437888,
      "step": 56210,
      "train_runtime": 546564.0289,
      "train_tokens_per_second": 29248.975
    },
    {
      "epoch": 1.9891753113065476,
      "grad_norm": 0.578125,
      "learning_rate": 1.943026372755469e-05,
      "loss": 0.38435850143432615,
      "num_input_tokens_seen": 15989271872,
      "step": 56220,
      "train_runtime": 546657.799,
      "train_tokens_per_second": 29249.143
    },
    {
      "epoch": 1.9895291315702766,
      "grad_norm": 0.59765625,
      "learning_rate": 1.942879677220032e-05,
      "loss": 0.38748531341552733,
      "num_input_tokens_seen": 15992126080,
      "step": 56230,
      "train_runtime": 546755.5419,
      "train_tokens_per_second": 29249.134
    },
    {
      "epoch": 1.989882951834005,
      "grad_norm": 0.59375,
      "learning_rate": 1.942733014905448e-05,
      "loss": 0.38696928024291993,
      "num_input_tokens_seen": 15994999744,
      "step": 56240,
      "train_runtime": 546853.9709,
      "train_tokens_per_second": 29249.124
    },
    {
      "epoch": 1.990236772097734,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9425863857991804e-05,
      "loss": 0.3852285385131836,
      "num_input_tokens_seen": 15997838912,
      "step": 56250,
      "train_runtime": 546950.1797,
      "train_tokens_per_second": 29249.17
    },
    {
      "epoch": 1.9905905923614626,
      "grad_norm": 0.625,
      "learning_rate": 1.9424397898886988e-05,
      "loss": 0.39238824844360354,
      "num_input_tokens_seen": 16000681920,
      "step": 56260,
      "train_runtime": 547048.5108,
      "train_tokens_per_second": 29249.11
    },
    {
      "epoch": 1.9909444126251916,
      "grad_norm": 0.640625,
      "learning_rate": 1.9422932271614797e-05,
      "loss": 0.3879425525665283,
      "num_input_tokens_seen": 16003485632,
      "step": 56270,
      "train_runtime": 547144.8376,
      "train_tokens_per_second": 29249.085
    },
    {
      "epoch": 1.9912982328889204,
      "grad_norm": 0.625,
      "learning_rate": 1.942146697605006e-05,
      "loss": 0.3862175941467285,
      "num_input_tokens_seen": 16006271936,
      "step": 56280,
      "train_runtime": 547237.6583,
      "train_tokens_per_second": 29249.215
    },
    {
      "epoch": 1.9916520531526491,
      "grad_norm": 0.59375,
      "learning_rate": 1.9420002012067674e-05,
      "loss": 0.3871592044830322,
      "num_input_tokens_seen": 16009110784,
      "step": 56290,
      "train_runtime": 547333.4235,
      "train_tokens_per_second": 29249.284
    },
    {
      "epoch": 1.992005873416378,
      "grad_norm": 0.61328125,
      "learning_rate": 1.94185373795426e-05,
      "loss": 0.3912575960159302,
      "num_input_tokens_seen": 16011932416,
      "step": 56300,
      "train_runtime": 547428.8662,
      "train_tokens_per_second": 29249.339
    },
    {
      "epoch": 1.9923596936801067,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9417073078349874e-05,
      "loss": 0.38839623928070066,
      "num_input_tokens_seen": 16014766144,
      "step": 56310,
      "train_runtime": 547526.2904,
      "train_tokens_per_second": 29249.31
    },
    {
      "epoch": 1.9927135139438354,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9415609108364576e-05,
      "loss": 0.3878905773162842,
      "num_input_tokens_seen": 16017600256,
      "step": 56320,
      "train_runtime": 547622.7303,
      "train_tokens_per_second": 29249.334
    },
    {
      "epoch": 1.9930673342075642,
      "grad_norm": 0.61328125,
      "learning_rate": 1.941414546946188e-05,
      "loss": 0.38534860610961913,
      "num_input_tokens_seen": 16020442688,
      "step": 56330,
      "train_runtime": 547717.2263,
      "train_tokens_per_second": 29249.477
    },
    {
      "epoch": 1.993421154471293,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9412682161517005e-05,
      "loss": 0.385499906539917,
      "num_input_tokens_seen": 16023284992,
      "step": 56340,
      "train_runtime": 547816.9663,
      "train_tokens_per_second": 29249.341
    },
    {
      "epoch": 1.9937749747350217,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9411219184405246e-05,
      "loss": 0.3872855186462402,
      "num_input_tokens_seen": 16026120832,
      "step": 56350,
      "train_runtime": 547913.5725,
      "train_tokens_per_second": 29249.359
    },
    {
      "epoch": 1.9941287949987507,
      "grad_norm": 0.59375,
      "learning_rate": 1.9409756538001956e-05,
      "loss": 0.38614506721496583,
      "num_input_tokens_seen": 16028994240,
      "step": 56360,
      "train_runtime": 548011.2381,
      "train_tokens_per_second": 29249.39
    },
    {
      "epoch": 1.9944826152624793,
      "grad_norm": 0.65625,
      "learning_rate": 1.940829422218256e-05,
      "loss": 0.3927909374237061,
      "num_input_tokens_seen": 16031783040,
      "step": 56370,
      "train_runtime": 548104.2196,
      "train_tokens_per_second": 29249.516
    },
    {
      "epoch": 1.9948364355262083,
      "grad_norm": 0.609375,
      "learning_rate": 1.9406832236822554e-05,
      "loss": 0.38695297241210935,
      "num_input_tokens_seen": 16034585152,
      "step": 56380,
      "train_runtime": 548200.7224,
      "train_tokens_per_second": 29249.478
    },
    {
      "epoch": 1.9951902557899368,
      "grad_norm": 0.60546875,
      "learning_rate": 1.940537058179749e-05,
      "loss": 0.39509925842285154,
      "num_input_tokens_seen": 16037380736,
      "step": 56390,
      "train_runtime": 548296.1867,
      "train_tokens_per_second": 29249.484
    },
    {
      "epoch": 1.9955440760536658,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9403909256982976e-05,
      "loss": 0.3866716384887695,
      "num_input_tokens_seen": 16040219648,
      "step": 56400,
      "train_runtime": 548394.1267,
      "train_tokens_per_second": 29249.437
    },
    {
      "epoch": 1.9958978963173943,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9402448262254713e-05,
      "loss": 0.3871128082275391,
      "num_input_tokens_seen": 16043105216,
      "step": 56410,
      "train_runtime": 548492.3908,
      "train_tokens_per_second": 29249.458
    },
    {
      "epoch": 1.9962517165811233,
      "grad_norm": 0.5859375,
      "learning_rate": 1.940098759748845e-05,
      "loss": 0.3847597599029541,
      "num_input_tokens_seen": 16045917184,
      "step": 56420,
      "train_runtime": 548585.5884,
      "train_tokens_per_second": 29249.615
    },
    {
      "epoch": 1.9966055368448519,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9399527262559997e-05,
      "loss": 0.39197144508361814,
      "num_input_tokens_seen": 16048747712,
      "step": 56430,
      "train_runtime": 548683.806,
      "train_tokens_per_second": 29249.538
    },
    {
      "epoch": 1.9969593571085809,
      "grad_norm": 0.609375,
      "learning_rate": 1.939806725734524e-05,
      "loss": 0.39178879261016847,
      "num_input_tokens_seen": 16051591232,
      "step": 56440,
      "train_runtime": 548783.5458,
      "train_tokens_per_second": 29249.403
    },
    {
      "epoch": 1.9973131773723096,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9396607581720126e-05,
      "loss": 0.38765707015991213,
      "num_input_tokens_seen": 16054406464,
      "step": 56450,
      "train_runtime": 548881.1302,
      "train_tokens_per_second": 29249.332
    },
    {
      "epoch": 1.9976669976360384,
      "grad_norm": 0.609375,
      "learning_rate": 1.9395148235560674e-05,
      "loss": 0.39126756191253664,
      "num_input_tokens_seen": 16057303168,
      "step": 56460,
      "train_runtime": 548982.8371,
      "train_tokens_per_second": 29249.19
    },
    {
      "epoch": 1.9980208178997672,
      "grad_norm": 0.609375,
      "learning_rate": 1.9393689218742955e-05,
      "loss": 0.3877740383148193,
      "num_input_tokens_seen": 16060162496,
      "step": 56470,
      "train_runtime": 549079.271,
      "train_tokens_per_second": 29249.26
    },
    {
      "epoch": 1.998374638163496,
      "grad_norm": 0.62109375,
      "learning_rate": 1.939223053114311e-05,
      "loss": 0.3860870361328125,
      "num_input_tokens_seen": 16062997696,
      "step": 56480,
      "train_runtime": 549176.1254,
      "train_tokens_per_second": 29249.264
    },
    {
      "epoch": 1.9987284584272247,
      "grad_norm": 0.609375,
      "learning_rate": 1.939077217263736e-05,
      "loss": 0.3871375560760498,
      "num_input_tokens_seen": 16065828288,
      "step": 56490,
      "train_runtime": 549273.7421,
      "train_tokens_per_second": 29249.22
    },
    {
      "epoch": 1.9990822786909535,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9389314143101967e-05,
      "loss": 0.39275059700012205,
      "num_input_tokens_seen": 16068675392,
      "step": 56500,
      "train_runtime": 549370.6268,
      "train_tokens_per_second": 29249.244
    },
    {
      "epoch": 1.9994360989546822,
      "grad_norm": 0.61328125,
      "learning_rate": 1.938785644241327e-05,
      "loss": 0.38933744430541994,
      "num_input_tokens_seen": 16071454080,
      "step": 56510,
      "train_runtime": 549462.9893,
      "train_tokens_per_second": 29249.384
    },
    {
      "epoch": 1.999789919218411,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9386399070447685e-05,
      "loss": 0.3833935260772705,
      "num_input_tokens_seen": 16074315392,
      "step": 56520,
      "train_runtime": 549561.6289,
      "train_tokens_per_second": 29249.341
    },
    {
      "epoch": 2.0001415281054915,
      "grad_norm": 0.5859375,
      "learning_rate": 1.938494202708167e-05,
      "loss": 0.38704848289489746,
      "num_input_tokens_seen": 16077140928,
      "step": 56530,
      "train_runtime": 549667.984,
      "train_tokens_per_second": 29248.822
    },
    {
      "epoch": 2.0004953483692205,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9383485312191764e-05,
      "loss": 0.38617887496948244,
      "num_input_tokens_seen": 16079929728,
      "step": 56540,
      "train_runtime": 549762.1155,
      "train_tokens_per_second": 29248.887
    },
    {
      "epoch": 2.000849168632949,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9382028925654567e-05,
      "loss": 0.3824448585510254,
      "num_input_tokens_seen": 16082804608,
      "step": 56550,
      "train_runtime": 549859.9417,
      "train_tokens_per_second": 29248.911
    },
    {
      "epoch": 2.001202988896678,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9380572867346734e-05,
      "loss": 0.387738037109375,
      "num_input_tokens_seen": 16085644544,
      "step": 56560,
      "train_runtime": 549956.9715,
      "train_tokens_per_second": 29248.915
    },
    {
      "epoch": 2.0015568091604066,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9379117137145007e-05,
      "loss": 0.38633456230163576,
      "num_input_tokens_seen": 16088528128,
      "step": 56570,
      "train_runtime": 550054.5191,
      "train_tokens_per_second": 29248.97
    },
    {
      "epoch": 2.0019106294241356,
      "grad_norm": 0.609375,
      "learning_rate": 1.937766173492617e-05,
      "loss": 0.3873283863067627,
      "num_input_tokens_seen": 16091350208,
      "step": 56580,
      "train_runtime": 550151.4144,
      "train_tokens_per_second": 29248.948
    },
    {
      "epoch": 2.002264449687864,
      "grad_norm": 0.59375,
      "learning_rate": 1.9376206660567083e-05,
      "loss": 0.388609790802002,
      "num_input_tokens_seen": 16094200960,
      "step": 56590,
      "train_runtime": 550250.2922,
      "train_tokens_per_second": 29248.873
    },
    {
      "epoch": 2.002618269951593,
      "grad_norm": 0.5859375,
      "learning_rate": 1.9374751913944676e-05,
      "loss": 0.3812508821487427,
      "num_input_tokens_seen": 16097080832,
      "step": 56600,
      "train_runtime": 550352.5456,
      "train_tokens_per_second": 29248.672
    },
    {
      "epoch": 2.0029720902153216,
      "grad_norm": 0.625,
      "learning_rate": 1.9373297494935928e-05,
      "loss": 0.38963875770568845,
      "num_input_tokens_seen": 16099895808,
      "step": 56610,
      "train_runtime": 550447.658,
      "train_tokens_per_second": 29248.732
    },
    {
      "epoch": 2.0033259104790506,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9371843403417894e-05,
      "loss": 0.38725159168243406,
      "num_input_tokens_seen": 16102718208,
      "step": 56620,
      "train_runtime": 550542.7941,
      "train_tokens_per_second": 29248.804
    },
    {
      "epoch": 2.003679730742779,
      "grad_norm": 0.59765625,
      "learning_rate": 1.937038963926769e-05,
      "loss": 0.3814414978027344,
      "num_input_tokens_seen": 16105603008,
      "step": 56630,
      "train_runtime": 550642.4855,
      "train_tokens_per_second": 29248.747
    },
    {
      "epoch": 2.004033551006508,
      "grad_norm": 0.63671875,
      "learning_rate": 1.93689362023625e-05,
      "loss": 0.38872013092041013,
      "num_input_tokens_seen": 16108478784,
      "step": 56640,
      "train_runtime": 550743.1313,
      "train_tokens_per_second": 29248.624
    },
    {
      "epoch": 2.0043873712702367,
      "grad_norm": 0.6171875,
      "learning_rate": 1.936748309257957e-05,
      "loss": 0.38713483810424804,
      "num_input_tokens_seen": 16111278848,
      "step": 56650,
      "train_runtime": 550835.4975,
      "train_tokens_per_second": 29248.803
    },
    {
      "epoch": 2.0047411915339657,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9366030309796208e-05,
      "loss": 0.38740978240966795,
      "num_input_tokens_seen": 16114109696,
      "step": 56660,
      "train_runtime": 550928.7259,
      "train_tokens_per_second": 29248.992
    },
    {
      "epoch": 2.0050950117976942,
      "grad_norm": 0.59765625,
      "learning_rate": 1.936457785388979e-05,
      "loss": 0.385851526260376,
      "num_input_tokens_seen": 16116990336,
      "step": 56670,
      "train_runtime": 551028.3389,
      "train_tokens_per_second": 29248.932
    },
    {
      "epoch": 2.005448832061423,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9363125724737757e-05,
      "loss": 0.38767757415771487,
      "num_input_tokens_seen": 16119840448,
      "step": 56680,
      "train_runtime": 551124.5255,
      "train_tokens_per_second": 29248.999
    },
    {
      "epoch": 2.0058026523251518,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9361673922217612e-05,
      "loss": 0.3847143888473511,
      "num_input_tokens_seen": 16122732800,
      "step": 56690,
      "train_runtime": 551222.4148,
      "train_tokens_per_second": 29249.052
    },
    {
      "epoch": 2.0061564725888807,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9360222446206923e-05,
      "loss": 0.38465657234191897,
      "num_input_tokens_seen": 16125596480,
      "step": 56700,
      "train_runtime": 551320.5566,
      "train_tokens_per_second": 29249.039
    },
    {
      "epoch": 2.0065102928526097,
      "grad_norm": 0.65234375,
      "learning_rate": 1.9358771296583317e-05,
      "loss": 0.38240666389465333,
      "num_input_tokens_seen": 16128472704,
      "step": 56710,
      "train_runtime": 551419.452,
      "train_tokens_per_second": 29249.009
    },
    {
      "epoch": 2.0068641131163383,
      "grad_norm": 0.62109375,
      "learning_rate": 1.935732047322449e-05,
      "loss": 0.3871786117553711,
      "num_input_tokens_seen": 16131364864,
      "step": 56720,
      "train_runtime": 551518.8231,
      "train_tokens_per_second": 29248.983
    },
    {
      "epoch": 2.0072179333800673,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9355869976008212e-05,
      "loss": 0.3847933292388916,
      "num_input_tokens_seen": 16134208448,
      "step": 56730,
      "train_runtime": 551617.4706,
      "train_tokens_per_second": 29248.908
    },
    {
      "epoch": 2.007571753643796,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9354419804812303e-05,
      "loss": 0.3835297107696533,
      "num_input_tokens_seen": 16137034688,
      "step": 56740,
      "train_runtime": 551714.0554,
      "train_tokens_per_second": 29248.91
    },
    {
      "epoch": 2.007925573907525,
      "grad_norm": 0.609375,
      "learning_rate": 1.935296995951465e-05,
      "loss": 0.38516807556152344,
      "num_input_tokens_seen": 16139858944,
      "step": 56750,
      "train_runtime": 551808.7734,
      "train_tokens_per_second": 29249.007
    },
    {
      "epoch": 2.0082793941712533,
      "grad_norm": 0.625,
      "learning_rate": 1.9351520439993207e-05,
      "loss": 0.3873849630355835,
      "num_input_tokens_seen": 16142652160,
      "step": 56760,
      "train_runtime": 551904.9945,
      "train_tokens_per_second": 29248.969
    },
    {
      "epoch": 2.0086332144349823,
      "grad_norm": 0.58984375,
      "learning_rate": 1.935007124612599e-05,
      "loss": 0.38938198089599607,
      "num_input_tokens_seen": 16145533824,
      "step": 56770,
      "train_runtime": 552003.5175,
      "train_tokens_per_second": 29248.969
    },
    {
      "epoch": 2.008987034698711,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9348622377791077e-05,
      "loss": 0.38253188133239746,
      "num_input_tokens_seen": 16148411392,
      "step": 56780,
      "train_runtime": 552100.5256,
      "train_tokens_per_second": 29249.042
    },
    {
      "epoch": 2.00934085496244,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9347173834866616e-05,
      "loss": 0.3878162860870361,
      "num_input_tokens_seen": 16151261248,
      "step": 56790,
      "train_runtime": 552199.3174,
      "train_tokens_per_second": 29248.97
    },
    {
      "epoch": 2.0096946752261684,
      "grad_norm": 0.6484375,
      "learning_rate": 1.9345725617230813e-05,
      "loss": 0.390401291847229,
      "num_input_tokens_seen": 16154109568,
      "step": 56800,
      "train_runtime": 552297.3459,
      "train_tokens_per_second": 29248.936
    },
    {
      "epoch": 2.0100484954898974,
      "grad_norm": 0.63671875,
      "learning_rate": 1.9344277724761943e-05,
      "loss": 0.38604035377502444,
      "num_input_tokens_seen": 16156944128,
      "step": 56810,
      "train_runtime": 552396.9767,
      "train_tokens_per_second": 29248.792
    },
    {
      "epoch": 2.010402315753626,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9342830157338338e-05,
      "loss": 0.3858720064163208,
      "num_input_tokens_seen": 16159769408,
      "step": 56820,
      "train_runtime": 552492.0817,
      "train_tokens_per_second": 29248.871
    },
    {
      "epoch": 2.010756136017355,
      "grad_norm": 0.6171875,
      "learning_rate": 1.93413829148384e-05,
      "loss": 0.38519110679626467,
      "num_input_tokens_seen": 16162587264,
      "step": 56830,
      "train_runtime": 552588.2256,
      "train_tokens_per_second": 29248.881
    },
    {
      "epoch": 2.0111099562810835,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9339935997140592e-05,
      "loss": 0.3876930236816406,
      "num_input_tokens_seen": 16165395328,
      "step": 56840,
      "train_runtime": 552682.9627,
      "train_tokens_per_second": 29248.948
    },
    {
      "epoch": 2.0114637765448125,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9338489404123445e-05,
      "loss": 0.384290885925293,
      "num_input_tokens_seen": 16168253568,
      "step": 56850,
      "train_runtime": 552783.4325,
      "train_tokens_per_second": 29248.803
    },
    {
      "epoch": 2.0118175968085414,
      "grad_norm": 0.640625,
      "learning_rate": 1.933704313566554e-05,
      "loss": 0.3906275510787964,
      "num_input_tokens_seen": 16171069824,
      "step": 56860,
      "train_runtime": 552878.7366,
      "train_tokens_per_second": 29248.855
    },
    {
      "epoch": 2.01217141707227,
      "grad_norm": 0.609375,
      "learning_rate": 1.9335597191645538e-05,
      "loss": 0.3834575414657593,
      "num_input_tokens_seen": 16173874944,
      "step": 56870,
      "train_runtime": 552975.2534,
      "train_tokens_per_second": 29248.822
    },
    {
      "epoch": 2.012525237335999,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9334151571942157e-05,
      "loss": 0.38457741737365725,
      "num_input_tokens_seen": 16176671680,
      "step": 56880,
      "train_runtime": 553069.6287,
      "train_tokens_per_second": 29248.888
    },
    {
      "epoch": 2.0128790575997275,
      "grad_norm": 0.6015625,
      "learning_rate": 1.933270627643417e-05,
      "loss": 0.3847815990447998,
      "num_input_tokens_seen": 16179527232,
      "step": 56890,
      "train_runtime": 553168.0118,
      "train_tokens_per_second": 29248.848
    },
    {
      "epoch": 2.0132328778634565,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9331261305000432e-05,
      "loss": 0.3897664546966553,
      "num_input_tokens_seen": 16182371008,
      "step": 56900,
      "train_runtime": 553264.691,
      "train_tokens_per_second": 29248.877
    },
    {
      "epoch": 2.013586698127185,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9329816657519845e-05,
      "loss": 0.3876787185668945,
      "num_input_tokens_seen": 16185209088,
      "step": 56910,
      "train_runtime": 553359.4992,
      "train_tokens_per_second": 29248.995
    },
    {
      "epoch": 2.013940518390914,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9328372333871378e-05,
      "loss": 0.38917956352233884,
      "num_input_tokens_seen": 16187992576,
      "step": 56920,
      "train_runtime": 553454.5417,
      "train_tokens_per_second": 29249.001
    },
    {
      "epoch": 2.0142943386546426,
      "grad_norm": 0.64453125,
      "learning_rate": 1.9326928333934072e-05,
      "loss": 0.3802932262420654,
      "num_input_tokens_seen": 16190857664,
      "step": 56930,
      "train_runtime": 553553.8492,
      "train_tokens_per_second": 29248.93
    },
    {
      "epoch": 2.0146481589183716,
      "grad_norm": 0.5859375,
      "learning_rate": 1.9325484657587018e-05,
      "loss": 0.38528952598571775,
      "num_input_tokens_seen": 16193695104,
      "step": 56940,
      "train_runtime": 553651.5095,
      "train_tokens_per_second": 29248.895
    },
    {
      "epoch": 2.0150019791821,
      "grad_norm": 0.59375,
      "learning_rate": 1.932404130470938e-05,
      "loss": 0.3865856409072876,
      "num_input_tokens_seen": 16196511360,
      "step": 56950,
      "train_runtime": 553746.0466,
      "train_tokens_per_second": 29248.988
    },
    {
      "epoch": 2.015355799445829,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9322598275180383e-05,
      "loss": 0.38748080730438234,
      "num_input_tokens_seen": 16199365248,
      "step": 56960,
      "train_runtime": 553844.8517,
      "train_tokens_per_second": 29248.923
    },
    {
      "epoch": 2.0157096197095576,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9321155568879315e-05,
      "loss": 0.38855226039886476,
      "num_input_tokens_seen": 16202188288,
      "step": 56970,
      "train_runtime": 553941.4092,
      "train_tokens_per_second": 29248.921
    },
    {
      "epoch": 2.0160634399732866,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9319713185685522e-05,
      "loss": 0.3852694511413574,
      "num_input_tokens_seen": 16205015936,
      "step": 56980,
      "train_runtime": 554039.6616,
      "train_tokens_per_second": 29248.837
    },
    {
      "epoch": 2.016417260237015,
      "grad_norm": 0.5859375,
      "learning_rate": 1.9318271125478422e-05,
      "loss": 0.391272759437561,
      "num_input_tokens_seen": 16207861376,
      "step": 56990,
      "train_runtime": 554134.221,
      "train_tokens_per_second": 29248.981
    },
    {
      "epoch": 2.016771080500744,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9316829388137487e-05,
      "loss": 0.38520827293396,
      "num_input_tokens_seen": 16210714240,
      "step": 57000,
      "train_runtime": 554232.7394,
      "train_tokens_per_second": 29248.929
    },
    {
      "epoch": 2.0171249007644727,
      "grad_norm": 0.6328125,
      "learning_rate": 1.931538797354226e-05,
      "loss": 0.3826314926147461,
      "num_input_tokens_seen": 16213538688,
      "step": 57010,
      "train_runtime": 554329.5303,
      "train_tokens_per_second": 29248.917
    },
    {
      "epoch": 2.0174787210282017,
      "grad_norm": 0.5859375,
      "learning_rate": 1.931394688157234e-05,
      "loss": 0.386994743347168,
      "num_input_tokens_seen": 16216409728,
      "step": 57020,
      "train_runtime": 554429.2333,
      "train_tokens_per_second": 29248.836
    },
    {
      "epoch": 2.0178325412919307,
      "grad_norm": 0.625,
      "learning_rate": 1.9312506112107394e-05,
      "loss": 0.3874991416931152,
      "num_input_tokens_seen": 16219295360,
      "step": 57030,
      "train_runtime": 554528.9717,
      "train_tokens_per_second": 29248.779
    },
    {
      "epoch": 2.0181863615556592,
      "grad_norm": 0.6640625,
      "learning_rate": 1.931106566502715e-05,
      "loss": 0.385809326171875,
      "num_input_tokens_seen": 16222161216,
      "step": 57040,
      "train_runtime": 554632.041,
      "train_tokens_per_second": 29248.511
    },
    {
      "epoch": 2.018540181819388,
      "grad_norm": 0.609375,
      "learning_rate": 1.9309625540211397e-05,
      "loss": 0.38689398765563965,
      "num_input_tokens_seen": 16224984512,
      "step": 57050,
      "train_runtime": 554729.5222,
      "train_tokens_per_second": 29248.46
    },
    {
      "epoch": 2.0188940020831168,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9308185737539988e-05,
      "loss": 0.3856795787811279,
      "num_input_tokens_seen": 16227826048,
      "step": 57060,
      "train_runtime": 554826.0232,
      "train_tokens_per_second": 29248.495
    },
    {
      "epoch": 2.0192478223468457,
      "grad_norm": 0.61328125,
      "learning_rate": 1.930674625689284e-05,
      "loss": 0.3907461404800415,
      "num_input_tokens_seen": 16230663616,
      "step": 57070,
      "train_runtime": 554922.5407,
      "train_tokens_per_second": 29248.521
    },
    {
      "epoch": 2.0196016426105743,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9305307098149935e-05,
      "loss": 0.38799228668212893,
      "num_input_tokens_seen": 16233473984,
      "step": 57080,
      "train_runtime": 555017.6253,
      "train_tokens_per_second": 29248.574
    },
    {
      "epoch": 2.0199554628743033,
      "grad_norm": 0.6171875,
      "learning_rate": 1.930386826119131e-05,
      "loss": 0.3867323875427246,
      "num_input_tokens_seen": 16236298688,
      "step": 57090,
      "train_runtime": 555112.0126,
      "train_tokens_per_second": 29248.689
    },
    {
      "epoch": 2.020309283138032,
      "grad_norm": 0.59765625,
      "learning_rate": 1.930242974589707e-05,
      "loss": 0.3862288475036621,
      "num_input_tokens_seen": 16239094976,
      "step": 57100,
      "train_runtime": 555206.5344,
      "train_tokens_per_second": 29248.746
    },
    {
      "epoch": 2.020663103401761,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9300991552147384e-05,
      "loss": 0.3861955165863037,
      "num_input_tokens_seen": 16241963392,
      "step": 57110,
      "train_runtime": 555306.6935,
      "train_tokens_per_second": 29248.636
    },
    {
      "epoch": 2.0210169236654894,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9299553679822472e-05,
      "loss": 0.38731913566589354,
      "num_input_tokens_seen": 16244818176,
      "step": 57120,
      "train_runtime": 555403.9317,
      "train_tokens_per_second": 29248.655
    },
    {
      "epoch": 2.0213707439292183,
      "grad_norm": 0.609375,
      "learning_rate": 1.929811612880264e-05,
      "loss": 0.3851296901702881,
      "num_input_tokens_seen": 16247708736,
      "step": 57130,
      "train_runtime": 555503.9843,
      "train_tokens_per_second": 29248.591
    },
    {
      "epoch": 2.021724564192947,
      "grad_norm": 0.60546875,
      "learning_rate": 1.929667889896823e-05,
      "loss": 0.3831844091415405,
      "num_input_tokens_seen": 16250485120,
      "step": 57140,
      "train_runtime": 555597.1178,
      "train_tokens_per_second": 29248.685
    },
    {
      "epoch": 2.022078384456676,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9295241990199666e-05,
      "loss": 0.3858208656311035,
      "num_input_tokens_seen": 16253317120,
      "step": 57150,
      "train_runtime": 555692.686,
      "train_tokens_per_second": 29248.751
    },
    {
      "epoch": 2.0224322047204044,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9293805402377416e-05,
      "loss": 0.3857236385345459,
      "num_input_tokens_seen": 16256130560,
      "step": 57160,
      "train_runtime": 555785.9703,
      "train_tokens_per_second": 29248.904
    },
    {
      "epoch": 2.0227860249841334,
      "grad_norm": 0.640625,
      "learning_rate": 1.9292369135382035e-05,
      "loss": 0.3830159902572632,
      "num_input_tokens_seen": 16258945856,
      "step": 57170,
      "train_runtime": 555882.5793,
      "train_tokens_per_second": 29248.885
    },
    {
      "epoch": 2.023139845247862,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9290933189094114e-05,
      "loss": 0.38844497203826905,
      "num_input_tokens_seen": 16261732672,
      "step": 57180,
      "train_runtime": 555976.2799,
      "train_tokens_per_second": 29248.968
    },
    {
      "epoch": 2.023493665511591,
      "grad_norm": 0.609375,
      "learning_rate": 1.928949756339432e-05,
      "loss": 0.38911459445953367,
      "num_input_tokens_seen": 16264601152,
      "step": 57190,
      "train_runtime": 556076.4171,
      "train_tokens_per_second": 29248.86
    },
    {
      "epoch": 2.02384748577532,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9288062258163386e-05,
      "loss": 0.3868520975112915,
      "num_input_tokens_seen": 16267415296,
      "step": 57200,
      "train_runtime": 556170.3384,
      "train_tokens_per_second": 29248.98
    },
    {
      "epoch": 2.0242013060390485,
      "grad_norm": 0.61328125,
      "learning_rate": 1.92866272732821e-05,
      "loss": 0.383447527885437,
      "num_input_tokens_seen": 16270270080,
      "step": 57210,
      "train_runtime": 556270.2683,
      "train_tokens_per_second": 29248.858
    },
    {
      "epoch": 2.0245551263027775,
      "grad_norm": 0.6015625,
      "learning_rate": 1.928519260863131e-05,
      "loss": 0.38642072677612305,
      "num_input_tokens_seen": 16273134592,
      "step": 57220,
      "train_runtime": 556368.4004,
      "train_tokens_per_second": 29248.848
    },
    {
      "epoch": 2.024908946566506,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9283758264091932e-05,
      "loss": 0.38658928871154785,
      "num_input_tokens_seen": 16275985856,
      "step": 57230,
      "train_runtime": 556464.4672,
      "train_tokens_per_second": 29248.922
    },
    {
      "epoch": 2.025262766830235,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9282324239544938e-05,
      "loss": 0.38594517707824705,
      "num_input_tokens_seen": 16278872896,
      "step": 57240,
      "train_runtime": 556566.1764,
      "train_tokens_per_second": 29248.764
    },
    {
      "epoch": 2.0256165870939635,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9280890534871375e-05,
      "loss": 0.38664746284484863,
      "num_input_tokens_seen": 16281744064,
      "step": 57250,
      "train_runtime": 556664.9385,
      "train_tokens_per_second": 29248.733
    },
    {
      "epoch": 2.0259704073576925,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9279457149952333e-05,
      "loss": 0.3837035894393921,
      "num_input_tokens_seen": 16284558848,
      "step": 57260,
      "train_runtime": 556760.7209,
      "train_tokens_per_second": 29248.757
    },
    {
      "epoch": 2.026324227621421,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9278024084668976e-05,
      "loss": 0.3891209363937378,
      "num_input_tokens_seen": 16287411648,
      "step": 57270,
      "train_runtime": 556858.9716,
      "train_tokens_per_second": 29248.719
    },
    {
      "epoch": 2.02667804788515,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9276591338902524e-05,
      "loss": 0.38507354259490967,
      "num_input_tokens_seen": 16290273344,
      "step": 57280,
      "train_runtime": 556957.0206,
      "train_tokens_per_second": 29248.708
    },
    {
      "epoch": 2.0270318681488786,
      "grad_norm": 0.63671875,
      "learning_rate": 1.9275158912534273e-05,
      "loss": 0.38678689002990724,
      "num_input_tokens_seen": 16293120320,
      "step": 57290,
      "train_runtime": 557054.4727,
      "train_tokens_per_second": 29248.702
    },
    {
      "epoch": 2.0273856884126076,
      "grad_norm": 0.625,
      "learning_rate": 1.9273726805445557e-05,
      "loss": 0.3900745868682861,
      "num_input_tokens_seen": 16295982912,
      "step": 57300,
      "train_runtime": 557151.7909,
      "train_tokens_per_second": 29248.731
    },
    {
      "epoch": 2.027739508676336,
      "grad_norm": 0.59375,
      "learning_rate": 1.927229501751779e-05,
      "loss": 0.3791146993637085,
      "num_input_tokens_seen": 16298892096,
      "step": 57310,
      "train_runtime": 557254.2572,
      "train_tokens_per_second": 29248.573
    },
    {
      "epoch": 2.028093328940065,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9270863548632443e-05,
      "loss": 0.38923916816711424,
      "num_input_tokens_seen": 16301703744,
      "step": 57320,
      "train_runtime": 557346.5851,
      "train_tokens_per_second": 29248.773
    },
    {
      "epoch": 2.0284471492037937,
      "grad_norm": 0.64453125,
      "learning_rate": 1.9269432398671045e-05,
      "loss": 0.38825297355651855,
      "num_input_tokens_seen": 16304540672,
      "step": 57330,
      "train_runtime": 557445.2754,
      "train_tokens_per_second": 29248.684
    },
    {
      "epoch": 2.0288009694675226,
      "grad_norm": 0.60546875,
      "learning_rate": 1.926800156751519e-05,
      "loss": 0.38906786441802976,
      "num_input_tokens_seen": 16307324480,
      "step": 57340,
      "train_runtime": 557539.0879,
      "train_tokens_per_second": 29248.756
    },
    {
      "epoch": 2.029154789731251,
      "grad_norm": 0.59375,
      "learning_rate": 1.9266571055046537e-05,
      "loss": 0.3896360158920288,
      "num_input_tokens_seen": 16310174400,
      "step": 57350,
      "train_runtime": 557636.9285,
      "train_tokens_per_second": 29248.734
    },
    {
      "epoch": 2.02950860999498,
      "grad_norm": 0.63671875,
      "learning_rate": 1.9265140861146795e-05,
      "loss": 0.3821599006652832,
      "num_input_tokens_seen": 16313056256,
      "step": 57360,
      "train_runtime": 557737.6963,
      "train_tokens_per_second": 29248.617
    },
    {
      "epoch": 2.029862430258709,
      "grad_norm": 0.625,
      "learning_rate": 1.9263710985697744e-05,
      "loss": 0.38727900981903074,
      "num_input_tokens_seen": 16315897216,
      "step": 57370,
      "train_runtime": 557833.805,
      "train_tokens_per_second": 29248.671
    },
    {
      "epoch": 2.0302162505224377,
      "grad_norm": 0.625,
      "learning_rate": 1.9262281428581227e-05,
      "loss": 0.3850520133972168,
      "num_input_tokens_seen": 16318746432,
      "step": 57380,
      "train_runtime": 557930.4874,
      "train_tokens_per_second": 29248.709
    },
    {
      "epoch": 2.0305700707861667,
      "grad_norm": 0.61328125,
      "learning_rate": 1.926085218967914e-05,
      "loss": 0.3860297203063965,
      "num_input_tokens_seen": 16321615360,
      "step": 57390,
      "train_runtime": 558027.1512,
      "train_tokens_per_second": 29248.784
    },
    {
      "epoch": 2.0309238910498952,
      "grad_norm": 0.6328125,
      "learning_rate": 1.925942326887345e-05,
      "loss": 0.3840751886367798,
      "num_input_tokens_seen": 16324460160,
      "step": 57400,
      "train_runtime": 558127.4625,
      "train_tokens_per_second": 29248.624
    },
    {
      "epoch": 2.0312777113136242,
      "grad_norm": 0.61328125,
      "learning_rate": 1.925799466604617e-05,
      "loss": 0.3841102123260498,
      "num_input_tokens_seen": 16327308032,
      "step": 57410,
      "train_runtime": 558225.264,
      "train_tokens_per_second": 29248.601
    },
    {
      "epoch": 2.0316315315773528,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9256566381079393e-05,
      "loss": 0.38485302925109866,
      "num_input_tokens_seen": 16330159360,
      "step": 57420,
      "train_runtime": 558323.6783,
      "train_tokens_per_second": 29248.552
    },
    {
      "epoch": 2.0319853518410818,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9255138413855265e-05,
      "loss": 0.38656368255615237,
      "num_input_tokens_seen": 16333020608,
      "step": 57430,
      "train_runtime": 558423.0674,
      "train_tokens_per_second": 29248.47
    },
    {
      "epoch": 2.0323391721048103,
      "grad_norm": 0.609375,
      "learning_rate": 1.9253710764255987e-05,
      "loss": 0.38820481300354004,
      "num_input_tokens_seen": 16335876224,
      "step": 57440,
      "train_runtime": 558518.9273,
      "train_tokens_per_second": 29248.563
    },
    {
      "epoch": 2.0326929923685393,
      "grad_norm": 0.60546875,
      "learning_rate": 1.925228343216383e-05,
      "loss": 0.3871767520904541,
      "num_input_tokens_seen": 16338737920,
      "step": 57450,
      "train_runtime": 558615.622,
      "train_tokens_per_second": 29248.623
    },
    {
      "epoch": 2.033046812632268,
      "grad_norm": 0.609375,
      "learning_rate": 1.925085641746113e-05,
      "loss": 0.3866570472717285,
      "num_input_tokens_seen": 16341589632,
      "step": 57460,
      "train_runtime": 558709.6202,
      "train_tokens_per_second": 29248.807
    },
    {
      "epoch": 2.033400632895997,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9249429720030262e-05,
      "loss": 0.3798818588256836,
      "num_input_tokens_seen": 16344415744,
      "step": 57470,
      "train_runtime": 558803.5216,
      "train_tokens_per_second": 29248.949
    },
    {
      "epoch": 2.0337544531597254,
      "grad_norm": 0.5859375,
      "learning_rate": 1.9248003339753688e-05,
      "loss": 0.3787179946899414,
      "num_input_tokens_seen": 16347214720,
      "step": 57480,
      "train_runtime": 558898.0564,
      "train_tokens_per_second": 29249.01
    },
    {
      "epoch": 2.0341082734234543,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9246577276513916e-05,
      "loss": 0.38461205959320066,
      "num_input_tokens_seen": 16350015872,
      "step": 57490,
      "train_runtime": 558996.377,
      "train_tokens_per_second": 29248.876
    },
    {
      "epoch": 2.034462093687183,
      "grad_norm": 0.6484375,
      "learning_rate": 1.9245151530193517e-05,
      "loss": 0.3852632999420166,
      "num_input_tokens_seen": 16352839488,
      "step": 57500,
      "train_runtime": 559093.7959,
      "train_tokens_per_second": 29248.83
    },
    {
      "epoch": 2.034815913950912,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9243726100675132e-05,
      "loss": 0.3867560863494873,
      "num_input_tokens_seen": 16355633728,
      "step": 57510,
      "train_runtime": 559188.6361,
      "train_tokens_per_second": 29248.866
    },
    {
      "epoch": 2.0351697342146404,
      "grad_norm": 0.6171875,
      "learning_rate": 1.924230098784145e-05,
      "loss": 0.3869407892227173,
      "num_input_tokens_seen": 16358462400,
      "step": 57520,
      "train_runtime": 559284.7853,
      "train_tokens_per_second": 29248.896
    },
    {
      "epoch": 2.0355235544783694,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9240876191575226e-05,
      "loss": 0.3871732234954834,
      "num_input_tokens_seen": 16361296448,
      "step": 57530,
      "train_runtime": 559379.9067,
      "train_tokens_per_second": 29248.988
    },
    {
      "epoch": 2.0358773747420984,
      "grad_norm": 0.6171875,
      "learning_rate": 1.923945171175928e-05,
      "loss": 0.38892321586608886,
      "num_input_tokens_seen": 16364104640,
      "step": 57540,
      "train_runtime": 559476.7551,
      "train_tokens_per_second": 29248.945
    },
    {
      "epoch": 2.036231195005827,
      "grad_norm": 0.64453125,
      "learning_rate": 1.9238027548276487e-05,
      "loss": 0.38225431442260743,
      "num_input_tokens_seen": 16366969344,
      "step": 57550,
      "train_runtime": 559573.6006,
      "train_tokens_per_second": 29249.002
    },
    {
      "epoch": 2.036585015269556,
      "grad_norm": 0.6171875,
      "learning_rate": 1.923660370100978e-05,
      "loss": 0.38768270015716555,
      "num_input_tokens_seen": 16369761984,
      "step": 57560,
      "train_runtime": 559667.6261,
      "train_tokens_per_second": 29249.078
    },
    {
      "epoch": 2.0369388355332845,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9235180169842168e-05,
      "loss": 0.3881745576858521,
      "num_input_tokens_seen": 16372637952,
      "step": 57570,
      "train_runtime": 559766.5694,
      "train_tokens_per_second": 29249.046
    },
    {
      "epoch": 2.0372926557970135,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9233756954656694e-05,
      "loss": 0.3880466938018799,
      "num_input_tokens_seen": 16375454912,
      "step": 57580,
      "train_runtime": 559862.882,
      "train_tokens_per_second": 29249.046
    },
    {
      "epoch": 2.037646476060742,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9232334055336495e-05,
      "loss": 0.3897970914840698,
      "num_input_tokens_seen": 16378272576,
      "step": 57590,
      "train_runtime": 559959.1872,
      "train_tokens_per_second": 29249.047
    },
    {
      "epoch": 2.038000296324471,
      "grad_norm": 0.61328125,
      "learning_rate": 1.923091147176474e-05,
      "loss": 0.3881336212158203,
      "num_input_tokens_seen": 16381104576,
      "step": 57600,
      "train_runtime": 560056.3161,
      "train_tokens_per_second": 29249.031
    },
    {
      "epoch": 2.0383541165881995,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9229489203824668e-05,
      "loss": 0.3915989398956299,
      "num_input_tokens_seen": 16383972736,
      "step": 57610,
      "train_runtime": 560156.3244,
      "train_tokens_per_second": 29248.929
    },
    {
      "epoch": 2.0387079368519285,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9228067251399588e-05,
      "loss": 0.3817268371582031,
      "num_input_tokens_seen": 16386880320,
      "step": 57620,
      "train_runtime": 560259.8743,
      "train_tokens_per_second": 29248.713
    },
    {
      "epoch": 2.039061757115657,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9226645614372854e-05,
      "loss": 0.3863487482070923,
      "num_input_tokens_seen": 16389696512,
      "step": 57630,
      "train_runtime": 560357.6068,
      "train_tokens_per_second": 29248.638
    },
    {
      "epoch": 2.039415577379386,
      "grad_norm": 0.64453125,
      "learning_rate": 1.922522429262789e-05,
      "loss": 0.39143540859222414,
      "num_input_tokens_seen": 16392534016,
      "step": 57640,
      "train_runtime": 560452.988,
      "train_tokens_per_second": 29248.723
    },
    {
      "epoch": 2.0397693976431146,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9223803286048182e-05,
      "loss": 0.39022886753082275,
      "num_input_tokens_seen": 16395392576,
      "step": 57650,
      "train_runtime": 560552.3168,
      "train_tokens_per_second": 29248.639
    },
    {
      "epoch": 2.0401232179068436,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9222382594517262e-05,
      "loss": 0.3853925228118896,
      "num_input_tokens_seen": 16398186176,
      "step": 57660,
      "train_runtime": 560643.8478,
      "train_tokens_per_second": 29248.847
    },
    {
      "epoch": 2.040477038170572,
      "grad_norm": 0.65234375,
      "learning_rate": 1.9220962217918742e-05,
      "loss": 0.38784065246582033,
      "num_input_tokens_seen": 16400999424,
      "step": 57670,
      "train_runtime": 560739.2843,
      "train_tokens_per_second": 29248.886
    },
    {
      "epoch": 2.040830858434301,
      "grad_norm": 0.640625,
      "learning_rate": 1.921954215613628e-05,
      "loss": 0.38124451637268064,
      "num_input_tokens_seen": 16403853632,
      "step": 57680,
      "train_runtime": 560832.849,
      "train_tokens_per_second": 29249.096
    },
    {
      "epoch": 2.0411846786980297,
      "grad_norm": 0.6015625,
      "learning_rate": 1.92181224090536e-05,
      "loss": 0.38401103019714355,
      "num_input_tokens_seen": 16406737536,
      "step": 57690,
      "train_runtime": 560932.3604,
      "train_tokens_per_second": 29249.048
    },
    {
      "epoch": 2.0415384989617587,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9216702976554487e-05,
      "loss": 0.38697223663330077,
      "num_input_tokens_seen": 16409616320,
      "step": 57700,
      "train_runtime": 561031.5844,
      "train_tokens_per_second": 29249.006
    },
    {
      "epoch": 2.0418923192254876,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9215283858522782e-05,
      "loss": 0.3856818199157715,
      "num_input_tokens_seen": 16412478848,
      "step": 57710,
      "train_runtime": 561129.7848,
      "train_tokens_per_second": 29248.989
    },
    {
      "epoch": 2.042246139489216,
      "grad_norm": 0.59375,
      "learning_rate": 1.9213865054842385e-05,
      "loss": 0.3871749401092529,
      "num_input_tokens_seen": 16415389632,
      "step": 57720,
      "train_runtime": 561229.7931,
      "train_tokens_per_second": 29248.963
    },
    {
      "epoch": 2.042599959752945,
      "grad_norm": 0.6171875,
      "learning_rate": 1.921244656539726e-05,
      "loss": 0.3837709426879883,
      "num_input_tokens_seen": 16418210240,
      "step": 57730,
      "train_runtime": 561324.8009,
      "train_tokens_per_second": 29249.038
    },
    {
      "epoch": 2.0429537800166737,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9211028390071435e-05,
      "loss": 0.38338398933410645,
      "num_input_tokens_seen": 16421070208,
      "step": 57740,
      "train_runtime": 561424.1885,
      "train_tokens_per_second": 29248.954
    },
    {
      "epoch": 2.0433076002804027,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9209610528748988e-05,
      "loss": 0.38839943408966066,
      "num_input_tokens_seen": 16423968064,
      "step": 57750,
      "train_runtime": 561524.4996,
      "train_tokens_per_second": 29248.89
    },
    {
      "epoch": 2.0436614205441312,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9208192981314058e-05,
      "loss": 0.3894335269927979,
      "num_input_tokens_seen": 16426784896,
      "step": 57760,
      "train_runtime": 561622.3918,
      "train_tokens_per_second": 29248.807
    },
    {
      "epoch": 2.0440152408078602,
      "grad_norm": 0.625,
      "learning_rate": 1.9206775747650858e-05,
      "loss": 0.38296008110046387,
      "num_input_tokens_seen": 16429569536,
      "step": 57770,
      "train_runtime": 561715.7375,
      "train_tokens_per_second": 29248.904
    },
    {
      "epoch": 2.0443690610715888,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9205358827643647e-05,
      "loss": 0.38352227210998535,
      "num_input_tokens_seen": 16432386752,
      "step": 57780,
      "train_runtime": 561809.0363,
      "train_tokens_per_second": 29249.061
    },
    {
      "epoch": 2.0447228813353178,
      "grad_norm": 0.609375,
      "learning_rate": 1.9203942221176744e-05,
      "loss": 0.38447539806365966,
      "num_input_tokens_seen": 16435227136,
      "step": 57790,
      "train_runtime": 561905.3125,
      "train_tokens_per_second": 29249.104
    },
    {
      "epoch": 2.0450767015990463,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9202525928134525e-05,
      "loss": 0.3876351356506348,
      "num_input_tokens_seen": 16438072832,
      "step": 57800,
      "train_runtime": 562004.4789,
      "train_tokens_per_second": 29249.007
    },
    {
      "epoch": 2.0454305218627753,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9201109948401445e-05,
      "loss": 0.38484692573547363,
      "num_input_tokens_seen": 16440924416,
      "step": 57810,
      "train_runtime": 562103.1082,
      "train_tokens_per_second": 29248.948
    },
    {
      "epoch": 2.045784342126504,
      "grad_norm": 0.640625,
      "learning_rate": 1.9199694281861996e-05,
      "loss": 0.3874545097351074,
      "num_input_tokens_seen": 16443785088,
      "step": 57820,
      "train_runtime": 562200.5191,
      "train_tokens_per_second": 29248.968
    },
    {
      "epoch": 2.046138162390233,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9198278928400747e-05,
      "loss": 0.38264055252075196,
      "num_input_tokens_seen": 16446580032,
      "step": 57830,
      "train_runtime": 562295.3919,
      "train_tokens_per_second": 29249.004
    },
    {
      "epoch": 2.0464919826539614,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9196863887902305e-05,
      "loss": 0.383734393119812,
      "num_input_tokens_seen": 16449394368,
      "step": 57840,
      "train_runtime": 562390.4881,
      "train_tokens_per_second": 29249.062
    },
    {
      "epoch": 2.0468458029176904,
      "grad_norm": 0.5859375,
      "learning_rate": 1.919544916025136e-05,
      "loss": 0.3878195285797119,
      "num_input_tokens_seen": 16452250240,
      "step": 57850,
      "train_runtime": 562486.8766,
      "train_tokens_per_second": 29249.127
    },
    {
      "epoch": 2.0471996231814193,
      "grad_norm": 0.62890625,
      "learning_rate": 1.919403474533265e-05,
      "loss": 0.38706605434417723,
      "num_input_tokens_seen": 16455117440,
      "step": 57860,
      "train_runtime": 562586.7438,
      "train_tokens_per_second": 29249.032
    },
    {
      "epoch": 2.047553443445148,
      "grad_norm": 0.609375,
      "learning_rate": 1.919262064303097e-05,
      "loss": 0.3906421661376953,
      "num_input_tokens_seen": 16457974784,
      "step": 57870,
      "train_runtime": 562684.496,
      "train_tokens_per_second": 29249.028
    },
    {
      "epoch": 2.047907263708877,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9191206853231182e-05,
      "loss": 0.38469743728637695,
      "num_input_tokens_seen": 16460806592,
      "step": 57880,
      "train_runtime": 562780.0141,
      "train_tokens_per_second": 29249.096
    },
    {
      "epoch": 2.0482610839726054,
      "grad_norm": 0.59375,
      "learning_rate": 1.9189793375818203e-05,
      "loss": 0.3847479343414307,
      "num_input_tokens_seen": 16463637056,
      "step": 57890,
      "train_runtime": 562879.5573,
      "train_tokens_per_second": 29248.952
    },
    {
      "epoch": 2.0486149042363344,
      "grad_norm": 0.625,
      "learning_rate": 1.9188380210677005e-05,
      "loss": 0.39026119709014895,
      "num_input_tokens_seen": 16466482944,
      "step": 57900,
      "train_runtime": 562974.5809,
      "train_tokens_per_second": 29249.07
    },
    {
      "epoch": 2.048968724500063,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9186967357692633e-05,
      "loss": 0.3886385440826416,
      "num_input_tokens_seen": 16469398272,
      "step": 57910,
      "train_runtime": 563077.4703,
      "train_tokens_per_second": 29248.903
    },
    {
      "epoch": 2.049322544763792,
      "grad_norm": 0.6171875,
      "learning_rate": 1.918555481675017e-05,
      "loss": 0.3855393648147583,
      "num_input_tokens_seen": 16472209088,
      "step": 57920,
      "train_runtime": 563174.7692,
      "train_tokens_per_second": 29248.841
    },
    {
      "epoch": 2.0496763650275205,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9184142587734786e-05,
      "loss": 0.3866295099258423,
      "num_input_tokens_seen": 16475035968,
      "step": 57930,
      "train_runtime": 563272.7156,
      "train_tokens_per_second": 29248.773
    },
    {
      "epoch": 2.0500301852912495,
      "grad_norm": 0.62109375,
      "learning_rate": 1.918273067053168e-05,
      "loss": 0.3828116893768311,
      "num_input_tokens_seen": 16477834560,
      "step": 57940,
      "train_runtime": 563366.9196,
      "train_tokens_per_second": 29248.85
    },
    {
      "epoch": 2.050384005554978,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9181319065026137e-05,
      "loss": 0.3829640865325928,
      "num_input_tokens_seen": 16480694144,
      "step": 57950,
      "train_runtime": 563465.1146,
      "train_tokens_per_second": 29248.828
    },
    {
      "epoch": 2.050737825818707,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9179907771103482e-05,
      "loss": 0.38607547283172605,
      "num_input_tokens_seen": 16483574144,
      "step": 57960,
      "train_runtime": 563567.8722,
      "train_tokens_per_second": 29248.605
    },
    {
      "epoch": 2.0510916460824355,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9178496788649106e-05,
      "loss": 0.384198522567749,
      "num_input_tokens_seen": 16486465920,
      "step": 57970,
      "train_runtime": 563664.3736,
      "train_tokens_per_second": 29248.728
    },
    {
      "epoch": 2.0514454663461645,
      "grad_norm": 0.609375,
      "learning_rate": 1.9177086117548462e-05,
      "loss": 0.38707704544067384,
      "num_input_tokens_seen": 16489395392,
      "step": 57980,
      "train_runtime": 563767.966,
      "train_tokens_per_second": 29248.55
    },
    {
      "epoch": 2.051799286609893,
      "grad_norm": 0.62109375,
      "learning_rate": 1.917567575768706e-05,
      "loss": 0.3893141269683838,
      "num_input_tokens_seen": 16492230464,
      "step": 57990,
      "train_runtime": 563864.1126,
      "train_tokens_per_second": 29248.59
    },
    {
      "epoch": 2.052153106873622,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9174265708950457e-05,
      "loss": 0.3840630531311035,
      "num_input_tokens_seen": 16495102592,
      "step": 58000,
      "train_runtime": 563964.7001,
      "train_tokens_per_second": 29248.466
    },
    {
      "epoch": 2.0525069271373506,
      "grad_norm": 0.609375,
      "learning_rate": 1.91728559712243e-05,
      "loss": 0.38563687801361085,
      "num_input_tokens_seen": 16497938880,
      "step": 58010,
      "train_runtime": 564063.0289,
      "train_tokens_per_second": 29248.396
    },
    {
      "epoch": 2.0528607474010796,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9171446544394252e-05,
      "loss": 0.3864439010620117,
      "num_input_tokens_seen": 16500778368,
      "step": 58020,
      "train_runtime": 564159.4754,
      "train_tokens_per_second": 29248.429
    },
    {
      "epoch": 2.0532145676648086,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9170037428346075e-05,
      "loss": 0.38301858901977537,
      "num_input_tokens_seen": 16503592896,
      "step": 58030,
      "train_runtime": 564256.5826,
      "train_tokens_per_second": 29248.383
    },
    {
      "epoch": 2.053568387928537,
      "grad_norm": 0.57421875,
      "learning_rate": 1.9168628622965565e-05,
      "loss": 0.38618757724761965,
      "num_input_tokens_seen": 16506471616,
      "step": 58040,
      "train_runtime": 564356.886,
      "train_tokens_per_second": 29248.286
    },
    {
      "epoch": 2.053922208192266,
      "grad_norm": 0.6171875,
      "learning_rate": 1.916722012813858e-05,
      "loss": 0.38683304786682127,
      "num_input_tokens_seen": 16509271680,
      "step": 58050,
      "train_runtime": 564452.3644,
      "train_tokens_per_second": 29248.299
    },
    {
      "epoch": 2.0542760284559947,
      "grad_norm": 0.60546875,
      "learning_rate": 1.916581194375105e-05,
      "loss": 0.3859408855438232,
      "num_input_tokens_seen": 16512101888,
      "step": 58060,
      "train_runtime": 564544.5039,
      "train_tokens_per_second": 29248.539
    },
    {
      "epoch": 2.0546298487197237,
      "grad_norm": 0.5703125,
      "learning_rate": 1.916440406968895e-05,
      "loss": 0.3859276294708252,
      "num_input_tokens_seen": 16514964032,
      "step": 58070,
      "train_runtime": 564642.1393,
      "train_tokens_per_second": 29248.55
    },
    {
      "epoch": 2.054983668983452,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9162996505838317e-05,
      "loss": 0.3842756271362305,
      "num_input_tokens_seen": 16517902080,
      "step": 58080,
      "train_runtime": 564744.0434,
      "train_tokens_per_second": 29248.475
    },
    {
      "epoch": 2.055337489247181,
      "grad_norm": 0.59375,
      "learning_rate": 1.9161589252085248e-05,
      "loss": 0.38589086532592776,
      "num_input_tokens_seen": 16520802944,
      "step": 58090,
      "train_runtime": 564845.5805,
      "train_tokens_per_second": 29248.353
    },
    {
      "epoch": 2.0556913095109097,
      "grad_norm": 0.609375,
      "learning_rate": 1.9160182308315897e-05,
      "loss": 0.3876638650894165,
      "num_input_tokens_seen": 16523652480,
      "step": 58100,
      "train_runtime": 564944.7787,
      "train_tokens_per_second": 29248.261
    },
    {
      "epoch": 2.0560451297746387,
      "grad_norm": 0.625,
      "learning_rate": 1.9158775674416478e-05,
      "loss": 0.38782439231872556,
      "num_input_tokens_seen": 16526456192,
      "step": 58110,
      "train_runtime": 565036.3657,
      "train_tokens_per_second": 29248.482
    },
    {
      "epoch": 2.0563989500383673,
      "grad_norm": 0.59375,
      "learning_rate": 1.9157369350273266e-05,
      "loss": 0.3871755838394165,
      "num_input_tokens_seen": 16529326656,
      "step": 58120,
      "train_runtime": 565136.1417,
      "train_tokens_per_second": 29248.398
    },
    {
      "epoch": 2.0567527703020962,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9155963335772588e-05,
      "loss": 0.3844400644302368,
      "num_input_tokens_seen": 16532169472,
      "step": 58130,
      "train_runtime": 565231.9643,
      "train_tokens_per_second": 29248.469
    },
    {
      "epoch": 2.057106590565825,
      "grad_norm": 0.64453125,
      "learning_rate": 1.9154557630800835e-05,
      "loss": 0.3873779535293579,
      "num_input_tokens_seen": 16535058240,
      "step": 58140,
      "train_runtime": 565330.8887,
      "train_tokens_per_second": 29248.461
    },
    {
      "epoch": 2.0574604108295538,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9153152235244455e-05,
      "loss": 0.3834215641021729,
      "num_input_tokens_seen": 16537861632,
      "step": 58150,
      "train_runtime": 565427.7113,
      "train_tokens_per_second": 29248.41
    },
    {
      "epoch": 2.0578142310932823,
      "grad_norm": 0.59375,
      "learning_rate": 1.9151747148989955e-05,
      "loss": 0.38964693546295165,
      "num_input_tokens_seen": 16540702272,
      "step": 58160,
      "train_runtime": 565525.4184,
      "train_tokens_per_second": 29248.38
    },
    {
      "epoch": 2.0581680513570113,
      "grad_norm": 0.65234375,
      "learning_rate": 1.915034237192389e-05,
      "loss": 0.3901151180267334,
      "num_input_tokens_seen": 16543565632,
      "step": 58170,
      "train_runtime": 565624.2022,
      "train_tokens_per_second": 29248.334
    },
    {
      "epoch": 2.05852187162074,
      "grad_norm": 0.625,
      "learning_rate": 1.9148937903932894e-05,
      "loss": 0.3880653142929077,
      "num_input_tokens_seen": 16546400960,
      "step": 58180,
      "train_runtime": 565720.4584,
      "train_tokens_per_second": 29248.369
    },
    {
      "epoch": 2.058875691884469,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9147533744903633e-05,
      "loss": 0.38370046615600584,
      "num_input_tokens_seen": 16549215936,
      "step": 58190,
      "train_runtime": 565814.1428,
      "train_tokens_per_second": 29248.502
    },
    {
      "epoch": 2.059229512148198,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9146129894722863e-05,
      "loss": 0.3874203681945801,
      "num_input_tokens_seen": 16552081728,
      "step": 58200,
      "train_runtime": 565911.0306,
      "train_tokens_per_second": 29248.558
    },
    {
      "epoch": 2.0595833324119264,
      "grad_norm": 0.63671875,
      "learning_rate": 1.9144726353277365e-05,
      "loss": 0.384489917755127,
      "num_input_tokens_seen": 16554895168,
      "step": 58210,
      "train_runtime": 566007.9633,
      "train_tokens_per_second": 29248.52
    },
    {
      "epoch": 2.0599371526756554,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9143323120453997e-05,
      "loss": 0.38760085105895997,
      "num_input_tokens_seen": 16557778880,
      "step": 58220,
      "train_runtime": 566110.256,
      "train_tokens_per_second": 29248.329
    },
    {
      "epoch": 2.060290972939384,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9141920196139683e-05,
      "loss": 0.3869471549987793,
      "num_input_tokens_seen": 16560576960,
      "step": 58230,
      "train_runtime": 566205.0338,
      "train_tokens_per_second": 29248.375
    },
    {
      "epoch": 2.060644793203113,
      "grad_norm": 0.62109375,
      "learning_rate": 1.914051758022138e-05,
      "loss": 0.38341827392578126,
      "num_input_tokens_seen": 16563413696,
      "step": 58240,
      "train_runtime": 566304.4096,
      "train_tokens_per_second": 29248.251
    },
    {
      "epoch": 2.0609986134668414,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9139115272586125e-05,
      "loss": 0.3877772092819214,
      "num_input_tokens_seen": 16566302912,
      "step": 58250,
      "train_runtime": 566402.1012,
      "train_tokens_per_second": 29248.308
    },
    {
      "epoch": 2.0613524337305704,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9137713273121004e-05,
      "loss": 0.3847397804260254,
      "num_input_tokens_seen": 16569132608,
      "step": 58260,
      "train_runtime": 566498.1859,
      "train_tokens_per_second": 29248.342
    },
    {
      "epoch": 2.061706253994299,
      "grad_norm": 0.609375,
      "learning_rate": 1.913631158171315e-05,
      "loss": 0.3809072017669678,
      "num_input_tokens_seen": 16571966016,
      "step": 58270,
      "train_runtime": 566595.0986,
      "train_tokens_per_second": 29248.34
    },
    {
      "epoch": 2.062060074258028,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9134910198249782e-05,
      "loss": 0.39027929306030273,
      "num_input_tokens_seen": 16574814464,
      "step": 58280,
      "train_runtime": 566694.2139,
      "train_tokens_per_second": 29248.251
    },
    {
      "epoch": 2.0624138945217565,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9133509122618147e-05,
      "loss": 0.3818178176879883,
      "num_input_tokens_seen": 16577656832,
      "step": 58290,
      "train_runtime": 566791.8585,
      "train_tokens_per_second": 29248.227
    },
    {
      "epoch": 2.0627677147854855,
      "grad_norm": 0.60546875,
      "learning_rate": 1.913210835470557e-05,
      "loss": 0.3854954242706299,
      "num_input_tokens_seen": 16580464512,
      "step": 58300,
      "train_runtime": 566888.2398,
      "train_tokens_per_second": 29248.207
    },
    {
      "epoch": 2.063121535049214,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9130707894399428e-05,
      "loss": 0.38894038200378417,
      "num_input_tokens_seen": 16583327232,
      "step": 58310,
      "train_runtime": 566987.1213,
      "train_tokens_per_second": 29248.155
    },
    {
      "epoch": 2.063475355312943,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9129307741587146e-05,
      "loss": 0.3856056690216064,
      "num_input_tokens_seen": 16586220096,
      "step": 58320,
      "train_runtime": 567088.732,
      "train_tokens_per_second": 29248.016
    },
    {
      "epoch": 2.0638291755766716,
      "grad_norm": 0.640625,
      "learning_rate": 1.9127907896156226e-05,
      "loss": 0.3851458549499512,
      "num_input_tokens_seen": 16589100800,
      "step": 58330,
      "train_runtime": 567188.6443,
      "train_tokens_per_second": 29247.942
    },
    {
      "epoch": 2.0641829958404005,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9126508357994205e-05,
      "loss": 0.38668527603149416,
      "num_input_tokens_seen": 16591993792,
      "step": 58340,
      "train_runtime": 567289.059,
      "train_tokens_per_second": 29247.865
    },
    {
      "epoch": 2.064536816104129,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9125109126988696e-05,
      "loss": 0.38441359996795654,
      "num_input_tokens_seen": 16594866560,
      "step": 58350,
      "train_runtime": 567387.2645,
      "train_tokens_per_second": 29247.866
    },
    {
      "epoch": 2.064890636367858,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9123710203027365e-05,
      "loss": 0.38750877380371096,
      "num_input_tokens_seen": 16597712704,
      "step": 58360,
      "train_runtime": 567485.7848,
      "train_tokens_per_second": 29247.803
    },
    {
      "epoch": 2.065244456631587,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9122311585997927e-05,
      "loss": 0.38785815238952637,
      "num_input_tokens_seen": 16600563392,
      "step": 58370,
      "train_runtime": 567582.9141,
      "train_tokens_per_second": 29247.821
    },
    {
      "epoch": 2.0655982768953156,
      "grad_norm": 0.609375,
      "learning_rate": 1.9120913275788162e-05,
      "loss": 0.3889042377471924,
      "num_input_tokens_seen": 16603433600,
      "step": 58380,
      "train_runtime": 567682.008,
      "train_tokens_per_second": 29247.771
    },
    {
      "epoch": 2.0659520971590446,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9119515272285913e-05,
      "loss": 0.3869378328323364,
      "num_input_tokens_seen": 16606308032,
      "step": 58390,
      "train_runtime": 567782.9588,
      "train_tokens_per_second": 29247.634
    },
    {
      "epoch": 2.066305917422773,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9118117575379065e-05,
      "loss": 0.3873006820678711,
      "num_input_tokens_seen": 16609180992,
      "step": 58400,
      "train_runtime": 567882.1006,
      "train_tokens_per_second": 29247.587
    },
    {
      "epoch": 2.066659737686502,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9116720184955574e-05,
      "loss": 0.38443694114685056,
      "num_input_tokens_seen": 16612066880,
      "step": 58410,
      "train_runtime": 567981.0292,
      "train_tokens_per_second": 29247.573
    },
    {
      "epoch": 2.0670135579502307,
      "grad_norm": 0.63671875,
      "learning_rate": 1.9115323100903446e-05,
      "loss": 0.3899965524673462,
      "num_input_tokens_seen": 16614880960,
      "step": 58420,
      "train_runtime": 568078.0556,
      "train_tokens_per_second": 29247.532
    },
    {
      "epoch": 2.0673673782139597,
      "grad_norm": 0.609375,
      "learning_rate": 1.9113926323110748e-05,
      "loss": 0.3834644317626953,
      "num_input_tokens_seen": 16617709824,
      "step": 58430,
      "train_runtime": 568173.0232,
      "train_tokens_per_second": 29247.622
    },
    {
      "epoch": 2.067721198477688,
      "grad_norm": 0.65625,
      "learning_rate": 1.9112529851465603e-05,
      "loss": 0.385329008102417,
      "num_input_tokens_seen": 16620580800,
      "step": 58440,
      "train_runtime": 568269.4465,
      "train_tokens_per_second": 29247.711
    },
    {
      "epoch": 2.068075018741417,
      "grad_norm": 0.61328125,
      "learning_rate": 1.911113368585619e-05,
      "loss": 0.3836948394775391,
      "num_input_tokens_seen": 16623380800,
      "step": 58450,
      "train_runtime": 568364.135,
      "train_tokens_per_second": 29247.765
    },
    {
      "epoch": 2.0684288390051457,
      "grad_norm": 0.60546875,
      "learning_rate": 1.910973782617074e-05,
      "loss": 0.38384337425231935,
      "num_input_tokens_seen": 16626224128,
      "step": 58460,
      "train_runtime": 568463.8353,
      "train_tokens_per_second": 29247.637
    },
    {
      "epoch": 2.0687826592688747,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9108342272297558e-05,
      "loss": 0.38511841297149657,
      "num_input_tokens_seen": 16629075392,
      "step": 58470,
      "train_runtime": 568561.4303,
      "train_tokens_per_second": 29247.632
    },
    {
      "epoch": 2.0691364795326033,
      "grad_norm": 0.62109375,
      "learning_rate": 1.910694702412499e-05,
      "loss": 0.38358519077301023,
      "num_input_tokens_seen": 16631960256,
      "step": 58480,
      "train_runtime": 568660.2095,
      "train_tokens_per_second": 29247.624
    },
    {
      "epoch": 2.0694902997963323,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9105552081541444e-05,
      "loss": 0.38654475212097167,
      "num_input_tokens_seen": 16634859328,
      "step": 58490,
      "train_runtime": 568760.3801,
      "train_tokens_per_second": 29247.571
    },
    {
      "epoch": 2.069844120060061,
      "grad_norm": 0.61328125,
      "learning_rate": 1.910415744443539e-05,
      "loss": 0.3828063249588013,
      "num_input_tokens_seen": 16637724160,
      "step": 58500,
      "train_runtime": 568857.6059,
      "train_tokens_per_second": 29247.608
    },
    {
      "epoch": 2.07019794032379,
      "grad_norm": 0.609375,
      "learning_rate": 1.9102763112695343e-05,
      "loss": 0.383486270904541,
      "num_input_tokens_seen": 16640587520,
      "step": 58510,
      "train_runtime": 568958.2697,
      "train_tokens_per_second": 29247.466
    },
    {
      "epoch": 2.0705517605875183,
      "grad_norm": 0.6171875,
      "learning_rate": 1.910136908620989e-05,
      "loss": 0.3876341342926025,
      "num_input_tokens_seen": 16643402432,
      "step": 58520,
      "train_runtime": 569053.6295,
      "train_tokens_per_second": 29247.511
    },
    {
      "epoch": 2.0709055808512473,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9099975364867663e-05,
      "loss": 0.3868858814239502,
      "num_input_tokens_seen": 16646235200,
      "step": 58530,
      "train_runtime": 569152.5102,
      "train_tokens_per_second": 29247.407
    },
    {
      "epoch": 2.0712594011149763,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9098581948557356e-05,
      "loss": 0.38619887828826904,
      "num_input_tokens_seen": 16649072384,
      "step": 58540,
      "train_runtime": 569249.3385,
      "train_tokens_per_second": 29247.416
    },
    {
      "epoch": 2.071613221378705,
      "grad_norm": 0.59765625,
      "learning_rate": 1.909718883716772e-05,
      "loss": 0.38712923526763915,
      "num_input_tokens_seen": 16651890624,
      "step": 58550,
      "train_runtime": 569347.113,
      "train_tokens_per_second": 29247.344
    },
    {
      "epoch": 2.071967041642434,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9095796030587556e-05,
      "loss": 0.3854062557220459,
      "num_input_tokens_seen": 16654746560,
      "step": 58560,
      "train_runtime": 569446.2075,
      "train_tokens_per_second": 29247.269
    },
    {
      "epoch": 2.0723208619061624,
      "grad_norm": 0.625,
      "learning_rate": 1.9094403528705734e-05,
      "loss": 0.39183869361877444,
      "num_input_tokens_seen": 16657629760,
      "step": 58570,
      "train_runtime": 569548.039,
      "train_tokens_per_second": 29247.102
    },
    {
      "epoch": 2.0726746821698914,
      "grad_norm": 0.6484375,
      "learning_rate": 1.9093011331411173e-05,
      "loss": 0.3855603694915771,
      "num_input_tokens_seen": 16660483520,
      "step": 58580,
      "train_runtime": 569645.0262,
      "train_tokens_per_second": 29247.132
    },
    {
      "epoch": 2.07302850243362,
      "grad_norm": 0.609375,
      "learning_rate": 1.9091619438592854e-05,
      "loss": 0.38959057331085206,
      "num_input_tokens_seen": 16663353984,
      "step": 58590,
      "train_runtime": 569743.7706,
      "train_tokens_per_second": 29247.102
    },
    {
      "epoch": 2.073382322697349,
      "grad_norm": 0.59765625,
      "learning_rate": 1.90902278501398e-05,
      "loss": 0.38801748752593995,
      "num_input_tokens_seen": 16666182080,
      "step": 58600,
      "train_runtime": 569837.3714,
      "train_tokens_per_second": 29247.261
    },
    {
      "epoch": 2.0737361429610774,
      "grad_norm": 0.6015625,
      "learning_rate": 1.908883656594111e-05,
      "loss": 0.38353424072265624,
      "num_input_tokens_seen": 16669053312,
      "step": 58610,
      "train_runtime": 569935.9164,
      "train_tokens_per_second": 29247.241
    },
    {
      "epoch": 2.0740899632248064,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9087445585885926e-05,
      "loss": 0.3850236892700195,
      "num_input_tokens_seen": 16671956672,
      "step": 58620,
      "train_runtime": 570038.1109,
      "train_tokens_per_second": 29247.091
    },
    {
      "epoch": 2.074443783488535,
      "grad_norm": 0.625,
      "learning_rate": 1.9086054909863457e-05,
      "loss": 0.3869431972503662,
      "num_input_tokens_seen": 16674774464,
      "step": 58630,
      "train_runtime": 570133.7539,
      "train_tokens_per_second": 29247.127
    },
    {
      "epoch": 2.074797603752264,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9084664537762958e-05,
      "loss": 0.382286810874939,
      "num_input_tokens_seen": 16677654784,
      "step": 58640,
      "train_runtime": 570233.7303,
      "train_tokens_per_second": 29247.051
    },
    {
      "epoch": 2.0751514240159925,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9083274469473747e-05,
      "loss": 0.3872214794158936,
      "num_input_tokens_seen": 16680489984,
      "step": 58650,
      "train_runtime": 570332.0646,
      "train_tokens_per_second": 29246.979
    },
    {
      "epoch": 2.0755052442797215,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9081884704885193e-05,
      "loss": 0.3864173889160156,
      "num_input_tokens_seen": 16683313088,
      "step": 58660,
      "train_runtime": 570428.5675,
      "train_tokens_per_second": 29246.98
    },
    {
      "epoch": 2.07585906454345,
      "grad_norm": 0.625,
      "learning_rate": 1.9080495243886733e-05,
      "loss": 0.3849653720855713,
      "num_input_tokens_seen": 16686174720,
      "step": 58670,
      "train_runtime": 570528.1999,
      "train_tokens_per_second": 29246.889
    },
    {
      "epoch": 2.076212884807179,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9079106086367845e-05,
      "loss": 0.38805761337280276,
      "num_input_tokens_seen": 16689068864,
      "step": 58680,
      "train_runtime": 570628.8018,
      "train_tokens_per_second": 29246.804
    },
    {
      "epoch": 2.076566705070908,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9077717232218072e-05,
      "loss": 0.3826923847198486,
      "num_input_tokens_seen": 16691909440,
      "step": 58690,
      "train_runtime": 570726.7727,
      "train_tokens_per_second": 29246.761
    },
    {
      "epoch": 2.0769205253346366,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9076328681327017e-05,
      "loss": 0.38638982772827146,
      "num_input_tokens_seen": 16694771648,
      "step": 58700,
      "train_runtime": 570824.6266,
      "train_tokens_per_second": 29246.761
    },
    {
      "epoch": 2.0772743455983655,
      "grad_norm": 0.61328125,
      "learning_rate": 1.907494043358433e-05,
      "loss": 0.3822152137756348,
      "num_input_tokens_seen": 16697627008,
      "step": 58710,
      "train_runtime": 570922.0915,
      "train_tokens_per_second": 29246.77
    },
    {
      "epoch": 2.077628165862094,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9073552488879724e-05,
      "loss": 0.38420515060424804,
      "num_input_tokens_seen": 16700488768,
      "step": 58720,
      "train_runtime": 571020.7905,
      "train_tokens_per_second": 29246.726
    },
    {
      "epoch": 2.077981986125823,
      "grad_norm": 0.609375,
      "learning_rate": 1.907216484710296e-05,
      "loss": 0.3886378288269043,
      "num_input_tokens_seen": 16703329920,
      "step": 58730,
      "train_runtime": 571115.4013,
      "train_tokens_per_second": 29246.856
    },
    {
      "epoch": 2.0783358063895516,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9070777508143868e-05,
      "loss": 0.3881584882736206,
      "num_input_tokens_seen": 16706194816,
      "step": 58740,
      "train_runtime": 571215.5316,
      "train_tokens_per_second": 29246.745
    },
    {
      "epoch": 2.0786896266532806,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9069390471892324e-05,
      "loss": 0.38559885025024415,
      "num_input_tokens_seen": 16708993728,
      "step": 58750,
      "train_runtime": 571309.2894,
      "train_tokens_per_second": 29246.844
    },
    {
      "epoch": 2.079043446917009,
      "grad_norm": 0.609375,
      "learning_rate": 1.9068003738238263e-05,
      "loss": 0.3888734817504883,
      "num_input_tokens_seen": 16711879168,
      "step": 58760,
      "train_runtime": 571407.3433,
      "train_tokens_per_second": 29246.875
    },
    {
      "epoch": 2.079397267180738,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9066617307071674e-05,
      "loss": 0.3883217811584473,
      "num_input_tokens_seen": 16714695232,
      "step": 58770,
      "train_runtime": 571506.2569,
      "train_tokens_per_second": 29246.741
    },
    {
      "epoch": 2.0797510874444667,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9065231178282607e-05,
      "loss": 0.3875225782394409,
      "num_input_tokens_seen": 16717574208,
      "step": 58780,
      "train_runtime": 571603.4339,
      "train_tokens_per_second": 29246.805
    },
    {
      "epoch": 2.0801049077081957,
      "grad_norm": 0.625,
      "learning_rate": 1.906384535176116e-05,
      "loss": 0.387648344039917,
      "num_input_tokens_seen": 16720437760,
      "step": 58790,
      "train_runtime": 571702.9974,
      "train_tokens_per_second": 29246.72
    },
    {
      "epoch": 2.080458727971924,
      "grad_norm": 0.640625,
      "learning_rate": 1.90624598273975e-05,
      "loss": 0.3830115795135498,
      "num_input_tokens_seen": 16723306368,
      "step": 58800,
      "train_runtime": 571802.8289,
      "train_tokens_per_second": 29246.631
    },
    {
      "epoch": 2.080812548235653,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9061074605081837e-05,
      "loss": 0.3860777378082275,
      "num_input_tokens_seen": 16726129088,
      "step": 58810,
      "train_runtime": 571898.008,
      "train_tokens_per_second": 29246.699
    },
    {
      "epoch": 2.0811663684993817,
      "grad_norm": 0.609375,
      "learning_rate": 1.905968968470444e-05,
      "loss": 0.38523015975952146,
      "num_input_tokens_seen": 16728959232,
      "step": 58820,
      "train_runtime": 571993.0954,
      "train_tokens_per_second": 29246.785
    },
    {
      "epoch": 2.0815201887631107,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9058305066155632e-05,
      "loss": 0.38867983818054197,
      "num_input_tokens_seen": 16731800960,
      "step": 58830,
      "train_runtime": 572091.0111,
      "train_tokens_per_second": 29246.747
    },
    {
      "epoch": 2.0818740090268393,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9056920749325805e-05,
      "loss": 0.38263487815856934,
      "num_input_tokens_seen": 16734670592,
      "step": 58840,
      "train_runtime": 572189.971,
      "train_tokens_per_second": 29246.704
    },
    {
      "epoch": 2.0822278292905683,
      "grad_norm": 0.58984375,
      "learning_rate": 1.905553673410539e-05,
      "loss": 0.38182716369628905,
      "num_input_tokens_seen": 16737600256,
      "step": 58850,
      "train_runtime": 572292.3695,
      "train_tokens_per_second": 29246.59
    },
    {
      "epoch": 2.082581649554297,
      "grad_norm": 0.61328125,
      "learning_rate": 1.905415302038488e-05,
      "loss": 0.3883997440338135,
      "num_input_tokens_seen": 16740470080,
      "step": 58860,
      "train_runtime": 572392.6356,
      "train_tokens_per_second": 29246.481
    },
    {
      "epoch": 2.082935469818026,
      "grad_norm": 0.60546875,
      "learning_rate": 1.905276960805483e-05,
      "loss": 0.3892531871795654,
      "num_input_tokens_seen": 16743316864,
      "step": 58870,
      "train_runtime": 572490.0627,
      "train_tokens_per_second": 29246.476
    },
    {
      "epoch": 2.083289290081755,
      "grad_norm": 0.63671875,
      "learning_rate": 1.9051386497005842e-05,
      "loss": 0.3850471019744873,
      "num_input_tokens_seen": 16746192640,
      "step": 58880,
      "train_runtime": 572589.2054,
      "train_tokens_per_second": 29246.434
    },
    {
      "epoch": 2.0836431103454833,
      "grad_norm": 0.609375,
      "learning_rate": 1.9050003687128574e-05,
      "loss": 0.38381457328796387,
      "num_input_tokens_seen": 16749062784,
      "step": 58890,
      "train_runtime": 572688.7732,
      "train_tokens_per_second": 29246.361
    },
    {
      "epoch": 2.0839969306092123,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9048621178313737e-05,
      "loss": 0.38531529903411865,
      "num_input_tokens_seen": 16751884544,
      "step": 58900,
      "train_runtime": 572787.0305,
      "train_tokens_per_second": 29246.271
    },
    {
      "epoch": 2.084350750872941,
      "grad_norm": 0.625,
      "learning_rate": 1.9047238970452116e-05,
      "loss": 0.38550586700439454,
      "num_input_tokens_seen": 16754719872,
      "step": 58910,
      "train_runtime": 572884.2049,
      "train_tokens_per_second": 29246.259
    },
    {
      "epoch": 2.08470457113667,
      "grad_norm": 0.61328125,
      "learning_rate": 1.904585706343453e-05,
      "loss": 0.3814000368118286,
      "num_input_tokens_seen": 16757534656,
      "step": 58920,
      "train_runtime": 572980.3912,
      "train_tokens_per_second": 29246.262
    },
    {
      "epoch": 2.0850583914003984,
      "grad_norm": 0.58203125,
      "learning_rate": 1.904447545715186e-05,
      "loss": 0.38686399459838866,
      "num_input_tokens_seen": 16760392192,
      "step": 58930,
      "train_runtime": 573078.6153,
      "train_tokens_per_second": 29246.236
    },
    {
      "epoch": 2.0854122116641274,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9043094151495045e-05,
      "loss": 0.3836365699768066,
      "num_input_tokens_seen": 16763251072,
      "step": 58940,
      "train_runtime": 573174.1386,
      "train_tokens_per_second": 29246.349
    },
    {
      "epoch": 2.085766031927856,
      "grad_norm": 0.64453125,
      "learning_rate": 1.904171314635508e-05,
      "loss": 0.38606197834014894,
      "num_input_tokens_seen": 16766037888,
      "step": 58950,
      "train_runtime": 573267.363,
      "train_tokens_per_second": 29246.455
    },
    {
      "epoch": 2.086119852191585,
      "grad_norm": 0.6171875,
      "learning_rate": 1.904033244162301e-05,
      "loss": 0.38819375038146975,
      "num_input_tokens_seen": 16768900480,
      "step": 58960,
      "train_runtime": 573365.602,
      "train_tokens_per_second": 29246.436
    },
    {
      "epoch": 2.0864736724553135,
      "grad_norm": 0.6171875,
      "learning_rate": 1.903895203718994e-05,
      "loss": 0.38612895011901854,
      "num_input_tokens_seen": 16771760064,
      "step": 58970,
      "train_runtime": 573461.1319,
      "train_tokens_per_second": 29246.551
    },
    {
      "epoch": 2.0868274927190424,
      "grad_norm": 0.6171875,
      "learning_rate": 1.903757193294703e-05,
      "loss": 0.3860905647277832,
      "num_input_tokens_seen": 16774568384,
      "step": 58980,
      "train_runtime": 573555.2946,
      "train_tokens_per_second": 29246.646
    },
    {
      "epoch": 2.087181312982771,
      "grad_norm": 0.58984375,
      "learning_rate": 1.9036192128785495e-05,
      "loss": 0.3890978336334229,
      "num_input_tokens_seen": 16777428352,
      "step": 58990,
      "train_runtime": 573656.3501,
      "train_tokens_per_second": 29246.479
    },
    {
      "epoch": 2.0875351332465,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9034812624596605e-05,
      "loss": 0.38856046199798583,
      "num_input_tokens_seen": 16780237632,
      "step": 59000,
      "train_runtime": 573751.5704,
      "train_tokens_per_second": 29246.521
    },
    {
      "epoch": 2.0878889535102285,
      "grad_norm": 0.61328125,
      "learning_rate": 1.903343342027168e-05,
      "loss": 0.3844883441925049,
      "num_input_tokens_seen": 16783086208,
      "step": 59010,
      "train_runtime": 573848.4889,
      "train_tokens_per_second": 29246.546
    },
    {
      "epoch": 2.0882427737739575,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9032054515702103e-05,
      "loss": 0.3868630170822144,
      "num_input_tokens_seen": 16785940544,
      "step": 59020,
      "train_runtime": 573946.2901,
      "train_tokens_per_second": 29246.535
    },
    {
      "epoch": 2.0885965940376865,
      "grad_norm": 0.640625,
      "learning_rate": 1.9030675910779307e-05,
      "loss": 0.3906904935836792,
      "num_input_tokens_seen": 16788787776,
      "step": 59030,
      "train_runtime": 574046.0853,
      "train_tokens_per_second": 29246.411
    },
    {
      "epoch": 2.088950414301415,
      "grad_norm": 0.625,
      "learning_rate": 1.9029297605394782e-05,
      "loss": 0.3896930456161499,
      "num_input_tokens_seen": 16791682304,
      "step": 59040,
      "train_runtime": 574147.1514,
      "train_tokens_per_second": 29246.304
    },
    {
      "epoch": 2.089304234565144,
      "grad_norm": 0.625,
      "learning_rate": 1.9027919599440077e-05,
      "loss": 0.3879457712173462,
      "num_input_tokens_seen": 16794538944,
      "step": 59050,
      "train_runtime": 574246.5517,
      "train_tokens_per_second": 29246.216
    },
    {
      "epoch": 2.0896580548288726,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9026541892806784e-05,
      "loss": 0.3865446805953979,
      "num_input_tokens_seen": 16797344256,
      "step": 59060,
      "train_runtime": 574340.3598,
      "train_tokens_per_second": 29246.324
    },
    {
      "epoch": 2.0900118750926016,
      "grad_norm": 0.59375,
      "learning_rate": 1.9025164485386562e-05,
      "loss": 0.3882107734680176,
      "num_input_tokens_seen": 16800192256,
      "step": 59070,
      "train_runtime": 574436.381,
      "train_tokens_per_second": 29246.393
    },
    {
      "epoch": 2.09036569535633,
      "grad_norm": 0.62109375,
      "learning_rate": 1.9023787377071123e-05,
      "loss": 0.37806999683380127,
      "num_input_tokens_seen": 16803091072,
      "step": 59080,
      "train_runtime": 574536.6708,
      "train_tokens_per_second": 29246.333
    },
    {
      "epoch": 2.090719515620059,
      "grad_norm": 0.61328125,
      "learning_rate": 1.902241056775223e-05,
      "loss": 0.3875662088394165,
      "num_input_tokens_seen": 16805860032,
      "step": 59090,
      "train_runtime": 574631.5893,
      "train_tokens_per_second": 29246.321
    },
    {
      "epoch": 2.0910733358837876,
      "grad_norm": 0.59765625,
      "learning_rate": 1.9021034057321694e-05,
      "loss": 0.3790420055389404,
      "num_input_tokens_seen": 16808715520,
      "step": 59100,
      "train_runtime": 574728.4199,
      "train_tokens_per_second": 29246.362
    },
    {
      "epoch": 2.0914271561475166,
      "grad_norm": 0.609375,
      "learning_rate": 1.90196578456714e-05,
      "loss": 0.38483433723449706,
      "num_input_tokens_seen": 16811623104,
      "step": 59110,
      "train_runtime": 574828.546,
      "train_tokens_per_second": 29246.326
    },
    {
      "epoch": 2.091780976411245,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9018281932693273e-05,
      "loss": 0.38743863105773924,
      "num_input_tokens_seen": 16814509888,
      "step": 59120,
      "train_runtime": 574927.043,
      "train_tokens_per_second": 29246.337
    },
    {
      "epoch": 2.092134796674974,
      "grad_norm": 0.625,
      "learning_rate": 1.9016906318279295e-05,
      "loss": 0.3846376895904541,
      "num_input_tokens_seen": 16817333248,
      "step": 59130,
      "train_runtime": 575023.0021,
      "train_tokens_per_second": 29246.366
    },
    {
      "epoch": 2.0924886169387027,
      "grad_norm": 0.609375,
      "learning_rate": 1.9015531002321503e-05,
      "loss": 0.3846463203430176,
      "num_input_tokens_seen": 16820180224,
      "step": 59140,
      "train_runtime": 575121.3335,
      "train_tokens_per_second": 29246.316
    },
    {
      "epoch": 2.0928424372024317,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9014155984711995e-05,
      "loss": 0.38582520484924315,
      "num_input_tokens_seen": 16823024000,
      "step": 59150,
      "train_runtime": 575220.0346,
      "train_tokens_per_second": 29246.241
    },
    {
      "epoch": 2.0931962574661602,
      "grad_norm": 0.625,
      "learning_rate": 1.9012781265342913e-05,
      "loss": 0.3891855955123901,
      "num_input_tokens_seen": 16825902592,
      "step": 59160,
      "train_runtime": 575320.6714,
      "train_tokens_per_second": 29246.129
    },
    {
      "epoch": 2.093550077729889,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9011406844106467e-05,
      "loss": 0.3889988899230957,
      "num_input_tokens_seen": 16828736576,
      "step": 59170,
      "train_runtime": 575417.0578,
      "train_tokens_per_second": 29246.155
    },
    {
      "epoch": 2.0939038979936178,
      "grad_norm": 0.59765625,
      "learning_rate": 1.90100327208949e-05,
      "loss": 0.3889432907104492,
      "num_input_tokens_seen": 16831600000,
      "step": 59180,
      "train_runtime": 575515.0192,
      "train_tokens_per_second": 29246.152
    },
    {
      "epoch": 2.0942577182573467,
      "grad_norm": 0.6171875,
      "learning_rate": 1.9008658895600537e-05,
      "loss": 0.3837843418121338,
      "num_input_tokens_seen": 16834408512,
      "step": 59190,
      "train_runtime": 575608.6488,
      "train_tokens_per_second": 29246.274
    },
    {
      "epoch": 2.0946115385210757,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9007285368115737e-05,
      "loss": 0.38916356563568116,
      "num_input_tokens_seen": 16837252480,
      "step": 59200,
      "train_runtime": 575706.2749,
      "train_tokens_per_second": 29246.255
    },
    {
      "epoch": 2.0949653587848043,
      "grad_norm": 0.609375,
      "learning_rate": 1.9005912138332918e-05,
      "loss": 0.3871171474456787,
      "num_input_tokens_seen": 16840087232,
      "step": 59210,
      "train_runtime": 575804.9652,
      "train_tokens_per_second": 29246.165
    },
    {
      "epoch": 2.0953191790485333,
      "grad_norm": 0.609375,
      "learning_rate": 1.9004539206144557e-05,
      "loss": 0.3865817546844482,
      "num_input_tokens_seen": 16842959360,
      "step": 59220,
      "train_runtime": 575906.016,
      "train_tokens_per_second": 29246.021
    },
    {
      "epoch": 2.095672999312262,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9003166571443183e-05,
      "loss": 0.39054596424102783,
      "num_input_tokens_seen": 16845817536,
      "step": 59230,
      "train_runtime": 576007.0958,
      "train_tokens_per_second": 29245.851
    },
    {
      "epoch": 2.096026819575991,
      "grad_norm": 0.61328125,
      "learning_rate": 1.900179423412138e-05,
      "loss": 0.3876781463623047,
      "num_input_tokens_seen": 16848680640,
      "step": 59240,
      "train_runtime": 576106.5664,
      "train_tokens_per_second": 29245.771
    },
    {
      "epoch": 2.0963806398397193,
      "grad_norm": 0.61328125,
      "learning_rate": 1.900042219407178e-05,
      "loss": 0.3834547758102417,
      "num_input_tokens_seen": 16851506560,
      "step": 59250,
      "train_runtime": 576201.7135,
      "train_tokens_per_second": 29245.846
    },
    {
      "epoch": 2.0967344601034483,
      "grad_norm": 0.609375,
      "learning_rate": 1.8999050451187084e-05,
      "loss": 0.3865455150604248,
      "num_input_tokens_seen": 16854356672,
      "step": 59260,
      "train_runtime": 576298.8162,
      "train_tokens_per_second": 29245.864
    },
    {
      "epoch": 2.097088280367177,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8997679005360026e-05,
      "loss": 0.38815932273864745,
      "num_input_tokens_seen": 16857248256,
      "step": 59270,
      "train_runtime": 576400.4358,
      "train_tokens_per_second": 29245.724
    },
    {
      "epoch": 2.097442100630906,
      "grad_norm": 0.609375,
      "learning_rate": 1.8996307856483418e-05,
      "loss": 0.3849550724029541,
      "num_input_tokens_seen": 16860063168,
      "step": 59280,
      "train_runtime": 576495.2226,
      "train_tokens_per_second": 29245.799
    },
    {
      "epoch": 2.0977959208946344,
      "grad_norm": 0.59375,
      "learning_rate": 1.8994937004450106e-05,
      "loss": 0.389237642288208,
      "num_input_tokens_seen": 16862879744,
      "step": 59290,
      "train_runtime": 576590.4013,
      "train_tokens_per_second": 29245.856
    },
    {
      "epoch": 2.0981497411583634,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8993566449153e-05,
      "loss": 0.38845076560974123,
      "num_input_tokens_seen": 16865706624,
      "step": 59300,
      "train_runtime": 576684.6441,
      "train_tokens_per_second": 29245.978
    },
    {
      "epoch": 2.098503561422092,
      "grad_norm": 0.6484375,
      "learning_rate": 1.8992196190485065e-05,
      "loss": 0.3912659645080566,
      "num_input_tokens_seen": 16868554560,
      "step": 59310,
      "train_runtime": 576783.3222,
      "train_tokens_per_second": 29245.912
    },
    {
      "epoch": 2.098857381685821,
      "grad_norm": 0.609375,
      "learning_rate": 1.8990826228339316e-05,
      "loss": 0.3865690231323242,
      "num_input_tokens_seen": 16871385536,
      "step": 59320,
      "train_runtime": 576877.9691,
      "train_tokens_per_second": 29246.022
    },
    {
      "epoch": 2.0992112019495495,
      "grad_norm": 0.59375,
      "learning_rate": 1.898945656260882e-05,
      "loss": 0.38594870567321776,
      "num_input_tokens_seen": 16874222016,
      "step": 59330,
      "train_runtime": 576975.5969,
      "train_tokens_per_second": 29245.989
    },
    {
      "epoch": 2.0995650222132785,
      "grad_norm": 0.6171875,
      "learning_rate": 1.89880871931867e-05,
      "loss": 0.38623223304748533,
      "num_input_tokens_seen": 16877072512,
      "step": 59340,
      "train_runtime": 577076.1244,
      "train_tokens_per_second": 29245.834
    },
    {
      "epoch": 2.099918842477007,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8986718119966142e-05,
      "loss": 0.38059368133544924,
      "num_input_tokens_seen": 16879917504,
      "step": 59350,
      "train_runtime": 577174.0837,
      "train_tokens_per_second": 29245.799
    },
    {
      "epoch": 2.100272662740736,
      "grad_norm": 0.62109375,
      "learning_rate": 1.898534934284038e-05,
      "loss": 0.3889148235321045,
      "num_input_tokens_seen": 16882767488,
      "step": 59360,
      "train_runtime": 577272.8232,
      "train_tokens_per_second": 29245.734
    },
    {
      "epoch": 2.100626483004465,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8983980861702692e-05,
      "loss": 0.38382296562194823,
      "num_input_tokens_seen": 16885574272,
      "step": 59370,
      "train_runtime": 577366.0794,
      "train_tokens_per_second": 29245.872
    },
    {
      "epoch": 2.1009803032681935,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8982612676446417e-05,
      "loss": 0.3874004364013672,
      "num_input_tokens_seen": 16888400384,
      "step": 59380,
      "train_runtime": 577459.3854,
      "train_tokens_per_second": 29246.04
    },
    {
      "epoch": 2.1013341235319225,
      "grad_norm": 0.609375,
      "learning_rate": 1.898124478696496e-05,
      "loss": 0.3878662586212158,
      "num_input_tokens_seen": 16891245568,
      "step": 59390,
      "train_runtime": 577557.2865,
      "train_tokens_per_second": 29246.009
    },
    {
      "epoch": 2.101687943795651,
      "grad_norm": 0.60546875,
      "learning_rate": 1.897987719315175e-05,
      "loss": 0.3853639602661133,
      "num_input_tokens_seen": 16894094848,
      "step": 59400,
      "train_runtime": 577654.9739,
      "train_tokens_per_second": 29245.996
    },
    {
      "epoch": 2.10204176405938,
      "grad_norm": 0.625,
      "learning_rate": 1.8978509894900306e-05,
      "loss": 0.38562397956848143,
      "num_input_tokens_seen": 16896904000,
      "step": 59410,
      "train_runtime": 577750.8945,
      "train_tokens_per_second": 29246.002
    },
    {
      "epoch": 2.1023955843231086,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8977142892104177e-05,
      "loss": 0.383879566192627,
      "num_input_tokens_seen": 16899750528,
      "step": 59420,
      "train_runtime": 577848.2879,
      "train_tokens_per_second": 29245.999
    },
    {
      "epoch": 2.1027494045868376,
      "grad_norm": 0.63671875,
      "learning_rate": 1.897577618465697e-05,
      "loss": 0.38712153434753416,
      "num_input_tokens_seen": 16902604160,
      "step": 59430,
      "train_runtime": 577947.9629,
      "train_tokens_per_second": 29245.893
    },
    {
      "epoch": 2.103103224850566,
      "grad_norm": 0.6640625,
      "learning_rate": 1.8974409772452347e-05,
      "loss": 0.3875401496887207,
      "num_input_tokens_seen": 16905459840,
      "step": 59440,
      "train_runtime": 578043.1975,
      "train_tokens_per_second": 29246.015
    },
    {
      "epoch": 2.103457045114295,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8973043655384025e-05,
      "loss": 0.3876309394836426,
      "num_input_tokens_seen": 16908293312,
      "step": 59450,
      "train_runtime": 578138.379,
      "train_tokens_per_second": 29246.101
    },
    {
      "epoch": 2.1038108653780236,
      "grad_norm": 0.609375,
      "learning_rate": 1.8971677833345776e-05,
      "loss": 0.38457136154174804,
      "num_input_tokens_seen": 16911197632,
      "step": 59460,
      "train_runtime": 578236.4712,
      "train_tokens_per_second": 29246.162
    },
    {
      "epoch": 2.1041646856417526,
      "grad_norm": 0.625,
      "learning_rate": 1.8970312306231413e-05,
      "loss": 0.39127962589263915,
      "num_input_tokens_seen": 16914028416,
      "step": 59470,
      "train_runtime": 578333.3764,
      "train_tokens_per_second": 29246.156
    },
    {
      "epoch": 2.104518505905481,
      "grad_norm": 0.6484375,
      "learning_rate": 1.896894707393483e-05,
      "loss": 0.38178465366363523,
      "num_input_tokens_seen": 16916836288,
      "step": 59480,
      "train_runtime": 578430.9873,
      "train_tokens_per_second": 29246.075
    },
    {
      "epoch": 2.10487232616921,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8967582136349938e-05,
      "loss": 0.3837705135345459,
      "num_input_tokens_seen": 16919666560,
      "step": 59490,
      "train_runtime": 578527.97,
      "train_tokens_per_second": 29246.065
    },
    {
      "epoch": 2.1052261464329387,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8966217493370726e-05,
      "loss": 0.3887350082397461,
      "num_input_tokens_seen": 16922553664,
      "step": 59500,
      "train_runtime": 578626.8082,
      "train_tokens_per_second": 29246.059
    },
    {
      "epoch": 2.1055799666966677,
      "grad_norm": 0.6171875,
      "learning_rate": 1.896485314489124e-05,
      "loss": 0.37958450317382814,
      "num_input_tokens_seen": 16925391424,
      "step": 59510,
      "train_runtime": 578723.8851,
      "train_tokens_per_second": 29246.056
    },
    {
      "epoch": 2.1059337869603967,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8963489090805563e-05,
      "loss": 0.3833619594573975,
      "num_input_tokens_seen": 16928190656,
      "step": 59520,
      "train_runtime": 578817.5683,
      "train_tokens_per_second": 29246.159
    },
    {
      "epoch": 2.1062876072241252,
      "grad_norm": 0.60546875,
      "learning_rate": 1.896212533100784e-05,
      "loss": 0.3876672267913818,
      "num_input_tokens_seen": 16931077696,
      "step": 59530,
      "train_runtime": 578915.4669,
      "train_tokens_per_second": 29246.2
    },
    {
      "epoch": 2.106641427487854,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8960761865392257e-05,
      "loss": 0.3884193181991577,
      "num_input_tokens_seen": 16933905408,
      "step": 59540,
      "train_runtime": 579011.2292,
      "train_tokens_per_second": 29246.247
    },
    {
      "epoch": 2.1069952477515828,
      "grad_norm": 0.62109375,
      "learning_rate": 1.895939869385308e-05,
      "loss": 0.38404288291931155,
      "num_input_tokens_seen": 16936739008,
      "step": 59550,
      "train_runtime": 579108.4406,
      "train_tokens_per_second": 29246.231
    },
    {
      "epoch": 2.1073490680153117,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8958035816284607e-05,
      "loss": 0.38599746227264403,
      "num_input_tokens_seen": 16939603072,
      "step": 59560,
      "train_runtime": 579205.0048,
      "train_tokens_per_second": 29246.3
    },
    {
      "epoch": 2.1077028882790403,
      "grad_norm": 0.609375,
      "learning_rate": 1.8956673232581184e-05,
      "loss": 0.3837181568145752,
      "num_input_tokens_seen": 16942526144,
      "step": 59570,
      "train_runtime": 579307.0688,
      "train_tokens_per_second": 29246.193
    },
    {
      "epoch": 2.1080567085427693,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8955310942637233e-05,
      "loss": 0.3881374835968018,
      "num_input_tokens_seen": 16945340224,
      "step": 59580,
      "train_runtime": 579401.9956,
      "train_tokens_per_second": 29246.258
    },
    {
      "epoch": 2.108410528806498,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8953948946347214e-05,
      "loss": 0.38920345306396487,
      "num_input_tokens_seen": 16948140032,
      "step": 59590,
      "train_runtime": 579496.0061,
      "train_tokens_per_second": 29246.345
    },
    {
      "epoch": 2.108764349070227,
      "grad_norm": 0.60546875,
      "learning_rate": 1.895258724360564e-05,
      "loss": 0.38628239631652833,
      "num_input_tokens_seen": 16950946176,
      "step": 59600,
      "train_runtime": 579589.8385,
      "train_tokens_per_second": 29246.452
    },
    {
      "epoch": 2.1091181693339554,
      "grad_norm": 0.625,
      "learning_rate": 1.8951225834307078e-05,
      "loss": 0.38569657802581786,
      "num_input_tokens_seen": 16953736832,
      "step": 59610,
      "train_runtime": 579682.637,
      "train_tokens_per_second": 29246.584
    },
    {
      "epoch": 2.1094719895976843,
      "grad_norm": 0.609375,
      "learning_rate": 1.8949864718346153e-05,
      "loss": 0.38288078308105467,
      "num_input_tokens_seen": 16956575488,
      "step": 59620,
      "train_runtime": 579776.5867,
      "train_tokens_per_second": 29246.741
    },
    {
      "epoch": 2.109825809861413,
      "grad_norm": 0.62109375,
      "learning_rate": 1.894850389561754e-05,
      "loss": 0.38653500080108644,
      "num_input_tokens_seen": 16959433536,
      "step": 59630,
      "train_runtime": 579873.3226,
      "train_tokens_per_second": 29246.79
    },
    {
      "epoch": 2.110179630125142,
      "grad_norm": 0.625,
      "learning_rate": 1.8947143366015962e-05,
      "loss": 0.3877106189727783,
      "num_input_tokens_seen": 16962239424,
      "step": 59640,
      "train_runtime": 579969.6623,
      "train_tokens_per_second": 29246.77
    },
    {
      "epoch": 2.1105334503888704,
      "grad_norm": 0.62109375,
      "learning_rate": 1.89457831294362e-05,
      "loss": 0.3871367692947388,
      "num_input_tokens_seen": 16965117632,
      "step": 59650,
      "train_runtime": 580069.7194,
      "train_tokens_per_second": 29246.687
    },
    {
      "epoch": 2.1108872706525994,
      "grad_norm": 0.640625,
      "learning_rate": 1.8944423185773093e-05,
      "loss": 0.3859116554260254,
      "num_input_tokens_seen": 16967958464,
      "step": 59660,
      "train_runtime": 580165.4415,
      "train_tokens_per_second": 29246.758
    },
    {
      "epoch": 2.111241090916328,
      "grad_norm": 0.6015625,
      "learning_rate": 1.894306353492153e-05,
      "loss": 0.3868103504180908,
      "num_input_tokens_seen": 16970785600,
      "step": 59670,
      "train_runtime": 580263.4264,
      "train_tokens_per_second": 29246.692
    },
    {
      "epoch": 2.111594911180057,
      "grad_norm": 0.625,
      "learning_rate": 1.8941704176776433e-05,
      "loss": 0.38725378513336184,
      "num_input_tokens_seen": 16973653568,
      "step": 59680,
      "train_runtime": 580362.9294,
      "train_tokens_per_second": 29246.619
    },
    {
      "epoch": 2.1119487314437855,
      "grad_norm": 0.59375,
      "learning_rate": 1.8940345111232806e-05,
      "loss": 0.38458704948425293,
      "num_input_tokens_seen": 16976514752,
      "step": 59690,
      "train_runtime": 580459.6854,
      "train_tokens_per_second": 29246.673
    },
    {
      "epoch": 2.1123025517075145,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8938986338185695e-05,
      "loss": 0.3891266107559204,
      "num_input_tokens_seen": 16979358528,
      "step": 59700,
      "train_runtime": 580559.4214,
      "train_tokens_per_second": 29246.547
    },
    {
      "epoch": 2.1126563719712435,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8937627857530196e-05,
      "loss": 0.38526318073272703,
      "num_input_tokens_seen": 16982169344,
      "step": 59710,
      "train_runtime": 580656.0782,
      "train_tokens_per_second": 29246.52
    },
    {
      "epoch": 2.113010192234972,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8936269669161455e-05,
      "loss": 0.38611767292022703,
      "num_input_tokens_seen": 16985047168,
      "step": 59720,
      "train_runtime": 580755.023,
      "train_tokens_per_second": 29246.492
    },
    {
      "epoch": 2.113364012498701,
      "grad_norm": 0.625,
      "learning_rate": 1.8934911772974672e-05,
      "loss": 0.38472647666931153,
      "num_input_tokens_seen": 16987827264,
      "step": 59730,
      "train_runtime": 580847.8788,
      "train_tokens_per_second": 29246.603
    },
    {
      "epoch": 2.1137178327624295,
      "grad_norm": 0.61328125,
      "learning_rate": 1.893355416886511e-05,
      "loss": 0.38174147605895997,
      "num_input_tokens_seen": 16990676416,
      "step": 59740,
      "train_runtime": 580945.9381,
      "train_tokens_per_second": 29246.571
    },
    {
      "epoch": 2.1140716530261585,
      "grad_norm": 0.6328125,
      "learning_rate": 1.893219685672807e-05,
      "loss": 0.38504838943481445,
      "num_input_tokens_seen": 16993544704,
      "step": 59750,
      "train_runtime": 581041.7241,
      "train_tokens_per_second": 29246.686
    },
    {
      "epoch": 2.114425473289887,
      "grad_norm": 0.625,
      "learning_rate": 1.893083983645892e-05,
      "loss": 0.3856698751449585,
      "num_input_tokens_seen": 16996417024,
      "step": 59760,
      "train_runtime": 581137.3846,
      "train_tokens_per_second": 29246.814
    },
    {
      "epoch": 2.114779293553616,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8929483107953064e-05,
      "loss": 0.383039665222168,
      "num_input_tokens_seen": 16999229504,
      "step": 59770,
      "train_runtime": 581233.0405,
      "train_tokens_per_second": 29246.84
    },
    {
      "epoch": 2.1151331138173446,
      "grad_norm": 0.6328125,
      "learning_rate": 1.892812667110597e-05,
      "loss": 0.38240551948547363,
      "num_input_tokens_seen": 17002028224,
      "step": 59780,
      "train_runtime": 581329.3576,
      "train_tokens_per_second": 29246.808
    },
    {
      "epoch": 2.1154869340810736,
      "grad_norm": 0.640625,
      "learning_rate": 1.8926770525813152e-05,
      "loss": 0.3890200614929199,
      "num_input_tokens_seen": 17004890368,
      "step": 59790,
      "train_runtime": 581430.2598,
      "train_tokens_per_second": 29246.655
    },
    {
      "epoch": 2.115840754344802,
      "grad_norm": 0.60546875,
      "learning_rate": 1.892541467197019e-05,
      "loss": 0.386551570892334,
      "num_input_tokens_seen": 17007705856,
      "step": 59800,
      "train_runtime": 581526.3966,
      "train_tokens_per_second": 29246.662
    },
    {
      "epoch": 2.116194574608531,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8924059109472695e-05,
      "loss": 0.38725121021270753,
      "num_input_tokens_seen": 17010587200,
      "step": 59810,
      "train_runtime": 581629.7743,
      "train_tokens_per_second": 29246.417
    },
    {
      "epoch": 2.1165483948722597,
      "grad_norm": 0.59765625,
      "learning_rate": 1.892270383821635e-05,
      "loss": 0.3862725257873535,
      "num_input_tokens_seen": 17013437824,
      "step": 59820,
      "train_runtime": 581725.8185,
      "train_tokens_per_second": 29246.489
    },
    {
      "epoch": 2.1169022151359886,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8921348858096877e-05,
      "loss": 0.3889531135559082,
      "num_input_tokens_seen": 17016261568,
      "step": 59830,
      "train_runtime": 581822.5161,
      "train_tokens_per_second": 29246.482
    },
    {
      "epoch": 2.117256035399717,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8919994169010055e-05,
      "loss": 0.386976432800293,
      "num_input_tokens_seen": 17019096064,
      "step": 59840,
      "train_runtime": 581917.9929,
      "train_tokens_per_second": 29246.554
    },
    {
      "epoch": 2.117609855663446,
      "grad_norm": 0.6171875,
      "learning_rate": 1.891863977085172e-05,
      "loss": 0.38829855918884276,
      "num_input_tokens_seen": 17021946176,
      "step": 59850,
      "train_runtime": 582012.4419,
      "train_tokens_per_second": 29246.705
    },
    {
      "epoch": 2.117963675927175,
      "grad_norm": 0.59375,
      "learning_rate": 1.8917285663517752e-05,
      "loss": 0.38342986106872556,
      "num_input_tokens_seen": 17024808960,
      "step": 59860,
      "train_runtime": 582108.9995,
      "train_tokens_per_second": 29246.772
    },
    {
      "epoch": 2.1183174961909037,
      "grad_norm": 0.625,
      "learning_rate": 1.8915931846904088e-05,
      "loss": 0.3851174831390381,
      "num_input_tokens_seen": 17027647168,
      "step": 59870,
      "train_runtime": 582202.9793,
      "train_tokens_per_second": 29246.926
    },
    {
      "epoch": 2.1186713164546327,
      "grad_norm": 0.609375,
      "learning_rate": 1.8914578320906713e-05,
      "loss": 0.38332548141479494,
      "num_input_tokens_seen": 17030481664,
      "step": 59880,
      "train_runtime": 582300.0427,
      "train_tokens_per_second": 29246.918
    },
    {
      "epoch": 2.1190251367183612,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8913225085421666e-05,
      "loss": 0.3865786552429199,
      "num_input_tokens_seen": 17033372480,
      "step": 59890,
      "train_runtime": 582402.4061,
      "train_tokens_per_second": 29246.741
    },
    {
      "epoch": 2.1193789569820902,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8911872140345043e-05,
      "loss": 0.3820863485336304,
      "num_input_tokens_seen": 17036237568,
      "step": 59900,
      "train_runtime": 582501.3795,
      "train_tokens_per_second": 29246.691
    },
    {
      "epoch": 2.1197327772458188,
      "grad_norm": 0.5703125,
      "learning_rate": 1.891051948557299e-05,
      "loss": 0.3850423574447632,
      "num_input_tokens_seen": 17039136448,
      "step": 59910,
      "train_runtime": 582603.0652,
      "train_tokens_per_second": 29246.562
    },
    {
      "epoch": 2.1200865975095478,
      "grad_norm": 0.609375,
      "learning_rate": 1.8909167121001694e-05,
      "loss": 0.3856400966644287,
      "num_input_tokens_seen": 17042093504,
      "step": 59920,
      "train_runtime": 582710.6242,
      "train_tokens_per_second": 29246.238
    },
    {
      "epoch": 2.1204404177732763,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8907815046527415e-05,
      "loss": 0.3830983638763428,
      "num_input_tokens_seen": 17044966144,
      "step": 59930,
      "train_runtime": 582807.5559,
      "train_tokens_per_second": 29246.303
    },
    {
      "epoch": 2.1207942380370053,
      "grad_norm": 0.609375,
      "learning_rate": 1.8906463262046445e-05,
      "loss": 0.3820266485214233,
      "num_input_tokens_seen": 17047717120,
      "step": 59940,
      "train_runtime": 582898.8403,
      "train_tokens_per_second": 29246.442
    },
    {
      "epoch": 2.121148058300734,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8905111767455133e-05,
      "loss": 0.387908411026001,
      "num_input_tokens_seen": 17050561792,
      "step": 59950,
      "train_runtime": 582996.2624,
      "train_tokens_per_second": 29246.434
    },
    {
      "epoch": 2.121501878564463,
      "grad_norm": 0.59765625,
      "learning_rate": 1.890376056264989e-05,
      "loss": 0.38885958194732667,
      "num_input_tokens_seen": 17053431360,
      "step": 59960,
      "train_runtime": 583093.5354,
      "train_tokens_per_second": 29246.476
    },
    {
      "epoch": 2.1218556988281914,
      "grad_norm": 0.5859375,
      "learning_rate": 1.890240964752717e-05,
      "loss": 0.39104886054992677,
      "num_input_tokens_seen": 17056203904,
      "step": 59970,
      "train_runtime": 583186.1891,
      "train_tokens_per_second": 29246.584
    },
    {
      "epoch": 2.1222095190919203,
      "grad_norm": 0.61328125,
      "learning_rate": 1.890105902198348e-05,
      "loss": 0.38394579887390134,
      "num_input_tokens_seen": 17059085696,
      "step": 59980,
      "train_runtime": 583284.3902,
      "train_tokens_per_second": 29246.601
    },
    {
      "epoch": 2.122563339355649,
      "grad_norm": 0.609375,
      "learning_rate": 1.889970868591537e-05,
      "loss": 0.3869048833847046,
      "num_input_tokens_seen": 17061870464,
      "step": 59990,
      "train_runtime": 583378.6621,
      "train_tokens_per_second": 29246.648
    },
    {
      "epoch": 2.122917159619378,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8898358639219462e-05,
      "loss": 0.3859964370727539,
      "num_input_tokens_seen": 17064681856,
      "step": 60000,
      "train_runtime": 583473.2994,
      "train_tokens_per_second": 29246.723
    },
    {
      "epoch": 2.122917159619378,
      "eval_loss": 0.33690783381462097,
      "eval_runtime": 8.4142,
      "eval_samples_per_second": 473.963,
      "eval_steps_per_second": 3.803,
      "num_input_tokens_seen": 17064681856,
      "step": 60000
    },
    {
      "epoch": 2.1232709798831064,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8897008881792415e-05,
      "loss": 0.38256158828735354,
      "num_input_tokens_seen": 17067488128,
      "step": 60010,
      "train_runtime": 583600.0222,
      "train_tokens_per_second": 29245.181
    },
    {
      "epoch": 2.1236248001468354,
      "grad_norm": 0.609375,
      "learning_rate": 1.889565941353094e-05,
      "loss": 0.3864023685455322,
      "num_input_tokens_seen": 17070313344,
      "step": 60020,
      "train_runtime": 583695.977,
      "train_tokens_per_second": 29245.213
    },
    {
      "epoch": 2.1239786204105644,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8894310234331808e-05,
      "loss": 0.39201419353485106,
      "num_input_tokens_seen": 17073132480,
      "step": 60030,
      "train_runtime": 583791.5338,
      "train_tokens_per_second": 29245.255
    },
    {
      "epoch": 2.124332440674293,
      "grad_norm": 0.6171875,
      "learning_rate": 1.889296134409183e-05,
      "loss": 0.3851889371871948,
      "num_input_tokens_seen": 17075946112,
      "step": 60040,
      "train_runtime": 583887.5879,
      "train_tokens_per_second": 29245.263
    },
    {
      "epoch": 2.124686260938022,
      "grad_norm": 0.625,
      "learning_rate": 1.8891612742707884e-05,
      "loss": 0.38455679416656496,
      "num_input_tokens_seen": 17078730368,
      "step": 60050,
      "train_runtime": 583982.5151,
      "train_tokens_per_second": 29245.277
    },
    {
      "epoch": 2.1250400812017505,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8890264430076882e-05,
      "loss": 0.38625550270080566,
      "num_input_tokens_seen": 17081572864,
      "step": 60060,
      "train_runtime": 584078.9958,
      "train_tokens_per_second": 29245.313
    },
    {
      "epoch": 2.1253939014654795,
      "grad_norm": 0.625,
      "learning_rate": 1.8888916406095796e-05,
      "loss": 0.38981564044952394,
      "num_input_tokens_seen": 17084360576,
      "step": 60070,
      "train_runtime": 584171.4315,
      "train_tokens_per_second": 29245.457
    },
    {
      "epoch": 2.125747721729208,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8887568670661654e-05,
      "loss": 0.38290517330169677,
      "num_input_tokens_seen": 17087122496,
      "step": 60080,
      "train_runtime": 584261.6864,
      "train_tokens_per_second": 29245.667
    },
    {
      "epoch": 2.126101541992937,
      "grad_norm": 0.62109375,
      "learning_rate": 1.888622122367153e-05,
      "loss": 0.385133171081543,
      "num_input_tokens_seen": 17089951680,
      "step": 60090,
      "train_runtime": 584359.6854,
      "train_tokens_per_second": 29245.604
    },
    {
      "epoch": 2.1264553622566655,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8884874065022547e-05,
      "loss": 0.38584825992584226,
      "num_input_tokens_seen": 17092807680,
      "step": 60100,
      "train_runtime": 584459.718,
      "train_tokens_per_second": 29245.485
    },
    {
      "epoch": 2.1268091825203945,
      "grad_norm": 0.6328125,
      "learning_rate": 1.888352719461188e-05,
      "loss": 0.388007664680481,
      "num_input_tokens_seen": 17095697664,
      "step": 60110,
      "train_runtime": 584561.7226,
      "train_tokens_per_second": 29245.325
    },
    {
      "epoch": 2.127163002784123,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8882180612336762e-05,
      "loss": 0.39150466918945315,
      "num_input_tokens_seen": 17098565440,
      "step": 60120,
      "train_runtime": 584660.2332,
      "train_tokens_per_second": 29245.303
    },
    {
      "epoch": 2.127516823047852,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8880834318094482e-05,
      "loss": 0.39028732776641845,
      "num_input_tokens_seen": 17101441408,
      "step": 60130,
      "train_runtime": 584756.8495,
      "train_tokens_per_second": 29245.389
    },
    {
      "epoch": 2.1278706433115806,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8879488311782353e-05,
      "loss": 0.38725595474243163,
      "num_input_tokens_seen": 17104317184,
      "step": 60140,
      "train_runtime": 584853.798,
      "train_tokens_per_second": 29245.458
    },
    {
      "epoch": 2.1282244635753096,
      "grad_norm": 0.59765625,
      "learning_rate": 1.887814259329777e-05,
      "loss": 0.3906563758850098,
      "num_input_tokens_seen": 17107165504,
      "step": 60150,
      "train_runtime": 584950.7248,
      "train_tokens_per_second": 29245.481
    },
    {
      "epoch": 2.128578283839038,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8876797162538166e-05,
      "loss": 0.38708858489990233,
      "num_input_tokens_seen": 17110032832,
      "step": 60160,
      "train_runtime": 585048.7171,
      "train_tokens_per_second": 29245.484
    },
    {
      "epoch": 2.128932104102767,
      "grad_norm": 0.6015625,
      "learning_rate": 1.887545201940102e-05,
      "loss": 0.38337829113006594,
      "num_input_tokens_seen": 17112893568,
      "step": 60170,
      "train_runtime": 585145.1411,
      "train_tokens_per_second": 29245.554
    },
    {
      "epoch": 2.1292859243664957,
      "grad_norm": 0.62890625,
      "learning_rate": 1.887410716378387e-05,
      "loss": 0.3826786994934082,
      "num_input_tokens_seen": 17115758336,
      "step": 60180,
      "train_runtime": 585244.1365,
      "train_tokens_per_second": 29245.502
    },
    {
      "epoch": 2.1296397446302247,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8872762595584308e-05,
      "loss": 0.3893911838531494,
      "num_input_tokens_seen": 17118615232,
      "step": 60190,
      "train_runtime": 585342.9324,
      "train_tokens_per_second": 29245.446
    },
    {
      "epoch": 2.1299935648939536,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8871418314699966e-05,
      "loss": 0.3834666728973389,
      "num_input_tokens_seen": 17121483968,
      "step": 60200,
      "train_runtime": 585441.7933,
      "train_tokens_per_second": 29245.408
    },
    {
      "epoch": 2.130347385157682,
      "grad_norm": 0.578125,
      "learning_rate": 1.887007432102854e-05,
      "loss": 0.39141502380371096,
      "num_input_tokens_seen": 17124333696,
      "step": 60210,
      "train_runtime": 585539.2879,
      "train_tokens_per_second": 29245.405
    },
    {
      "epoch": 2.130701205421411,
      "grad_norm": 0.609375,
      "learning_rate": 1.8868730614467763e-05,
      "loss": 0.3861417770385742,
      "num_input_tokens_seen": 17127250240,
      "step": 60220,
      "train_runtime": 585637.6815,
      "train_tokens_per_second": 29245.472
    },
    {
      "epoch": 2.1310550256851397,
      "grad_norm": 0.62109375,
      "learning_rate": 1.886738719491543e-05,
      "loss": 0.3890763521194458,
      "num_input_tokens_seen": 17130068096,
      "step": 60230,
      "train_runtime": 585735.0049,
      "train_tokens_per_second": 29245.423
    },
    {
      "epoch": 2.1314088459488687,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8866044062269382e-05,
      "loss": 0.3886754035949707,
      "num_input_tokens_seen": 17132911360,
      "step": 60240,
      "train_runtime": 585832.1138,
      "train_tokens_per_second": 29245.429
    },
    {
      "epoch": 2.1317626662125972,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8864701216427515e-05,
      "loss": 0.38402643203735354,
      "num_input_tokens_seen": 17135699264,
      "step": 60250,
      "train_runtime": 585924.8573,
      "train_tokens_per_second": 29245.558
    },
    {
      "epoch": 2.1321164864763262,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8863358657287767e-05,
      "loss": 0.3857043504714966,
      "num_input_tokens_seen": 17138549440,
      "step": 60260,
      "train_runtime": 586022.2117,
      "train_tokens_per_second": 29245.563
    },
    {
      "epoch": 2.1324703067400548,
      "grad_norm": 0.6171875,
      "learning_rate": 1.886201638474814e-05,
      "loss": 0.38955609798431395,
      "num_input_tokens_seen": 17141375680,
      "step": 60270,
      "train_runtime": 586118.7647,
      "train_tokens_per_second": 29245.567
    },
    {
      "epoch": 2.1328241270037838,
      "grad_norm": 0.625,
      "learning_rate": 1.886067439870667e-05,
      "loss": 0.38744549751281737,
      "num_input_tokens_seen": 17144211968,
      "step": 60280,
      "train_runtime": 586216.7559,
      "train_tokens_per_second": 29245.517
    },
    {
      "epoch": 2.1331779472675123,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8859332699061463e-05,
      "loss": 0.38938026428222655,
      "num_input_tokens_seen": 17147082240,
      "step": 60290,
      "train_runtime": 586313.5677,
      "train_tokens_per_second": 29245.583
    },
    {
      "epoch": 2.1335317675312413,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8857991285710668e-05,
      "loss": 0.3856241703033447,
      "num_input_tokens_seen": 17149895104,
      "step": 60300,
      "train_runtime": 586410.0609,
      "train_tokens_per_second": 29245.568
    },
    {
      "epoch": 2.13388558779497,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8856650158552467e-05,
      "loss": 0.38742842674255373,
      "num_input_tokens_seen": 17152704384,
      "step": 60310,
      "train_runtime": 586504.1008,
      "train_tokens_per_second": 29245.668
    },
    {
      "epoch": 2.134239408058699,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8855309317485123e-05,
      "loss": 0.38872356414794923,
      "num_input_tokens_seen": 17155585408,
      "step": 60320,
      "train_runtime": 586603.804,
      "train_tokens_per_second": 29245.609
    },
    {
      "epoch": 2.1345932283224274,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8853968762406928e-05,
      "loss": 0.38028364181518554,
      "num_input_tokens_seen": 17158439360,
      "step": 60330,
      "train_runtime": 586700.1198,
      "train_tokens_per_second": 29245.672
    },
    {
      "epoch": 2.1349470485861564,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8852628493216236e-05,
      "loss": 0.3856541156768799,
      "num_input_tokens_seen": 17161269120,
      "step": 60340,
      "train_runtime": 586796.539,
      "train_tokens_per_second": 29245.689
    },
    {
      "epoch": 2.1353008688498853,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8851288509811446e-05,
      "loss": 0.3877675533294678,
      "num_input_tokens_seen": 17164128000,
      "step": 60350,
      "train_runtime": 586893.7037,
      "train_tokens_per_second": 29245.718
    },
    {
      "epoch": 2.135654689113614,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8849948812091007e-05,
      "loss": 0.38853323459625244,
      "num_input_tokens_seen": 17166980032,
      "step": 60360,
      "train_runtime": 586991.0606,
      "train_tokens_per_second": 29245.727
    },
    {
      "epoch": 2.136008509377343,
      "grad_norm": 0.6015625,
      "learning_rate": 1.884860939995342e-05,
      "loss": 0.38944363594055176,
      "num_input_tokens_seen": 17169832896,
      "step": 60370,
      "train_runtime": 587088.5613,
      "train_tokens_per_second": 29245.729
    },
    {
      "epoch": 2.1363623296410714,
      "grad_norm": 0.59375,
      "learning_rate": 1.884727027329724e-05,
      "loss": 0.39067137241363525,
      "num_input_tokens_seen": 17172664512,
      "step": 60380,
      "train_runtime": 587184.6077,
      "train_tokens_per_second": 29245.767
    },
    {
      "epoch": 2.1367161499048004,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8845931432021067e-05,
      "loss": 0.3879371404647827,
      "num_input_tokens_seen": 17175466560,
      "step": 60390,
      "train_runtime": 587277.8665,
      "train_tokens_per_second": 29245.895
    },
    {
      "epoch": 2.137069970168529,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8844592876023555e-05,
      "loss": 0.3840583086013794,
      "num_input_tokens_seen": 17178302656,
      "step": 60400,
      "train_runtime": 587374.6415,
      "train_tokens_per_second": 29245.904
    },
    {
      "epoch": 2.137423790432258,
      "grad_norm": 0.625,
      "learning_rate": 1.88432546052034e-05,
      "loss": 0.38749213218688966,
      "num_input_tokens_seen": 17181206528,
      "step": 60410,
      "train_runtime": 587473.9305,
      "train_tokens_per_second": 29245.905
    },
    {
      "epoch": 2.1377776106959865,
      "grad_norm": 0.64453125,
      "learning_rate": 1.884191661945937e-05,
      "loss": 0.38306283950805664,
      "num_input_tokens_seen": 17184034752,
      "step": 60420,
      "train_runtime": 587568.2646,
      "train_tokens_per_second": 29246.023
    },
    {
      "epoch": 2.1381314309597155,
      "grad_norm": 0.625,
      "learning_rate": 1.884057891869025e-05,
      "loss": 0.38718523979187014,
      "num_input_tokens_seen": 17186925248,
      "step": 60430,
      "train_runtime": 587669.4673,
      "train_tokens_per_second": 29245.905
    },
    {
      "epoch": 2.138485251223444,
      "grad_norm": 0.6015625,
      "learning_rate": 1.883924150279491e-05,
      "loss": 0.39154276847839353,
      "num_input_tokens_seen": 17189764032,
      "step": 60440,
      "train_runtime": 587764.1462,
      "train_tokens_per_second": 29246.024
    },
    {
      "epoch": 2.138839071487173,
      "grad_norm": 0.59375,
      "learning_rate": 1.8837904371672252e-05,
      "loss": 0.38042216300964354,
      "num_input_tokens_seen": 17192595904,
      "step": 60450,
      "train_runtime": 587861.0143,
      "train_tokens_per_second": 29246.022
    },
    {
      "epoch": 2.1391928917509015,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8836567525221222e-05,
      "loss": 0.38723273277282716,
      "num_input_tokens_seen": 17195496768,
      "step": 60460,
      "train_runtime": 587960.6276,
      "train_tokens_per_second": 29246.0
    },
    {
      "epoch": 2.1395467120146305,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8835230963340828e-05,
      "loss": 0.38415482044219973,
      "num_input_tokens_seen": 17198369216,
      "step": 60470,
      "train_runtime": 588059.2478,
      "train_tokens_per_second": 29245.98
    },
    {
      "epoch": 2.139900532278359,
      "grad_norm": 0.62109375,
      "learning_rate": 1.883389468593013e-05,
      "loss": 0.3883045673370361,
      "num_input_tokens_seen": 17201179520,
      "step": 60480,
      "train_runtime": 588153.1828,
      "train_tokens_per_second": 29246.088
    },
    {
      "epoch": 2.140254352542088,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8832558692888226e-05,
      "loss": 0.38452730178833006,
      "num_input_tokens_seen": 17204054720,
      "step": 60490,
      "train_runtime": 588253.6834,
      "train_tokens_per_second": 29245.979
    },
    {
      "epoch": 2.1406081728058166,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8831222984114276e-05,
      "loss": 0.38687431812286377,
      "num_input_tokens_seen": 17206879424,
      "step": 60500,
      "train_runtime": 588349.8404,
      "train_tokens_per_second": 29246.0
    },
    {
      "epoch": 2.1409619930695456,
      "grad_norm": 0.67578125,
      "learning_rate": 1.882988755950748e-05,
      "loss": 0.39170589447021487,
      "num_input_tokens_seen": 17209693952,
      "step": 60510,
      "train_runtime": 588444.9904,
      "train_tokens_per_second": 29246.054
    },
    {
      "epoch": 2.141315813333274,
      "grad_norm": 0.6015625,
      "learning_rate": 1.88285524189671e-05,
      "loss": 0.3879119396209717,
      "num_input_tokens_seen": 17212594368,
      "step": 60520,
      "train_runtime": 588551.4936,
      "train_tokens_per_second": 29245.69
    },
    {
      "epoch": 2.141669633597003,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8827217562392433e-05,
      "loss": 0.386278772354126,
      "num_input_tokens_seen": 17215391488,
      "step": 60530,
      "train_runtime": 588645.6942,
      "train_tokens_per_second": 29245.761
    },
    {
      "epoch": 2.142023453860732,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8825882989682843e-05,
      "loss": 0.3828026294708252,
      "num_input_tokens_seen": 17218271232,
      "step": 60540,
      "train_runtime": 588745.4907,
      "train_tokens_per_second": 29245.695
    },
    {
      "epoch": 2.1423772741244607,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8824548700737723e-05,
      "loss": 0.3884097099304199,
      "num_input_tokens_seen": 17221112832,
      "step": 60550,
      "train_runtime": 588840.9073,
      "train_tokens_per_second": 29245.782
    },
    {
      "epoch": 2.1427310943881896,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8823214695456538e-05,
      "loss": 0.3851633548736572,
      "num_input_tokens_seen": 17223947904,
      "step": 60560,
      "train_runtime": 588935.3099,
      "train_tokens_per_second": 29245.908
    },
    {
      "epoch": 2.143084914651918,
      "grad_norm": 0.59375,
      "learning_rate": 1.8821880973738792e-05,
      "loss": 0.3876689910888672,
      "num_input_tokens_seen": 17226748480,
      "step": 60570,
      "train_runtime": 589029.0583,
      "train_tokens_per_second": 29246.008
    },
    {
      "epoch": 2.143438734915647,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8820547535484033e-05,
      "loss": 0.38605761528015137,
      "num_input_tokens_seen": 17229556736,
      "step": 60580,
      "train_runtime": 589124.5894,
      "train_tokens_per_second": 29246.032
    },
    {
      "epoch": 2.1437925551793757,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8819214380591876e-05,
      "loss": 0.3882163047790527,
      "num_input_tokens_seen": 17232463680,
      "step": 60590,
      "train_runtime": 589224.7377,
      "train_tokens_per_second": 29245.995
    },
    {
      "epoch": 2.1441463754431047,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8817881508961965e-05,
      "loss": 0.3880929708480835,
      "num_input_tokens_seen": 17235324608,
      "step": 60600,
      "train_runtime": 589320.2934,
      "train_tokens_per_second": 29246.107
    },
    {
      "epoch": 2.1445001957068333,
      "grad_norm": 0.59375,
      "learning_rate": 1.881654892049401e-05,
      "loss": 0.38804771900177004,
      "num_input_tokens_seen": 17238223424,
      "step": 60610,
      "train_runtime": 589423.5462,
      "train_tokens_per_second": 29245.902
    },
    {
      "epoch": 2.1448540159705622,
      "grad_norm": 0.609375,
      "learning_rate": 1.881521661508776e-05,
      "loss": 0.3868117809295654,
      "num_input_tokens_seen": 17241020416,
      "step": 60620,
      "train_runtime": 589517.9086,
      "train_tokens_per_second": 29245.965
    },
    {
      "epoch": 2.145207836234291,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8813884592643023e-05,
      "loss": 0.38586370944976806,
      "num_input_tokens_seen": 17243888512,
      "step": 60630,
      "train_runtime": 589619.9387,
      "train_tokens_per_second": 29245.769
    },
    {
      "epoch": 2.1455616564980198,
      "grad_norm": 0.625,
      "learning_rate": 1.8812552853059646e-05,
      "loss": 0.39091432094573975,
      "num_input_tokens_seen": 17246744000,
      "step": 60640,
      "train_runtime": 589715.6254,
      "train_tokens_per_second": 29245.866
    },
    {
      "epoch": 2.1459154767617483,
      "grad_norm": 0.6328125,
      "learning_rate": 1.881122139623754e-05,
      "loss": 0.3884439468383789,
      "num_input_tokens_seen": 17249580224,
      "step": 60650,
      "train_runtime": 589814.1781,
      "train_tokens_per_second": 29245.788
    },
    {
      "epoch": 2.1462692970254773,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8809890222076647e-05,
      "loss": 0.3859309911727905,
      "num_input_tokens_seen": 17252459328,
      "step": 60660,
      "train_runtime": 589911.6224,
      "train_tokens_per_second": 29245.837
    },
    {
      "epoch": 2.146623117289206,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8808559330476978e-05,
      "loss": 0.3882889747619629,
      "num_input_tokens_seen": 17255289664,
      "step": 60670,
      "train_runtime": 590007.4005,
      "train_tokens_per_second": 29245.887
    },
    {
      "epoch": 2.146976937552935,
      "grad_norm": 0.62109375,
      "learning_rate": 1.880722872133858e-05,
      "loss": 0.3860327243804932,
      "num_input_tokens_seen": 17258201984,
      "step": 60680,
      "train_runtime": 590104.544,
      "train_tokens_per_second": 29246.008
    },
    {
      "epoch": 2.147330757816664,
      "grad_norm": 0.625,
      "learning_rate": 1.8805898394561554e-05,
      "loss": 0.3851497650146484,
      "num_input_tokens_seen": 17261048000,
      "step": 60690,
      "train_runtime": 590201.5445,
      "train_tokens_per_second": 29246.023
    },
    {
      "epoch": 2.1476845780803924,
      "grad_norm": 0.640625,
      "learning_rate": 1.8804568350046047e-05,
      "loss": 0.38547773361206056,
      "num_input_tokens_seen": 17263837120,
      "step": 60700,
      "train_runtime": 590296.4701,
      "train_tokens_per_second": 29246.045
    },
    {
      "epoch": 2.1480383983441214,
      "grad_norm": 0.64453125,
      "learning_rate": 1.8803238587692266e-05,
      "loss": 0.38608880043029786,
      "num_input_tokens_seen": 17266761024,
      "step": 60710,
      "train_runtime": 590400.1299,
      "train_tokens_per_second": 29245.863
    },
    {
      "epoch": 2.14839221860785,
      "grad_norm": 0.61328125,
      "learning_rate": 1.880190910740045e-05,
      "loss": 0.3900933265686035,
      "num_input_tokens_seen": 17269636992,
      "step": 60720,
      "train_runtime": 590499.8074,
      "train_tokens_per_second": 29245.796
    },
    {
      "epoch": 2.148746038871579,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8800579909070908e-05,
      "loss": 0.38424131870269773,
      "num_input_tokens_seen": 17272521472,
      "step": 60730,
      "train_runtime": 590598.9428,
      "train_tokens_per_second": 29245.771
    },
    {
      "epoch": 2.1490998591353074,
      "grad_norm": 0.62890625,
      "learning_rate": 1.879925099260398e-05,
      "loss": 0.38953866958618166,
      "num_input_tokens_seen": 17275307584,
      "step": 60740,
      "train_runtime": 590690.031,
      "train_tokens_per_second": 29245.978
    },
    {
      "epoch": 2.1494536793990364,
      "grad_norm": 0.6015625,
      "learning_rate": 1.879792235790007e-05,
      "loss": 0.3831746578216553,
      "num_input_tokens_seen": 17278189248,
      "step": 60750,
      "train_runtime": 590791.4273,
      "train_tokens_per_second": 29245.836
    },
    {
      "epoch": 2.149807499662765,
      "grad_norm": 0.58203125,
      "learning_rate": 1.8796594004859614e-05,
      "loss": 0.3884650707244873,
      "num_input_tokens_seen": 17281056064,
      "step": 60760,
      "train_runtime": 590888.5332,
      "train_tokens_per_second": 29245.882
    },
    {
      "epoch": 2.150161319926494,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8795265933383115e-05,
      "loss": 0.385513162612915,
      "num_input_tokens_seen": 17283865152,
      "step": 60770,
      "train_runtime": 590985.5877,
      "train_tokens_per_second": 29245.832
    },
    {
      "epoch": 2.1505151401902225,
      "grad_norm": 0.609375,
      "learning_rate": 1.8793938143371118e-05,
      "loss": 0.3835740566253662,
      "num_input_tokens_seen": 17286665344,
      "step": 60780,
      "train_runtime": 591079.0888,
      "train_tokens_per_second": 29245.943
    },
    {
      "epoch": 2.1508689604539515,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8792610634724215e-05,
      "loss": 0.3861441135406494,
      "num_input_tokens_seen": 17289489664,
      "step": 60790,
      "train_runtime": 591174.4455,
      "train_tokens_per_second": 29246.003
    },
    {
      "epoch": 2.15122278071768,
      "grad_norm": 0.609375,
      "learning_rate": 1.8791283407343048e-05,
      "loss": 0.38686556816101075,
      "num_input_tokens_seen": 17292283776,
      "step": 60800,
      "train_runtime": 591268.8176,
      "train_tokens_per_second": 29246.061
    },
    {
      "epoch": 2.151576600981409,
      "grad_norm": 0.6484375,
      "learning_rate": 1.8789956461128312e-05,
      "loss": 0.3895624876022339,
      "num_input_tokens_seen": 17295177216,
      "step": 60810,
      "train_runtime": 591367.7868,
      "train_tokens_per_second": 29246.059
    },
    {
      "epoch": 2.1519304212451376,
      "grad_norm": 0.61328125,
      "learning_rate": 1.878862979598074e-05,
      "loss": 0.38757944107055664,
      "num_input_tokens_seen": 17297993216,
      "step": 60820,
      "train_runtime": 591463.4299,
      "train_tokens_per_second": 29246.091
    },
    {
      "epoch": 2.1522842415088665,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8787303411801134e-05,
      "loss": 0.387520956993103,
      "num_input_tokens_seen": 17300816320,
      "step": 60830,
      "train_runtime": 591557.4557,
      "train_tokens_per_second": 29246.215
    },
    {
      "epoch": 2.152638061772595,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8785977308490328e-05,
      "loss": 0.3854555368423462,
      "num_input_tokens_seen": 17303713856,
      "step": 60840,
      "train_runtime": 591654.5406,
      "train_tokens_per_second": 29246.313
    },
    {
      "epoch": 2.152991882036324,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8784651485949205e-05,
      "loss": 0.38765678405761717,
      "num_input_tokens_seen": 17306572992,
      "step": 60850,
      "train_runtime": 591752.2406,
      "train_tokens_per_second": 29246.316
    },
    {
      "epoch": 2.1533457023000526,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8783325944078713e-05,
      "loss": 0.3828439712524414,
      "num_input_tokens_seen": 17309422016,
      "step": 60860,
      "train_runtime": 591849.3837,
      "train_tokens_per_second": 29246.329
    },
    {
      "epoch": 2.1536995225637816,
      "grad_norm": 0.61328125,
      "learning_rate": 1.878200068277983e-05,
      "loss": 0.38779449462890625,
      "num_input_tokens_seen": 17312235456,
      "step": 60870,
      "train_runtime": 591944.913,
      "train_tokens_per_second": 29246.362
    },
    {
      "epoch": 2.1540533428275106,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8780675701953594e-05,
      "loss": 0.38764224052429197,
      "num_input_tokens_seen": 17315097280,
      "step": 60880,
      "train_runtime": 592042.8286,
      "train_tokens_per_second": 29246.359
    },
    {
      "epoch": 2.154407163091239,
      "grad_norm": 0.640625,
      "learning_rate": 1.877935100150109e-05,
      "loss": 0.38846454620361326,
      "num_input_tokens_seen": 17317914176,
      "step": 60890,
      "train_runtime": 592139.9772,
      "train_tokens_per_second": 29246.318
    },
    {
      "epoch": 2.154760983354968,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8778026581323445e-05,
      "loss": 0.388960599899292,
      "num_input_tokens_seen": 17320673664,
      "step": 60900,
      "train_runtime": 592232.7427,
      "train_tokens_per_second": 29246.397
    },
    {
      "epoch": 2.1551148036186967,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8776702441321843e-05,
      "loss": 0.38748199939727784,
      "num_input_tokens_seen": 17323522560,
      "step": 60910,
      "train_runtime": 592327.8702,
      "train_tokens_per_second": 29246.509
    },
    {
      "epoch": 2.1554686238824257,
      "grad_norm": 0.5859375,
      "learning_rate": 1.8775378581397523e-05,
      "loss": 0.3825246334075928,
      "num_input_tokens_seen": 17326356480,
      "step": 60920,
      "train_runtime": 592426.945,
      "train_tokens_per_second": 29246.402
    },
    {
      "epoch": 2.155822444146154,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8774055001451754e-05,
      "loss": 0.38863024711608884,
      "num_input_tokens_seen": 17329271552,
      "step": 60930,
      "train_runtime": 592528.7453,
      "train_tokens_per_second": 29246.297
    },
    {
      "epoch": 2.156176264409883,
      "grad_norm": 0.5859375,
      "learning_rate": 1.8772731701385865e-05,
      "loss": 0.38425326347351074,
      "num_input_tokens_seen": 17332025280,
      "step": 60940,
      "train_runtime": 592621.0365,
      "train_tokens_per_second": 29246.389
    },
    {
      "epoch": 2.1565300846736117,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8771408681101236e-05,
      "loss": 0.3854637384414673,
      "num_input_tokens_seen": 17334866112,
      "step": 60950,
      "train_runtime": 592718.0602,
      "train_tokens_per_second": 29246.394
    },
    {
      "epoch": 2.1568839049373407,
      "grad_norm": 0.64453125,
      "learning_rate": 1.877008594049929e-05,
      "loss": 0.3821453809738159,
      "num_input_tokens_seen": 17337678848,
      "step": 60960,
      "train_runtime": 592812.9728,
      "train_tokens_per_second": 29246.457
    },
    {
      "epoch": 2.1572377252010693,
      "grad_norm": 0.61328125,
      "learning_rate": 1.87687634794815e-05,
      "loss": 0.3865180730819702,
      "num_input_tokens_seen": 17340461952,
      "step": 60970,
      "train_runtime": 592907.6713,
      "train_tokens_per_second": 29246.479
    },
    {
      "epoch": 2.1575915454647983,
      "grad_norm": 0.61328125,
      "learning_rate": 1.876744129794939e-05,
      "loss": 0.38787617683410647,
      "num_input_tokens_seen": 17343278528,
      "step": 60980,
      "train_runtime": 593003.7045,
      "train_tokens_per_second": 29246.493
    },
    {
      "epoch": 2.157945365728527,
      "grad_norm": 0.59375,
      "learning_rate": 1.8766119395804525e-05,
      "loss": 0.3859707355499268,
      "num_input_tokens_seen": 17346139328,
      "step": 60990,
      "train_runtime": 593103.2686,
      "train_tokens_per_second": 29246.407
    },
    {
      "epoch": 2.158299185992256,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8764797772948537e-05,
      "loss": 0.3854742288589478,
      "num_input_tokens_seen": 17348953856,
      "step": 61000,
      "train_runtime": 593198.9228,
      "train_tokens_per_second": 29246.435
    },
    {
      "epoch": 2.1586530062559843,
      "grad_norm": 0.59375,
      "learning_rate": 1.8763476429283083e-05,
      "loss": 0.381731653213501,
      "num_input_tokens_seen": 17351814208,
      "step": 61010,
      "train_runtime": 593295.9978,
      "train_tokens_per_second": 29246.471
    },
    {
      "epoch": 2.1590068265197133,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8762155364709885e-05,
      "loss": 0.3908177614212036,
      "num_input_tokens_seen": 17354639808,
      "step": 61020,
      "train_runtime": 593393.2845,
      "train_tokens_per_second": 29246.438
    },
    {
      "epoch": 2.1593606467834423,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8760834579130705e-05,
      "loss": 0.3879721641540527,
      "num_input_tokens_seen": 17357488192,
      "step": 61030,
      "train_runtime": 593489.658,
      "train_tokens_per_second": 29246.488
    },
    {
      "epoch": 2.159714467047171,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8759514072447352e-05,
      "loss": 0.3863555908203125,
      "num_input_tokens_seen": 17360398464,
      "step": 61040,
      "train_runtime": 593587.9768,
      "train_tokens_per_second": 29246.547
    },
    {
      "epoch": 2.1600682873109,
      "grad_norm": 0.640625,
      "learning_rate": 1.8758193844561695e-05,
      "loss": 0.38376669883728026,
      "num_input_tokens_seen": 17363253888,
      "step": 61050,
      "train_runtime": 593687.6612,
      "train_tokens_per_second": 29246.446
    },
    {
      "epoch": 2.1604221075746284,
      "grad_norm": 0.6171875,
      "learning_rate": 1.875687389537564e-05,
      "loss": 0.3873939037322998,
      "num_input_tokens_seen": 17366058752,
      "step": 61060,
      "train_runtime": 593781.5523,
      "train_tokens_per_second": 29246.545
    },
    {
      "epoch": 2.1607759278383574,
      "grad_norm": 0.58203125,
      "learning_rate": 1.8755554224791144e-05,
      "loss": 0.3876068353652954,
      "num_input_tokens_seen": 17368965504,
      "step": 61070,
      "train_runtime": 593882.3837,
      "train_tokens_per_second": 29246.474
    },
    {
      "epoch": 2.161129748102086,
      "grad_norm": 0.609375,
      "learning_rate": 1.8754234832710217e-05,
      "loss": 0.38747658729553225,
      "num_input_tokens_seen": 17371842752,
      "step": 61080,
      "train_runtime": 593983.0066,
      "train_tokens_per_second": 29246.363
    },
    {
      "epoch": 2.161483568365815,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8752915719034914e-05,
      "loss": 0.38890810012817384,
      "num_input_tokens_seen": 17374629824,
      "step": 61090,
      "train_runtime": 594076.3849,
      "train_tokens_per_second": 29246.458
    },
    {
      "epoch": 2.1618373886295434,
      "grad_norm": 0.5859375,
      "learning_rate": 1.8751596883667337e-05,
      "loss": 0.3885769844055176,
      "num_input_tokens_seen": 17377418176,
      "step": 61100,
      "train_runtime": 594171.4768,
      "train_tokens_per_second": 29246.47
    },
    {
      "epoch": 2.1621912088932724,
      "grad_norm": 0.625,
      "learning_rate": 1.8750278326509628e-05,
      "loss": 0.3861593246459961,
      "num_input_tokens_seen": 17380196800,
      "step": 61110,
      "train_runtime": 594261.7217,
      "train_tokens_per_second": 29246.704
    },
    {
      "epoch": 2.162545029157001,
      "grad_norm": 0.59375,
      "learning_rate": 1.8748960047464e-05,
      "loss": 0.3824602127075195,
      "num_input_tokens_seen": 17383017792,
      "step": 61120,
      "train_runtime": 594356.8389,
      "train_tokens_per_second": 29246.77
    },
    {
      "epoch": 2.16289884942073,
      "grad_norm": 0.625,
      "learning_rate": 1.8747642046432698e-05,
      "loss": 0.38700094223022463,
      "num_input_tokens_seen": 17385819328,
      "step": 61130,
      "train_runtime": 594448.7053,
      "train_tokens_per_second": 29246.963
    },
    {
      "epoch": 2.1632526696844585,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8746324323318008e-05,
      "loss": 0.3900285720825195,
      "num_input_tokens_seen": 17388649984,
      "step": 61140,
      "train_runtime": 594544.2303,
      "train_tokens_per_second": 29247.025
    },
    {
      "epoch": 2.1636064899481875,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8745006878022285e-05,
      "loss": 0.3876815319061279,
      "num_input_tokens_seen": 17391516032,
      "step": 61150,
      "train_runtime": 594641.7453,
      "train_tokens_per_second": 29247.049
    },
    {
      "epoch": 2.163960310211916,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8743689710447914e-05,
      "loss": 0.3892263650894165,
      "num_input_tokens_seen": 17394293696,
      "step": 61160,
      "train_runtime": 594737.6579,
      "train_tokens_per_second": 29247.002
    },
    {
      "epoch": 2.164314130475645,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8742372820497335e-05,
      "loss": 0.3843223571777344,
      "num_input_tokens_seen": 17397146880,
      "step": 61170,
      "train_runtime": 594835.393,
      "train_tokens_per_second": 29246.994
    },
    {
      "epoch": 2.164667950739374,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8741056208073036e-05,
      "loss": 0.3839869976043701,
      "num_input_tokens_seen": 17400026752,
      "step": 61180,
      "train_runtime": 594934.8474,
      "train_tokens_per_second": 29246.945
    },
    {
      "epoch": 2.1650217710031026,
      "grad_norm": 0.59375,
      "learning_rate": 1.873973987307755e-05,
      "loss": 0.3889696359634399,
      "num_input_tokens_seen": 17402869504,
      "step": 61190,
      "train_runtime": 595032.4639,
      "train_tokens_per_second": 29246.924
    },
    {
      "epoch": 2.1653755912668315,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8738423815413468e-05,
      "loss": 0.3881132364273071,
      "num_input_tokens_seen": 17405652992,
      "step": 61200,
      "train_runtime": 595125.3724,
      "train_tokens_per_second": 29247.036
    },
    {
      "epoch": 2.16572941153056,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8737108034983415e-05,
      "loss": 0.39003844261169435,
      "num_input_tokens_seen": 17408465152,
      "step": 61210,
      "train_runtime": 595221.0184,
      "train_tokens_per_second": 29247.061
    },
    {
      "epoch": 2.166083231794289,
      "grad_norm": 0.5703125,
      "learning_rate": 1.873579253169007e-05,
      "loss": 0.38290090560913087,
      "num_input_tokens_seen": 17411338880,
      "step": 61220,
      "train_runtime": 595315.9282,
      "train_tokens_per_second": 29247.225
    },
    {
      "epoch": 2.1664370520580176,
      "grad_norm": 0.609375,
      "learning_rate": 1.8734477305436166e-05,
      "loss": 0.38256587982177737,
      "num_input_tokens_seen": 17414242496,
      "step": 61230,
      "train_runtime": 595416.7842,
      "train_tokens_per_second": 29247.147
    },
    {
      "epoch": 2.1667908723217466,
      "grad_norm": 0.6171875,
      "learning_rate": 1.873316235612447e-05,
      "loss": 0.3900241136550903,
      "num_input_tokens_seen": 17417080704,
      "step": 61240,
      "train_runtime": 595514.7058,
      "train_tokens_per_second": 29247.104
    },
    {
      "epoch": 2.167144692585475,
      "grad_norm": 0.60546875,
      "learning_rate": 1.873184768365781e-05,
      "loss": 0.38803949356079104,
      "num_input_tokens_seen": 17419943680,
      "step": 61250,
      "train_runtime": 595615.3844,
      "train_tokens_per_second": 29246.967
    },
    {
      "epoch": 2.167498512849204,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8730533287939052e-05,
      "loss": 0.386526894569397,
      "num_input_tokens_seen": 17422755584,
      "step": 61260,
      "train_runtime": 595709.132,
      "train_tokens_per_second": 29247.085
    },
    {
      "epoch": 2.1678523331129327,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8729219168871118e-05,
      "loss": 0.3834963083267212,
      "num_input_tokens_seen": 17425571776,
      "step": 61270,
      "train_runtime": 595805.0437,
      "train_tokens_per_second": 29247.103
    },
    {
      "epoch": 2.1682061533766617,
      "grad_norm": 0.609375,
      "learning_rate": 1.872790532635697e-05,
      "loss": 0.3876005172729492,
      "num_input_tokens_seen": 17428384256,
      "step": 61280,
      "train_runtime": 595901.7086,
      "train_tokens_per_second": 29247.079
    },
    {
      "epoch": 2.16855997364039,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8726591760299626e-05,
      "loss": 0.38833298683166506,
      "num_input_tokens_seen": 17431207488,
      "step": 61290,
      "train_runtime": 595995.4931,
      "train_tokens_per_second": 29247.214
    },
    {
      "epoch": 2.168913793904119,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8725278470602142e-05,
      "loss": 0.3874513149261475,
      "num_input_tokens_seen": 17434045888,
      "step": 61300,
      "train_runtime": 596093.0135,
      "train_tokens_per_second": 29247.19
    },
    {
      "epoch": 2.1692676141678477,
      "grad_norm": 0.625,
      "learning_rate": 1.872396545716763e-05,
      "loss": 0.3876046895980835,
      "num_input_tokens_seen": 17436891264,
      "step": 61310,
      "train_runtime": 596190.2179,
      "train_tokens_per_second": 29247.195
    },
    {
      "epoch": 2.1696214344315767,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8722652719899246e-05,
      "loss": 0.38645663261413576,
      "num_input_tokens_seen": 17439702144,
      "step": 61320,
      "train_runtime": 596286.3965,
      "train_tokens_per_second": 29247.191
    },
    {
      "epoch": 2.1699752546953053,
      "grad_norm": 0.59765625,
      "learning_rate": 1.872134025870019e-05,
      "loss": 0.3838057518005371,
      "num_input_tokens_seen": 17442579904,
      "step": 61330,
      "train_runtime": 596385.6629,
      "train_tokens_per_second": 29247.148
    },
    {
      "epoch": 2.1703290749590343,
      "grad_norm": 0.59375,
      "learning_rate": 1.872002807347371e-05,
      "loss": 0.38198373317718504,
      "num_input_tokens_seen": 17445395392,
      "step": 61340,
      "train_runtime": 596480.0234,
      "train_tokens_per_second": 29247.242
    },
    {
      "epoch": 2.170682895222763,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8718716164123114e-05,
      "loss": 0.3841330289840698,
      "num_input_tokens_seen": 17448275072,
      "step": 61350,
      "train_runtime": 596580.2192,
      "train_tokens_per_second": 29247.157
    },
    {
      "epoch": 2.171036715486492,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8717404530551747e-05,
      "loss": 0.3877728939056396,
      "num_input_tokens_seen": 17451117248,
      "step": 61360,
      "train_runtime": 596678.0557,
      "train_tokens_per_second": 29247.124
    },
    {
      "epoch": 2.171390535750221,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8716093172662993e-05,
      "loss": 0.3879808187484741,
      "num_input_tokens_seen": 17453973312,
      "step": 61370,
      "train_runtime": 596776.3798,
      "train_tokens_per_second": 29247.091
    },
    {
      "epoch": 2.1717443560139493,
      "grad_norm": 0.60546875,
      "learning_rate": 1.87147820903603e-05,
      "loss": 0.38223164081573485,
      "num_input_tokens_seen": 17456835520,
      "step": 61380,
      "train_runtime": 596876.917,
      "train_tokens_per_second": 29246.96
    },
    {
      "epoch": 2.1720981762776783,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8713471283547156e-05,
      "loss": 0.3881186008453369,
      "num_input_tokens_seen": 17459705216,
      "step": 61390,
      "train_runtime": 596977.6474,
      "train_tokens_per_second": 29246.832
    },
    {
      "epoch": 2.172451996541407,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8712160752127094e-05,
      "loss": 0.38828539848327637,
      "num_input_tokens_seen": 17462594880,
      "step": 61400,
      "train_runtime": 597077.7358,
      "train_tokens_per_second": 29246.769
    },
    {
      "epoch": 2.172805816805136,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8710850496003697e-05,
      "loss": 0.38840174674987793,
      "num_input_tokens_seen": 17465466816,
      "step": 61410,
      "train_runtime": 597177.5508,
      "train_tokens_per_second": 29246.69
    },
    {
      "epoch": 2.1731596370688644,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8709540515080597e-05,
      "loss": 0.38725717067718507,
      "num_input_tokens_seen": 17468312512,
      "step": 61420,
      "train_runtime": 597275.2683,
      "train_tokens_per_second": 29246.67
    },
    {
      "epoch": 2.1735134573325934,
      "grad_norm": 0.609375,
      "learning_rate": 1.8708230809261465e-05,
      "loss": 0.3879450082778931,
      "num_input_tokens_seen": 17471146880,
      "step": 61430,
      "train_runtime": 597370.3696,
      "train_tokens_per_second": 29246.758
    },
    {
      "epoch": 2.173867277596322,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8706921378450034e-05,
      "loss": 0.3891200065612793,
      "num_input_tokens_seen": 17473970816,
      "step": 61440,
      "train_runtime": 597470.9696,
      "train_tokens_per_second": 29246.56
    },
    {
      "epoch": 2.174221097860051,
      "grad_norm": 0.6328125,
      "learning_rate": 1.870561222255007e-05,
      "loss": 0.38735768795013426,
      "num_input_tokens_seen": 17476795328,
      "step": 61450,
      "train_runtime": 597566.1268,
      "train_tokens_per_second": 29246.63
    },
    {
      "epoch": 2.1745749181237795,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8704303341465393e-05,
      "loss": 0.3866576194763184,
      "num_input_tokens_seen": 17479608960,
      "step": 61460,
      "train_runtime": 597660.7614,
      "train_tokens_per_second": 29246.707
    },
    {
      "epoch": 2.1749287383875084,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8702994735099868e-05,
      "loss": 0.3878880500793457,
      "num_input_tokens_seen": 17482478592,
      "step": 61470,
      "train_runtime": 597756.386,
      "train_tokens_per_second": 29246.829
    },
    {
      "epoch": 2.175282558651237,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8701686403357407e-05,
      "loss": 0.3896933078765869,
      "num_input_tokens_seen": 17485279424,
      "step": 61480,
      "train_runtime": 597852.4748,
      "train_tokens_per_second": 29246.813
    },
    {
      "epoch": 2.175636378914966,
      "grad_norm": 0.609375,
      "learning_rate": 1.8700378346141973e-05,
      "loss": 0.3861856460571289,
      "num_input_tokens_seen": 17488142464,
      "step": 61490,
      "train_runtime": 597951.6453,
      "train_tokens_per_second": 29246.75
    },
    {
      "epoch": 2.1759901991786945,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8699070563357574e-05,
      "loss": 0.38948023319244385,
      "num_input_tokens_seen": 17490942848,
      "step": 61500,
      "train_runtime": 598046.2986,
      "train_tokens_per_second": 29246.804
    },
    {
      "epoch": 2.1763440194424235,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8697763054908255e-05,
      "loss": 0.3894660472869873,
      "num_input_tokens_seen": 17493765120,
      "step": 61510,
      "train_runtime": 598143.1164,
      "train_tokens_per_second": 29246.788
    },
    {
      "epoch": 2.1766978397061525,
      "grad_norm": 0.65234375,
      "learning_rate": 1.8696455820698124e-05,
      "loss": 0.38852667808532715,
      "num_input_tokens_seen": 17496527616,
      "step": 61520,
      "train_runtime": 598236.8218,
      "train_tokens_per_second": 29246.825
    },
    {
      "epoch": 2.177051659969881,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8695148860631327e-05,
      "loss": 0.3853722095489502,
      "num_input_tokens_seen": 17499360960,
      "step": 61530,
      "train_runtime": 598331.1446,
      "train_tokens_per_second": 29246.95
    },
    {
      "epoch": 2.17740548023361,
      "grad_norm": 0.62109375,
      "learning_rate": 1.869384217461206e-05,
      "loss": 0.3896056652069092,
      "num_input_tokens_seen": 17502238720,
      "step": 61540,
      "train_runtime": 598433.1918,
      "train_tokens_per_second": 29246.771
    },
    {
      "epoch": 2.1777593004973386,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8692535762544565e-05,
      "loss": 0.3850872993469238,
      "num_input_tokens_seen": 17505065216,
      "step": 61550,
      "train_runtime": 598531.1551,
      "train_tokens_per_second": 29246.707
    },
    {
      "epoch": 2.1781131207610676,
      "grad_norm": 0.609375,
      "learning_rate": 1.8691229624333126e-05,
      "loss": 0.38930888175964357,
      "num_input_tokens_seen": 17507923456,
      "step": 61560,
      "train_runtime": 598630.1993,
      "train_tokens_per_second": 29246.643
    },
    {
      "epoch": 2.178466941024796,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8689923759882078e-05,
      "loss": 0.3874336004257202,
      "num_input_tokens_seen": 17510774400,
      "step": 61570,
      "train_runtime": 598725.3105,
      "train_tokens_per_second": 29246.758
    },
    {
      "epoch": 2.178820761288525,
      "grad_norm": 0.60546875,
      "learning_rate": 1.868861816909581e-05,
      "loss": 0.38097991943359377,
      "num_input_tokens_seen": 17513635840,
      "step": 61580,
      "train_runtime": 598821.0504,
      "train_tokens_per_second": 29246.861
    },
    {
      "epoch": 2.1791745815522536,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8687312851878747e-05,
      "loss": 0.38545732498168944,
      "num_input_tokens_seen": 17516477248,
      "step": 61590,
      "train_runtime": 598916.7381,
      "train_tokens_per_second": 29246.932
    },
    {
      "epoch": 2.1795284018159826,
      "grad_norm": 0.6171875,
      "learning_rate": 1.868600780813536e-05,
      "loss": 0.38284945487976074,
      "num_input_tokens_seen": 17519313344,
      "step": 61600,
      "train_runtime": 599011.0879,
      "train_tokens_per_second": 29247.06
    },
    {
      "epoch": 2.179882222079711,
      "grad_norm": 0.5625,
      "learning_rate": 1.8684703037770177e-05,
      "loss": 0.3869945049285889,
      "num_input_tokens_seen": 17522141632,
      "step": 61610,
      "train_runtime": 599107.534,
      "train_tokens_per_second": 29247.073
    },
    {
      "epoch": 2.18023604234344,
      "grad_norm": 0.60546875,
      "learning_rate": 1.868339854068776e-05,
      "loss": 0.38443565368652344,
      "num_input_tokens_seen": 17525007360,
      "step": 61620,
      "train_runtime": 599206.5513,
      "train_tokens_per_second": 29247.022
    },
    {
      "epoch": 2.1805898626071687,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8682094316792737e-05,
      "loss": 0.38622493743896485,
      "num_input_tokens_seen": 17527868992,
      "step": 61630,
      "train_runtime": 599304.6675,
      "train_tokens_per_second": 29247.009
    },
    {
      "epoch": 2.1809436828708977,
      "grad_norm": 0.62109375,
      "learning_rate": 1.868079036598975e-05,
      "loss": 0.3877041578292847,
      "num_input_tokens_seen": 17530742272,
      "step": 61640,
      "train_runtime": 599405.3695,
      "train_tokens_per_second": 29246.889
    },
    {
      "epoch": 2.1812975031346262,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8679486688183527e-05,
      "loss": 0.39104065895080564,
      "num_input_tokens_seen": 17533594048,
      "step": 61650,
      "train_runtime": 599503.877,
      "train_tokens_per_second": 29246.84
    },
    {
      "epoch": 2.181651323398355,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8678183283278816e-05,
      "loss": 0.38863177299499513,
      "num_input_tokens_seen": 17536464256,
      "step": 61660,
      "train_runtime": 599601.7379,
      "train_tokens_per_second": 29246.854
    },
    {
      "epoch": 2.1820051436620838,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8676880151180415e-05,
      "loss": 0.38354628086090087,
      "num_input_tokens_seen": 17539378816,
      "step": 61670,
      "train_runtime": 599701.7234,
      "train_tokens_per_second": 29246.837
    },
    {
      "epoch": 2.1823589639258127,
      "grad_norm": 0.609375,
      "learning_rate": 1.8675577291793172e-05,
      "loss": 0.38083949089050295,
      "num_input_tokens_seen": 17542225280,
      "step": 61680,
      "train_runtime": 599797.7277,
      "train_tokens_per_second": 29246.902
    },
    {
      "epoch": 2.1827127841895413,
      "grad_norm": 0.59375,
      "learning_rate": 1.867427470502199e-05,
      "loss": 0.3875278472900391,
      "num_input_tokens_seen": 17545053888,
      "step": 61690,
      "train_runtime": 599894.651,
      "train_tokens_per_second": 29246.892
    },
    {
      "epoch": 2.1830666044532703,
      "grad_norm": 0.609375,
      "learning_rate": 1.86729723907718e-05,
      "loss": 0.3849807262420654,
      "num_input_tokens_seen": 17547918080,
      "step": 61700,
      "train_runtime": 599992.0655,
      "train_tokens_per_second": 29246.917
    },
    {
      "epoch": 2.1834204247169993,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8671670348947597e-05,
      "loss": 0.3840256452560425,
      "num_input_tokens_seen": 17550791040,
      "step": 61710,
      "train_runtime": 600090.3697,
      "train_tokens_per_second": 29246.913
    },
    {
      "epoch": 2.183774244980728,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8670368579454405e-05,
      "loss": 0.38533930778503417,
      "num_input_tokens_seen": 17553607104,
      "step": 61720,
      "train_runtime": 600187.5026,
      "train_tokens_per_second": 29246.872
    },
    {
      "epoch": 2.184128065244457,
      "grad_norm": 0.609375,
      "learning_rate": 1.8669067082197318e-05,
      "loss": 0.3853305339813232,
      "num_input_tokens_seen": 17556536064,
      "step": 61730,
      "train_runtime": 600289.9262,
      "train_tokens_per_second": 29246.761
    },
    {
      "epoch": 2.1844818855081853,
      "grad_norm": 0.65234375,
      "learning_rate": 1.866776585708145e-05,
      "loss": 0.3883723258972168,
      "num_input_tokens_seen": 17559385856,
      "step": 61740,
      "train_runtime": 600383.7392,
      "train_tokens_per_second": 29246.938
    },
    {
      "epoch": 2.1848357057719143,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8666464904011983e-05,
      "loss": 0.3842418909072876,
      "num_input_tokens_seen": 17562246080,
      "step": 61750,
      "train_runtime": 600483.492,
      "train_tokens_per_second": 29246.842
    },
    {
      "epoch": 2.185189526035643,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8665164222894125e-05,
      "loss": 0.38751270771026614,
      "num_input_tokens_seen": 17565140800,
      "step": 61760,
      "train_runtime": 600582.504,
      "train_tokens_per_second": 29246.841
    },
    {
      "epoch": 2.185543346299372,
      "grad_norm": 0.625,
      "learning_rate": 1.8663863813633152e-05,
      "loss": 0.3861790895462036,
      "num_input_tokens_seen": 17567945856,
      "step": 61770,
      "train_runtime": 600676.0565,
      "train_tokens_per_second": 29246.955
    },
    {
      "epoch": 2.1858971665631004,
      "grad_norm": 0.625,
      "learning_rate": 1.866256367613437e-05,
      "loss": 0.38616693019866943,
      "num_input_tokens_seen": 17570781184,
      "step": 61780,
      "train_runtime": 600773.0557,
      "train_tokens_per_second": 29246.953
    },
    {
      "epoch": 2.1862509868268294,
      "grad_norm": 0.609375,
      "learning_rate": 1.8661263810303138e-05,
      "loss": 0.38463492393493653,
      "num_input_tokens_seen": 17573646208,
      "step": 61790,
      "train_runtime": 600870.4557,
      "train_tokens_per_second": 29246.98
    },
    {
      "epoch": 2.186604807090558,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8659964216044853e-05,
      "loss": 0.3852818012237549,
      "num_input_tokens_seen": 17576526592,
      "step": 61800,
      "train_runtime": 600969.5101,
      "train_tokens_per_second": 29246.952
    },
    {
      "epoch": 2.186958627354287,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8658664893264978e-05,
      "loss": 0.390532922744751,
      "num_input_tokens_seen": 17579403136,
      "step": 61810,
      "train_runtime": 601071.3939,
      "train_tokens_per_second": 29246.781
    },
    {
      "epoch": 2.1873124476180155,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8657365841868997e-05,
      "loss": 0.3843573570251465,
      "num_input_tokens_seen": 17582272640,
      "step": 61820,
      "train_runtime": 601168.5734,
      "train_tokens_per_second": 29246.826
    },
    {
      "epoch": 2.1876662678817445,
      "grad_norm": 0.60546875,
      "learning_rate": 1.865606706176246e-05,
      "loss": 0.388136625289917,
      "num_input_tokens_seen": 17585155648,
      "step": 61830,
      "train_runtime": 601267.5946,
      "train_tokens_per_second": 29246.804
    },
    {
      "epoch": 2.188020088145473,
      "grad_norm": 0.609375,
      "learning_rate": 1.8654768552850943e-05,
      "loss": 0.3837641716003418,
      "num_input_tokens_seen": 17588005056,
      "step": 61840,
      "train_runtime": 601363.2962,
      "train_tokens_per_second": 29246.888
    },
    {
      "epoch": 2.188373908409202,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8653470315040096e-05,
      "loss": 0.38420701026916504,
      "num_input_tokens_seen": 17590807744,
      "step": 61850,
      "train_runtime": 601457.8784,
      "train_tokens_per_second": 29246.949
    },
    {
      "epoch": 2.188727728672931,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8652172348235588e-05,
      "loss": 0.38457636833190917,
      "num_input_tokens_seen": 17593702784,
      "step": 61860,
      "train_runtime": 601555.5593,
      "train_tokens_per_second": 29247.012
    },
    {
      "epoch": 2.1890815489366595,
      "grad_norm": 0.64453125,
      "learning_rate": 1.8650874652343146e-05,
      "loss": 0.38947651386260984,
      "num_input_tokens_seen": 17596551360,
      "step": 61870,
      "train_runtime": 601651.8834,
      "train_tokens_per_second": 29247.064
    },
    {
      "epoch": 2.1894353692003885,
      "grad_norm": 0.62109375,
      "learning_rate": 1.864957722726855e-05,
      "loss": 0.3834806442260742,
      "num_input_tokens_seen": 17599421888,
      "step": 61880,
      "train_runtime": 601751.389,
      "train_tokens_per_second": 29246.998
    },
    {
      "epoch": 2.189789189464117,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8648280072917605e-05,
      "loss": 0.3866006851196289,
      "num_input_tokens_seen": 17602278592,
      "step": 61890,
      "train_runtime": 601846.6763,
      "train_tokens_per_second": 29247.114
    },
    {
      "epoch": 2.190143009727846,
      "grad_norm": 0.61328125,
      "learning_rate": 1.864698318919618e-05,
      "loss": 0.38844497203826905,
      "num_input_tokens_seen": 17605110528,
      "step": 61900,
      "train_runtime": 601942.0069,
      "train_tokens_per_second": 29247.187
    },
    {
      "epoch": 2.1904968299915746,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8645686576010185e-05,
      "loss": 0.386997127532959,
      "num_input_tokens_seen": 17607923200,
      "step": 61910,
      "train_runtime": 602040.4114,
      "train_tokens_per_second": 29247.079
    },
    {
      "epoch": 2.1908506502553036,
      "grad_norm": 0.625,
      "learning_rate": 1.8644390233265576e-05,
      "loss": 0.3856661319732666,
      "num_input_tokens_seen": 17610732992,
      "step": 61920,
      "train_runtime": 602135.3385,
      "train_tokens_per_second": 29247.134
    },
    {
      "epoch": 2.191204470519032,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8643094160868353e-05,
      "loss": 0.3890425682067871,
      "num_input_tokens_seen": 17613603008,
      "step": 61930,
      "train_runtime": 602232.1773,
      "train_tokens_per_second": 29247.197
    },
    {
      "epoch": 2.191558290782761,
      "grad_norm": 0.59375,
      "learning_rate": 1.864179835872456e-05,
      "loss": 0.3832890033721924,
      "num_input_tokens_seen": 17616494016,
      "step": 61940,
      "train_runtime": 602332.8183,
      "train_tokens_per_second": 29247.11
    },
    {
      "epoch": 2.1919121110464896,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8640502826740292e-05,
      "loss": 0.3883743047714233,
      "num_input_tokens_seen": 17619335488,
      "step": 61950,
      "train_runtime": 602429.0226,
      "train_tokens_per_second": 29247.156
    },
    {
      "epoch": 2.1922659313102186,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8639207564821686e-05,
      "loss": 0.38341598510742186,
      "num_input_tokens_seen": 17622172864,
      "step": 61960,
      "train_runtime": 602524.6057,
      "train_tokens_per_second": 29247.225
    },
    {
      "epoch": 2.192619751573947,
      "grad_norm": 0.609375,
      "learning_rate": 1.8637912572874924e-05,
      "loss": 0.3869826316833496,
      "num_input_tokens_seen": 17625011712,
      "step": 61970,
      "train_runtime": 602619.6093,
      "train_tokens_per_second": 29247.325
    },
    {
      "epoch": 2.192973571837676,
      "grad_norm": 0.625,
      "learning_rate": 1.863661785080624e-05,
      "loss": 0.3922722339630127,
      "num_input_tokens_seen": 17627814336,
      "step": 61980,
      "train_runtime": 602715.2134,
      "train_tokens_per_second": 29247.336
    },
    {
      "epoch": 2.1933273921014047,
      "grad_norm": 0.63671875,
      "learning_rate": 1.86353233985219e-05,
      "loss": 0.3880609035491943,
      "num_input_tokens_seen": 17630649280,
      "step": 61990,
      "train_runtime": 602813.6206,
      "train_tokens_per_second": 29247.264
    },
    {
      "epoch": 2.1936812123651337,
      "grad_norm": 0.609375,
      "learning_rate": 1.8634029215928235e-05,
      "loss": 0.38416998386383056,
      "num_input_tokens_seen": 17633497792,
      "step": 62000,
      "train_runtime": 602910.2793,
      "train_tokens_per_second": 29247.3
    },
    {
      "epoch": 2.1940350326288627,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8632735302931604e-05,
      "loss": 0.38510818481445314,
      "num_input_tokens_seen": 17636320640,
      "step": 62010,
      "train_runtime": 603006.1169,
      "train_tokens_per_second": 29247.333
    },
    {
      "epoch": 2.1943888528925912,
      "grad_norm": 0.64453125,
      "learning_rate": 1.8631441659438417e-05,
      "loss": 0.3807917356491089,
      "num_input_tokens_seen": 17639196672,
      "step": 62020,
      "train_runtime": 603102.8197,
      "train_tokens_per_second": 29247.412
    },
    {
      "epoch": 2.19474267315632,
      "grad_norm": 0.61328125,
      "learning_rate": 1.863014828535514e-05,
      "loss": 0.38839921951293943,
      "num_input_tokens_seen": 17642081216,
      "step": 62030,
      "train_runtime": 603200.2926,
      "train_tokens_per_second": 29247.468
    },
    {
      "epoch": 2.1950964934200488,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8628855180588262e-05,
      "loss": 0.38147544860839844,
      "num_input_tokens_seen": 17644976064,
      "step": 62040,
      "train_runtime": 603299.4058,
      "train_tokens_per_second": 29247.461
    },
    {
      "epoch": 2.1954503136837777,
      "grad_norm": 0.6328125,
      "learning_rate": 1.862756234504434e-05,
      "loss": 0.3902407646179199,
      "num_input_tokens_seen": 17647833216,
      "step": 62050,
      "train_runtime": 603395.9347,
      "train_tokens_per_second": 29247.518
    },
    {
      "epoch": 2.1958041339475063,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8626269778629965e-05,
      "loss": 0.38625283241271974,
      "num_input_tokens_seen": 17650638208,
      "step": 62060,
      "train_runtime": 603490.1566,
      "train_tokens_per_second": 29247.599
    },
    {
      "epoch": 2.1961579542112353,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8624977481251777e-05,
      "loss": 0.3855902671813965,
      "num_input_tokens_seen": 17653478400,
      "step": 62070,
      "train_runtime": 603587.1114,
      "train_tokens_per_second": 29247.607
    },
    {
      "epoch": 2.196511774474964,
      "grad_norm": 0.5859375,
      "learning_rate": 1.862368545281646e-05,
      "loss": 0.3816222667694092,
      "num_input_tokens_seen": 17656320896,
      "step": 62080,
      "train_runtime": 603682.9316,
      "train_tokens_per_second": 29247.673
    },
    {
      "epoch": 2.196865594738693,
      "grad_norm": 0.609375,
      "learning_rate": 1.8622393693230738e-05,
      "loss": 0.3865756273269653,
      "num_input_tokens_seen": 17659203712,
      "step": 62090,
      "train_runtime": 603783.9889,
      "train_tokens_per_second": 29247.552
    },
    {
      "epoch": 2.1972194150024213,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8621102202401388e-05,
      "loss": 0.3871957302093506,
      "num_input_tokens_seen": 17662076288,
      "step": 62100,
      "train_runtime": 603882.0943,
      "train_tokens_per_second": 29247.558
    },
    {
      "epoch": 2.1975732352661503,
      "grad_norm": 0.609375,
      "learning_rate": 1.8619810980235236e-05,
      "loss": 0.38620591163635254,
      "num_input_tokens_seen": 17664887296,
      "step": 62110,
      "train_runtime": 603973.8743,
      "train_tokens_per_second": 29247.767
    },
    {
      "epoch": 2.197927055529879,
      "grad_norm": 0.64453125,
      "learning_rate": 1.861852002663913e-05,
      "loss": 0.3887690544128418,
      "num_input_tokens_seen": 17667683200,
      "step": 62120,
      "train_runtime": 604067.617,
      "train_tokens_per_second": 29247.857
    },
    {
      "epoch": 2.198280875793608,
      "grad_norm": 0.59375,
      "learning_rate": 1.8617229341520003e-05,
      "loss": 0.38804523944854735,
      "num_input_tokens_seen": 17670567744,
      "step": 62130,
      "train_runtime": 604164.1993,
      "train_tokens_per_second": 29247.956
    },
    {
      "epoch": 2.1986346960573364,
      "grad_norm": 0.58984375,
      "learning_rate": 1.861593892478479e-05,
      "loss": 0.3862619400024414,
      "num_input_tokens_seen": 17673351808,
      "step": 62140,
      "train_runtime": 604254.694,
      "train_tokens_per_second": 29248.183
    },
    {
      "epoch": 2.1989885163210654,
      "grad_norm": 0.64453125,
      "learning_rate": 1.861464877634051e-05,
      "loss": 0.38628759384155276,
      "num_input_tokens_seen": 17676174976,
      "step": 62150,
      "train_runtime": 604351.8363,
      "train_tokens_per_second": 29248.153
    },
    {
      "epoch": 2.199342336584794,
      "grad_norm": 0.609375,
      "learning_rate": 1.861335889609419e-05,
      "loss": 0.3862605571746826,
      "num_input_tokens_seen": 17679012096,
      "step": 62160,
      "train_runtime": 604450.0359,
      "train_tokens_per_second": 29248.095
    },
    {
      "epoch": 2.199696156848523,
      "grad_norm": 0.61328125,
      "learning_rate": 1.861206928395293e-05,
      "loss": 0.38453800678253175,
      "num_input_tokens_seen": 17681857024,
      "step": 62170,
      "train_runtime": 604548.7779,
      "train_tokens_per_second": 29248.024
    },
    {
      "epoch": 2.2000499771122515,
      "grad_norm": 0.609375,
      "learning_rate": 1.8610779939823866e-05,
      "loss": 0.3866212606430054,
      "num_input_tokens_seen": 17684691456,
      "step": 62180,
      "train_runtime": 604648.1554,
      "train_tokens_per_second": 29247.904
    },
    {
      "epoch": 2.2004037973759805,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8609490863614178e-05,
      "loss": 0.39065597057342527,
      "num_input_tokens_seen": 17687516544,
      "step": 62190,
      "train_runtime": 604743.1645,
      "train_tokens_per_second": 29247.981
    },
    {
      "epoch": 2.2007576176397095,
      "grad_norm": 0.61328125,
      "learning_rate": 1.860820205523109e-05,
      "loss": 0.3849517583847046,
      "num_input_tokens_seen": 17690320320,
      "step": 62200,
      "train_runtime": 604837.9408,
      "train_tokens_per_second": 29248.033
    },
    {
      "epoch": 2.201111437903438,
      "grad_norm": 0.61328125,
      "learning_rate": 1.860691351458187e-05,
      "loss": 0.39128837585449217,
      "num_input_tokens_seen": 17693185216,
      "step": 62210,
      "train_runtime": 604937.7764,
      "train_tokens_per_second": 29247.942
    },
    {
      "epoch": 2.201465258167167,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8605625241573843e-05,
      "loss": 0.3887389421463013,
      "num_input_tokens_seen": 17696017216,
      "step": 62220,
      "train_runtime": 605034.9224,
      "train_tokens_per_second": 29247.927
    },
    {
      "epoch": 2.2018190784308955,
      "grad_norm": 0.625,
      "learning_rate": 1.860433723611436e-05,
      "loss": 0.38534297943115237,
      "num_input_tokens_seen": 17698888960,
      "step": 62230,
      "train_runtime": 605131.8595,
      "train_tokens_per_second": 29247.987
    },
    {
      "epoch": 2.2021728986946245,
      "grad_norm": 0.62109375,
      "learning_rate": 1.860304949811083e-05,
      "loss": 0.37967557907104493,
      "num_input_tokens_seen": 17701728960,
      "step": 62240,
      "train_runtime": 605232.2659,
      "train_tokens_per_second": 29247.828
    },
    {
      "epoch": 2.202526718958353,
      "grad_norm": 0.64453125,
      "learning_rate": 1.8601762027470703e-05,
      "loss": 0.3878924369812012,
      "num_input_tokens_seen": 17704541120,
      "step": 62250,
      "train_runtime": 605329.6438,
      "train_tokens_per_second": 29247.768
    },
    {
      "epoch": 2.202880539222082,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8600474824101473e-05,
      "loss": 0.38958141803741453,
      "num_input_tokens_seen": 17707415232,
      "step": 62260,
      "train_runtime": 605428.2997,
      "train_tokens_per_second": 29247.749
    },
    {
      "epoch": 2.2032343594858106,
      "grad_norm": 0.6484375,
      "learning_rate": 1.8599187887910684e-05,
      "loss": 0.38509092330932615,
      "num_input_tokens_seen": 17710243136,
      "step": 62270,
      "train_runtime": 605525.9695,
      "train_tokens_per_second": 29247.702
    },
    {
      "epoch": 2.2035881797495396,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8597901218805913e-05,
      "loss": 0.392694878578186,
      "num_input_tokens_seen": 17713073152,
      "step": 62280,
      "train_runtime": 605621.957,
      "train_tokens_per_second": 29247.739
    },
    {
      "epoch": 2.203942000013268,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8596614816694798e-05,
      "loss": 0.38522379398345946,
      "num_input_tokens_seen": 17715918400,
      "step": 62290,
      "train_runtime": 605718.6219,
      "train_tokens_per_second": 29247.769
    },
    {
      "epoch": 2.204295820276997,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8595328681485004e-05,
      "loss": 0.38733806610107424,
      "num_input_tokens_seen": 17718743104,
      "step": 62300,
      "train_runtime": 605815.5704,
      "train_tokens_per_second": 29247.751
    },
    {
      "epoch": 2.2046496405407257,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8594042813084256e-05,
      "loss": 0.3868782758712769,
      "num_input_tokens_seen": 17721597248,
      "step": 62310,
      "train_runtime": 605915.173,
      "train_tokens_per_second": 29247.654
    },
    {
      "epoch": 2.2050034608044546,
      "grad_norm": 0.6171875,
      "learning_rate": 1.859275721140032e-05,
      "loss": 0.3843999862670898,
      "num_input_tokens_seen": 17724442304,
      "step": 62320,
      "train_runtime": 606012.1567,
      "train_tokens_per_second": 29247.668
    },
    {
      "epoch": 2.205357281068183,
      "grad_norm": 0.58203125,
      "learning_rate": 1.8591471876341e-05,
      "loss": 0.3873064279556274,
      "num_input_tokens_seen": 17727305152,
      "step": 62330,
      "train_runtime": 606107.481,
      "train_tokens_per_second": 29247.791
    },
    {
      "epoch": 2.205711101331912,
      "grad_norm": 0.59765625,
      "learning_rate": 1.859018680781414e-05,
      "loss": 0.38848042488098145,
      "num_input_tokens_seen": 17730153856,
      "step": 62340,
      "train_runtime": 606203.4464,
      "train_tokens_per_second": 29247.861
    },
    {
      "epoch": 2.206064921595641,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8588902005727657e-05,
      "loss": 0.3906503677368164,
      "num_input_tokens_seen": 17732989184,
      "step": 62350,
      "train_runtime": 606298.5508,
      "train_tokens_per_second": 29247.949
    },
    {
      "epoch": 2.2064187418593697,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8587617469989476e-05,
      "loss": 0.38793971538543703,
      "num_input_tokens_seen": 17735848512,
      "step": 62360,
      "train_runtime": 606398.6225,
      "train_tokens_per_second": 29247.838
    },
    {
      "epoch": 2.2067725621230987,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8586333200507588e-05,
      "loss": 0.3845386981964111,
      "num_input_tokens_seen": 17738637504,
      "step": 62370,
      "train_runtime": 606493.5293,
      "train_tokens_per_second": 29247.859
    },
    {
      "epoch": 2.2071263823868272,
      "grad_norm": 0.60546875,
      "learning_rate": 1.858504919719003e-05,
      "loss": 0.3884584903717041,
      "num_input_tokens_seen": 17741497344,
      "step": 62380,
      "train_runtime": 606591.6658,
      "train_tokens_per_second": 29247.842
    },
    {
      "epoch": 2.2074802026505562,
      "grad_norm": 0.6328125,
      "learning_rate": 1.858376545994487e-05,
      "loss": 0.3864442348480225,
      "num_input_tokens_seen": 17744365440,
      "step": 62390,
      "train_runtime": 606690.8469,
      "train_tokens_per_second": 29247.788
    },
    {
      "epoch": 2.2078340229142848,
      "grad_norm": 0.62109375,
      "learning_rate": 1.858248198868023e-05,
      "loss": 0.3872101068496704,
      "num_input_tokens_seen": 17747151168,
      "step": 62400,
      "train_runtime": 606785.3896,
      "train_tokens_per_second": 29247.822
    },
    {
      "epoch": 2.2081878431780138,
      "grad_norm": 0.59375,
      "learning_rate": 1.8581198783304274e-05,
      "loss": 0.38182380199432375,
      "num_input_tokens_seen": 17749970624,
      "step": 62410,
      "train_runtime": 606880.3499,
      "train_tokens_per_second": 29247.891
    },
    {
      "epoch": 2.2085416634417423,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8579915843725208e-05,
      "loss": 0.3883842468261719,
      "num_input_tokens_seen": 17752786432,
      "step": 62420,
      "train_runtime": 606978.4378,
      "train_tokens_per_second": 29247.804
    },
    {
      "epoch": 2.2088954837054713,
      "grad_norm": 0.6015625,
      "learning_rate": 1.857863316985129e-05,
      "loss": 0.3855446815490723,
      "num_input_tokens_seen": 17755608000,
      "step": 62430,
      "train_runtime": 607071.7665,
      "train_tokens_per_second": 29247.955
    },
    {
      "epoch": 2.2092493039692,
      "grad_norm": 0.59765625,
      "learning_rate": 1.857735076159082e-05,
      "loss": 0.3832533359527588,
      "num_input_tokens_seen": 17758458496,
      "step": 62440,
      "train_runtime": 607170.0545,
      "train_tokens_per_second": 29247.916
    },
    {
      "epoch": 2.209603124232929,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8576068618852123e-05,
      "loss": 0.3876873254776001,
      "num_input_tokens_seen": 17761298176,
      "step": 62450,
      "train_runtime": 607268.5017,
      "train_tokens_per_second": 29247.85
    },
    {
      "epoch": 2.2099569444966574,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8574786741543605e-05,
      "loss": 0.38634462356567384,
      "num_input_tokens_seen": 17764157248,
      "step": 62460,
      "train_runtime": 607367.0004,
      "train_tokens_per_second": 29247.814
    },
    {
      "epoch": 2.2103107647603863,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8573505129573684e-05,
      "loss": 0.38374159336090086,
      "num_input_tokens_seen": 17766932928,
      "step": 62470,
      "train_runtime": 607459.2752,
      "train_tokens_per_second": 29247.941
    },
    {
      "epoch": 2.210664585024115,
      "grad_norm": 0.61328125,
      "learning_rate": 1.857222378285084e-05,
      "loss": 0.3854056835174561,
      "num_input_tokens_seen": 17769756160,
      "step": 62480,
      "train_runtime": 607554.1835,
      "train_tokens_per_second": 29248.019
    },
    {
      "epoch": 2.211018405287844,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8570942701283585e-05,
      "loss": 0.38499021530151367,
      "num_input_tokens_seen": 17772630336,
      "step": 62490,
      "train_runtime": 607652.1745,
      "train_tokens_per_second": 29248.032
    },
    {
      "epoch": 2.2113722255515724,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8569661884780485e-05,
      "loss": 0.38618247509002684,
      "num_input_tokens_seen": 17775517248,
      "step": 62500,
      "train_runtime": 607752.4337,
      "train_tokens_per_second": 29247.957
    },
    {
      "epoch": 2.2117260458153014,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8568381333250152e-05,
      "loss": 0.3884061098098755,
      "num_input_tokens_seen": 17778377984,
      "step": 62510,
      "train_runtime": 607849.1252,
      "train_tokens_per_second": 29248.011
    },
    {
      "epoch": 2.21207986607903,
      "grad_norm": 0.625,
      "learning_rate": 1.856710104660123e-05,
      "loss": 0.38503437042236327,
      "num_input_tokens_seen": 17781199488,
      "step": 62520,
      "train_runtime": 607946.0832,
      "train_tokens_per_second": 29247.988
    },
    {
      "epoch": 2.212433686342759,
      "grad_norm": 0.625,
      "learning_rate": 1.8565821024742414e-05,
      "loss": 0.39023537635803224,
      "num_input_tokens_seen": 17784016896,
      "step": 62530,
      "train_runtime": 608041.5571,
      "train_tokens_per_second": 29248.029
    },
    {
      "epoch": 2.212787506606488,
      "grad_norm": 0.62890625,
      "learning_rate": 1.856454126758245e-05,
      "loss": 0.3851016521453857,
      "num_input_tokens_seen": 17786886592,
      "step": 62540,
      "train_runtime": 608140.767,
      "train_tokens_per_second": 29247.976
    },
    {
      "epoch": 2.2131413268702165,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8563261775030108e-05,
      "loss": 0.3857923984527588,
      "num_input_tokens_seen": 17789750720,
      "step": 62550,
      "train_runtime": 608239.8742,
      "train_tokens_per_second": 29247.919
    },
    {
      "epoch": 2.2134951471339455,
      "grad_norm": 0.578125,
      "learning_rate": 1.856198254699423e-05,
      "loss": 0.38740196228027346,
      "num_input_tokens_seen": 17792588096,
      "step": 62560,
      "train_runtime": 608338.613,
      "train_tokens_per_second": 29247.836
    },
    {
      "epoch": 2.213848967397674,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8560703583383675e-05,
      "loss": 0.38474035263061523,
      "num_input_tokens_seen": 17795441536,
      "step": 62570,
      "train_runtime": 608434.9886,
      "train_tokens_per_second": 29247.893
    },
    {
      "epoch": 2.214202787661403,
      "grad_norm": 0.62109375,
      "learning_rate": 1.855942488410737e-05,
      "loss": 0.38535232543945314,
      "num_input_tokens_seen": 17798296512,
      "step": 62580,
      "train_runtime": 608532.7341,
      "train_tokens_per_second": 29247.887
    },
    {
      "epoch": 2.2145566079251315,
      "grad_norm": 0.61328125,
      "learning_rate": 1.855814644907426e-05,
      "loss": 0.3838515281677246,
      "num_input_tokens_seen": 17801200640,
      "step": 62590,
      "train_runtime": 608634.6649,
      "train_tokens_per_second": 29247.76
    },
    {
      "epoch": 2.2149104281888605,
      "grad_norm": 0.625,
      "learning_rate": 1.855686827819336e-05,
      "loss": 0.3851050138473511,
      "num_input_tokens_seen": 17804031872,
      "step": 62600,
      "train_runtime": 608731.584,
      "train_tokens_per_second": 29247.754
    },
    {
      "epoch": 2.215264248452589,
      "grad_norm": 0.61328125,
      "learning_rate": 1.855559037137371e-05,
      "loss": 0.3896501064300537,
      "num_input_tokens_seen": 17806885504,
      "step": 62610,
      "train_runtime": 608828.1516,
      "train_tokens_per_second": 29247.802
    },
    {
      "epoch": 2.215618068716318,
      "grad_norm": 0.64453125,
      "learning_rate": 1.85543127285244e-05,
      "loss": 0.38946242332458497,
      "num_input_tokens_seen": 17809684032,
      "step": 62620,
      "train_runtime": 608924.3744,
      "train_tokens_per_second": 29247.777
    },
    {
      "epoch": 2.2159718889800466,
      "grad_norm": 0.609375,
      "learning_rate": 1.8553035349554565e-05,
      "loss": 0.38803634643554685,
      "num_input_tokens_seen": 17812522176,
      "step": 62630,
      "train_runtime": 609019.8604,
      "train_tokens_per_second": 29247.851
    },
    {
      "epoch": 2.2163257092437756,
      "grad_norm": 0.6328125,
      "learning_rate": 1.855175823437339e-05,
      "loss": 0.38423309326171873,
      "num_input_tokens_seen": 17815373440,
      "step": 62640,
      "train_runtime": 609117.5216,
      "train_tokens_per_second": 29247.843
    },
    {
      "epoch": 2.216679529507504,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8550481382890086e-05,
      "loss": 0.38359885215759276,
      "num_input_tokens_seen": 17818197824,
      "step": 62650,
      "train_runtime": 609211.9786,
      "train_tokens_per_second": 29247.944
    },
    {
      "epoch": 2.217033349771233,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8549204795013926e-05,
      "loss": 0.389111852645874,
      "num_input_tokens_seen": 17821069632,
      "step": 62660,
      "train_runtime": 609310.1439,
      "train_tokens_per_second": 29247.945
    },
    {
      "epoch": 2.2173871700349617,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8547928470654215e-05,
      "loss": 0.3860340118408203,
      "num_input_tokens_seen": 17823951168,
      "step": 62670,
      "train_runtime": 609411.3084,
      "train_tokens_per_second": 29247.818
    },
    {
      "epoch": 2.2177409902986907,
      "grad_norm": 0.6015625,
      "learning_rate": 1.854665240972031e-05,
      "loss": 0.3835591793060303,
      "num_input_tokens_seen": 17826824576,
      "step": 62680,
      "train_runtime": 609509.8536,
      "train_tokens_per_second": 29247.804
    },
    {
      "epoch": 2.2180948105624196,
      "grad_norm": 0.625,
      "learning_rate": 1.8545376612121605e-05,
      "loss": 0.3858494281768799,
      "num_input_tokens_seen": 17829648384,
      "step": 62690,
      "train_runtime": 609607.5202,
      "train_tokens_per_second": 29247.75
    },
    {
      "epoch": 2.218448630826148,
      "grad_norm": 0.6328125,
      "learning_rate": 1.854410107776754e-05,
      "loss": 0.3890784740447998,
      "num_input_tokens_seen": 17832494912,
      "step": 62700,
      "train_runtime": 609704.5337,
      "train_tokens_per_second": 29247.765
    },
    {
      "epoch": 2.218802451089877,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8542825806567598e-05,
      "loss": 0.387264347076416,
      "num_input_tokens_seen": 17835344064,
      "step": 62710,
      "train_runtime": 609800.2778,
      "train_tokens_per_second": 29247.845
    },
    {
      "epoch": 2.2191562713536057,
      "grad_norm": 0.6171875,
      "learning_rate": 1.854155079843131e-05,
      "loss": 0.38875930309295653,
      "num_input_tokens_seen": 17838160576,
      "step": 62720,
      "train_runtime": 609895.4197,
      "train_tokens_per_second": 29247.901
    },
    {
      "epoch": 2.2195100916173347,
      "grad_norm": 0.625,
      "learning_rate": 1.8540276053268246e-05,
      "loss": 0.38609349727630615,
      "num_input_tokens_seen": 17840954304,
      "step": 62730,
      "train_runtime": 609991.6282,
      "train_tokens_per_second": 29247.867
    },
    {
      "epoch": 2.2198639118810632,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8539001570988016e-05,
      "loss": 0.38794708251953125,
      "num_input_tokens_seen": 17843745280,
      "step": 62740,
      "train_runtime": 610086.7993,
      "train_tokens_per_second": 29247.88
    },
    {
      "epoch": 2.2202177321447922,
      "grad_norm": 0.64453125,
      "learning_rate": 1.8537727351500283e-05,
      "loss": 0.38756680488586426,
      "num_input_tokens_seen": 17846575936,
      "step": 62750,
      "train_runtime": 610183.5299,
      "train_tokens_per_second": 29247.882
    },
    {
      "epoch": 2.2205715524085208,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8536453394714744e-05,
      "loss": 0.3864884376525879,
      "num_input_tokens_seen": 17849360512,
      "step": 62760,
      "train_runtime": 610276.7602,
      "train_tokens_per_second": 29247.977
    },
    {
      "epoch": 2.2209253726722498,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8535179700541148e-05,
      "loss": 0.38657164573669434,
      "num_input_tokens_seen": 17852226816,
      "step": 62770,
      "train_runtime": 610375.875,
      "train_tokens_per_second": 29247.923
    },
    {
      "epoch": 2.2212791929359783,
      "grad_norm": 0.6171875,
      "learning_rate": 1.853390626888928e-05,
      "loss": 0.3843152284622192,
      "num_input_tokens_seen": 17855051712,
      "step": 62780,
      "train_runtime": 610474.2793,
      "train_tokens_per_second": 29247.836
    },
    {
      "epoch": 2.2216330131997073,
      "grad_norm": 0.640625,
      "learning_rate": 1.8532633099668977e-05,
      "loss": 0.38683643341064455,
      "num_input_tokens_seen": 17857865280,
      "step": 62790,
      "train_runtime": 610569.9174,
      "train_tokens_per_second": 29247.863
    },
    {
      "epoch": 2.221986833463436,
      "grad_norm": 0.625,
      "learning_rate": 1.8531360192790107e-05,
      "loss": 0.3838510990142822,
      "num_input_tokens_seen": 17860658048,
      "step": 62800,
      "train_runtime": 610662.4757,
      "train_tokens_per_second": 29248.003
    },
    {
      "epoch": 2.222340653727165,
      "grad_norm": 0.62890625,
      "learning_rate": 1.853008754816259e-05,
      "loss": 0.38716635704040525,
      "num_input_tokens_seen": 17863485696,
      "step": 62810,
      "train_runtime": 610758.7341,
      "train_tokens_per_second": 29248.023
    },
    {
      "epoch": 2.2226944739908934,
      "grad_norm": 0.62109375,
      "learning_rate": 1.852881516569639e-05,
      "loss": 0.38870112895965575,
      "num_input_tokens_seen": 17866356800,
      "step": 62820,
      "train_runtime": 610859.9269,
      "train_tokens_per_second": 29247.878
    },
    {
      "epoch": 2.2230482942546224,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8527543045301513e-05,
      "loss": 0.38657517433166505,
      "num_input_tokens_seen": 17869209344,
      "step": 62830,
      "train_runtime": 610954.5554,
      "train_tokens_per_second": 29248.017
    },
    {
      "epoch": 2.223402114518351,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8526271186888e-05,
      "loss": 0.3889434337615967,
      "num_input_tokens_seen": 17872057152,
      "step": 62840,
      "train_runtime": 611052.1161,
      "train_tokens_per_second": 29248.008
    },
    {
      "epoch": 2.22375593478208,
      "grad_norm": 0.62890625,
      "learning_rate": 1.852499959036595e-05,
      "loss": 0.38601605892181395,
      "num_input_tokens_seen": 17874869568,
      "step": 62850,
      "train_runtime": 611145.1393,
      "train_tokens_per_second": 29248.158
    },
    {
      "epoch": 2.2241097550458084,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8523728255645494e-05,
      "loss": 0.38413076400756835,
      "num_input_tokens_seen": 17877708864,
      "step": 62860,
      "train_runtime": 611241.4438,
      "train_tokens_per_second": 29248.195
    },
    {
      "epoch": 2.2244635753095374,
      "grad_norm": 0.609375,
      "learning_rate": 1.8522457182636814e-05,
      "loss": 0.38504643440246583,
      "num_input_tokens_seen": 17880557632,
      "step": 62870,
      "train_runtime": 611339.6248,
      "train_tokens_per_second": 29248.157
    },
    {
      "epoch": 2.2248173955732664,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8521186371250122e-05,
      "loss": 0.38976006507873534,
      "num_input_tokens_seen": 17883430784,
      "step": 62880,
      "train_runtime": 611434.1718,
      "train_tokens_per_second": 29248.334
    },
    {
      "epoch": 2.225171215836995,
      "grad_norm": 0.609375,
      "learning_rate": 1.851991582139569e-05,
      "loss": 0.38901619911193847,
      "num_input_tokens_seen": 17886279104,
      "step": 62890,
      "train_runtime": 611527.9339,
      "train_tokens_per_second": 29248.507
    },
    {
      "epoch": 2.225525036100724,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8518645532983816e-05,
      "loss": 0.38570528030395507,
      "num_input_tokens_seen": 17889097344,
      "step": 62900,
      "train_runtime": 611624.0357,
      "train_tokens_per_second": 29248.519
    },
    {
      "epoch": 2.2258788563644525,
      "grad_norm": 0.625,
      "learning_rate": 1.851737550592486e-05,
      "loss": 0.38550467491149903,
      "num_input_tokens_seen": 17891929792,
      "step": 62910,
      "train_runtime": 611720.1888,
      "train_tokens_per_second": 29248.552
    },
    {
      "epoch": 2.2262326766281815,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8516105740129215e-05,
      "loss": 0.38628244400024414,
      "num_input_tokens_seen": 17894842304,
      "step": 62920,
      "train_runtime": 611820.5967,
      "train_tokens_per_second": 29248.512
    },
    {
      "epoch": 2.22658649689191,
      "grad_norm": 0.65234375,
      "learning_rate": 1.8514836235507305e-05,
      "loss": 0.390250039100647,
      "num_input_tokens_seen": 17897694336,
      "step": 62930,
      "train_runtime": 611917.809,
      "train_tokens_per_second": 29248.527
    },
    {
      "epoch": 2.226940317155639,
      "grad_norm": 0.5859375,
      "learning_rate": 1.8513566991969618e-05,
      "loss": 0.3838039875030518,
      "num_input_tokens_seen": 17900512064,
      "step": 62940,
      "train_runtime": 612012.9569,
      "train_tokens_per_second": 29248.583
    },
    {
      "epoch": 2.2272941374193675,
      "grad_norm": 0.6015625,
      "learning_rate": 1.851229800942668e-05,
      "loss": 0.3902816534042358,
      "num_input_tokens_seen": 17903365056,
      "step": 62950,
      "train_runtime": 612108.2113,
      "train_tokens_per_second": 29248.693
    },
    {
      "epoch": 2.2276479576830965,
      "grad_norm": 0.59375,
      "learning_rate": 1.851102928778905e-05,
      "loss": 0.38694334030151367,
      "num_input_tokens_seen": 17906207488,
      "step": 62960,
      "train_runtime": 612204.7139,
      "train_tokens_per_second": 29248.725
    },
    {
      "epoch": 2.228001777946825,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8509760826967337e-05,
      "loss": 0.38334312438964846,
      "num_input_tokens_seen": 17909064768,
      "step": 62970,
      "train_runtime": 612302.7836,
      "train_tokens_per_second": 29248.707
    },
    {
      "epoch": 2.228355598210554,
      "grad_norm": 0.62890625,
      "learning_rate": 1.850849262687219e-05,
      "loss": 0.38725640773773196,
      "num_input_tokens_seen": 17911911680,
      "step": 62980,
      "train_runtime": 612397.2952,
      "train_tokens_per_second": 29248.842
    },
    {
      "epoch": 2.2287094184742826,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8507224687414308e-05,
      "loss": 0.38622047901153567,
      "num_input_tokens_seen": 17914744320,
      "step": 62990,
      "train_runtime": 612493.8376,
      "train_tokens_per_second": 29248.856
    },
    {
      "epoch": 2.2290632387380116,
      "grad_norm": 0.609375,
      "learning_rate": 1.8505957008504415e-05,
      "loss": 0.3843966007232666,
      "num_input_tokens_seen": 17917577600,
      "step": 63000,
      "train_runtime": 612588.1447,
      "train_tokens_per_second": 29248.979
    },
    {
      "epoch": 2.22941705900174,
      "grad_norm": 0.59375,
      "learning_rate": 1.8504689590053308e-05,
      "loss": 0.38288173675537107,
      "num_input_tokens_seen": 17920431744,
      "step": 63010,
      "train_runtime": 612687.612,
      "train_tokens_per_second": 29248.889
    },
    {
      "epoch": 2.229770879265469,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8503422431971794e-05,
      "loss": 0.38563108444213867,
      "num_input_tokens_seen": 17923229248,
      "step": 63020,
      "train_runtime": 612784.4382,
      "train_tokens_per_second": 29248.832
    },
    {
      "epoch": 2.230124699529198,
      "grad_norm": 0.6015625,
      "learning_rate": 1.850215553417074e-05,
      "loss": 0.38161664009094237,
      "num_input_tokens_seen": 17926048128,
      "step": 63030,
      "train_runtime": 612879.0707,
      "train_tokens_per_second": 29248.915
    },
    {
      "epoch": 2.2304785197929267,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8500888896561066e-05,
      "loss": 0.3875094413757324,
      "num_input_tokens_seen": 17928859264,
      "step": 63040,
      "train_runtime": 612975.8531,
      "train_tokens_per_second": 29248.883
    },
    {
      "epoch": 2.2308323400566556,
      "grad_norm": 0.73828125,
      "learning_rate": 1.8499622519053708e-05,
      "loss": 0.38614673614501954,
      "num_input_tokens_seen": 17931673024,
      "step": 63050,
      "train_runtime": 613071.8507,
      "train_tokens_per_second": 29248.893
    },
    {
      "epoch": 2.231186160320384,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8498356401559667e-05,
      "loss": 0.3868428707122803,
      "num_input_tokens_seen": 17934523840,
      "step": 63060,
      "train_runtime": 613169.0365,
      "train_tokens_per_second": 29248.907
    },
    {
      "epoch": 2.231539980584113,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8497090543989972e-05,
      "loss": 0.38914055824279786,
      "num_input_tokens_seen": 17937379968,
      "step": 63070,
      "train_runtime": 613265.6988,
      "train_tokens_per_second": 29248.954
    },
    {
      "epoch": 2.2318938008478417,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8495824946255704e-05,
      "loss": 0.38773550987243655,
      "num_input_tokens_seen": 17940220032,
      "step": 63080,
      "train_runtime": 613362.1471,
      "train_tokens_per_second": 29248.985
    },
    {
      "epoch": 2.2322476211115707,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8494559608267985e-05,
      "loss": 0.38642306327819825,
      "num_input_tokens_seen": 17943035136,
      "step": 63090,
      "train_runtime": 613457.4787,
      "train_tokens_per_second": 29249.028
    },
    {
      "epoch": 2.2326014413752993,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8493294529937976e-05,
      "loss": 0.3919111728668213,
      "num_input_tokens_seen": 17945947456,
      "step": 63100,
      "train_runtime": 613559.7063,
      "train_tokens_per_second": 29248.902
    },
    {
      "epoch": 2.2329552616390282,
      "grad_norm": 0.625,
      "learning_rate": 1.849202971117688e-05,
      "loss": 0.389154052734375,
      "num_input_tokens_seen": 17948796544,
      "step": 63110,
      "train_runtime": 613656.1787,
      "train_tokens_per_second": 29248.946
    },
    {
      "epoch": 2.233309081902757,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8490765151895952e-05,
      "loss": 0.392545747756958,
      "num_input_tokens_seen": 17951693056,
      "step": 63120,
      "train_runtime": 613754.3092,
      "train_tokens_per_second": 29248.989
    },
    {
      "epoch": 2.2336629021664858,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8489500852006484e-05,
      "loss": 0.3844566345214844,
      "num_input_tokens_seen": 17954511552,
      "step": 63130,
      "train_runtime": 613850.3295,
      "train_tokens_per_second": 29249.005
    },
    {
      "epoch": 2.2340167224302143,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8488236811419795e-05,
      "loss": 0.3873509645462036,
      "num_input_tokens_seen": 17957332800,
      "step": 63140,
      "train_runtime": 613943.4574,
      "train_tokens_per_second": 29249.164
    },
    {
      "epoch": 2.2343705426939433,
      "grad_norm": 0.65625,
      "learning_rate": 1.8486973030047278e-05,
      "loss": 0.38297343254089355,
      "num_input_tokens_seen": 17960199232,
      "step": 63150,
      "train_runtime": 614041.9337,
      "train_tokens_per_second": 29249.141
    },
    {
      "epoch": 2.234724362957672,
      "grad_norm": 0.58203125,
      "learning_rate": 1.8485709507800337e-05,
      "loss": 0.38829975128173827,
      "num_input_tokens_seen": 17963042176,
      "step": 63160,
      "train_runtime": 614140.0573,
      "train_tokens_per_second": 29249.097
    },
    {
      "epoch": 2.235078183221401,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8484446244590438e-05,
      "loss": 0.38270604610443115,
      "num_input_tokens_seen": 17965905216,
      "step": 63170,
      "train_runtime": 614242.3739,
      "train_tokens_per_second": 29248.886
    },
    {
      "epoch": 2.23543200348513,
      "grad_norm": 0.6015625,
      "learning_rate": 1.848318324032909e-05,
      "loss": 0.38156564235687257,
      "num_input_tokens_seen": 17968737600,
      "step": 63180,
      "train_runtime": 614339.7398,
      "train_tokens_per_second": 29248.861
    },
    {
      "epoch": 2.2357858237488584,
      "grad_norm": 0.625,
      "learning_rate": 1.8481920494927827e-05,
      "loss": 0.38745296001434326,
      "num_input_tokens_seen": 17971527744,
      "step": 63190,
      "train_runtime": 614434.6463,
      "train_tokens_per_second": 29248.884
    },
    {
      "epoch": 2.2361396440125874,
      "grad_norm": 0.625,
      "learning_rate": 1.8480658008298242e-05,
      "loss": 0.3892880916595459,
      "num_input_tokens_seen": 17974361408,
      "step": 63200,
      "train_runtime": 614534.984,
      "train_tokens_per_second": 29248.72
    },
    {
      "epoch": 2.236493464276316,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8479395780351963e-05,
      "loss": 0.3830745697021484,
      "num_input_tokens_seen": 17977231808,
      "step": 63210,
      "train_runtime": 614632.0735,
      "train_tokens_per_second": 29248.769
    },
    {
      "epoch": 2.236847284540045,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8478133811000662e-05,
      "loss": 0.3844662427902222,
      "num_input_tokens_seen": 17980107904,
      "step": 63220,
      "train_runtime": 614730.1412,
      "train_tokens_per_second": 29248.782
    },
    {
      "epoch": 2.2372011048037734,
      "grad_norm": 0.609375,
      "learning_rate": 1.8476872100156055e-05,
      "loss": 0.3875835657119751,
      "num_input_tokens_seen": 17982972992,
      "step": 63230,
      "train_runtime": 614826.6736,
      "train_tokens_per_second": 29248.85
    },
    {
      "epoch": 2.2375549250675024,
      "grad_norm": 0.6484375,
      "learning_rate": 1.8475610647729893e-05,
      "loss": 0.38609728813171384,
      "num_input_tokens_seen": 17985832064,
      "step": 63240,
      "train_runtime": 614925.7825,
      "train_tokens_per_second": 29248.785
    },
    {
      "epoch": 2.237908745331231,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8474349453633978e-05,
      "loss": 0.38477334976196287,
      "num_input_tokens_seen": 17988681024,
      "step": 63250,
      "train_runtime": 615024.0059,
      "train_tokens_per_second": 29248.746
    },
    {
      "epoch": 2.23826256559496,
      "grad_norm": 0.625,
      "learning_rate": 1.8473088517780156e-05,
      "loss": 0.3886557102203369,
      "num_input_tokens_seen": 17991489792,
      "step": 63260,
      "train_runtime": 615118.4382,
      "train_tokens_per_second": 29248.822
    },
    {
      "epoch": 2.2386163858586885,
      "grad_norm": 0.6328125,
      "learning_rate": 1.84718278400803e-05,
      "loss": 0.38866229057312013,
      "num_input_tokens_seen": 17994283456,
      "step": 63270,
      "train_runtime": 615212.4775,
      "train_tokens_per_second": 29248.892
    },
    {
      "epoch": 2.2389702061224175,
      "grad_norm": 0.62890625,
      "learning_rate": 1.847056742044634e-05,
      "loss": 0.3846184730529785,
      "num_input_tokens_seen": 17997138560,
      "step": 63280,
      "train_runtime": 615310.5759,
      "train_tokens_per_second": 29248.869
    },
    {
      "epoch": 2.239324026386146,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8469307258790238e-05,
      "loss": 0.3878648281097412,
      "num_input_tokens_seen": 18000014592,
      "step": 63290,
      "train_runtime": 615412.7538,
      "train_tokens_per_second": 29248.686
    },
    {
      "epoch": 2.239677846649875,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8468047355024013e-05,
      "loss": 0.3866232395172119,
      "num_input_tokens_seen": 18002834944,
      "step": 63300,
      "train_runtime": 615509.0316,
      "train_tokens_per_second": 29248.693
    },
    {
      "epoch": 2.2400316669136036,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8466787709059705e-05,
      "loss": 0.39187049865722656,
      "num_input_tokens_seen": 18005689024,
      "step": 63310,
      "train_runtime": 615607.4936,
      "train_tokens_per_second": 29248.651
    },
    {
      "epoch": 2.2403854871773325,
      "grad_norm": 0.625,
      "learning_rate": 1.846552832080941e-05,
      "loss": 0.39432082176208494,
      "num_input_tokens_seen": 18008566912,
      "step": 63320,
      "train_runtime": 615708.9379,
      "train_tokens_per_second": 29248.507
    },
    {
      "epoch": 2.240739307441061,
      "grad_norm": 0.609375,
      "learning_rate": 1.8464269190185267e-05,
      "loss": 0.3885178804397583,
      "num_input_tokens_seen": 18011385536,
      "step": 63330,
      "train_runtime": 615804.3236,
      "train_tokens_per_second": 29248.553
    },
    {
      "epoch": 2.24109312770479,
      "grad_norm": 0.640625,
      "learning_rate": 1.8463010317099445e-05,
      "loss": 0.386199951171875,
      "num_input_tokens_seen": 18014273600,
      "step": 63340,
      "train_runtime": 615903.306,
      "train_tokens_per_second": 29248.542
    },
    {
      "epoch": 2.2414469479685186,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8461751701464168e-05,
      "loss": 0.39035494327545167,
      "num_input_tokens_seen": 18017111104,
      "step": 63350,
      "train_runtime": 615997.6359,
      "train_tokens_per_second": 29248.669
    },
    {
      "epoch": 2.2418007682322476,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8460493343191697e-05,
      "loss": 0.3860316276550293,
      "num_input_tokens_seen": 18019982592,
      "step": 63360,
      "train_runtime": 616096.8328,
      "train_tokens_per_second": 29248.621
    },
    {
      "epoch": 2.2421545884959766,
      "grad_norm": 0.59375,
      "learning_rate": 1.8459235242194336e-05,
      "loss": 0.3914928913116455,
      "num_input_tokens_seen": 18022809728,
      "step": 63370,
      "train_runtime": 616193.5266,
      "train_tokens_per_second": 29248.619
    },
    {
      "epoch": 2.242508408759705,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8457977398384422e-05,
      "loss": 0.3819234848022461,
      "num_input_tokens_seen": 18025575296,
      "step": 63380,
      "train_runtime": 616288.094,
      "train_tokens_per_second": 29248.618
    },
    {
      "epoch": 2.242862229023434,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8456719811674344e-05,
      "loss": 0.3863274097442627,
      "num_input_tokens_seen": 18028389056,
      "step": 63390,
      "train_runtime": 616382.6946,
      "train_tokens_per_second": 29248.694
    },
    {
      "epoch": 2.2432160492871627,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8455462481976532e-05,
      "loss": 0.3876702308654785,
      "num_input_tokens_seen": 18031265984,
      "step": 63400,
      "train_runtime": 616480.2311,
      "train_tokens_per_second": 29248.733
    },
    {
      "epoch": 2.2435698695508917,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8454205409203457e-05,
      "loss": 0.3863736629486084,
      "num_input_tokens_seen": 18034121600,
      "step": 63410,
      "train_runtime": 616581.6094,
      "train_tokens_per_second": 29248.556
    },
    {
      "epoch": 2.24392368981462,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8452948593267625e-05,
      "loss": 0.38455352783203123,
      "num_input_tokens_seen": 18037002880,
      "step": 63420,
      "train_runtime": 616681.6374,
      "train_tokens_per_second": 29248.484
    },
    {
      "epoch": 2.244277510078349,
      "grad_norm": 0.5703125,
      "learning_rate": 1.845169203408159e-05,
      "loss": 0.3856102466583252,
      "num_input_tokens_seen": 18039850752,
      "step": 63430,
      "train_runtime": 616779.0066,
      "train_tokens_per_second": 29248.484
    },
    {
      "epoch": 2.2446313303420777,
      "grad_norm": 0.62890625,
      "learning_rate": 1.845043573155795e-05,
      "loss": 0.38246469497680663,
      "num_input_tokens_seen": 18042645056,
      "step": 63440,
      "train_runtime": 616876.221,
      "train_tokens_per_second": 29248.404
    },
    {
      "epoch": 2.2449851506058067,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8449179685609337e-05,
      "loss": 0.38494458198547366,
      "num_input_tokens_seen": 18045479104,
      "step": 63450,
      "train_runtime": 616970.7627,
      "train_tokens_per_second": 29248.516
    },
    {
      "epoch": 2.2453389708695353,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8447923896148434e-05,
      "loss": 0.38051161766052244,
      "num_input_tokens_seen": 18048314304,
      "step": 63460,
      "train_runtime": 617066.7148,
      "train_tokens_per_second": 29248.562
    },
    {
      "epoch": 2.2456927911332643,
      "grad_norm": 0.69921875,
      "learning_rate": 1.8446668363087957e-05,
      "loss": 0.3884440422058105,
      "num_input_tokens_seen": 18051161152,
      "step": 63470,
      "train_runtime": 617166.1079,
      "train_tokens_per_second": 29248.465
    },
    {
      "epoch": 2.246046611396993,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8445413086340664e-05,
      "loss": 0.38826732635498046,
      "num_input_tokens_seen": 18054002368,
      "step": 63480,
      "train_runtime": 617262.593,
      "train_tokens_per_second": 29248.496
    },
    {
      "epoch": 2.246400431660722,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8444158065819366e-05,
      "loss": 0.3854444742202759,
      "num_input_tokens_seen": 18056809152,
      "step": 63490,
      "train_runtime": 617358.7415,
      "train_tokens_per_second": 29248.487
    },
    {
      "epoch": 2.2467542519244503,
      "grad_norm": 0.59375,
      "learning_rate": 1.8442903301436902e-05,
      "loss": 0.3874361991882324,
      "num_input_tokens_seen": 18059634944,
      "step": 63500,
      "train_runtime": 617457.3552,
      "train_tokens_per_second": 29248.392
    },
    {
      "epoch": 2.2471080721881793,
      "grad_norm": 0.625,
      "learning_rate": 1.844164879310616e-05,
      "loss": 0.3885998010635376,
      "num_input_tokens_seen": 18062452928,
      "step": 63510,
      "train_runtime": 617552.5647,
      "train_tokens_per_second": 29248.446
    },
    {
      "epoch": 2.2474618924519083,
      "grad_norm": 0.625,
      "learning_rate": 1.844039454074006e-05,
      "loss": 0.3909266471862793,
      "num_input_tokens_seen": 18065313152,
      "step": 63520,
      "train_runtime": 617650.4032,
      "train_tokens_per_second": 29248.444
    },
    {
      "epoch": 2.247815712715637,
      "grad_norm": 0.640625,
      "learning_rate": 1.8439140544251578e-05,
      "loss": 0.3861539840698242,
      "num_input_tokens_seen": 18068128896,
      "step": 63530,
      "train_runtime": 617748.0946,
      "train_tokens_per_second": 29248.377
    },
    {
      "epoch": 2.248169532979366,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8437886803553726e-05,
      "loss": 0.3860292434692383,
      "num_input_tokens_seen": 18070927296,
      "step": 63540,
      "train_runtime": 617839.3269,
      "train_tokens_per_second": 29248.587
    },
    {
      "epoch": 2.2485233532430944,
      "grad_norm": 0.59765625,
      "learning_rate": 1.843663331855955e-05,
      "loss": 0.3855440139770508,
      "num_input_tokens_seen": 18073789568,
      "step": 63550,
      "train_runtime": 617939.1781,
      "train_tokens_per_second": 29248.493
    },
    {
      "epoch": 2.2488771735068234,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8435380089182142e-05,
      "loss": 0.38881330490112304,
      "num_input_tokens_seen": 18076627072,
      "step": 63560,
      "train_runtime": 618037.3166,
      "train_tokens_per_second": 29248.44
    },
    {
      "epoch": 2.249230993770552,
      "grad_norm": 0.59375,
      "learning_rate": 1.8434127115334636e-05,
      "loss": 0.38506574630737306,
      "num_input_tokens_seen": 18079494784,
      "step": 63570,
      "train_runtime": 618136.0165,
      "train_tokens_per_second": 29248.409
    },
    {
      "epoch": 2.249584814034281,
      "grad_norm": 0.61328125,
      "learning_rate": 1.843287439693021e-05,
      "loss": 0.3893226146697998,
      "num_input_tokens_seen": 18082270528,
      "step": 63580,
      "train_runtime": 618229.0636,
      "train_tokens_per_second": 29248.496
    },
    {
      "epoch": 2.2499386342980094,
      "grad_norm": 0.640625,
      "learning_rate": 1.8431621933882087e-05,
      "loss": 0.38431107997894287,
      "num_input_tokens_seen": 18085124736,
      "step": 63590,
      "train_runtime": 618327.2981,
      "train_tokens_per_second": 29248.466
    },
    {
      "epoch": 2.2502924545617384,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8430369726103514e-05,
      "loss": 0.3872710704803467,
      "num_input_tokens_seen": 18087948160,
      "step": 63600,
      "train_runtime": 618424.1864,
      "train_tokens_per_second": 29248.449
    },
    {
      "epoch": 2.250646274825467,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8429117773507797e-05,
      "loss": 0.387183403968811,
      "num_input_tokens_seen": 18090803840,
      "step": 63610,
      "train_runtime": 618521.4649,
      "train_tokens_per_second": 29248.466
    },
    {
      "epoch": 2.251000095089196,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8427866076008265e-05,
      "loss": 0.3877406597137451,
      "num_input_tokens_seen": 18093636864,
      "step": 63620,
      "train_runtime": 618616.229,
      "train_tokens_per_second": 29248.565
    },
    {
      "epoch": 2.2513539153529245,
      "grad_norm": 0.609375,
      "learning_rate": 1.842661463351832e-05,
      "loss": 0.38566179275512696,
      "num_input_tokens_seen": 18096458176,
      "step": 63630,
      "train_runtime": 618712.2165,
      "train_tokens_per_second": 29248.587
    },
    {
      "epoch": 2.2517077356166535,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8425363445951368e-05,
      "loss": 0.38597493171691893,
      "num_input_tokens_seen": 18099305088,
      "step": 63640,
      "train_runtime": 618810.5609,
      "train_tokens_per_second": 29248.539
    },
    {
      "epoch": 2.252061555880382,
      "grad_norm": 0.6328125,
      "learning_rate": 1.842411251322088e-05,
      "loss": 0.38338935375213623,
      "num_input_tokens_seen": 18102157248,
      "step": 63650,
      "train_runtime": 618909.4243,
      "train_tokens_per_second": 29248.476
    },
    {
      "epoch": 2.252415376144111,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8422861835240357e-05,
      "loss": 0.3890397071838379,
      "num_input_tokens_seen": 18105021696,
      "step": 63660,
      "train_runtime": 619006.2779,
      "train_tokens_per_second": 29248.527
    },
    {
      "epoch": 2.25276919640784,
      "grad_norm": 0.59375,
      "learning_rate": 1.842161141192335e-05,
      "loss": 0.38200969696044923,
      "num_input_tokens_seen": 18107910592,
      "step": 63670,
      "train_runtime": 619105.5955,
      "train_tokens_per_second": 29248.501
    },
    {
      "epoch": 2.2531230166715686,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8420361243183444e-05,
      "loss": 0.3853623628616333,
      "num_input_tokens_seen": 18110765696,
      "step": 63680,
      "train_runtime": 619200.921,
      "train_tokens_per_second": 29248.609
    },
    {
      "epoch": 2.253476836935297,
      "grad_norm": 0.61328125,
      "learning_rate": 1.841911132893427e-05,
      "loss": 0.3850614070892334,
      "num_input_tokens_seen": 18113634752,
      "step": 63690,
      "train_runtime": 619300.1032,
      "train_tokens_per_second": 29248.558
    },
    {
      "epoch": 2.253830657199026,
      "grad_norm": 0.6171875,
      "learning_rate": 1.841786166908949e-05,
      "loss": 0.38353285789489744,
      "num_input_tokens_seen": 18116427968,
      "step": 63700,
      "train_runtime": 619393.3731,
      "train_tokens_per_second": 29248.663
    },
    {
      "epoch": 2.254184477462755,
      "grad_norm": 0.6328125,
      "learning_rate": 1.841661226356282e-05,
      "loss": 0.38275372982025146,
      "num_input_tokens_seen": 18119236352,
      "step": 63710,
      "train_runtime": 619488.8798,
      "train_tokens_per_second": 29248.687
    },
    {
      "epoch": 2.2545382977264836,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8415363112268012e-05,
      "loss": 0.3870521545410156,
      "num_input_tokens_seen": 18122137728,
      "step": 63720,
      "train_runtime": 619589.5425,
      "train_tokens_per_second": 29248.618
    },
    {
      "epoch": 2.2548921179902126,
      "grad_norm": 0.6015625,
      "learning_rate": 1.841411421511886e-05,
      "loss": 0.3854475736618042,
      "num_input_tokens_seen": 18124975104,
      "step": 63730,
      "train_runtime": 619685.3409,
      "train_tokens_per_second": 29248.675
    },
    {
      "epoch": 2.255245938253941,
      "grad_norm": 0.6015625,
      "learning_rate": 1.841286557202919e-05,
      "loss": 0.3885016679763794,
      "num_input_tokens_seen": 18127781952,
      "step": 63740,
      "train_runtime": 619780.9295,
      "train_tokens_per_second": 29248.693
    },
    {
      "epoch": 2.25559975851767,
      "grad_norm": 0.6015625,
      "learning_rate": 1.841161718291288e-05,
      "loss": 0.37965714931488037,
      "num_input_tokens_seen": 18130655552,
      "step": 63750,
      "train_runtime": 619879.7819,
      "train_tokens_per_second": 29248.664
    },
    {
      "epoch": 2.2559535787813987,
      "grad_norm": 0.625,
      "learning_rate": 1.8410369047683846e-05,
      "loss": 0.3882360219955444,
      "num_input_tokens_seen": 18133477952,
      "step": 63760,
      "train_runtime": 619975.9904,
      "train_tokens_per_second": 29248.678
    },
    {
      "epoch": 2.2563073990451277,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8409121166256037e-05,
      "loss": 0.38308262825012207,
      "num_input_tokens_seen": 18136394880,
      "step": 63770,
      "train_runtime": 620075.9373,
      "train_tokens_per_second": 29248.667
    },
    {
      "epoch": 2.256661219308856,
      "grad_norm": 0.63671875,
      "learning_rate": 1.840787353854346e-05,
      "loss": 0.3947603702545166,
      "num_input_tokens_seen": 18139277696,
      "step": 63780,
      "train_runtime": 620173.7832,
      "train_tokens_per_second": 29248.701
    },
    {
      "epoch": 2.257015039572585,
      "grad_norm": 0.6171875,
      "learning_rate": 1.840662616446015e-05,
      "loss": 0.3855008125305176,
      "num_input_tokens_seen": 18142099264,
      "step": 63790,
      "train_runtime": 620267.5793,
      "train_tokens_per_second": 29248.827
    },
    {
      "epoch": 2.2573688598363137,
      "grad_norm": 0.609375,
      "learning_rate": 1.8405379043920177e-05,
      "loss": 0.3874178886413574,
      "num_input_tokens_seen": 18144960192,
      "step": 63800,
      "train_runtime": 620365.8154,
      "train_tokens_per_second": 29248.807
    },
    {
      "epoch": 2.2577226801000427,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8404132176837667e-05,
      "loss": 0.38170671463012695,
      "num_input_tokens_seen": 18147823744,
      "step": 63810,
      "train_runtime": 620461.957,
      "train_tokens_per_second": 29248.89
    },
    {
      "epoch": 2.2580765003637713,
      "grad_norm": 0.609375,
      "learning_rate": 1.840288556312678e-05,
      "loss": 0.3856589078903198,
      "num_input_tokens_seen": 18150667328,
      "step": 63820,
      "train_runtime": 620560.6893,
      "train_tokens_per_second": 29248.819
    },
    {
      "epoch": 2.2584303206275003,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8401639202701713e-05,
      "loss": 0.3918909549713135,
      "num_input_tokens_seen": 18153507776,
      "step": 63830,
      "train_runtime": 620658.3348,
      "train_tokens_per_second": 29248.794
    },
    {
      "epoch": 2.258784140891229,
      "grad_norm": 0.625,
      "learning_rate": 1.8400393095476712e-05,
      "loss": 0.3859292268753052,
      "num_input_tokens_seen": 18156356992,
      "step": 63840,
      "train_runtime": 620755.3692,
      "train_tokens_per_second": 29248.812
    },
    {
      "epoch": 2.259137961154958,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8399147241366053e-05,
      "loss": 0.38292129039764405,
      "num_input_tokens_seen": 18159206912,
      "step": 63850,
      "train_runtime": 620853.4952,
      "train_tokens_per_second": 29248.779
    },
    {
      "epoch": 2.259491781418687,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8397901640284058e-05,
      "loss": 0.3852531433105469,
      "num_input_tokens_seen": 18162015296,
      "step": 63860,
      "train_runtime": 620948.0599,
      "train_tokens_per_second": 29248.848
    },
    {
      "epoch": 2.2598456016824153,
      "grad_norm": 0.640625,
      "learning_rate": 1.8396656292145095e-05,
      "loss": 0.3920015811920166,
      "num_input_tokens_seen": 18164865728,
      "step": 63870,
      "train_runtime": 621045.0674,
      "train_tokens_per_second": 29248.869
    },
    {
      "epoch": 2.2601994219461443,
      "grad_norm": 0.625,
      "learning_rate": 1.8395411196863562e-05,
      "loss": 0.38411126136779783,
      "num_input_tokens_seen": 18167747456,
      "step": 63880,
      "train_runtime": 621144.9787,
      "train_tokens_per_second": 29248.804
    },
    {
      "epoch": 2.260553242209873,
      "grad_norm": 0.62890625,
      "learning_rate": 1.839416635435391e-05,
      "loss": 0.3886701583862305,
      "num_input_tokens_seen": 18170612736,
      "step": 63890,
      "train_runtime": 621243.452,
      "train_tokens_per_second": 29248.779
    },
    {
      "epoch": 2.260907062473602,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8392921764530616e-05,
      "loss": 0.3931838274002075,
      "num_input_tokens_seen": 18173508032,
      "step": 63900,
      "train_runtime": 621341.3544,
      "train_tokens_per_second": 29248.831
    },
    {
      "epoch": 2.2612608827373304,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8391677427308206e-05,
      "loss": 0.3892277717590332,
      "num_input_tokens_seen": 18176356672,
      "step": 63910,
      "train_runtime": 621438.4389,
      "train_tokens_per_second": 29248.845
    },
    {
      "epoch": 2.2616147030010594,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8390433342601252e-05,
      "loss": 0.38676550388336184,
      "num_input_tokens_seen": 18179242432,
      "step": 63920,
      "train_runtime": 621537.1414,
      "train_tokens_per_second": 29248.843
    },
    {
      "epoch": 2.261968523264788,
      "grad_norm": 0.59375,
      "learning_rate": 1.838918951032435e-05,
      "loss": 0.3876307487487793,
      "num_input_tokens_seen": 18182086272,
      "step": 63930,
      "train_runtime": 621638.1417,
      "train_tokens_per_second": 29248.666
    },
    {
      "epoch": 2.262322343528517,
      "grad_norm": 0.59375,
      "learning_rate": 1.8387945930392157e-05,
      "loss": 0.3839359521865845,
      "num_input_tokens_seen": 18184978560,
      "step": 63940,
      "train_runtime": 621739.1856,
      "train_tokens_per_second": 29248.564
    },
    {
      "epoch": 2.2626761637922455,
      "grad_norm": 0.6015625,
      "learning_rate": 1.838670260271935e-05,
      "loss": 0.3845299482345581,
      "num_input_tokens_seen": 18187870912,
      "step": 63950,
      "train_runtime": 621838.598,
      "train_tokens_per_second": 29248.54
    },
    {
      "epoch": 2.2630299840559744,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8385459527220665e-05,
      "loss": 0.3811118841171265,
      "num_input_tokens_seen": 18190690944,
      "step": 63960,
      "train_runtime": 621936.2747,
      "train_tokens_per_second": 29248.48
    },
    {
      "epoch": 2.263383804319703,
      "grad_norm": 0.64453125,
      "learning_rate": 1.838421670381086e-05,
      "loss": 0.38582501411437986,
      "num_input_tokens_seen": 18193527552,
      "step": 63970,
      "train_runtime": 622032.0523,
      "train_tokens_per_second": 29248.537
    },
    {
      "epoch": 2.263737624583432,
      "grad_norm": 0.60546875,
      "learning_rate": 1.838297413240475e-05,
      "loss": 0.3884643316268921,
      "num_input_tokens_seen": 18196320960,
      "step": 63980,
      "train_runtime": 622127.734,
      "train_tokens_per_second": 29248.529
    },
    {
      "epoch": 2.2640914448471605,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8381731812917183e-05,
      "loss": 0.38035154342651367,
      "num_input_tokens_seen": 18199177536,
      "step": 63990,
      "train_runtime": 622224.5776,
      "train_tokens_per_second": 29248.567
    },
    {
      "epoch": 2.2644452651108895,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8380489745263044e-05,
      "loss": 0.39151365756988527,
      "num_input_tokens_seen": 18202115584,
      "step": 64000,
      "train_runtime": 622327.9893,
      "train_tokens_per_second": 29248.428
    },
    {
      "epoch": 2.2647990853746185,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8379247929357262e-05,
      "loss": 0.3870481014251709,
      "num_input_tokens_seen": 18204937984,
      "step": 64010,
      "train_runtime": 622423.9877,
      "train_tokens_per_second": 29248.452
    },
    {
      "epoch": 2.265152905638347,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8378006365114808e-05,
      "loss": 0.38342905044555664,
      "num_input_tokens_seen": 18207782784,
      "step": 64020,
      "train_runtime": 622522.8633,
      "train_tokens_per_second": 29248.376
    },
    {
      "epoch": 2.2655067259020756,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8376765052450693e-05,
      "loss": 0.38654041290283203,
      "num_input_tokens_seen": 18210612032,
      "step": 64030,
      "train_runtime": 622618.4137,
      "train_tokens_per_second": 29248.432
    },
    {
      "epoch": 2.2658605461658046,
      "grad_norm": 0.609375,
      "learning_rate": 1.8375523991279958e-05,
      "loss": 0.388966703414917,
      "num_input_tokens_seen": 18213459136,
      "step": 64040,
      "train_runtime": 622716.1569,
      "train_tokens_per_second": 29248.413
    },
    {
      "epoch": 2.2662143664295336,
      "grad_norm": 0.61328125,
      "learning_rate": 1.83742831815177e-05,
      "loss": 0.38438925743103025,
      "num_input_tokens_seen": 18216316096,
      "step": 64050,
      "train_runtime": 622813.759,
      "train_tokens_per_second": 29248.416
    },
    {
      "epoch": 2.266568186693262,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8373042623079042e-05,
      "loss": 0.38289847373962405,
      "num_input_tokens_seen": 18219144064,
      "step": 64060,
      "train_runtime": 622908.4088,
      "train_tokens_per_second": 29248.512
    },
    {
      "epoch": 2.266922006956991,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8371802315879157e-05,
      "loss": 0.38445069789886477,
      "num_input_tokens_seen": 18222026368,
      "step": 64070,
      "train_runtime": 623008.2321,
      "train_tokens_per_second": 29248.452
    },
    {
      "epoch": 2.2672758272207196,
      "grad_norm": 0.62109375,
      "learning_rate": 1.837056225983326e-05,
      "loss": 0.38940114974975587,
      "num_input_tokens_seen": 18224849856,
      "step": 64080,
      "train_runtime": 623104.9458,
      "train_tokens_per_second": 29248.444
    },
    {
      "epoch": 2.2676296474844486,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8369322454856595e-05,
      "loss": 0.38449721336364745,
      "num_input_tokens_seen": 18227699200,
      "step": 64090,
      "train_runtime": 623203.2159,
      "train_tokens_per_second": 29248.404
    },
    {
      "epoch": 2.267983467748177,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8368082900864454e-05,
      "loss": 0.3882309436798096,
      "num_input_tokens_seen": 18230607296,
      "step": 64100,
      "train_runtime": 623303.6239,
      "train_tokens_per_second": 29248.358
    },
    {
      "epoch": 2.268337288011906,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8366843597772163e-05,
      "loss": 0.3835292339324951,
      "num_input_tokens_seen": 18233416512,
      "step": 64110,
      "train_runtime": 623398.0001,
      "train_tokens_per_second": 29248.436
    },
    {
      "epoch": 2.2686911082756347,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8365604545495093e-05,
      "loss": 0.384830904006958,
      "num_input_tokens_seen": 18236270336,
      "step": 64120,
      "train_runtime": 623496.3096,
      "train_tokens_per_second": 29248.401
    },
    {
      "epoch": 2.2690449285393637,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8364365743948656e-05,
      "loss": 0.38359675407409666,
      "num_input_tokens_seen": 18239149376,
      "step": 64130,
      "train_runtime": 623594.1746,
      "train_tokens_per_second": 29248.428
    },
    {
      "epoch": 2.2693987488030922,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8363127193048303e-05,
      "loss": 0.3859595775604248,
      "num_input_tokens_seen": 18241989056,
      "step": 64140,
      "train_runtime": 623691.6708,
      "train_tokens_per_second": 29248.409
    },
    {
      "epoch": 2.269752569066821,
      "grad_norm": 0.64453125,
      "learning_rate": 1.836188889270952e-05,
      "loss": 0.38685386180877684,
      "num_input_tokens_seen": 18244820672,
      "step": 64150,
      "train_runtime": 623790.5176,
      "train_tokens_per_second": 29248.314
    },
    {
      "epoch": 2.2701063893305498,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8360650842847834e-05,
      "loss": 0.3923333644866943,
      "num_input_tokens_seen": 18247666816,
      "step": 64160,
      "train_runtime": 623890.0459,
      "train_tokens_per_second": 29248.21
    },
    {
      "epoch": 2.2704602095942787,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8359413043378822e-05,
      "loss": 0.3894052505493164,
      "num_input_tokens_seen": 18250471552,
      "step": 64170,
      "train_runtime": 623987.1168,
      "train_tokens_per_second": 29248.154
    },
    {
      "epoch": 2.2708140298580073,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8358175494218088e-05,
      "loss": 0.38121535778045657,
      "num_input_tokens_seen": 18253279872,
      "step": 64180,
      "train_runtime": 624080.5487,
      "train_tokens_per_second": 29248.276
    },
    {
      "epoch": 2.2711678501217363,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8356938195281282e-05,
      "loss": 0.3870903491973877,
      "num_input_tokens_seen": 18256188672,
      "step": 64190,
      "train_runtime": 624183.1258,
      "train_tokens_per_second": 29248.129
    },
    {
      "epoch": 2.2715216703854653,
      "grad_norm": 0.609375,
      "learning_rate": 1.8355701146484093e-05,
      "loss": 0.38402590751647947,
      "num_input_tokens_seen": 18258988480,
      "step": 64200,
      "train_runtime": 624278.4668,
      "train_tokens_per_second": 29248.147
    },
    {
      "epoch": 2.271875490649194,
      "grad_norm": 0.6328125,
      "learning_rate": 1.835446434774225e-05,
      "loss": 0.38783631324768064,
      "num_input_tokens_seen": 18261761728,
      "step": 64210,
      "train_runtime": 624370.0392,
      "train_tokens_per_second": 29248.299
    },
    {
      "epoch": 2.272229310912923,
      "grad_norm": 0.6171875,
      "learning_rate": 1.835322779897152e-05,
      "loss": 0.38504955768585203,
      "num_input_tokens_seen": 18264589440,
      "step": 64220,
      "train_runtime": 624468.065,
      "train_tokens_per_second": 29248.236
    },
    {
      "epoch": 2.2725831311766513,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8351991500087712e-05,
      "loss": 0.3877102851867676,
      "num_input_tokens_seen": 18267451840,
      "step": 64230,
      "train_runtime": 624564.492,
      "train_tokens_per_second": 29248.304
    },
    {
      "epoch": 2.2729369514403803,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8350755451006675e-05,
      "loss": 0.391158390045166,
      "num_input_tokens_seen": 18270272704,
      "step": 64240,
      "train_runtime": 624661.5345,
      "train_tokens_per_second": 29248.276
    },
    {
      "epoch": 2.273290771704109,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8349519651644293e-05,
      "loss": 0.38818509578704835,
      "num_input_tokens_seen": 18273148800,
      "step": 64250,
      "train_runtime": 624760.5452,
      "train_tokens_per_second": 29248.244
    },
    {
      "epoch": 2.273644591967838,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8348284101916496e-05,
      "loss": 0.38398194313049316,
      "num_input_tokens_seen": 18275962176,
      "step": 64260,
      "train_runtime": 624855.2861,
      "train_tokens_per_second": 29248.312
    },
    {
      "epoch": 2.2739984122315664,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8347048801739246e-05,
      "loss": 0.38775057792663575,
      "num_input_tokens_seen": 18278836416,
      "step": 64270,
      "train_runtime": 624951.3426,
      "train_tokens_per_second": 29248.415
    },
    {
      "epoch": 2.2743522324952954,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8345813751028558e-05,
      "loss": 0.3825794219970703,
      "num_input_tokens_seen": 18281634432,
      "step": 64280,
      "train_runtime": 625044.7715,
      "train_tokens_per_second": 29248.52
    },
    {
      "epoch": 2.274706052759024,
      "grad_norm": 0.59375,
      "learning_rate": 1.834457894970047e-05,
      "loss": 0.3875119686126709,
      "num_input_tokens_seen": 18284517248,
      "step": 64290,
      "train_runtime": 625145.7634,
      "train_tokens_per_second": 29248.406
    },
    {
      "epoch": 2.275059873022753,
      "grad_norm": 0.62890625,
      "learning_rate": 1.834334439767107e-05,
      "loss": 0.390265154838562,
      "num_input_tokens_seen": 18287369728,
      "step": 64300,
      "train_runtime": 625245.0624,
      "train_tokens_per_second": 29248.323
    },
    {
      "epoch": 2.2754136932864815,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8342110094856478e-05,
      "loss": 0.3876230239868164,
      "num_input_tokens_seen": 18290188480,
      "step": 64310,
      "train_runtime": 625341.3633,
      "train_tokens_per_second": 29248.327
    },
    {
      "epoch": 2.2757675135502105,
      "grad_norm": 0.640625,
      "learning_rate": 1.834087604117287e-05,
      "loss": 0.3871523141860962,
      "num_input_tokens_seen": 18293081152,
      "step": 64320,
      "train_runtime": 625440.6346,
      "train_tokens_per_second": 29248.309
    },
    {
      "epoch": 2.276121333813939,
      "grad_norm": 0.60546875,
      "learning_rate": 1.833964223653644e-05,
      "loss": 0.38451662063598635,
      "num_input_tokens_seen": 18295940352,
      "step": 64330,
      "train_runtime": 625538.3515,
      "train_tokens_per_second": 29248.311
    },
    {
      "epoch": 2.276475154077668,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8338408680863432e-05,
      "loss": 0.3850873470306396,
      "num_input_tokens_seen": 18298808640,
      "step": 64340,
      "train_runtime": 625635.1343,
      "train_tokens_per_second": 29248.371
    },
    {
      "epoch": 2.276828974341397,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8337175374070135e-05,
      "loss": 0.3885332107543945,
      "num_input_tokens_seen": 18301680768,
      "step": 64350,
      "train_runtime": 625732.3507,
      "train_tokens_per_second": 29248.417
    },
    {
      "epoch": 2.2771827946051255,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8335942316072865e-05,
      "loss": 0.38517260551452637,
      "num_input_tokens_seen": 18304524160,
      "step": 64360,
      "train_runtime": 625829.1708,
      "train_tokens_per_second": 29248.436
    },
    {
      "epoch": 2.277536614868854,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8334709506787985e-05,
      "loss": 0.38727407455444335,
      "num_input_tokens_seen": 18307434048,
      "step": 64370,
      "train_runtime": 625930.5692,
      "train_tokens_per_second": 29248.346
    },
    {
      "epoch": 2.277890435132583,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8333476946131896e-05,
      "loss": 0.3823495388031006,
      "num_input_tokens_seen": 18310294400,
      "step": 64380,
      "train_runtime": 626029.2378,
      "train_tokens_per_second": 29248.306
    },
    {
      "epoch": 2.278244255396312,
      "grad_norm": 0.63671875,
      "learning_rate": 1.833224463402104e-05,
      "loss": 0.38387060165405273,
      "num_input_tokens_seen": 18313157696,
      "step": 64390,
      "train_runtime": 626128.0501,
      "train_tokens_per_second": 29248.263
    },
    {
      "epoch": 2.2785980756600406,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8331012570371896e-05,
      "loss": 0.38391757011413574,
      "num_input_tokens_seen": 18316027776,
      "step": 64400,
      "train_runtime": 626225.6232,
      "train_tokens_per_second": 29248.289
    },
    {
      "epoch": 2.2789518959237696,
      "grad_norm": 0.625,
      "learning_rate": 1.8329780755100982e-05,
      "loss": 0.3866096019744873,
      "num_input_tokens_seen": 18318848320,
      "step": 64410,
      "train_runtime": 626319.5249,
      "train_tokens_per_second": 29248.407
    },
    {
      "epoch": 2.279305716187498,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8328549188124855e-05,
      "loss": 0.3855460166931152,
      "num_input_tokens_seen": 18321744320,
      "step": 64420,
      "train_runtime": 626421.7309,
      "train_tokens_per_second": 29248.258
    },
    {
      "epoch": 2.279659536451227,
      "grad_norm": 0.609375,
      "learning_rate": 1.8327317869360112e-05,
      "loss": 0.3851401567459106,
      "num_input_tokens_seen": 18324578176,
      "step": 64430,
      "train_runtime": 626518.6012,
      "train_tokens_per_second": 29248.259
    },
    {
      "epoch": 2.2800133567149556,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8326086798723394e-05,
      "loss": 0.386049485206604,
      "num_input_tokens_seen": 18327486656,
      "step": 64440,
      "train_runtime": 626618.9161,
      "train_tokens_per_second": 29248.218
    },
    {
      "epoch": 2.2803671769786846,
      "grad_norm": 0.625,
      "learning_rate": 1.832485597613137e-05,
      "loss": 0.39018571376800537,
      "num_input_tokens_seen": 18330381120,
      "step": 64450,
      "train_runtime": 626719.1844,
      "train_tokens_per_second": 29248.157
    },
    {
      "epoch": 2.280720997242413,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8323625401500763e-05,
      "loss": 0.38712291717529296,
      "num_input_tokens_seen": 18333246272,
      "step": 64460,
      "train_runtime": 626817.8772,
      "train_tokens_per_second": 29248.123
    },
    {
      "epoch": 2.281074817506142,
      "grad_norm": 0.60546875,
      "learning_rate": 1.832239507474832e-05,
      "loss": 0.38573760986328126,
      "num_input_tokens_seen": 18336156800,
      "step": 64470,
      "train_runtime": 626919.7318,
      "train_tokens_per_second": 29248.014
    },
    {
      "epoch": 2.2814286377698707,
      "grad_norm": 0.59375,
      "learning_rate": 1.8321164995790837e-05,
      "loss": 0.38653221130371096,
      "num_input_tokens_seen": 18339035840,
      "step": 64480,
      "train_runtime": 627018.4884,
      "train_tokens_per_second": 29247.999
    },
    {
      "epoch": 2.2817824580335997,
      "grad_norm": 0.6171875,
      "learning_rate": 1.831993516454515e-05,
      "loss": 0.3872544765472412,
      "num_input_tokens_seen": 18341846016,
      "step": 64490,
      "train_runtime": 627115.3202,
      "train_tokens_per_second": 29247.964
    },
    {
      "epoch": 2.2821362782973287,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8318705580928116e-05,
      "loss": 0.38541855812072756,
      "num_input_tokens_seen": 18344666752,
      "step": 64500,
      "train_runtime": 627211.2121,
      "train_tokens_per_second": 29247.989
    },
    {
      "epoch": 2.2824900985610572,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8317476244856666e-05,
      "loss": 0.38570847511291506,
      "num_input_tokens_seen": 18347492224,
      "step": 64510,
      "train_runtime": 627305.8565,
      "train_tokens_per_second": 29248.081
    },
    {
      "epoch": 2.2828439188247858,
      "grad_norm": 0.609375,
      "learning_rate": 1.8316247156247734e-05,
      "loss": 0.38433194160461426,
      "num_input_tokens_seen": 18350386304,
      "step": 64520,
      "train_runtime": 627405.4278,
      "train_tokens_per_second": 29248.052
    },
    {
      "epoch": 2.2831977390885148,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8315018315018315e-05,
      "loss": 0.384813928604126,
      "num_input_tokens_seen": 18353217600,
      "step": 64530,
      "train_runtime": 627503.2391,
      "train_tokens_per_second": 29248.005
    },
    {
      "epoch": 2.2835515593522437,
      "grad_norm": 0.609375,
      "learning_rate": 1.8313789721085437e-05,
      "loss": 0.38196754455566406,
      "num_input_tokens_seen": 18356073664,
      "step": 64540,
      "train_runtime": 627600.0699,
      "train_tokens_per_second": 29248.043
    },
    {
      "epoch": 2.2839053796159723,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8312561374366162e-05,
      "loss": 0.3838608741760254,
      "num_input_tokens_seen": 18358872704,
      "step": 64550,
      "train_runtime": 627694.189,
      "train_tokens_per_second": 29248.116
    },
    {
      "epoch": 2.2842591998797013,
      "grad_norm": 0.609375,
      "learning_rate": 1.83113332747776e-05,
      "loss": 0.3866115570068359,
      "num_input_tokens_seen": 18361735744,
      "step": 64560,
      "train_runtime": 627794.5986,
      "train_tokens_per_second": 29247.999
    },
    {
      "epoch": 2.28461302014343,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8310105422236888e-05,
      "loss": 0.3874118089675903,
      "num_input_tokens_seen": 18364562880,
      "step": 64570,
      "train_runtime": 627892.2596,
      "train_tokens_per_second": 29247.952
    },
    {
      "epoch": 2.284966840407159,
      "grad_norm": 0.625,
      "learning_rate": 1.8308877816661216e-05,
      "loss": 0.3888392448425293,
      "num_input_tokens_seen": 18367401024,
      "step": 64580,
      "train_runtime": 627990.4738,
      "train_tokens_per_second": 29247.898
    },
    {
      "epoch": 2.2853206606708873,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8307650457967805e-05,
      "loss": 0.3871408224105835,
      "num_input_tokens_seen": 18370255744,
      "step": 64590,
      "train_runtime": 628090.709,
      "train_tokens_per_second": 29247.775
    },
    {
      "epoch": 2.2856744809346163,
      "grad_norm": 0.640625,
      "learning_rate": 1.8306423346073912e-05,
      "loss": 0.3865889310836792,
      "num_input_tokens_seen": 18373099648,
      "step": 64600,
      "train_runtime": 628185.8544,
      "train_tokens_per_second": 29247.872
    },
    {
      "epoch": 2.286028301198345,
      "grad_norm": 0.62890625,
      "learning_rate": 1.830519648089684e-05,
      "loss": 0.3860872745513916,
      "num_input_tokens_seen": 18375949440,
      "step": 64610,
      "train_runtime": 628279.7378,
      "train_tokens_per_second": 29248.038
    },
    {
      "epoch": 2.286382121462074,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8303969862353928e-05,
      "loss": 0.3834192752838135,
      "num_input_tokens_seen": 18378796160,
      "step": 64620,
      "train_runtime": 628375.5314,
      "train_tokens_per_second": 29248.109
    },
    {
      "epoch": 2.2867359417258024,
      "grad_norm": 0.640625,
      "learning_rate": 1.830274349036255e-05,
      "loss": 0.38586907386779784,
      "num_input_tokens_seen": 18381682240,
      "step": 64630,
      "train_runtime": 628476.7142,
      "train_tokens_per_second": 29247.993
    },
    {
      "epoch": 2.2870897619895314,
      "grad_norm": 0.62890625,
      "learning_rate": 1.830151736484013e-05,
      "loss": 0.3856481075286865,
      "num_input_tokens_seen": 18384510848,
      "step": 64640,
      "train_runtime": 628575.2069,
      "train_tokens_per_second": 29247.91
    },
    {
      "epoch": 2.28744358225326,
      "grad_norm": 0.5859375,
      "learning_rate": 1.830029148570411e-05,
      "loss": 0.38550620079040526,
      "num_input_tokens_seen": 18387382016,
      "step": 64650,
      "train_runtime": 628674.9517,
      "train_tokens_per_second": 29247.836
    },
    {
      "epoch": 2.287797402516989,
      "grad_norm": 0.61328125,
      "learning_rate": 1.829906585287199e-05,
      "loss": 0.39180865287780764,
      "num_input_tokens_seen": 18390254336,
      "step": 64660,
      "train_runtime": 628772.6892,
      "train_tokens_per_second": 29247.858
    },
    {
      "epoch": 2.2881512227807175,
      "grad_norm": 0.609375,
      "learning_rate": 1.8297840466261302e-05,
      "loss": 0.38927016258239744,
      "num_input_tokens_seen": 18393092352,
      "step": 64670,
      "train_runtime": 628871.6133,
      "train_tokens_per_second": 29247.77
    },
    {
      "epoch": 2.2885050430444465,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8296615325789615e-05,
      "loss": 0.3878291845321655,
      "num_input_tokens_seen": 18395909632,
      "step": 64680,
      "train_runtime": 628965.0934,
      "train_tokens_per_second": 29247.902
    },
    {
      "epoch": 2.2888588633081755,
      "grad_norm": 0.6171875,
      "learning_rate": 1.829539043137454e-05,
      "loss": 0.387098503112793,
      "num_input_tokens_seen": 18398797824,
      "step": 64690,
      "train_runtime": 629062.4303,
      "train_tokens_per_second": 29247.968
    },
    {
      "epoch": 2.289212683571904,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8294165782933725e-05,
      "loss": 0.38688108921051023,
      "num_input_tokens_seen": 18401685376,
      "step": 64700,
      "train_runtime": 629163.5467,
      "train_tokens_per_second": 29247.857
    },
    {
      "epoch": 2.289566503835633,
      "grad_norm": 0.65234375,
      "learning_rate": 1.8292941380384852e-05,
      "loss": 0.38908910751342773,
      "num_input_tokens_seen": 18404513216,
      "step": 64710,
      "train_runtime": 629260.0233,
      "train_tokens_per_second": 29247.867
    },
    {
      "epoch": 2.2899203240993615,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8291717223645656e-05,
      "loss": 0.3844440460205078,
      "num_input_tokens_seen": 18407399104,
      "step": 64720,
      "train_runtime": 629360.1143,
      "train_tokens_per_second": 29247.801
    },
    {
      "epoch": 2.2902741443630905,
      "grad_norm": 0.625,
      "learning_rate": 1.8290493312633896e-05,
      "loss": 0.3866072177886963,
      "num_input_tokens_seen": 18410233344,
      "step": 64730,
      "train_runtime": 629456.4705,
      "train_tokens_per_second": 29247.826
    },
    {
      "epoch": 2.290627964626819,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8289269647267365e-05,
      "loss": 0.3871736526489258,
      "num_input_tokens_seen": 18413004224,
      "step": 64740,
      "train_runtime": 629550.1764,
      "train_tokens_per_second": 29247.874
    },
    {
      "epoch": 2.290981784890548,
      "grad_norm": 0.6484375,
      "learning_rate": 1.8288046227463916e-05,
      "loss": 0.3888375997543335,
      "num_input_tokens_seen": 18415917824,
      "step": 64750,
      "train_runtime": 629651.3246,
      "train_tokens_per_second": 29247.803
    },
    {
      "epoch": 2.2913356051542766,
      "grad_norm": 0.609375,
      "learning_rate": 1.828682305314142e-05,
      "loss": 0.38705532550811766,
      "num_input_tokens_seen": 18418796032,
      "step": 64760,
      "train_runtime": 629748.6961,
      "train_tokens_per_second": 29247.851
    },
    {
      "epoch": 2.2916894254180056,
      "grad_norm": 0.64453125,
      "learning_rate": 1.82856001242178e-05,
      "loss": 0.38695216178894043,
      "num_input_tokens_seen": 18421703296,
      "step": 64770,
      "train_runtime": 629850.934,
      "train_tokens_per_second": 29247.719
    },
    {
      "epoch": 2.292043245681734,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8284377440611006e-05,
      "loss": 0.38610775470733644,
      "num_input_tokens_seen": 18424564416,
      "step": 64780,
      "train_runtime": 629948.5526,
      "train_tokens_per_second": 29247.729
    },
    {
      "epoch": 2.292397065945463,
      "grad_norm": 0.58984375,
      "learning_rate": 1.828315500223904e-05,
      "loss": 0.3819850206375122,
      "num_input_tokens_seen": 18427471424,
      "step": 64790,
      "train_runtime": 630046.852,
      "train_tokens_per_second": 29247.78
    },
    {
      "epoch": 2.2927508862091917,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8281932809019927e-05,
      "loss": 0.38526899814605714,
      "num_input_tokens_seen": 18430335168,
      "step": 64800,
      "train_runtime": 630141.5097,
      "train_tokens_per_second": 29247.931
    },
    {
      "epoch": 2.2931047064729206,
      "grad_norm": 0.6328125,
      "learning_rate": 1.828071086087174e-05,
      "loss": 0.38569176197052,
      "num_input_tokens_seen": 18433147648,
      "step": 64810,
      "train_runtime": 630238.3758,
      "train_tokens_per_second": 29247.898
    },
    {
      "epoch": 2.293458526736649,
      "grad_norm": 0.6171875,
      "learning_rate": 1.827948915771259e-05,
      "loss": 0.38973560333251955,
      "num_input_tokens_seen": 18436049856,
      "step": 64820,
      "train_runtime": 630337.5761,
      "train_tokens_per_second": 29247.899
    },
    {
      "epoch": 2.293812347000378,
      "grad_norm": 0.6015625,
      "learning_rate": 1.827826769946063e-05,
      "loss": 0.3876962184906006,
      "num_input_tokens_seen": 18438903360,
      "step": 64830,
      "train_runtime": 630437.5369,
      "train_tokens_per_second": 29247.788
    },
    {
      "epoch": 2.294166167264107,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8277046486034035e-05,
      "loss": 0.3843767404556274,
      "num_input_tokens_seen": 18441781184,
      "step": 64840,
      "train_runtime": 630537.4013,
      "train_tokens_per_second": 29247.72
    },
    {
      "epoch": 2.2945199875278357,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8275825517351035e-05,
      "loss": 0.3874641418457031,
      "num_input_tokens_seen": 18444648704,
      "step": 64850,
      "train_runtime": 630636.1729,
      "train_tokens_per_second": 29247.686
    },
    {
      "epoch": 2.2948738077915642,
      "grad_norm": 0.58984375,
      "learning_rate": 1.827460479332989e-05,
      "loss": 0.38566668033599855,
      "num_input_tokens_seen": 18447546560,
      "step": 64860,
      "train_runtime": 630736.3539,
      "train_tokens_per_second": 29247.635
    },
    {
      "epoch": 2.2952276280552932,
      "grad_norm": 0.61328125,
      "learning_rate": 1.827338431388891e-05,
      "loss": 0.38759870529174806,
      "num_input_tokens_seen": 18450421440,
      "step": 64870,
      "train_runtime": 630838.9224,
      "train_tokens_per_second": 29247.437
    },
    {
      "epoch": 2.295581448319022,
      "grad_norm": 0.609375,
      "learning_rate": 1.8272164078946418e-05,
      "loss": 0.3854562520980835,
      "num_input_tokens_seen": 18453212928,
      "step": 64880,
      "train_runtime": 630930.6381,
      "train_tokens_per_second": 29247.61
    },
    {
      "epoch": 2.2959352685827508,
      "grad_norm": 0.63671875,
      "learning_rate": 1.82709440884208e-05,
      "loss": 0.38477783203125,
      "num_input_tokens_seen": 18456091264,
      "step": 64890,
      "train_runtime": 631029.7588,
      "train_tokens_per_second": 29247.577
    },
    {
      "epoch": 2.2962890888464798,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8269724342230474e-05,
      "loss": 0.38562769889831544,
      "num_input_tokens_seen": 18458993792,
      "step": 64900,
      "train_runtime": 631128.6558,
      "train_tokens_per_second": 29247.593
    },
    {
      "epoch": 2.2966429091102083,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8268504840293886e-05,
      "loss": 0.3923305034637451,
      "num_input_tokens_seen": 18461815168,
      "step": 64910,
      "train_runtime": 631223.0299,
      "train_tokens_per_second": 29247.689
    },
    {
      "epoch": 2.2969967293739373,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8267285582529535e-05,
      "loss": 0.3861689567565918,
      "num_input_tokens_seen": 18464662144,
      "step": 64920,
      "train_runtime": 631318.2256,
      "train_tokens_per_second": 29247.789
    },
    {
      "epoch": 2.297350549637666,
      "grad_norm": 0.6484375,
      "learning_rate": 1.826606656885594e-05,
      "loss": 0.38799750804901123,
      "num_input_tokens_seen": 18467464896,
      "step": 64930,
      "train_runtime": 631413.51,
      "train_tokens_per_second": 29247.814
    },
    {
      "epoch": 2.297704369901395,
      "grad_norm": 0.59765625,
      "learning_rate": 1.826484779919168e-05,
      "loss": 0.38740129470825196,
      "num_input_tokens_seen": 18470356672,
      "step": 64940,
      "train_runtime": 631513.9313,
      "train_tokens_per_second": 29247.742
    },
    {
      "epoch": 2.2980581901651234,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8263629273455353e-05,
      "loss": 0.388030481338501,
      "num_input_tokens_seen": 18473203840,
      "step": 64950,
      "train_runtime": 631614.4037,
      "train_tokens_per_second": 29247.597
    },
    {
      "epoch": 2.2984120104288523,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8262410991565605e-05,
      "loss": 0.3871011734008789,
      "num_input_tokens_seen": 18476093120,
      "step": 64960,
      "train_runtime": 631717.5783,
      "train_tokens_per_second": 29247.394
    },
    {
      "epoch": 2.298765830692581,
      "grad_norm": 0.6484375,
      "learning_rate": 1.826119295344112e-05,
      "loss": 0.38274321556091306,
      "num_input_tokens_seen": 18478951936,
      "step": 64970,
      "train_runtime": 631816.5242,
      "train_tokens_per_second": 29247.339
    },
    {
      "epoch": 2.29911965095631,
      "grad_norm": 0.59375,
      "learning_rate": 1.8259975159000613e-05,
      "loss": 0.3868141174316406,
      "num_input_tokens_seen": 18481807872,
      "step": 64980,
      "train_runtime": 631913.135,
      "train_tokens_per_second": 29247.387
    },
    {
      "epoch": 2.2994734712200384,
      "grad_norm": 0.59765625,
      "learning_rate": 1.825875760816284e-05,
      "loss": 0.3884878158569336,
      "num_input_tokens_seen": 18484662720,
      "step": 64990,
      "train_runtime": 632009.3297,
      "train_tokens_per_second": 29247.452
    },
    {
      "epoch": 2.2998272914837674,
      "grad_norm": 0.65625,
      "learning_rate": 1.8257540300846603e-05,
      "loss": 0.3876064300537109,
      "num_input_tokens_seen": 18487515328,
      "step": 65000,
      "train_runtime": 632109.7985,
      "train_tokens_per_second": 29247.316
    },
    {
      "epoch": 2.2998272914837674,
      "eval_loss": 0.3366485834121704,
      "eval_runtime": 8.4285,
      "eval_samples_per_second": 473.157,
      "eval_steps_per_second": 3.797,
      "num_input_tokens_seen": 18487515328,
      "step": 65000
    },
    {
      "epoch": 2.300181111747496,
      "grad_norm": 0.609375,
      "learning_rate": 1.825632323697073e-05,
      "loss": 0.38701891899108887,
      "num_input_tokens_seen": 18490398528,
      "step": 65010,
      "train_runtime": 632237.6177,
      "train_tokens_per_second": 29245.964
    },
    {
      "epoch": 2.300534932011225,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8255106416454093e-05,
      "loss": 0.3864434719085693,
      "num_input_tokens_seen": 18493255488,
      "step": 65020,
      "train_runtime": 632334.9735,
      "train_tokens_per_second": 29245.979
    },
    {
      "epoch": 2.300888752274954,
      "grad_norm": 0.63671875,
      "learning_rate": 1.82538898392156e-05,
      "loss": 0.3889376163482666,
      "num_input_tokens_seen": 18496091968,
      "step": 65030,
      "train_runtime": 632434.5849,
      "train_tokens_per_second": 29245.858
    },
    {
      "epoch": 2.3012425725386825,
      "grad_norm": 0.625,
      "learning_rate": 1.8252673505174206e-05,
      "loss": 0.3881181240081787,
      "num_input_tokens_seen": 18498931456,
      "step": 65040,
      "train_runtime": 632531.4566,
      "train_tokens_per_second": 29245.868
    },
    {
      "epoch": 2.3015963928024115,
      "grad_norm": 0.6328125,
      "learning_rate": 1.825145741424888e-05,
      "loss": 0.38629789352416993,
      "num_input_tokens_seen": 18501787392,
      "step": 65050,
      "train_runtime": 632631.6518,
      "train_tokens_per_second": 29245.75
    },
    {
      "epoch": 2.30195021306614,
      "grad_norm": 0.625,
      "learning_rate": 1.8250241566358662e-05,
      "loss": 0.3839745044708252,
      "num_input_tokens_seen": 18504658240,
      "step": 65060,
      "train_runtime": 632731.0523,
      "train_tokens_per_second": 29245.693
    },
    {
      "epoch": 2.302304033329869,
      "grad_norm": 0.609375,
      "learning_rate": 1.8249025961422598e-05,
      "loss": 0.38858942985534667,
      "num_input_tokens_seen": 18507565504,
      "step": 65070,
      "train_runtime": 632831.5762,
      "train_tokens_per_second": 29245.642
    },
    {
      "epoch": 2.3026578535935975,
      "grad_norm": 0.62890625,
      "learning_rate": 1.824781059935979e-05,
      "loss": 0.38565118312835694,
      "num_input_tokens_seen": 18510434112,
      "step": 65080,
      "train_runtime": 632931.5554,
      "train_tokens_per_second": 29245.554
    },
    {
      "epoch": 2.3030116738573265,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8246595480089376e-05,
      "loss": 0.38530526161193845,
      "num_input_tokens_seen": 18513277376,
      "step": 65090,
      "train_runtime": 633028.3542,
      "train_tokens_per_second": 29245.574
    },
    {
      "epoch": 2.303365494121055,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8245380603530526e-05,
      "loss": 0.3832041501998901,
      "num_input_tokens_seen": 18516184256,
      "step": 65100,
      "train_runtime": 633127.7731,
      "train_tokens_per_second": 29245.573
    },
    {
      "epoch": 2.303719314384784,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8244165969602455e-05,
      "loss": 0.3861818313598633,
      "num_input_tokens_seen": 18519012096,
      "step": 65110,
      "train_runtime": 633224.2649,
      "train_tokens_per_second": 29245.582
    },
    {
      "epoch": 2.3040731346485126,
      "grad_norm": 0.609375,
      "learning_rate": 1.824295157822441e-05,
      "loss": 0.38622608184814455,
      "num_input_tokens_seen": 18521816896,
      "step": 65120,
      "train_runtime": 633318.3804,
      "train_tokens_per_second": 29245.665
    },
    {
      "epoch": 2.3044269549122416,
      "grad_norm": 0.625,
      "learning_rate": 1.8241737429315674e-05,
      "loss": 0.3872580051422119,
      "num_input_tokens_seen": 18524689920,
      "step": 65130,
      "train_runtime": 633418.3835,
      "train_tokens_per_second": 29245.583
    },
    {
      "epoch": 2.30478077517597,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8240523522795575e-05,
      "loss": 0.3836192607879639,
      "num_input_tokens_seen": 18527584000,
      "step": 65140,
      "train_runtime": 633523.1267,
      "train_tokens_per_second": 29245.316
    },
    {
      "epoch": 2.305134595439699,
      "grad_norm": 0.640625,
      "learning_rate": 1.8239309858583466e-05,
      "loss": 0.38173096179962157,
      "num_input_tokens_seen": 18530475648,
      "step": 65150,
      "train_runtime": 633621.0424,
      "train_tokens_per_second": 29245.36
    },
    {
      "epoch": 2.3054884157034277,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8238096436598755e-05,
      "loss": 0.38759703636169435,
      "num_input_tokens_seen": 18533333888,
      "step": 65160,
      "train_runtime": 633719.5944,
      "train_tokens_per_second": 29245.322
    },
    {
      "epoch": 2.3058422359671567,
      "grad_norm": 0.625,
      "learning_rate": 1.8236883256760882e-05,
      "loss": 0.3836358070373535,
      "num_input_tokens_seen": 18536214272,
      "step": 65170,
      "train_runtime": 633819.6206,
      "train_tokens_per_second": 29245.252
    },
    {
      "epoch": 2.3061960562308856,
      "grad_norm": 0.6640625,
      "learning_rate": 1.8235670318989305e-05,
      "loss": 0.3821556806564331,
      "num_input_tokens_seen": 18539085120,
      "step": 65180,
      "train_runtime": 633917.3366,
      "train_tokens_per_second": 29245.272
    },
    {
      "epoch": 2.306549876494614,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8234457623203548e-05,
      "loss": 0.38454060554504393,
      "num_input_tokens_seen": 18541982144,
      "step": 65190,
      "train_runtime": 634015.2402,
      "train_tokens_per_second": 29245.326
    },
    {
      "epoch": 2.3069036967583427,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8233245169323163e-05,
      "loss": 0.3883782386779785,
      "num_input_tokens_seen": 18544886080,
      "step": 65200,
      "train_runtime": 634116.1726,
      "train_tokens_per_second": 29245.25
    },
    {
      "epoch": 2.3072575170220717,
      "grad_norm": 0.58984375,
      "learning_rate": 1.823203295726772e-05,
      "loss": 0.38774886131286623,
      "num_input_tokens_seen": 18547720512,
      "step": 65210,
      "train_runtime": 634210.7352,
      "train_tokens_per_second": 29245.359
    },
    {
      "epoch": 2.3076113372858007,
      "grad_norm": 0.62109375,
      "learning_rate": 1.823082098695686e-05,
      "loss": 0.3898928642272949,
      "num_input_tokens_seen": 18550568192,
      "step": 65220,
      "train_runtime": 634310.5853,
      "train_tokens_per_second": 29245.245
    },
    {
      "epoch": 2.3079651575495292,
      "grad_norm": 0.59375,
      "learning_rate": 1.8229609258310233e-05,
      "loss": 0.3822164535522461,
      "num_input_tokens_seen": 18553420544,
      "step": 65230,
      "train_runtime": 634408.2978,
      "train_tokens_per_second": 29245.236
    },
    {
      "epoch": 2.3083189778132582,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8228397771247542e-05,
      "loss": 0.3885348320007324,
      "num_input_tokens_seen": 18556261376,
      "step": 65240,
      "train_runtime": 634506.3229,
      "train_tokens_per_second": 29245.195
    },
    {
      "epoch": 2.3086727980769868,
      "grad_norm": 0.61328125,
      "learning_rate": 1.822718652568852e-05,
      "loss": 0.38760967254638673,
      "num_input_tokens_seen": 18559124864,
      "step": 65250,
      "train_runtime": 634601.0271,
      "train_tokens_per_second": 29245.343
    },
    {
      "epoch": 2.3090266183407158,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8225975521552944e-05,
      "loss": 0.38417651653289797,
      "num_input_tokens_seen": 18561957888,
      "step": 65260,
      "train_runtime": 634696.8587,
      "train_tokens_per_second": 29245.391
    },
    {
      "epoch": 2.3093804386044443,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8224764758760622e-05,
      "loss": 0.39082984924316405,
      "num_input_tokens_seen": 18564790272,
      "step": 65270,
      "train_runtime": 634794.0302,
      "train_tokens_per_second": 29245.376
    },
    {
      "epoch": 2.3097342588681733,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8223554237231398e-05,
      "loss": 0.38508620262146,
      "num_input_tokens_seen": 18567663936,
      "step": 65280,
      "train_runtime": 634894.2937,
      "train_tokens_per_second": 29245.284
    },
    {
      "epoch": 2.310088079131902,
      "grad_norm": 0.6328125,
      "learning_rate": 1.822234395688517e-05,
      "loss": 0.3839262008666992,
      "num_input_tokens_seen": 18570522944,
      "step": 65290,
      "train_runtime": 634993.4044,
      "train_tokens_per_second": 29245.222
    },
    {
      "epoch": 2.310441899395631,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8221133917641845e-05,
      "loss": 0.38491172790527345,
      "num_input_tokens_seen": 18573323968,
      "step": 65300,
      "train_runtime": 635085.8851,
      "train_tokens_per_second": 29245.374
    },
    {
      "epoch": 2.3107957196593594,
      "grad_norm": 0.609375,
      "learning_rate": 1.821992411942139e-05,
      "loss": 0.38651888370513915,
      "num_input_tokens_seen": 18576142400,
      "step": 65310,
      "train_runtime": 635181.5934,
      "train_tokens_per_second": 29245.404
    },
    {
      "epoch": 2.3111495399230884,
      "grad_norm": 0.609375,
      "learning_rate": 1.8218714562143804e-05,
      "loss": 0.385313892364502,
      "num_input_tokens_seen": 18578974528,
      "step": 65320,
      "train_runtime": 635277.6333,
      "train_tokens_per_second": 29245.441
    },
    {
      "epoch": 2.3115033601868173,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8217505245729113e-05,
      "loss": 0.3890071392059326,
      "num_input_tokens_seen": 18581874560,
      "step": 65330,
      "train_runtime": 635377.6016,
      "train_tokens_per_second": 29245.404
    },
    {
      "epoch": 2.311857180450546,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8216296170097395e-05,
      "loss": 0.38903064727783204,
      "num_input_tokens_seen": 18584736768,
      "step": 65340,
      "train_runtime": 635478.7432,
      "train_tokens_per_second": 29245.253
    },
    {
      "epoch": 2.3122110007142744,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8215087335168758e-05,
      "loss": 0.3895043134689331,
      "num_input_tokens_seen": 18587584192,
      "step": 65350,
      "train_runtime": 635575.2941,
      "train_tokens_per_second": 29245.291
    },
    {
      "epoch": 2.3125648209780034,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8213878740863342e-05,
      "loss": 0.3860654592514038,
      "num_input_tokens_seen": 18590437696,
      "step": 65360,
      "train_runtime": 635672.8997,
      "train_tokens_per_second": 29245.289
    },
    {
      "epoch": 2.3129186412417324,
      "grad_norm": 0.6171875,
      "learning_rate": 1.821267038710133e-05,
      "loss": 0.39289758205413816,
      "num_input_tokens_seen": 18593262144,
      "step": 65370,
      "train_runtime": 635766.4709,
      "train_tokens_per_second": 29245.427
    },
    {
      "epoch": 2.313272461505461,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8211462273802946e-05,
      "loss": 0.388733434677124,
      "num_input_tokens_seen": 18596127232,
      "step": 65380,
      "train_runtime": 635866.6835,
      "train_tokens_per_second": 29245.324
    },
    {
      "epoch": 2.31362628176919,
      "grad_norm": 0.63671875,
      "learning_rate": 1.821025440088844e-05,
      "loss": 0.3889653444290161,
      "num_input_tokens_seen": 18598984320,
      "step": 65390,
      "train_runtime": 635965.4346,
      "train_tokens_per_second": 29245.275
    },
    {
      "epoch": 2.3139801020329185,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8209046768278115e-05,
      "loss": 0.3817785978317261,
      "num_input_tokens_seen": 18601814464,
      "step": 65400,
      "train_runtime": 636062.8022,
      "train_tokens_per_second": 29245.248
    },
    {
      "epoch": 2.3143339222966475,
      "grad_norm": 0.6484375,
      "learning_rate": 1.8207839375892292e-05,
      "loss": 0.391879940032959,
      "num_input_tokens_seen": 18604680256,
      "step": 65410,
      "train_runtime": 636161.922,
      "train_tokens_per_second": 29245.196
    },
    {
      "epoch": 2.314687742560376,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8206632223651342e-05,
      "loss": 0.3825816631317139,
      "num_input_tokens_seen": 18607483072,
      "step": 65420,
      "train_runtime": 636255.1977,
      "train_tokens_per_second": 29245.314
    },
    {
      "epoch": 2.315041562824105,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8205425311475674e-05,
      "loss": 0.3869927167892456,
      "num_input_tokens_seen": 18610298048,
      "step": 65430,
      "train_runtime": 636349.6136,
      "train_tokens_per_second": 29245.399
    },
    {
      "epoch": 2.3153953830878335,
      "grad_norm": 0.60546875,
      "learning_rate": 1.820421863928572e-05,
      "loss": 0.3871757507324219,
      "num_input_tokens_seen": 18613147456,
      "step": 65440,
      "train_runtime": 636449.7369,
      "train_tokens_per_second": 29245.275
    },
    {
      "epoch": 2.3157492033515625,
      "grad_norm": 0.625,
      "learning_rate": 1.8203012207001965e-05,
      "loss": 0.38469738960266114,
      "num_input_tokens_seen": 18616037376,
      "step": 65450,
      "train_runtime": 636551.7329,
      "train_tokens_per_second": 29245.129
    },
    {
      "epoch": 2.316103023615291,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8201806014544923e-05,
      "loss": 0.3902743816375732,
      "num_input_tokens_seen": 18618861184,
      "step": 65460,
      "train_runtime": 636644.9673,
      "train_tokens_per_second": 29245.281
    },
    {
      "epoch": 2.31645684387902,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8200600061835143e-05,
      "loss": 0.3905118465423584,
      "num_input_tokens_seen": 18621689664,
      "step": 65470,
      "train_runtime": 636741.4679,
      "train_tokens_per_second": 29245.291
    },
    {
      "epoch": 2.3168106641427486,
      "grad_norm": 0.609375,
      "learning_rate": 1.8199394348793213e-05,
      "loss": 0.3878653526306152,
      "num_input_tokens_seen": 18624549248,
      "step": 65480,
      "train_runtime": 636838.7872,
      "train_tokens_per_second": 29245.312
    },
    {
      "epoch": 2.3171644844064776,
      "grad_norm": 0.609375,
      "learning_rate": 1.8198188875339767e-05,
      "loss": 0.3901439428329468,
      "num_input_tokens_seen": 18627419968,
      "step": 65490,
      "train_runtime": 636935.7204,
      "train_tokens_per_second": 29245.369
    },
    {
      "epoch": 2.317518304670206,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8196983641395458e-05,
      "loss": 0.3875009298324585,
      "num_input_tokens_seen": 18630259328,
      "step": 65500,
      "train_runtime": 637030.9285,
      "train_tokens_per_second": 29245.455
    },
    {
      "epoch": 2.317872124933935,
      "grad_norm": 0.6015625,
      "learning_rate": 1.819577864688099e-05,
      "loss": 0.3852522611618042,
      "num_input_tokens_seen": 18633127296,
      "step": 65510,
      "train_runtime": 637127.8564,
      "train_tokens_per_second": 29245.507
    },
    {
      "epoch": 2.318225945197664,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8194573891717096e-05,
      "loss": 0.38724446296691895,
      "num_input_tokens_seen": 18635986944,
      "step": 65520,
      "train_runtime": 637224.3972,
      "train_tokens_per_second": 29245.564
    },
    {
      "epoch": 2.3185797654613927,
      "grad_norm": 0.609375,
      "learning_rate": 1.819336937582455e-05,
      "loss": 0.38571667671203613,
      "num_input_tokens_seen": 18638905088,
      "step": 65530,
      "train_runtime": 637327.0636,
      "train_tokens_per_second": 29245.432
    },
    {
      "epoch": 2.3189335857251216,
      "grad_norm": 0.640625,
      "learning_rate": 1.8192165099124165e-05,
      "loss": 0.3846283435821533,
      "num_input_tokens_seen": 18641765184,
      "step": 65540,
      "train_runtime": 637425.5079,
      "train_tokens_per_second": 29245.402
    },
    {
      "epoch": 2.31928740598885,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8190961061536783e-05,
      "loss": 0.38587141036987305,
      "num_input_tokens_seen": 18644563072,
      "step": 65550,
      "train_runtime": 637519.7526,
      "train_tokens_per_second": 29245.467
    },
    {
      "epoch": 2.319641226252579,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8189757262983285e-05,
      "loss": 0.3852789878845215,
      "num_input_tokens_seen": 18647349824,
      "step": 65560,
      "train_runtime": 637612.4318,
      "train_tokens_per_second": 29245.587
    },
    {
      "epoch": 2.3199950465163077,
      "grad_norm": 0.63671875,
      "learning_rate": 1.818855370338459e-05,
      "loss": 0.38773860931396487,
      "num_input_tokens_seen": 18650167424,
      "step": 65570,
      "train_runtime": 637707.8408,
      "train_tokens_per_second": 29245.63
    },
    {
      "epoch": 2.3203488667800367,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8187350382661664e-05,
      "loss": 0.38150548934936523,
      "num_input_tokens_seen": 18653026496,
      "step": 65580,
      "train_runtime": 637805.0661,
      "train_tokens_per_second": 29245.654
    },
    {
      "epoch": 2.3207026870437653,
      "grad_norm": 0.62109375,
      "learning_rate": 1.818614730073549e-05,
      "loss": 0.3881141901016235,
      "num_input_tokens_seen": 18655831808,
      "step": 65590,
      "train_runtime": 637898.5944,
      "train_tokens_per_second": 29245.764
    },
    {
      "epoch": 2.3210565073074942,
      "grad_norm": 0.64453125,
      "learning_rate": 1.81849444575271e-05,
      "loss": 0.38222854137420653,
      "num_input_tokens_seen": 18658701440,
      "step": 65600,
      "train_runtime": 637996.3308,
      "train_tokens_per_second": 29245.782
    },
    {
      "epoch": 2.321410327571223,
      "grad_norm": 0.62109375,
      "learning_rate": 1.818374185295756e-05,
      "loss": 0.3893751621246338,
      "num_input_tokens_seen": 18661548544,
      "step": 65610,
      "train_runtime": 638090.8655,
      "train_tokens_per_second": 29245.911
    },
    {
      "epoch": 2.3217641478349518,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8182539486947967e-05,
      "loss": 0.38494224548339845,
      "num_input_tokens_seen": 18664386688,
      "step": 65620,
      "train_runtime": 638185.4847,
      "train_tokens_per_second": 29246.022
    },
    {
      "epoch": 2.3221179680986803,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8181337359419467e-05,
      "loss": 0.3871985912322998,
      "num_input_tokens_seen": 18667199936,
      "step": 65630,
      "train_runtime": 638280.192,
      "train_tokens_per_second": 29246.09
    },
    {
      "epoch": 2.3224717883624093,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8180135470293233e-05,
      "loss": 0.3882903099060059,
      "num_input_tokens_seen": 18670070528,
      "step": 65640,
      "train_runtime": 638380.8818,
      "train_tokens_per_second": 29245.974
    },
    {
      "epoch": 2.322825608626138,
      "grad_norm": 0.6171875,
      "learning_rate": 1.817893381949048e-05,
      "loss": 0.3839714527130127,
      "num_input_tokens_seen": 18672928768,
      "step": 65650,
      "train_runtime": 638477.9911,
      "train_tokens_per_second": 29246.002
    },
    {
      "epoch": 2.323179428889867,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8177732406932444e-05,
      "loss": 0.38971834182739257,
      "num_input_tokens_seen": 18675744448,
      "step": 65660,
      "train_runtime": 638570.83,
      "train_tokens_per_second": 29246.16
    },
    {
      "epoch": 2.323533249153596,
      "grad_norm": 0.625,
      "learning_rate": 1.8176531232540427e-05,
      "loss": 0.389362096786499,
      "num_input_tokens_seen": 18678559424,
      "step": 65670,
      "train_runtime": 638665.0724,
      "train_tokens_per_second": 29246.252
    },
    {
      "epoch": 2.3238870694173244,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8175330296235734e-05,
      "loss": 0.3869276762008667,
      "num_input_tokens_seen": 18681412992,
      "step": 65680,
      "train_runtime": 638762.8011,
      "train_tokens_per_second": 29246.244
    },
    {
      "epoch": 2.324240889681053,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8174129597939733e-05,
      "loss": 0.38761887550354,
      "num_input_tokens_seen": 18684223744,
      "step": 65690,
      "train_runtime": 638857.9893,
      "train_tokens_per_second": 29246.286
    },
    {
      "epoch": 2.324594709944782,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8172929137573813e-05,
      "loss": 0.3872941017150879,
      "num_input_tokens_seen": 18687058176,
      "step": 65700,
      "train_runtime": 638956.4801,
      "train_tokens_per_second": 29246.214
    },
    {
      "epoch": 2.324948530208511,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8171728915059404e-05,
      "loss": 0.38887743949890136,
      "num_input_tokens_seen": 18689947136,
      "step": 65710,
      "train_runtime": 639055.4016,
      "train_tokens_per_second": 29246.208
    },
    {
      "epoch": 2.3253023504722394,
      "grad_norm": 0.61328125,
      "learning_rate": 1.817052893031798e-05,
      "loss": 0.3858642578125,
      "num_input_tokens_seen": 18692801856,
      "step": 65720,
      "train_runtime": 639151.0217,
      "train_tokens_per_second": 29246.299
    },
    {
      "epoch": 2.3256561707359684,
      "grad_norm": 0.609375,
      "learning_rate": 1.816932918327103e-05,
      "loss": 0.38456680774688723,
      "num_input_tokens_seen": 18695612672,
      "step": 65730,
      "train_runtime": 639246.767,
      "train_tokens_per_second": 29246.316
    },
    {
      "epoch": 2.326009990999697,
      "grad_norm": 0.65234375,
      "learning_rate": 1.8168129673840103e-05,
      "loss": 0.3868723392486572,
      "num_input_tokens_seen": 18698471744,
      "step": 65740,
      "train_runtime": 639344.4729,
      "train_tokens_per_second": 29246.318
    },
    {
      "epoch": 2.326363811263426,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8166930401946775e-05,
      "loss": 0.38702569007873533,
      "num_input_tokens_seen": 18701272576,
      "step": 65750,
      "train_runtime": 639437.5184,
      "train_tokens_per_second": 29246.442
    },
    {
      "epoch": 2.3267176315271545,
      "grad_norm": 0.6171875,
      "learning_rate": 1.816573136751265e-05,
      "loss": 0.38517940044403076,
      "num_input_tokens_seen": 18704139328,
      "step": 65760,
      "train_runtime": 639533.5852,
      "train_tokens_per_second": 29246.532
    },
    {
      "epoch": 2.3270714517908835,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8164532570459383e-05,
      "loss": 0.3821824073791504,
      "num_input_tokens_seen": 18707006848,
      "step": 65770,
      "train_runtime": 639631.086,
      "train_tokens_per_second": 29246.557
    },
    {
      "epoch": 2.327425272054612,
      "grad_norm": 0.64453125,
      "learning_rate": 1.8163334010708654e-05,
      "loss": 0.38336355686187745,
      "num_input_tokens_seen": 18709892928,
      "step": 65780,
      "train_runtime": 639728.5832,
      "train_tokens_per_second": 29246.611
    },
    {
      "epoch": 2.327779092318341,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8162135688182183e-05,
      "loss": 0.38662965297698976,
      "num_input_tokens_seen": 18712777408,
      "step": 65790,
      "train_runtime": 639827.0812,
      "train_tokens_per_second": 29246.617
    },
    {
      "epoch": 2.3281329125820696,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8160937602801726e-05,
      "loss": 0.3911834478378296,
      "num_input_tokens_seen": 18715626432,
      "step": 65800,
      "train_runtime": 639925.2329,
      "train_tokens_per_second": 29246.583
    },
    {
      "epoch": 2.3284867328457985,
      "grad_norm": 0.640625,
      "learning_rate": 1.815973975448908e-05,
      "loss": 0.38278281688690186,
      "num_input_tokens_seen": 18718445888,
      "step": 65810,
      "train_runtime": 640021.4856,
      "train_tokens_per_second": 29246.59
    },
    {
      "epoch": 2.328840553109527,
      "grad_norm": 0.6484375,
      "learning_rate": 1.8158542143166065e-05,
      "loss": 0.38755595684051514,
      "num_input_tokens_seen": 18721253184,
      "step": 65820,
      "train_runtime": 640116.7987,
      "train_tokens_per_second": 29246.621
    },
    {
      "epoch": 2.329194373373256,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8157344768754553e-05,
      "loss": 0.3937510013580322,
      "num_input_tokens_seen": 18724112320,
      "step": 65830,
      "train_runtime": 640214.1708,
      "train_tokens_per_second": 29246.638
    },
    {
      "epoch": 2.3295481936369846,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8156147631176442e-05,
      "loss": 0.38613576889038087,
      "num_input_tokens_seen": 18726930880,
      "step": 65840,
      "train_runtime": 640308.373,
      "train_tokens_per_second": 29246.737
    },
    {
      "epoch": 2.3299020139007136,
      "grad_norm": 0.640625,
      "learning_rate": 1.815495073035367e-05,
      "loss": 0.3827350616455078,
      "num_input_tokens_seen": 18729769408,
      "step": 65850,
      "train_runtime": 640404.8797,
      "train_tokens_per_second": 29246.762
    },
    {
      "epoch": 2.3302558341644426,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8153754066208205e-05,
      "loss": 0.3805245876312256,
      "num_input_tokens_seen": 18732633152,
      "step": 65860,
      "train_runtime": 640501.7894,
      "train_tokens_per_second": 29246.808
    },
    {
      "epoch": 2.330609654428171,
      "grad_norm": 0.6171875,
      "learning_rate": 1.815255763866206e-05,
      "loss": 0.3851560354232788,
      "num_input_tokens_seen": 18735486784,
      "step": 65870,
      "train_runtime": 640599.2072,
      "train_tokens_per_second": 29246.815
    },
    {
      "epoch": 2.3309634746919,
      "grad_norm": 0.62890625,
      "learning_rate": 1.815136144763728e-05,
      "loss": 0.3921637058258057,
      "num_input_tokens_seen": 18738361472,
      "step": 65880,
      "train_runtime": 640698.4072,
      "train_tokens_per_second": 29246.774
    },
    {
      "epoch": 2.3313172949556287,
      "grad_norm": 0.625,
      "learning_rate": 1.815016549305594e-05,
      "loss": 0.3898458480834961,
      "num_input_tokens_seen": 18741237696,
      "step": 65890,
      "train_runtime": 640800.4233,
      "train_tokens_per_second": 29246.606
    },
    {
      "epoch": 2.3316711152193577,
      "grad_norm": 0.5859375,
      "learning_rate": 1.8148969774840164e-05,
      "loss": 0.3838940143585205,
      "num_input_tokens_seen": 18744061760,
      "step": 65900,
      "train_runtime": 640893.4113,
      "train_tokens_per_second": 29246.769
    },
    {
      "epoch": 2.332024935483086,
      "grad_norm": 0.60546875,
      "learning_rate": 1.81477742929121e-05,
      "loss": 0.38811526298522947,
      "num_input_tokens_seen": 18746900800,
      "step": 65910,
      "train_runtime": 640988.6602,
      "train_tokens_per_second": 29246.853
    },
    {
      "epoch": 2.332378755746815,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8146579047193934e-05,
      "loss": 0.3865293025970459,
      "num_input_tokens_seen": 18749736960,
      "step": 65920,
      "train_runtime": 641083.4656,
      "train_tokens_per_second": 29246.951
    },
    {
      "epoch": 2.3327325760105437,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8145384037607897e-05,
      "loss": 0.39019017219543456,
      "num_input_tokens_seen": 18752535488,
      "step": 65930,
      "train_runtime": 641177.5922,
      "train_tokens_per_second": 29247.023
    },
    {
      "epoch": 2.3330863962742727,
      "grad_norm": 0.625,
      "learning_rate": 1.814418926407624e-05,
      "loss": 0.38755249977111816,
      "num_input_tokens_seen": 18755344768,
      "step": 65940,
      "train_runtime": 641274.6151,
      "train_tokens_per_second": 29246.978
    },
    {
      "epoch": 2.3334402165380013,
      "grad_norm": 0.609375,
      "learning_rate": 1.814299472652127e-05,
      "loss": 0.3913054227828979,
      "num_input_tokens_seen": 18758153088,
      "step": 65950,
      "train_runtime": 641369.1982,
      "train_tokens_per_second": 29247.044
    },
    {
      "epoch": 2.3337940368017303,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8141800424865312e-05,
      "loss": 0.38626272678375245,
      "num_input_tokens_seen": 18761029760,
      "step": 65960,
      "train_runtime": 641471.081,
      "train_tokens_per_second": 29246.883
    },
    {
      "epoch": 2.334147857065459,
      "grad_norm": 0.62890625,
      "learning_rate": 1.814060635903073e-05,
      "loss": 0.3903139352798462,
      "num_input_tokens_seen": 18763859392,
      "step": 65970,
      "train_runtime": 641566.7502,
      "train_tokens_per_second": 29246.932
    },
    {
      "epoch": 2.334501677329188,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8139412528939936e-05,
      "loss": 0.3872283220291138,
      "num_input_tokens_seen": 18766739456,
      "step": 65980,
      "train_runtime": 641663.7352,
      "train_tokens_per_second": 29247.0
    },
    {
      "epoch": 2.3348554975929163,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8138218934515365e-05,
      "loss": 0.38320231437683105,
      "num_input_tokens_seen": 18769582080,
      "step": 65990,
      "train_runtime": 641758.1845,
      "train_tokens_per_second": 29247.125
    },
    {
      "epoch": 2.3352093178566453,
      "grad_norm": 0.625,
      "learning_rate": 1.8137025575679486e-05,
      "loss": 0.38507232666015623,
      "num_input_tokens_seen": 18772463872,
      "step": 66000,
      "train_runtime": 641859.2814,
      "train_tokens_per_second": 29247.009
    },
    {
      "epoch": 2.3355631381203743,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8135832452354816e-05,
      "loss": 0.38588366508483884,
      "num_input_tokens_seen": 18775306816,
      "step": 66010,
      "train_runtime": 641958.3687,
      "train_tokens_per_second": 29246.923
    },
    {
      "epoch": 2.335916958384103,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8134639564463902e-05,
      "loss": 0.3845822811126709,
      "num_input_tokens_seen": 18778154816,
      "step": 66020,
      "train_runtime": 642053.3225,
      "train_tokens_per_second": 29247.033
    },
    {
      "epoch": 2.3362707786478314,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8133446911929316e-05,
      "loss": 0.3836571216583252,
      "num_input_tokens_seen": 18781012096,
      "step": 66030,
      "train_runtime": 642151.9568,
      "train_tokens_per_second": 29246.99
    },
    {
      "epoch": 2.3366245989115604,
      "grad_norm": 0.6015625,
      "learning_rate": 1.813225449467369e-05,
      "loss": 0.3847710132598877,
      "num_input_tokens_seen": 18783775872,
      "step": 66040,
      "train_runtime": 642245.0448,
      "train_tokens_per_second": 29247.055
    },
    {
      "epoch": 2.3369784191752894,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8131062312619663e-05,
      "loss": 0.38940436840057374,
      "num_input_tokens_seen": 18786629696,
      "step": 66050,
      "train_runtime": 642344.5377,
      "train_tokens_per_second": 29246.967
    },
    {
      "epoch": 2.337332239439018,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8129870365689933e-05,
      "loss": 0.3868520736694336,
      "num_input_tokens_seen": 18789471104,
      "step": 66060,
      "train_runtime": 642442.6671,
      "train_tokens_per_second": 29246.923
    },
    {
      "epoch": 2.337686059702747,
      "grad_norm": 0.625,
      "learning_rate": 1.812867865380722e-05,
      "loss": 0.38232736587524413,
      "num_input_tokens_seen": 18792219776,
      "step": 66070,
      "train_runtime": 642532.2414,
      "train_tokens_per_second": 29247.123
    },
    {
      "epoch": 2.3380398799664754,
      "grad_norm": 0.64453125,
      "learning_rate": 1.8127487176894286e-05,
      "loss": 0.3905600070953369,
      "num_input_tokens_seen": 18795002688,
      "step": 66080,
      "train_runtime": 642626.7007,
      "train_tokens_per_second": 29247.155
    },
    {
      "epoch": 2.3383937002302044,
      "grad_norm": 0.625,
      "learning_rate": 1.8126295934873926e-05,
      "loss": 0.3846308708190918,
      "num_input_tokens_seen": 18797805056,
      "step": 66090,
      "train_runtime": 642721.4829,
      "train_tokens_per_second": 29247.202
    },
    {
      "epoch": 2.338747520493933,
      "grad_norm": 0.609375,
      "learning_rate": 1.8125104927668964e-05,
      "loss": 0.38696794509887694,
      "num_input_tokens_seen": 18800643456,
      "step": 66100,
      "train_runtime": 642819.4424,
      "train_tokens_per_second": 29247.161
    },
    {
      "epoch": 2.339101340757662,
      "grad_norm": 0.59375,
      "learning_rate": 1.8123914155202273e-05,
      "loss": 0.3825960636138916,
      "num_input_tokens_seen": 18803516992,
      "step": 66110,
      "train_runtime": 642918.9863,
      "train_tokens_per_second": 29247.102
    },
    {
      "epoch": 2.3394551610213905,
      "grad_norm": 0.609375,
      "learning_rate": 1.8122723617396757e-05,
      "loss": 0.3834648609161377,
      "num_input_tokens_seen": 18806350144,
      "step": 66120,
      "train_runtime": 643016.338,
      "train_tokens_per_second": 29247.08
    },
    {
      "epoch": 2.3398089812851195,
      "grad_norm": 0.64453125,
      "learning_rate": 1.8121533314175345e-05,
      "loss": 0.385701847076416,
      "num_input_tokens_seen": 18809254528,
      "step": 66130,
      "train_runtime": 643114.8653,
      "train_tokens_per_second": 29247.115
    },
    {
      "epoch": 2.340162801548848,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8120343245461015e-05,
      "loss": 0.38894429206848147,
      "num_input_tokens_seen": 18812083584,
      "step": 66140,
      "train_runtime": 643209.6227,
      "train_tokens_per_second": 29247.205
    },
    {
      "epoch": 2.340516621812577,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8119153411176775e-05,
      "loss": 0.38922088146209716,
      "num_input_tokens_seen": 18814965056,
      "step": 66150,
      "train_runtime": 643308.0302,
      "train_tokens_per_second": 29247.21
    },
    {
      "epoch": 2.3408704420763056,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8117963811245663e-05,
      "loss": 0.39129436016082764,
      "num_input_tokens_seen": 18817774016,
      "step": 66160,
      "train_runtime": 643404.259,
      "train_tokens_per_second": 29247.202
    },
    {
      "epoch": 2.3412242623400346,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8116774445590763e-05,
      "loss": 0.3829531192779541,
      "num_input_tokens_seen": 18820576064,
      "step": 66170,
      "train_runtime": 643498.3129,
      "train_tokens_per_second": 29247.281
    },
    {
      "epoch": 2.341578082603763,
      "grad_norm": 0.6171875,
      "learning_rate": 1.811558531413519e-05,
      "loss": 0.3856523036956787,
      "num_input_tokens_seen": 18823359872,
      "step": 66180,
      "train_runtime": 643593.1088,
      "train_tokens_per_second": 29247.299
    },
    {
      "epoch": 2.341931902867492,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8114396416802094e-05,
      "loss": 0.3844825267791748,
      "num_input_tokens_seen": 18826234368,
      "step": 66190,
      "train_runtime": 643692.166,
      "train_tokens_per_second": 29247.263
    },
    {
      "epoch": 2.342285723131221,
      "grad_norm": 0.58984375,
      "learning_rate": 1.8113207753514652e-05,
      "loss": 0.38792896270751953,
      "num_input_tokens_seen": 18829086336,
      "step": 66200,
      "train_runtime": 643788.881,
      "train_tokens_per_second": 29247.3
    },
    {
      "epoch": 2.3426395433949496,
      "grad_norm": 0.609375,
      "learning_rate": 1.8112019324196088e-05,
      "loss": 0.39226531982421875,
      "num_input_tokens_seen": 18831936704,
      "step": 66210,
      "train_runtime": 643885.464,
      "train_tokens_per_second": 29247.339
    },
    {
      "epoch": 2.3429933636586786,
      "grad_norm": 0.60546875,
      "learning_rate": 1.811083112876966e-05,
      "loss": 0.3846435546875,
      "num_input_tokens_seen": 18834722496,
      "step": 66220,
      "train_runtime": 643977.9275,
      "train_tokens_per_second": 29247.466
    },
    {
      "epoch": 2.343347183922407,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8109643167158653e-05,
      "loss": 0.38691723346710205,
      "num_input_tokens_seen": 18837561088,
      "step": 66230,
      "train_runtime": 644074.8959,
      "train_tokens_per_second": 29247.47
    },
    {
      "epoch": 2.343701004186136,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8108455439286402e-05,
      "loss": 0.3878159999847412,
      "num_input_tokens_seen": 18840350976,
      "step": 66240,
      "train_runtime": 644167.466,
      "train_tokens_per_second": 29247.598
    },
    {
      "epoch": 2.3440548244498647,
      "grad_norm": 0.625,
      "learning_rate": 1.8107267945076254e-05,
      "loss": 0.38343660831451415,
      "num_input_tokens_seen": 18843199936,
      "step": 66250,
      "train_runtime": 644265.0498,
      "train_tokens_per_second": 29247.59
    },
    {
      "epoch": 2.3444086447135937,
      "grad_norm": 0.62109375,
      "learning_rate": 1.810608068445162e-05,
      "loss": 0.38767142295837403,
      "num_input_tokens_seen": 18846032704,
      "step": 66260,
      "train_runtime": 644359.0772,
      "train_tokens_per_second": 29247.718
    },
    {
      "epoch": 2.344762464977322,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8104893657335918e-05,
      "loss": 0.3838646411895752,
      "num_input_tokens_seen": 18848898560,
      "step": 66270,
      "train_runtime": 644456.4573,
      "train_tokens_per_second": 29247.746
    },
    {
      "epoch": 2.345116285241051,
      "grad_norm": 0.63671875,
      "learning_rate": 1.810370686365262e-05,
      "loss": 0.3861142635345459,
      "num_input_tokens_seen": 18851771072,
      "step": 66280,
      "train_runtime": 644556.4751,
      "train_tokens_per_second": 29247.664
    },
    {
      "epoch": 2.3454701055047797,
      "grad_norm": 0.6015625,
      "learning_rate": 1.810252030332523e-05,
      "loss": 0.38788437843322754,
      "num_input_tokens_seen": 18854587648,
      "step": 66290,
      "train_runtime": 644653.0426,
      "train_tokens_per_second": 29247.652
    },
    {
      "epoch": 2.3458239257685087,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8101333976277282e-05,
      "loss": 0.38442020416259765,
      "num_input_tokens_seen": 18857386176,
      "step": 66300,
      "train_runtime": 644748.6271,
      "train_tokens_per_second": 29247.656
    },
    {
      "epoch": 2.3461777460322373,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8100147882432346e-05,
      "loss": 0.3865343570709229,
      "num_input_tokens_seen": 18860248448,
      "step": 66310,
      "train_runtime": 644847.002,
      "train_tokens_per_second": 29247.633
    },
    {
      "epoch": 2.3465315662959663,
      "grad_norm": 0.625,
      "learning_rate": 1.8098962021714032e-05,
      "loss": 0.3854925870895386,
      "num_input_tokens_seen": 18863064320,
      "step": 66320,
      "train_runtime": 644943.0405,
      "train_tokens_per_second": 29247.644
    },
    {
      "epoch": 2.346885386559695,
      "grad_norm": 0.609375,
      "learning_rate": 1.8097776394045977e-05,
      "loss": 0.3840511322021484,
      "num_input_tokens_seen": 18865859712,
      "step": 66330,
      "train_runtime": 645036.4179,
      "train_tokens_per_second": 29247.744
    },
    {
      "epoch": 2.347239206823424,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8096590999351865e-05,
      "loss": 0.3879375457763672,
      "num_input_tokens_seen": 18868693952,
      "step": 66340,
      "train_runtime": 645131.5674,
      "train_tokens_per_second": 29247.823
    },
    {
      "epoch": 2.347593027087153,
      "grad_norm": 0.625,
      "learning_rate": 1.8095405837555397e-05,
      "loss": 0.3843435764312744,
      "num_input_tokens_seen": 18871519552,
      "step": 66350,
      "train_runtime": 645230.1334,
      "train_tokens_per_second": 29247.734
    },
    {
      "epoch": 2.3479468473508813,
      "grad_norm": 0.62109375,
      "learning_rate": 1.809422090858033e-05,
      "loss": 0.3864013195037842,
      "num_input_tokens_seen": 18874326400,
      "step": 66360,
      "train_runtime": 645329.0799,
      "train_tokens_per_second": 29247.599
    },
    {
      "epoch": 2.3483006676146103,
      "grad_norm": 0.60546875,
      "learning_rate": 1.809303621235044e-05,
      "loss": 0.38490939140319824,
      "num_input_tokens_seen": 18877182784,
      "step": 66370,
      "train_runtime": 645428.3269,
      "train_tokens_per_second": 29247.528
    },
    {
      "epoch": 2.348654487878339,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8091851748789548e-05,
      "loss": 0.38621280193328855,
      "num_input_tokens_seen": 18880003776,
      "step": 66380,
      "train_runtime": 645525.1182,
      "train_tokens_per_second": 29247.512
    },
    {
      "epoch": 2.349008308142068,
      "grad_norm": 0.625,
      "learning_rate": 1.80906675178215e-05,
      "loss": 0.3870787382125854,
      "num_input_tokens_seen": 18882870144,
      "step": 66390,
      "train_runtime": 645622.1936,
      "train_tokens_per_second": 29247.554
    },
    {
      "epoch": 2.3493621284057964,
      "grad_norm": 0.625,
      "learning_rate": 1.808948351937018e-05,
      "loss": 0.387070894241333,
      "num_input_tokens_seen": 18885742848,
      "step": 66400,
      "train_runtime": 645722.4224,
      "train_tokens_per_second": 29247.463
    },
    {
      "epoch": 2.3497159486695254,
      "grad_norm": 0.58984375,
      "learning_rate": 1.808829975335952e-05,
      "loss": 0.38469786643981935,
      "num_input_tokens_seen": 18888581888,
      "step": 66410,
      "train_runtime": 645818.56,
      "train_tokens_per_second": 29247.506
    },
    {
      "epoch": 2.350069768933254,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8087116219713467e-05,
      "loss": 0.3849670648574829,
      "num_input_tokens_seen": 18891465728,
      "step": 66420,
      "train_runtime": 645920.2036,
      "train_tokens_per_second": 29247.368
    },
    {
      "epoch": 2.350423589196983,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8085932918356015e-05,
      "loss": 0.39301486015319825,
      "num_input_tokens_seen": 18894308736,
      "step": 66430,
      "train_runtime": 646016.2967,
      "train_tokens_per_second": 29247.418
    },
    {
      "epoch": 2.3507774094607115,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8084749849211188e-05,
      "loss": 0.3868676424026489,
      "num_input_tokens_seen": 18897165888,
      "step": 66440,
      "train_runtime": 646110.6441,
      "train_tokens_per_second": 29247.569
    },
    {
      "epoch": 2.3511312297244404,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8083567012203052e-05,
      "loss": 0.3886558055877686,
      "num_input_tokens_seen": 18900017216,
      "step": 66450,
      "train_runtime": 646209.5855,
      "train_tokens_per_second": 29247.504
    },
    {
      "epoch": 2.351485049988169,
      "grad_norm": 0.609375,
      "learning_rate": 1.808238440725569e-05,
      "loss": 0.3784471988677979,
      "num_input_tokens_seen": 18902877056,
      "step": 66460,
      "train_runtime": 646306.3435,
      "train_tokens_per_second": 29247.55
    },
    {
      "epoch": 2.351838870251898,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8081202034293245e-05,
      "loss": 0.3872998237609863,
      "num_input_tokens_seen": 18905723136,
      "step": 66470,
      "train_runtime": 646404.2408,
      "train_tokens_per_second": 29247.523
    },
    {
      "epoch": 2.3521926905156265,
      "grad_norm": 0.6171875,
      "learning_rate": 1.808001989323987e-05,
      "loss": 0.3838960647583008,
      "num_input_tokens_seen": 18908575936,
      "step": 66480,
      "train_runtime": 646500.4465,
      "train_tokens_per_second": 29247.584
    },
    {
      "epoch": 2.3525465107793555,
      "grad_norm": 0.60546875,
      "learning_rate": 1.807883798401978e-05,
      "loss": 0.38517446517944337,
      "num_input_tokens_seen": 18911356288,
      "step": 66490,
      "train_runtime": 646593.5346,
      "train_tokens_per_second": 29247.673
    },
    {
      "epoch": 2.3529003310430845,
      "grad_norm": 0.609375,
      "learning_rate": 1.8077656306557196e-05,
      "loss": 0.3873758316040039,
      "num_input_tokens_seen": 18914208448,
      "step": 66500,
      "train_runtime": 646692.2892,
      "train_tokens_per_second": 29247.617
    },
    {
      "epoch": 2.353254151306813,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8076474860776387e-05,
      "loss": 0.38544299602508547,
      "num_input_tokens_seen": 18917053952,
      "step": 66510,
      "train_runtime": 646788.3514,
      "train_tokens_per_second": 29247.673
    },
    {
      "epoch": 2.3536079715705416,
      "grad_norm": 0.609375,
      "learning_rate": 1.807529364660166e-05,
      "loss": 0.38534984588623045,
      "num_input_tokens_seen": 18919881536,
      "step": 66520,
      "train_runtime": 646886.3583,
      "train_tokens_per_second": 29247.612
    },
    {
      "epoch": 2.3539617918342706,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8074112663957353e-05,
      "loss": 0.3883091449737549,
      "num_input_tokens_seen": 18922675456,
      "step": 66530,
      "train_runtime": 646980.6405,
      "train_tokens_per_second": 29247.669
    },
    {
      "epoch": 2.3543156120979996,
      "grad_norm": 0.609375,
      "learning_rate": 1.8072931912767843e-05,
      "loss": 0.384459924697876,
      "num_input_tokens_seen": 18925517760,
      "step": 66540,
      "train_runtime": 647077.0155,
      "train_tokens_per_second": 29247.705
    },
    {
      "epoch": 2.354669432361728,
      "grad_norm": 0.6171875,
      "learning_rate": 1.807175139295753e-05,
      "loss": 0.38231282234191893,
      "num_input_tokens_seen": 18928376128,
      "step": 66550,
      "train_runtime": 647176.8396,
      "train_tokens_per_second": 29247.61
    },
    {
      "epoch": 2.355023252625457,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8070571104450855e-05,
      "loss": 0.3841726303100586,
      "num_input_tokens_seen": 18931240256,
      "step": 66560,
      "train_runtime": 647277.1317,
      "train_tokens_per_second": 29247.504
    },
    {
      "epoch": 2.3553770728891856,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8069391047172297e-05,
      "loss": 0.3860349416732788,
      "num_input_tokens_seen": 18934033152,
      "step": 66570,
      "train_runtime": 647371.6176,
      "train_tokens_per_second": 29247.549
    },
    {
      "epoch": 2.3557308931529146,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8068211221046367e-05,
      "loss": 0.38725550174713136,
      "num_input_tokens_seen": 18936829376,
      "step": 66580,
      "train_runtime": 647465.053,
      "train_tokens_per_second": 29247.647
    },
    {
      "epoch": 2.356084713416643,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8067031625997616e-05,
      "loss": 0.3895761489868164,
      "num_input_tokens_seen": 18939639552,
      "step": 66590,
      "train_runtime": 647558.7382,
      "train_tokens_per_second": 29247.755
    },
    {
      "epoch": 2.356438533680372,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8065852261950612e-05,
      "loss": 0.3881476879119873,
      "num_input_tokens_seen": 18942461440,
      "step": 66600,
      "train_runtime": 647653.8605,
      "train_tokens_per_second": 29247.817
    },
    {
      "epoch": 2.3567923539441007,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8064673128829977e-05,
      "loss": 0.38766913414001464,
      "num_input_tokens_seen": 18945258304,
      "step": 66610,
      "train_runtime": 647747.509,
      "train_tokens_per_second": 29247.906
    },
    {
      "epoch": 2.3571461742078297,
      "grad_norm": 0.625,
      "learning_rate": 1.8063494226560355e-05,
      "loss": 0.38472020626068115,
      "num_input_tokens_seen": 18948082688,
      "step": 66620,
      "train_runtime": 647842.5085,
      "train_tokens_per_second": 29247.977
    },
    {
      "epoch": 2.3574999944715582,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8062315555066432e-05,
      "loss": 0.3879100799560547,
      "num_input_tokens_seen": 18950908864,
      "step": 66630,
      "train_runtime": 647937.4518,
      "train_tokens_per_second": 29248.053
    },
    {
      "epoch": 2.357853814735287,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8061137114272934e-05,
      "loss": 0.3836753606796265,
      "num_input_tokens_seen": 18953769920,
      "step": 66640,
      "train_runtime": 648036.104,
      "train_tokens_per_second": 29248.015
    },
    {
      "epoch": 2.3582076349990158,
      "grad_norm": 0.609375,
      "learning_rate": 1.8059958904104592e-05,
      "loss": 0.3854692459106445,
      "num_input_tokens_seen": 18956558976,
      "step": 66650,
      "train_runtime": 648127.7377,
      "train_tokens_per_second": 29248.183
    },
    {
      "epoch": 2.3585614552627447,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8058780924486212e-05,
      "loss": 0.3840113639831543,
      "num_input_tokens_seen": 18959454272,
      "step": 66660,
      "train_runtime": 648228.805,
      "train_tokens_per_second": 29248.09
    },
    {
      "epoch": 2.3589152755264733,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8057603175342603e-05,
      "loss": 0.3865053176879883,
      "num_input_tokens_seen": 18962273984,
      "step": 66670,
      "train_runtime": 648324.1628,
      "train_tokens_per_second": 29248.137
    },
    {
      "epoch": 2.3592690957902023,
      "grad_norm": 0.5703125,
      "learning_rate": 1.8056425656598623e-05,
      "loss": 0.38563454151153564,
      "num_input_tokens_seen": 18965111808,
      "step": 66680,
      "train_runtime": 648418.8958,
      "train_tokens_per_second": 29248.24
    },
    {
      "epoch": 2.3596229160539313,
      "grad_norm": 0.6328125,
      "learning_rate": 1.805524836817916e-05,
      "loss": 0.3880745887756348,
      "num_input_tokens_seen": 18967973248,
      "step": 66690,
      "train_runtime": 648518.2619,
      "train_tokens_per_second": 29248.171
    },
    {
      "epoch": 2.35997673631766,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8054071310009144e-05,
      "loss": 0.3890956401824951,
      "num_input_tokens_seen": 18970783296,
      "step": 66700,
      "train_runtime": 648615.0094,
      "train_tokens_per_second": 29248.141
    },
    {
      "epoch": 2.360330556581389,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8052894482013526e-05,
      "loss": 0.3844252347946167,
      "num_input_tokens_seen": 18973634560,
      "step": 66710,
      "train_runtime": 648713.7525,
      "train_tokens_per_second": 29248.084
    },
    {
      "epoch": 2.3606843768451173,
      "grad_norm": 0.62890625,
      "learning_rate": 1.80517178841173e-05,
      "loss": 0.3907251596450806,
      "num_input_tokens_seen": 18976445888,
      "step": 66720,
      "train_runtime": 648808.1675,
      "train_tokens_per_second": 29248.161
    },
    {
      "epoch": 2.3610381971088463,
      "grad_norm": 0.6015625,
      "learning_rate": 1.805054151624549e-05,
      "loss": 0.3880284070968628,
      "num_input_tokens_seen": 18979350976,
      "step": 66730,
      "train_runtime": 648908.7899,
      "train_tokens_per_second": 29248.103
    },
    {
      "epoch": 2.361392017372575,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8049365378323154e-05,
      "loss": 0.38728818893432615,
      "num_input_tokens_seen": 18982182400,
      "step": 66740,
      "train_runtime": 649005.3534,
      "train_tokens_per_second": 29248.114
    },
    {
      "epoch": 2.361745837636304,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8048189470275397e-05,
      "loss": 0.3859985828399658,
      "num_input_tokens_seen": 18985068800,
      "step": 66750,
      "train_runtime": 649102.5821,
      "train_tokens_per_second": 29248.179
    },
    {
      "epoch": 2.3620996579000324,
      "grad_norm": 0.59765625,
      "learning_rate": 1.804701379202734e-05,
      "loss": 0.38720026016235354,
      "num_input_tokens_seen": 18987914560,
      "step": 66760,
      "train_runtime": 649200.1359,
      "train_tokens_per_second": 29248.168
    },
    {
      "epoch": 2.3624534781637614,
      "grad_norm": 0.609375,
      "learning_rate": 1.8045838343504142e-05,
      "loss": 0.3851181983947754,
      "num_input_tokens_seen": 18990767360,
      "step": 66770,
      "train_runtime": 649297.6939,
      "train_tokens_per_second": 29248.167
    },
    {
      "epoch": 2.36280729842749,
      "grad_norm": 0.578125,
      "learning_rate": 1.8044663124631007e-05,
      "loss": 0.3856996536254883,
      "num_input_tokens_seen": 18993610112,
      "step": 66780,
      "train_runtime": 649397.4844,
      "train_tokens_per_second": 29248.05
    },
    {
      "epoch": 2.363161118691219,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8043488135333166e-05,
      "loss": 0.3841752052307129,
      "num_input_tokens_seen": 18996413312,
      "step": 66790,
      "train_runtime": 649492.3725,
      "train_tokens_per_second": 29248.093
    },
    {
      "epoch": 2.3635149389549475,
      "grad_norm": 0.63671875,
      "learning_rate": 1.804231337553588e-05,
      "loss": 0.3858665466308594,
      "num_input_tokens_seen": 18999240960,
      "step": 66800,
      "train_runtime": 649588.3122,
      "train_tokens_per_second": 29248.126
    },
    {
      "epoch": 2.3638687592186765,
      "grad_norm": 0.6015625,
      "learning_rate": 1.804113884516445e-05,
      "loss": 0.38831095695495604,
      "num_input_tokens_seen": 19002115392,
      "step": 66810,
      "train_runtime": 649685.1296,
      "train_tokens_per_second": 29248.192
    },
    {
      "epoch": 2.364222579482405,
      "grad_norm": 0.609375,
      "learning_rate": 1.803996454414421e-05,
      "loss": 0.3856593132019043,
      "num_input_tokens_seen": 19004939456,
      "step": 66820,
      "train_runtime": 649780.1918,
      "train_tokens_per_second": 29248.259
    },
    {
      "epoch": 2.364576399746134,
      "grad_norm": 0.6015625,
      "learning_rate": 1.803879047240052e-05,
      "loss": 0.3851649761199951,
      "num_input_tokens_seen": 19007806976,
      "step": 66830,
      "train_runtime": 649879.3818,
      "train_tokens_per_second": 29248.207
    },
    {
      "epoch": 2.364930220009863,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8037616629858794e-05,
      "loss": 0.3904259443283081,
      "num_input_tokens_seen": 19010648704,
      "step": 66840,
      "train_runtime": 649977.205,
      "train_tokens_per_second": 29248.178
    },
    {
      "epoch": 2.3652840402735915,
      "grad_norm": 0.6015625,
      "learning_rate": 1.803644301644446e-05,
      "loss": 0.38640198707580564,
      "num_input_tokens_seen": 19013491520,
      "step": 66850,
      "train_runtime": 650070.1276,
      "train_tokens_per_second": 29248.37
    },
    {
      "epoch": 2.36563786053732,
      "grad_norm": 0.59765625,
      "learning_rate": 1.8035269632082984e-05,
      "loss": 0.38723571300506593,
      "num_input_tokens_seen": 19016368320,
      "step": 66860,
      "train_runtime": 650170.072,
      "train_tokens_per_second": 29248.298
    },
    {
      "epoch": 2.365991680801049,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8034096476699873e-05,
      "loss": 0.3838082790374756,
      "num_input_tokens_seen": 19019231680,
      "step": 66870,
      "train_runtime": 650269.4669,
      "train_tokens_per_second": 29248.231
    },
    {
      "epoch": 2.366345501064778,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8032923550220667e-05,
      "loss": 0.38194947242736815,
      "num_input_tokens_seen": 19022032128,
      "step": 66880,
      "train_runtime": 650365.9442,
      "train_tokens_per_second": 29248.198
    },
    {
      "epoch": 2.3666993213285066,
      "grad_norm": 0.6484375,
      "learning_rate": 1.803175085257093e-05,
      "loss": 0.38920769691467283,
      "num_input_tokens_seen": 19024924416,
      "step": 66890,
      "train_runtime": 650466.4228,
      "train_tokens_per_second": 29248.127
    },
    {
      "epoch": 2.3670531415922356,
      "grad_norm": 0.65625,
      "learning_rate": 1.8030578383676272e-05,
      "loss": 0.3835707902908325,
      "num_input_tokens_seen": 19027770240,
      "step": 66900,
      "train_runtime": 650565.5491,
      "train_tokens_per_second": 29248.045
    },
    {
      "epoch": 2.367406961855964,
      "grad_norm": 0.6484375,
      "learning_rate": 1.8029406143462326e-05,
      "loss": 0.39029593467712403,
      "num_input_tokens_seen": 19030609344,
      "step": 66910,
      "train_runtime": 650662.2479,
      "train_tokens_per_second": 29248.061
    },
    {
      "epoch": 2.367760782119693,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8028234131854774e-05,
      "loss": 0.3825751781463623,
      "num_input_tokens_seen": 19033486464,
      "step": 66920,
      "train_runtime": 650759.6986,
      "train_tokens_per_second": 29248.103
    },
    {
      "epoch": 2.3681146023834216,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8027062348779313e-05,
      "loss": 0.3817171573638916,
      "num_input_tokens_seen": 19036277568,
      "step": 66930,
      "train_runtime": 650855.6204,
      "train_tokens_per_second": 29248.08
    },
    {
      "epoch": 2.3684684226471506,
      "grad_norm": 0.61328125,
      "learning_rate": 1.802589079416168e-05,
      "loss": 0.38561415672302246,
      "num_input_tokens_seen": 19039119552,
      "step": 66940,
      "train_runtime": 650948.9906,
      "train_tokens_per_second": 29248.251
    },
    {
      "epoch": 2.368822242910879,
      "grad_norm": 0.59375,
      "learning_rate": 1.8024719467927664e-05,
      "loss": 0.38842480182647704,
      "num_input_tokens_seen": 19042012416,
      "step": 66950,
      "train_runtime": 651051.8933,
      "train_tokens_per_second": 29248.072
    },
    {
      "epoch": 2.369176063174608,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8023548370003062e-05,
      "loss": 0.38610038757324217,
      "num_input_tokens_seen": 19044805184,
      "step": 66960,
      "train_runtime": 651145.5184,
      "train_tokens_per_second": 29248.155
    },
    {
      "epoch": 2.3695298834383367,
      "grad_norm": 0.6015625,
      "learning_rate": 1.8022377500313714e-05,
      "loss": 0.38412020206451414,
      "num_input_tokens_seen": 19047623872,
      "step": 66970,
      "train_runtime": 651240.1106,
      "train_tokens_per_second": 29248.235
    },
    {
      "epoch": 2.3698837037020657,
      "grad_norm": 0.59375,
      "learning_rate": 1.80212068587855e-05,
      "loss": 0.387067985534668,
      "num_input_tokens_seen": 19050525824,
      "step": 66980,
      "train_runtime": 651340.3006,
      "train_tokens_per_second": 29248.191
    },
    {
      "epoch": 2.3702375239657942,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8020036445344325e-05,
      "loss": 0.3839916467666626,
      "num_input_tokens_seen": 19053351168,
      "step": 66990,
      "train_runtime": 651436.9939,
      "train_tokens_per_second": 29248.187
    },
    {
      "epoch": 2.3705913442295232,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8018866259916134e-05,
      "loss": 0.3910531044006348,
      "num_input_tokens_seen": 19056236480,
      "step": 67000,
      "train_runtime": 651534.7595,
      "train_tokens_per_second": 29248.227
    },
    {
      "epoch": 2.3709451644932518,
      "grad_norm": 0.60546875,
      "learning_rate": 1.8017696302426904e-05,
      "loss": 0.3840883016586304,
      "num_input_tokens_seen": 19059105792,
      "step": 67010,
      "train_runtime": 651632.3382,
      "train_tokens_per_second": 29248.25
    },
    {
      "epoch": 2.3712989847569808,
      "grad_norm": 0.64453125,
      "learning_rate": 1.801652657280264e-05,
      "loss": 0.38502113819122313,
      "num_input_tokens_seen": 19061936128,
      "step": 67020,
      "train_runtime": 651727.936,
      "train_tokens_per_second": 29248.303
    },
    {
      "epoch": 2.3716528050207097,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8015357070969387e-05,
      "loss": 0.3863154172897339,
      "num_input_tokens_seen": 19064763584,
      "step": 67030,
      "train_runtime": 651823.9772,
      "train_tokens_per_second": 29248.331
    },
    {
      "epoch": 2.3720066252844383,
      "grad_norm": 0.6171875,
      "learning_rate": 1.801418779685323e-05,
      "loss": 0.3895847320556641,
      "num_input_tokens_seen": 19067641472,
      "step": 67040,
      "train_runtime": 651921.2878,
      "train_tokens_per_second": 29248.38
    },
    {
      "epoch": 2.3723604455481673,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8013018750380265e-05,
      "loss": 0.38856494426727295,
      "num_input_tokens_seen": 19070533248,
      "step": 67050,
      "train_runtime": 652022.4864,
      "train_tokens_per_second": 29248.275
    },
    {
      "epoch": 2.372714265811896,
      "grad_norm": 0.62890625,
      "learning_rate": 1.8011849931476647e-05,
      "loss": 0.3876237154006958,
      "num_input_tokens_seen": 19073411136,
      "step": 67060,
      "train_runtime": 652122.7214,
      "train_tokens_per_second": 29248.193
    },
    {
      "epoch": 2.373068086075625,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8010681340068548e-05,
      "loss": 0.38921217918395995,
      "num_input_tokens_seen": 19076215360,
      "step": 67070,
      "train_runtime": 652220.7375,
      "train_tokens_per_second": 29248.097
    },
    {
      "epoch": 2.3734219063393533,
      "grad_norm": 0.61328125,
      "learning_rate": 1.800951297608218e-05,
      "loss": 0.3862889766693115,
      "num_input_tokens_seen": 19079047936,
      "step": 67080,
      "train_runtime": 652315.2522,
      "train_tokens_per_second": 29248.201
    },
    {
      "epoch": 2.3737757266030823,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8008344839443795e-05,
      "loss": 0.37821016311645506,
      "num_input_tokens_seen": 19081876416,
      "step": 67090,
      "train_runtime": 652412.6667,
      "train_tokens_per_second": 29248.17
    },
    {
      "epoch": 2.374129546866811,
      "grad_norm": 0.62109375,
      "learning_rate": 1.8007176930079663e-05,
      "loss": 0.3829841613769531,
      "num_input_tokens_seen": 19084723968,
      "step": 67100,
      "train_runtime": 652507.6098,
      "train_tokens_per_second": 29248.278
    },
    {
      "epoch": 2.37448336713054,
      "grad_norm": 0.64453125,
      "learning_rate": 1.8006009247916096e-05,
      "loss": 0.3886546611785889,
      "num_input_tokens_seen": 19087596672,
      "step": 67110,
      "train_runtime": 652605.0066,
      "train_tokens_per_second": 29248.315
    },
    {
      "epoch": 2.3748371873942684,
      "grad_norm": 0.6015625,
      "learning_rate": 1.800484179287944e-05,
      "loss": 0.38506135940551756,
      "num_input_tokens_seen": 19090493248,
      "step": 67120,
      "train_runtime": 652705.2186,
      "train_tokens_per_second": 29248.262
    },
    {
      "epoch": 2.3751910076579974,
      "grad_norm": 0.63671875,
      "learning_rate": 1.8003674564896075e-05,
      "loss": 0.3862583637237549,
      "num_input_tokens_seen": 19093306240,
      "step": 67130,
      "train_runtime": 652800.0633,
      "train_tokens_per_second": 29248.322
    },
    {
      "epoch": 2.375544827921726,
      "grad_norm": 0.625,
      "learning_rate": 1.8002507563892413e-05,
      "loss": 0.3825004816055298,
      "num_input_tokens_seen": 19096105600,
      "step": 67140,
      "train_runtime": 652893.9238,
      "train_tokens_per_second": 29248.405
    },
    {
      "epoch": 2.375898648185455,
      "grad_norm": 0.6171875,
      "learning_rate": 1.8001340789794895e-05,
      "loss": 0.3888943910598755,
      "num_input_tokens_seen": 19098944000,
      "step": 67150,
      "train_runtime": 652990.8187,
      "train_tokens_per_second": 29248.411
    },
    {
      "epoch": 2.3762524684491835,
      "grad_norm": 0.61328125,
      "learning_rate": 1.800017424253001e-05,
      "loss": 0.3856736421585083,
      "num_input_tokens_seen": 19101779264,
      "step": 67160,
      "train_runtime": 653091.9635,
      "train_tokens_per_second": 29248.223
    },
    {
      "epoch": 2.3766062887129125,
      "grad_norm": 0.67578125,
      "learning_rate": 1.7999007922024254e-05,
      "loss": 0.3889627933502197,
      "num_input_tokens_seen": 19104616576,
      "step": 67170,
      "train_runtime": 653187.2613,
      "train_tokens_per_second": 29248.299
    },
    {
      "epoch": 2.3769601089766415,
      "grad_norm": 0.59375,
      "learning_rate": 1.7997841828204185e-05,
      "loss": 0.386491584777832,
      "num_input_tokens_seen": 19107457088,
      "step": 67180,
      "train_runtime": 653282.6978,
      "train_tokens_per_second": 29248.375
    },
    {
      "epoch": 2.37731392924037,
      "grad_norm": 0.59765625,
      "learning_rate": 1.799667596099638e-05,
      "loss": 0.3883882999420166,
      "num_input_tokens_seen": 19110291264,
      "step": 67190,
      "train_runtime": 653377.4336,
      "train_tokens_per_second": 29248.472
    },
    {
      "epoch": 2.3776677495040985,
      "grad_norm": 0.640625,
      "learning_rate": 1.7995510320327445e-05,
      "loss": 0.3875779628753662,
      "num_input_tokens_seen": 19113143040,
      "step": 67200,
      "train_runtime": 653473.3777,
      "train_tokens_per_second": 29248.541
    },
    {
      "epoch": 2.3780215697678275,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7994344906124028e-05,
      "loss": 0.3921213150024414,
      "num_input_tokens_seen": 19115954752,
      "step": 67210,
      "train_runtime": 653571.7862,
      "train_tokens_per_second": 29248.439
    },
    {
      "epoch": 2.3783753900315565,
      "grad_norm": 0.625,
      "learning_rate": 1.799317971831281e-05,
      "loss": 0.38783869743347166,
      "num_input_tokens_seen": 19118786944,
      "step": 67220,
      "train_runtime": 653669.0166,
      "train_tokens_per_second": 29248.422
    },
    {
      "epoch": 2.378729210295285,
      "grad_norm": 0.61328125,
      "learning_rate": 1.79920147568205e-05,
      "loss": 0.3880669116973877,
      "num_input_tokens_seen": 19121651968,
      "step": 67230,
      "train_runtime": 653767.5396,
      "train_tokens_per_second": 29248.396
    },
    {
      "epoch": 2.379083030559014,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7990850021573845e-05,
      "loss": 0.3857611656188965,
      "num_input_tokens_seen": 19124483200,
      "step": 67240,
      "train_runtime": 653863.8638,
      "train_tokens_per_second": 29248.417
    },
    {
      "epoch": 2.3794368508227426,
      "grad_norm": 0.62890625,
      "learning_rate": 1.798968551249962e-05,
      "loss": 0.3844860076904297,
      "num_input_tokens_seen": 19127322624,
      "step": 67250,
      "train_runtime": 653960.4245,
      "train_tokens_per_second": 29248.441
    },
    {
      "epoch": 2.3797906710864716,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7988521229524637e-05,
      "loss": 0.386318302154541,
      "num_input_tokens_seen": 19130141760,
      "step": 67260,
      "train_runtime": 654053.9467,
      "train_tokens_per_second": 29248.569
    },
    {
      "epoch": 2.3801444913502,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7987357172575745e-05,
      "loss": 0.3859975814819336,
      "num_input_tokens_seen": 19132988224,
      "step": 67270,
      "train_runtime": 654150.7836,
      "train_tokens_per_second": 29248.59
    },
    {
      "epoch": 2.380498311613929,
      "grad_norm": 0.66796875,
      "learning_rate": 1.7986193341579814e-05,
      "loss": 0.3843409061431885,
      "num_input_tokens_seen": 19135846528,
      "step": 67280,
      "train_runtime": 654248.8822,
      "train_tokens_per_second": 29248.574
    },
    {
      "epoch": 2.3808521318776577,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7985029736463757e-05,
      "loss": 0.3858004570007324,
      "num_input_tokens_seen": 19138757312,
      "step": 67290,
      "train_runtime": 654353.1232,
      "train_tokens_per_second": 29248.362
    },
    {
      "epoch": 2.3812059521413866,
      "grad_norm": 0.625,
      "learning_rate": 1.798386635715452e-05,
      "loss": 0.38802061080932615,
      "num_input_tokens_seen": 19141531136,
      "step": 67300,
      "train_runtime": 654445.0724,
      "train_tokens_per_second": 29248.491
    },
    {
      "epoch": 2.381559772405115,
      "grad_norm": 0.62109375,
      "learning_rate": 1.798270320357908e-05,
      "loss": 0.38294553756713867,
      "num_input_tokens_seen": 19144349440,
      "step": 67310,
      "train_runtime": 654543.2547,
      "train_tokens_per_second": 29248.41
    },
    {
      "epoch": 2.381913592668844,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7981540275664447e-05,
      "loss": 0.38640589714050294,
      "num_input_tokens_seen": 19147203008,
      "step": 67320,
      "train_runtime": 654639.8297,
      "train_tokens_per_second": 29248.454
    },
    {
      "epoch": 2.382267412932573,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7980377573337658e-05,
      "loss": 0.38403520584106443,
      "num_input_tokens_seen": 19150074432,
      "step": 67330,
      "train_runtime": 654737.4114,
      "train_tokens_per_second": 29248.481
    },
    {
      "epoch": 2.3826212331963017,
      "grad_norm": 0.66796875,
      "learning_rate": 1.7979215096525794e-05,
      "loss": 0.38894870281219485,
      "num_input_tokens_seen": 19152964736,
      "step": 67340,
      "train_runtime": 654836.1498,
      "train_tokens_per_second": 29248.484
    },
    {
      "epoch": 2.3829750534600302,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7978052845155965e-05,
      "loss": 0.3874341011047363,
      "num_input_tokens_seen": 19155801792,
      "step": 67350,
      "train_runtime": 654934.7389,
      "train_tokens_per_second": 29248.413
    },
    {
      "epoch": 2.3833288737237592,
      "grad_norm": 0.63671875,
      "learning_rate": 1.797689081915531e-05,
      "loss": 0.38392815589904783,
      "num_input_tokens_seen": 19158648192,
      "step": 67360,
      "train_runtime": 655032.1756,
      "train_tokens_per_second": 29248.408
    },
    {
      "epoch": 2.383682693987488,
      "grad_norm": 0.58984375,
      "learning_rate": 1.7975729018451004e-05,
      "loss": 0.38665311336517333,
      "num_input_tokens_seen": 19161527488,
      "step": 67370,
      "train_runtime": 655134.0268,
      "train_tokens_per_second": 29248.256
    },
    {
      "epoch": 2.3840365142512168,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7974567442970254e-05,
      "loss": 0.38753490447998046,
      "num_input_tokens_seen": 19164337920,
      "step": 67380,
      "train_runtime": 655228.2903,
      "train_tokens_per_second": 29248.337
    },
    {
      "epoch": 2.3843903345149458,
      "grad_norm": 0.625,
      "learning_rate": 1.7973406092640308e-05,
      "loss": 0.38848047256469725,
      "num_input_tokens_seen": 19167200960,
      "step": 67390,
      "train_runtime": 655327.7185,
      "train_tokens_per_second": 29248.268
    },
    {
      "epoch": 2.3847441547786743,
      "grad_norm": 0.59375,
      "learning_rate": 1.797224496738843e-05,
      "loss": 0.38537015914916994,
      "num_input_tokens_seen": 19170035712,
      "step": 67400,
      "train_runtime": 655426.0759,
      "train_tokens_per_second": 29248.204
    },
    {
      "epoch": 2.3850979750424033,
      "grad_norm": 0.65234375,
      "learning_rate": 1.797108406714193e-05,
      "loss": 0.3868121147155762,
      "num_input_tokens_seen": 19172876992,
      "step": 67410,
      "train_runtime": 655522.7833,
      "train_tokens_per_second": 29248.224
    },
    {
      "epoch": 2.385451795306132,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7969923391828146e-05,
      "loss": 0.3907296657562256,
      "num_input_tokens_seen": 19175719744,
      "step": 67420,
      "train_runtime": 655618.7415,
      "train_tokens_per_second": 29248.279
    },
    {
      "epoch": 2.385805615569861,
      "grad_norm": 0.65234375,
      "learning_rate": 1.7968762941374456e-05,
      "loss": 0.3842211961746216,
      "num_input_tokens_seen": 19178526016,
      "step": 67430,
      "train_runtime": 655716.0888,
      "train_tokens_per_second": 29248.216
    },
    {
      "epoch": 2.3861594358335894,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7967602715708257e-05,
      "loss": 0.3899972677230835,
      "num_input_tokens_seen": 19181360832,
      "step": 67440,
      "train_runtime": 655812.1668,
      "train_tokens_per_second": 29248.254
    },
    {
      "epoch": 2.3865132560973183,
      "grad_norm": 0.58203125,
      "learning_rate": 1.7966442714756992e-05,
      "loss": 0.3831056594848633,
      "num_input_tokens_seen": 19184188736,
      "step": 67450,
      "train_runtime": 655906.9262,
      "train_tokens_per_second": 29248.34
    },
    {
      "epoch": 2.386867076361047,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7965282938448134e-05,
      "loss": 0.3842476844787598,
      "num_input_tokens_seen": 19186973888,
      "step": 67460,
      "train_runtime": 655998.3292,
      "train_tokens_per_second": 29248.51
    },
    {
      "epoch": 2.387220896624776,
      "grad_norm": 0.609375,
      "learning_rate": 1.7964123386709175e-05,
      "loss": 0.3880655765533447,
      "num_input_tokens_seen": 19189801984,
      "step": 67470,
      "train_runtime": 656093.8372,
      "train_tokens_per_second": 29248.563
    },
    {
      "epoch": 2.3875747168885044,
      "grad_norm": 0.6484375,
      "learning_rate": 1.796296405946766e-05,
      "loss": 0.3874183177947998,
      "num_input_tokens_seen": 19192624448,
      "step": 67480,
      "train_runtime": 656187.9472,
      "train_tokens_per_second": 29248.67
    },
    {
      "epoch": 2.3879285371522334,
      "grad_norm": 0.61328125,
      "learning_rate": 1.796180495665116e-05,
      "loss": 0.3858750820159912,
      "num_input_tokens_seen": 19195485120,
      "step": 67490,
      "train_runtime": 656285.4672,
      "train_tokens_per_second": 29248.682
    },
    {
      "epoch": 2.388282357415962,
      "grad_norm": 0.62109375,
      "learning_rate": 1.796064607818727e-05,
      "loss": 0.38378736972808836,
      "num_input_tokens_seen": 19198248000,
      "step": 67500,
      "train_runtime": 656380.3029,
      "train_tokens_per_second": 29248.666
    },
    {
      "epoch": 2.388636177679691,
      "grad_norm": 0.625,
      "learning_rate": 1.795948742400363e-05,
      "loss": 0.3820474624633789,
      "num_input_tokens_seen": 19201038656,
      "step": 67510,
      "train_runtime": 656473.2496,
      "train_tokens_per_second": 29248.775
    },
    {
      "epoch": 2.38898999794342,
      "grad_norm": 0.66015625,
      "learning_rate": 1.7958328994027903e-05,
      "loss": 0.3854734182357788,
      "num_input_tokens_seen": 19203872192,
      "step": 67520,
      "train_runtime": 656566.678,
      "train_tokens_per_second": 29248.929
    },
    {
      "epoch": 2.3893438182071485,
      "grad_norm": 0.58984375,
      "learning_rate": 1.7957170788187792e-05,
      "loss": 0.3885235548019409,
      "num_input_tokens_seen": 19206691648,
      "step": 67530,
      "train_runtime": 656661.9185,
      "train_tokens_per_second": 29248.98
    },
    {
      "epoch": 2.3896976384708775,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7956012806411028e-05,
      "loss": 0.38292100429534914,
      "num_input_tokens_seen": 19209527552,
      "step": 67540,
      "train_runtime": 656762.8867,
      "train_tokens_per_second": 29248.802
    },
    {
      "epoch": 2.390051458734606,
      "grad_norm": 0.62890625,
      "learning_rate": 1.795485504862537e-05,
      "loss": 0.38591823577880857,
      "num_input_tokens_seen": 19212346624,
      "step": 67550,
      "train_runtime": 656859.1842,
      "train_tokens_per_second": 29248.806
    },
    {
      "epoch": 2.390405278998335,
      "grad_norm": 0.60546875,
      "learning_rate": 1.795369751475862e-05,
      "loss": 0.38375487327575686,
      "num_input_tokens_seen": 19215182976,
      "step": 67560,
      "train_runtime": 656954.3784,
      "train_tokens_per_second": 29248.885
    },
    {
      "epoch": 2.3907590992620635,
      "grad_norm": 0.6328125,
      "learning_rate": 1.795254020473861e-05,
      "loss": 0.39004030227661135,
      "num_input_tokens_seen": 19218037120,
      "step": 67570,
      "train_runtime": 657053.248,
      "train_tokens_per_second": 29248.828
    },
    {
      "epoch": 2.3911129195257925,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7951383118493203e-05,
      "loss": 0.38717434406280515,
      "num_input_tokens_seen": 19220905088,
      "step": 67580,
      "train_runtime": 657155.582,
      "train_tokens_per_second": 29248.637
    },
    {
      "epoch": 2.391466739789521,
      "grad_norm": 0.609375,
      "learning_rate": 1.795022625595029e-05,
      "loss": 0.38304524421691893,
      "num_input_tokens_seen": 19223711296,
      "step": 67590,
      "train_runtime": 657250.3483,
      "train_tokens_per_second": 29248.689
    },
    {
      "epoch": 2.39182056005325,
      "grad_norm": 0.62109375,
      "learning_rate": 1.79490696170378e-05,
      "loss": 0.38646512031555175,
      "num_input_tokens_seen": 19226493888,
      "step": 67600,
      "train_runtime": 657342.8652,
      "train_tokens_per_second": 29248.806
    },
    {
      "epoch": 2.3921743803169786,
      "grad_norm": 0.609375,
      "learning_rate": 1.79479132016837e-05,
      "loss": 0.38769145011901857,
      "num_input_tokens_seen": 19229300800,
      "step": 67610,
      "train_runtime": 657436.241,
      "train_tokens_per_second": 29248.921
    },
    {
      "epoch": 2.3925282005807076,
      "grad_norm": 0.62109375,
      "learning_rate": 1.794675700981597e-05,
      "loss": 0.38835821151733396,
      "num_input_tokens_seen": 19232120256,
      "step": 67620,
      "train_runtime": 657534.3171,
      "train_tokens_per_second": 29248.846
    },
    {
      "epoch": 2.392882020844436,
      "grad_norm": 0.60546875,
      "learning_rate": 1.794560104136264e-05,
      "loss": 0.38687279224395754,
      "num_input_tokens_seen": 19234951936,
      "step": 67630,
      "train_runtime": 657628.7664,
      "train_tokens_per_second": 29248.952
    },
    {
      "epoch": 2.393235841108165,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7944445296251776e-05,
      "loss": 0.3826420307159424,
      "num_input_tokens_seen": 19237797504,
      "step": 67640,
      "train_runtime": 657726.5015,
      "train_tokens_per_second": 29248.932
    },
    {
      "epoch": 2.3935896613718937,
      "grad_norm": 0.58984375,
      "learning_rate": 1.794328977441146e-05,
      "loss": 0.384617280960083,
      "num_input_tokens_seen": 19240642304,
      "step": 67650,
      "train_runtime": 657821.5443,
      "train_tokens_per_second": 29249.03
    },
    {
      "epoch": 2.3939434816356226,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7942134475769813e-05,
      "loss": 0.3850719690322876,
      "num_input_tokens_seen": 19243497152,
      "step": 67660,
      "train_runtime": 657922.1269,
      "train_tokens_per_second": 29248.898
    },
    {
      "epoch": 2.3942973018993516,
      "grad_norm": 0.625,
      "learning_rate": 1.794097940025499e-05,
      "loss": 0.3867626190185547,
      "num_input_tokens_seen": 19246382720,
      "step": 67670,
      "train_runtime": 658021.6857,
      "train_tokens_per_second": 29248.858
    },
    {
      "epoch": 2.39465112216308,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7939824547795183e-05,
      "loss": 0.387221097946167,
      "num_input_tokens_seen": 19249216896,
      "step": 67680,
      "train_runtime": 658116.8658,
      "train_tokens_per_second": 29248.934
    },
    {
      "epoch": 2.3950049424268087,
      "grad_norm": 0.6171875,
      "learning_rate": 1.793866991831861e-05,
      "loss": 0.38462803363800047,
      "num_input_tokens_seen": 19252052992,
      "step": 67690,
      "train_runtime": 658210.0308,
      "train_tokens_per_second": 29249.103
    },
    {
      "epoch": 2.3953587626905377,
      "grad_norm": 0.640625,
      "learning_rate": 1.793751551175352e-05,
      "loss": 0.382578182220459,
      "num_input_tokens_seen": 19254910272,
      "step": 67700,
      "train_runtime": 658305.8863,
      "train_tokens_per_second": 29249.184
    },
    {
      "epoch": 2.3957125829542667,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7936361328028196e-05,
      "loss": 0.3824758529663086,
      "num_input_tokens_seen": 19257707904,
      "step": 67710,
      "train_runtime": 658398.7477,
      "train_tokens_per_second": 29249.308
    },
    {
      "epoch": 2.3960664032179952,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7935207367070964e-05,
      "loss": 0.38828563690185547,
      "num_input_tokens_seen": 19260580096,
      "step": 67720,
      "train_runtime": 658496.2851,
      "train_tokens_per_second": 29249.338
    },
    {
      "epoch": 2.3964202234817242,
      "grad_norm": 0.6328125,
      "learning_rate": 1.793405362881016e-05,
      "loss": 0.3870595693588257,
      "num_input_tokens_seen": 19263461888,
      "step": 67730,
      "train_runtime": 658595.9488,
      "train_tokens_per_second": 29249.287
    },
    {
      "epoch": 2.3967740437454528,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7932900113174182e-05,
      "loss": 0.3890698909759521,
      "num_input_tokens_seen": 19266301248,
      "step": 67740,
      "train_runtime": 658692.086,
      "train_tokens_per_second": 29249.329
    },
    {
      "epoch": 2.3971278640091818,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7931746820091423e-05,
      "loss": 0.3873075723648071,
      "num_input_tokens_seen": 19269187200,
      "step": 67750,
      "train_runtime": 658793.2023,
      "train_tokens_per_second": 29249.22
    },
    {
      "epoch": 2.3974816842729103,
      "grad_norm": 0.609375,
      "learning_rate": 1.793059374949034e-05,
      "loss": 0.38792710304260253,
      "num_input_tokens_seen": 19272044224,
      "step": 67760,
      "train_runtime": 658890.4124,
      "train_tokens_per_second": 29249.241
    },
    {
      "epoch": 2.3978355045366393,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7929440901299405e-05,
      "loss": 0.38840744495391843,
      "num_input_tokens_seen": 19274895168,
      "step": 67770,
      "train_runtime": 658987.086,
      "train_tokens_per_second": 29249.276
    },
    {
      "epoch": 2.398189324800368,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7928288275447135e-05,
      "loss": 0.3856579542160034,
      "num_input_tokens_seen": 19277718016,
      "step": 67780,
      "train_runtime": 659083.6087,
      "train_tokens_per_second": 29249.275
    },
    {
      "epoch": 2.398543145064097,
      "grad_norm": 0.6171875,
      "learning_rate": 1.792713587186207e-05,
      "loss": 0.3879667282104492,
      "num_input_tokens_seen": 19280601024,
      "step": 67790,
      "train_runtime": 659183.7444,
      "train_tokens_per_second": 29249.206
    },
    {
      "epoch": 2.3988969653278254,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7925983690472782e-05,
      "loss": 0.38777642250061034,
      "num_input_tokens_seen": 19283403200,
      "step": 67800,
      "train_runtime": 659278.0277,
      "train_tokens_per_second": 29249.273
    },
    {
      "epoch": 2.3992507855915544,
      "grad_norm": 0.59375,
      "learning_rate": 1.7924831731207873e-05,
      "loss": 0.3857454776763916,
      "num_input_tokens_seen": 19286276032,
      "step": 67810,
      "train_runtime": 659377.6345,
      "train_tokens_per_second": 29249.212
    },
    {
      "epoch": 2.399604605855283,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7923679993995988e-05,
      "loss": 0.3831601619720459,
      "num_input_tokens_seen": 19289053632,
      "step": 67820,
      "train_runtime": 659469.2148,
      "train_tokens_per_second": 29249.362
    },
    {
      "epoch": 2.399958426119012,
      "grad_norm": 0.62890625,
      "learning_rate": 1.79225284787658e-05,
      "loss": 0.39115471839904786,
      "num_input_tokens_seen": 19291869760,
      "step": 67830,
      "train_runtime": 659564.5288,
      "train_tokens_per_second": 29249.405
    },
    {
      "epoch": 2.4003122463827404,
      "grad_norm": 0.64453125,
      "learning_rate": 1.7921377185446003e-05,
      "loss": 0.3869039297103882,
      "num_input_tokens_seen": 19294748224,
      "step": 67840,
      "train_runtime": 659664.713,
      "train_tokens_per_second": 29249.326
    },
    {
      "epoch": 2.4006660666464694,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7920226113965338e-05,
      "loss": 0.3882428169250488,
      "num_input_tokens_seen": 19297604224,
      "step": 67850,
      "train_runtime": 659761.8706,
      "train_tokens_per_second": 29249.348
    },
    {
      "epoch": 2.4010198869101984,
      "grad_norm": 0.58984375,
      "learning_rate": 1.791907526425257e-05,
      "loss": 0.3835087060928345,
      "num_input_tokens_seen": 19300442624,
      "step": 67860,
      "train_runtime": 659859.3756,
      "train_tokens_per_second": 29249.327
    },
    {
      "epoch": 2.401373707173927,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7917924636236495e-05,
      "loss": 0.39130091667175293,
      "num_input_tokens_seen": 19303286912,
      "step": 67870,
      "train_runtime": 659956.5227,
      "train_tokens_per_second": 29249.331
    },
    {
      "epoch": 2.401727527437656,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7916774229845944e-05,
      "loss": 0.39014620780944825,
      "num_input_tokens_seen": 19306127872,
      "step": 67880,
      "train_runtime": 660054.073,
      "train_tokens_per_second": 29249.313
    },
    {
      "epoch": 2.4020813477013845,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7915624045009787e-05,
      "loss": 0.3851542711257935,
      "num_input_tokens_seen": 19308988352,
      "step": 67890,
      "train_runtime": 660154.3482,
      "train_tokens_per_second": 29249.203
    },
    {
      "epoch": 2.4024351679651135,
      "grad_norm": 0.640625,
      "learning_rate": 1.791447408165691e-05,
      "loss": 0.38439435958862306,
      "num_input_tokens_seen": 19311830400,
      "step": 67900,
      "train_runtime": 660251.0498,
      "train_tokens_per_second": 29249.223
    },
    {
      "epoch": 2.402788988228842,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7913324339716244e-05,
      "loss": 0.3864377498626709,
      "num_input_tokens_seen": 19314628096,
      "step": 67910,
      "train_runtime": 660348.8875,
      "train_tokens_per_second": 29249.126
    },
    {
      "epoch": 2.403142808492571,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7912174819116744e-05,
      "loss": 0.39105536937713625,
      "num_input_tokens_seen": 19317486464,
      "step": 67920,
      "train_runtime": 660443.6466,
      "train_tokens_per_second": 29249.258
    },
    {
      "epoch": 2.4034966287562995,
      "grad_norm": 0.65625,
      "learning_rate": 1.7911025519787406e-05,
      "loss": 0.3873642921447754,
      "num_input_tokens_seen": 19320343936,
      "step": 67930,
      "train_runtime": 660542.2593,
      "train_tokens_per_second": 29249.217
    },
    {
      "epoch": 2.4038504490200285,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7909876441657247e-05,
      "loss": 0.3838647365570068,
      "num_input_tokens_seen": 19323255232,
      "step": 67940,
      "train_runtime": 660642.8159,
      "train_tokens_per_second": 29249.172
    },
    {
      "epoch": 2.404204269283757,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7908727584655324e-05,
      "loss": 0.3887533664703369,
      "num_input_tokens_seen": 19326011456,
      "step": 67950,
      "train_runtime": 660735.479,
      "train_tokens_per_second": 29249.241
    },
    {
      "epoch": 2.404558089547486,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7907578948710722e-05,
      "loss": 0.38205652236938475,
      "num_input_tokens_seen": 19328812416,
      "step": 67960,
      "train_runtime": 660830.0656,
      "train_tokens_per_second": 29249.293
    },
    {
      "epoch": 2.4049119098112146,
      "grad_norm": 0.6015625,
      "learning_rate": 1.790643053375256e-05,
      "loss": 0.3907313823699951,
      "num_input_tokens_seen": 19331645056,
      "step": 67970,
      "train_runtime": 660925.1271,
      "train_tokens_per_second": 29249.372
    },
    {
      "epoch": 2.4052657300749436,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7905282339709987e-05,
      "loss": 0.38699769973754883,
      "num_input_tokens_seen": 19334499904,
      "step": 67980,
      "train_runtime": 661021.6204,
      "train_tokens_per_second": 29249.421
    },
    {
      "epoch": 2.405619550338672,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7904134366512183e-05,
      "loss": 0.3880749702453613,
      "num_input_tokens_seen": 19337347840,
      "step": 67990,
      "train_runtime": 661118.1021,
      "train_tokens_per_second": 29249.461
    },
    {
      "epoch": 2.405973370602401,
      "grad_norm": 0.60546875,
      "learning_rate": 1.790298661408836e-05,
      "loss": 0.3875674486160278,
      "num_input_tokens_seen": 19340199680,
      "step": 68000,
      "train_runtime": 661213.2493,
      "train_tokens_per_second": 29249.565
    },
    {
      "epoch": 2.40632719086613,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7901839082367766e-05,
      "loss": 0.3829047203063965,
      "num_input_tokens_seen": 19343053632,
      "step": 68010,
      "train_runtime": 661310.6351,
      "train_tokens_per_second": 29249.573
    },
    {
      "epoch": 2.4066810111298587,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7900691771279682e-05,
      "loss": 0.38587212562561035,
      "num_input_tokens_seen": 19345908864,
      "step": 68020,
      "train_runtime": 661406.2284,
      "train_tokens_per_second": 29249.662
    },
    {
      "epoch": 2.407034831393587,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7899544680753404e-05,
      "loss": 0.3864143371582031,
      "num_input_tokens_seen": 19348721856,
      "step": 68030,
      "train_runtime": 661502.3503,
      "train_tokens_per_second": 29249.665
    },
    {
      "epoch": 2.407388651657316,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7898397810718286e-05,
      "loss": 0.38871009349823,
      "num_input_tokens_seen": 19351526976,
      "step": 68040,
      "train_runtime": 661599.8308,
      "train_tokens_per_second": 29249.595
    },
    {
      "epoch": 2.407742471921045,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7897251161103693e-05,
      "loss": 0.3876418352127075,
      "num_input_tokens_seen": 19354386368,
      "step": 68050,
      "train_runtime": 661696.8537,
      "train_tokens_per_second": 29249.627
    },
    {
      "epoch": 2.4080962921847737,
      "grad_norm": 0.625,
      "learning_rate": 1.789610473183903e-05,
      "loss": 0.38550117015838625,
      "num_input_tokens_seen": 19357238848,
      "step": 68060,
      "train_runtime": 661791.8658,
      "train_tokens_per_second": 29249.738
    },
    {
      "epoch": 2.4084501124485027,
      "grad_norm": 0.62109375,
      "learning_rate": 1.789495852285373e-05,
      "loss": 0.3871241807937622,
      "num_input_tokens_seen": 19360095936,
      "step": 68070,
      "train_runtime": 661889.7004,
      "train_tokens_per_second": 29249.731
    },
    {
      "epoch": 2.4088039327122313,
      "grad_norm": 0.62890625,
      "learning_rate": 1.789381253407726e-05,
      "loss": 0.3843658447265625,
      "num_input_tokens_seen": 19362983168,
      "step": 68080,
      "train_runtime": 661989.3366,
      "train_tokens_per_second": 29249.69
    },
    {
      "epoch": 2.4091577529759602,
      "grad_norm": 0.59375,
      "learning_rate": 1.7892666765439117e-05,
      "loss": 0.38748888969421386,
      "num_input_tokens_seen": 19365830400,
      "step": 68090,
      "train_runtime": 662084.8793,
      "train_tokens_per_second": 29249.77
    },
    {
      "epoch": 2.409511573239689,
      "grad_norm": 0.62109375,
      "learning_rate": 1.789152121686884e-05,
      "loss": 0.384832763671875,
      "num_input_tokens_seen": 19368627456,
      "step": 68100,
      "train_runtime": 662176.325,
      "train_tokens_per_second": 29249.955
    },
    {
      "epoch": 2.4098653935034178,
      "grad_norm": 0.640625,
      "learning_rate": 1.789037588829598e-05,
      "loss": 0.3864151954650879,
      "num_input_tokens_seen": 19371494976,
      "step": 68110,
      "train_runtime": 662275.0771,
      "train_tokens_per_second": 29249.923
    },
    {
      "epoch": 2.4102192137671463,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7889230779650135e-05,
      "loss": 0.38594019412994385,
      "num_input_tokens_seen": 19374391936,
      "step": 68120,
      "train_runtime": 662376.4841,
      "train_tokens_per_second": 29249.818
    },
    {
      "epoch": 2.4105730340308753,
      "grad_norm": 0.58203125,
      "learning_rate": 1.7888085890860927e-05,
      "loss": 0.38209917545318606,
      "num_input_tokens_seen": 19377238656,
      "step": 68130,
      "train_runtime": 662472.6451,
      "train_tokens_per_second": 29249.87
    },
    {
      "epoch": 2.410926854294604,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7886941221858022e-05,
      "loss": 0.38693718910217284,
      "num_input_tokens_seen": 19380071296,
      "step": 68140,
      "train_runtime": 662570.1248,
      "train_tokens_per_second": 29249.842
    },
    {
      "epoch": 2.411280674558333,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7885796772571094e-05,
      "loss": 0.38875608444213866,
      "num_input_tokens_seen": 19382972800,
      "step": 68150,
      "train_runtime": 662672.2939,
      "train_tokens_per_second": 29249.711
    },
    {
      "epoch": 2.411634494822062,
      "grad_norm": 0.625,
      "learning_rate": 1.788465254292987e-05,
      "loss": 0.3817655086517334,
      "num_input_tokens_seen": 19385770304,
      "step": 68160,
      "train_runtime": 662768.091,
      "train_tokens_per_second": 29249.704
    },
    {
      "epoch": 2.4119883150857904,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7883508532864095e-05,
      "loss": 0.38323497772216797,
      "num_input_tokens_seen": 19388657024,
      "step": 68170,
      "train_runtime": 662868.5361,
      "train_tokens_per_second": 29249.626
    },
    {
      "epoch": 2.412342135349519,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7882364742303563e-05,
      "loss": 0.38757858276367185,
      "num_input_tokens_seen": 19391486080,
      "step": 68180,
      "train_runtime": 662965.2125,
      "train_tokens_per_second": 29249.628
    },
    {
      "epoch": 2.412695955613248,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7881221171178072e-05,
      "loss": 0.38554298877716064,
      "num_input_tokens_seen": 19394326528,
      "step": 68190,
      "train_runtime": 663061.4661,
      "train_tokens_per_second": 29249.666
    },
    {
      "epoch": 2.413049775876977,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7880077819417483e-05,
      "loss": 0.3880670785903931,
      "num_input_tokens_seen": 19397190464,
      "step": 68200,
      "train_runtime": 663159.4791,
      "train_tokens_per_second": 29249.662
    },
    {
      "epoch": 2.4134035961407054,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7878934686951657e-05,
      "loss": 0.3865138053894043,
      "num_input_tokens_seen": 19399980352,
      "step": 68210,
      "train_runtime": 663251.5265,
      "train_tokens_per_second": 29249.809
    },
    {
      "epoch": 2.4137574164044344,
      "grad_norm": 0.62890625,
      "learning_rate": 1.787779177371051e-05,
      "loss": 0.3825418472290039,
      "num_input_tokens_seen": 19402815616,
      "step": 68220,
      "train_runtime": 663349.6357,
      "train_tokens_per_second": 29249.757
    },
    {
      "epoch": 2.414111236668163,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7876649079623982e-05,
      "loss": 0.3875403642654419,
      "num_input_tokens_seen": 19405657088,
      "step": 68230,
      "train_runtime": 663446.6959,
      "train_tokens_per_second": 29249.761
    },
    {
      "epoch": 2.414465056931892,
      "grad_norm": 0.62109375,
      "learning_rate": 1.787550660462204e-05,
      "loss": 0.3883230209350586,
      "num_input_tokens_seen": 19408556928,
      "step": 68240,
      "train_runtime": 663548.9657,
      "train_tokens_per_second": 29249.623
    },
    {
      "epoch": 2.4148188771956205,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7874364348634686e-05,
      "loss": 0.38724329471588137,
      "num_input_tokens_seen": 19411388096,
      "step": 68250,
      "train_runtime": 663644.3295,
      "train_tokens_per_second": 29249.686
    },
    {
      "epoch": 2.4151726974593495,
      "grad_norm": 0.61328125,
      "learning_rate": 1.787322231159196e-05,
      "loss": 0.3832000732421875,
      "num_input_tokens_seen": 19414268544,
      "step": 68260,
      "train_runtime": 663743.4242,
      "train_tokens_per_second": 29249.659
    },
    {
      "epoch": 2.415526517723078,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7872080493423915e-05,
      "loss": 0.3849701642990112,
      "num_input_tokens_seen": 19417073856,
      "step": 68270,
      "train_runtime": 663840.2457,
      "train_tokens_per_second": 29249.618
    },
    {
      "epoch": 2.415880337986807,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7870938894060653e-05,
      "loss": 0.3862785816192627,
      "num_input_tokens_seen": 19419968512,
      "step": 68280,
      "train_runtime": 663937.9209,
      "train_tokens_per_second": 29249.675
    },
    {
      "epoch": 2.4162341582505356,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7869797513432305e-05,
      "loss": 0.3864187717437744,
      "num_input_tokens_seen": 19422819328,
      "step": 68290,
      "train_runtime": 664035.0492,
      "train_tokens_per_second": 29249.69
    },
    {
      "epoch": 2.4165879785142645,
      "grad_norm": 0.58203125,
      "learning_rate": 1.7868656351469017e-05,
      "loss": 0.3882637023925781,
      "num_input_tokens_seen": 19425674496,
      "step": 68300,
      "train_runtime": 664134.0223,
      "train_tokens_per_second": 29249.63
    },
    {
      "epoch": 2.416941798777993,
      "grad_norm": 0.62109375,
      "learning_rate": 1.786751540810099e-05,
      "loss": 0.38697137832641604,
      "num_input_tokens_seen": 19428499456,
      "step": 68310,
      "train_runtime": 664229.8259,
      "train_tokens_per_second": 29249.664
    },
    {
      "epoch": 2.417295619041722,
      "grad_norm": 0.6328125,
      "learning_rate": 1.786637468325844e-05,
      "loss": 0.3823454141616821,
      "num_input_tokens_seen": 19431315072,
      "step": 68320,
      "train_runtime": 664322.3507,
      "train_tokens_per_second": 29249.829
    },
    {
      "epoch": 2.4176494393054506,
      "grad_norm": 0.62890625,
      "learning_rate": 1.786523417687162e-05,
      "loss": 0.38868486881256104,
      "num_input_tokens_seen": 19434190656,
      "step": 68330,
      "train_runtime": 664425.3466,
      "train_tokens_per_second": 29249.623
    },
    {
      "epoch": 2.4180032595691796,
      "grad_norm": 0.609375,
      "learning_rate": 1.7864093888870813e-05,
      "loss": 0.38460938930511473,
      "num_input_tokens_seen": 19437104192,
      "step": 68340,
      "train_runtime": 664531.0014,
      "train_tokens_per_second": 29249.357
    },
    {
      "epoch": 2.4183570798329086,
      "grad_norm": 0.59375,
      "learning_rate": 1.7862953819186332e-05,
      "loss": 0.3870745420455933,
      "num_input_tokens_seen": 19439931200,
      "step": 68350,
      "train_runtime": 664628.9457,
      "train_tokens_per_second": 29249.3
    },
    {
      "epoch": 2.418710900096637,
      "grad_norm": 0.59765625,
      "learning_rate": 1.786181396774852e-05,
      "loss": 0.3881475210189819,
      "num_input_tokens_seen": 19442805888,
      "step": 68360,
      "train_runtime": 664725.3103,
      "train_tokens_per_second": 29249.384
    },
    {
      "epoch": 2.419064720360366,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7860674334487757e-05,
      "loss": 0.382158088684082,
      "num_input_tokens_seen": 19445669952,
      "step": 68370,
      "train_runtime": 664822.3962,
      "train_tokens_per_second": 29249.421
    },
    {
      "epoch": 2.4194185406240947,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7859534919334447e-05,
      "loss": 0.3840055465698242,
      "num_input_tokens_seen": 19448465280,
      "step": 68380,
      "train_runtime": 664914.7205,
      "train_tokens_per_second": 29249.563
    },
    {
      "epoch": 2.4197723608878237,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7858395722219032e-05,
      "loss": 0.3886756658554077,
      "num_input_tokens_seen": 19451290176,
      "step": 68390,
      "train_runtime": 665011.1215,
      "train_tokens_per_second": 29249.571
    },
    {
      "epoch": 2.420126181151552,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7857256743071978e-05,
      "loss": 0.3851438522338867,
      "num_input_tokens_seen": 19454168704,
      "step": 68400,
      "train_runtime": 665111.5759,
      "train_tokens_per_second": 29249.481
    },
    {
      "epoch": 2.420480001415281,
      "grad_norm": 0.65234375,
      "learning_rate": 1.785611798182379e-05,
      "loss": 0.38528194427490237,
      "num_input_tokens_seen": 19456994560,
      "step": 68410,
      "train_runtime": 665206.4897,
      "train_tokens_per_second": 29249.556
    },
    {
      "epoch": 2.4208338216790097,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7854979438405e-05,
      "loss": 0.38706820011138915,
      "num_input_tokens_seen": 19459841792,
      "step": 68420,
      "train_runtime": 665301.2115,
      "train_tokens_per_second": 29249.671
    },
    {
      "epoch": 2.4211876419427387,
      "grad_norm": 0.58984375,
      "learning_rate": 1.7853841112746157e-05,
      "loss": 0.3822739839553833,
      "num_input_tokens_seen": 19462673920,
      "step": 68430,
      "train_runtime": 665397.9831,
      "train_tokens_per_second": 29249.674
    },
    {
      "epoch": 2.4215414622064673,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7852703004777873e-05,
      "loss": 0.38995659351348877,
      "num_input_tokens_seen": 19465500608,
      "step": 68440,
      "train_runtime": 665493.3342,
      "train_tokens_per_second": 29249.73
    },
    {
      "epoch": 2.4218952824701963,
      "grad_norm": 0.625,
      "learning_rate": 1.7851565114430764e-05,
      "loss": 0.385045599937439,
      "num_input_tokens_seen": 19468332288,
      "step": 68450,
      "train_runtime": 665591.9177,
      "train_tokens_per_second": 29249.652
    },
    {
      "epoch": 2.422249102733925,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7850427441635482e-05,
      "loss": 0.38572192192077637,
      "num_input_tokens_seen": 19471153856,
      "step": 68460,
      "train_runtime": 665686.3888,
      "train_tokens_per_second": 29249.74
    },
    {
      "epoch": 2.422602922997654,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7849289986322723e-05,
      "loss": 0.38976345062255857,
      "num_input_tokens_seen": 19473960384,
      "step": 68470,
      "train_runtime": 665780.5846,
      "train_tokens_per_second": 29249.817
    },
    {
      "epoch": 2.4229567432613823,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7848152748423196e-05,
      "loss": 0.384015417098999,
      "num_input_tokens_seen": 19476810240,
      "step": 68480,
      "train_runtime": 665878.4752,
      "train_tokens_per_second": 29249.797
    },
    {
      "epoch": 2.4233105635251113,
      "grad_norm": 0.62109375,
      "learning_rate": 1.784701572786765e-05,
      "loss": 0.3880761623382568,
      "num_input_tokens_seen": 19479679872,
      "step": 68490,
      "train_runtime": 665974.2185,
      "train_tokens_per_second": 29249.901
    },
    {
      "epoch": 2.4236643837888403,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7845878924586875e-05,
      "loss": 0.38806092739105225,
      "num_input_tokens_seen": 19482488192,
      "step": 68500,
      "train_runtime": 666068.814,
      "train_tokens_per_second": 29249.963
    },
    {
      "epoch": 2.424018204052569,
      "grad_norm": 0.59375,
      "learning_rate": 1.7844742338511663e-05,
      "loss": 0.38856942653656007,
      "num_input_tokens_seen": 19485348736,
      "step": 68510,
      "train_runtime": 666167.8738,
      "train_tokens_per_second": 29249.908
    },
    {
      "epoch": 2.4243720243162974,
      "grad_norm": 0.609375,
      "learning_rate": 1.7843605969572865e-05,
      "loss": 0.385299015045166,
      "num_input_tokens_seen": 19488182592,
      "step": 68520,
      "train_runtime": 666264.1823,
      "train_tokens_per_second": 29249.933
    },
    {
      "epoch": 2.4247258445800264,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7842469817701354e-05,
      "loss": 0.38481812477111815,
      "num_input_tokens_seen": 19491078592,
      "step": 68530,
      "train_runtime": 666363.2866,
      "train_tokens_per_second": 29249.929
    },
    {
      "epoch": 2.4250796648437554,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7841333882828033e-05,
      "loss": 0.38563559055328367,
      "num_input_tokens_seen": 19493887616,
      "step": 68540,
      "train_runtime": 666460.125,
      "train_tokens_per_second": 29249.893
    },
    {
      "epoch": 2.425433485107484,
      "grad_norm": 0.640625,
      "learning_rate": 1.7840198164883827e-05,
      "loss": 0.3876136541366577,
      "num_input_tokens_seen": 19496733952,
      "step": 68550,
      "train_runtime": 666555.5352,
      "train_tokens_per_second": 29249.977
    },
    {
      "epoch": 2.425787305371213,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7839062663799703e-05,
      "loss": 0.3833590030670166,
      "num_input_tokens_seen": 19499576576,
      "step": 68560,
      "train_runtime": 666653.7683,
      "train_tokens_per_second": 29249.931
    },
    {
      "epoch": 2.4261411256349414,
      "grad_norm": 0.625,
      "learning_rate": 1.7837927379506662e-05,
      "loss": 0.38374671936035154,
      "num_input_tokens_seen": 19502451520,
      "step": 68570,
      "train_runtime": 666750.7974,
      "train_tokens_per_second": 29249.986
    },
    {
      "epoch": 2.4264949458986704,
      "grad_norm": 0.609375,
      "learning_rate": 1.7836792311935722e-05,
      "loss": 0.38574895858764646,
      "num_input_tokens_seen": 19505334848,
      "step": 68580,
      "train_runtime": 666851.2098,
      "train_tokens_per_second": 29249.905
    },
    {
      "epoch": 2.426848766162399,
      "grad_norm": 0.65625,
      "learning_rate": 1.7835657461017948e-05,
      "loss": 0.3871032238006592,
      "num_input_tokens_seen": 19508146112,
      "step": 68590,
      "train_runtime": 666948.2138,
      "train_tokens_per_second": 29249.866
    },
    {
      "epoch": 2.427202586426128,
      "grad_norm": 0.59375,
      "learning_rate": 1.7834522826684414e-05,
      "loss": 0.38348524570465087,
      "num_input_tokens_seen": 19511018624,
      "step": 68600,
      "train_runtime": 667045.8508,
      "train_tokens_per_second": 29249.891
    },
    {
      "epoch": 2.4275564066898565,
      "grad_norm": 0.6328125,
      "learning_rate": 1.783338840886625e-05,
      "loss": 0.3856853485107422,
      "num_input_tokens_seen": 19513878080,
      "step": 68610,
      "train_runtime": 667142.7645,
      "train_tokens_per_second": 29249.928
    },
    {
      "epoch": 2.4279102269535855,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7832254207494598e-05,
      "loss": 0.3850064277648926,
      "num_input_tokens_seen": 19516727616,
      "step": 68620,
      "train_runtime": 667243.9464,
      "train_tokens_per_second": 29249.763
    },
    {
      "epoch": 2.428264047217314,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7831120222500635e-05,
      "loss": 0.39541604518890383,
      "num_input_tokens_seen": 19519535680,
      "step": 68630,
      "train_runtime": 667340.0371,
      "train_tokens_per_second": 29249.76
    },
    {
      "epoch": 2.428617867481043,
      "grad_norm": 0.63671875,
      "learning_rate": 1.782998645381558e-05,
      "loss": 0.3894403696060181,
      "num_input_tokens_seen": 19522431680,
      "step": 68640,
      "train_runtime": 667439.2528,
      "train_tokens_per_second": 29249.751
    },
    {
      "epoch": 2.4289716877447716,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7828852901370663e-05,
      "loss": 0.38569400310516355,
      "num_input_tokens_seen": 19525270144,
      "step": 68650,
      "train_runtime": 667535.6113,
      "train_tokens_per_second": 29249.781
    },
    {
      "epoch": 2.4293255080085006,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7827719565097157e-05,
      "loss": 0.3869462966918945,
      "num_input_tokens_seen": 19528156928,
      "step": 68660,
      "train_runtime": 667635.5715,
      "train_tokens_per_second": 29249.725
    },
    {
      "epoch": 2.429679328272229,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7826586444926364e-05,
      "loss": 0.3893005847930908,
      "num_input_tokens_seen": 19531035392,
      "step": 68670,
      "train_runtime": 667733.3684,
      "train_tokens_per_second": 29249.752
    },
    {
      "epoch": 2.430033148535958,
      "grad_norm": 0.609375,
      "learning_rate": 1.782545354078962e-05,
      "loss": 0.3851781845092773,
      "num_input_tokens_seen": 19533874624,
      "step": 68680,
      "train_runtime": 667830.1509,
      "train_tokens_per_second": 29249.764
    },
    {
      "epoch": 2.430386968799687,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7824320852618285e-05,
      "loss": 0.384342098236084,
      "num_input_tokens_seen": 19536690112,
      "step": 68690,
      "train_runtime": 667927.1153,
      "train_tokens_per_second": 29249.733
    },
    {
      "epoch": 2.4307407890634156,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7823188380343746e-05,
      "loss": 0.3890566110610962,
      "num_input_tokens_seen": 19539581504,
      "step": 68700,
      "train_runtime": 668026.3404,
      "train_tokens_per_second": 29249.717
    },
    {
      "epoch": 2.4310946093271446,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7822056123897433e-05,
      "loss": 0.3845391750335693,
      "num_input_tokens_seen": 19542478080,
      "step": 68710,
      "train_runtime": 668127.6735,
      "train_tokens_per_second": 29249.616
    },
    {
      "epoch": 2.431448429590873,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7820924083210802e-05,
      "loss": 0.38262989521026614,
      "num_input_tokens_seen": 19545280320,
      "step": 68720,
      "train_runtime": 668220.0452,
      "train_tokens_per_second": 29249.767
    },
    {
      "epoch": 2.431802249854602,
      "grad_norm": 0.62109375,
      "learning_rate": 1.781979225821533e-05,
      "loss": 0.38478808403015136,
      "num_input_tokens_seen": 19548097024,
      "step": 68730,
      "train_runtime": 668317.4254,
      "train_tokens_per_second": 29249.719
    },
    {
      "epoch": 2.4321560701183307,
      "grad_norm": 0.6328125,
      "learning_rate": 1.781866064884254e-05,
      "loss": 0.38600692749023435,
      "num_input_tokens_seen": 19550942208,
      "step": 68740,
      "train_runtime": 668413.4443,
      "train_tokens_per_second": 29249.774
    },
    {
      "epoch": 2.4325098903820597,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7817529255023968e-05,
      "loss": 0.39072885513305666,
      "num_input_tokens_seen": 19553779264,
      "step": 68750,
      "train_runtime": 668508.1131,
      "train_tokens_per_second": 29249.876
    },
    {
      "epoch": 2.432863710645788,
      "grad_norm": 0.59765625,
      "learning_rate": 1.78163980766912e-05,
      "loss": 0.38450586795806885,
      "num_input_tokens_seen": 19556599360,
      "step": 68760,
      "train_runtime": 668605.5211,
      "train_tokens_per_second": 29249.832
    },
    {
      "epoch": 2.433217530909517,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7815267113775832e-05,
      "loss": 0.38595972061157224,
      "num_input_tokens_seen": 19559495680,
      "step": 68770,
      "train_runtime": 668707.8685,
      "train_tokens_per_second": 29249.687
    },
    {
      "epoch": 2.4335713511732457,
      "grad_norm": 0.61328125,
      "learning_rate": 1.781413636620951e-05,
      "loss": 0.38293497562408446,
      "num_input_tokens_seen": 19562304384,
      "step": 68780,
      "train_runtime": 668802.4918,
      "train_tokens_per_second": 29249.748
    },
    {
      "epoch": 2.4339251714369747,
      "grad_norm": 0.59765625,
      "learning_rate": 1.781300583392389e-05,
      "loss": 0.3914920806884766,
      "num_input_tokens_seen": 19565156416,
      "step": 68790,
      "train_runtime": 668900.683,
      "train_tokens_per_second": 29249.718
    },
    {
      "epoch": 2.4342789917007033,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7811875516850678e-05,
      "loss": 0.3848621606826782,
      "num_input_tokens_seen": 19568039168,
      "step": 68800,
      "train_runtime": 669000.243,
      "train_tokens_per_second": 29249.674
    },
    {
      "epoch": 2.4346328119644323,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7810745414921605e-05,
      "loss": 0.38750481605529785,
      "num_input_tokens_seen": 19570895680,
      "step": 68810,
      "train_runtime": 669097.9941,
      "train_tokens_per_second": 29249.67
    },
    {
      "epoch": 2.434986632228161,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7809615528068414e-05,
      "loss": 0.3826527833938599,
      "num_input_tokens_seen": 19573712576,
      "step": 68820,
      "train_runtime": 669193.8068,
      "train_tokens_per_second": 29249.692
    },
    {
      "epoch": 2.43534045249189,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7808485856222905e-05,
      "loss": 0.38694062232971194,
      "num_input_tokens_seen": 19576527808,
      "step": 68830,
      "train_runtime": 669288.9496,
      "train_tokens_per_second": 29249.74
    },
    {
      "epoch": 2.435694272755619,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7807356399316896e-05,
      "loss": 0.38736538887023925,
      "num_input_tokens_seen": 19579367232,
      "step": 68840,
      "train_runtime": 669386.3943,
      "train_tokens_per_second": 29249.724
    },
    {
      "epoch": 2.4360480930193473,
      "grad_norm": 0.6328125,
      "learning_rate": 1.780622715728223e-05,
      "loss": 0.3896360397338867,
      "num_input_tokens_seen": 19582191040,
      "step": 68850,
      "train_runtime": 669481.6135,
      "train_tokens_per_second": 29249.782
    },
    {
      "epoch": 2.436401913283076,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7805098130050797e-05,
      "loss": 0.3867495536804199,
      "num_input_tokens_seen": 19585090240,
      "step": 68860,
      "train_runtime": 669582.4338,
      "train_tokens_per_second": 29249.707
    },
    {
      "epoch": 2.436755733546805,
      "grad_norm": 0.625,
      "learning_rate": 1.780396931755449e-05,
      "loss": 0.3905338287353516,
      "num_input_tokens_seen": 19587910144,
      "step": 68870,
      "train_runtime": 669678.1423,
      "train_tokens_per_second": 29249.738
    },
    {
      "epoch": 2.437109553810534,
      "grad_norm": 0.609375,
      "learning_rate": 1.7802840719725262e-05,
      "loss": 0.38450236320495607,
      "num_input_tokens_seen": 19590721216,
      "step": 68880,
      "train_runtime": 669774.2965,
      "train_tokens_per_second": 29249.736
    },
    {
      "epoch": 2.4374633740742624,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7801712336495076e-05,
      "loss": 0.3825587511062622,
      "num_input_tokens_seen": 19593571584,
      "step": 68890,
      "train_runtime": 669873.8151,
      "train_tokens_per_second": 29249.645
    },
    {
      "epoch": 2.4378171943379914,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7800584167795936e-05,
      "loss": 0.38755455017089846,
      "num_input_tokens_seen": 19596410624,
      "step": 68900,
      "train_runtime": 669971.3997,
      "train_tokens_per_second": 29249.623
    },
    {
      "epoch": 2.43817101460172,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7799456213559863e-05,
      "loss": 0.37914950847625734,
      "num_input_tokens_seen": 19599232384,
      "step": 68910,
      "train_runtime": 670068.2046,
      "train_tokens_per_second": 29249.608
    },
    {
      "epoch": 2.438524834865449,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7798328473718933e-05,
      "loss": 0.3909867525100708,
      "num_input_tokens_seen": 19602052352,
      "step": 68920,
      "train_runtime": 670164.2452,
      "train_tokens_per_second": 29249.624
    },
    {
      "epoch": 2.4388786551291775,
      "grad_norm": 0.60546875,
      "learning_rate": 1.779720094820522e-05,
      "loss": 0.3860414743423462,
      "num_input_tokens_seen": 19604874368,
      "step": 68930,
      "train_runtime": 670257.9732,
      "train_tokens_per_second": 29249.744
    },
    {
      "epoch": 2.4392324753929064,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7796073636950857e-05,
      "loss": 0.38661651611328124,
      "num_input_tokens_seen": 19607754176,
      "step": 68940,
      "train_runtime": 670357.2167,
      "train_tokens_per_second": 29249.71
    },
    {
      "epoch": 2.439586295656635,
      "grad_norm": 0.625,
      "learning_rate": 1.7794946539887983e-05,
      "loss": 0.38774547576904295,
      "num_input_tokens_seen": 19610550208,
      "step": 68950,
      "train_runtime": 670453.2073,
      "train_tokens_per_second": 29249.693
    },
    {
      "epoch": 2.439940115920364,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7793819656948787e-05,
      "loss": 0.3843644142150879,
      "num_input_tokens_seen": 19613431296,
      "step": 68960,
      "train_runtime": 670549.666,
      "train_tokens_per_second": 29249.782
    },
    {
      "epoch": 2.4402939361840925,
      "grad_norm": 0.625,
      "learning_rate": 1.7792692988065476e-05,
      "loss": 0.3909768581390381,
      "num_input_tokens_seen": 19616247744,
      "step": 68970,
      "train_runtime": 670646.0564,
      "train_tokens_per_second": 29249.777
    },
    {
      "epoch": 2.4406477564478215,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7791566533170294e-05,
      "loss": 0.38463563919067384,
      "num_input_tokens_seen": 19619128000,
      "step": 68980,
      "train_runtime": 670746.0025,
      "train_tokens_per_second": 29249.713
    },
    {
      "epoch": 2.44100157671155,
      "grad_norm": 0.609375,
      "learning_rate": 1.7790440292195505e-05,
      "loss": 0.3878877878189087,
      "num_input_tokens_seen": 19621999168,
      "step": 68990,
      "train_runtime": 670844.7747,
      "train_tokens_per_second": 29249.686
    },
    {
      "epoch": 2.441355396975279,
      "grad_norm": 0.6328125,
      "learning_rate": 1.778931426507342e-05,
      "loss": 0.39021735191345214,
      "num_input_tokens_seen": 19624835136,
      "step": 69000,
      "train_runtime": 670942.8044,
      "train_tokens_per_second": 29249.639
    },
    {
      "epoch": 2.4417092172390076,
      "grad_norm": 0.6171875,
      "learning_rate": 1.778818845173636e-05,
      "loss": 0.3840382814407349,
      "num_input_tokens_seen": 19627691904,
      "step": 69010,
      "train_runtime": 671040.5282,
      "train_tokens_per_second": 29249.637
    },
    {
      "epoch": 2.4420630375027366,
      "grad_norm": 0.625,
      "learning_rate": 1.7787062852116687e-05,
      "loss": 0.3801845073699951,
      "num_input_tokens_seen": 19630551680,
      "step": 69020,
      "train_runtime": 671137.9418,
      "train_tokens_per_second": 29249.653
    },
    {
      "epoch": 2.4424168577664656,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7785937466146795e-05,
      "loss": 0.3876810550689697,
      "num_input_tokens_seen": 19633376384,
      "step": 69030,
      "train_runtime": 671234.0637,
      "train_tokens_per_second": 29249.672
    },
    {
      "epoch": 2.442770678030194,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7784812293759106e-05,
      "loss": 0.3851534128189087,
      "num_input_tokens_seen": 19636211712,
      "step": 69040,
      "train_runtime": 671329.8676,
      "train_tokens_per_second": 29249.722
    },
    {
      "epoch": 2.443124498293923,
      "grad_norm": 0.58203125,
      "learning_rate": 1.7783687334886065e-05,
      "loss": 0.38785078525543215,
      "num_input_tokens_seen": 19639016256,
      "step": 69050,
      "train_runtime": 671424.5537,
      "train_tokens_per_second": 29249.774
    },
    {
      "epoch": 2.4434783185576516,
      "grad_norm": 0.625,
      "learning_rate": 1.7782562589460156e-05,
      "loss": 0.38778104782104494,
      "num_input_tokens_seen": 19641962752,
      "step": 69060,
      "train_runtime": 671528.1726,
      "train_tokens_per_second": 29249.648
    },
    {
      "epoch": 2.4438321388213806,
      "grad_norm": 0.6015625,
      "learning_rate": 1.778143805741389e-05,
      "loss": 0.3882192611694336,
      "num_input_tokens_seen": 19644806592,
      "step": 69070,
      "train_runtime": 671627.499,
      "train_tokens_per_second": 29249.557
    },
    {
      "epoch": 2.444185959085109,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7780313738679804e-05,
      "loss": 0.38394689559936523,
      "num_input_tokens_seen": 19647662272,
      "step": 69080,
      "train_runtime": 671722.4818,
      "train_tokens_per_second": 29249.672
    },
    {
      "epoch": 2.444539779348838,
      "grad_norm": 0.62890625,
      "learning_rate": 1.777918963319047e-05,
      "loss": 0.38447554111480714,
      "num_input_tokens_seen": 19650563968,
      "step": 69090,
      "train_runtime": 671821.6218,
      "train_tokens_per_second": 29249.675
    },
    {
      "epoch": 2.4448935996125667,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7778065740878485e-05,
      "loss": 0.3888410091400146,
      "num_input_tokens_seen": 19653403136,
      "step": 69100,
      "train_runtime": 671921.0224,
      "train_tokens_per_second": 29249.573
    },
    {
      "epoch": 2.4452474198762957,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7776942061676485e-05,
      "loss": 0.3863215923309326,
      "num_input_tokens_seen": 19656230784,
      "step": 69110,
      "train_runtime": 672018.5213,
      "train_tokens_per_second": 29249.537
    },
    {
      "epoch": 2.4456012401400242,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7775818595517125e-05,
      "loss": 0.3869148254394531,
      "num_input_tokens_seen": 19659065408,
      "step": 69120,
      "train_runtime": 672114.4271,
      "train_tokens_per_second": 29249.581
    },
    {
      "epoch": 2.445955060403753,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7774695342333093e-05,
      "loss": 0.38787178993225097,
      "num_input_tokens_seen": 19661943488,
      "step": 69130,
      "train_runtime": 672211.7971,
      "train_tokens_per_second": 29249.626
    },
    {
      "epoch": 2.4463088806674818,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7773572302057108e-05,
      "loss": 0.3808140754699707,
      "num_input_tokens_seen": 19664729728,
      "step": 69140,
      "train_runtime": 672303.7223,
      "train_tokens_per_second": 29249.771
    },
    {
      "epoch": 2.4466627009312107,
      "grad_norm": 0.6328125,
      "learning_rate": 1.777244947462192e-05,
      "loss": 0.3911693572998047,
      "num_input_tokens_seen": 19667571456,
      "step": 69150,
      "train_runtime": 672402.1277,
      "train_tokens_per_second": 29249.716
    },
    {
      "epoch": 2.4470165211949393,
      "grad_norm": 0.609375,
      "learning_rate": 1.777132685996031e-05,
      "loss": 0.3874667167663574,
      "num_input_tokens_seen": 19670420928,
      "step": 69160,
      "train_runtime": 672498.7597,
      "train_tokens_per_second": 29249.75
    },
    {
      "epoch": 2.4473703414586683,
      "grad_norm": 0.609375,
      "learning_rate": 1.7770204458005086e-05,
      "loss": 0.38370957374572756,
      "num_input_tokens_seen": 19673216000,
      "step": 69170,
      "train_runtime": 672592.5247,
      "train_tokens_per_second": 29249.829
    },
    {
      "epoch": 2.4477241617223973,
      "grad_norm": 0.6328125,
      "learning_rate": 1.776908226868908e-05,
      "loss": 0.3922628402709961,
      "num_input_tokens_seen": 19676064512,
      "step": 69180,
      "train_runtime": 672690.8619,
      "train_tokens_per_second": 29249.787
    },
    {
      "epoch": 2.448077981986126,
      "grad_norm": 0.6328125,
      "learning_rate": 1.776796029194517e-05,
      "loss": 0.38692190647125246,
      "num_input_tokens_seen": 19678938944,
      "step": 69190,
      "train_runtime": 672788.6467,
      "train_tokens_per_second": 29249.808
    },
    {
      "epoch": 2.448431802249855,
      "grad_norm": 0.65625,
      "learning_rate": 1.776683852770624e-05,
      "loss": 0.38929319381713867,
      "num_input_tokens_seen": 19681750464,
      "step": 69200,
      "train_runtime": 672884.4013,
      "train_tokens_per_second": 29249.824
    },
    {
      "epoch": 2.4487856225135833,
      "grad_norm": 0.625,
      "learning_rate": 1.7765716975905224e-05,
      "loss": 0.3889157772064209,
      "num_input_tokens_seen": 19684566592,
      "step": 69210,
      "train_runtime": 672981.2231,
      "train_tokens_per_second": 29249.801
    },
    {
      "epoch": 2.4491394427773123,
      "grad_norm": 0.62109375,
      "learning_rate": 1.776459563647508e-05,
      "loss": 0.3855783462524414,
      "num_input_tokens_seen": 19687365696,
      "step": 69220,
      "train_runtime": 673074.3676,
      "train_tokens_per_second": 29249.912
    },
    {
      "epoch": 2.449493263041041,
      "grad_norm": 0.59375,
      "learning_rate": 1.7763474509348795e-05,
      "loss": 0.3851781368255615,
      "num_input_tokens_seen": 19690192640,
      "step": 69230,
      "train_runtime": 673170.1943,
      "train_tokens_per_second": 29249.947
    },
    {
      "epoch": 2.44984708330477,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7762353594459377e-05,
      "loss": 0.38280205726623534,
      "num_input_tokens_seen": 19693018944,
      "step": 69240,
      "train_runtime": 673264.6716,
      "train_tokens_per_second": 29250.041
    },
    {
      "epoch": 2.4502009035684984,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7761232891739878e-05,
      "loss": 0.3867254018783569,
      "num_input_tokens_seen": 19695877760,
      "step": 69250,
      "train_runtime": 673364.6092,
      "train_tokens_per_second": 29249.945
    },
    {
      "epoch": 2.4505547238322274,
      "grad_norm": 0.640625,
      "learning_rate": 1.776011240112337e-05,
      "loss": 0.3847388505935669,
      "num_input_tokens_seen": 19698677056,
      "step": 69260,
      "train_runtime": 673461.9936,
      "train_tokens_per_second": 29249.872
    },
    {
      "epoch": 2.450908544095956,
      "grad_norm": 0.59375,
      "learning_rate": 1.7758992122542966e-05,
      "loss": 0.38564984798431395,
      "num_input_tokens_seen": 19701504960,
      "step": 69270,
      "train_runtime": 673558.1779,
      "train_tokens_per_second": 29249.893
    },
    {
      "epoch": 2.451262364359685,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7757872055931786e-05,
      "loss": 0.38860111236572265,
      "num_input_tokens_seen": 19704340224,
      "step": 69280,
      "train_runtime": 673654.0166,
      "train_tokens_per_second": 29249.941
    },
    {
      "epoch": 2.4516161846234135,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7756752201223005e-05,
      "loss": 0.3863933563232422,
      "num_input_tokens_seen": 19707155328,
      "step": 69290,
      "train_runtime": 673748.2219,
      "train_tokens_per_second": 29250.029
    },
    {
      "epoch": 2.4519700048871425,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7755632558349808e-05,
      "loss": 0.38688414096832274,
      "num_input_tokens_seen": 19710033664,
      "step": 69300,
      "train_runtime": 673847.4195,
      "train_tokens_per_second": 29249.995
    },
    {
      "epoch": 2.452323825150871,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7754513127245423e-05,
      "loss": 0.38744819164276123,
      "num_input_tokens_seen": 19712820608,
      "step": 69310,
      "train_runtime": 673940.2089,
      "train_tokens_per_second": 29250.103
    },
    {
      "epoch": 2.4526776454146,
      "grad_norm": 0.60546875,
      "learning_rate": 1.77533939078431e-05,
      "loss": 0.38521599769592285,
      "num_input_tokens_seen": 19715692160,
      "step": 69320,
      "train_runtime": 674037.5017,
      "train_tokens_per_second": 29250.141
    },
    {
      "epoch": 2.453031465678329,
      "grad_norm": 0.65625,
      "learning_rate": 1.7752274900076117e-05,
      "loss": 0.38271634578704833,
      "num_input_tokens_seen": 19718515712,
      "step": 69330,
      "train_runtime": 674134.1946,
      "train_tokens_per_second": 29250.134
    },
    {
      "epoch": 2.4533852859420575,
      "grad_norm": 0.62109375,
      "learning_rate": 1.775115610387779e-05,
      "loss": 0.37929162979125974,
      "num_input_tokens_seen": 19721335040,
      "step": 69340,
      "train_runtime": 674228.3903,
      "train_tokens_per_second": 29250.229
    },
    {
      "epoch": 2.453739106205786,
      "grad_norm": 0.609375,
      "learning_rate": 1.7750037519181458e-05,
      "loss": 0.38574793338775637,
      "num_input_tokens_seen": 19724223744,
      "step": 69350,
      "train_runtime": 674329.6789,
      "train_tokens_per_second": 29250.12
    },
    {
      "epoch": 2.454092926469515,
      "grad_norm": 0.61328125,
      "learning_rate": 1.774891914592049e-05,
      "loss": 0.3858103036880493,
      "num_input_tokens_seen": 19727068800,
      "step": 69360,
      "train_runtime": 674427.1838,
      "train_tokens_per_second": 29250.109
    },
    {
      "epoch": 2.454446746733244,
      "grad_norm": 0.640625,
      "learning_rate": 1.7747800984028288e-05,
      "loss": 0.38794875144958496,
      "num_input_tokens_seen": 19729890368,
      "step": 69370,
      "train_runtime": 674524.3261,
      "train_tokens_per_second": 29250.08
    },
    {
      "epoch": 2.4548005669969726,
      "grad_norm": 0.609375,
      "learning_rate": 1.7746683033438273e-05,
      "loss": 0.38625426292419435,
      "num_input_tokens_seen": 19732693376,
      "step": 69380,
      "train_runtime": 674622.4596,
      "train_tokens_per_second": 29249.98
    },
    {
      "epoch": 2.4551543872607016,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7745565294083913e-05,
      "loss": 0.3860450744628906,
      "num_input_tokens_seen": 19735536192,
      "step": 69390,
      "train_runtime": 674720.6276,
      "train_tokens_per_second": 29249.938
    },
    {
      "epoch": 2.45550820752443,
      "grad_norm": 0.58984375,
      "learning_rate": 1.7744447765898684e-05,
      "loss": 0.38929362297058107,
      "num_input_tokens_seen": 19738395840,
      "step": 69400,
      "train_runtime": 674817.8123,
      "train_tokens_per_second": 29249.963
    },
    {
      "epoch": 2.455862027788159,
      "grad_norm": 0.65234375,
      "learning_rate": 1.774333044881611e-05,
      "loss": 0.3885905027389526,
      "num_input_tokens_seen": 19741271552,
      "step": 69410,
      "train_runtime": 674915.8755,
      "train_tokens_per_second": 29249.974
    },
    {
      "epoch": 2.4562158480518876,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7742213342769734e-05,
      "loss": 0.38676636219024657,
      "num_input_tokens_seen": 19744131584,
      "step": 69420,
      "train_runtime": 675013.0214,
      "train_tokens_per_second": 29250.001
    },
    {
      "epoch": 2.4565696683156166,
      "grad_norm": 0.625,
      "learning_rate": 1.7741096447693133e-05,
      "loss": 0.3845911741256714,
      "num_input_tokens_seen": 19747028800,
      "step": 69430,
      "train_runtime": 675113.8845,
      "train_tokens_per_second": 29249.923
    },
    {
      "epoch": 2.456923488579345,
      "grad_norm": 0.625,
      "learning_rate": 1.7739979763519905e-05,
      "loss": 0.3830073833465576,
      "num_input_tokens_seen": 19749886016,
      "step": 69440,
      "train_runtime": 675209.0608,
      "train_tokens_per_second": 29250.031
    },
    {
      "epoch": 2.457277308843074,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7738863290183695e-05,
      "loss": 0.37986533641815184,
      "num_input_tokens_seen": 19752744512,
      "step": 69450,
      "train_runtime": 675308.4248,
      "train_tokens_per_second": 29249.96
    },
    {
      "epoch": 2.4576311291068027,
      "grad_norm": 0.609375,
      "learning_rate": 1.7737747027618156e-05,
      "loss": 0.38946533203125,
      "num_input_tokens_seen": 19755592768,
      "step": 69460,
      "train_runtime": 675405.4356,
      "train_tokens_per_second": 29249.976
    },
    {
      "epoch": 2.4579849493705317,
      "grad_norm": 0.609375,
      "learning_rate": 1.7736630975756982e-05,
      "loss": 0.3870500087738037,
      "num_input_tokens_seen": 19758432768,
      "step": 69470,
      "train_runtime": 675503.4735,
      "train_tokens_per_second": 29249.935
    },
    {
      "epoch": 2.4583387696342602,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7735515134533902e-05,
      "loss": 0.3837573528289795,
      "num_input_tokens_seen": 19761284928,
      "step": 69480,
      "train_runtime": 675602.5976,
      "train_tokens_per_second": 29249.865
    },
    {
      "epoch": 2.4586925898979892,
      "grad_norm": 0.61328125,
      "learning_rate": 1.773439950388265e-05,
      "loss": 0.38840084075927733,
      "num_input_tokens_seen": 19764132992,
      "step": 69490,
      "train_runtime": 675700.018,
      "train_tokens_per_second": 29249.863
    },
    {
      "epoch": 2.4590464101617178,
      "grad_norm": 0.66796875,
      "learning_rate": 1.773328408373702e-05,
      "loss": 0.3912993907928467,
      "num_input_tokens_seen": 19766993728,
      "step": 69500,
      "train_runtime": 675798.5429,
      "train_tokens_per_second": 29249.832
    },
    {
      "epoch": 2.4594002304254468,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7732168874030813e-05,
      "loss": 0.3795579433441162,
      "num_input_tokens_seen": 19769818432,
      "step": 69510,
      "train_runtime": 675893.3566,
      "train_tokens_per_second": 29249.908
    },
    {
      "epoch": 2.4597540506891757,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7731053874697875e-05,
      "loss": 0.3902834177017212,
      "num_input_tokens_seen": 19772655424,
      "step": 69520,
      "train_runtime": 675990.9674,
      "train_tokens_per_second": 29249.881
    },
    {
      "epoch": 2.4601078709529043,
      "grad_norm": 0.625,
      "learning_rate": 1.7729939085672064e-05,
      "loss": 0.3869187593460083,
      "num_input_tokens_seen": 19775456576,
      "step": 69530,
      "train_runtime": 676085.493,
      "train_tokens_per_second": 29249.935
    },
    {
      "epoch": 2.4604616912166333,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7728824506887278e-05,
      "loss": 0.3830282211303711,
      "num_input_tokens_seen": 19778306176,
      "step": 69540,
      "train_runtime": 676182.4485,
      "train_tokens_per_second": 29249.955
    },
    {
      "epoch": 2.460815511480362,
      "grad_norm": 0.625,
      "learning_rate": 1.7727710138277448e-05,
      "loss": 0.38532142639160155,
      "num_input_tokens_seen": 19781158528,
      "step": 69550,
      "train_runtime": 676279.2555,
      "train_tokens_per_second": 29249.986
    },
    {
      "epoch": 2.461169331744091,
      "grad_norm": 0.65234375,
      "learning_rate": 1.7726595979776522e-05,
      "loss": 0.38169758319854735,
      "num_input_tokens_seen": 19783951744,
      "step": 69560,
      "train_runtime": 676373.044,
      "train_tokens_per_second": 29250.059
    },
    {
      "epoch": 2.4615231520078193,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7725482031318487e-05,
      "loss": 0.38606102466583253,
      "num_input_tokens_seen": 19786827456,
      "step": 69570,
      "train_runtime": 676475.1798,
      "train_tokens_per_second": 29249.894
    },
    {
      "epoch": 2.4618769722715483,
      "grad_norm": 0.6015625,
      "learning_rate": 1.772436829283735e-05,
      "loss": 0.38149051666259765,
      "num_input_tokens_seen": 19789700992,
      "step": 69580,
      "train_runtime": 676574.3379,
      "train_tokens_per_second": 29249.855
    },
    {
      "epoch": 2.462230792535277,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7723254764267157e-05,
      "loss": 0.3839037179946899,
      "num_input_tokens_seen": 19792513024,
      "step": 69590,
      "train_runtime": 676670.2785,
      "train_tokens_per_second": 29249.863
    },
    {
      "epoch": 2.462584612799006,
      "grad_norm": 0.62890625,
      "learning_rate": 1.772214144554198e-05,
      "loss": 0.3879279613494873,
      "num_input_tokens_seen": 19795376896,
      "step": 69600,
      "train_runtime": 676767.6194,
      "train_tokens_per_second": 29249.888
    },
    {
      "epoch": 2.4629384330627344,
      "grad_norm": 0.58984375,
      "learning_rate": 1.772102833659591e-05,
      "loss": 0.392243480682373,
      "num_input_tokens_seen": 19798204352,
      "step": 69610,
      "train_runtime": 676864.2648,
      "train_tokens_per_second": 29249.889
    },
    {
      "epoch": 2.4632922533264634,
      "grad_norm": 0.60546875,
      "learning_rate": 1.771991543736308e-05,
      "loss": 0.38358900547027586,
      "num_input_tokens_seen": 19801081152,
      "step": 69620,
      "train_runtime": 676961.8303,
      "train_tokens_per_second": 29249.923
    },
    {
      "epoch": 2.463646073590192,
      "grad_norm": 0.66796875,
      "learning_rate": 1.7718802747777656e-05,
      "loss": 0.38689651489257815,
      "num_input_tokens_seen": 19803958720,
      "step": 69630,
      "train_runtime": 677062.4267,
      "train_tokens_per_second": 29249.827
    },
    {
      "epoch": 2.463999893853921,
      "grad_norm": 0.61328125,
      "learning_rate": 1.771769026777381e-05,
      "loss": 0.39035146236419677,
      "num_input_tokens_seen": 19806807360,
      "step": 69640,
      "train_runtime": 677158.941,
      "train_tokens_per_second": 29249.865
    },
    {
      "epoch": 2.4643537141176495,
      "grad_norm": 0.609375,
      "learning_rate": 1.7716577997285763e-05,
      "loss": 0.38914761543273924,
      "num_input_tokens_seen": 19809599360,
      "step": 69650,
      "train_runtime": 677255.5234,
      "train_tokens_per_second": 29249.816
    },
    {
      "epoch": 2.4647075343813785,
      "grad_norm": 0.62890625,
      "learning_rate": 1.771546593624776e-05,
      "loss": 0.3914644241333008,
      "num_input_tokens_seen": 19812505984,
      "step": 69660,
      "train_runtime": 677353.529,
      "train_tokens_per_second": 29249.875
    },
    {
      "epoch": 2.4650613546451074,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7714354084594077e-05,
      "loss": 0.3835927963256836,
      "num_input_tokens_seen": 19815300608,
      "step": 69670,
      "train_runtime": 677446.42,
      "train_tokens_per_second": 29249.989
    },
    {
      "epoch": 2.465415174908836,
      "grad_norm": 0.58984375,
      "learning_rate": 1.7713242442259005e-05,
      "loss": 0.38383283615112307,
      "num_input_tokens_seen": 19818156096,
      "step": 69680,
      "train_runtime": 677544.0532,
      "train_tokens_per_second": 29249.989
    },
    {
      "epoch": 2.4657689951725645,
      "grad_norm": 0.60546875,
      "learning_rate": 1.771213100917688e-05,
      "loss": 0.3838203907012939,
      "num_input_tokens_seen": 19821056256,
      "step": 69690,
      "train_runtime": 677644.1262,
      "train_tokens_per_second": 29249.949
    },
    {
      "epoch": 2.4661228154362935,
      "grad_norm": 0.6171875,
      "learning_rate": 1.771101978528207e-05,
      "loss": 0.3864152431488037,
      "num_input_tokens_seen": 19823867648,
      "step": 69700,
      "train_runtime": 677738.432,
      "train_tokens_per_second": 29250.027
    },
    {
      "epoch": 2.4664766357000225,
      "grad_norm": 0.6171875,
      "learning_rate": 1.770990877050895e-05,
      "loss": 0.38721094131469724,
      "num_input_tokens_seen": 19826690560,
      "step": 69710,
      "train_runtime": 677835.239,
      "train_tokens_per_second": 29250.014
    },
    {
      "epoch": 2.466830455963751,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7708797964791944e-05,
      "loss": 0.383952260017395,
      "num_input_tokens_seen": 19829525056,
      "step": 69720,
      "train_runtime": 677931.3943,
      "train_tokens_per_second": 29250.047
    },
    {
      "epoch": 2.46718427622748,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7707687368065497e-05,
      "loss": 0.38641607761383057,
      "num_input_tokens_seen": 19832363904,
      "step": 69730,
      "train_runtime": 678030.0622,
      "train_tokens_per_second": 29249.977
    },
    {
      "epoch": 2.4675380964912086,
      "grad_norm": 0.625,
      "learning_rate": 1.7706576980264086e-05,
      "loss": 0.38266696929931643,
      "num_input_tokens_seen": 19835189952,
      "step": 69740,
      "train_runtime": 678125.3478,
      "train_tokens_per_second": 29250.035
    },
    {
      "epoch": 2.4678919167549376,
      "grad_norm": 0.63671875,
      "learning_rate": 1.770546680132221e-05,
      "loss": 0.3842828035354614,
      "num_input_tokens_seen": 19838050304,
      "step": 69750,
      "train_runtime": 678223.5087,
      "train_tokens_per_second": 29250.019
    },
    {
      "epoch": 2.468245737018666,
      "grad_norm": 0.5859375,
      "learning_rate": 1.7704356831174402e-05,
      "loss": 0.3810943841934204,
      "num_input_tokens_seen": 19840935616,
      "step": 69760,
      "train_runtime": 678322.278,
      "train_tokens_per_second": 29250.013
    },
    {
      "epoch": 2.468599557282395,
      "grad_norm": 0.59375,
      "learning_rate": 1.770324706975523e-05,
      "loss": 0.3846139907836914,
      "num_input_tokens_seen": 19843782208,
      "step": 69770,
      "train_runtime": 678417.6898,
      "train_tokens_per_second": 29250.095
    },
    {
      "epoch": 2.4689533775461237,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7702137516999267e-05,
      "loss": 0.3798831462860107,
      "num_input_tokens_seen": 19846603840,
      "step": 69780,
      "train_runtime": 678513.006,
      "train_tokens_per_second": 29250.145
    },
    {
      "epoch": 2.4693071978098526,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7701028172841147e-05,
      "loss": 0.38324086666107177,
      "num_input_tokens_seen": 19849473472,
      "step": 69790,
      "train_runtime": 678613.9837,
      "train_tokens_per_second": 29250.021
    },
    {
      "epoch": 2.469661018073581,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7699919037215512e-05,
      "loss": 0.38632800579071047,
      "num_input_tokens_seen": 19852302208,
      "step": 69800,
      "train_runtime": 678709.1298,
      "train_tokens_per_second": 29250.089
    },
    {
      "epoch": 2.47001483833731,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7698810110057035e-05,
      "loss": 0.38391814231872556,
      "num_input_tokens_seen": 19855086656,
      "step": 69810,
      "train_runtime": 678804.4673,
      "train_tokens_per_second": 29250.082
    },
    {
      "epoch": 2.4703686586010387,
      "grad_norm": 0.62890625,
      "learning_rate": 1.769770139130042e-05,
      "loss": 0.3902827501296997,
      "num_input_tokens_seen": 19857882560,
      "step": 69820,
      "train_runtime": 678900.4313,
      "train_tokens_per_second": 29250.066
    },
    {
      "epoch": 2.4707224788647677,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7696592880880408e-05,
      "loss": 0.3875112056732178,
      "num_input_tokens_seen": 19860724608,
      "step": 69830,
      "train_runtime": 678998.5732,
      "train_tokens_per_second": 29250.024
    },
    {
      "epoch": 2.4710762991284962,
      "grad_norm": 0.66796875,
      "learning_rate": 1.769548457873175e-05,
      "loss": 0.3854401111602783,
      "num_input_tokens_seen": 19863600896,
      "step": 69840,
      "train_runtime": 679098.1213,
      "train_tokens_per_second": 29249.972
    },
    {
      "epoch": 2.4714301193922252,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7694376484789247e-05,
      "loss": 0.3867147207260132,
      "num_input_tokens_seen": 19866412352,
      "step": 69850,
      "train_runtime": 679193.9332,
      "train_tokens_per_second": 29249.985
    },
    {
      "epoch": 2.471783939655954,
      "grad_norm": 0.59375,
      "learning_rate": 1.7693268598987705e-05,
      "loss": 0.3834913969039917,
      "num_input_tokens_seen": 19869221248,
      "step": 69860,
      "train_runtime": 679291.8624,
      "train_tokens_per_second": 29249.903
    },
    {
      "epoch": 2.4721377599196828,
      "grad_norm": 0.65625,
      "learning_rate": 1.769216092126198e-05,
      "loss": 0.38513617515563964,
      "num_input_tokens_seen": 19872089536,
      "step": 69870,
      "train_runtime": 679389.6569,
      "train_tokens_per_second": 29249.915
    },
    {
      "epoch": 2.4724915801834118,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7691053451546948e-05,
      "loss": 0.38790097236633303,
      "num_input_tokens_seen": 19874967808,
      "step": 69880,
      "train_runtime": 679488.2907,
      "train_tokens_per_second": 29249.905
    },
    {
      "epoch": 2.4728454004471403,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7689946189777502e-05,
      "loss": 0.38289918899536135,
      "num_input_tokens_seen": 19877793088,
      "step": 69890,
      "train_runtime": 679582.0751,
      "train_tokens_per_second": 29250.026
    },
    {
      "epoch": 2.4731992207108693,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7688839135888594e-05,
      "loss": 0.3907016277313232,
      "num_input_tokens_seen": 19880589376,
      "step": 69900,
      "train_runtime": 679680.713,
      "train_tokens_per_second": 29249.895
    },
    {
      "epoch": 2.473553040974598,
      "grad_norm": 0.62890625,
      "learning_rate": 1.768773228981517e-05,
      "loss": 0.38656439781188967,
      "num_input_tokens_seen": 19883474944,
      "step": 69910,
      "train_runtime": 679778.0774,
      "train_tokens_per_second": 29249.95
    },
    {
      "epoch": 2.473906861238327,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7686625651492226e-05,
      "loss": 0.3838407754898071,
      "num_input_tokens_seen": 19886367424,
      "step": 69920,
      "train_runtime": 679875.7261,
      "train_tokens_per_second": 29250.004
    },
    {
      "epoch": 2.4742606815020554,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7685519220854775e-05,
      "loss": 0.3877925634384155,
      "num_input_tokens_seen": 19889233600,
      "step": 69930,
      "train_runtime": 679977.644,
      "train_tokens_per_second": 29249.835
    },
    {
      "epoch": 2.4746145017657843,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7684412997837872e-05,
      "loss": 0.38650951385498045,
      "num_input_tokens_seen": 19892081472,
      "step": 69940,
      "train_runtime": 680077.006,
      "train_tokens_per_second": 29249.749
    },
    {
      "epoch": 2.474968322029513,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7683306982376585e-05,
      "loss": 0.3828622102737427,
      "num_input_tokens_seen": 19894934656,
      "step": 69950,
      "train_runtime": 680175.6842,
      "train_tokens_per_second": 29249.7
    },
    {
      "epoch": 2.475322142293242,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7682201174406023e-05,
      "loss": 0.3817132949829102,
      "num_input_tokens_seen": 19897847616,
      "step": 69960,
      "train_runtime": 680276.9111,
      "train_tokens_per_second": 29249.63
    },
    {
      "epoch": 2.4756759625569704,
      "grad_norm": 0.65234375,
      "learning_rate": 1.7681095573861314e-05,
      "loss": 0.3854166269302368,
      "num_input_tokens_seen": 19900666432,
      "step": 69970,
      "train_runtime": 680373.374,
      "train_tokens_per_second": 29249.626
    },
    {
      "epoch": 2.4760297828206994,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7679990180677623e-05,
      "loss": 0.38974804878234864,
      "num_input_tokens_seen": 19903458944,
      "step": 69980,
      "train_runtime": 680466.8818,
      "train_tokens_per_second": 29249.71
    },
    {
      "epoch": 2.476383603084428,
      "grad_norm": 0.62890625,
      "learning_rate": 1.767888499479013e-05,
      "loss": 0.3887038707733154,
      "num_input_tokens_seen": 19906279488,
      "step": 69990,
      "train_runtime": 680563.327,
      "train_tokens_per_second": 29249.709
    },
    {
      "epoch": 2.476737423348157,
      "grad_norm": 0.59765625,
      "learning_rate": 1.767778001613406e-05,
      "loss": 0.3845175266265869,
      "num_input_tokens_seen": 19909106304,
      "step": 70000,
      "train_runtime": 680660.7774,
      "train_tokens_per_second": 29249.675
    },
    {
      "epoch": 2.476737423348157,
      "eval_loss": 0.3363865911960602,
      "eval_runtime": 8.4244,
      "eval_samples_per_second": 473.389,
      "eval_steps_per_second": 3.799,
      "num_input_tokens_seen": 19909106304,
      "step": 70000
    },
    {
      "epoch": 2.477091243611886,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7676675244644658e-05,
      "loss": 0.38898544311523436,
      "num_input_tokens_seen": 19911947520,
      "step": 70010,
      "train_runtime": 680785.4424,
      "train_tokens_per_second": 29248.492
    },
    {
      "epoch": 2.4774450638756145,
      "grad_norm": 0.65234375,
      "learning_rate": 1.7675570680257193e-05,
      "loss": 0.38664755821228025,
      "num_input_tokens_seen": 19914752768,
      "step": 70020,
      "train_runtime": 680879.5656,
      "train_tokens_per_second": 29248.569
    },
    {
      "epoch": 2.477798884139343,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7674466322906973e-05,
      "loss": 0.38545091152191163,
      "num_input_tokens_seen": 19917633088,
      "step": 70030,
      "train_runtime": 680979.0432,
      "train_tokens_per_second": 29248.526
    },
    {
      "epoch": 2.478152704403072,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7673362172529324e-05,
      "loss": 0.38661556243896483,
      "num_input_tokens_seen": 19920416832,
      "step": 70040,
      "train_runtime": 681071.5682,
      "train_tokens_per_second": 29248.64
    },
    {
      "epoch": 2.478506524666801,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7672258229059605e-05,
      "loss": 0.38671653270721434,
      "num_input_tokens_seen": 19923288640,
      "step": 70050,
      "train_runtime": 681166.3388,
      "train_tokens_per_second": 29248.786
    },
    {
      "epoch": 2.4788603449305295,
      "grad_norm": 0.6015625,
      "learning_rate": 1.76711544924332e-05,
      "loss": 0.3889922142028809,
      "num_input_tokens_seen": 19926118016,
      "step": 70060,
      "train_runtime": 681260.3676,
      "train_tokens_per_second": 29248.902
    },
    {
      "epoch": 2.4792141651942585,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7670050962585536e-05,
      "loss": 0.3879117012023926,
      "num_input_tokens_seen": 19928952320,
      "step": 70070,
      "train_runtime": 681357.4297,
      "train_tokens_per_second": 29248.896
    },
    {
      "epoch": 2.479567985457987,
      "grad_norm": 0.609375,
      "learning_rate": 1.7668947639452045e-05,
      "loss": 0.39149041175842286,
      "num_input_tokens_seen": 19931790464,
      "step": 70080,
      "train_runtime": 681455.0622,
      "train_tokens_per_second": 29248.87
    },
    {
      "epoch": 2.479921805721716,
      "grad_norm": 0.62890625,
      "learning_rate": 1.76678445229682e-05,
      "loss": 0.3885554313659668,
      "num_input_tokens_seen": 19934580160,
      "step": 70090,
      "train_runtime": 681548.5404,
      "train_tokens_per_second": 29248.951
    },
    {
      "epoch": 2.4802756259854446,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7666741613069505e-05,
      "loss": 0.3846467971801758,
      "num_input_tokens_seen": 19937440128,
      "step": 70100,
      "train_runtime": 681645.3253,
      "train_tokens_per_second": 29248.994
    },
    {
      "epoch": 2.4806294462491736,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7665638909691484e-05,
      "loss": 0.38697223663330077,
      "num_input_tokens_seen": 19940313664,
      "step": 70110,
      "train_runtime": 681746.0083,
      "train_tokens_per_second": 29248.889
    },
    {
      "epoch": 2.480983266512902,
      "grad_norm": 0.60546875,
      "learning_rate": 1.766453641276969e-05,
      "loss": 0.3926079273223877,
      "num_input_tokens_seen": 19943147776,
      "step": 70120,
      "train_runtime": 681844.3728,
      "train_tokens_per_second": 29248.827
    },
    {
      "epoch": 2.481337086776631,
      "grad_norm": 0.625,
      "learning_rate": 1.7663434122239723e-05,
      "loss": 0.3873988151550293,
      "num_input_tokens_seen": 19945951232,
      "step": 70130,
      "train_runtime": 681940.3578,
      "train_tokens_per_second": 29248.821
    },
    {
      "epoch": 2.4816909070403597,
      "grad_norm": 0.578125,
      "learning_rate": 1.7662332038037176e-05,
      "loss": 0.38596668243408205,
      "num_input_tokens_seen": 19948818112,
      "step": 70140,
      "train_runtime": 682039.9796,
      "train_tokens_per_second": 29248.752
    },
    {
      "epoch": 2.4820447273040886,
      "grad_norm": 0.6328125,
      "learning_rate": 1.76612301600977e-05,
      "loss": 0.3911670446395874,
      "num_input_tokens_seen": 19951633920,
      "step": 70150,
      "train_runtime": 682134.1734,
      "train_tokens_per_second": 29248.841
    },
    {
      "epoch": 2.4823985475678176,
      "grad_norm": 0.640625,
      "learning_rate": 1.7660128488356968e-05,
      "loss": 0.3846114635467529,
      "num_input_tokens_seen": 19954501632,
      "step": 70160,
      "train_runtime": 682232.3839,
      "train_tokens_per_second": 29248.834
    },
    {
      "epoch": 2.482752367831546,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7659027022750667e-05,
      "loss": 0.38332276344299315,
      "num_input_tokens_seen": 19957396288,
      "step": 70170,
      "train_runtime": 682331.8456,
      "train_tokens_per_second": 29248.813
    },
    {
      "epoch": 2.4831061880952747,
      "grad_norm": 0.62890625,
      "learning_rate": 1.765792576321452e-05,
      "loss": 0.3880870819091797,
      "num_input_tokens_seen": 19960188992,
      "step": 70180,
      "train_runtime": 682426.8258,
      "train_tokens_per_second": 29248.834
    },
    {
      "epoch": 2.4834600083590037,
      "grad_norm": 0.625,
      "learning_rate": 1.7656824709684287e-05,
      "loss": 0.38588049411773684,
      "num_input_tokens_seen": 19963016256,
      "step": 70190,
      "train_runtime": 682520.8258,
      "train_tokens_per_second": 29248.948
    },
    {
      "epoch": 2.4838138286227327,
      "grad_norm": 0.6171875,
      "learning_rate": 1.765572386209575e-05,
      "loss": 0.3874520778656006,
      "num_input_tokens_seen": 19965854144,
      "step": 70200,
      "train_runtime": 682619.522,
      "train_tokens_per_second": 29248.877
    },
    {
      "epoch": 2.4841676488864612,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7654623220384713e-05,
      "loss": 0.39153075218200684,
      "num_input_tokens_seen": 19968755072,
      "step": 70210,
      "train_runtime": 682719.049,
      "train_tokens_per_second": 29248.862
    },
    {
      "epoch": 2.4845214691501902,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7653522784487014e-05,
      "loss": 0.38494038581848145,
      "num_input_tokens_seen": 19971626880,
      "step": 70220,
      "train_runtime": 682820.5947,
      "train_tokens_per_second": 29248.718
    },
    {
      "epoch": 2.4848752894139188,
      "grad_norm": 0.609375,
      "learning_rate": 1.765242255433852e-05,
      "loss": 0.3874661445617676,
      "num_input_tokens_seen": 19974504704,
      "step": 70230,
      "train_runtime": 682918.4584,
      "train_tokens_per_second": 29248.74
    },
    {
      "epoch": 2.4852291096776478,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7651322529875126e-05,
      "loss": 0.38773882389068604,
      "num_input_tokens_seen": 19977307968,
      "step": 70240,
      "train_runtime": 683013.4007,
      "train_tokens_per_second": 29248.779
    },
    {
      "epoch": 2.4855829299413763,
      "grad_norm": 0.625,
      "learning_rate": 1.7650222711032744e-05,
      "loss": 0.38625500202178953,
      "num_input_tokens_seen": 19980154560,
      "step": 70250,
      "train_runtime": 683110.5157,
      "train_tokens_per_second": 29248.788
    },
    {
      "epoch": 2.4859367502051053,
      "grad_norm": 0.62890625,
      "learning_rate": 1.764912309774733e-05,
      "loss": 0.38571677207946775,
      "num_input_tokens_seen": 19982925312,
      "step": 70260,
      "train_runtime": 683202.7717,
      "train_tokens_per_second": 29248.894
    },
    {
      "epoch": 2.486290570468834,
      "grad_norm": 0.60546875,
      "learning_rate": 1.764802368995486e-05,
      "loss": 0.3853276252746582,
      "num_input_tokens_seen": 19985811008,
      "step": 70270,
      "train_runtime": 683300.5099,
      "train_tokens_per_second": 29248.933
    },
    {
      "epoch": 2.486644390732563,
      "grad_norm": 0.61328125,
      "learning_rate": 1.764692448759134e-05,
      "loss": 0.38460555076599123,
      "num_input_tokens_seen": 19988728576,
      "step": 70280,
      "train_runtime": 683400.6412,
      "train_tokens_per_second": 29248.917
    },
    {
      "epoch": 2.4869982109962914,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7645825490592797e-05,
      "loss": 0.38587079048156736,
      "num_input_tokens_seen": 19991587584,
      "step": 70290,
      "train_runtime": 683501.763,
      "train_tokens_per_second": 29248.773
    },
    {
      "epoch": 2.4873520312600204,
      "grad_norm": 0.609375,
      "learning_rate": 1.7644726698895296e-05,
      "loss": 0.38921346664428713,
      "num_input_tokens_seen": 19994424064,
      "step": 70300,
      "train_runtime": 683597.2451,
      "train_tokens_per_second": 29248.837
    },
    {
      "epoch": 2.487705851523749,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7643628112434918e-05,
      "loss": 0.38572556972503663,
      "num_input_tokens_seen": 19997281344,
      "step": 70310,
      "train_runtime": 683694.1972,
      "train_tokens_per_second": 29248.868
    },
    {
      "epoch": 2.488059671787478,
      "grad_norm": 0.640625,
      "learning_rate": 1.764252973114779e-05,
      "loss": 0.38218538761138915,
      "num_input_tokens_seen": 20000128448,
      "step": 70320,
      "train_runtime": 683791.4072,
      "train_tokens_per_second": 29248.874
    },
    {
      "epoch": 2.4884134920512064,
      "grad_norm": 0.59375,
      "learning_rate": 1.764143155497005e-05,
      "loss": 0.3856635093688965,
      "num_input_tokens_seen": 20002925120,
      "step": 70330,
      "train_runtime": 683885.9125,
      "train_tokens_per_second": 29248.921
    },
    {
      "epoch": 2.4887673123149354,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7640333583837874e-05,
      "loss": 0.3883447408676147,
      "num_input_tokens_seen": 20005764160,
      "step": 70340,
      "train_runtime": 683983.4951,
      "train_tokens_per_second": 29248.899
    },
    {
      "epoch": 2.4891211325786644,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7639235817687455e-05,
      "loss": 0.38215930461883546,
      "num_input_tokens_seen": 20008615616,
      "step": 70350,
      "train_runtime": 684081.5611,
      "train_tokens_per_second": 29248.874
    },
    {
      "epoch": 2.489474952842393,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7638138256455024e-05,
      "loss": 0.3882733345031738,
      "num_input_tokens_seen": 20011465024,
      "step": 70360,
      "train_runtime": 684178.6784,
      "train_tokens_per_second": 29248.887
    },
    {
      "epoch": 2.489828773106122,
      "grad_norm": 0.609375,
      "learning_rate": 1.7637040900076837e-05,
      "loss": 0.3836606740951538,
      "num_input_tokens_seen": 20014335744,
      "step": 70370,
      "train_runtime": 684276.0441,
      "train_tokens_per_second": 29248.921
    },
    {
      "epoch": 2.4901825933698505,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7635943748489174e-05,
      "loss": 0.38716998100280764,
      "num_input_tokens_seen": 20017177792,
      "step": 70380,
      "train_runtime": 684374.4691,
      "train_tokens_per_second": 29248.867
    },
    {
      "epoch": 2.4905364136335795,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7634846801628348e-05,
      "loss": 0.37952208518981934,
      "num_input_tokens_seen": 20020005760,
      "step": 70390,
      "train_runtime": 684470.0219,
      "train_tokens_per_second": 29248.915
    },
    {
      "epoch": 2.490890233897308,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7633750059430698e-05,
      "loss": 0.3838933229446411,
      "num_input_tokens_seen": 20022855936,
      "step": 70400,
      "train_runtime": 684566.9677,
      "train_tokens_per_second": 29248.937
    },
    {
      "epoch": 2.491244054161037,
      "grad_norm": 0.62109375,
      "learning_rate": 1.763265352183259e-05,
      "loss": 0.38312368392944335,
      "num_input_tokens_seen": 20025694464,
      "step": 70410,
      "train_runtime": 684660.9776,
      "train_tokens_per_second": 29249.066
    },
    {
      "epoch": 2.4915978744247655,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7631557188770415e-05,
      "loss": 0.38325514793396,
      "num_input_tokens_seen": 20028502272,
      "step": 70420,
      "train_runtime": 684755.0196,
      "train_tokens_per_second": 29249.15
    },
    {
      "epoch": 2.4919516946884945,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7630461060180593e-05,
      "loss": 0.38318958282470705,
      "num_input_tokens_seen": 20031350080,
      "step": 70430,
      "train_runtime": 684849.4729,
      "train_tokens_per_second": 29249.274
    },
    {
      "epoch": 2.492305514952223,
      "grad_norm": 0.609375,
      "learning_rate": 1.762936513599958e-05,
      "loss": 0.3877424240112305,
      "num_input_tokens_seen": 20034196416,
      "step": 70440,
      "train_runtime": 684948.1704,
      "train_tokens_per_second": 29249.215
    },
    {
      "epoch": 2.492659335215952,
      "grad_norm": 0.609375,
      "learning_rate": 1.7628269416163852e-05,
      "loss": 0.3866433620452881,
      "num_input_tokens_seen": 20036979200,
      "step": 70450,
      "train_runtime": 685042.9028,
      "train_tokens_per_second": 29249.233
    },
    {
      "epoch": 2.4930131554796806,
      "grad_norm": 0.625,
      "learning_rate": 1.7627173900609905e-05,
      "loss": 0.38688628673553466,
      "num_input_tokens_seen": 20039828608,
      "step": 70460,
      "train_runtime": 685142.0571,
      "train_tokens_per_second": 29249.158
    },
    {
      "epoch": 2.4933669757434096,
      "grad_norm": 0.62109375,
      "learning_rate": 1.762607858927428e-05,
      "loss": 0.38215885162353513,
      "num_input_tokens_seen": 20042683840,
      "step": 70470,
      "train_runtime": 685238.9604,
      "train_tokens_per_second": 29249.189
    },
    {
      "epoch": 2.493720796007138,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7624983482093533e-05,
      "loss": 0.3901267766952515,
      "num_input_tokens_seen": 20045562304,
      "step": 70480,
      "train_runtime": 685340.836,
      "train_tokens_per_second": 29249.041
    },
    {
      "epoch": 2.494074616270867,
      "grad_norm": 0.62890625,
      "learning_rate": 1.762388857900425e-05,
      "loss": 0.38214433193206787,
      "num_input_tokens_seen": 20048406016,
      "step": 70490,
      "train_runtime": 685437.9874,
      "train_tokens_per_second": 29249.044
    },
    {
      "epoch": 2.494428436534596,
      "grad_norm": 0.609375,
      "learning_rate": 1.762279387994305e-05,
      "loss": 0.3872185230255127,
      "num_input_tokens_seen": 20051217472,
      "step": 70500,
      "train_runtime": 685531.6317,
      "train_tokens_per_second": 29249.15
    },
    {
      "epoch": 2.4947822567983247,
      "grad_norm": 0.609375,
      "learning_rate": 1.7621699384846572e-05,
      "loss": 0.38647022247314455,
      "num_input_tokens_seen": 20054047872,
      "step": 70510,
      "train_runtime": 685626.8472,
      "train_tokens_per_second": 29249.216
    },
    {
      "epoch": 2.495136077062053,
      "grad_norm": 0.59375,
      "learning_rate": 1.7620605093651485e-05,
      "loss": 0.38982081413269043,
      "num_input_tokens_seen": 20056927616,
      "step": 70520,
      "train_runtime": 685728.5697,
      "train_tokens_per_second": 29249.077
    },
    {
      "epoch": 2.495489897325782,
      "grad_norm": 0.62109375,
      "learning_rate": 1.761951100629449e-05,
      "loss": 0.3834831714630127,
      "num_input_tokens_seen": 20059779904,
      "step": 70530,
      "train_runtime": 685826.1375,
      "train_tokens_per_second": 29249.075
    },
    {
      "epoch": 2.495843717589511,
      "grad_norm": 0.625,
      "learning_rate": 1.7618417122712306e-05,
      "loss": 0.3838294506072998,
      "num_input_tokens_seen": 20062643328,
      "step": 70540,
      "train_runtime": 685923.7298,
      "train_tokens_per_second": 29249.088
    },
    {
      "epoch": 2.4961975378532397,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7617323442841694e-05,
      "loss": 0.3840170860290527,
      "num_input_tokens_seen": 20065479360,
      "step": 70550,
      "train_runtime": 686020.4895,
      "train_tokens_per_second": 29249.096
    },
    {
      "epoch": 2.4965513581169687,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7616229966619426e-05,
      "loss": 0.3809061527252197,
      "num_input_tokens_seen": 20068353216,
      "step": 70560,
      "train_runtime": 686120.5821,
      "train_tokens_per_second": 29249.018
    },
    {
      "epoch": 2.4969051783806973,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7615136693982312e-05,
      "loss": 0.39016356468200686,
      "num_input_tokens_seen": 20071205120,
      "step": 70570,
      "train_runtime": 686219.9509,
      "train_tokens_per_second": 29248.938
    },
    {
      "epoch": 2.4972589986444262,
      "grad_norm": 0.671875,
      "learning_rate": 1.761404362486719e-05,
      "loss": 0.3852728605270386,
      "num_input_tokens_seen": 20073973632,
      "step": 70580,
      "train_runtime": 686313.1148,
      "train_tokens_per_second": 29249.002
    },
    {
      "epoch": 2.497612818908155,
      "grad_norm": 0.6171875,
      "learning_rate": 1.761295075921092e-05,
      "loss": 0.3809437036514282,
      "num_input_tokens_seen": 20076802560,
      "step": 70590,
      "train_runtime": 686410.7386,
      "train_tokens_per_second": 29248.963
    },
    {
      "epoch": 2.4979666391718838,
      "grad_norm": 0.61328125,
      "learning_rate": 1.761185809695039e-05,
      "loss": 0.3876835823059082,
      "num_input_tokens_seen": 20079634944,
      "step": 70600,
      "train_runtime": 686512.6742,
      "train_tokens_per_second": 29248.746
    },
    {
      "epoch": 2.4983204594356123,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7610765638022518e-05,
      "loss": 0.3828484535217285,
      "num_input_tokens_seen": 20082539520,
      "step": 70610,
      "train_runtime": 686614.2438,
      "train_tokens_per_second": 29248.65
    },
    {
      "epoch": 2.4986742796993413,
      "grad_norm": 0.6171875,
      "learning_rate": 1.760967338236425e-05,
      "loss": 0.38417739868164064,
      "num_input_tokens_seen": 20085392384,
      "step": 70620,
      "train_runtime": 686711.1073,
      "train_tokens_per_second": 29248.678
    },
    {
      "epoch": 2.49902809996307,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7608581329912554e-05,
      "loss": 0.38393497467041016,
      "num_input_tokens_seen": 20088244544,
      "step": 70630,
      "train_runtime": 686807.9649,
      "train_tokens_per_second": 29248.706
    },
    {
      "epoch": 2.499381920226799,
      "grad_norm": 0.609375,
      "learning_rate": 1.7607489480604435e-05,
      "loss": 0.3870813846588135,
      "num_input_tokens_seen": 20091088832,
      "step": 70640,
      "train_runtime": 686905.8724,
      "train_tokens_per_second": 29248.678
    },
    {
      "epoch": 2.4997357404905274,
      "grad_norm": 0.62890625,
      "learning_rate": 1.760639783437692e-05,
      "loss": 0.3891451835632324,
      "num_input_tokens_seen": 20093947968,
      "step": 70650,
      "train_runtime": 687004.7149,
      "train_tokens_per_second": 29248.632
    },
    {
      "epoch": 2.5000895607542564,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7605306391167054e-05,
      "loss": 0.39011578559875487,
      "num_input_tokens_seen": 20096796864,
      "step": 70660,
      "train_runtime": 687103.7261,
      "train_tokens_per_second": 29248.563
    },
    {
      "epoch": 2.500443381017985,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7604215150911924e-05,
      "loss": 0.3902756690979004,
      "num_input_tokens_seen": 20099653888,
      "step": 70670,
      "train_runtime": 687203.6403,
      "train_tokens_per_second": 29248.468
    },
    {
      "epoch": 2.500797201281714,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7603124113548636e-05,
      "loss": 0.38825139999389646,
      "num_input_tokens_seen": 20102535744,
      "step": 70680,
      "train_runtime": 687304.2315,
      "train_tokens_per_second": 29248.381
    },
    {
      "epoch": 2.501151021545443,
      "grad_norm": 0.625,
      "learning_rate": 1.7602033279014336e-05,
      "loss": 0.3888502597808838,
      "num_input_tokens_seen": 20105282176,
      "step": 70690,
      "train_runtime": 687396.3185,
      "train_tokens_per_second": 29248.458
    },
    {
      "epoch": 2.5015048418091714,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7600942647246167e-05,
      "loss": 0.38906421661376955,
      "num_input_tokens_seen": 20108109184,
      "step": 70700,
      "train_runtime": 687493.6239,
      "train_tokens_per_second": 29248.43
    },
    {
      "epoch": 2.5018586620729,
      "grad_norm": 0.625,
      "learning_rate": 1.759985221818134e-05,
      "loss": 0.3870364189147949,
      "num_input_tokens_seen": 20110959168,
      "step": 70710,
      "train_runtime": 687588.247,
      "train_tokens_per_second": 29248.55
    },
    {
      "epoch": 2.502212482336629,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7598761991757063e-05,
      "loss": 0.3857097148895264,
      "num_input_tokens_seen": 20113792832,
      "step": 70720,
      "train_runtime": 687682.6265,
      "train_tokens_per_second": 29248.656
    },
    {
      "epoch": 2.502566302600358,
      "grad_norm": 0.64453125,
      "learning_rate": 1.7597671967910576e-05,
      "loss": 0.3848094463348389,
      "num_input_tokens_seen": 20116706048,
      "step": 70730,
      "train_runtime": 687784.033,
      "train_tokens_per_second": 29248.58
    },
    {
      "epoch": 2.5029201228640865,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7596582146579158e-05,
      "loss": 0.3874844551086426,
      "num_input_tokens_seen": 20119553664,
      "step": 70740,
      "train_runtime": 687882.4337,
      "train_tokens_per_second": 29248.535
    },
    {
      "epoch": 2.5032739431278155,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7595492527700113e-05,
      "loss": 0.382014536857605,
      "num_input_tokens_seen": 20122383680,
      "step": 70750,
      "train_runtime": 687980.3416,
      "train_tokens_per_second": 29248.486
    },
    {
      "epoch": 2.503627763391544,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7594403111210754e-05,
      "loss": 0.3867045879364014,
      "num_input_tokens_seen": 20125241664,
      "step": 70760,
      "train_runtime": 688077.0482,
      "train_tokens_per_second": 29248.529
    },
    {
      "epoch": 2.503981583655273,
      "grad_norm": 0.58984375,
      "learning_rate": 1.759331389704844e-05,
      "loss": 0.38740832805633546,
      "num_input_tokens_seen": 20128067520,
      "step": 70770,
      "train_runtime": 688172.1706,
      "train_tokens_per_second": 29248.593
    },
    {
      "epoch": 2.5043354039190016,
      "grad_norm": 0.64453125,
      "learning_rate": 1.7592224885150554e-05,
      "loss": 0.3863245964050293,
      "num_input_tokens_seen": 20130945472,
      "step": 70780,
      "train_runtime": 688269.1459,
      "train_tokens_per_second": 29248.653
    },
    {
      "epoch": 2.5046892241827305,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7591136075454505e-05,
      "loss": 0.3898874282836914,
      "num_input_tokens_seen": 20133811840,
      "step": 70790,
      "train_runtime": 688367.4727,
      "train_tokens_per_second": 29248.639
    },
    {
      "epoch": 2.505043044446459,
      "grad_norm": 0.59765625,
      "learning_rate": 1.759004746789772e-05,
      "loss": 0.3844128608703613,
      "num_input_tokens_seen": 20136662016,
      "step": 70800,
      "train_runtime": 688464.5969,
      "train_tokens_per_second": 29248.653
    },
    {
      "epoch": 2.505396864710188,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7588959062417677e-05,
      "loss": 0.3871614933013916,
      "num_input_tokens_seen": 20139489600,
      "step": 70810,
      "train_runtime": 688560.2503,
      "train_tokens_per_second": 29248.696
    },
    {
      "epoch": 2.5057506849739166,
      "grad_norm": 0.609375,
      "learning_rate": 1.7587870858951845e-05,
      "loss": 0.38546171188354494,
      "num_input_tokens_seen": 20142356672,
      "step": 70820,
      "train_runtime": 688657.9891,
      "train_tokens_per_second": 29248.708
    },
    {
      "epoch": 2.5061045052376456,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7586782857437747e-05,
      "loss": 0.38804049491882325,
      "num_input_tokens_seen": 20145186560,
      "step": 70830,
      "train_runtime": 688755.1921,
      "train_tokens_per_second": 29248.689
    },
    {
      "epoch": 2.5064583255013746,
      "grad_norm": 0.59375,
      "learning_rate": 1.7585695057812936e-05,
      "loss": 0.38854146003723145,
      "num_input_tokens_seen": 20148037056,
      "step": 70840,
      "train_runtime": 688854.7817,
      "train_tokens_per_second": 29248.599
    },
    {
      "epoch": 2.506812145765103,
      "grad_norm": 0.625,
      "learning_rate": 1.7584607460014974e-05,
      "loss": 0.3861550807952881,
      "num_input_tokens_seen": 20150856640,
      "step": 70850,
      "train_runtime": 688952.2104,
      "train_tokens_per_second": 29248.555
    },
    {
      "epoch": 2.5071659660288317,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7583520063981454e-05,
      "loss": 0.382442569732666,
      "num_input_tokens_seen": 20153722560,
      "step": 70860,
      "train_runtime": 689049.1265,
      "train_tokens_per_second": 29248.6
    },
    {
      "epoch": 2.5075197862925607,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7582432869650005e-05,
      "loss": 0.38774971961975097,
      "num_input_tokens_seen": 20156559872,
      "step": 70870,
      "train_runtime": 689147.6676,
      "train_tokens_per_second": 29248.535
    },
    {
      "epoch": 2.5078736065562897,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7581345876958283e-05,
      "loss": 0.38747391700744627,
      "num_input_tokens_seen": 20159407296,
      "step": 70880,
      "train_runtime": 689245.3035,
      "train_tokens_per_second": 29248.523
    },
    {
      "epoch": 2.508227426820018,
      "grad_norm": 0.625,
      "learning_rate": 1.758025908584396e-05,
      "loss": 0.3846898078918457,
      "num_input_tokens_seen": 20162231168,
      "step": 70890,
      "train_runtime": 689342.8341,
      "train_tokens_per_second": 29248.482
    },
    {
      "epoch": 2.508581247083747,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7579172496244738e-05,
      "loss": 0.3857410907745361,
      "num_input_tokens_seen": 20165083776,
      "step": 70900,
      "train_runtime": 689438.1967,
      "train_tokens_per_second": 29248.574
    },
    {
      "epoch": 2.5089350673474757,
      "grad_norm": 0.59375,
      "learning_rate": 1.7578086108098357e-05,
      "loss": 0.3873237371444702,
      "num_input_tokens_seen": 20167938752,
      "step": 70910,
      "train_runtime": 689533.1872,
      "train_tokens_per_second": 29248.685
    },
    {
      "epoch": 2.5092888876112047,
      "grad_norm": 0.58203125,
      "learning_rate": 1.7576999921342574e-05,
      "loss": 0.3880675554275513,
      "num_input_tokens_seen": 20170824000,
      "step": 70920,
      "train_runtime": 689632.4546,
      "train_tokens_per_second": 29248.658
    },
    {
      "epoch": 2.5096427078749333,
      "grad_norm": 0.60546875,
      "learning_rate": 1.757591393591517e-05,
      "loss": 0.3835528135299683,
      "num_input_tokens_seen": 20173757184,
      "step": 70930,
      "train_runtime": 689736.366,
      "train_tokens_per_second": 29248.504
    },
    {
      "epoch": 2.5099965281386623,
      "grad_norm": 0.62109375,
      "learning_rate": 1.757482815175397e-05,
      "loss": 0.3894948959350586,
      "num_input_tokens_seen": 20176624384,
      "step": 70940,
      "train_runtime": 689836.9419,
      "train_tokens_per_second": 29248.396
    },
    {
      "epoch": 2.510350348402391,
      "grad_norm": 0.6640625,
      "learning_rate": 1.7573742568796796e-05,
      "loss": 0.3857612371444702,
      "num_input_tokens_seen": 20179397632,
      "step": 70950,
      "train_runtime": 689926.9181,
      "train_tokens_per_second": 29248.602
    },
    {
      "epoch": 2.51070416866612,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7572657186981528e-05,
      "loss": 0.38314089775085447,
      "num_input_tokens_seen": 20182239744,
      "step": 70960,
      "train_runtime": 690022.9101,
      "train_tokens_per_second": 29248.652
    },
    {
      "epoch": 2.5110579889298483,
      "grad_norm": 0.640625,
      "learning_rate": 1.7571572006246058e-05,
      "loss": 0.3839996337890625,
      "num_input_tokens_seen": 20185052096,
      "step": 70970,
      "train_runtime": 690119.1853,
      "train_tokens_per_second": 29248.647
    },
    {
      "epoch": 2.5114118091935773,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7570487026528298e-05,
      "loss": 0.38002169132232666,
      "num_input_tokens_seen": 20187876608,
      "step": 70980,
      "train_runtime": 690212.8905,
      "train_tokens_per_second": 29248.768
    },
    {
      "epoch": 2.5117656294573063,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7569402247766204e-05,
      "loss": 0.38715014457702634,
      "num_input_tokens_seen": 20190655872,
      "step": 70990,
      "train_runtime": 690309.2956,
      "train_tokens_per_second": 29248.709
    },
    {
      "epoch": 2.512119449721035,
      "grad_norm": 0.62890625,
      "learning_rate": 1.756831766989775e-05,
      "loss": 0.3922735214233398,
      "num_input_tokens_seen": 20193493696,
      "step": 71000,
      "train_runtime": 690404.6582,
      "train_tokens_per_second": 29248.78
    },
    {
      "epoch": 2.5124732699847634,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7567233292860927e-05,
      "loss": 0.3836193084716797,
      "num_input_tokens_seen": 20196396096,
      "step": 71010,
      "train_runtime": 690504.7227,
      "train_tokens_per_second": 29248.744
    },
    {
      "epoch": 2.5128270902484924,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7566149116593772e-05,
      "loss": 0.38806490898132323,
      "num_input_tokens_seen": 20199170688,
      "step": 71020,
      "train_runtime": 690596.4621,
      "train_tokens_per_second": 29248.877
    },
    {
      "epoch": 2.5131809105122214,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7565065141034336e-05,
      "loss": 0.3897226333618164,
      "num_input_tokens_seen": 20202050368,
      "step": 71030,
      "train_runtime": 690696.5447,
      "train_tokens_per_second": 29248.808
    },
    {
      "epoch": 2.51353473077595,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7563981366120702e-05,
      "loss": 0.38810908794403076,
      "num_input_tokens_seen": 20204947520,
      "step": 71040,
      "train_runtime": 690800.6839,
      "train_tokens_per_second": 29248.592
    },
    {
      "epoch": 2.5138885510396785,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7562897791790976e-05,
      "loss": 0.38892536163330077,
      "num_input_tokens_seen": 20207766720,
      "step": 71050,
      "train_runtime": 690896.1682,
      "train_tokens_per_second": 29248.63
    },
    {
      "epoch": 2.5142423713034074,
      "grad_norm": 0.625,
      "learning_rate": 1.7561814417983292e-05,
      "loss": 0.39044952392578125,
      "num_input_tokens_seen": 20210606912,
      "step": 71060,
      "train_runtime": 690991.5395,
      "train_tokens_per_second": 29248.704
    },
    {
      "epoch": 2.5145961915671364,
      "grad_norm": 0.61328125,
      "learning_rate": 1.756073124463581e-05,
      "loss": 0.38892350196838377,
      "num_input_tokens_seen": 20213471168,
      "step": 71070,
      "train_runtime": 691091.4736,
      "train_tokens_per_second": 29248.619
    },
    {
      "epoch": 2.514950011830865,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7559648271686728e-05,
      "loss": 0.3855701684951782,
      "num_input_tokens_seen": 20216309632,
      "step": 71080,
      "train_runtime": 691186.9743,
      "train_tokens_per_second": 29248.684
    },
    {
      "epoch": 2.515303832094594,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7558565499074244e-05,
      "loss": 0.38713946342468264,
      "num_input_tokens_seen": 20219150528,
      "step": 71090,
      "train_runtime": 691283.6061,
      "train_tokens_per_second": 29248.705
    },
    {
      "epoch": 2.5156576523583225,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7557482926736613e-05,
      "loss": 0.3840949535369873,
      "num_input_tokens_seen": 20221969920,
      "step": 71100,
      "train_runtime": 691380.7785,
      "train_tokens_per_second": 29248.672
    },
    {
      "epoch": 2.5160114726220515,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7556400554612095e-05,
      "loss": 0.389097785949707,
      "num_input_tokens_seen": 20224859456,
      "step": 71110,
      "train_runtime": 691479.4911,
      "train_tokens_per_second": 29248.676
    },
    {
      "epoch": 2.51636529288578,
      "grad_norm": 0.6171875,
      "learning_rate": 1.755531838263899e-05,
      "loss": 0.385307240486145,
      "num_input_tokens_seen": 20227731328,
      "step": 71120,
      "train_runtime": 691577.1014,
      "train_tokens_per_second": 29248.7
    },
    {
      "epoch": 2.516719113149509,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7554236410755618e-05,
      "loss": 0.3875169515609741,
      "num_input_tokens_seen": 20230561728,
      "step": 71130,
      "train_runtime": 691672.6368,
      "train_tokens_per_second": 29248.752
    },
    {
      "epoch": 2.517072933413238,
      "grad_norm": 0.640625,
      "learning_rate": 1.7553154638900324e-05,
      "loss": 0.3899528980255127,
      "num_input_tokens_seen": 20233369024,
      "step": 71140,
      "train_runtime": 691767.9708,
      "train_tokens_per_second": 29248.78
    },
    {
      "epoch": 2.5174267536769666,
      "grad_norm": 0.6015625,
      "learning_rate": 1.755207306701148e-05,
      "loss": 0.3865573644638062,
      "num_input_tokens_seen": 20236241920,
      "step": 71150,
      "train_runtime": 691866.8622,
      "train_tokens_per_second": 29248.751
    },
    {
      "epoch": 2.517780573940695,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7550991695027495e-05,
      "loss": 0.3876651763916016,
      "num_input_tokens_seen": 20239093888,
      "step": 71160,
      "train_runtime": 691962.571,
      "train_tokens_per_second": 29248.828
    },
    {
      "epoch": 2.518134394204424,
      "grad_norm": 0.6171875,
      "learning_rate": 1.754991052288679e-05,
      "loss": 0.38529953956604,
      "num_input_tokens_seen": 20241943488,
      "step": 71170,
      "train_runtime": 692058.406,
      "train_tokens_per_second": 29248.895
    },
    {
      "epoch": 2.518488214468153,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7548829550527827e-05,
      "loss": 0.3858953475952148,
      "num_input_tokens_seen": 20244810752,
      "step": 71180,
      "train_runtime": 692155.112,
      "train_tokens_per_second": 29248.951
    },
    {
      "epoch": 2.5188420347318816,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7547748777889075e-05,
      "loss": 0.3898101091384888,
      "num_input_tokens_seen": 20247703104,
      "step": 71190,
      "train_runtime": 692255.2837,
      "train_tokens_per_second": 29248.896
    },
    {
      "epoch": 2.51919585499561,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7546668204909052e-05,
      "loss": 0.39031836986541746,
      "num_input_tokens_seen": 20250545472,
      "step": 71200,
      "train_runtime": 692353.2982,
      "train_tokens_per_second": 29248.861
    },
    {
      "epoch": 2.519549675259339,
      "grad_norm": 0.609375,
      "learning_rate": 1.7545587831526283e-05,
      "loss": 0.3870730400085449,
      "num_input_tokens_seen": 20253413440,
      "step": 71210,
      "train_runtime": 692452.4596,
      "train_tokens_per_second": 29248.814
    },
    {
      "epoch": 2.519903495523068,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7544507657679332e-05,
      "loss": 0.38480219841003416,
      "num_input_tokens_seen": 20256302592,
      "step": 71220,
      "train_runtime": 692551.7593,
      "train_tokens_per_second": 29248.792
    },
    {
      "epoch": 2.5202573157867967,
      "grad_norm": 0.62890625,
      "learning_rate": 1.754342768330679e-05,
      "loss": 0.386008620262146,
      "num_input_tokens_seen": 20259174336,
      "step": 71230,
      "train_runtime": 692650.1319,
      "train_tokens_per_second": 29248.784
    },
    {
      "epoch": 2.5206111360505257,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7542347908347262e-05,
      "loss": 0.38831191062927245,
      "num_input_tokens_seen": 20262036224,
      "step": 71240,
      "train_runtime": 692748.8535,
      "train_tokens_per_second": 29248.747
    },
    {
      "epoch": 2.520964956314254,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7541268332739394e-05,
      "loss": 0.38004751205444337,
      "num_input_tokens_seen": 20264869056,
      "step": 71250,
      "train_runtime": 692847.7769,
      "train_tokens_per_second": 29248.66
    },
    {
      "epoch": 2.521318776577983,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7540188956421847e-05,
      "loss": 0.3855795383453369,
      "num_input_tokens_seen": 20267714688,
      "step": 71260,
      "train_runtime": 692943.6113,
      "train_tokens_per_second": 29248.721
    },
    {
      "epoch": 2.5216725968417117,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7539109779333315e-05,
      "loss": 0.3830139636993408,
      "num_input_tokens_seen": 20270540288,
      "step": 71270,
      "train_runtime": 693040.3268,
      "train_tokens_per_second": 29248.717
    },
    {
      "epoch": 2.5220264171054407,
      "grad_norm": 0.62109375,
      "learning_rate": 1.753803080141252e-05,
      "loss": 0.38417460918426516,
      "num_input_tokens_seen": 20273355968,
      "step": 71280,
      "train_runtime": 693133.9695,
      "train_tokens_per_second": 29248.828
    },
    {
      "epoch": 2.5223802373691693,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7536952022598202e-05,
      "loss": 0.3841129779815674,
      "num_input_tokens_seen": 20276167552,
      "step": 71290,
      "train_runtime": 693230.5216,
      "train_tokens_per_second": 29248.81
    },
    {
      "epoch": 2.5227340576328983,
      "grad_norm": 0.66796875,
      "learning_rate": 1.7535873442829135e-05,
      "loss": 0.3805593490600586,
      "num_input_tokens_seen": 20278973056,
      "step": 71300,
      "train_runtime": 693324.9608,
      "train_tokens_per_second": 29248.872
    },
    {
      "epoch": 2.523087877896627,
      "grad_norm": 0.625,
      "learning_rate": 1.753479506204412e-05,
      "loss": 0.38619811534881593,
      "num_input_tokens_seen": 20281819072,
      "step": 71310,
      "train_runtime": 693420.402,
      "train_tokens_per_second": 29248.951
    },
    {
      "epoch": 2.523441698160356,
      "grad_norm": 0.64453125,
      "learning_rate": 1.7533716880181973e-05,
      "loss": 0.38463554382324217,
      "num_input_tokens_seen": 20284649920,
      "step": 71320,
      "train_runtime": 693515.5695,
      "train_tokens_per_second": 29249.019
    },
    {
      "epoch": 2.523795518424085,
      "grad_norm": 0.6171875,
      "learning_rate": 1.753263889718155e-05,
      "loss": 0.386564826965332,
      "num_input_tokens_seen": 20287518336,
      "step": 71330,
      "train_runtime": 693616.5028,
      "train_tokens_per_second": 29248.898
    },
    {
      "epoch": 2.5241493386878133,
      "grad_norm": 0.6015625,
      "learning_rate": 1.753156111298173e-05,
      "loss": 0.38729922771453856,
      "num_input_tokens_seen": 20290399488,
      "step": 71340,
      "train_runtime": 693714.2334,
      "train_tokens_per_second": 29248.931
    },
    {
      "epoch": 2.524503158951542,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7530483527521414e-05,
      "loss": 0.38954720497131345,
      "num_input_tokens_seen": 20293213312,
      "step": 71350,
      "train_runtime": 693808.9855,
      "train_tokens_per_second": 29248.992
    },
    {
      "epoch": 2.524856979215271,
      "grad_norm": 0.62890625,
      "learning_rate": 1.752940614073953e-05,
      "loss": 0.3870147943496704,
      "num_input_tokens_seen": 20295999424,
      "step": 71360,
      "train_runtime": 693905.1482,
      "train_tokens_per_second": 29248.954
    },
    {
      "epoch": 2.525210799479,
      "grad_norm": 0.62890625,
      "learning_rate": 1.752832895257503e-05,
      "loss": 0.3876980781555176,
      "num_input_tokens_seen": 20298866944,
      "step": 71370,
      "train_runtime": 694003.7238,
      "train_tokens_per_second": 29248.931
    },
    {
      "epoch": 2.5255646197427284,
      "grad_norm": 0.61328125,
      "learning_rate": 1.752725196296691e-05,
      "loss": 0.38730497360229493,
      "num_input_tokens_seen": 20301717888,
      "step": 71380,
      "train_runtime": 694099.454,
      "train_tokens_per_second": 29249.004
    },
    {
      "epoch": 2.5259184400064574,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7526175171854162e-05,
      "loss": 0.38428587913513185,
      "num_input_tokens_seen": 20304618112,
      "step": 71390,
      "train_runtime": 694199.4714,
      "train_tokens_per_second": 29248.968
    },
    {
      "epoch": 2.526272260270186,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7525098579175827e-05,
      "loss": 0.3889008522033691,
      "num_input_tokens_seen": 20307481920,
      "step": 71400,
      "train_runtime": 694299.9816,
      "train_tokens_per_second": 29248.859
    },
    {
      "epoch": 2.526626080533915,
      "grad_norm": 0.59375,
      "learning_rate": 1.752402218487097e-05,
      "loss": 0.3824108600616455,
      "num_input_tokens_seen": 20310337024,
      "step": 71410,
      "train_runtime": 694396.3827,
      "train_tokens_per_second": 29248.91
    },
    {
      "epoch": 2.5269799007976435,
      "grad_norm": 0.625,
      "learning_rate": 1.752294598887867e-05,
      "loss": 0.38058953285217284,
      "num_input_tokens_seen": 20313203264,
      "step": 71420,
      "train_runtime": 694491.9797,
      "train_tokens_per_second": 29249.011
    },
    {
      "epoch": 2.5273337210613724,
      "grad_norm": 0.609375,
      "learning_rate": 1.7521869991138047e-05,
      "loss": 0.38137240409851075,
      "num_input_tokens_seen": 20316021696,
      "step": 71430,
      "train_runtime": 694587.9346,
      "train_tokens_per_second": 29249.028
    },
    {
      "epoch": 2.527687541325101,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7520794191588235e-05,
      "loss": 0.38843822479248047,
      "num_input_tokens_seen": 20318850752,
      "step": 71440,
      "train_runtime": 694681.6505,
      "train_tokens_per_second": 29249.154
    },
    {
      "epoch": 2.52804136158883,
      "grad_norm": 0.59765625,
      "learning_rate": 1.75197185901684e-05,
      "loss": 0.3843752145767212,
      "num_input_tokens_seen": 20321689920,
      "step": 71450,
      "train_runtime": 694777.8408,
      "train_tokens_per_second": 29249.191
    },
    {
      "epoch": 2.5283951818525585,
      "grad_norm": 0.625,
      "learning_rate": 1.7518643186817734e-05,
      "loss": 0.3882627487182617,
      "num_input_tokens_seen": 20324493056,
      "step": 71460,
      "train_runtime": 694872.5298,
      "train_tokens_per_second": 29249.24
    },
    {
      "epoch": 2.5287490021162875,
      "grad_norm": 0.640625,
      "learning_rate": 1.751756798147546e-05,
      "loss": 0.38930478096008303,
      "num_input_tokens_seen": 20327357952,
      "step": 71470,
      "train_runtime": 694970.2748,
      "train_tokens_per_second": 29249.248
    },
    {
      "epoch": 2.5291028223800165,
      "grad_norm": 0.609375,
      "learning_rate": 1.7516492974080814e-05,
      "loss": 0.38165066242218015,
      "num_input_tokens_seen": 20330247104,
      "step": 71480,
      "train_runtime": 695070.0549,
      "train_tokens_per_second": 29249.206
    },
    {
      "epoch": 2.529456642643745,
      "grad_norm": 0.640625,
      "learning_rate": 1.7515418164573068e-05,
      "loss": 0.383614706993103,
      "num_input_tokens_seen": 20333112576,
      "step": 71490,
      "train_runtime": 695168.0907,
      "train_tokens_per_second": 29249.203
    },
    {
      "epoch": 2.5298104629074736,
      "grad_norm": 0.6484375,
      "learning_rate": 1.751434355289152e-05,
      "loss": 0.38726081848144533,
      "num_input_tokens_seen": 20335925760,
      "step": 71500,
      "train_runtime": 695260.9833,
      "train_tokens_per_second": 29249.341
    },
    {
      "epoch": 2.5301642831712026,
      "grad_norm": 0.640625,
      "learning_rate": 1.751326913897549e-05,
      "loss": 0.3896699190139771,
      "num_input_tokens_seen": 20338758848,
      "step": 71510,
      "train_runtime": 695357.861,
      "train_tokens_per_second": 29249.341
    },
    {
      "epoch": 2.5305181034349316,
      "grad_norm": 0.61328125,
      "learning_rate": 1.751219492276433e-05,
      "loss": 0.3855555534362793,
      "num_input_tokens_seen": 20341625664,
      "step": 71520,
      "train_runtime": 695454.958,
      "train_tokens_per_second": 29249.379
    },
    {
      "epoch": 2.53087192369866,
      "grad_norm": 0.65234375,
      "learning_rate": 1.7511120904197406e-05,
      "loss": 0.38286623954772947,
      "num_input_tokens_seen": 20344459520,
      "step": 71530,
      "train_runtime": 695551.6955,
      "train_tokens_per_second": 29249.385
    },
    {
      "epoch": 2.5312257439623886,
      "grad_norm": 0.609375,
      "learning_rate": 1.7510047083214125e-05,
      "loss": 0.39187235832214357,
      "num_input_tokens_seen": 20347315264,
      "step": 71540,
      "train_runtime": 695647.1014,
      "train_tokens_per_second": 29249.479
    },
    {
      "epoch": 2.5315795642261176,
      "grad_norm": 0.58984375,
      "learning_rate": 1.750897345975391e-05,
      "loss": 0.382798433303833,
      "num_input_tokens_seen": 20350156416,
      "step": 71550,
      "train_runtime": 695744.8487,
      "train_tokens_per_second": 29249.453
    },
    {
      "epoch": 2.5319333844898466,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7507900033756213e-05,
      "loss": 0.3854855537414551,
      "num_input_tokens_seen": 20352940096,
      "step": 71560,
      "train_runtime": 695838.2717,
      "train_tokens_per_second": 29249.527
    },
    {
      "epoch": 2.532287204753575,
      "grad_norm": 0.62109375,
      "learning_rate": 1.750682680516052e-05,
      "loss": 0.3845286130905151,
      "num_input_tokens_seen": 20355880000,
      "step": 71570,
      "train_runtime": 695940.4388,
      "train_tokens_per_second": 29249.457
    },
    {
      "epoch": 2.532641025017304,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7505753773906322e-05,
      "loss": 0.38743271827697756,
      "num_input_tokens_seen": 20358729344,
      "step": 71580,
      "train_runtime": 696040.6809,
      "train_tokens_per_second": 29249.338
    },
    {
      "epoch": 2.5329948452810327,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7504680939933153e-05,
      "loss": 0.38537399768829345,
      "num_input_tokens_seen": 20361543488,
      "step": 71590,
      "train_runtime": 696135.6167,
      "train_tokens_per_second": 29249.392
    },
    {
      "epoch": 2.5333486655447617,
      "grad_norm": 0.625,
      "learning_rate": 1.7503608303180574e-05,
      "loss": 0.3836169481277466,
      "num_input_tokens_seen": 20364404352,
      "step": 71600,
      "train_runtime": 696234.9218,
      "train_tokens_per_second": 29249.329
    },
    {
      "epoch": 2.5337024858084902,
      "grad_norm": 0.640625,
      "learning_rate": 1.7502535863588165e-05,
      "loss": 0.38952980041503904,
      "num_input_tokens_seen": 20367230016,
      "step": 71610,
      "train_runtime": 696330.4075,
      "train_tokens_per_second": 29249.376
    },
    {
      "epoch": 2.534056306072219,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7501463621095532e-05,
      "loss": 0.38968582153320314,
      "num_input_tokens_seen": 20370069696,
      "step": 71620,
      "train_runtime": 696430.742,
      "train_tokens_per_second": 29249.24
    },
    {
      "epoch": 2.5344101263359478,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7500391575642305e-05,
      "loss": 0.38724427223205565,
      "num_input_tokens_seen": 20372935360,
      "step": 71630,
      "train_runtime": 696528.7902,
      "train_tokens_per_second": 29249.237
    },
    {
      "epoch": 2.5347639465996767,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7499319727168155e-05,
      "loss": 0.38423311710357666,
      "num_input_tokens_seen": 20375760320,
      "step": 71640,
      "train_runtime": 696626.9106,
      "train_tokens_per_second": 29249.172
    },
    {
      "epoch": 2.5351177668634053,
      "grad_norm": 0.59375,
      "learning_rate": 1.7498248075612753e-05,
      "loss": 0.3837351083755493,
      "num_input_tokens_seen": 20378570176,
      "step": 71650,
      "train_runtime": 696721.4654,
      "train_tokens_per_second": 29249.235
    },
    {
      "epoch": 2.5354715871271343,
      "grad_norm": 0.609375,
      "learning_rate": 1.749717662091582e-05,
      "loss": 0.38747596740722656,
      "num_input_tokens_seen": 20381461056,
      "step": 71660,
      "train_runtime": 696818.4778,
      "train_tokens_per_second": 29249.312
    },
    {
      "epoch": 2.5358254073908633,
      "grad_norm": 0.609375,
      "learning_rate": 1.749610536301709e-05,
      "loss": 0.3857890129089355,
      "num_input_tokens_seen": 20384278976,
      "step": 71670,
      "train_runtime": 696914.4905,
      "train_tokens_per_second": 29249.326
    },
    {
      "epoch": 2.536179227654592,
      "grad_norm": 0.6484375,
      "learning_rate": 1.749503430185633e-05,
      "loss": 0.38343987464904783,
      "num_input_tokens_seen": 20387070080,
      "step": 71680,
      "train_runtime": 697010.9799,
      "train_tokens_per_second": 29249.281
    },
    {
      "epoch": 2.5365330479183203,
      "grad_norm": 0.6015625,
      "learning_rate": 1.749396343737332e-05,
      "loss": 0.38547072410583494,
      "num_input_tokens_seen": 20389926656,
      "step": 71690,
      "train_runtime": 697110.6057,
      "train_tokens_per_second": 29249.199
    },
    {
      "epoch": 2.5368868681820493,
      "grad_norm": 0.640625,
      "learning_rate": 1.7492892769507885e-05,
      "loss": 0.3897873401641846,
      "num_input_tokens_seen": 20392709184,
      "step": 71700,
      "train_runtime": 697204.406,
      "train_tokens_per_second": 29249.255
    },
    {
      "epoch": 2.5372406884457783,
      "grad_norm": 0.59375,
      "learning_rate": 1.7491822298199854e-05,
      "loss": 0.3842735767364502,
      "num_input_tokens_seen": 20395600960,
      "step": 71710,
      "train_runtime": 697305.3707,
      "train_tokens_per_second": 29249.167
    },
    {
      "epoch": 2.537594508709507,
      "grad_norm": 0.6171875,
      "learning_rate": 1.74907520233891e-05,
      "loss": 0.38124186992645265,
      "num_input_tokens_seen": 20398470400,
      "step": 71720,
      "train_runtime": 697404.6002,
      "train_tokens_per_second": 29249.119
    },
    {
      "epoch": 2.537948328973236,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7489681945015517e-05,
      "loss": 0.38555612564086916,
      "num_input_tokens_seen": 20401296832,
      "step": 71730,
      "train_runtime": 697499.6705,
      "train_tokens_per_second": 29249.185
    },
    {
      "epoch": 2.5383021492369644,
      "grad_norm": 0.65234375,
      "learning_rate": 1.7488612063019018e-05,
      "loss": 0.3859811067581177,
      "num_input_tokens_seen": 20404107264,
      "step": 71740,
      "train_runtime": 697593.8178,
      "train_tokens_per_second": 29249.266
    },
    {
      "epoch": 2.5386559695006934,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7487542377339547e-05,
      "loss": 0.3880645751953125,
      "num_input_tokens_seen": 20406953792,
      "step": 71750,
      "train_runtime": 697689.9534,
      "train_tokens_per_second": 29249.316
    },
    {
      "epoch": 2.539009789764422,
      "grad_norm": 0.609375,
      "learning_rate": 1.7486472887917072e-05,
      "loss": 0.3881432771682739,
      "num_input_tokens_seen": 20409816768,
      "step": 71760,
      "train_runtime": 697787.7304,
      "train_tokens_per_second": 29249.32
    },
    {
      "epoch": 2.539363610028151,
      "grad_norm": 0.625,
      "learning_rate": 1.7485403594691592e-05,
      "loss": 0.3873645305633545,
      "num_input_tokens_seen": 20412693888,
      "step": 71770,
      "train_runtime": 697887.4533,
      "train_tokens_per_second": 29249.263
    },
    {
      "epoch": 2.5397174302918795,
      "grad_norm": 0.65234375,
      "learning_rate": 1.7484334497603126e-05,
      "loss": 0.38535351753234864,
      "num_input_tokens_seen": 20415487872,
      "step": 71780,
      "train_runtime": 697982.8577,
      "train_tokens_per_second": 29249.268
    },
    {
      "epoch": 2.5400712505556085,
      "grad_norm": 0.609375,
      "learning_rate": 1.7483265596591712e-05,
      "loss": 0.38485443592071533,
      "num_input_tokens_seen": 20418321472,
      "step": 71790,
      "train_runtime": 698079.1639,
      "train_tokens_per_second": 29249.292
    },
    {
      "epoch": 2.540425070819337,
      "grad_norm": 0.640625,
      "learning_rate": 1.7482196891597433e-05,
      "loss": 0.38553314208984374,
      "num_input_tokens_seen": 20421154816,
      "step": 71800,
      "train_runtime": 698175.0823,
      "train_tokens_per_second": 29249.332
    },
    {
      "epoch": 2.540778891083066,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7481128382560382e-05,
      "loss": 0.3869482517242432,
      "num_input_tokens_seen": 20423964608,
      "step": 71810,
      "train_runtime": 698272.9826,
      "train_tokens_per_second": 29249.255
    },
    {
      "epoch": 2.541132711346795,
      "grad_norm": 0.58984375,
      "learning_rate": 1.748006006942068e-05,
      "loss": 0.3878339290618896,
      "num_input_tokens_seen": 20426861632,
      "step": 71820,
      "train_runtime": 698371.8545,
      "train_tokens_per_second": 29249.262
    },
    {
      "epoch": 2.5414865316105235,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7478991952118476e-05,
      "loss": 0.3841610670089722,
      "num_input_tokens_seen": 20429681792,
      "step": 71830,
      "train_runtime": 698467.528,
      "train_tokens_per_second": 29249.294
    },
    {
      "epoch": 2.541840351874252,
      "grad_norm": 0.6484375,
      "learning_rate": 1.747792403059395e-05,
      "loss": 0.38683784008026123,
      "num_input_tokens_seen": 20432497280,
      "step": 71840,
      "train_runtime": 698561.9039,
      "train_tokens_per_second": 29249.372
    },
    {
      "epoch": 2.542194172137981,
      "grad_norm": 0.6328125,
      "learning_rate": 1.747685630478729e-05,
      "loss": 0.3898813486099243,
      "num_input_tokens_seen": 20435354624,
      "step": 71850,
      "train_runtime": 698658.9605,
      "train_tokens_per_second": 29249.399
    },
    {
      "epoch": 2.54254799240171,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7475788774638732e-05,
      "loss": 0.38735437393188477,
      "num_input_tokens_seen": 20438161600,
      "step": 71860,
      "train_runtime": 698752.2788,
      "train_tokens_per_second": 29249.51
    },
    {
      "epoch": 2.5429018126654386,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7474721440088522e-05,
      "loss": 0.38495023250579835,
      "num_input_tokens_seen": 20441053888,
      "step": 71870,
      "train_runtime": 698851.4249,
      "train_tokens_per_second": 29249.499
    },
    {
      "epoch": 2.543255632929167,
      "grad_norm": 0.63671875,
      "learning_rate": 1.747365430107694e-05,
      "loss": 0.388193941116333,
      "num_input_tokens_seen": 20443936192,
      "step": 71880,
      "train_runtime": 698949.8783,
      "train_tokens_per_second": 29249.502
    },
    {
      "epoch": 2.543609453192896,
      "grad_norm": 0.609375,
      "learning_rate": 1.7472587357544283e-05,
      "loss": 0.38282017707824706,
      "num_input_tokens_seen": 20446803712,
      "step": 71890,
      "train_runtime": 699045.8926,
      "train_tokens_per_second": 29249.587
    },
    {
      "epoch": 2.543963273456625,
      "grad_norm": 0.640625,
      "learning_rate": 1.7471520609430883e-05,
      "loss": 0.388105583190918,
      "num_input_tokens_seen": 20449636224,
      "step": 71900,
      "train_runtime": 699141.3227,
      "train_tokens_per_second": 29249.646
    },
    {
      "epoch": 2.5443170937203536,
      "grad_norm": 0.60546875,
      "learning_rate": 1.747045405667709e-05,
      "loss": 0.38689699172973635,
      "num_input_tokens_seen": 20452428224,
      "step": 71910,
      "train_runtime": 699234.5971,
      "train_tokens_per_second": 29249.737
    },
    {
      "epoch": 2.5446709139840826,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7469387699223282e-05,
      "loss": 0.3821892261505127,
      "num_input_tokens_seen": 20455261440,
      "step": 71920,
      "train_runtime": 699329.6725,
      "train_tokens_per_second": 29249.812
    },
    {
      "epoch": 2.545024734247811,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7468321537009867e-05,
      "loss": 0.3824774265289307,
      "num_input_tokens_seen": 20458134784,
      "step": 71930,
      "train_runtime": 699429.13,
      "train_tokens_per_second": 29249.761
    },
    {
      "epoch": 2.54537855451154,
      "grad_norm": 0.625,
      "learning_rate": 1.746725556997727e-05,
      "loss": 0.38945980072021485,
      "num_input_tokens_seen": 20460955776,
      "step": 71940,
      "train_runtime": 699527.1202,
      "train_tokens_per_second": 29249.696
    },
    {
      "epoch": 2.5457323747752687,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7466189798065948e-05,
      "loss": 0.38823580741882324,
      "num_input_tokens_seen": 20463774080,
      "step": 71950,
      "train_runtime": 699623.8114,
      "train_tokens_per_second": 29249.682
    },
    {
      "epoch": 2.5460861950389977,
      "grad_norm": 0.671875,
      "learning_rate": 1.7465124221216382e-05,
      "loss": 0.3891043186187744,
      "num_input_tokens_seen": 20466610112,
      "step": 71960,
      "train_runtime": 699718.5001,
      "train_tokens_per_second": 29249.777
    },
    {
      "epoch": 2.5464400153027267,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7464058839369074e-05,
      "loss": 0.3851813554763794,
      "num_input_tokens_seen": 20469394560,
      "step": 71970,
      "train_runtime": 699812.7036,
      "train_tokens_per_second": 29249.818
    },
    {
      "epoch": 2.546793835566455,
      "grad_norm": 0.609375,
      "learning_rate": 1.7462993652464556e-05,
      "loss": 0.3852069616317749,
      "num_input_tokens_seen": 20472254144,
      "step": 71980,
      "train_runtime": 699910.6959,
      "train_tokens_per_second": 29249.809
    },
    {
      "epoch": 2.5471476558301838,
      "grad_norm": 0.59765625,
      "learning_rate": 1.746192866044339e-05,
      "loss": 0.3870324373245239,
      "num_input_tokens_seen": 20475150016,
      "step": 71990,
      "train_runtime": 700011.282,
      "train_tokens_per_second": 29249.743
    },
    {
      "epoch": 2.5475014760939128,
      "grad_norm": 0.578125,
      "learning_rate": 1.746086386324615e-05,
      "loss": 0.3841458797454834,
      "num_input_tokens_seen": 20477997248,
      "step": 72000,
      "train_runtime": 700108.2143,
      "train_tokens_per_second": 29249.76
    },
    {
      "epoch": 2.5478552963576417,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7459799260813448e-05,
      "loss": 0.38673877716064453,
      "num_input_tokens_seen": 20480808576,
      "step": 72010,
      "train_runtime": 700203.8994,
      "train_tokens_per_second": 29249.778
    },
    {
      "epoch": 2.5482091166213703,
      "grad_norm": 0.65234375,
      "learning_rate": 1.7458734853085918e-05,
      "loss": 0.3884261608123779,
      "num_input_tokens_seen": 20483689472,
      "step": 72020,
      "train_runtime": 700303.4733,
      "train_tokens_per_second": 29249.733
    },
    {
      "epoch": 2.548562936885099,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7457670640004212e-05,
      "loss": 0.38671445846557617,
      "num_input_tokens_seen": 20486498944,
      "step": 72030,
      "train_runtime": 700397.1172,
      "train_tokens_per_second": 29249.833
    },
    {
      "epoch": 2.548916757148828,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7456606621509013e-05,
      "loss": 0.38740196228027346,
      "num_input_tokens_seen": 20489338176,
      "step": 72040,
      "train_runtime": 700495.338,
      "train_tokens_per_second": 29249.785
    },
    {
      "epoch": 2.549270577412557,
      "grad_norm": 0.64453125,
      "learning_rate": 1.745554279754104e-05,
      "loss": 0.3833166599273682,
      "num_input_tokens_seen": 20492240576,
      "step": 72050,
      "train_runtime": 700596.783,
      "train_tokens_per_second": 29249.693
    },
    {
      "epoch": 2.5496243976762853,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7454479168041013e-05,
      "loss": 0.38718838691711427,
      "num_input_tokens_seen": 20495117568,
      "step": 72060,
      "train_runtime": 700696.7751,
      "train_tokens_per_second": 29249.625
    },
    {
      "epoch": 2.5499782179400143,
      "grad_norm": 0.6015625,
      "learning_rate": 1.74534157329497e-05,
      "loss": 0.38707432746887205,
      "num_input_tokens_seen": 20497946432,
      "step": 72070,
      "train_runtime": 700791.8608,
      "train_tokens_per_second": 29249.692
    },
    {
      "epoch": 2.550332038203743,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7452352492207886e-05,
      "loss": 0.3846604347229004,
      "num_input_tokens_seen": 20500781440,
      "step": 72080,
      "train_runtime": 700886.5944,
      "train_tokens_per_second": 29249.784
    },
    {
      "epoch": 2.550685858467472,
      "grad_norm": 0.625,
      "learning_rate": 1.7451289445756374e-05,
      "loss": 0.3814728260040283,
      "num_input_tokens_seen": 20503619904,
      "step": 72090,
      "train_runtime": 700984.4335,
      "train_tokens_per_second": 29249.751
    },
    {
      "epoch": 2.5510396787312004,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7450226593536005e-05,
      "loss": 0.3872837543487549,
      "num_input_tokens_seen": 20506504192,
      "step": 72100,
      "train_runtime": 701082.679,
      "train_tokens_per_second": 29249.766
    },
    {
      "epoch": 2.5513934989949294,
      "grad_norm": 0.61328125,
      "learning_rate": 1.744916393548763e-05,
      "loss": 0.384901762008667,
      "num_input_tokens_seen": 20509361536,
      "step": 72110,
      "train_runtime": 701182.4991,
      "train_tokens_per_second": 29249.677
    },
    {
      "epoch": 2.551747319258658,
      "grad_norm": 0.6328125,
      "learning_rate": 1.744810147155215e-05,
      "loss": 0.3845955848693848,
      "num_input_tokens_seen": 20512159936,
      "step": 72120,
      "train_runtime": 701280.902,
      "train_tokens_per_second": 29249.563
    },
    {
      "epoch": 2.552101139522387,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7447039201670458e-05,
      "loss": 0.38531951904296874,
      "num_input_tokens_seen": 20515013056,
      "step": 72130,
      "train_runtime": 701378.3319,
      "train_tokens_per_second": 29249.568
    },
    {
      "epoch": 2.5524549597861155,
      "grad_norm": 0.625,
      "learning_rate": 1.7445977125783497e-05,
      "loss": 0.387006950378418,
      "num_input_tokens_seen": 20517825984,
      "step": 72140,
      "train_runtime": 701476.1195,
      "train_tokens_per_second": 29249.5
    },
    {
      "epoch": 2.5528087800498445,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7444915243832234e-05,
      "loss": 0.3883934497833252,
      "num_input_tokens_seen": 20520697984,
      "step": 72150,
      "train_runtime": 701576.9798,
      "train_tokens_per_second": 29249.389
    },
    {
      "epoch": 2.5531626003135734,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7443853555757645e-05,
      "loss": 0.3865645408630371,
      "num_input_tokens_seen": 20523585344,
      "step": 72160,
      "train_runtime": 701675.1793,
      "train_tokens_per_second": 29249.41
    },
    {
      "epoch": 2.553516420577302,
      "grad_norm": 0.640625,
      "learning_rate": 1.744279206150074e-05,
      "loss": 0.3867837905883789,
      "num_input_tokens_seen": 20526424384,
      "step": 72170,
      "train_runtime": 701772.4606,
      "train_tokens_per_second": 29249.401
    },
    {
      "epoch": 2.5538702408410305,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7441730761002567e-05,
      "loss": 0.39119277000427244,
      "num_input_tokens_seen": 20529302208,
      "step": 72180,
      "train_runtime": 701872.238,
      "train_tokens_per_second": 29249.344
    },
    {
      "epoch": 2.5542240611047595,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7440669654204185e-05,
      "loss": 0.3830560207366943,
      "num_input_tokens_seen": 20532160512,
      "step": 72190,
      "train_runtime": 701970.7846,
      "train_tokens_per_second": 29249.309
    },
    {
      "epoch": 2.5545778813684885,
      "grad_norm": 0.640625,
      "learning_rate": 1.7439608741046666e-05,
      "loss": 0.3865551471710205,
      "num_input_tokens_seen": 20535015680,
      "step": 72200,
      "train_runtime": 702069.4557,
      "train_tokens_per_second": 29249.265
    },
    {
      "epoch": 2.554931701632217,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7438548021471134e-05,
      "loss": 0.3892939805984497,
      "num_input_tokens_seen": 20537910336,
      "step": 72210,
      "train_runtime": 702171.1597,
      "train_tokens_per_second": 29249.151
    },
    {
      "epoch": 2.555285521895946,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7437487495418724e-05,
      "loss": 0.38391158580780027,
      "num_input_tokens_seen": 20540782016,
      "step": 72220,
      "train_runtime": 702268.7552,
      "train_tokens_per_second": 29249.175
    },
    {
      "epoch": 2.5556393421596746,
      "grad_norm": 0.6640625,
      "learning_rate": 1.7436427162830597e-05,
      "loss": 0.38832964897155764,
      "num_input_tokens_seen": 20543592128,
      "step": 72230,
      "train_runtime": 702365.4667,
      "train_tokens_per_second": 29249.149
    },
    {
      "epoch": 2.5559931624234036,
      "grad_norm": 0.63671875,
      "learning_rate": 1.743536702364794e-05,
      "loss": 0.3866887092590332,
      "num_input_tokens_seen": 20546470208,
      "step": 72240,
      "train_runtime": 702464.5948,
      "train_tokens_per_second": 29249.119
    },
    {
      "epoch": 2.556346982687132,
      "grad_norm": 0.59765625,
      "learning_rate": 1.743430707781196e-05,
      "loss": 0.39111597537994386,
      "num_input_tokens_seen": 20549301952,
      "step": 72250,
      "train_runtime": 702563.8962,
      "train_tokens_per_second": 29249.015
    },
    {
      "epoch": 2.556700802950861,
      "grad_norm": 0.6328125,
      "learning_rate": 1.74332473252639e-05,
      "loss": 0.38591115474700927,
      "num_input_tokens_seen": 20552154176,
      "step": 72260,
      "train_runtime": 702661.3964,
      "train_tokens_per_second": 29249.016
    },
    {
      "epoch": 2.5570546232145897,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7432187765945016e-05,
      "loss": 0.38828890323638915,
      "num_input_tokens_seen": 20554992704,
      "step": 72270,
      "train_runtime": 702760.2105,
      "train_tokens_per_second": 29248.942
    },
    {
      "epoch": 2.5574084434783186,
      "grad_norm": 0.62109375,
      "learning_rate": 1.74311283997966e-05,
      "loss": 0.3892641305923462,
      "num_input_tokens_seen": 20557903744,
      "step": 72280,
      "train_runtime": 702860.0418,
      "train_tokens_per_second": 29248.929
    },
    {
      "epoch": 2.557762263742047,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7430069226759962e-05,
      "loss": 0.38228960037231446,
      "num_input_tokens_seen": 20560747520,
      "step": 72290,
      "train_runtime": 702955.0139,
      "train_tokens_per_second": 29249.023
    },
    {
      "epoch": 2.558116084005776,
      "grad_norm": 0.58984375,
      "learning_rate": 1.7429010246776435e-05,
      "loss": 0.38678109645843506,
      "num_input_tokens_seen": 20563542208,
      "step": 72300,
      "train_runtime": 703047.809,
      "train_tokens_per_second": 29249.138
    },
    {
      "epoch": 2.558469904269505,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7427951459787386e-05,
      "loss": 0.3943061828613281,
      "num_input_tokens_seen": 20566405568,
      "step": 72310,
      "train_runtime": 703146.2275,
      "train_tokens_per_second": 29249.116
    },
    {
      "epoch": 2.5588237245332337,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7426892865734197e-05,
      "loss": 0.38891263008117677,
      "num_input_tokens_seen": 20569259520,
      "step": 72320,
      "train_runtime": 703245.4019,
      "train_tokens_per_second": 29249.049
    },
    {
      "epoch": 2.5591775447969622,
      "grad_norm": 0.609375,
      "learning_rate": 1.742583446455828e-05,
      "loss": 0.3832040309906006,
      "num_input_tokens_seen": 20572132544,
      "step": 72330,
      "train_runtime": 703340.1749,
      "train_tokens_per_second": 29249.193
    },
    {
      "epoch": 2.5595313650606912,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7424776256201074e-05,
      "loss": 0.39090604782104493,
      "num_input_tokens_seen": 20574947520,
      "step": 72340,
      "train_runtime": 703435.2092,
      "train_tokens_per_second": 29249.243
    },
    {
      "epoch": 2.55988518532442,
      "grad_norm": 0.640625,
      "learning_rate": 1.7423718240604035e-05,
      "loss": 0.38827416896820066,
      "num_input_tokens_seen": 20577857664,
      "step": 72350,
      "train_runtime": 703537.5926,
      "train_tokens_per_second": 29249.123
    },
    {
      "epoch": 2.5602390055881488,
      "grad_norm": 0.625,
      "learning_rate": 1.7422660417708656e-05,
      "loss": 0.38462128639221194,
      "num_input_tokens_seen": 20580678912,
      "step": 72360,
      "train_runtime": 703634.5961,
      "train_tokens_per_second": 29249.1
    },
    {
      "epoch": 2.5605928258518773,
      "grad_norm": 0.609375,
      "learning_rate": 1.7421602787456446e-05,
      "loss": 0.3870999336242676,
      "num_input_tokens_seen": 20583452032,
      "step": 72370,
      "train_runtime": 703726.5436,
      "train_tokens_per_second": 29249.219
    },
    {
      "epoch": 2.5609466461156063,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7420545349788937e-05,
      "loss": 0.3895162582397461,
      "num_input_tokens_seen": 20586235200,
      "step": 72380,
      "train_runtime": 703821.1367,
      "train_tokens_per_second": 29249.243
    },
    {
      "epoch": 2.5613004663793353,
      "grad_norm": 0.609375,
      "learning_rate": 1.7419488104647694e-05,
      "loss": 0.38846116065979003,
      "num_input_tokens_seen": 20589092928,
      "step": 72390,
      "train_runtime": 703921.3168,
      "train_tokens_per_second": 29249.14
    },
    {
      "epoch": 2.561654286643064,
      "grad_norm": 0.6484375,
      "learning_rate": 1.7418431051974298e-05,
      "loss": 0.38369078636169435,
      "num_input_tokens_seen": 20591972160,
      "step": 72400,
      "train_runtime": 704018.3517,
      "train_tokens_per_second": 29249.198
    },
    {
      "epoch": 2.562008106906793,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7417374191710362e-05,
      "loss": 0.3863586187362671,
      "num_input_tokens_seen": 20594829568,
      "step": 72410,
      "train_runtime": 704115.8396,
      "train_tokens_per_second": 29249.206
    },
    {
      "epoch": 2.5623619271705214,
      "grad_norm": 0.59375,
      "learning_rate": 1.7416317523797522e-05,
      "loss": 0.38245353698730467,
      "num_input_tokens_seen": 20597679360,
      "step": 72420,
      "train_runtime": 704210.167,
      "train_tokens_per_second": 29249.335
    },
    {
      "epoch": 2.5627157474342503,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7415261048177434e-05,
      "loss": 0.38879098892211916,
      "num_input_tokens_seen": 20600511872,
      "step": 72430,
      "train_runtime": 704306.3703,
      "train_tokens_per_second": 29249.362
    },
    {
      "epoch": 2.563069567697979,
      "grad_norm": 0.65625,
      "learning_rate": 1.7414204764791785e-05,
      "loss": 0.3847618103027344,
      "num_input_tokens_seen": 20603403456,
      "step": 72440,
      "train_runtime": 704405.4807,
      "train_tokens_per_second": 29249.351
    },
    {
      "epoch": 2.563423387961708,
      "grad_norm": 0.59375,
      "learning_rate": 1.741314867358229e-05,
      "loss": 0.38361034393310545,
      "num_input_tokens_seen": 20606216256,
      "step": 72450,
      "train_runtime": 704499.9682,
      "train_tokens_per_second": 29249.421
    },
    {
      "epoch": 2.5637772082254364,
      "grad_norm": 0.58984375,
      "learning_rate": 1.7412092774490668e-05,
      "loss": 0.3859847068786621,
      "num_input_tokens_seen": 20609104064,
      "step": 72460,
      "train_runtime": 704601.5791,
      "train_tokens_per_second": 29249.302
    },
    {
      "epoch": 2.5641310284891654,
      "grad_norm": 0.60546875,
      "learning_rate": 1.741103706745869e-05,
      "loss": 0.386204195022583,
      "num_input_tokens_seen": 20611943552,
      "step": 72470,
      "train_runtime": 704702.6505,
      "train_tokens_per_second": 29249.136
    },
    {
      "epoch": 2.564484848752894,
      "grad_norm": 0.6484375,
      "learning_rate": 1.740998155242814e-05,
      "loss": 0.38925986289978026,
      "num_input_tokens_seen": 20614718400,
      "step": 72480,
      "train_runtime": 704796.515,
      "train_tokens_per_second": 29249.178
    },
    {
      "epoch": 2.564838669016623,
      "grad_norm": 0.6015625,
      "learning_rate": 1.740892622934082e-05,
      "loss": 0.3884198904037476,
      "num_input_tokens_seen": 20617540800,
      "step": 72490,
      "train_runtime": 704891.1115,
      "train_tokens_per_second": 29249.256
    },
    {
      "epoch": 2.565192489280352,
      "grad_norm": 0.59765625,
      "learning_rate": 1.740787109813857e-05,
      "loss": 0.387147855758667,
      "num_input_tokens_seen": 20620348352,
      "step": 72500,
      "train_runtime": 704985.8545,
      "train_tokens_per_second": 29249.308
    },
    {
      "epoch": 2.5655463095440805,
      "grad_norm": 0.6484375,
      "learning_rate": 1.7406816158763242e-05,
      "loss": 0.3906718254089355,
      "num_input_tokens_seen": 20623187264,
      "step": 72510,
      "train_runtime": 705082.8747,
      "train_tokens_per_second": 29249.31
    },
    {
      "epoch": 2.565900129807809,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7405761411156717e-05,
      "loss": 0.3823741674423218,
      "num_input_tokens_seen": 20626035456,
      "step": 72520,
      "train_runtime": 705179.3947,
      "train_tokens_per_second": 29249.345
    },
    {
      "epoch": 2.566253950071538,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7404706855260906e-05,
      "loss": 0.3907054662704468,
      "num_input_tokens_seen": 20628858880,
      "step": 72530,
      "train_runtime": 705276.0853,
      "train_tokens_per_second": 29249.338
    },
    {
      "epoch": 2.566607770335267,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7403652491017738e-05,
      "loss": 0.38681812286376954,
      "num_input_tokens_seen": 20631688512,
      "step": 72540,
      "train_runtime": 705372.1867,
      "train_tokens_per_second": 29249.365
    },
    {
      "epoch": 2.5669615905989955,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7402598318369174e-05,
      "loss": 0.3863770246505737,
      "num_input_tokens_seen": 20634507200,
      "step": 72550,
      "train_runtime": 705465.7844,
      "train_tokens_per_second": 29249.48
    },
    {
      "epoch": 2.5673154108627245,
      "grad_norm": 0.58984375,
      "learning_rate": 1.7401544337257186e-05,
      "loss": 0.3860642910003662,
      "num_input_tokens_seen": 20637348352,
      "step": 72560,
      "train_runtime": 705563.5365,
      "train_tokens_per_second": 29249.454
    },
    {
      "epoch": 2.567669231126453,
      "grad_norm": 0.609375,
      "learning_rate": 1.740049054762379e-05,
      "loss": 0.3843374729156494,
      "num_input_tokens_seen": 20640215424,
      "step": 72570,
      "train_runtime": 705661.4099,
      "train_tokens_per_second": 29249.46
    },
    {
      "epoch": 2.568023051390182,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7399436949411007e-05,
      "loss": 0.38776202201843263,
      "num_input_tokens_seen": 20643027648,
      "step": 72580,
      "train_runtime": 705754.2139,
      "train_tokens_per_second": 29249.599
    },
    {
      "epoch": 2.5683768716539106,
      "grad_norm": 0.64453125,
      "learning_rate": 1.7398383542560892e-05,
      "loss": 0.3860463619232178,
      "num_input_tokens_seen": 20645891072,
      "step": 72590,
      "train_runtime": 705852.5086,
      "train_tokens_per_second": 29249.582
    },
    {
      "epoch": 2.5687306919176396,
      "grad_norm": 0.63671875,
      "learning_rate": 1.739733032701553e-05,
      "loss": 0.3863032579421997,
      "num_input_tokens_seen": 20648709120,
      "step": 72600,
      "train_runtime": 705947.9263,
      "train_tokens_per_second": 29249.621
    },
    {
      "epoch": 2.569084512181368,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7396277302717024e-05,
      "loss": 0.3927278995513916,
      "num_input_tokens_seen": 20651559168,
      "step": 72610,
      "train_runtime": 706045.0998,
      "train_tokens_per_second": 29249.632
    },
    {
      "epoch": 2.569438332445097,
      "grad_norm": 0.58203125,
      "learning_rate": 1.7395224469607495e-05,
      "loss": 0.38797450065612793,
      "num_input_tokens_seen": 20654400000,
      "step": 72620,
      "train_runtime": 706143.5269,
      "train_tokens_per_second": 29249.578
    },
    {
      "epoch": 2.5697921527088257,
      "grad_norm": 0.62109375,
      "learning_rate": 1.73941718276291e-05,
      "loss": 0.38426249027252196,
      "num_input_tokens_seen": 20657272192,
      "step": 72630,
      "train_runtime": 706243.9945,
      "train_tokens_per_second": 29249.484
    },
    {
      "epoch": 2.5701459729725546,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7393119376724022e-05,
      "loss": 0.38627171516418457,
      "num_input_tokens_seen": 20660119040,
      "step": 72640,
      "train_runtime": 706339.557,
      "train_tokens_per_second": 29249.557
    },
    {
      "epoch": 2.5704997932362836,
      "grad_norm": 0.6640625,
      "learning_rate": 1.739206711683445e-05,
      "loss": 0.38833160400390626,
      "num_input_tokens_seen": 20662932928,
      "step": 72650,
      "train_runtime": 706437.1436,
      "train_tokens_per_second": 29249.5
    },
    {
      "epoch": 2.570853613500012,
      "grad_norm": 0.58984375,
      "learning_rate": 1.739101504790262e-05,
      "loss": 0.3859187126159668,
      "num_input_tokens_seen": 20665755456,
      "step": 72660,
      "train_runtime": 706532.4764,
      "train_tokens_per_second": 29249.548
    },
    {
      "epoch": 2.5712074337637407,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7389963169870784e-05,
      "loss": 0.3849137783050537,
      "num_input_tokens_seen": 20668651456,
      "step": 72670,
      "train_runtime": 706632.5511,
      "train_tokens_per_second": 29249.504
    },
    {
      "epoch": 2.5715612540274697,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7388911482681212e-05,
      "loss": 0.3879070281982422,
      "num_input_tokens_seen": 20671458752,
      "step": 72680,
      "train_runtime": 706727.6465,
      "train_tokens_per_second": 29249.54
    },
    {
      "epoch": 2.5719150742911987,
      "grad_norm": 0.609375,
      "learning_rate": 1.7387859986276197e-05,
      "loss": 0.3888691425323486,
      "num_input_tokens_seen": 20674322048,
      "step": 72690,
      "train_runtime": 706827.4297,
      "train_tokens_per_second": 29249.462
    },
    {
      "epoch": 2.5722688945549272,
      "grad_norm": 0.609375,
      "learning_rate": 1.7386808680598078e-05,
      "loss": 0.3865368366241455,
      "num_input_tokens_seen": 20677222848,
      "step": 72700,
      "train_runtime": 706927.8599,
      "train_tokens_per_second": 29249.41
    },
    {
      "epoch": 2.572622714818656,
      "grad_norm": 0.625,
      "learning_rate": 1.7385757565589192e-05,
      "loss": 0.3862831830978394,
      "num_input_tokens_seen": 20680008704,
      "step": 72710,
      "train_runtime": 707023.4399,
      "train_tokens_per_second": 29249.396
    },
    {
      "epoch": 2.5729765350823848,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7384706641191913e-05,
      "loss": 0.3844675779342651,
      "num_input_tokens_seen": 20682933632,
      "step": 72720,
      "train_runtime": 707126.9081,
      "train_tokens_per_second": 29249.253
    },
    {
      "epoch": 2.5733303553461138,
      "grad_norm": 0.64453125,
      "learning_rate": 1.7383655907348643e-05,
      "loss": 0.38379473686218263,
      "num_input_tokens_seen": 20685755072,
      "step": 72730,
      "train_runtime": 707223.4991,
      "train_tokens_per_second": 29249.247
    },
    {
      "epoch": 2.5736841756098423,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7382605364001803e-05,
      "loss": 0.38410191535949706,
      "num_input_tokens_seen": 20688576000,
      "step": 72740,
      "train_runtime": 707321.4104,
      "train_tokens_per_second": 29249.187
    },
    {
      "epoch": 2.5740379958735713,
      "grad_norm": 0.6171875,
      "learning_rate": 1.738155501109383e-05,
      "loss": 0.38686318397521974,
      "num_input_tokens_seen": 20691433728,
      "step": 72750,
      "train_runtime": 707419.358,
      "train_tokens_per_second": 29249.177
    },
    {
      "epoch": 2.5743918161373,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7380504848567207e-05,
      "loss": 0.38775198459625243,
      "num_input_tokens_seen": 20694294400,
      "step": 72760,
      "train_runtime": 707515.5769,
      "train_tokens_per_second": 29249.242
    },
    {
      "epoch": 2.574745636401029,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7379454876364413e-05,
      "loss": 0.3854989528656006,
      "num_input_tokens_seen": 20697143680,
      "step": 72770,
      "train_runtime": 707612.2355,
      "train_tokens_per_second": 29249.273
    },
    {
      "epoch": 2.5750994566647574,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7378405094427978e-05,
      "loss": 0.3907506465911865,
      "num_input_tokens_seen": 20700026816,
      "step": 72780,
      "train_runtime": 707712.194,
      "train_tokens_per_second": 29249.216
    },
    {
      "epoch": 2.5754532769284864,
      "grad_norm": 0.59765625,
      "learning_rate": 1.737735550270045e-05,
      "loss": 0.3860790252685547,
      "num_input_tokens_seen": 20702885952,
      "step": 72790,
      "train_runtime": 707810.2234,
      "train_tokens_per_second": 29249.204
    },
    {
      "epoch": 2.5758070971922153,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7376306101124385e-05,
      "loss": 0.3875294208526611,
      "num_input_tokens_seen": 20705759360,
      "step": 72800,
      "train_runtime": 707907.0578,
      "train_tokens_per_second": 29249.262
    },
    {
      "epoch": 2.576160917455944,
      "grad_norm": 0.6484375,
      "learning_rate": 1.7375256889642376e-05,
      "loss": 0.38815035820007326,
      "num_input_tokens_seen": 20708582592,
      "step": 72810,
      "train_runtime": 708003.3812,
      "train_tokens_per_second": 29249.271
    },
    {
      "epoch": 2.5765147377196724,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7374207868197044e-05,
      "loss": 0.3845224380493164,
      "num_input_tokens_seen": 20711415936,
      "step": 72820,
      "train_runtime": 708099.9531,
      "train_tokens_per_second": 29249.283
    },
    {
      "epoch": 2.5768685579834014,
      "grad_norm": 0.609375,
      "learning_rate": 1.7373159036731028e-05,
      "loss": 0.3894948959350586,
      "num_input_tokens_seen": 20714234880,
      "step": 72830,
      "train_runtime": 708195.7337,
      "train_tokens_per_second": 29249.308
    },
    {
      "epoch": 2.5772223782471304,
      "grad_norm": 0.609375,
      "learning_rate": 1.737211039518699e-05,
      "loss": 0.3889970064163208,
      "num_input_tokens_seen": 20717062464,
      "step": 72840,
      "train_runtime": 708294.0493,
      "train_tokens_per_second": 29249.24
    },
    {
      "epoch": 2.577576198510859,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7371061943507615e-05,
      "loss": 0.38866562843322755,
      "num_input_tokens_seen": 20719861312,
      "step": 72850,
      "train_runtime": 708387.205,
      "train_tokens_per_second": 29249.344
    },
    {
      "epoch": 2.5779300187745875,
      "grad_norm": 0.65625,
      "learning_rate": 1.7370013681635626e-05,
      "loss": 0.38962316513061523,
      "num_input_tokens_seen": 20722699136,
      "step": 72860,
      "train_runtime": 708484.5341,
      "train_tokens_per_second": 29249.332
    },
    {
      "epoch": 2.5782838390383165,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7368965609513755e-05,
      "loss": 0.3829775094985962,
      "num_input_tokens_seen": 20725508736,
      "step": 72870,
      "train_runtime": 708583.5109,
      "train_tokens_per_second": 29249.211
    },
    {
      "epoch": 2.5786376593020455,
      "grad_norm": 0.62109375,
      "learning_rate": 1.736791772708476e-05,
      "loss": 0.3895395755767822,
      "num_input_tokens_seen": 20728335552,
      "step": 72880,
      "train_runtime": 708679.5834,
      "train_tokens_per_second": 29249.235
    },
    {
      "epoch": 2.578991479565774,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7366870034291427e-05,
      "loss": 0.3847679615020752,
      "num_input_tokens_seen": 20731187840,
      "step": 72890,
      "train_runtime": 708777.086,
      "train_tokens_per_second": 29249.235
    },
    {
      "epoch": 2.579345299829503,
      "grad_norm": 0.625,
      "learning_rate": 1.736582253107657e-05,
      "loss": 0.3882322072982788,
      "num_input_tokens_seen": 20733978560,
      "step": 72900,
      "train_runtime": 708870.4904,
      "train_tokens_per_second": 29249.318
    },
    {
      "epoch": 2.5796991200932315,
      "grad_norm": 0.609375,
      "learning_rate": 1.736477521738302e-05,
      "loss": 0.38532466888427735,
      "num_input_tokens_seen": 20736831488,
      "step": 72910,
      "train_runtime": 708968.0909,
      "train_tokens_per_second": 29249.316
    },
    {
      "epoch": 2.5800529403569605,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7363728093153632e-05,
      "loss": 0.3869034767150879,
      "num_input_tokens_seen": 20739722624,
      "step": 72920,
      "train_runtime": 709069.509,
      "train_tokens_per_second": 29249.21
    },
    {
      "epoch": 2.580406760620689,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7362681158331295e-05,
      "loss": 0.3879739761352539,
      "num_input_tokens_seen": 20742627072,
      "step": 72930,
      "train_runtime": 709168.9954,
      "train_tokens_per_second": 29249.202
    },
    {
      "epoch": 2.580760580884418,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7361634412858908e-05,
      "loss": 0.38549802303314207,
      "num_input_tokens_seen": 20745470464,
      "step": 72940,
      "train_runtime": 709268.268,
      "train_tokens_per_second": 29249.117
    },
    {
      "epoch": 2.5811144011481466,
      "grad_norm": 0.59375,
      "learning_rate": 1.7360587856679405e-05,
      "loss": 0.3879418849945068,
      "num_input_tokens_seen": 20748280128,
      "step": 72950,
      "train_runtime": 709363.5986,
      "train_tokens_per_second": 29249.147
    },
    {
      "epoch": 2.5814682214118756,
      "grad_norm": 0.6171875,
      "learning_rate": 1.735954148973574e-05,
      "loss": 0.3875317335128784,
      "num_input_tokens_seen": 20751092480,
      "step": 72960,
      "train_runtime": 709459.1305,
      "train_tokens_per_second": 29249.172
    },
    {
      "epoch": 2.581822041675604,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7358495311970887e-05,
      "loss": 0.3866284370422363,
      "num_input_tokens_seen": 20753923712,
      "step": 72970,
      "train_runtime": 709552.9986,
      "train_tokens_per_second": 29249.293
    },
    {
      "epoch": 2.582175861939333,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7357449323327853e-05,
      "loss": 0.3865057468414307,
      "num_input_tokens_seen": 20756733696,
      "step": 72980,
      "train_runtime": 709649.6532,
      "train_tokens_per_second": 29249.269
    },
    {
      "epoch": 2.582529682203062,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7356403523749667e-05,
      "loss": 0.3865872859954834,
      "num_input_tokens_seen": 20759522816,
      "step": 72990,
      "train_runtime": 709743.9765,
      "train_tokens_per_second": 29249.312
    },
    {
      "epoch": 2.5828835024667907,
      "grad_norm": 0.59375,
      "learning_rate": 1.7355357913179375e-05,
      "loss": 0.38580405712127686,
      "num_input_tokens_seen": 20762470848,
      "step": 73000,
      "train_runtime": 709845.707,
      "train_tokens_per_second": 29249.273
    },
    {
      "epoch": 2.583237322730519,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7354312491560043e-05,
      "loss": 0.38407533168792723,
      "num_input_tokens_seen": 20765296640,
      "step": 73010,
      "train_runtime": 709940.1239,
      "train_tokens_per_second": 29249.363
    },
    {
      "epoch": 2.583591142994248,
      "grad_norm": 0.625,
      "learning_rate": 1.7353267258834785e-05,
      "loss": 0.38304691314697265,
      "num_input_tokens_seen": 20768109120,
      "step": 73020,
      "train_runtime": 710035.7575,
      "train_tokens_per_second": 29249.385
    },
    {
      "epoch": 2.583944963257977,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7352222214946716e-05,
      "loss": 0.3822983741760254,
      "num_input_tokens_seen": 20770975936,
      "step": 73030,
      "train_runtime": 710135.7982,
      "train_tokens_per_second": 29249.301
    },
    {
      "epoch": 2.5842987835217057,
      "grad_norm": 0.609375,
      "learning_rate": 1.735117735983898e-05,
      "loss": 0.38666086196899413,
      "num_input_tokens_seen": 20773825472,
      "step": 73040,
      "train_runtime": 710237.5475,
      "train_tokens_per_second": 29249.123
    },
    {
      "epoch": 2.5846526037854347,
      "grad_norm": 0.625,
      "learning_rate": 1.735013269345475e-05,
      "loss": 0.38639349937438966,
      "num_input_tokens_seen": 20776646144,
      "step": 73050,
      "train_runtime": 710334.1833,
      "train_tokens_per_second": 29249.115
    },
    {
      "epoch": 2.5850064240491633,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7349088215737225e-05,
      "loss": 0.3866072416305542,
      "num_input_tokens_seen": 20779489664,
      "step": 73060,
      "train_runtime": 710431.9329,
      "train_tokens_per_second": 29249.093
    },
    {
      "epoch": 2.5853602443128922,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7348043926629614e-05,
      "loss": 0.38664243221282957,
      "num_input_tokens_seen": 20782366720,
      "step": 73070,
      "train_runtime": 710529.5534,
      "train_tokens_per_second": 29249.124
    },
    {
      "epoch": 2.585714064576621,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7346999826075162e-05,
      "loss": 0.38595821857452395,
      "num_input_tokens_seen": 20785225088,
      "step": 73080,
      "train_runtime": 710628.3667,
      "train_tokens_per_second": 29249.079
    },
    {
      "epoch": 2.5860678848403498,
      "grad_norm": 0.59375,
      "learning_rate": 1.7345955914017143e-05,
      "loss": 0.38643383979797363,
      "num_input_tokens_seen": 20788115840,
      "step": 73090,
      "train_runtime": 710727.0954,
      "train_tokens_per_second": 29249.083
    },
    {
      "epoch": 2.5864217051040783,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7344912190398834e-05,
      "loss": 0.38464994430541993,
      "num_input_tokens_seen": 20790963584,
      "step": 73100,
      "train_runtime": 710824.5474,
      "train_tokens_per_second": 29249.079
    },
    {
      "epoch": 2.5867755253678073,
      "grad_norm": 0.60546875,
      "learning_rate": 1.734386865516356e-05,
      "loss": 0.38803610801696775,
      "num_input_tokens_seen": 20793819776,
      "step": 73110,
      "train_runtime": 710920.4085,
      "train_tokens_per_second": 29249.153
    },
    {
      "epoch": 2.587129345631536,
      "grad_norm": 0.62109375,
      "learning_rate": 1.734282530825465e-05,
      "loss": 0.3899891138076782,
      "num_input_tokens_seen": 20796683840,
      "step": 73120,
      "train_runtime": 711021.1756,
      "train_tokens_per_second": 29249.036
    },
    {
      "epoch": 2.587483165895265,
      "grad_norm": 0.6015625,
      "learning_rate": 1.734178214961547e-05,
      "loss": 0.3866103172302246,
      "num_input_tokens_seen": 20799500928,
      "step": 73130,
      "train_runtime": 711115.58,
      "train_tokens_per_second": 29249.114
    },
    {
      "epoch": 2.587836986158994,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7340739179189408e-05,
      "loss": 0.38913352489471437,
      "num_input_tokens_seen": 20802316864,
      "step": 73140,
      "train_runtime": 711210.8509,
      "train_tokens_per_second": 29249.156
    },
    {
      "epoch": 2.5881908064227224,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7339696396919868e-05,
      "loss": 0.388286828994751,
      "num_input_tokens_seen": 20805153536,
      "step": 73150,
      "train_runtime": 711309.6948,
      "train_tokens_per_second": 29249.079
    },
    {
      "epoch": 2.588544626686451,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7338653802750284e-05,
      "loss": 0.38502509593963624,
      "num_input_tokens_seen": 20808015808,
      "step": 73160,
      "train_runtime": 711406.7687,
      "train_tokens_per_second": 29249.111
    },
    {
      "epoch": 2.58889844695018,
      "grad_norm": 0.625,
      "learning_rate": 1.7337611396624114e-05,
      "loss": 0.38944072723388673,
      "num_input_tokens_seen": 20810881280,
      "step": 73170,
      "train_runtime": 711505.1033,
      "train_tokens_per_second": 29249.096
    },
    {
      "epoch": 2.589252267213909,
      "grad_norm": 0.61328125,
      "learning_rate": 1.733656917848484e-05,
      "loss": 0.38742449283599856,
      "num_input_tokens_seen": 20813741248,
      "step": 73180,
      "train_runtime": 711600.366,
      "train_tokens_per_second": 29249.2
    },
    {
      "epoch": 2.5896060874776374,
      "grad_norm": 0.62890625,
      "learning_rate": 1.733552714827596e-05,
      "loss": 0.38803579807281496,
      "num_input_tokens_seen": 20816625088,
      "step": 73190,
      "train_runtime": 711702.2839,
      "train_tokens_per_second": 29249.063
    },
    {
      "epoch": 2.589959907741366,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7334485305941013e-05,
      "loss": 0.38478214740753175,
      "num_input_tokens_seen": 20819518080,
      "step": 73200,
      "train_runtime": 711804.3458,
      "train_tokens_per_second": 29248.934
    },
    {
      "epoch": 2.590313728005095,
      "grad_norm": 0.609375,
      "learning_rate": 1.7333443651423543e-05,
      "loss": 0.3860636234283447,
      "num_input_tokens_seen": 20822403584,
      "step": 73210,
      "train_runtime": 711904.1,
      "train_tokens_per_second": 29248.888
    },
    {
      "epoch": 2.590667548268824,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7332402184667127e-05,
      "loss": 0.38175268173217775,
      "num_input_tokens_seen": 20825248640,
      "step": 73220,
      "train_runtime": 711999.9099,
      "train_tokens_per_second": 29248.948
    },
    {
      "epoch": 2.5910213685325525,
      "grad_norm": 0.609375,
      "learning_rate": 1.7331360905615364e-05,
      "loss": 0.38857741355895997,
      "num_input_tokens_seen": 20828073856,
      "step": 73230,
      "train_runtime": 712096.7565,
      "train_tokens_per_second": 29248.938
    },
    {
      "epoch": 2.5913751887962815,
      "grad_norm": 0.63671875,
      "learning_rate": 1.733031981421188e-05,
      "loss": 0.3916304111480713,
      "num_input_tokens_seen": 20830918720,
      "step": 73240,
      "train_runtime": 712196.0154,
      "train_tokens_per_second": 29248.856
    },
    {
      "epoch": 2.59172900906001,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7329278910400316e-05,
      "loss": 0.38816845417022705,
      "num_input_tokens_seen": 20833775936,
      "step": 73250,
      "train_runtime": 712292.6427,
      "train_tokens_per_second": 29248.9
    },
    {
      "epoch": 2.592082829323739,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7328238194124348e-05,
      "loss": 0.3832723617553711,
      "num_input_tokens_seen": 20836626816,
      "step": 73260,
      "train_runtime": 712390.7221,
      "train_tokens_per_second": 29248.874
    },
    {
      "epoch": 2.5924366495874676,
      "grad_norm": 0.609375,
      "learning_rate": 1.7327197665327667e-05,
      "loss": 0.3866168975830078,
      "num_input_tokens_seen": 20839408512,
      "step": 73270,
      "train_runtime": 712483.4933,
      "train_tokens_per_second": 29248.97
    },
    {
      "epoch": 2.5927904698511965,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7326157323953994e-05,
      "loss": 0.3852705717086792,
      "num_input_tokens_seen": 20842184896,
      "step": 73280,
      "train_runtime": 712574.5521,
      "train_tokens_per_second": 29249.129
    },
    {
      "epoch": 2.593144290114925,
      "grad_norm": 0.609375,
      "learning_rate": 1.732511716994707e-05,
      "loss": 0.384720516204834,
      "num_input_tokens_seen": 20845055808,
      "step": 73290,
      "train_runtime": 712671.8941,
      "train_tokens_per_second": 29249.162
    },
    {
      "epoch": 2.593498110378654,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7324077203250654e-05,
      "loss": 0.3839640855789185,
      "num_input_tokens_seen": 20847931264,
      "step": 73300,
      "train_runtime": 712770.1137,
      "train_tokens_per_second": 29249.166
    },
    {
      "epoch": 2.5938519306423826,
      "grad_norm": 0.625,
      "learning_rate": 1.7323037423808545e-05,
      "loss": 0.38370201587677,
      "num_input_tokens_seen": 20850769152,
      "step": 73310,
      "train_runtime": 712864.9451,
      "train_tokens_per_second": 29249.256
    },
    {
      "epoch": 2.5942057509061116,
      "grad_norm": 0.609375,
      "learning_rate": 1.7321997831564546e-05,
      "loss": 0.38873872756958006,
      "num_input_tokens_seen": 20853610880,
      "step": 73320,
      "train_runtime": 712962.1339,
      "train_tokens_per_second": 29249.254
    },
    {
      "epoch": 2.5945595711698406,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7320958426462496e-05,
      "loss": 0.386285138130188,
      "num_input_tokens_seen": 20856452352,
      "step": 73330,
      "train_runtime": 713057.6688,
      "train_tokens_per_second": 29249.321
    },
    {
      "epoch": 2.594913391433569,
      "grad_norm": 0.6171875,
      "learning_rate": 1.731991920844626e-05,
      "loss": 0.3880527257919312,
      "num_input_tokens_seen": 20859311808,
      "step": 73340,
      "train_runtime": 713156.5236,
      "train_tokens_per_second": 29249.276
    },
    {
      "epoch": 2.5952672116972977,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7318880177459714e-05,
      "loss": 0.3857563495635986,
      "num_input_tokens_seen": 20862189568,
      "step": 73350,
      "train_runtime": 713256.1177,
      "train_tokens_per_second": 29249.226
    },
    {
      "epoch": 2.5956210319610267,
      "grad_norm": 0.609375,
      "learning_rate": 1.7317841333446765e-05,
      "loss": 0.38396296501159666,
      "num_input_tokens_seen": 20865010560,
      "step": 73360,
      "train_runtime": 713351.5974,
      "train_tokens_per_second": 29249.266
    },
    {
      "epoch": 2.5959748522247557,
      "grad_norm": 0.62109375,
      "learning_rate": 1.731680267635135e-05,
      "loss": 0.3856297492980957,
      "num_input_tokens_seen": 20867867264,
      "step": 73370,
      "train_runtime": 713449.3101,
      "train_tokens_per_second": 29249.264
    },
    {
      "epoch": 2.596328672488484,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7315764206117417e-05,
      "loss": 0.3855828523635864,
      "num_input_tokens_seen": 20870682048,
      "step": 73380,
      "train_runtime": 713544.7762,
      "train_tokens_per_second": 29249.296
    },
    {
      "epoch": 2.596682492752213,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7314725922688945e-05,
      "loss": 0.3825018882751465,
      "num_input_tokens_seen": 20873590400,
      "step": 73390,
      "train_runtime": 713644.7422,
      "train_tokens_per_second": 29249.274
    },
    {
      "epoch": 2.5970363130159417,
      "grad_norm": 0.6171875,
      "learning_rate": 1.731368782600993e-05,
      "loss": 0.38587749004364014,
      "num_input_tokens_seen": 20876455424,
      "step": 73400,
      "train_runtime": 713741.4389,
      "train_tokens_per_second": 29249.325
    },
    {
      "epoch": 2.5973901332796707,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7312649916024402e-05,
      "loss": 0.38906397819519045,
      "num_input_tokens_seen": 20879320896,
      "step": 73410,
      "train_runtime": 713838.5318,
      "train_tokens_per_second": 29249.361
    },
    {
      "epoch": 2.5977439535433993,
      "grad_norm": 0.6171875,
      "learning_rate": 1.731161219267641e-05,
      "loss": 0.3845083713531494,
      "num_input_tokens_seen": 20882183040,
      "step": 73420,
      "train_runtime": 713935.9455,
      "train_tokens_per_second": 29249.379
    },
    {
      "epoch": 2.5980977738071283,
      "grad_norm": 0.60546875,
      "learning_rate": 1.731057465591002e-05,
      "loss": 0.3824885368347168,
      "num_input_tokens_seen": 20885040192,
      "step": 73430,
      "train_runtime": 714037.7764,
      "train_tokens_per_second": 29249.209
    },
    {
      "epoch": 2.598451594070857,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7309537305669334e-05,
      "loss": 0.38613746166229246,
      "num_input_tokens_seen": 20887862656,
      "step": 73440,
      "train_runtime": 714131.2075,
      "train_tokens_per_second": 29249.335
    },
    {
      "epoch": 2.598805414334586,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7308500141898458e-05,
      "loss": 0.3843658447265625,
      "num_input_tokens_seen": 20890725696,
      "step": 73450,
      "train_runtime": 714225.9561,
      "train_tokens_per_second": 29249.463
    },
    {
      "epoch": 2.5991592345983143,
      "grad_norm": 0.62890625,
      "learning_rate": 1.730746316454155e-05,
      "loss": 0.37988944053649903,
      "num_input_tokens_seen": 20893585792,
      "step": 73460,
      "train_runtime": 714326.1766,
      "train_tokens_per_second": 29249.363
    },
    {
      "epoch": 2.5995130548620433,
      "grad_norm": 0.65234375,
      "learning_rate": 1.730642637354276e-05,
      "loss": 0.38753297328948977,
      "num_input_tokens_seen": 20896413568,
      "step": 73470,
      "train_runtime": 714420.8511,
      "train_tokens_per_second": 29249.445
    },
    {
      "epoch": 2.5998668751257723,
      "grad_norm": 0.609375,
      "learning_rate": 1.7305389768846286e-05,
      "loss": 0.38197689056396483,
      "num_input_tokens_seen": 20899254592,
      "step": 73480,
      "train_runtime": 714517.1223,
      "train_tokens_per_second": 29249.48
    },
    {
      "epoch": 2.600220695389501,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7304353350396332e-05,
      "loss": 0.38238523006439207,
      "num_input_tokens_seen": 20902064384,
      "step": 73490,
      "train_runtime": 714612.0366,
      "train_tokens_per_second": 29249.527
    },
    {
      "epoch": 2.6005745156532294,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7303317118137143e-05,
      "loss": 0.3910677433013916,
      "num_input_tokens_seen": 20904853440,
      "step": 73500,
      "train_runtime": 714705.8951,
      "train_tokens_per_second": 29249.589
    },
    {
      "epoch": 2.6009283359169584,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7302281072012973e-05,
      "loss": 0.38762850761413575,
      "num_input_tokens_seen": 20907649664,
      "step": 73510,
      "train_runtime": 714798.7037,
      "train_tokens_per_second": 29249.703
    },
    {
      "epoch": 2.6012821561806874,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7301245211968094e-05,
      "loss": 0.38384621143341063,
      "num_input_tokens_seen": 20910440256,
      "step": 73520,
      "train_runtime": 714894.2926,
      "train_tokens_per_second": 29249.695
    },
    {
      "epoch": 2.601635976444416,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7300209537946832e-05,
      "loss": 0.38538570404052735,
      "num_input_tokens_seen": 20913234880,
      "step": 73530,
      "train_runtime": 714987.3653,
      "train_tokens_per_second": 29249.796
    },
    {
      "epoch": 2.6019897967081445,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7299174049893497e-05,
      "loss": 0.3852902173995972,
      "num_input_tokens_seen": 20916057088,
      "step": 73540,
      "train_runtime": 715084.7642,
      "train_tokens_per_second": 29249.759
    },
    {
      "epoch": 2.6023436169718734,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7298138747752453e-05,
      "loss": 0.38555431365966797,
      "num_input_tokens_seen": 20918907264,
      "step": 73550,
      "train_runtime": 715182.9946,
      "train_tokens_per_second": 29249.727
    },
    {
      "epoch": 2.6026974372356024,
      "grad_norm": 0.625,
      "learning_rate": 1.729710363146807e-05,
      "loss": 0.38380913734436034,
      "num_input_tokens_seen": 20921776704,
      "step": 73560,
      "train_runtime": 715282.5639,
      "train_tokens_per_second": 29249.667
    },
    {
      "epoch": 2.603051257499331,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7296068700984743e-05,
      "loss": 0.3886416912078857,
      "num_input_tokens_seen": 20924649920,
      "step": 73570,
      "train_runtime": 715380.6677,
      "train_tokens_per_second": 29249.672
    },
    {
      "epoch": 2.60340507776306,
      "grad_norm": 0.625,
      "learning_rate": 1.7295033956246902e-05,
      "loss": 0.39080562591552737,
      "num_input_tokens_seen": 20927540288,
      "step": 73580,
      "train_runtime": 715481.2225,
      "train_tokens_per_second": 29249.601
    },
    {
      "epoch": 2.6037588980267885,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7293999397198985e-05,
      "loss": 0.3841590881347656,
      "num_input_tokens_seen": 20930353536,
      "step": 73590,
      "train_runtime": 715576.1632,
      "train_tokens_per_second": 29249.652
    },
    {
      "epoch": 2.6041127182905175,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7292965023785464e-05,
      "loss": 0.38810131549835203,
      "num_input_tokens_seen": 20933269120,
      "step": 73600,
      "train_runtime": 715676.3002,
      "train_tokens_per_second": 29249.633
    },
    {
      "epoch": 2.604466538554246,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7291930835950827e-05,
      "loss": 0.385075044631958,
      "num_input_tokens_seen": 20936127552,
      "step": 73610,
      "train_runtime": 715772.8776,
      "train_tokens_per_second": 29249.68
    },
    {
      "epoch": 2.604820358817975,
      "grad_norm": 0.62109375,
      "learning_rate": 1.72908968336396e-05,
      "loss": 0.39043853282928465,
      "num_input_tokens_seen": 20938932160,
      "step": 73620,
      "train_runtime": 715871.9027,
      "train_tokens_per_second": 29249.552
    },
    {
      "epoch": 2.6051741790817036,
      "grad_norm": 0.609375,
      "learning_rate": 1.728986301679631e-05,
      "loss": 0.3904495477676392,
      "num_input_tokens_seen": 20941741568,
      "step": 73630,
      "train_runtime": 715966.6537,
      "train_tokens_per_second": 29249.605
    },
    {
      "epoch": 2.6055279993454326,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7288829385365522e-05,
      "loss": 0.384967565536499,
      "num_input_tokens_seen": 20944624768,
      "step": 73640,
      "train_runtime": 716066.6603,
      "train_tokens_per_second": 29249.546
    },
    {
      "epoch": 2.605881819609161,
      "grad_norm": 0.59375,
      "learning_rate": 1.7287795939291814e-05,
      "loss": 0.38532323837280275,
      "num_input_tokens_seen": 20947419776,
      "step": 73650,
      "train_runtime": 716162.4076,
      "train_tokens_per_second": 29249.538
    },
    {
      "epoch": 2.60623563987289,
      "grad_norm": 0.62890625,
      "learning_rate": 1.728676267851981e-05,
      "loss": 0.38766984939575194,
      "num_input_tokens_seen": 20950244352,
      "step": 73660,
      "train_runtime": 716259.4315,
      "train_tokens_per_second": 29249.52
    },
    {
      "epoch": 2.606589460136619,
      "grad_norm": 0.640625,
      "learning_rate": 1.728572960299412e-05,
      "loss": 0.3832993507385254,
      "num_input_tokens_seen": 20953037376,
      "step": 73670,
      "train_runtime": 716353.1972,
      "train_tokens_per_second": 29249.59
    },
    {
      "epoch": 2.6069432804003476,
      "grad_norm": 0.65234375,
      "learning_rate": 1.7284696712659413e-05,
      "loss": 0.3878342628479004,
      "num_input_tokens_seen": 20955934848,
      "step": 73680,
      "train_runtime": 716453.0024,
      "train_tokens_per_second": 29249.56
    },
    {
      "epoch": 2.607297100664076,
      "grad_norm": 0.6015625,
      "learning_rate": 1.728366400746036e-05,
      "loss": 0.38535079956054685,
      "num_input_tokens_seen": 20958793728,
      "step": 73690,
      "train_runtime": 716550.4947,
      "train_tokens_per_second": 29249.57
    },
    {
      "epoch": 2.607650920927805,
      "grad_norm": 0.65234375,
      "learning_rate": 1.7282631487341668e-05,
      "loss": 0.3885688304901123,
      "num_input_tokens_seen": 20961629568,
      "step": 73700,
      "train_runtime": 716646.4262,
      "train_tokens_per_second": 29249.612
    },
    {
      "epoch": 2.608004741191534,
      "grad_norm": 0.6015625,
      "learning_rate": 1.728159915224805e-05,
      "loss": 0.38648903369903564,
      "num_input_tokens_seen": 20964455104,
      "step": 73710,
      "train_runtime": 716742.7275,
      "train_tokens_per_second": 29249.624
    },
    {
      "epoch": 2.6083585614552627,
      "grad_norm": 0.62109375,
      "learning_rate": 1.728056700212426e-05,
      "loss": 0.38951883316040037,
      "num_input_tokens_seen": 20967304832,
      "step": 73720,
      "train_runtime": 716838.9429,
      "train_tokens_per_second": 29249.673
    },
    {
      "epoch": 2.6087123817189917,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7279535036915067e-05,
      "loss": 0.3854218006134033,
      "num_input_tokens_seen": 20970106624,
      "step": 73730,
      "train_runtime": 716934.0214,
      "train_tokens_per_second": 29249.702
    },
    {
      "epoch": 2.60906620198272,
      "grad_norm": 0.62109375,
      "learning_rate": 1.727850325656526e-05,
      "loss": 0.3818650245666504,
      "num_input_tokens_seen": 20972956032,
      "step": 73740,
      "train_runtime": 717029.4234,
      "train_tokens_per_second": 29249.784
    },
    {
      "epoch": 2.609420022246449,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7277471661019656e-05,
      "loss": 0.3885868310928345,
      "num_input_tokens_seen": 20975826112,
      "step": 73750,
      "train_runtime": 717128.0438,
      "train_tokens_per_second": 29249.764
    },
    {
      "epoch": 2.6097738425101777,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7276440250223096e-05,
      "loss": 0.3880035161972046,
      "num_input_tokens_seen": 20978662528,
      "step": 73760,
      "train_runtime": 717225.1514,
      "train_tokens_per_second": 29249.759
    },
    {
      "epoch": 2.6101276627739067,
      "grad_norm": 0.625,
      "learning_rate": 1.727540902412044e-05,
      "loss": 0.3850111484527588,
      "num_input_tokens_seen": 20981486208,
      "step": 73770,
      "train_runtime": 717319.8713,
      "train_tokens_per_second": 29249.833
    },
    {
      "epoch": 2.6104814830376353,
      "grad_norm": 0.609375,
      "learning_rate": 1.7274377982656575e-05,
      "loss": 0.38850417137146,
      "num_input_tokens_seen": 20984376768,
      "step": 73780,
      "train_runtime": 717418.0371,
      "train_tokens_per_second": 29249.859
    },
    {
      "epoch": 2.6108353033013643,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7273347125776408e-05,
      "loss": 0.3834697723388672,
      "num_input_tokens_seen": 20987196416,
      "step": 73790,
      "train_runtime": 717515.5469,
      "train_tokens_per_second": 29249.814
    },
    {
      "epoch": 2.611189123565093,
      "grad_norm": 0.62109375,
      "learning_rate": 1.727231645342487e-05,
      "loss": 0.38417582511901854,
      "num_input_tokens_seen": 20990048384,
      "step": 73800,
      "train_runtime": 717614.0842,
      "train_tokens_per_second": 29249.772
    },
    {
      "epoch": 2.611542943828822,
      "grad_norm": 0.62109375,
      "learning_rate": 1.727128596554691e-05,
      "loss": 0.38305504322052003,
      "num_input_tokens_seen": 20992882624,
      "step": 73810,
      "train_runtime": 717711.8541,
      "train_tokens_per_second": 29249.737
    },
    {
      "epoch": 2.611896764092551,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7270255662087515e-05,
      "loss": 0.3844930410385132,
      "num_input_tokens_seen": 20995766208,
      "step": 73820,
      "train_runtime": 717810.7786,
      "train_tokens_per_second": 29249.723
    },
    {
      "epoch": 2.6122505843562793,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7269225542991676e-05,
      "loss": 0.3874272108078003,
      "num_input_tokens_seen": 20998585920,
      "step": 73830,
      "train_runtime": 717905.2163,
      "train_tokens_per_second": 29249.803
    },
    {
      "epoch": 2.612604404620008,
      "grad_norm": 0.62890625,
      "learning_rate": 1.726819560820442e-05,
      "loss": 0.3849189281463623,
      "num_input_tokens_seen": 21001371392,
      "step": 73840,
      "train_runtime": 717998.062,
      "train_tokens_per_second": 29249.9
    },
    {
      "epoch": 2.612958224883737,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7267165857670794e-05,
      "loss": 0.386501407623291,
      "num_input_tokens_seen": 21004203648,
      "step": 73850,
      "train_runtime": 718096.8867,
      "train_tokens_per_second": 29249.819
    },
    {
      "epoch": 2.613312045147466,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7266136291335867e-05,
      "loss": 0.3880645990371704,
      "num_input_tokens_seen": 21007036032,
      "step": 73860,
      "train_runtime": 718191.7931,
      "train_tokens_per_second": 29249.897
    },
    {
      "epoch": 2.6136658654111944,
      "grad_norm": 0.64453125,
      "learning_rate": 1.726510690914472e-05,
      "loss": 0.38773393630981445,
      "num_input_tokens_seen": 21009850816,
      "step": 73870,
      "train_runtime": 718285.8432,
      "train_tokens_per_second": 29249.986
    },
    {
      "epoch": 2.614019685674923,
      "grad_norm": 0.64453125,
      "learning_rate": 1.7264077711042482e-05,
      "loss": 0.3821838140487671,
      "num_input_tokens_seen": 21012699328,
      "step": 73880,
      "train_runtime": 718384.9695,
      "train_tokens_per_second": 29249.915
    },
    {
      "epoch": 2.614373505938652,
      "grad_norm": 0.625,
      "learning_rate": 1.7263048696974286e-05,
      "loss": 0.3879333257675171,
      "num_input_tokens_seen": 21015604288,
      "step": 73890,
      "train_runtime": 718484.2853,
      "train_tokens_per_second": 29249.915
    },
    {
      "epoch": 2.614727326202381,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7262019866885286e-05,
      "loss": 0.38918299674987794,
      "num_input_tokens_seen": 21018489536,
      "step": 73900,
      "train_runtime": 718582.2685,
      "train_tokens_per_second": 29249.942
    },
    {
      "epoch": 2.6150811464661095,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7260991220720677e-05,
      "loss": 0.38948636054992675,
      "num_input_tokens_seen": 21021335488,
      "step": 73910,
      "train_runtime": 718676.7568,
      "train_tokens_per_second": 29250.056
    },
    {
      "epoch": 2.6154349667298384,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7259962758425654e-05,
      "loss": 0.3825868844985962,
      "num_input_tokens_seen": 21024135552,
      "step": 73920,
      "train_runtime": 718769.5055,
      "train_tokens_per_second": 29250.177
    },
    {
      "epoch": 2.615788786993567,
      "grad_norm": 0.62109375,
      "learning_rate": 1.725893447994545e-05,
      "loss": 0.38635573387145994,
      "num_input_tokens_seen": 21026949824,
      "step": 73930,
      "train_runtime": 718863.8151,
      "train_tokens_per_second": 29250.255
    },
    {
      "epoch": 2.616142607257296,
      "grad_norm": 0.59765625,
      "learning_rate": 1.725790638522532e-05,
      "loss": 0.3859381198883057,
      "num_input_tokens_seen": 21029835200,
      "step": 73940,
      "train_runtime": 718963.9428,
      "train_tokens_per_second": 29250.195
    },
    {
      "epoch": 2.6164964275210245,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7256878474210535e-05,
      "loss": 0.3851091146469116,
      "num_input_tokens_seen": 21032660032,
      "step": 73950,
      "train_runtime": 719060.2858,
      "train_tokens_per_second": 29250.204
    },
    {
      "epoch": 2.6168502477847535,
      "grad_norm": 0.625,
      "learning_rate": 1.7255850746846394e-05,
      "loss": 0.3885054588317871,
      "num_input_tokens_seen": 21035511104,
      "step": 73960,
      "train_runtime": 719159.0054,
      "train_tokens_per_second": 29250.153
    },
    {
      "epoch": 2.6172040680484825,
      "grad_norm": 0.66015625,
      "learning_rate": 1.7254823203078215e-05,
      "loss": 0.38359837532043456,
      "num_input_tokens_seen": 21038345728,
      "step": 73970,
      "train_runtime": 719260.6023,
      "train_tokens_per_second": 29249.963
    },
    {
      "epoch": 2.617557888312211,
      "grad_norm": 0.6484375,
      "learning_rate": 1.7253795842851346e-05,
      "loss": 0.3879091739654541,
      "num_input_tokens_seen": 21041184256,
      "step": 73980,
      "train_runtime": 719355.7815,
      "train_tokens_per_second": 29250.038
    },
    {
      "epoch": 2.6179117085759396,
      "grad_norm": 0.640625,
      "learning_rate": 1.7252768666111146e-05,
      "loss": 0.39066300392150877,
      "num_input_tokens_seen": 21044001600,
      "step": 73990,
      "train_runtime": 719451.7995,
      "train_tokens_per_second": 29250.051
    },
    {
      "epoch": 2.6182655288396686,
      "grad_norm": 0.62109375,
      "learning_rate": 1.725174167280301e-05,
      "loss": 0.3870569705963135,
      "num_input_tokens_seen": 21046825920,
      "step": 74000,
      "train_runtime": 719548.5494,
      "train_tokens_per_second": 29250.043
    },
    {
      "epoch": 2.6186193491033976,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7250714862872344e-05,
      "loss": 0.38666629791259766,
      "num_input_tokens_seen": 21049622976,
      "step": 74010,
      "train_runtime": 719640.2701,
      "train_tokens_per_second": 29250.202
    },
    {
      "epoch": 2.618973169367126,
      "grad_norm": 0.625,
      "learning_rate": 1.7249688236264586e-05,
      "loss": 0.388053297996521,
      "num_input_tokens_seen": 21052515648,
      "step": 74020,
      "train_runtime": 719736.5742,
      "train_tokens_per_second": 29250.307
    },
    {
      "epoch": 2.6193269896308546,
      "grad_norm": 0.609375,
      "learning_rate": 1.7248661792925187e-05,
      "loss": 0.3832382678985596,
      "num_input_tokens_seen": 21055384128,
      "step": 74030,
      "train_runtime": 719833.6914,
      "train_tokens_per_second": 29250.345
    },
    {
      "epoch": 2.6196808098945836,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7247635532799638e-05,
      "loss": 0.3932713747024536,
      "num_input_tokens_seen": 21058265024,
      "step": 74040,
      "train_runtime": 719933.8618,
      "train_tokens_per_second": 29250.277
    },
    {
      "epoch": 2.6200346301583126,
      "grad_norm": 0.625,
      "learning_rate": 1.7246609455833433e-05,
      "loss": 0.39045188426971433,
      "num_input_tokens_seen": 21061118592,
      "step": 74050,
      "train_runtime": 720032.4665,
      "train_tokens_per_second": 29250.235
    },
    {
      "epoch": 2.620388450422041,
      "grad_norm": 0.640625,
      "learning_rate": 1.7245583561972093e-05,
      "loss": 0.38668007850646974,
      "num_input_tokens_seen": 21063940864,
      "step": 74060,
      "train_runtime": 720127.1913,
      "train_tokens_per_second": 29250.306
    },
    {
      "epoch": 2.62074227068577,
      "grad_norm": 0.625,
      "learning_rate": 1.724455785116118e-05,
      "loss": 0.38700165748596194,
      "num_input_tokens_seen": 21066691904,
      "step": 74070,
      "train_runtime": 720217.7977,
      "train_tokens_per_second": 29250.446
    },
    {
      "epoch": 2.6210960909494987,
      "grad_norm": 0.59375,
      "learning_rate": 1.7243532323346245e-05,
      "loss": 0.3821300983428955,
      "num_input_tokens_seen": 21069529344,
      "step": 74080,
      "train_runtime": 720315.8648,
      "train_tokens_per_second": 29250.403
    },
    {
      "epoch": 2.6214499112132277,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7242506978472894e-05,
      "loss": 0.38727333545684817,
      "num_input_tokens_seen": 21072362368,
      "step": 74090,
      "train_runtime": 720414.6346,
      "train_tokens_per_second": 29250.325
    },
    {
      "epoch": 2.6218037314769562,
      "grad_norm": 0.609375,
      "learning_rate": 1.724148181648674e-05,
      "loss": 0.3851685762405396,
      "num_input_tokens_seen": 21075212032,
      "step": 74100,
      "train_runtime": 720512.9842,
      "train_tokens_per_second": 29250.288
    },
    {
      "epoch": 2.622157551740685,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7240456837333422e-05,
      "loss": 0.3895851135253906,
      "num_input_tokens_seen": 21078059712,
      "step": 74110,
      "train_runtime": 720611.6324,
      "train_tokens_per_second": 29250.235
    },
    {
      "epoch": 2.6225113720044138,
      "grad_norm": 0.6015625,
      "learning_rate": 1.72394320409586e-05,
      "loss": 0.38266630172729493,
      "num_input_tokens_seen": 21080842944,
      "step": 74120,
      "train_runtime": 720705.1064,
      "train_tokens_per_second": 29250.303
    },
    {
      "epoch": 2.6228651922681427,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7238407427307952e-05,
      "loss": 0.3894216775894165,
      "num_input_tokens_seen": 21083691840,
      "step": 74130,
      "train_runtime": 720802.75,
      "train_tokens_per_second": 29250.293
    },
    {
      "epoch": 2.6232190125318713,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7237382996327196e-05,
      "loss": 0.38794474601745604,
      "num_input_tokens_seen": 21086498176,
      "step": 74140,
      "train_runtime": 720898.5875,
      "train_tokens_per_second": 29250.298
    },
    {
      "epoch": 2.6235728327956003,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7236358747962054e-05,
      "loss": 0.38649075031280516,
      "num_input_tokens_seen": 21089361856,
      "step": 74150,
      "train_runtime": 720998.8592,
      "train_tokens_per_second": 29250.201
    },
    {
      "epoch": 2.6239266530593293,
      "grad_norm": 0.640625,
      "learning_rate": 1.723533468215827e-05,
      "loss": 0.3895872116088867,
      "num_input_tokens_seen": 21092201920,
      "step": 74160,
      "train_runtime": 721095.2183,
      "train_tokens_per_second": 29250.231
    },
    {
      "epoch": 2.624280473323058,
      "grad_norm": 0.66796875,
      "learning_rate": 1.7234310798861627e-05,
      "loss": 0.38542847633361815,
      "num_input_tokens_seen": 21095045888,
      "step": 74170,
      "train_runtime": 721194.1609,
      "train_tokens_per_second": 29250.162
    },
    {
      "epoch": 2.6246342935867863,
      "grad_norm": 0.61328125,
      "learning_rate": 1.723328709801792e-05,
      "loss": 0.3887073516845703,
      "num_input_tokens_seen": 21097907136,
      "step": 74180,
      "train_runtime": 721290.8753,
      "train_tokens_per_second": 29250.207
    },
    {
      "epoch": 2.6249881138505153,
      "grad_norm": 0.59375,
      "learning_rate": 1.7232263579572966e-05,
      "loss": 0.3836081027984619,
      "num_input_tokens_seen": 21100776960,
      "step": 74190,
      "train_runtime": 721386.928,
      "train_tokens_per_second": 29250.29
    },
    {
      "epoch": 2.6253419341142443,
      "grad_norm": 0.61328125,
      "learning_rate": 1.723124024347261e-05,
      "loss": 0.3843811511993408,
      "num_input_tokens_seen": 21103612416,
      "step": 74200,
      "train_runtime": 721485.4236,
      "train_tokens_per_second": 29250.227
    },
    {
      "epoch": 2.625695754377973,
      "grad_norm": 0.6328125,
      "learning_rate": 1.723021708966271e-05,
      "loss": 0.38996295928955077,
      "num_input_tokens_seen": 21106506368,
      "step": 74210,
      "train_runtime": 721586.903,
      "train_tokens_per_second": 29250.124
    },
    {
      "epoch": 2.626049574641702,
      "grad_norm": 0.640625,
      "learning_rate": 1.7229194118089154e-05,
      "loss": 0.3860771179199219,
      "num_input_tokens_seen": 21109353856,
      "step": 74220,
      "train_runtime": 721687.1291,
      "train_tokens_per_second": 29250.007
    },
    {
      "epoch": 2.6264033949054304,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7228171328697853e-05,
      "loss": 0.3857696771621704,
      "num_input_tokens_seen": 21112211456,
      "step": 74230,
      "train_runtime": 721784.4995,
      "train_tokens_per_second": 29250.021
    },
    {
      "epoch": 2.6267572151691594,
      "grad_norm": 0.609375,
      "learning_rate": 1.7227148721434737e-05,
      "loss": 0.38657779693603517,
      "num_input_tokens_seen": 21115057344,
      "step": 74240,
      "train_runtime": 721880.3118,
      "train_tokens_per_second": 29250.081
    },
    {
      "epoch": 2.627111035432888,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7226126296245764e-05,
      "loss": 0.38780524730682375,
      "num_input_tokens_seen": 21117978752,
      "step": 74250,
      "train_runtime": 721983.4398,
      "train_tokens_per_second": 29249.949
    },
    {
      "epoch": 2.627464855696617,
      "grad_norm": 0.625,
      "learning_rate": 1.7225104053076904e-05,
      "loss": 0.38394529819488527,
      "num_input_tokens_seen": 21120835712,
      "step": 74260,
      "train_runtime": 722082.154,
      "train_tokens_per_second": 29249.907
    },
    {
      "epoch": 2.6278186759603455,
      "grad_norm": 0.6171875,
      "learning_rate": 1.722408199187416e-05,
      "loss": 0.3886481761932373,
      "num_input_tokens_seen": 21123661312,
      "step": 74270,
      "train_runtime": 722179.004,
      "train_tokens_per_second": 29249.897
    },
    {
      "epoch": 2.6281724962240745,
      "grad_norm": 0.6171875,
      "learning_rate": 1.722306011258355e-05,
      "loss": 0.3893352746963501,
      "num_input_tokens_seen": 21126457792,
      "step": 74280,
      "train_runtime": 722273.9977,
      "train_tokens_per_second": 29249.922
    },
    {
      "epoch": 2.628526316487803,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7222038415151115e-05,
      "loss": 0.385011100769043,
      "num_input_tokens_seen": 21129291136,
      "step": 74290,
      "train_runtime": 722371.5036,
      "train_tokens_per_second": 29249.896
    },
    {
      "epoch": 2.628880136751532,
      "grad_norm": 0.609375,
      "learning_rate": 1.7221016899522925e-05,
      "loss": 0.3886619806289673,
      "num_input_tokens_seen": 21132155264,
      "step": 74300,
      "train_runtime": 722469.4719,
      "train_tokens_per_second": 29249.894
    },
    {
      "epoch": 2.629233957015261,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7219995565645076e-05,
      "loss": 0.3862520694732666,
      "num_input_tokens_seen": 21135021504,
      "step": 74310,
      "train_runtime": 722567.19,
      "train_tokens_per_second": 29249.905
    },
    {
      "epoch": 2.6295877772789895,
      "grad_norm": 0.625,
      "learning_rate": 1.721897441346366e-05,
      "loss": 0.387230658531189,
      "num_input_tokens_seen": 21137851712,
      "step": 74320,
      "train_runtime": 722661.3885,
      "train_tokens_per_second": 29250.008
    },
    {
      "epoch": 2.629941597542718,
      "grad_norm": 0.5859375,
      "learning_rate": 1.721795344292483e-05,
      "loss": 0.3869665861129761,
      "num_input_tokens_seen": 21140660928,
      "step": 74330,
      "train_runtime": 722756.6684,
      "train_tokens_per_second": 29250.039
    },
    {
      "epoch": 2.630295417806447,
      "grad_norm": 0.65234375,
      "learning_rate": 1.7216932653974727e-05,
      "loss": 0.38742647171020506,
      "num_input_tokens_seen": 21143533184,
      "step": 74340,
      "train_runtime": 722854.2076,
      "train_tokens_per_second": 29250.066
    },
    {
      "epoch": 2.630649238070176,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7215912046559533e-05,
      "loss": 0.3892971992492676,
      "num_input_tokens_seen": 21146320576,
      "step": 74350,
      "train_runtime": 722950.9025,
      "train_tokens_per_second": 29250.009
    },
    {
      "epoch": 2.6310030583339046,
      "grad_norm": 0.64453125,
      "learning_rate": 1.7214891620625454e-05,
      "loss": 0.39136877059936526,
      "num_input_tokens_seen": 21149176960,
      "step": 74360,
      "train_runtime": 723048.1644,
      "train_tokens_per_second": 29250.025
    },
    {
      "epoch": 2.631356878597633,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7213871376118704e-05,
      "loss": 0.38932342529296876,
      "num_input_tokens_seen": 21152039488,
      "step": 74370,
      "train_runtime": 723143.9373,
      "train_tokens_per_second": 29250.11
    },
    {
      "epoch": 2.631710698861362,
      "grad_norm": 0.60546875,
      "learning_rate": 1.721285131298553e-05,
      "loss": 0.3894200801849365,
      "num_input_tokens_seen": 21154890176,
      "step": 74380,
      "train_runtime": 723240.5038,
      "train_tokens_per_second": 29250.146
    },
    {
      "epoch": 2.632064519125091,
      "grad_norm": 0.6015625,
      "learning_rate": 1.72118314311722e-05,
      "loss": 0.3884094476699829,
      "num_input_tokens_seen": 21157692672,
      "step": 74390,
      "train_runtime": 723336.071,
      "train_tokens_per_second": 29250.156
    },
    {
      "epoch": 2.6324183393888196,
      "grad_norm": 0.609375,
      "learning_rate": 1.7210811730625004e-05,
      "loss": 0.3872340679168701,
      "num_input_tokens_seen": 21160574016,
      "step": 74400,
      "train_runtime": 723435.6168,
      "train_tokens_per_second": 29250.114
    },
    {
      "epoch": 2.6327721596525486,
      "grad_norm": 0.62890625,
      "learning_rate": 1.720979221129026e-05,
      "loss": 0.38279070854187014,
      "num_input_tokens_seen": 21163414784,
      "step": 74410,
      "train_runtime": 723531.4542,
      "train_tokens_per_second": 29250.166
    },
    {
      "epoch": 2.633125979916277,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7208772873114285e-05,
      "loss": 0.3878284454345703,
      "num_input_tokens_seen": 21166249216,
      "step": 74420,
      "train_runtime": 723626.5484,
      "train_tokens_per_second": 29250.239
    },
    {
      "epoch": 2.633479800180006,
      "grad_norm": 0.61328125,
      "learning_rate": 1.720775371604345e-05,
      "loss": 0.38469562530517576,
      "num_input_tokens_seen": 21169072000,
      "step": 74430,
      "train_runtime": 723721.837,
      "train_tokens_per_second": 29250.288
    },
    {
      "epoch": 2.6338336204437347,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7206734740024123e-05,
      "loss": 0.38969926834106444,
      "num_input_tokens_seen": 21171826496,
      "step": 74440,
      "train_runtime": 723816.3905,
      "train_tokens_per_second": 29250.272
    },
    {
      "epoch": 2.6341874407074637,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7205715945002714e-05,
      "loss": 0.385995888710022,
      "num_input_tokens_seen": 21174732928,
      "step": 74450,
      "train_runtime": 723917.2761,
      "train_tokens_per_second": 29250.211
    },
    {
      "epoch": 2.6345412609711922,
      "grad_norm": 0.6328125,
      "learning_rate": 1.720469733092564e-05,
      "loss": 0.38598835468292236,
      "num_input_tokens_seen": 21177572160,
      "step": 74460,
      "train_runtime": 724010.6643,
      "train_tokens_per_second": 29250.359
    },
    {
      "epoch": 2.634895081234921,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7203678897739344e-05,
      "loss": 0.3873575687408447,
      "num_input_tokens_seen": 21180421568,
      "step": 74470,
      "train_runtime": 724109.7653,
      "train_tokens_per_second": 29250.291
    },
    {
      "epoch": 2.6352489014986498,
      "grad_norm": 0.65625,
      "learning_rate": 1.72026606453903e-05,
      "loss": 0.3840164661407471,
      "num_input_tokens_seen": 21183226624,
      "step": 74480,
      "train_runtime": 724205.8702,
      "train_tokens_per_second": 29250.283
    },
    {
      "epoch": 2.6356027217623788,
      "grad_norm": 0.58984375,
      "learning_rate": 1.7201642573824997e-05,
      "loss": 0.3837096691131592,
      "num_input_tokens_seen": 21186133120,
      "step": 74490,
      "train_runtime": 724304.8472,
      "train_tokens_per_second": 29250.299
    },
    {
      "epoch": 2.6359565420261077,
      "grad_norm": 0.625,
      "learning_rate": 1.7200624682989938e-05,
      "loss": 0.38433451652526857,
      "num_input_tokens_seen": 21188933696,
      "step": 74500,
      "train_runtime": 724400.8244,
      "train_tokens_per_second": 29250.289
    },
    {
      "epoch": 2.6363103622898363,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7199606972831665e-05,
      "loss": 0.38067963123321535,
      "num_input_tokens_seen": 21191812480,
      "step": 74510,
      "train_runtime": 724498.8213,
      "train_tokens_per_second": 29250.306
    },
    {
      "epoch": 2.636664182553565,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7198589443296727e-05,
      "loss": 0.3851020336151123,
      "num_input_tokens_seen": 21194666944,
      "step": 74520,
      "train_runtime": 724596.4691,
      "train_tokens_per_second": 29250.304
    },
    {
      "epoch": 2.637018002817294,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7197572094331708e-05,
      "loss": 0.3846184730529785,
      "num_input_tokens_seen": 21197567488,
      "step": 74530,
      "train_runtime": 724697.0592,
      "train_tokens_per_second": 29250.246
    },
    {
      "epoch": 2.637371823081023,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7196554925883205e-05,
      "loss": 0.3866997957229614,
      "num_input_tokens_seen": 21200438720,
      "step": 74540,
      "train_runtime": 724796.5692,
      "train_tokens_per_second": 29250.192
    },
    {
      "epoch": 2.6377256433447513,
      "grad_norm": 0.625,
      "learning_rate": 1.719553793789784e-05,
      "loss": 0.3820904016494751,
      "num_input_tokens_seen": 21203302528,
      "step": 74550,
      "train_runtime": 724895.4851,
      "train_tokens_per_second": 29250.151
    },
    {
      "epoch": 2.6380794636084803,
      "grad_norm": 0.609375,
      "learning_rate": 1.7194521130322257e-05,
      "loss": 0.38410923480987547,
      "num_input_tokens_seen": 21206112320,
      "step": 74560,
      "train_runtime": 724991.5268,
      "train_tokens_per_second": 29250.152
    },
    {
      "epoch": 2.638433283872209,
      "grad_norm": 0.625,
      "learning_rate": 1.719350450310313e-05,
      "loss": 0.3834213733673096,
      "num_input_tokens_seen": 21208951360,
      "step": 74570,
      "train_runtime": 725088.4165,
      "train_tokens_per_second": 29250.159
    },
    {
      "epoch": 2.638787104135938,
      "grad_norm": 0.6328125,
      "learning_rate": 1.719248805618713e-05,
      "loss": 0.38759140968322753,
      "num_input_tokens_seen": 21211793536,
      "step": 74580,
      "train_runtime": 725183.3092,
      "train_tokens_per_second": 29250.251
    },
    {
      "epoch": 2.6391409243996664,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7191471789520986e-05,
      "loss": 0.3907924652099609,
      "num_input_tokens_seen": 21214634752,
      "step": 74590,
      "train_runtime": 725281.0295,
      "train_tokens_per_second": 29250.227
    },
    {
      "epoch": 2.6394947446633954,
      "grad_norm": 0.625,
      "learning_rate": 1.7190455703051418e-05,
      "loss": 0.3893891334533691,
      "num_input_tokens_seen": 21217478016,
      "step": 74600,
      "train_runtime": 725378.3345,
      "train_tokens_per_second": 29250.223
    },
    {
      "epoch": 2.639848564927124,
      "grad_norm": 0.62109375,
      "learning_rate": 1.718943979672519e-05,
      "loss": 0.38451220989227297,
      "num_input_tokens_seen": 21220320320,
      "step": 74610,
      "train_runtime": 725475.615,
      "train_tokens_per_second": 29250.219
    },
    {
      "epoch": 2.640202385190853,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7188424070489067e-05,
      "loss": 0.38333878517150877,
      "num_input_tokens_seen": 21223128576,
      "step": 74620,
      "train_runtime": 725571.6979,
      "train_tokens_per_second": 29250.216
    },
    {
      "epoch": 2.6405562054545815,
      "grad_norm": 0.61328125,
      "learning_rate": 1.718740852428986e-05,
      "loss": 0.39389753341674805,
      "num_input_tokens_seen": 21226022720,
      "step": 74630,
      "train_runtime": 725670.6523,
      "train_tokens_per_second": 29250.215
    },
    {
      "epoch": 2.6409100257183105,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7186393158074376e-05,
      "loss": 0.38661441802978513,
      "num_input_tokens_seen": 21228902464,
      "step": 74640,
      "train_runtime": 725771.621,
      "train_tokens_per_second": 29250.114
    },
    {
      "epoch": 2.6412638459820394,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7185377971789472e-05,
      "loss": 0.38412113189697267,
      "num_input_tokens_seen": 21231730240,
      "step": 74650,
      "train_runtime": 725866.4709,
      "train_tokens_per_second": 29250.187
    },
    {
      "epoch": 2.641617666245768,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7184362965382004e-05,
      "loss": 0.38910865783691406,
      "num_input_tokens_seen": 21234550016,
      "step": 74660,
      "train_runtime": 725960.4556,
      "train_tokens_per_second": 29250.285
    },
    {
      "epoch": 2.6419714865094965,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7183348138798857e-05,
      "loss": 0.3888072490692139,
      "num_input_tokens_seen": 21237429248,
      "step": 74670,
      "train_runtime": 726059.1722,
      "train_tokens_per_second": 29250.273
    },
    {
      "epoch": 2.6423253067732255,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7182333491986944e-05,
      "loss": 0.3864339590072632,
      "num_input_tokens_seen": 21240264064,
      "step": 74680,
      "train_runtime": 726156.2351,
      "train_tokens_per_second": 29250.267
    },
    {
      "epoch": 2.6426791270369545,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7181319024893197e-05,
      "loss": 0.38910503387451173,
      "num_input_tokens_seen": 21243128064,
      "step": 74690,
      "train_runtime": 726254.749,
      "train_tokens_per_second": 29250.243
    },
    {
      "epoch": 2.643032947300683,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7180304737464563e-05,
      "loss": 0.38635969161987305,
      "num_input_tokens_seen": 21245930688,
      "step": 74700,
      "train_runtime": 726351.2687,
      "train_tokens_per_second": 29250.215
    },
    {
      "epoch": 2.6433867675644116,
      "grad_norm": 0.63671875,
      "learning_rate": 1.717929062964802e-05,
      "loss": 0.3903571605682373,
      "num_input_tokens_seen": 21248833792,
      "step": 74710,
      "train_runtime": 726452.2539,
      "train_tokens_per_second": 29250.145
    },
    {
      "epoch": 2.6437405878281406,
      "grad_norm": 0.625,
      "learning_rate": 1.7178276701390557e-05,
      "loss": 0.38762946128845216,
      "num_input_tokens_seen": 21251690944,
      "step": 74720,
      "train_runtime": 726549.0079,
      "train_tokens_per_second": 29250.182
    },
    {
      "epoch": 2.6440944080918696,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7177262952639204e-05,
      "loss": 0.3850399017333984,
      "num_input_tokens_seen": 21254541952,
      "step": 74730,
      "train_runtime": 726648.3875,
      "train_tokens_per_second": 29250.105
    },
    {
      "epoch": 2.644448228355598,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7176249383340994e-05,
      "loss": 0.3854344129562378,
      "num_input_tokens_seen": 21257403136,
      "step": 74740,
      "train_runtime": 726745.6114,
      "train_tokens_per_second": 29250.129
    },
    {
      "epoch": 2.644802048619327,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7175235993442985e-05,
      "loss": 0.39130709171295164,
      "num_input_tokens_seen": 21260179008,
      "step": 74750,
      "train_runtime": 726836.1137,
      "train_tokens_per_second": 29250.306
    },
    {
      "epoch": 2.6451558688830556,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7174222782892264e-05,
      "loss": 0.3896197319030762,
      "num_input_tokens_seen": 21262949312,
      "step": 74760,
      "train_runtime": 726930.9648,
      "train_tokens_per_second": 29250.301
    },
    {
      "epoch": 2.6455096891467846,
      "grad_norm": 0.6328125,
      "learning_rate": 1.717320975163594e-05,
      "loss": 0.38497314453125,
      "num_input_tokens_seen": 21265767488,
      "step": 74770,
      "train_runtime": 727026.3698,
      "train_tokens_per_second": 29250.339
    },
    {
      "epoch": 2.645863509410513,
      "grad_norm": 0.62890625,
      "learning_rate": 1.717219689962114e-05,
      "loss": 0.3902001857757568,
      "num_input_tokens_seen": 21268573504,
      "step": 74780,
      "train_runtime": 727121.5073,
      "train_tokens_per_second": 29250.371
    },
    {
      "epoch": 2.646217329674242,
      "grad_norm": 0.63671875,
      "learning_rate": 1.717118422679501e-05,
      "loss": 0.3815563201904297,
      "num_input_tokens_seen": 21271434688,
      "step": 74790,
      "train_runtime": 727217.5015,
      "train_tokens_per_second": 29250.444
    },
    {
      "epoch": 2.646571149937971,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7170171733104716e-05,
      "loss": 0.3863277196884155,
      "num_input_tokens_seen": 21274258048,
      "step": 74800,
      "train_runtime": 727313.467,
      "train_tokens_per_second": 29250.466
    },
    {
      "epoch": 2.6469249702016997,
      "grad_norm": 0.609375,
      "learning_rate": 1.7169159418497464e-05,
      "loss": 0.3851405143737793,
      "num_input_tokens_seen": 21277078912,
      "step": 74810,
      "train_runtime": 727410.0677,
      "train_tokens_per_second": 29250.46
    },
    {
      "epoch": 2.6472787904654282,
      "grad_norm": 0.61328125,
      "learning_rate": 1.716814728292046e-05,
      "loss": 0.3886796712875366,
      "num_input_tokens_seen": 21279938368,
      "step": 74820,
      "train_runtime": 727506.3182,
      "train_tokens_per_second": 29250.52
    },
    {
      "epoch": 2.6476326107291572,
      "grad_norm": 0.609375,
      "learning_rate": 1.716713532632094e-05,
      "loss": 0.3882636785507202,
      "num_input_tokens_seen": 21282810560,
      "step": 74830,
      "train_runtime": 727604.8263,
      "train_tokens_per_second": 29250.508
    },
    {
      "epoch": 2.647986430992886,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7166123548646162e-05,
      "loss": 0.39129340648651123,
      "num_input_tokens_seen": 21285690624,
      "step": 74840,
      "train_runtime": 727705.8786,
      "train_tokens_per_second": 29250.404
    },
    {
      "epoch": 2.6483402512566148,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7165111949843414e-05,
      "loss": 0.3841498613357544,
      "num_input_tokens_seen": 21288509504,
      "step": 74850,
      "train_runtime": 727804.0111,
      "train_tokens_per_second": 29250.333
    },
    {
      "epoch": 2.6486940715203433,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7164100529859983e-05,
      "loss": 0.38797218799591066,
      "num_input_tokens_seen": 21291339008,
      "step": 74860,
      "train_runtime": 727899.496,
      "train_tokens_per_second": 29250.383
    },
    {
      "epoch": 2.6490478917840723,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7163089288643202e-05,
      "loss": 0.3857181787490845,
      "num_input_tokens_seen": 21294160896,
      "step": 74870,
      "train_runtime": 727997.6187,
      "train_tokens_per_second": 29250.317
    },
    {
      "epoch": 2.6494017120478013,
      "grad_norm": 0.6171875,
      "learning_rate": 1.716207822614042e-05,
      "loss": 0.3836466073989868,
      "num_input_tokens_seen": 21296970304,
      "step": 74880,
      "train_runtime": 728093.2564,
      "train_tokens_per_second": 29250.333
    },
    {
      "epoch": 2.64975553231153,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7161067342298994e-05,
      "loss": 0.3877160310745239,
      "num_input_tokens_seen": 21299887296,
      "step": 74890,
      "train_runtime": 728194.6684,
      "train_tokens_per_second": 29250.265
    },
    {
      "epoch": 2.650109352575259,
      "grad_norm": 0.61328125,
      "learning_rate": 1.716005663706632e-05,
      "loss": 0.3892143726348877,
      "num_input_tokens_seen": 21302715392,
      "step": 74900,
      "train_runtime": 728292.0218,
      "train_tokens_per_second": 29250.239
    },
    {
      "epoch": 2.6504631728389874,
      "grad_norm": 0.6171875,
      "learning_rate": 1.71590461103898e-05,
      "loss": 0.3806128025054932,
      "num_input_tokens_seen": 21305607168,
      "step": 74910,
      "train_runtime": 728391.0476,
      "train_tokens_per_second": 29250.232
    },
    {
      "epoch": 2.6508169931027163,
      "grad_norm": 0.65625,
      "learning_rate": 1.7158035762216875e-05,
      "loss": 0.39104580879211426,
      "num_input_tokens_seen": 21308404736,
      "step": 74920,
      "train_runtime": 728485.5838,
      "train_tokens_per_second": 29250.276
    },
    {
      "epoch": 2.651170813366445,
      "grad_norm": 0.625,
      "learning_rate": 1.715702559249499e-05,
      "loss": 0.39225139617919924,
      "num_input_tokens_seen": 21311278016,
      "step": 74930,
      "train_runtime": 728583.7065,
      "train_tokens_per_second": 29250.281
    },
    {
      "epoch": 2.651524633630174,
      "grad_norm": 0.609375,
      "learning_rate": 1.715601560117163e-05,
      "loss": 0.38332881927490237,
      "num_input_tokens_seen": 21314162560,
      "step": 74940,
      "train_runtime": 728685.6618,
      "train_tokens_per_second": 29250.147
    },
    {
      "epoch": 2.6518784538939024,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7155005788194283e-05,
      "loss": 0.387650203704834,
      "num_input_tokens_seen": 21317031872,
      "step": 74950,
      "train_runtime": 728785.7536,
      "train_tokens_per_second": 29250.067
    },
    {
      "epoch": 2.6522322741576314,
      "grad_norm": 0.65625,
      "learning_rate": 1.7153996153510472e-05,
      "loss": 0.38602781295776367,
      "num_input_tokens_seen": 21319871040,
      "step": 74960,
      "train_runtime": 728884.1089,
      "train_tokens_per_second": 29250.015
    },
    {
      "epoch": 2.65258609442136,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7152986697067734e-05,
      "loss": 0.3877532958984375,
      "num_input_tokens_seen": 21322697536,
      "step": 74970,
      "train_runtime": 728982.092,
      "train_tokens_per_second": 29249.961
    },
    {
      "epoch": 2.652939914685089,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7151977418813636e-05,
      "loss": 0.3876424551010132,
      "num_input_tokens_seen": 21325580480,
      "step": 74980,
      "train_runtime": 729081.5034,
      "train_tokens_per_second": 29249.927
    },
    {
      "epoch": 2.653293734948818,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7150968318695755e-05,
      "loss": 0.38400471210479736,
      "num_input_tokens_seen": 21328386432,
      "step": 74990,
      "train_runtime": 729176.1206,
      "train_tokens_per_second": 29249.979
    },
    {
      "epoch": 2.6536475552125465,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7149959396661695e-05,
      "loss": 0.38516504764556886,
      "num_input_tokens_seen": 21331230016,
      "step": 75000,
      "train_runtime": 729272.3207,
      "train_tokens_per_second": 29250.02
    },
    {
      "epoch": 2.6536475552125465,
      "eval_loss": 0.33609068393707275,
      "eval_runtime": 8.43,
      "eval_samples_per_second": 473.072,
      "eval_steps_per_second": 3.796,
      "num_input_tokens_seen": 21331230016,
      "step": 75000
    },
    {
      "epoch": 2.654001375476275,
      "grad_norm": 0.61328125,
      "learning_rate": 1.714895065265909e-05,
      "loss": 0.38525369167327883,
      "num_input_tokens_seen": 21334045504,
      "step": 75010,
      "train_runtime": 729397.8039,
      "train_tokens_per_second": 29248.848
    },
    {
      "epoch": 2.654355195740004,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7147942086635584e-05,
      "loss": 0.3812159299850464,
      "num_input_tokens_seen": 21336892864,
      "step": 75020,
      "train_runtime": 729496.1851,
      "train_tokens_per_second": 29248.807
    },
    {
      "epoch": 2.654709016003733,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7146933698538846e-05,
      "loss": 0.38089590072631835,
      "num_input_tokens_seen": 21339728320,
      "step": 75030,
      "train_runtime": 729589.9907,
      "train_tokens_per_second": 29248.932
    },
    {
      "epoch": 2.6550628362674615,
      "grad_norm": 0.62890625,
      "learning_rate": 1.714592548831657e-05,
      "loss": 0.384178900718689,
      "num_input_tokens_seen": 21342609664,
      "step": 75040,
      "train_runtime": 729690.0074,
      "train_tokens_per_second": 29248.872
    },
    {
      "epoch": 2.6554166565311905,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7144917455916462e-05,
      "loss": 0.3812519311904907,
      "num_input_tokens_seen": 21345477696,
      "step": 75050,
      "train_runtime": 729788.6767,
      "train_tokens_per_second": 29248.847
    },
    {
      "epoch": 2.655770476794919,
      "grad_norm": 0.59375,
      "learning_rate": 1.7143909601286258e-05,
      "loss": 0.3848569393157959,
      "num_input_tokens_seen": 21348371520,
      "step": 75060,
      "train_runtime": 729889.5555,
      "train_tokens_per_second": 29248.77
    },
    {
      "epoch": 2.656124297058648,
      "grad_norm": 0.578125,
      "learning_rate": 1.714290192437372e-05,
      "loss": 0.3827219486236572,
      "num_input_tokens_seen": 21351214976,
      "step": 75070,
      "train_runtime": 729989.5144,
      "train_tokens_per_second": 29248.66
    },
    {
      "epoch": 2.6564781173223766,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7141894425126622e-05,
      "loss": 0.3897125244140625,
      "num_input_tokens_seen": 21354111424,
      "step": 75080,
      "train_runtime": 730090.7149,
      "train_tokens_per_second": 29248.573
    },
    {
      "epoch": 2.6568319375861056,
      "grad_norm": 0.6171875,
      "learning_rate": 1.714088710349276e-05,
      "loss": 0.3853651285171509,
      "num_input_tokens_seen": 21356917824,
      "step": 75090,
      "train_runtime": 730184.6421,
      "train_tokens_per_second": 29248.654
    },
    {
      "epoch": 2.657185757849834,
      "grad_norm": 0.59375,
      "learning_rate": 1.7139879959419956e-05,
      "loss": 0.38927493095397947,
      "num_input_tokens_seen": 21359767360,
      "step": 75100,
      "train_runtime": 730281.1719,
      "train_tokens_per_second": 29248.69
    },
    {
      "epoch": 2.657539578113563,
      "grad_norm": 0.58984375,
      "learning_rate": 1.7138872992856052e-05,
      "loss": 0.38738260269165037,
      "num_input_tokens_seen": 21362590848,
      "step": 75110,
      "train_runtime": 730377.8378,
      "train_tokens_per_second": 29248.684
    },
    {
      "epoch": 2.6578933983772917,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7137866203748908e-05,
      "loss": 0.3886084079742432,
      "num_input_tokens_seen": 21365424768,
      "step": 75120,
      "train_runtime": 730472.7639,
      "train_tokens_per_second": 29248.763
    },
    {
      "epoch": 2.6582472186410206,
      "grad_norm": 0.625,
      "learning_rate": 1.7136859592046407e-05,
      "loss": 0.3856485366821289,
      "num_input_tokens_seen": 21368230656,
      "step": 75130,
      "train_runtime": 730568.0745,
      "train_tokens_per_second": 29248.788
    },
    {
      "epoch": 2.6586010389047496,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7135853157696462e-05,
      "loss": 0.3815124988555908,
      "num_input_tokens_seen": 21371087424,
      "step": 75140,
      "train_runtime": 730668.3444,
      "train_tokens_per_second": 29248.684
    },
    {
      "epoch": 2.658954859168478,
      "grad_norm": 0.64453125,
      "learning_rate": 1.7134846900647e-05,
      "loss": 0.38549108505249025,
      "num_input_tokens_seen": 21373933824,
      "step": 75150,
      "train_runtime": 730764.2045,
      "train_tokens_per_second": 29248.742
    },
    {
      "epoch": 2.6593086794322067,
      "grad_norm": 0.58203125,
      "learning_rate": 1.713384082084596e-05,
      "loss": 0.38742890357971194,
      "num_input_tokens_seen": 21376792704,
      "step": 75160,
      "train_runtime": 730860.7498,
      "train_tokens_per_second": 29248.79
    },
    {
      "epoch": 2.6596624996959357,
      "grad_norm": 0.61328125,
      "learning_rate": 1.713283491824132e-05,
      "loss": 0.3921644687652588,
      "num_input_tokens_seen": 21379616256,
      "step": 75170,
      "train_runtime": 730953.7484,
      "train_tokens_per_second": 29248.932
    },
    {
      "epoch": 2.6600163199596647,
      "grad_norm": 0.61328125,
      "learning_rate": 1.713182919278107e-05,
      "loss": 0.3826498031616211,
      "num_input_tokens_seen": 21382498624,
      "step": 75180,
      "train_runtime": 731056.1849,
      "train_tokens_per_second": 29248.776
    },
    {
      "epoch": 2.6603701402233932,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7130823644413222e-05,
      "loss": 0.3903824806213379,
      "num_input_tokens_seen": 21385420160,
      "step": 75190,
      "train_runtime": 731156.5335,
      "train_tokens_per_second": 29248.758
    },
    {
      "epoch": 2.660723960487122,
      "grad_norm": 0.62890625,
      "learning_rate": 1.712981827308581e-05,
      "loss": 0.384019660949707,
      "num_input_tokens_seen": 21388258624,
      "step": 75200,
      "train_runtime": 731253.0597,
      "train_tokens_per_second": 29248.778
    },
    {
      "epoch": 2.6610777807508508,
      "grad_norm": 0.625,
      "learning_rate": 1.712881307874689e-05,
      "loss": 0.3902750968933105,
      "num_input_tokens_seen": 21391087168,
      "step": 75210,
      "train_runtime": 731351.5875,
      "train_tokens_per_second": 29248.705
    },
    {
      "epoch": 2.6614316010145798,
      "grad_norm": 0.65625,
      "learning_rate": 1.712780806134454e-05,
      "loss": 0.3889188766479492,
      "num_input_tokens_seen": 21393891648,
      "step": 75220,
      "train_runtime": 731445.3455,
      "train_tokens_per_second": 29248.79
    },
    {
      "epoch": 2.6617854212783083,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7126803220826854e-05,
      "loss": 0.3854790687561035,
      "num_input_tokens_seen": 21396704512,
      "step": 75230,
      "train_runtime": 731541.3849,
      "train_tokens_per_second": 29248.796
    },
    {
      "epoch": 2.6621392415420373,
      "grad_norm": 0.609375,
      "learning_rate": 1.7125798557141954e-05,
      "loss": 0.3865359783172607,
      "num_input_tokens_seen": 21399525952,
      "step": 75240,
      "train_runtime": 731637.1957,
      "train_tokens_per_second": 29248.822
    },
    {
      "epoch": 2.662493061805766,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7124794070237987e-05,
      "loss": 0.38853440284729,
      "num_input_tokens_seen": 21402350976,
      "step": 75250,
      "train_runtime": 731731.2279,
      "train_tokens_per_second": 29248.924
    },
    {
      "epoch": 2.662846882069495,
      "grad_norm": 0.65625,
      "learning_rate": 1.71237897600631e-05,
      "loss": 0.3857755184173584,
      "num_input_tokens_seen": 21405216576,
      "step": 75260,
      "train_runtime": 731829.119,
      "train_tokens_per_second": 29248.927
    },
    {
      "epoch": 2.6632007023332234,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7122785626565498e-05,
      "loss": 0.3863802909851074,
      "num_input_tokens_seen": 21408015360,
      "step": 75270,
      "train_runtime": 731923.1369,
      "train_tokens_per_second": 29248.994
    },
    {
      "epoch": 2.6635545225969524,
      "grad_norm": 0.625,
      "learning_rate": 1.7121781669693363e-05,
      "loss": 0.3834370613098145,
      "num_input_tokens_seen": 21410867392,
      "step": 75280,
      "train_runtime": 732021.8677,
      "train_tokens_per_second": 29248.945
    },
    {
      "epoch": 2.663908342860681,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7120777889394933e-05,
      "loss": 0.38622612953186036,
      "num_input_tokens_seen": 21413760192,
      "step": 75290,
      "train_runtime": 732120.2308,
      "train_tokens_per_second": 29248.967
    },
    {
      "epoch": 2.66426216312441,
      "grad_norm": 0.62890625,
      "learning_rate": 1.711977428561845e-05,
      "loss": 0.38889827728271487,
      "num_input_tokens_seen": 21416683264,
      "step": 75300,
      "train_runtime": 732222.6123,
      "train_tokens_per_second": 29248.869
    },
    {
      "epoch": 2.6646159833881384,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7118770858312188e-05,
      "loss": 0.38661072254180906,
      "num_input_tokens_seen": 21419519360,
      "step": 75310,
      "train_runtime": 732320.4496,
      "train_tokens_per_second": 29248.834
    },
    {
      "epoch": 2.6649698036518674,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7117767607424434e-05,
      "loss": 0.38426618576049804,
      "num_input_tokens_seen": 21422421760,
      "step": 75320,
      "train_runtime": 732419.5605,
      "train_tokens_per_second": 29248.839
    },
    {
      "epoch": 2.6653236239155964,
      "grad_norm": 0.59765625,
      "learning_rate": 1.71167645329035e-05,
      "loss": 0.38428621292114257,
      "num_input_tokens_seen": 21425259456,
      "step": 75330,
      "train_runtime": 732513.224,
      "train_tokens_per_second": 29248.973
    },
    {
      "epoch": 2.665677444179325,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7115761634697715e-05,
      "loss": 0.3917688846588135,
      "num_input_tokens_seen": 21428091328,
      "step": 75340,
      "train_runtime": 732610.3699,
      "train_tokens_per_second": 29248.96
    },
    {
      "epoch": 2.6660312644430535,
      "grad_norm": 0.625,
      "learning_rate": 1.7114758912755432e-05,
      "loss": 0.38284454345703123,
      "num_input_tokens_seen": 21430920960,
      "step": 75350,
      "train_runtime": 732709.8049,
      "train_tokens_per_second": 29248.852
    },
    {
      "epoch": 2.6663850847067825,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7113756367025033e-05,
      "loss": 0.386495566368103,
      "num_input_tokens_seen": 21433734336,
      "step": 75360,
      "train_runtime": 732804.12,
      "train_tokens_per_second": 29248.927
    },
    {
      "epoch": 2.6667389049705115,
      "grad_norm": 0.609375,
      "learning_rate": 1.71127539974549e-05,
      "loss": 0.3837636947631836,
      "num_input_tokens_seen": 21436544064,
      "step": 75370,
      "train_runtime": 732898.6519,
      "train_tokens_per_second": 29248.988
    },
    {
      "epoch": 2.66709272523424,
      "grad_norm": 0.6484375,
      "learning_rate": 1.711175180399346e-05,
      "loss": 0.38471112251281736,
      "num_input_tokens_seen": 21439364224,
      "step": 75380,
      "train_runtime": 732991.7175,
      "train_tokens_per_second": 29249.122
    },
    {
      "epoch": 2.667446545497969,
      "grad_norm": 0.640625,
      "learning_rate": 1.7110749786589146e-05,
      "loss": 0.3879976749420166,
      "num_input_tokens_seen": 21442223744,
      "step": 75390,
      "train_runtime": 733091.8995,
      "train_tokens_per_second": 29249.026
    },
    {
      "epoch": 2.6678003657616975,
      "grad_norm": 0.64453125,
      "learning_rate": 1.7109747945190424e-05,
      "loss": 0.38988304138183594,
      "num_input_tokens_seen": 21445103360,
      "step": 75400,
      "train_runtime": 733193.5064,
      "train_tokens_per_second": 29248.9
    },
    {
      "epoch": 2.6681541860254265,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7108746279745763e-05,
      "loss": 0.3864877223968506,
      "num_input_tokens_seen": 21447920640,
      "step": 75410,
      "train_runtime": 733289.3641,
      "train_tokens_per_second": 29248.918
    },
    {
      "epoch": 2.668508006289155,
      "grad_norm": 0.625,
      "learning_rate": 1.7107744790203676e-05,
      "loss": 0.3839066505432129,
      "num_input_tokens_seen": 21450836416,
      "step": 75420,
      "train_runtime": 733392.1899,
      "train_tokens_per_second": 29248.793
    },
    {
      "epoch": 2.668861826552884,
      "grad_norm": 0.625,
      "learning_rate": 1.7106743476512673e-05,
      "loss": 0.3884526491165161,
      "num_input_tokens_seen": 21453674560,
      "step": 75430,
      "train_runtime": 733487.0433,
      "train_tokens_per_second": 29248.88
    },
    {
      "epoch": 2.6692156468166126,
      "grad_norm": 0.65625,
      "learning_rate": 1.71057423386213e-05,
      "loss": 0.387351393699646,
      "num_input_tokens_seen": 21456513792,
      "step": 75440,
      "train_runtime": 733580.8536,
      "train_tokens_per_second": 29249.01
    },
    {
      "epoch": 2.6695694670803416,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7104741376478137e-05,
      "loss": 0.3889933586120605,
      "num_input_tokens_seen": 21459358976,
      "step": 75450,
      "train_runtime": 733678.4985,
      "train_tokens_per_second": 29248.995
    },
    {
      "epoch": 2.66992328734407,
      "grad_norm": 0.609375,
      "learning_rate": 1.710374059003175e-05,
      "loss": 0.3874448299407959,
      "num_input_tokens_seen": 21462132352,
      "step": 75460,
      "train_runtime": 733772.7353,
      "train_tokens_per_second": 29249.019
    },
    {
      "epoch": 2.670277107607799,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7102739979230753e-05,
      "loss": 0.3866525173187256,
      "num_input_tokens_seen": 21464969152,
      "step": 75470,
      "train_runtime": 733871.4709,
      "train_tokens_per_second": 29248.949
    },
    {
      "epoch": 2.670630927871528,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7101739544023772e-05,
      "loss": 0.38673057556152346,
      "num_input_tokens_seen": 21467774336,
      "step": 75480,
      "train_runtime": 733965.1097,
      "train_tokens_per_second": 29249.039
    },
    {
      "epoch": 2.6709847481352567,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7100739284359464e-05,
      "loss": 0.3853473663330078,
      "num_input_tokens_seen": 21470656000,
      "step": 75490,
      "train_runtime": 734063.6275,
      "train_tokens_per_second": 29249.039
    },
    {
      "epoch": 2.671338568398985,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7099739200186484e-05,
      "loss": 0.38706963062286376,
      "num_input_tokens_seen": 21473521728,
      "step": 75500,
      "train_runtime": 734159.6334,
      "train_tokens_per_second": 29249.118
    },
    {
      "epoch": 2.671692388662714,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7098739291453537e-05,
      "loss": 0.38556981086730957,
      "num_input_tokens_seen": 21476320704,
      "step": 75510,
      "train_runtime": 734254.1466,
      "train_tokens_per_second": 29249.165
    },
    {
      "epoch": 2.672046208926443,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7097739558109325e-05,
      "loss": 0.3881751537322998,
      "num_input_tokens_seen": 21479090432,
      "step": 75520,
      "train_runtime": 734346.9563,
      "train_tokens_per_second": 29249.24
    },
    {
      "epoch": 2.6724000291901717,
      "grad_norm": 0.63671875,
      "learning_rate": 1.709674000010258e-05,
      "loss": 0.38916056156158446,
      "num_input_tokens_seen": 21482007488,
      "step": 75530,
      "train_runtime": 734451.0781,
      "train_tokens_per_second": 29249.065
    },
    {
      "epoch": 2.6727538494539003,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7095740617382063e-05,
      "loss": 0.3858955860137939,
      "num_input_tokens_seen": 21484823488,
      "step": 75540,
      "train_runtime": 734544.1078,
      "train_tokens_per_second": 29249.195
    },
    {
      "epoch": 2.6731076697176293,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7094741409896546e-05,
      "loss": 0.3860058307647705,
      "num_input_tokens_seen": 21487602816,
      "step": 75550,
      "train_runtime": 734637.1399,
      "train_tokens_per_second": 29249.274
    },
    {
      "epoch": 2.6734614899813582,
      "grad_norm": 0.61328125,
      "learning_rate": 1.709374237759482e-05,
      "loss": 0.3865677356719971,
      "num_input_tokens_seen": 21490430208,
      "step": 75560,
      "train_runtime": 734734.4625,
      "train_tokens_per_second": 29249.248
    },
    {
      "epoch": 2.673815310245087,
      "grad_norm": 0.5859375,
      "learning_rate": 1.7092743520425707e-05,
      "loss": 0.3832091808319092,
      "num_input_tokens_seen": 21493338944,
      "step": 75570,
      "train_runtime": 734836.4788,
      "train_tokens_per_second": 29249.145
    },
    {
      "epoch": 2.6741691305088158,
      "grad_norm": 0.609375,
      "learning_rate": 1.709174483833804e-05,
      "loss": 0.39235196113586424,
      "num_input_tokens_seen": 21496185408,
      "step": 75580,
      "train_runtime": 734932.0085,
      "train_tokens_per_second": 29249.216
    },
    {
      "epoch": 2.6745229507725443,
      "grad_norm": 0.625,
      "learning_rate": 1.7090746331280686e-05,
      "loss": 0.3864452838897705,
      "num_input_tokens_seen": 21499051008,
      "step": 75590,
      "train_runtime": 735029.5198,
      "train_tokens_per_second": 29249.235
    },
    {
      "epoch": 2.6748767710362733,
      "grad_norm": 0.60546875,
      "learning_rate": 1.708974799920251e-05,
      "loss": 0.38420858383178713,
      "num_input_tokens_seen": 21501881216,
      "step": 75600,
      "train_runtime": 735126.5475,
      "train_tokens_per_second": 29249.224
    },
    {
      "epoch": 2.675230591300002,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7088749842052426e-05,
      "loss": 0.38848147392272947,
      "num_input_tokens_seen": 21504745920,
      "step": 75610,
      "train_runtime": 735226.1882,
      "train_tokens_per_second": 29249.157
    },
    {
      "epoch": 2.675584411563731,
      "grad_norm": 0.62890625,
      "learning_rate": 1.708775185977935e-05,
      "loss": 0.3859638452529907,
      "num_input_tokens_seen": 21507583360,
      "step": 75620,
      "train_runtime": 735322.6778,
      "train_tokens_per_second": 29249.177
    },
    {
      "epoch": 2.67593823182746,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7086754052332214e-05,
      "loss": 0.38794145584106443,
      "num_input_tokens_seen": 21510452928,
      "step": 75630,
      "train_runtime": 735421.0076,
      "train_tokens_per_second": 29249.168
    },
    {
      "epoch": 2.6762920520911884,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7085756419659995e-05,
      "loss": 0.3890834331512451,
      "num_input_tokens_seen": 21513343104,
      "step": 75640,
      "train_runtime": 735517.6428,
      "train_tokens_per_second": 29249.255
    },
    {
      "epoch": 2.676645872354917,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7084758961711672e-05,
      "loss": 0.38825201988220215,
      "num_input_tokens_seen": 21516211328,
      "step": 75650,
      "train_runtime": 735616.2345,
      "train_tokens_per_second": 29249.234
    },
    {
      "epoch": 2.676999692618646,
      "grad_norm": 0.63671875,
      "learning_rate": 1.708376167843625e-05,
      "loss": 0.385071325302124,
      "num_input_tokens_seen": 21519057920,
      "step": 75660,
      "train_runtime": 735712.5886,
      "train_tokens_per_second": 29249.272
    },
    {
      "epoch": 2.677353512882375,
      "grad_norm": 0.625,
      "learning_rate": 1.708276456978275e-05,
      "loss": 0.3842024803161621,
      "num_input_tokens_seen": 21521913728,
      "step": 75670,
      "train_runtime": 735808.9764,
      "train_tokens_per_second": 29249.322
    },
    {
      "epoch": 2.6777073331461034,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7081767635700222e-05,
      "loss": 0.38241674900054934,
      "num_input_tokens_seen": 21524808512,
      "step": 75680,
      "train_runtime": 735908.1923,
      "train_tokens_per_second": 29249.312
    },
    {
      "epoch": 2.678061153409832,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7080770876137737e-05,
      "loss": 0.3874392509460449,
      "num_input_tokens_seen": 21527647104,
      "step": 75690,
      "train_runtime": 736004.7949,
      "train_tokens_per_second": 29249.33
    },
    {
      "epoch": 2.678414973673561,
      "grad_norm": 0.63671875,
      "learning_rate": 1.707977429104437e-05,
      "loss": 0.3857190847396851,
      "num_input_tokens_seen": 21530510784,
      "step": 75700,
      "train_runtime": 736101.7731,
      "train_tokens_per_second": 29249.367
    },
    {
      "epoch": 2.67876879393729,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7078777880369244e-05,
      "loss": 0.3850836277008057,
      "num_input_tokens_seen": 21533340544,
      "step": 75710,
      "train_runtime": 736195.6083,
      "train_tokens_per_second": 29249.482
    },
    {
      "epoch": 2.6791226142010185,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7077781644061477e-05,
      "loss": 0.3871301651000977,
      "num_input_tokens_seen": 21536198336,
      "step": 75720,
      "train_runtime": 736291.3738,
      "train_tokens_per_second": 29249.559
    },
    {
      "epoch": 2.6794764344647475,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7076785582070228e-05,
      "loss": 0.38590035438537595,
      "num_input_tokens_seen": 21539043904,
      "step": 75730,
      "train_runtime": 736388.1521,
      "train_tokens_per_second": 29249.58
    },
    {
      "epoch": 2.679830254728476,
      "grad_norm": 0.59375,
      "learning_rate": 1.7075789694344664e-05,
      "loss": 0.38796091079711914,
      "num_input_tokens_seen": 21541908672,
      "step": 75740,
      "train_runtime": 736485.0552,
      "train_tokens_per_second": 29249.621
    },
    {
      "epoch": 2.680184074992205,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7074793980833974e-05,
      "loss": 0.38872604370117186,
      "num_input_tokens_seen": 21544763392,
      "step": 75750,
      "train_runtime": 736584.0186,
      "train_tokens_per_second": 29249.567
    },
    {
      "epoch": 2.6805378952559336,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7073798441487372e-05,
      "loss": 0.3832289934158325,
      "num_input_tokens_seen": 21547567424,
      "step": 75760,
      "train_runtime": 736677.5667,
      "train_tokens_per_second": 29249.659
    },
    {
      "epoch": 2.6808917155196625,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7072803076254092e-05,
      "loss": 0.3870734214782715,
      "num_input_tokens_seen": 21550345856,
      "step": 75770,
      "train_runtime": 736769.8855,
      "train_tokens_per_second": 29249.765
    },
    {
      "epoch": 2.681245535783391,
      "grad_norm": 0.625,
      "learning_rate": 1.707180788508339e-05,
      "loss": 0.38754122257232665,
      "num_input_tokens_seen": 21553202816,
      "step": 75780,
      "train_runtime": 736866.9507,
      "train_tokens_per_second": 29249.789
    },
    {
      "epoch": 2.68159935604712,
      "grad_norm": 0.60546875,
      "learning_rate": 1.707081286792454e-05,
      "loss": 0.38608517646789553,
      "num_input_tokens_seen": 21556086656,
      "step": 75790,
      "train_runtime": 736966.306,
      "train_tokens_per_second": 29249.759
    },
    {
      "epoch": 2.6819531763108486,
      "grad_norm": 0.64453125,
      "learning_rate": 1.7069818024726833e-05,
      "loss": 0.38488450050354006,
      "num_input_tokens_seen": 21558929152,
      "step": 75800,
      "train_runtime": 737063.7975,
      "train_tokens_per_second": 29249.746
    },
    {
      "epoch": 2.6823069965745776,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7068823355439583e-05,
      "loss": 0.38973703384399416,
      "num_input_tokens_seen": 21561794176,
      "step": 75810,
      "train_runtime": 737159.3135,
      "train_tokens_per_second": 29249.843
    },
    {
      "epoch": 2.6826608168383066,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7067828860012138e-05,
      "loss": 0.3831465244293213,
      "num_input_tokens_seen": 21564658624,
      "step": 75820,
      "train_runtime": 737259.1402,
      "train_tokens_per_second": 29249.768
    },
    {
      "epoch": 2.683014637102035,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7066834538393844e-05,
      "loss": 0.38516356945037844,
      "num_input_tokens_seen": 21567477376,
      "step": 75830,
      "train_runtime": 737356.0075,
      "train_tokens_per_second": 29249.748
    },
    {
      "epoch": 2.6833684573657637,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7065840390534083e-05,
      "loss": 0.3875086307525635,
      "num_input_tokens_seen": 21570292608,
      "step": 75840,
      "train_runtime": 737452.2396,
      "train_tokens_per_second": 29249.749
    },
    {
      "epoch": 2.6837222776294927,
      "grad_norm": 0.6484375,
      "learning_rate": 1.7064846416382256e-05,
      "loss": 0.38725829124450684,
      "num_input_tokens_seen": 21573104320,
      "step": 75850,
      "train_runtime": 737547.3803,
      "train_tokens_per_second": 29249.788
    },
    {
      "epoch": 2.6840760978932217,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7063852615887773e-05,
      "loss": 0.38848841190338135,
      "num_input_tokens_seen": 21575921984,
      "step": 75860,
      "train_runtime": 737642.3321,
      "train_tokens_per_second": 29249.842
    },
    {
      "epoch": 2.68442991815695,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7062858989000085e-05,
      "loss": 0.3879056930541992,
      "num_input_tokens_seen": 21578773568,
      "step": 75870,
      "train_runtime": 737740.1067,
      "train_tokens_per_second": 29249.831
    },
    {
      "epoch": 2.684783738420679,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7061865535668645e-05,
      "loss": 0.38703012466430664,
      "num_input_tokens_seen": 21581601600,
      "step": 75880,
      "train_runtime": 737834.5643,
      "train_tokens_per_second": 29249.919
    },
    {
      "epoch": 2.6851375586844077,
      "grad_norm": 0.609375,
      "learning_rate": 1.706087225584294e-05,
      "loss": 0.38229732513427733,
      "num_input_tokens_seen": 21584469952,
      "step": 75890,
      "train_runtime": 737933.8026,
      "train_tokens_per_second": 29249.873
    },
    {
      "epoch": 2.6854913789481367,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7059879149472464e-05,
      "loss": 0.38459124565124514,
      "num_input_tokens_seen": 21587296640,
      "step": 75900,
      "train_runtime": 738029.3646,
      "train_tokens_per_second": 29249.916
    },
    {
      "epoch": 2.6858451992118653,
      "grad_norm": 0.5859375,
      "learning_rate": 1.7058886216506746e-05,
      "loss": 0.38746132850646975,
      "num_input_tokens_seen": 21590111808,
      "step": 75910,
      "train_runtime": 738122.8953,
      "train_tokens_per_second": 29250.023
    },
    {
      "epoch": 2.6861990194755943,
      "grad_norm": 0.609375,
      "learning_rate": 1.705789345689532e-05,
      "loss": 0.38462421894073484,
      "num_input_tokens_seen": 21592972352,
      "step": 75920,
      "train_runtime": 738221.9809,
      "train_tokens_per_second": 29249.972
    },
    {
      "epoch": 2.686552839739323,
      "grad_norm": 0.63671875,
      "learning_rate": 1.705690087058776e-05,
      "loss": 0.38400309085845946,
      "num_input_tokens_seen": 21595827328,
      "step": 75930,
      "train_runtime": 738319.6262,
      "train_tokens_per_second": 29249.971
    },
    {
      "epoch": 2.686906660003052,
      "grad_norm": 0.6484375,
      "learning_rate": 1.7055908457533644e-05,
      "loss": 0.39088256359100343,
      "num_input_tokens_seen": 21598685184,
      "step": 75940,
      "train_runtime": 738419.7646,
      "train_tokens_per_second": 29249.874
    },
    {
      "epoch": 2.6872604802667803,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7054916217682577e-05,
      "loss": 0.386277437210083,
      "num_input_tokens_seen": 21601495360,
      "step": 75950,
      "train_runtime": 738516.6758,
      "train_tokens_per_second": 29249.841
    },
    {
      "epoch": 2.6876143005305093,
      "grad_norm": 0.640625,
      "learning_rate": 1.7053924150984184e-05,
      "loss": 0.38320021629333495,
      "num_input_tokens_seen": 21604312000,
      "step": 75960,
      "train_runtime": 738613.4068,
      "train_tokens_per_second": 29249.824
    },
    {
      "epoch": 2.6879681207942383,
      "grad_norm": 0.609375,
      "learning_rate": 1.7052932257388107e-05,
      "loss": 0.3878506660461426,
      "num_input_tokens_seen": 21607158528,
      "step": 75970,
      "train_runtime": 738710.598,
      "train_tokens_per_second": 29249.829
    },
    {
      "epoch": 2.688321941057967,
      "grad_norm": 0.64453125,
      "learning_rate": 1.7051940536844015e-05,
      "loss": 0.3887681484222412,
      "num_input_tokens_seen": 21609980864,
      "step": 75980,
      "train_runtime": 738806.5576,
      "train_tokens_per_second": 29249.85
    },
    {
      "epoch": 2.6886757613216954,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7050948989301595e-05,
      "loss": 0.3843262195587158,
      "num_input_tokens_seen": 21612790976,
      "step": 75990,
      "train_runtime": 738903.3475,
      "train_tokens_per_second": 29249.821
    },
    {
      "epoch": 2.6890295815854244,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7049957614710553e-05,
      "loss": 0.38376855850219727,
      "num_input_tokens_seen": 21615594560,
      "step": 76000,
      "train_runtime": 738997.8071,
      "train_tokens_per_second": 29249.876
    },
    {
      "epoch": 2.6893834018491534,
      "grad_norm": 0.60546875,
      "learning_rate": 1.7048966413020613e-05,
      "loss": 0.38604605197906494,
      "num_input_tokens_seen": 21618415616,
      "step": 76010,
      "train_runtime": 739094.7889,
      "train_tokens_per_second": 29249.855
    },
    {
      "epoch": 2.689737222112882,
      "grad_norm": 0.62109375,
      "learning_rate": 1.704797538418153e-05,
      "loss": 0.38668999671936033,
      "num_input_tokens_seen": 21621306880,
      "step": 76020,
      "train_runtime": 739192.3123,
      "train_tokens_per_second": 29249.908
    },
    {
      "epoch": 2.6900910423766105,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7046984528143065e-05,
      "loss": 0.38678865432739257,
      "num_input_tokens_seen": 21624121792,
      "step": 76030,
      "train_runtime": 739286.5256,
      "train_tokens_per_second": 29249.988
    },
    {
      "epoch": 2.6904448626403394,
      "grad_norm": 0.609375,
      "learning_rate": 1.7045993844855012e-05,
      "loss": 0.385788106918335,
      "num_input_tokens_seen": 21626982592,
      "step": 76040,
      "train_runtime": 739382.4385,
      "train_tokens_per_second": 29250.063
    },
    {
      "epoch": 2.6907986829040684,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7045003334267175e-05,
      "loss": 0.3811150550842285,
      "num_input_tokens_seen": 21629820416,
      "step": 76050,
      "train_runtime": 739480.0209,
      "train_tokens_per_second": 29250.04
    },
    {
      "epoch": 2.691152503167797,
      "grad_norm": 0.6328125,
      "learning_rate": 1.704401299632939e-05,
      "loss": 0.38633582592010496,
      "num_input_tokens_seen": 21632654976,
      "step": 76060,
      "train_runtime": 739577.6041,
      "train_tokens_per_second": 29250.014
    },
    {
      "epoch": 2.691506323431526,
      "grad_norm": 0.64453125,
      "learning_rate": 1.7043022830991498e-05,
      "loss": 0.379053521156311,
      "num_input_tokens_seen": 21635567616,
      "step": 76070,
      "train_runtime": 739678.8977,
      "train_tokens_per_second": 29249.946
    },
    {
      "epoch": 2.6918601436952545,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7042032838203372e-05,
      "loss": 0.3882865905761719,
      "num_input_tokens_seen": 21638366592,
      "step": 76080,
      "train_runtime": 739774.4354,
      "train_tokens_per_second": 29249.952
    },
    {
      "epoch": 2.6922139639589835,
      "grad_norm": 0.609375,
      "learning_rate": 1.7041043017914912e-05,
      "loss": 0.3874474048614502,
      "num_input_tokens_seen": 21641229952,
      "step": 76090,
      "train_runtime": 739872.5146,
      "train_tokens_per_second": 29249.944
    },
    {
      "epoch": 2.692567784222712,
      "grad_norm": 0.625,
      "learning_rate": 1.7040053370076014e-05,
      "loss": 0.38727507591247556,
      "num_input_tokens_seen": 21644022464,
      "step": 76100,
      "train_runtime": 739968.0229,
      "train_tokens_per_second": 29249.943
    },
    {
      "epoch": 2.692921604486441,
      "grad_norm": 0.62109375,
      "learning_rate": 1.703906389463662e-05,
      "loss": 0.3891221284866333,
      "num_input_tokens_seen": 21646869952,
      "step": 76110,
      "train_runtime": 740066.1051,
      "train_tokens_per_second": 29249.914
    },
    {
      "epoch": 2.6932754247501696,
      "grad_norm": 0.6328125,
      "learning_rate": 1.703807459154668e-05,
      "loss": 0.38599109649658203,
      "num_input_tokens_seen": 21649738432,
      "step": 76120,
      "train_runtime": 740166.7922,
      "train_tokens_per_second": 29249.811
    },
    {
      "epoch": 2.6936292450138986,
      "grad_norm": 0.625,
      "learning_rate": 1.703708546075616e-05,
      "loss": 0.388598108291626,
      "num_input_tokens_seen": 21652608960,
      "step": 76130,
      "train_runtime": 740265.112,
      "train_tokens_per_second": 29249.803
    },
    {
      "epoch": 2.693983065277627,
      "grad_norm": 0.6171875,
      "learning_rate": 1.7036096502215064e-05,
      "loss": 0.385296106338501,
      "num_input_tokens_seen": 21655470272,
      "step": 76140,
      "train_runtime": 740361.3112,
      "train_tokens_per_second": 29249.868
    },
    {
      "epoch": 2.694336885541356,
      "grad_norm": 0.625,
      "learning_rate": 1.703510771587339e-05,
      "loss": 0.3901038646697998,
      "num_input_tokens_seen": 21658332672,
      "step": 76150,
      "train_runtime": 740456.8869,
      "train_tokens_per_second": 29249.958
    },
    {
      "epoch": 2.694690705805085,
      "grad_norm": 0.61328125,
      "learning_rate": 1.703411910168118e-05,
      "loss": 0.38502204418182373,
      "num_input_tokens_seen": 21661167552,
      "step": 76160,
      "train_runtime": 740551.2079,
      "train_tokens_per_second": 29250.06
    },
    {
      "epoch": 2.6950445260688136,
      "grad_norm": 0.640625,
      "learning_rate": 1.7033130659588486e-05,
      "loss": 0.3832906007766724,
      "num_input_tokens_seen": 21663979008,
      "step": 76170,
      "train_runtime": 740645.8342,
      "train_tokens_per_second": 29250.119
    },
    {
      "epoch": 2.695398346332542,
      "grad_norm": 0.60546875,
      "learning_rate": 1.703214238954538e-05,
      "loss": 0.38453123569488523,
      "num_input_tokens_seen": 21666785984,
      "step": 76180,
      "train_runtime": 740739.2266,
      "train_tokens_per_second": 29250.221
    },
    {
      "epoch": 2.695752166596271,
      "grad_norm": 0.59375,
      "learning_rate": 1.7031154291501962e-05,
      "loss": 0.3841399192810059,
      "num_input_tokens_seen": 21669590656,
      "step": 76190,
      "train_runtime": 740833.1431,
      "train_tokens_per_second": 29250.299
    },
    {
      "epoch": 2.69610598686,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7030166365408335e-05,
      "loss": 0.39055495262145995,
      "num_input_tokens_seen": 21672408704,
      "step": 76200,
      "train_runtime": 740927.035,
      "train_tokens_per_second": 29250.395
    },
    {
      "epoch": 2.6964598071237287,
      "grad_norm": 0.62109375,
      "learning_rate": 1.702917861121464e-05,
      "loss": 0.3875312328338623,
      "num_input_tokens_seen": 21675227200,
      "step": 76210,
      "train_runtime": 741020.3775,
      "train_tokens_per_second": 29250.514
    },
    {
      "epoch": 2.6968136273874577,
      "grad_norm": 0.63671875,
      "learning_rate": 1.7028191028871033e-05,
      "loss": 0.38333899974823,
      "num_input_tokens_seen": 21678019072,
      "step": 76220,
      "train_runtime": 741113.9961,
      "train_tokens_per_second": 29250.587
    },
    {
      "epoch": 2.697167447651186,
      "grad_norm": 0.640625,
      "learning_rate": 1.7027203618327685e-05,
      "loss": 0.3813507080078125,
      "num_input_tokens_seen": 21680876416,
      "step": 76230,
      "train_runtime": 741213.6737,
      "train_tokens_per_second": 29250.508
    },
    {
      "epoch": 2.697521267914915,
      "grad_norm": 0.6796875,
      "learning_rate": 1.7026216379534793e-05,
      "loss": 0.3822087049484253,
      "num_input_tokens_seen": 21683763584,
      "step": 76240,
      "train_runtime": 741315.0653,
      "train_tokens_per_second": 29250.402
    },
    {
      "epoch": 2.6978750881786437,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7025229312442577e-05,
      "loss": 0.38779416084289553,
      "num_input_tokens_seen": 21686638080,
      "step": 76250,
      "train_runtime": 741414.8495,
      "train_tokens_per_second": 29250.342
    },
    {
      "epoch": 2.6982289084423727,
      "grad_norm": 0.60546875,
      "learning_rate": 1.702424241700126e-05,
      "loss": 0.3850146770477295,
      "num_input_tokens_seen": 21689497024,
      "step": 76260,
      "train_runtime": 741512.6553,
      "train_tokens_per_second": 29250.34
    },
    {
      "epoch": 2.6985827287061013,
      "grad_norm": 0.58984375,
      "learning_rate": 1.702325569316111e-05,
      "loss": 0.3861421585083008,
      "num_input_tokens_seen": 21692301504,
      "step": 76270,
      "train_runtime": 741607.6974,
      "train_tokens_per_second": 29250.373
    },
    {
      "epoch": 2.6989365489698303,
      "grad_norm": 0.640625,
      "learning_rate": 1.702226914087239e-05,
      "loss": 0.38536500930786133,
      "num_input_tokens_seen": 21695201088,
      "step": 76280,
      "train_runtime": 741708.8022,
      "train_tokens_per_second": 29250.295
    },
    {
      "epoch": 2.699290369233559,
      "grad_norm": 0.65625,
      "learning_rate": 1.7021282760085415e-05,
      "loss": 0.38678975105285646,
      "num_input_tokens_seen": 21698053248,
      "step": 76290,
      "train_runtime": 741810.9373,
      "train_tokens_per_second": 29250.112
    },
    {
      "epoch": 2.699644189497288,
      "grad_norm": 0.59765625,
      "learning_rate": 1.7020296550750485e-05,
      "loss": 0.38616104125976564,
      "num_input_tokens_seen": 21700865984,
      "step": 76300,
      "train_runtime": 741907.1691,
      "train_tokens_per_second": 29250.11
    },
    {
      "epoch": 2.699998009761017,
      "grad_norm": 0.625,
      "learning_rate": 1.7019310512817936e-05,
      "loss": 0.38514184951782227,
      "num_input_tokens_seen": 21703666688,
      "step": 76310,
      "train_runtime": 742002.9235,
      "train_tokens_per_second": 29250.109
    },
    {
      "epoch": 2.7003518300247453,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7018324646238134e-05,
      "loss": 0.38914012908935547,
      "num_input_tokens_seen": 21706505600,
      "step": 76320,
      "train_runtime": 742100.5754,
      "train_tokens_per_second": 29250.086
    },
    {
      "epoch": 2.700705650288474,
      "grad_norm": 0.5859375,
      "learning_rate": 1.7017338950961455e-05,
      "loss": 0.3854983329772949,
      "num_input_tokens_seen": 21709358976,
      "step": 76330,
      "train_runtime": 742198.521,
      "train_tokens_per_second": 29250.07
    },
    {
      "epoch": 2.701059470552203,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7016353426938287e-05,
      "loss": 0.3853592872619629,
      "num_input_tokens_seen": 21712215616,
      "step": 76340,
      "train_runtime": 742297.553,
      "train_tokens_per_second": 29250.016
    },
    {
      "epoch": 2.701413290815932,
      "grad_norm": 0.63671875,
      "learning_rate": 1.701536807411905e-05,
      "loss": 0.38556020259857177,
      "num_input_tokens_seen": 21715044352,
      "step": 76350,
      "train_runtime": 742394.2885,
      "train_tokens_per_second": 29250.015
    },
    {
      "epoch": 2.7017671110796604,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7014382892454186e-05,
      "loss": 0.39275150299072265,
      "num_input_tokens_seen": 21717864448,
      "step": 76360,
      "train_runtime": 742488.8151,
      "train_tokens_per_second": 29250.09
    },
    {
      "epoch": 2.702120931343389,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7013397881894147e-05,
      "loss": 0.386823034286499,
      "num_input_tokens_seen": 21720737600,
      "step": 76370,
      "train_runtime": 742587.7526,
      "train_tokens_per_second": 29250.062
    },
    {
      "epoch": 2.702474751607118,
      "grad_norm": 0.6328125,
      "learning_rate": 1.701241304238941e-05,
      "loss": 0.3873495578765869,
      "num_input_tokens_seen": 21723578752,
      "step": 76380,
      "train_runtime": 742688.2047,
      "train_tokens_per_second": 29249.931
    },
    {
      "epoch": 2.702828571870847,
      "grad_norm": 0.625,
      "learning_rate": 1.7011428373890478e-05,
      "loss": 0.3863509178161621,
      "num_input_tokens_seen": 21726396992,
      "step": 76390,
      "train_runtime": 742784.1663,
      "train_tokens_per_second": 29249.946
    },
    {
      "epoch": 2.7031823921345755,
      "grad_norm": 0.640625,
      "learning_rate": 1.7010443876347863e-05,
      "loss": 0.3863941431045532,
      "num_input_tokens_seen": 21729171008,
      "step": 76400,
      "train_runtime": 742877.763,
      "train_tokens_per_second": 29249.995
    },
    {
      "epoch": 2.7035362123983044,
      "grad_norm": 0.6328125,
      "learning_rate": 1.70094595497121e-05,
      "loss": 0.38391761779785155,
      "num_input_tokens_seen": 21731987520,
      "step": 76410,
      "train_runtime": 742974.2451,
      "train_tokens_per_second": 29249.988
    },
    {
      "epoch": 2.703890032662033,
      "grad_norm": 0.60546875,
      "learning_rate": 1.700847539393375e-05,
      "loss": 0.39025249481201174,
      "num_input_tokens_seen": 21734839360,
      "step": 76420,
      "train_runtime": 743074.4538,
      "train_tokens_per_second": 29249.881
    },
    {
      "epoch": 2.704243852925762,
      "grad_norm": 0.61328125,
      "learning_rate": 1.7007491408963392e-05,
      "loss": 0.3919255256652832,
      "num_input_tokens_seen": 21737673280,
      "step": 76430,
      "train_runtime": 743170.6731,
      "train_tokens_per_second": 29249.907
    },
    {
      "epoch": 2.7045976731894905,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7006507594751617e-05,
      "loss": 0.3894176959991455,
      "num_input_tokens_seen": 21740469248,
      "step": 76440,
      "train_runtime": 743264.5521,
      "train_tokens_per_second": 29249.975
    },
    {
      "epoch": 2.7049514934532195,
      "grad_norm": 0.62890625,
      "learning_rate": 1.7005523951249047e-05,
      "loss": 0.3877555847167969,
      "num_input_tokens_seen": 21743335360,
      "step": 76450,
      "train_runtime": 743361.1243,
      "train_tokens_per_second": 29250.03
    },
    {
      "epoch": 2.705305313716948,
      "grad_norm": 0.62890625,
      "learning_rate": 1.700454047840632e-05,
      "loss": 0.3856799125671387,
      "num_input_tokens_seen": 21746228608,
      "step": 76460,
      "train_runtime": 743463.7112,
      "train_tokens_per_second": 29249.886
    },
    {
      "epoch": 2.705659133980677,
      "grad_norm": 0.62109375,
      "learning_rate": 1.700355717617409e-05,
      "loss": 0.3885051250457764,
      "num_input_tokens_seen": 21749065088,
      "step": 76470,
      "train_runtime": 743561.4273,
      "train_tokens_per_second": 29249.857
    },
    {
      "epoch": 2.7060129542444056,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7002574044503038e-05,
      "loss": 0.3870185613632202,
      "num_input_tokens_seen": 21751911296,
      "step": 76480,
      "train_runtime": 743659.804,
      "train_tokens_per_second": 29249.814
    },
    {
      "epoch": 2.7063667745081346,
      "grad_norm": 0.6328125,
      "learning_rate": 1.7001591083343853e-05,
      "loss": 0.38483624458312987,
      "num_input_tokens_seen": 21754742208,
      "step": 76490,
      "train_runtime": 743757.1695,
      "train_tokens_per_second": 29249.792
    },
    {
      "epoch": 2.7067205947718636,
      "grad_norm": 0.61328125,
      "learning_rate": 1.700060829264726e-05,
      "loss": 0.38583059310913087,
      "num_input_tokens_seen": 21757582848,
      "step": 76500,
      "train_runtime": 743853.5949,
      "train_tokens_per_second": 29249.819
    },
    {
      "epoch": 2.707074415035592,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6999625672363997e-05,
      "loss": 0.38498797416687014,
      "num_input_tokens_seen": 21760409728,
      "step": 76510,
      "train_runtime": 743950.4095,
      "train_tokens_per_second": 29249.812
    },
    {
      "epoch": 2.7074282352993206,
      "grad_norm": 0.6328125,
      "learning_rate": 1.699864322244482e-05,
      "loss": 0.38719921112060546,
      "num_input_tokens_seen": 21763204672,
      "step": 76520,
      "train_runtime": 744045.2425,
      "train_tokens_per_second": 29249.841
    },
    {
      "epoch": 2.7077820555630496,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6997660942840497e-05,
      "loss": 0.38466825485229494,
      "num_input_tokens_seen": 21766003712,
      "step": 76530,
      "train_runtime": 744137.7869,
      "train_tokens_per_second": 29249.964
    },
    {
      "epoch": 2.7081358758267786,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6996678833501832e-05,
      "loss": 0.38578386306762696,
      "num_input_tokens_seen": 21768833600,
      "step": 76540,
      "train_runtime": 744234.5362,
      "train_tokens_per_second": 29249.964
    },
    {
      "epoch": 2.708489696090507,
      "grad_norm": 0.62109375,
      "learning_rate": 1.699569689437965e-05,
      "loss": 0.38596031665802,
      "num_input_tokens_seen": 21771657984,
      "step": 76550,
      "train_runtime": 744330.6372,
      "train_tokens_per_second": 29249.982
    },
    {
      "epoch": 2.708843516354236,
      "grad_norm": 0.58984375,
      "learning_rate": 1.699471512542477e-05,
      "loss": 0.38004016876220703,
      "num_input_tokens_seen": 21774478912,
      "step": 76560,
      "train_runtime": 744427.5824,
      "train_tokens_per_second": 29249.963
    },
    {
      "epoch": 2.7091973366179647,
      "grad_norm": 0.66796875,
      "learning_rate": 1.6993733526588064e-05,
      "loss": 0.38650059700012207,
      "num_input_tokens_seen": 21777322304,
      "step": 76570,
      "train_runtime": 744524.8378,
      "train_tokens_per_second": 29249.961
    },
    {
      "epoch": 2.7095511568816937,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6992752097820396e-05,
      "loss": 0.38565430641174314,
      "num_input_tokens_seen": 21780193088,
      "step": 76580,
      "train_runtime": 744623.2876,
      "train_tokens_per_second": 29249.949
    },
    {
      "epoch": 2.7099049771454222,
      "grad_norm": 0.609375,
      "learning_rate": 1.699177083907267e-05,
      "loss": 0.3872755765914917,
      "num_input_tokens_seen": 21783066048,
      "step": 76590,
      "train_runtime": 744720.8747,
      "train_tokens_per_second": 29249.974
    },
    {
      "epoch": 2.710258797409151,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6990789750295798e-05,
      "loss": 0.3843536853790283,
      "num_input_tokens_seen": 21785906688,
      "step": 76600,
      "train_runtime": 744817.0564,
      "train_tokens_per_second": 29250.01
    },
    {
      "epoch": 2.7106126176728798,
      "grad_norm": 0.640625,
      "learning_rate": 1.6989808831440724e-05,
      "loss": 0.3836094856262207,
      "num_input_tokens_seen": 21788675264,
      "step": 76610,
      "train_runtime": 744910.1321,
      "train_tokens_per_second": 29250.072
    },
    {
      "epoch": 2.7109664379366087,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6988828082458395e-05,
      "loss": 0.38330371379852296,
      "num_input_tokens_seen": 21791558400,
      "step": 76620,
      "train_runtime": 745008.4164,
      "train_tokens_per_second": 29250.084
    },
    {
      "epoch": 2.7113202582003373,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6987847503299786e-05,
      "loss": 0.3911462306976318,
      "num_input_tokens_seen": 21794476480,
      "step": 76630,
      "train_runtime": 745109.581,
      "train_tokens_per_second": 29250.029
    },
    {
      "epoch": 2.7116740784640663,
      "grad_norm": 0.625,
      "learning_rate": 1.6986867093915897e-05,
      "loss": 0.3838812828063965,
      "num_input_tokens_seen": 21797305600,
      "step": 76640,
      "train_runtime": 745204.5384,
      "train_tokens_per_second": 29250.098
    },
    {
      "epoch": 2.7120278987277953,
      "grad_norm": 0.625,
      "learning_rate": 1.6985886854257746e-05,
      "loss": 0.38353514671325684,
      "num_input_tokens_seen": 21800140032,
      "step": 76650,
      "train_runtime": 745301.4313,
      "train_tokens_per_second": 29250.098
    },
    {
      "epoch": 2.712381718991524,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6984906784276358e-05,
      "loss": 0.38346195220947266,
      "num_input_tokens_seen": 21802975936,
      "step": 76660,
      "train_runtime": 745397.5088,
      "train_tokens_per_second": 29250.133
    },
    {
      "epoch": 2.7127355392552523,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6983926883922794e-05,
      "loss": 0.3906045913696289,
      "num_input_tokens_seen": 21805830272,
      "step": 76670,
      "train_runtime": 745493.0547,
      "train_tokens_per_second": 29250.213
    },
    {
      "epoch": 2.7130893595189813,
      "grad_norm": 0.62890625,
      "learning_rate": 1.698294715314813e-05,
      "loss": 0.3925213575363159,
      "num_input_tokens_seen": 21808682048,
      "step": 76680,
      "train_runtime": 745592.4224,
      "train_tokens_per_second": 29250.139
    },
    {
      "epoch": 2.7134431797827103,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6981967591903456e-05,
      "loss": 0.3881517887115479,
      "num_input_tokens_seen": 21811552832,
      "step": 76690,
      "train_runtime": 745691.328,
      "train_tokens_per_second": 29250.109
    },
    {
      "epoch": 2.713797000046439,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6980988200139892e-05,
      "loss": 0.3868730068206787,
      "num_input_tokens_seen": 21814358976,
      "step": 76700,
      "train_runtime": 745783.3234,
      "train_tokens_per_second": 29250.264
    },
    {
      "epoch": 2.7141508203101674,
      "grad_norm": 0.6484375,
      "learning_rate": 1.698000897780856e-05,
      "loss": 0.3928858757019043,
      "num_input_tokens_seen": 21817212736,
      "step": 76710,
      "train_runtime": 745881.1323,
      "train_tokens_per_second": 29250.254
    },
    {
      "epoch": 2.7145046405738964,
      "grad_norm": 0.625,
      "learning_rate": 1.6979029924860625e-05,
      "loss": 0.3907364845275879,
      "num_input_tokens_seen": 21820037248,
      "step": 76720,
      "train_runtime": 745977.6706,
      "train_tokens_per_second": 29250.255
    },
    {
      "epoch": 2.7148584608376254,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6978051041247257e-05,
      "loss": 0.3829702854156494,
      "num_input_tokens_seen": 21822866944,
      "step": 76730,
      "train_runtime": 746072.5581,
      "train_tokens_per_second": 29250.328
    },
    {
      "epoch": 2.715212281101354,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6977072326919648e-05,
      "loss": 0.3784936904907227,
      "num_input_tokens_seen": 21825739008,
      "step": 76740,
      "train_runtime": 746171.6832,
      "train_tokens_per_second": 29250.291
    },
    {
      "epoch": 2.715566101365083,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6976093781829007e-05,
      "loss": 0.38403844833374023,
      "num_input_tokens_seen": 21828614784,
      "step": 76750,
      "train_runtime": 746271.5175,
      "train_tokens_per_second": 29250.232
    },
    {
      "epoch": 2.7159199216288115,
      "grad_norm": 0.609375,
      "learning_rate": 1.697511540592657e-05,
      "loss": 0.3903098821640015,
      "num_input_tokens_seen": 21831440576,
      "step": 76760,
      "train_runtime": 746370.6037,
      "train_tokens_per_second": 29250.135
    },
    {
      "epoch": 2.7162737418925404,
      "grad_norm": 0.59375,
      "learning_rate": 1.6974137199163594e-05,
      "loss": 0.3891082763671875,
      "num_input_tokens_seen": 21834287296,
      "step": 76770,
      "train_runtime": 746468.3147,
      "train_tokens_per_second": 29250.119
    },
    {
      "epoch": 2.716627562156269,
      "grad_norm": 0.625,
      "learning_rate": 1.697315916149134e-05,
      "loss": 0.38273158073425295,
      "num_input_tokens_seen": 21837165440,
      "step": 76780,
      "train_runtime": 746569.9871,
      "train_tokens_per_second": 29249.991
    },
    {
      "epoch": 2.716981382419998,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6972181292861104e-05,
      "loss": 0.38473920822143554,
      "num_input_tokens_seen": 21839997568,
      "step": 76790,
      "train_runtime": 746666.3219,
      "train_tokens_per_second": 29250.01
    },
    {
      "epoch": 2.717335202683727,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6971203593224196e-05,
      "loss": 0.38906278610229494,
      "num_input_tokens_seen": 21842827200,
      "step": 76800,
      "train_runtime": 746761.8053,
      "train_tokens_per_second": 29250.059
    },
    {
      "epoch": 2.7176890229474555,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6970226062531953e-05,
      "loss": 0.3868934869766235,
      "num_input_tokens_seen": 21845663168,
      "step": 76810,
      "train_runtime": 746856.0429,
      "train_tokens_per_second": 29250.166
    },
    {
      "epoch": 2.718042843211184,
      "grad_norm": 0.609375,
      "learning_rate": 1.6969248700735715e-05,
      "loss": 0.3909526109695435,
      "num_input_tokens_seen": 21848520832,
      "step": 76820,
      "train_runtime": 746956.4496,
      "train_tokens_per_second": 29250.06
    },
    {
      "epoch": 2.718396663474913,
      "grad_norm": 0.625,
      "learning_rate": 1.6968271507786855e-05,
      "loss": 0.38682026863098146,
      "num_input_tokens_seen": 21851369344,
      "step": 76830,
      "train_runtime": 747055.7265,
      "train_tokens_per_second": 29249.986
    },
    {
      "epoch": 2.718750483738642,
      "grad_norm": 0.58203125,
      "learning_rate": 1.696729448363677e-05,
      "loss": 0.3850741386413574,
      "num_input_tokens_seen": 21854241984,
      "step": 76840,
      "train_runtime": 747152.8527,
      "train_tokens_per_second": 29250.028
    },
    {
      "epoch": 2.7191043040023706,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6966317628236856e-05,
      "loss": 0.3836874008178711,
      "num_input_tokens_seen": 21857128640,
      "step": 76850,
      "train_runtime": 747251.3663,
      "train_tokens_per_second": 29250.035
    },
    {
      "epoch": 2.719458124266099,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6965340941538545e-05,
      "loss": 0.3817173957824707,
      "num_input_tokens_seen": 21860015104,
      "step": 76860,
      "train_runtime": 747350.8908,
      "train_tokens_per_second": 29250.002
    },
    {
      "epoch": 2.719811944529828,
      "grad_norm": 0.58203125,
      "learning_rate": 1.6964364423493297e-05,
      "loss": 0.38294711112976076,
      "num_input_tokens_seen": 21862857920,
      "step": 76870,
      "train_runtime": 747448.484,
      "train_tokens_per_second": 29249.986
    },
    {
      "epoch": 2.720165764793557,
      "grad_norm": 0.640625,
      "learning_rate": 1.6963388074052565e-05,
      "loss": 0.3841033220291138,
      "num_input_tokens_seen": 21865689920,
      "step": 76880,
      "train_runtime": 747545.2345,
      "train_tokens_per_second": 29249.989
    },
    {
      "epoch": 2.7205195850572856,
      "grad_norm": 0.609375,
      "learning_rate": 1.696241189316784e-05,
      "loss": 0.38262276649475097,
      "num_input_tokens_seen": 21868568512,
      "step": 76890,
      "train_runtime": 747644.7313,
      "train_tokens_per_second": 29249.947
    },
    {
      "epoch": 2.7208734053210146,
      "grad_norm": 0.60546875,
      "learning_rate": 1.696143588079063e-05,
      "loss": 0.3831918239593506,
      "num_input_tokens_seen": 21871420864,
      "step": 76900,
      "train_runtime": 747744.3292,
      "train_tokens_per_second": 29249.865
    },
    {
      "epoch": 2.721227225584743,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6960460036872468e-05,
      "loss": 0.3868118762969971,
      "num_input_tokens_seen": 21874252096,
      "step": 76910,
      "train_runtime": 747838.2979,
      "train_tokens_per_second": 29249.976
    },
    {
      "epoch": 2.721581045848472,
      "grad_norm": 0.58984375,
      "learning_rate": 1.6959484361364888e-05,
      "loss": 0.3924058437347412,
      "num_input_tokens_seen": 21877093760,
      "step": 76920,
      "train_runtime": 747933.5293,
      "train_tokens_per_second": 29250.051
    },
    {
      "epoch": 2.7219348661122007,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6958508854219458e-05,
      "loss": 0.3860179901123047,
      "num_input_tokens_seen": 21879918656,
      "step": 76930,
      "train_runtime": 748030.5029,
      "train_tokens_per_second": 29250.035
    },
    {
      "epoch": 2.7222886863759297,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6957533515387773e-05,
      "loss": 0.3884253025054932,
      "num_input_tokens_seen": 21882779904,
      "step": 76940,
      "train_runtime": 748127.9297,
      "train_tokens_per_second": 29250.051
    },
    {
      "epoch": 2.7226425066396582,
      "grad_norm": 0.65234375,
      "learning_rate": 1.6956558344821422e-05,
      "loss": 0.3865114688873291,
      "num_input_tokens_seen": 21885640960,
      "step": 76950,
      "train_runtime": 748226.2766,
      "train_tokens_per_second": 29250.03
    },
    {
      "epoch": 2.722996326903387,
      "grad_norm": 0.6328125,
      "learning_rate": 1.695558334247204e-05,
      "loss": 0.39028191566467285,
      "num_input_tokens_seen": 21888424064,
      "step": 76960,
      "train_runtime": 748319.5775,
      "train_tokens_per_second": 29250.102
    },
    {
      "epoch": 2.7233501471671158,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6954608508291262e-05,
      "loss": 0.3842151165008545,
      "num_input_tokens_seen": 21891334464,
      "step": 76970,
      "train_runtime": 748422.6426,
      "train_tokens_per_second": 29249.963
    },
    {
      "epoch": 2.7237039674308448,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6953633842230755e-05,
      "loss": 0.38736732006073,
      "num_input_tokens_seen": 21894168512,
      "step": 76980,
      "train_runtime": 748518.0486,
      "train_tokens_per_second": 29250.021
    },
    {
      "epoch": 2.7240577876945737,
      "grad_norm": 0.59375,
      "learning_rate": 1.6952659344242205e-05,
      "loss": 0.38971450328826907,
      "num_input_tokens_seen": 21896982784,
      "step": 76990,
      "train_runtime": 748615.3145,
      "train_tokens_per_second": 29249.98
    },
    {
      "epoch": 2.7244116079583023,
      "grad_norm": 0.640625,
      "learning_rate": 1.6951685014277307e-05,
      "loss": 0.3848890781402588,
      "num_input_tokens_seen": 21899842432,
      "step": 77000,
      "train_runtime": 748714.9969,
      "train_tokens_per_second": 29249.905
    },
    {
      "epoch": 2.724765428222031,
      "grad_norm": 0.66015625,
      "learning_rate": 1.6950710852287785e-05,
      "loss": 0.38681445121765134,
      "num_input_tokens_seen": 21902723712,
      "step": 77010,
      "train_runtime": 748814.7742,
      "train_tokens_per_second": 29249.855
    },
    {
      "epoch": 2.72511924848576,
      "grad_norm": 0.625,
      "learning_rate": 1.694973685822538e-05,
      "loss": 0.391303563117981,
      "num_input_tokens_seen": 21905556736,
      "step": 77020,
      "train_runtime": 748915.068,
      "train_tokens_per_second": 29249.721
    },
    {
      "epoch": 2.725473068749489,
      "grad_norm": 0.62109375,
      "learning_rate": 1.694876303204185e-05,
      "loss": 0.3900762557983398,
      "num_input_tokens_seen": 21908415424,
      "step": 77030,
      "train_runtime": 749011.9566,
      "train_tokens_per_second": 29249.754
    },
    {
      "epoch": 2.7258268890132173,
      "grad_norm": 0.609375,
      "learning_rate": 1.694778937368897e-05,
      "loss": 0.3882468223571777,
      "num_input_tokens_seen": 21911277504,
      "step": 77040,
      "train_runtime": 749111.3282,
      "train_tokens_per_second": 29249.695
    },
    {
      "epoch": 2.7261807092769463,
      "grad_norm": 0.58984375,
      "learning_rate": 1.694681588311855e-05,
      "loss": 0.38852438926696775,
      "num_input_tokens_seen": 21914132480,
      "step": 77050,
      "train_runtime": 749209.0818,
      "train_tokens_per_second": 29249.689
    },
    {
      "epoch": 2.726534529540675,
      "grad_norm": 0.65625,
      "learning_rate": 1.69458425602824e-05,
      "loss": 0.38608083724975584,
      "num_input_tokens_seen": 21917014848,
      "step": 77060,
      "train_runtime": 749308.3396,
      "train_tokens_per_second": 29249.661
    },
    {
      "epoch": 2.726888349804404,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6944869405132355e-05,
      "loss": 0.38509140014648435,
      "num_input_tokens_seen": 21919817088,
      "step": 77070,
      "train_runtime": 749403.5247,
      "train_tokens_per_second": 29249.685
    },
    {
      "epoch": 2.7272421700681324,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6943896417620272e-05,
      "loss": 0.38679819107055663,
      "num_input_tokens_seen": 21922673088,
      "step": 77080,
      "train_runtime": 749503.2182,
      "train_tokens_per_second": 29249.605
    },
    {
      "epoch": 2.7275959903318614,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6942923597698035e-05,
      "loss": 0.38493216037750244,
      "num_input_tokens_seen": 21925522880,
      "step": 77090,
      "train_runtime": 749599.3769,
      "train_tokens_per_second": 29249.655
    },
    {
      "epoch": 2.72794981059559,
      "grad_norm": 0.6484375,
      "learning_rate": 1.6941950945317536e-05,
      "loss": 0.3853883743286133,
      "num_input_tokens_seen": 21928275456,
      "step": 77100,
      "train_runtime": 749689.7128,
      "train_tokens_per_second": 29249.802
    },
    {
      "epoch": 2.728303630859319,
      "grad_norm": 0.625,
      "learning_rate": 1.6940978460430684e-05,
      "loss": 0.38822784423828127,
      "num_input_tokens_seen": 21931067648,
      "step": 77110,
      "train_runtime": 749782.9327,
      "train_tokens_per_second": 29249.889
    },
    {
      "epoch": 2.7286574511230475,
      "grad_norm": 0.65234375,
      "learning_rate": 1.694000614298942e-05,
      "loss": 0.3830754518508911,
      "num_input_tokens_seen": 21933895808,
      "step": 77120,
      "train_runtime": 749878.0472,
      "train_tokens_per_second": 29249.951
    },
    {
      "epoch": 2.7290112713867765,
      "grad_norm": 0.6640625,
      "learning_rate": 1.6939033992945697e-05,
      "loss": 0.3825803756713867,
      "num_input_tokens_seen": 21936745856,
      "step": 77130,
      "train_runtime": 749974.8465,
      "train_tokens_per_second": 29249.975
    },
    {
      "epoch": 2.7293650916505054,
      "grad_norm": 0.578125,
      "learning_rate": 1.6938062010251483e-05,
      "loss": 0.3856812953948975,
      "num_input_tokens_seen": 21939569088,
      "step": 77140,
      "train_runtime": 750070.6797,
      "train_tokens_per_second": 29250.002
    },
    {
      "epoch": 2.729718911914234,
      "grad_norm": 0.59375,
      "learning_rate": 1.693709019485877e-05,
      "loss": 0.38559327125549314,
      "num_input_tokens_seen": 21942403072,
      "step": 77150,
      "train_runtime": 750165.7906,
      "train_tokens_per_second": 29250.072
    },
    {
      "epoch": 2.7300727321779625,
      "grad_norm": 0.640625,
      "learning_rate": 1.6936118546719572e-05,
      "loss": 0.3913443565368652,
      "num_input_tokens_seen": 21945177536,
      "step": 77160,
      "train_runtime": 750259.1952,
      "train_tokens_per_second": 29250.128
    },
    {
      "epoch": 2.7304265524416915,
      "grad_norm": 0.640625,
      "learning_rate": 1.6935147065785917e-05,
      "loss": 0.3923962116241455,
      "num_input_tokens_seen": 21948041216,
      "step": 77170,
      "train_runtime": 750356.5018,
      "train_tokens_per_second": 29250.151
    },
    {
      "epoch": 2.7307803727054205,
      "grad_norm": 0.62890625,
      "learning_rate": 1.693417575200986e-05,
      "loss": 0.3857344388961792,
      "num_input_tokens_seen": 21950913408,
      "step": 77180,
      "train_runtime": 750453.6191,
      "train_tokens_per_second": 29250.193
    },
    {
      "epoch": 2.731134192969149,
      "grad_norm": 0.58984375,
      "learning_rate": 1.6933204605343466e-05,
      "loss": 0.3833878993988037,
      "num_input_tokens_seen": 21953740544,
      "step": 77190,
      "train_runtime": 750550.7575,
      "train_tokens_per_second": 29250.174
    },
    {
      "epoch": 2.7314880132328776,
      "grad_norm": 0.63671875,
      "learning_rate": 1.693223362573882e-05,
      "loss": 0.3907898426055908,
      "num_input_tokens_seen": 21956518336,
      "step": 77200,
      "train_runtime": 750642.4725,
      "train_tokens_per_second": 29250.301
    },
    {
      "epoch": 2.7318418334966066,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6931262813148036e-05,
      "loss": 0.39191856384277346,
      "num_input_tokens_seen": 21959352768,
      "step": 77210,
      "train_runtime": 750740.714,
      "train_tokens_per_second": 29250.249
    },
    {
      "epoch": 2.7321956537603356,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6930292167523236e-05,
      "loss": 0.3813400030136108,
      "num_input_tokens_seen": 21962230144,
      "step": 77220,
      "train_runtime": 750839.7142,
      "train_tokens_per_second": 29250.224
    },
    {
      "epoch": 2.732549474024064,
      "grad_norm": 0.609375,
      "learning_rate": 1.6929321688816573e-05,
      "loss": 0.3807516098022461,
      "num_input_tokens_seen": 21965091328,
      "step": 77230,
      "train_runtime": 750933.5386,
      "train_tokens_per_second": 29250.38
    },
    {
      "epoch": 2.732903294287793,
      "grad_norm": 0.640625,
      "learning_rate": 1.6928351376980197e-05,
      "loss": 0.38213391304016114,
      "num_input_tokens_seen": 21967972992,
      "step": 77240,
      "train_runtime": 751031.5699,
      "train_tokens_per_second": 29250.399
    },
    {
      "epoch": 2.7332571145515216,
      "grad_norm": 0.6171875,
      "learning_rate": 1.692738123196631e-05,
      "loss": 0.3840660572052002,
      "num_input_tokens_seen": 21970800576,
      "step": 77250,
      "train_runtime": 751128.4832,
      "train_tokens_per_second": 29250.389
    },
    {
      "epoch": 2.7336109348152506,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6926411253727104e-05,
      "loss": 0.38460583686828614,
      "num_input_tokens_seen": 21973649280,
      "step": 77260,
      "train_runtime": 751225.0384,
      "train_tokens_per_second": 29250.422
    },
    {
      "epoch": 2.733964755078979,
      "grad_norm": 0.61328125,
      "learning_rate": 1.69254414422148e-05,
      "loss": 0.3856522798538208,
      "num_input_tokens_seen": 21976449216,
      "step": 77270,
      "train_runtime": 751318.7683,
      "train_tokens_per_second": 29250.499
    },
    {
      "epoch": 2.734318575342708,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6924471797381652e-05,
      "loss": 0.38483476638793945,
      "num_input_tokens_seen": 21979263296,
      "step": 77280,
      "train_runtime": 751414.4084,
      "train_tokens_per_second": 29250.521
    },
    {
      "epoch": 2.7346723956064367,
      "grad_norm": 0.62109375,
      "learning_rate": 1.692350231917991e-05,
      "loss": 0.38748822212219236,
      "num_input_tokens_seen": 21982126528,
      "step": 77290,
      "train_runtime": 751510.7036,
      "train_tokens_per_second": 29250.583
    },
    {
      "epoch": 2.7350262158701657,
      "grad_norm": 0.609375,
      "learning_rate": 1.692253300756186e-05,
      "loss": 0.3898794174194336,
      "num_input_tokens_seen": 21985036480,
      "step": 77300,
      "train_runtime": 751611.5396,
      "train_tokens_per_second": 29250.531
    },
    {
      "epoch": 2.7353800361338942,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6921563862479795e-05,
      "loss": 0.3850629568099976,
      "num_input_tokens_seen": 21987895872,
      "step": 77310,
      "train_runtime": 751707.2659,
      "train_tokens_per_second": 29250.61
    },
    {
      "epoch": 2.7357338563976232,
      "grad_norm": 0.609375,
      "learning_rate": 1.692059488388604e-05,
      "loss": 0.38352408409118655,
      "num_input_tokens_seen": 21990741056,
      "step": 77320,
      "train_runtime": 751803.5255,
      "train_tokens_per_second": 29250.649
    },
    {
      "epoch": 2.736087676661352,
      "grad_norm": 0.625,
      "learning_rate": 1.6919626071732924e-05,
      "loss": 0.38570780754089357,
      "num_input_tokens_seen": 21993612672,
      "step": 77330,
      "train_runtime": 751903.1957,
      "train_tokens_per_second": 29250.591
    },
    {
      "epoch": 2.7364414969250808,
      "grad_norm": 0.609375,
      "learning_rate": 1.6918657425972812e-05,
      "loss": 0.3877967596054077,
      "num_input_tokens_seen": 21996454272,
      "step": 77340,
      "train_runtime": 752001.1087,
      "train_tokens_per_second": 29250.561
    },
    {
      "epoch": 2.7367953171888093,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6917688946558078e-05,
      "loss": 0.3909803867340088,
      "num_input_tokens_seen": 21999303488,
      "step": 77350,
      "train_runtime": 752098.5884,
      "train_tokens_per_second": 29250.558
    },
    {
      "epoch": 2.7371491374525383,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6916720633441113e-05,
      "loss": 0.3863527297973633,
      "num_input_tokens_seen": 22002172608,
      "step": 77360,
      "train_runtime": 752195.9116,
      "train_tokens_per_second": 29250.588
    },
    {
      "epoch": 2.7375029577162673,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6915752486574336e-05,
      "loss": 0.3840712785720825,
      "num_input_tokens_seen": 22005088576,
      "step": 77370,
      "train_runtime": 752294.9756,
      "train_tokens_per_second": 29250.612
    },
    {
      "epoch": 2.737856777979996,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6914784505910173e-05,
      "loss": 0.38370304107666015,
      "num_input_tokens_seen": 22007950400,
      "step": 77380,
      "train_runtime": 752388.907,
      "train_tokens_per_second": 29250.764
    },
    {
      "epoch": 2.738210598243725,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6913816691401083e-05,
      "loss": 0.3804908514022827,
      "num_input_tokens_seen": 22010779520,
      "step": 77390,
      "train_runtime": 752484.7852,
      "train_tokens_per_second": 29250.797
    },
    {
      "epoch": 2.7385644185074534,
      "grad_norm": 0.625,
      "learning_rate": 1.691284904299953e-05,
      "loss": 0.3805437803268433,
      "num_input_tokens_seen": 22013619264,
      "step": 77400,
      "train_runtime": 752579.5399,
      "train_tokens_per_second": 29250.887
    },
    {
      "epoch": 2.7389182387711823,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6911881560658007e-05,
      "loss": 0.38747193813323977,
      "num_input_tokens_seen": 22016483264,
      "step": 77410,
      "train_runtime": 752677.4287,
      "train_tokens_per_second": 29250.888
    },
    {
      "epoch": 2.739272059034911,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6910914244329028e-05,
      "loss": 0.38733615875244143,
      "num_input_tokens_seen": 22019272000,
      "step": 77420,
      "train_runtime": 752771.7801,
      "train_tokens_per_second": 29250.926
    },
    {
      "epoch": 2.73962587929864,
      "grad_norm": 0.62890625,
      "learning_rate": 1.690994709396511e-05,
      "loss": 0.38250150680541994,
      "num_input_tokens_seen": 22022173440,
      "step": 77430,
      "train_runtime": 752874.0059,
      "train_tokens_per_second": 29250.809
    },
    {
      "epoch": 2.7399796995623684,
      "grad_norm": 0.609375,
      "learning_rate": 1.690898010951881e-05,
      "loss": 0.384661865234375,
      "num_input_tokens_seen": 22024978752,
      "step": 77440,
      "train_runtime": 752969.4686,
      "train_tokens_per_second": 29250.826
    },
    {
      "epoch": 2.7403335198260974,
      "grad_norm": 0.60546875,
      "learning_rate": 1.690801329094269e-05,
      "loss": 0.38803393840789796,
      "num_input_tokens_seen": 22027876416,
      "step": 77450,
      "train_runtime": 753067.8373,
      "train_tokens_per_second": 29250.853
    },
    {
      "epoch": 2.740687340089826,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6907046638189336e-05,
      "loss": 0.38559856414794924,
      "num_input_tokens_seen": 22030670016,
      "step": 77460,
      "train_runtime": 753162.3707,
      "train_tokens_per_second": 29250.89
    },
    {
      "epoch": 2.741041160353555,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6906080151211352e-05,
      "loss": 0.38303604125976565,
      "num_input_tokens_seen": 22033492032,
      "step": 77470,
      "train_runtime": 753258.5108,
      "train_tokens_per_second": 29250.904
    },
    {
      "epoch": 2.741394980617284,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6905113829961355e-05,
      "loss": 0.3902784824371338,
      "num_input_tokens_seen": 22036393792,
      "step": 77480,
      "train_runtime": 753359.9816,
      "train_tokens_per_second": 29250.815
    },
    {
      "epoch": 2.7417488008810125,
      "grad_norm": 0.609375,
      "learning_rate": 1.6904147674391995e-05,
      "loss": 0.38965511322021484,
      "num_input_tokens_seen": 22039244224,
      "step": 77490,
      "train_runtime": 753457.1231,
      "train_tokens_per_second": 29250.827
    },
    {
      "epoch": 2.742102621144741,
      "grad_norm": 0.6171875,
      "learning_rate": 1.690318168445593e-05,
      "loss": 0.38219285011291504,
      "num_input_tokens_seen": 22042118848,
      "step": 77500,
      "train_runtime": 753557.1493,
      "train_tokens_per_second": 29250.759
    },
    {
      "epoch": 2.74245644140847,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6902215860105837e-05,
      "loss": 0.3888564109802246,
      "num_input_tokens_seen": 22044959296,
      "step": 77510,
      "train_runtime": 753651.5898,
      "train_tokens_per_second": 29250.863
    },
    {
      "epoch": 2.742810261672199,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6901250201294422e-05,
      "loss": 0.38782639503479005,
      "num_input_tokens_seen": 22047794688,
      "step": 77520,
      "train_runtime": 753749.1399,
      "train_tokens_per_second": 29250.839
    },
    {
      "epoch": 2.7431640819359275,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6900284707974394e-05,
      "loss": 0.3811646461486816,
      "num_input_tokens_seen": 22050627136,
      "step": 77530,
      "train_runtime": 753845.4471,
      "train_tokens_per_second": 29250.859
    },
    {
      "epoch": 2.743517902199656,
      "grad_norm": 0.62109375,
      "learning_rate": 1.689931938009849e-05,
      "loss": 0.39369497299194334,
      "num_input_tokens_seen": 22053482688,
      "step": 77540,
      "train_runtime": 753941.214,
      "train_tokens_per_second": 29250.931
    },
    {
      "epoch": 2.743871722463385,
      "grad_norm": 0.609375,
      "learning_rate": 1.6898354217619476e-05,
      "loss": 0.3870671272277832,
      "num_input_tokens_seen": 22056334528,
      "step": 77550,
      "train_runtime": 754036.6702,
      "train_tokens_per_second": 29251.01
    },
    {
      "epoch": 2.744225542727114,
      "grad_norm": 0.62109375,
      "learning_rate": 1.689738922049011e-05,
      "loss": 0.38261919021606444,
      "num_input_tokens_seen": 22059175488,
      "step": 77560,
      "train_runtime": 754135.1853,
      "train_tokens_per_second": 29250.956
    },
    {
      "epoch": 2.7445793629908426,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6896424388663198e-05,
      "loss": 0.3870835304260254,
      "num_input_tokens_seen": 22061974016,
      "step": 77570,
      "train_runtime": 754231.244,
      "train_tokens_per_second": 29250.942
    },
    {
      "epoch": 2.7449331832545716,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6895459722091548e-05,
      "loss": 0.38804211616516116,
      "num_input_tokens_seen": 22064835392,
      "step": 77580,
      "train_runtime": 754329.4637,
      "train_tokens_per_second": 29250.926
    },
    {
      "epoch": 2.7452870035183,
      "grad_norm": 0.65234375,
      "learning_rate": 1.689449522072799e-05,
      "loss": 0.3837564468383789,
      "num_input_tokens_seen": 22067711104,
      "step": 77590,
      "train_runtime": 754427.2535,
      "train_tokens_per_second": 29250.946
    },
    {
      "epoch": 2.745640823782029,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6893530884525372e-05,
      "loss": 0.3875821590423584,
      "num_input_tokens_seen": 22070538496,
      "step": 77600,
      "train_runtime": 754521.6255,
      "train_tokens_per_second": 29251.035
    },
    {
      "epoch": 2.7459946440457577,
      "grad_norm": 0.609375,
      "learning_rate": 1.6892566713436568e-05,
      "loss": 0.38980393409729003,
      "num_input_tokens_seen": 22073471424,
      "step": 77610,
      "train_runtime": 754627.0618,
      "train_tokens_per_second": 29250.835
    },
    {
      "epoch": 2.7463484643094866,
      "grad_norm": 0.6640625,
      "learning_rate": 1.6891602707414458e-05,
      "loss": 0.3899756908416748,
      "num_input_tokens_seen": 22076286656,
      "step": 77620,
      "train_runtime": 754722.0824,
      "train_tokens_per_second": 29250.882
    },
    {
      "epoch": 2.7467022845732156,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6890638866411958e-05,
      "loss": 0.3900710105895996,
      "num_input_tokens_seen": 22079122304,
      "step": 77630,
      "train_runtime": 754819.7119,
      "train_tokens_per_second": 29250.855
    },
    {
      "epoch": 2.747056104836944,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6889675190381985e-05,
      "loss": 0.390013313293457,
      "num_input_tokens_seen": 22081886336,
      "step": 77640,
      "train_runtime": 754912.078,
      "train_tokens_per_second": 29250.938
    },
    {
      "epoch": 2.7474099251006727,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6888711679277482e-05,
      "loss": 0.38519601821899413,
      "num_input_tokens_seen": 22084674624,
      "step": 77650,
      "train_runtime": 755008.7423,
      "train_tokens_per_second": 29250.886
    },
    {
      "epoch": 2.7477637453644017,
      "grad_norm": 0.6171875,
      "learning_rate": 1.688774833305142e-05,
      "loss": 0.38426783084869387,
      "num_input_tokens_seen": 22087477760,
      "step": 77660,
      "train_runtime": 755103.0701,
      "train_tokens_per_second": 29250.944
    },
    {
      "epoch": 2.7481175656281307,
      "grad_norm": 0.62109375,
      "learning_rate": 1.688678515165677e-05,
      "loss": 0.3895281791687012,
      "num_input_tokens_seen": 22090309504,
      "step": 77670,
      "train_runtime": 755200.9052,
      "train_tokens_per_second": 29250.904
    },
    {
      "epoch": 2.7484713858918592,
      "grad_norm": 0.609375,
      "learning_rate": 1.6885822135046542e-05,
      "loss": 0.38850746154785154,
      "num_input_tokens_seen": 22093155520,
      "step": 77680,
      "train_runtime": 755300.3605,
      "train_tokens_per_second": 29250.821
    },
    {
      "epoch": 2.748825206155588,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6884859283173742e-05,
      "loss": 0.3878913164138794,
      "num_input_tokens_seen": 22095957504,
      "step": 77690,
      "train_runtime": 755393.2069,
      "train_tokens_per_second": 29250.935
    },
    {
      "epoch": 2.7491790264193168,
      "grad_norm": 0.62890625,
      "learning_rate": 1.688389659599142e-05,
      "loss": 0.38462255001068113,
      "num_input_tokens_seen": 22098825664,
      "step": 77700,
      "train_runtime": 755493.6739,
      "train_tokens_per_second": 29250.841
    },
    {
      "epoch": 2.7495328466830458,
      "grad_norm": 0.60546875,
      "learning_rate": 1.688293407345263e-05,
      "loss": 0.3852888822555542,
      "num_input_tokens_seen": 22101652608,
      "step": 77710,
      "train_runtime": 755591.7813,
      "train_tokens_per_second": 29250.785
    },
    {
      "epoch": 2.7498866669467743,
      "grad_norm": 0.63671875,
      "learning_rate": 1.688197171551044e-05,
      "loss": 0.3814253568649292,
      "num_input_tokens_seen": 22104494784,
      "step": 77720,
      "train_runtime": 755689.4385,
      "train_tokens_per_second": 29250.766
    },
    {
      "epoch": 2.7502404872105033,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6881009522117954e-05,
      "loss": 0.3821877002716064,
      "num_input_tokens_seen": 22107361344,
      "step": 77730,
      "train_runtime": 755790.6466,
      "train_tokens_per_second": 29250.642
    },
    {
      "epoch": 2.750594307474232,
      "grad_norm": 0.64453125,
      "learning_rate": 1.688004749322828e-05,
      "loss": 0.38443262577056886,
      "num_input_tokens_seen": 22110234624,
      "step": 77740,
      "train_runtime": 755890.2035,
      "train_tokens_per_second": 29250.59
    },
    {
      "epoch": 2.750948127737961,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6879085628794544e-05,
      "loss": 0.3903075695037842,
      "num_input_tokens_seen": 22113058496,
      "step": 77750,
      "train_runtime": 755988.8854,
      "train_tokens_per_second": 29250.507
    },
    {
      "epoch": 2.7513019480016894,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6878123928769905e-05,
      "loss": 0.3827351093292236,
      "num_input_tokens_seen": 22115896384,
      "step": 77760,
      "train_runtime": 756085.6973,
      "train_tokens_per_second": 29250.515
    },
    {
      "epoch": 2.7516557682654184,
      "grad_norm": 0.6640625,
      "learning_rate": 1.6877162393107524e-05,
      "loss": 0.38554117679595945,
      "num_input_tokens_seen": 22118697984,
      "step": 77770,
      "train_runtime": 756180.304,
      "train_tokens_per_second": 29250.561
    },
    {
      "epoch": 2.752009588529147,
      "grad_norm": 0.63671875,
      "learning_rate": 1.687620102176059e-05,
      "loss": 0.38048548698425294,
      "num_input_tokens_seen": 22121535360,
      "step": 77780,
      "train_runtime": 756277.6664,
      "train_tokens_per_second": 29250.547
    },
    {
      "epoch": 2.752363408792876,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6875239814682313e-05,
      "loss": 0.38638551235198976,
      "num_input_tokens_seen": 22124367936,
      "step": 77790,
      "train_runtime": 756375.6677,
      "train_tokens_per_second": 29250.502
    },
    {
      "epoch": 2.7527172290566044,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6874278771825917e-05,
      "loss": 0.3864504337310791,
      "num_input_tokens_seen": 22127225408,
      "step": 77800,
      "train_runtime": 756474.4407,
      "train_tokens_per_second": 29250.46
    },
    {
      "epoch": 2.7530710493203334,
      "grad_norm": 0.625,
      "learning_rate": 1.6873317893144635e-05,
      "loss": 0.38992536067962646,
      "num_input_tokens_seen": 22130098176,
      "step": 77810,
      "train_runtime": 756575.7715,
      "train_tokens_per_second": 29250.34
    },
    {
      "epoch": 2.7534248695840624,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6872357178591744e-05,
      "loss": 0.38590755462646487,
      "num_input_tokens_seen": 22132897664,
      "step": 77820,
      "train_runtime": 756668.3787,
      "train_tokens_per_second": 29250.459
    },
    {
      "epoch": 2.753778689847791,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6871396628120516e-05,
      "loss": 0.3891230583190918,
      "num_input_tokens_seen": 22135731520,
      "step": 77830,
      "train_runtime": 756767.6185,
      "train_tokens_per_second": 29250.368
    },
    {
      "epoch": 2.7541325101115195,
      "grad_norm": 0.609375,
      "learning_rate": 1.6870436241684255e-05,
      "loss": 0.38344383239746094,
      "num_input_tokens_seen": 22138625280,
      "step": 77840,
      "train_runtime": 756867.1567,
      "train_tokens_per_second": 29250.345
    },
    {
      "epoch": 2.7544863303752485,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6869476019236267e-05,
      "loss": 0.3902973413467407,
      "num_input_tokens_seen": 22141391040,
      "step": 77850,
      "train_runtime": 756961.1538,
      "train_tokens_per_second": 29250.366
    },
    {
      "epoch": 2.7548401506389775,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6868515960729903e-05,
      "loss": 0.38764266967773436,
      "num_input_tokens_seen": 22144226176,
      "step": 77860,
      "train_runtime": 757055.6285,
      "train_tokens_per_second": 29250.461
    },
    {
      "epoch": 2.755193970902706,
      "grad_norm": 0.6328125,
      "learning_rate": 1.686755606611851e-05,
      "loss": 0.3846998453140259,
      "num_input_tokens_seen": 22147074304,
      "step": 77870,
      "train_runtime": 757150.3233,
      "train_tokens_per_second": 29250.564
    },
    {
      "epoch": 2.755547791166435,
      "grad_norm": 0.625,
      "learning_rate": 1.6866596335355463e-05,
      "loss": 0.3827911138534546,
      "num_input_tokens_seen": 22149905152,
      "step": 77880,
      "train_runtime": 757247.042,
      "train_tokens_per_second": 29250.567
    },
    {
      "epoch": 2.7559016114301635,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6865636768394156e-05,
      "loss": 0.3889758110046387,
      "num_input_tokens_seen": 22152753536,
      "step": 77890,
      "train_runtime": 757344.7885,
      "train_tokens_per_second": 29250.553
    },
    {
      "epoch": 2.7562554316938925,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6864677365187998e-05,
      "loss": 0.37844281196594237,
      "num_input_tokens_seen": 22155593856,
      "step": 77900,
      "train_runtime": 757441.8161,
      "train_tokens_per_second": 29250.555
    },
    {
      "epoch": 2.756609251957621,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6863718125690418e-05,
      "loss": 0.3849318027496338,
      "num_input_tokens_seen": 22158424960,
      "step": 77910,
      "train_runtime": 757537.1247,
      "train_tokens_per_second": 29250.613
    },
    {
      "epoch": 2.75696307222135,
      "grad_norm": 0.6171875,
      "learning_rate": 1.686275904985486e-05,
      "loss": 0.3863871097564697,
      "num_input_tokens_seen": 22161254208,
      "step": 77920,
      "train_runtime": 757634.4553,
      "train_tokens_per_second": 29250.589
    },
    {
      "epoch": 2.7573168924850786,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6861800137634795e-05,
      "loss": 0.3870380163192749,
      "num_input_tokens_seen": 22164182720,
      "step": 77930,
      "train_runtime": 757736.6095,
      "train_tokens_per_second": 29250.511
    },
    {
      "epoch": 2.7576707127488076,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6860841388983708e-05,
      "loss": 0.3864006042480469,
      "num_input_tokens_seen": 22167008064,
      "step": 77940,
      "train_runtime": 757831.0429,
      "train_tokens_per_second": 29250.594
    },
    {
      "epoch": 2.758024533012536,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6859882803855102e-05,
      "loss": 0.3868141174316406,
      "num_input_tokens_seen": 22169825856,
      "step": 77950,
      "train_runtime": 757925.5393,
      "train_tokens_per_second": 29250.665
    },
    {
      "epoch": 2.758378353276265,
      "grad_norm": 0.640625,
      "learning_rate": 1.6858924382202497e-05,
      "loss": 0.3833159923553467,
      "num_input_tokens_seen": 22172692160,
      "step": 77960,
      "train_runtime": 758024.2703,
      "train_tokens_per_second": 29250.636
    },
    {
      "epoch": 2.758732173539994,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6857966123979434e-05,
      "loss": 0.3863433837890625,
      "num_input_tokens_seen": 22175520448,
      "step": 77970,
      "train_runtime": 758120.08,
      "train_tokens_per_second": 29250.67
    },
    {
      "epoch": 2.7590859938037227,
      "grad_norm": 0.6328125,
      "learning_rate": 1.685700802913947e-05,
      "loss": 0.38769395351409913,
      "num_input_tokens_seen": 22178426944,
      "step": 77980,
      "train_runtime": 758219.2038,
      "train_tokens_per_second": 29250.68
    },
    {
      "epoch": 2.759439814067451,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6856050097636186e-05,
      "loss": 0.3921926975250244,
      "num_input_tokens_seen": 22181314688,
      "step": 77990,
      "train_runtime": 758321.1588,
      "train_tokens_per_second": 29250.555
    },
    {
      "epoch": 2.75979363433118,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6855092329423174e-05,
      "loss": 0.3871100902557373,
      "num_input_tokens_seen": 22184108224,
      "step": 78000,
      "train_runtime": 758412.7332,
      "train_tokens_per_second": 29250.706
    },
    {
      "epoch": 2.760147454594909,
      "grad_norm": 0.6328125,
      "learning_rate": 1.685413472445405e-05,
      "loss": 0.383712100982666,
      "num_input_tokens_seen": 22186936768,
      "step": 78010,
      "train_runtime": 758509.242,
      "train_tokens_per_second": 29250.714
    },
    {
      "epoch": 2.7605012748586377,
      "grad_norm": 0.59765625,
      "learning_rate": 1.685317728268245e-05,
      "loss": 0.38512454032897947,
      "num_input_tokens_seen": 22189785280,
      "step": 78020,
      "train_runtime": 758609.2897,
      "train_tokens_per_second": 29250.611
    },
    {
      "epoch": 2.7608550951223663,
      "grad_norm": 0.62890625,
      "learning_rate": 1.685222000406202e-05,
      "loss": 0.38504347801208494,
      "num_input_tokens_seen": 22192593792,
      "step": 78030,
      "train_runtime": 758705.6,
      "train_tokens_per_second": 29250.6
    },
    {
      "epoch": 2.7612089153860953,
      "grad_norm": 0.59765625,
      "learning_rate": 1.685126288854643e-05,
      "loss": 0.38506505489349363,
      "num_input_tokens_seen": 22195471872,
      "step": 78040,
      "train_runtime": 758804.4313,
      "train_tokens_per_second": 29250.583
    },
    {
      "epoch": 2.7615627356498242,
      "grad_norm": 0.625,
      "learning_rate": 1.6850305936089365e-05,
      "loss": 0.38674774169921877,
      "num_input_tokens_seen": 22198294656,
      "step": 78050,
      "train_runtime": 758899.343,
      "train_tokens_per_second": 29250.644
    },
    {
      "epoch": 2.761916555913553,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6849349146644543e-05,
      "loss": 0.38572063446044924,
      "num_input_tokens_seen": 22201098240,
      "step": 78060,
      "train_runtime": 758994.7973,
      "train_tokens_per_second": 29250.659
    },
    {
      "epoch": 2.7622703761772818,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6848392520165675e-05,
      "loss": 0.3863166093826294,
      "num_input_tokens_seen": 22203973952,
      "step": 78070,
      "train_runtime": 759092.8568,
      "train_tokens_per_second": 29250.669
    },
    {
      "epoch": 2.7626241964410103,
      "grad_norm": 0.60546875,
      "learning_rate": 1.684743605660651e-05,
      "loss": 0.3838151693344116,
      "num_input_tokens_seen": 22206807104,
      "step": 78080,
      "train_runtime": 759194.3946,
      "train_tokens_per_second": 29250.489
    },
    {
      "epoch": 2.7629780167047393,
      "grad_norm": 0.62109375,
      "learning_rate": 1.684647975592081e-05,
      "loss": 0.3818851709365845,
      "num_input_tokens_seen": 22209710528,
      "step": 78090,
      "train_runtime": 759292.4956,
      "train_tokens_per_second": 29250.533
    },
    {
      "epoch": 2.763331836968468,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6845523618062353e-05,
      "loss": 0.38626275062561033,
      "num_input_tokens_seen": 22212573568,
      "step": 78100,
      "train_runtime": 759390.0674,
      "train_tokens_per_second": 29250.545
    },
    {
      "epoch": 2.763685657232197,
      "grad_norm": 0.62109375,
      "learning_rate": 1.684456764298494e-05,
      "loss": 0.38286261558532714,
      "num_input_tokens_seen": 22215399360,
      "step": 78110,
      "train_runtime": 759486.69,
      "train_tokens_per_second": 29250.545
    },
    {
      "epoch": 2.7640394774959254,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6843611830642385e-05,
      "loss": 0.38418700695037844,
      "num_input_tokens_seen": 22218234368,
      "step": 78120,
      "train_runtime": 759583.6029,
      "train_tokens_per_second": 29250.545
    },
    {
      "epoch": 2.7643932977596544,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6842656180988522e-05,
      "loss": 0.38846583366394044,
      "num_input_tokens_seen": 22221042496,
      "step": 78130,
      "train_runtime": 759679.9268,
      "train_tokens_per_second": 29250.533
    },
    {
      "epoch": 2.764747118023383,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6841700693977213e-05,
      "loss": 0.38442559242248536,
      "num_input_tokens_seen": 22223843904,
      "step": 78140,
      "train_runtime": 759774.4951,
      "train_tokens_per_second": 29250.579
    },
    {
      "epoch": 2.765100938287112,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6840745369562314e-05,
      "loss": 0.3828442573547363,
      "num_input_tokens_seen": 22226673472,
      "step": 78150,
      "train_runtime": 759869.0405,
      "train_tokens_per_second": 29250.663
    },
    {
      "epoch": 2.765454758550841,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6839790207697727e-05,
      "loss": 0.38646955490112306,
      "num_input_tokens_seen": 22229570880,
      "step": 78160,
      "train_runtime": 759968.713,
      "train_tokens_per_second": 29250.64
    },
    {
      "epoch": 2.7658085788145694,
      "grad_norm": 0.609375,
      "learning_rate": 1.6838835208337357e-05,
      "loss": 0.38210015296936034,
      "num_input_tokens_seen": 22232420608,
      "step": 78170,
      "train_runtime": 760064.9304,
      "train_tokens_per_second": 29250.686
    },
    {
      "epoch": 2.766162399078298,
      "grad_norm": 0.609375,
      "learning_rate": 1.6837880371435127e-05,
      "loss": 0.3843751668930054,
      "num_input_tokens_seen": 22235267136,
      "step": 78180,
      "train_runtime": 760163.2679,
      "train_tokens_per_second": 29250.647
    },
    {
      "epoch": 2.766516219342027,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6836925696944987e-05,
      "loss": 0.38655805587768555,
      "num_input_tokens_seen": 22238079104,
      "step": 78190,
      "train_runtime": 760257.9899,
      "train_tokens_per_second": 29250.701
    },
    {
      "epoch": 2.766870039605756,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6835971184820895e-05,
      "loss": 0.39056613445281985,
      "num_input_tokens_seen": 22240920832,
      "step": 78200,
      "train_runtime": 760353.5365,
      "train_tokens_per_second": 29250.763
    },
    {
      "epoch": 2.7672238598694845,
      "grad_norm": 0.625,
      "learning_rate": 1.6835016835016834e-05,
      "loss": 0.3845752000808716,
      "num_input_tokens_seen": 22243733824,
      "step": 78210,
      "train_runtime": 760448.7896,
      "train_tokens_per_second": 29250.798
    },
    {
      "epoch": 2.7675776801332135,
      "grad_norm": 0.60546875,
      "learning_rate": 1.683406264748681e-05,
      "loss": 0.38459124565124514,
      "num_input_tokens_seen": 22246553728,
      "step": 78220,
      "train_runtime": 760543.6462,
      "train_tokens_per_second": 29250.857
    },
    {
      "epoch": 2.767931500396942,
      "grad_norm": 0.609375,
      "learning_rate": 1.6833108622184828e-05,
      "loss": 0.38901629447937014,
      "num_input_tokens_seen": 22249420352,
      "step": 78230,
      "train_runtime": 760644.2516,
      "train_tokens_per_second": 29250.757
    },
    {
      "epoch": 2.768285320660671,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6832154759064936e-05,
      "loss": 0.3875391960144043,
      "num_input_tokens_seen": 22252240832,
      "step": 78240,
      "train_runtime": 760740.2973,
      "train_tokens_per_second": 29250.772
    },
    {
      "epoch": 2.7686391409243996,
      "grad_norm": 0.65625,
      "learning_rate": 1.6831201058081183e-05,
      "loss": 0.38439579010009767,
      "num_input_tokens_seen": 22255082176,
      "step": 78250,
      "train_runtime": 760835.9149,
      "train_tokens_per_second": 29250.83
    },
    {
      "epoch": 2.7689929611881285,
      "grad_norm": 0.59375,
      "learning_rate": 1.6830247519187638e-05,
      "loss": 0.389575719833374,
      "num_input_tokens_seen": 22257962304,
      "step": 78260,
      "train_runtime": 760934.5693,
      "train_tokens_per_second": 29250.823
    },
    {
      "epoch": 2.769346781451857,
      "grad_norm": 0.6328125,
      "learning_rate": 1.68292941423384e-05,
      "loss": 0.3874102354049683,
      "num_input_tokens_seen": 22260806144,
      "step": 78270,
      "train_runtime": 761033.2437,
      "train_tokens_per_second": 29250.767
    },
    {
      "epoch": 2.769700601715586,
      "grad_norm": 0.6015625,
      "learning_rate": 1.682834092748757e-05,
      "loss": 0.38518362045288085,
      "num_input_tokens_seen": 22263636032,
      "step": 78280,
      "train_runtime": 761129.2154,
      "train_tokens_per_second": 29250.797
    },
    {
      "epoch": 2.7700544219793146,
      "grad_norm": 0.62109375,
      "learning_rate": 1.682738787458928e-05,
      "loss": 0.38323662281036375,
      "num_input_tokens_seen": 22266491520,
      "step": 78290,
      "train_runtime": 761225.9895,
      "train_tokens_per_second": 29250.829
    },
    {
      "epoch": 2.7704082422430436,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6826434983597672e-05,
      "loss": 0.386376953125,
      "num_input_tokens_seen": 22269292672,
      "step": 78300,
      "train_runtime": 761318.5664,
      "train_tokens_per_second": 29250.952
    },
    {
      "epoch": 2.7707620625067726,
      "grad_norm": 0.609375,
      "learning_rate": 1.6825482254466916e-05,
      "loss": 0.3856979846954346,
      "num_input_tokens_seen": 22272114496,
      "step": 78310,
      "train_runtime": 761415.1428,
      "train_tokens_per_second": 29250.948
    },
    {
      "epoch": 2.771115882770501,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6824529687151183e-05,
      "loss": 0.3857269287109375,
      "num_input_tokens_seen": 22274962176,
      "step": 78320,
      "train_runtime": 761512.7167,
      "train_tokens_per_second": 29250.939
    },
    {
      "epoch": 2.7714697030342297,
      "grad_norm": 0.625,
      "learning_rate": 1.682357728160468e-05,
      "loss": 0.38968968391418457,
      "num_input_tokens_seen": 22277834496,
      "step": 78330,
      "train_runtime": 761612.3398,
      "train_tokens_per_second": 29250.884
    },
    {
      "epoch": 2.7718235232979587,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6822625037781626e-05,
      "loss": 0.3861381530761719,
      "num_input_tokens_seen": 22280728768,
      "step": 78340,
      "train_runtime": 761712.5502,
      "train_tokens_per_second": 29250.836
    },
    {
      "epoch": 2.7721773435616877,
      "grad_norm": 0.59375,
      "learning_rate": 1.6821672955636254e-05,
      "loss": 0.3871183156967163,
      "num_input_tokens_seen": 22283664064,
      "step": 78350,
      "train_runtime": 761816.6357,
      "train_tokens_per_second": 29250.692
    },
    {
      "epoch": 2.772531163825416,
      "grad_norm": 0.640625,
      "learning_rate": 1.6820721035122815e-05,
      "loss": 0.38559417724609374,
      "num_input_tokens_seen": 22286453632,
      "step": 78360,
      "train_runtime": 761908.3887,
      "train_tokens_per_second": 29250.831
    },
    {
      "epoch": 2.7728849840891447,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6819769276195584e-05,
      "loss": 0.3849762439727783,
      "num_input_tokens_seen": 22289323072,
      "step": 78370,
      "train_runtime": 762006.2301,
      "train_tokens_per_second": 29250.841
    },
    {
      "epoch": 2.7732388043528737,
      "grad_norm": 0.625,
      "learning_rate": 1.6818817678808855e-05,
      "loss": 0.38625340461730956,
      "num_input_tokens_seen": 22292172864,
      "step": 78380,
      "train_runtime": 762102.3342,
      "train_tokens_per_second": 29250.892
    },
    {
      "epoch": 2.7735926246166027,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6817866242916934e-05,
      "loss": 0.3882770538330078,
      "num_input_tokens_seen": 22295007808,
      "step": 78390,
      "train_runtime": 762199.5133,
      "train_tokens_per_second": 29250.882
    },
    {
      "epoch": 2.7739464448803313,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6816914968474148e-05,
      "loss": 0.3871619701385498,
      "num_input_tokens_seen": 22297791168,
      "step": 78400,
      "train_runtime": 762292.1313,
      "train_tokens_per_second": 29250.979
    },
    {
      "epoch": 2.7743002651440603,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6815963855434838e-05,
      "loss": 0.3849196910858154,
      "num_input_tokens_seen": 22300622016,
      "step": 78410,
      "train_runtime": 762387.2175,
      "train_tokens_per_second": 29251.044
    },
    {
      "epoch": 2.774654085407789,
      "grad_norm": 0.62890625,
      "learning_rate": 1.681501290375337e-05,
      "loss": 0.386171555519104,
      "num_input_tokens_seen": 22303400320,
      "step": 78420,
      "train_runtime": 762480.0351,
      "train_tokens_per_second": 29251.127
    },
    {
      "epoch": 2.775007905671518,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6814062113384127e-05,
      "loss": 0.38807241916656493,
      "num_input_tokens_seen": 22306210880,
      "step": 78430,
      "train_runtime": 762576.5861,
      "train_tokens_per_second": 29251.109
    },
    {
      "epoch": 2.7753617259352463,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6813111484281506e-05,
      "loss": 0.3816505432128906,
      "num_input_tokens_seen": 22309084032,
      "step": 78440,
      "train_runtime": 762679.4806,
      "train_tokens_per_second": 29250.93
    },
    {
      "epoch": 2.7757155461989753,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6812161016399918e-05,
      "loss": 0.3843584775924683,
      "num_input_tokens_seen": 22311947648,
      "step": 78450,
      "train_runtime": 762779.4376,
      "train_tokens_per_second": 29250.851
    },
    {
      "epoch": 2.7760693664627043,
      "grad_norm": 0.59375,
      "learning_rate": 1.681121070969381e-05,
      "loss": 0.38947181701660155,
      "num_input_tokens_seen": 22314863168,
      "step": 78460,
      "train_runtime": 762880.4765,
      "train_tokens_per_second": 29250.799
    },
    {
      "epoch": 2.776423186726433,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6810260564117625e-05,
      "loss": 0.3871855020523071,
      "num_input_tokens_seen": 22317715904,
      "step": 78470,
      "train_runtime": 762980.0693,
      "train_tokens_per_second": 29250.719
    },
    {
      "epoch": 2.7767770069901614,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6809310579625838e-05,
      "loss": 0.38584098815917967,
      "num_input_tokens_seen": 22320463168,
      "step": 78480,
      "train_runtime": 763072.3837,
      "train_tokens_per_second": 29250.781
    },
    {
      "epoch": 2.7771308272538904,
      "grad_norm": 0.6171875,
      "learning_rate": 1.680836075617293e-05,
      "loss": 0.38796844482421877,
      "num_input_tokens_seen": 22323344384,
      "step": 78490,
      "train_runtime": 763172.0173,
      "train_tokens_per_second": 29250.738
    },
    {
      "epoch": 2.7774846475176194,
      "grad_norm": 0.609375,
      "learning_rate": 1.6807411093713423e-05,
      "loss": 0.3851827621459961,
      "num_input_tokens_seen": 22326178176,
      "step": 78500,
      "train_runtime": 763270.9909,
      "train_tokens_per_second": 29250.657
    },
    {
      "epoch": 2.777838467781348,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6806461592201837e-05,
      "loss": 0.383903956413269,
      "num_input_tokens_seen": 22329054592,
      "step": 78510,
      "train_runtime": 763371.2938,
      "train_tokens_per_second": 29250.582
    },
    {
      "epoch": 2.7781922880450765,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6805512251592702e-05,
      "loss": 0.3865769863128662,
      "num_input_tokens_seen": 22331912768,
      "step": 78520,
      "train_runtime": 763470.0557,
      "train_tokens_per_second": 29250.542
    },
    {
      "epoch": 2.7785461083088054,
      "grad_norm": 0.625,
      "learning_rate": 1.6804563071840595e-05,
      "loss": 0.387351393699646,
      "num_input_tokens_seen": 22334710336,
      "step": 78530,
      "train_runtime": 763565.5118,
      "train_tokens_per_second": 29250.549
    },
    {
      "epoch": 2.7788999285725344,
      "grad_norm": 0.609375,
      "learning_rate": 1.680361405290009e-05,
      "loss": 0.3864955186843872,
      "num_input_tokens_seen": 22337572160,
      "step": 78540,
      "train_runtime": 763663.8241,
      "train_tokens_per_second": 29250.531
    },
    {
      "epoch": 2.779253748836263,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6802665194725778e-05,
      "loss": 0.39133076667785643,
      "num_input_tokens_seen": 22340434496,
      "step": 78550,
      "train_runtime": 763760.4965,
      "train_tokens_per_second": 29250.576
    },
    {
      "epoch": 2.779607569099992,
      "grad_norm": 0.59375,
      "learning_rate": 1.6801716497272283e-05,
      "loss": 0.3841114044189453,
      "num_input_tokens_seen": 22343290432,
      "step": 78560,
      "train_runtime": 763856.7578,
      "train_tokens_per_second": 29250.629
    },
    {
      "epoch": 2.7799613893637205,
      "grad_norm": 0.58203125,
      "learning_rate": 1.6800767960494234e-05,
      "loss": 0.38547325134277344,
      "num_input_tokens_seen": 22346138368,
      "step": 78570,
      "train_runtime": 763952.8508,
      "train_tokens_per_second": 29250.677
    },
    {
      "epoch": 2.7803152096274495,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6799819584346275e-05,
      "loss": 0.3842307567596436,
      "num_input_tokens_seen": 22348997440,
      "step": 78580,
      "train_runtime": 764048.1975,
      "train_tokens_per_second": 29250.769
    },
    {
      "epoch": 2.780669029891178,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6798871368783085e-05,
      "loss": 0.39015960693359375,
      "num_input_tokens_seen": 22351825024,
      "step": 78590,
      "train_runtime": 764142.8209,
      "train_tokens_per_second": 29250.847
    },
    {
      "epoch": 2.781022850154907,
      "grad_norm": 0.59375,
      "learning_rate": 1.6797923313759344e-05,
      "loss": 0.3816447019577026,
      "num_input_tokens_seen": 22354714048,
      "step": 78600,
      "train_runtime": 764242.0161,
      "train_tokens_per_second": 29250.831
    },
    {
      "epoch": 2.7813766704186356,
      "grad_norm": 0.60546875,
      "learning_rate": 1.679697541922976e-05,
      "loss": 0.3905625581741333,
      "num_input_tokens_seen": 22357579840,
      "step": 78610,
      "train_runtime": 764341.4414,
      "train_tokens_per_second": 29250.775
    },
    {
      "epoch": 2.7817304906823646,
      "grad_norm": 0.62890625,
      "learning_rate": 1.679602768514905e-05,
      "loss": 0.3888481378555298,
      "num_input_tokens_seen": 22360363328,
      "step": 78620,
      "train_runtime": 764435.5736,
      "train_tokens_per_second": 29250.815
    },
    {
      "epoch": 2.782084310946093,
      "grad_norm": 0.609375,
      "learning_rate": 1.679508011147196e-05,
      "loss": 0.38836982250213625,
      "num_input_tokens_seen": 22363190912,
      "step": 78630,
      "train_runtime": 764531.3634,
      "train_tokens_per_second": 29250.848
    },
    {
      "epoch": 2.782438131209822,
      "grad_norm": 0.62109375,
      "learning_rate": 1.679413269815325e-05,
      "loss": 0.3852013111114502,
      "num_input_tokens_seen": 22366037632,
      "step": 78640,
      "train_runtime": 764627.4857,
      "train_tokens_per_second": 29250.894
    },
    {
      "epoch": 2.782791951473551,
      "grad_norm": 0.59765625,
      "learning_rate": 1.679318544514769e-05,
      "loss": 0.38367300033569335,
      "num_input_tokens_seen": 22368912640,
      "step": 78650,
      "train_runtime": 764726.6005,
      "train_tokens_per_second": 29250.862
    },
    {
      "epoch": 2.7831457717372796,
      "grad_norm": 0.6328125,
      "learning_rate": 1.679223835241007e-05,
      "loss": 0.3868675708770752,
      "num_input_tokens_seen": 22371757440,
      "step": 78660,
      "train_runtime": 764823.4942,
      "train_tokens_per_second": 29250.876
    },
    {
      "epoch": 2.783499592001008,
      "grad_norm": 0.640625,
      "learning_rate": 1.679129141989521e-05,
      "loss": 0.3858372688293457,
      "num_input_tokens_seen": 22374578816,
      "step": 78670,
      "train_runtime": 764918.652,
      "train_tokens_per_second": 29250.926
    },
    {
      "epoch": 2.783853412264737,
      "grad_norm": 0.58984375,
      "learning_rate": 1.6790344647557936e-05,
      "loss": 0.3847068786621094,
      "num_input_tokens_seen": 22377440960,
      "step": 78680,
      "train_runtime": 765018.1488,
      "train_tokens_per_second": 29250.863
    },
    {
      "epoch": 2.784207232528466,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6789398035353095e-05,
      "loss": 0.38566620349884034,
      "num_input_tokens_seen": 22380344832,
      "step": 78690,
      "train_runtime": 765118.0009,
      "train_tokens_per_second": 29250.841
    },
    {
      "epoch": 2.7845610527921947,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6788451583235557e-05,
      "loss": 0.38476803302764895,
      "num_input_tokens_seen": 22383213120,
      "step": 78700,
      "train_runtime": 765216.1778,
      "train_tokens_per_second": 29250.836
    },
    {
      "epoch": 2.7849148730559237,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6787505291160198e-05,
      "loss": 0.38550920486450196,
      "num_input_tokens_seen": 22386001152,
      "step": 78710,
      "train_runtime": 765306.815,
      "train_tokens_per_second": 29251.015
    },
    {
      "epoch": 2.785268693319652,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6786559159081923e-05,
      "loss": 0.3879872798919678,
      "num_input_tokens_seen": 22388851200,
      "step": 78720,
      "train_runtime": 765405.6575,
      "train_tokens_per_second": 29250.961
    },
    {
      "epoch": 2.785622513583381,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6785613186955645e-05,
      "loss": 0.38557913303375246,
      "num_input_tokens_seen": 22391718272,
      "step": 78730,
      "train_runtime": 765504.1252,
      "train_tokens_per_second": 29250.944
    },
    {
      "epoch": 2.7859763338471097,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6784667374736305e-05,
      "loss": 0.3802063226699829,
      "num_input_tokens_seen": 22394603136,
      "step": 78740,
      "train_runtime": 765604.2003,
      "train_tokens_per_second": 29250.889
    },
    {
      "epoch": 2.7863301541108387,
      "grad_norm": 0.61328125,
      "learning_rate": 1.678372172237886e-05,
      "loss": 0.3919590950012207,
      "num_input_tokens_seen": 22397408832,
      "step": 78750,
      "train_runtime": 765700.2599,
      "train_tokens_per_second": 29250.883
    },
    {
      "epoch": 2.7866839743745673,
      "grad_norm": 0.63671875,
      "learning_rate": 1.678277622983827e-05,
      "loss": 0.3880882501602173,
      "num_input_tokens_seen": 22400283968,
      "step": 78760,
      "train_runtime": 765800.2557,
      "train_tokens_per_second": 29250.818
    },
    {
      "epoch": 2.7870377946382963,
      "grad_norm": 0.61328125,
      "learning_rate": 1.678183089706954e-05,
      "loss": 0.3854238033294678,
      "num_input_tokens_seen": 22403159552,
      "step": 78770,
      "train_runtime": 765898.8163,
      "train_tokens_per_second": 29250.808
    },
    {
      "epoch": 2.787391614902025,
      "grad_norm": 0.58984375,
      "learning_rate": 1.6780885724027666e-05,
      "loss": 0.38970532417297366,
      "num_input_tokens_seen": 22405994944,
      "step": 78780,
      "train_runtime": 765997.5295,
      "train_tokens_per_second": 29250.74
    },
    {
      "epoch": 2.787745435165754,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6779940710667678e-05,
      "loss": 0.3870706558227539,
      "num_input_tokens_seen": 22408906560,
      "step": 78790,
      "train_runtime": 766098.8858,
      "train_tokens_per_second": 29250.671
    },
    {
      "epoch": 2.788099255429483,
      "grad_norm": 0.609375,
      "learning_rate": 1.677899585694461e-05,
      "loss": 0.38471298217773436,
      "num_input_tokens_seen": 22411748224,
      "step": 78800,
      "train_runtime": 766194.8742,
      "train_tokens_per_second": 29250.715
    },
    {
      "epoch": 2.7884530756932113,
      "grad_norm": 0.60546875,
      "learning_rate": 1.677805116281354e-05,
      "loss": 0.3853740692138672,
      "num_input_tokens_seen": 22414616128,
      "step": 78810,
      "train_runtime": 766294.7032,
      "train_tokens_per_second": 29250.647
    },
    {
      "epoch": 2.78880689595694,
      "grad_norm": 0.65234375,
      "learning_rate": 1.6777106628229533e-05,
      "loss": 0.38826632499694824,
      "num_input_tokens_seen": 22417387840,
      "step": 78820,
      "train_runtime": 766388.8791,
      "train_tokens_per_second": 29250.67
    },
    {
      "epoch": 2.789160716220669,
      "grad_norm": 0.62109375,
      "learning_rate": 1.677616225314769e-05,
      "loss": 0.3869755506515503,
      "num_input_tokens_seen": 22420179392,
      "step": 78830,
      "train_runtime": 766482.6747,
      "train_tokens_per_second": 29250.732
    },
    {
      "epoch": 2.789514536484398,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6775218037523117e-05,
      "loss": 0.38819022178649903,
      "num_input_tokens_seen": 22423023808,
      "step": 78840,
      "train_runtime": 766579.0447,
      "train_tokens_per_second": 29250.765
    },
    {
      "epoch": 2.7898683567481264,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6774273981310952e-05,
      "loss": 0.38688805103302004,
      "num_input_tokens_seen": 22425856768,
      "step": 78850,
      "train_runtime": 766676.0618,
      "train_tokens_per_second": 29250.759
    },
    {
      "epoch": 2.790222177011855,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6773330084466342e-05,
      "loss": 0.386118221282959,
      "num_input_tokens_seen": 22428683712,
      "step": 78860,
      "train_runtime": 766773.9637,
      "train_tokens_per_second": 29250.711
    },
    {
      "epoch": 2.790575997275584,
      "grad_norm": 0.609375,
      "learning_rate": 1.6772386346944455e-05,
      "loss": 0.3862916469573975,
      "num_input_tokens_seen": 22431505600,
      "step": 78870,
      "train_runtime": 766868.901,
      "train_tokens_per_second": 29250.77
    },
    {
      "epoch": 2.790929817539313,
      "grad_norm": 0.609375,
      "learning_rate": 1.6771442768700477e-05,
      "loss": 0.38166322708129885,
      "num_input_tokens_seen": 22434373568,
      "step": 78880,
      "train_runtime": 766965.4522,
      "train_tokens_per_second": 29250.827
    },
    {
      "epoch": 2.7912836378030415,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6770499349689606e-05,
      "loss": 0.38252265453338624,
      "num_input_tokens_seen": 22437308160,
      "step": 78890,
      "train_runtime": 767066.1437,
      "train_tokens_per_second": 29250.813
    },
    {
      "epoch": 2.7916374580667704,
      "grad_norm": 0.61328125,
      "learning_rate": 1.676955608986706e-05,
      "loss": 0.38495802879333496,
      "num_input_tokens_seen": 22440142976,
      "step": 78900,
      "train_runtime": 767161.2054,
      "train_tokens_per_second": 29250.883
    },
    {
      "epoch": 2.791991278330499,
      "grad_norm": 0.62890625,
      "learning_rate": 1.676861298918808e-05,
      "loss": 0.3879550457000732,
      "num_input_tokens_seen": 22442945728,
      "step": 78910,
      "train_runtime": 767255.816,
      "train_tokens_per_second": 29250.929
    },
    {
      "epoch": 2.792345098594228,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6767670047607924e-05,
      "loss": 0.38320116996765136,
      "num_input_tokens_seen": 22445788992,
      "step": 78920,
      "train_runtime": 767353.7433,
      "train_tokens_per_second": 29250.902
    },
    {
      "epoch": 2.7926989188579565,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6766727265081858e-05,
      "loss": 0.38472561836242675,
      "num_input_tokens_seen": 22448639232,
      "step": 78930,
      "train_runtime": 767452.1006,
      "train_tokens_per_second": 29250.867
    },
    {
      "epoch": 2.7930527391216855,
      "grad_norm": 0.6015625,
      "learning_rate": 1.676578464156517e-05,
      "loss": 0.38680319786071776,
      "num_input_tokens_seen": 22451453376,
      "step": 78940,
      "train_runtime": 767547.9433,
      "train_tokens_per_second": 29250.881
    },
    {
      "epoch": 2.793406559385414,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6764842177013176e-05,
      "loss": 0.3873145341873169,
      "num_input_tokens_seen": 22454338048,
      "step": 78950,
      "train_runtime": 767649.2603,
      "train_tokens_per_second": 29250.778
    },
    {
      "epoch": 2.793760379649143,
      "grad_norm": 0.6484375,
      "learning_rate": 1.67638998713812e-05,
      "loss": 0.38494348526000977,
      "num_input_tokens_seen": 22457192384,
      "step": 78960,
      "train_runtime": 767744.8764,
      "train_tokens_per_second": 29250.853
    },
    {
      "epoch": 2.7941141999128716,
      "grad_norm": 0.6171875,
      "learning_rate": 1.676295772462457e-05,
      "loss": 0.38083271980285643,
      "num_input_tokens_seen": 22460072896,
      "step": 78970,
      "train_runtime": 767846.4679,
      "train_tokens_per_second": 29250.734
    },
    {
      "epoch": 2.7944680201766006,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6762015736698667e-05,
      "loss": 0.39101157188415525,
      "num_input_tokens_seen": 22462947456,
      "step": 78980,
      "train_runtime": 767947.068,
      "train_tokens_per_second": 29250.646
    },
    {
      "epoch": 2.7948218404403296,
      "grad_norm": 0.60546875,
      "learning_rate": 1.676107390755886e-05,
      "loss": 0.38897204399108887,
      "num_input_tokens_seen": 22465819968,
      "step": 78990,
      "train_runtime": 768047.2326,
      "train_tokens_per_second": 29250.571
    },
    {
      "epoch": 2.795175660704058,
      "grad_norm": 0.625,
      "learning_rate": 1.6760132237160538e-05,
      "loss": 0.38744144439697265,
      "num_input_tokens_seen": 22468670784,
      "step": 79000,
      "train_runtime": 768143.3612,
      "train_tokens_per_second": 29250.622
    },
    {
      "epoch": 2.7955294809677866,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6759190725459125e-05,
      "loss": 0.38637101650238037,
      "num_input_tokens_seen": 22471500608,
      "step": 79010,
      "train_runtime": 768240.8708,
      "train_tokens_per_second": 29250.592
    },
    {
      "epoch": 2.7958833012315156,
      "grad_norm": 0.60546875,
      "learning_rate": 1.675824937241005e-05,
      "loss": 0.38822498321533205,
      "num_input_tokens_seen": 22474348096,
      "step": 79020,
      "train_runtime": 768338.8414,
      "train_tokens_per_second": 29250.569
    },
    {
      "epoch": 2.7962371214952446,
      "grad_norm": 0.6171875,
      "learning_rate": 1.675730817796875e-05,
      "loss": 0.3906718254089355,
      "num_input_tokens_seen": 22477195520,
      "step": 79030,
      "train_runtime": 768434.8025,
      "train_tokens_per_second": 29250.621
    },
    {
      "epoch": 2.796590941758973,
      "grad_norm": 0.61328125,
      "learning_rate": 1.67563671420907e-05,
      "loss": 0.3826432228088379,
      "num_input_tokens_seen": 22480027648,
      "step": 79040,
      "train_runtime": 768531.8048,
      "train_tokens_per_second": 29250.615
    },
    {
      "epoch": 2.796944762022702,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6755426264731383e-05,
      "loss": 0.38129429817199706,
      "num_input_tokens_seen": 22482931776,
      "step": 79050,
      "train_runtime": 768632.3431,
      "train_tokens_per_second": 29250.567
    },
    {
      "epoch": 2.7972985822864307,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6754485545846296e-05,
      "loss": 0.38255622386932375,
      "num_input_tokens_seen": 22485762688,
      "step": 79060,
      "train_runtime": 768729.2989,
      "train_tokens_per_second": 29250.56
    },
    {
      "epoch": 2.7976524025501597,
      "grad_norm": 0.59765625,
      "learning_rate": 1.675354498539096e-05,
      "loss": 0.3849782943725586,
      "num_input_tokens_seen": 22488650496,
      "step": 79070,
      "train_runtime": 768827.3218,
      "train_tokens_per_second": 29250.587
    },
    {
      "epoch": 2.798006222813888,
      "grad_norm": 0.62109375,
      "learning_rate": 1.675260458332091e-05,
      "loss": 0.3899967670440674,
      "num_input_tokens_seen": 22491476864,
      "step": 79080,
      "train_runtime": 768925.8452,
      "train_tokens_per_second": 29250.515
    },
    {
      "epoch": 2.798360043077617,
      "grad_norm": 0.640625,
      "learning_rate": 1.6751664339591696e-05,
      "loss": 0.38746883869171145,
      "num_input_tokens_seen": 22494350720,
      "step": 79090,
      "train_runtime": 769027.1071,
      "train_tokens_per_second": 29250.4
    },
    {
      "epoch": 2.7987138633413458,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6750724254158888e-05,
      "loss": 0.38402130603790285,
      "num_input_tokens_seen": 22497189248,
      "step": 79100,
      "train_runtime": 769124.5857,
      "train_tokens_per_second": 29250.384
    },
    {
      "epoch": 2.7990676836050747,
      "grad_norm": 0.6328125,
      "learning_rate": 1.674978432697808e-05,
      "loss": 0.38719642162323,
      "num_input_tokens_seen": 22500000512,
      "step": 79110,
      "train_runtime": 769219.649,
      "train_tokens_per_second": 29250.424
    },
    {
      "epoch": 2.7994215038688033,
      "grad_norm": 0.59375,
      "learning_rate": 1.6748844558004877e-05,
      "loss": 0.38563075065612795,
      "num_input_tokens_seen": 22502837696,
      "step": 79120,
      "train_runtime": 769315.7808,
      "train_tokens_per_second": 29250.456
    },
    {
      "epoch": 2.7997753241325323,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6747904947194892e-05,
      "loss": 0.38919844627380373,
      "num_input_tokens_seen": 22505659008,
      "step": 79130,
      "train_runtime": 769412.7603,
      "train_tokens_per_second": 29250.436
    },
    {
      "epoch": 2.8001291443962613,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6746965494503776e-05,
      "loss": 0.3818293571472168,
      "num_input_tokens_seen": 22508518272,
      "step": 79140,
      "train_runtime": 769514.9417,
      "train_tokens_per_second": 29250.268
    },
    {
      "epoch": 2.80048296465999,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6746026199887178e-05,
      "loss": 0.385685133934021,
      "num_input_tokens_seen": 22511400896,
      "step": 79150,
      "train_runtime": 769614.656,
      "train_tokens_per_second": 29250.224
    },
    {
      "epoch": 2.8008367849237183,
      "grad_norm": 0.625,
      "learning_rate": 1.674508706330078e-05,
      "loss": 0.38182744979858396,
      "num_input_tokens_seen": 22514250368,
      "step": 79160,
      "train_runtime": 769711.907,
      "train_tokens_per_second": 29250.23
    },
    {
      "epoch": 2.8011906051874473,
      "grad_norm": 0.62109375,
      "learning_rate": 1.674414808470027e-05,
      "loss": 0.3875946760177612,
      "num_input_tokens_seen": 22517091392,
      "step": 79170,
      "train_runtime": 769808.3316,
      "train_tokens_per_second": 29250.257
    },
    {
      "epoch": 2.8015444254511763,
      "grad_norm": 0.6328125,
      "learning_rate": 1.674320926404136e-05,
      "loss": 0.38591713905334474,
      "num_input_tokens_seen": 22519973952,
      "step": 79180,
      "train_runtime": 769910.0173,
      "train_tokens_per_second": 29250.138
    },
    {
      "epoch": 2.801898245714905,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6742270601279773e-05,
      "loss": 0.38675990104675295,
      "num_input_tokens_seen": 22522788032,
      "step": 79190,
      "train_runtime": 770005.1674,
      "train_tokens_per_second": 29250.178
    },
    {
      "epoch": 2.8022520659786334,
      "grad_norm": 0.60546875,
      "learning_rate": 1.674133209637126e-05,
      "loss": 0.38520815372467043,
      "num_input_tokens_seen": 22525527232,
      "step": 79200,
      "train_runtime": 770097.6322,
      "train_tokens_per_second": 29250.223
    },
    {
      "epoch": 2.8026058862423624,
      "grad_norm": 0.625,
      "learning_rate": 1.6740393749271583e-05,
      "loss": 0.3826102018356323,
      "num_input_tokens_seen": 22528348864,
      "step": 79210,
      "train_runtime": 770193.5533,
      "train_tokens_per_second": 29250.243
    },
    {
      "epoch": 2.8029597065060914,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6739455559936515e-05,
      "loss": 0.38376452922821047,
      "num_input_tokens_seen": 22531238848,
      "step": 79220,
      "train_runtime": 770295.2241,
      "train_tokens_per_second": 29250.134
    },
    {
      "epoch": 2.80331352676982,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6738517528321858e-05,
      "loss": 0.3872870922088623,
      "num_input_tokens_seen": 22534054976,
      "step": 79230,
      "train_runtime": 770389.2595,
      "train_tokens_per_second": 29250.22
    },
    {
      "epoch": 2.803667347033549,
      "grad_norm": 0.64453125,
      "learning_rate": 1.673757965438342e-05,
      "loss": 0.3831596851348877,
      "num_input_tokens_seen": 22536885504,
      "step": 79240,
      "train_runtime": 770484.0029,
      "train_tokens_per_second": 29250.296
    },
    {
      "epoch": 2.8040211672972775,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6736641938077036e-05,
      "loss": 0.38428702354431155,
      "num_input_tokens_seen": 22539703424,
      "step": 79250,
      "train_runtime": 770580.5287,
      "train_tokens_per_second": 29250.289
    },
    {
      "epoch": 2.8043749875610064,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6735704379358556e-05,
      "loss": 0.38142881393432615,
      "num_input_tokens_seen": 22542576768,
      "step": 79260,
      "train_runtime": 770678.0908,
      "train_tokens_per_second": 29250.315
    },
    {
      "epoch": 2.804728807824735,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6734766978183844e-05,
      "loss": 0.38763766288757323,
      "num_input_tokens_seen": 22545414400,
      "step": 79270,
      "train_runtime": 770775.9565,
      "train_tokens_per_second": 29250.282
    },
    {
      "epoch": 2.805082628088464,
      "grad_norm": 0.578125,
      "learning_rate": 1.6733829734508786e-05,
      "loss": 0.38669252395629883,
      "num_input_tokens_seen": 22548244032,
      "step": 79280,
      "train_runtime": 770870.6481,
      "train_tokens_per_second": 29250.36
    },
    {
      "epoch": 2.8054364483521925,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6732892648289272e-05,
      "loss": 0.38689565658569336,
      "num_input_tokens_seen": 22551078080,
      "step": 79290,
      "train_runtime": 770966.7723,
      "train_tokens_per_second": 29250.389
    },
    {
      "epoch": 2.8057902686159215,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6731955719481236e-05,
      "loss": 0.38439764976501467,
      "num_input_tokens_seen": 22553890176,
      "step": 79300,
      "train_runtime": 771060.6516,
      "train_tokens_per_second": 29250.475
    },
    {
      "epoch": 2.80614408887965,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6731018948040602e-05,
      "loss": 0.38398351669311526,
      "num_input_tokens_seen": 22556735360,
      "step": 79310,
      "train_runtime": 771158.8378,
      "train_tokens_per_second": 29250.44
    },
    {
      "epoch": 2.806497909143379,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6730082333923324e-05,
      "loss": 0.38662004470825195,
      "num_input_tokens_seen": 22559638784,
      "step": 79320,
      "train_runtime": 771261.0718,
      "train_tokens_per_second": 29250.327
    },
    {
      "epoch": 2.806851729407108,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6729145877085377e-05,
      "loss": 0.38518533706665037,
      "num_input_tokens_seen": 22562438080,
      "step": 79330,
      "train_runtime": 771355.3018,
      "train_tokens_per_second": 29250.383
    },
    {
      "epoch": 2.8072055496708366,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6728209577482738e-05,
      "loss": 0.383742618560791,
      "num_input_tokens_seen": 22565311872,
      "step": 79340,
      "train_runtime": 771452.5265,
      "train_tokens_per_second": 29250.422
    },
    {
      "epoch": 2.807559369934565,
      "grad_norm": 0.6171875,
      "learning_rate": 1.672727343507142e-05,
      "loss": 0.3861359119415283,
      "num_input_tokens_seen": 22568143872,
      "step": 79350,
      "train_runtime": 771549.3219,
      "train_tokens_per_second": 29250.423
    },
    {
      "epoch": 2.807913190198294,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6726337449807442e-05,
      "loss": 0.3795956611633301,
      "num_input_tokens_seen": 22570998912,
      "step": 79360,
      "train_runtime": 771645.9935,
      "train_tokens_per_second": 29250.458
    },
    {
      "epoch": 2.808267010462023,
      "grad_norm": 0.625,
      "learning_rate": 1.6725401621646838e-05,
      "loss": 0.38527793884277345,
      "num_input_tokens_seen": 22573867392,
      "step": 79370,
      "train_runtime": 771744.4834,
      "train_tokens_per_second": 29250.442
    },
    {
      "epoch": 2.8086208307257516,
      "grad_norm": 0.59375,
      "learning_rate": 1.672446595054567e-05,
      "loss": 0.3826130867004395,
      "num_input_tokens_seen": 22576701120,
      "step": 79380,
      "train_runtime": 771840.792,
      "train_tokens_per_second": 29250.464
    },
    {
      "epoch": 2.8089746509894806,
      "grad_norm": 0.640625,
      "learning_rate": 1.6723530436460007e-05,
      "loss": 0.3877525806427002,
      "num_input_tokens_seen": 22579582016,
      "step": 79390,
      "train_runtime": 771940.6443,
      "train_tokens_per_second": 29250.412
    },
    {
      "epoch": 2.809328471253209,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6722595079345938e-05,
      "loss": 0.3821340799331665,
      "num_input_tokens_seen": 22582426944,
      "step": 79400,
      "train_runtime": 772036.1112,
      "train_tokens_per_second": 29250.48
    },
    {
      "epoch": 2.809682291516938,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6721659879159576e-05,
      "loss": 0.3870774984359741,
      "num_input_tokens_seen": 22585227584,
      "step": 79410,
      "train_runtime": 772129.5818,
      "train_tokens_per_second": 29250.566
    },
    {
      "epoch": 2.8100361117806667,
      "grad_norm": 0.6171875,
      "learning_rate": 1.672072483585704e-05,
      "loss": 0.3856063365936279,
      "num_input_tokens_seen": 22588139392,
      "step": 79420,
      "train_runtime": 772229.4001,
      "train_tokens_per_second": 29250.556
    },
    {
      "epoch": 2.8103899320443957,
      "grad_norm": 0.625,
      "learning_rate": 1.6719789949394475e-05,
      "loss": 0.38336544036865233,
      "num_input_tokens_seen": 22590980032,
      "step": 79430,
      "train_runtime": 772328.7449,
      "train_tokens_per_second": 29250.472
    },
    {
      "epoch": 2.8107437523081242,
      "grad_norm": 0.609375,
      "learning_rate": 1.6718855219728038e-05,
      "loss": 0.3862900972366333,
      "num_input_tokens_seen": 22593824960,
      "step": 79440,
      "train_runtime": 772425.5995,
      "train_tokens_per_second": 29250.487
    },
    {
      "epoch": 2.811097572571853,
      "grad_norm": 0.59375,
      "learning_rate": 1.6717920646813904e-05,
      "loss": 0.3862600564956665,
      "num_input_tokens_seen": 22596675008,
      "step": 79450,
      "train_runtime": 772521.7824,
      "train_tokens_per_second": 29250.534
    },
    {
      "epoch": 2.8114513928355818,
      "grad_norm": 0.6328125,
      "learning_rate": 1.671698623060827e-05,
      "loss": 0.3853559970855713,
      "num_input_tokens_seen": 22599531776,
      "step": 79460,
      "train_runtime": 772620.4066,
      "train_tokens_per_second": 29250.498
    },
    {
      "epoch": 2.8118052130993108,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6716051971067343e-05,
      "loss": 0.3849400520324707,
      "num_input_tokens_seen": 22602311936,
      "step": 79470,
      "train_runtime": 772713.2379,
      "train_tokens_per_second": 29250.582
    },
    {
      "epoch": 2.8121590333630397,
      "grad_norm": 0.6328125,
      "learning_rate": 1.671511786814735e-05,
      "loss": 0.3867997646331787,
      "num_input_tokens_seen": 22605161536,
      "step": 79480,
      "train_runtime": 772812.6185,
      "train_tokens_per_second": 29250.508
    },
    {
      "epoch": 2.8125128536267683,
      "grad_norm": 0.59765625,
      "learning_rate": 1.671418392180454e-05,
      "loss": 0.38906450271606446,
      "num_input_tokens_seen": 22608042176,
      "step": 79490,
      "train_runtime": 772913.7003,
      "train_tokens_per_second": 29250.409
    },
    {
      "epoch": 2.812866673890497,
      "grad_norm": 0.625,
      "learning_rate": 1.671325013199517e-05,
      "loss": 0.38559508323669434,
      "num_input_tokens_seen": 22610894592,
      "step": 79500,
      "train_runtime": 773010.0466,
      "train_tokens_per_second": 29250.454
    },
    {
      "epoch": 2.813220494154226,
      "grad_norm": 0.62890625,
      "learning_rate": 1.671231649867552e-05,
      "loss": 0.39032411575317383,
      "num_input_tokens_seen": 22613708032,
      "step": 79510,
      "train_runtime": 773107.7141,
      "train_tokens_per_second": 29250.398
    },
    {
      "epoch": 2.813574314417955,
      "grad_norm": 0.61328125,
      "learning_rate": 1.671138302180189e-05,
      "loss": 0.38908555507659914,
      "num_input_tokens_seen": 22616538560,
      "step": 79520,
      "train_runtime": 773204.8063,
      "train_tokens_per_second": 29250.385
    },
    {
      "epoch": 2.8139281346816833,
      "grad_norm": 0.625,
      "learning_rate": 1.6710449701330584e-05,
      "loss": 0.38480467796325685,
      "num_input_tokens_seen": 22619327744,
      "step": 79530,
      "train_runtime": 773296.5492,
      "train_tokens_per_second": 29250.522
    },
    {
      "epoch": 2.814281954945412,
      "grad_norm": 0.62890625,
      "learning_rate": 1.670951653721794e-05,
      "loss": 0.386159610748291,
      "num_input_tokens_seen": 22622145408,
      "step": 79540,
      "train_runtime": 773391.6603,
      "train_tokens_per_second": 29250.568
    },
    {
      "epoch": 2.814635775209141,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6708583529420303e-05,
      "loss": 0.389597749710083,
      "num_input_tokens_seen": 22624937088,
      "step": 79550,
      "train_runtime": 773485.379,
      "train_tokens_per_second": 29250.633
    },
    {
      "epoch": 2.81498959547287,
      "grad_norm": 0.65625,
      "learning_rate": 1.6707650677894036e-05,
      "loss": 0.3874568700790405,
      "num_input_tokens_seen": 22627746688,
      "step": 79560,
      "train_runtime": 773579.6011,
      "train_tokens_per_second": 29250.702
    },
    {
      "epoch": 2.8153434157365984,
      "grad_norm": 0.63671875,
      "learning_rate": 1.670671798259552e-05,
      "loss": 0.38398401737213134,
      "num_input_tokens_seen": 22630539840,
      "step": 79570,
      "train_runtime": 773674.6064,
      "train_tokens_per_second": 29250.721
    },
    {
      "epoch": 2.8156972360003274,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6705785443481153e-05,
      "loss": 0.3836381196975708,
      "num_input_tokens_seen": 22633376000,
      "step": 79580,
      "train_runtime": 773768.2135,
      "train_tokens_per_second": 29250.847
    },
    {
      "epoch": 2.816051056264056,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6704853060507353e-05,
      "loss": 0.3806574821472168,
      "num_input_tokens_seen": 22636252992,
      "step": 79590,
      "train_runtime": 773867.8512,
      "train_tokens_per_second": 29250.799
    },
    {
      "epoch": 2.816404876527785,
      "grad_norm": 0.65234375,
      "learning_rate": 1.6703920833630547e-05,
      "loss": 0.38527727127075195,
      "num_input_tokens_seen": 22639025728,
      "step": 79600,
      "train_runtime": 773960.6244,
      "train_tokens_per_second": 29250.875
    },
    {
      "epoch": 2.8167586967915135,
      "grad_norm": 0.625,
      "learning_rate": 1.670298876280719e-05,
      "loss": 0.38628311157226564,
      "num_input_tokens_seen": 22641890048,
      "step": 79610,
      "train_runtime": 774057.9569,
      "train_tokens_per_second": 29250.898
    },
    {
      "epoch": 2.8171125170552425,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6702056847993747e-05,
      "loss": 0.3895313262939453,
      "num_input_tokens_seen": 22644749760,
      "step": 79620,
      "train_runtime": 774154.8705,
      "train_tokens_per_second": 29250.93
    },
    {
      "epoch": 2.8174663373189714,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6701125089146698e-05,
      "loss": 0.38973727226257326,
      "num_input_tokens_seen": 22647593088,
      "step": 79630,
      "train_runtime": 774251.9179,
      "train_tokens_per_second": 29250.936
    },
    {
      "epoch": 2.8178201575827,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6700193486222547e-05,
      "loss": 0.3906029224395752,
      "num_input_tokens_seen": 22650394176,
      "step": 79640,
      "train_runtime": 774346.1597,
      "train_tokens_per_second": 29250.993
    },
    {
      "epoch": 2.8181739778464285,
      "grad_norm": 0.62109375,
      "learning_rate": 1.669926203917781e-05,
      "loss": 0.38452539443969724,
      "num_input_tokens_seen": 22653288576,
      "step": 79650,
      "train_runtime": 774442.431,
      "train_tokens_per_second": 29251.094
    },
    {
      "epoch": 2.8185277981101575,
      "grad_norm": 0.609375,
      "learning_rate": 1.669833074796902e-05,
      "loss": 0.38055756092071535,
      "num_input_tokens_seen": 22656080832,
      "step": 79660,
      "train_runtime": 774536.5878,
      "train_tokens_per_second": 29251.143
    },
    {
      "epoch": 2.8188816183738865,
      "grad_norm": 0.63671875,
      "learning_rate": 1.669739961255273e-05,
      "loss": 0.38433547019958497,
      "num_input_tokens_seen": 22658895168,
      "step": 79670,
      "train_runtime": 774633.1622,
      "train_tokens_per_second": 29251.13
    },
    {
      "epoch": 2.819235438637615,
      "grad_norm": 0.625,
      "learning_rate": 1.669646863288551e-05,
      "loss": 0.38738934993743895,
      "num_input_tokens_seen": 22661759104,
      "step": 79680,
      "train_runtime": 774728.8904,
      "train_tokens_per_second": 29251.212
    },
    {
      "epoch": 2.8195892589013436,
      "grad_norm": 0.61328125,
      "learning_rate": 1.669553780892394e-05,
      "loss": 0.3875512838363647,
      "num_input_tokens_seen": 22664598336,
      "step": 79690,
      "train_runtime": 774828.0474,
      "train_tokens_per_second": 29251.133
    },
    {
      "epoch": 2.8199430791650726,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6694607140624624e-05,
      "loss": 0.3931875467300415,
      "num_input_tokens_seen": 22667450240,
      "step": 79700,
      "train_runtime": 774925.6416,
      "train_tokens_per_second": 29251.129
    },
    {
      "epoch": 2.8202968994288016,
      "grad_norm": 0.625,
      "learning_rate": 1.6693676627944183e-05,
      "loss": 0.3834944486618042,
      "num_input_tokens_seen": 22670263360,
      "step": 79710,
      "train_runtime": 775021.9081,
      "train_tokens_per_second": 29251.126
    },
    {
      "epoch": 2.82065071969253,
      "grad_norm": 0.625,
      "learning_rate": 1.6692746270839252e-05,
      "loss": 0.3909069299697876,
      "num_input_tokens_seen": 22673060800,
      "step": 79720,
      "train_runtime": 775117.582,
      "train_tokens_per_second": 29251.124
    },
    {
      "epoch": 2.821004539956259,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6691816069266487e-05,
      "loss": 0.38351175785064695,
      "num_input_tokens_seen": 22675910144,
      "step": 79730,
      "train_runtime": 775214.9135,
      "train_tokens_per_second": 29251.127
    },
    {
      "epoch": 2.8213583602199876,
      "grad_norm": 0.640625,
      "learning_rate": 1.6690886023182548e-05,
      "loss": 0.38737714290618896,
      "num_input_tokens_seen": 22678804544,
      "step": 79740,
      "train_runtime": 775312.7158,
      "train_tokens_per_second": 29251.171
    },
    {
      "epoch": 2.8217121804837166,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6689956132544133e-05,
      "loss": 0.3835658073425293,
      "num_input_tokens_seen": 22681628544,
      "step": 79750,
      "train_runtime": 775411.3488,
      "train_tokens_per_second": 29251.092
    },
    {
      "epoch": 2.822066000747445,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6689026397307936e-05,
      "loss": 0.3828464984893799,
      "num_input_tokens_seen": 22684480640,
      "step": 79760,
      "train_runtime": 775508.2772,
      "train_tokens_per_second": 29251.114
    },
    {
      "epoch": 2.822419821011174,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6688096817430687e-05,
      "loss": 0.38687493801116946,
      "num_input_tokens_seen": 22687366912,
      "step": 79770,
      "train_runtime": 775607.9198,
      "train_tokens_per_second": 29251.077
    },
    {
      "epoch": 2.8227736412749027,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6687167392869115e-05,
      "loss": 0.38324871063232424,
      "num_input_tokens_seen": 22690203008,
      "step": 79780,
      "train_runtime": 775704.5829,
      "train_tokens_per_second": 29251.088
    },
    {
      "epoch": 2.8231274615386317,
      "grad_norm": 0.62890625,
      "learning_rate": 1.668623812357998e-05,
      "loss": 0.38308563232421877,
      "num_input_tokens_seen": 22693049152,
      "step": 79790,
      "train_runtime": 775799.6588,
      "train_tokens_per_second": 29251.172
    },
    {
      "epoch": 2.8234812818023602,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6685309009520046e-05,
      "loss": 0.38517184257507325,
      "num_input_tokens_seen": 22695839424,
      "step": 79800,
      "train_runtime": 775893.286,
      "train_tokens_per_second": 29251.238
    },
    {
      "epoch": 2.8238351020660892,
      "grad_norm": 0.61328125,
      "learning_rate": 1.668438005064611e-05,
      "loss": 0.38472585678100585,
      "num_input_tokens_seen": 22698688832,
      "step": 79810,
      "train_runtime": 775994.041,
      "train_tokens_per_second": 29251.112
    },
    {
      "epoch": 2.824188922329818,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6683451246914973e-05,
      "loss": 0.3862781047821045,
      "num_input_tokens_seen": 22701506496,
      "step": 79820,
      "train_runtime": 776089.4126,
      "train_tokens_per_second": 29251.148
    },
    {
      "epoch": 2.8245427425935468,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6682522598283452e-05,
      "loss": 0.38753132820129393,
      "num_input_tokens_seen": 22704423808,
      "step": 79830,
      "train_runtime": 776189.2958,
      "train_tokens_per_second": 29251.143
    },
    {
      "epoch": 2.8248965628572753,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6681594104708393e-05,
      "loss": 0.38221988677978513,
      "num_input_tokens_seen": 22707222912,
      "step": 79840,
      "train_runtime": 776284.2988,
      "train_tokens_per_second": 29251.169
    },
    {
      "epoch": 2.8252503831210043,
      "grad_norm": 0.640625,
      "learning_rate": 1.6680665766146643e-05,
      "loss": 0.38983840942382814,
      "num_input_tokens_seen": 22710062464,
      "step": 79850,
      "train_runtime": 776381.5522,
      "train_tokens_per_second": 29251.162
    },
    {
      "epoch": 2.8256042033847333,
      "grad_norm": 0.60546875,
      "learning_rate": 1.667973758255508e-05,
      "loss": 0.38589110374450686,
      "num_input_tokens_seen": 22712959424,
      "step": 79860,
      "train_runtime": 776480.9717,
      "train_tokens_per_second": 29251.147
    },
    {
      "epoch": 2.825958023648462,
      "grad_norm": 0.62109375,
      "learning_rate": 1.667880955389059e-05,
      "loss": 0.38367085456848143,
      "num_input_tokens_seen": 22715825792,
      "step": 79870,
      "train_runtime": 776579.961,
      "train_tokens_per_second": 29251.11
    },
    {
      "epoch": 2.826311843912191,
      "grad_norm": 0.57421875,
      "learning_rate": 1.6677881680110082e-05,
      "loss": 0.3864089012145996,
      "num_input_tokens_seen": 22718702272,
      "step": 79880,
      "train_runtime": 776679.1135,
      "train_tokens_per_second": 29251.079
    },
    {
      "epoch": 2.8266656641759194,
      "grad_norm": 0.68359375,
      "learning_rate": 1.6676953961170472e-05,
      "loss": 0.3819018602371216,
      "num_input_tokens_seen": 22721534784,
      "step": 79890,
      "train_runtime": 776775.0115,
      "train_tokens_per_second": 29251.114
    },
    {
      "epoch": 2.8270194844396483,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6676026397028705e-05,
      "loss": 0.3894267797470093,
      "num_input_tokens_seen": 22724360256,
      "step": 79900,
      "train_runtime": 776871.0666,
      "train_tokens_per_second": 29251.135
    },
    {
      "epoch": 2.827373304703377,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6675098987641735e-05,
      "loss": 0.3883958101272583,
      "num_input_tokens_seen": 22727262016,
      "step": 79910,
      "train_runtime": 776974.9496,
      "train_tokens_per_second": 29250.959
    },
    {
      "epoch": 2.827727124967106,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6674171732966538e-05,
      "loss": 0.386810827255249,
      "num_input_tokens_seen": 22730093376,
      "step": 79920,
      "train_runtime": 777071.0968,
      "train_tokens_per_second": 29250.983
    },
    {
      "epoch": 2.8280809452308344,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6673244632960093e-05,
      "loss": 0.38858642578125,
      "num_input_tokens_seen": 22732973632,
      "step": 79930,
      "train_runtime": 777173.3332,
      "train_tokens_per_second": 29250.841
    },
    {
      "epoch": 2.8284347654945634,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6672317687579418e-05,
      "loss": 0.38652629852294923,
      "num_input_tokens_seen": 22735793600,
      "step": 79940,
      "train_runtime": 777268.3467,
      "train_tokens_per_second": 29250.893
    },
    {
      "epoch": 2.828788585758292,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6671390896781527e-05,
      "loss": 0.3870053291320801,
      "num_input_tokens_seen": 22738639808,
      "step": 79950,
      "train_runtime": 777364.8108,
      "train_tokens_per_second": 29250.925
    },
    {
      "epoch": 2.829142406022021,
      "grad_norm": 0.609375,
      "learning_rate": 1.6670464260523466e-05,
      "loss": 0.3891130447387695,
      "num_input_tokens_seen": 22741458816,
      "step": 79960,
      "train_runtime": 777455.8513,
      "train_tokens_per_second": 29251.126
    },
    {
      "epoch": 2.82949622628575,
      "grad_norm": 0.62109375,
      "learning_rate": 1.666953777876229e-05,
      "loss": 0.38746280670166017,
      "num_input_tokens_seen": 22744280000,
      "step": 79970,
      "train_runtime": 777549.9918,
      "train_tokens_per_second": 29251.212
    },
    {
      "epoch": 2.8298500465494785,
      "grad_norm": 0.609375,
      "learning_rate": 1.6668611451455068e-05,
      "loss": 0.3845673084259033,
      "num_input_tokens_seen": 22747096128,
      "step": 79980,
      "train_runtime": 777646.0949,
      "train_tokens_per_second": 29251.219
    },
    {
      "epoch": 2.830203866813207,
      "grad_norm": 0.609375,
      "learning_rate": 1.6667685278558893e-05,
      "loss": 0.38469481468200684,
      "num_input_tokens_seen": 22749936704,
      "step": 79990,
      "train_runtime": 777744.0866,
      "train_tokens_per_second": 29251.186
    },
    {
      "epoch": 2.830557687076936,
      "grad_norm": 0.61328125,
      "learning_rate": 1.666675926003087e-05,
      "loss": 0.3861361503601074,
      "num_input_tokens_seen": 22752762176,
      "step": 80000,
      "train_runtime": 777842.3534,
      "train_tokens_per_second": 29251.123
    },
    {
      "epoch": 2.830557687076936,
      "eval_loss": 0.33632194995880127,
      "eval_runtime": 8.4203,
      "eval_samples_per_second": 473.62,
      "eval_steps_per_second": 3.8,
      "num_input_tokens_seen": 22752762176,
      "step": 80000
    },
    {
      "epoch": 2.830911507340665,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6665833395828124e-05,
      "loss": 0.3849381446838379,
      "num_input_tokens_seen": 22755631936,
      "step": 80010,
      "train_runtime": 777968.77,
      "train_tokens_per_second": 29250.058
    },
    {
      "epoch": 2.8312653276043935,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6664907685907795e-05,
      "loss": 0.3885098218917847,
      "num_input_tokens_seen": 22758462976,
      "step": 80020,
      "train_runtime": 778067.0432,
      "train_tokens_per_second": 29250.003
    },
    {
      "epoch": 2.831619147868122,
      "grad_norm": 0.64453125,
      "learning_rate": 1.666398213022704e-05,
      "loss": 0.3882602691650391,
      "num_input_tokens_seen": 22761258944,
      "step": 80030,
      "train_runtime": 778160.0273,
      "train_tokens_per_second": 29250.1
    },
    {
      "epoch": 2.831972968131851,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6663056728743024e-05,
      "loss": 0.38812813758850095,
      "num_input_tokens_seen": 22764086592,
      "step": 80040,
      "train_runtime": 778256.4386,
      "train_tokens_per_second": 29250.11
    },
    {
      "epoch": 2.83232678839558,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6662131481412947e-05,
      "loss": 0.384420108795166,
      "num_input_tokens_seen": 22766929536,
      "step": 80050,
      "train_runtime": 778354.7476,
      "train_tokens_per_second": 29250.068
    },
    {
      "epoch": 2.8326806086593086,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6661206388194006e-05,
      "loss": 0.3842109203338623,
      "num_input_tokens_seen": 22769697472,
      "step": 80060,
      "train_runtime": 778449.7291,
      "train_tokens_per_second": 29250.055
    },
    {
      "epoch": 2.8330344289230376,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6660281449043436e-05,
      "loss": 0.3840348720550537,
      "num_input_tokens_seen": 22772522304,
      "step": 80070,
      "train_runtime": 778548.0338,
      "train_tokens_per_second": 29249.99
    },
    {
      "epoch": 2.833388249186766,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6659356663918466e-05,
      "loss": 0.38687548637390134,
      "num_input_tokens_seen": 22775296320,
      "step": 80080,
      "train_runtime": 778640.3477,
      "train_tokens_per_second": 29250.085
    },
    {
      "epoch": 2.833742069450495,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6658432032776354e-05,
      "loss": 0.38970940113067626,
      "num_input_tokens_seen": 22778134208,
      "step": 80090,
      "train_runtime": 778736.5431,
      "train_tokens_per_second": 29250.116
    },
    {
      "epoch": 2.8340958897142237,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6657507555574375e-05,
      "loss": 0.39075944423675535,
      "num_input_tokens_seen": 22780978304,
      "step": 80100,
      "train_runtime": 778832.5269,
      "train_tokens_per_second": 29250.163
    },
    {
      "epoch": 2.8344497099779526,
      "grad_norm": 0.5859375,
      "learning_rate": 1.6656583232269816e-05,
      "loss": 0.38437342643737793,
      "num_input_tokens_seen": 22783811328,
      "step": 80110,
      "train_runtime": 778926.3621,
      "train_tokens_per_second": 29250.276
    },
    {
      "epoch": 2.834803530241681,
      "grad_norm": 0.625,
      "learning_rate": 1.665565906281999e-05,
      "loss": 0.3807594537734985,
      "num_input_tokens_seen": 22786625408,
      "step": 80120,
      "train_runtime": 779021.4333,
      "train_tokens_per_second": 29250.319
    },
    {
      "epoch": 2.83515735050541,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6654735047182213e-05,
      "loss": 0.3859172105789185,
      "num_input_tokens_seen": 22789406720,
      "step": 80130,
      "train_runtime": 779116.6409,
      "train_tokens_per_second": 29250.314
    },
    {
      "epoch": 2.8355111707691387,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6653811185313824e-05,
      "loss": 0.38521318435668944,
      "num_input_tokens_seen": 22792252736,
      "step": 80140,
      "train_runtime": 779213.4352,
      "train_tokens_per_second": 29250.333
    },
    {
      "epoch": 2.8358649910328677,
      "grad_norm": 0.62890625,
      "learning_rate": 1.665288747717218e-05,
      "loss": 0.3856534481048584,
      "num_input_tokens_seen": 22795048704,
      "step": 80150,
      "train_runtime": 779306.3575,
      "train_tokens_per_second": 29250.433
    },
    {
      "epoch": 2.8362188112965967,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6651963922714655e-05,
      "loss": 0.38876359462738036,
      "num_input_tokens_seen": 22797884928,
      "step": 80160,
      "train_runtime": 779404.6228,
      "train_tokens_per_second": 29250.385
    },
    {
      "epoch": 2.8365726315603252,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6651040521898635e-05,
      "loss": 0.3902108192443848,
      "num_input_tokens_seen": 22800727424,
      "step": 80170,
      "train_runtime": 779499.9339,
      "train_tokens_per_second": 29250.455
    },
    {
      "epoch": 2.836926451824054,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6650117274681528e-05,
      "loss": 0.3851343631744385,
      "num_input_tokens_seen": 22803598144,
      "step": 80180,
      "train_runtime": 779596.4794,
      "train_tokens_per_second": 29250.515
    },
    {
      "epoch": 2.8372802720877828,
      "grad_norm": 0.609375,
      "learning_rate": 1.6649194181020756e-05,
      "loss": 0.3855489730834961,
      "num_input_tokens_seen": 22806523392,
      "step": 80190,
      "train_runtime": 779698.1872,
      "train_tokens_per_second": 29250.451
    },
    {
      "epoch": 2.8376340923515118,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6648271240873753e-05,
      "loss": 0.3873851776123047,
      "num_input_tokens_seen": 22809368256,
      "step": 80200,
      "train_runtime": 779796.3105,
      "train_tokens_per_second": 29250.418
    },
    {
      "epoch": 2.8379879126152403,
      "grad_norm": 0.62890625,
      "learning_rate": 1.664734845419798e-05,
      "loss": 0.3816483736038208,
      "num_input_tokens_seen": 22812157952,
      "step": 80210,
      "train_runtime": 779888.013,
      "train_tokens_per_second": 29250.556
    },
    {
      "epoch": 2.8383417328789693,
      "grad_norm": 0.62109375,
      "learning_rate": 1.66464258209509e-05,
      "loss": 0.3907389402389526,
      "num_input_tokens_seen": 22815019264,
      "step": 80220,
      "train_runtime": 779987.0567,
      "train_tokens_per_second": 29250.51
    },
    {
      "epoch": 2.838695553142698,
      "grad_norm": 0.6015625,
      "learning_rate": 1.664550334109001e-05,
      "loss": 0.38678407669067383,
      "num_input_tokens_seen": 22817832448,
      "step": 80230,
      "train_runtime": 780082.5395,
      "train_tokens_per_second": 29250.536
    },
    {
      "epoch": 2.839049373406427,
      "grad_norm": 0.63671875,
      "learning_rate": 1.664458101457281e-05,
      "loss": 0.3886272430419922,
      "num_input_tokens_seen": 22820726848,
      "step": 80240,
      "train_runtime": 780183.7718,
      "train_tokens_per_second": 29250.451
    },
    {
      "epoch": 2.8394031936701554,
      "grad_norm": 0.61328125,
      "learning_rate": 1.664365884135682e-05,
      "loss": 0.39076919555664064,
      "num_input_tokens_seen": 22823606464,
      "step": 80250,
      "train_runtime": 780281.8615,
      "train_tokens_per_second": 29250.464
    },
    {
      "epoch": 2.8397570139338844,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6642736821399583e-05,
      "loss": 0.3917582035064697,
      "num_input_tokens_seen": 22826425216,
      "step": 80260,
      "train_runtime": 780377.1104,
      "train_tokens_per_second": 29250.506
    },
    {
      "epoch": 2.840110834197613,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6641814954658643e-05,
      "loss": 0.38414387702941893,
      "num_input_tokens_seen": 22829255552,
      "step": 80270,
      "train_runtime": 780472.5692,
      "train_tokens_per_second": 29250.555
    },
    {
      "epoch": 2.840464654461342,
      "grad_norm": 0.63671875,
      "learning_rate": 1.664089324109158e-05,
      "loss": 0.3921938180923462,
      "num_input_tokens_seen": 22832113664,
      "step": 80280,
      "train_runtime": 780570.7698,
      "train_tokens_per_second": 29250.536
    },
    {
      "epoch": 2.8408184747250704,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6639971680655975e-05,
      "loss": 0.38543407917022704,
      "num_input_tokens_seen": 22834964416,
      "step": 80290,
      "train_runtime": 780667.5209,
      "train_tokens_per_second": 29250.563
    },
    {
      "epoch": 2.8411722949887994,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6639050273309432e-05,
      "loss": 0.3855253219604492,
      "num_input_tokens_seen": 22837856320,
      "step": 80300,
      "train_runtime": 780767.5143,
      "train_tokens_per_second": 29250.521
    },
    {
      "epoch": 2.8415261152525284,
      "grad_norm": 0.609375,
      "learning_rate": 1.6638129019009573e-05,
      "loss": 0.3859180212020874,
      "num_input_tokens_seen": 22840750912,
      "step": 80310,
      "train_runtime": 780869.1977,
      "train_tokens_per_second": 29250.419
    },
    {
      "epoch": 2.841879935516257,
      "grad_norm": 0.6171875,
      "learning_rate": 1.663720791771403e-05,
      "loss": 0.3858411550521851,
      "num_input_tokens_seen": 22843562560,
      "step": 80320,
      "train_runtime": 780967.7743,
      "train_tokens_per_second": 29250.327
    },
    {
      "epoch": 2.8422337557799855,
      "grad_norm": 0.6171875,
      "learning_rate": 1.663628696938046e-05,
      "loss": 0.3882434368133545,
      "num_input_tokens_seen": 22846482752,
      "step": 80330,
      "train_runtime": 781069.9396,
      "train_tokens_per_second": 29250.239
    },
    {
      "epoch": 2.8425875760437145,
      "grad_norm": 0.62109375,
      "learning_rate": 1.663536617396653e-05,
      "loss": 0.3832517623901367,
      "num_input_tokens_seen": 22849334080,
      "step": 80340,
      "train_runtime": 781171.1469,
      "train_tokens_per_second": 29250.1
    },
    {
      "epoch": 2.8429413963074435,
      "grad_norm": 0.6484375,
      "learning_rate": 1.6634445531429922e-05,
      "loss": 0.38891263008117677,
      "num_input_tokens_seen": 22852179392,
      "step": 80350,
      "train_runtime": 781268.8711,
      "train_tokens_per_second": 29250.083
    },
    {
      "epoch": 2.843295216571172,
      "grad_norm": 0.6171875,
      "learning_rate": 1.663352504172834e-05,
      "loss": 0.3878302097320557,
      "num_input_tokens_seen": 22854962240,
      "step": 80360,
      "train_runtime": 781361.3607,
      "train_tokens_per_second": 29250.182
    },
    {
      "epoch": 2.8436490368349006,
      "grad_norm": 0.62109375,
      "learning_rate": 1.663260470481951e-05,
      "loss": 0.3905947208404541,
      "num_input_tokens_seen": 22857796608,
      "step": 80370,
      "train_runtime": 781457.9163,
      "train_tokens_per_second": 29250.195
    },
    {
      "epoch": 2.8440028570986295,
      "grad_norm": 0.60546875,
      "learning_rate": 1.663168452066115e-05,
      "loss": 0.38461761474609374,
      "num_input_tokens_seen": 22860663616,
      "step": 80380,
      "train_runtime": 781555.6871,
      "train_tokens_per_second": 29250.204
    },
    {
      "epoch": 2.8443566773623585,
      "grad_norm": 0.609375,
      "learning_rate": 1.663076448921102e-05,
      "loss": 0.3854233980178833,
      "num_input_tokens_seen": 22863503296,
      "step": 80390,
      "train_runtime": 781651.8041,
      "train_tokens_per_second": 29250.241
    },
    {
      "epoch": 2.844710497626087,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6629844610426892e-05,
      "loss": 0.39031667709350587,
      "num_input_tokens_seen": 22866358464,
      "step": 80400,
      "train_runtime": 781751.1409,
      "train_tokens_per_second": 29250.176
    },
    {
      "epoch": 2.845064317889816,
      "grad_norm": 0.66015625,
      "learning_rate": 1.6628924884266542e-05,
      "loss": 0.389051628112793,
      "num_input_tokens_seen": 22869242816,
      "step": 80410,
      "train_runtime": 781850.0502,
      "train_tokens_per_second": 29250.165
    },
    {
      "epoch": 2.8454181381535446,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6628005310687773e-05,
      "loss": 0.39028325080871584,
      "num_input_tokens_seen": 22872063808,
      "step": 80420,
      "train_runtime": 781944.7537,
      "train_tokens_per_second": 29250.23
    },
    {
      "epoch": 2.8457719584172736,
      "grad_norm": 0.625,
      "learning_rate": 1.6627085889648396e-05,
      "loss": 0.38775925636291503,
      "num_input_tokens_seen": 22874942272,
      "step": 80430,
      "train_runtime": 782040.7286,
      "train_tokens_per_second": 29250.321
    },
    {
      "epoch": 2.846125778681002,
      "grad_norm": 0.609375,
      "learning_rate": 1.6626166621106252e-05,
      "loss": 0.38816003799438475,
      "num_input_tokens_seen": 22877804864,
      "step": 80440,
      "train_runtime": 782138.7298,
      "train_tokens_per_second": 29250.316
    },
    {
      "epoch": 2.846479598944731,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6625247505019183e-05,
      "loss": 0.385836124420166,
      "num_input_tokens_seen": 22880627520,
      "step": 80450,
      "train_runtime": 782232.8879,
      "train_tokens_per_second": 29250.403
    },
    {
      "epoch": 2.84683341920846,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6624328541345054e-05,
      "loss": 0.3842759132385254,
      "num_input_tokens_seen": 22883450304,
      "step": 80460,
      "train_runtime": 782328.0565,
      "train_tokens_per_second": 29250.453
    },
    {
      "epoch": 2.8471872394721887,
      "grad_norm": 0.62890625,
      "learning_rate": 1.662340973004175e-05,
      "loss": 0.38395133018493655,
      "num_input_tokens_seen": 22886343104,
      "step": 80470,
      "train_runtime": 782428.7274,
      "train_tokens_per_second": 29250.387
    },
    {
      "epoch": 2.847541059735917,
      "grad_norm": 0.609375,
      "learning_rate": 1.6622491071067173e-05,
      "loss": 0.3870535850524902,
      "num_input_tokens_seen": 22889160512,
      "step": 80480,
      "train_runtime": 782522.6449,
      "train_tokens_per_second": 29250.477
    },
    {
      "epoch": 2.847894879999646,
      "grad_norm": 0.62890625,
      "learning_rate": 1.662157256437923e-05,
      "loss": 0.38376150131225584,
      "num_input_tokens_seen": 22892025600,
      "step": 80490,
      "train_runtime": 782620.9043,
      "train_tokens_per_second": 29250.465
    },
    {
      "epoch": 2.848248700263375,
      "grad_norm": 0.609375,
      "learning_rate": 1.6620654209935848e-05,
      "loss": 0.3873354434967041,
      "num_input_tokens_seen": 22894854976,
      "step": 80500,
      "train_runtime": 782717.777,
      "train_tokens_per_second": 29250.46
    },
    {
      "epoch": 2.8486025205271037,
      "grad_norm": 0.62109375,
      "learning_rate": 1.661973600769498e-05,
      "loss": 0.3883077383041382,
      "num_input_tokens_seen": 22897708736,
      "step": 80510,
      "train_runtime": 782816.3256,
      "train_tokens_per_second": 29250.423
    },
    {
      "epoch": 2.8489563407908323,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6618817957614587e-05,
      "loss": 0.38348290920257566,
      "num_input_tokens_seen": 22900593600,
      "step": 80520,
      "train_runtime": 782915.7654,
      "train_tokens_per_second": 29250.393
    },
    {
      "epoch": 2.8493101610545613,
      "grad_norm": 0.60546875,
      "learning_rate": 1.661790005965265e-05,
      "loss": 0.3868986129760742,
      "num_input_tokens_seen": 22903405696,
      "step": 80530,
      "train_runtime": 783011.7308,
      "train_tokens_per_second": 29250.399
    },
    {
      "epoch": 2.8496639813182902,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6616982313767158e-05,
      "loss": 0.38521132469177244,
      "num_input_tokens_seen": 22906273536,
      "step": 80540,
      "train_runtime": 783110.5185,
      "train_tokens_per_second": 29250.371
    },
    {
      "epoch": 2.850017801582019,
      "grad_norm": 0.625,
      "learning_rate": 1.661606471991613e-05,
      "loss": 0.38873720169067383,
      "num_input_tokens_seen": 22909100416,
      "step": 80550,
      "train_runtime": 783205.2885,
      "train_tokens_per_second": 29250.441
    },
    {
      "epoch": 2.8503716218457478,
      "grad_norm": 0.63671875,
      "learning_rate": 1.661514727805759e-05,
      "loss": 0.39095287322998046,
      "num_input_tokens_seen": 22911954944,
      "step": 80560,
      "train_runtime": 783305.2964,
      "train_tokens_per_second": 29250.351
    },
    {
      "epoch": 2.8507254421094763,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6614229988149582e-05,
      "loss": 0.38513381481170655,
      "num_input_tokens_seen": 22914805888,
      "step": 80570,
      "train_runtime": 783402.2149,
      "train_tokens_per_second": 29250.372
    },
    {
      "epoch": 2.8510792623732053,
      "grad_norm": 0.625,
      "learning_rate": 1.6613312850150166e-05,
      "loss": 0.38893725872039797,
      "num_input_tokens_seen": 22917642432,
      "step": 80580,
      "train_runtime": 783497.692,
      "train_tokens_per_second": 29250.427
    },
    {
      "epoch": 2.851433082636934,
      "grad_norm": 0.6171875,
      "learning_rate": 1.661239586401742e-05,
      "loss": 0.3853666543960571,
      "num_input_tokens_seen": 22920466432,
      "step": 80590,
      "train_runtime": 783595.5057,
      "train_tokens_per_second": 29250.38
    },
    {
      "epoch": 2.851786902900663,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6611479029709438e-05,
      "loss": 0.3924888610839844,
      "num_input_tokens_seen": 22923306688,
      "step": 80600,
      "train_runtime": 783690.092,
      "train_tokens_per_second": 29250.474
    },
    {
      "epoch": 2.8521407231643914,
      "grad_norm": 0.625,
      "learning_rate": 1.6610562347184323e-05,
      "loss": 0.38466897010803225,
      "num_input_tokens_seen": 22926170240,
      "step": 80610,
      "train_runtime": 783786.5151,
      "train_tokens_per_second": 29250.529
    },
    {
      "epoch": 2.8524945434281204,
      "grad_norm": 0.6328125,
      "learning_rate": 1.66096458164002e-05,
      "loss": 0.38309307098388673,
      "num_input_tokens_seen": 22928970432,
      "step": 80620,
      "train_runtime": 783882.9271,
      "train_tokens_per_second": 29250.504
    },
    {
      "epoch": 2.852848363691849,
      "grad_norm": 0.609375,
      "learning_rate": 1.6608729437315218e-05,
      "loss": 0.3849660873413086,
      "num_input_tokens_seen": 22931792896,
      "step": 80630,
      "train_runtime": 783978.8945,
      "train_tokens_per_second": 29250.523
    },
    {
      "epoch": 2.853202183955578,
      "grad_norm": 0.59765625,
      "learning_rate": 1.660781320988753e-05,
      "loss": 0.3867595434188843,
      "num_input_tokens_seen": 22934705536,
      "step": 80640,
      "train_runtime": 784080.3947,
      "train_tokens_per_second": 29250.451
    },
    {
      "epoch": 2.853556004219307,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6606897134075307e-05,
      "loss": 0.38904967308044436,
      "num_input_tokens_seen": 22937512192,
      "step": 80650,
      "train_runtime": 784174.6519,
      "train_tokens_per_second": 29250.515
    },
    {
      "epoch": 2.8539098244830354,
      "grad_norm": 0.640625,
      "learning_rate": 1.660598120983674e-05,
      "loss": 0.39061775207519533,
      "num_input_tokens_seen": 22940352448,
      "step": 80660,
      "train_runtime": 784271.1141,
      "train_tokens_per_second": 29250.539
    },
    {
      "epoch": 2.854263644746764,
      "grad_norm": 0.62109375,
      "learning_rate": 1.660506543713004e-05,
      "loss": 0.3834700107574463,
      "num_input_tokens_seen": 22943216320,
      "step": 80670,
      "train_runtime": 784369.9723,
      "train_tokens_per_second": 29250.503
    },
    {
      "epoch": 2.854617465010493,
      "grad_norm": 0.62109375,
      "learning_rate": 1.660414981591342e-05,
      "loss": 0.38445167541503905,
      "num_input_tokens_seen": 22946051328,
      "step": 80680,
      "train_runtime": 784467.4124,
      "train_tokens_per_second": 29250.484
    },
    {
      "epoch": 2.854971285274222,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6603234346145123e-05,
      "loss": 0.38368687629699705,
      "num_input_tokens_seen": 22948868032,
      "step": 80690,
      "train_runtime": 784561.2666,
      "train_tokens_per_second": 29250.575
    },
    {
      "epoch": 2.8553251055379505,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6602319027783403e-05,
      "loss": 0.38330755233764646,
      "num_input_tokens_seen": 22951708672,
      "step": 80700,
      "train_runtime": 784656.5964,
      "train_tokens_per_second": 29250.641
    },
    {
      "epoch": 2.8556789258016795,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6601403860786528e-05,
      "loss": 0.3859090805053711,
      "num_input_tokens_seen": 22954602496,
      "step": 80710,
      "train_runtime": 784756.0554,
      "train_tokens_per_second": 29250.622
    },
    {
      "epoch": 2.856032746065408,
      "grad_norm": 0.671875,
      "learning_rate": 1.660048884511279e-05,
      "loss": 0.3851149082183838,
      "num_input_tokens_seen": 22957379648,
      "step": 80720,
      "train_runtime": 784849.9043,
      "train_tokens_per_second": 29250.662
    },
    {
      "epoch": 2.856386566329137,
      "grad_norm": 0.609375,
      "learning_rate": 1.6599573980720478e-05,
      "loss": 0.38347327709198,
      "num_input_tokens_seen": 22960240256,
      "step": 80730,
      "train_runtime": 784945.5008,
      "train_tokens_per_second": 29250.744
    },
    {
      "epoch": 2.8567403865928656,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6598659267567925e-05,
      "loss": 0.3873884916305542,
      "num_input_tokens_seen": 22963055232,
      "step": 80740,
      "train_runtime": 785041.8243,
      "train_tokens_per_second": 29250.741
    },
    {
      "epoch": 2.8570942068565945,
      "grad_norm": 0.625,
      "learning_rate": 1.6597744705613464e-05,
      "loss": 0.38725402355194094,
      "num_input_tokens_seen": 22965878144,
      "step": 80750,
      "train_runtime": 785139.5512,
      "train_tokens_per_second": 29250.696
    },
    {
      "epoch": 2.857448027120323,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6596830294815434e-05,
      "loss": 0.38797409534454347,
      "num_input_tokens_seen": 22968705728,
      "step": 80760,
      "train_runtime": 785236.4918,
      "train_tokens_per_second": 29250.686
    },
    {
      "epoch": 2.857801847384052,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6595916035132218e-05,
      "loss": 0.38525421619415284,
      "num_input_tokens_seen": 22971548928,
      "step": 80770,
      "train_runtime": 785336.157,
      "train_tokens_per_second": 29250.594
    },
    {
      "epoch": 2.8581556676477806,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6595001926522184e-05,
      "loss": 0.3828424453735352,
      "num_input_tokens_seen": 22974430848,
      "step": 80780,
      "train_runtime": 785435.9423,
      "train_tokens_per_second": 29250.547
    },
    {
      "epoch": 2.8585094879115096,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6594087968943735e-05,
      "loss": 0.3842684268951416,
      "num_input_tokens_seen": 22977273280,
      "step": 80790,
      "train_runtime": 785531.8988,
      "train_tokens_per_second": 29250.592
    },
    {
      "epoch": 2.8588633081752386,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6593174162355295e-05,
      "loss": 0.38341169357299804,
      "num_input_tokens_seen": 22980155840,
      "step": 80800,
      "train_runtime": 785630.5183,
      "train_tokens_per_second": 29250.59
    },
    {
      "epoch": 2.859217128438967,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6592260506715286e-05,
      "loss": 0.38517658710479735,
      "num_input_tokens_seen": 22983013056,
      "step": 80810,
      "train_runtime": 785728.2441,
      "train_tokens_per_second": 29250.588
    },
    {
      "epoch": 2.8595709487026957,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6591347001982155e-05,
      "loss": 0.38637361526489256,
      "num_input_tokens_seen": 22985856384,
      "step": 80820,
      "train_runtime": 785825.4351,
      "train_tokens_per_second": 29250.588
    },
    {
      "epoch": 2.8599247689664247,
      "grad_norm": 0.62890625,
      "learning_rate": 1.659043364811436e-05,
      "loss": 0.38329670429229734,
      "num_input_tokens_seen": 22988643200,
      "step": 80830,
      "train_runtime": 785919.5308,
      "train_tokens_per_second": 29250.632
    },
    {
      "epoch": 2.8602785892301537,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6589520445070395e-05,
      "loss": 0.3859461784362793,
      "num_input_tokens_seen": 22991561792,
      "step": 80840,
      "train_runtime": 786017.2471,
      "train_tokens_per_second": 29250.709
    },
    {
      "epoch": 2.860632409493882,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6588607392808746e-05,
      "loss": 0.38774890899658204,
      "num_input_tokens_seen": 22994382656,
      "step": 80850,
      "train_runtime": 786112.7548,
      "train_tokens_per_second": 29250.744
    },
    {
      "epoch": 2.8609862297576107,
      "grad_norm": 0.58984375,
      "learning_rate": 1.658769449128792e-05,
      "loss": 0.3839534521102905,
      "num_input_tokens_seen": 22997277952,
      "step": 80860,
      "train_runtime": 786215.1305,
      "train_tokens_per_second": 29250.617
    },
    {
      "epoch": 2.8613400500213397,
      "grad_norm": 0.625,
      "learning_rate": 1.6586781740466448e-05,
      "loss": 0.385223913192749,
      "num_input_tokens_seen": 23000124160,
      "step": 80870,
      "train_runtime": 786312.3651,
      "train_tokens_per_second": 29250.62
    },
    {
      "epoch": 2.8616938702850687,
      "grad_norm": 0.65625,
      "learning_rate": 1.6585869140302875e-05,
      "loss": 0.3855762958526611,
      "num_input_tokens_seen": 23002947008,
      "step": 80880,
      "train_runtime": 786407.8233,
      "train_tokens_per_second": 29250.659
    },
    {
      "epoch": 2.8620476905487973,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6584956690755752e-05,
      "loss": 0.39006495475769043,
      "num_input_tokens_seen": 23005774592,
      "step": 80890,
      "train_runtime": 786505.5582,
      "train_tokens_per_second": 29250.619
    },
    {
      "epoch": 2.8624015108125263,
      "grad_norm": 0.62890625,
      "learning_rate": 1.658404439178366e-05,
      "loss": 0.38556675910949706,
      "num_input_tokens_seen": 23008555840,
      "step": 80900,
      "train_runtime": 786598.0095,
      "train_tokens_per_second": 29250.717
    },
    {
      "epoch": 2.862755331076255,
      "grad_norm": 0.58984375,
      "learning_rate": 1.6583132243345195e-05,
      "loss": 0.3841893672943115,
      "num_input_tokens_seen": 23011404608,
      "step": 80910,
      "train_runtime": 786694.0004,
      "train_tokens_per_second": 29250.769
    },
    {
      "epoch": 2.863109151339984,
      "grad_norm": 0.625,
      "learning_rate": 1.6582220245398953e-05,
      "loss": 0.3896572828292847,
      "num_input_tokens_seen": 23014274560,
      "step": 80920,
      "train_runtime": 786792.3161,
      "train_tokens_per_second": 29250.762
    },
    {
      "epoch": 2.8634629716037123,
      "grad_norm": 0.6015625,
      "learning_rate": 1.658130839790356e-05,
      "loss": 0.38740460872650145,
      "num_input_tokens_seen": 23017134336,
      "step": 80930,
      "train_runtime": 786888.6637,
      "train_tokens_per_second": 29250.815
    },
    {
      "epoch": 2.8638167918674413,
      "grad_norm": 0.63671875,
      "learning_rate": 1.658039670081766e-05,
      "loss": 0.37956216335296633,
      "num_input_tokens_seen": 23019984512,
      "step": 80940,
      "train_runtime": 786986.1573,
      "train_tokens_per_second": 29250.812
    },
    {
      "epoch": 2.86417061213117,
      "grad_norm": 0.609375,
      "learning_rate": 1.65794851540999e-05,
      "loss": 0.38520843982696534,
      "num_input_tokens_seen": 23022766400,
      "step": 80950,
      "train_runtime": 787079.0905,
      "train_tokens_per_second": 29250.893
    },
    {
      "epoch": 2.864524432394899,
      "grad_norm": 0.609375,
      "learning_rate": 1.6578573757708953e-05,
      "loss": 0.38732295036315917,
      "num_input_tokens_seen": 23025656960,
      "step": 80960,
      "train_runtime": 787179.0435,
      "train_tokens_per_second": 29250.851
    },
    {
      "epoch": 2.8648782526586274,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6577662511603507e-05,
      "loss": 0.3871471881866455,
      "num_input_tokens_seen": 23028472640,
      "step": 80970,
      "train_runtime": 787274.3673,
      "train_tokens_per_second": 29250.886
    },
    {
      "epoch": 2.8652320729223564,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6576751415742262e-05,
      "loss": 0.3873793363571167,
      "num_input_tokens_seen": 23031304192,
      "step": 80980,
      "train_runtime": 787368.0609,
      "train_tokens_per_second": 29251.001
    },
    {
      "epoch": 2.8655858931860854,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6575840470083934e-05,
      "loss": 0.38881549835205076,
      "num_input_tokens_seen": 23034158272,
      "step": 80990,
      "train_runtime": 787465.0715,
      "train_tokens_per_second": 29251.022
    },
    {
      "epoch": 2.865939713449814,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6574929674587263e-05,
      "loss": 0.3861287355422974,
      "num_input_tokens_seen": 23037005184,
      "step": 81000,
      "train_runtime": 787563.2652,
      "train_tokens_per_second": 29250.99
    },
    {
      "epoch": 2.8662935337135425,
      "grad_norm": 0.6640625,
      "learning_rate": 1.6574019029210995e-05,
      "loss": 0.3872844696044922,
      "num_input_tokens_seen": 23039829248,
      "step": 81010,
      "train_runtime": 787659.067,
      "train_tokens_per_second": 29251.018
    },
    {
      "epoch": 2.8666473539772714,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6573108533913894e-05,
      "loss": 0.38566126823425295,
      "num_input_tokens_seen": 23042716608,
      "step": 81020,
      "train_runtime": 787760.1879,
      "train_tokens_per_second": 29250.928
    },
    {
      "epoch": 2.8670011742410004,
      "grad_norm": 0.6171875,
      "learning_rate": 1.657219818865474e-05,
      "loss": 0.38692626953125,
      "num_input_tokens_seen": 23045577920,
      "step": 81030,
      "train_runtime": 787857.7704,
      "train_tokens_per_second": 29250.937
    },
    {
      "epoch": 2.867354994504729,
      "grad_norm": 0.609375,
      "learning_rate": 1.6571287993392336e-05,
      "loss": 0.3842721700668335,
      "num_input_tokens_seen": 23048399296,
      "step": 81040,
      "train_runtime": 787953.1177,
      "train_tokens_per_second": 29250.978
    },
    {
      "epoch": 2.867708814768458,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6570377948085496e-05,
      "loss": 0.3845519065856934,
      "num_input_tokens_seen": 23051287040,
      "step": 81050,
      "train_runtime": 788050.9936,
      "train_tokens_per_second": 29251.009
    },
    {
      "epoch": 2.8680626350321865,
      "grad_norm": 0.66015625,
      "learning_rate": 1.6569468052693042e-05,
      "loss": 0.39033515453338624,
      "num_input_tokens_seen": 23054120704,
      "step": 81060,
      "train_runtime": 788148.1429,
      "train_tokens_per_second": 29250.999
    },
    {
      "epoch": 2.8684164552959155,
      "grad_norm": 0.609375,
      "learning_rate": 1.6568558307173823e-05,
      "loss": 0.38727493286132814,
      "num_input_tokens_seen": 23056946880,
      "step": 81070,
      "train_runtime": 788246.4344,
      "train_tokens_per_second": 29250.937
    },
    {
      "epoch": 2.868770275559644,
      "grad_norm": 0.640625,
      "learning_rate": 1.65676487114867e-05,
      "loss": 0.38291044235229493,
      "num_input_tokens_seen": 23059841856,
      "step": 81080,
      "train_runtime": 788346.273,
      "train_tokens_per_second": 29250.905
    },
    {
      "epoch": 2.869124095823373,
      "grad_norm": 0.625,
      "learning_rate": 1.6566739265590547e-05,
      "loss": 0.3887549877166748,
      "num_input_tokens_seen": 23062668928,
      "step": 81090,
      "train_runtime": 788442.8903,
      "train_tokens_per_second": 29250.906
    },
    {
      "epoch": 2.8694779160871016,
      "grad_norm": 0.62890625,
      "learning_rate": 1.656582996944426e-05,
      "loss": 0.3819906234741211,
      "num_input_tokens_seen": 23065522304,
      "step": 81100,
      "train_runtime": 788540.3032,
      "train_tokens_per_second": 29250.911
    },
    {
      "epoch": 2.8698317363508306,
      "grad_norm": 0.625,
      "learning_rate": 1.6564920823006745e-05,
      "loss": 0.386607027053833,
      "num_input_tokens_seen": 23068403328,
      "step": 81110,
      "train_runtime": 788638.8854,
      "train_tokens_per_second": 29250.908
    },
    {
      "epoch": 2.870185556614559,
      "grad_norm": 0.59375,
      "learning_rate": 1.6564011826236926e-05,
      "loss": 0.3872599363327026,
      "num_input_tokens_seen": 23071303168,
      "step": 81120,
      "train_runtime": 788739.9712,
      "train_tokens_per_second": 29250.836
    },
    {
      "epoch": 2.870539376878288,
      "grad_norm": 0.6328125,
      "learning_rate": 1.656310297909374e-05,
      "loss": 0.38534929752349856,
      "num_input_tokens_seen": 23074100224,
      "step": 81130,
      "train_runtime": 788835.5325,
      "train_tokens_per_second": 29250.838
    },
    {
      "epoch": 2.870893197142017,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6562194281536146e-05,
      "loss": 0.38730440139770506,
      "num_input_tokens_seen": 23076961088,
      "step": 81140,
      "train_runtime": 788933.5368,
      "train_tokens_per_second": 29250.83
    },
    {
      "epoch": 2.8712470174057456,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6561285733523108e-05,
      "loss": 0.3849898338317871,
      "num_input_tokens_seen": 23079791232,
      "step": 81150,
      "train_runtime": 789030.6827,
      "train_tokens_per_second": 29250.816
    },
    {
      "epoch": 2.871600837669474,
      "grad_norm": 0.609375,
      "learning_rate": 1.6560377335013625e-05,
      "loss": 0.38543477058410647,
      "num_input_tokens_seen": 23082620160,
      "step": 81160,
      "train_runtime": 789125.0493,
      "train_tokens_per_second": 29250.903
    },
    {
      "epoch": 2.871954657933203,
      "grad_norm": 0.59375,
      "learning_rate": 1.655946908596669e-05,
      "loss": 0.38542652130126953,
      "num_input_tokens_seen": 23085475136,
      "step": 81170,
      "train_runtime": 789224.3364,
      "train_tokens_per_second": 29250.84
    },
    {
      "epoch": 2.872308478196932,
      "grad_norm": 0.6484375,
      "learning_rate": 1.6558560986341324e-05,
      "loss": 0.3886594295501709,
      "num_input_tokens_seen": 23088293056,
      "step": 81180,
      "train_runtime": 789319.4661,
      "train_tokens_per_second": 29250.885
    },
    {
      "epoch": 2.8726622984606607,
      "grad_norm": 0.625,
      "learning_rate": 1.6557653036096564e-05,
      "loss": 0.39289045333862305,
      "num_input_tokens_seen": 23091135424,
      "step": 81190,
      "train_runtime": 789416.4736,
      "train_tokens_per_second": 29250.891
    },
    {
      "epoch": 2.8730161187243892,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6556745235191454e-05,
      "loss": 0.38567628860473635,
      "num_input_tokens_seen": 23093985664,
      "step": 81200,
      "train_runtime": 789511.9572,
      "train_tokens_per_second": 29250.964
    },
    {
      "epoch": 2.873369938988118,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6555837583585066e-05,
      "loss": 0.39205853939056395,
      "num_input_tokens_seen": 23096877056,
      "step": 81210,
      "train_runtime": 789612.3049,
      "train_tokens_per_second": 29250.908
    },
    {
      "epoch": 2.873723759251847,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6554930081236475e-05,
      "loss": 0.3855047464370728,
      "num_input_tokens_seen": 23099723456,
      "step": 81220,
      "train_runtime": 789709.8876,
      "train_tokens_per_second": 29250.898
    },
    {
      "epoch": 2.8740775795155757,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6554022728104783e-05,
      "loss": 0.38598003387451174,
      "num_input_tokens_seen": 23102569664,
      "step": 81230,
      "train_runtime": 789807.4231,
      "train_tokens_per_second": 29250.89
    },
    {
      "epoch": 2.8744313997793047,
      "grad_norm": 0.625,
      "learning_rate": 1.65531155241491e-05,
      "loss": 0.3862329959869385,
      "num_input_tokens_seen": 23105362432,
      "step": 81240,
      "train_runtime": 789900.9452,
      "train_tokens_per_second": 29250.962
    },
    {
      "epoch": 2.8747852200430333,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6552208469328555e-05,
      "loss": 0.3869790077209473,
      "num_input_tokens_seen": 23108133120,
      "step": 81250,
      "train_runtime": 789992.5089,
      "train_tokens_per_second": 29251.079
    },
    {
      "epoch": 2.8751390403067623,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6551301563602294e-05,
      "loss": 0.3815010547637939,
      "num_input_tokens_seen": 23110986048,
      "step": 81260,
      "train_runtime": 790091.2287,
      "train_tokens_per_second": 29251.035
    },
    {
      "epoch": 2.875492860570491,
      "grad_norm": 0.6328125,
      "learning_rate": 1.655039480692947e-05,
      "loss": 0.3899769067764282,
      "num_input_tokens_seen": 23113758464,
      "step": 81270,
      "train_runtime": 790183.9305,
      "train_tokens_per_second": 29251.112
    },
    {
      "epoch": 2.87584668083422,
      "grad_norm": 0.609375,
      "learning_rate": 1.6549488199269263e-05,
      "loss": 0.3841294765472412,
      "num_input_tokens_seen": 23116559744,
      "step": 81280,
      "train_runtime": 790278.4974,
      "train_tokens_per_second": 29251.156
    },
    {
      "epoch": 2.876200501097949,
      "grad_norm": 0.60546875,
      "learning_rate": 1.654858174058087e-05,
      "loss": 0.3816652536392212,
      "num_input_tokens_seen": 23119414528,
      "step": 81290,
      "train_runtime": 790376.6878,
      "train_tokens_per_second": 29251.134
    },
    {
      "epoch": 2.8765543213616773,
      "grad_norm": 0.640625,
      "learning_rate": 1.6547675430823484e-05,
      "loss": 0.38713769912719725,
      "num_input_tokens_seen": 23122306944,
      "step": 81300,
      "train_runtime": 790476.0557,
      "train_tokens_per_second": 29251.116
    },
    {
      "epoch": 2.876908141625406,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6546769269956337e-05,
      "loss": 0.3860367774963379,
      "num_input_tokens_seen": 23125164416,
      "step": 81310,
      "train_runtime": 790575.2789,
      "train_tokens_per_second": 29251.059
    },
    {
      "epoch": 2.877261961889135,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6545863257938662e-05,
      "loss": 0.3824927806854248,
      "num_input_tokens_seen": 23128015424,
      "step": 81320,
      "train_runtime": 790671.7099,
      "train_tokens_per_second": 29251.098
    },
    {
      "epoch": 2.877615782152864,
      "grad_norm": 0.6328125,
      "learning_rate": 1.654495739472972e-05,
      "loss": 0.38840570449829104,
      "num_input_tokens_seen": 23130818240,
      "step": 81330,
      "train_runtime": 790767.5978,
      "train_tokens_per_second": 29251.095
    },
    {
      "epoch": 2.8779696024165924,
      "grad_norm": 0.6171875,
      "learning_rate": 1.654405168028877e-05,
      "loss": 0.38400247097015383,
      "num_input_tokens_seen": 23133666240,
      "step": 81340,
      "train_runtime": 790866.0679,
      "train_tokens_per_second": 29251.054
    },
    {
      "epoch": 2.878323422680321,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6543146114575108e-05,
      "loss": 0.38600523471832277,
      "num_input_tokens_seen": 23136511360,
      "step": 81350,
      "train_runtime": 790962.5959,
      "train_tokens_per_second": 29251.082
    },
    {
      "epoch": 2.87867724294405,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6542240697548022e-05,
      "loss": 0.382672119140625,
      "num_input_tokens_seen": 23139309760,
      "step": 81360,
      "train_runtime": 791058.0198,
      "train_tokens_per_second": 29251.091
    },
    {
      "epoch": 2.879031063207779,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6541335429166833e-05,
      "loss": 0.38936915397644045,
      "num_input_tokens_seen": 23142062976,
      "step": 81370,
      "train_runtime": 791148.3108,
      "train_tokens_per_second": 29251.232
    },
    {
      "epoch": 2.8793848834715075,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6540430309390877e-05,
      "loss": 0.38384013175964354,
      "num_input_tokens_seen": 23144899776,
      "step": 81380,
      "train_runtime": 791245.5605,
      "train_tokens_per_second": 29251.222
    },
    {
      "epoch": 2.8797387037352364,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6539525338179495e-05,
      "loss": 0.3860579013824463,
      "num_input_tokens_seen": 23147722432,
      "step": 81390,
      "train_runtime": 791343.3789,
      "train_tokens_per_second": 29251.173
    },
    {
      "epoch": 2.880092523998965,
      "grad_norm": 0.625,
      "learning_rate": 1.6538620515492054e-05,
      "loss": 0.3901110887527466,
      "num_input_tokens_seen": 23150635136,
      "step": 81400,
      "train_runtime": 791447.8889,
      "train_tokens_per_second": 29250.991
    },
    {
      "epoch": 2.880446344262694,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6537715841287933e-05,
      "loss": 0.38684787750244143,
      "num_input_tokens_seen": 23153445824,
      "step": 81410,
      "train_runtime": 791546.531,
      "train_tokens_per_second": 29250.897
    },
    {
      "epoch": 2.8808001645264225,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6536811315526516e-05,
      "loss": 0.39116668701171875,
      "num_input_tokens_seen": 23156338304,
      "step": 81420,
      "train_runtime": 791646.5693,
      "train_tokens_per_second": 29250.854
    },
    {
      "epoch": 2.8811539847901515,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6535906938167223e-05,
      "loss": 0.38677234649658204,
      "num_input_tokens_seen": 23159170432,
      "step": 81430,
      "train_runtime": 791743.5291,
      "train_tokens_per_second": 29250.849
    },
    {
      "epoch": 2.88150780505388,
      "grad_norm": 0.65234375,
      "learning_rate": 1.6535002709169474e-05,
      "loss": 0.38967745304107665,
      "num_input_tokens_seen": 23162008640,
      "step": 81440,
      "train_runtime": 791837.0017,
      "train_tokens_per_second": 29250.98
    },
    {
      "epoch": 2.881861625317609,
      "grad_norm": 0.63671875,
      "learning_rate": 1.653409862849271e-05,
      "loss": 0.3861429691314697,
      "num_input_tokens_seen": 23164825472,
      "step": 81450,
      "train_runtime": 791934.299,
      "train_tokens_per_second": 29250.944
    },
    {
      "epoch": 2.8822154455813376,
      "grad_norm": 0.63671875,
      "learning_rate": 1.653319469609639e-05,
      "loss": 0.3831209182739258,
      "num_input_tokens_seen": 23167639168,
      "step": 81460,
      "train_runtime": 792029.261,
      "train_tokens_per_second": 29250.989
    },
    {
      "epoch": 2.8825692658450666,
      "grad_norm": 0.6484375,
      "learning_rate": 1.6532290911939978e-05,
      "loss": 0.3878655910491943,
      "num_input_tokens_seen": 23170475456,
      "step": 81470,
      "train_runtime": 792125.1391,
      "train_tokens_per_second": 29251.029
    },
    {
      "epoch": 2.8829230861087956,
      "grad_norm": 0.625,
      "learning_rate": 1.6531387275982972e-05,
      "loss": 0.38855006694793703,
      "num_input_tokens_seen": 23173259136,
      "step": 81480,
      "train_runtime": 792220.4839,
      "train_tokens_per_second": 29251.022
    },
    {
      "epoch": 2.883276906372524,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6530483788184863e-05,
      "loss": 0.3821551322937012,
      "num_input_tokens_seen": 23176068160,
      "step": 81490,
      "train_runtime": 792314.8089,
      "train_tokens_per_second": 29251.085
    },
    {
      "epoch": 2.8836307266362526,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6529580448505172e-05,
      "loss": 0.3886830806732178,
      "num_input_tokens_seen": 23178903680,
      "step": 81500,
      "train_runtime": 792409.9868,
      "train_tokens_per_second": 29251.15
    },
    {
      "epoch": 2.8839845468999816,
      "grad_norm": 0.640625,
      "learning_rate": 1.6528677256903436e-05,
      "loss": 0.3892590761184692,
      "num_input_tokens_seen": 23181723136,
      "step": 81510,
      "train_runtime": 792506.2284,
      "train_tokens_per_second": 29251.156
    },
    {
      "epoch": 2.8843383671637106,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6527774213339204e-05,
      "loss": 0.3896749496459961,
      "num_input_tokens_seen": 23184502272,
      "step": 81520,
      "train_runtime": 792598.6911,
      "train_tokens_per_second": 29251.25
    },
    {
      "epoch": 2.884692187427439,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6526871317772037e-05,
      "loss": 0.38340129852294924,
      "num_input_tokens_seen": 23187387136,
      "step": 81530,
      "train_runtime": 792694.2099,
      "train_tokens_per_second": 29251.364
    },
    {
      "epoch": 2.885046007691168,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6525968570161512e-05,
      "loss": 0.3903005599975586,
      "num_input_tokens_seen": 23190172480,
      "step": 81540,
      "train_runtime": 792789.6181,
      "train_tokens_per_second": 29251.357
    },
    {
      "epoch": 2.8853998279548967,
      "grad_norm": 0.640625,
      "learning_rate": 1.6525065970467236e-05,
      "loss": 0.38321237564086913,
      "num_input_tokens_seen": 23193086080,
      "step": 81550,
      "train_runtime": 792893.3739,
      "train_tokens_per_second": 29251.204
    },
    {
      "epoch": 2.8857536482186257,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6524163518648803e-05,
      "loss": 0.3898627281188965,
      "num_input_tokens_seen": 23195886528,
      "step": 81560,
      "train_runtime": 792988.7158,
      "train_tokens_per_second": 29251.219
    },
    {
      "epoch": 2.886107468482354,
      "grad_norm": 0.65625,
      "learning_rate": 1.6523261214665855e-05,
      "loss": 0.38414835929870605,
      "num_input_tokens_seen": 23198753536,
      "step": 81570,
      "train_runtime": 793086.8064,
      "train_tokens_per_second": 29251.216
    },
    {
      "epoch": 2.886461288746083,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6522359058478024e-05,
      "loss": 0.384470796585083,
      "num_input_tokens_seen": 23201578240,
      "step": 81580,
      "train_runtime": 793180.8015,
      "train_tokens_per_second": 29251.311
    },
    {
      "epoch": 2.8868151090098118,
      "grad_norm": 0.640625,
      "learning_rate": 1.652145705004497e-05,
      "loss": 0.3838948249816895,
      "num_input_tokens_seen": 23204418304,
      "step": 81590,
      "train_runtime": 793276.5534,
      "train_tokens_per_second": 29251.36
    },
    {
      "epoch": 2.8871689292735407,
      "grad_norm": 0.6171875,
      "learning_rate": 1.652055518932636e-05,
      "loss": 0.38438825607299804,
      "num_input_tokens_seen": 23207239872,
      "step": 81600,
      "train_runtime": 793371.9011,
      "train_tokens_per_second": 29251.401
    },
    {
      "epoch": 2.8875227495372693,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6519653476281895e-05,
      "loss": 0.386983060836792,
      "num_input_tokens_seen": 23210166912,
      "step": 81610,
      "train_runtime": 793472.3377,
      "train_tokens_per_second": 29251.388
    },
    {
      "epoch": 2.8878765698009983,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6518751910871267e-05,
      "loss": 0.38485558032989503,
      "num_input_tokens_seen": 23212979328,
      "step": 81620,
      "train_runtime": 793566.3105,
      "train_tokens_per_second": 29251.468
    },
    {
      "epoch": 2.8882303900647273,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6517850493054192e-05,
      "loss": 0.3835980176925659,
      "num_input_tokens_seen": 23215794752,
      "step": 81630,
      "train_runtime": 793658.6734,
      "train_tokens_per_second": 29251.611
    },
    {
      "epoch": 2.888584210328456,
      "grad_norm": 0.6484375,
      "learning_rate": 1.6516949222790416e-05,
      "loss": 0.38127636909484863,
      "num_input_tokens_seen": 23218609280,
      "step": 81640,
      "train_runtime": 793756.0186,
      "train_tokens_per_second": 29251.569
    },
    {
      "epoch": 2.8889380305921843,
      "grad_norm": 0.6171875,
      "learning_rate": 1.651604810003968e-05,
      "loss": 0.3875267744064331,
      "num_input_tokens_seen": 23221403840,
      "step": 81650,
      "train_runtime": 793850.1855,
      "train_tokens_per_second": 29251.62
    },
    {
      "epoch": 2.8892918508559133,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6515147124761745e-05,
      "loss": 0.38620743751525877,
      "num_input_tokens_seen": 23224249472,
      "step": 81660,
      "train_runtime": 793946.3548,
      "train_tokens_per_second": 29251.661
    },
    {
      "epoch": 2.8896456711196423,
      "grad_norm": 0.640625,
      "learning_rate": 1.6514246296916406e-05,
      "loss": 0.3855967283248901,
      "num_input_tokens_seen": 23227105728,
      "step": 81670,
      "train_runtime": 794042.8205,
      "train_tokens_per_second": 29251.704
    },
    {
      "epoch": 2.889999491383371,
      "grad_norm": 0.609375,
      "learning_rate": 1.6513345616463443e-05,
      "loss": 0.38955535888671877,
      "num_input_tokens_seen": 23229927488,
      "step": 81680,
      "train_runtime": 794140.607,
      "train_tokens_per_second": 29251.656
    },
    {
      "epoch": 2.8903533116470994,
      "grad_norm": 0.6015625,
      "learning_rate": 1.651244508336267e-05,
      "loss": 0.3868558406829834,
      "num_input_tokens_seen": 23232781248,
      "step": 81690,
      "train_runtime": 794236.0817,
      "train_tokens_per_second": 29251.732
    },
    {
      "epoch": 2.8907071319108284,
      "grad_norm": 0.62890625,
      "learning_rate": 1.651154469757392e-05,
      "loss": 0.3861047029495239,
      "num_input_tokens_seen": 23235630400,
      "step": 81700,
      "train_runtime": 794331.8317,
      "train_tokens_per_second": 29251.793
    },
    {
      "epoch": 2.8910609521745574,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6510644459057032e-05,
      "loss": 0.3819290637969971,
      "num_input_tokens_seen": 23238496384,
      "step": 81710,
      "train_runtime": 794429.9231,
      "train_tokens_per_second": 29251.789
    },
    {
      "epoch": 2.891414772438286,
      "grad_norm": 0.609375,
      "learning_rate": 1.6509744367771856e-05,
      "loss": 0.38619470596313477,
      "num_input_tokens_seen": 23241361984,
      "step": 81720,
      "train_runtime": 794532.7109,
      "train_tokens_per_second": 29251.611
    },
    {
      "epoch": 2.891768592702015,
      "grad_norm": 0.61328125,
      "learning_rate": 1.650884442367827e-05,
      "loss": 0.38799047470092773,
      "num_input_tokens_seen": 23244226048,
      "step": 81730,
      "train_runtime": 794629.7972,
      "train_tokens_per_second": 29251.642
    },
    {
      "epoch": 2.8921224129657435,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6507944626736166e-05,
      "loss": 0.3849436521530151,
      "num_input_tokens_seen": 23247043072,
      "step": 81740,
      "train_runtime": 794724.3475,
      "train_tokens_per_second": 29251.706
    },
    {
      "epoch": 2.8924762332294724,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6507044976905436e-05,
      "loss": 0.38478143215179444,
      "num_input_tokens_seen": 23249923328,
      "step": 81750,
      "train_runtime": 794823.3739,
      "train_tokens_per_second": 29251.685
    },
    {
      "epoch": 2.892830053493201,
      "grad_norm": 0.5859375,
      "learning_rate": 1.6506145474146002e-05,
      "loss": 0.3904749631881714,
      "num_input_tokens_seen": 23252808448,
      "step": 81760,
      "train_runtime": 794923.4818,
      "train_tokens_per_second": 29251.631
    },
    {
      "epoch": 2.89318387375693,
      "grad_norm": 0.625,
      "learning_rate": 1.6505246118417803e-05,
      "loss": 0.3853966951370239,
      "num_input_tokens_seen": 23255655424,
      "step": 81770,
      "train_runtime": 795018.8708,
      "train_tokens_per_second": 29251.702
    },
    {
      "epoch": 2.8935376940206585,
      "grad_norm": 0.609375,
      "learning_rate": 1.650434690968078e-05,
      "loss": 0.385425066947937,
      "num_input_tokens_seen": 23258495040,
      "step": 81780,
      "train_runtime": 795114.4154,
      "train_tokens_per_second": 29251.759
    },
    {
      "epoch": 2.8938915142843875,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6503447847894902e-05,
      "loss": 0.3871616363525391,
      "num_input_tokens_seen": 23261335040,
      "step": 81790,
      "train_runtime": 795208.7604,
      "train_tokens_per_second": 29251.86
    },
    {
      "epoch": 2.894245334548116,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6502548933020142e-05,
      "loss": 0.390057897567749,
      "num_input_tokens_seen": 23264181824,
      "step": 81800,
      "train_runtime": 795306.5143,
      "train_tokens_per_second": 29251.844
    },
    {
      "epoch": 2.894599154811845,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6501650165016504e-05,
      "loss": 0.38388235569000245,
      "num_input_tokens_seen": 23267051008,
      "step": 81810,
      "train_runtime": 795408.3919,
      "train_tokens_per_second": 29251.704
    },
    {
      "epoch": 2.894952975075574,
      "grad_norm": 0.609375,
      "learning_rate": 1.6500751543843988e-05,
      "loss": 0.38326637744903563,
      "num_input_tokens_seen": 23269968704,
      "step": 81820,
      "train_runtime": 795509.8269,
      "train_tokens_per_second": 29251.642
    },
    {
      "epoch": 2.8953067953393026,
      "grad_norm": 0.6640625,
      "learning_rate": 1.6499853069462623e-05,
      "loss": 0.3844102621078491,
      "num_input_tokens_seen": 23272819136,
      "step": 81830,
      "train_runtime": 795608.6372,
      "train_tokens_per_second": 29251.592
    },
    {
      "epoch": 2.895660615603031,
      "grad_norm": 0.6171875,
      "learning_rate": 1.649895474183245e-05,
      "loss": 0.3845831871032715,
      "num_input_tokens_seen": 23275730752,
      "step": 81840,
      "train_runtime": 795706.8968,
      "train_tokens_per_second": 29251.639
    },
    {
      "epoch": 2.89601443586676,
      "grad_norm": 0.6015625,
      "learning_rate": 1.649805656091352e-05,
      "loss": 0.3881673812866211,
      "num_input_tokens_seen": 23278545536,
      "step": 81850,
      "train_runtime": 795802.7138,
      "train_tokens_per_second": 29251.654
    },
    {
      "epoch": 2.896368256130489,
      "grad_norm": 0.6484375,
      "learning_rate": 1.649715852666591e-05,
      "loss": 0.38451552391052246,
      "num_input_tokens_seen": 23281337024,
      "step": 81860,
      "train_runtime": 795897.4076,
      "train_tokens_per_second": 29251.681
    },
    {
      "epoch": 2.8967220763942176,
      "grad_norm": 0.609375,
      "learning_rate": 1.6496260639049703e-05,
      "loss": 0.3868791341781616,
      "num_input_tokens_seen": 23284166976,
      "step": 81870,
      "train_runtime": 795994.4553,
      "train_tokens_per_second": 29251.67
    },
    {
      "epoch": 2.8970758966579466,
      "grad_norm": 0.625,
      "learning_rate": 1.6495362898024995e-05,
      "loss": 0.3884408950805664,
      "num_input_tokens_seen": 23286988288,
      "step": 81880,
      "train_runtime": 796090.5999,
      "train_tokens_per_second": 29251.681
    },
    {
      "epoch": 2.897429716921675,
      "grad_norm": 0.59765625,
      "learning_rate": 1.649446530355191e-05,
      "loss": 0.3881901741027832,
      "num_input_tokens_seen": 23289779840,
      "step": 81890,
      "train_runtime": 796186.3836,
      "train_tokens_per_second": 29251.668
    },
    {
      "epoch": 2.897783537185404,
      "grad_norm": 0.609375,
      "learning_rate": 1.6493567855590576e-05,
      "loss": 0.3867985486984253,
      "num_input_tokens_seen": 23292648000,
      "step": 81900,
      "train_runtime": 796281.432,
      "train_tokens_per_second": 29251.778
    },
    {
      "epoch": 2.8981373574491327,
      "grad_norm": 0.640625,
      "learning_rate": 1.649267055410114e-05,
      "loss": 0.3887564897537231,
      "num_input_tokens_seen": 23295458304,
      "step": 81910,
      "train_runtime": 796376.1053,
      "train_tokens_per_second": 29251.83
    },
    {
      "epoch": 2.8984911777128617,
      "grad_norm": 0.59765625,
      "learning_rate": 1.649177339904376e-05,
      "loss": 0.3849599361419678,
      "num_input_tokens_seen": 23298307712,
      "step": 81920,
      "train_runtime": 796471.9361,
      "train_tokens_per_second": 29251.888
    },
    {
      "epoch": 2.8988449979765902,
      "grad_norm": 0.625,
      "learning_rate": 1.649087639037862e-05,
      "loss": 0.3815911769866943,
      "num_input_tokens_seen": 23301108992,
      "step": 81930,
      "train_runtime": 796563.892,
      "train_tokens_per_second": 29252.028
    },
    {
      "epoch": 2.899198818240319,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6489979528065905e-05,
      "loss": 0.38560535907745364,
      "num_input_tokens_seen": 23303952384,
      "step": 81940,
      "train_runtime": 796662.055,
      "train_tokens_per_second": 29251.992
    },
    {
      "epoch": 2.8995526385040478,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6489082812065826e-05,
      "loss": 0.38556230068206787,
      "num_input_tokens_seen": 23306748480,
      "step": 81950,
      "train_runtime": 796758.1956,
      "train_tokens_per_second": 29251.972
    },
    {
      "epoch": 2.8999064587677768,
      "grad_norm": 0.58203125,
      "learning_rate": 1.64881862423386e-05,
      "loss": 0.3902272701263428,
      "num_input_tokens_seen": 23309584064,
      "step": 81960,
      "train_runtime": 796853.2062,
      "train_tokens_per_second": 29252.043
    },
    {
      "epoch": 2.9002602790315057,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6487289818844477e-05,
      "loss": 0.3831901550292969,
      "num_input_tokens_seen": 23312365312,
      "step": 81970,
      "train_runtime": 796946.3862,
      "train_tokens_per_second": 29252.112
    },
    {
      "epoch": 2.9006140992952343,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6486393541543697e-05,
      "loss": 0.38457527160644533,
      "num_input_tokens_seen": 23315219072,
      "step": 81980,
      "train_runtime": 797049.0178,
      "train_tokens_per_second": 29251.926
    },
    {
      "epoch": 2.900967919558963,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6485497410396535e-05,
      "loss": 0.3861251354217529,
      "num_input_tokens_seen": 23318028480,
      "step": 81990,
      "train_runtime": 797141.4714,
      "train_tokens_per_second": 29252.058
    },
    {
      "epoch": 2.901321739822692,
      "grad_norm": 0.625,
      "learning_rate": 1.6484601425363267e-05,
      "loss": 0.38825788497924807,
      "num_input_tokens_seen": 23320873664,
      "step": 82000,
      "train_runtime": 797238.6041,
      "train_tokens_per_second": 29252.063
    },
    {
      "epoch": 2.901675560086421,
      "grad_norm": 0.609375,
      "learning_rate": 1.6483705586404196e-05,
      "loss": 0.3879213333129883,
      "num_input_tokens_seen": 23323738368,
      "step": 82010,
      "train_runtime": 797334.7913,
      "train_tokens_per_second": 29252.127
    },
    {
      "epoch": 2.9020293803501493,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6482809893479635e-05,
      "loss": 0.38478999137878417,
      "num_input_tokens_seen": 23326582464,
      "step": 82020,
      "train_runtime": 797433.0514,
      "train_tokens_per_second": 29252.089
    },
    {
      "epoch": 2.902383200613878,
      "grad_norm": 0.6015625,
      "learning_rate": 1.648191434654991e-05,
      "loss": 0.3881343126296997,
      "num_input_tokens_seen": 23329444608,
      "step": 82030,
      "train_runtime": 797532.3221,
      "train_tokens_per_second": 29252.037
    },
    {
      "epoch": 2.902737020877607,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6481018945575367e-05,
      "loss": 0.3858876943588257,
      "num_input_tokens_seen": 23332333568,
      "step": 82040,
      "train_runtime": 797633.876,
      "train_tokens_per_second": 29251.934
    },
    {
      "epoch": 2.903090841141336,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6480123690516364e-05,
      "loss": 0.387485408782959,
      "num_input_tokens_seen": 23335177600,
      "step": 82050,
      "train_runtime": 797732.1834,
      "train_tokens_per_second": 29251.894
    },
    {
      "epoch": 2.9034446614050644,
      "grad_norm": 0.625,
      "learning_rate": 1.6479228581333274e-05,
      "loss": 0.38678512573242185,
      "num_input_tokens_seen": 23338055424,
      "step": 82060,
      "train_runtime": 797832.0491,
      "train_tokens_per_second": 29251.84
    },
    {
      "epoch": 2.9037984816687934,
      "grad_norm": 0.63671875,
      "learning_rate": 1.647833361798648e-05,
      "loss": 0.38803179264068605,
      "num_input_tokens_seen": 23340829952,
      "step": 82070,
      "train_runtime": 797926.1842,
      "train_tokens_per_second": 29251.866
    },
    {
      "epoch": 2.904152301932522,
      "grad_norm": 0.625,
      "learning_rate": 1.6477438800436394e-05,
      "loss": 0.3876200199127197,
      "num_input_tokens_seen": 23343640960,
      "step": 82080,
      "train_runtime": 798020.2788,
      "train_tokens_per_second": 29251.94
    },
    {
      "epoch": 2.904506122196251,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6476544128643425e-05,
      "loss": 0.3908067226409912,
      "num_input_tokens_seen": 23346495808,
      "step": 82090,
      "train_runtime": 798118.9086,
      "train_tokens_per_second": 29251.902
    },
    {
      "epoch": 2.9048599424599795,
      "grad_norm": 0.625,
      "learning_rate": 1.6475649602568016e-05,
      "loss": 0.388978099822998,
      "num_input_tokens_seen": 23349385728,
      "step": 82100,
      "train_runtime": 798222.0452,
      "train_tokens_per_second": 29251.743
    },
    {
      "epoch": 2.9052137627237085,
      "grad_norm": 0.609375,
      "learning_rate": 1.6474755222170613e-05,
      "loss": 0.3860910892486572,
      "num_input_tokens_seen": 23352194688,
      "step": 82110,
      "train_runtime": 798317.4213,
      "train_tokens_per_second": 29251.766
    },
    {
      "epoch": 2.905567582987437,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6473860987411676e-05,
      "loss": 0.38558456897735593,
      "num_input_tokens_seen": 23355018176,
      "step": 82120,
      "train_runtime": 798413.1412,
      "train_tokens_per_second": 29251.796
    },
    {
      "epoch": 2.905921403251166,
      "grad_norm": 0.65625,
      "learning_rate": 1.6472966898251688e-05,
      "loss": 0.3880720853805542,
      "num_input_tokens_seen": 23357856768,
      "step": 82130,
      "train_runtime": 798508.8794,
      "train_tokens_per_second": 29251.843
    },
    {
      "epoch": 2.9062752235148945,
      "grad_norm": 0.61328125,
      "learning_rate": 1.647207295465114e-05,
      "loss": 0.3878506660461426,
      "num_input_tokens_seen": 23360679424,
      "step": 82140,
      "train_runtime": 798604.314,
      "train_tokens_per_second": 29251.882
    },
    {
      "epoch": 2.9066290437786235,
      "grad_norm": 0.6171875,
      "learning_rate": 1.647117915657054e-05,
      "loss": 0.3875114917755127,
      "num_input_tokens_seen": 23363557888,
      "step": 82150,
      "train_runtime": 798702.0068,
      "train_tokens_per_second": 29251.908
    },
    {
      "epoch": 2.9069828640423525,
      "grad_norm": 0.57421875,
      "learning_rate": 1.6470285503970414e-05,
      "loss": 0.3877086639404297,
      "num_input_tokens_seen": 23366426304,
      "step": 82160,
      "train_runtime": 798802.766,
      "train_tokens_per_second": 29251.809
    },
    {
      "epoch": 2.907336684306081,
      "grad_norm": 0.63671875,
      "learning_rate": 1.64693919968113e-05,
      "loss": 0.38842005729675294,
      "num_input_tokens_seen": 23369299520,
      "step": 82170,
      "train_runtime": 798902.7782,
      "train_tokens_per_second": 29251.744
    },
    {
      "epoch": 2.9076905045698096,
      "grad_norm": 0.609375,
      "learning_rate": 1.646849863505375e-05,
      "loss": 0.3870576858520508,
      "num_input_tokens_seen": 23372104064,
      "step": 82180,
      "train_runtime": 798998.3905,
      "train_tokens_per_second": 29251.754
    },
    {
      "epoch": 2.9080443248335386,
      "grad_norm": 0.62890625,
      "learning_rate": 1.646760541865834e-05,
      "loss": 0.38545010089874265,
      "num_input_tokens_seen": 23374971840,
      "step": 82190,
      "train_runtime": 799095.4987,
      "train_tokens_per_second": 29251.788
    },
    {
      "epoch": 2.9083981450972676,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6466712347585644e-05,
      "loss": 0.38867011070251467,
      "num_input_tokens_seen": 23377801472,
      "step": 82200,
      "train_runtime": 799191.6627,
      "train_tokens_per_second": 29251.808
    },
    {
      "epoch": 2.908751965360996,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6465819421796264e-05,
      "loss": 0.38352065086364745,
      "num_input_tokens_seen": 23380662528,
      "step": 82210,
      "train_runtime": 799291.8311,
      "train_tokens_per_second": 29251.722
    },
    {
      "epoch": 2.909105785624725,
      "grad_norm": 0.59375,
      "learning_rate": 1.6464926641250812e-05,
      "loss": 0.38645453453063966,
      "num_input_tokens_seen": 23383518848,
      "step": 82220,
      "train_runtime": 799390.4051,
      "train_tokens_per_second": 29251.688
    },
    {
      "epoch": 2.9094596058884536,
      "grad_norm": 0.65625,
      "learning_rate": 1.646403400590992e-05,
      "loss": 0.38330936431884766,
      "num_input_tokens_seen": 23386361024,
      "step": 82230,
      "train_runtime": 799486.5176,
      "train_tokens_per_second": 29251.727
    },
    {
      "epoch": 2.9098134261521826,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6463141515734232e-05,
      "loss": 0.3893889427185059,
      "num_input_tokens_seen": 23389224576,
      "step": 82240,
      "train_runtime": 799586.4815,
      "train_tokens_per_second": 29251.651
    },
    {
      "epoch": 2.910167246415911,
      "grad_norm": 0.65234375,
      "learning_rate": 1.6462249170684403e-05,
      "loss": 0.38858442306518554,
      "num_input_tokens_seen": 23392087744,
      "step": 82250,
      "train_runtime": 799684.0798,
      "train_tokens_per_second": 29251.661
    },
    {
      "epoch": 2.91052106667964,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6461356970721107e-05,
      "loss": 0.38663530349731445,
      "num_input_tokens_seen": 23394902336,
      "step": 82260,
      "train_runtime": 799778.373,
      "train_tokens_per_second": 29251.732
    },
    {
      "epoch": 2.9108748869433687,
      "grad_norm": 0.6171875,
      "learning_rate": 1.646046491580503e-05,
      "loss": 0.38488140106201174,
      "num_input_tokens_seen": 23397752896,
      "step": 82270,
      "train_runtime": 799875.8841,
      "train_tokens_per_second": 29251.729
    },
    {
      "epoch": 2.9112287072070977,
      "grad_norm": 0.62109375,
      "learning_rate": 1.645957300589688e-05,
      "loss": 0.3893092632293701,
      "num_input_tokens_seen": 23400557440,
      "step": 82280,
      "train_runtime": 799973.2886,
      "train_tokens_per_second": 29251.673
    },
    {
      "epoch": 2.9115825274708262,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6458681240957374e-05,
      "loss": 0.38267061710357664,
      "num_input_tokens_seen": 23403410944,
      "step": 82290,
      "train_runtime": 800071.1152,
      "train_tokens_per_second": 29251.663
    },
    {
      "epoch": 2.9119363477345552,
      "grad_norm": 0.58984375,
      "learning_rate": 1.645778962094724e-05,
      "loss": 0.38626751899719236,
      "num_input_tokens_seen": 23406277184,
      "step": 82300,
      "train_runtime": 800170.0068,
      "train_tokens_per_second": 29251.63
    },
    {
      "epoch": 2.912290167998284,
      "grad_norm": 0.625,
      "learning_rate": 1.645689814582723e-05,
      "loss": 0.38466930389404297,
      "num_input_tokens_seen": 23409114752,
      "step": 82310,
      "train_runtime": 800267.402,
      "train_tokens_per_second": 29251.616
    },
    {
      "epoch": 2.9126439882620128,
      "grad_norm": 0.65234375,
      "learning_rate": 1.645600681555811e-05,
      "loss": 0.3833747863769531,
      "num_input_tokens_seen": 23411988544,
      "step": 82320,
      "train_runtime": 800367.7889,
      "train_tokens_per_second": 29251.538
    },
    {
      "epoch": 2.9129978085257413,
      "grad_norm": 0.62109375,
      "learning_rate": 1.645511563010065e-05,
      "loss": 0.3836660385131836,
      "num_input_tokens_seen": 23414833600,
      "step": 82330,
      "train_runtime": 800462.4365,
      "train_tokens_per_second": 29251.633
    },
    {
      "epoch": 2.9133516287894703,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6454224589415647e-05,
      "loss": 0.3863020896911621,
      "num_input_tokens_seen": 23417700224,
      "step": 82340,
      "train_runtime": 800565.5209,
      "train_tokens_per_second": 29251.447
    },
    {
      "epoch": 2.9137054490531993,
      "grad_norm": 0.59375,
      "learning_rate": 1.6453333693463906e-05,
      "loss": 0.3894108772277832,
      "num_input_tokens_seen": 23420516672,
      "step": 82350,
      "train_runtime": 800661.063,
      "train_tokens_per_second": 29251.475
    },
    {
      "epoch": 2.914059269316928,
      "grad_norm": 0.6171875,
      "learning_rate": 1.645244294220625e-05,
      "loss": 0.3824916362762451,
      "num_input_tokens_seen": 23423308992,
      "step": 82360,
      "train_runtime": 800755.2406,
      "train_tokens_per_second": 29251.521
    },
    {
      "epoch": 2.914413089580657,
      "grad_norm": 0.63671875,
      "learning_rate": 1.645155233560352e-05,
      "loss": 0.38927502632141114,
      "num_input_tokens_seen": 23426087808,
      "step": 82370,
      "train_runtime": 800849.6012,
      "train_tokens_per_second": 29251.545
    },
    {
      "epoch": 2.9147669098443854,
      "grad_norm": 0.625,
      "learning_rate": 1.645066187361656e-05,
      "loss": 0.38840441703796386,
      "num_input_tokens_seen": 23428857280,
      "step": 82380,
      "train_runtime": 800942.9596,
      "train_tokens_per_second": 29251.593
    },
    {
      "epoch": 2.9151207301081143,
      "grad_norm": 0.640625,
      "learning_rate": 1.6449771556206242e-05,
      "loss": 0.38769676685333254,
      "num_input_tokens_seen": 23431687424,
      "step": 82390,
      "train_runtime": 801041.7426,
      "train_tokens_per_second": 29251.519
    },
    {
      "epoch": 2.915474550371843,
      "grad_norm": 0.640625,
      "learning_rate": 1.6448881383333446e-05,
      "loss": 0.38501365184783937,
      "num_input_tokens_seen": 23434521024,
      "step": 82400,
      "train_runtime": 801135.9172,
      "train_tokens_per_second": 29251.617
    },
    {
      "epoch": 2.915828370635572,
      "grad_norm": 0.625,
      "learning_rate": 1.644799135495907e-05,
      "loss": 0.38813204765319825,
      "num_input_tokens_seen": 23437367872,
      "step": 82410,
      "train_runtime": 801234.5991,
      "train_tokens_per_second": 29251.567
    },
    {
      "epoch": 2.9161821908993004,
      "grad_norm": 0.6015625,
      "learning_rate": 1.644710147104402e-05,
      "loss": 0.3833182334899902,
      "num_input_tokens_seen": 23440213440,
      "step": 82420,
      "train_runtime": 801333.9401,
      "train_tokens_per_second": 29251.492
    },
    {
      "epoch": 2.9165360111630294,
      "grad_norm": 0.61328125,
      "learning_rate": 1.644621173154923e-05,
      "loss": 0.3859065055847168,
      "num_input_tokens_seen": 23443040192,
      "step": 82430,
      "train_runtime": 801431.143,
      "train_tokens_per_second": 29251.471
    },
    {
      "epoch": 2.916889831426758,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6445322136435636e-05,
      "loss": 0.38240962028503417,
      "num_input_tokens_seen": 23445897728,
      "step": 82440,
      "train_runtime": 801528.214,
      "train_tokens_per_second": 29251.494
    },
    {
      "epoch": 2.917243651690487,
      "grad_norm": 0.6171875,
      "learning_rate": 1.644443268566419e-05,
      "loss": 0.3901880979537964,
      "num_input_tokens_seen": 23448717952,
      "step": 82450,
      "train_runtime": 801623.2574,
      "train_tokens_per_second": 29251.544
    },
    {
      "epoch": 2.917597471954216,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6443543379195868e-05,
      "loss": 0.38433401584625243,
      "num_input_tokens_seen": 23451563712,
      "step": 82460,
      "train_runtime": 801720.5405,
      "train_tokens_per_second": 29251.544
    },
    {
      "epoch": 2.9179512922179445,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6442654216991655e-05,
      "loss": 0.3919672966003418,
      "num_input_tokens_seen": 23454397248,
      "step": 82470,
      "train_runtime": 801817.1071,
      "train_tokens_per_second": 29251.555
    },
    {
      "epoch": 2.918305112481673,
      "grad_norm": 0.5859375,
      "learning_rate": 1.6441765199012542e-05,
      "loss": 0.3850131034851074,
      "num_input_tokens_seen": 23457219136,
      "step": 82480,
      "train_runtime": 801912.2591,
      "train_tokens_per_second": 29251.603
    },
    {
      "epoch": 2.918658932745402,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6440876325219555e-05,
      "loss": 0.3848023176193237,
      "num_input_tokens_seen": 23460053376,
      "step": 82490,
      "train_runtime": 802010.9696,
      "train_tokens_per_second": 29251.537
    },
    {
      "epoch": 2.919012753009131,
      "grad_norm": 0.64453125,
      "learning_rate": 1.643998759557372e-05,
      "loss": 0.38811097145080564,
      "num_input_tokens_seen": 23462918400,
      "step": 82500,
      "train_runtime": 802111.4252,
      "train_tokens_per_second": 29251.445
    },
    {
      "epoch": 2.9193665732728595,
      "grad_norm": 0.58984375,
      "learning_rate": 1.6439099010036077e-05,
      "loss": 0.3836625814437866,
      "num_input_tokens_seen": 23465773888,
      "step": 82510,
      "train_runtime": 802211.9389,
      "train_tokens_per_second": 29251.34
    },
    {
      "epoch": 2.919720393536588,
      "grad_norm": 0.6328125,
      "learning_rate": 1.643821056856769e-05,
      "loss": 0.38422541618347167,
      "num_input_tokens_seen": 23468648320,
      "step": 82520,
      "train_runtime": 802310.2545,
      "train_tokens_per_second": 29251.338
    },
    {
      "epoch": 2.920074213800317,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6437322271129625e-05,
      "loss": 0.38550510406494143,
      "num_input_tokens_seen": 23471500992,
      "step": 82530,
      "train_runtime": 802410.2736,
      "train_tokens_per_second": 29251.247
    },
    {
      "epoch": 2.920428034064046,
      "grad_norm": 0.609375,
      "learning_rate": 1.6436434117682977e-05,
      "loss": 0.383764386177063,
      "num_input_tokens_seen": 23474317632,
      "step": 82540,
      "train_runtime": 802506.311,
      "train_tokens_per_second": 29251.256
    },
    {
      "epoch": 2.9207818543277746,
      "grad_norm": 0.6328125,
      "learning_rate": 1.643554610818885e-05,
      "loss": 0.3846308469772339,
      "num_input_tokens_seen": 23477154112,
      "step": 82550,
      "train_runtime": 802603.7564,
      "train_tokens_per_second": 29251.239
    },
    {
      "epoch": 2.9211356745915036,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6434658242608358e-05,
      "loss": 0.3879562854766846,
      "num_input_tokens_seen": 23479999040,
      "step": 82560,
      "train_runtime": 802701.8921,
      "train_tokens_per_second": 29251.207
    },
    {
      "epoch": 2.921489494855232,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6433770520902633e-05,
      "loss": 0.38408846855163575,
      "num_input_tokens_seen": 23482883584,
      "step": 82570,
      "train_runtime": 802798.8625,
      "train_tokens_per_second": 29251.267
    },
    {
      "epoch": 2.921843315118961,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6432882943032823e-05,
      "loss": 0.3868762969970703,
      "num_input_tokens_seen": 23485719744,
      "step": 82580,
      "train_runtime": 802895.6746,
      "train_tokens_per_second": 29251.272
    },
    {
      "epoch": 2.9221971353826897,
      "grad_norm": 0.625,
      "learning_rate": 1.643199550896009e-05,
      "loss": 0.38322787284851073,
      "num_input_tokens_seen": 23488491712,
      "step": 82590,
      "train_runtime": 802988.3141,
      "train_tokens_per_second": 29251.349
    },
    {
      "epoch": 2.9225509556464186,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6431108218645612e-05,
      "loss": 0.3878592014312744,
      "num_input_tokens_seen": 23491391424,
      "step": 82600,
      "train_runtime": 803089.3654,
      "train_tokens_per_second": 29251.279
    },
    {
      "epoch": 2.922904775910147,
      "grad_norm": 0.625,
      "learning_rate": 1.643022107205058e-05,
      "loss": 0.3882612228393555,
      "num_input_tokens_seen": 23494246912,
      "step": 82610,
      "train_runtime": 803189.2487,
      "train_tokens_per_second": 29251.197
    },
    {
      "epoch": 2.923258596173876,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6429334069136193e-05,
      "loss": 0.38557631969451905,
      "num_input_tokens_seen": 23497134400,
      "step": 82620,
      "train_runtime": 803288.1373,
      "train_tokens_per_second": 29251.191
    },
    {
      "epoch": 2.9236124164376047,
      "grad_norm": 0.62890625,
      "learning_rate": 1.642844720986368e-05,
      "loss": 0.3846468448638916,
      "num_input_tokens_seen": 23499930368,
      "step": 82630,
      "train_runtime": 803381.1145,
      "train_tokens_per_second": 29251.286
    },
    {
      "epoch": 2.9239662367013337,
      "grad_norm": 0.63671875,
      "learning_rate": 1.642756049419427e-05,
      "loss": 0.38690791130065916,
      "num_input_tokens_seen": 23502771072,
      "step": 82640,
      "train_runtime": 803477.3236,
      "train_tokens_per_second": 29251.318
    },
    {
      "epoch": 2.9243200569650627,
      "grad_norm": 0.625,
      "learning_rate": 1.6426673922089217e-05,
      "loss": 0.3903887033462524,
      "num_input_tokens_seen": 23505573696,
      "step": 82650,
      "train_runtime": 803571.9645,
      "train_tokens_per_second": 29251.361
    },
    {
      "epoch": 2.9246738772287912,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6425787493509783e-05,
      "loss": 0.3897097110748291,
      "num_input_tokens_seen": 23508397952,
      "step": 82660,
      "train_runtime": 803668.7908,
      "train_tokens_per_second": 29251.351
    },
    {
      "epoch": 2.92502769749252,
      "grad_norm": 0.61328125,
      "learning_rate": 1.642490120841725e-05,
      "loss": 0.3898149013519287,
      "num_input_tokens_seen": 23511237120,
      "step": 82670,
      "train_runtime": 803765.9287,
      "train_tokens_per_second": 29251.348
    },
    {
      "epoch": 2.9253815177562488,
      "grad_norm": 0.5859375,
      "learning_rate": 1.6424015066772907e-05,
      "loss": 0.38415255546569826,
      "num_input_tokens_seen": 23514068032,
      "step": 82680,
      "train_runtime": 803861.7196,
      "train_tokens_per_second": 29251.384
    },
    {
      "epoch": 2.9257353380199778,
      "grad_norm": 0.609375,
      "learning_rate": 1.6423129068538065e-05,
      "loss": 0.38874664306640627,
      "num_input_tokens_seen": 23516875840,
      "step": 82690,
      "train_runtime": 803954.4711,
      "train_tokens_per_second": 29251.502
    },
    {
      "epoch": 2.9260891582837063,
      "grad_norm": 0.61328125,
      "learning_rate": 1.642224321367405e-05,
      "loss": 0.3848646402359009,
      "num_input_tokens_seen": 23519660288,
      "step": 82700,
      "train_runtime": 804051.3578,
      "train_tokens_per_second": 29251.44
    },
    {
      "epoch": 2.9264429785474353,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6421357502142192e-05,
      "loss": 0.3840176582336426,
      "num_input_tokens_seen": 23522507648,
      "step": 82710,
      "train_runtime": 804149.8971,
      "train_tokens_per_second": 29251.397
    },
    {
      "epoch": 2.926796798811164,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6420471933903847e-05,
      "loss": 0.38656115531921387,
      "num_input_tokens_seen": 23525359104,
      "step": 82720,
      "train_runtime": 804247.9308,
      "train_tokens_per_second": 29251.377
    },
    {
      "epoch": 2.927150619074893,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6419586508920383e-05,
      "loss": 0.3846739768981934,
      "num_input_tokens_seen": 23528215360,
      "step": 82730,
      "train_runtime": 804345.542,
      "train_tokens_per_second": 29251.378
    },
    {
      "epoch": 2.9275044393386214,
      "grad_norm": 0.58984375,
      "learning_rate": 1.641870122715318e-05,
      "loss": 0.3805502414703369,
      "num_input_tokens_seen": 23531081984,
      "step": 82740,
      "train_runtime": 804444.6396,
      "train_tokens_per_second": 29251.338
    },
    {
      "epoch": 2.9278582596023504,
      "grad_norm": 0.62109375,
      "learning_rate": 1.641781608856363e-05,
      "loss": 0.3851936817169189,
      "num_input_tokens_seen": 23533993408,
      "step": 82750,
      "train_runtime": 804549.1638,
      "train_tokens_per_second": 29251.156
    },
    {
      "epoch": 2.928212079866079,
      "grad_norm": 0.66015625,
      "learning_rate": 1.6416931093113156e-05,
      "loss": 0.3823270320892334,
      "num_input_tokens_seen": 23536830208,
      "step": 82760,
      "train_runtime": 804646.945,
      "train_tokens_per_second": 29251.127
    },
    {
      "epoch": 2.928565900129808,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6416046240763165e-05,
      "loss": 0.38697922229766846,
      "num_input_tokens_seen": 23539626496,
      "step": 82770,
      "train_runtime": 804741.2888,
      "train_tokens_per_second": 29251.173
    },
    {
      "epoch": 2.9289197203935364,
      "grad_norm": 0.63671875,
      "learning_rate": 1.641516153147511e-05,
      "loss": 0.38668863773345946,
      "num_input_tokens_seen": 23542492032,
      "step": 82780,
      "train_runtime": 804839.8677,
      "train_tokens_per_second": 29251.15
    },
    {
      "epoch": 2.9292735406572654,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6414276965210436e-05,
      "loss": 0.3821967124938965,
      "num_input_tokens_seen": 23545410752,
      "step": 82790,
      "train_runtime": 804939.2334,
      "train_tokens_per_second": 29251.166
    },
    {
      "epoch": 2.9296273609209944,
      "grad_norm": 0.62109375,
      "learning_rate": 1.641339254193062e-05,
      "loss": 0.3810905456542969,
      "num_input_tokens_seen": 23548268608,
      "step": 82800,
      "train_runtime": 805037.9791,
      "train_tokens_per_second": 29251.128
    },
    {
      "epoch": 2.929981181184723,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6412508261597135e-05,
      "loss": 0.3860907554626465,
      "num_input_tokens_seen": 23551140928,
      "step": 82810,
      "train_runtime": 805136.7942,
      "train_tokens_per_second": 29251.105
    },
    {
      "epoch": 2.9303350014484515,
      "grad_norm": 0.63671875,
      "learning_rate": 1.641162412417149e-05,
      "loss": 0.3856302261352539,
      "num_input_tokens_seen": 23554037248,
      "step": 82820,
      "train_runtime": 805236.3193,
      "train_tokens_per_second": 29251.087
    },
    {
      "epoch": 2.9306888217121805,
      "grad_norm": 0.6015625,
      "learning_rate": 1.641074012961519e-05,
      "loss": 0.38336546421051027,
      "num_input_tokens_seen": 23556947584,
      "step": 82830,
      "train_runtime": 805338.2179,
      "train_tokens_per_second": 29250.999
    },
    {
      "epoch": 2.9310426419759095,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6409856277889765e-05,
      "loss": 0.38657488822937014,
      "num_input_tokens_seen": 23559806400,
      "step": 82840,
      "train_runtime": 805433.7414,
      "train_tokens_per_second": 29251.079
    },
    {
      "epoch": 2.931396462239638,
      "grad_norm": 0.625,
      "learning_rate": 1.6408972568956752e-05,
      "loss": 0.3888406276702881,
      "num_input_tokens_seen": 23562656192,
      "step": 82850,
      "train_runtime": 805532.0673,
      "train_tokens_per_second": 29251.047
    },
    {
      "epoch": 2.9317502825033666,
      "grad_norm": 0.609375,
      "learning_rate": 1.640808900277771e-05,
      "loss": 0.3857844352722168,
      "num_input_tokens_seen": 23565470976,
      "step": 82860,
      "train_runtime": 805628.9688,
      "train_tokens_per_second": 29251.022
    },
    {
      "epoch": 2.9321041027670955,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6407205579314206e-05,
      "loss": 0.3853294372558594,
      "num_input_tokens_seen": 23568335808,
      "step": 82870,
      "train_runtime": 805727.3849,
      "train_tokens_per_second": 29251.005
    },
    {
      "epoch": 2.9324579230308245,
      "grad_norm": 0.62109375,
      "learning_rate": 1.640632229852783e-05,
      "loss": 0.388051700592041,
      "num_input_tokens_seen": 23571196224,
      "step": 82880,
      "train_runtime": 805827.8079,
      "train_tokens_per_second": 29250.909
    },
    {
      "epoch": 2.932811743294553,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6405439160380177e-05,
      "loss": 0.38526253700256347,
      "num_input_tokens_seen": 23574055680,
      "step": 82890,
      "train_runtime": 805927.4114,
      "train_tokens_per_second": 29250.842
    },
    {
      "epoch": 2.933165563558282,
      "grad_norm": 0.6328125,
      "learning_rate": 1.640455616483286e-05,
      "loss": 0.3887608051300049,
      "num_input_tokens_seen": 23576903680,
      "step": 82900,
      "train_runtime": 806023.5991,
      "train_tokens_per_second": 29250.885
    },
    {
      "epoch": 2.9335193838220106,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6403673311847505e-05,
      "loss": 0.3862152099609375,
      "num_input_tokens_seen": 23579735552,
      "step": 82910,
      "train_runtime": 806118.0537,
      "train_tokens_per_second": 29250.971
    },
    {
      "epoch": 2.9338732040857396,
      "grad_norm": 0.65625,
      "learning_rate": 1.6402790601385763e-05,
      "loss": 0.38983745574951173,
      "num_input_tokens_seen": 23582601088,
      "step": 82920,
      "train_runtime": 806214.7457,
      "train_tokens_per_second": 29251.017
    },
    {
      "epoch": 2.934227024349468,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6401908033409285e-05,
      "loss": 0.38552117347717285,
      "num_input_tokens_seen": 23585431296,
      "step": 82930,
      "train_runtime": 806308.9158,
      "train_tokens_per_second": 29251.111
    },
    {
      "epoch": 2.934580844613197,
      "grad_norm": 0.60546875,
      "learning_rate": 1.640102560787974e-05,
      "loss": 0.3851055145263672,
      "num_input_tokens_seen": 23588249344,
      "step": 82940,
      "train_runtime": 806406.6898,
      "train_tokens_per_second": 29251.059
    },
    {
      "epoch": 2.9349346648769257,
      "grad_norm": 0.60546875,
      "learning_rate": 1.640014332475882e-05,
      "loss": 0.3832894802093506,
      "num_input_tokens_seen": 23591092352,
      "step": 82950,
      "train_runtime": 806502.8358,
      "train_tokens_per_second": 29251.097
    },
    {
      "epoch": 2.9352884851406547,
      "grad_norm": 0.61328125,
      "learning_rate": 1.639926118400822e-05,
      "loss": 0.3889326810836792,
      "num_input_tokens_seen": 23593946496,
      "step": 82960,
      "train_runtime": 806600.5491,
      "train_tokens_per_second": 29251.092
    },
    {
      "epoch": 2.935642305404383,
      "grad_norm": 0.625,
      "learning_rate": 1.6398379185589653e-05,
      "loss": 0.385534405708313,
      "num_input_tokens_seen": 23596771136,
      "step": 82970,
      "train_runtime": 806696.8554,
      "train_tokens_per_second": 29251.101
    },
    {
      "epoch": 2.935996125668112,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6397497329464858e-05,
      "loss": 0.3899168729782104,
      "num_input_tokens_seen": 23599582208,
      "step": 82980,
      "train_runtime": 806794.4251,
      "train_tokens_per_second": 29251.048
    },
    {
      "epoch": 2.936349945931841,
      "grad_norm": 0.60546875,
      "learning_rate": 1.639661561559557e-05,
      "loss": 0.38686366081237794,
      "num_input_tokens_seen": 23602429056,
      "step": 82990,
      "train_runtime": 806894.4242,
      "train_tokens_per_second": 29250.951
    },
    {
      "epoch": 2.9367037661955697,
      "grad_norm": 0.6015625,
      "learning_rate": 1.639573404394355e-05,
      "loss": 0.38504624366760254,
      "num_input_tokens_seen": 23605236096,
      "step": 83000,
      "train_runtime": 806988.7567,
      "train_tokens_per_second": 29251.01
    },
    {
      "epoch": 2.9370575864592983,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6394852614470568e-05,
      "loss": 0.38621604442596436,
      "num_input_tokens_seen": 23608039360,
      "step": 83010,
      "train_runtime": 807083.1546,
      "train_tokens_per_second": 29251.062
    },
    {
      "epoch": 2.9374114067230273,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6393971327138412e-05,
      "loss": 0.38556957244873047,
      "num_input_tokens_seen": 23610894848,
      "step": 83020,
      "train_runtime": 807180.9453,
      "train_tokens_per_second": 29251.056
    },
    {
      "epoch": 2.9377652269867562,
      "grad_norm": 0.625,
      "learning_rate": 1.639309018190888e-05,
      "loss": 0.38628058433532714,
      "num_input_tokens_seen": 23613715840,
      "step": 83030,
      "train_runtime": 807278.9878,
      "train_tokens_per_second": 29250.998
    },
    {
      "epoch": 2.938119047250485,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6392209178743793e-05,
      "loss": 0.37987167835235597,
      "num_input_tokens_seen": 23616599040,
      "step": 83040,
      "train_runtime": 807378.2615,
      "train_tokens_per_second": 29250.972
    },
    {
      "epoch": 2.9384728675142138,
      "grad_norm": 0.609375,
      "learning_rate": 1.639132831760497e-05,
      "loss": 0.3892202377319336,
      "num_input_tokens_seen": 23619449024,
      "step": 83050,
      "train_runtime": 807474.8821,
      "train_tokens_per_second": 29251.002
    },
    {
      "epoch": 2.9388266877779423,
      "grad_norm": 0.609375,
      "learning_rate": 1.6390447598454276e-05,
      "loss": 0.38800644874572754,
      "num_input_tokens_seen": 23622251072,
      "step": 83060,
      "train_runtime": 807567.1711,
      "train_tokens_per_second": 29251.128
    },
    {
      "epoch": 2.9391805080416713,
      "grad_norm": 0.609375,
      "learning_rate": 1.6389567021253547e-05,
      "loss": 0.3839103698730469,
      "num_input_tokens_seen": 23625106624,
      "step": 83070,
      "train_runtime": 807661.7853,
      "train_tokens_per_second": 29251.237
    },
    {
      "epoch": 2.9395343283054,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6388686585964667e-05,
      "loss": 0.38768534660339354,
      "num_input_tokens_seen": 23627990528,
      "step": 83080,
      "train_runtime": 807762.3764,
      "train_tokens_per_second": 29251.165
    },
    {
      "epoch": 2.939888148569129,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6387806292549522e-05,
      "loss": 0.3853769302368164,
      "num_input_tokens_seen": 23630893568,
      "step": 83090,
      "train_runtime": 807863.1703,
      "train_tokens_per_second": 29251.109
    },
    {
      "epoch": 2.9402419688328574,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6386926140970015e-05,
      "loss": 0.38523569107055666,
      "num_input_tokens_seen": 23633754176,
      "step": 83100,
      "train_runtime": 807960.258,
      "train_tokens_per_second": 29251.134
    },
    {
      "epoch": 2.9405957890965864,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6386046131188054e-05,
      "loss": 0.38545637130737304,
      "num_input_tokens_seen": 23636573824,
      "step": 83110,
      "train_runtime": 808055.1514,
      "train_tokens_per_second": 29251.189
    },
    {
      "epoch": 2.940949609360315,
      "grad_norm": 0.63671875,
      "learning_rate": 1.638516626316558e-05,
      "loss": 0.3828999996185303,
      "num_input_tokens_seen": 23639506816,
      "step": 83120,
      "train_runtime": 808158.5608,
      "train_tokens_per_second": 29251.075
    },
    {
      "epoch": 2.941303429624044,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6384286536864525e-05,
      "loss": 0.38341171741485597,
      "num_input_tokens_seen": 23642375104,
      "step": 83130,
      "train_runtime": 808257.5624,
      "train_tokens_per_second": 29251.041
    },
    {
      "epoch": 2.941657249887773,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6383406952246856e-05,
      "loss": 0.38609349727630615,
      "num_input_tokens_seen": 23645265600,
      "step": 83140,
      "train_runtime": 808359.3333,
      "train_tokens_per_second": 29250.934
    },
    {
      "epoch": 2.9420110701515014,
      "grad_norm": 0.640625,
      "learning_rate": 1.638252750927455e-05,
      "loss": 0.3891077995300293,
      "num_input_tokens_seen": 23648169152,
      "step": 83150,
      "train_runtime": 808462.3777,
      "train_tokens_per_second": 29250.797
    },
    {
      "epoch": 2.94236489041523,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6381648207909584e-05,
      "loss": 0.38407163619995116,
      "num_input_tokens_seen": 23651006656,
      "step": 83160,
      "train_runtime": 808560.7752,
      "train_tokens_per_second": 29250.747
    },
    {
      "epoch": 2.942718710678959,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6380769048113964e-05,
      "loss": 0.38840465545654296,
      "num_input_tokens_seen": 23653862656,
      "step": 83170,
      "train_runtime": 808659.2289,
      "train_tokens_per_second": 29250.717
    },
    {
      "epoch": 2.943072530942688,
      "grad_norm": 0.640625,
      "learning_rate": 1.6379890029849707e-05,
      "loss": 0.3854518890380859,
      "num_input_tokens_seen": 23656685696,
      "step": 83180,
      "train_runtime": 808753.499,
      "train_tokens_per_second": 29250.799
    },
    {
      "epoch": 2.9434263512064165,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6379011153078844e-05,
      "loss": 0.38362760543823243,
      "num_input_tokens_seen": 23659521088,
      "step": 83190,
      "train_runtime": 808851.3075,
      "train_tokens_per_second": 29250.767
    },
    {
      "epoch": 2.943780171470145,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6378132417763415e-05,
      "loss": 0.3878089427947998,
      "num_input_tokens_seen": 23662329792,
      "step": 83200,
      "train_runtime": 808944.838,
      "train_tokens_per_second": 29250.857
    },
    {
      "epoch": 2.944133991733874,
      "grad_norm": 0.6171875,
      "learning_rate": 1.637725382386548e-05,
      "loss": 0.3869441509246826,
      "num_input_tokens_seen": 23665161344,
      "step": 83210,
      "train_runtime": 809040.6236,
      "train_tokens_per_second": 29250.894
    },
    {
      "epoch": 2.944487811997603,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6376375371347118e-05,
      "loss": 0.3846909046173096,
      "num_input_tokens_seen": 23668013440,
      "step": 83220,
      "train_runtime": 809137.3557,
      "train_tokens_per_second": 29250.922
    },
    {
      "epoch": 2.9448416322613316,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6375497060170403e-05,
      "loss": 0.3894462585449219,
      "num_input_tokens_seen": 23670866432,
      "step": 83230,
      "train_runtime": 809235.3718,
      "train_tokens_per_second": 29250.904
    },
    {
      "epoch": 2.9451954525250605,
      "grad_norm": 0.609375,
      "learning_rate": 1.637461889029745e-05,
      "loss": 0.39057221412658694,
      "num_input_tokens_seen": 23673660032,
      "step": 83240,
      "train_runtime": 809329.3318,
      "train_tokens_per_second": 29250.96
    },
    {
      "epoch": 2.945549272788789,
      "grad_norm": 0.625,
      "learning_rate": 1.637374086169037e-05,
      "loss": 0.3871551513671875,
      "num_input_tokens_seen": 23676509120,
      "step": 83250,
      "train_runtime": 809428.5724,
      "train_tokens_per_second": 29250.894
    },
    {
      "epoch": 2.945903093052518,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6372862974311286e-05,
      "loss": 0.3927941083908081,
      "num_input_tokens_seen": 23679353472,
      "step": 83260,
      "train_runtime": 809525.959,
      "train_tokens_per_second": 29250.888
    },
    {
      "epoch": 2.9462569133162466,
      "grad_norm": 0.609375,
      "learning_rate": 1.637198522812235e-05,
      "loss": 0.38093230724334715,
      "num_input_tokens_seen": 23682166592,
      "step": 83270,
      "train_runtime": 809620.0608,
      "train_tokens_per_second": 29250.963
    },
    {
      "epoch": 2.9466107335799756,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6371107623085722e-05,
      "loss": 0.38582792282104494,
      "num_input_tokens_seen": 23684986752,
      "step": 83280,
      "train_runtime": 809713.6848,
      "train_tokens_per_second": 29251.064
    },
    {
      "epoch": 2.9469645538437046,
      "grad_norm": 0.63671875,
      "learning_rate": 1.637023015916356e-05,
      "loss": 0.3854196786880493,
      "num_input_tokens_seen": 23687790464,
      "step": 83290,
      "train_runtime": 809809.1079,
      "train_tokens_per_second": 29251.079
    },
    {
      "epoch": 2.947318374107433,
      "grad_norm": 0.6171875,
      "learning_rate": 1.636935283631807e-05,
      "loss": 0.3870986461639404,
      "num_input_tokens_seen": 23690656576,
      "step": 83300,
      "train_runtime": 809908.5148,
      "train_tokens_per_second": 29251.028
    },
    {
      "epoch": 2.9476721943711617,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6368475654511438e-05,
      "loss": 0.3840007781982422,
      "num_input_tokens_seen": 23693496896,
      "step": 83310,
      "train_runtime": 810006.6314,
      "train_tokens_per_second": 29250.991
    },
    {
      "epoch": 2.9480260146348907,
      "grad_norm": 0.640625,
      "learning_rate": 1.6367598613705887e-05,
      "loss": 0.38996109962463377,
      "num_input_tokens_seen": 23696350464,
      "step": 83320,
      "train_runtime": 810106.3755,
      "train_tokens_per_second": 29250.912
    },
    {
      "epoch": 2.9483798348986197,
      "grad_norm": 0.62109375,
      "learning_rate": 1.636672171386364e-05,
      "loss": 0.3854548454284668,
      "num_input_tokens_seen": 23699161920,
      "step": 83330,
      "train_runtime": 810200.3888,
      "train_tokens_per_second": 29250.988
    },
    {
      "epoch": 2.948733655162348,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6365844954946946e-05,
      "loss": 0.38654074668884275,
      "num_input_tokens_seen": 23702015040,
      "step": 83340,
      "train_runtime": 810298.678,
      "train_tokens_per_second": 29250.961
    },
    {
      "epoch": 2.9490874754260767,
      "grad_norm": 0.609375,
      "learning_rate": 1.6364968336918064e-05,
      "loss": 0.38648381233215334,
      "num_input_tokens_seen": 23704874816,
      "step": 83350,
      "train_runtime": 810395.9266,
      "train_tokens_per_second": 29250.98
    },
    {
      "epoch": 2.9494412956898057,
      "grad_norm": 0.6484375,
      "learning_rate": 1.636409185973925e-05,
      "loss": 0.3865505218505859,
      "num_input_tokens_seen": 23707676672,
      "step": 83360,
      "train_runtime": 810491.4689,
      "train_tokens_per_second": 29250.989
    },
    {
      "epoch": 2.9497951159535347,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6363215523372812e-05,
      "loss": 0.38443286418914796,
      "num_input_tokens_seen": 23710491712,
      "step": 83370,
      "train_runtime": 810586.5557,
      "train_tokens_per_second": 29251.03
    },
    {
      "epoch": 2.9501489362172633,
      "grad_norm": 0.609375,
      "learning_rate": 1.6362339327781036e-05,
      "loss": 0.38669571876525877,
      "num_input_tokens_seen": 23713341312,
      "step": 83380,
      "train_runtime": 810683.6683,
      "train_tokens_per_second": 29251.041
    },
    {
      "epoch": 2.9505027564809923,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6361463272926238e-05,
      "loss": 0.3865213394165039,
      "num_input_tokens_seen": 23716154880,
      "step": 83390,
      "train_runtime": 810779.7494,
      "train_tokens_per_second": 29251.045
    },
    {
      "epoch": 2.950856576744721,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6360587358770747e-05,
      "loss": 0.3836618423461914,
      "num_input_tokens_seen": 23719030144,
      "step": 83400,
      "train_runtime": 810878.6613,
      "train_tokens_per_second": 29251.023
    },
    {
      "epoch": 2.95121039700845,
      "grad_norm": 0.625,
      "learning_rate": 1.635971158527691e-05,
      "loss": 0.38424320220947267,
      "num_input_tokens_seen": 23721824896,
      "step": 83410,
      "train_runtime": 810973.1139,
      "train_tokens_per_second": 29251.062
    },
    {
      "epoch": 2.9515642172721783,
      "grad_norm": 0.625,
      "learning_rate": 1.6358835952407074e-05,
      "loss": 0.3885367393493652,
      "num_input_tokens_seen": 23724640000,
      "step": 83420,
      "train_runtime": 811069.0067,
      "train_tokens_per_second": 29251.075
    },
    {
      "epoch": 2.9519180375359073,
      "grad_norm": 0.59375,
      "learning_rate": 1.6357960460123614e-05,
      "loss": 0.38461220264434814,
      "num_input_tokens_seen": 23727508160,
      "step": 83430,
      "train_runtime": 811165.983,
      "train_tokens_per_second": 29251.113
    },
    {
      "epoch": 2.952271857799636,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6357085108388913e-05,
      "loss": 0.38738555908203126,
      "num_input_tokens_seen": 23730346240,
      "step": 83440,
      "train_runtime": 811264.0204,
      "train_tokens_per_second": 29251.077
    },
    {
      "epoch": 2.952625678063365,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6356209897165374e-05,
      "loss": 0.39089107513427734,
      "num_input_tokens_seen": 23733172224,
      "step": 83450,
      "train_runtime": 811360.8403,
      "train_tokens_per_second": 29251.069
    },
    {
      "epoch": 2.9529794983270934,
      "grad_norm": 0.62890625,
      "learning_rate": 1.635533482641541e-05,
      "loss": 0.38880579471588134,
      "num_input_tokens_seen": 23736043904,
      "step": 83460,
      "train_runtime": 811460.4281,
      "train_tokens_per_second": 29251.018
    },
    {
      "epoch": 2.9533333185908224,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6354459896101436e-05,
      "loss": 0.38423309326171873,
      "num_input_tokens_seen": 23738905920,
      "step": 83470,
      "train_runtime": 811561.4152,
      "train_tokens_per_second": 29250.905
    },
    {
      "epoch": 2.9536871388545514,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6353585106185908e-05,
      "loss": 0.38716278076171873,
      "num_input_tokens_seen": 23741739648,
      "step": 83480,
      "train_runtime": 811659.9734,
      "train_tokens_per_second": 29250.844
    },
    {
      "epoch": 2.95404095911828,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6352710456631275e-05,
      "loss": 0.38989255428314207,
      "num_input_tokens_seen": 23744573248,
      "step": 83490,
      "train_runtime": 811754.8302,
      "train_tokens_per_second": 29250.917
    },
    {
      "epoch": 2.9543947793820085,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6351835947400003e-05,
      "loss": 0.38204116821289064,
      "num_input_tokens_seen": 23747442240,
      "step": 83500,
      "train_runtime": 811852.6835,
      "train_tokens_per_second": 29250.925
    },
    {
      "epoch": 2.9547485996457374,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6350961578454578e-05,
      "loss": 0.3851689577102661,
      "num_input_tokens_seen": 23750253312,
      "step": 83510,
      "train_runtime": 811946.1042,
      "train_tokens_per_second": 29251.022
    },
    {
      "epoch": 2.9551024199094664,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6350087349757494e-05,
      "loss": 0.38490025997161864,
      "num_input_tokens_seen": 23752986560,
      "step": 83520,
      "train_runtime": 812038.1159,
      "train_tokens_per_second": 29251.073
    },
    {
      "epoch": 2.955456240173195,
      "grad_norm": 0.625,
      "learning_rate": 1.6349213261271268e-05,
      "loss": 0.3878573656082153,
      "num_input_tokens_seen": 23755793728,
      "step": 83530,
      "train_runtime": 812131.7532,
      "train_tokens_per_second": 29251.157
    },
    {
      "epoch": 2.955810060436924,
      "grad_norm": 0.62890625,
      "learning_rate": 1.634833931295842e-05,
      "loss": 0.38826005458831786,
      "num_input_tokens_seen": 23758717696,
      "step": 83540,
      "train_runtime": 812233.6767,
      "train_tokens_per_second": 29251.087
    },
    {
      "epoch": 2.9561638807006525,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6347465504781485e-05,
      "loss": 0.38635029792785647,
      "num_input_tokens_seen": 23761537792,
      "step": 83550,
      "train_runtime": 812327.9192,
      "train_tokens_per_second": 29251.165
    },
    {
      "epoch": 2.9565177009643815,
      "grad_norm": 0.62890625,
      "learning_rate": 1.634659183670303e-05,
      "loss": 0.38464970588684083,
      "num_input_tokens_seen": 23764342720,
      "step": 83560,
      "train_runtime": 812422.5798,
      "train_tokens_per_second": 29251.209
    },
    {
      "epoch": 2.95687152122811,
      "grad_norm": 0.640625,
      "learning_rate": 1.634571830868561e-05,
      "loss": 0.38938307762145996,
      "num_input_tokens_seen": 23767170432,
      "step": 83570,
      "train_runtime": 812518.6613,
      "train_tokens_per_second": 29251.23
    },
    {
      "epoch": 2.957225341491839,
      "grad_norm": 0.62109375,
      "learning_rate": 1.634484492069181e-05,
      "loss": 0.38609766960144043,
      "num_input_tokens_seen": 23770029248,
      "step": 83580,
      "train_runtime": 812618.9286,
      "train_tokens_per_second": 29251.139
    },
    {
      "epoch": 2.9575791617555676,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6343971672684228e-05,
      "loss": 0.3823888063430786,
      "num_input_tokens_seen": 23772902400,
      "step": 83590,
      "train_runtime": 812717.1496,
      "train_tokens_per_second": 29251.139
    },
    {
      "epoch": 2.9579329820192966,
      "grad_norm": 0.65625,
      "learning_rate": 1.634309856462547e-05,
      "loss": 0.38741140365600585,
      "num_input_tokens_seen": 23775763584,
      "step": 83600,
      "train_runtime": 812816.5878,
      "train_tokens_per_second": 29251.081
    },
    {
      "epoch": 2.958286802283025,
      "grad_norm": 0.609375,
      "learning_rate": 1.6342225596478154e-05,
      "loss": 0.39187190532684324,
      "num_input_tokens_seen": 23778612032,
      "step": 83610,
      "train_runtime": 812912.3357,
      "train_tokens_per_second": 29251.139
    },
    {
      "epoch": 2.958640622546754,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6341352768204925e-05,
      "loss": 0.3843673229217529,
      "num_input_tokens_seen": 23781454720,
      "step": 83620,
      "train_runtime": 813011.2587,
      "train_tokens_per_second": 29251.077
    },
    {
      "epoch": 2.958994442810483,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6340480079768432e-05,
      "loss": 0.3842111110687256,
      "num_input_tokens_seen": 23784243904,
      "step": 83630,
      "train_runtime": 813107.931,
      "train_tokens_per_second": 29251.029
    },
    {
      "epoch": 2.9593482630742116,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6339607531131343e-05,
      "loss": 0.38688862323760986,
      "num_input_tokens_seen": 23787060416,
      "step": 83640,
      "train_runtime": 813205.2827,
      "train_tokens_per_second": 29250.991
    },
    {
      "epoch": 2.95970208333794,
      "grad_norm": 0.59375,
      "learning_rate": 1.633873512225633e-05,
      "loss": 0.3859706401824951,
      "num_input_tokens_seen": 23789838016,
      "step": 83650,
      "train_runtime": 813298.9852,
      "train_tokens_per_second": 29251.036
    },
    {
      "epoch": 2.960055903601669,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6337862853106086e-05,
      "loss": 0.385085129737854,
      "num_input_tokens_seen": 23792624640,
      "step": 83660,
      "train_runtime": 813393.4514,
      "train_tokens_per_second": 29251.065
    },
    {
      "epoch": 2.960409723865398,
      "grad_norm": 0.6171875,
      "learning_rate": 1.633699072364333e-05,
      "loss": 0.3840926170349121,
      "num_input_tokens_seen": 23795411712,
      "step": 83670,
      "train_runtime": 813488.1029,
      "train_tokens_per_second": 29251.088
    },
    {
      "epoch": 2.9607635441291267,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6336118733830765e-05,
      "loss": 0.39058752059936525,
      "num_input_tokens_seen": 23798338304,
      "step": 83680,
      "train_runtime": 813589.8391,
      "train_tokens_per_second": 29251.027
    },
    {
      "epoch": 2.9611173643928552,
      "grad_norm": 0.6015625,
      "learning_rate": 1.633524688363114e-05,
      "loss": 0.385579252243042,
      "num_input_tokens_seen": 23801185152,
      "step": 83690,
      "train_runtime": 813687.262,
      "train_tokens_per_second": 29251.023
    },
    {
      "epoch": 2.961471184656584,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6334375173007193e-05,
      "loss": 0.383055591583252,
      "num_input_tokens_seen": 23804061376,
      "step": 83700,
      "train_runtime": 813785.6328,
      "train_tokens_per_second": 29251.022
    },
    {
      "epoch": 2.961825004920313,
      "grad_norm": 0.625,
      "learning_rate": 1.6333503601921697e-05,
      "loss": 0.38640170097351073,
      "num_input_tokens_seen": 23806873920,
      "step": 83710,
      "train_runtime": 813884.2769,
      "train_tokens_per_second": 29250.932
    },
    {
      "epoch": 2.9621788251840417,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6332632170337422e-05,
      "loss": 0.3885246753692627,
      "num_input_tokens_seen": 23809774912,
      "step": 83720,
      "train_runtime": 813987.0272,
      "train_tokens_per_second": 29250.804
    },
    {
      "epoch": 2.9625326454477707,
      "grad_norm": 0.609375,
      "learning_rate": 1.6331760878217162e-05,
      "loss": 0.38816304206848146,
      "num_input_tokens_seen": 23812611520,
      "step": 83730,
      "train_runtime": 814080.9216,
      "train_tokens_per_second": 29250.915
    },
    {
      "epoch": 2.9628864657114993,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6330889725523713e-05,
      "loss": 0.38534929752349856,
      "num_input_tokens_seen": 23815486784,
      "step": 83740,
      "train_runtime": 814181.2573,
      "train_tokens_per_second": 29250.841
    },
    {
      "epoch": 2.9632402859752283,
      "grad_norm": 0.6484375,
      "learning_rate": 1.6330018712219902e-05,
      "loss": 0.38374896049499513,
      "num_input_tokens_seen": 23818351040,
      "step": 83750,
      "train_runtime": 814278.0851,
      "train_tokens_per_second": 29250.881
    },
    {
      "epoch": 2.963594106238957,
      "grad_norm": 0.62109375,
      "learning_rate": 1.632914783826856e-05,
      "loss": 0.38723015785217285,
      "num_input_tokens_seen": 23821142784,
      "step": 83760,
      "train_runtime": 814371.8529,
      "train_tokens_per_second": 29250.941
    },
    {
      "epoch": 2.963947926502686,
      "grad_norm": 0.6171875,
      "learning_rate": 1.632827710363253e-05,
      "loss": 0.386868953704834,
      "num_input_tokens_seen": 23823973440,
      "step": 83770,
      "train_runtime": 814469.9264,
      "train_tokens_per_second": 29250.894
    },
    {
      "epoch": 2.9643017467664143,
      "grad_norm": 0.61328125,
      "learning_rate": 1.632740650827467e-05,
      "loss": 0.387319278717041,
      "num_input_tokens_seen": 23826783360,
      "step": 83780,
      "train_runtime": 814564.5968,
      "train_tokens_per_second": 29250.944
    },
    {
      "epoch": 2.9646555670301433,
      "grad_norm": 0.64453125,
      "learning_rate": 1.632653605215786e-05,
      "loss": 0.38904705047607424,
      "num_input_tokens_seen": 23829662080,
      "step": 83790,
      "train_runtime": 814665.6683,
      "train_tokens_per_second": 29250.849
    },
    {
      "epoch": 2.965009387293872,
      "grad_norm": 0.609375,
      "learning_rate": 1.6325665735244984e-05,
      "loss": 0.38100247383117675,
      "num_input_tokens_seen": 23832484480,
      "step": 83800,
      "train_runtime": 814762.2296,
      "train_tokens_per_second": 29250.846
    },
    {
      "epoch": 2.965363207557601,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6324795557498935e-05,
      "loss": 0.38549013137817384,
      "num_input_tokens_seen": 23835282624,
      "step": 83810,
      "train_runtime": 814856.1657,
      "train_tokens_per_second": 29250.908
    },
    {
      "epoch": 2.96571702782133,
      "grad_norm": 0.6484375,
      "learning_rate": 1.6323925518882646e-05,
      "loss": 0.39046125411987304,
      "num_input_tokens_seen": 23838149952,
      "step": 83820,
      "train_runtime": 814954.5309,
      "train_tokens_per_second": 29250.896
    },
    {
      "epoch": 2.9660708480850584,
      "grad_norm": 0.625,
      "learning_rate": 1.632305561935903e-05,
      "loss": 0.3899209499359131,
      "num_input_tokens_seen": 23840988480,
      "step": 83830,
      "train_runtime": 815052.8829,
      "train_tokens_per_second": 29250.849
    },
    {
      "epoch": 2.966424668348787,
      "grad_norm": 0.62890625,
      "learning_rate": 1.632218585889104e-05,
      "loss": 0.384243369102478,
      "num_input_tokens_seen": 23843888640,
      "step": 83840,
      "train_runtime": 815152.8016,
      "train_tokens_per_second": 29250.821
    },
    {
      "epoch": 2.966778488612516,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6321316237441624e-05,
      "loss": 0.38520216941833496,
      "num_input_tokens_seen": 23846769472,
      "step": 83850,
      "train_runtime": 815252.6967,
      "train_tokens_per_second": 29250.77
    },
    {
      "epoch": 2.967132308876245,
      "grad_norm": 0.6171875,
      "learning_rate": 1.632044675497376e-05,
      "loss": 0.38357346057891845,
      "num_input_tokens_seen": 23849633408,
      "step": 83860,
      "train_runtime": 815349.9518,
      "train_tokens_per_second": 29250.794
    },
    {
      "epoch": 2.9674861291399734,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6319577411450427e-05,
      "loss": 0.38504045009613036,
      "num_input_tokens_seen": 23852474496,
      "step": 83870,
      "train_runtime": 815448.7905,
      "train_tokens_per_second": 29250.733
    },
    {
      "epoch": 2.9678399494037024,
      "grad_norm": 0.62109375,
      "learning_rate": 1.631870820683463e-05,
      "loss": 0.38596034049987793,
      "num_input_tokens_seen": 23855295936,
      "step": 83880,
      "train_runtime": 815543.3712,
      "train_tokens_per_second": 29250.8
    },
    {
      "epoch": 2.968193769667431,
      "grad_norm": 0.625,
      "learning_rate": 1.6317839141089375e-05,
      "loss": 0.3841878414154053,
      "num_input_tokens_seen": 23858154432,
      "step": 83890,
      "train_runtime": 815642.4243,
      "train_tokens_per_second": 29250.752
    },
    {
      "epoch": 2.96854758993116,
      "grad_norm": 0.66796875,
      "learning_rate": 1.6316970214177686e-05,
      "loss": 0.3918689966201782,
      "num_input_tokens_seen": 23861043520,
      "step": 83900,
      "train_runtime": 815741.8705,
      "train_tokens_per_second": 29250.728
    },
    {
      "epoch": 2.9689014101948885,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6316101426062604e-05,
      "loss": 0.38515558242797854,
      "num_input_tokens_seen": 23863870976,
      "step": 83910,
      "train_runtime": 815840.8788,
      "train_tokens_per_second": 29250.644
    },
    {
      "epoch": 2.9692552304586175,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6315232776707187e-05,
      "loss": 0.387478232383728,
      "num_input_tokens_seen": 23866709632,
      "step": 83920,
      "train_runtime": 815936.0057,
      "train_tokens_per_second": 29250.713
    },
    {
      "epoch": 2.969609050722346,
      "grad_norm": 0.62109375,
      "learning_rate": 1.63143642660745e-05,
      "loss": 0.38737921714782714,
      "num_input_tokens_seen": 23869575552,
      "step": 83930,
      "train_runtime": 816033.6368,
      "train_tokens_per_second": 29250.725
    },
    {
      "epoch": 2.969962870986075,
      "grad_norm": 0.6015625,
      "learning_rate": 1.631349589412762e-05,
      "loss": 0.38399181365966795,
      "num_input_tokens_seen": 23872414336,
      "step": 83940,
      "train_runtime": 816130.8162,
      "train_tokens_per_second": 29250.72
    },
    {
      "epoch": 2.9703166912498036,
      "grad_norm": 0.61328125,
      "learning_rate": 1.631262766082964e-05,
      "loss": 0.3832873344421387,
      "num_input_tokens_seen": 23875271040,
      "step": 83950,
      "train_runtime": 816228.2014,
      "train_tokens_per_second": 29250.73
    },
    {
      "epoch": 2.9706705115135326,
      "grad_norm": 0.6328125,
      "learning_rate": 1.631175956614368e-05,
      "loss": 0.39100937843322753,
      "num_input_tokens_seen": 23878081408,
      "step": 83960,
      "train_runtime": 816324.8059,
      "train_tokens_per_second": 29250.712
    },
    {
      "epoch": 2.9710243317772616,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6310891610032843e-05,
      "loss": 0.3873640775680542,
      "num_input_tokens_seen": 23880967424,
      "step": 83970,
      "train_runtime": 816423.2244,
      "train_tokens_per_second": 29250.72
    },
    {
      "epoch": 2.97137815204099,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6310023792460282e-05,
      "loss": 0.3853854417800903,
      "num_input_tokens_seen": 23883809984,
      "step": 83980,
      "train_runtime": 816517.7085,
      "train_tokens_per_second": 29250.817
    },
    {
      "epoch": 2.9717319723047186,
      "grad_norm": 0.6015625,
      "learning_rate": 1.630915611338914e-05,
      "loss": 0.384059476852417,
      "num_input_tokens_seen": 23886686784,
      "step": 83990,
      "train_runtime": 816618.8333,
      "train_tokens_per_second": 29250.717
    },
    {
      "epoch": 2.9720857925684476,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6308288572782576e-05,
      "loss": 0.382511568069458,
      "num_input_tokens_seen": 23889513152,
      "step": 84000,
      "train_runtime": 816716.6254,
      "train_tokens_per_second": 29250.676
    },
    {
      "epoch": 2.9724396128321766,
      "grad_norm": 0.57421875,
      "learning_rate": 1.6307421170603776e-05,
      "loss": 0.3868354320526123,
      "num_input_tokens_seen": 23892358592,
      "step": 84010,
      "train_runtime": 816813.8951,
      "train_tokens_per_second": 29250.676
    },
    {
      "epoch": 2.972793433095905,
      "grad_norm": 0.62109375,
      "learning_rate": 1.630655390681592e-05,
      "loss": 0.38753175735473633,
      "num_input_tokens_seen": 23895242368,
      "step": 84020,
      "train_runtime": 816912.2422,
      "train_tokens_per_second": 29250.685
    },
    {
      "epoch": 2.9731472533596337,
      "grad_norm": 0.6015625,
      "learning_rate": 1.6305686781382223e-05,
      "loss": 0.3871962785720825,
      "num_input_tokens_seen": 23898074240,
      "step": 84030,
      "train_runtime": 817007.2581,
      "train_tokens_per_second": 29250.749
    },
    {
      "epoch": 2.9735010736233627,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6304819794265894e-05,
      "loss": 0.38687567710876464,
      "num_input_tokens_seen": 23900947776,
      "step": 84040,
      "train_runtime": 817107.5339,
      "train_tokens_per_second": 29250.676
    },
    {
      "epoch": 2.9738548938870917,
      "grad_norm": 0.63671875,
      "learning_rate": 1.630395294543017e-05,
      "loss": 0.38591742515563965,
      "num_input_tokens_seen": 23903795648,
      "step": 84050,
      "train_runtime": 817205.3436,
      "train_tokens_per_second": 29250.66
    },
    {
      "epoch": 2.97420871415082,
      "grad_norm": 0.609375,
      "learning_rate": 1.6303086234838297e-05,
      "loss": 0.3876797199249268,
      "num_input_tokens_seen": 23906608384,
      "step": 84060,
      "train_runtime": 817302.2332,
      "train_tokens_per_second": 29250.634
    },
    {
      "epoch": 2.974562534414549,
      "grad_norm": 0.609375,
      "learning_rate": 1.6302219662453528e-05,
      "loss": 0.38749096393585203,
      "num_input_tokens_seen": 23909471936,
      "step": 84070,
      "train_runtime": 817399.5166,
      "train_tokens_per_second": 29250.656
    },
    {
      "epoch": 2.9749163546782778,
      "grad_norm": 0.640625,
      "learning_rate": 1.6301353228239137e-05,
      "loss": 0.3863779067993164,
      "num_input_tokens_seen": 23912263040,
      "step": 84080,
      "train_runtime": 817494.1015,
      "train_tokens_per_second": 29250.686
    },
    {
      "epoch": 2.9752701749420067,
      "grad_norm": 0.61328125,
      "learning_rate": 1.630048693215842e-05,
      "loss": 0.3879044532775879,
      "num_input_tokens_seen": 23915095168,
      "step": 84090,
      "train_runtime": 817591.5215,
      "train_tokens_per_second": 29250.664
    },
    {
      "epoch": 2.9756239952057353,
      "grad_norm": 0.58984375,
      "learning_rate": 1.6299620774174663e-05,
      "loss": 0.38594398498535154,
      "num_input_tokens_seen": 23917917952,
      "step": 84100,
      "train_runtime": 817684.8408,
      "train_tokens_per_second": 29250.778
    },
    {
      "epoch": 2.9759778154694643,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6298754754251192e-05,
      "loss": 0.38448729515075686,
      "num_input_tokens_seen": 23920775040,
      "step": 84110,
      "train_runtime": 817784.7256,
      "train_tokens_per_second": 29250.699
    },
    {
      "epoch": 2.9763316357331933,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6297888872351324e-05,
      "loss": 0.3839802503585815,
      "num_input_tokens_seen": 23923597440,
      "step": 84120,
      "train_runtime": 817879.9603,
      "train_tokens_per_second": 29250.744
    },
    {
      "epoch": 2.976685455996922,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6297023128438407e-05,
      "loss": 0.3864637851715088,
      "num_input_tokens_seen": 23926424320,
      "step": 84130,
      "train_runtime": 817974.4292,
      "train_tokens_per_second": 29250.822
    },
    {
      "epoch": 2.9770392762606503,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6296157522475794e-05,
      "loss": 0.381453800201416,
      "num_input_tokens_seen": 23929324800,
      "step": 84140,
      "train_runtime": 818076.6973,
      "train_tokens_per_second": 29250.711
    },
    {
      "epoch": 2.9773930965243793,
      "grad_norm": 0.62890625,
      "learning_rate": 1.629529205442685e-05,
      "loss": 0.38701021671295166,
      "num_input_tokens_seen": 23932135424,
      "step": 84150,
      "train_runtime": 818170.1628,
      "train_tokens_per_second": 29250.804
    },
    {
      "epoch": 2.9777469167881083,
      "grad_norm": 0.62109375,
      "learning_rate": 1.629442672425496e-05,
      "loss": 0.3835440635681152,
      "num_input_tokens_seen": 23934969664,
      "step": 84160,
      "train_runtime": 818268.1814,
      "train_tokens_per_second": 29250.764
    },
    {
      "epoch": 2.978100737051837,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6293561531923517e-05,
      "loss": 0.38739843368530275,
      "num_input_tokens_seen": 23937865792,
      "step": 84170,
      "train_runtime": 818369.1988,
      "train_tokens_per_second": 29250.693
    },
    {
      "epoch": 2.9784545573155654,
      "grad_norm": 0.609375,
      "learning_rate": 1.6292696477395934e-05,
      "loss": 0.3834614515304565,
      "num_input_tokens_seen": 23940704256,
      "step": 84180,
      "train_runtime": 818463.4353,
      "train_tokens_per_second": 29250.793
    },
    {
      "epoch": 2.9788083775792944,
      "grad_norm": 0.609375,
      "learning_rate": 1.6291831560635628e-05,
      "loss": 0.384566330909729,
      "num_input_tokens_seen": 23943607104,
      "step": 84190,
      "train_runtime": 818565.4569,
      "train_tokens_per_second": 29250.693
    },
    {
      "epoch": 2.9791621978430234,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6290966781606043e-05,
      "loss": 0.38975088596343993,
      "num_input_tokens_seen": 23946468992,
      "step": 84200,
      "train_runtime": 818665.2585,
      "train_tokens_per_second": 29250.623
    },
    {
      "epoch": 2.979516018106752,
      "grad_norm": 0.6484375,
      "learning_rate": 1.6290102140270622e-05,
      "loss": 0.38160152435302735,
      "num_input_tokens_seen": 23949304768,
      "step": 84210,
      "train_runtime": 818763.0072,
      "train_tokens_per_second": 29250.595
    },
    {
      "epoch": 2.979869838370481,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6289237636592827e-05,
      "loss": 0.3898234605789185,
      "num_input_tokens_seen": 23952142016,
      "step": 84220,
      "train_runtime": 818861.5797,
      "train_tokens_per_second": 29250.538
    },
    {
      "epoch": 2.9802236586342095,
      "grad_norm": 0.62109375,
      "learning_rate": 1.628837327053614e-05,
      "loss": 0.3887813568115234,
      "num_input_tokens_seen": 23955013248,
      "step": 84230,
      "train_runtime": 818960.0758,
      "train_tokens_per_second": 29250.526
    },
    {
      "epoch": 2.9805774788979384,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6287509042064053e-05,
      "loss": 0.38473570346832275,
      "num_input_tokens_seen": 23957892288,
      "step": 84240,
      "train_runtime": 819061.6066,
      "train_tokens_per_second": 29250.415
    },
    {
      "epoch": 2.980931299161667,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6286644951140067e-05,
      "loss": 0.38391947746276855,
      "num_input_tokens_seen": 23960771200,
      "step": 84250,
      "train_runtime": 819160.1108,
      "train_tokens_per_second": 29250.413
    },
    {
      "epoch": 2.981285119425396,
      "grad_norm": 0.6484375,
      "learning_rate": 1.6285780997727696e-05,
      "loss": 0.38719513416290285,
      "num_input_tokens_seen": 23963668736,
      "step": 84260,
      "train_runtime": 819257.5901,
      "train_tokens_per_second": 29250.469
    },
    {
      "epoch": 2.9816389396891245,
      "grad_norm": 0.640625,
      "learning_rate": 1.6284917181790476e-05,
      "loss": 0.3832190990447998,
      "num_input_tokens_seen": 23966484160,
      "step": 84270,
      "train_runtime": 819352.6332,
      "train_tokens_per_second": 29250.512
    },
    {
      "epoch": 2.9819927599528535,
      "grad_norm": 0.6171875,
      "learning_rate": 1.628405350329195e-05,
      "loss": 0.3852893590927124,
      "num_input_tokens_seen": 23969362176,
      "step": 84280,
      "train_runtime": 819451.2226,
      "train_tokens_per_second": 29250.505
    },
    {
      "epoch": 2.982346580216582,
      "grad_norm": 0.609375,
      "learning_rate": 1.628318996219568e-05,
      "loss": 0.3811798572540283,
      "num_input_tokens_seen": 23972235456,
      "step": 84290,
      "train_runtime": 819551.1124,
      "train_tokens_per_second": 29250.446
    },
    {
      "epoch": 2.982700400480311,
      "grad_norm": 0.640625,
      "learning_rate": 1.628232655846523e-05,
      "loss": 0.38569204807281493,
      "num_input_tokens_seen": 23975004352,
      "step": 84300,
      "train_runtime": 819646.7948,
      "train_tokens_per_second": 29250.409
    },
    {
      "epoch": 2.98305422074404,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6281463292064193e-05,
      "loss": 0.3844261646270752,
      "num_input_tokens_seen": 23977789120,
      "step": 84310,
      "train_runtime": 819741.5913,
      "train_tokens_per_second": 29250.424
    },
    {
      "epoch": 2.9834080410077686,
      "grad_norm": 0.625,
      "learning_rate": 1.628060016295616e-05,
      "loss": 0.38418993949890134,
      "num_input_tokens_seen": 23980638016,
      "step": 84320,
      "train_runtime": 819842.4247,
      "train_tokens_per_second": 29250.301
    },
    {
      "epoch": 2.983761861271497,
      "grad_norm": 0.63671875,
      "learning_rate": 1.6279737171104752e-05,
      "loss": 0.38731942176818845,
      "num_input_tokens_seen": 23983545536,
      "step": 84330,
      "train_runtime": 819941.5167,
      "train_tokens_per_second": 29250.312
    },
    {
      "epoch": 2.984115681535226,
      "grad_norm": 0.625,
      "learning_rate": 1.6278874316473593e-05,
      "loss": 0.3855140686035156,
      "num_input_tokens_seen": 23986342400,
      "step": 84340,
      "train_runtime": 820036.4802,
      "train_tokens_per_second": 29250.336
    },
    {
      "epoch": 2.984469501798955,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6278011599026314e-05,
      "loss": 0.3823753595352173,
      "num_input_tokens_seen": 23989209536,
      "step": 84350,
      "train_runtime": 820135.6798,
      "train_tokens_per_second": 29250.294
    },
    {
      "epoch": 2.9848233220626836,
      "grad_norm": 0.62890625,
      "learning_rate": 1.627714901872657e-05,
      "loss": 0.3882883071899414,
      "num_input_tokens_seen": 23992052352,
      "step": 84360,
      "train_runtime": 820230.2152,
      "train_tokens_per_second": 29250.388
    },
    {
      "epoch": 2.9851771423264126,
      "grad_norm": 0.62890625,
      "learning_rate": 1.627628657553804e-05,
      "loss": 0.3877525806427002,
      "num_input_tokens_seen": 23994967104,
      "step": 84370,
      "train_runtime": 820332.5306,
      "train_tokens_per_second": 29250.293
    },
    {
      "epoch": 2.985530962590141,
      "grad_norm": 0.62890625,
      "learning_rate": 1.627542426942439e-05,
      "loss": 0.386457633972168,
      "num_input_tokens_seen": 23997814592,
      "step": 84380,
      "train_runtime": 820427.9195,
      "train_tokens_per_second": 29250.363
    },
    {
      "epoch": 2.98588478285387,
      "grad_norm": 0.609375,
      "learning_rate": 1.627456210034932e-05,
      "loss": 0.38124253749847414,
      "num_input_tokens_seen": 24000678208,
      "step": 84390,
      "train_runtime": 820526.0061,
      "train_tokens_per_second": 29250.357
    },
    {
      "epoch": 2.9862386031175987,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6273700068276534e-05,
      "loss": 0.38640170097351073,
      "num_input_tokens_seen": 24003519680,
      "step": 84400,
      "train_runtime": 820623.7711,
      "train_tokens_per_second": 29250.334
    },
    {
      "epoch": 2.9865924233813277,
      "grad_norm": 0.609375,
      "learning_rate": 1.6272838173169752e-05,
      "loss": 0.3883213043212891,
      "num_input_tokens_seen": 24006336896,
      "step": 84410,
      "train_runtime": 820714.6258,
      "train_tokens_per_second": 29250.529
    },
    {
      "epoch": 2.9869462436450562,
      "grad_norm": 0.59375,
      "learning_rate": 1.627197641499271e-05,
      "loss": 0.38307795524597166,
      "num_input_tokens_seen": 24009124224,
      "step": 84420,
      "train_runtime": 820809.05,
      "train_tokens_per_second": 29250.56
    },
    {
      "epoch": 2.987300063908785,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6271114793709146e-05,
      "loss": 0.3844077825546265,
      "num_input_tokens_seen": 24011981888,
      "step": 84430,
      "train_runtime": 820904.8797,
      "train_tokens_per_second": 29250.626
    },
    {
      "epoch": 2.9876538841725138,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6270253309282836e-05,
      "loss": 0.38652212619781495,
      "num_input_tokens_seen": 24014828672,
      "step": 84440,
      "train_runtime": 821002.4014,
      "train_tokens_per_second": 29250.619
    },
    {
      "epoch": 2.9880077044362428,
      "grad_norm": 0.62109375,
      "learning_rate": 1.626939196167754e-05,
      "loss": 0.3835011959075928,
      "num_input_tokens_seen": 24017713280,
      "step": 84450,
      "train_runtime": 821101.0595,
      "train_tokens_per_second": 29250.618
    },
    {
      "epoch": 2.9883615246999717,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6268530750857053e-05,
      "loss": 0.3885365962982178,
      "num_input_tokens_seen": 24020520128,
      "step": 84460,
      "train_runtime": 821196.3123,
      "train_tokens_per_second": 29250.643
    },
    {
      "epoch": 2.9887153449637003,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6267669676785175e-05,
      "loss": 0.38040938377380373,
      "num_input_tokens_seen": 24023329728,
      "step": 84470,
      "train_runtime": 821292.3313,
      "train_tokens_per_second": 29250.644
    },
    {
      "epoch": 2.989069165227429,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6266808739425716e-05,
      "loss": 0.3820831775665283,
      "num_input_tokens_seen": 24026236160,
      "step": 84480,
      "train_runtime": 821394.8832,
      "train_tokens_per_second": 29250.531
    },
    {
      "epoch": 2.989422985491158,
      "grad_norm": 0.64453125,
      "learning_rate": 1.6265947938742507e-05,
      "loss": 0.38855204582214353,
      "num_input_tokens_seen": 24029062336,
      "step": 84490,
      "train_runtime": 821493.0443,
      "train_tokens_per_second": 29250.476
    },
    {
      "epoch": 2.989776805754887,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6265087274699387e-05,
      "loss": 0.3898318767547607,
      "num_input_tokens_seen": 24031888384,
      "step": 84500,
      "train_runtime": 821593.2205,
      "train_tokens_per_second": 29250.349
    },
    {
      "epoch": 2.9901306260186153,
      "grad_norm": 0.625,
      "learning_rate": 1.626422674726021e-05,
      "loss": 0.3839479923248291,
      "num_input_tokens_seen": 24034728064,
      "step": 84510,
      "train_runtime": 821689.304,
      "train_tokens_per_second": 29250.384
    },
    {
      "epoch": 2.990484446282344,
      "grad_norm": 0.59375,
      "learning_rate": 1.6263366356388846e-05,
      "loss": 0.38756752014160156,
      "num_input_tokens_seen": 24037534720,
      "step": 84520,
      "train_runtime": 821785.21,
      "train_tokens_per_second": 29250.386
    },
    {
      "epoch": 2.990838266546073,
      "grad_norm": 0.609375,
      "learning_rate": 1.626250610204918e-05,
      "loss": 0.3872845649719238,
      "num_input_tokens_seen": 24040405760,
      "step": 84530,
      "train_runtime": 821885.0155,
      "train_tokens_per_second": 29250.327
    },
    {
      "epoch": 2.991192086809802,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6261645984205095e-05,
      "loss": 0.39045803546905516,
      "num_input_tokens_seen": 24043297024,
      "step": 84540,
      "train_runtime": 821987.6068,
      "train_tokens_per_second": 29250.194
    },
    {
      "epoch": 2.9915459070735304,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6260786002820508e-05,
      "loss": 0.3870557308197021,
      "num_input_tokens_seen": 24046108160,
      "step": 84550,
      "train_runtime": 822083.9736,
      "train_tokens_per_second": 29250.185
    },
    {
      "epoch": 2.9918997273372594,
      "grad_norm": 0.60546875,
      "learning_rate": 1.6259926157859336e-05,
      "loss": 0.3919745922088623,
      "num_input_tokens_seen": 24049044672,
      "step": 84560,
      "train_runtime": 822185.994,
      "train_tokens_per_second": 29250.127
    },
    {
      "epoch": 2.992253547600988,
      "grad_norm": 0.59375,
      "learning_rate": 1.625906644928552e-05,
      "loss": 0.38787250518798827,
      "num_input_tokens_seen": 24051938048,
      "step": 84570,
      "train_runtime": 822287.6618,
      "train_tokens_per_second": 29250.029
    },
    {
      "epoch": 2.992607367864717,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6258206877062995e-05,
      "loss": 0.38698816299438477,
      "num_input_tokens_seen": 24054786176,
      "step": 84580,
      "train_runtime": 822384.409,
      "train_tokens_per_second": 29250.051
    },
    {
      "epoch": 2.9929611881284455,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6257347441155738e-05,
      "loss": 0.38751916885375975,
      "num_input_tokens_seen": 24057639360,
      "step": 84590,
      "train_runtime": 822481.9447,
      "train_tokens_per_second": 29250.052
    },
    {
      "epoch": 2.9933150083921745,
      "grad_norm": 0.61328125,
      "learning_rate": 1.625648814152771e-05,
      "loss": 0.38276000022888185,
      "num_input_tokens_seen": 24060529600,
      "step": 84600,
      "train_runtime": 822582.7049,
      "train_tokens_per_second": 29249.982
    },
    {
      "epoch": 2.993668828655903,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6255628978142908e-05,
      "loss": 0.3855485439300537,
      "num_input_tokens_seen": 24063329024,
      "step": 84610,
      "train_runtime": 822675.0288,
      "train_tokens_per_second": 29250.103
    },
    {
      "epoch": 2.994022648919632,
      "grad_norm": 0.6328125,
      "learning_rate": 1.6254769950965327e-05,
      "loss": 0.38649394512176516,
      "num_input_tokens_seen": 24066189824,
      "step": 84620,
      "train_runtime": 822772.0373,
      "train_tokens_per_second": 29250.131
    },
    {
      "epoch": 2.9943764691833605,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6253911059958986e-05,
      "loss": 0.3890997886657715,
      "num_input_tokens_seen": 24069068992,
      "step": 84630,
      "train_runtime": 822870.7955,
      "train_tokens_per_second": 29250.119
    },
    {
      "epoch": 2.9947302894470895,
      "grad_norm": 0.62109375,
      "learning_rate": 1.625305230508791e-05,
      "loss": 0.38601021766662597,
      "num_input_tokens_seen": 24071887808,
      "step": 84640,
      "train_runtime": 822964.6293,
      "train_tokens_per_second": 29250.209
    },
    {
      "epoch": 2.9950841097108185,
      "grad_norm": 0.63671875,
      "learning_rate": 1.625219368631614e-05,
      "loss": 0.3883702754974365,
      "num_input_tokens_seen": 24074699776,
      "step": 84650,
      "train_runtime": 823058.8071,
      "train_tokens_per_second": 29250.279
    },
    {
      "epoch": 2.995437929974547,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6251335203607734e-05,
      "loss": 0.38436412811279297,
      "num_input_tokens_seen": 24077558400,
      "step": 84660,
      "train_runtime": 823157.9823,
      "train_tokens_per_second": 29250.228
    },
    {
      "epoch": 2.9957917502382756,
      "grad_norm": 0.61328125,
      "learning_rate": 1.6250476856926757e-05,
      "loss": 0.3865625858306885,
      "num_input_tokens_seen": 24080440832,
      "step": 84670,
      "train_runtime": 823257.6809,
      "train_tokens_per_second": 29250.187
    },
    {
      "epoch": 2.9961455705020046,
      "grad_norm": 0.62109375,
      "learning_rate": 1.6249618646237292e-05,
      "loss": 0.3846981287002563,
      "num_input_tokens_seen": 24083276032,
      "step": 84680,
      "train_runtime": 823356.4856,
      "train_tokens_per_second": 29250.12
    },
    {
      "epoch": 2.9964993907657336,
      "grad_norm": 0.61328125,
      "learning_rate": 1.624876057150343e-05,
      "loss": 0.3845958232879639,
      "num_input_tokens_seen": 24086081344,
      "step": 84690,
      "train_runtime": 823451.4212,
      "train_tokens_per_second": 29250.155
    },
    {
      "epoch": 2.996853211029462,
      "grad_norm": 0.59765625,
      "learning_rate": 1.624790263268928e-05,
      "loss": 0.3860167503356934,
      "num_input_tokens_seen": 24088907200,
      "step": 84700,
      "train_runtime": 823545.9534,
      "train_tokens_per_second": 29250.228
    },
    {
      "epoch": 2.997207031293191,
      "grad_norm": 0.6171875,
      "learning_rate": 1.6247044829758963e-05,
      "loss": 0.3877938985824585,
      "num_input_tokens_seen": 24091731584,
      "step": 84710,
      "train_runtime": 823641.9437,
      "train_tokens_per_second": 29250.249
    },
    {
      "epoch": 2.9975608515569196,
      "grad_norm": 0.59765625,
      "learning_rate": 1.6246187162676616e-05,
      "loss": 0.3880586624145508,
      "num_input_tokens_seen": 24094547328,
      "step": 84720,
      "train_runtime": 823734.922,
      "train_tokens_per_second": 29250.365
    },
    {
      "epoch": 2.9979146718206486,
      "grad_norm": 0.59375,
      "learning_rate": 1.624532963140638e-05,
      "loss": 0.38626956939697266,
      "num_input_tokens_seen": 24097437120,
      "step": 84730,
      "train_runtime": 823837.1031,
      "train_tokens_per_second": 29250.245
    },
    {
      "epoch": 2.998268492084377,
      "grad_norm": 0.5859375,
      "learning_rate": 1.6244472235912418e-05,
      "loss": 0.38243086338043214,
      "num_input_tokens_seen": 24100316032,
      "step": 84740,
      "train_runtime": 823936.3314,
      "train_tokens_per_second": 29250.216
    },
    {
      "epoch": 2.998622312348106,
      "grad_norm": 0.58203125,
      "learning_rate": 1.624361497615891e-05,
      "loss": 0.3870294809341431,
      "num_input_tokens_seen": 24103207808,
      "step": 84750,
      "train_runtime": 824037.6102,
      "train_tokens_per_second": 29250.131
    },
    {
      "epoch": 2.9989761326118347,
      "grad_norm": 0.62890625,
      "learning_rate": 1.624275785211003e-05,
      "loss": 0.39350252151489257,
      "num_input_tokens_seen": 24106077248,
      "step": 84760,
      "train_runtime": 824137.1856,
      "train_tokens_per_second": 29250.078
    },
    {
      "epoch": 2.9993299528755637,
      "grad_norm": 0.6171875,
      "learning_rate": 1.624190086372999e-05,
      "loss": 0.3870228290557861,
      "num_input_tokens_seen": 24108912320,
      "step": 84770,
      "train_runtime": 824233.0925,
      "train_tokens_per_second": 29250.114
    },
    {
      "epoch": 2.9996837731392922,
      "grad_norm": 0.62890625,
      "learning_rate": 1.6241044010982998e-05,
      "loss": 0.3872014045715332,
      "num_input_tokens_seen": 24111758528,
      "step": 84780,
      "train_runtime": 824328.4964,
      "train_tokens_per_second": 29250.182
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.3361869752407074,
      "eval_runtime": 8.4623,
      "eval_samples_per_second": 471.268,
      "eval_steps_per_second": 3.781,
      "num_input_tokens_seen": 24114289216,
      "step": 84789
    },
    {
      "epoch": 3.0,
      "num_input_tokens_seen": 24114289216,
      "step": 84789,
      "total_flos": 4.014701818973667e+20,
      "train_loss": 0.4004683384819235,
      "train_runtime": 824447.1535,
      "train_samples_per_second": 105.311,
      "train_steps_per_second": 0.103
    }
  ],
  "logging_steps": 10,
  "max_steps": 84789,
  "num_input_tokens_seen": 24114289216,
  "num_train_epochs": 3,
  "save_steps": 5000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.014701818973667e+20,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}