{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.1839080459770115,
  "eval_steps": 200,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 9.195402298850575e-05,
      "grad_norm": 29.12846565246582,
      "learning_rate": 0.0,
      "loss": 1.0049550533294678,
      "num_input_tokens_seen": 592,
      "step": 1,
      "train_runtime": 7.3694,
      "train_tokens_per_second": 80.332
    },
    {
      "epoch": 0.0001839080459770115,
      "grad_norm": 28.490211486816406,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.0715882778167725,
      "num_input_tokens_seen": 1180,
      "step": 2,
      "train_runtime": 9.2983,
      "train_tokens_per_second": 126.904
    },
    {
      "epoch": 0.00027586206896551725,
      "grad_norm": 25.38025665283203,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.9221532344818115,
      "num_input_tokens_seen": 1864,
      "step": 3,
      "train_runtime": 11.2178,
      "train_tokens_per_second": 166.164
    },
    {
      "epoch": 0.000367816091954023,
      "grad_norm": 33.270748138427734,
      "learning_rate": 1.2e-05,
      "loss": 0.7833482623100281,
      "num_input_tokens_seen": 2712,
      "step": 4,
      "train_runtime": 13.1157,
      "train_tokens_per_second": 206.775
    },
    {
      "epoch": 0.0004597701149425287,
      "grad_norm": 20.2131290435791,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.7940477132797241,
      "num_input_tokens_seen": 3504,
      "step": 5,
      "train_runtime": 15.0072,
      "train_tokens_per_second": 233.488
    },
    {
      "epoch": 0.0005517241379310345,
      "grad_norm": 17.375965118408203,
      "learning_rate": 2e-05,
      "loss": 0.6811378598213196,
      "num_input_tokens_seen": 4204,
      "step": 6,
      "train_runtime": 16.9237,
      "train_tokens_per_second": 248.41
    },
    {
      "epoch": 0.0006436781609195402,
      "grad_norm": 19.164548873901367,
      "learning_rate": 1.998997493734336e-05,
      "loss": 0.5850780010223389,
      "num_input_tokens_seen": 4872,
      "step": 7,
      "train_runtime": 18.8068,
      "train_tokens_per_second": 259.055
    },
    {
      "epoch": 0.000735632183908046,
      "grad_norm": 9.285567283630371,
      "learning_rate": 1.997994987468672e-05,
      "loss": 0.6868342161178589,
      "num_input_tokens_seen": 6356,
      "step": 8,
      "train_runtime": 20.7877,
      "train_tokens_per_second": 305.758
    },
    {
      "epoch": 0.0008275862068965517,
      "grad_norm": 11.42162036895752,
      "learning_rate": 1.9969924812030075e-05,
      "loss": 0.6028171181678772,
      "num_input_tokens_seen": 7144,
      "step": 9,
      "train_runtime": 22.6773,
      "train_tokens_per_second": 315.028
    },
    {
      "epoch": 0.0009195402298850574,
      "grad_norm": 10.177748680114746,
      "learning_rate": 1.9959899749373436e-05,
      "loss": 0.6105828881263733,
      "num_input_tokens_seen": 7820,
      "step": 10,
      "train_runtime": 24.5836,
      "train_tokens_per_second": 318.098
    },
    {
      "epoch": 0.0010114942528735632,
      "grad_norm": 9.14651107788086,
      "learning_rate": 1.9949874686716792e-05,
      "loss": 0.6887412071228027,
      "num_input_tokens_seen": 8676,
      "step": 11,
      "train_runtime": 26.5012,
      "train_tokens_per_second": 327.382
    },
    {
      "epoch": 0.001103448275862069,
      "grad_norm": 9.492165565490723,
      "learning_rate": 1.9939849624060153e-05,
      "loss": 0.7147798538208008,
      "num_input_tokens_seen": 9560,
      "step": 12,
      "train_runtime": 28.4657,
      "train_tokens_per_second": 335.843
    },
    {
      "epoch": 0.0011954022988505746,
      "grad_norm": 8.382616996765137,
      "learning_rate": 1.992982456140351e-05,
      "loss": 0.5508676171302795,
      "num_input_tokens_seen": 10224,
      "step": 13,
      "train_runtime": 30.3593,
      "train_tokens_per_second": 336.766
    },
    {
      "epoch": 0.0012873563218390805,
      "grad_norm": 8.802109718322754,
      "learning_rate": 1.991979949874687e-05,
      "loss": 0.5035771131515503,
      "num_input_tokens_seen": 10784,
      "step": 14,
      "train_runtime": 32.2457,
      "train_tokens_per_second": 334.433
    },
    {
      "epoch": 0.001379310344827586,
      "grad_norm": 8.765376091003418,
      "learning_rate": 1.9909774436090226e-05,
      "loss": 0.5817534327507019,
      "num_input_tokens_seen": 11580,
      "step": 15,
      "train_runtime": 34.1678,
      "train_tokens_per_second": 338.915
    },
    {
      "epoch": 0.001471264367816092,
      "grad_norm": 8.994365692138672,
      "learning_rate": 1.9899749373433587e-05,
      "loss": 0.6585532426834106,
      "num_input_tokens_seen": 12340,
      "step": 16,
      "train_runtime": 36.0534,
      "train_tokens_per_second": 342.27
    },
    {
      "epoch": 0.0015632183908045978,
      "grad_norm": 8.071599006652832,
      "learning_rate": 1.9889724310776943e-05,
      "loss": 0.5290766954421997,
      "num_input_tokens_seen": 12940,
      "step": 17,
      "train_runtime": 37.9406,
      "train_tokens_per_second": 341.06
    },
    {
      "epoch": 0.0016551724137931034,
      "grad_norm": 9.644457817077637,
      "learning_rate": 1.9879699248120303e-05,
      "loss": 0.7164220809936523,
      "num_input_tokens_seen": 14640,
      "step": 18,
      "train_runtime": 40.0076,
      "train_tokens_per_second": 365.93
    },
    {
      "epoch": 0.0017471264367816092,
      "grad_norm": 8.754335403442383,
      "learning_rate": 1.986967418546366e-05,
      "loss": 0.6070724725723267,
      "num_input_tokens_seen": 15492,
      "step": 19,
      "train_runtime": 41.9315,
      "train_tokens_per_second": 369.459
    },
    {
      "epoch": 0.0018390804597701149,
      "grad_norm": 8.747954368591309,
      "learning_rate": 1.9859649122807017e-05,
      "loss": 0.5036498308181763,
      "num_input_tokens_seen": 16340,
      "step": 20,
      "train_runtime": 43.8604,
      "train_tokens_per_second": 372.546
    },
    {
      "epoch": 0.0019310344827586207,
      "grad_norm": 8.61439323425293,
      "learning_rate": 1.9849624060150377e-05,
      "loss": 0.47874048352241516,
      "num_input_tokens_seen": 16896,
      "step": 21,
      "train_runtime": 45.741,
      "train_tokens_per_second": 369.384
    },
    {
      "epoch": 0.0020229885057471263,
      "grad_norm": 8.474122047424316,
      "learning_rate": 1.9839598997493737e-05,
      "loss": 0.5896688103675842,
      "num_input_tokens_seen": 17892,
      "step": 22,
      "train_runtime": 47.6675,
      "train_tokens_per_second": 375.35
    },
    {
      "epoch": 0.0021149425287356324,
      "grad_norm": 9.135046005249023,
      "learning_rate": 1.9829573934837094e-05,
      "loss": 0.4764246344566345,
      "num_input_tokens_seen": 18576,
      "step": 23,
      "train_runtime": 49.5549,
      "train_tokens_per_second": 374.857
    },
    {
      "epoch": 0.002206896551724138,
      "grad_norm": 8.152758598327637,
      "learning_rate": 1.981954887218045e-05,
      "loss": 0.5896083116531372,
      "num_input_tokens_seen": 19380,
      "step": 24,
      "train_runtime": 51.4758,
      "train_tokens_per_second": 376.488
    },
    {
      "epoch": 0.0022988505747126436,
      "grad_norm": 7.929178714752197,
      "learning_rate": 1.980952380952381e-05,
      "loss": 0.7542365789413452,
      "num_input_tokens_seen": 20720,
      "step": 25,
      "train_runtime": 53.4594,
      "train_tokens_per_second": 387.584
    },
    {
      "epoch": 0.0023908045977011493,
      "grad_norm": 9.76752758026123,
      "learning_rate": 1.9799498746867168e-05,
      "loss": 0.6881550550460815,
      "num_input_tokens_seen": 21436,
      "step": 26,
      "train_runtime": 55.3448,
      "train_tokens_per_second": 387.318
    },
    {
      "epoch": 0.0024827586206896553,
      "grad_norm": 8.284303665161133,
      "learning_rate": 1.9789473684210528e-05,
      "loss": 0.5384884476661682,
      "num_input_tokens_seen": 21988,
      "step": 27,
      "train_runtime": 57.228,
      "train_tokens_per_second": 384.218
    },
    {
      "epoch": 0.002574712643678161,
      "grad_norm": 7.906507968902588,
      "learning_rate": 1.977944862155389e-05,
      "loss": 0.635601282119751,
      "num_input_tokens_seen": 22812,
      "step": 28,
      "train_runtime": 59.15,
      "train_tokens_per_second": 385.663
    },
    {
      "epoch": 0.0026666666666666666,
      "grad_norm": 8.403658866882324,
      "learning_rate": 1.9769423558897245e-05,
      "loss": 0.5284605026245117,
      "num_input_tokens_seen": 23504,
      "step": 29,
      "train_runtime": 61.0362,
      "train_tokens_per_second": 385.083
    },
    {
      "epoch": 0.002758620689655172,
      "grad_norm": 8.070573806762695,
      "learning_rate": 1.9759398496240602e-05,
      "loss": 0.6003434658050537,
      "num_input_tokens_seen": 24304,
      "step": 30,
      "train_runtime": 62.9134,
      "train_tokens_per_second": 386.309
    },
    {
      "epoch": 0.0028505747126436783,
      "grad_norm": 7.809580326080322,
      "learning_rate": 1.974937343358396e-05,
      "loss": 0.5435101985931396,
      "num_input_tokens_seen": 24980,
      "step": 31,
      "train_runtime": 71.4956,
      "train_tokens_per_second": 349.392
    },
    {
      "epoch": 0.002942528735632184,
      "grad_norm": 9.469648361206055,
      "learning_rate": 1.973934837092732e-05,
      "loss": 0.6884284019470215,
      "num_input_tokens_seen": 25968,
      "step": 32,
      "train_runtime": 73.4456,
      "train_tokens_per_second": 353.568
    },
    {
      "epoch": 0.0030344827586206895,
      "grad_norm": 7.747436046600342,
      "learning_rate": 1.972932330827068e-05,
      "loss": 0.4698449969291687,
      "num_input_tokens_seen": 26592,
      "step": 33,
      "train_runtime": 75.3303,
      "train_tokens_per_second": 353.005
    },
    {
      "epoch": 0.0031264367816091956,
      "grad_norm": 8.132604598999023,
      "learning_rate": 1.9719298245614036e-05,
      "loss": 0.5784815549850464,
      "num_input_tokens_seen": 27296,
      "step": 34,
      "train_runtime": 77.2232,
      "train_tokens_per_second": 353.469
    },
    {
      "epoch": 0.003218390804597701,
      "grad_norm": 7.501383304595947,
      "learning_rate": 1.9709273182957396e-05,
      "loss": 0.7156097292900085,
      "num_input_tokens_seen": 28108,
      "step": 35,
      "train_runtime": 79.1498,
      "train_tokens_per_second": 355.124
    },
    {
      "epoch": 0.003310344827586207,
      "grad_norm": 8.279403686523438,
      "learning_rate": 1.9699248120300753e-05,
      "loss": 0.5377761125564575,
      "num_input_tokens_seen": 28740,
      "step": 36,
      "train_runtime": 81.0569,
      "train_tokens_per_second": 354.566
    },
    {
      "epoch": 0.0034022988505747124,
      "grad_norm": 7.578561305999756,
      "learning_rate": 1.968922305764411e-05,
      "loss": 0.5211909413337708,
      "num_input_tokens_seen": 29340,
      "step": 37,
      "train_runtime": 82.9417,
      "train_tokens_per_second": 353.742
    },
    {
      "epoch": 0.0034942528735632185,
      "grad_norm": 7.668713092803955,
      "learning_rate": 1.967919799498747e-05,
      "loss": 0.6027074456214905,
      "num_input_tokens_seen": 29956,
      "step": 38,
      "train_runtime": 84.8345,
      "train_tokens_per_second": 353.111
    },
    {
      "epoch": 0.003586206896551724,
      "grad_norm": 7.965372085571289,
      "learning_rate": 1.966917293233083e-05,
      "loss": 0.5476580262184143,
      "num_input_tokens_seen": 30508,
      "step": 39,
      "train_runtime": 86.723,
      "train_tokens_per_second": 351.787
    },
    {
      "epoch": 0.0036781609195402297,
      "grad_norm": 8.622340202331543,
      "learning_rate": 1.9659147869674187e-05,
      "loss": 0.599757730960846,
      "num_input_tokens_seen": 31160,
      "step": 40,
      "train_runtime": 88.6236,
      "train_tokens_per_second": 351.599
    },
    {
      "epoch": 0.003770114942528736,
      "grad_norm": 8.414080619812012,
      "learning_rate": 1.9649122807017544e-05,
      "loss": 0.5053638815879822,
      "num_input_tokens_seen": 31792,
      "step": 41,
      "train_runtime": 90.5165,
      "train_tokens_per_second": 351.229
    },
    {
      "epoch": 0.0038620689655172414,
      "grad_norm": 8.579339027404785,
      "learning_rate": 1.9639097744360904e-05,
      "loss": 0.6343808174133301,
      "num_input_tokens_seen": 32860,
      "step": 42,
      "train_runtime": 92.4696,
      "train_tokens_per_second": 355.36
    },
    {
      "epoch": 0.0039540229885057475,
      "grad_norm": 7.376124382019043,
      "learning_rate": 1.962907268170426e-05,
      "loss": 0.48202601075172424,
      "num_input_tokens_seen": 33820,
      "step": 43,
      "train_runtime": 94.4085,
      "train_tokens_per_second": 358.23
    },
    {
      "epoch": 0.004045977011494253,
      "grad_norm": 8.69480037689209,
      "learning_rate": 1.961904761904762e-05,
      "loss": 0.586205244064331,
      "num_input_tokens_seen": 34464,
      "step": 44,
      "train_runtime": 96.314,
      "train_tokens_per_second": 357.83
    },
    {
      "epoch": 0.004137931034482759,
      "grad_norm": 7.8471527099609375,
      "learning_rate": 1.960902255639098e-05,
      "loss": 0.6266012191772461,
      "num_input_tokens_seen": 35312,
      "step": 45,
      "train_runtime": 98.3444,
      "train_tokens_per_second": 359.065
    },
    {
      "epoch": 0.004229885057471265,
      "grad_norm": 9.146677017211914,
      "learning_rate": 1.9598997493734338e-05,
      "loss": 0.6910461187362671,
      "num_input_tokens_seen": 36540,
      "step": 46,
      "train_runtime": 100.3279,
      "train_tokens_per_second": 364.206
    },
    {
      "epoch": 0.00432183908045977,
      "grad_norm": 7.88541841506958,
      "learning_rate": 1.9588972431077695e-05,
      "loss": 0.6365747451782227,
      "num_input_tokens_seen": 37644,
      "step": 47,
      "train_runtime": 102.3224,
      "train_tokens_per_second": 367.896
    },
    {
      "epoch": 0.004413793103448276,
      "grad_norm": 8.206524848937988,
      "learning_rate": 1.9578947368421055e-05,
      "loss": 0.5315179228782654,
      "num_input_tokens_seen": 38412,
      "step": 48,
      "train_runtime": 104.2731,
      "train_tokens_per_second": 368.379
    },
    {
      "epoch": 0.004505747126436781,
      "grad_norm": 8.704801559448242,
      "learning_rate": 1.956892230576441e-05,
      "loss": 0.5973373055458069,
      "num_input_tokens_seen": 38976,
      "step": 49,
      "train_runtime": 106.1861,
      "train_tokens_per_second": 367.054
    },
    {
      "epoch": 0.004597701149425287,
      "grad_norm": 8.045174598693848,
      "learning_rate": 1.9558897243107772e-05,
      "loss": 0.6078757047653198,
      "num_input_tokens_seen": 39648,
      "step": 50,
      "train_runtime": 108.0766,
      "train_tokens_per_second": 366.851
    },
    {
      "epoch": 0.004689655172413793,
      "grad_norm": 9.606939315795898,
      "learning_rate": 1.954887218045113e-05,
      "loss": 0.5741446018218994,
      "num_input_tokens_seen": 40364,
      "step": 51,
      "train_runtime": 109.9935,
      "train_tokens_per_second": 366.967
    },
    {
      "epoch": 0.0047816091954022985,
      "grad_norm": 7.9645676612854,
      "learning_rate": 1.953884711779449e-05,
      "loss": 0.5054519176483154,
      "num_input_tokens_seen": 40972,
      "step": 52,
      "train_runtime": 111.8772,
      "train_tokens_per_second": 366.223
    },
    {
      "epoch": 0.004873563218390805,
      "grad_norm": 9.297037124633789,
      "learning_rate": 1.9528822055137845e-05,
      "loss": 0.5256612300872803,
      "num_input_tokens_seen": 42188,
      "step": 53,
      "train_runtime": 113.8307,
      "train_tokens_per_second": 370.621
    },
    {
      "epoch": 0.004965517241379311,
      "grad_norm": 8.875506401062012,
      "learning_rate": 1.9518796992481202e-05,
      "loss": 0.5609273314476013,
      "num_input_tokens_seen": 42772,
      "step": 54,
      "train_runtime": 115.7116,
      "train_tokens_per_second": 369.643
    },
    {
      "epoch": 0.005057471264367816,
      "grad_norm": 7.938648223876953,
      "learning_rate": 1.9508771929824562e-05,
      "loss": 0.6343899369239807,
      "num_input_tokens_seen": 43780,
      "step": 55,
      "train_runtime": 117.6838,
      "train_tokens_per_second": 372.014
    },
    {
      "epoch": 0.005149425287356322,
      "grad_norm": 7.548194408416748,
      "learning_rate": 1.9498746867167923e-05,
      "loss": 0.6220943331718445,
      "num_input_tokens_seen": 45076,
      "step": 56,
      "train_runtime": 119.6727,
      "train_tokens_per_second": 376.661
    },
    {
      "epoch": 0.005241379310344828,
      "grad_norm": 8.334528923034668,
      "learning_rate": 1.948872180451128e-05,
      "loss": 0.6186922788619995,
      "num_input_tokens_seen": 45728,
      "step": 57,
      "train_runtime": 121.5646,
      "train_tokens_per_second": 376.162
    },
    {
      "epoch": 0.005333333333333333,
      "grad_norm": 8.271797180175781,
      "learning_rate": 1.947869674185464e-05,
      "loss": 0.5532838702201843,
      "num_input_tokens_seen": 46460,
      "step": 58,
      "train_runtime": 123.4469,
      "train_tokens_per_second": 376.356
    },
    {
      "epoch": 0.005425287356321839,
      "grad_norm": 7.5711846351623535,
      "learning_rate": 1.9468671679197996e-05,
      "loss": 0.4910699129104614,
      "num_input_tokens_seen": 47132,
      "step": 59,
      "train_runtime": 125.3402,
      "train_tokens_per_second": 376.032
    },
    {
      "epoch": 0.005517241379310344,
      "grad_norm": 8.477574348449707,
      "learning_rate": 1.9458646616541353e-05,
      "loss": 0.5690719485282898,
      "num_input_tokens_seen": 47780,
      "step": 60,
      "train_runtime": 127.216,
      "train_tokens_per_second": 375.582
    },
    {
      "epoch": 0.0056091954022988505,
      "grad_norm": 7.364357948303223,
      "learning_rate": 1.9448621553884713e-05,
      "loss": 0.605834424495697,
      "num_input_tokens_seen": 48944,
      "step": 61,
      "train_runtime": 135.7442,
      "train_tokens_per_second": 360.561
    },
    {
      "epoch": 0.0057011494252873565,
      "grad_norm": 8.096819877624512,
      "learning_rate": 1.9438596491228074e-05,
      "loss": 0.49838387966156006,
      "num_input_tokens_seen": 49484,
      "step": 62,
      "train_runtime": 137.6251,
      "train_tokens_per_second": 359.556
    },
    {
      "epoch": 0.005793103448275862,
      "grad_norm": 8.219697952270508,
      "learning_rate": 1.942857142857143e-05,
      "loss": 0.5258055925369263,
      "num_input_tokens_seen": 50200,
      "step": 63,
      "train_runtime": 139.5812,
      "train_tokens_per_second": 359.647
    },
    {
      "epoch": 0.005885057471264368,
      "grad_norm": 8.2337007522583,
      "learning_rate": 1.9418546365914787e-05,
      "loss": 0.54833984375,
      "num_input_tokens_seen": 50980,
      "step": 64,
      "train_runtime": 141.5768,
      "train_tokens_per_second": 360.087
    },
    {
      "epoch": 0.005977011494252874,
      "grad_norm": 7.676711082458496,
      "learning_rate": 1.9408521303258147e-05,
      "loss": 0.43378540873527527,
      "num_input_tokens_seen": 51580,
      "step": 65,
      "train_runtime": 143.4945,
      "train_tokens_per_second": 359.456
    },
    {
      "epoch": 0.006068965517241379,
      "grad_norm": 8.562492370605469,
      "learning_rate": 1.9398496240601504e-05,
      "loss": 0.49912506341934204,
      "num_input_tokens_seen": 52200,
      "step": 66,
      "train_runtime": 145.421,
      "train_tokens_per_second": 358.958
    },
    {
      "epoch": 0.006160919540229885,
      "grad_norm": 9.452392578125,
      "learning_rate": 1.9388471177944864e-05,
      "loss": 0.6256085634231567,
      "num_input_tokens_seen": 53200,
      "step": 67,
      "train_runtime": 147.4133,
      "train_tokens_per_second": 360.89
    },
    {
      "epoch": 0.006252873563218391,
      "grad_norm": 7.979230880737305,
      "learning_rate": 1.9378446115288224e-05,
      "loss": 0.4097839593887329,
      "num_input_tokens_seen": 53780,
      "step": 68,
      "train_runtime": 149.2951,
      "train_tokens_per_second": 360.226
    },
    {
      "epoch": 0.006344827586206896,
      "grad_norm": 7.691853046417236,
      "learning_rate": 1.936842105263158e-05,
      "loss": 0.41319534182548523,
      "num_input_tokens_seen": 54340,
      "step": 69,
      "train_runtime": 151.2002,
      "train_tokens_per_second": 359.391
    },
    {
      "epoch": 0.006436781609195402,
      "grad_norm": 7.792855739593506,
      "learning_rate": 1.9358395989974938e-05,
      "loss": 0.5904505848884583,
      "num_input_tokens_seen": 55136,
      "step": 70,
      "train_runtime": 153.1216,
      "train_tokens_per_second": 360.08
    },
    {
      "epoch": 0.0065287356321839084,
      "grad_norm": 7.832664966583252,
      "learning_rate": 1.9348370927318295e-05,
      "loss": 0.6427686214447021,
      "num_input_tokens_seen": 55992,
      "step": 71,
      "train_runtime": 155.0516,
      "train_tokens_per_second": 361.118
    },
    {
      "epoch": 0.006620689655172414,
      "grad_norm": 9.28006649017334,
      "learning_rate": 1.9338345864661655e-05,
      "loss": 0.49627256393432617,
      "num_input_tokens_seen": 56600,
      "step": 72,
      "train_runtime": 156.9191,
      "train_tokens_per_second": 360.695
    },
    {
      "epoch": 0.00671264367816092,
      "grad_norm": 9.48000717163086,
      "learning_rate": 1.9328320802005015e-05,
      "loss": 0.5595530867576599,
      "num_input_tokens_seen": 57360,
      "step": 73,
      "train_runtime": 158.8057,
      "train_tokens_per_second": 361.196
    },
    {
      "epoch": 0.006804597701149425,
      "grad_norm": 7.642348289489746,
      "learning_rate": 1.9318295739348372e-05,
      "loss": 0.5516754388809204,
      "num_input_tokens_seen": 59168,
      "step": 74,
      "train_runtime": 160.8679,
      "train_tokens_per_second": 367.805
    },
    {
      "epoch": 0.006896551724137931,
      "grad_norm": 8.76455307006836,
      "learning_rate": 1.9308270676691732e-05,
      "loss": 0.6654430031776428,
      "num_input_tokens_seen": 59984,
      "step": 75,
      "train_runtime": 162.7537,
      "train_tokens_per_second": 368.557
    },
    {
      "epoch": 0.006988505747126437,
      "grad_norm": 8.426393508911133,
      "learning_rate": 1.929824561403509e-05,
      "loss": 0.5090984106063843,
      "num_input_tokens_seen": 60672,
      "step": 76,
      "train_runtime": 164.6432,
      "train_tokens_per_second": 368.506
    },
    {
      "epoch": 0.007080459770114942,
      "grad_norm": 7.846069812774658,
      "learning_rate": 1.9288220551378446e-05,
      "loss": 0.6218851804733276,
      "num_input_tokens_seen": 61332,
      "step": 77,
      "train_runtime": 166.5232,
      "train_tokens_per_second": 368.309
    },
    {
      "epoch": 0.007172413793103448,
      "grad_norm": 8.188020706176758,
      "learning_rate": 1.9278195488721806e-05,
      "loss": 0.5355145335197449,
      "num_input_tokens_seen": 62092,
      "step": 78,
      "train_runtime": 168.4406,
      "train_tokens_per_second": 368.629
    },
    {
      "epoch": 0.007264367816091954,
      "grad_norm": 7.9347429275512695,
      "learning_rate": 1.9268170426065166e-05,
      "loss": 0.6026157736778259,
      "num_input_tokens_seen": 63364,
      "step": 79,
      "train_runtime": 170.386,
      "train_tokens_per_second": 371.885
    },
    {
      "epoch": 0.0073563218390804595,
      "grad_norm": 8.449371337890625,
      "learning_rate": 1.9258145363408523e-05,
      "loss": 0.5600061416625977,
      "num_input_tokens_seen": 64152,
      "step": 80,
      "train_runtime": 172.2723,
      "train_tokens_per_second": 372.387
    },
    {
      "epoch": 0.0074482758620689656,
      "grad_norm": 8.201337814331055,
      "learning_rate": 1.924812030075188e-05,
      "loss": 0.4756973087787628,
      "num_input_tokens_seen": 64848,
      "step": 81,
      "train_runtime": 174.1521,
      "train_tokens_per_second": 372.364
    },
    {
      "epoch": 0.007540229885057472,
      "grad_norm": 7.353557586669922,
      "learning_rate": 1.923809523809524e-05,
      "loss": 0.6228072643280029,
      "num_input_tokens_seen": 65848,
      "step": 82,
      "train_runtime": 176.0762,
      "train_tokens_per_second": 373.974
    },
    {
      "epoch": 0.007632183908045977,
      "grad_norm": 7.903290748596191,
      "learning_rate": 1.9228070175438597e-05,
      "loss": 0.5884923934936523,
      "num_input_tokens_seen": 66920,
      "step": 83,
      "train_runtime": 178.0269,
      "train_tokens_per_second": 375.898
    },
    {
      "epoch": 0.007724137931034483,
      "grad_norm": 6.583453178405762,
      "learning_rate": 1.9218045112781957e-05,
      "loss": 0.5183554887771606,
      "num_input_tokens_seen": 68648,
      "step": 84,
      "train_runtime": 180.1074,
      "train_tokens_per_second": 381.15
    },
    {
      "epoch": 0.007816091954022988,
      "grad_norm": 7.549685955047607,
      "learning_rate": 1.9208020050125317e-05,
      "loss": 0.5646144151687622,
      "num_input_tokens_seen": 69404,
      "step": 85,
      "train_runtime": 182.0299,
      "train_tokens_per_second": 381.278
    },
    {
      "epoch": 0.007908045977011495,
      "grad_norm": 7.347597122192383,
      "learning_rate": 1.9197994987468674e-05,
      "loss": 0.5947625637054443,
      "num_input_tokens_seen": 70088,
      "step": 86,
      "train_runtime": 183.9222,
      "train_tokens_per_second": 381.074
    },
    {
      "epoch": 0.008,
      "grad_norm": 7.432178020477295,
      "learning_rate": 1.918796992481203e-05,
      "loss": 0.4872491955757141,
      "num_input_tokens_seen": 70704,
      "step": 87,
      "train_runtime": 185.8038,
      "train_tokens_per_second": 380.53
    },
    {
      "epoch": 0.008091954022988505,
      "grad_norm": 6.862610340118408,
      "learning_rate": 1.917794486215539e-05,
      "loss": 0.4244173765182495,
      "num_input_tokens_seen": 71380,
      "step": 88,
      "train_runtime": 187.691,
      "train_tokens_per_second": 380.306
    },
    {
      "epoch": 0.008183908045977012,
      "grad_norm": 8.37441635131836,
      "learning_rate": 1.9167919799498748e-05,
      "loss": 0.6755821704864502,
      "num_input_tokens_seen": 73004,
      "step": 89,
      "train_runtime": 189.712,
      "train_tokens_per_second": 384.815
    },
    {
      "epoch": 0.008275862068965517,
      "grad_norm": 7.625087261199951,
      "learning_rate": 1.9157894736842108e-05,
      "loss": 0.5150968432426453,
      "num_input_tokens_seen": 73688,
      "step": 90,
      "train_runtime": 191.614,
      "train_tokens_per_second": 384.565
    },
    {
      "epoch": 0.008367816091954023,
      "grad_norm": 7.352773189544678,
      "learning_rate": 1.9147869674185465e-05,
      "loss": 0.5014460682868958,
      "num_input_tokens_seen": 74528,
      "step": 91,
      "train_runtime": 199.7209,
      "train_tokens_per_second": 373.161
    },
    {
      "epoch": 0.00845977011494253,
      "grad_norm": 7.710674285888672,
      "learning_rate": 1.9137844611528825e-05,
      "loss": 0.5669783353805542,
      "num_input_tokens_seen": 75320,
      "step": 92,
      "train_runtime": 201.7281,
      "train_tokens_per_second": 373.374
    },
    {
      "epoch": 0.008551724137931035,
      "grad_norm": 8.01150131225586,
      "learning_rate": 1.912781954887218e-05,
      "loss": 0.515295147895813,
      "num_input_tokens_seen": 76000,
      "step": 93,
      "train_runtime": 203.6283,
      "train_tokens_per_second": 373.229
    },
    {
      "epoch": 0.00864367816091954,
      "grad_norm": 6.6490678787231445,
      "learning_rate": 1.911779448621554e-05,
      "loss": 0.3702370524406433,
      "num_input_tokens_seen": 76568,
      "step": 94,
      "train_runtime": 205.5128,
      "train_tokens_per_second": 372.57
    },
    {
      "epoch": 0.008735632183908045,
      "grad_norm": 7.624765872955322,
      "learning_rate": 1.91077694235589e-05,
      "loss": 0.5499355792999268,
      "num_input_tokens_seen": 77332,
      "step": 95,
      "train_runtime": 207.4163,
      "train_tokens_per_second": 372.835
    },
    {
      "epoch": 0.008827586206896552,
      "grad_norm": 8.19070816040039,
      "learning_rate": 1.909774436090226e-05,
      "loss": 0.4549538791179657,
      "num_input_tokens_seen": 77996,
      "step": 96,
      "train_runtime": 209.3374,
      "train_tokens_per_second": 372.585
    },
    {
      "epoch": 0.008919540229885057,
      "grad_norm": 7.147436141967773,
      "learning_rate": 1.9087719298245616e-05,
      "loss": 0.47513991594314575,
      "num_input_tokens_seen": 78568,
      "step": 97,
      "train_runtime": 211.23,
      "train_tokens_per_second": 371.955
    },
    {
      "epoch": 0.009011494252873562,
      "grad_norm": 7.666897773742676,
      "learning_rate": 1.9077694235588976e-05,
      "loss": 0.5075606107711792,
      "num_input_tokens_seen": 79212,
      "step": 98,
      "train_runtime": 213.1298,
      "train_tokens_per_second": 371.661
    },
    {
      "epoch": 0.00910344827586207,
      "grad_norm": 7.700671195983887,
      "learning_rate": 1.9067669172932333e-05,
      "loss": 0.5114626288414001,
      "num_input_tokens_seen": 79956,
      "step": 99,
      "train_runtime": 215.0445,
      "train_tokens_per_second": 371.811
    },
    {
      "epoch": 0.009195402298850575,
      "grad_norm": 8.367555618286133,
      "learning_rate": 1.905764411027569e-05,
      "loss": 0.4240979850292206,
      "num_input_tokens_seen": 80552,
      "step": 100,
      "train_runtime": 216.9449,
      "train_tokens_per_second": 371.302
    },
    {
      "epoch": 0.00928735632183908,
      "grad_norm": 8.55939769744873,
      "learning_rate": 1.904761904761905e-05,
      "loss": 0.5901445746421814,
      "num_input_tokens_seen": 81288,
      "step": 101,
      "train_runtime": 218.8644,
      "train_tokens_per_second": 371.408
    },
    {
      "epoch": 0.009379310344827587,
      "grad_norm": 7.81965446472168,
      "learning_rate": 1.903759398496241e-05,
      "loss": 0.5190936326980591,
      "num_input_tokens_seen": 81896,
      "step": 102,
      "train_runtime": 220.7905,
      "train_tokens_per_second": 370.922
    },
    {
      "epoch": 0.009471264367816092,
      "grad_norm": 13.143085479736328,
      "learning_rate": 1.9027568922305766e-05,
      "loss": 0.5595793724060059,
      "num_input_tokens_seen": 82568,
      "step": 103,
      "train_runtime": 222.687,
      "train_tokens_per_second": 370.781
    },
    {
      "epoch": 0.009563218390804597,
      "grad_norm": 8.601846694946289,
      "learning_rate": 1.9017543859649123e-05,
      "loss": 0.47885704040527344,
      "num_input_tokens_seen": 83164,
      "step": 104,
      "train_runtime": 224.5775,
      "train_tokens_per_second": 370.313
    },
    {
      "epoch": 0.009655172413793104,
      "grad_norm": 9.756768226623535,
      "learning_rate": 1.9007518796992483e-05,
      "loss": 0.5794239640235901,
      "num_input_tokens_seen": 83896,
      "step": 105,
      "train_runtime": 226.4862,
      "train_tokens_per_second": 370.424
    },
    {
      "epoch": 0.00974712643678161,
      "grad_norm": 10.664995193481445,
      "learning_rate": 1.899749373433584e-05,
      "loss": 0.651731014251709,
      "num_input_tokens_seen": 84464,
      "step": 106,
      "train_runtime": 228.3805,
      "train_tokens_per_second": 369.839
    },
    {
      "epoch": 0.009839080459770114,
      "grad_norm": 8.48730754852295,
      "learning_rate": 1.89874686716792e-05,
      "loss": 0.5737159848213196,
      "num_input_tokens_seen": 85132,
      "step": 107,
      "train_runtime": 230.3206,
      "train_tokens_per_second": 369.624
    },
    {
      "epoch": 0.009931034482758621,
      "grad_norm": 8.69455337524414,
      "learning_rate": 1.8977443609022557e-05,
      "loss": 0.5830780863761902,
      "num_input_tokens_seen": 85960,
      "step": 108,
      "train_runtime": 232.2347,
      "train_tokens_per_second": 370.143
    },
    {
      "epoch": 0.010022988505747127,
      "grad_norm": 8.370550155639648,
      "learning_rate": 1.8967418546365917e-05,
      "loss": 0.6618383526802063,
      "num_input_tokens_seen": 87276,
      "step": 109,
      "train_runtime": 234.1896,
      "train_tokens_per_second": 372.672
    },
    {
      "epoch": 0.010114942528735632,
      "grad_norm": 7.473726272583008,
      "learning_rate": 1.8957393483709274e-05,
      "loss": 0.6731764674186707,
      "num_input_tokens_seen": 88400,
      "step": 110,
      "train_runtime": 236.1578,
      "train_tokens_per_second": 374.326
    },
    {
      "epoch": 0.010206896551724139,
      "grad_norm": 8.688715934753418,
      "learning_rate": 1.894736842105263e-05,
      "loss": 0.596034049987793,
      "num_input_tokens_seen": 89320,
      "step": 111,
      "train_runtime": 238.0888,
      "train_tokens_per_second": 375.154
    },
    {
      "epoch": 0.010298850574712644,
      "grad_norm": 7.928012847900391,
      "learning_rate": 1.893734335839599e-05,
      "loss": 0.5787239074707031,
      "num_input_tokens_seen": 89964,
      "step": 112,
      "train_runtime": 239.9884,
      "train_tokens_per_second": 374.868
    },
    {
      "epoch": 0.010390804597701149,
      "grad_norm": 9.923861503601074,
      "learning_rate": 1.892731829573935e-05,
      "loss": 0.5914971232414246,
      "num_input_tokens_seen": 90668,
      "step": 113,
      "train_runtime": 241.8905,
      "train_tokens_per_second": 374.831
    },
    {
      "epoch": 0.010482758620689656,
      "grad_norm": 6.971408843994141,
      "learning_rate": 1.8917293233082708e-05,
      "loss": 0.5332297086715698,
      "num_input_tokens_seen": 91472,
      "step": 114,
      "train_runtime": 243.8157,
      "train_tokens_per_second": 375.169
    },
    {
      "epoch": 0.010574712643678161,
      "grad_norm": 8.14550495147705,
      "learning_rate": 1.890726817042607e-05,
      "loss": 0.6091741323471069,
      "num_input_tokens_seen": 92200,
      "step": 115,
      "train_runtime": 245.7518,
      "train_tokens_per_second": 375.175
    },
    {
      "epoch": 0.010666666666666666,
      "grad_norm": 6.912510395050049,
      "learning_rate": 1.8897243107769425e-05,
      "loss": 0.49012452363967896,
      "num_input_tokens_seen": 92828,
      "step": 116,
      "train_runtime": 247.633,
      "train_tokens_per_second": 374.861
    },
    {
      "epoch": 0.010758620689655173,
      "grad_norm": 7.863478660583496,
      "learning_rate": 1.8887218045112782e-05,
      "loss": 0.5282617211341858,
      "num_input_tokens_seen": 93792,
      "step": 117,
      "train_runtime": 249.5678,
      "train_tokens_per_second": 375.818
    },
    {
      "epoch": 0.010850574712643678,
      "grad_norm": 8.387312889099121,
      "learning_rate": 1.8877192982456142e-05,
      "loss": 0.6196162700653076,
      "num_input_tokens_seen": 94400,
      "step": 118,
      "train_runtime": 251.4511,
      "train_tokens_per_second": 375.421
    },
    {
      "epoch": 0.010942528735632184,
      "grad_norm": 9.462424278259277,
      "learning_rate": 1.88671679197995e-05,
      "loss": 0.6195122003555298,
      "num_input_tokens_seen": 95144,
      "step": 119,
      "train_runtime": 253.3371,
      "train_tokens_per_second": 375.563
    },
    {
      "epoch": 0.011034482758620689,
      "grad_norm": 7.591942310333252,
      "learning_rate": 1.885714285714286e-05,
      "loss": 0.4428488612174988,
      "num_input_tokens_seen": 95780,
      "step": 120,
      "train_runtime": 255.2228,
      "train_tokens_per_second": 375.28
    },
    {
      "epoch": 0.011126436781609196,
      "grad_norm": 7.117181777954102,
      "learning_rate": 1.8847117794486216e-05,
      "loss": 0.565083920955658,
      "num_input_tokens_seen": 97264,
      "step": 121,
      "train_runtime": 263.4535,
      "train_tokens_per_second": 369.188
    },
    {
      "epoch": 0.011218390804597701,
      "grad_norm": 7.480065822601318,
      "learning_rate": 1.8837092731829576e-05,
      "loss": 0.6817567348480225,
      "num_input_tokens_seen": 98964,
      "step": 122,
      "train_runtime": 265.617,
      "train_tokens_per_second": 372.582
    },
    {
      "epoch": 0.011310344827586206,
      "grad_norm": 7.171160697937012,
      "learning_rate": 1.8827067669172933e-05,
      "loss": 0.4195317029953003,
      "num_input_tokens_seen": 99484,
      "step": 123,
      "train_runtime": 267.4988,
      "train_tokens_per_second": 371.905
    },
    {
      "epoch": 0.011402298850574713,
      "grad_norm": 7.548673629760742,
      "learning_rate": 1.8817042606516293e-05,
      "loss": 0.512393593788147,
      "num_input_tokens_seen": 100088,
      "step": 124,
      "train_runtime": 269.4043,
      "train_tokens_per_second": 371.516
    },
    {
      "epoch": 0.011494252873563218,
      "grad_norm": 7.146829128265381,
      "learning_rate": 1.880701754385965e-05,
      "loss": 0.5014789700508118,
      "num_input_tokens_seen": 100696,
      "step": 125,
      "train_runtime": 271.2988,
      "train_tokens_per_second": 371.163
    },
    {
      "epoch": 0.011586206896551723,
      "grad_norm": 7.781763553619385,
      "learning_rate": 1.879699248120301e-05,
      "loss": 0.6027175784111023,
      "num_input_tokens_seen": 101424,
      "step": 126,
      "train_runtime": 273.1785,
      "train_tokens_per_second": 371.274
    },
    {
      "epoch": 0.01167816091954023,
      "grad_norm": 7.295479774475098,
      "learning_rate": 1.8786967418546367e-05,
      "loss": 0.5001509189605713,
      "num_input_tokens_seen": 102172,
      "step": 127,
      "train_runtime": 275.0744,
      "train_tokens_per_second": 371.434
    },
    {
      "epoch": 0.011770114942528736,
      "grad_norm": 7.57470703125,
      "learning_rate": 1.8776942355889727e-05,
      "loss": 0.5681148767471313,
      "num_input_tokens_seen": 103564,
      "step": 128,
      "train_runtime": 277.0633,
      "train_tokens_per_second": 373.792
    },
    {
      "epoch": 0.01186206896551724,
      "grad_norm": 7.474367141723633,
      "learning_rate": 1.8766917293233084e-05,
      "loss": 0.4590882658958435,
      "num_input_tokens_seen": 104224,
      "step": 129,
      "train_runtime": 278.9515,
      "train_tokens_per_second": 373.628
    },
    {
      "epoch": 0.011954022988505748,
      "grad_norm": 6.8280510902404785,
      "learning_rate": 1.8756892230576444e-05,
      "loss": 0.5731692910194397,
      "num_input_tokens_seen": 105212,
      "step": 130,
      "train_runtime": 280.8758,
      "train_tokens_per_second": 374.585
    },
    {
      "epoch": 0.012045977011494253,
      "grad_norm": 7.685474872589111,
      "learning_rate": 1.87468671679198e-05,
      "loss": 0.5310137271881104,
      "num_input_tokens_seen": 105796,
      "step": 131,
      "train_runtime": 282.7592,
      "train_tokens_per_second": 374.156
    },
    {
      "epoch": 0.012137931034482758,
      "grad_norm": 7.769192695617676,
      "learning_rate": 1.873684210526316e-05,
      "loss": 0.49234193563461304,
      "num_input_tokens_seen": 106456,
      "step": 132,
      "train_runtime": 284.6456,
      "train_tokens_per_second": 373.995
    },
    {
      "epoch": 0.012229885057471265,
      "grad_norm": 7.137904167175293,
      "learning_rate": 1.8726817042606518e-05,
      "loss": 0.49794071912765503,
      "num_input_tokens_seen": 107404,
      "step": 133,
      "train_runtime": 286.6011,
      "train_tokens_per_second": 374.751
    },
    {
      "epoch": 0.01232183908045977,
      "grad_norm": 7.115078926086426,
      "learning_rate": 1.8716791979949875e-05,
      "loss": 0.5229765176773071,
      "num_input_tokens_seen": 108220,
      "step": 134,
      "train_runtime": 288.5295,
      "train_tokens_per_second": 375.074
    },
    {
      "epoch": 0.012413793103448275,
      "grad_norm": 6.738773822784424,
      "learning_rate": 1.8706766917293235e-05,
      "loss": 0.5511250495910645,
      "num_input_tokens_seen": 108980,
      "step": 135,
      "train_runtime": 290.4219,
      "train_tokens_per_second": 375.247
    },
    {
      "epoch": 0.012505747126436782,
      "grad_norm": 7.377186298370361,
      "learning_rate": 1.869674185463659e-05,
      "loss": 0.54349684715271,
      "num_input_tokens_seen": 109844,
      "step": 136,
      "train_runtime": 292.3474,
      "train_tokens_per_second": 375.731
    },
    {
      "epoch": 0.012597701149425287,
      "grad_norm": 7.415782928466797,
      "learning_rate": 1.868671679197995e-05,
      "loss": 0.481075257062912,
      "num_input_tokens_seen": 110660,
      "step": 137,
      "train_runtime": 294.2855,
      "train_tokens_per_second": 376.029
    },
    {
      "epoch": 0.012689655172413793,
      "grad_norm": 7.624503135681152,
      "learning_rate": 1.8676691729323312e-05,
      "loss": 0.6776347160339355,
      "num_input_tokens_seen": 111668,
      "step": 138,
      "train_runtime": 296.2393,
      "train_tokens_per_second": 376.952
    },
    {
      "epoch": 0.0127816091954023,
      "grad_norm": 7.655098915100098,
      "learning_rate": 1.866666666666667e-05,
      "loss": 0.4513550102710724,
      "num_input_tokens_seen": 112296,
      "step": 139,
      "train_runtime": 298.1362,
      "train_tokens_per_second": 376.66
    },
    {
      "epoch": 0.012873563218390805,
      "grad_norm": 7.344651222229004,
      "learning_rate": 1.8656641604010025e-05,
      "loss": 0.5229818820953369,
      "num_input_tokens_seen": 112892,
      "step": 140,
      "train_runtime": 300.0228,
      "train_tokens_per_second": 376.278
    },
    {
      "epoch": 0.01296551724137931,
      "grad_norm": 7.522806167602539,
      "learning_rate": 1.8646616541353386e-05,
      "loss": 0.5736265778541565,
      "num_input_tokens_seen": 113760,
      "step": 141,
      "train_runtime": 301.9488,
      "train_tokens_per_second": 376.753
    },
    {
      "epoch": 0.013057471264367817,
      "grad_norm": 7.7204742431640625,
      "learning_rate": 1.8636591478696742e-05,
      "loss": 0.5658729672431946,
      "num_input_tokens_seen": 114956,
      "step": 142,
      "train_runtime": 303.9391,
      "train_tokens_per_second": 378.221
    },
    {
      "epoch": 0.013149425287356322,
      "grad_norm": 6.6276984214782715,
      "learning_rate": 1.8626566416040103e-05,
      "loss": 0.4700078070163727,
      "num_input_tokens_seen": 115932,
      "step": 143,
      "train_runtime": 305.8725,
      "train_tokens_per_second": 379.021
    },
    {
      "epoch": 0.013241379310344827,
      "grad_norm": 7.029630661010742,
      "learning_rate": 1.861654135338346e-05,
      "loss": 0.5873706340789795,
      "num_input_tokens_seen": 116604,
      "step": 144,
      "train_runtime": 307.7698,
      "train_tokens_per_second": 378.868
    },
    {
      "epoch": 0.013333333333333334,
      "grad_norm": 7.206961631774902,
      "learning_rate": 1.860651629072682e-05,
      "loss": 0.6351268291473389,
      "num_input_tokens_seen": 117608,
      "step": 145,
      "train_runtime": 309.72,
      "train_tokens_per_second": 379.724
    },
    {
      "epoch": 0.01342528735632184,
      "grad_norm": 6.574268817901611,
      "learning_rate": 1.8596491228070176e-05,
      "loss": 0.4774665832519531,
      "num_input_tokens_seen": 118228,
      "step": 146,
      "train_runtime": 311.6299,
      "train_tokens_per_second": 379.386
    },
    {
      "epoch": 0.013517241379310345,
      "grad_norm": 7.435815811157227,
      "learning_rate": 1.8586466165413533e-05,
      "loss": 0.43258947134017944,
      "num_input_tokens_seen": 118776,
      "step": 147,
      "train_runtime": 313.5359,
      "train_tokens_per_second": 378.827
    },
    {
      "epoch": 0.01360919540229885,
      "grad_norm": 7.4443888664245605,
      "learning_rate": 1.8576441102756893e-05,
      "loss": 0.5305068492889404,
      "num_input_tokens_seen": 119380,
      "step": 148,
      "train_runtime": 315.4398,
      "train_tokens_per_second": 378.456
    },
    {
      "epoch": 0.013701149425287357,
      "grad_norm": 7.344997882843018,
      "learning_rate": 1.8566416040100254e-05,
      "loss": 0.5557343363761902,
      "num_input_tokens_seen": 120012,
      "step": 149,
      "train_runtime": 317.3617,
      "train_tokens_per_second": 378.155
    },
    {
      "epoch": 0.013793103448275862,
      "grad_norm": 7.7498955726623535,
      "learning_rate": 1.855639097744361e-05,
      "loss": 0.46552395820617676,
      "num_input_tokens_seen": 120808,
      "step": 150,
      "train_runtime": 319.2852,
      "train_tokens_per_second": 378.37
    },
    {
      "epoch": 0.013885057471264367,
      "grad_norm": 8.965892791748047,
      "learning_rate": 1.8546365914786967e-05,
      "loss": 0.6639130115509033,
      "num_input_tokens_seen": 121472,
      "step": 151,
      "train_runtime": 327.5003,
      "train_tokens_per_second": 370.907
    },
    {
      "epoch": 0.013977011494252874,
      "grad_norm": 7.571508884429932,
      "learning_rate": 1.8536340852130327e-05,
      "loss": 0.4892725944519043,
      "num_input_tokens_seen": 122152,
      "step": 152,
      "train_runtime": 329.4378,
      "train_tokens_per_second": 370.789
    },
    {
      "epoch": 0.01406896551724138,
      "grad_norm": 8.367015838623047,
      "learning_rate": 1.8526315789473684e-05,
      "loss": 0.5385444760322571,
      "num_input_tokens_seen": 122776,
      "step": 153,
      "train_runtime": 331.3947,
      "train_tokens_per_second": 370.483
    },
    {
      "epoch": 0.014160919540229884,
      "grad_norm": 6.705316543579102,
      "learning_rate": 1.8516290726817044e-05,
      "loss": 0.5696083903312683,
      "num_input_tokens_seen": 123688,
      "step": 154,
      "train_runtime": 333.3162,
      "train_tokens_per_second": 371.083
    },
    {
      "epoch": 0.014252873563218391,
      "grad_norm": 7.952955722808838,
      "learning_rate": 1.8506265664160404e-05,
      "loss": 0.5096924901008606,
      "num_input_tokens_seen": 124320,
      "step": 155,
      "train_runtime": 335.2047,
      "train_tokens_per_second": 370.878
    },
    {
      "epoch": 0.014344827586206896,
      "grad_norm": 8.218250274658203,
      "learning_rate": 1.849624060150376e-05,
      "loss": 0.5339059829711914,
      "num_input_tokens_seen": 124964,
      "step": 156,
      "train_runtime": 337.0894,
      "train_tokens_per_second": 370.715
    },
    {
      "epoch": 0.014436781609195402,
      "grad_norm": 7.730725288391113,
      "learning_rate": 1.8486215538847118e-05,
      "loss": 0.5613219141960144,
      "num_input_tokens_seen": 125960,
      "step": 157,
      "train_runtime": 339.0384,
      "train_tokens_per_second": 371.521
    },
    {
      "epoch": 0.014528735632183909,
      "grad_norm": 7.79622220993042,
      "learning_rate": 1.8476190476190478e-05,
      "loss": 0.620633602142334,
      "num_input_tokens_seen": 126764,
      "step": 158,
      "train_runtime": 340.9557,
      "train_tokens_per_second": 371.79
    },
    {
      "epoch": 0.014620689655172414,
      "grad_norm": 7.506601810455322,
      "learning_rate": 1.8466165413533835e-05,
      "loss": 0.5521153211593628,
      "num_input_tokens_seen": 127476,
      "step": 159,
      "train_runtime": 342.8699,
      "train_tokens_per_second": 371.791
    },
    {
      "epoch": 0.014712643678160919,
      "grad_norm": 7.675780773162842,
      "learning_rate": 1.8456140350877195e-05,
      "loss": 0.5243714451789856,
      "num_input_tokens_seen": 128148,
      "step": 160,
      "train_runtime": 344.7562,
      "train_tokens_per_second": 371.706
    },
    {
      "epoch": 0.014804597701149426,
      "grad_norm": 7.011389255523682,
      "learning_rate": 1.8446115288220552e-05,
      "loss": 0.522988498210907,
      "num_input_tokens_seen": 128792,
      "step": 161,
      "train_runtime": 346.6336,
      "train_tokens_per_second": 371.551
    },
    {
      "epoch": 0.014896551724137931,
      "grad_norm": 9.304445266723633,
      "learning_rate": 1.8436090225563912e-05,
      "loss": 0.5413367748260498,
      "num_input_tokens_seen": 129340,
      "step": 162,
      "train_runtime": 348.4954,
      "train_tokens_per_second": 371.138
    },
    {
      "epoch": 0.014988505747126436,
      "grad_norm": 8.286294937133789,
      "learning_rate": 1.842606516290727e-05,
      "loss": 0.45945218205451965,
      "num_input_tokens_seen": 130000,
      "step": 163,
      "train_runtime": 350.3797,
      "train_tokens_per_second": 371.026
    },
    {
      "epoch": 0.015080459770114943,
      "grad_norm": 7.830119609832764,
      "learning_rate": 1.8416040100250626e-05,
      "loss": 0.5834083557128906,
      "num_input_tokens_seen": 130652,
      "step": 164,
      "train_runtime": 352.2848,
      "train_tokens_per_second": 370.87
    },
    {
      "epoch": 0.015172413793103448,
      "grad_norm": 6.1280012130737305,
      "learning_rate": 1.8406015037593986e-05,
      "loss": 0.5597648620605469,
      "num_input_tokens_seen": 132264,
      "step": 165,
      "train_runtime": 354.3074,
      "train_tokens_per_second": 373.303
    },
    {
      "epoch": 0.015264367816091954,
      "grad_norm": 6.631570816040039,
      "learning_rate": 1.8395989974937346e-05,
      "loss": 0.4796145558357239,
      "num_input_tokens_seen": 132920,
      "step": 166,
      "train_runtime": 356.2071,
      "train_tokens_per_second": 373.154
    },
    {
      "epoch": 0.01535632183908046,
      "grad_norm": 7.558768272399902,
      "learning_rate": 1.8385964912280703e-05,
      "loss": 0.5823076367378235,
      "num_input_tokens_seen": 133732,
      "step": 167,
      "train_runtime": 358.1485,
      "train_tokens_per_second": 373.398
    },
    {
      "epoch": 0.015448275862068966,
      "grad_norm": 8.231863021850586,
      "learning_rate": 1.8375939849624063e-05,
      "loss": 0.5344194173812866,
      "num_input_tokens_seen": 134456,
      "step": 168,
      "train_runtime": 360.0532,
      "train_tokens_per_second": 373.434
    },
    {
      "epoch": 0.015540229885057471,
      "grad_norm": 7.6874918937683105,
      "learning_rate": 1.836591478696742e-05,
      "loss": 0.49539342522621155,
      "num_input_tokens_seen": 135232,
      "step": 169,
      "train_runtime": 361.9481,
      "train_tokens_per_second": 373.623
    },
    {
      "epoch": 0.015632183908045976,
      "grad_norm": 7.448578834533691,
      "learning_rate": 1.8355889724310777e-05,
      "loss": 0.5806928873062134,
      "num_input_tokens_seen": 135920,
      "step": 170,
      "train_runtime": 363.8556,
      "train_tokens_per_second": 373.555
    },
    {
      "epoch": 0.01572413793103448,
      "grad_norm": 7.790022373199463,
      "learning_rate": 1.8345864661654137e-05,
      "loss": 0.5202723145484924,
      "num_input_tokens_seen": 136640,
      "step": 171,
      "train_runtime": 365.8044,
      "train_tokens_per_second": 373.533
    },
    {
      "epoch": 0.01581609195402299,
      "grad_norm": 8.534542083740234,
      "learning_rate": 1.8335839598997497e-05,
      "loss": 0.5159137845039368,
      "num_input_tokens_seen": 137316,
      "step": 172,
      "train_runtime": 367.7069,
      "train_tokens_per_second": 373.439
    },
    {
      "epoch": 0.015908045977011495,
      "grad_norm": 7.472785949707031,
      "learning_rate": 1.8325814536340854e-05,
      "loss": 0.5066118836402893,
      "num_input_tokens_seen": 137852,
      "step": 173,
      "train_runtime": 369.6042,
      "train_tokens_per_second": 372.972
    },
    {
      "epoch": 0.016,
      "grad_norm": 8.05160903930664,
      "learning_rate": 1.831578947368421e-05,
      "loss": 0.5389904975891113,
      "num_input_tokens_seen": 138592,
      "step": 174,
      "train_runtime": 371.4986,
      "train_tokens_per_second": 373.062
    },
    {
      "epoch": 0.016091954022988506,
      "grad_norm": 7.358802318572998,
      "learning_rate": 1.830576441102757e-05,
      "loss": 0.4707336723804474,
      "num_input_tokens_seen": 139180,
      "step": 175,
      "train_runtime": 373.3832,
      "train_tokens_per_second": 372.754
    },
    {
      "epoch": 0.01618390804597701,
      "grad_norm": 8.020739555358887,
      "learning_rate": 1.8295739348370928e-05,
      "loss": 0.6465672850608826,
      "num_input_tokens_seen": 140032,
      "step": 176,
      "train_runtime": 375.3123,
      "train_tokens_per_second": 373.108
    },
    {
      "epoch": 0.016275862068965516,
      "grad_norm": 8.557355880737305,
      "learning_rate": 1.8285714285714288e-05,
      "loss": 0.5434992909431458,
      "num_input_tokens_seen": 140760,
      "step": 177,
      "train_runtime": 377.2407,
      "train_tokens_per_second": 373.13
    },
    {
      "epoch": 0.016367816091954025,
      "grad_norm": 7.8487749099731445,
      "learning_rate": 1.8275689223057648e-05,
      "loss": 0.5491044521331787,
      "num_input_tokens_seen": 141376,
      "step": 178,
      "train_runtime": 379.1302,
      "train_tokens_per_second": 372.896
    },
    {
      "epoch": 0.01645977011494253,
      "grad_norm": 8.528973579406738,
      "learning_rate": 1.8265664160401005e-05,
      "loss": 0.5439373850822449,
      "num_input_tokens_seen": 142224,
      "step": 179,
      "train_runtime": 381.0765,
      "train_tokens_per_second": 373.216
    },
    {
      "epoch": 0.016551724137931035,
      "grad_norm": 8.664113998413086,
      "learning_rate": 1.825563909774436e-05,
      "loss": 0.5221170783042908,
      "num_input_tokens_seen": 143128,
      "step": 180,
      "train_runtime": 383.0288,
      "train_tokens_per_second": 373.674
    },
    {
      "epoch": 0.01664367816091954,
      "grad_norm": 7.165023326873779,
      "learning_rate": 1.824561403508772e-05,
      "loss": 0.5653952956199646,
      "num_input_tokens_seen": 143724,
      "step": 181,
      "train_runtime": 391.4096,
      "train_tokens_per_second": 367.196
    },
    {
      "epoch": 0.016735632183908045,
      "grad_norm": 7.648165702819824,
      "learning_rate": 1.823558897243108e-05,
      "loss": 0.6471481919288635,
      "num_input_tokens_seen": 144416,
      "step": 182,
      "train_runtime": 393.2982,
      "train_tokens_per_second": 367.192
    },
    {
      "epoch": 0.01682758620689655,
      "grad_norm": 7.569953918457031,
      "learning_rate": 1.822556390977444e-05,
      "loss": 0.5339801907539368,
      "num_input_tokens_seen": 145112,
      "step": 183,
      "train_runtime": 395.1992,
      "train_tokens_per_second": 367.187
    },
    {
      "epoch": 0.01691954022988506,
      "grad_norm": 6.800056457519531,
      "learning_rate": 1.8215538847117796e-05,
      "loss": 0.5550782084465027,
      "num_input_tokens_seen": 146044,
      "step": 184,
      "train_runtime": 397.1537,
      "train_tokens_per_second": 367.727
    },
    {
      "epoch": 0.017011494252873564,
      "grad_norm": 8.55329704284668,
      "learning_rate": 1.8205513784461156e-05,
      "loss": 0.5194864869117737,
      "num_input_tokens_seen": 146600,
      "step": 185,
      "train_runtime": 399.0518,
      "train_tokens_per_second": 367.371
    },
    {
      "epoch": 0.01710344827586207,
      "grad_norm": 7.240639686584473,
      "learning_rate": 1.8195488721804512e-05,
      "loss": 0.5830307006835938,
      "num_input_tokens_seen": 147380,
      "step": 186,
      "train_runtime": 401.0028,
      "train_tokens_per_second": 367.529
    },
    {
      "epoch": 0.017195402298850575,
      "grad_norm": 8.107155799865723,
      "learning_rate": 1.818546365914787e-05,
      "loss": 0.5524142980575562,
      "num_input_tokens_seen": 148104,
      "step": 187,
      "train_runtime": 402.964,
      "train_tokens_per_second": 367.537
    },
    {
      "epoch": 0.01728735632183908,
      "grad_norm": 7.435492038726807,
      "learning_rate": 1.817543859649123e-05,
      "loss": 0.6043378710746765,
      "num_input_tokens_seen": 149284,
      "step": 188,
      "train_runtime": 404.9412,
      "train_tokens_per_second": 368.656
    },
    {
      "epoch": 0.017379310344827585,
      "grad_norm": 6.442080020904541,
      "learning_rate": 1.816541353383459e-05,
      "loss": 0.46551936864852905,
      "num_input_tokens_seen": 150324,
      "step": 189,
      "train_runtime": 406.875,
      "train_tokens_per_second": 369.46
    },
    {
      "epoch": 0.01747126436781609,
      "grad_norm": 6.798178195953369,
      "learning_rate": 1.8155388471177946e-05,
      "loss": 0.49806857109069824,
      "num_input_tokens_seen": 151148,
      "step": 190,
      "train_runtime": 408.8109,
      "train_tokens_per_second": 369.726
    },
    {
      "epoch": 0.0175632183908046,
      "grad_norm": 7.303373336791992,
      "learning_rate": 1.8145363408521303e-05,
      "loss": 0.45829498767852783,
      "num_input_tokens_seen": 151860,
      "step": 191,
      "train_runtime": 410.7081,
      "train_tokens_per_second": 369.752
    },
    {
      "epoch": 0.017655172413793104,
      "grad_norm": 7.943329811096191,
      "learning_rate": 1.8135338345864663e-05,
      "loss": 0.5377188920974731,
      "num_input_tokens_seen": 152984,
      "step": 192,
      "train_runtime": 412.6619,
      "train_tokens_per_second": 370.725
    },
    {
      "epoch": 0.01774712643678161,
      "grad_norm": 7.833362579345703,
      "learning_rate": 1.812531328320802e-05,
      "loss": 0.5730334520339966,
      "num_input_tokens_seen": 153640,
      "step": 193,
      "train_runtime": 414.5476,
      "train_tokens_per_second": 370.621
    },
    {
      "epoch": 0.017839080459770115,
      "grad_norm": 7.654214382171631,
      "learning_rate": 1.811528822055138e-05,
      "loss": 0.5064957737922668,
      "num_input_tokens_seen": 154212,
      "step": 194,
      "train_runtime": 416.4986,
      "train_tokens_per_second": 370.258
    },
    {
      "epoch": 0.01793103448275862,
      "grad_norm": 7.190835952758789,
      "learning_rate": 1.810526315789474e-05,
      "loss": 0.498201847076416,
      "num_input_tokens_seen": 154832,
      "step": 195,
      "train_runtime": 418.3974,
      "train_tokens_per_second": 370.06
    },
    {
      "epoch": 0.018022988505747125,
      "grad_norm": 7.107874393463135,
      "learning_rate": 1.8095238095238097e-05,
      "loss": 0.547589123249054,
      "num_input_tokens_seen": 155480,
      "step": 196,
      "train_runtime": 420.3576,
      "train_tokens_per_second": 369.876
    },
    {
      "epoch": 0.018114942528735634,
      "grad_norm": 7.128254413604736,
      "learning_rate": 1.8085213032581454e-05,
      "loss": 0.4057043790817261,
      "num_input_tokens_seen": 156068,
      "step": 197,
      "train_runtime": 422.2485,
      "train_tokens_per_second": 369.612
    },
    {
      "epoch": 0.01820689655172414,
      "grad_norm": 6.993947982788086,
      "learning_rate": 1.8075187969924814e-05,
      "loss": 0.4971397817134857,
      "num_input_tokens_seen": 156796,
      "step": 198,
      "train_runtime": 424.1386,
      "train_tokens_per_second": 369.681
    },
    {
      "epoch": 0.018298850574712644,
      "grad_norm": 7.305413246154785,
      "learning_rate": 1.806516290726817e-05,
      "loss": 0.5406436920166016,
      "num_input_tokens_seen": 157392,
      "step": 199,
      "train_runtime": 426.0503,
      "train_tokens_per_second": 369.421
    },
    {
      "epoch": 0.01839080459770115,
      "grad_norm": 7.552101135253906,
      "learning_rate": 1.805513784461153e-05,
      "loss": 0.585086464881897,
      "num_input_tokens_seen": 158068,
      "step": 200,
      "train_runtime": 427.941,
      "train_tokens_per_second": 369.369
    },
    {
      "epoch": 0.01839080459770115,
      "eval_loss": 2.040510416030884,
      "eval_runtime": 28.0516,
      "eval_samples_per_second": 35.649,
      "eval_steps_per_second": 8.912,
      "num_input_tokens_seen": 158068,
      "step": 200
    },
    {
      "epoch": 0.018482758620689654,
      "grad_norm": 8.1217622756958,
      "learning_rate": 1.8045112781954888e-05,
      "loss": 0.5849893689155579,
      "num_input_tokens_seen": 158748,
      "step": 201,
      "train_runtime": 457.9204,
      "train_tokens_per_second": 346.672
    },
    {
      "epoch": 0.01857471264367816,
      "grad_norm": 7.835590839385986,
      "learning_rate": 1.8035087719298248e-05,
      "loss": 0.6012417674064636,
      "num_input_tokens_seen": 159464,
      "step": 202,
      "train_runtime": 459.8195,
      "train_tokens_per_second": 346.797
    },
    {
      "epoch": 0.018666666666666668,
      "grad_norm": 7.545464515686035,
      "learning_rate": 1.8025062656641605e-05,
      "loss": 0.6498401165008545,
      "num_input_tokens_seen": 160492,
      "step": 203,
      "train_runtime": 461.76,
      "train_tokens_per_second": 347.566
    },
    {
      "epoch": 0.018758620689655173,
      "grad_norm": 6.976754188537598,
      "learning_rate": 1.8015037593984962e-05,
      "loss": 0.4431565999984741,
      "num_input_tokens_seen": 161056,
      "step": 204,
      "train_runtime": 463.6636,
      "train_tokens_per_second": 347.355
    },
    {
      "epoch": 0.01885057471264368,
      "grad_norm": 7.590615749359131,
      "learning_rate": 1.8005012531328322e-05,
      "loss": 0.5572191476821899,
      "num_input_tokens_seen": 162040,
      "step": 205,
      "train_runtime": 465.6237,
      "train_tokens_per_second": 348.006
    },
    {
      "epoch": 0.018942528735632184,
      "grad_norm": 7.885451793670654,
      "learning_rate": 1.7994987468671682e-05,
      "loss": 0.5881732702255249,
      "num_input_tokens_seen": 162736,
      "step": 206,
      "train_runtime": 467.5271,
      "train_tokens_per_second": 348.078
    },
    {
      "epoch": 0.01903448275862069,
      "grad_norm": 7.598711013793945,
      "learning_rate": 1.798496240601504e-05,
      "loss": 0.5935019850730896,
      "num_input_tokens_seen": 163956,
      "step": 207,
      "train_runtime": 469.5279,
      "train_tokens_per_second": 349.193
    },
    {
      "epoch": 0.019126436781609194,
      "grad_norm": 6.595381736755371,
      "learning_rate": 1.79749373433584e-05,
      "loss": 0.4594612121582031,
      "num_input_tokens_seen": 164600,
      "step": 208,
      "train_runtime": 471.4212,
      "train_tokens_per_second": 349.157
    },
    {
      "epoch": 0.019218390804597703,
      "grad_norm": 8.24846076965332,
      "learning_rate": 1.7964912280701756e-05,
      "loss": 0.5824248194694519,
      "num_input_tokens_seen": 165280,
      "step": 209,
      "train_runtime": 473.3561,
      "train_tokens_per_second": 349.166
    },
    {
      "epoch": 0.019310344827586208,
      "grad_norm": 7.064051151275635,
      "learning_rate": 1.7954887218045113e-05,
      "loss": 0.4960821270942688,
      "num_input_tokens_seen": 165936,
      "step": 210,
      "train_runtime": 475.255,
      "train_tokens_per_second": 349.152
    },
    {
      "epoch": 0.019402298850574713,
      "grad_norm": 7.1449971199035645,
      "learning_rate": 1.7944862155388473e-05,
      "loss": 0.532036304473877,
      "num_input_tokens_seen": 166496,
      "step": 211,
      "train_runtime": 483.4971,
      "train_tokens_per_second": 344.358
    },
    {
      "epoch": 0.01949425287356322,
      "grad_norm": 7.421840667724609,
      "learning_rate": 1.7934837092731833e-05,
      "loss": 0.6296423077583313,
      "num_input_tokens_seen": 167480,
      "step": 212,
      "train_runtime": 485.468,
      "train_tokens_per_second": 344.987
    },
    {
      "epoch": 0.019586206896551724,
      "grad_norm": 7.221124649047852,
      "learning_rate": 1.792481203007519e-05,
      "loss": 0.47323280572891235,
      "num_input_tokens_seen": 168028,
      "step": 213,
      "train_runtime": 487.3766,
      "train_tokens_per_second": 344.76
    },
    {
      "epoch": 0.01967816091954023,
      "grad_norm": 7.000731468200684,
      "learning_rate": 1.7914786967418547e-05,
      "loss": 0.514447808265686,
      "num_input_tokens_seen": 168684,
      "step": 214,
      "train_runtime": 489.2893,
      "train_tokens_per_second": 344.753
    },
    {
      "epoch": 0.019770114942528734,
      "grad_norm": 7.600850582122803,
      "learning_rate": 1.7904761904761907e-05,
      "loss": 0.5671390891075134,
      "num_input_tokens_seen": 169720,
      "step": 215,
      "train_runtime": 491.264,
      "train_tokens_per_second": 345.476
    },
    {
      "epoch": 0.019862068965517243,
      "grad_norm": 7.677800178527832,
      "learning_rate": 1.7894736842105264e-05,
      "loss": 0.5139601826667786,
      "num_input_tokens_seen": 170640,
      "step": 216,
      "train_runtime": 493.1983,
      "train_tokens_per_second": 345.987
    },
    {
      "epoch": 0.019954022988505748,
      "grad_norm": 7.964269161224365,
      "learning_rate": 1.7884711779448624e-05,
      "loss": 0.5507964491844177,
      "num_input_tokens_seen": 171284,
      "step": 217,
      "train_runtime": 495.09,
      "train_tokens_per_second": 345.965
    },
    {
      "epoch": 0.020045977011494253,
      "grad_norm": 7.0535807609558105,
      "learning_rate": 1.7874686716791984e-05,
      "loss": 0.5258899331092834,
      "num_input_tokens_seen": 171928,
      "step": 218,
      "train_runtime": 496.9832,
      "train_tokens_per_second": 345.943
    },
    {
      "epoch": 0.020137931034482758,
      "grad_norm": 7.253530979156494,
      "learning_rate": 1.786466165413534e-05,
      "loss": 0.47745299339294434,
      "num_input_tokens_seen": 172484,
      "step": 219,
      "train_runtime": 498.8784,
      "train_tokens_per_second": 345.744
    },
    {
      "epoch": 0.020229885057471263,
      "grad_norm": 7.037721633911133,
      "learning_rate": 1.7854636591478698e-05,
      "loss": 0.5528386831283569,
      "num_input_tokens_seen": 173252,
      "step": 220,
      "train_runtime": 500.7811,
      "train_tokens_per_second": 345.964
    },
    {
      "epoch": 0.02032183908045977,
      "grad_norm": 8.044549942016602,
      "learning_rate": 1.7844611528822054e-05,
      "loss": 0.5703632831573486,
      "num_input_tokens_seen": 174108,
      "step": 221,
      "train_runtime": 502.7378,
      "train_tokens_per_second": 346.32
    },
    {
      "epoch": 0.020413793103448277,
      "grad_norm": 8.174488067626953,
      "learning_rate": 1.7834586466165415e-05,
      "loss": 0.647956132888794,
      "num_input_tokens_seen": 174944,
      "step": 222,
      "train_runtime": 504.6707,
      "train_tokens_per_second": 346.65
    },
    {
      "epoch": 0.020505747126436782,
      "grad_norm": 6.857749938964844,
      "learning_rate": 1.7824561403508775e-05,
      "loss": 0.42070627212524414,
      "num_input_tokens_seen": 175608,
      "step": 223,
      "train_runtime": 506.5865,
      "train_tokens_per_second": 346.65
    },
    {
      "epoch": 0.020597701149425288,
      "grad_norm": 7.1977972984313965,
      "learning_rate": 1.781453634085213e-05,
      "loss": 0.4759255647659302,
      "num_input_tokens_seen": 176304,
      "step": 224,
      "train_runtime": 508.5197,
      "train_tokens_per_second": 346.7
    },
    {
      "epoch": 0.020689655172413793,
      "grad_norm": 8.122084617614746,
      "learning_rate": 1.7804511278195492e-05,
      "loss": 0.5117164850234985,
      "num_input_tokens_seen": 176988,
      "step": 225,
      "train_runtime": 510.4634,
      "train_tokens_per_second": 346.72
    },
    {
      "epoch": 0.020781609195402298,
      "grad_norm": 8.753883361816406,
      "learning_rate": 1.779448621553885e-05,
      "loss": 0.6285914778709412,
      "num_input_tokens_seen": 177868,
      "step": 226,
      "train_runtime": 512.4213,
      "train_tokens_per_second": 347.113
    },
    {
      "epoch": 0.020873563218390803,
      "grad_norm": 8.017762184143066,
      "learning_rate": 1.7784461152882205e-05,
      "loss": 0.4385458827018738,
      "num_input_tokens_seen": 178496,
      "step": 227,
      "train_runtime": 514.3449,
      "train_tokens_per_second": 347.036
    },
    {
      "epoch": 0.020965517241379312,
      "grad_norm": 6.637861251831055,
      "learning_rate": 1.7774436090225566e-05,
      "loss": 0.41486579179763794,
      "num_input_tokens_seen": 179028,
      "step": 228,
      "train_runtime": 516.2433,
      "train_tokens_per_second": 346.79
    },
    {
      "epoch": 0.021057471264367817,
      "grad_norm": 7.022968769073486,
      "learning_rate": 1.7764411027568926e-05,
      "loss": 0.5290236473083496,
      "num_input_tokens_seen": 179752,
      "step": 229,
      "train_runtime": 518.1484,
      "train_tokens_per_second": 346.912
    },
    {
      "epoch": 0.021149425287356322,
      "grad_norm": 7.590112686157227,
      "learning_rate": 1.7754385964912283e-05,
      "loss": 0.5051460862159729,
      "num_input_tokens_seen": 180596,
      "step": 230,
      "train_runtime": 520.0747,
      "train_tokens_per_second": 347.25
    },
    {
      "epoch": 0.021241379310344827,
      "grad_norm": 7.5514702796936035,
      "learning_rate": 1.774436090225564e-05,
      "loss": 0.4634629189968109,
      "num_input_tokens_seen": 181284,
      "step": 231,
      "train_runtime": 521.9776,
      "train_tokens_per_second": 347.302
    },
    {
      "epoch": 0.021333333333333333,
      "grad_norm": 5.992765426635742,
      "learning_rate": 1.7734335839599e-05,
      "loss": 0.47778037190437317,
      "num_input_tokens_seen": 182468,
      "step": 232,
      "train_runtime": 523.9507,
      "train_tokens_per_second": 348.254
    },
    {
      "epoch": 0.021425287356321838,
      "grad_norm": 7.337306976318359,
      "learning_rate": 1.7724310776942356e-05,
      "loss": 0.5078150629997253,
      "num_input_tokens_seen": 183060,
      "step": 233,
      "train_runtime": 525.8389,
      "train_tokens_per_second": 348.129
    },
    {
      "epoch": 0.021517241379310346,
      "grad_norm": 8.190838813781738,
      "learning_rate": 1.7714285714285717e-05,
      "loss": 0.6096634864807129,
      "num_input_tokens_seen": 183764,
      "step": 234,
      "train_runtime": 527.7719,
      "train_tokens_per_second": 348.188
    },
    {
      "epoch": 0.02160919540229885,
      "grad_norm": 8.298602104187012,
      "learning_rate": 1.7704260651629073e-05,
      "loss": 0.5493342280387878,
      "num_input_tokens_seen": 184504,
      "step": 235,
      "train_runtime": 529.7032,
      "train_tokens_per_second": 348.316
    },
    {
      "epoch": 0.021701149425287357,
      "grad_norm": 6.748563766479492,
      "learning_rate": 1.7694235588972433e-05,
      "loss": 0.5589863061904907,
      "num_input_tokens_seen": 185332,
      "step": 236,
      "train_runtime": 531.6396,
      "train_tokens_per_second": 348.605
    },
    {
      "epoch": 0.021793103448275862,
      "grad_norm": 7.466015338897705,
      "learning_rate": 1.768421052631579e-05,
      "loss": 0.5016068816184998,
      "num_input_tokens_seen": 185900,
      "step": 237,
      "train_runtime": 533.5417,
      "train_tokens_per_second": 348.426
    },
    {
      "epoch": 0.021885057471264367,
      "grad_norm": 6.749599933624268,
      "learning_rate": 1.767418546365915e-05,
      "loss": 0.5446072220802307,
      "num_input_tokens_seen": 186916,
      "step": 238,
      "train_runtime": 535.5178,
      "train_tokens_per_second": 349.038
    },
    {
      "epoch": 0.021977011494252872,
      "grad_norm": 8.142538070678711,
      "learning_rate": 1.7664160401002507e-05,
      "loss": 0.47896608710289,
      "num_input_tokens_seen": 187520,
      "step": 239,
      "train_runtime": 537.467,
      "train_tokens_per_second": 348.896
    },
    {
      "epoch": 0.022068965517241378,
      "grad_norm": 7.028704643249512,
      "learning_rate": 1.7654135338345867e-05,
      "loss": 0.5260388255119324,
      "num_input_tokens_seen": 188244,
      "step": 240,
      "train_runtime": 539.4071,
      "train_tokens_per_second": 348.983
    },
    {
      "epoch": 0.022160919540229886,
      "grad_norm": 6.732629299163818,
      "learning_rate": 1.7644110275689224e-05,
      "loss": 0.6071557998657227,
      "num_input_tokens_seen": 190136,
      "step": 241,
      "train_runtime": 547.7035,
      "train_tokens_per_second": 347.151
    },
    {
      "epoch": 0.02225287356321839,
      "grad_norm": 6.396223545074463,
      "learning_rate": 1.7634085213032584e-05,
      "loss": 0.47636085748672485,
      "num_input_tokens_seen": 190776,
      "step": 242,
      "train_runtime": 549.6296,
      "train_tokens_per_second": 347.099
    },
    {
      "epoch": 0.022344827586206897,
      "grad_norm": 7.123258113861084,
      "learning_rate": 1.762406015037594e-05,
      "loss": 0.5651535987854004,
      "num_input_tokens_seen": 191492,
      "step": 243,
      "train_runtime": 551.6421,
      "train_tokens_per_second": 347.131
    },
    {
      "epoch": 0.022436781609195402,
      "grad_norm": 7.592582702636719,
      "learning_rate": 1.7614035087719298e-05,
      "loss": 0.5869500041007996,
      "num_input_tokens_seen": 192244,
      "step": 244,
      "train_runtime": 553.5882,
      "train_tokens_per_second": 347.269
    },
    {
      "epoch": 0.022528735632183907,
      "grad_norm": 7.659783363342285,
      "learning_rate": 1.7604010025062658e-05,
      "loss": 0.5372046232223511,
      "num_input_tokens_seen": 193056,
      "step": 245,
      "train_runtime": 555.5172,
      "train_tokens_per_second": 347.525
    },
    {
      "epoch": 0.022620689655172412,
      "grad_norm": 7.2951788902282715,
      "learning_rate": 1.7593984962406015e-05,
      "loss": 0.5913550853729248,
      "num_input_tokens_seen": 193632,
      "step": 246,
      "train_runtime": 557.4088,
      "train_tokens_per_second": 347.379
    },
    {
      "epoch": 0.02271264367816092,
      "grad_norm": 7.493248462677002,
      "learning_rate": 1.7583959899749375e-05,
      "loss": 0.648095428943634,
      "num_input_tokens_seen": 194468,
      "step": 247,
      "train_runtime": 559.366,
      "train_tokens_per_second": 347.658
    },
    {
      "epoch": 0.022804597701149426,
      "grad_norm": 6.816006183624268,
      "learning_rate": 1.7573934837092735e-05,
      "loss": 0.5106541514396667,
      "num_input_tokens_seen": 195100,
      "step": 248,
      "train_runtime": 561.2475,
      "train_tokens_per_second": 347.619
    },
    {
      "epoch": 0.02289655172413793,
      "grad_norm": 8.779420852661133,
      "learning_rate": 1.7563909774436092e-05,
      "loss": 0.5156307220458984,
      "num_input_tokens_seen": 195844,
      "step": 249,
      "train_runtime": 563.1295,
      "train_tokens_per_second": 347.778
    },
    {
      "epoch": 0.022988505747126436,
      "grad_norm": 6.330947399139404,
      "learning_rate": 1.755388471177945e-05,
      "loss": 0.566486656665802,
      "num_input_tokens_seen": 196664,
      "step": 250,
      "train_runtime": 565.074,
      "train_tokens_per_second": 348.032
    },
    {
      "epoch": 0.02308045977011494,
      "grad_norm": 7.991855621337891,
      "learning_rate": 1.754385964912281e-05,
      "loss": 0.5192869901657104,
      "num_input_tokens_seen": 197168,
      "step": 251,
      "train_runtime": 567.021,
      "train_tokens_per_second": 347.726
    },
    {
      "epoch": 0.023172413793103447,
      "grad_norm": 9.131534576416016,
      "learning_rate": 1.7533834586466166e-05,
      "loss": 0.5779814124107361,
      "num_input_tokens_seen": 198076,
      "step": 252,
      "train_runtime": 568.9457,
      "train_tokens_per_second": 348.146
    },
    {
      "epoch": 0.023264367816091956,
      "grad_norm": 6.700856685638428,
      "learning_rate": 1.7523809523809526e-05,
      "loss": 0.5244114398956299,
      "num_input_tokens_seen": 198616,
      "step": 253,
      "train_runtime": 570.8471,
      "train_tokens_per_second": 347.932
    },
    {
      "epoch": 0.02335632183908046,
      "grad_norm": 7.709068298339844,
      "learning_rate": 1.7513784461152883e-05,
      "loss": 0.5432565808296204,
      "num_input_tokens_seen": 199348,
      "step": 254,
      "train_runtime": 572.7795,
      "train_tokens_per_second": 348.036
    },
    {
      "epoch": 0.023448275862068966,
      "grad_norm": 7.840363502502441,
      "learning_rate": 1.7503759398496243e-05,
      "loss": 0.6463830471038818,
      "num_input_tokens_seen": 200348,
      "step": 255,
      "train_runtime": 574.7446,
      "train_tokens_per_second": 348.586
    },
    {
      "epoch": 0.02354022988505747,
      "grad_norm": 7.466499328613281,
      "learning_rate": 1.74937343358396e-05,
      "loss": 0.6219689249992371,
      "num_input_tokens_seen": 201288,
      "step": 256,
      "train_runtime": 576.7192,
      "train_tokens_per_second": 349.023
    },
    {
      "epoch": 0.023632183908045976,
      "grad_norm": 7.196322441101074,
      "learning_rate": 1.7483709273182957e-05,
      "loss": 0.4431312382221222,
      "num_input_tokens_seen": 202012,
      "step": 257,
      "train_runtime": 578.6366,
      "train_tokens_per_second": 349.117
    },
    {
      "epoch": 0.02372413793103448,
      "grad_norm": 6.883704662322998,
      "learning_rate": 1.7473684210526317e-05,
      "loss": 0.5374019742012024,
      "num_input_tokens_seen": 202716,
      "step": 258,
      "train_runtime": 580.5259,
      "train_tokens_per_second": 349.194
    },
    {
      "epoch": 0.02381609195402299,
      "grad_norm": 7.998088359832764,
      "learning_rate": 1.7463659147869677e-05,
      "loss": 0.5854915976524353,
      "num_input_tokens_seen": 203360,
      "step": 259,
      "train_runtime": 582.4157,
      "train_tokens_per_second": 349.166
    },
    {
      "epoch": 0.023908045977011495,
      "grad_norm": 7.955687999725342,
      "learning_rate": 1.7453634085213034e-05,
      "loss": 0.6497164964675903,
      "num_input_tokens_seen": 204056,
      "step": 260,
      "train_runtime": 584.2994,
      "train_tokens_per_second": 349.232
    },
    {
      "epoch": 0.024,
      "grad_norm": 7.1547322273254395,
      "learning_rate": 1.744360902255639e-05,
      "loss": 0.41980886459350586,
      "num_input_tokens_seen": 204944,
      "step": 261,
      "train_runtime": 586.2316,
      "train_tokens_per_second": 349.596
    },
    {
      "epoch": 0.024091954022988506,
      "grad_norm": 7.54673433303833,
      "learning_rate": 1.743358395989975e-05,
      "loss": 0.5536658763885498,
      "num_input_tokens_seen": 205500,
      "step": 262,
      "train_runtime": 588.1355,
      "train_tokens_per_second": 349.409
    },
    {
      "epoch": 0.02418390804597701,
      "grad_norm": 7.382690906524658,
      "learning_rate": 1.7423558897243108e-05,
      "loss": 0.6529373526573181,
      "num_input_tokens_seen": 206492,
      "step": 263,
      "train_runtime": 590.1008,
      "train_tokens_per_second": 349.927
    },
    {
      "epoch": 0.024275862068965516,
      "grad_norm": 7.5943098068237305,
      "learning_rate": 1.7413533834586468e-05,
      "loss": 0.5493870377540588,
      "num_input_tokens_seen": 207152,
      "step": 264,
      "train_runtime": 592.0444,
      "train_tokens_per_second": 349.893
    },
    {
      "epoch": 0.02436781609195402,
      "grad_norm": 7.41045618057251,
      "learning_rate": 1.7403508771929828e-05,
      "loss": 0.5508020520210266,
      "num_input_tokens_seen": 207860,
      "step": 265,
      "train_runtime": 593.9381,
      "train_tokens_per_second": 349.969
    },
    {
      "epoch": 0.02445977011494253,
      "grad_norm": 6.776822566986084,
      "learning_rate": 1.7393483709273185e-05,
      "loss": 0.5031247735023499,
      "num_input_tokens_seen": 208508,
      "step": 266,
      "train_runtime": 595.833,
      "train_tokens_per_second": 349.944
    },
    {
      "epoch": 0.024551724137931035,
      "grad_norm": 8.271478652954102,
      "learning_rate": 1.738345864661654e-05,
      "loss": 0.4709621071815491,
      "num_input_tokens_seen": 209376,
      "step": 267,
      "train_runtime": 597.7989,
      "train_tokens_per_second": 350.245
    },
    {
      "epoch": 0.02464367816091954,
      "grad_norm": 6.908759593963623,
      "learning_rate": 1.7373433583959902e-05,
      "loss": 0.5502421259880066,
      "num_input_tokens_seen": 210492,
      "step": 268,
      "train_runtime": 599.7666,
      "train_tokens_per_second": 350.957
    },
    {
      "epoch": 0.024735632183908045,
      "grad_norm": 7.29292106628418,
      "learning_rate": 1.736340852130326e-05,
      "loss": 0.557747483253479,
      "num_input_tokens_seen": 211200,
      "step": 269,
      "train_runtime": 601.663,
      "train_tokens_per_second": 351.027
    },
    {
      "epoch": 0.02482758620689655,
      "grad_norm": 6.774946689605713,
      "learning_rate": 1.735338345864662e-05,
      "loss": 0.5118838548660278,
      "num_input_tokens_seen": 211868,
      "step": 270,
      "train_runtime": 603.5524,
      "train_tokens_per_second": 351.035
    },
    {
      "epoch": 0.024919540229885056,
      "grad_norm": 7.238329887390137,
      "learning_rate": 1.7343358395989975e-05,
      "loss": 0.614051342010498,
      "num_input_tokens_seen": 212612,
      "step": 271,
      "train_runtime": 611.6827,
      "train_tokens_per_second": 347.585
    },
    {
      "epoch": 0.025011494252873565,
      "grad_norm": 6.524905204772949,
      "learning_rate": 1.7333333333333336e-05,
      "loss": 0.5046440362930298,
      "num_input_tokens_seen": 213204,
      "step": 272,
      "train_runtime": 613.9334,
      "train_tokens_per_second": 347.275
    },
    {
      "epoch": 0.02510344827586207,
      "grad_norm": 8.400614738464355,
      "learning_rate": 1.7323308270676692e-05,
      "loss": 0.6119961738586426,
      "num_input_tokens_seen": 213896,
      "step": 273,
      "train_runtime": 615.8391,
      "train_tokens_per_second": 347.324
    },
    {
      "epoch": 0.025195402298850575,
      "grad_norm": 7.2303900718688965,
      "learning_rate": 1.731328320802005e-05,
      "loss": 0.5299623608589172,
      "num_input_tokens_seen": 214556,
      "step": 274,
      "train_runtime": 617.7405,
      "train_tokens_per_second": 347.324
    },
    {
      "epoch": 0.02528735632183908,
      "grad_norm": 6.81049108505249,
      "learning_rate": 1.730325814536341e-05,
      "loss": 0.528329610824585,
      "num_input_tokens_seen": 215248,
      "step": 275,
      "train_runtime": 619.6613,
      "train_tokens_per_second": 347.364
    },
    {
      "epoch": 0.025379310344827585,
      "grad_norm": 7.140843868255615,
      "learning_rate": 1.729323308270677e-05,
      "loss": 0.4379081428050995,
      "num_input_tokens_seen": 216176,
      "step": 276,
      "train_runtime": 621.5847,
      "train_tokens_per_second": 347.782
    },
    {
      "epoch": 0.02547126436781609,
      "grad_norm": 8.193724632263184,
      "learning_rate": 1.7283208020050126e-05,
      "loss": 0.6148580312728882,
      "num_input_tokens_seen": 217052,
      "step": 277,
      "train_runtime": 623.5334,
      "train_tokens_per_second": 348.1
    },
    {
      "epoch": 0.0255632183908046,
      "grad_norm": 8.05411148071289,
      "learning_rate": 1.7273182957393487e-05,
      "loss": 0.5443435907363892,
      "num_input_tokens_seen": 217816,
      "step": 278,
      "train_runtime": 625.4334,
      "train_tokens_per_second": 348.264
    },
    {
      "epoch": 0.025655172413793104,
      "grad_norm": 7.705923080444336,
      "learning_rate": 1.7263157894736843e-05,
      "loss": 0.5824005603790283,
      "num_input_tokens_seen": 218520,
      "step": 279,
      "train_runtime": 627.3256,
      "train_tokens_per_second": 348.336
    },
    {
      "epoch": 0.02574712643678161,
      "grad_norm": 7.342845916748047,
      "learning_rate": 1.72531328320802e-05,
      "loss": 0.5593616962432861,
      "num_input_tokens_seen": 219480,
      "step": 280,
      "train_runtime": 629.2501,
      "train_tokens_per_second": 348.796
    },
    {
      "epoch": 0.025839080459770115,
      "grad_norm": 7.2280778884887695,
      "learning_rate": 1.724310776942356e-05,
      "loss": 0.6934468746185303,
      "num_input_tokens_seen": 220408,
      "step": 281,
      "train_runtime": 631.1774,
      "train_tokens_per_second": 349.201
    },
    {
      "epoch": 0.02593103448275862,
      "grad_norm": 9.139898300170898,
      "learning_rate": 1.723308270676692e-05,
      "loss": 0.5340681672096252,
      "num_input_tokens_seen": 220996,
      "step": 282,
      "train_runtime": 633.0838,
      "train_tokens_per_second": 349.079
    },
    {
      "epoch": 0.026022988505747125,
      "grad_norm": 7.518560409545898,
      "learning_rate": 1.7223057644110277e-05,
      "loss": 0.5287448763847351,
      "num_input_tokens_seen": 221592,
      "step": 283,
      "train_runtime": 634.9804,
      "train_tokens_per_second": 348.975
    },
    {
      "epoch": 0.026114942528735634,
      "grad_norm": 6.808462142944336,
      "learning_rate": 1.7213032581453634e-05,
      "loss": 0.47242191433906555,
      "num_input_tokens_seen": 222288,
      "step": 284,
      "train_runtime": 636.8682,
      "train_tokens_per_second": 349.033
    },
    {
      "epoch": 0.02620689655172414,
      "grad_norm": 7.224859714508057,
      "learning_rate": 1.7203007518796994e-05,
      "loss": 0.5434243679046631,
      "num_input_tokens_seen": 222948,
      "step": 285,
      "train_runtime": 638.777,
      "train_tokens_per_second": 349.023
    },
    {
      "epoch": 0.026298850574712644,
      "grad_norm": 7.309472560882568,
      "learning_rate": 1.719298245614035e-05,
      "loss": 0.49765464663505554,
      "num_input_tokens_seen": 223592,
      "step": 286,
      "train_runtime": 640.6969,
      "train_tokens_per_second": 348.982
    },
    {
      "epoch": 0.02639080459770115,
      "grad_norm": 7.0516862869262695,
      "learning_rate": 1.718295739348371e-05,
      "loss": 0.4650920629501343,
      "num_input_tokens_seen": 224504,
      "step": 287,
      "train_runtime": 642.6571,
      "train_tokens_per_second": 349.337
    },
    {
      "epoch": 0.026482758620689655,
      "grad_norm": 7.711924076080322,
      "learning_rate": 1.717293233082707e-05,
      "loss": 0.5413693189620972,
      "num_input_tokens_seen": 225156,
      "step": 288,
      "train_runtime": 644.5908,
      "train_tokens_per_second": 349.301
    },
    {
      "epoch": 0.02657471264367816,
      "grad_norm": 6.597391605377197,
      "learning_rate": 1.7162907268170428e-05,
      "loss": 0.47959834337234497,
      "num_input_tokens_seen": 225716,
      "step": 289,
      "train_runtime": 646.5015,
      "train_tokens_per_second": 349.135
    },
    {
      "epoch": 0.02666666666666667,
      "grad_norm": 6.937802314758301,
      "learning_rate": 1.7152882205513785e-05,
      "loss": 0.5246532559394836,
      "num_input_tokens_seen": 226472,
      "step": 290,
      "train_runtime": 648.4354,
      "train_tokens_per_second": 349.259
    },
    {
      "epoch": 0.026758620689655174,
      "grad_norm": 7.025156497955322,
      "learning_rate": 1.7142857142857142e-05,
      "loss": 0.4767950773239136,
      "num_input_tokens_seen": 227260,
      "step": 291,
      "train_runtime": 650.366,
      "train_tokens_per_second": 349.434
    },
    {
      "epoch": 0.02685057471264368,
      "grad_norm": 6.792970180511475,
      "learning_rate": 1.7132832080200502e-05,
      "loss": 0.4855608344078064,
      "num_input_tokens_seen": 227876,
      "step": 292,
      "train_runtime": 652.2724,
      "train_tokens_per_second": 349.357
    },
    {
      "epoch": 0.026942528735632184,
      "grad_norm": 6.304465293884277,
      "learning_rate": 1.7122807017543862e-05,
      "loss": 0.4634402394294739,
      "num_input_tokens_seen": 228468,
      "step": 293,
      "train_runtime": 654.1767,
      "train_tokens_per_second": 349.245
    },
    {
      "epoch": 0.02703448275862069,
      "grad_norm": 7.347883224487305,
      "learning_rate": 1.711278195488722e-05,
      "loss": 0.49251794815063477,
      "num_input_tokens_seen": 229120,
      "step": 294,
      "train_runtime": 656.0684,
      "train_tokens_per_second": 349.232
    },
    {
      "epoch": 0.027126436781609194,
      "grad_norm": 7.306178569793701,
      "learning_rate": 1.710275689223058e-05,
      "loss": 0.4296570420265198,
      "num_input_tokens_seen": 229800,
      "step": 295,
      "train_runtime": 657.9788,
      "train_tokens_per_second": 349.251
    },
    {
      "epoch": 0.0272183908045977,
      "grad_norm": 7.8655219078063965,
      "learning_rate": 1.7092731829573936e-05,
      "loss": 0.418113112449646,
      "num_input_tokens_seen": 230396,
      "step": 296,
      "train_runtime": 659.9753,
      "train_tokens_per_second": 349.098
    },
    {
      "epoch": 0.027310344827586208,
      "grad_norm": 6.831302642822266,
      "learning_rate": 1.7082706766917293e-05,
      "loss": 0.5011525750160217,
      "num_input_tokens_seen": 231020,
      "step": 297,
      "train_runtime": 661.8719,
      "train_tokens_per_second": 349.04
    },
    {
      "epoch": 0.027402298850574713,
      "grad_norm": 7.1317243576049805,
      "learning_rate": 1.7072681704260653e-05,
      "loss": 0.5581461787223816,
      "num_input_tokens_seen": 231636,
      "step": 298,
      "train_runtime": 663.7711,
      "train_tokens_per_second": 348.97
    },
    {
      "epoch": 0.02749425287356322,
      "grad_norm": 8.014022827148438,
      "learning_rate": 1.7062656641604013e-05,
      "loss": 0.4860498905181885,
      "num_input_tokens_seen": 232208,
      "step": 299,
      "train_runtime": 665.66,
      "train_tokens_per_second": 348.839
    },
    {
      "epoch": 0.027586206896551724,
      "grad_norm": 7.223250389099121,
      "learning_rate": 1.705263157894737e-05,
      "loss": 0.5932018756866455,
      "num_input_tokens_seen": 233212,
      "step": 300,
      "train_runtime": 667.6343,
      "train_tokens_per_second": 349.311
    },
    {
      "epoch": 0.02767816091954023,
      "grad_norm": 6.203488826751709,
      "learning_rate": 1.7042606516290727e-05,
      "loss": 0.4114920496940613,
      "num_input_tokens_seen": 233744,
      "step": 301,
      "train_runtime": 676.0733,
      "train_tokens_per_second": 345.738
    },
    {
      "epoch": 0.027770114942528734,
      "grad_norm": 7.068984508514404,
      "learning_rate": 1.7032581453634087e-05,
      "loss": 0.48409998416900635,
      "num_input_tokens_seen": 234376,
      "step": 302,
      "train_runtime": 678.0913,
      "train_tokens_per_second": 345.641
    },
    {
      "epoch": 0.027862068965517243,
      "grad_norm": 6.957635402679443,
      "learning_rate": 1.7022556390977444e-05,
      "loss": 0.5025308728218079,
      "num_input_tokens_seen": 235152,
      "step": 303,
      "train_runtime": 680.1595,
      "train_tokens_per_second": 345.731
    },
    {
      "epoch": 0.027954022988505748,
      "grad_norm": 7.4896650314331055,
      "learning_rate": 1.7012531328320804e-05,
      "loss": 0.5307778120040894,
      "num_input_tokens_seen": 235736,
      "step": 304,
      "train_runtime": 682.0385,
      "train_tokens_per_second": 345.634
    },
    {
      "epoch": 0.028045977011494253,
      "grad_norm": 6.645881652832031,
      "learning_rate": 1.7002506265664164e-05,
      "loss": 0.42069652676582336,
      "num_input_tokens_seen": 236504,
      "step": 305,
      "train_runtime": 683.9612,
      "train_tokens_per_second": 345.786
    },
    {
      "epoch": 0.02813793103448276,
      "grad_norm": 6.69299840927124,
      "learning_rate": 1.699248120300752e-05,
      "loss": 0.5573474764823914,
      "num_input_tokens_seen": 237360,
      "step": 306,
      "train_runtime": 685.9102,
      "train_tokens_per_second": 346.051
    },
    {
      "epoch": 0.028229885057471264,
      "grad_norm": 15.929272651672363,
      "learning_rate": 1.6982456140350878e-05,
      "loss": 0.475852906703949,
      "num_input_tokens_seen": 238012,
      "step": 307,
      "train_runtime": 687.8612,
      "train_tokens_per_second": 346.017
    },
    {
      "epoch": 0.02832183908045977,
      "grad_norm": 8.076292037963867,
      "learning_rate": 1.6972431077694238e-05,
      "loss": 0.6119599938392639,
      "num_input_tokens_seen": 238804,
      "step": 308,
      "train_runtime": 689.7721,
      "train_tokens_per_second": 346.207
    },
    {
      "epoch": 0.028413793103448277,
      "grad_norm": 7.994293212890625,
      "learning_rate": 1.6962406015037595e-05,
      "loss": 0.5304892659187317,
      "num_input_tokens_seen": 239544,
      "step": 309,
      "train_runtime": 691.7,
      "train_tokens_per_second": 346.312
    },
    {
      "epoch": 0.028505747126436783,
      "grad_norm": 7.314886093139648,
      "learning_rate": 1.6952380952380955e-05,
      "loss": 0.5019823908805847,
      "num_input_tokens_seen": 240252,
      "step": 310,
      "train_runtime": 693.5857,
      "train_tokens_per_second": 346.391
    },
    {
      "epoch": 0.028597701149425288,
      "grad_norm": 7.226398468017578,
      "learning_rate": 1.694235588972431e-05,
      "loss": 0.5056269764900208,
      "num_input_tokens_seen": 241092,
      "step": 311,
      "train_runtime": 695.5132,
      "train_tokens_per_second": 346.639
    },
    {
      "epoch": 0.028689655172413793,
      "grad_norm": 7.674276828765869,
      "learning_rate": 1.6932330827067672e-05,
      "loss": 0.5091602802276611,
      "num_input_tokens_seen": 241748,
      "step": 312,
      "train_runtime": 697.4204,
      "train_tokens_per_second": 346.632
    },
    {
      "epoch": 0.028781609195402298,
      "grad_norm": 7.352941989898682,
      "learning_rate": 1.692230576441103e-05,
      "loss": 0.6187993884086609,
      "num_input_tokens_seen": 242908,
      "step": 313,
      "train_runtime": 699.4205,
      "train_tokens_per_second": 347.299
    },
    {
      "epoch": 0.028873563218390803,
      "grad_norm": 6.615414619445801,
      "learning_rate": 1.6912280701754385e-05,
      "loss": 0.5045319199562073,
      "num_input_tokens_seen": 243740,
      "step": 314,
      "train_runtime": 701.3581,
      "train_tokens_per_second": 347.526
    },
    {
      "epoch": 0.028965517241379312,
      "grad_norm": 7.64569616317749,
      "learning_rate": 1.6902255639097746e-05,
      "loss": 0.5935924649238586,
      "num_input_tokens_seen": 244772,
      "step": 315,
      "train_runtime": 703.3276,
      "train_tokens_per_second": 348.02
    },
    {
      "epoch": 0.029057471264367817,
      "grad_norm": 7.15838623046875,
      "learning_rate": 1.6892230576441106e-05,
      "loss": 0.5581420063972473,
      "num_input_tokens_seen": 246072,
      "step": 316,
      "train_runtime": 705.331,
      "train_tokens_per_second": 348.874
    },
    {
      "epoch": 0.029149425287356322,
      "grad_norm": 7.60428524017334,
      "learning_rate": 1.6882205513784463e-05,
      "loss": 0.5789986848831177,
      "num_input_tokens_seen": 246808,
      "step": 317,
      "train_runtime": 707.2303,
      "train_tokens_per_second": 348.978
    },
    {
      "epoch": 0.029241379310344828,
      "grad_norm": 8.331788063049316,
      "learning_rate": 1.6872180451127823e-05,
      "loss": 0.4795679450035095,
      "num_input_tokens_seen": 247472,
      "step": 318,
      "train_runtime": 709.1367,
      "train_tokens_per_second": 348.976
    },
    {
      "epoch": 0.029333333333333333,
      "grad_norm": 6.896542549133301,
      "learning_rate": 1.686215538847118e-05,
      "loss": 0.49073925614356995,
      "num_input_tokens_seen": 248160,
      "step": 319,
      "train_runtime": 711.0311,
      "train_tokens_per_second": 349.014
    },
    {
      "epoch": 0.029425287356321838,
      "grad_norm": 6.418603420257568,
      "learning_rate": 1.6852130325814536e-05,
      "loss": 0.5961854457855225,
      "num_input_tokens_seen": 249272,
      "step": 320,
      "train_runtime": 712.9908,
      "train_tokens_per_second": 349.615
    },
    {
      "epoch": 0.029517241379310343,
      "grad_norm": 7.81927490234375,
      "learning_rate": 1.6842105263157896e-05,
      "loss": 0.5365660786628723,
      "num_input_tokens_seen": 249932,
      "step": 321,
      "train_runtime": 714.8848,
      "train_tokens_per_second": 349.612
    },
    {
      "epoch": 0.029609195402298852,
      "grad_norm": 6.75691032409668,
      "learning_rate": 1.6832080200501257e-05,
      "loss": 0.6306750774383545,
      "num_input_tokens_seen": 250836,
      "step": 322,
      "train_runtime": 716.8207,
      "train_tokens_per_second": 349.929
    },
    {
      "epoch": 0.029701149425287357,
      "grad_norm": 7.404897689819336,
      "learning_rate": 1.6822055137844613e-05,
      "loss": 0.3980158567428589,
      "num_input_tokens_seen": 251360,
      "step": 323,
      "train_runtime": 718.7182,
      "train_tokens_per_second": 349.734
    },
    {
      "epoch": 0.029793103448275862,
      "grad_norm": 7.205636024475098,
      "learning_rate": 1.681203007518797e-05,
      "loss": 0.49449974298477173,
      "num_input_tokens_seen": 252000,
      "step": 324,
      "train_runtime": 720.6201,
      "train_tokens_per_second": 349.699
    },
    {
      "epoch": 0.029885057471264367,
      "grad_norm": 6.470882415771484,
      "learning_rate": 1.680200501253133e-05,
      "loss": 0.4838469922542572,
      "num_input_tokens_seen": 252848,
      "step": 325,
      "train_runtime": 722.555,
      "train_tokens_per_second": 349.936
    },
    {
      "epoch": 0.029977011494252873,
      "grad_norm": 7.9414286613464355,
      "learning_rate": 1.6791979949874687e-05,
      "loss": 0.5206264853477478,
      "num_input_tokens_seen": 253912,
      "step": 326,
      "train_runtime": 724.5384,
      "train_tokens_per_second": 350.447
    },
    {
      "epoch": 0.030068965517241378,
      "grad_norm": 6.74638032913208,
      "learning_rate": 1.6781954887218047e-05,
      "loss": 0.48208141326904297,
      "num_input_tokens_seen": 254576,
      "step": 327,
      "train_runtime": 726.4314,
      "train_tokens_per_second": 350.447
    },
    {
      "epoch": 0.030160919540229886,
      "grad_norm": 6.492095470428467,
      "learning_rate": 1.6771929824561408e-05,
      "loss": 0.5265463590621948,
      "num_input_tokens_seen": 255440,
      "step": 328,
      "train_runtime": 728.3501,
      "train_tokens_per_second": 350.71
    },
    {
      "epoch": 0.03025287356321839,
      "grad_norm": 7.264855861663818,
      "learning_rate": 1.6761904761904764e-05,
      "loss": 0.5523508191108704,
      "num_input_tokens_seen": 256132,
      "step": 329,
      "train_runtime": 730.2381,
      "train_tokens_per_second": 350.751
    },
    {
      "epoch": 0.030344827586206897,
      "grad_norm": 7.530397891998291,
      "learning_rate": 1.675187969924812e-05,
      "loss": 0.524081826210022,
      "num_input_tokens_seen": 256948,
      "step": 330,
      "train_runtime": 732.1609,
      "train_tokens_per_second": 350.945
    },
    {
      "epoch": 0.030436781609195402,
      "grad_norm": 6.863246440887451,
      "learning_rate": 1.6741854636591478e-05,
      "loss": 0.4773816168308258,
      "num_input_tokens_seen": 257584,
      "step": 331,
      "train_runtime": 740.3818,
      "train_tokens_per_second": 347.907
    },
    {
      "epoch": 0.030528735632183907,
      "grad_norm": 7.0147294998168945,
      "learning_rate": 1.6731829573934838e-05,
      "loss": 0.6112295389175415,
      "num_input_tokens_seen": 258228,
      "step": 332,
      "train_runtime": 742.3362,
      "train_tokens_per_second": 347.859
    },
    {
      "epoch": 0.030620689655172412,
      "grad_norm": 8.865488052368164,
      "learning_rate": 1.67218045112782e-05,
      "loss": 0.48941272497177124,
      "num_input_tokens_seen": 259052,
      "step": 333,
      "train_runtime": 744.4107,
      "train_tokens_per_second": 347.996
    },
    {
      "epoch": 0.03071264367816092,
      "grad_norm": 6.03111457824707,
      "learning_rate": 1.6711779448621555e-05,
      "loss": 0.4099096953868866,
      "num_input_tokens_seen": 259704,
      "step": 334,
      "train_runtime": 746.3563,
      "train_tokens_per_second": 347.963
    },
    {
      "epoch": 0.030804597701149426,
      "grad_norm": 7.29164457321167,
      "learning_rate": 1.6701754385964915e-05,
      "loss": 0.5145337581634521,
      "num_input_tokens_seen": 260576,
      "step": 335,
      "train_runtime": 748.3101,
      "train_tokens_per_second": 348.219
    },
    {
      "epoch": 0.03089655172413793,
      "grad_norm": 7.4277777671813965,
      "learning_rate": 1.6691729323308272e-05,
      "loss": 0.49768251180648804,
      "num_input_tokens_seen": 261320,
      "step": 336,
      "train_runtime": 750.2052,
      "train_tokens_per_second": 348.331
    },
    {
      "epoch": 0.030988505747126437,
      "grad_norm": 6.960900783538818,
      "learning_rate": 1.668170426065163e-05,
      "loss": 0.4325927495956421,
      "num_input_tokens_seen": 261948,
      "step": 337,
      "train_runtime": 752.1629,
      "train_tokens_per_second": 348.26
    },
    {
      "epoch": 0.031080459770114942,
      "grad_norm": 6.691569805145264,
      "learning_rate": 1.667167919799499e-05,
      "loss": 0.4351831376552582,
      "num_input_tokens_seen": 262576,
      "step": 338,
      "train_runtime": 754.0653,
      "train_tokens_per_second": 348.214
    },
    {
      "epoch": 0.031172413793103447,
      "grad_norm": 6.913660049438477,
      "learning_rate": 1.666165413533835e-05,
      "loss": 0.5391501188278198,
      "num_input_tokens_seen": 263340,
      "step": 339,
      "train_runtime": 755.9776,
      "train_tokens_per_second": 348.344
    },
    {
      "epoch": 0.03126436781609195,
      "grad_norm": 6.391702651977539,
      "learning_rate": 1.6651629072681706e-05,
      "loss": 0.49873214960098267,
      "num_input_tokens_seen": 264336,
      "step": 340,
      "train_runtime": 757.913,
      "train_tokens_per_second": 348.768
    },
    {
      "epoch": 0.03135632183908046,
      "grad_norm": 7.701982498168945,
      "learning_rate": 1.6641604010025063e-05,
      "loss": 0.4868852198123932,
      "num_input_tokens_seen": 264988,
      "step": 341,
      "train_runtime": 759.803,
      "train_tokens_per_second": 348.759
    },
    {
      "epoch": 0.03144827586206896,
      "grad_norm": 7.031266689300537,
      "learning_rate": 1.6631578947368423e-05,
      "loss": 0.502269446849823,
      "num_input_tokens_seen": 265580,
      "step": 342,
      "train_runtime": 761.7374,
      "train_tokens_per_second": 348.65
    },
    {
      "epoch": 0.031540229885057475,
      "grad_norm": 6.966207027435303,
      "learning_rate": 1.662155388471178e-05,
      "loss": 0.4639718234539032,
      "num_input_tokens_seen": 266412,
      "step": 343,
      "train_runtime": 763.6914,
      "train_tokens_per_second": 348.848
    },
    {
      "epoch": 0.03163218390804598,
      "grad_norm": 7.524121284484863,
      "learning_rate": 1.661152882205514e-05,
      "loss": 0.5666488409042358,
      "num_input_tokens_seen": 267528,
      "step": 344,
      "train_runtime": 765.7296,
      "train_tokens_per_second": 349.377
    },
    {
      "epoch": 0.031724137931034485,
      "grad_norm": 7.829931259155273,
      "learning_rate": 1.6601503759398497e-05,
      "loss": 0.5500428676605225,
      "num_input_tokens_seen": 268236,
      "step": 345,
      "train_runtime": 767.6511,
      "train_tokens_per_second": 349.424
    },
    {
      "epoch": 0.03181609195402299,
      "grad_norm": 7.847830295562744,
      "learning_rate": 1.6591478696741857e-05,
      "loss": 0.5698979496955872,
      "num_input_tokens_seen": 268996,
      "step": 346,
      "train_runtime": 769.541,
      "train_tokens_per_second": 349.554
    },
    {
      "epoch": 0.031908045977011495,
      "grad_norm": 6.956431865692139,
      "learning_rate": 1.6581453634085214e-05,
      "loss": 0.46126455068588257,
      "num_input_tokens_seen": 269616,
      "step": 347,
      "train_runtime": 771.4533,
      "train_tokens_per_second": 349.491
    },
    {
      "epoch": 0.032,
      "grad_norm": 7.762545585632324,
      "learning_rate": 1.6571428571428574e-05,
      "loss": 0.5739911198616028,
      "num_input_tokens_seen": 270672,
      "step": 348,
      "train_runtime": 773.4333,
      "train_tokens_per_second": 349.962
    },
    {
      "epoch": 0.032091954022988506,
      "grad_norm": 7.1435699462890625,
      "learning_rate": 1.656140350877193e-05,
      "loss": 0.4908761978149414,
      "num_input_tokens_seen": 271392,
      "step": 349,
      "train_runtime": 775.3249,
      "train_tokens_per_second": 350.036
    },
    {
      "epoch": 0.03218390804597701,
      "grad_norm": 7.2212724685668945,
      "learning_rate": 1.655137844611529e-05,
      "loss": 0.4466068148612976,
      "num_input_tokens_seen": 271944,
      "step": 350,
      "train_runtime": 777.2137,
      "train_tokens_per_second": 349.896
    },
    {
      "epoch": 0.032275862068965516,
      "grad_norm": 7.709162712097168,
      "learning_rate": 1.6541353383458648e-05,
      "loss": 0.4983013868331909,
      "num_input_tokens_seen": 272548,
      "step": 351,
      "train_runtime": 779.1033,
      "train_tokens_per_second": 349.823
    },
    {
      "epoch": 0.03236781609195402,
      "grad_norm": 7.1974005699157715,
      "learning_rate": 1.6531328320802008e-05,
      "loss": 0.5227024555206299,
      "num_input_tokens_seen": 273388,
      "step": 352,
      "train_runtime": 781.0284,
      "train_tokens_per_second": 350.036
    },
    {
      "epoch": 0.03245977011494253,
      "grad_norm": 7.129546642303467,
      "learning_rate": 1.6521303258145365e-05,
      "loss": 0.5885041952133179,
      "num_input_tokens_seen": 274252,
      "step": 353,
      "train_runtime": 782.9604,
      "train_tokens_per_second": 350.276
    },
    {
      "epoch": 0.03255172413793103,
      "grad_norm": 7.138310432434082,
      "learning_rate": 1.651127819548872e-05,
      "loss": 0.6486871242523193,
      "num_input_tokens_seen": 275204,
      "step": 354,
      "train_runtime": 784.895,
      "train_tokens_per_second": 350.625
    },
    {
      "epoch": 0.03264367816091954,
      "grad_norm": 7.11072301864624,
      "learning_rate": 1.650125313283208e-05,
      "loss": 0.47041159868240356,
      "num_input_tokens_seen": 275812,
      "step": 355,
      "train_runtime": 786.7857,
      "train_tokens_per_second": 350.555
    },
    {
      "epoch": 0.03273563218390805,
      "grad_norm": 8.056539535522461,
      "learning_rate": 1.649122807017544e-05,
      "loss": 0.5420215725898743,
      "num_input_tokens_seen": 276460,
      "step": 356,
      "train_runtime": 788.6773,
      "train_tokens_per_second": 350.536
    },
    {
      "epoch": 0.032827586206896554,
      "grad_norm": 12.881464004516602,
      "learning_rate": 1.64812030075188e-05,
      "loss": 0.6470522880554199,
      "num_input_tokens_seen": 277332,
      "step": 357,
      "train_runtime": 790.6098,
      "train_tokens_per_second": 350.782
    },
    {
      "epoch": 0.03291954022988506,
      "grad_norm": 8.421309471130371,
      "learning_rate": 1.647117794486216e-05,
      "loss": 0.5130929946899414,
      "num_input_tokens_seen": 278088,
      "step": 358,
      "train_runtime": 792.5274,
      "train_tokens_per_second": 350.888
    },
    {
      "epoch": 0.033011494252873565,
      "grad_norm": 5.847091197967529,
      "learning_rate": 1.6461152882205516e-05,
      "loss": 0.39224687218666077,
      "num_input_tokens_seen": 278716,
      "step": 359,
      "train_runtime": 794.4104,
      "train_tokens_per_second": 350.846
    },
    {
      "epoch": 0.03310344827586207,
      "grad_norm": 7.019738674163818,
      "learning_rate": 1.6451127819548872e-05,
      "loss": 0.5132017135620117,
      "num_input_tokens_seen": 279480,
      "step": 360,
      "train_runtime": 796.319,
      "train_tokens_per_second": 350.965
    },
    {
      "epoch": 0.033195402298850575,
      "grad_norm": 6.658227443695068,
      "learning_rate": 1.6441102756892233e-05,
      "loss": 0.4876769781112671,
      "num_input_tokens_seen": 280152,
      "step": 361,
      "train_runtime": 804.5307,
      "train_tokens_per_second": 348.218
    },
    {
      "epoch": 0.03328735632183908,
      "grad_norm": 6.251015663146973,
      "learning_rate": 1.643107769423559e-05,
      "loss": 0.5464555621147156,
      "num_input_tokens_seen": 281624,
      "step": 362,
      "train_runtime": 806.5943,
      "train_tokens_per_second": 349.152
    },
    {
      "epoch": 0.033379310344827585,
      "grad_norm": 5.939515113830566,
      "learning_rate": 1.642105263157895e-05,
      "loss": 0.4470316767692566,
      "num_input_tokens_seen": 282292,
      "step": 363,
      "train_runtime": 808.5787,
      "train_tokens_per_second": 349.121
    },
    {
      "epoch": 0.03347126436781609,
      "grad_norm": 6.443390369415283,
      "learning_rate": 1.6411027568922306e-05,
      "loss": 0.388321191072464,
      "num_input_tokens_seen": 282944,
      "step": 364,
      "train_runtime": 810.4796,
      "train_tokens_per_second": 349.107
    },
    {
      "epoch": 0.033563218390804596,
      "grad_norm": 6.1911492347717285,
      "learning_rate": 1.6401002506265667e-05,
      "loss": 0.5065643191337585,
      "num_input_tokens_seen": 283636,
      "step": 365,
      "train_runtime": 812.4084,
      "train_tokens_per_second": 349.13
    },
    {
      "epoch": 0.0336551724137931,
      "grad_norm": 7.345110893249512,
      "learning_rate": 1.6390977443609023e-05,
      "loss": 0.5557699799537659,
      "num_input_tokens_seen": 284292,
      "step": 366,
      "train_runtime": 814.2967,
      "train_tokens_per_second": 349.126
    },
    {
      "epoch": 0.033747126436781606,
      "grad_norm": 6.9500226974487305,
      "learning_rate": 1.6380952380952384e-05,
      "loss": 0.5253832340240479,
      "num_input_tokens_seen": 284828,
      "step": 367,
      "train_runtime": 816.1715,
      "train_tokens_per_second": 348.981
    },
    {
      "epoch": 0.03383908045977012,
      "grad_norm": 7.6618523597717285,
      "learning_rate": 1.637092731829574e-05,
      "loss": 0.5470195412635803,
      "num_input_tokens_seen": 285500,
      "step": 368,
      "train_runtime": 818.051,
      "train_tokens_per_second": 349.0
    },
    {
      "epoch": 0.033931034482758624,
      "grad_norm": 7.579816818237305,
      "learning_rate": 1.63609022556391e-05,
      "loss": 0.5715543031692505,
      "num_input_tokens_seen": 286120,
      "step": 369,
      "train_runtime": 819.9256,
      "train_tokens_per_second": 348.958
    },
    {
      "epoch": 0.03402298850574713,
      "grad_norm": 7.251576900482178,
      "learning_rate": 1.6350877192982457e-05,
      "loss": 0.41737809777259827,
      "num_input_tokens_seen": 286676,
      "step": 370,
      "train_runtime": 821.7969,
      "train_tokens_per_second": 348.84
    },
    {
      "epoch": 0.034114942528735634,
      "grad_norm": 7.673832416534424,
      "learning_rate": 1.6340852130325814e-05,
      "loss": 0.4248308837413788,
      "num_input_tokens_seen": 287340,
      "step": 371,
      "train_runtime": 823.6856,
      "train_tokens_per_second": 348.847
    },
    {
      "epoch": 0.03420689655172414,
      "grad_norm": 7.723014831542969,
      "learning_rate": 1.6330827067669174e-05,
      "loss": 0.5945442914962769,
      "num_input_tokens_seen": 288192,
      "step": 372,
      "train_runtime": 825.6152,
      "train_tokens_per_second": 349.063
    },
    {
      "epoch": 0.034298850574712644,
      "grad_norm": 7.674216270446777,
      "learning_rate": 1.632080200501253e-05,
      "loss": 0.518921434879303,
      "num_input_tokens_seen": 288844,
      "step": 373,
      "train_runtime": 827.517,
      "train_tokens_per_second": 349.049
    },
    {
      "epoch": 0.03439080459770115,
      "grad_norm": 7.32497501373291,
      "learning_rate": 1.631077694235589e-05,
      "loss": 0.6034227609634399,
      "num_input_tokens_seen": 290068,
      "step": 374,
      "train_runtime": 829.4698,
      "train_tokens_per_second": 349.703
    },
    {
      "epoch": 0.034482758620689655,
      "grad_norm": 8.127471923828125,
      "learning_rate": 1.630075187969925e-05,
      "loss": 0.5146776437759399,
      "num_input_tokens_seen": 291048,
      "step": 375,
      "train_runtime": 831.3926,
      "train_tokens_per_second": 350.073
    },
    {
      "epoch": 0.03457471264367816,
      "grad_norm": 6.988226890563965,
      "learning_rate": 1.6290726817042608e-05,
      "loss": 0.4996544122695923,
      "num_input_tokens_seen": 291704,
      "step": 376,
      "train_runtime": 833.3642,
      "train_tokens_per_second": 350.032
    },
    {
      "epoch": 0.034666666666666665,
      "grad_norm": 6.708095550537109,
      "learning_rate": 1.6280701754385965e-05,
      "loss": 0.48568931221961975,
      "num_input_tokens_seen": 292804,
      "step": 377,
      "train_runtime": 835.3228,
      "train_tokens_per_second": 350.528
    },
    {
      "epoch": 0.03475862068965517,
      "grad_norm": 7.753971099853516,
      "learning_rate": 1.6270676691729325e-05,
      "loss": 0.5069139003753662,
      "num_input_tokens_seen": 293420,
      "step": 378,
      "train_runtime": 837.2175,
      "train_tokens_per_second": 350.47
    },
    {
      "epoch": 0.034850574712643675,
      "grad_norm": 6.488057613372803,
      "learning_rate": 1.6260651629072682e-05,
      "loss": 0.47667884826660156,
      "num_input_tokens_seen": 294024,
      "step": 379,
      "train_runtime": 839.1322,
      "train_tokens_per_second": 350.391
    },
    {
      "epoch": 0.03494252873563218,
      "grad_norm": 6.502963066101074,
      "learning_rate": 1.6250626566416042e-05,
      "loss": 0.48642775416374207,
      "num_input_tokens_seen": 294936,
      "step": 380,
      "train_runtime": 841.0563,
      "train_tokens_per_second": 350.673
    },
    {
      "epoch": 0.03503448275862069,
      "grad_norm": 7.4670562744140625,
      "learning_rate": 1.62406015037594e-05,
      "loss": 0.46354836225509644,
      "num_input_tokens_seen": 295632,
      "step": 381,
      "train_runtime": 842.9546,
      "train_tokens_per_second": 350.709
    },
    {
      "epoch": 0.0351264367816092,
      "grad_norm": 6.839841842651367,
      "learning_rate": 1.623057644110276e-05,
      "loss": 0.4796101450920105,
      "num_input_tokens_seen": 296268,
      "step": 382,
      "train_runtime": 844.8762,
      "train_tokens_per_second": 350.664
    },
    {
      "epoch": 0.0352183908045977,
      "grad_norm": 7.7321577072143555,
      "learning_rate": 1.6220551378446116e-05,
      "loss": 0.6368752717971802,
      "num_input_tokens_seen": 297088,
      "step": 383,
      "train_runtime": 846.7974,
      "train_tokens_per_second": 350.837
    },
    {
      "epoch": 0.03531034482758621,
      "grad_norm": 6.988363265991211,
      "learning_rate": 1.6210526315789473e-05,
      "loss": 0.5102667808532715,
      "num_input_tokens_seen": 297756,
      "step": 384,
      "train_runtime": 848.6868,
      "train_tokens_per_second": 350.843
    },
    {
      "epoch": 0.035402298850574714,
      "grad_norm": 5.6896586418151855,
      "learning_rate": 1.6200501253132833e-05,
      "loss": 0.4484521150588989,
      "num_input_tokens_seen": 298500,
      "step": 385,
      "train_runtime": 850.6177,
      "train_tokens_per_second": 350.921
    },
    {
      "epoch": 0.03549425287356322,
      "grad_norm": 6.464532852172852,
      "learning_rate": 1.6190476190476193e-05,
      "loss": 0.5552738904953003,
      "num_input_tokens_seen": 299140,
      "step": 386,
      "train_runtime": 852.5017,
      "train_tokens_per_second": 350.897
    },
    {
      "epoch": 0.035586206896551724,
      "grad_norm": 7.861773490905762,
      "learning_rate": 1.618045112781955e-05,
      "loss": 0.5233801007270813,
      "num_input_tokens_seen": 299820,
      "step": 387,
      "train_runtime": 854.3992,
      "train_tokens_per_second": 350.913
    },
    {
      "epoch": 0.03567816091954023,
      "grad_norm": 6.787221431732178,
      "learning_rate": 1.617042606516291e-05,
      "loss": 0.551461935043335,
      "num_input_tokens_seen": 300432,
      "step": 388,
      "train_runtime": 856.2855,
      "train_tokens_per_second": 350.855
    },
    {
      "epoch": 0.035770114942528734,
      "grad_norm": 6.840238094329834,
      "learning_rate": 1.6160401002506267e-05,
      "loss": 0.4511426091194153,
      "num_input_tokens_seen": 301136,
      "step": 389,
      "train_runtime": 858.1776,
      "train_tokens_per_second": 350.902
    },
    {
      "epoch": 0.03586206896551724,
      "grad_norm": 6.89091157913208,
      "learning_rate": 1.6150375939849624e-05,
      "loss": 0.49777084589004517,
      "num_input_tokens_seen": 301856,
      "step": 390,
      "train_runtime": 860.0736,
      "train_tokens_per_second": 350.965
    },
    {
      "epoch": 0.035954022988505745,
      "grad_norm": 7.600098133087158,
      "learning_rate": 1.6140350877192984e-05,
      "loss": 0.5933030247688293,
      "num_input_tokens_seen": 302624,
      "step": 391,
      "train_runtime": 868.2261,
      "train_tokens_per_second": 348.554
    },
    {
      "epoch": 0.03604597701149425,
      "grad_norm": 7.551762104034424,
      "learning_rate": 1.6130325814536344e-05,
      "loss": 0.46076130867004395,
      "num_input_tokens_seen": 303352,
      "step": 392,
      "train_runtime": 870.1267,
      "train_tokens_per_second": 348.63
    },
    {
      "epoch": 0.03613793103448276,
      "grad_norm": 6.803619861602783,
      "learning_rate": 1.61203007518797e-05,
      "loss": 0.4574781358242035,
      "num_input_tokens_seen": 303964,
      "step": 393,
      "train_runtime": 872.2061,
      "train_tokens_per_second": 348.5
    },
    {
      "epoch": 0.03622988505747127,
      "grad_norm": 6.719482898712158,
      "learning_rate": 1.6110275689223058e-05,
      "loss": 0.5644866228103638,
      "num_input_tokens_seen": 304624,
      "step": 394,
      "train_runtime": 874.0997,
      "train_tokens_per_second": 348.5
    },
    {
      "epoch": 0.03632183908045977,
      "grad_norm": 8.121594429016113,
      "learning_rate": 1.6100250626566418e-05,
      "loss": 0.4595955014228821,
      "num_input_tokens_seen": 305188,
      "step": 395,
      "train_runtime": 875.9938,
      "train_tokens_per_second": 348.391
    },
    {
      "epoch": 0.03641379310344828,
      "grad_norm": 6.627795219421387,
      "learning_rate": 1.6090225563909775e-05,
      "loss": 0.47989457845687866,
      "num_input_tokens_seen": 305804,
      "step": 396,
      "train_runtime": 877.891,
      "train_tokens_per_second": 348.339
    },
    {
      "epoch": 0.03650574712643678,
      "grad_norm": 7.286935806274414,
      "learning_rate": 1.6080200501253135e-05,
      "loss": 0.6367931365966797,
      "num_input_tokens_seen": 307136,
      "step": 397,
      "train_runtime": 879.8924,
      "train_tokens_per_second": 349.061
    },
    {
      "epoch": 0.03659770114942529,
      "grad_norm": 7.091745853424072,
      "learning_rate": 1.6070175438596495e-05,
      "loss": 0.5370696783065796,
      "num_input_tokens_seen": 308344,
      "step": 398,
      "train_runtime": 881.9306,
      "train_tokens_per_second": 349.624
    },
    {
      "epoch": 0.03668965517241379,
      "grad_norm": 6.8993377685546875,
      "learning_rate": 1.6060150375939852e-05,
      "loss": 0.47272491455078125,
      "num_input_tokens_seen": 309188,
      "step": 399,
      "train_runtime": 883.8732,
      "train_tokens_per_second": 349.81
    },
    {
      "epoch": 0.0367816091954023,
      "grad_norm": 9.172377586364746,
      "learning_rate": 1.605012531328321e-05,
      "loss": 0.6651483774185181,
      "num_input_tokens_seen": 309892,
      "step": 400,
      "train_runtime": 885.7652,
      "train_tokens_per_second": 349.858
    },
    {
      "epoch": 0.0367816091954023,
      "eval_loss": 1.9598596096038818,
      "eval_runtime": 27.5975,
      "eval_samples_per_second": 36.235,
      "eval_steps_per_second": 9.059,
      "num_input_tokens_seen": 309892,
      "step": 400
    },
    {
      "epoch": 0.036873563218390804,
      "grad_norm": 7.518550872802734,
      "learning_rate": 1.6040100250626565e-05,
      "loss": 0.45641279220581055,
      "num_input_tokens_seen": 310892,
      "step": 401,
      "train_runtime": 915.3197,
      "train_tokens_per_second": 339.654
    },
    {
      "epoch": 0.03696551724137931,
      "grad_norm": 6.781665325164795,
      "learning_rate": 1.6030075187969926e-05,
      "loss": 0.4673669636249542,
      "num_input_tokens_seen": 311500,
      "step": 402,
      "train_runtime": 917.2008,
      "train_tokens_per_second": 339.62
    },
    {
      "epoch": 0.037057471264367814,
      "grad_norm": 6.372986316680908,
      "learning_rate": 1.6020050125313286e-05,
      "loss": 0.5332642793655396,
      "num_input_tokens_seen": 312148,
      "step": 403,
      "train_runtime": 919.0813,
      "train_tokens_per_second": 339.63
    },
    {
      "epoch": 0.03714942528735632,
      "grad_norm": 6.82274866104126,
      "learning_rate": 1.6010025062656642e-05,
      "loss": 0.5302132368087769,
      "num_input_tokens_seen": 312844,
      "step": 404,
      "train_runtime": 920.9834,
      "train_tokens_per_second": 339.685
    },
    {
      "epoch": 0.037241379310344824,
      "grad_norm": 6.736145496368408,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.39181384444236755,
      "num_input_tokens_seen": 313408,
      "step": 405,
      "train_runtime": 922.8715,
      "train_tokens_per_second": 339.601
    },
    {
      "epoch": 0.037333333333333336,
      "grad_norm": 6.29167366027832,
      "learning_rate": 1.598997493734336e-05,
      "loss": 0.4519786834716797,
      "num_input_tokens_seen": 314056,
      "step": 406,
      "train_runtime": 924.7587,
      "train_tokens_per_second": 339.609
    },
    {
      "epoch": 0.03742528735632184,
      "grad_norm": 6.830358028411865,
      "learning_rate": 1.5979949874686716e-05,
      "loss": 0.5322299003601074,
      "num_input_tokens_seen": 314756,
      "step": 407,
      "train_runtime": 926.6487,
      "train_tokens_per_second": 339.671
    },
    {
      "epoch": 0.03751724137931035,
      "grad_norm": 6.957012176513672,
      "learning_rate": 1.5969924812030076e-05,
      "loss": 0.5151546001434326,
      "num_input_tokens_seen": 315372,
      "step": 408,
      "train_runtime": 928.5232,
      "train_tokens_per_second": 339.649
    },
    {
      "epoch": 0.03760919540229885,
      "grad_norm": 7.0232930183410645,
      "learning_rate": 1.5959899749373437e-05,
      "loss": 0.5350349545478821,
      "num_input_tokens_seen": 316036,
      "step": 409,
      "train_runtime": 930.4045,
      "train_tokens_per_second": 339.676
    },
    {
      "epoch": 0.03770114942528736,
      "grad_norm": 6.103466033935547,
      "learning_rate": 1.5949874686716793e-05,
      "loss": 0.47409480810165405,
      "num_input_tokens_seen": 316632,
      "step": 410,
      "train_runtime": 932.2832,
      "train_tokens_per_second": 339.631
    },
    {
      "epoch": 0.03779310344827586,
      "grad_norm": 7.183594226837158,
      "learning_rate": 1.593984962406015e-05,
      "loss": 0.5453521013259888,
      "num_input_tokens_seen": 317320,
      "step": 411,
      "train_runtime": 934.1647,
      "train_tokens_per_second": 339.683
    },
    {
      "epoch": 0.03788505747126437,
      "grad_norm": 7.0533447265625,
      "learning_rate": 1.592982456140351e-05,
      "loss": 0.5170130729675293,
      "num_input_tokens_seen": 317928,
      "step": 412,
      "train_runtime": 936.0357,
      "train_tokens_per_second": 339.654
    },
    {
      "epoch": 0.03797701149425287,
      "grad_norm": 6.611300945281982,
      "learning_rate": 1.5919799498746867e-05,
      "loss": 0.3984178900718689,
      "num_input_tokens_seen": 318452,
      "step": 413,
      "train_runtime": 937.9105,
      "train_tokens_per_second": 339.533
    },
    {
      "epoch": 0.03806896551724138,
      "grad_norm": 7.702702045440674,
      "learning_rate": 1.5909774436090227e-05,
      "loss": 0.5222338438034058,
      "num_input_tokens_seen": 319132,
      "step": 414,
      "train_runtime": 939.7931,
      "train_tokens_per_second": 339.577
    },
    {
      "epoch": 0.03816091954022988,
      "grad_norm": 6.963353157043457,
      "learning_rate": 1.5899749373433588e-05,
      "loss": 0.5179736614227295,
      "num_input_tokens_seen": 319940,
      "step": 415,
      "train_runtime": 941.7035,
      "train_tokens_per_second": 339.746
    },
    {
      "epoch": 0.03825287356321839,
      "grad_norm": 6.797804832458496,
      "learning_rate": 1.5889724310776944e-05,
      "loss": 0.5295509099960327,
      "num_input_tokens_seen": 320552,
      "step": 416,
      "train_runtime": 943.5741,
      "train_tokens_per_second": 339.721
    },
    {
      "epoch": 0.038344827586206894,
      "grad_norm": 7.196625709533691,
      "learning_rate": 1.58796992481203e-05,
      "loss": 0.5286198854446411,
      "num_input_tokens_seen": 321352,
      "step": 417,
      "train_runtime": 945.4807,
      "train_tokens_per_second": 339.882
    },
    {
      "epoch": 0.038436781609195406,
      "grad_norm": 7.270211696624756,
      "learning_rate": 1.5869674185463658e-05,
      "loss": 0.4256896376609802,
      "num_input_tokens_seen": 322080,
      "step": 418,
      "train_runtime": 947.3712,
      "train_tokens_per_second": 339.972
    },
    {
      "epoch": 0.03852873563218391,
      "grad_norm": 7.343602657318115,
      "learning_rate": 1.5859649122807018e-05,
      "loss": 0.4776989817619324,
      "num_input_tokens_seen": 322700,
      "step": 419,
      "train_runtime": 949.2596,
      "train_tokens_per_second": 339.949
    },
    {
      "epoch": 0.038620689655172416,
      "grad_norm": 6.887912750244141,
      "learning_rate": 1.5849624060150378e-05,
      "loss": 0.45532381534576416,
      "num_input_tokens_seen": 323328,
      "step": 420,
      "train_runtime": 951.1473,
      "train_tokens_per_second": 339.935
    },
    {
      "epoch": 0.03871264367816092,
      "grad_norm": 6.504075050354004,
      "learning_rate": 1.5839598997493735e-05,
      "loss": 0.41847917437553406,
      "num_input_tokens_seen": 324092,
      "step": 421,
      "train_runtime": 959.2486,
      "train_tokens_per_second": 337.86
    },
    {
      "epoch": 0.038804597701149426,
      "grad_norm": 6.65957498550415,
      "learning_rate": 1.5829573934837095e-05,
      "loss": 0.4958955645561218,
      "num_input_tokens_seen": 324604,
      "step": 422,
      "train_runtime": 961.2212,
      "train_tokens_per_second": 337.7
    },
    {
      "epoch": 0.03889655172413793,
      "grad_norm": 6.565097808837891,
      "learning_rate": 1.5819548872180452e-05,
      "loss": 0.42420342564582825,
      "num_input_tokens_seen": 325232,
      "step": 423,
      "train_runtime": 963.1153,
      "train_tokens_per_second": 337.688
    },
    {
      "epoch": 0.03898850574712644,
      "grad_norm": 7.877225399017334,
      "learning_rate": 1.580952380952381e-05,
      "loss": 0.6448503732681274,
      "num_input_tokens_seen": 326136,
      "step": 424,
      "train_runtime": 965.0444,
      "train_tokens_per_second": 337.949
    },
    {
      "epoch": 0.03908045977011494,
      "grad_norm": 6.912954330444336,
      "learning_rate": 1.579949874686717e-05,
      "loss": 0.5450248122215271,
      "num_input_tokens_seen": 326740,
      "step": 425,
      "train_runtime": 966.9381,
      "train_tokens_per_second": 337.912
    },
    {
      "epoch": 0.03917241379310345,
      "grad_norm": 7.217955112457275,
      "learning_rate": 1.578947368421053e-05,
      "loss": 0.46079760789871216,
      "num_input_tokens_seen": 327380,
      "step": 426,
      "train_runtime": 968.8266,
      "train_tokens_per_second": 337.914
    },
    {
      "epoch": 0.03926436781609195,
      "grad_norm": 8.541083335876465,
      "learning_rate": 1.5779448621553886e-05,
      "loss": 0.5666900277137756,
      "num_input_tokens_seen": 328380,
      "step": 427,
      "train_runtime": 970.7469,
      "train_tokens_per_second": 338.276
    },
    {
      "epoch": 0.03935632183908046,
      "grad_norm": 7.42299222946167,
      "learning_rate": 1.5769423558897243e-05,
      "loss": 0.5644451379776001,
      "num_input_tokens_seen": 329068,
      "step": 428,
      "train_runtime": 972.6311,
      "train_tokens_per_second": 338.328
    },
    {
      "epoch": 0.03944827586206896,
      "grad_norm": 6.829087257385254,
      "learning_rate": 1.5759398496240603e-05,
      "loss": 0.46858036518096924,
      "num_input_tokens_seen": 329672,
      "step": 429,
      "train_runtime": 974.531,
      "train_tokens_per_second": 338.288
    },
    {
      "epoch": 0.03954022988505747,
      "grad_norm": 7.258794784545898,
      "learning_rate": 1.574937343358396e-05,
      "loss": 0.4189729690551758,
      "num_input_tokens_seen": 330316,
      "step": 430,
      "train_runtime": 976.4375,
      "train_tokens_per_second": 338.287
    },
    {
      "epoch": 0.03963218390804598,
      "grad_norm": 6.7284064292907715,
      "learning_rate": 1.573934837092732e-05,
      "loss": 0.4453396499156952,
      "num_input_tokens_seen": 330940,
      "step": 431,
      "train_runtime": 978.3328,
      "train_tokens_per_second": 338.269
    },
    {
      "epoch": 0.039724137931034485,
      "grad_norm": 7.023474216461182,
      "learning_rate": 1.572932330827068e-05,
      "loss": 0.5812249779701233,
      "num_input_tokens_seen": 331580,
      "step": 432,
      "train_runtime": 980.2252,
      "train_tokens_per_second": 338.269
    },
    {
      "epoch": 0.03981609195402299,
      "grad_norm": 6.450960159301758,
      "learning_rate": 1.5719298245614037e-05,
      "loss": 0.5113633871078491,
      "num_input_tokens_seen": 332208,
      "step": 433,
      "train_runtime": 982.122,
      "train_tokens_per_second": 338.255
    },
    {
      "epoch": 0.039908045977011496,
      "grad_norm": 6.986124038696289,
      "learning_rate": 1.5709273182957394e-05,
      "loss": 0.4958512783050537,
      "num_input_tokens_seen": 332816,
      "step": 434,
      "train_runtime": 984.0321,
      "train_tokens_per_second": 338.217
    },
    {
      "epoch": 0.04,
      "grad_norm": 6.635585308074951,
      "learning_rate": 1.5699248120300754e-05,
      "loss": 0.5527627468109131,
      "num_input_tokens_seen": 333972,
      "step": 435,
      "train_runtime": 986.0673,
      "train_tokens_per_second": 338.691
    },
    {
      "epoch": 0.040091954022988506,
      "grad_norm": 7.397719383239746,
      "learning_rate": 1.568922305764411e-05,
      "loss": 0.5977311730384827,
      "num_input_tokens_seen": 334836,
      "step": 436,
      "train_runtime": 988.0303,
      "train_tokens_per_second": 338.892
    },
    {
      "epoch": 0.04018390804597701,
      "grad_norm": 7.226812362670898,
      "learning_rate": 1.567919799498747e-05,
      "loss": 0.5547776222229004,
      "num_input_tokens_seen": 335580,
      "step": 437,
      "train_runtime": 989.9561,
      "train_tokens_per_second": 338.985
    },
    {
      "epoch": 0.040275862068965516,
      "grad_norm": 7.948246002197266,
      "learning_rate": 1.5669172932330828e-05,
      "loss": 0.571733295917511,
      "num_input_tokens_seen": 336312,
      "step": 438,
      "train_runtime": 991.8528,
      "train_tokens_per_second": 339.075
    },
    {
      "epoch": 0.04036781609195402,
      "grad_norm": 8.043785095214844,
      "learning_rate": 1.5659147869674188e-05,
      "loss": 0.5533593893051147,
      "num_input_tokens_seen": 337140,
      "step": 439,
      "train_runtime": 993.8152,
      "train_tokens_per_second": 339.238
    },
    {
      "epoch": 0.04045977011494253,
      "grad_norm": 6.651862144470215,
      "learning_rate": 1.5649122807017545e-05,
      "loss": 0.490252286195755,
      "num_input_tokens_seen": 337860,
      "step": 440,
      "train_runtime": 995.7001,
      "train_tokens_per_second": 339.319
    },
    {
      "epoch": 0.04055172413793103,
      "grad_norm": 6.687456130981445,
      "learning_rate": 1.56390977443609e-05,
      "loss": 0.5104302167892456,
      "num_input_tokens_seen": 338488,
      "step": 441,
      "train_runtime": 997.5875,
      "train_tokens_per_second": 339.307
    },
    {
      "epoch": 0.04064367816091954,
      "grad_norm": 7.131701469421387,
      "learning_rate": 1.562907268170426e-05,
      "loss": 0.47217661142349243,
      "num_input_tokens_seen": 339076,
      "step": 442,
      "train_runtime": 999.4676,
      "train_tokens_per_second": 339.257
    },
    {
      "epoch": 0.04073563218390805,
      "grad_norm": 7.258623123168945,
      "learning_rate": 1.5619047619047622e-05,
      "loss": 0.49841707944869995,
      "num_input_tokens_seen": 339704,
      "step": 443,
      "train_runtime": 1001.4076,
      "train_tokens_per_second": 339.227
    },
    {
      "epoch": 0.040827586206896554,
      "grad_norm": 6.1737961769104,
      "learning_rate": 1.560902255639098e-05,
      "loss": 0.5507537126541138,
      "num_input_tokens_seen": 341264,
      "step": 444,
      "train_runtime": 1003.4544,
      "train_tokens_per_second": 340.089
    },
    {
      "epoch": 0.04091954022988506,
      "grad_norm": 7.9596638679504395,
      "learning_rate": 1.559899749373434e-05,
      "loss": 0.543859601020813,
      "num_input_tokens_seen": 341904,
      "step": 445,
      "train_runtime": 1005.3426,
      "train_tokens_per_second": 340.087
    },
    {
      "epoch": 0.041011494252873565,
      "grad_norm": 7.282109260559082,
      "learning_rate": 1.5588972431077696e-05,
      "loss": 0.6562669277191162,
      "num_input_tokens_seen": 342880,
      "step": 446,
      "train_runtime": 1007.3023,
      "train_tokens_per_second": 340.394
    },
    {
      "epoch": 0.04110344827586207,
      "grad_norm": 6.994173526763916,
      "learning_rate": 1.5578947368421052e-05,
      "loss": 0.3848903179168701,
      "num_input_tokens_seen": 343408,
      "step": 447,
      "train_runtime": 1009.1897,
      "train_tokens_per_second": 340.281
    },
    {
      "epoch": 0.041195402298850575,
      "grad_norm": 6.728405952453613,
      "learning_rate": 1.5568922305764413e-05,
      "loss": 0.43232136964797974,
      "num_input_tokens_seen": 343996,
      "step": 448,
      "train_runtime": 1011.1096,
      "train_tokens_per_second": 340.216
    },
    {
      "epoch": 0.04128735632183908,
      "grad_norm": 7.072303295135498,
      "learning_rate": 1.5558897243107773e-05,
      "loss": 0.49546629190444946,
      "num_input_tokens_seen": 344668,
      "step": 449,
      "train_runtime": 1013.0351,
      "train_tokens_per_second": 340.233
    },
    {
      "epoch": 0.041379310344827586,
      "grad_norm": 6.667862415313721,
      "learning_rate": 1.554887218045113e-05,
      "loss": 0.5161006450653076,
      "num_input_tokens_seen": 345404,
      "step": 450,
      "train_runtime": 1014.9477,
      "train_tokens_per_second": 340.317
    },
    {
      "epoch": 0.04147126436781609,
      "grad_norm": 7.385157585144043,
      "learning_rate": 1.5538847117794486e-05,
      "loss": 0.46987441182136536,
      "num_input_tokens_seen": 346044,
      "step": 451,
      "train_runtime": 1023.0972,
      "train_tokens_per_second": 338.232
    },
    {
      "epoch": 0.041563218390804596,
      "grad_norm": 7.392758846282959,
      "learning_rate": 1.5528822055137847e-05,
      "loss": 0.571925163269043,
      "num_input_tokens_seen": 346728,
      "step": 452,
      "train_runtime": 1025.2028,
      "train_tokens_per_second": 338.204
    },
    {
      "epoch": 0.0416551724137931,
      "grad_norm": 6.2191572189331055,
      "learning_rate": 1.5518796992481203e-05,
      "loss": 0.480907678604126,
      "num_input_tokens_seen": 347488,
      "step": 453,
      "train_runtime": 1027.1362,
      "train_tokens_per_second": 338.308
    },
    {
      "epoch": 0.041747126436781606,
      "grad_norm": 7.119461536407471,
      "learning_rate": 1.5508771929824563e-05,
      "loss": 0.4267956614494324,
      "num_input_tokens_seen": 348148,
      "step": 454,
      "train_runtime": 1029.04,
      "train_tokens_per_second": 338.323
    },
    {
      "epoch": 0.04183908045977011,
      "grad_norm": 7.01524019241333,
      "learning_rate": 1.5498746867167924e-05,
      "loss": 0.4772740602493286,
      "num_input_tokens_seen": 348864,
      "step": 455,
      "train_runtime": 1030.959,
      "train_tokens_per_second": 338.388
    },
    {
      "epoch": 0.041931034482758624,
      "grad_norm": 7.350391387939453,
      "learning_rate": 1.548872180451128e-05,
      "loss": 0.5528818368911743,
      "num_input_tokens_seen": 349640,
      "step": 456,
      "train_runtime": 1032.9029,
      "train_tokens_per_second": 338.502
    },
    {
      "epoch": 0.04202298850574713,
      "grad_norm": 7.847813606262207,
      "learning_rate": 1.5478696741854637e-05,
      "loss": 0.48816895484924316,
      "num_input_tokens_seen": 350220,
      "step": 457,
      "train_runtime": 1034.8334,
      "train_tokens_per_second": 338.431
    },
    {
      "epoch": 0.042114942528735634,
      "grad_norm": 6.420209884643555,
      "learning_rate": 1.5468671679197994e-05,
      "loss": 0.3913109302520752,
      "num_input_tokens_seen": 350840,
      "step": 458,
      "train_runtime": 1036.7479,
      "train_tokens_per_second": 338.404
    },
    {
      "epoch": 0.04220689655172414,
      "grad_norm": 6.909530162811279,
      "learning_rate": 1.5458646616541354e-05,
      "loss": 0.47092413902282715,
      "num_input_tokens_seen": 351484,
      "step": 459,
      "train_runtime": 1038.6446,
      "train_tokens_per_second": 338.406
    },
    {
      "epoch": 0.042298850574712644,
      "grad_norm": 7.583240032196045,
      "learning_rate": 1.5448621553884714e-05,
      "loss": 0.5202780961990356,
      "num_input_tokens_seen": 352184,
      "step": 460,
      "train_runtime": 1040.5485,
      "train_tokens_per_second": 338.46
    },
    {
      "epoch": 0.04239080459770115,
      "grad_norm": 7.578042030334473,
      "learning_rate": 1.543859649122807e-05,
      "loss": 0.45263203978538513,
      "num_input_tokens_seen": 352828,
      "step": 461,
      "train_runtime": 1042.4493,
      "train_tokens_per_second": 338.461
    },
    {
      "epoch": 0.042482758620689655,
      "grad_norm": 8.038250923156738,
      "learning_rate": 1.542857142857143e-05,
      "loss": 0.5287742614746094,
      "num_input_tokens_seen": 353524,
      "step": 462,
      "train_runtime": 1044.3473,
      "train_tokens_per_second": 338.512
    },
    {
      "epoch": 0.04257471264367816,
      "grad_norm": 7.293453216552734,
      "learning_rate": 1.5418546365914788e-05,
      "loss": 0.5775200128555298,
      "num_input_tokens_seen": 354688,
      "step": 463,
      "train_runtime": 1046.3764,
      "train_tokens_per_second": 338.968
    },
    {
      "epoch": 0.042666666666666665,
      "grad_norm": 6.730240345001221,
      "learning_rate": 1.5408521303258145e-05,
      "loss": 0.5127347707748413,
      "num_input_tokens_seen": 355448,
      "step": 464,
      "train_runtime": 1048.2898,
      "train_tokens_per_second": 339.074
    },
    {
      "epoch": 0.04275862068965517,
      "grad_norm": 7.285630226135254,
      "learning_rate": 1.5398496240601505e-05,
      "loss": 0.5023565292358398,
      "num_input_tokens_seen": 356124,
      "step": 465,
      "train_runtime": 1050.1896,
      "train_tokens_per_second": 339.104
    },
    {
      "epoch": 0.042850574712643676,
      "grad_norm": 6.86665153503418,
      "learning_rate": 1.5388471177944865e-05,
      "loss": 0.5498976707458496,
      "num_input_tokens_seen": 356848,
      "step": 466,
      "train_runtime": 1052.0953,
      "train_tokens_per_second": 339.178
    },
    {
      "epoch": 0.04294252873563218,
      "grad_norm": 7.294552803039551,
      "learning_rate": 1.5378446115288222e-05,
      "loss": 0.44851136207580566,
      "num_input_tokens_seen": 357460,
      "step": 467,
      "train_runtime": 1053.9797,
      "train_tokens_per_second": 339.153
    },
    {
      "epoch": 0.04303448275862069,
      "grad_norm": 7.41213321685791,
      "learning_rate": 1.536842105263158e-05,
      "loss": 0.5168503522872925,
      "num_input_tokens_seen": 358144,
      "step": 468,
      "train_runtime": 1055.89,
      "train_tokens_per_second": 339.187
    },
    {
      "epoch": 0.0431264367816092,
      "grad_norm": 7.41863489151001,
      "learning_rate": 1.535839598997494e-05,
      "loss": 0.5424340963363647,
      "num_input_tokens_seen": 358816,
      "step": 469,
      "train_runtime": 1057.8372,
      "train_tokens_per_second": 339.198
    },
    {
      "epoch": 0.0432183908045977,
      "grad_norm": 8.371504783630371,
      "learning_rate": 1.5348370927318296e-05,
      "loss": 0.537324070930481,
      "num_input_tokens_seen": 359420,
      "step": 470,
      "train_runtime": 1059.7345,
      "train_tokens_per_second": 339.16
    },
    {
      "epoch": 0.04331034482758621,
      "grad_norm": 7.062784671783447,
      "learning_rate": 1.5338345864661656e-05,
      "loss": 0.5428791642189026,
      "num_input_tokens_seen": 360168,
      "step": 471,
      "train_runtime": 1061.657,
      "train_tokens_per_second": 339.251
    },
    {
      "epoch": 0.043402298850574714,
      "grad_norm": 6.576425075531006,
      "learning_rate": 1.5328320802005013e-05,
      "loss": 0.4823741614818573,
      "num_input_tokens_seen": 360840,
      "step": 472,
      "train_runtime": 1063.5554,
      "train_tokens_per_second": 339.277
    },
    {
      "epoch": 0.04349425287356322,
      "grad_norm": 6.060364246368408,
      "learning_rate": 1.5318295739348373e-05,
      "loss": 0.3691994547843933,
      "num_input_tokens_seen": 361532,
      "step": 473,
      "train_runtime": 1065.4594,
      "train_tokens_per_second": 339.32
    },
    {
      "epoch": 0.043586206896551724,
      "grad_norm": 6.288186550140381,
      "learning_rate": 1.530827067669173e-05,
      "loss": 0.4629441499710083,
      "num_input_tokens_seen": 362200,
      "step": 474,
      "train_runtime": 1067.341,
      "train_tokens_per_second": 339.348
    },
    {
      "epoch": 0.04367816091954023,
      "grad_norm": 7.6472554206848145,
      "learning_rate": 1.529824561403509e-05,
      "loss": 0.5968223810195923,
      "num_input_tokens_seen": 362824,
      "step": 475,
      "train_runtime": 1069.2298,
      "train_tokens_per_second": 339.332
    },
    {
      "epoch": 0.043770114942528734,
      "grad_norm": 9.063471794128418,
      "learning_rate": 1.5288220551378447e-05,
      "loss": 0.5535184741020203,
      "num_input_tokens_seen": 363520,
      "step": 476,
      "train_runtime": 1071.1651,
      "train_tokens_per_second": 339.369
    },
    {
      "epoch": 0.04386206896551724,
      "grad_norm": 6.4986572265625,
      "learning_rate": 1.5278195488721807e-05,
      "loss": 0.47492700815200806,
      "num_input_tokens_seen": 364088,
      "step": 477,
      "train_runtime": 1073.0609,
      "train_tokens_per_second": 339.299
    },
    {
      "epoch": 0.043954022988505745,
      "grad_norm": 6.825013637542725,
      "learning_rate": 1.5268170426065164e-05,
      "loss": 0.5102417469024658,
      "num_input_tokens_seen": 365156,
      "step": 478,
      "train_runtime": 1075.0211,
      "train_tokens_per_second": 339.673
    },
    {
      "epoch": 0.04404597701149425,
      "grad_norm": 6.955339431762695,
      "learning_rate": 1.5258145363408522e-05,
      "loss": 0.5035444498062134,
      "num_input_tokens_seen": 365840,
      "step": 479,
      "train_runtime": 1076.9043,
      "train_tokens_per_second": 339.714
    },
    {
      "epoch": 0.044137931034482755,
      "grad_norm": 7.432355880737305,
      "learning_rate": 1.524812030075188e-05,
      "loss": 0.5037747025489807,
      "num_input_tokens_seen": 366700,
      "step": 480,
      "train_runtime": 1078.8369,
      "train_tokens_per_second": 339.903
    },
    {
      "epoch": 0.04422988505747127,
      "grad_norm": 7.469895839691162,
      "learning_rate": 1.523809523809524e-05,
      "loss": 0.5288439393043518,
      "num_input_tokens_seen": 367300,
      "step": 481,
      "train_runtime": 1086.9333,
      "train_tokens_per_second": 337.923
    },
    {
      "epoch": 0.04432183908045977,
      "grad_norm": 7.748864650726318,
      "learning_rate": 1.5228070175438598e-05,
      "loss": 0.5014351010322571,
      "num_input_tokens_seen": 367876,
      "step": 482,
      "train_runtime": 1088.8175,
      "train_tokens_per_second": 337.867
    },
    {
      "epoch": 0.04441379310344828,
      "grad_norm": 6.835473537445068,
      "learning_rate": 1.5218045112781956e-05,
      "loss": 0.4579215347766876,
      "num_input_tokens_seen": 368536,
      "step": 483,
      "train_runtime": 1090.7755,
      "train_tokens_per_second": 337.866
    },
    {
      "epoch": 0.04450574712643678,
      "grad_norm": 6.053252220153809,
      "learning_rate": 1.5208020050125315e-05,
      "loss": 0.4907318353652954,
      "num_input_tokens_seen": 369384,
      "step": 484,
      "train_runtime": 1092.6799,
      "train_tokens_per_second": 338.053
    },
    {
      "epoch": 0.04459770114942529,
      "grad_norm": 7.192389488220215,
      "learning_rate": 1.5197994987468673e-05,
      "loss": 0.40747565031051636,
      "num_input_tokens_seen": 370044,
      "step": 485,
      "train_runtime": 1094.5944,
      "train_tokens_per_second": 338.065
    },
    {
      "epoch": 0.04468965517241379,
      "grad_norm": 8.78969669342041,
      "learning_rate": 1.5187969924812032e-05,
      "loss": 0.5660673975944519,
      "num_input_tokens_seen": 370748,
      "step": 486,
      "train_runtime": 1096.4827,
      "train_tokens_per_second": 338.125
    },
    {
      "epoch": 0.0447816091954023,
      "grad_norm": 6.862427711486816,
      "learning_rate": 1.5177944862155388e-05,
      "loss": 0.7227067947387695,
      "num_input_tokens_seen": 371808,
      "step": 487,
      "train_runtime": 1098.41,
      "train_tokens_per_second": 338.497
    },
    {
      "epoch": 0.044873563218390804,
      "grad_norm": 7.217940807342529,
      "learning_rate": 1.5167919799498747e-05,
      "loss": 0.5614221096038818,
      "num_input_tokens_seen": 372516,
      "step": 488,
      "train_runtime": 1100.3087,
      "train_tokens_per_second": 338.556
    },
    {
      "epoch": 0.04496551724137931,
      "grad_norm": 6.643126487731934,
      "learning_rate": 1.5157894736842107e-05,
      "loss": 0.43227484822273254,
      "num_input_tokens_seen": 373208,
      "step": 489,
      "train_runtime": 1102.2096,
      "train_tokens_per_second": 338.6
    },
    {
      "epoch": 0.045057471264367814,
      "grad_norm": 7.370142459869385,
      "learning_rate": 1.5147869674185464e-05,
      "loss": 0.5297635197639465,
      "num_input_tokens_seen": 373884,
      "step": 490,
      "train_runtime": 1104.107,
      "train_tokens_per_second": 338.63
    },
    {
      "epoch": 0.04514942528735632,
      "grad_norm": 6.208383560180664,
      "learning_rate": 1.5137844611528822e-05,
      "loss": 0.5463951230049133,
      "num_input_tokens_seen": 374784,
      "step": 491,
      "train_runtime": 1106.0678,
      "train_tokens_per_second": 338.844
    },
    {
      "epoch": 0.045241379310344824,
      "grad_norm": 7.913474082946777,
      "learning_rate": 1.5127819548872183e-05,
      "loss": 0.48958340287208557,
      "num_input_tokens_seen": 375484,
      "step": 492,
      "train_runtime": 1107.9672,
      "train_tokens_per_second": 338.895
    },
    {
      "epoch": 0.04533333333333334,
      "grad_norm": 6.947763442993164,
      "learning_rate": 1.511779448621554e-05,
      "loss": 0.5918217897415161,
      "num_input_tokens_seen": 376292,
      "step": 493,
      "train_runtime": 1109.9038,
      "train_tokens_per_second": 339.031
    },
    {
      "epoch": 0.04542528735632184,
      "grad_norm": 7.253250598907471,
      "learning_rate": 1.5107769423558898e-05,
      "loss": 0.48761123418807983,
      "num_input_tokens_seen": 376852,
      "step": 494,
      "train_runtime": 1111.8869,
      "train_tokens_per_second": 338.93
    },
    {
      "epoch": 0.04551724137931035,
      "grad_norm": 6.873199939727783,
      "learning_rate": 1.5097744360902258e-05,
      "loss": 0.5429773926734924,
      "num_input_tokens_seen": 377628,
      "step": 495,
      "train_runtime": 1113.7885,
      "train_tokens_per_second": 339.048
    },
    {
      "epoch": 0.04560919540229885,
      "grad_norm": 7.205342769622803,
      "learning_rate": 1.5087719298245615e-05,
      "loss": 0.5807958841323853,
      "num_input_tokens_seen": 378456,
      "step": 496,
      "train_runtime": 1115.7442,
      "train_tokens_per_second": 339.196
    },
    {
      "epoch": 0.04570114942528736,
      "grad_norm": 6.8208136558532715,
      "learning_rate": 1.5077694235588973e-05,
      "loss": 0.4977841377258301,
      "num_input_tokens_seen": 379240,
      "step": 497,
      "train_runtime": 1117.6718,
      "train_tokens_per_second": 339.312
    },
    {
      "epoch": 0.04579310344827586,
      "grad_norm": 6.9195122718811035,
      "learning_rate": 1.506766917293233e-05,
      "loss": 0.5864359140396118,
      "num_input_tokens_seen": 380128,
      "step": 498,
      "train_runtime": 1119.6232,
      "train_tokens_per_second": 339.514
    },
    {
      "epoch": 0.04588505747126437,
      "grad_norm": 6.805229187011719,
      "learning_rate": 1.505764411027569e-05,
      "loss": 0.5215876698493958,
      "num_input_tokens_seen": 380824,
      "step": 499,
      "train_runtime": 1121.5691,
      "train_tokens_per_second": 339.546
    },
    {
      "epoch": 0.04597701149425287,
      "grad_norm": 9.82595443725586,
      "learning_rate": 1.5047619047619049e-05,
      "loss": 0.5936279296875,
      "num_input_tokens_seen": 381636,
      "step": 500,
      "train_runtime": 1123.5033,
      "train_tokens_per_second": 339.684
    },
    {
      "epoch": 0.04606896551724138,
      "grad_norm": 7.440690040588379,
      "learning_rate": 1.5037593984962406e-05,
      "loss": 0.5643438100814819,
      "num_input_tokens_seen": 382312,
      "step": 501,
      "train_runtime": 1125.44,
      "train_tokens_per_second": 339.7
    },
    {
      "epoch": 0.04616091954022988,
      "grad_norm": 6.9187331199646,
      "learning_rate": 1.5027568922305766e-05,
      "loss": 0.5397284626960754,
      "num_input_tokens_seen": 383204,
      "step": 502,
      "train_runtime": 1127.3951,
      "train_tokens_per_second": 339.902
    },
    {
      "epoch": 0.04625287356321839,
      "grad_norm": 6.451259136199951,
      "learning_rate": 1.5017543859649124e-05,
      "loss": 0.44190657138824463,
      "num_input_tokens_seen": 383884,
      "step": 503,
      "train_runtime": 1129.3123,
      "train_tokens_per_second": 339.927
    },
    {
      "epoch": 0.046344827586206894,
      "grad_norm": 6.243966579437256,
      "learning_rate": 1.5007518796992481e-05,
      "loss": 0.41493821144104004,
      "num_input_tokens_seen": 384620,
      "step": 504,
      "train_runtime": 1131.2851,
      "train_tokens_per_second": 339.985
    },
    {
      "epoch": 0.0464367816091954,
      "grad_norm": 7.39512300491333,
      "learning_rate": 1.4997493734335841e-05,
      "loss": 0.44311580061912537,
      "num_input_tokens_seen": 385204,
      "step": 505,
      "train_runtime": 1133.1717,
      "train_tokens_per_second": 339.934
    },
    {
      "epoch": 0.04652873563218391,
      "grad_norm": 6.167356491088867,
      "learning_rate": 1.49874686716792e-05,
      "loss": 0.40335023403167725,
      "num_input_tokens_seen": 385972,
      "step": 506,
      "train_runtime": 1135.11,
      "train_tokens_per_second": 340.03
    },
    {
      "epoch": 0.046620689655172416,
      "grad_norm": 6.923739433288574,
      "learning_rate": 1.4977443609022557e-05,
      "loss": 0.5292226672172546,
      "num_input_tokens_seen": 386780,
      "step": 507,
      "train_runtime": 1137.033,
      "train_tokens_per_second": 340.166
    },
    {
      "epoch": 0.04671264367816092,
      "grad_norm": 6.463278770446777,
      "learning_rate": 1.4967418546365915e-05,
      "loss": 0.466069757938385,
      "num_input_tokens_seen": 387512,
      "step": 508,
      "train_runtime": 1138.9544,
      "train_tokens_per_second": 340.235
    },
    {
      "epoch": 0.04680459770114943,
      "grad_norm": 6.766587734222412,
      "learning_rate": 1.4957393483709275e-05,
      "loss": 0.5540851950645447,
      "num_input_tokens_seen": 388624,
      "step": 509,
      "train_runtime": 1140.9151,
      "train_tokens_per_second": 340.625
    },
    {
      "epoch": 0.04689655172413793,
      "grad_norm": 6.9168806076049805,
      "learning_rate": 1.4947368421052632e-05,
      "loss": 0.44126710295677185,
      "num_input_tokens_seen": 389448,
      "step": 510,
      "train_runtime": 1142.8088,
      "train_tokens_per_second": 340.781
    },
    {
      "epoch": 0.04698850574712644,
      "grad_norm": 7.043146133422852,
      "learning_rate": 1.493734335839599e-05,
      "loss": 0.5624634623527527,
      "num_input_tokens_seen": 391272,
      "step": 511,
      "train_runtime": 1151.2832,
      "train_tokens_per_second": 339.857
    },
    {
      "epoch": 0.04708045977011494,
      "grad_norm": 5.854249477386475,
      "learning_rate": 1.492731829573935e-05,
      "loss": 0.3640488088130951,
      "num_input_tokens_seen": 391788,
      "step": 512,
      "train_runtime": 1153.1669,
      "train_tokens_per_second": 339.75
    },
    {
      "epoch": 0.04717241379310345,
      "grad_norm": 6.783285617828369,
      "learning_rate": 1.4917293233082707e-05,
      "loss": 0.3959953784942627,
      "num_input_tokens_seen": 392368,
      "step": 513,
      "train_runtime": 1155.0514,
      "train_tokens_per_second": 339.697
    },
    {
      "epoch": 0.04726436781609195,
      "grad_norm": 6.844997406005859,
      "learning_rate": 1.4907268170426066e-05,
      "loss": 0.457014799118042,
      "num_input_tokens_seen": 392984,
      "step": 514,
      "train_runtime": 1156.9385,
      "train_tokens_per_second": 339.676
    },
    {
      "epoch": 0.04735632183908046,
      "grad_norm": 6.397000312805176,
      "learning_rate": 1.4897243107769426e-05,
      "loss": 0.4044094383716583,
      "num_input_tokens_seen": 393648,
      "step": 515,
      "train_runtime": 1158.8609,
      "train_tokens_per_second": 339.685
    },
    {
      "epoch": 0.04744827586206896,
      "grad_norm": 7.328312397003174,
      "learning_rate": 1.4887218045112783e-05,
      "loss": 0.480017751455307,
      "num_input_tokens_seen": 394428,
      "step": 516,
      "train_runtime": 1160.7991,
      "train_tokens_per_second": 339.79
    },
    {
      "epoch": 0.04754022988505747,
      "grad_norm": 7.213738441467285,
      "learning_rate": 1.4877192982456141e-05,
      "loss": 0.4687790274620056,
      "num_input_tokens_seen": 395116,
      "step": 517,
      "train_runtime": 1162.7129,
      "train_tokens_per_second": 339.822
    },
    {
      "epoch": 0.04763218390804598,
      "grad_norm": 7.760647296905518,
      "learning_rate": 1.4867167919799498e-05,
      "loss": 0.600262463092804,
      "num_input_tokens_seen": 395808,
      "step": 518,
      "train_runtime": 1164.6098,
      "train_tokens_per_second": 339.863
    },
    {
      "epoch": 0.047724137931034485,
      "grad_norm": 8.408520698547363,
      "learning_rate": 1.4857142857142858e-05,
      "loss": 0.5764064788818359,
      "num_input_tokens_seen": 396572,
      "step": 519,
      "train_runtime": 1166.5371,
      "train_tokens_per_second": 339.957
    },
    {
      "epoch": 0.04781609195402299,
      "grad_norm": 7.4145612716674805,
      "learning_rate": 1.4847117794486217e-05,
      "loss": 0.4971860945224762,
      "num_input_tokens_seen": 397400,
      "step": 520,
      "train_runtime": 1168.4576,
      "train_tokens_per_second": 340.106
    },
    {
      "epoch": 0.047908045977011496,
      "grad_norm": 6.525569915771484,
      "learning_rate": 1.4837092731829574e-05,
      "loss": 0.36636286973953247,
      "num_input_tokens_seen": 397920,
      "step": 521,
      "train_runtime": 1170.3383,
      "train_tokens_per_second": 340.004
    },
    {
      "epoch": 0.048,
      "grad_norm": 6.48974084854126,
      "learning_rate": 1.4827067669172934e-05,
      "loss": 0.3712833523750305,
      "num_input_tokens_seen": 398624,
      "step": 522,
      "train_runtime": 1172.223,
      "train_tokens_per_second": 340.058
    },
    {
      "epoch": 0.048091954022988506,
      "grad_norm": 6.563835620880127,
      "learning_rate": 1.4817042606516292e-05,
      "loss": 0.4676303565502167,
      "num_input_tokens_seen": 399460,
      "step": 523,
      "train_runtime": 1174.1338,
      "train_tokens_per_second": 340.217
    },
    {
      "epoch": 0.04818390804597701,
      "grad_norm": 7.273741245269775,
      "learning_rate": 1.4807017543859649e-05,
      "loss": 0.5030032992362976,
      "num_input_tokens_seen": 400264,
      "step": 524,
      "train_runtime": 1176.0418,
      "train_tokens_per_second": 340.348
    },
    {
      "epoch": 0.04827586206896552,
      "grad_norm": 6.6886115074157715,
      "learning_rate": 1.479699248120301e-05,
      "loss": 0.5100569725036621,
      "num_input_tokens_seen": 400944,
      "step": 525,
      "train_runtime": 1178.0781,
      "train_tokens_per_second": 340.337
    },
    {
      "epoch": 0.04836781609195402,
      "grad_norm": 7.2770233154296875,
      "learning_rate": 1.4786967418546368e-05,
      "loss": 0.4868226647377014,
      "num_input_tokens_seen": 401476,
      "step": 526,
      "train_runtime": 1179.9553,
      "train_tokens_per_second": 340.247
    },
    {
      "epoch": 0.04845977011494253,
      "grad_norm": 7.772225856781006,
      "learning_rate": 1.4776942355889725e-05,
      "loss": 0.5350745916366577,
      "num_input_tokens_seen": 402412,
      "step": 527,
      "train_runtime": 1181.8793,
      "train_tokens_per_second": 340.485
    },
    {
      "epoch": 0.04855172413793103,
      "grad_norm": 7.460968494415283,
      "learning_rate": 1.4766917293233083e-05,
      "loss": 0.6134611368179321,
      "num_input_tokens_seen": 403108,
      "step": 528,
      "train_runtime": 1183.7734,
      "train_tokens_per_second": 340.528
    },
    {
      "epoch": 0.04864367816091954,
      "grad_norm": 7.340336322784424,
      "learning_rate": 1.4756892230576443e-05,
      "loss": 0.46978676319122314,
      "num_input_tokens_seen": 403736,
      "step": 529,
      "train_runtime": 1185.6759,
      "train_tokens_per_second": 340.511
    },
    {
      "epoch": 0.04873563218390804,
      "grad_norm": 7.76758337020874,
      "learning_rate": 1.47468671679198e-05,
      "loss": 0.5982768535614014,
      "num_input_tokens_seen": 404300,
      "step": 530,
      "train_runtime": 1187.5791,
      "train_tokens_per_second": 340.44
    },
    {
      "epoch": 0.048827586206896555,
      "grad_norm": 6.966864109039307,
      "learning_rate": 1.4736842105263159e-05,
      "loss": 0.5222433805465698,
      "num_input_tokens_seen": 405168,
      "step": 531,
      "train_runtime": 1189.5104,
      "train_tokens_per_second": 340.617
    },
    {
      "epoch": 0.04891954022988506,
      "grad_norm": 7.772345542907715,
      "learning_rate": 1.4726817042606519e-05,
      "loss": 0.5692169666290283,
      "num_input_tokens_seen": 405864,
      "step": 532,
      "train_runtime": 1191.4174,
      "train_tokens_per_second": 340.656
    },
    {
      "epoch": 0.049011494252873565,
      "grad_norm": 7.645069122314453,
      "learning_rate": 1.4716791979949876e-05,
      "loss": 0.4851095676422119,
      "num_input_tokens_seen": 406556,
      "step": 533,
      "train_runtime": 1193.297,
      "train_tokens_per_second": 340.7
    },
    {
      "epoch": 0.04910344827586207,
      "grad_norm": 7.193717002868652,
      "learning_rate": 1.4706766917293234e-05,
      "loss": 0.5600373148918152,
      "num_input_tokens_seen": 407384,
      "step": 534,
      "train_runtime": 1195.2217,
      "train_tokens_per_second": 340.844
    },
    {
      "epoch": 0.049195402298850575,
      "grad_norm": 9.166099548339844,
      "learning_rate": 1.4696741854636594e-05,
      "loss": 0.48763924837112427,
      "num_input_tokens_seen": 407952,
      "step": 535,
      "train_runtime": 1197.1062,
      "train_tokens_per_second": 340.782
    },
    {
      "epoch": 0.04928735632183908,
      "grad_norm": 7.534561634063721,
      "learning_rate": 1.4686716791979951e-05,
      "loss": 0.47965383529663086,
      "num_input_tokens_seen": 408540,
      "step": 536,
      "train_runtime": 1198.9918,
      "train_tokens_per_second": 340.736
    },
    {
      "epoch": 0.049379310344827586,
      "grad_norm": 6.627105236053467,
      "learning_rate": 1.467669172932331e-05,
      "loss": 0.405030220746994,
      "num_input_tokens_seen": 409064,
      "step": 537,
      "train_runtime": 1200.8814,
      "train_tokens_per_second": 340.636
    },
    {
      "epoch": 0.04947126436781609,
      "grad_norm": 7.8374409675598145,
      "learning_rate": 1.4666666666666666e-05,
      "loss": 0.5932339429855347,
      "num_input_tokens_seen": 409784,
      "step": 538,
      "train_runtime": 1202.7989,
      "train_tokens_per_second": 340.692
    },
    {
      "epoch": 0.049563218390804596,
      "grad_norm": 6.745156764984131,
      "learning_rate": 1.4656641604010026e-05,
      "loss": 0.48866891860961914,
      "num_input_tokens_seen": 410744,
      "step": 539,
      "train_runtime": 1204.7106,
      "train_tokens_per_second": 340.948
    },
    {
      "epoch": 0.0496551724137931,
      "grad_norm": 7.251765251159668,
      "learning_rate": 1.4646616541353385e-05,
      "loss": 0.38917776942253113,
      "num_input_tokens_seen": 411364,
      "step": 540,
      "train_runtime": 1206.5887,
      "train_tokens_per_second": 340.931
    },
    {
      "epoch": 0.04974712643678161,
      "grad_norm": 6.308721542358398,
      "learning_rate": 1.4636591478696742e-05,
      "loss": 0.4285096526145935,
      "num_input_tokens_seen": 411964,
      "step": 541,
      "train_runtime": 1215.1066,
      "train_tokens_per_second": 339.035
    },
    {
      "epoch": 0.04983908045977011,
      "grad_norm": 6.954815864562988,
      "learning_rate": 1.4626566416040102e-05,
      "loss": 0.4781310558319092,
      "num_input_tokens_seen": 412772,
      "step": 542,
      "train_runtime": 1217.0158,
      "train_tokens_per_second": 339.167
    },
    {
      "epoch": 0.049931034482758624,
      "grad_norm": 6.529648780822754,
      "learning_rate": 1.461654135338346e-05,
      "loss": 0.5092747211456299,
      "num_input_tokens_seen": 413304,
      "step": 543,
      "train_runtime": 1218.8999,
      "train_tokens_per_second": 339.08
    },
    {
      "epoch": 0.05002298850574713,
      "grad_norm": 6.731094837188721,
      "learning_rate": 1.4606516290726817e-05,
      "loss": 0.46955224871635437,
      "num_input_tokens_seen": 413900,
      "step": 544,
      "train_runtime": 1220.9783,
      "train_tokens_per_second": 338.99
    },
    {
      "epoch": 0.050114942528735634,
      "grad_norm": 7.204357624053955,
      "learning_rate": 1.4596491228070177e-05,
      "loss": 0.5351137518882751,
      "num_input_tokens_seen": 414556,
      "step": 545,
      "train_runtime": 1222.8707,
      "train_tokens_per_second": 339.002
    },
    {
      "epoch": 0.05020689655172414,
      "grad_norm": 6.740512371063232,
      "learning_rate": 1.4586466165413536e-05,
      "loss": 0.4871787428855896,
      "num_input_tokens_seen": 415116,
      "step": 546,
      "train_runtime": 1224.7612,
      "train_tokens_per_second": 338.936
    },
    {
      "epoch": 0.050298850574712645,
      "grad_norm": 6.851949214935303,
      "learning_rate": 1.4576441102756893e-05,
      "loss": 0.3844849169254303,
      "num_input_tokens_seen": 415732,
      "step": 547,
      "train_runtime": 1226.6582,
      "train_tokens_per_second": 338.914
    },
    {
      "epoch": 0.05039080459770115,
      "grad_norm": 6.723283290863037,
      "learning_rate": 1.4566416040100251e-05,
      "loss": 0.5672256946563721,
      "num_input_tokens_seen": 416528,
      "step": 548,
      "train_runtime": 1228.6518,
      "train_tokens_per_second": 339.012
    },
    {
      "epoch": 0.050482758620689655,
      "grad_norm": 7.665659427642822,
      "learning_rate": 1.4556390977443611e-05,
      "loss": 0.5598821640014648,
      "num_input_tokens_seen": 417108,
      "step": 549,
      "train_runtime": 1230.5523,
      "train_tokens_per_second": 338.96
    },
    {
      "epoch": 0.05057471264367816,
      "grad_norm": 7.521910190582275,
      "learning_rate": 1.4546365914786968e-05,
      "loss": 0.5603845119476318,
      "num_input_tokens_seen": 417860,
      "step": 550,
      "train_runtime": 1232.4428,
      "train_tokens_per_second": 339.05
    },
    {
      "epoch": 0.050666666666666665,
      "grad_norm": 7.453955173492432,
      "learning_rate": 1.4536340852130327e-05,
      "loss": 0.5222201347351074,
      "num_input_tokens_seen": 418544,
      "step": 551,
      "train_runtime": 1234.36,
      "train_tokens_per_second": 339.078
    },
    {
      "epoch": 0.05075862068965517,
      "grad_norm": 7.16868782043457,
      "learning_rate": 1.4526315789473687e-05,
      "loss": 0.45393678545951843,
      "num_input_tokens_seen": 419120,
      "step": 552,
      "train_runtime": 1236.2603,
      "train_tokens_per_second": 339.022
    },
    {
      "epoch": 0.050850574712643676,
      "grad_norm": 8.150325775146484,
      "learning_rate": 1.4516290726817044e-05,
      "loss": 0.5021551847457886,
      "num_input_tokens_seen": 419832,
      "step": 553,
      "train_runtime": 1238.1494,
      "train_tokens_per_second": 339.08
    },
    {
      "epoch": 0.05094252873563218,
      "grad_norm": 8.25429630279541,
      "learning_rate": 1.4506265664160402e-05,
      "loss": 0.509451687335968,
      "num_input_tokens_seen": 420460,
      "step": 554,
      "train_runtime": 1240.0395,
      "train_tokens_per_second": 339.07
    },
    {
      "epoch": 0.05103448275862069,
      "grad_norm": 7.019073963165283,
      "learning_rate": 1.4496240601503762e-05,
      "loss": 0.4084882140159607,
      "num_input_tokens_seen": 421124,
      "step": 555,
      "train_runtime": 1241.981,
      "train_tokens_per_second": 339.074
    },
    {
      "epoch": 0.0511264367816092,
      "grad_norm": 6.161911964416504,
      "learning_rate": 1.4486215538847119e-05,
      "loss": 0.4584411382675171,
      "num_input_tokens_seen": 421816,
      "step": 556,
      "train_runtime": 1243.883,
      "train_tokens_per_second": 339.112
    },
    {
      "epoch": 0.051218390804597703,
      "grad_norm": 7.898520469665527,
      "learning_rate": 1.4476190476190478e-05,
      "loss": 0.5508074760437012,
      "num_input_tokens_seen": 422556,
      "step": 557,
      "train_runtime": 1245.7764,
      "train_tokens_per_second": 339.191
    },
    {
      "epoch": 0.05131034482758621,
      "grad_norm": 7.782658100128174,
      "learning_rate": 1.4466165413533834e-05,
      "loss": 0.5297361016273499,
      "num_input_tokens_seen": 423264,
      "step": 558,
      "train_runtime": 1247.6927,
      "train_tokens_per_second": 339.237
    },
    {
      "epoch": 0.051402298850574714,
      "grad_norm": 7.9233574867248535,
      "learning_rate": 1.4456140350877195e-05,
      "loss": 0.5354646444320679,
      "num_input_tokens_seen": 423932,
      "step": 559,
      "train_runtime": 1249.608,
      "train_tokens_per_second": 339.252
    },
    {
      "epoch": 0.05149425287356322,
      "grad_norm": 7.549014568328857,
      "learning_rate": 1.4446115288220553e-05,
      "loss": 0.5111515522003174,
      "num_input_tokens_seen": 424620,
      "step": 560,
      "train_runtime": 1251.4837,
      "train_tokens_per_second": 339.293
    },
    {
      "epoch": 0.051586206896551724,
      "grad_norm": 7.239883899688721,
      "learning_rate": 1.443609022556391e-05,
      "loss": 0.47565650939941406,
      "num_input_tokens_seen": 425456,
      "step": 561,
      "train_runtime": 1253.4081,
      "train_tokens_per_second": 339.439
    },
    {
      "epoch": 0.05167816091954023,
      "grad_norm": 7.050983428955078,
      "learning_rate": 1.442606516290727e-05,
      "loss": 0.6090424060821533,
      "num_input_tokens_seen": 427156,
      "step": 562,
      "train_runtime": 1255.4411,
      "train_tokens_per_second": 340.244
    },
    {
      "epoch": 0.051770114942528735,
      "grad_norm": 8.222979545593262,
      "learning_rate": 1.4416040100250628e-05,
      "loss": 0.48763030767440796,
      "num_input_tokens_seen": 427768,
      "step": 563,
      "train_runtime": 1257.3115,
      "train_tokens_per_second": 340.224
    },
    {
      "epoch": 0.05186206896551724,
      "grad_norm": 6.865083694458008,
      "learning_rate": 1.4406015037593985e-05,
      "loss": 0.45680394768714905,
      "num_input_tokens_seen": 428572,
      "step": 564,
      "train_runtime": 1259.246,
      "train_tokens_per_second": 340.34
    },
    {
      "epoch": 0.051954022988505745,
      "grad_norm": 6.205889701843262,
      "learning_rate": 1.4395989974937345e-05,
      "loss": 0.4514808654785156,
      "num_input_tokens_seen": 429556,
      "step": 565,
      "train_runtime": 1261.1552,
      "train_tokens_per_second": 340.605
    },
    {
      "epoch": 0.05204597701149425,
      "grad_norm": 6.261847496032715,
      "learning_rate": 1.4385964912280704e-05,
      "loss": 0.391244113445282,
      "num_input_tokens_seen": 430128,
      "step": 566,
      "train_runtime": 1263.0367,
      "train_tokens_per_second": 340.551
    },
    {
      "epoch": 0.052137931034482755,
      "grad_norm": 7.326925754547119,
      "learning_rate": 1.437593984962406e-05,
      "loss": 0.5507190823554993,
      "num_input_tokens_seen": 430912,
      "step": 567,
      "train_runtime": 1264.915,
      "train_tokens_per_second": 340.665
    },
    {
      "epoch": 0.05222988505747127,
      "grad_norm": 6.737486362457275,
      "learning_rate": 1.436591478696742e-05,
      "loss": 0.5284934639930725,
      "num_input_tokens_seen": 431840,
      "step": 568,
      "train_runtime": 1266.8229,
      "train_tokens_per_second": 340.884
    },
    {
      "epoch": 0.05232183908045977,
      "grad_norm": 7.514186382293701,
      "learning_rate": 1.435588972431078e-05,
      "loss": 0.5544549226760864,
      "num_input_tokens_seen": 432600,
      "step": 569,
      "train_runtime": 1268.7077,
      "train_tokens_per_second": 340.977
    },
    {
      "epoch": 0.05241379310344828,
      "grad_norm": 7.510397434234619,
      "learning_rate": 1.4345864661654136e-05,
      "loss": 0.41615110635757446,
      "num_input_tokens_seen": 433148,
      "step": 570,
      "train_runtime": 1270.5817,
      "train_tokens_per_second": 340.905
    },
    {
      "epoch": 0.05250574712643678,
      "grad_norm": 6.880468845367432,
      "learning_rate": 1.4335839598997495e-05,
      "loss": 0.4594903886318207,
      "num_input_tokens_seen": 433784,
      "step": 571,
      "train_runtime": 1279.5805,
      "train_tokens_per_second": 339.005
    },
    {
      "epoch": 0.05259770114942529,
      "grad_norm": 7.571253299713135,
      "learning_rate": 1.4325814536340855e-05,
      "loss": 0.5042752027511597,
      "num_input_tokens_seen": 434416,
      "step": 572,
      "train_runtime": 1281.5123,
      "train_tokens_per_second": 338.987
    },
    {
      "epoch": 0.052689655172413793,
      "grad_norm": 6.492729663848877,
      "learning_rate": 1.4315789473684212e-05,
      "loss": 0.48361361026763916,
      "num_input_tokens_seen": 435320,
      "step": 573,
      "train_runtime": 1283.4323,
      "train_tokens_per_second": 339.184
    },
    {
      "epoch": 0.0527816091954023,
      "grad_norm": 6.326514720916748,
      "learning_rate": 1.430576441102757e-05,
      "loss": 0.4741998016834259,
      "num_input_tokens_seen": 436088,
      "step": 574,
      "train_runtime": 1285.3277,
      "train_tokens_per_second": 339.282
    },
    {
      "epoch": 0.052873563218390804,
      "grad_norm": 6.165903568267822,
      "learning_rate": 1.4295739348370929e-05,
      "loss": 0.4289013147354126,
      "num_input_tokens_seen": 436632,
      "step": 575,
      "train_runtime": 1287.1992,
      "train_tokens_per_second": 339.211
    },
    {
      "epoch": 0.05296551724137931,
      "grad_norm": 7.48010778427124,
      "learning_rate": 1.4285714285714287e-05,
      "loss": 0.5268988013267517,
      "num_input_tokens_seen": 437312,
      "step": 576,
      "train_runtime": 1289.1189,
      "train_tokens_per_second": 339.233
    },
    {
      "epoch": 0.053057471264367814,
      "grad_norm": 7.232751369476318,
      "learning_rate": 1.4275689223057646e-05,
      "loss": 0.5748152136802673,
      "num_input_tokens_seen": 438096,
      "step": 577,
      "train_runtime": 1291.0259,
      "train_tokens_per_second": 339.339
    },
    {
      "epoch": 0.05314942528735632,
      "grad_norm": 5.850233554840088,
      "learning_rate": 1.4265664160401002e-05,
      "loss": 0.3801572620868683,
      "num_input_tokens_seen": 438744,
      "step": 578,
      "train_runtime": 1292.9025,
      "train_tokens_per_second": 339.348
    },
    {
      "epoch": 0.053241379310344825,
      "grad_norm": 7.358694553375244,
      "learning_rate": 1.4255639097744363e-05,
      "loss": 0.5200968384742737,
      "num_input_tokens_seen": 439996,
      "step": 579,
      "train_runtime": 1294.841,
      "train_tokens_per_second": 339.807
    },
    {
      "epoch": 0.05333333333333334,
      "grad_norm": 7.193248271942139,
      "learning_rate": 1.4245614035087721e-05,
      "loss": 0.6091095805168152,
      "num_input_tokens_seen": 441044,
      "step": 580,
      "train_runtime": 1296.7586,
      "train_tokens_per_second": 340.113
    },
    {
      "epoch": 0.05342528735632184,
      "grad_norm": 6.751431465148926,
      "learning_rate": 1.4235588972431078e-05,
      "loss": 0.5009475946426392,
      "num_input_tokens_seen": 441592,
      "step": 581,
      "train_runtime": 1298.6382,
      "train_tokens_per_second": 340.042
    },
    {
      "epoch": 0.05351724137931035,
      "grad_norm": 7.5356831550598145,
      "learning_rate": 1.4225563909774438e-05,
      "loss": 0.4519139528274536,
      "num_input_tokens_seen": 442288,
      "step": 582,
      "train_runtime": 1300.5207,
      "train_tokens_per_second": 340.085
    },
    {
      "epoch": 0.05360919540229885,
      "grad_norm": 6.881683826446533,
      "learning_rate": 1.4215538847117797e-05,
      "loss": 0.6669802069664001,
      "num_input_tokens_seen": 443224,
      "step": 583,
      "train_runtime": 1302.4548,
      "train_tokens_per_second": 340.299
    },
    {
      "epoch": 0.05370114942528736,
      "grad_norm": 7.067171096801758,
      "learning_rate": 1.4205513784461153e-05,
      "loss": 0.42816635966300964,
      "num_input_tokens_seen": 444032,
      "step": 584,
      "train_runtime": 1304.4902,
      "train_tokens_per_second": 340.387
    },
    {
      "epoch": 0.05379310344827586,
      "grad_norm": 6.609787464141846,
      "learning_rate": 1.4195488721804514e-05,
      "loss": 0.5364288091659546,
      "num_input_tokens_seen": 444736,
      "step": 585,
      "train_runtime": 1306.4324,
      "train_tokens_per_second": 340.42
    },
    {
      "epoch": 0.05388505747126437,
      "grad_norm": 7.067381858825684,
      "learning_rate": 1.4185463659147872e-05,
      "loss": 0.5622860193252563,
      "num_input_tokens_seen": 445672,
      "step": 586,
      "train_runtime": 1308.3677,
      "train_tokens_per_second": 340.632
    },
    {
      "epoch": 0.05397701149425287,
      "grad_norm": 6.087325096130371,
      "learning_rate": 1.4175438596491229e-05,
      "loss": 0.4185801148414612,
      "num_input_tokens_seen": 446316,
      "step": 587,
      "train_runtime": 1310.2514,
      "train_tokens_per_second": 340.634
    },
    {
      "epoch": 0.05406896551724138,
      "grad_norm": 6.607666015625,
      "learning_rate": 1.4165413533834587e-05,
      "loss": 0.4877302348613739,
      "num_input_tokens_seen": 446916,
      "step": 588,
      "train_runtime": 1312.1457,
      "train_tokens_per_second": 340.599
    },
    {
      "epoch": 0.054160919540229883,
      "grad_norm": 6.90864372253418,
      "learning_rate": 1.4155388471177946e-05,
      "loss": 0.4787876605987549,
      "num_input_tokens_seen": 447448,
      "step": 589,
      "train_runtime": 1314.0595,
      "train_tokens_per_second": 340.508
    },
    {
      "epoch": 0.05425287356321839,
      "grad_norm": 6.653378009796143,
      "learning_rate": 1.4145363408521304e-05,
      "loss": 0.45052623748779297,
      "num_input_tokens_seen": 448144,
      "step": 590,
      "train_runtime": 1315.9691,
      "train_tokens_per_second": 340.543
    },
    {
      "epoch": 0.054344827586206894,
      "grad_norm": 7.320746898651123,
      "learning_rate": 1.4135338345864663e-05,
      "loss": 0.5776675343513489,
      "num_input_tokens_seen": 448920,
      "step": 591,
      "train_runtime": 1317.9018,
      "train_tokens_per_second": 340.632
    },
    {
      "epoch": 0.0544367816091954,
      "grad_norm": 6.116318225860596,
      "learning_rate": 1.4125313283208021e-05,
      "loss": 0.4263470768928528,
      "num_input_tokens_seen": 449716,
      "step": 592,
      "train_runtime": 1319.8314,
      "train_tokens_per_second": 340.737
    },
    {
      "epoch": 0.05452873563218391,
      "grad_norm": 7.996311187744141,
      "learning_rate": 1.411528822055138e-05,
      "loss": 0.5323801040649414,
      "num_input_tokens_seen": 450356,
      "step": 593,
      "train_runtime": 1321.7236,
      "train_tokens_per_second": 340.734
    },
    {
      "epoch": 0.054620689655172416,
      "grad_norm": 6.50635290145874,
      "learning_rate": 1.4105263157894738e-05,
      "loss": 0.4133082628250122,
      "num_input_tokens_seen": 450952,
      "step": 594,
      "train_runtime": 1323.6055,
      "train_tokens_per_second": 340.7
    },
    {
      "epoch": 0.05471264367816092,
      "grad_norm": 6.741985321044922,
      "learning_rate": 1.4095238095238097e-05,
      "loss": 0.5009517669677734,
      "num_input_tokens_seen": 451908,
      "step": 595,
      "train_runtime": 1325.5491,
      "train_tokens_per_second": 340.921
    },
    {
      "epoch": 0.05480459770114943,
      "grad_norm": 7.827582359313965,
      "learning_rate": 1.4085213032581455e-05,
      "loss": 0.5611375570297241,
      "num_input_tokens_seen": 452680,
      "step": 596,
      "train_runtime": 1327.4278,
      "train_tokens_per_second": 341.02
    },
    {
      "epoch": 0.05489655172413793,
      "grad_norm": 7.473243236541748,
      "learning_rate": 1.4075187969924814e-05,
      "loss": 0.4487534165382385,
      "num_input_tokens_seen": 453380,
      "step": 597,
      "train_runtime": 1329.3336,
      "train_tokens_per_second": 341.058
    },
    {
      "epoch": 0.05498850574712644,
      "grad_norm": 11.061628341674805,
      "learning_rate": 1.406516290726817e-05,
      "loss": 0.48190802335739136,
      "num_input_tokens_seen": 454044,
      "step": 598,
      "train_runtime": 1331.2129,
      "train_tokens_per_second": 341.075
    },
    {
      "epoch": 0.05508045977011494,
      "grad_norm": 6.291112899780273,
      "learning_rate": 1.405513784461153e-05,
      "loss": 0.6732971668243408,
      "num_input_tokens_seen": 455540,
      "step": 599,
      "train_runtime": 1333.2299,
      "train_tokens_per_second": 341.681
    },
    {
      "epoch": 0.05517241379310345,
      "grad_norm": 6.915909767150879,
      "learning_rate": 1.4045112781954887e-05,
      "loss": 0.4161275625228882,
      "num_input_tokens_seen": 456248,
      "step": 600,
      "train_runtime": 1335.1361,
      "train_tokens_per_second": 341.724
    },
    {
      "epoch": 0.05517241379310345,
      "eval_loss": 1.9245530366897583,
      "eval_runtime": 27.8262,
      "eval_samples_per_second": 35.937,
      "eval_steps_per_second": 8.984,
      "num_input_tokens_seen": 456248,
      "step": 600
    },
    {
      "epoch": 0.05526436781609195,
      "grad_norm": 7.501951217651367,
      "learning_rate": 1.4035087719298246e-05,
      "loss": 0.5070131421089172,
      "num_input_tokens_seen": 456844,
      "step": 601,
      "train_runtime": 1371.7412,
      "train_tokens_per_second": 333.039
    },
    {
      "epoch": 0.05535632183908046,
      "grad_norm": 7.221893787384033,
      "learning_rate": 1.4025062656641606e-05,
      "loss": 0.500610888004303,
      "num_input_tokens_seen": 457516,
      "step": 602,
      "train_runtime": 1373.7721,
      "train_tokens_per_second": 333.036
    },
    {
      "epoch": 0.05544827586206896,
      "grad_norm": 7.347901821136475,
      "learning_rate": 1.4015037593984963e-05,
      "loss": 0.6231392025947571,
      "num_input_tokens_seen": 458384,
      "step": 603,
      "train_runtime": 1375.6946,
      "train_tokens_per_second": 333.202
    },
    {
      "epoch": 0.05554022988505747,
      "grad_norm": 6.535465240478516,
      "learning_rate": 1.4005012531328321e-05,
      "loss": 0.38742274045944214,
      "num_input_tokens_seen": 458928,
      "step": 604,
      "train_runtime": 1377.584,
      "train_tokens_per_second": 333.14
    },
    {
      "epoch": 0.05563218390804598,
      "grad_norm": 6.457494735717773,
      "learning_rate": 1.3994987468671682e-05,
      "loss": 0.47244036197662354,
      "num_input_tokens_seen": 459684,
      "step": 605,
      "train_runtime": 1379.5472,
      "train_tokens_per_second": 333.214
    },
    {
      "epoch": 0.055724137931034486,
      "grad_norm": 6.482058525085449,
      "learning_rate": 1.3984962406015038e-05,
      "loss": 0.429859459400177,
      "num_input_tokens_seen": 460348,
      "step": 606,
      "train_runtime": 1381.4705,
      "train_tokens_per_second": 333.23
    },
    {
      "epoch": 0.05581609195402299,
      "grad_norm": 8.471290588378906,
      "learning_rate": 1.3974937343358397e-05,
      "loss": 0.5313338041305542,
      "num_input_tokens_seen": 461072,
      "step": 607,
      "train_runtime": 1383.3572,
      "train_tokens_per_second": 333.299
    },
    {
      "epoch": 0.055908045977011496,
      "grad_norm": 6.4897003173828125,
      "learning_rate": 1.3964912280701755e-05,
      "loss": 0.4652232527732849,
      "num_input_tokens_seen": 461680,
      "step": 608,
      "train_runtime": 1385.2346,
      "train_tokens_per_second": 333.287
    },
    {
      "epoch": 0.056,
      "grad_norm": 7.194956302642822,
      "learning_rate": 1.3954887218045114e-05,
      "loss": 0.4530368447303772,
      "num_input_tokens_seen": 462492,
      "step": 609,
      "train_runtime": 1387.1447,
      "train_tokens_per_second": 333.413
    },
    {
      "epoch": 0.056091954022988506,
      "grad_norm": 7.285534858703613,
      "learning_rate": 1.3944862155388472e-05,
      "loss": 0.4804021120071411,
      "num_input_tokens_seen": 463204,
      "step": 610,
      "train_runtime": 1389.0595,
      "train_tokens_per_second": 333.466
    },
    {
      "epoch": 0.05618390804597701,
      "grad_norm": 6.3361687660217285,
      "learning_rate": 1.3934837092731829e-05,
      "loss": 0.47082218527793884,
      "num_input_tokens_seen": 463976,
      "step": 611,
      "train_runtime": 1390.989,
      "train_tokens_per_second": 333.558
    },
    {
      "epoch": 0.05627586206896552,
      "grad_norm": 6.84358549118042,
      "learning_rate": 1.392481203007519e-05,
      "loss": 0.4161568284034729,
      "num_input_tokens_seen": 464488,
      "step": 612,
      "train_runtime": 1392.8722,
      "train_tokens_per_second": 333.475
    },
    {
      "epoch": 0.05636781609195402,
      "grad_norm": 6.668210029602051,
      "learning_rate": 1.3914786967418548e-05,
      "loss": 0.5246766209602356,
      "num_input_tokens_seen": 465308,
      "step": 613,
      "train_runtime": 1394.8056,
      "train_tokens_per_second": 333.601
    },
    {
      "epoch": 0.05645977011494253,
      "grad_norm": 6.792135238647461,
      "learning_rate": 1.3904761904761905e-05,
      "loss": 0.502672016620636,
      "num_input_tokens_seen": 465952,
      "step": 614,
      "train_runtime": 1396.7219,
      "train_tokens_per_second": 333.604
    },
    {
      "epoch": 0.05655172413793103,
      "grad_norm": 8.083513259887695,
      "learning_rate": 1.3894736842105265e-05,
      "loss": 0.4832074046134949,
      "num_input_tokens_seen": 466960,
      "step": 615,
      "train_runtime": 1398.6947,
      "train_tokens_per_second": 333.854
    },
    {
      "epoch": 0.05664367816091954,
      "grad_norm": 7.883939266204834,
      "learning_rate": 1.3884711779448623e-05,
      "loss": 0.45869264006614685,
      "num_input_tokens_seen": 467584,
      "step": 616,
      "train_runtime": 1400.5931,
      "train_tokens_per_second": 333.847
    },
    {
      "epoch": 0.05673563218390804,
      "grad_norm": 6.502721309661865,
      "learning_rate": 1.387468671679198e-05,
      "loss": 0.3988117575645447,
      "num_input_tokens_seen": 468228,
      "step": 617,
      "train_runtime": 1402.4785,
      "train_tokens_per_second": 333.858
    },
    {
      "epoch": 0.056827586206896555,
      "grad_norm": 7.88711404800415,
      "learning_rate": 1.3864661654135339e-05,
      "loss": 0.48905035853385925,
      "num_input_tokens_seen": 468772,
      "step": 618,
      "train_runtime": 1404.4042,
      "train_tokens_per_second": 333.787
    },
    {
      "epoch": 0.05691954022988506,
      "grad_norm": 6.885398864746094,
      "learning_rate": 1.3854636591478699e-05,
      "loss": 0.5362621545791626,
      "num_input_tokens_seen": 469608,
      "step": 619,
      "train_runtime": 1406.3305,
      "train_tokens_per_second": 333.924
    },
    {
      "epoch": 0.057011494252873565,
      "grad_norm": 6.992614269256592,
      "learning_rate": 1.3844611528822056e-05,
      "loss": 0.43546825647354126,
      "num_input_tokens_seen": 470188,
      "step": 620,
      "train_runtime": 1408.2209,
      "train_tokens_per_second": 333.888
    },
    {
      "epoch": 0.05710344827586207,
      "grad_norm": 7.322522163391113,
      "learning_rate": 1.3834586466165414e-05,
      "loss": 0.40986236929893494,
      "num_input_tokens_seen": 470740,
      "step": 621,
      "train_runtime": 1410.1226,
      "train_tokens_per_second": 333.829
    },
    {
      "epoch": 0.057195402298850576,
      "grad_norm": 6.906339645385742,
      "learning_rate": 1.3824561403508774e-05,
      "loss": 0.48197853565216064,
      "num_input_tokens_seen": 471544,
      "step": 622,
      "train_runtime": 1412.0454,
      "train_tokens_per_second": 333.944
    },
    {
      "epoch": 0.05728735632183908,
      "grad_norm": 6.975005149841309,
      "learning_rate": 1.3814536340852131e-05,
      "loss": 0.45380908250808716,
      "num_input_tokens_seen": 472196,
      "step": 623,
      "train_runtime": 1413.9404,
      "train_tokens_per_second": 333.958
    },
    {
      "epoch": 0.057379310344827586,
      "grad_norm": 6.444450855255127,
      "learning_rate": 1.380451127819549e-05,
      "loss": 0.45575663447380066,
      "num_input_tokens_seen": 472880,
      "step": 624,
      "train_runtime": 1415.8372,
      "train_tokens_per_second": 333.993
    },
    {
      "epoch": 0.05747126436781609,
      "grad_norm": 6.794878005981445,
      "learning_rate": 1.379448621553885e-05,
      "loss": 0.41316524147987366,
      "num_input_tokens_seen": 473432,
      "step": 625,
      "train_runtime": 1417.7243,
      "train_tokens_per_second": 333.938
    },
    {
      "epoch": 0.057563218390804596,
      "grad_norm": 7.611130237579346,
      "learning_rate": 1.3784461152882206e-05,
      "loss": 0.5504496693611145,
      "num_input_tokens_seen": 474304,
      "step": 626,
      "train_runtime": 1419.7027,
      "train_tokens_per_second": 334.087
    },
    {
      "epoch": 0.0576551724137931,
      "grad_norm": 7.361907958984375,
      "learning_rate": 1.3774436090225565e-05,
      "loss": 0.603052020072937,
      "num_input_tokens_seen": 474832,
      "step": 627,
      "train_runtime": 1421.5911,
      "train_tokens_per_second": 334.014
    },
    {
      "epoch": 0.05774712643678161,
      "grad_norm": 7.018294811248779,
      "learning_rate": 1.3764411027568922e-05,
      "loss": 0.49843698740005493,
      "num_input_tokens_seen": 475436,
      "step": 628,
      "train_runtime": 1423.4811,
      "train_tokens_per_second": 333.995
    },
    {
      "epoch": 0.05783908045977011,
      "grad_norm": 6.863377571105957,
      "learning_rate": 1.3754385964912282e-05,
      "loss": 0.5012993216514587,
      "num_input_tokens_seen": 476196,
      "step": 629,
      "train_runtime": 1425.3856,
      "train_tokens_per_second": 334.082
    },
    {
      "epoch": 0.057931034482758624,
      "grad_norm": 7.29086446762085,
      "learning_rate": 1.374436090225564e-05,
      "loss": 0.41038984060287476,
      "num_input_tokens_seen": 476824,
      "step": 630,
      "train_runtime": 1427.2687,
      "train_tokens_per_second": 334.081
    },
    {
      "epoch": 0.05802298850574713,
      "grad_norm": 6.348538398742676,
      "learning_rate": 1.3734335839598997e-05,
      "loss": 0.41164064407348633,
      "num_input_tokens_seen": 477396,
      "step": 631,
      "train_runtime": 1436.0596,
      "train_tokens_per_second": 332.435
    },
    {
      "epoch": 0.058114942528735634,
      "grad_norm": 8.6748046875,
      "learning_rate": 1.3724310776942357e-05,
      "loss": 0.42310094833374023,
      "num_input_tokens_seen": 478000,
      "step": 632,
      "train_runtime": 1438.0276,
      "train_tokens_per_second": 332.4
    },
    {
      "epoch": 0.05820689655172414,
      "grad_norm": 7.081576824188232,
      "learning_rate": 1.3714285714285716e-05,
      "loss": 0.563265860080719,
      "num_input_tokens_seen": 478700,
      "step": 633,
      "train_runtime": 1439.9265,
      "train_tokens_per_second": 332.448
    },
    {
      "epoch": 0.058298850574712645,
      "grad_norm": 7.547027587890625,
      "learning_rate": 1.3704260651629073e-05,
      "loss": 0.5364609956741333,
      "num_input_tokens_seen": 479380,
      "step": 634,
      "train_runtime": 1441.9256,
      "train_tokens_per_second": 332.458
    },
    {
      "epoch": 0.05839080459770115,
      "grad_norm": 7.5485429763793945,
      "learning_rate": 1.3694235588972433e-05,
      "loss": 0.4307040572166443,
      "num_input_tokens_seen": 480088,
      "step": 635,
      "train_runtime": 1443.8814,
      "train_tokens_per_second": 332.498
    },
    {
      "epoch": 0.058482758620689655,
      "grad_norm": 8.463400840759277,
      "learning_rate": 1.3684210526315791e-05,
      "loss": 0.5067605376243591,
      "num_input_tokens_seen": 480764,
      "step": 636,
      "train_runtime": 1445.9187,
      "train_tokens_per_second": 332.497
    },
    {
      "epoch": 0.05857471264367816,
      "grad_norm": 6.971987247467041,
      "learning_rate": 1.3674185463659148e-05,
      "loss": 0.48865610361099243,
      "num_input_tokens_seen": 481488,
      "step": 637,
      "train_runtime": 1447.8676,
      "train_tokens_per_second": 332.55
    },
    {
      "epoch": 0.058666666666666666,
      "grad_norm": 6.737226963043213,
      "learning_rate": 1.3664160401002507e-05,
      "loss": 0.4488855302333832,
      "num_input_tokens_seen": 482432,
      "step": 638,
      "train_runtime": 1449.8477,
      "train_tokens_per_second": 332.747
    },
    {
      "epoch": 0.05875862068965517,
      "grad_norm": 6.816892623901367,
      "learning_rate": 1.3654135338345867e-05,
      "loss": 0.46124267578125,
      "num_input_tokens_seen": 483012,
      "step": 639,
      "train_runtime": 1451.7729,
      "train_tokens_per_second": 332.705
    },
    {
      "epoch": 0.058850574712643676,
      "grad_norm": 7.571601390838623,
      "learning_rate": 1.3644110275689224e-05,
      "loss": 0.5857945084571838,
      "num_input_tokens_seen": 483728,
      "step": 640,
      "train_runtime": 1453.692,
      "train_tokens_per_second": 332.758
    },
    {
      "epoch": 0.05894252873563218,
      "grad_norm": 6.198338508605957,
      "learning_rate": 1.3634085213032582e-05,
      "loss": 0.4566928446292877,
      "num_input_tokens_seen": 484372,
      "step": 641,
      "train_runtime": 1455.5907,
      "train_tokens_per_second": 332.767
    },
    {
      "epoch": 0.059034482758620686,
      "grad_norm": 7.301205635070801,
      "learning_rate": 1.3624060150375942e-05,
      "loss": 0.50412917137146,
      "num_input_tokens_seen": 484992,
      "step": 642,
      "train_runtime": 1457.4654,
      "train_tokens_per_second": 332.764
    },
    {
      "epoch": 0.0591264367816092,
      "grad_norm": 6.268556594848633,
      "learning_rate": 1.3614035087719299e-05,
      "loss": 0.45484834909439087,
      "num_input_tokens_seen": 486216,
      "step": 643,
      "train_runtime": 1459.4441,
      "train_tokens_per_second": 333.152
    },
    {
      "epoch": 0.059218390804597704,
      "grad_norm": 7.198829650878906,
      "learning_rate": 1.3604010025062658e-05,
      "loss": 0.4686107039451599,
      "num_input_tokens_seen": 486900,
      "step": 644,
      "train_runtime": 1461.3529,
      "train_tokens_per_second": 333.184
    },
    {
      "epoch": 0.05931034482758621,
      "grad_norm": 6.827565670013428,
      "learning_rate": 1.3593984962406018e-05,
      "loss": 0.46443143486976624,
      "num_input_tokens_seen": 487716,
      "step": 645,
      "train_runtime": 1463.2791,
      "train_tokens_per_second": 333.303
    },
    {
      "epoch": 0.059402298850574714,
      "grad_norm": 6.095017910003662,
      "learning_rate": 1.3583959899749374e-05,
      "loss": 0.46966350078582764,
      "num_input_tokens_seen": 488416,
      "step": 646,
      "train_runtime": 1465.1834,
      "train_tokens_per_second": 333.348
    },
    {
      "epoch": 0.05949425287356322,
      "grad_norm": 7.52510404586792,
      "learning_rate": 1.3573934837092733e-05,
      "loss": 0.5057723522186279,
      "num_input_tokens_seen": 489248,
      "step": 647,
      "train_runtime": 1467.1037,
      "train_tokens_per_second": 333.479
    },
    {
      "epoch": 0.059586206896551724,
      "grad_norm": 6.387228965759277,
      "learning_rate": 1.356390977443609e-05,
      "loss": 0.5528343915939331,
      "num_input_tokens_seen": 490452,
      "step": 648,
      "train_runtime": 1469.0557,
      "train_tokens_per_second": 333.855
    },
    {
      "epoch": 0.05967816091954023,
      "grad_norm": 7.294100284576416,
      "learning_rate": 1.355388471177945e-05,
      "loss": 0.5030719041824341,
      "num_input_tokens_seen": 491080,
      "step": 649,
      "train_runtime": 1470.9525,
      "train_tokens_per_second": 333.852
    },
    {
      "epoch": 0.059770114942528735,
      "grad_norm": 6.579063415527344,
      "learning_rate": 1.3543859649122808e-05,
      "loss": 0.41523224115371704,
      "num_input_tokens_seen": 491824,
      "step": 650,
      "train_runtime": 1472.8282,
      "train_tokens_per_second": 333.932
    },
    {
      "epoch": 0.05986206896551724,
      "grad_norm": 6.683007717132568,
      "learning_rate": 1.3533834586466165e-05,
      "loss": 0.48073697090148926,
      "num_input_tokens_seen": 492464,
      "step": 651,
      "train_runtime": 1474.7119,
      "train_tokens_per_second": 333.939
    },
    {
      "epoch": 0.059954022988505745,
      "grad_norm": 7.796069145202637,
      "learning_rate": 1.3523809523809525e-05,
      "loss": 0.6073784828186035,
      "num_input_tokens_seen": 493376,
      "step": 652,
      "train_runtime": 1476.6561,
      "train_tokens_per_second": 334.117
    },
    {
      "epoch": 0.06004597701149425,
      "grad_norm": 6.322948455810547,
      "learning_rate": 1.3513784461152884e-05,
      "loss": 0.6217048168182373,
      "num_input_tokens_seen": 494208,
      "step": 653,
      "train_runtime": 1478.6082,
      "train_tokens_per_second": 334.239
    },
    {
      "epoch": 0.060137931034482756,
      "grad_norm": 6.643691062927246,
      "learning_rate": 1.350375939849624e-05,
      "loss": 0.4755760729312897,
      "num_input_tokens_seen": 494820,
      "step": 654,
      "train_runtime": 1480.5004,
      "train_tokens_per_second": 334.225
    },
    {
      "epoch": 0.06022988505747127,
      "grad_norm": 6.905515670776367,
      "learning_rate": 1.3493734335839601e-05,
      "loss": 0.5002609491348267,
      "num_input_tokens_seen": 495404,
      "step": 655,
      "train_runtime": 1482.3805,
      "train_tokens_per_second": 334.195
    },
    {
      "epoch": 0.06032183908045977,
      "grad_norm": 7.0027666091918945,
      "learning_rate": 1.348370927318296e-05,
      "loss": 0.5450930595397949,
      "num_input_tokens_seen": 496092,
      "step": 656,
      "train_runtime": 1484.2702,
      "train_tokens_per_second": 334.233
    },
    {
      "epoch": 0.06041379310344828,
      "grad_norm": 6.325944900512695,
      "learning_rate": 1.3473684210526316e-05,
      "loss": 0.4743611812591553,
      "num_input_tokens_seen": 496636,
      "step": 657,
      "train_runtime": 1486.1459,
      "train_tokens_per_second": 334.177
    },
    {
      "epoch": 0.06050574712643678,
      "grad_norm": 7.365074157714844,
      "learning_rate": 1.3463659147869675e-05,
      "loss": 0.4588896632194519,
      "num_input_tokens_seen": 497492,
      "step": 658,
      "train_runtime": 1488.033,
      "train_tokens_per_second": 334.329
    },
    {
      "epoch": 0.06059770114942529,
      "grad_norm": 5.5434088706970215,
      "learning_rate": 1.3453634085213035e-05,
      "loss": 0.43437156081199646,
      "num_input_tokens_seen": 499976,
      "step": 659,
      "train_runtime": 1490.1784,
      "train_tokens_per_second": 335.514
    },
    {
      "epoch": 0.060689655172413794,
      "grad_norm": 6.805002689361572,
      "learning_rate": 1.3443609022556392e-05,
      "loss": 0.44768649339675903,
      "num_input_tokens_seen": 500716,
      "step": 660,
      "train_runtime": 1492.0889,
      "train_tokens_per_second": 335.581
    },
    {
      "epoch": 0.0607816091954023,
      "grad_norm": 7.561191082000732,
      "learning_rate": 1.343358395989975e-05,
      "loss": 0.5202498435974121,
      "num_input_tokens_seen": 501336,
      "step": 661,
      "train_runtime": 1500.5241,
      "train_tokens_per_second": 334.107
    },
    {
      "epoch": 0.060873563218390804,
      "grad_norm": 7.143009185791016,
      "learning_rate": 1.342355889724311e-05,
      "loss": 0.46148866415023804,
      "num_input_tokens_seen": 501952,
      "step": 662,
      "train_runtime": 1502.5227,
      "train_tokens_per_second": 334.073
    },
    {
      "epoch": 0.06096551724137931,
      "grad_norm": 6.5299859046936035,
      "learning_rate": 1.3413533834586467e-05,
      "loss": 0.4477870464324951,
      "num_input_tokens_seen": 502632,
      "step": 663,
      "train_runtime": 1504.4094,
      "train_tokens_per_second": 334.106
    },
    {
      "epoch": 0.061057471264367814,
      "grad_norm": 7.125858306884766,
      "learning_rate": 1.3403508771929826e-05,
      "loss": 0.6004577875137329,
      "num_input_tokens_seen": 503324,
      "step": 664,
      "train_runtime": 1506.3018,
      "train_tokens_per_second": 334.146
    },
    {
      "epoch": 0.06114942528735632,
      "grad_norm": 7.250263690948486,
      "learning_rate": 1.3393483709273186e-05,
      "loss": 0.4981212913990021,
      "num_input_tokens_seen": 503956,
      "step": 665,
      "train_runtime": 1508.1885,
      "train_tokens_per_second": 334.147
    },
    {
      "epoch": 0.061241379310344825,
      "grad_norm": 7.098365306854248,
      "learning_rate": 1.3383458646616543e-05,
      "loss": 0.510361909866333,
      "num_input_tokens_seen": 504520,
      "step": 666,
      "train_runtime": 1510.085,
      "train_tokens_per_second": 334.1
    },
    {
      "epoch": 0.06133333333333333,
      "grad_norm": 6.275176525115967,
      "learning_rate": 1.3373433583959901e-05,
      "loss": 0.4923957884311676,
      "num_input_tokens_seen": 505248,
      "step": 667,
      "train_runtime": 1511.9878,
      "train_tokens_per_second": 334.161
    },
    {
      "epoch": 0.06142528735632184,
      "grad_norm": 6.961394309997559,
      "learning_rate": 1.3363408521303258e-05,
      "loss": 0.48068487644195557,
      "num_input_tokens_seen": 505916,
      "step": 668,
      "train_runtime": 1513.8834,
      "train_tokens_per_second": 334.184
    },
    {
      "epoch": 0.06151724137931035,
      "grad_norm": 6.190309524536133,
      "learning_rate": 1.3353383458646618e-05,
      "loss": 0.4339936077594757,
      "num_input_tokens_seen": 506564,
      "step": 669,
      "train_runtime": 1515.7905,
      "train_tokens_per_second": 334.191
    },
    {
      "epoch": 0.06160919540229885,
      "grad_norm": 7.2936110496521,
      "learning_rate": 1.3343358395989977e-05,
      "loss": 0.5688676238059998,
      "num_input_tokens_seen": 507416,
      "step": 670,
      "train_runtime": 1517.72,
      "train_tokens_per_second": 334.328
    },
    {
      "epoch": 0.06170114942528736,
      "grad_norm": 8.014532089233398,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.47973912954330444,
      "num_input_tokens_seen": 508124,
      "step": 671,
      "train_runtime": 1519.6395,
      "train_tokens_per_second": 334.371
    },
    {
      "epoch": 0.06179310344827586,
      "grad_norm": 7.369278907775879,
      "learning_rate": 1.3323308270676693e-05,
      "loss": 0.5662002563476562,
      "num_input_tokens_seen": 508928,
      "step": 672,
      "train_runtime": 1521.5806,
      "train_tokens_per_second": 334.473
    },
    {
      "epoch": 0.06188505747126437,
      "grad_norm": 6.326199531555176,
      "learning_rate": 1.3313283208020052e-05,
      "loss": 0.5147907733917236,
      "num_input_tokens_seen": 509632,
      "step": 673,
      "train_runtime": 1523.4741,
      "train_tokens_per_second": 334.52
    },
    {
      "epoch": 0.06197701149425287,
      "grad_norm": 6.8698015213012695,
      "learning_rate": 1.3303258145363409e-05,
      "loss": 0.521502673625946,
      "num_input_tokens_seen": 510564,
      "step": 674,
      "train_runtime": 1525.4458,
      "train_tokens_per_second": 334.698
    },
    {
      "epoch": 0.06206896551724138,
      "grad_norm": 7.055248260498047,
      "learning_rate": 1.3293233082706769e-05,
      "loss": 0.49516522884368896,
      "num_input_tokens_seen": 511180,
      "step": 675,
      "train_runtime": 1527.3351,
      "train_tokens_per_second": 334.688
    },
    {
      "epoch": 0.062160919540229884,
      "grad_norm": 7.518498420715332,
      "learning_rate": 1.3283208020050127e-05,
      "loss": 0.5852192044258118,
      "num_input_tokens_seen": 512044,
      "step": 676,
      "train_runtime": 1529.2661,
      "train_tokens_per_second": 334.83
    },
    {
      "epoch": 0.06225287356321839,
      "grad_norm": 5.807666301727295,
      "learning_rate": 1.3273182957393484e-05,
      "loss": 0.49576497077941895,
      "num_input_tokens_seen": 513420,
      "step": 677,
      "train_runtime": 1531.2502,
      "train_tokens_per_second": 335.295
    },
    {
      "epoch": 0.062344827586206894,
      "grad_norm": 8.027884483337402,
      "learning_rate": 1.3263157894736843e-05,
      "loss": 0.5500902533531189,
      "num_input_tokens_seen": 514204,
      "step": 678,
      "train_runtime": 1533.1688,
      "train_tokens_per_second": 335.386
    },
    {
      "epoch": 0.0624367816091954,
      "grad_norm": 6.813953399658203,
      "learning_rate": 1.3253132832080203e-05,
      "loss": 0.4119127690792084,
      "num_input_tokens_seen": 514844,
      "step": 679,
      "train_runtime": 1535.0529,
      "train_tokens_per_second": 335.392
    },
    {
      "epoch": 0.0625287356321839,
      "grad_norm": 7.3369011878967285,
      "learning_rate": 1.324310776942356e-05,
      "loss": 0.49944669008255005,
      "num_input_tokens_seen": 515852,
      "step": 680,
      "train_runtime": 1537.0046,
      "train_tokens_per_second": 335.622
    },
    {
      "epoch": 0.06262068965517241,
      "grad_norm": 6.819929122924805,
      "learning_rate": 1.3233082706766918e-05,
      "loss": 0.46720486879348755,
      "num_input_tokens_seen": 516672,
      "step": 681,
      "train_runtime": 1538.9407,
      "train_tokens_per_second": 335.732
    },
    {
      "epoch": 0.06271264367816091,
      "grad_norm": 7.631635665893555,
      "learning_rate": 1.3223057644110278e-05,
      "loss": 0.5723601579666138,
      "num_input_tokens_seen": 517516,
      "step": 682,
      "train_runtime": 1540.8935,
      "train_tokens_per_second": 335.854
    },
    {
      "epoch": 0.06280459770114942,
      "grad_norm": 6.981370449066162,
      "learning_rate": 1.3213032581453635e-05,
      "loss": 0.5557063221931458,
      "num_input_tokens_seen": 518096,
      "step": 683,
      "train_runtime": 1542.7841,
      "train_tokens_per_second": 335.819
    },
    {
      "epoch": 0.06289655172413793,
      "grad_norm": 6.457602500915527,
      "learning_rate": 1.3203007518796994e-05,
      "loss": 0.46246883273124695,
      "num_input_tokens_seen": 518796,
      "step": 684,
      "train_runtime": 1544.6934,
      "train_tokens_per_second": 335.857
    },
    {
      "epoch": 0.06298850574712643,
      "grad_norm": 7.061386585235596,
      "learning_rate": 1.3192982456140354e-05,
      "loss": 0.5143232345581055,
      "num_input_tokens_seen": 519392,
      "step": 685,
      "train_runtime": 1546.5914,
      "train_tokens_per_second": 335.83
    },
    {
      "epoch": 0.06308045977011495,
      "grad_norm": 7.075534343719482,
      "learning_rate": 1.318295739348371e-05,
      "loss": 0.42988720536231995,
      "num_input_tokens_seen": 520132,
      "step": 686,
      "train_runtime": 1548.4884,
      "train_tokens_per_second": 335.897
    },
    {
      "epoch": 0.06317241379310345,
      "grad_norm": 6.861327171325684,
      "learning_rate": 1.3172932330827069e-05,
      "loss": 0.5197246670722961,
      "num_input_tokens_seen": 520916,
      "step": 687,
      "train_runtime": 1550.4232,
      "train_tokens_per_second": 335.983
    },
    {
      "epoch": 0.06326436781609196,
      "grad_norm": 9.0070219039917,
      "learning_rate": 1.3162907268170426e-05,
      "loss": 0.5759735107421875,
      "num_input_tokens_seen": 521644,
      "step": 688,
      "train_runtime": 1552.3332,
      "train_tokens_per_second": 336.039
    },
    {
      "epoch": 0.06335632183908046,
      "grad_norm": 7.550922393798828,
      "learning_rate": 1.3152882205513786e-05,
      "loss": 0.4897928535938263,
      "num_input_tokens_seen": 522600,
      "step": 689,
      "train_runtime": 1554.3084,
      "train_tokens_per_second": 336.227
    },
    {
      "epoch": 0.06344827586206897,
      "grad_norm": 6.247206211090088,
      "learning_rate": 1.3142857142857145e-05,
      "loss": 0.5541774034500122,
      "num_input_tokens_seen": 523932,
      "step": 690,
      "train_runtime": 1556.3604,
      "train_tokens_per_second": 336.639
    },
    {
      "epoch": 0.06354022988505748,
      "grad_norm": 7.213705539703369,
      "learning_rate": 1.3132832080200501e-05,
      "loss": 0.4968850016593933,
      "num_input_tokens_seen": 524616,
      "step": 691,
      "train_runtime": 1565.1313,
      "train_tokens_per_second": 335.19
    },
    {
      "epoch": 0.06363218390804598,
      "grad_norm": 6.573073387145996,
      "learning_rate": 1.3122807017543862e-05,
      "loss": 0.4320947527885437,
      "num_input_tokens_seen": 525172,
      "step": 692,
      "train_runtime": 1567.1796,
      "train_tokens_per_second": 335.106
    },
    {
      "epoch": 0.06372413793103449,
      "grad_norm": 6.980879306793213,
      "learning_rate": 1.311278195488722e-05,
      "loss": 0.502927303314209,
      "num_input_tokens_seen": 525796,
      "step": 693,
      "train_runtime": 1569.0886,
      "train_tokens_per_second": 335.096
    },
    {
      "epoch": 0.06381609195402299,
      "grad_norm": 7.194272041320801,
      "learning_rate": 1.3102756892230577e-05,
      "loss": 0.4879305064678192,
      "num_input_tokens_seen": 526352,
      "step": 694,
      "train_runtime": 1571.0386,
      "train_tokens_per_second": 335.034
    },
    {
      "epoch": 0.0639080459770115,
      "grad_norm": 5.791599750518799,
      "learning_rate": 1.3092731829573937e-05,
      "loss": 0.4360908269882202,
      "num_input_tokens_seen": 526888,
      "step": 695,
      "train_runtime": 1572.9798,
      "train_tokens_per_second": 334.962
    },
    {
      "epoch": 0.064,
      "grad_norm": 7.671477317810059,
      "learning_rate": 1.3082706766917295e-05,
      "loss": 0.5835292935371399,
      "num_input_tokens_seen": 527900,
      "step": 696,
      "train_runtime": 1574.9703,
      "train_tokens_per_second": 335.181
    },
    {
      "epoch": 0.0640919540229885,
      "grad_norm": 7.398995876312256,
      "learning_rate": 1.3072681704260652e-05,
      "loss": 0.5330015420913696,
      "num_input_tokens_seen": 528808,
      "step": 697,
      "train_runtime": 1576.9475,
      "train_tokens_per_second": 335.336
    },
    {
      "epoch": 0.06418390804597701,
      "grad_norm": 6.887153148651123,
      "learning_rate": 1.306265664160401e-05,
      "loss": 0.6172720789909363,
      "num_input_tokens_seen": 529544,
      "step": 698,
      "train_runtime": 1578.9174,
      "train_tokens_per_second": 335.384
    },
    {
      "epoch": 0.06427586206896552,
      "grad_norm": 6.892273426055908,
      "learning_rate": 1.305263157894737e-05,
      "loss": 0.505545437335968,
      "num_input_tokens_seen": 530676,
      "step": 699,
      "train_runtime": 1580.9355,
      "train_tokens_per_second": 335.672
    },
    {
      "epoch": 0.06436781609195402,
      "grad_norm": 6.331185817718506,
      "learning_rate": 1.3042606516290728e-05,
      "loss": 0.4878854751586914,
      "num_input_tokens_seen": 531472,
      "step": 700,
      "train_runtime": 1582.9317,
      "train_tokens_per_second": 335.752
    },
    {
      "epoch": 0.06445977011494253,
      "grad_norm": 6.539681911468506,
      "learning_rate": 1.3032581453634086e-05,
      "loss": 0.5967066287994385,
      "num_input_tokens_seen": 532592,
      "step": 701,
      "train_runtime": 1585.0363,
      "train_tokens_per_second": 336.012
    },
    {
      "epoch": 0.06455172413793103,
      "grad_norm": 7.155969142913818,
      "learning_rate": 1.3022556390977445e-05,
      "loss": 0.5363156795501709,
      "num_input_tokens_seen": 533604,
      "step": 702,
      "train_runtime": 1587.0372,
      "train_tokens_per_second": 336.227
    },
    {
      "epoch": 0.06464367816091954,
      "grad_norm": 7.249508857727051,
      "learning_rate": 1.3012531328320803e-05,
      "loss": 0.570030927658081,
      "num_input_tokens_seen": 534604,
      "step": 703,
      "train_runtime": 1589.0663,
      "train_tokens_per_second": 336.426
    },
    {
      "epoch": 0.06473563218390804,
      "grad_norm": 6.429741859436035,
      "learning_rate": 1.3002506265664162e-05,
      "loss": 0.5177555084228516,
      "num_input_tokens_seen": 535212,
      "step": 704,
      "train_runtime": 1591.0705,
      "train_tokens_per_second": 336.385
    },
    {
      "epoch": 0.06482758620689655,
      "grad_norm": 6.571535110473633,
      "learning_rate": 1.299248120300752e-05,
      "loss": 0.584343433380127,
      "num_input_tokens_seen": 535956,
      "step": 705,
      "train_runtime": 1593.0127,
      "train_tokens_per_second": 336.442
    },
    {
      "epoch": 0.06491954022988505,
      "grad_norm": 7.391951084136963,
      "learning_rate": 1.2982456140350879e-05,
      "loss": 0.49518847465515137,
      "num_input_tokens_seen": 536788,
      "step": 706,
      "train_runtime": 1594.987,
      "train_tokens_per_second": 336.547
    },
    {
      "epoch": 0.06501149425287356,
      "grad_norm": 7.156089782714844,
      "learning_rate": 1.2972431077694237e-05,
      "loss": 0.48724111914634705,
      "num_input_tokens_seen": 537544,
      "step": 707,
      "train_runtime": 1596.9956,
      "train_tokens_per_second": 336.597
    },
    {
      "epoch": 0.06510344827586206,
      "grad_norm": 6.625319004058838,
      "learning_rate": 1.2962406015037594e-05,
      "loss": 0.47585195302963257,
      "num_input_tokens_seen": 538064,
      "step": 708,
      "train_runtime": 1598.9342,
      "train_tokens_per_second": 336.514
    },
    {
      "epoch": 0.06519540229885057,
      "grad_norm": 7.106809616088867,
      "learning_rate": 1.2952380952380954e-05,
      "loss": 0.520663857460022,
      "num_input_tokens_seen": 538956,
      "step": 709,
      "train_runtime": 1600.9127,
      "train_tokens_per_second": 336.655
    },
    {
      "epoch": 0.06528735632183907,
      "grad_norm": 6.606273174285889,
      "learning_rate": 1.2942355889724313e-05,
      "loss": 0.4652138948440552,
      "num_input_tokens_seen": 539504,
      "step": 710,
      "train_runtime": 1602.8735,
      "train_tokens_per_second": 336.586
    },
    {
      "epoch": 0.0653793103448276,
      "grad_norm": 6.744972229003906,
      "learning_rate": 1.293233082706767e-05,
      "loss": 0.5241913199424744,
      "num_input_tokens_seen": 540496,
      "step": 711,
      "train_runtime": 1604.9125,
      "train_tokens_per_second": 336.776
    },
    {
      "epoch": 0.0654712643678161,
      "grad_norm": 7.374359130859375,
      "learning_rate": 1.292230576441103e-05,
      "loss": 0.5033762454986572,
      "num_input_tokens_seen": 541140,
      "step": 712,
      "train_runtime": 1606.8829,
      "train_tokens_per_second": 336.764
    },
    {
      "epoch": 0.0655632183908046,
      "grad_norm": 6.294163227081299,
      "learning_rate": 1.2912280701754386e-05,
      "loss": 0.4169749617576599,
      "num_input_tokens_seen": 541992,
      "step": 713,
      "train_runtime": 1608.8905,
      "train_tokens_per_second": 336.873
    },
    {
      "epoch": 0.06565517241379311,
      "grad_norm": 6.865566253662109,
      "learning_rate": 1.2902255639097745e-05,
      "loss": 0.5325437188148499,
      "num_input_tokens_seen": 543256,
      "step": 714,
      "train_runtime": 1610.9148,
      "train_tokens_per_second": 337.234
    },
    {
      "epoch": 0.06574712643678161,
      "grad_norm": 6.749205589294434,
      "learning_rate": 1.2892230576441105e-05,
      "loss": 0.47279292345046997,
      "num_input_tokens_seen": 543916,
      "step": 715,
      "train_runtime": 1612.8835,
      "train_tokens_per_second": 337.232
    },
    {
      "epoch": 0.06583908045977012,
      "grad_norm": 6.720696449279785,
      "learning_rate": 1.2882205513784462e-05,
      "loss": 0.48280414938926697,
      "num_input_tokens_seen": 544688,
      "step": 716,
      "train_runtime": 1614.8576,
      "train_tokens_per_second": 337.298
    },
    {
      "epoch": 0.06593103448275862,
      "grad_norm": 8.794879913330078,
      "learning_rate": 1.287218045112782e-05,
      "loss": 0.5291348099708557,
      "num_input_tokens_seen": 545472,
      "step": 717,
      "train_runtime": 1616.8019,
      "train_tokens_per_second": 337.377
    },
    {
      "epoch": 0.06602298850574713,
      "grad_norm": 6.2707719802856445,
      "learning_rate": 1.2862155388471179e-05,
      "loss": 0.4864928126335144,
      "num_input_tokens_seen": 546060,
      "step": 718,
      "train_runtime": 1618.7249,
      "train_tokens_per_second": 337.34
    },
    {
      "epoch": 0.06611494252873563,
      "grad_norm": 7.030215740203857,
      "learning_rate": 1.2852130325814537e-05,
      "loss": 0.4947786331176758,
      "num_input_tokens_seen": 546808,
      "step": 719,
      "train_runtime": 1620.6869,
      "train_tokens_per_second": 337.393
    },
    {
      "epoch": 0.06620689655172414,
      "grad_norm": 6.65582799911499,
      "learning_rate": 1.2842105263157896e-05,
      "loss": 0.5161091089248657,
      "num_input_tokens_seen": 547440,
      "step": 720,
      "train_runtime": 1622.6139,
      "train_tokens_per_second": 337.382
    },
    {
      "epoch": 0.06629885057471264,
      "grad_norm": 6.280404090881348,
      "learning_rate": 1.2832080200501254e-05,
      "loss": 0.44713616371154785,
      "num_input_tokens_seen": 548160,
      "step": 721,
      "train_runtime": 1631.4892,
      "train_tokens_per_second": 335.988
    },
    {
      "epoch": 0.06639080459770115,
      "grad_norm": 7.807005882263184,
      "learning_rate": 1.2822055137844613e-05,
      "loss": 0.5657007694244385,
      "num_input_tokens_seen": 548812,
      "step": 722,
      "train_runtime": 1633.4444,
      "train_tokens_per_second": 335.984
    },
    {
      "epoch": 0.06648275862068966,
      "grad_norm": 6.831088066101074,
      "learning_rate": 1.2812030075187971e-05,
      "loss": 0.429911345243454,
      "num_input_tokens_seen": 549380,
      "step": 723,
      "train_runtime": 1635.368,
      "train_tokens_per_second": 335.937
    },
    {
      "epoch": 0.06657471264367816,
      "grad_norm": 5.718352317810059,
      "learning_rate": 1.2802005012531328e-05,
      "loss": 0.412373423576355,
      "num_input_tokens_seen": 551236,
      "step": 724,
      "train_runtime": 1637.4296,
      "train_tokens_per_second": 336.647
    },
    {
      "epoch": 0.06666666666666667,
      "grad_norm": 7.15751314163208,
      "learning_rate": 1.2791979949874688e-05,
      "loss": 0.40920761227607727,
      "num_input_tokens_seen": 551944,
      "step": 725,
      "train_runtime": 1639.3586,
      "train_tokens_per_second": 336.683
    },
    {
      "epoch": 0.06675862068965517,
      "grad_norm": 6.765956401824951,
      "learning_rate": 1.2781954887218047e-05,
      "loss": 0.4629984498023987,
      "num_input_tokens_seen": 553276,
      "step": 726,
      "train_runtime": 1641.4808,
      "train_tokens_per_second": 337.059
    },
    {
      "epoch": 0.06685057471264368,
      "grad_norm": 7.850322246551514,
      "learning_rate": 1.2771929824561404e-05,
      "loss": 0.46668586134910583,
      "num_input_tokens_seen": 553928,
      "step": 727,
      "train_runtime": 1643.4299,
      "train_tokens_per_second": 337.056
    },
    {
      "epoch": 0.06694252873563218,
      "grad_norm": 6.204847812652588,
      "learning_rate": 1.2761904761904762e-05,
      "loss": 0.45401400327682495,
      "num_input_tokens_seen": 554452,
      "step": 728,
      "train_runtime": 1645.3516,
      "train_tokens_per_second": 336.981
    },
    {
      "epoch": 0.06703448275862069,
      "grad_norm": 7.187865257263184,
      "learning_rate": 1.2751879699248122e-05,
      "loss": 0.4337291717529297,
      "num_input_tokens_seen": 555292,
      "step": 729,
      "train_runtime": 1647.3034,
      "train_tokens_per_second": 337.092
    },
    {
      "epoch": 0.06712643678160919,
      "grad_norm": 6.8617844581604,
      "learning_rate": 1.2741854636591479e-05,
      "loss": 0.5119789242744446,
      "num_input_tokens_seen": 555944,
      "step": 730,
      "train_runtime": 1649.2402,
      "train_tokens_per_second": 337.091
    },
    {
      "epoch": 0.0672183908045977,
      "grad_norm": 6.7105841636657715,
      "learning_rate": 1.2731829573934837e-05,
      "loss": 0.5011810064315796,
      "num_input_tokens_seen": 556584,
      "step": 731,
      "train_runtime": 1651.1496,
      "train_tokens_per_second": 337.089
    },
    {
      "epoch": 0.0673103448275862,
      "grad_norm": 7.510727882385254,
      "learning_rate": 1.2721804511278198e-05,
      "loss": 0.6270932555198669,
      "num_input_tokens_seen": 557844,
      "step": 732,
      "train_runtime": 1653.1757,
      "train_tokens_per_second": 337.438
    },
    {
      "epoch": 0.06740229885057471,
      "grad_norm": 6.8630595207214355,
      "learning_rate": 1.2711779448621554e-05,
      "loss": 0.4799724817276001,
      "num_input_tokens_seen": 558528,
      "step": 733,
      "train_runtime": 1655.108,
      "train_tokens_per_second": 337.457
    },
    {
      "epoch": 0.06749425287356321,
      "grad_norm": 6.472028732299805,
      "learning_rate": 1.2701754385964913e-05,
      "loss": 0.5073041915893555,
      "num_input_tokens_seen": 559064,
      "step": 734,
      "train_runtime": 1657.103,
      "train_tokens_per_second": 337.374
    },
    {
      "epoch": 0.06758620689655172,
      "grad_norm": 6.790537357330322,
      "learning_rate": 1.2691729323308273e-05,
      "loss": 0.6256628632545471,
      "num_input_tokens_seen": 560108,
      "step": 735,
      "train_runtime": 1659.1651,
      "train_tokens_per_second": 337.584
    },
    {
      "epoch": 0.06767816091954024,
      "grad_norm": 7.035492897033691,
      "learning_rate": 1.268170426065163e-05,
      "loss": 0.4381142556667328,
      "num_input_tokens_seen": 560728,
      "step": 736,
      "train_runtime": 1661.0869,
      "train_tokens_per_second": 337.567
    },
    {
      "epoch": 0.06777011494252874,
      "grad_norm": 7.550258159637451,
      "learning_rate": 1.2671679197994988e-05,
      "loss": 0.46850088238716125,
      "num_input_tokens_seen": 561380,
      "step": 737,
      "train_runtime": 1663.0278,
      "train_tokens_per_second": 337.565
    },
    {
      "epoch": 0.06786206896551725,
      "grad_norm": 6.153674125671387,
      "learning_rate": 1.2661654135338345e-05,
      "loss": 0.45343711972236633,
      "num_input_tokens_seen": 561984,
      "step": 738,
      "train_runtime": 1664.9469,
      "train_tokens_per_second": 337.539
    },
    {
      "epoch": 0.06795402298850575,
      "grad_norm": 7.199565410614014,
      "learning_rate": 1.2651629072681705e-05,
      "loss": 0.4830615520477295,
      "num_input_tokens_seen": 562680,
      "step": 739,
      "train_runtime": 1666.8798,
      "train_tokens_per_second": 337.565
    },
    {
      "epoch": 0.06804597701149426,
      "grad_norm": 7.291567325592041,
      "learning_rate": 1.2641604010025064e-05,
      "loss": 0.5234308242797852,
      "num_input_tokens_seen": 563344,
      "step": 740,
      "train_runtime": 1668.8178,
      "train_tokens_per_second": 337.571
    },
    {
      "epoch": 0.06813793103448276,
      "grad_norm": 7.292444705963135,
      "learning_rate": 1.263157894736842e-05,
      "loss": 0.565500020980835,
      "num_input_tokens_seen": 564068,
      "step": 741,
      "train_runtime": 1670.7482,
      "train_tokens_per_second": 337.614
    },
    {
      "epoch": 0.06822988505747127,
      "grad_norm": 7.037001132965088,
      "learning_rate": 1.262155388471178e-05,
      "loss": 0.46853190660476685,
      "num_input_tokens_seen": 564764,
      "step": 742,
      "train_runtime": 1672.7809,
      "train_tokens_per_second": 337.62
    },
    {
      "epoch": 0.06832183908045977,
      "grad_norm": 6.376340389251709,
      "learning_rate": 1.261152882205514e-05,
      "loss": 0.5655608177185059,
      "num_input_tokens_seen": 565700,
      "step": 743,
      "train_runtime": 1674.7417,
      "train_tokens_per_second": 337.783
    },
    {
      "epoch": 0.06841379310344828,
      "grad_norm": 7.177804470062256,
      "learning_rate": 1.2601503759398496e-05,
      "loss": 0.49037617444992065,
      "num_input_tokens_seen": 566368,
      "step": 744,
      "train_runtime": 1676.6737,
      "train_tokens_per_second": 337.793
    },
    {
      "epoch": 0.06850574712643678,
      "grad_norm": 6.068066596984863,
      "learning_rate": 1.2591478696741856e-05,
      "loss": 0.39159297943115234,
      "num_input_tokens_seen": 566976,
      "step": 745,
      "train_runtime": 1678.5987,
      "train_tokens_per_second": 337.767
    },
    {
      "epoch": 0.06859770114942529,
      "grad_norm": 6.261608600616455,
      "learning_rate": 1.2581453634085215e-05,
      "loss": 0.5073766708374023,
      "num_input_tokens_seen": 568312,
      "step": 746,
      "train_runtime": 1680.6311,
      "train_tokens_per_second": 338.154
    },
    {
      "epoch": 0.0686896551724138,
      "grad_norm": 8.0780668258667,
      "learning_rate": 1.2571428571428572e-05,
      "loss": 0.6464099287986755,
      "num_input_tokens_seen": 568928,
      "step": 747,
      "train_runtime": 1682.559,
      "train_tokens_per_second": 338.133
    },
    {
      "epoch": 0.0687816091954023,
      "grad_norm": 6.58332633972168,
      "learning_rate": 1.256140350877193e-05,
      "loss": 0.43471819162368774,
      "num_input_tokens_seen": 569544,
      "step": 748,
      "train_runtime": 1684.4879,
      "train_tokens_per_second": 338.111
    },
    {
      "epoch": 0.0688735632183908,
      "grad_norm": 6.317200183868408,
      "learning_rate": 1.255137844611529e-05,
      "loss": 0.4880402982234955,
      "num_input_tokens_seen": 570352,
      "step": 749,
      "train_runtime": 1686.4616,
      "train_tokens_per_second": 338.194
    },
    {
      "epoch": 0.06896551724137931,
      "grad_norm": 6.550645351409912,
      "learning_rate": 1.2541353383458647e-05,
      "loss": 0.5459703803062439,
      "num_input_tokens_seen": 571376,
      "step": 750,
      "train_runtime": 1688.4486,
      "train_tokens_per_second": 338.403
    },
    {
      "epoch": 0.06905747126436781,
      "grad_norm": 6.884756565093994,
      "learning_rate": 1.2531328320802006e-05,
      "loss": 0.4800689220428467,
      "num_input_tokens_seen": 572200,
      "step": 751,
      "train_runtime": 1697.3154,
      "train_tokens_per_second": 337.121
    },
    {
      "epoch": 0.06914942528735632,
      "grad_norm": 7.125977993011475,
      "learning_rate": 1.2521303258145366e-05,
      "loss": 0.5538926124572754,
      "num_input_tokens_seen": 572980,
      "step": 752,
      "train_runtime": 1699.3056,
      "train_tokens_per_second": 337.185
    },
    {
      "epoch": 0.06924137931034482,
      "grad_norm": 6.5595574378967285,
      "learning_rate": 1.2511278195488723e-05,
      "loss": 0.5036803483963013,
      "num_input_tokens_seen": 573712,
      "step": 753,
      "train_runtime": 1701.2344,
      "train_tokens_per_second": 337.233
    },
    {
      "epoch": 0.06933333333333333,
      "grad_norm": 6.945261478424072,
      "learning_rate": 1.2501253132832081e-05,
      "loss": 0.42304864525794983,
      "num_input_tokens_seen": 574240,
      "step": 754,
      "train_runtime": 1703.1613,
      "train_tokens_per_second": 337.161
    },
    {
      "epoch": 0.06942528735632184,
      "grad_norm": 8.34179973602295,
      "learning_rate": 1.2491228070175441e-05,
      "loss": 0.5464489459991455,
      "num_input_tokens_seen": 575192,
      "step": 755,
      "train_runtime": 1705.2184,
      "train_tokens_per_second": 337.313
    },
    {
      "epoch": 0.06951724137931034,
      "grad_norm": 7.022257328033447,
      "learning_rate": 1.2481203007518798e-05,
      "loss": 0.5530050992965698,
      "num_input_tokens_seen": 576100,
      "step": 756,
      "train_runtime": 1707.2174,
      "train_tokens_per_second": 337.45
    },
    {
      "epoch": 0.06960919540229885,
      "grad_norm": 6.397459506988525,
      "learning_rate": 1.2471177944862156e-05,
      "loss": 0.46334466338157654,
      "num_input_tokens_seen": 576744,
      "step": 757,
      "train_runtime": 1709.144,
      "train_tokens_per_second": 337.446
    },
    {
      "epoch": 0.06970114942528735,
      "grad_norm": 6.368110179901123,
      "learning_rate": 1.2461152882205513e-05,
      "loss": 0.43705421686172485,
      "num_input_tokens_seen": 577436,
      "step": 758,
      "train_runtime": 1711.1179,
      "train_tokens_per_second": 337.461
    },
    {
      "epoch": 0.06979310344827586,
      "grad_norm": 7.558487892150879,
      "learning_rate": 1.2451127819548873e-05,
      "loss": 0.4765796661376953,
      "num_input_tokens_seen": 578072,
      "step": 759,
      "train_runtime": 1713.0453,
      "train_tokens_per_second": 337.453
    },
    {
      "epoch": 0.06988505747126436,
      "grad_norm": 7.0516533851623535,
      "learning_rate": 1.2441102756892232e-05,
      "loss": 0.5368772745132446,
      "num_input_tokens_seen": 578868,
      "step": 760,
      "train_runtime": 1714.9924,
      "train_tokens_per_second": 337.534
    },
    {
      "epoch": 0.06997701149425288,
      "grad_norm": 6.277525424957275,
      "learning_rate": 1.2431077694235589e-05,
      "loss": 0.36886775493621826,
      "num_input_tokens_seen": 579620,
      "step": 761,
      "train_runtime": 1716.952,
      "train_tokens_per_second": 337.587
    },
    {
      "epoch": 0.07006896551724139,
      "grad_norm": 7.795188903808594,
      "learning_rate": 1.2421052631578949e-05,
      "loss": 0.4997369050979614,
      "num_input_tokens_seen": 580360,
      "step": 762,
      "train_runtime": 1718.9349,
      "train_tokens_per_second": 337.628
    },
    {
      "epoch": 0.07016091954022989,
      "grad_norm": 6.9216132164001465,
      "learning_rate": 1.2411027568922307e-05,
      "loss": 0.4790230393409729,
      "num_input_tokens_seen": 580980,
      "step": 763,
      "train_runtime": 1720.8851,
      "train_tokens_per_second": 337.605
    },
    {
      "epoch": 0.0702528735632184,
      "grad_norm": 6.596647262573242,
      "learning_rate": 1.2401002506265664e-05,
      "loss": 0.4051140546798706,
      "num_input_tokens_seen": 581576,
      "step": 764,
      "train_runtime": 1722.8138,
      "train_tokens_per_second": 337.573
    },
    {
      "epoch": 0.0703448275862069,
      "grad_norm": 7.1984076499938965,
      "learning_rate": 1.2390977443609024e-05,
      "loss": 0.4946567714214325,
      "num_input_tokens_seen": 582340,
      "step": 765,
      "train_runtime": 1724.7467,
      "train_tokens_per_second": 337.638
    },
    {
      "epoch": 0.0704367816091954,
      "grad_norm": 6.350241661071777,
      "learning_rate": 1.2380952380952383e-05,
      "loss": 0.5237550735473633,
      "num_input_tokens_seen": 583264,
      "step": 766,
      "train_runtime": 1726.7142,
      "train_tokens_per_second": 337.788
    },
    {
      "epoch": 0.07052873563218391,
      "grad_norm": 7.541025161743164,
      "learning_rate": 1.237092731829574e-05,
      "loss": 0.4568924605846405,
      "num_input_tokens_seen": 583896,
      "step": 767,
      "train_runtime": 1728.6403,
      "train_tokens_per_second": 337.778
    },
    {
      "epoch": 0.07062068965517242,
      "grad_norm": 6.652822017669678,
      "learning_rate": 1.2360902255639098e-05,
      "loss": 0.40945571660995483,
      "num_input_tokens_seen": 584680,
      "step": 768,
      "train_runtime": 1730.6044,
      "train_tokens_per_second": 337.847
    },
    {
      "epoch": 0.07071264367816092,
      "grad_norm": 6.930812835693359,
      "learning_rate": 1.2350877192982458e-05,
      "loss": 0.5204863548278809,
      "num_input_tokens_seen": 586168,
      "step": 769,
      "train_runtime": 1732.6709,
      "train_tokens_per_second": 338.303
    },
    {
      "epoch": 0.07080459770114943,
      "grad_norm": 7.797668933868408,
      "learning_rate": 1.2340852130325815e-05,
      "loss": 0.6980093717575073,
      "num_input_tokens_seen": 586836,
      "step": 770,
      "train_runtime": 1734.5988,
      "train_tokens_per_second": 338.312
    },
    {
      "epoch": 0.07089655172413793,
      "grad_norm": 6.9763898849487305,
      "learning_rate": 1.2330827067669174e-05,
      "loss": 0.4698995351791382,
      "num_input_tokens_seen": 588052,
      "step": 771,
      "train_runtime": 1736.5856,
      "train_tokens_per_second": 338.625
    },
    {
      "epoch": 0.07098850574712644,
      "grad_norm": 6.0939178466796875,
      "learning_rate": 1.2320802005012534e-05,
      "loss": 0.41389238834381104,
      "num_input_tokens_seen": 588796,
      "step": 772,
      "train_runtime": 1738.5425,
      "train_tokens_per_second": 338.672
    },
    {
      "epoch": 0.07108045977011494,
      "grad_norm": 6.996767520904541,
      "learning_rate": 1.231077694235589e-05,
      "loss": 0.4361867308616638,
      "num_input_tokens_seen": 589440,
      "step": 773,
      "train_runtime": 1740.4911,
      "train_tokens_per_second": 338.663
    },
    {
      "epoch": 0.07117241379310345,
      "grad_norm": 6.896433353424072,
      "learning_rate": 1.2300751879699249e-05,
      "loss": 0.4922274351119995,
      "num_input_tokens_seen": 590072,
      "step": 774,
      "train_runtime": 1742.4207,
      "train_tokens_per_second": 338.651
    },
    {
      "epoch": 0.07126436781609195,
      "grad_norm": 7.727783679962158,
      "learning_rate": 1.229072681704261e-05,
      "loss": 0.5629440546035767,
      "num_input_tokens_seen": 590904,
      "step": 775,
      "train_runtime": 1744.44,
      "train_tokens_per_second": 338.736
    },
    {
      "epoch": 0.07135632183908046,
      "grad_norm": 7.155002117156982,
      "learning_rate": 1.2280701754385966e-05,
      "loss": 0.4994778037071228,
      "num_input_tokens_seen": 591592,
      "step": 776,
      "train_runtime": 1746.3834,
      "train_tokens_per_second": 338.753
    },
    {
      "epoch": 0.07144827586206896,
      "grad_norm": 6.324461460113525,
      "learning_rate": 1.2270676691729325e-05,
      "loss": 0.3676263093948364,
      "num_input_tokens_seen": 592392,
      "step": 777,
      "train_runtime": 1748.3815,
      "train_tokens_per_second": 338.823
    },
    {
      "epoch": 0.07154022988505747,
      "grad_norm": 7.361177921295166,
      "learning_rate": 1.2260651629072681e-05,
      "loss": 0.5257224440574646,
      "num_input_tokens_seen": 593304,
      "step": 778,
      "train_runtime": 1750.3634,
      "train_tokens_per_second": 338.96
    },
    {
      "epoch": 0.07163218390804597,
      "grad_norm": 7.276780128479004,
      "learning_rate": 1.2250626566416041e-05,
      "loss": 0.4875470995903015,
      "num_input_tokens_seen": 593864,
      "step": 779,
      "train_runtime": 1752.2796,
      "train_tokens_per_second": 338.909
    },
    {
      "epoch": 0.07172413793103448,
      "grad_norm": 6.369472503662109,
      "learning_rate": 1.22406015037594e-05,
      "loss": 0.42024746537208557,
      "num_input_tokens_seen": 594428,
      "step": 780,
      "train_runtime": 1754.1979,
      "train_tokens_per_second": 338.86
    },
    {
      "epoch": 0.07181609195402298,
      "grad_norm": 6.957118988037109,
      "learning_rate": 1.2230576441102757e-05,
      "loss": 0.5289345979690552,
      "num_input_tokens_seen": 595080,
      "step": 781,
      "train_runtime": 1763.0355,
      "train_tokens_per_second": 337.531
    },
    {
      "epoch": 0.07190804597701149,
      "grad_norm": 6.785836696624756,
      "learning_rate": 1.2220551378446117e-05,
      "loss": 0.4399987757205963,
      "num_input_tokens_seen": 595692,
      "step": 782,
      "train_runtime": 1764.9625,
      "train_tokens_per_second": 337.51
    },
    {
      "epoch": 0.072,
      "grad_norm": 6.67493200302124,
      "learning_rate": 1.2210526315789475e-05,
      "loss": 0.4305074214935303,
      "num_input_tokens_seen": 596852,
      "step": 783,
      "train_runtime": 1766.9915,
      "train_tokens_per_second": 337.779
    },
    {
      "epoch": 0.0720919540229885,
      "grad_norm": 6.000429630279541,
      "learning_rate": 1.2200501253132832e-05,
      "loss": 0.45768213272094727,
      "num_input_tokens_seen": 597488,
      "step": 784,
      "train_runtime": 1768.9435,
      "train_tokens_per_second": 337.765
    },
    {
      "epoch": 0.072183908045977,
      "grad_norm": 5.904489517211914,
      "learning_rate": 1.2190476190476192e-05,
      "loss": 0.5369732975959778,
      "num_input_tokens_seen": 598436,
      "step": 785,
      "train_runtime": 1770.9035,
      "train_tokens_per_second": 337.927
    },
    {
      "epoch": 0.07227586206896552,
      "grad_norm": 7.443309783935547,
      "learning_rate": 1.2180451127819551e-05,
      "loss": 0.49624449014663696,
      "num_input_tokens_seen": 599000,
      "step": 786,
      "train_runtime": 1772.839,
      "train_tokens_per_second": 337.876
    },
    {
      "epoch": 0.07236781609195403,
      "grad_norm": 6.947629451751709,
      "learning_rate": 1.2170426065162908e-05,
      "loss": 0.47903862595558167,
      "num_input_tokens_seen": 599692,
      "step": 787,
      "train_runtime": 1774.7656,
      "train_tokens_per_second": 337.899
    },
    {
      "epoch": 0.07245977011494253,
      "grad_norm": 6.876119136810303,
      "learning_rate": 1.2160401002506266e-05,
      "loss": 0.5088704824447632,
      "num_input_tokens_seen": 600560,
      "step": 788,
      "train_runtime": 1776.7337,
      "train_tokens_per_second": 338.014
    },
    {
      "epoch": 0.07255172413793104,
      "grad_norm": 8.312402725219727,
      "learning_rate": 1.2150375939849626e-05,
      "loss": 0.46357420086860657,
      "num_input_tokens_seen": 601216,
      "step": 789,
      "train_runtime": 1778.7976,
      "train_tokens_per_second": 337.99
    },
    {
      "epoch": 0.07264367816091954,
      "grad_norm": 7.5309014320373535,
      "learning_rate": 1.2140350877192983e-05,
      "loss": 0.4875525236129761,
      "num_input_tokens_seen": 601828,
      "step": 790,
      "train_runtime": 1780.8088,
      "train_tokens_per_second": 337.952
    },
    {
      "epoch": 0.07273563218390805,
      "grad_norm": 6.479823112487793,
      "learning_rate": 1.2130325814536342e-05,
      "loss": 0.4365723431110382,
      "num_input_tokens_seen": 602428,
      "step": 791,
      "train_runtime": 1782.7738,
      "train_tokens_per_second": 337.916
    },
    {
      "epoch": 0.07282758620689656,
      "grad_norm": 6.727955341339111,
      "learning_rate": 1.2120300751879702e-05,
      "loss": 0.4399411678314209,
      "num_input_tokens_seen": 602992,
      "step": 792,
      "train_runtime": 1784.6858,
      "train_tokens_per_second": 337.87
    },
    {
      "epoch": 0.07291954022988506,
      "grad_norm": 7.448745250701904,
      "learning_rate": 1.2110275689223059e-05,
      "loss": 0.5305242538452148,
      "num_input_tokens_seen": 603716,
      "step": 793,
      "train_runtime": 1786.6041,
      "train_tokens_per_second": 337.913
    },
    {
      "epoch": 0.07301149425287357,
      "grad_norm": 7.2634806632995605,
      "learning_rate": 1.2100250626566417e-05,
      "loss": 0.6570507884025574,
      "num_input_tokens_seen": 604496,
      "step": 794,
      "train_runtime": 1788.5597,
      "train_tokens_per_second": 337.979
    },
    {
      "epoch": 0.07310344827586207,
      "grad_norm": 6.13576602935791,
      "learning_rate": 1.2090225563909777e-05,
      "loss": 0.41936028003692627,
      "num_input_tokens_seen": 605140,
      "step": 795,
      "train_runtime": 1790.4811,
      "train_tokens_per_second": 337.976
    },
    {
      "epoch": 0.07319540229885058,
      "grad_norm": 8.78651237487793,
      "learning_rate": 1.2080200501253134e-05,
      "loss": 0.5948994159698486,
      "num_input_tokens_seen": 605908,
      "step": 796,
      "train_runtime": 1792.4324,
      "train_tokens_per_second": 338.037
    },
    {
      "epoch": 0.07328735632183908,
      "grad_norm": 6.633975505828857,
      "learning_rate": 1.2070175438596493e-05,
      "loss": 0.4693654775619507,
      "num_input_tokens_seen": 606708,
      "step": 797,
      "train_runtime": 1794.3597,
      "train_tokens_per_second": 338.12
    },
    {
      "epoch": 0.07337931034482759,
      "grad_norm": 6.151555061340332,
      "learning_rate": 1.206015037593985e-05,
      "loss": 0.4316493570804596,
      "num_input_tokens_seen": 607320,
      "step": 798,
      "train_runtime": 1796.2865,
      "train_tokens_per_second": 338.098
    },
    {
      "epoch": 0.07347126436781609,
      "grad_norm": 6.873909950256348,
      "learning_rate": 1.205012531328321e-05,
      "loss": 0.3840366005897522,
      "num_input_tokens_seen": 607868,
      "step": 799,
      "train_runtime": 1798.2086,
      "train_tokens_per_second": 338.041
    },
    {
      "epoch": 0.0735632183908046,
      "grad_norm": 6.15745735168457,
      "learning_rate": 1.2040100250626568e-05,
      "loss": 0.43657514452934265,
      "num_input_tokens_seen": 608576,
      "step": 800,
      "train_runtime": 1800.1532,
      "train_tokens_per_second": 338.069
    },
    {
      "epoch": 0.0735632183908046,
      "eval_loss": 1.8885889053344727,
      "eval_runtime": 27.7822,
      "eval_samples_per_second": 35.994,
      "eval_steps_per_second": 8.999,
      "num_input_tokens_seen": 608576,
      "step": 800
    },
    {
      "epoch": 0.0736551724137931,
      "grad_norm": 6.745829105377197,
      "learning_rate": 1.2030075187969925e-05,
      "loss": 0.5598949193954468,
      "num_input_tokens_seen": 609524,
      "step": 801,
      "train_runtime": 1829.9183,
      "train_tokens_per_second": 333.088
    },
    {
      "epoch": 0.07374712643678161,
      "grad_norm": 6.7891435623168945,
      "learning_rate": 1.2020050125313285e-05,
      "loss": 0.4976637363433838,
      "num_input_tokens_seen": 610128,
      "step": 802,
      "train_runtime": 1831.8397,
      "train_tokens_per_second": 333.068
    },
    {
      "epoch": 0.07383908045977011,
      "grad_norm": 6.635828971862793,
      "learning_rate": 1.2010025062656644e-05,
      "loss": 0.42081525921821594,
      "num_input_tokens_seen": 610828,
      "step": 803,
      "train_runtime": 1833.7565,
      "train_tokens_per_second": 333.102
    },
    {
      "epoch": 0.07393103448275862,
      "grad_norm": 8.774009704589844,
      "learning_rate": 1.2e-05,
      "loss": 0.5076240301132202,
      "num_input_tokens_seen": 611836,
      "step": 804,
      "train_runtime": 1835.7381,
      "train_tokens_per_second": 333.292
    },
    {
      "epoch": 0.07402298850574712,
      "grad_norm": 6.756108283996582,
      "learning_rate": 1.1989974937343359e-05,
      "loss": 0.4481182098388672,
      "num_input_tokens_seen": 612480,
      "step": 805,
      "train_runtime": 1837.6441,
      "train_tokens_per_second": 333.296
    },
    {
      "epoch": 0.07411494252873563,
      "grad_norm": 8.837931632995605,
      "learning_rate": 1.1979949874686719e-05,
      "loss": 0.5549317598342896,
      "num_input_tokens_seen": 613136,
      "step": 806,
      "train_runtime": 1839.5563,
      "train_tokens_per_second": 333.306
    },
    {
      "epoch": 0.07420689655172413,
      "grad_norm": 8.184959411621094,
      "learning_rate": 1.1969924812030076e-05,
      "loss": 0.5435487031936646,
      "num_input_tokens_seen": 613828,
      "step": 807,
      "train_runtime": 1841.4644,
      "train_tokens_per_second": 333.337
    },
    {
      "epoch": 0.07429885057471264,
      "grad_norm": 8.38587760925293,
      "learning_rate": 1.1959899749373434e-05,
      "loss": 0.5304284691810608,
      "num_input_tokens_seen": 614508,
      "step": 808,
      "train_runtime": 1843.3763,
      "train_tokens_per_second": 333.36
    },
    {
      "epoch": 0.07439080459770114,
      "grad_norm": 8.372071266174316,
      "learning_rate": 1.1949874686716794e-05,
      "loss": 0.6614456176757812,
      "num_input_tokens_seen": 615200,
      "step": 809,
      "train_runtime": 1845.3267,
      "train_tokens_per_second": 333.383
    },
    {
      "epoch": 0.07448275862068965,
      "grad_norm": 8.287668228149414,
      "learning_rate": 1.1939849624060151e-05,
      "loss": 0.4698237478733063,
      "num_input_tokens_seen": 615896,
      "step": 810,
      "train_runtime": 1847.3082,
      "train_tokens_per_second": 333.402
    },
    {
      "epoch": 0.07457471264367817,
      "grad_norm": 6.464916229248047,
      "learning_rate": 1.192982456140351e-05,
      "loss": 0.5125246644020081,
      "num_input_tokens_seen": 616840,
      "step": 811,
      "train_runtime": 1856.0997,
      "train_tokens_per_second": 332.331
    },
    {
      "epoch": 0.07466666666666667,
      "grad_norm": 6.795294761657715,
      "learning_rate": 1.1919799498746868e-05,
      "loss": 0.5421081185340881,
      "num_input_tokens_seen": 617604,
      "step": 812,
      "train_runtime": 1858.0393,
      "train_tokens_per_second": 332.396
    },
    {
      "epoch": 0.07475862068965518,
      "grad_norm": 7.088548183441162,
      "learning_rate": 1.1909774436090227e-05,
      "loss": 0.467110276222229,
      "num_input_tokens_seen": 618220,
      "step": 813,
      "train_runtime": 1860.1864,
      "train_tokens_per_second": 332.343
    },
    {
      "epoch": 0.07485057471264368,
      "grad_norm": 7.344161510467529,
      "learning_rate": 1.1899749373433585e-05,
      "loss": 0.4739065170288086,
      "num_input_tokens_seen": 618972,
      "step": 814,
      "train_runtime": 1862.1494,
      "train_tokens_per_second": 332.397
    },
    {
      "epoch": 0.07494252873563219,
      "grad_norm": 7.0441484451293945,
      "learning_rate": 1.1889724310776942e-05,
      "loss": 0.45308488607406616,
      "num_input_tokens_seen": 619636,
      "step": 815,
      "train_runtime": 1864.0787,
      "train_tokens_per_second": 332.409
    },
    {
      "epoch": 0.0750344827586207,
      "grad_norm": 6.296428203582764,
      "learning_rate": 1.1879699248120302e-05,
      "loss": 0.42479604482650757,
      "num_input_tokens_seen": 620176,
      "step": 816,
      "train_runtime": 1865.9977,
      "train_tokens_per_second": 332.356
    },
    {
      "epoch": 0.0751264367816092,
      "grad_norm": 6.567066669464111,
      "learning_rate": 1.186967418546366e-05,
      "loss": 0.5601491332054138,
      "num_input_tokens_seen": 620900,
      "step": 817,
      "train_runtime": 1867.9469,
      "train_tokens_per_second": 332.397
    },
    {
      "epoch": 0.0752183908045977,
      "grad_norm": 6.935360908508301,
      "learning_rate": 1.1859649122807017e-05,
      "loss": 0.5025911927223206,
      "num_input_tokens_seen": 621528,
      "step": 818,
      "train_runtime": 1869.8617,
      "train_tokens_per_second": 332.392
    },
    {
      "epoch": 0.07531034482758621,
      "grad_norm": 6.330498218536377,
      "learning_rate": 1.1849624060150378e-05,
      "loss": 0.4385153651237488,
      "num_input_tokens_seen": 622152,
      "step": 819,
      "train_runtime": 1871.7717,
      "train_tokens_per_second": 332.387
    },
    {
      "epoch": 0.07540229885057471,
      "grad_norm": 6.7535552978515625,
      "learning_rate": 1.1839598997493736e-05,
      "loss": 0.4697404205799103,
      "num_input_tokens_seen": 622868,
      "step": 820,
      "train_runtime": 1873.7108,
      "train_tokens_per_second": 332.425
    },
    {
      "epoch": 0.07549425287356322,
      "grad_norm": 8.974464416503906,
      "learning_rate": 1.1829573934837093e-05,
      "loss": 0.445892333984375,
      "num_input_tokens_seen": 623364,
      "step": 821,
      "train_runtime": 1875.6189,
      "train_tokens_per_second": 332.351
    },
    {
      "epoch": 0.07558620689655172,
      "grad_norm": 5.514608860015869,
      "learning_rate": 1.1819548872180453e-05,
      "loss": 0.46127551794052124,
      "num_input_tokens_seen": 623996,
      "step": 822,
      "train_runtime": 1877.518,
      "train_tokens_per_second": 332.352
    },
    {
      "epoch": 0.07567816091954023,
      "grad_norm": 6.168216228485107,
      "learning_rate": 1.180952380952381e-05,
      "loss": 0.46345341205596924,
      "num_input_tokens_seen": 624712,
      "step": 823,
      "train_runtime": 1879.4229,
      "train_tokens_per_second": 332.396
    },
    {
      "epoch": 0.07577011494252874,
      "grad_norm": 6.788818836212158,
      "learning_rate": 1.1799498746867168e-05,
      "loss": 0.48941296339035034,
      "num_input_tokens_seen": 625888,
      "step": 824,
      "train_runtime": 1881.4051,
      "train_tokens_per_second": 332.671
    },
    {
      "epoch": 0.07586206896551724,
      "grad_norm": 7.571475982666016,
      "learning_rate": 1.1789473684210527e-05,
      "loss": 0.5315030813217163,
      "num_input_tokens_seen": 626768,
      "step": 825,
      "train_runtime": 1883.4487,
      "train_tokens_per_second": 332.777
    },
    {
      "epoch": 0.07595402298850575,
      "grad_norm": 6.428432941436768,
      "learning_rate": 1.1779448621553885e-05,
      "loss": 0.38236087560653687,
      "num_input_tokens_seen": 627548,
      "step": 826,
      "train_runtime": 1885.4018,
      "train_tokens_per_second": 332.846
    },
    {
      "epoch": 0.07604597701149425,
      "grad_norm": 6.466652870178223,
      "learning_rate": 1.1769423558897244e-05,
      "loss": 0.4763678312301636,
      "num_input_tokens_seen": 629068,
      "step": 827,
      "train_runtime": 1887.4518,
      "train_tokens_per_second": 333.29
    },
    {
      "epoch": 0.07613793103448276,
      "grad_norm": 6.652420997619629,
      "learning_rate": 1.1759398496240602e-05,
      "loss": 0.43790727853775024,
      "num_input_tokens_seen": 629608,
      "step": 828,
      "train_runtime": 1889.3464,
      "train_tokens_per_second": 333.241
    },
    {
      "epoch": 0.07622988505747126,
      "grad_norm": 7.144771575927734,
      "learning_rate": 1.174937343358396e-05,
      "loss": 0.5762131214141846,
      "num_input_tokens_seen": 630360,
      "step": 829,
      "train_runtime": 1891.2668,
      "train_tokens_per_second": 333.3
    },
    {
      "epoch": 0.07632183908045977,
      "grad_norm": 7.68195104598999,
      "learning_rate": 1.173934837092732e-05,
      "loss": 0.5741535425186157,
      "num_input_tokens_seen": 631076,
      "step": 830,
      "train_runtime": 1893.2182,
      "train_tokens_per_second": 333.335
    },
    {
      "epoch": 0.07641379310344827,
      "grad_norm": 7.149155139923096,
      "learning_rate": 1.1729323308270678e-05,
      "loss": 0.4585779905319214,
      "num_input_tokens_seen": 631732,
      "step": 831,
      "train_runtime": 1895.1613,
      "train_tokens_per_second": 333.339
    },
    {
      "epoch": 0.07650574712643678,
      "grad_norm": 7.119060516357422,
      "learning_rate": 1.1719298245614036e-05,
      "loss": 0.5880687236785889,
      "num_input_tokens_seen": 632812,
      "step": 832,
      "train_runtime": 1897.1405,
      "train_tokens_per_second": 333.561
    },
    {
      "epoch": 0.07659770114942528,
      "grad_norm": 7.2561750411987305,
      "learning_rate": 1.1709273182957395e-05,
      "loss": 0.5442973971366882,
      "num_input_tokens_seen": 633628,
      "step": 833,
      "train_runtime": 1899.1184,
      "train_tokens_per_second": 333.643
    },
    {
      "epoch": 0.07668965517241379,
      "grad_norm": 6.054276943206787,
      "learning_rate": 1.1699248120300753e-05,
      "loss": 0.4302482604980469,
      "num_input_tokens_seen": 634352,
      "step": 834,
      "train_runtime": 1901.0686,
      "train_tokens_per_second": 333.682
    },
    {
      "epoch": 0.07678160919540229,
      "grad_norm": 6.72073221206665,
      "learning_rate": 1.168922305764411e-05,
      "loss": 0.5162347555160522,
      "num_input_tokens_seen": 635000,
      "step": 835,
      "train_runtime": 1902.987,
      "train_tokens_per_second": 333.686
    },
    {
      "epoch": 0.07687356321839081,
      "grad_norm": 6.960562229156494,
      "learning_rate": 1.167919799498747e-05,
      "loss": 0.420468807220459,
      "num_input_tokens_seen": 635644,
      "step": 836,
      "train_runtime": 1904.9042,
      "train_tokens_per_second": 333.688
    },
    {
      "epoch": 0.07696551724137932,
      "grad_norm": 6.8537750244140625,
      "learning_rate": 1.1669172932330827e-05,
      "loss": 0.46297162771224976,
      "num_input_tokens_seen": 636256,
      "step": 837,
      "train_runtime": 1906.8568,
      "train_tokens_per_second": 333.667
    },
    {
      "epoch": 0.07705747126436782,
      "grad_norm": 6.908999919891357,
      "learning_rate": 1.1659147869674185e-05,
      "loss": 0.4330160617828369,
      "num_input_tokens_seen": 636920,
      "step": 838,
      "train_runtime": 1908.801,
      "train_tokens_per_second": 333.675
    },
    {
      "epoch": 0.07714942528735633,
      "grad_norm": 7.5881147384643555,
      "learning_rate": 1.1649122807017546e-05,
      "loss": 0.4976758062839508,
      "num_input_tokens_seen": 637816,
      "step": 839,
      "train_runtime": 1910.7814,
      "train_tokens_per_second": 333.799
    },
    {
      "epoch": 0.07724137931034483,
      "grad_norm": 7.4209747314453125,
      "learning_rate": 1.1639097744360902e-05,
      "loss": 0.44853490591049194,
      "num_input_tokens_seen": 638508,
      "step": 840,
      "train_runtime": 1912.7069,
      "train_tokens_per_second": 333.824
    },
    {
      "epoch": 0.07733333333333334,
      "grad_norm": 8.012955665588379,
      "learning_rate": 1.1629072681704261e-05,
      "loss": 0.5154711604118347,
      "num_input_tokens_seen": 639356,
      "step": 841,
      "train_runtime": 1921.2892,
      "train_tokens_per_second": 332.774
    },
    {
      "epoch": 0.07742528735632184,
      "grad_norm": 6.575119972229004,
      "learning_rate": 1.1619047619047621e-05,
      "loss": 0.41180163621902466,
      "num_input_tokens_seen": 639932,
      "step": 842,
      "train_runtime": 1923.2031,
      "train_tokens_per_second": 332.743
    },
    {
      "epoch": 0.07751724137931035,
      "grad_norm": 7.115898132324219,
      "learning_rate": 1.1609022556390978e-05,
      "loss": 0.5225002765655518,
      "num_input_tokens_seen": 640772,
      "step": 843,
      "train_runtime": 1925.3575,
      "train_tokens_per_second": 332.807
    },
    {
      "epoch": 0.07760919540229885,
      "grad_norm": 8.289604187011719,
      "learning_rate": 1.1598997493734336e-05,
      "loss": 0.5227071046829224,
      "num_input_tokens_seen": 641400,
      "step": 844,
      "train_runtime": 1927.2611,
      "train_tokens_per_second": 332.804
    },
    {
      "epoch": 0.07770114942528736,
      "grad_norm": 8.993541717529297,
      "learning_rate": 1.1588972431077695e-05,
      "loss": 0.5554720163345337,
      "num_input_tokens_seen": 642492,
      "step": 845,
      "train_runtime": 1929.2504,
      "train_tokens_per_second": 333.027
    },
    {
      "epoch": 0.07779310344827586,
      "grad_norm": 10.20373249053955,
      "learning_rate": 1.1578947368421053e-05,
      "loss": 0.5579010844230652,
      "num_input_tokens_seen": 643508,
      "step": 846,
      "train_runtime": 1931.2594,
      "train_tokens_per_second": 333.206
    },
    {
      "epoch": 0.07788505747126437,
      "grad_norm": 7.8156328201293945,
      "learning_rate": 1.1568922305764412e-05,
      "loss": 0.5619121789932251,
      "num_input_tokens_seen": 644208,
      "step": 847,
      "train_runtime": 1933.2068,
      "train_tokens_per_second": 333.233
    },
    {
      "epoch": 0.07797701149425287,
      "grad_norm": 7.236008167266846,
      "learning_rate": 1.1558897243107769e-05,
      "loss": 0.5577964186668396,
      "num_input_tokens_seen": 644920,
      "step": 848,
      "train_runtime": 1935.1306,
      "train_tokens_per_second": 333.269
    },
    {
      "epoch": 0.07806896551724138,
      "grad_norm": 6.673439025878906,
      "learning_rate": 1.1548872180451129e-05,
      "loss": 0.39545294642448425,
      "num_input_tokens_seen": 645452,
      "step": 849,
      "train_runtime": 1937.0371,
      "train_tokens_per_second": 333.216
    },
    {
      "epoch": 0.07816091954022988,
      "grad_norm": 8.605973243713379,
      "learning_rate": 1.1538847117794487e-05,
      "loss": 0.4736593961715698,
      "num_input_tokens_seen": 646036,
      "step": 850,
      "train_runtime": 1938.9508,
      "train_tokens_per_second": 333.188
    },
    {
      "epoch": 0.07825287356321839,
      "grad_norm": 6.712059020996094,
      "learning_rate": 1.1528822055137844e-05,
      "loss": 0.4098498821258545,
      "num_input_tokens_seen": 646576,
      "step": 851,
      "train_runtime": 1940.8521,
      "train_tokens_per_second": 333.14
    },
    {
      "epoch": 0.0783448275862069,
      "grad_norm": 6.2568206787109375,
      "learning_rate": 1.1518796992481204e-05,
      "loss": 0.3715309500694275,
      "num_input_tokens_seen": 647120,
      "step": 852,
      "train_runtime": 1942.7527,
      "train_tokens_per_second": 333.094
    },
    {
      "epoch": 0.0784367816091954,
      "grad_norm": 7.1600565910339355,
      "learning_rate": 1.1508771929824563e-05,
      "loss": 0.4333030581474304,
      "num_input_tokens_seen": 647904,
      "step": 853,
      "train_runtime": 1944.6604,
      "train_tokens_per_second": 333.171
    },
    {
      "epoch": 0.0785287356321839,
      "grad_norm": 6.429007053375244,
      "learning_rate": 1.149874686716792e-05,
      "loss": 0.3695297837257385,
      "num_input_tokens_seen": 648512,
      "step": 854,
      "train_runtime": 1946.5679,
      "train_tokens_per_second": 333.157
    },
    {
      "epoch": 0.07862068965517241,
      "grad_norm": 6.61116361618042,
      "learning_rate": 1.1488721804511278e-05,
      "loss": 0.5142643451690674,
      "num_input_tokens_seen": 649256,
      "step": 855,
      "train_runtime": 1948.4773,
      "train_tokens_per_second": 333.212
    },
    {
      "epoch": 0.07871264367816092,
      "grad_norm": 5.623296737670898,
      "learning_rate": 1.1478696741854638e-05,
      "loss": 0.38079267740249634,
      "num_input_tokens_seen": 649904,
      "step": 856,
      "train_runtime": 1950.3949,
      "train_tokens_per_second": 333.217
    },
    {
      "epoch": 0.07880459770114942,
      "grad_norm": 7.264566421508789,
      "learning_rate": 1.1468671679197995e-05,
      "loss": 0.5372199416160583,
      "num_input_tokens_seen": 650600,
      "step": 857,
      "train_runtime": 1952.2987,
      "train_tokens_per_second": 333.248
    },
    {
      "epoch": 0.07889655172413793,
      "grad_norm": 7.014410018920898,
      "learning_rate": 1.1458646616541354e-05,
      "loss": 0.5414127707481384,
      "num_input_tokens_seen": 651508,
      "step": 858,
      "train_runtime": 1954.2731,
      "train_tokens_per_second": 333.376
    },
    {
      "epoch": 0.07898850574712643,
      "grad_norm": 7.4664306640625,
      "learning_rate": 1.1448621553884714e-05,
      "loss": 0.5058953762054443,
      "num_input_tokens_seen": 652496,
      "step": 859,
      "train_runtime": 1956.2115,
      "train_tokens_per_second": 333.551
    },
    {
      "epoch": 0.07908045977011494,
      "grad_norm": 6.7541117668151855,
      "learning_rate": 1.143859649122807e-05,
      "loss": 0.39631015062332153,
      "num_input_tokens_seen": 653028,
      "step": 860,
      "train_runtime": 1958.105,
      "train_tokens_per_second": 333.5
    },
    {
      "epoch": 0.07917241379310345,
      "grad_norm": 7.931762218475342,
      "learning_rate": 1.1428571428571429e-05,
      "loss": 0.5153555274009705,
      "num_input_tokens_seen": 653676,
      "step": 861,
      "train_runtime": 1960.0172,
      "train_tokens_per_second": 333.505
    },
    {
      "epoch": 0.07926436781609196,
      "grad_norm": 6.375082015991211,
      "learning_rate": 1.141854636591479e-05,
      "loss": 0.4492836594581604,
      "num_input_tokens_seen": 654324,
      "step": 862,
      "train_runtime": 1961.9875,
      "train_tokens_per_second": 333.501
    },
    {
      "epoch": 0.07935632183908047,
      "grad_norm": 7.007679462432861,
      "learning_rate": 1.1408521303258146e-05,
      "loss": 0.45718562602996826,
      "num_input_tokens_seen": 655020,
      "step": 863,
      "train_runtime": 1963.8989,
      "train_tokens_per_second": 333.53
    },
    {
      "epoch": 0.07944827586206897,
      "grad_norm": 6.726501941680908,
      "learning_rate": 1.1398496240601504e-05,
      "loss": 0.47400668263435364,
      "num_input_tokens_seen": 655608,
      "step": 864,
      "train_runtime": 1965.8383,
      "train_tokens_per_second": 333.5
    },
    {
      "epoch": 0.07954022988505748,
      "grad_norm": 6.898266792297363,
      "learning_rate": 1.1388471177944861e-05,
      "loss": 0.47314706444740295,
      "num_input_tokens_seen": 656136,
      "step": 865,
      "train_runtime": 1967.7478,
      "train_tokens_per_second": 333.445
    },
    {
      "epoch": 0.07963218390804598,
      "grad_norm": 6.916375637054443,
      "learning_rate": 1.1378446115288221e-05,
      "loss": 0.40405744314193726,
      "num_input_tokens_seen": 656808,
      "step": 866,
      "train_runtime": 1969.6776,
      "train_tokens_per_second": 333.46
    },
    {
      "epoch": 0.07972413793103449,
      "grad_norm": 6.643969535827637,
      "learning_rate": 1.136842105263158e-05,
      "loss": 0.43459606170654297,
      "num_input_tokens_seen": 657364,
      "step": 867,
      "train_runtime": 1971.6123,
      "train_tokens_per_second": 333.414
    },
    {
      "epoch": 0.07981609195402299,
      "grad_norm": 7.246833801269531,
      "learning_rate": 1.1358395989974937e-05,
      "loss": 0.5171590447425842,
      "num_input_tokens_seen": 657992,
      "step": 868,
      "train_runtime": 1973.5583,
      "train_tokens_per_second": 333.404
    },
    {
      "epoch": 0.0799080459770115,
      "grad_norm": 6.035616874694824,
      "learning_rate": 1.1348370927318297e-05,
      "loss": 0.4002436399459839,
      "num_input_tokens_seen": 658644,
      "step": 869,
      "train_runtime": 1975.473,
      "train_tokens_per_second": 333.411
    },
    {
      "epoch": 0.08,
      "grad_norm": 6.385766983032227,
      "learning_rate": 1.1338345864661655e-05,
      "loss": 0.49872082471847534,
      "num_input_tokens_seen": 659340,
      "step": 870,
      "train_runtime": 1977.4045,
      "train_tokens_per_second": 333.437
    },
    {
      "epoch": 0.0800919540229885,
      "grad_norm": 6.394904613494873,
      "learning_rate": 1.1328320802005012e-05,
      "loss": 0.49301695823669434,
      "num_input_tokens_seen": 660572,
      "step": 871,
      "train_runtime": 1985.9584,
      "train_tokens_per_second": 332.621
    },
    {
      "epoch": 0.08018390804597701,
      "grad_norm": 6.4015092849731445,
      "learning_rate": 1.1318295739348372e-05,
      "loss": 0.46217477321624756,
      "num_input_tokens_seen": 661224,
      "step": 872,
      "train_runtime": 1987.8841,
      "train_tokens_per_second": 332.627
    },
    {
      "epoch": 0.08027586206896552,
      "grad_norm": 6.900290489196777,
      "learning_rate": 1.1308270676691731e-05,
      "loss": 0.5131678581237793,
      "num_input_tokens_seen": 661996,
      "step": 873,
      "train_runtime": 1989.8417,
      "train_tokens_per_second": 332.688
    },
    {
      "epoch": 0.08036781609195402,
      "grad_norm": 6.40717077255249,
      "learning_rate": 1.1298245614035088e-05,
      "loss": 0.47580698132514954,
      "num_input_tokens_seen": 662536,
      "step": 874,
      "train_runtime": 1991.7705,
      "train_tokens_per_second": 332.637
    },
    {
      "epoch": 0.08045977011494253,
      "grad_norm": 6.060063362121582,
      "learning_rate": 1.1288220551378446e-05,
      "loss": 0.37694984674453735,
      "num_input_tokens_seen": 663124,
      "step": 875,
      "train_runtime": 1993.6831,
      "train_tokens_per_second": 332.613
    },
    {
      "epoch": 0.08055172413793103,
      "grad_norm": 6.391973495483398,
      "learning_rate": 1.1278195488721806e-05,
      "loss": 0.4877256453037262,
      "num_input_tokens_seen": 663828,
      "step": 876,
      "train_runtime": 1995.6021,
      "train_tokens_per_second": 332.645
    },
    {
      "epoch": 0.08064367816091954,
      "grad_norm": 7.020805358886719,
      "learning_rate": 1.1268170426065163e-05,
      "loss": 0.4744494557380676,
      "num_input_tokens_seen": 664832,
      "step": 877,
      "train_runtime": 1997.5559,
      "train_tokens_per_second": 332.823
    },
    {
      "epoch": 0.08073563218390804,
      "grad_norm": 6.907968044281006,
      "learning_rate": 1.1258145363408522e-05,
      "loss": 0.44763144850730896,
      "num_input_tokens_seen": 665408,
      "step": 878,
      "train_runtime": 1999.4846,
      "train_tokens_per_second": 332.79
    },
    {
      "epoch": 0.08082758620689655,
      "grad_norm": 5.961269855499268,
      "learning_rate": 1.1248120300751882e-05,
      "loss": 0.38856619596481323,
      "num_input_tokens_seen": 665984,
      "step": 879,
      "train_runtime": 2001.4023,
      "train_tokens_per_second": 332.759
    },
    {
      "epoch": 0.08091954022988505,
      "grad_norm": 6.160533428192139,
      "learning_rate": 1.1238095238095239e-05,
      "loss": 0.5681909918785095,
      "num_input_tokens_seen": 666908,
      "step": 880,
      "train_runtime": 2003.3519,
      "train_tokens_per_second": 332.896
    },
    {
      "epoch": 0.08101149425287356,
      "grad_norm": 6.835758686065674,
      "learning_rate": 1.1228070175438597e-05,
      "loss": 0.41134172677993774,
      "num_input_tokens_seen": 667668,
      "step": 881,
      "train_runtime": 2005.289,
      "train_tokens_per_second": 332.954
    },
    {
      "epoch": 0.08110344827586206,
      "grad_norm": 6.545367240905762,
      "learning_rate": 1.1218045112781957e-05,
      "loss": 0.3771873116493225,
      "num_input_tokens_seen": 668372,
      "step": 882,
      "train_runtime": 2007.2165,
      "train_tokens_per_second": 332.985
    },
    {
      "epoch": 0.08119540229885057,
      "grad_norm": 6.41802453994751,
      "learning_rate": 1.1208020050125314e-05,
      "loss": 0.44758257269859314,
      "num_input_tokens_seen": 669256,
      "step": 883,
      "train_runtime": 2009.1714,
      "train_tokens_per_second": 333.101
    },
    {
      "epoch": 0.08128735632183907,
      "grad_norm": 6.513979434967041,
      "learning_rate": 1.1197994987468673e-05,
      "loss": 0.44916343688964844,
      "num_input_tokens_seen": 669952,
      "step": 884,
      "train_runtime": 2011.1194,
      "train_tokens_per_second": 333.124
    },
    {
      "epoch": 0.08137931034482758,
      "grad_norm": 6.685109615325928,
      "learning_rate": 1.118796992481203e-05,
      "loss": 0.49495604634284973,
      "num_input_tokens_seen": 670624,
      "step": 885,
      "train_runtime": 2013.0315,
      "train_tokens_per_second": 333.141
    },
    {
      "epoch": 0.0814712643678161,
      "grad_norm": 6.367468357086182,
      "learning_rate": 1.117794486215539e-05,
      "loss": 0.4811069965362549,
      "num_input_tokens_seen": 671316,
      "step": 886,
      "train_runtime": 2014.9433,
      "train_tokens_per_second": 333.169
    },
    {
      "epoch": 0.0815632183908046,
      "grad_norm": 6.819706916809082,
      "learning_rate": 1.1167919799498748e-05,
      "loss": 0.42282989621162415,
      "num_input_tokens_seen": 672064,
      "step": 887,
      "train_runtime": 2016.8867,
      "train_tokens_per_second": 333.219
    },
    {
      "epoch": 0.08165517241379311,
      "grad_norm": 5.85295295715332,
      "learning_rate": 1.1157894736842105e-05,
      "loss": 0.3618812561035156,
      "num_input_tokens_seen": 672880,
      "step": 888,
      "train_runtime": 2018.8282,
      "train_tokens_per_second": 333.302
    },
    {
      "epoch": 0.08174712643678161,
      "grad_norm": 6.7765913009643555,
      "learning_rate": 1.1147869674185465e-05,
      "loss": 0.46744728088378906,
      "num_input_tokens_seen": 673440,
      "step": 889,
      "train_runtime": 2020.7379,
      "train_tokens_per_second": 333.264
    },
    {
      "epoch": 0.08183908045977012,
      "grad_norm": 7.506669998168945,
      "learning_rate": 1.1137844611528823e-05,
      "loss": 0.47970443964004517,
      "num_input_tokens_seen": 674164,
      "step": 890,
      "train_runtime": 2022.6668,
      "train_tokens_per_second": 333.305
    },
    {
      "epoch": 0.08193103448275862,
      "grad_norm": 6.59580659866333,
      "learning_rate": 1.112781954887218e-05,
      "loss": 0.42027968168258667,
      "num_input_tokens_seen": 674772,
      "step": 891,
      "train_runtime": 2024.5891,
      "train_tokens_per_second": 333.288
    },
    {
      "epoch": 0.08202298850574713,
      "grad_norm": 7.922469139099121,
      "learning_rate": 1.111779448621554e-05,
      "loss": 0.5631629228591919,
      "num_input_tokens_seen": 675492,
      "step": 892,
      "train_runtime": 2026.5207,
      "train_tokens_per_second": 333.326
    },
    {
      "epoch": 0.08211494252873563,
      "grad_norm": 7.59067440032959,
      "learning_rate": 1.1107769423558899e-05,
      "loss": 0.5644431710243225,
      "num_input_tokens_seen": 676152,
      "step": 893,
      "train_runtime": 2028.4569,
      "train_tokens_per_second": 333.333
    },
    {
      "epoch": 0.08220689655172414,
      "grad_norm": 6.923701763153076,
      "learning_rate": 1.1097744360902256e-05,
      "loss": 0.46577513217926025,
      "num_input_tokens_seen": 676924,
      "step": 894,
      "train_runtime": 2030.3974,
      "train_tokens_per_second": 333.395
    },
    {
      "epoch": 0.08229885057471265,
      "grad_norm": 6.746239185333252,
      "learning_rate": 1.1087719298245614e-05,
      "loss": 0.6171218156814575,
      "num_input_tokens_seen": 677864,
      "step": 895,
      "train_runtime": 2032.4181,
      "train_tokens_per_second": 333.526
    },
    {
      "epoch": 0.08239080459770115,
      "grad_norm": 6.714953422546387,
      "learning_rate": 1.1077694235588974e-05,
      "loss": 0.4712778329849243,
      "num_input_tokens_seen": 678744,
      "step": 896,
      "train_runtime": 2034.359,
      "train_tokens_per_second": 333.64
    },
    {
      "epoch": 0.08248275862068966,
      "grad_norm": 7.062094211578369,
      "learning_rate": 1.1067669172932331e-05,
      "loss": 0.5276885032653809,
      "num_input_tokens_seen": 679472,
      "step": 897,
      "train_runtime": 2036.3004,
      "train_tokens_per_second": 333.68
    },
    {
      "epoch": 0.08257471264367816,
      "grad_norm": 6.546316623687744,
      "learning_rate": 1.105764411027569e-05,
      "loss": 0.40527454018592834,
      "num_input_tokens_seen": 680064,
      "step": 898,
      "train_runtime": 2038.3067,
      "train_tokens_per_second": 333.642
    },
    {
      "epoch": 0.08266666666666667,
      "grad_norm": 6.58282470703125,
      "learning_rate": 1.104761904761905e-05,
      "loss": 0.41458383202552795,
      "num_input_tokens_seen": 680804,
      "step": 899,
      "train_runtime": 2040.2651,
      "train_tokens_per_second": 333.684
    },
    {
      "epoch": 0.08275862068965517,
      "grad_norm": 6.260782718658447,
      "learning_rate": 1.1037593984962407e-05,
      "loss": 0.5471869111061096,
      "num_input_tokens_seen": 681448,
      "step": 900,
      "train_runtime": 2042.1806,
      "train_tokens_per_second": 333.686
    },
    {
      "epoch": 0.08285057471264368,
      "grad_norm": 7.149852752685547,
      "learning_rate": 1.1027568922305765e-05,
      "loss": 0.4660685658454895,
      "num_input_tokens_seen": 682076,
      "step": 901,
      "train_runtime": 2050.6772,
      "train_tokens_per_second": 332.61
    },
    {
      "epoch": 0.08294252873563218,
      "grad_norm": 6.504394054412842,
      "learning_rate": 1.1017543859649125e-05,
      "loss": 0.5031466484069824,
      "num_input_tokens_seen": 682968,
      "step": 902,
      "train_runtime": 2052.8674,
      "train_tokens_per_second": 332.69
    },
    {
      "epoch": 0.08303448275862069,
      "grad_norm": 6.417966842651367,
      "learning_rate": 1.1007518796992482e-05,
      "loss": 0.3842109441757202,
      "num_input_tokens_seen": 683544,
      "step": 903,
      "train_runtime": 2054.7968,
      "train_tokens_per_second": 332.658
    },
    {
      "epoch": 0.08312643678160919,
      "grad_norm": 5.998277187347412,
      "learning_rate": 1.099749373433584e-05,
      "loss": 0.5193146467208862,
      "num_input_tokens_seen": 684420,
      "step": 904,
      "train_runtime": 2056.7639,
      "train_tokens_per_second": 332.765
    },
    {
      "epoch": 0.0832183908045977,
      "grad_norm": 6.862845420837402,
      "learning_rate": 1.0987468671679197e-05,
      "loss": 0.4847695827484131,
      "num_input_tokens_seen": 685048,
      "step": 905,
      "train_runtime": 2058.7083,
      "train_tokens_per_second": 332.756
    },
    {
      "epoch": 0.0833103448275862,
      "grad_norm": 6.598841667175293,
      "learning_rate": 1.0977443609022558e-05,
      "loss": 0.5620836019515991,
      "num_input_tokens_seen": 685896,
      "step": 906,
      "train_runtime": 2060.6561,
      "train_tokens_per_second": 332.853
    },
    {
      "epoch": 0.08340229885057471,
      "grad_norm": 5.996801376342773,
      "learning_rate": 1.0967418546365916e-05,
      "loss": 0.4808143973350525,
      "num_input_tokens_seen": 686628,
      "step": 907,
      "train_runtime": 2062.5797,
      "train_tokens_per_second": 332.898
    },
    {
      "epoch": 0.08349425287356321,
      "grad_norm": 6.211836338043213,
      "learning_rate": 1.0957393483709273e-05,
      "loss": 0.49074992537498474,
      "num_input_tokens_seen": 687568,
      "step": 908,
      "train_runtime": 2064.5337,
      "train_tokens_per_second": 333.038
    },
    {
      "epoch": 0.08358620689655172,
      "grad_norm": 7.279179573059082,
      "learning_rate": 1.0947368421052633e-05,
      "loss": 0.5068899989128113,
      "num_input_tokens_seen": 688208,
      "step": 909,
      "train_runtime": 2066.4414,
      "train_tokens_per_second": 333.04
    },
    {
      "epoch": 0.08367816091954022,
      "grad_norm": 10.764835357666016,
      "learning_rate": 1.0937343358395992e-05,
      "loss": 0.3864727318286896,
      "num_input_tokens_seen": 688928,
      "step": 910,
      "train_runtime": 2068.3762,
      "train_tokens_per_second": 333.077
    },
    {
      "epoch": 0.08377011494252874,
      "grad_norm": 5.784451484680176,
      "learning_rate": 1.0927318295739348e-05,
      "loss": 0.38656336069107056,
      "num_input_tokens_seen": 689648,
      "step": 911,
      "train_runtime": 2070.3169,
      "train_tokens_per_second": 333.112
    },
    {
      "epoch": 0.08386206896551725,
      "grad_norm": 6.182672023773193,
      "learning_rate": 1.0917293233082709e-05,
      "loss": 0.42912742495536804,
      "num_input_tokens_seen": 690216,
      "step": 912,
      "train_runtime": 2072.2239,
      "train_tokens_per_second": 333.08
    },
    {
      "epoch": 0.08395402298850575,
      "grad_norm": 7.196264266967773,
      "learning_rate": 1.0907268170426067e-05,
      "loss": 0.5714797973632812,
      "num_input_tokens_seen": 691316,
      "step": 913,
      "train_runtime": 2074.1931,
      "train_tokens_per_second": 333.294
    },
    {
      "epoch": 0.08404597701149426,
      "grad_norm": 5.99977970123291,
      "learning_rate": 1.0897243107769424e-05,
      "loss": 0.4422471523284912,
      "num_input_tokens_seen": 692048,
      "step": 914,
      "train_runtime": 2076.1055,
      "train_tokens_per_second": 333.34
    },
    {
      "epoch": 0.08413793103448276,
      "grad_norm": 7.258148193359375,
      "learning_rate": 1.0887218045112782e-05,
      "loss": 0.508287250995636,
      "num_input_tokens_seen": 692848,
      "step": 915,
      "train_runtime": 2078.0763,
      "train_tokens_per_second": 333.408
    },
    {
      "epoch": 0.08422988505747127,
      "grad_norm": 7.077605247497559,
      "learning_rate": 1.0877192982456142e-05,
      "loss": 0.5383909940719604,
      "num_input_tokens_seen": 693920,
      "step": 916,
      "train_runtime": 2080.1235,
      "train_tokens_per_second": 333.596
    },
    {
      "epoch": 0.08432183908045977,
      "grad_norm": 7.42640495300293,
      "learning_rate": 1.08671679197995e-05,
      "loss": 0.5203660726547241,
      "num_input_tokens_seen": 694544,
      "step": 917,
      "train_runtime": 2082.0655,
      "train_tokens_per_second": 333.584
    },
    {
      "epoch": 0.08441379310344828,
      "grad_norm": 7.465845108032227,
      "learning_rate": 1.0857142857142858e-05,
      "loss": 0.48671990633010864,
      "num_input_tokens_seen": 695228,
      "step": 918,
      "train_runtime": 2084.0367,
      "train_tokens_per_second": 333.597
    },
    {
      "epoch": 0.08450574712643678,
      "grad_norm": 6.276941776275635,
      "learning_rate": 1.0847117794486218e-05,
      "loss": 0.39228111505508423,
      "num_input_tokens_seen": 695916,
      "step": 919,
      "train_runtime": 2085.9518,
      "train_tokens_per_second": 333.62
    },
    {
      "epoch": 0.08459770114942529,
      "grad_norm": 8.05283260345459,
      "learning_rate": 1.0837092731829575e-05,
      "loss": 0.49192559719085693,
      "num_input_tokens_seen": 696468,
      "step": 920,
      "train_runtime": 2087.88,
      "train_tokens_per_second": 333.577
    },
    {
      "epoch": 0.0846896551724138,
      "grad_norm": 7.042476177215576,
      "learning_rate": 1.0827067669172933e-05,
      "loss": 0.5734916925430298,
      "num_input_tokens_seen": 697356,
      "step": 921,
      "train_runtime": 2089.8969,
      "train_tokens_per_second": 333.68
    },
    {
      "epoch": 0.0847816091954023,
      "grad_norm": 6.639822959899902,
      "learning_rate": 1.0817042606516293e-05,
      "loss": 0.47092488408088684,
      "num_input_tokens_seen": 698036,
      "step": 922,
      "train_runtime": 2091.8192,
      "train_tokens_per_second": 333.698
    },
    {
      "epoch": 0.0848735632183908,
      "grad_norm": 7.225997447967529,
      "learning_rate": 1.080701754385965e-05,
      "loss": 0.4256677031517029,
      "num_input_tokens_seen": 698776,
      "step": 923,
      "train_runtime": 2093.9929,
      "train_tokens_per_second": 333.705
    },
    {
      "epoch": 0.08496551724137931,
      "grad_norm": 6.923762798309326,
      "learning_rate": 1.0796992481203009e-05,
      "loss": 0.5419110655784607,
      "num_input_tokens_seen": 699452,
      "step": 924,
      "train_runtime": 2096.0254,
      "train_tokens_per_second": 333.704
    },
    {
      "epoch": 0.08505747126436781,
      "grad_norm": 6.349163055419922,
      "learning_rate": 1.0786967418546365e-05,
      "loss": 0.5122674703598022,
      "num_input_tokens_seen": 700132,
      "step": 925,
      "train_runtime": 2097.9464,
      "train_tokens_per_second": 333.723
    },
    {
      "epoch": 0.08514942528735632,
      "grad_norm": 7.829336166381836,
      "learning_rate": 1.0776942355889726e-05,
      "loss": 0.522594690322876,
      "num_input_tokens_seen": 700764,
      "step": 926,
      "train_runtime": 2099.8556,
      "train_tokens_per_second": 333.72
    },
    {
      "epoch": 0.08524137931034483,
      "grad_norm": 7.369714260101318,
      "learning_rate": 1.0766917293233084e-05,
      "loss": 0.45676571130752563,
      "num_input_tokens_seen": 701408,
      "step": 927,
      "train_runtime": 2101.8478,
      "train_tokens_per_second": 333.71
    },
    {
      "epoch": 0.08533333333333333,
      "grad_norm": 6.920671463012695,
      "learning_rate": 1.0756892230576441e-05,
      "loss": 0.413580060005188,
      "num_input_tokens_seen": 702060,
      "step": 928,
      "train_runtime": 2103.7675,
      "train_tokens_per_second": 333.716
    },
    {
      "epoch": 0.08542528735632184,
      "grad_norm": 10.892513275146484,
      "learning_rate": 1.0746867167919801e-05,
      "loss": 0.5898758769035339,
      "num_input_tokens_seen": 702684,
      "step": 929,
      "train_runtime": 2105.6902,
      "train_tokens_per_second": 333.707
    },
    {
      "epoch": 0.08551724137931034,
      "grad_norm": 6.803252696990967,
      "learning_rate": 1.073684210526316e-05,
      "loss": 0.4193257689476013,
      "num_input_tokens_seen": 703260,
      "step": 930,
      "train_runtime": 2107.6118,
      "train_tokens_per_second": 333.676
    },
    {
      "epoch": 0.08560919540229885,
      "grad_norm": 6.986143112182617,
      "learning_rate": 1.0726817042606516e-05,
      "loss": 0.5487346649169922,
      "num_input_tokens_seen": 704160,
      "step": 931,
      "train_runtime": 2116.4739,
      "train_tokens_per_second": 332.704
    },
    {
      "epoch": 0.08570114942528735,
      "grad_norm": 7.126340866088867,
      "learning_rate": 1.0716791979949877e-05,
      "loss": 0.36908119916915894,
      "num_input_tokens_seen": 704704,
      "step": 932,
      "train_runtime": 2118.3939,
      "train_tokens_per_second": 332.66
    },
    {
      "epoch": 0.08579310344827586,
      "grad_norm": 6.432652473449707,
      "learning_rate": 1.0706766917293235e-05,
      "loss": 0.48489218950271606,
      "num_input_tokens_seen": 705972,
      "step": 933,
      "train_runtime": 2120.4204,
      "train_tokens_per_second": 332.94
    },
    {
      "epoch": 0.08588505747126436,
      "grad_norm": 7.483302593231201,
      "learning_rate": 1.0696741854636592e-05,
      "loss": 0.4927271902561188,
      "num_input_tokens_seen": 706736,
      "step": 934,
      "train_runtime": 2122.3768,
      "train_tokens_per_second": 332.993
    },
    {
      "epoch": 0.08597701149425287,
      "grad_norm": 7.03338098526001,
      "learning_rate": 1.068671679197995e-05,
      "loss": 0.5359400510787964,
      "num_input_tokens_seen": 707328,
      "step": 935,
      "train_runtime": 2124.2771,
      "train_tokens_per_second": 332.974
    },
    {
      "epoch": 0.08606896551724139,
      "grad_norm": 6.798832416534424,
      "learning_rate": 1.0676691729323309e-05,
      "loss": 0.456082284450531,
      "num_input_tokens_seen": 707852,
      "step": 936,
      "train_runtime": 2126.1803,
      "train_tokens_per_second": 332.922
    },
    {
      "epoch": 0.08616091954022989,
      "grad_norm": 6.522405624389648,
      "learning_rate": 1.0666666666666667e-05,
      "loss": 0.32978537678718567,
      "num_input_tokens_seen": 708380,
      "step": 937,
      "train_runtime": 2128.0939,
      "train_tokens_per_second": 332.871
    },
    {
      "epoch": 0.0862528735632184,
      "grad_norm": 5.998535633087158,
      "learning_rate": 1.0656641604010026e-05,
      "loss": 0.44136589765548706,
      "num_input_tokens_seen": 709296,
      "step": 938,
      "train_runtime": 2130.0448,
      "train_tokens_per_second": 332.996
    },
    {
      "epoch": 0.0863448275862069,
      "grad_norm": 7.186913013458252,
      "learning_rate": 1.0646616541353384e-05,
      "loss": 0.5467941761016846,
      "num_input_tokens_seen": 710024,
      "step": 939,
      "train_runtime": 2131.9635,
      "train_tokens_per_second": 333.038
    },
    {
      "epoch": 0.0864367816091954,
      "grad_norm": 7.097897529602051,
      "learning_rate": 1.0636591478696743e-05,
      "loss": 0.5385963320732117,
      "num_input_tokens_seen": 711136,
      "step": 940,
      "train_runtime": 2133.9513,
      "train_tokens_per_second": 333.248
    },
    {
      "epoch": 0.08652873563218391,
      "grad_norm": 7.345585346221924,
      "learning_rate": 1.0626566416040101e-05,
      "loss": 0.5527505278587341,
      "num_input_tokens_seen": 711860,
      "step": 941,
      "train_runtime": 2135.876,
      "train_tokens_per_second": 333.287
    },
    {
      "epoch": 0.08662068965517242,
      "grad_norm": 6.7262091636657715,
      "learning_rate": 1.061654135338346e-05,
      "loss": 0.47531142830848694,
      "num_input_tokens_seen": 712632,
      "step": 942,
      "train_runtime": 2137.8432,
      "train_tokens_per_second": 333.342
    },
    {
      "epoch": 0.08671264367816092,
      "grad_norm": 6.64491081237793,
      "learning_rate": 1.0606516290726818e-05,
      "loss": 0.39258521795272827,
      "num_input_tokens_seen": 713316,
      "step": 943,
      "train_runtime": 2139.7785,
      "train_tokens_per_second": 333.36
    },
    {
      "epoch": 0.08680459770114943,
      "grad_norm": 6.810537338256836,
      "learning_rate": 1.0596491228070177e-05,
      "loss": 0.6050618886947632,
      "num_input_tokens_seen": 714004,
      "step": 944,
      "train_runtime": 2141.7049,
      "train_tokens_per_second": 333.381
    },
    {
      "epoch": 0.08689655172413793,
      "grad_norm": 6.627172470092773,
      "learning_rate": 1.0586466165413534e-05,
      "loss": 0.46951693296432495,
      "num_input_tokens_seen": 714800,
      "step": 945,
      "train_runtime": 2143.6533,
      "train_tokens_per_second": 333.449
    },
    {
      "epoch": 0.08698850574712644,
      "grad_norm": 6.956730842590332,
      "learning_rate": 1.0576441102756894e-05,
      "loss": 0.4267975091934204,
      "num_input_tokens_seen": 715328,
      "step": 946,
      "train_runtime": 2145.6099,
      "train_tokens_per_second": 333.391
    },
    {
      "epoch": 0.08708045977011494,
      "grad_norm": 7.196598052978516,
      "learning_rate": 1.0566416040100252e-05,
      "loss": 0.4994352459907532,
      "num_input_tokens_seen": 715928,
      "step": 947,
      "train_runtime": 2147.5366,
      "train_tokens_per_second": 333.372
    },
    {
      "epoch": 0.08717241379310345,
      "grad_norm": 6.782207012176514,
      "learning_rate": 1.0556390977443609e-05,
      "loss": 0.4833005964756012,
      "num_input_tokens_seen": 716688,
      "step": 948,
      "train_runtime": 2149.4796,
      "train_tokens_per_second": 333.424
    },
    {
      "epoch": 0.08726436781609195,
      "grad_norm": 6.9549078941345215,
      "learning_rate": 1.054636591478697e-05,
      "loss": 0.5216882228851318,
      "num_input_tokens_seen": 717364,
      "step": 949,
      "train_runtime": 2151.4126,
      "train_tokens_per_second": 333.439
    },
    {
      "epoch": 0.08735632183908046,
      "grad_norm": 6.560072898864746,
      "learning_rate": 1.0536340852130326e-05,
      "loss": 0.5301787257194519,
      "num_input_tokens_seen": 718072,
      "step": 950,
      "train_runtime": 2153.3403,
      "train_tokens_per_second": 333.469
    },
    {
      "epoch": 0.08744827586206896,
      "grad_norm": 6.456213474273682,
      "learning_rate": 1.0526315789473684e-05,
      "loss": 0.4645478129386902,
      "num_input_tokens_seen": 718744,
      "step": 951,
      "train_runtime": 2155.2471,
      "train_tokens_per_second": 333.486
    },
    {
      "epoch": 0.08754022988505747,
      "grad_norm": 6.61658239364624,
      "learning_rate": 1.0516290726817045e-05,
      "loss": 0.5712957382202148,
      "num_input_tokens_seen": 719720,
      "step": 952,
      "train_runtime": 2157.194,
      "train_tokens_per_second": 333.637
    },
    {
      "epoch": 0.08763218390804597,
      "grad_norm": 7.224437713623047,
      "learning_rate": 1.0506265664160401e-05,
      "loss": 0.5404413938522339,
      "num_input_tokens_seen": 720388,
      "step": 953,
      "train_runtime": 2159.1051,
      "train_tokens_per_second": 333.651
    },
    {
      "epoch": 0.08772413793103448,
      "grad_norm": 6.876253128051758,
      "learning_rate": 1.049624060150376e-05,
      "loss": 0.4262760281562805,
      "num_input_tokens_seen": 721200,
      "step": 954,
      "train_runtime": 2161.1182,
      "train_tokens_per_second": 333.716
    },
    {
      "epoch": 0.08781609195402298,
      "grad_norm": 6.844622611999512,
      "learning_rate": 1.0486215538847118e-05,
      "loss": 0.40760573744773865,
      "num_input_tokens_seen": 722076,
      "step": 955,
      "train_runtime": 2163.0585,
      "train_tokens_per_second": 333.822
    },
    {
      "epoch": 0.08790804597701149,
      "grad_norm": 6.745266914367676,
      "learning_rate": 1.0476190476190477e-05,
      "loss": 0.49876174330711365,
      "num_input_tokens_seen": 722884,
      "step": 956,
      "train_runtime": 2164.9884,
      "train_tokens_per_second": 333.897
    },
    {
      "epoch": 0.088,
      "grad_norm": 7.548923969268799,
      "learning_rate": 1.0466165413533835e-05,
      "loss": 0.41772881150245667,
      "num_input_tokens_seen": 723728,
      "step": 957,
      "train_runtime": 2166.9611,
      "train_tokens_per_second": 333.983
    },
    {
      "epoch": 0.0880919540229885,
      "grad_norm": 6.423858642578125,
      "learning_rate": 1.0456140350877194e-05,
      "loss": 0.45678162574768066,
      "num_input_tokens_seen": 724616,
      "step": 958,
      "train_runtime": 2168.9082,
      "train_tokens_per_second": 334.093
    },
    {
      "epoch": 0.088183908045977,
      "grad_norm": 6.32468843460083,
      "learning_rate": 1.0446115288220552e-05,
      "loss": 0.4759761691093445,
      "num_input_tokens_seen": 725176,
      "step": 959,
      "train_runtime": 2170.8174,
      "train_tokens_per_second": 334.057
    },
    {
      "epoch": 0.08827586206896551,
      "grad_norm": 8.623871803283691,
      "learning_rate": 1.043609022556391e-05,
      "loss": 0.4497523307800293,
      "num_input_tokens_seen": 725992,
      "step": 960,
      "train_runtime": 2172.7712,
      "train_tokens_per_second": 334.132
    },
    {
      "epoch": 0.08836781609195403,
      "grad_norm": 6.408360481262207,
      "learning_rate": 1.0426065162907268e-05,
      "loss": 0.47394728660583496,
      "num_input_tokens_seen": 726620,
      "step": 961,
      "train_runtime": 2181.5509,
      "train_tokens_per_second": 333.075
    },
    {
      "epoch": 0.08845977011494253,
      "grad_norm": 7.175382614135742,
      "learning_rate": 1.0416040100250628e-05,
      "loss": 0.4901235103607178,
      "num_input_tokens_seen": 727352,
      "step": 962,
      "train_runtime": 2183.4621,
      "train_tokens_per_second": 333.119
    },
    {
      "epoch": 0.08855172413793104,
      "grad_norm": 6.749276161193848,
      "learning_rate": 1.0406015037593986e-05,
      "loss": 0.4491744041442871,
      "num_input_tokens_seen": 728012,
      "step": 963,
      "train_runtime": 2185.3676,
      "train_tokens_per_second": 333.13
    },
    {
      "epoch": 0.08864367816091955,
      "grad_norm": 6.454928398132324,
      "learning_rate": 1.0395989974937343e-05,
      "loss": 0.494658887386322,
      "num_input_tokens_seen": 729052,
      "step": 964,
      "train_runtime": 2187.3494,
      "train_tokens_per_second": 333.304
    },
    {
      "epoch": 0.08873563218390805,
      "grad_norm": 6.768080711364746,
      "learning_rate": 1.0385964912280702e-05,
      "loss": 0.4335147440433502,
      "num_input_tokens_seen": 730704,
      "step": 965,
      "train_runtime": 2189.3893,
      "train_tokens_per_second": 333.748
    },
    {
      "epoch": 0.08882758620689656,
      "grad_norm": 7.455011367797852,
      "learning_rate": 1.0375939849624062e-05,
      "loss": 0.5612114071846008,
      "num_input_tokens_seen": 731508,
      "step": 966,
      "train_runtime": 2191.3087,
      "train_tokens_per_second": 333.822
    },
    {
      "epoch": 0.08891954022988506,
      "grad_norm": 7.381419658660889,
      "learning_rate": 1.0365914786967419e-05,
      "loss": 0.5774643421173096,
      "num_input_tokens_seen": 732140,
      "step": 967,
      "train_runtime": 2193.2334,
      "train_tokens_per_second": 333.818
    },
    {
      "epoch": 0.08901149425287357,
      "grad_norm": 7.418336391448975,
      "learning_rate": 1.0355889724310777e-05,
      "loss": 0.4607248306274414,
      "num_input_tokens_seen": 732912,
      "step": 968,
      "train_runtime": 2195.1667,
      "train_tokens_per_second": 333.875
    },
    {
      "epoch": 0.08910344827586207,
      "grad_norm": 7.282698154449463,
      "learning_rate": 1.0345864661654137e-05,
      "loss": 0.5161183476448059,
      "num_input_tokens_seen": 733484,
      "step": 969,
      "train_runtime": 2197.0837,
      "train_tokens_per_second": 333.844
    },
    {
      "epoch": 0.08919540229885058,
      "grad_norm": 6.587283134460449,
      "learning_rate": 1.0335839598997494e-05,
      "loss": 0.4503616988658905,
      "num_input_tokens_seen": 734236,
      "step": 970,
      "train_runtime": 2199.1932,
      "train_tokens_per_second": 333.866
    },
    {
      "epoch": 0.08928735632183908,
      "grad_norm": 6.250298976898193,
      "learning_rate": 1.0325814536340853e-05,
      "loss": 0.5449129343032837,
      "num_input_tokens_seen": 735100,
      "step": 971,
      "train_runtime": 2201.1607,
      "train_tokens_per_second": 333.96
    },
    {
      "epoch": 0.08937931034482759,
      "grad_norm": 6.994154930114746,
      "learning_rate": 1.0315789473684213e-05,
      "loss": 0.4635671079158783,
      "num_input_tokens_seen": 735740,
      "step": 972,
      "train_runtime": 2203.0748,
      "train_tokens_per_second": 333.961
    },
    {
      "epoch": 0.08947126436781609,
      "grad_norm": 7.00700569152832,
      "learning_rate": 1.030576441102757e-05,
      "loss": 0.47756463289260864,
      "num_input_tokens_seen": 736600,
      "step": 973,
      "train_runtime": 2205.0136,
      "train_tokens_per_second": 334.057
    },
    {
      "epoch": 0.0895632183908046,
      "grad_norm": 6.1424126625061035,
      "learning_rate": 1.0295739348370928e-05,
      "loss": 0.457974910736084,
      "num_input_tokens_seen": 737128,
      "step": 974,
      "train_runtime": 2206.9115,
      "train_tokens_per_second": 334.009
    },
    {
      "epoch": 0.0896551724137931,
      "grad_norm": 6.277089595794678,
      "learning_rate": 1.0285714285714285e-05,
      "loss": 0.47886407375335693,
      "num_input_tokens_seen": 737712,
      "step": 975,
      "train_runtime": 2208.8354,
      "train_tokens_per_second": 333.982
    },
    {
      "epoch": 0.08974712643678161,
      "grad_norm": 7.152872085571289,
      "learning_rate": 1.0275689223057645e-05,
      "loss": 0.5071455836296082,
      "num_input_tokens_seen": 738496,
      "step": 976,
      "train_runtime": 2210.7804,
      "train_tokens_per_second": 334.043
    },
    {
      "epoch": 0.08983908045977011,
      "grad_norm": 6.826153755187988,
      "learning_rate": 1.0265664160401003e-05,
      "loss": 0.4909060001373291,
      "num_input_tokens_seen": 739424,
      "step": 977,
      "train_runtime": 2212.7298,
      "train_tokens_per_second": 334.168
    },
    {
      "epoch": 0.08993103448275862,
      "grad_norm": 7.014489650726318,
      "learning_rate": 1.025563909774436e-05,
      "loss": 0.7470390796661377,
      "num_input_tokens_seen": 740280,
      "step": 978,
      "train_runtime": 2214.6685,
      "train_tokens_per_second": 334.262
    },
    {
      "epoch": 0.09002298850574712,
      "grad_norm": 6.223145484924316,
      "learning_rate": 1.024561403508772e-05,
      "loss": 0.4532124102115631,
      "num_input_tokens_seen": 741056,
      "step": 979,
      "train_runtime": 2216.6006,
      "train_tokens_per_second": 334.321
    },
    {
      "epoch": 0.09011494252873563,
      "grad_norm": 6.236148357391357,
      "learning_rate": 1.0235588972431079e-05,
      "loss": 0.47539493441581726,
      "num_input_tokens_seen": 741672,
      "step": 980,
      "train_runtime": 2218.5179,
      "train_tokens_per_second": 334.31
    },
    {
      "epoch": 0.09020689655172413,
      "grad_norm": 7.284667491912842,
      "learning_rate": 1.0225563909774436e-05,
      "loss": 0.5616993308067322,
      "num_input_tokens_seen": 742476,
      "step": 981,
      "train_runtime": 2220.4934,
      "train_tokens_per_second": 334.374
    },
    {
      "epoch": 0.09029885057471264,
      "grad_norm": 7.2242865562438965,
      "learning_rate": 1.0215538847117796e-05,
      "loss": 0.44696980714797974,
      "num_input_tokens_seen": 743076,
      "step": 982,
      "train_runtime": 2222.4054,
      "train_tokens_per_second": 334.357
    },
    {
      "epoch": 0.09039080459770114,
      "grad_norm": 6.407309055328369,
      "learning_rate": 1.0205513784461154e-05,
      "loss": 0.45359712839126587,
      "num_input_tokens_seen": 743724,
      "step": 983,
      "train_runtime": 2224.3194,
      "train_tokens_per_second": 334.36
    },
    {
      "epoch": 0.09048275862068965,
      "grad_norm": 6.019703388214111,
      "learning_rate": 1.0195488721804511e-05,
      "loss": 0.4540659189224243,
      "num_input_tokens_seen": 744484,
      "step": 984,
      "train_runtime": 2226.2369,
      "train_tokens_per_second": 334.414
    },
    {
      "epoch": 0.09057471264367815,
      "grad_norm": 6.675520896911621,
      "learning_rate": 1.018546365914787e-05,
      "loss": 0.4840589761734009,
      "num_input_tokens_seen": 745088,
      "step": 985,
      "train_runtime": 2228.2202,
      "train_tokens_per_second": 334.387
    },
    {
      "epoch": 0.09066666666666667,
      "grad_norm": 7.436343193054199,
      "learning_rate": 1.017543859649123e-05,
      "loss": 0.544686496257782,
      "num_input_tokens_seen": 745768,
      "step": 986,
      "train_runtime": 2230.1297,
      "train_tokens_per_second": 334.406
    },
    {
      "epoch": 0.09075862068965518,
      "grad_norm": 6.213075160980225,
      "learning_rate": 1.0165413533834587e-05,
      "loss": 0.5067095160484314,
      "num_input_tokens_seen": 746480,
      "step": 987,
      "train_runtime": 2232.0565,
      "train_tokens_per_second": 334.436
    },
    {
      "epoch": 0.09085057471264368,
      "grad_norm": 6.276841640472412,
      "learning_rate": 1.0155388471177945e-05,
      "loss": 0.4725772738456726,
      "num_input_tokens_seen": 747052,
      "step": 988,
      "train_runtime": 2233.9724,
      "train_tokens_per_second": 334.405
    },
    {
      "epoch": 0.09094252873563219,
      "grad_norm": 6.9957685470581055,
      "learning_rate": 1.0145363408521305e-05,
      "loss": 0.46687793731689453,
      "num_input_tokens_seen": 747804,
      "step": 989,
      "train_runtime": 2235.8855,
      "train_tokens_per_second": 334.455
    },
    {
      "epoch": 0.0910344827586207,
      "grad_norm": 8.03308391571045,
      "learning_rate": 1.0135338345864662e-05,
      "loss": 0.6137831211090088,
      "num_input_tokens_seen": 748572,
      "step": 990,
      "train_runtime": 2237.7989,
      "train_tokens_per_second": 334.513
    },
    {
      "epoch": 0.0911264367816092,
      "grad_norm": 6.838526248931885,
      "learning_rate": 1.012531328320802e-05,
      "loss": 0.5152433514595032,
      "num_input_tokens_seen": 749308,
      "step": 991,
      "train_runtime": 2246.6315,
      "train_tokens_per_second": 333.525
    },
    {
      "epoch": 0.0912183908045977,
      "grad_norm": 6.151811599731445,
      "learning_rate": 1.011528822055138e-05,
      "loss": 0.4851880669593811,
      "num_input_tokens_seen": 750168,
      "step": 992,
      "train_runtime": 2248.5729,
      "train_tokens_per_second": 333.62
    },
    {
      "epoch": 0.09131034482758621,
      "grad_norm": 6.756059646606445,
      "learning_rate": 1.0105263157894738e-05,
      "loss": 0.5420541763305664,
      "num_input_tokens_seen": 750828,
      "step": 993,
      "train_runtime": 2250.5831,
      "train_tokens_per_second": 333.615
    },
    {
      "epoch": 0.09140229885057471,
      "grad_norm": 6.793330192565918,
      "learning_rate": 1.0095238095238096e-05,
      "loss": 0.4886292517185211,
      "num_input_tokens_seen": 751688,
      "step": 994,
      "train_runtime": 2252.5376,
      "train_tokens_per_second": 333.707
    },
    {
      "epoch": 0.09149425287356322,
      "grad_norm": 6.752333164215088,
      "learning_rate": 1.0085213032581453e-05,
      "loss": 0.5576735734939575,
      "num_input_tokens_seen": 752396,
      "step": 995,
      "train_runtime": 2254.4697,
      "train_tokens_per_second": 333.735
    },
    {
      "epoch": 0.09158620689655173,
      "grad_norm": 6.473622798919678,
      "learning_rate": 1.0075187969924813e-05,
      "loss": 0.44832098484039307,
      "num_input_tokens_seen": 753084,
      "step": 996,
      "train_runtime": 2256.4224,
      "train_tokens_per_second": 333.751
    },
    {
      "epoch": 0.09167816091954023,
      "grad_norm": 6.3503241539001465,
      "learning_rate": 1.0065162907268171e-05,
      "loss": 0.4546602964401245,
      "num_input_tokens_seen": 754052,
      "step": 997,
      "train_runtime": 2258.4717,
      "train_tokens_per_second": 333.877
    },
    {
      "epoch": 0.09177011494252874,
      "grad_norm": 7.291573524475098,
      "learning_rate": 1.0055137844611528e-05,
      "loss": 0.6066989898681641,
      "num_input_tokens_seen": 754880,
      "step": 998,
      "train_runtime": 2260.403,
      "train_tokens_per_second": 333.958
    },
    {
      "epoch": 0.09186206896551724,
      "grad_norm": 7.389471530914307,
      "learning_rate": 1.0045112781954888e-05,
      "loss": 0.41600704193115234,
      "num_input_tokens_seen": 755516,
      "step": 999,
      "train_runtime": 2262.3346,
      "train_tokens_per_second": 333.954
    },
    {
      "epoch": 0.09195402298850575,
      "grad_norm": 6.413146495819092,
      "learning_rate": 1.0035087719298247e-05,
      "loss": 0.39765024185180664,
      "num_input_tokens_seen": 756312,
      "step": 1000,
      "train_runtime": 2264.28,
      "train_tokens_per_second": 334.019
    },
    {
      "epoch": 0.09195402298850575,
      "eval_loss": 1.8523097038269043,
      "eval_runtime": 27.9112,
      "eval_samples_per_second": 35.828,
      "eval_steps_per_second": 8.957,
      "num_input_tokens_seen": 756312,
      "step": 1000
    },
    {
      "epoch": 0.09204597701149425,
      "grad_norm": 6.664978981018066,
      "learning_rate": 1.0025062656641604e-05,
      "loss": 0.5257925987243652,
      "num_input_tokens_seen": 756928,
      "step": 1001,
      "train_runtime": 2294.1764,
      "train_tokens_per_second": 329.935
    },
    {
      "epoch": 0.09213793103448276,
      "grad_norm": 7.344442844390869,
      "learning_rate": 1.0015037593984964e-05,
      "loss": 0.46458640694618225,
      "num_input_tokens_seen": 757640,
      "step": 1002,
      "train_runtime": 2296.1199,
      "train_tokens_per_second": 329.965
    },
    {
      "epoch": 0.09222988505747126,
      "grad_norm": 7.410263538360596,
      "learning_rate": 1.0005012531328322e-05,
      "loss": 0.5652271509170532,
      "num_input_tokens_seen": 758324,
      "step": 1003,
      "train_runtime": 2298.0398,
      "train_tokens_per_second": 329.987
    },
    {
      "epoch": 0.09232183908045977,
      "grad_norm": 8.379242897033691,
      "learning_rate": 9.99498746867168e-06,
      "loss": 0.5291005373001099,
      "num_input_tokens_seen": 758924,
      "step": 1004,
      "train_runtime": 2299.9606,
      "train_tokens_per_second": 329.973
    },
    {
      "epoch": 0.09241379310344827,
      "grad_norm": 8.128753662109375,
      "learning_rate": 9.984962406015038e-06,
      "loss": 0.6162893772125244,
      "num_input_tokens_seen": 759480,
      "step": 1005,
      "train_runtime": 2301.8825,
      "train_tokens_per_second": 329.939
    },
    {
      "epoch": 0.09250574712643678,
      "grad_norm": 6.83087682723999,
      "learning_rate": 9.974937343358396e-06,
      "loss": 0.4586988687515259,
      "num_input_tokens_seen": 760144,
      "step": 1006,
      "train_runtime": 2303.8003,
      "train_tokens_per_second": 329.952
    },
    {
      "epoch": 0.09259770114942528,
      "grad_norm": 6.153554439544678,
      "learning_rate": 9.964912280701755e-06,
      "loss": 0.4497118592262268,
      "num_input_tokens_seen": 760904,
      "step": 1007,
      "train_runtime": 2305.7437,
      "train_tokens_per_second": 330.004
    },
    {
      "epoch": 0.09268965517241379,
      "grad_norm": 6.233102321624756,
      "learning_rate": 9.954887218045113e-06,
      "loss": 0.4667837917804718,
      "num_input_tokens_seen": 761492,
      "step": 1008,
      "train_runtime": 2307.6686,
      "train_tokens_per_second": 329.983
    },
    {
      "epoch": 0.09278160919540229,
      "grad_norm": 6.957793235778809,
      "learning_rate": 9.944862155388472e-06,
      "loss": 0.4331594407558441,
      "num_input_tokens_seen": 762456,
      "step": 1009,
      "train_runtime": 2309.6656,
      "train_tokens_per_second": 330.115
    },
    {
      "epoch": 0.0928735632183908,
      "grad_norm": 9.187891006469727,
      "learning_rate": 9.93483709273183e-06,
      "loss": 0.5834053754806519,
      "num_input_tokens_seen": 763152,
      "step": 1010,
      "train_runtime": 2311.5771,
      "train_tokens_per_second": 330.143
    },
    {
      "epoch": 0.09296551724137932,
      "grad_norm": 7.78358268737793,
      "learning_rate": 9.924812030075189e-06,
      "loss": 0.46204400062561035,
      "num_input_tokens_seen": 763780,
      "step": 1011,
      "train_runtime": 2313.4962,
      "train_tokens_per_second": 330.141
    },
    {
      "epoch": 0.09305747126436782,
      "grad_norm": 6.727712631225586,
      "learning_rate": 9.914786967418547e-06,
      "loss": 0.3884506821632385,
      "num_input_tokens_seen": 764636,
      "step": 1012,
      "train_runtime": 2315.4506,
      "train_tokens_per_second": 330.232
    },
    {
      "epoch": 0.09314942528735633,
      "grad_norm": 6.322958946228027,
      "learning_rate": 9.904761904761906e-06,
      "loss": 0.42635682225227356,
      "num_input_tokens_seen": 765280,
      "step": 1013,
      "train_runtime": 2317.3708,
      "train_tokens_per_second": 330.236
    },
    {
      "epoch": 0.09324137931034483,
      "grad_norm": 6.7299909591674805,
      "learning_rate": 9.894736842105264e-06,
      "loss": 0.6027582883834839,
      "num_input_tokens_seen": 766368,
      "step": 1014,
      "train_runtime": 2319.357,
      "train_tokens_per_second": 330.423
    },
    {
      "epoch": 0.09333333333333334,
      "grad_norm": 6.988217830657959,
      "learning_rate": 9.884711779448623e-06,
      "loss": 0.4799932837486267,
      "num_input_tokens_seen": 767000,
      "step": 1015,
      "train_runtime": 2321.2654,
      "train_tokens_per_second": 330.423
    },
    {
      "epoch": 0.09342528735632184,
      "grad_norm": 6.878663539886475,
      "learning_rate": 9.87468671679198e-06,
      "loss": 0.5265244245529175,
      "num_input_tokens_seen": 767704,
      "step": 1016,
      "train_runtime": 2323.1791,
      "train_tokens_per_second": 330.454
    },
    {
      "epoch": 0.09351724137931035,
      "grad_norm": 6.759012222290039,
      "learning_rate": 9.86466165413534e-06,
      "loss": 0.41557538509368896,
      "num_input_tokens_seen": 768352,
      "step": 1017,
      "train_runtime": 2325.0907,
      "train_tokens_per_second": 330.461
    },
    {
      "epoch": 0.09360919540229885,
      "grad_norm": 7.261196613311768,
      "learning_rate": 9.854636591478698e-06,
      "loss": 0.5402182936668396,
      "num_input_tokens_seen": 769004,
      "step": 1018,
      "train_runtime": 2327.0003,
      "train_tokens_per_second": 330.47
    },
    {
      "epoch": 0.09370114942528736,
      "grad_norm": 7.715246677398682,
      "learning_rate": 9.844611528822055e-06,
      "loss": 0.5894691944122314,
      "num_input_tokens_seen": 769816,
      "step": 1019,
      "train_runtime": 2328.9826,
      "train_tokens_per_second": 330.537
    },
    {
      "epoch": 0.09379310344827586,
      "grad_norm": 9.650315284729004,
      "learning_rate": 9.834586466165415e-06,
      "loss": 0.48343488574028015,
      "num_input_tokens_seen": 770492,
      "step": 1020,
      "train_runtime": 2330.9136,
      "train_tokens_per_second": 330.554
    },
    {
      "epoch": 0.09388505747126437,
      "grad_norm": 12.389888763427734,
      "learning_rate": 9.824561403508772e-06,
      "loss": 0.49280035495758057,
      "num_input_tokens_seen": 771328,
      "step": 1021,
      "train_runtime": 2339.7748,
      "train_tokens_per_second": 329.659
    },
    {
      "epoch": 0.09397701149425287,
      "grad_norm": 6.612550735473633,
      "learning_rate": 9.81453634085213e-06,
      "loss": 0.48790669441223145,
      "num_input_tokens_seen": 772216,
      "step": 1022,
      "train_runtime": 2341.7279,
      "train_tokens_per_second": 329.763
    },
    {
      "epoch": 0.09406896551724138,
      "grad_norm": 6.0658440589904785,
      "learning_rate": 9.80451127819549e-06,
      "loss": 0.4938814640045166,
      "num_input_tokens_seen": 773264,
      "step": 1023,
      "train_runtime": 2343.8558,
      "train_tokens_per_second": 329.911
    },
    {
      "epoch": 0.09416091954022988,
      "grad_norm": 5.443805694580078,
      "learning_rate": 9.794486215538847e-06,
      "loss": 0.4375538229942322,
      "num_input_tokens_seen": 774336,
      "step": 1024,
      "train_runtime": 2345.8428,
      "train_tokens_per_second": 330.089
    },
    {
      "epoch": 0.09425287356321839,
      "grad_norm": 7.415752410888672,
      "learning_rate": 9.784461152882206e-06,
      "loss": 0.5282442569732666,
      "num_input_tokens_seen": 775188,
      "step": 1025,
      "train_runtime": 2347.7896,
      "train_tokens_per_second": 330.178
    },
    {
      "epoch": 0.0943448275862069,
      "grad_norm": 6.058030128479004,
      "learning_rate": 9.774436090225564e-06,
      "loss": 0.3603585958480835,
      "num_input_tokens_seen": 775800,
      "step": 1026,
      "train_runtime": 2349.6978,
      "train_tokens_per_second": 330.17
    },
    {
      "epoch": 0.0944367816091954,
      "grad_norm": 6.946608066558838,
      "learning_rate": 9.764411027568923e-06,
      "loss": 0.5203933715820312,
      "num_input_tokens_seen": 776552,
      "step": 1027,
      "train_runtime": 2351.6785,
      "train_tokens_per_second": 330.212
    },
    {
      "epoch": 0.0945287356321839,
      "grad_norm": 6.649303913116455,
      "learning_rate": 9.754385964912281e-06,
      "loss": 0.47817522287368774,
      "num_input_tokens_seen": 777548,
      "step": 1028,
      "train_runtime": 2353.6686,
      "train_tokens_per_second": 330.356
    },
    {
      "epoch": 0.09462068965517241,
      "grad_norm": 6.833725452423096,
      "learning_rate": 9.74436090225564e-06,
      "loss": 0.4406696856021881,
      "num_input_tokens_seen": 778188,
      "step": 1029,
      "train_runtime": 2355.5837,
      "train_tokens_per_second": 330.359
    },
    {
      "epoch": 0.09471264367816092,
      "grad_norm": 6.373605251312256,
      "learning_rate": 9.734335839598998e-06,
      "loss": 0.43867525458335876,
      "num_input_tokens_seen": 778788,
      "step": 1030,
      "train_runtime": 2357.4893,
      "train_tokens_per_second": 330.346
    },
    {
      "epoch": 0.09480459770114942,
      "grad_norm": 6.011414051055908,
      "learning_rate": 9.724310776942357e-06,
      "loss": 0.4227856993675232,
      "num_input_tokens_seen": 779704,
      "step": 1031,
      "train_runtime": 2359.4771,
      "train_tokens_per_second": 330.456
    },
    {
      "epoch": 0.09489655172413793,
      "grad_norm": 6.286978244781494,
      "learning_rate": 9.714285714285715e-06,
      "loss": 0.36833497881889343,
      "num_input_tokens_seen": 780308,
      "step": 1032,
      "train_runtime": 2361.3896,
      "train_tokens_per_second": 330.444
    },
    {
      "epoch": 0.09498850574712643,
      "grad_norm": 8.112756729125977,
      "learning_rate": 9.704260651629074e-06,
      "loss": 0.47532498836517334,
      "num_input_tokens_seen": 781060,
      "step": 1033,
      "train_runtime": 2363.3331,
      "train_tokens_per_second": 330.491
    },
    {
      "epoch": 0.09508045977011494,
      "grad_norm": 7.036821365356445,
      "learning_rate": 9.694235588972432e-06,
      "loss": 0.5230494737625122,
      "num_input_tokens_seen": 781756,
      "step": 1034,
      "train_runtime": 2365.243,
      "train_tokens_per_second": 330.518
    },
    {
      "epoch": 0.09517241379310344,
      "grad_norm": 8.099452018737793,
      "learning_rate": 9.68421052631579e-06,
      "loss": 0.503882884979248,
      "num_input_tokens_seen": 782404,
      "step": 1035,
      "train_runtime": 2367.1472,
      "train_tokens_per_second": 330.526
    },
    {
      "epoch": 0.09526436781609196,
      "grad_norm": 6.2233076095581055,
      "learning_rate": 9.674185463659147e-06,
      "loss": 0.43777409195899963,
      "num_input_tokens_seen": 783036,
      "step": 1036,
      "train_runtime": 2369.0585,
      "train_tokens_per_second": 330.526
    },
    {
      "epoch": 0.09535632183908047,
      "grad_norm": 7.426332473754883,
      "learning_rate": 9.664160401002508e-06,
      "loss": 0.4811877906322479,
      "num_input_tokens_seen": 783604,
      "step": 1037,
      "train_runtime": 2371.0943,
      "train_tokens_per_second": 330.482
    },
    {
      "epoch": 0.09544827586206897,
      "grad_norm": 7.485308647155762,
      "learning_rate": 9.654135338345866e-06,
      "loss": 0.5360345840454102,
      "num_input_tokens_seen": 784320,
      "step": 1038,
      "train_runtime": 2373.0419,
      "train_tokens_per_second": 330.512
    },
    {
      "epoch": 0.09554022988505748,
      "grad_norm": 7.009185314178467,
      "learning_rate": 9.644110275689223e-06,
      "loss": 0.5393097400665283,
      "num_input_tokens_seen": 785092,
      "step": 1039,
      "train_runtime": 2374.9493,
      "train_tokens_per_second": 330.572
    },
    {
      "epoch": 0.09563218390804598,
      "grad_norm": 6.129383563995361,
      "learning_rate": 9.634085213032583e-06,
      "loss": 0.4603983759880066,
      "num_input_tokens_seen": 785788,
      "step": 1040,
      "train_runtime": 2376.9207,
      "train_tokens_per_second": 330.591
    },
    {
      "epoch": 0.09572413793103449,
      "grad_norm": 6.2865519523620605,
      "learning_rate": 9.62406015037594e-06,
      "loss": 0.42140841484069824,
      "num_input_tokens_seen": 786404,
      "step": 1041,
      "train_runtime": 2378.8442,
      "train_tokens_per_second": 330.582
    },
    {
      "epoch": 0.09581609195402299,
      "grad_norm": 8.007585525512695,
      "learning_rate": 9.614035087719298e-06,
      "loss": 0.5779958367347717,
      "num_input_tokens_seen": 787304,
      "step": 1042,
      "train_runtime": 2380.7954,
      "train_tokens_per_second": 330.689
    },
    {
      "epoch": 0.0959080459770115,
      "grad_norm": 6.392016410827637,
      "learning_rate": 9.604010025062659e-06,
      "loss": 0.5943008661270142,
      "num_input_tokens_seen": 788388,
      "step": 1043,
      "train_runtime": 2382.7425,
      "train_tokens_per_second": 330.874
    },
    {
      "epoch": 0.096,
      "grad_norm": 6.230937957763672,
      "learning_rate": 9.593984962406015e-06,
      "loss": 0.5474150776863098,
      "num_input_tokens_seen": 790628,
      "step": 1044,
      "train_runtime": 2384.8541,
      "train_tokens_per_second": 331.52
    },
    {
      "epoch": 0.09609195402298851,
      "grad_norm": 7.178816318511963,
      "learning_rate": 9.583959899749374e-06,
      "loss": 0.46896564960479736,
      "num_input_tokens_seen": 791384,
      "step": 1045,
      "train_runtime": 2386.8215,
      "train_tokens_per_second": 331.564
    },
    {
      "epoch": 0.09618390804597701,
      "grad_norm": 6.814177513122559,
      "learning_rate": 9.573934837092732e-06,
      "loss": 0.4675101637840271,
      "num_input_tokens_seen": 792124,
      "step": 1046,
      "train_runtime": 2388.7697,
      "train_tokens_per_second": 331.603
    },
    {
      "epoch": 0.09627586206896552,
      "grad_norm": 8.564444541931152,
      "learning_rate": 9.56390977443609e-06,
      "loss": 0.4232885241508484,
      "num_input_tokens_seen": 792688,
      "step": 1047,
      "train_runtime": 2390.6854,
      "train_tokens_per_second": 331.574
    },
    {
      "epoch": 0.09636781609195402,
      "grad_norm": 10.102237701416016,
      "learning_rate": 9.55388471177945e-06,
      "loss": 0.4183449149131775,
      "num_input_tokens_seen": 793280,
      "step": 1048,
      "train_runtime": 2392.5987,
      "train_tokens_per_second": 331.556
    },
    {
      "epoch": 0.09645977011494253,
      "grad_norm": 7.193587779998779,
      "learning_rate": 9.543859649122808e-06,
      "loss": 0.5160861015319824,
      "num_input_tokens_seen": 794048,
      "step": 1049,
      "train_runtime": 2394.5092,
      "train_tokens_per_second": 331.612
    },
    {
      "epoch": 0.09655172413793103,
      "grad_norm": 7.2332024574279785,
      "learning_rate": 9.533834586466166e-06,
      "loss": 0.5566169023513794,
      "num_input_tokens_seen": 794764,
      "step": 1050,
      "train_runtime": 2396.4189,
      "train_tokens_per_second": 331.647
    },
    {
      "epoch": 0.09664367816091954,
      "grad_norm": 7.799232006072998,
      "learning_rate": 9.523809523809525e-06,
      "loss": 0.5159203410148621,
      "num_input_tokens_seen": 795508,
      "step": 1051,
      "train_runtime": 2405.4812,
      "train_tokens_per_second": 330.706
    },
    {
      "epoch": 0.09673563218390804,
      "grad_norm": 14.995935440063477,
      "learning_rate": 9.513784461152883e-06,
      "loss": 0.45017606019973755,
      "num_input_tokens_seen": 796120,
      "step": 1052,
      "train_runtime": 2407.388,
      "train_tokens_per_second": 330.699
    },
    {
      "epoch": 0.09682758620689655,
      "grad_norm": 6.2709503173828125,
      "learning_rate": 9.503759398496242e-06,
      "loss": 0.4437541365623474,
      "num_input_tokens_seen": 797056,
      "step": 1053,
      "train_runtime": 2409.3635,
      "train_tokens_per_second": 330.816
    },
    {
      "epoch": 0.09691954022988505,
      "grad_norm": 6.873055458068848,
      "learning_rate": 9.4937343358396e-06,
      "loss": 0.5023096799850464,
      "num_input_tokens_seen": 797724,
      "step": 1054,
      "train_runtime": 2411.2752,
      "train_tokens_per_second": 330.831
    },
    {
      "epoch": 0.09701149425287356,
      "grad_norm": 6.0630202293396,
      "learning_rate": 9.483709273182959e-06,
      "loss": 0.4279667139053345,
      "num_input_tokens_seen": 798576,
      "step": 1055,
      "train_runtime": 2413.2174,
      "train_tokens_per_second": 330.918
    },
    {
      "epoch": 0.09710344827586206,
      "grad_norm": 5.470777988433838,
      "learning_rate": 9.473684210526315e-06,
      "loss": 0.32489827275276184,
      "num_input_tokens_seen": 799192,
      "step": 1056,
      "train_runtime": 2415.1235,
      "train_tokens_per_second": 330.911
    },
    {
      "epoch": 0.09719540229885057,
      "grad_norm": 6.455587387084961,
      "learning_rate": 9.463659147869676e-06,
      "loss": 0.5397228002548218,
      "num_input_tokens_seen": 799772,
      "step": 1057,
      "train_runtime": 2417.0295,
      "train_tokens_per_second": 330.89
    },
    {
      "epoch": 0.09728735632183907,
      "grad_norm": 6.85917854309082,
      "learning_rate": 9.453634085213034e-06,
      "loss": 0.47662365436553955,
      "num_input_tokens_seen": 800536,
      "step": 1058,
      "train_runtime": 2418.9768,
      "train_tokens_per_second": 330.94
    },
    {
      "epoch": 0.09737931034482758,
      "grad_norm": 7.232198715209961,
      "learning_rate": 9.443609022556391e-06,
      "loss": 0.3814907670021057,
      "num_input_tokens_seen": 801092,
      "step": 1059,
      "train_runtime": 2420.8882,
      "train_tokens_per_second": 330.908
    },
    {
      "epoch": 0.09747126436781609,
      "grad_norm": 7.328190803527832,
      "learning_rate": 9.43358395989975e-06,
      "loss": 0.5666010975837708,
      "num_input_tokens_seen": 801832,
      "step": 1060,
      "train_runtime": 2422.8456,
      "train_tokens_per_second": 330.946
    },
    {
      "epoch": 0.0975632183908046,
      "grad_norm": 6.977006435394287,
      "learning_rate": 9.423558897243108e-06,
      "loss": 0.477314829826355,
      "num_input_tokens_seen": 802572,
      "step": 1061,
      "train_runtime": 2424.7861,
      "train_tokens_per_second": 330.987
    },
    {
      "epoch": 0.09765517241379311,
      "grad_norm": 7.465899467468262,
      "learning_rate": 9.413533834586466e-06,
      "loss": 0.44286948442459106,
      "num_input_tokens_seen": 803200,
      "step": 1062,
      "train_runtime": 2426.703,
      "train_tokens_per_second": 330.984
    },
    {
      "epoch": 0.09774712643678161,
      "grad_norm": 7.511911869049072,
      "learning_rate": 9.403508771929825e-06,
      "loss": 0.5399251580238342,
      "num_input_tokens_seen": 804012,
      "step": 1063,
      "train_runtime": 2428.6794,
      "train_tokens_per_second": 331.049
    },
    {
      "epoch": 0.09783908045977012,
      "grad_norm": 6.484589099884033,
      "learning_rate": 9.393483709273183e-06,
      "loss": 0.46299657225608826,
      "num_input_tokens_seen": 804604,
      "step": 1064,
      "train_runtime": 2430.745,
      "train_tokens_per_second": 331.011
    },
    {
      "epoch": 0.09793103448275862,
      "grad_norm": 7.001509189605713,
      "learning_rate": 9.383458646616542e-06,
      "loss": 0.5671294331550598,
      "num_input_tokens_seen": 805624,
      "step": 1065,
      "train_runtime": 2432.7273,
      "train_tokens_per_second": 331.161
    },
    {
      "epoch": 0.09802298850574713,
      "grad_norm": 6.694453716278076,
      "learning_rate": 9.3734335839599e-06,
      "loss": 0.4465494155883789,
      "num_input_tokens_seen": 806336,
      "step": 1066,
      "train_runtime": 2434.6512,
      "train_tokens_per_second": 331.192
    },
    {
      "epoch": 0.09811494252873564,
      "grad_norm": 8.25455379486084,
      "learning_rate": 9.363408521303259e-06,
      "loss": 0.4501800537109375,
      "num_input_tokens_seen": 807000,
      "step": 1067,
      "train_runtime": 2436.563,
      "train_tokens_per_second": 331.204
    },
    {
      "epoch": 0.09820689655172414,
      "grad_norm": 6.203482627868652,
      "learning_rate": 9.353383458646617e-06,
      "loss": 0.45742660760879517,
      "num_input_tokens_seen": 807544,
      "step": 1068,
      "train_runtime": 2438.4891,
      "train_tokens_per_second": 331.166
    },
    {
      "epoch": 0.09829885057471265,
      "grad_norm": 6.2609944343566895,
      "learning_rate": 9.343358395989976e-06,
      "loss": 0.454520583152771,
      "num_input_tokens_seen": 808232,
      "step": 1069,
      "train_runtime": 2440.4183,
      "train_tokens_per_second": 331.186
    },
    {
      "epoch": 0.09839080459770115,
      "grad_norm": 7.969869613647461,
      "learning_rate": 9.333333333333334e-06,
      "loss": 0.45502176880836487,
      "num_input_tokens_seen": 809000,
      "step": 1070,
      "train_runtime": 2442.3543,
      "train_tokens_per_second": 331.238
    },
    {
      "epoch": 0.09848275862068966,
      "grad_norm": 8.279662132263184,
      "learning_rate": 9.323308270676693e-06,
      "loss": 0.5213327407836914,
      "num_input_tokens_seen": 809616,
      "step": 1071,
      "train_runtime": 2444.2928,
      "train_tokens_per_second": 331.227
    },
    {
      "epoch": 0.09857471264367816,
      "grad_norm": 6.863427639007568,
      "learning_rate": 9.313283208020051e-06,
      "loss": 0.35063788294792175,
      "num_input_tokens_seen": 810140,
      "step": 1072,
      "train_runtime": 2446.2168,
      "train_tokens_per_second": 331.181
    },
    {
      "epoch": 0.09866666666666667,
      "grad_norm": 6.921423435211182,
      "learning_rate": 9.30325814536341e-06,
      "loss": 0.43684956431388855,
      "num_input_tokens_seen": 810724,
      "step": 1073,
      "train_runtime": 2448.15,
      "train_tokens_per_second": 331.158
    },
    {
      "epoch": 0.09875862068965517,
      "grad_norm": 8.909449577331543,
      "learning_rate": 9.293233082706767e-06,
      "loss": 0.3925761580467224,
      "num_input_tokens_seen": 811896,
      "step": 1074,
      "train_runtime": 2450.2017,
      "train_tokens_per_second": 331.359
    },
    {
      "epoch": 0.09885057471264368,
      "grad_norm": 7.516204833984375,
      "learning_rate": 9.283208020050127e-06,
      "loss": 0.4987984895706177,
      "num_input_tokens_seen": 812572,
      "step": 1075,
      "train_runtime": 2452.1206,
      "train_tokens_per_second": 331.375
    },
    {
      "epoch": 0.09894252873563218,
      "grad_norm": 6.935107707977295,
      "learning_rate": 9.273182957393484e-06,
      "loss": 0.4580703377723694,
      "num_input_tokens_seen": 813196,
      "step": 1076,
      "train_runtime": 2454.0342,
      "train_tokens_per_second": 331.371
    },
    {
      "epoch": 0.09903448275862069,
      "grad_norm": 5.472903728485107,
      "learning_rate": 9.263157894736842e-06,
      "loss": 0.37355905771255493,
      "num_input_tokens_seen": 813836,
      "step": 1077,
      "train_runtime": 2455.9524,
      "train_tokens_per_second": 331.373
    },
    {
      "epoch": 0.09912643678160919,
      "grad_norm": 6.800434589385986,
      "learning_rate": 9.253132832080202e-06,
      "loss": 0.4778006970882416,
      "num_input_tokens_seen": 814488,
      "step": 1078,
      "train_runtime": 2457.8663,
      "train_tokens_per_second": 331.38
    },
    {
      "epoch": 0.0992183908045977,
      "grad_norm": 6.6643290519714355,
      "learning_rate": 9.243107769423559e-06,
      "loss": 0.53313809633255,
      "num_input_tokens_seen": 815840,
      "step": 1079,
      "train_runtime": 2459.8754,
      "train_tokens_per_second": 331.659
    },
    {
      "epoch": 0.0993103448275862,
      "grad_norm": 6.268790245056152,
      "learning_rate": 9.233082706766918e-06,
      "loss": 0.4480176568031311,
      "num_input_tokens_seen": 816696,
      "step": 1080,
      "train_runtime": 2461.8252,
      "train_tokens_per_second": 331.744
    },
    {
      "epoch": 0.09940229885057471,
      "grad_norm": 6.700995445251465,
      "learning_rate": 9.223057644110276e-06,
      "loss": 0.3908253312110901,
      "num_input_tokens_seen": 817340,
      "step": 1081,
      "train_runtime": 2470.706,
      "train_tokens_per_second": 330.812
    },
    {
      "epoch": 0.09949425287356321,
      "grad_norm": 7.005143165588379,
      "learning_rate": 9.213032581453634e-06,
      "loss": 0.5249738693237305,
      "num_input_tokens_seen": 817884,
      "step": 1082,
      "train_runtime": 2472.6545,
      "train_tokens_per_second": 330.772
    },
    {
      "epoch": 0.09958620689655172,
      "grad_norm": 6.341146469116211,
      "learning_rate": 9.203007518796993e-06,
      "loss": 0.4519422650337219,
      "num_input_tokens_seen": 818568,
      "step": 1083,
      "train_runtime": 2474.5714,
      "train_tokens_per_second": 330.792
    },
    {
      "epoch": 0.09967816091954022,
      "grad_norm": 6.9556708335876465,
      "learning_rate": 9.192982456140351e-06,
      "loss": 0.5711106061935425,
      "num_input_tokens_seen": 819528,
      "step": 1084,
      "train_runtime": 2476.5501,
      "train_tokens_per_second": 330.915
    },
    {
      "epoch": 0.09977011494252873,
      "grad_norm": 6.9345269203186035,
      "learning_rate": 9.18295739348371e-06,
      "loss": 0.5837572813034058,
      "num_input_tokens_seen": 820140,
      "step": 1085,
      "train_runtime": 2478.641,
      "train_tokens_per_second": 330.883
    },
    {
      "epoch": 0.09986206896551725,
      "grad_norm": 7.81229305267334,
      "learning_rate": 9.172932330827068e-06,
      "loss": 0.5010105967521667,
      "num_input_tokens_seen": 820840,
      "step": 1086,
      "train_runtime": 2480.5816,
      "train_tokens_per_second": 330.906
    },
    {
      "epoch": 0.09995402298850575,
      "grad_norm": 6.963803768157959,
      "learning_rate": 9.162907268170427e-06,
      "loss": 0.45326730608940125,
      "num_input_tokens_seen": 821532,
      "step": 1087,
      "train_runtime": 2482.5694,
      "train_tokens_per_second": 330.92
    },
    {
      "epoch": 0.10004597701149426,
      "grad_norm": 6.994347095489502,
      "learning_rate": 9.152882205513785e-06,
      "loss": 0.6128959655761719,
      "num_input_tokens_seen": 822252,
      "step": 1088,
      "train_runtime": 2484.4887,
      "train_tokens_per_second": 330.954
    },
    {
      "epoch": 0.10013793103448276,
      "grad_norm": 6.74661111831665,
      "learning_rate": 9.142857142857144e-06,
      "loss": 0.49878284335136414,
      "num_input_tokens_seen": 822928,
      "step": 1089,
      "train_runtime": 2486.414,
      "train_tokens_per_second": 330.97
    },
    {
      "epoch": 0.10022988505747127,
      "grad_norm": 7.087134838104248,
      "learning_rate": 9.132832080200502e-06,
      "loss": 0.43717753887176514,
      "num_input_tokens_seen": 823580,
      "step": 1090,
      "train_runtime": 2488.4213,
      "train_tokens_per_second": 330.965
    },
    {
      "epoch": 0.10032183908045977,
      "grad_norm": 5.783131122589111,
      "learning_rate": 9.12280701754386e-06,
      "loss": 0.3840745687484741,
      "num_input_tokens_seen": 824268,
      "step": 1091,
      "train_runtime": 2490.3564,
      "train_tokens_per_second": 330.984
    },
    {
      "epoch": 0.10041379310344828,
      "grad_norm": 7.296253204345703,
      "learning_rate": 9.11278195488722e-06,
      "loss": 0.43705883622169495,
      "num_input_tokens_seen": 824904,
      "step": 1092,
      "train_runtime": 2492.3035,
      "train_tokens_per_second": 330.981
    },
    {
      "epoch": 0.10050574712643678,
      "grad_norm": 6.83408784866333,
      "learning_rate": 9.102756892230578e-06,
      "loss": 0.44784992933273315,
      "num_input_tokens_seen": 825472,
      "step": 1093,
      "train_runtime": 2494.2832,
      "train_tokens_per_second": 330.946
    },
    {
      "epoch": 0.10059770114942529,
      "grad_norm": 5.716932773590088,
      "learning_rate": 9.092731829573935e-06,
      "loss": 0.4516908526420593,
      "num_input_tokens_seen": 826596,
      "step": 1094,
      "train_runtime": 2496.2588,
      "train_tokens_per_second": 331.134
    },
    {
      "epoch": 0.1006896551724138,
      "grad_norm": 6.829396724700928,
      "learning_rate": 9.082706766917295e-06,
      "loss": 0.4212993085384369,
      "num_input_tokens_seen": 827232,
      "step": 1095,
      "train_runtime": 2498.1779,
      "train_tokens_per_second": 331.134
    },
    {
      "epoch": 0.1007816091954023,
      "grad_norm": 7.256698131561279,
      "learning_rate": 9.072681704260652e-06,
      "loss": 0.4646654427051544,
      "num_input_tokens_seen": 827912,
      "step": 1096,
      "train_runtime": 2500.1032,
      "train_tokens_per_second": 331.151
    },
    {
      "epoch": 0.1008735632183908,
      "grad_norm": 6.697038650512695,
      "learning_rate": 9.06265664160401e-06,
      "loss": 0.5193271040916443,
      "num_input_tokens_seen": 828964,
      "step": 1097,
      "train_runtime": 2502.0918,
      "train_tokens_per_second": 331.308
    },
    {
      "epoch": 0.10096551724137931,
      "grad_norm": 6.856689453125,
      "learning_rate": 9.05263157894737e-06,
      "loss": 0.5617572069168091,
      "num_input_tokens_seen": 829588,
      "step": 1098,
      "train_runtime": 2504.0147,
      "train_tokens_per_second": 331.303
    },
    {
      "epoch": 0.10105747126436782,
      "grad_norm": 6.135411739349365,
      "learning_rate": 9.042606516290727e-06,
      "loss": 0.3666629493236542,
      "num_input_tokens_seen": 830120,
      "step": 1099,
      "train_runtime": 2505.932,
      "train_tokens_per_second": 331.262
    },
    {
      "epoch": 0.10114942528735632,
      "grad_norm": 7.073989391326904,
      "learning_rate": 9.032581453634086e-06,
      "loss": 0.4835805296897888,
      "num_input_tokens_seen": 830876,
      "step": 1100,
      "train_runtime": 2507.882,
      "train_tokens_per_second": 331.306
    },
    {
      "epoch": 0.10124137931034483,
      "grad_norm": 6.260268211364746,
      "learning_rate": 9.022556390977444e-06,
      "loss": 0.5108278393745422,
      "num_input_tokens_seen": 831800,
      "step": 1101,
      "train_runtime": 2509.8337,
      "train_tokens_per_second": 331.416
    },
    {
      "epoch": 0.10133333333333333,
      "grad_norm": 6.927764892578125,
      "learning_rate": 9.012531328320803e-06,
      "loss": 0.43247172236442566,
      "num_input_tokens_seen": 832444,
      "step": 1102,
      "train_runtime": 2511.7489,
      "train_tokens_per_second": 331.42
    },
    {
      "epoch": 0.10142528735632184,
      "grad_norm": 7.856224536895752,
      "learning_rate": 9.002506265664161e-06,
      "loss": 0.45604825019836426,
      "num_input_tokens_seen": 833036,
      "step": 1103,
      "train_runtime": 2513.6687,
      "train_tokens_per_second": 331.402
    },
    {
      "epoch": 0.10151724137931034,
      "grad_norm": 5.877823352813721,
      "learning_rate": 8.99248120300752e-06,
      "loss": 0.40084534883499146,
      "num_input_tokens_seen": 833608,
      "step": 1104,
      "train_runtime": 2515.582,
      "train_tokens_per_second": 331.378
    },
    {
      "epoch": 0.10160919540229885,
      "grad_norm": 7.92399263381958,
      "learning_rate": 8.982456140350878e-06,
      "loss": 0.6011634469032288,
      "num_input_tokens_seen": 834436,
      "step": 1105,
      "train_runtime": 2517.5498,
      "train_tokens_per_second": 331.448
    },
    {
      "epoch": 0.10170114942528735,
      "grad_norm": 6.29843282699585,
      "learning_rate": 8.972431077694236e-06,
      "loss": 0.48987555503845215,
      "num_input_tokens_seen": 835324,
      "step": 1106,
      "train_runtime": 2519.4913,
      "train_tokens_per_second": 331.545
    },
    {
      "epoch": 0.10179310344827586,
      "grad_norm": 6.689895153045654,
      "learning_rate": 8.962406015037595e-06,
      "loss": 0.40902847051620483,
      "num_input_tokens_seen": 835936,
      "step": 1107,
      "train_runtime": 2521.4095,
      "train_tokens_per_second": 331.535
    },
    {
      "epoch": 0.10188505747126436,
      "grad_norm": 6.9143147468566895,
      "learning_rate": 8.952380952380953e-06,
      "loss": 0.5424303412437439,
      "num_input_tokens_seen": 836648,
      "step": 1108,
      "train_runtime": 2523.3206,
      "train_tokens_per_second": 331.566
    },
    {
      "epoch": 0.10197701149425287,
      "grad_norm": 7.003936290740967,
      "learning_rate": 8.942355889724312e-06,
      "loss": 0.44123268127441406,
      "num_input_tokens_seen": 837352,
      "step": 1109,
      "train_runtime": 2525.2621,
      "train_tokens_per_second": 331.59
    },
    {
      "epoch": 0.10206896551724139,
      "grad_norm": 6.4153008460998535,
      "learning_rate": 8.93233082706767e-06,
      "loss": 0.4268386960029602,
      "num_input_tokens_seen": 838628,
      "step": 1110,
      "train_runtime": 2527.2351,
      "train_tokens_per_second": 331.836
    },
    {
      "epoch": 0.10216091954022989,
      "grad_norm": 5.765989780426025,
      "learning_rate": 8.922305764411027e-06,
      "loss": 0.5652371644973755,
      "num_input_tokens_seen": 840140,
      "step": 1111,
      "train_runtime": 2536.2102,
      "train_tokens_per_second": 331.258
    },
    {
      "epoch": 0.1022528735632184,
      "grad_norm": 6.961856842041016,
      "learning_rate": 8.912280701754387e-06,
      "loss": 0.3952780067920685,
      "num_input_tokens_seen": 840828,
      "step": 1112,
      "train_runtime": 2538.1877,
      "train_tokens_per_second": 331.271
    },
    {
      "epoch": 0.1023448275862069,
      "grad_norm": 7.673656463623047,
      "learning_rate": 8.902255639097746e-06,
      "loss": 0.5512932538986206,
      "num_input_tokens_seen": 841480,
      "step": 1113,
      "train_runtime": 2540.1737,
      "train_tokens_per_second": 331.269
    },
    {
      "epoch": 0.10243678160919541,
      "grad_norm": 11.50959300994873,
      "learning_rate": 8.892230576441103e-06,
      "loss": 0.423117458820343,
      "num_input_tokens_seen": 842056,
      "step": 1114,
      "train_runtime": 2542.1072,
      "train_tokens_per_second": 331.243
    },
    {
      "epoch": 0.10252873563218391,
      "grad_norm": 6.463172912597656,
      "learning_rate": 8.882205513784463e-06,
      "loss": 0.4713388681411743,
      "num_input_tokens_seen": 842692,
      "step": 1115,
      "train_runtime": 2544.0231,
      "train_tokens_per_second": 331.244
    },
    {
      "epoch": 0.10262068965517242,
      "grad_norm": 6.369577407836914,
      "learning_rate": 8.87218045112782e-06,
      "loss": 0.4382193088531494,
      "num_input_tokens_seen": 843528,
      "step": 1116,
      "train_runtime": 2545.9715,
      "train_tokens_per_second": 331.319
    },
    {
      "epoch": 0.10271264367816092,
      "grad_norm": 6.405300617218018,
      "learning_rate": 8.862155388471178e-06,
      "loss": 0.4564143419265747,
      "num_input_tokens_seen": 844264,
      "step": 1117,
      "train_runtime": 2547.9274,
      "train_tokens_per_second": 331.353
    },
    {
      "epoch": 0.10280459770114943,
      "grad_norm": 6.811070442199707,
      "learning_rate": 8.852130325814537e-06,
      "loss": 0.602678120136261,
      "num_input_tokens_seen": 845160,
      "step": 1118,
      "train_runtime": 2549.8913,
      "train_tokens_per_second": 331.449
    },
    {
      "epoch": 0.10289655172413793,
      "grad_norm": 6.943862438201904,
      "learning_rate": 8.842105263157895e-06,
      "loss": 0.421866238117218,
      "num_input_tokens_seen": 845820,
      "step": 1119,
      "train_runtime": 2551.8374,
      "train_tokens_per_second": 331.455
    },
    {
      "epoch": 0.10298850574712644,
      "grad_norm": 6.22981071472168,
      "learning_rate": 8.832080200501254e-06,
      "loss": 0.4855208992958069,
      "num_input_tokens_seen": 846392,
      "step": 1120,
      "train_runtime": 2553.7755,
      "train_tokens_per_second": 331.428
    },
    {
      "epoch": 0.10308045977011494,
      "grad_norm": 9.345059394836426,
      "learning_rate": 8.822055137844612e-06,
      "loss": 0.5808264017105103,
      "num_input_tokens_seen": 848220,
      "step": 1121,
      "train_runtime": 2555.8766,
      "train_tokens_per_second": 331.87
    },
    {
      "epoch": 0.10317241379310345,
      "grad_norm": 6.821878433227539,
      "learning_rate": 8.81203007518797e-06,
      "loss": 0.4558633267879486,
      "num_input_tokens_seen": 848888,
      "step": 1122,
      "train_runtime": 2557.8287,
      "train_tokens_per_second": 331.878
    },
    {
      "epoch": 0.10326436781609195,
      "grad_norm": 7.337451457977295,
      "learning_rate": 8.802005012531329e-06,
      "loss": 0.49908947944641113,
      "num_input_tokens_seen": 849748,
      "step": 1123,
      "train_runtime": 2559.7814,
      "train_tokens_per_second": 331.961
    },
    {
      "epoch": 0.10335632183908046,
      "grad_norm": 6.414590835571289,
      "learning_rate": 8.791979949874688e-06,
      "loss": 0.45316675305366516,
      "num_input_tokens_seen": 850480,
      "step": 1124,
      "train_runtime": 2561.6977,
      "train_tokens_per_second": 331.999
    },
    {
      "epoch": 0.10344827586206896,
      "grad_norm": 5.787412166595459,
      "learning_rate": 8.781954887218046e-06,
      "loss": 0.37481051683425903,
      "num_input_tokens_seen": 851084,
      "step": 1125,
      "train_runtime": 2563.6018,
      "train_tokens_per_second": 331.988
    },
    {
      "epoch": 0.10354022988505747,
      "grad_norm": 6.6846842765808105,
      "learning_rate": 8.771929824561405e-06,
      "loss": 0.44032424688339233,
      "num_input_tokens_seen": 851752,
      "step": 1126,
      "train_runtime": 2565.5326,
      "train_tokens_per_second": 331.998
    },
    {
      "epoch": 0.10363218390804597,
      "grad_norm": 7.464715957641602,
      "learning_rate": 8.761904761904763e-06,
      "loss": 0.40682342648506165,
      "num_input_tokens_seen": 852344,
      "step": 1127,
      "train_runtime": 2567.4486,
      "train_tokens_per_second": 331.981
    },
    {
      "epoch": 0.10372413793103448,
      "grad_norm": 6.334898948669434,
      "learning_rate": 8.751879699248122e-06,
      "loss": 0.47571703791618347,
      "num_input_tokens_seen": 853560,
      "step": 1128,
      "train_runtime": 2569.4544,
      "train_tokens_per_second": 332.195
    },
    {
      "epoch": 0.10381609195402298,
      "grad_norm": 6.915487289428711,
      "learning_rate": 8.741854636591478e-06,
      "loss": 0.4608912169933319,
      "num_input_tokens_seen": 854812,
      "step": 1129,
      "train_runtime": 2571.4341,
      "train_tokens_per_second": 332.426
    },
    {
      "epoch": 0.10390804597701149,
      "grad_norm": 7.502654075622559,
      "learning_rate": 8.731829573934839e-06,
      "loss": 0.46037107706069946,
      "num_input_tokens_seen": 855540,
      "step": 1130,
      "train_runtime": 2573.4207,
      "train_tokens_per_second": 332.452
    },
    {
      "epoch": 0.104,
      "grad_norm": 6.771470546722412,
      "learning_rate": 8.721804511278195e-06,
      "loss": 0.5191470384597778,
      "num_input_tokens_seen": 856396,
      "step": 1131,
      "train_runtime": 2575.5216,
      "train_tokens_per_second": 332.514
    },
    {
      "epoch": 0.1040919540229885,
      "grad_norm": 6.378627777099609,
      "learning_rate": 8.711779448621554e-06,
      "loss": 0.4394569993019104,
      "num_input_tokens_seen": 857096,
      "step": 1132,
      "train_runtime": 2577.4414,
      "train_tokens_per_second": 332.538
    },
    {
      "epoch": 0.104183908045977,
      "grad_norm": 6.649139404296875,
      "learning_rate": 8.701754385964914e-06,
      "loss": 0.5085347890853882,
      "num_input_tokens_seen": 857760,
      "step": 1133,
      "train_runtime": 2579.3566,
      "train_tokens_per_second": 332.548
    },
    {
      "epoch": 0.10427586206896551,
      "grad_norm": 6.40639066696167,
      "learning_rate": 8.69172932330827e-06,
      "loss": 0.5946711301803589,
      "num_input_tokens_seen": 859220,
      "step": 1134,
      "train_runtime": 2581.3642,
      "train_tokens_per_second": 332.855
    },
    {
      "epoch": 0.10436781609195403,
      "grad_norm": 8.070779800415039,
      "learning_rate": 8.68170426065163e-06,
      "loss": 0.5779873132705688,
      "num_input_tokens_seen": 860148,
      "step": 1135,
      "train_runtime": 2583.3163,
      "train_tokens_per_second": 332.963
    },
    {
      "epoch": 0.10445977011494254,
      "grad_norm": 7.178983688354492,
      "learning_rate": 8.671679197994988e-06,
      "loss": 0.43063604831695557,
      "num_input_tokens_seen": 860680,
      "step": 1136,
      "train_runtime": 2585.2274,
      "train_tokens_per_second": 332.922
    },
    {
      "epoch": 0.10455172413793104,
      "grad_norm": 7.161444664001465,
      "learning_rate": 8.661654135338346e-06,
      "loss": 0.4682421088218689,
      "num_input_tokens_seen": 861344,
      "step": 1137,
      "train_runtime": 2587.1461,
      "train_tokens_per_second": 332.932
    },
    {
      "epoch": 0.10464367816091955,
      "grad_norm": 6.621373176574707,
      "learning_rate": 8.651629072681705e-06,
      "loss": 0.3778235614299774,
      "num_input_tokens_seen": 861904,
      "step": 1138,
      "train_runtime": 2589.0599,
      "train_tokens_per_second": 332.902
    },
    {
      "epoch": 0.10473563218390805,
      "grad_norm": 6.741207122802734,
      "learning_rate": 8.641604010025063e-06,
      "loss": 0.500791072845459,
      "num_input_tokens_seen": 862960,
      "step": 1139,
      "train_runtime": 2591.0595,
      "train_tokens_per_second": 333.053
    },
    {
      "epoch": 0.10482758620689656,
      "grad_norm": 7.26300048828125,
      "learning_rate": 8.631578947368422e-06,
      "loss": 0.5782602429389954,
      "num_input_tokens_seen": 863716,
      "step": 1140,
      "train_runtime": 2593.0137,
      "train_tokens_per_second": 333.094
    },
    {
      "epoch": 0.10491954022988506,
      "grad_norm": 7.038207530975342,
      "learning_rate": 8.62155388471178e-06,
      "loss": 0.46371889114379883,
      "num_input_tokens_seen": 864972,
      "step": 1141,
      "train_runtime": 2601.6965,
      "train_tokens_per_second": 332.465
    },
    {
      "epoch": 0.10501149425287357,
      "grad_norm": 6.5405192375183105,
      "learning_rate": 8.611528822055139e-06,
      "loss": 0.5100393891334534,
      "num_input_tokens_seen": 865836,
      "step": 1142,
      "train_runtime": 2603.6423,
      "train_tokens_per_second": 332.548
    },
    {
      "epoch": 0.10510344827586207,
      "grad_norm": 8.58717155456543,
      "learning_rate": 8.601503759398497e-06,
      "loss": 0.5306845307350159,
      "num_input_tokens_seen": 866456,
      "step": 1143,
      "train_runtime": 2605.6254,
      "train_tokens_per_second": 332.533
    },
    {
      "epoch": 0.10519540229885058,
      "grad_norm": 6.847975254058838,
      "learning_rate": 8.591478696741856e-06,
      "loss": 0.5252816081047058,
      "num_input_tokens_seen": 867268,
      "step": 1144,
      "train_runtime": 2607.5767,
      "train_tokens_per_second": 332.595
    },
    {
      "epoch": 0.10528735632183908,
      "grad_norm": 7.272277355194092,
      "learning_rate": 8.581453634085214e-06,
      "loss": 0.554922342300415,
      "num_input_tokens_seen": 868228,
      "step": 1145,
      "train_runtime": 2609.555,
      "train_tokens_per_second": 332.711
    },
    {
      "epoch": 0.10537931034482759,
      "grad_norm": 6.600028991699219,
      "learning_rate": 8.571428571428571e-06,
      "loss": 0.47623491287231445,
      "num_input_tokens_seen": 868812,
      "step": 1146,
      "train_runtime": 2611.4597,
      "train_tokens_per_second": 332.692
    },
    {
      "epoch": 0.10547126436781609,
      "grad_norm": 8.430718421936035,
      "learning_rate": 8.561403508771931e-06,
      "loss": 0.4833829998970032,
      "num_input_tokens_seen": 869404,
      "step": 1147,
      "train_runtime": 2613.4042,
      "train_tokens_per_second": 332.671
    },
    {
      "epoch": 0.1055632183908046,
      "grad_norm": 6.189053535461426,
      "learning_rate": 8.55137844611529e-06,
      "loss": 0.401858389377594,
      "num_input_tokens_seen": 870064,
      "step": 1148,
      "train_runtime": 2615.3139,
      "train_tokens_per_second": 332.681
    },
    {
      "epoch": 0.1056551724137931,
      "grad_norm": 8.391399383544922,
      "learning_rate": 8.541353383458646e-06,
      "loss": 0.534809410572052,
      "num_input_tokens_seen": 870740,
      "step": 1149,
      "train_runtime": 2617.2292,
      "train_tokens_per_second": 332.695
    },
    {
      "epoch": 0.10574712643678161,
      "grad_norm": 7.162993431091309,
      "learning_rate": 8.531328320802007e-06,
      "loss": 0.48639699816703796,
      "num_input_tokens_seen": 871544,
      "step": 1150,
      "train_runtime": 2619.1697,
      "train_tokens_per_second": 332.756
    },
    {
      "epoch": 0.10583908045977011,
      "grad_norm": 5.858705997467041,
      "learning_rate": 8.521303258145363e-06,
      "loss": 0.4374701976776123,
      "num_input_tokens_seen": 872508,
      "step": 1151,
      "train_runtime": 2621.1555,
      "train_tokens_per_second": 332.872
    },
    {
      "epoch": 0.10593103448275862,
      "grad_norm": 6.275646209716797,
      "learning_rate": 8.511278195488722e-06,
      "loss": 0.49831047654151917,
      "num_input_tokens_seen": 873292,
      "step": 1152,
      "train_runtime": 2623.0981,
      "train_tokens_per_second": 332.924
    },
    {
      "epoch": 0.10602298850574712,
      "grad_norm": 6.347123622894287,
      "learning_rate": 8.501253132832082e-06,
      "loss": 0.4552074074745178,
      "num_input_tokens_seen": 873944,
      "step": 1153,
      "train_runtime": 2625.0113,
      "train_tokens_per_second": 332.93
    },
    {
      "epoch": 0.10611494252873563,
      "grad_norm": 8.130861282348633,
      "learning_rate": 8.491228070175439e-06,
      "loss": 0.6418692469596863,
      "num_input_tokens_seen": 875456,
      "step": 1154,
      "train_runtime": 2627.0299,
      "train_tokens_per_second": 333.249
    },
    {
      "epoch": 0.10620689655172413,
      "grad_norm": 6.69249963760376,
      "learning_rate": 8.481203007518797e-06,
      "loss": 0.377086877822876,
      "num_input_tokens_seen": 876152,
      "step": 1155,
      "train_runtime": 2628.9819,
      "train_tokens_per_second": 333.267
    },
    {
      "epoch": 0.10629885057471264,
      "grad_norm": 5.914850234985352,
      "learning_rate": 8.471177944862156e-06,
      "loss": 0.4858471751213074,
      "num_input_tokens_seen": 877164,
      "step": 1156,
      "train_runtime": 2630.9646,
      "train_tokens_per_second": 333.4
    },
    {
      "epoch": 0.10639080459770114,
      "grad_norm": 6.8667168617248535,
      "learning_rate": 8.461152882205514e-06,
      "loss": 0.47280052304267883,
      "num_input_tokens_seen": 877992,
      "step": 1157,
      "train_runtime": 2632.9249,
      "train_tokens_per_second": 333.466
    },
    {
      "epoch": 0.10648275862068965,
      "grad_norm": 6.664647579193115,
      "learning_rate": 8.451127819548873e-06,
      "loss": 0.4369109869003296,
      "num_input_tokens_seen": 878632,
      "step": 1158,
      "train_runtime": 2634.8346,
      "train_tokens_per_second": 333.468
    },
    {
      "epoch": 0.10657471264367815,
      "grad_norm": 6.69909143447876,
      "learning_rate": 8.441102756892231e-06,
      "loss": 0.5258703231811523,
      "num_input_tokens_seen": 879296,
      "step": 1159,
      "train_runtime": 2636.7554,
      "train_tokens_per_second": 333.477
    },
    {
      "epoch": 0.10666666666666667,
      "grad_norm": 6.778830528259277,
      "learning_rate": 8.43107769423559e-06,
      "loss": 0.5211995244026184,
      "num_input_tokens_seen": 880040,
      "step": 1160,
      "train_runtime": 2638.674,
      "train_tokens_per_second": 333.516
    },
    {
      "epoch": 0.10675862068965518,
      "grad_norm": 7.172182559967041,
      "learning_rate": 8.421052631578948e-06,
      "loss": 0.49945908784866333,
      "num_input_tokens_seen": 880796,
      "step": 1161,
      "train_runtime": 2640.5989,
      "train_tokens_per_second": 333.559
    },
    {
      "epoch": 0.10685057471264368,
      "grad_norm": 6.542418479919434,
      "learning_rate": 8.411027568922307e-06,
      "loss": 0.5056060552597046,
      "num_input_tokens_seen": 881676,
      "step": 1162,
      "train_runtime": 2642.545,
      "train_tokens_per_second": 333.647
    },
    {
      "epoch": 0.10694252873563219,
      "grad_norm": 6.930642604827881,
      "learning_rate": 8.401002506265665e-06,
      "loss": 0.5065448880195618,
      "num_input_tokens_seen": 882676,
      "step": 1163,
      "train_runtime": 2644.5277,
      "train_tokens_per_second": 333.775
    },
    {
      "epoch": 0.1070344827586207,
      "grad_norm": 6.357349395751953,
      "learning_rate": 8.390977443609024e-06,
      "loss": 0.4251636266708374,
      "num_input_tokens_seen": 883556,
      "step": 1164,
      "train_runtime": 2646.4774,
      "train_tokens_per_second": 333.861
    },
    {
      "epoch": 0.1071264367816092,
      "grad_norm": 6.012309551239014,
      "learning_rate": 8.380952380952382e-06,
      "loss": 0.37506213784217834,
      "num_input_tokens_seen": 884100,
      "step": 1165,
      "train_runtime": 2648.4139,
      "train_tokens_per_second": 333.822
    },
    {
      "epoch": 0.1072183908045977,
      "grad_norm": 6.938154220581055,
      "learning_rate": 8.370927318295739e-06,
      "loss": 0.5116899609565735,
      "num_input_tokens_seen": 884892,
      "step": 1166,
      "train_runtime": 2650.356,
      "train_tokens_per_second": 333.877
    },
    {
      "epoch": 0.10731034482758621,
      "grad_norm": 5.789228916168213,
      "learning_rate": 8.3609022556391e-06,
      "loss": 0.47559186816215515,
      "num_input_tokens_seen": 886440,
      "step": 1167,
      "train_runtime": 2652.4012,
      "train_tokens_per_second": 334.203
    },
    {
      "epoch": 0.10740229885057472,
      "grad_norm": 7.105683326721191,
      "learning_rate": 8.350877192982458e-06,
      "loss": 0.4797236919403076,
      "num_input_tokens_seen": 887448,
      "step": 1168,
      "train_runtime": 2654.3786,
      "train_tokens_per_second": 334.334
    },
    {
      "epoch": 0.10749425287356322,
      "grad_norm": 6.87125825881958,
      "learning_rate": 8.340852130325814e-06,
      "loss": 0.4814605712890625,
      "num_input_tokens_seen": 888048,
      "step": 1169,
      "train_runtime": 2656.2971,
      "train_tokens_per_second": 334.318
    },
    {
      "epoch": 0.10758620689655173,
      "grad_norm": 6.2805986404418945,
      "learning_rate": 8.330827067669175e-06,
      "loss": 0.3974026143550873,
      "num_input_tokens_seen": 888680,
      "step": 1170,
      "train_runtime": 2658.2043,
      "train_tokens_per_second": 334.316
    },
    {
      "epoch": 0.10767816091954023,
      "grad_norm": 6.526704788208008,
      "learning_rate": 8.320802005012531e-06,
      "loss": 0.44288185238838196,
      "num_input_tokens_seen": 889284,
      "step": 1171,
      "train_runtime": 2667.0342,
      "train_tokens_per_second": 333.436
    },
    {
      "epoch": 0.10777011494252874,
      "grad_norm": 6.377283573150635,
      "learning_rate": 8.31077694235589e-06,
      "loss": 0.4232429265975952,
      "num_input_tokens_seen": 889836,
      "step": 1172,
      "train_runtime": 2668.9832,
      "train_tokens_per_second": 333.399
    },
    {
      "epoch": 0.10786206896551724,
      "grad_norm": 7.443594455718994,
      "learning_rate": 8.300751879699248e-06,
      "loss": 0.5917798280715942,
      "num_input_tokens_seen": 890616,
      "step": 1173,
      "train_runtime": 2670.9192,
      "train_tokens_per_second": 333.449
    },
    {
      "epoch": 0.10795402298850575,
      "grad_norm": 7.515342712402344,
      "learning_rate": 8.290726817042607e-06,
      "loss": 0.5746567845344543,
      "num_input_tokens_seen": 891592,
      "step": 1174,
      "train_runtime": 2672.8873,
      "train_tokens_per_second": 333.569
    },
    {
      "epoch": 0.10804597701149425,
      "grad_norm": 6.263985633850098,
      "learning_rate": 8.280701754385965e-06,
      "loss": 0.40477776527404785,
      "num_input_tokens_seen": 892196,
      "step": 1175,
      "train_runtime": 2674.7982,
      "train_tokens_per_second": 333.556
    },
    {
      "epoch": 0.10813793103448276,
      "grad_norm": 6.493422031402588,
      "learning_rate": 8.270676691729324e-06,
      "loss": 0.41471216082572937,
      "num_input_tokens_seen": 892932,
      "step": 1176,
      "train_runtime": 2676.731,
      "train_tokens_per_second": 333.59
    },
    {
      "epoch": 0.10822988505747126,
      "grad_norm": 10.124730110168457,
      "learning_rate": 8.260651629072682e-06,
      "loss": 0.5155556797981262,
      "num_input_tokens_seen": 893656,
      "step": 1177,
      "train_runtime": 2678.6435,
      "train_tokens_per_second": 333.623
    },
    {
      "epoch": 0.10832183908045977,
      "grad_norm": 6.713027477264404,
      "learning_rate": 8.25062656641604e-06,
      "loss": 0.4645010828971863,
      "num_input_tokens_seen": 894368,
      "step": 1178,
      "train_runtime": 2680.5852,
      "train_tokens_per_second": 333.647
    },
    {
      "epoch": 0.10841379310344827,
      "grad_norm": 7.475142002105713,
      "learning_rate": 8.2406015037594e-06,
      "loss": 0.3595820367336273,
      "num_input_tokens_seen": 894960,
      "step": 1179,
      "train_runtime": 2682.5029,
      "train_tokens_per_second": 333.629
    },
    {
      "epoch": 0.10850574712643678,
      "grad_norm": 8.378244400024414,
      "learning_rate": 8.230576441102758e-06,
      "loss": 0.4922295808792114,
      "num_input_tokens_seen": 895668,
      "step": 1180,
      "train_runtime": 2684.4189,
      "train_tokens_per_second": 333.654
    },
    {
      "epoch": 0.10859770114942528,
      "grad_norm": 7.011902332305908,
      "learning_rate": 8.220551378446116e-06,
      "loss": 0.4168221056461334,
      "num_input_tokens_seen": 896356,
      "step": 1181,
      "train_runtime": 2686.3328,
      "train_tokens_per_second": 333.673
    },
    {
      "epoch": 0.10868965517241379,
      "grad_norm": 7.045474052429199,
      "learning_rate": 8.210526315789475e-06,
      "loss": 0.5761998891830444,
      "num_input_tokens_seen": 897256,
      "step": 1182,
      "train_runtime": 2688.2752,
      "train_tokens_per_second": 333.766
    },
    {
      "epoch": 0.10878160919540229,
      "grad_norm": 7.988003730773926,
      "learning_rate": 8.200501253132833e-06,
      "loss": 0.53521728515625,
      "num_input_tokens_seen": 897932,
      "step": 1183,
      "train_runtime": 2690.2398,
      "train_tokens_per_second": 333.774
    },
    {
      "epoch": 0.1088735632183908,
      "grad_norm": 7.01152229309082,
      "learning_rate": 8.190476190476192e-06,
      "loss": 0.4423326551914215,
      "num_input_tokens_seen": 898776,
      "step": 1184,
      "train_runtime": 2692.2319,
      "train_tokens_per_second": 333.84
    },
    {
      "epoch": 0.10896551724137932,
      "grad_norm": 7.029073238372803,
      "learning_rate": 8.18045112781955e-06,
      "loss": 0.5665289163589478,
      "num_input_tokens_seen": 899572,
      "step": 1185,
      "train_runtime": 2694.2814,
      "train_tokens_per_second": 333.882
    },
    {
      "epoch": 0.10905747126436782,
      "grad_norm": 7.208654403686523,
      "learning_rate": 8.170426065162907e-06,
      "loss": 0.3946235775947571,
      "num_input_tokens_seen": 900160,
      "step": 1186,
      "train_runtime": 2696.1993,
      "train_tokens_per_second": 333.863
    },
    {
      "epoch": 0.10914942528735633,
      "grad_norm": 7.401321887969971,
      "learning_rate": 8.160401002506266e-06,
      "loss": 0.5328009724617004,
      "num_input_tokens_seen": 900812,
      "step": 1187,
      "train_runtime": 2698.1093,
      "train_tokens_per_second": 333.868
    },
    {
      "epoch": 0.10924137931034483,
      "grad_norm": 6.7505879402160645,
      "learning_rate": 8.150375939849626e-06,
      "loss": 0.5387961268424988,
      "num_input_tokens_seen": 901768,
      "step": 1188,
      "train_runtime": 2700.0659,
      "train_tokens_per_second": 333.98
    },
    {
      "epoch": 0.10933333333333334,
      "grad_norm": 6.317501544952393,
      "learning_rate": 8.140350877192983e-06,
      "loss": 0.5291674137115479,
      "num_input_tokens_seen": 902548,
      "step": 1189,
      "train_runtime": 2702.0193,
      "train_tokens_per_second": 334.027
    },
    {
      "epoch": 0.10942528735632184,
      "grad_norm": 7.332826614379883,
      "learning_rate": 8.130325814536341e-06,
      "loss": 0.509458601474762,
      "num_input_tokens_seen": 903644,
      "step": 1190,
      "train_runtime": 2704.0013,
      "train_tokens_per_second": 334.188
    },
    {
      "epoch": 0.10951724137931035,
      "grad_norm": 6.344303131103516,
      "learning_rate": 8.1203007518797e-06,
      "loss": 0.40950438380241394,
      "num_input_tokens_seen": 904380,
      "step": 1191,
      "train_runtime": 2705.9656,
      "train_tokens_per_second": 334.217
    },
    {
      "epoch": 0.10960919540229885,
      "grad_norm": 6.608860015869141,
      "learning_rate": 8.110275689223058e-06,
      "loss": 0.4933936893939972,
      "num_input_tokens_seen": 905072,
      "step": 1192,
      "train_runtime": 2707.9096,
      "train_tokens_per_second": 334.233
    },
    {
      "epoch": 0.10970114942528736,
      "grad_norm": 6.558511734008789,
      "learning_rate": 8.100250626566416e-06,
      "loss": 0.5003643035888672,
      "num_input_tokens_seen": 905848,
      "step": 1193,
      "train_runtime": 2709.869,
      "train_tokens_per_second": 334.277
    },
    {
      "epoch": 0.10979310344827586,
      "grad_norm": 6.844476699829102,
      "learning_rate": 8.090225563909775e-06,
      "loss": 0.5392162799835205,
      "num_input_tokens_seen": 906656,
      "step": 1194,
      "train_runtime": 2711.8417,
      "train_tokens_per_second": 334.332
    },
    {
      "epoch": 0.10988505747126437,
      "grad_norm": 6.837486267089844,
      "learning_rate": 8.080200501253133e-06,
      "loss": 0.46220025420188904,
      "num_input_tokens_seen": 907380,
      "step": 1195,
      "train_runtime": 2713.7757,
      "train_tokens_per_second": 334.361
    },
    {
      "epoch": 0.10997701149425287,
      "grad_norm": 6.648373603820801,
      "learning_rate": 8.070175438596492e-06,
      "loss": 0.5654021501541138,
      "num_input_tokens_seen": 908124,
      "step": 1196,
      "train_runtime": 2715.7471,
      "train_tokens_per_second": 334.392
    },
    {
      "epoch": 0.11006896551724138,
      "grad_norm": 6.451106548309326,
      "learning_rate": 8.06015037593985e-06,
      "loss": 0.4598565697669983,
      "num_input_tokens_seen": 908884,
      "step": 1197,
      "train_runtime": 2717.6835,
      "train_tokens_per_second": 334.433
    },
    {
      "epoch": 0.11016091954022988,
      "grad_norm": 6.6288957595825195,
      "learning_rate": 8.050125313283209e-06,
      "loss": 0.4231450855731964,
      "num_input_tokens_seen": 909656,
      "step": 1198,
      "train_runtime": 2719.5952,
      "train_tokens_per_second": 334.482
    },
    {
      "epoch": 0.11025287356321839,
      "grad_norm": 6.884613513946533,
      "learning_rate": 8.040100250626567e-06,
      "loss": 0.5025525689125061,
      "num_input_tokens_seen": 910616,
      "step": 1199,
      "train_runtime": 2721.5412,
      "train_tokens_per_second": 334.596
    },
    {
      "epoch": 0.1103448275862069,
      "grad_norm": 6.784577369689941,
      "learning_rate": 8.030075187969926e-06,
      "loss": 0.4666309356689453,
      "num_input_tokens_seen": 911164,
      "step": 1200,
      "train_runtime": 2723.4617,
      "train_tokens_per_second": 334.561
    },
    {
      "epoch": 0.1103448275862069,
      "eval_loss": 1.8325949907302856,
      "eval_runtime": 28.023,
      "eval_samples_per_second": 35.685,
      "eval_steps_per_second": 8.921,
      "num_input_tokens_seen": 911164,
      "step": 1200
    },
    {
      "epoch": 0.1104367816091954,
      "grad_norm": 7.389218807220459,
      "learning_rate": 8.020050125313283e-06,
      "loss": 0.4460408389568329,
      "num_input_tokens_seen": 911708,
      "step": 1201,
      "train_runtime": 2759.9673,
      "train_tokens_per_second": 330.333
    },
    {
      "epoch": 0.1105287356321839,
      "grad_norm": 6.948616981506348,
      "learning_rate": 8.010025062656643e-06,
      "loss": 0.4718795120716095,
      "num_input_tokens_seen": 912324,
      "step": 1202,
      "train_runtime": 2761.8909,
      "train_tokens_per_second": 330.326
    },
    {
      "epoch": 0.11062068965517241,
      "grad_norm": 6.183836936950684,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.37865975499153137,
      "num_input_tokens_seen": 913012,
      "step": 1203,
      "train_runtime": 2763.9892,
      "train_tokens_per_second": 330.324
    },
    {
      "epoch": 0.11071264367816092,
      "grad_norm": 6.418726921081543,
      "learning_rate": 7.989974937343358e-06,
      "loss": 0.43318432569503784,
      "num_input_tokens_seen": 913600,
      "step": 1204,
      "train_runtime": 2765.9107,
      "train_tokens_per_second": 330.307
    },
    {
      "epoch": 0.11080459770114942,
      "grad_norm": 6.6056742668151855,
      "learning_rate": 7.979949874686718e-06,
      "loss": 0.4426375925540924,
      "num_input_tokens_seen": 914228,
      "step": 1205,
      "train_runtime": 2767.8236,
      "train_tokens_per_second": 330.306
    },
    {
      "epoch": 0.11089655172413793,
      "grad_norm": 6.80130672454834,
      "learning_rate": 7.969924812030075e-06,
      "loss": 0.550324022769928,
      "num_input_tokens_seen": 914928,
      "step": 1206,
      "train_runtime": 2769.7346,
      "train_tokens_per_second": 330.331
    },
    {
      "epoch": 0.11098850574712643,
      "grad_norm": 6.935859680175781,
      "learning_rate": 7.959899749373434e-06,
      "loss": 0.5600872039794922,
      "num_input_tokens_seen": 915964,
      "step": 1207,
      "train_runtime": 2771.7074,
      "train_tokens_per_second": 330.469
    },
    {
      "epoch": 0.11108045977011494,
      "grad_norm": 6.1385087966918945,
      "learning_rate": 7.949874686716794e-06,
      "loss": 0.4273529052734375,
      "num_input_tokens_seen": 916872,
      "step": 1208,
      "train_runtime": 2773.6419,
      "train_tokens_per_second": 330.566
    },
    {
      "epoch": 0.11117241379310344,
      "grad_norm": 5.904097080230713,
      "learning_rate": 7.93984962406015e-06,
      "loss": 0.472692608833313,
      "num_input_tokens_seen": 917884,
      "step": 1209,
      "train_runtime": 2775.6237,
      "train_tokens_per_second": 330.695
    },
    {
      "epoch": 0.11126436781609196,
      "grad_norm": 7.022626876831055,
      "learning_rate": 7.929824561403509e-06,
      "loss": 0.4611586630344391,
      "num_input_tokens_seen": 918556,
      "step": 1210,
      "train_runtime": 2777.5339,
      "train_tokens_per_second": 330.709
    },
    {
      "epoch": 0.11135632183908047,
      "grad_norm": 6.110097408294678,
      "learning_rate": 7.919799498746868e-06,
      "loss": 0.45371317863464355,
      "num_input_tokens_seen": 919208,
      "step": 1211,
      "train_runtime": 2779.4778,
      "train_tokens_per_second": 330.712
    },
    {
      "epoch": 0.11144827586206897,
      "grad_norm": 6.828897476196289,
      "learning_rate": 7.909774436090226e-06,
      "loss": 0.44778716564178467,
      "num_input_tokens_seen": 919764,
      "step": 1212,
      "train_runtime": 2781.3898,
      "train_tokens_per_second": 330.685
    },
    {
      "epoch": 0.11154022988505748,
      "grad_norm": 7.322358131408691,
      "learning_rate": 7.899749373433585e-06,
      "loss": 0.4146677851676941,
      "num_input_tokens_seen": 920636,
      "step": 1213,
      "train_runtime": 2783.3356,
      "train_tokens_per_second": 330.767
    },
    {
      "epoch": 0.11163218390804598,
      "grad_norm": 7.033532619476318,
      "learning_rate": 7.889724310776943e-06,
      "loss": 0.5266438126564026,
      "num_input_tokens_seen": 921500,
      "step": 1214,
      "train_runtime": 2785.2893,
      "train_tokens_per_second": 330.845
    },
    {
      "epoch": 0.11172413793103449,
      "grad_norm": 6.854131698608398,
      "learning_rate": 7.879699248120301e-06,
      "loss": 0.38787639141082764,
      "num_input_tokens_seen": 922152,
      "step": 1215,
      "train_runtime": 2787.2151,
      "train_tokens_per_second": 330.851
    },
    {
      "epoch": 0.11181609195402299,
      "grad_norm": 5.913403034210205,
      "learning_rate": 7.86967418546366e-06,
      "loss": 0.4330481290817261,
      "num_input_tokens_seen": 922832,
      "step": 1216,
      "train_runtime": 2789.1347,
      "train_tokens_per_second": 330.867
    },
    {
      "epoch": 0.1119080459770115,
      "grad_norm": 6.958983421325684,
      "learning_rate": 7.859649122807018e-06,
      "loss": 0.5390165448188782,
      "num_input_tokens_seen": 924340,
      "step": 1217,
      "train_runtime": 2791.1779,
      "train_tokens_per_second": 331.165
    },
    {
      "epoch": 0.112,
      "grad_norm": 6.044406890869141,
      "learning_rate": 7.849624060150377e-06,
      "loss": 0.3765028715133667,
      "num_input_tokens_seen": 925084,
      "step": 1218,
      "train_runtime": 2793.1356,
      "train_tokens_per_second": 331.199
    },
    {
      "epoch": 0.11209195402298851,
      "grad_norm": 6.692909240722656,
      "learning_rate": 7.839598997493735e-06,
      "loss": 0.4812348484992981,
      "num_input_tokens_seen": 925720,
      "step": 1219,
      "train_runtime": 2795.0454,
      "train_tokens_per_second": 331.2
    },
    {
      "epoch": 0.11218390804597701,
      "grad_norm": 5.939342498779297,
      "learning_rate": 7.829573934837094e-06,
      "loss": 0.3884960412979126,
      "num_input_tokens_seen": 926544,
      "step": 1220,
      "train_runtime": 2796.9973,
      "train_tokens_per_second": 331.264
    },
    {
      "epoch": 0.11227586206896552,
      "grad_norm": 5.711117267608643,
      "learning_rate": 7.81954887218045e-06,
      "loss": 0.38959741592407227,
      "num_input_tokens_seen": 927772,
      "step": 1221,
      "train_runtime": 2799.0019,
      "train_tokens_per_second": 331.465
    },
    {
      "epoch": 0.11236781609195402,
      "grad_norm": 6.349226951599121,
      "learning_rate": 7.809523809523811e-06,
      "loss": 0.4741501212120056,
      "num_input_tokens_seen": 928452,
      "step": 1222,
      "train_runtime": 2800.9274,
      "train_tokens_per_second": 331.48
    },
    {
      "epoch": 0.11245977011494253,
      "grad_norm": 6.910517692565918,
      "learning_rate": 7.79949874686717e-06,
      "loss": 0.46652236580848694,
      "num_input_tokens_seen": 929504,
      "step": 1223,
      "train_runtime": 2802.9518,
      "train_tokens_per_second": 331.616
    },
    {
      "epoch": 0.11255172413793103,
      "grad_norm": 6.269015789031982,
      "learning_rate": 7.789473684210526e-06,
      "loss": 0.4816398620605469,
      "num_input_tokens_seen": 930964,
      "step": 1224,
      "train_runtime": 2804.9737,
      "train_tokens_per_second": 331.898
    },
    {
      "epoch": 0.11264367816091954,
      "grad_norm": 6.046663284301758,
      "learning_rate": 7.779448621553886e-06,
      "loss": 0.42682701349258423,
      "num_input_tokens_seen": 931808,
      "step": 1225,
      "train_runtime": 2806.9188,
      "train_tokens_per_second": 331.968
    },
    {
      "epoch": 0.11273563218390804,
      "grad_norm": 8.843550682067871,
      "learning_rate": 7.769423558897243e-06,
      "loss": 0.4442460238933563,
      "num_input_tokens_seen": 932388,
      "step": 1226,
      "train_runtime": 2808.826,
      "train_tokens_per_second": 331.949
    },
    {
      "epoch": 0.11282758620689655,
      "grad_norm": 7.7472357749938965,
      "learning_rate": 7.759398496240602e-06,
      "loss": 0.5236173868179321,
      "num_input_tokens_seen": 933988,
      "step": 1227,
      "train_runtime": 2810.8848,
      "train_tokens_per_second": 332.275
    },
    {
      "epoch": 0.11291954022988505,
      "grad_norm": 6.5638508796691895,
      "learning_rate": 7.749373433583962e-06,
      "loss": 0.45537734031677246,
      "num_input_tokens_seen": 934712,
      "step": 1228,
      "train_runtime": 2812.798,
      "train_tokens_per_second": 332.307
    },
    {
      "epoch": 0.11301149425287356,
      "grad_norm": 6.816479682922363,
      "learning_rate": 7.739348370927319e-06,
      "loss": 0.572117269039154,
      "num_input_tokens_seen": 935404,
      "step": 1229,
      "train_runtime": 2814.708,
      "train_tokens_per_second": 332.327
    },
    {
      "epoch": 0.11310344827586206,
      "grad_norm": 6.51023530960083,
      "learning_rate": 7.729323308270677e-06,
      "loss": 0.4204486906528473,
      "num_input_tokens_seen": 936116,
      "step": 1230,
      "train_runtime": 2816.6454,
      "train_tokens_per_second": 332.351
    },
    {
      "epoch": 0.11319540229885057,
      "grad_norm": 6.704290866851807,
      "learning_rate": 7.719298245614036e-06,
      "loss": 0.4270913600921631,
      "num_input_tokens_seen": 936816,
      "step": 1231,
      "train_runtime": 2825.1471,
      "train_tokens_per_second": 331.599
    },
    {
      "epoch": 0.11328735632183908,
      "grad_norm": 6.436280250549316,
      "learning_rate": 7.709273182957394e-06,
      "loss": 0.41451695561408997,
      "num_input_tokens_seen": 938312,
      "step": 1232,
      "train_runtime": 2827.191,
      "train_tokens_per_second": 331.888
    },
    {
      "epoch": 0.11337931034482758,
      "grad_norm": 5.83052921295166,
      "learning_rate": 7.699248120300753e-06,
      "loss": 0.38611364364624023,
      "num_input_tokens_seen": 938952,
      "step": 1233,
      "train_runtime": 2829.2456,
      "train_tokens_per_second": 331.874
    },
    {
      "epoch": 0.11347126436781609,
      "grad_norm": 6.24282169342041,
      "learning_rate": 7.689223057644111e-06,
      "loss": 0.36962324380874634,
      "num_input_tokens_seen": 939560,
      "step": 1234,
      "train_runtime": 2831.1614,
      "train_tokens_per_second": 331.864
    },
    {
      "epoch": 0.1135632183908046,
      "grad_norm": 5.778984069824219,
      "learning_rate": 7.67919799498747e-06,
      "loss": 0.3523557782173157,
      "num_input_tokens_seen": 940188,
      "step": 1235,
      "train_runtime": 2833.0813,
      "train_tokens_per_second": 331.861
    },
    {
      "epoch": 0.11365517241379311,
      "grad_norm": 7.877459526062012,
      "learning_rate": 7.669172932330828e-06,
      "loss": 0.6218960881233215,
      "num_input_tokens_seen": 940956,
      "step": 1236,
      "train_runtime": 2835.0017,
      "train_tokens_per_second": 331.907
    },
    {
      "epoch": 0.11374712643678161,
      "grad_norm": 6.774043560028076,
      "learning_rate": 7.659147869674187e-06,
      "loss": 0.42774122953414917,
      "num_input_tokens_seen": 941708,
      "step": 1237,
      "train_runtime": 2836.9211,
      "train_tokens_per_second": 331.947
    },
    {
      "epoch": 0.11383908045977012,
      "grad_norm": 6.163104057312012,
      "learning_rate": 7.649122807017545e-06,
      "loss": 0.42159563302993774,
      "num_input_tokens_seen": 942544,
      "step": 1238,
      "train_runtime": 2838.8814,
      "train_tokens_per_second": 332.012
    },
    {
      "epoch": 0.11393103448275863,
      "grad_norm": 7.027411937713623,
      "learning_rate": 7.639097744360904e-06,
      "loss": 0.4619765281677246,
      "num_input_tokens_seen": 943472,
      "step": 1239,
      "train_runtime": 2840.8528,
      "train_tokens_per_second": 332.109
    },
    {
      "epoch": 0.11402298850574713,
      "grad_norm": 6.717901229858398,
      "learning_rate": 7.629072681704261e-06,
      "loss": 0.4811328053474426,
      "num_input_tokens_seen": 944436,
      "step": 1240,
      "train_runtime": 2842.8124,
      "train_tokens_per_second": 332.219
    },
    {
      "epoch": 0.11411494252873564,
      "grad_norm": 7.712195873260498,
      "learning_rate": 7.61904761904762e-06,
      "loss": 0.4902462065219879,
      "num_input_tokens_seen": 945216,
      "step": 1241,
      "train_runtime": 2844.7583,
      "train_tokens_per_second": 332.266
    },
    {
      "epoch": 0.11420689655172414,
      "grad_norm": 6.58721399307251,
      "learning_rate": 7.609022556390978e-06,
      "loss": 0.41334906220436096,
      "num_input_tokens_seen": 945764,
      "step": 1242,
      "train_runtime": 2846.6615,
      "train_tokens_per_second": 332.236
    },
    {
      "epoch": 0.11429885057471265,
      "grad_norm": 5.698861122131348,
      "learning_rate": 7.598997493734337e-06,
      "loss": 0.3368165194988251,
      "num_input_tokens_seen": 946316,
      "step": 1243,
      "train_runtime": 2848.564,
      "train_tokens_per_second": 332.208
    },
    {
      "epoch": 0.11439080459770115,
      "grad_norm": 7.710287094116211,
      "learning_rate": 7.588972431077694e-06,
      "loss": 0.5103346705436707,
      "num_input_tokens_seen": 947040,
      "step": 1244,
      "train_runtime": 2850.4722,
      "train_tokens_per_second": 332.24
    },
    {
      "epoch": 0.11448275862068966,
      "grad_norm": 6.46506929397583,
      "learning_rate": 7.578947368421054e-06,
      "loss": 0.3795013129711151,
      "num_input_tokens_seen": 947764,
      "step": 1245,
      "train_runtime": 2852.3874,
      "train_tokens_per_second": 332.27
    },
    {
      "epoch": 0.11457471264367816,
      "grad_norm": 5.176570415496826,
      "learning_rate": 7.568922305764411e-06,
      "loss": 0.3259194791316986,
      "num_input_tokens_seen": 948872,
      "step": 1246,
      "train_runtime": 2854.3658,
      "train_tokens_per_second": 332.428
    },
    {
      "epoch": 0.11466666666666667,
      "grad_norm": 6.8006391525268555,
      "learning_rate": 7.55889724310777e-06,
      "loss": 0.487837016582489,
      "num_input_tokens_seen": 949640,
      "step": 1247,
      "train_runtime": 2856.3058,
      "train_tokens_per_second": 332.471
    },
    {
      "epoch": 0.11475862068965517,
      "grad_norm": 7.8169379234313965,
      "learning_rate": 7.548872180451129e-06,
      "loss": 0.47706305980682373,
      "num_input_tokens_seen": 950192,
      "step": 1248,
      "train_runtime": 2858.2077,
      "train_tokens_per_second": 332.443
    },
    {
      "epoch": 0.11485057471264368,
      "grad_norm": 7.041769027709961,
      "learning_rate": 7.538847117794487e-06,
      "loss": 0.5111751556396484,
      "num_input_tokens_seen": 952244,
      "step": 1249,
      "train_runtime": 2860.3189,
      "train_tokens_per_second": 332.915
    },
    {
      "epoch": 0.11494252873563218,
      "grad_norm": 8.484130859375,
      "learning_rate": 7.528822055137845e-06,
      "loss": 0.3944612145423889,
      "num_input_tokens_seen": 952840,
      "step": 1250,
      "train_runtime": 2862.29,
      "train_tokens_per_second": 332.894
    },
    {
      "epoch": 0.11503448275862069,
      "grad_norm": 7.208199977874756,
      "learning_rate": 7.518796992481203e-06,
      "loss": 0.453296035528183,
      "num_input_tokens_seen": 953632,
      "step": 1251,
      "train_runtime": 2864.3212,
      "train_tokens_per_second": 332.935
    },
    {
      "epoch": 0.11512643678160919,
      "grad_norm": 5.981998920440674,
      "learning_rate": 7.508771929824562e-06,
      "loss": 0.4200296700000763,
      "num_input_tokens_seen": 954248,
      "step": 1252,
      "train_runtime": 2866.3438,
      "train_tokens_per_second": 332.915
    },
    {
      "epoch": 0.1152183908045977,
      "grad_norm": 6.765686988830566,
      "learning_rate": 7.498746867167921e-06,
      "loss": 0.63531893491745,
      "num_input_tokens_seen": 955628,
      "step": 1253,
      "train_runtime": 2868.3814,
      "train_tokens_per_second": 333.159
    },
    {
      "epoch": 0.1153103448275862,
      "grad_norm": 6.0461578369140625,
      "learning_rate": 7.488721804511278e-06,
      "loss": 0.37273716926574707,
      "num_input_tokens_seen": 956296,
      "step": 1254,
      "train_runtime": 2870.2817,
      "train_tokens_per_second": 333.171
    },
    {
      "epoch": 0.11540229885057471,
      "grad_norm": 6.15737247467041,
      "learning_rate": 7.478696741854638e-06,
      "loss": 0.4166964590549469,
      "num_input_tokens_seen": 956992,
      "step": 1255,
      "train_runtime": 2872.2439,
      "train_tokens_per_second": 333.186
    },
    {
      "epoch": 0.11549425287356321,
      "grad_norm": 7.3590288162231445,
      "learning_rate": 7.468671679197995e-06,
      "loss": 0.48610079288482666,
      "num_input_tokens_seen": 957616,
      "step": 1256,
      "train_runtime": 2874.151,
      "train_tokens_per_second": 333.182
    },
    {
      "epoch": 0.11558620689655172,
      "grad_norm": 7.055130958557129,
      "learning_rate": 7.458646616541354e-06,
      "loss": 0.4384363889694214,
      "num_input_tokens_seen": 958360,
      "step": 1257,
      "train_runtime": 2876.094,
      "train_tokens_per_second": 333.216
    },
    {
      "epoch": 0.11567816091954022,
      "grad_norm": 6.699975967407227,
      "learning_rate": 7.448621553884713e-06,
      "loss": 0.5209221243858337,
      "num_input_tokens_seen": 958952,
      "step": 1258,
      "train_runtime": 2878.0156,
      "train_tokens_per_second": 333.199
    },
    {
      "epoch": 0.11577011494252873,
      "grad_norm": 6.583015441894531,
      "learning_rate": 7.438596491228071e-06,
      "loss": 0.49776387214660645,
      "num_input_tokens_seen": 959580,
      "step": 1259,
      "train_runtime": 2879.9265,
      "train_tokens_per_second": 333.196
    },
    {
      "epoch": 0.11586206896551725,
      "grad_norm": 7.058274745941162,
      "learning_rate": 7.428571428571429e-06,
      "loss": 0.5116837620735168,
      "num_input_tokens_seen": 960348,
      "step": 1260,
      "train_runtime": 2881.8483,
      "train_tokens_per_second": 333.24
    },
    {
      "epoch": 0.11595402298850575,
      "grad_norm": 8.371155738830566,
      "learning_rate": 7.418546365914787e-06,
      "loss": 0.6046146154403687,
      "num_input_tokens_seen": 961084,
      "step": 1261,
      "train_runtime": 2890.2659,
      "train_tokens_per_second": 332.524
    },
    {
      "epoch": 0.11604597701149426,
      "grad_norm": 6.7046427726745605,
      "learning_rate": 7.408521303258146e-06,
      "loss": 0.43961718678474426,
      "num_input_tokens_seen": 961840,
      "step": 1262,
      "train_runtime": 2892.1769,
      "train_tokens_per_second": 332.566
    },
    {
      "epoch": 0.11613793103448276,
      "grad_norm": 6.21190071105957,
      "learning_rate": 7.398496240601505e-06,
      "loss": 0.45463913679122925,
      "num_input_tokens_seen": 962788,
      "step": 1263,
      "train_runtime": 2894.2452,
      "train_tokens_per_second": 332.656
    },
    {
      "epoch": 0.11622988505747127,
      "grad_norm": 7.1875152587890625,
      "learning_rate": 7.388471177944862e-06,
      "loss": 0.5096670389175415,
      "num_input_tokens_seen": 963580,
      "step": 1264,
      "train_runtime": 2896.1827,
      "train_tokens_per_second": 332.707
    },
    {
      "epoch": 0.11632183908045977,
      "grad_norm": 6.639959812164307,
      "learning_rate": 7.378446115288222e-06,
      "loss": 0.49722224473953247,
      "num_input_tokens_seen": 964224,
      "step": 1265,
      "train_runtime": 2898.0953,
      "train_tokens_per_second": 332.71
    },
    {
      "epoch": 0.11641379310344828,
      "grad_norm": 6.688169002532959,
      "learning_rate": 7.368421052631579e-06,
      "loss": 0.45049408078193665,
      "num_input_tokens_seen": 964752,
      "step": 1266,
      "train_runtime": 2900.0519,
      "train_tokens_per_second": 332.667
    },
    {
      "epoch": 0.11650574712643678,
      "grad_norm": 5.245498180389404,
      "learning_rate": 7.358395989974938e-06,
      "loss": 0.35358959436416626,
      "num_input_tokens_seen": 965900,
      "step": 1267,
      "train_runtime": 2902.0334,
      "train_tokens_per_second": 332.836
    },
    {
      "epoch": 0.11659770114942529,
      "grad_norm": 6.420312404632568,
      "learning_rate": 7.348370927318297e-06,
      "loss": 0.3846766948699951,
      "num_input_tokens_seen": 966436,
      "step": 1268,
      "train_runtime": 2903.9291,
      "train_tokens_per_second": 332.803
    },
    {
      "epoch": 0.1166896551724138,
      "grad_norm": 5.828828811645508,
      "learning_rate": 7.338345864661655e-06,
      "loss": 0.39027321338653564,
      "num_input_tokens_seen": 967100,
      "step": 1269,
      "train_runtime": 2905.8341,
      "train_tokens_per_second": 332.813
    },
    {
      "epoch": 0.1167816091954023,
      "grad_norm": 7.703383445739746,
      "learning_rate": 7.328320802005013e-06,
      "loss": 0.5670377016067505,
      "num_input_tokens_seen": 968004,
      "step": 1270,
      "train_runtime": 2907.7779,
      "train_tokens_per_second": 332.902
    },
    {
      "epoch": 0.1168735632183908,
      "grad_norm": 7.101981163024902,
      "learning_rate": 7.318295739348371e-06,
      "loss": 0.5447335243225098,
      "num_input_tokens_seen": 968724,
      "step": 1271,
      "train_runtime": 2909.743,
      "train_tokens_per_second": 332.924
    },
    {
      "epoch": 0.11696551724137931,
      "grad_norm": 7.2557172775268555,
      "learning_rate": 7.30827067669173e-06,
      "loss": 0.4155055582523346,
      "num_input_tokens_seen": 969320,
      "step": 1272,
      "train_runtime": 2911.651,
      "train_tokens_per_second": 332.911
    },
    {
      "epoch": 0.11705747126436782,
      "grad_norm": 6.135560989379883,
      "learning_rate": 7.298245614035089e-06,
      "loss": 0.39388903975486755,
      "num_input_tokens_seen": 970140,
      "step": 1273,
      "train_runtime": 2913.5828,
      "train_tokens_per_second": 332.971
    },
    {
      "epoch": 0.11714942528735632,
      "grad_norm": 6.169620037078857,
      "learning_rate": 7.288220551378446e-06,
      "loss": 0.41742026805877686,
      "num_input_tokens_seen": 970768,
      "step": 1274,
      "train_runtime": 2915.4927,
      "train_tokens_per_second": 332.969
    },
    {
      "epoch": 0.11724137931034483,
      "grad_norm": 6.049920558929443,
      "learning_rate": 7.278195488721806e-06,
      "loss": 0.4158029854297638,
      "num_input_tokens_seen": 971812,
      "step": 1275,
      "train_runtime": 2917.4721,
      "train_tokens_per_second": 333.101
    },
    {
      "epoch": 0.11733333333333333,
      "grad_norm": 6.8199944496154785,
      "learning_rate": 7.268170426065163e-06,
      "loss": 0.4261828660964966,
      "num_input_tokens_seen": 972356,
      "step": 1276,
      "train_runtime": 2919.374,
      "train_tokens_per_second": 333.07
    },
    {
      "epoch": 0.11742528735632184,
      "grad_norm": 6.459235668182373,
      "learning_rate": 7.258145363408522e-06,
      "loss": 0.4272160828113556,
      "num_input_tokens_seen": 973120,
      "step": 1277,
      "train_runtime": 2921.3163,
      "train_tokens_per_second": 333.11
    },
    {
      "epoch": 0.11751724137931034,
      "grad_norm": 6.01685905456543,
      "learning_rate": 7.248120300751881e-06,
      "loss": 0.3409920632839203,
      "num_input_tokens_seen": 973756,
      "step": 1278,
      "train_runtime": 2923.2171,
      "train_tokens_per_second": 333.111
    },
    {
      "epoch": 0.11760919540229885,
      "grad_norm": 6.02919340133667,
      "learning_rate": 7.238095238095239e-06,
      "loss": 0.39888471364974976,
      "num_input_tokens_seen": 974424,
      "step": 1279,
      "train_runtime": 2925.1221,
      "train_tokens_per_second": 333.123
    },
    {
      "epoch": 0.11770114942528735,
      "grad_norm": 6.491978168487549,
      "learning_rate": 7.228070175438597e-06,
      "loss": 0.4698963463306427,
      "num_input_tokens_seen": 975280,
      "step": 1280,
      "train_runtime": 2927.059,
      "train_tokens_per_second": 333.195
    },
    {
      "epoch": 0.11779310344827586,
      "grad_norm": 7.6572418212890625,
      "learning_rate": 7.218045112781955e-06,
      "loss": 0.5386727452278137,
      "num_input_tokens_seen": 975904,
      "step": 1281,
      "train_runtime": 2928.9553,
      "train_tokens_per_second": 333.192
    },
    {
      "epoch": 0.11788505747126436,
      "grad_norm": 7.648731231689453,
      "learning_rate": 7.208020050125314e-06,
      "loss": 0.4813305735588074,
      "num_input_tokens_seen": 976560,
      "step": 1282,
      "train_runtime": 2930.8896,
      "train_tokens_per_second": 333.196
    },
    {
      "epoch": 0.11797701149425287,
      "grad_norm": 6.699535369873047,
      "learning_rate": 7.197994987468673e-06,
      "loss": 0.5627443194389343,
      "num_input_tokens_seen": 977800,
      "step": 1283,
      "train_runtime": 2932.8772,
      "train_tokens_per_second": 333.393
    },
    {
      "epoch": 0.11806896551724137,
      "grad_norm": 6.436405658721924,
      "learning_rate": 7.18796992481203e-06,
      "loss": 0.3652324676513672,
      "num_input_tokens_seen": 978380,
      "step": 1284,
      "train_runtime": 2934.78,
      "train_tokens_per_second": 333.374
    },
    {
      "epoch": 0.11816091954022989,
      "grad_norm": 6.360345363616943,
      "learning_rate": 7.17794486215539e-06,
      "loss": 0.41607725620269775,
      "num_input_tokens_seen": 979136,
      "step": 1285,
      "train_runtime": 2936.7313,
      "train_tokens_per_second": 333.41
    },
    {
      "epoch": 0.1182528735632184,
      "grad_norm": 7.502152919769287,
      "learning_rate": 7.167919799498747e-06,
      "loss": 0.47135108709335327,
      "num_input_tokens_seen": 979696,
      "step": 1286,
      "train_runtime": 2938.6534,
      "train_tokens_per_second": 333.383
    },
    {
      "epoch": 0.1183448275862069,
      "grad_norm": 7.678892612457275,
      "learning_rate": 7.157894736842106e-06,
      "loss": 0.5936618447303772,
      "num_input_tokens_seen": 980712,
      "step": 1287,
      "train_runtime": 2940.641,
      "train_tokens_per_second": 333.503
    },
    {
      "epoch": 0.11843678160919541,
      "grad_norm": 7.467742919921875,
      "learning_rate": 7.147869674185464e-06,
      "loss": 0.48023247718811035,
      "num_input_tokens_seen": 981540,
      "step": 1288,
      "train_runtime": 2942.6101,
      "train_tokens_per_second": 333.561
    },
    {
      "epoch": 0.11852873563218391,
      "grad_norm": 7.7129716873168945,
      "learning_rate": 7.137844611528823e-06,
      "loss": 0.4181150794029236,
      "num_input_tokens_seen": 982072,
      "step": 1289,
      "train_runtime": 2944.5141,
      "train_tokens_per_second": 333.526
    },
    {
      "epoch": 0.11862068965517242,
      "grad_norm": 5.898501873016357,
      "learning_rate": 7.127819548872181e-06,
      "loss": 0.405730277299881,
      "num_input_tokens_seen": 982752,
      "step": 1290,
      "train_runtime": 2946.4299,
      "train_tokens_per_second": 333.54
    },
    {
      "epoch": 0.11871264367816092,
      "grad_norm": 7.654778957366943,
      "learning_rate": 7.117794486215539e-06,
      "loss": 0.45681047439575195,
      "num_input_tokens_seen": 983404,
      "step": 1291,
      "train_runtime": 2954.6326,
      "train_tokens_per_second": 332.835
    },
    {
      "epoch": 0.11880459770114943,
      "grad_norm": 6.504164218902588,
      "learning_rate": 7.107769423558898e-06,
      "loss": 0.4210878014564514,
      "num_input_tokens_seen": 984096,
      "step": 1292,
      "train_runtime": 2956.5496,
      "train_tokens_per_second": 332.853
    },
    {
      "epoch": 0.11889655172413793,
      "grad_norm": 6.645524501800537,
      "learning_rate": 7.097744360902257e-06,
      "loss": 0.5178579092025757,
      "num_input_tokens_seen": 984968,
      "step": 1293,
      "train_runtime": 2958.6199,
      "train_tokens_per_second": 332.915
    },
    {
      "epoch": 0.11898850574712644,
      "grad_norm": 7.195277690887451,
      "learning_rate": 7.087719298245614e-06,
      "loss": 0.4999891519546509,
      "num_input_tokens_seen": 985684,
      "step": 1294,
      "train_runtime": 2960.5407,
      "train_tokens_per_second": 332.941
    },
    {
      "epoch": 0.11908045977011494,
      "grad_norm": 8.545036315917969,
      "learning_rate": 7.077694235588973e-06,
      "loss": 0.4457113444805145,
      "num_input_tokens_seen": 986236,
      "step": 1295,
      "train_runtime": 2962.4355,
      "train_tokens_per_second": 332.914
    },
    {
      "epoch": 0.11917241379310345,
      "grad_norm": 6.367618083953857,
      "learning_rate": 7.067669172932331e-06,
      "loss": 0.39394238591194153,
      "num_input_tokens_seen": 987160,
      "step": 1296,
      "train_runtime": 2964.3753,
      "train_tokens_per_second": 333.008
    },
    {
      "epoch": 0.11926436781609195,
      "grad_norm": 6.343026161193848,
      "learning_rate": 7.05764411027569e-06,
      "loss": 0.46257346868515015,
      "num_input_tokens_seen": 987916,
      "step": 1297,
      "train_runtime": 2966.2829,
      "train_tokens_per_second": 333.048
    },
    {
      "epoch": 0.11935632183908046,
      "grad_norm": 6.313677787780762,
      "learning_rate": 7.047619047619048e-06,
      "loss": 0.4674181342124939,
      "num_input_tokens_seen": 988696,
      "step": 1298,
      "train_runtime": 2968.2228,
      "train_tokens_per_second": 333.094
    },
    {
      "epoch": 0.11944827586206896,
      "grad_norm": 6.0055670738220215,
      "learning_rate": 7.037593984962407e-06,
      "loss": 0.4524652361869812,
      "num_input_tokens_seen": 989496,
      "step": 1299,
      "train_runtime": 2970.1839,
      "train_tokens_per_second": 333.143
    },
    {
      "epoch": 0.11954022988505747,
      "grad_norm": 7.333286285400391,
      "learning_rate": 7.027568922305765e-06,
      "loss": 0.501343846321106,
      "num_input_tokens_seen": 990224,
      "step": 1300,
      "train_runtime": 2972.139,
      "train_tokens_per_second": 333.169
    },
    {
      "epoch": 0.11963218390804597,
      "grad_norm": 6.726962089538574,
      "learning_rate": 7.017543859649123e-06,
      "loss": 0.42660602927207947,
      "num_input_tokens_seen": 991196,
      "step": 1301,
      "train_runtime": 2974.1136,
      "train_tokens_per_second": 333.274
    },
    {
      "epoch": 0.11972413793103448,
      "grad_norm": 7.37306022644043,
      "learning_rate": 7.0075187969924815e-06,
      "loss": 0.5156997442245483,
      "num_input_tokens_seen": 991896,
      "step": 1302,
      "train_runtime": 2976.0606,
      "train_tokens_per_second": 333.292
    },
    {
      "epoch": 0.11981609195402299,
      "grad_norm": 5.937094688415527,
      "learning_rate": 6.997493734335841e-06,
      "loss": 0.4371004104614258,
      "num_input_tokens_seen": 992536,
      "step": 1303,
      "train_runtime": 2978.0514,
      "train_tokens_per_second": 333.284
    },
    {
      "epoch": 0.11990804597701149,
      "grad_norm": 7.102948188781738,
      "learning_rate": 6.9874686716791984e-06,
      "loss": 0.5632842183113098,
      "num_input_tokens_seen": 993284,
      "step": 1304,
      "train_runtime": 2979.9936,
      "train_tokens_per_second": 333.317
    },
    {
      "epoch": 0.12,
      "grad_norm": 7.164879322052002,
      "learning_rate": 6.977443609022557e-06,
      "loss": 0.6001710891723633,
      "num_input_tokens_seen": 994356,
      "step": 1305,
      "train_runtime": 2981.9767,
      "train_tokens_per_second": 333.455
    },
    {
      "epoch": 0.1200919540229885,
      "grad_norm": 7.405031681060791,
      "learning_rate": 6.9674185463659146e-06,
      "loss": 0.5065048933029175,
      "num_input_tokens_seen": 995140,
      "step": 1306,
      "train_runtime": 2983.9312,
      "train_tokens_per_second": 333.5
    },
    {
      "epoch": 0.120183908045977,
      "grad_norm": 6.406159400939941,
      "learning_rate": 6.957393483709274e-06,
      "loss": 0.49356621503829956,
      "num_input_tokens_seen": 995996,
      "step": 1307,
      "train_runtime": 2985.8675,
      "train_tokens_per_second": 333.57
    },
    {
      "epoch": 0.12027586206896551,
      "grad_norm": 7.691381931304932,
      "learning_rate": 6.947368421052632e-06,
      "loss": 0.5442687273025513,
      "num_input_tokens_seen": 996680,
      "step": 1308,
      "train_runtime": 2987.7861,
      "train_tokens_per_second": 333.585
    },
    {
      "epoch": 0.12036781609195402,
      "grad_norm": 6.470928192138672,
      "learning_rate": 6.93734335839599e-06,
      "loss": 0.4957112669944763,
      "num_input_tokens_seen": 997352,
      "step": 1309,
      "train_runtime": 2989.7051,
      "train_tokens_per_second": 333.595
    },
    {
      "epoch": 0.12045977011494254,
      "grad_norm": 6.938445568084717,
      "learning_rate": 6.927318295739349e-06,
      "loss": 0.47365546226501465,
      "num_input_tokens_seen": 998180,
      "step": 1310,
      "train_runtime": 2991.6573,
      "train_tokens_per_second": 333.655
    },
    {
      "epoch": 0.12055172413793104,
      "grad_norm": 7.241644382476807,
      "learning_rate": 6.917293233082707e-06,
      "loss": 0.5160254240036011,
      "num_input_tokens_seen": 999008,
      "step": 1311,
      "train_runtime": 2993.5955,
      "train_tokens_per_second": 333.715
    },
    {
      "epoch": 0.12064367816091955,
      "grad_norm": 6.897066116333008,
      "learning_rate": 6.9072681704260655e-06,
      "loss": 0.41822683811187744,
      "num_input_tokens_seen": 999748,
      "step": 1312,
      "train_runtime": 2995.5064,
      "train_tokens_per_second": 333.749
    },
    {
      "epoch": 0.12073563218390805,
      "grad_norm": 6.118927955627441,
      "learning_rate": 6.897243107769425e-06,
      "loss": 0.4357345700263977,
      "num_input_tokens_seen": 1000256,
      "step": 1313,
      "train_runtime": 2997.4297,
      "train_tokens_per_second": 333.705
    },
    {
      "epoch": 0.12082758620689656,
      "grad_norm": 6.424326419830322,
      "learning_rate": 6.8872180451127825e-06,
      "loss": 0.40614956617355347,
      "num_input_tokens_seen": 1000924,
      "step": 1314,
      "train_runtime": 2999.3728,
      "train_tokens_per_second": 333.711
    },
    {
      "epoch": 0.12091954022988506,
      "grad_norm": 6.810154438018799,
      "learning_rate": 6.877192982456141e-06,
      "loss": 0.4233674108982086,
      "num_input_tokens_seen": 1001508,
      "step": 1315,
      "train_runtime": 3001.2942,
      "train_tokens_per_second": 333.692
    },
    {
      "epoch": 0.12101149425287357,
      "grad_norm": 7.00108528137207,
      "learning_rate": 6.867167919799499e-06,
      "loss": 0.5287919044494629,
      "num_input_tokens_seen": 1002272,
      "step": 1316,
      "train_runtime": 3003.2455,
      "train_tokens_per_second": 333.73
    },
    {
      "epoch": 0.12110344827586207,
      "grad_norm": 6.731994152069092,
      "learning_rate": 6.857142857142858e-06,
      "loss": 0.47859540581703186,
      "num_input_tokens_seen": 1002940,
      "step": 1317,
      "train_runtime": 3005.1695,
      "train_tokens_per_second": 333.738
    },
    {
      "epoch": 0.12119540229885058,
      "grad_norm": 6.614876747131348,
      "learning_rate": 6.847117794486216e-06,
      "loss": 0.4953509569168091,
      "num_input_tokens_seen": 1003656,
      "step": 1318,
      "train_runtime": 3007.118,
      "train_tokens_per_second": 333.76
    },
    {
      "epoch": 0.12128735632183908,
      "grad_norm": 6.834334373474121,
      "learning_rate": 6.837092731829574e-06,
      "loss": 0.45668864250183105,
      "num_input_tokens_seen": 1004560,
      "step": 1319,
      "train_runtime": 3009.0766,
      "train_tokens_per_second": 333.843
    },
    {
      "epoch": 0.12137931034482759,
      "grad_norm": 6.399694919586182,
      "learning_rate": 6.827067669172933e-06,
      "loss": 0.4552806615829468,
      "num_input_tokens_seen": 1005216,
      "step": 1320,
      "train_runtime": 3010.9833,
      "train_tokens_per_second": 333.85
    },
    {
      "epoch": 0.12147126436781609,
      "grad_norm": 6.645393371582031,
      "learning_rate": 6.817042606516291e-06,
      "loss": 0.4933241009712219,
      "num_input_tokens_seen": 1006024,
      "step": 1321,
      "train_runtime": 3019.5186,
      "train_tokens_per_second": 333.174
    },
    {
      "epoch": 0.1215632183908046,
      "grad_norm": 6.884487628936768,
      "learning_rate": 6.8070175438596495e-06,
      "loss": 0.5027730464935303,
      "num_input_tokens_seen": 1006796,
      "step": 1322,
      "train_runtime": 3021.5673,
      "train_tokens_per_second": 333.203
    },
    {
      "epoch": 0.1216551724137931,
      "grad_norm": 6.07246732711792,
      "learning_rate": 6.796992481203009e-06,
      "loss": 0.4355407953262329,
      "num_input_tokens_seen": 1007412,
      "step": 1323,
      "train_runtime": 3023.4839,
      "train_tokens_per_second": 333.196
    },
    {
      "epoch": 0.12174712643678161,
      "grad_norm": 6.737724304199219,
      "learning_rate": 6.7869674185463665e-06,
      "loss": 0.3820093274116516,
      "num_input_tokens_seen": 1008040,
      "step": 1324,
      "train_runtime": 3025.3977,
      "train_tokens_per_second": 333.193
    },
    {
      "epoch": 0.12183908045977011,
      "grad_norm": 6.211243152618408,
      "learning_rate": 6.776942355889725e-06,
      "loss": 0.41384297609329224,
      "num_input_tokens_seen": 1008664,
      "step": 1325,
      "train_runtime": 3027.3225,
      "train_tokens_per_second": 333.187
    },
    {
      "epoch": 0.12193103448275862,
      "grad_norm": 7.142533302307129,
      "learning_rate": 6.766917293233083e-06,
      "loss": 0.5230511426925659,
      "num_input_tokens_seen": 1009344,
      "step": 1326,
      "train_runtime": 3029.2401,
      "train_tokens_per_second": 333.2
    },
    {
      "epoch": 0.12202298850574712,
      "grad_norm": 7.864989280700684,
      "learning_rate": 6.756892230576442e-06,
      "loss": 0.48035430908203125,
      "num_input_tokens_seen": 1009924,
      "step": 1327,
      "train_runtime": 3031.1697,
      "train_tokens_per_second": 333.18
    },
    {
      "epoch": 0.12211494252873563,
      "grad_norm": 5.697092056274414,
      "learning_rate": 6.7468671679198004e-06,
      "loss": 0.35426461696624756,
      "num_input_tokens_seen": 1010496,
      "step": 1328,
      "train_runtime": 3033.1224,
      "train_tokens_per_second": 333.154
    },
    {
      "epoch": 0.12220689655172413,
      "grad_norm": 5.4057769775390625,
      "learning_rate": 6.736842105263158e-06,
      "loss": 0.36207208037376404,
      "num_input_tokens_seen": 1012652,
      "step": 1329,
      "train_runtime": 3035.3075,
      "train_tokens_per_second": 333.624
    },
    {
      "epoch": 0.12229885057471264,
      "grad_norm": 6.612935543060303,
      "learning_rate": 6.726817042606517e-06,
      "loss": 0.390036404132843,
      "num_input_tokens_seen": 1013384,
      "step": 1330,
      "train_runtime": 3037.2674,
      "train_tokens_per_second": 333.65
    },
    {
      "epoch": 0.12239080459770114,
      "grad_norm": 6.282397270202637,
      "learning_rate": 6.716791979949875e-06,
      "loss": 0.5557735562324524,
      "num_input_tokens_seen": 1014008,
      "step": 1331,
      "train_runtime": 3039.2309,
      "train_tokens_per_second": 333.64
    },
    {
      "epoch": 0.12248275862068965,
      "grad_norm": 7.2630205154418945,
      "learning_rate": 6.7067669172932335e-06,
      "loss": 0.5023081302642822,
      "num_input_tokens_seen": 1014708,
      "step": 1332,
      "train_runtime": 3041.2158,
      "train_tokens_per_second": 333.652
    },
    {
      "epoch": 0.12257471264367815,
      "grad_norm": 7.504981517791748,
      "learning_rate": 6.696741854636593e-06,
      "loss": 0.5782328248023987,
      "num_input_tokens_seen": 1015332,
      "step": 1333,
      "train_runtime": 3043.1424,
      "train_tokens_per_second": 333.646
    },
    {
      "epoch": 0.12266666666666666,
      "grad_norm": 7.2758469581604,
      "learning_rate": 6.6867167919799505e-06,
      "loss": 0.5410507917404175,
      "num_input_tokens_seen": 1016096,
      "step": 1334,
      "train_runtime": 3045.0864,
      "train_tokens_per_second": 333.684
    },
    {
      "epoch": 0.12275862068965518,
      "grad_norm": 6.432806491851807,
      "learning_rate": 6.676691729323309e-06,
      "loss": 0.4315190315246582,
      "num_input_tokens_seen": 1016772,
      "step": 1335,
      "train_runtime": 3047.0245,
      "train_tokens_per_second": 333.693
    },
    {
      "epoch": 0.12285057471264368,
      "grad_norm": 6.5717949867248535,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.4417974650859833,
      "num_input_tokens_seen": 1017432,
      "step": 1336,
      "train_runtime": 3048.9766,
      "train_tokens_per_second": 333.696
    },
    {
      "epoch": 0.12294252873563219,
      "grad_norm": 6.500550270080566,
      "learning_rate": 6.656641604010026e-06,
      "loss": 0.4049522578716278,
      "num_input_tokens_seen": 1018016,
      "step": 1337,
      "train_runtime": 3050.8932,
      "train_tokens_per_second": 333.678
    },
    {
      "epoch": 0.1230344827586207,
      "grad_norm": 6.5369133949279785,
      "learning_rate": 6.6466165413533845e-06,
      "loss": 0.5170691013336182,
      "num_input_tokens_seen": 1018748,
      "step": 1338,
      "train_runtime": 3052.8676,
      "train_tokens_per_second": 333.702
    },
    {
      "epoch": 0.1231264367816092,
      "grad_norm": 6.739131450653076,
      "learning_rate": 6.636591478696742e-06,
      "loss": 0.4723159372806549,
      "num_input_tokens_seen": 1019424,
      "step": 1339,
      "train_runtime": 3054.7944,
      "train_tokens_per_second": 333.713
    },
    {
      "epoch": 0.1232183908045977,
      "grad_norm": 6.277287006378174,
      "learning_rate": 6.6265664160401014e-06,
      "loss": 0.4405796527862549,
      "num_input_tokens_seen": 1020172,
      "step": 1340,
      "train_runtime": 3056.7242,
      "train_tokens_per_second": 333.747
    },
    {
      "epoch": 0.12331034482758621,
      "grad_norm": 7.16923189163208,
      "learning_rate": 6.616541353383459e-06,
      "loss": 0.5072904229164124,
      "num_input_tokens_seen": 1020976,
      "step": 1341,
      "train_runtime": 3058.6876,
      "train_tokens_per_second": 333.795
    },
    {
      "epoch": 0.12340229885057472,
      "grad_norm": 7.278201103210449,
      "learning_rate": 6.6065162907268176e-06,
      "loss": 0.5480138063430786,
      "num_input_tokens_seen": 1021620,
      "step": 1342,
      "train_runtime": 3060.6039,
      "train_tokens_per_second": 333.797
    },
    {
      "epoch": 0.12349425287356322,
      "grad_norm": 7.211563587188721,
      "learning_rate": 6.596491228070177e-06,
      "loss": 0.49704140424728394,
      "num_input_tokens_seen": 1022536,
      "step": 1343,
      "train_runtime": 3062.5565,
      "train_tokens_per_second": 333.883
    },
    {
      "epoch": 0.12358620689655173,
      "grad_norm": 5.980576992034912,
      "learning_rate": 6.5864661654135345e-06,
      "loss": 0.38621148467063904,
      "num_input_tokens_seen": 1023336,
      "step": 1344,
      "train_runtime": 3064.4769,
      "train_tokens_per_second": 333.935
    },
    {
      "epoch": 0.12367816091954023,
      "grad_norm": 6.77341365814209,
      "learning_rate": 6.576441102756893e-06,
      "loss": 0.3904351592063904,
      "num_input_tokens_seen": 1023992,
      "step": 1345,
      "train_runtime": 3066.47,
      "train_tokens_per_second": 333.932
    },
    {
      "epoch": 0.12377011494252874,
      "grad_norm": 6.9706315994262695,
      "learning_rate": 6.566416040100251e-06,
      "loss": 0.43598905205726624,
      "num_input_tokens_seen": 1024908,
      "step": 1346,
      "train_runtime": 3068.4171,
      "train_tokens_per_second": 334.018
    },
    {
      "epoch": 0.12386206896551724,
      "grad_norm": 6.686893463134766,
      "learning_rate": 6.55639097744361e-06,
      "loss": 0.4698653221130371,
      "num_input_tokens_seen": 1025832,
      "step": 1347,
      "train_runtime": 3070.416,
      "train_tokens_per_second": 334.102
    },
    {
      "epoch": 0.12395402298850575,
      "grad_norm": 6.886135101318359,
      "learning_rate": 6.5463659147869685e-06,
      "loss": 0.454531729221344,
      "num_input_tokens_seen": 1026528,
      "step": 1348,
      "train_runtime": 3072.3435,
      "train_tokens_per_second": 334.119
    },
    {
      "epoch": 0.12404597701149425,
      "grad_norm": 7.164763927459717,
      "learning_rate": 6.536340852130326e-06,
      "loss": 0.51418137550354,
      "num_input_tokens_seen": 1027468,
      "step": 1349,
      "train_runtime": 3074.3162,
      "train_tokens_per_second": 334.21
    },
    {
      "epoch": 0.12413793103448276,
      "grad_norm": 7.003333568572998,
      "learning_rate": 6.526315789473685e-06,
      "loss": 0.48889005184173584,
      "num_input_tokens_seen": 1028212,
      "step": 1350,
      "train_runtime": 3076.2264,
      "train_tokens_per_second": 334.245
    },
    {
      "epoch": 0.12422988505747126,
      "grad_norm": 6.748643398284912,
      "learning_rate": 6.516290726817043e-06,
      "loss": 0.4787951707839966,
      "num_input_tokens_seen": 1029088,
      "step": 1351,
      "train_runtime": 3084.7446,
      "train_tokens_per_second": 333.606
    },
    {
      "epoch": 0.12432183908045977,
      "grad_norm": 6.4219746589660645,
      "learning_rate": 6.506265664160402e-06,
      "loss": 0.5620080232620239,
      "num_input_tokens_seen": 1030104,
      "step": 1352,
      "train_runtime": 3086.7219,
      "train_tokens_per_second": 333.721
    },
    {
      "epoch": 0.12441379310344827,
      "grad_norm": 6.431827545166016,
      "learning_rate": 6.49624060150376e-06,
      "loss": 0.48305195569992065,
      "num_input_tokens_seen": 1030736,
      "step": 1353,
      "train_runtime": 3088.6306,
      "train_tokens_per_second": 333.719
    },
    {
      "epoch": 0.12450574712643678,
      "grad_norm": 5.823015213012695,
      "learning_rate": 6.486215538847119e-06,
      "loss": 0.4215491712093353,
      "num_input_tokens_seen": 1031332,
      "step": 1354,
      "train_runtime": 3090.5332,
      "train_tokens_per_second": 333.707
    },
    {
      "epoch": 0.12459770114942528,
      "grad_norm": 5.891067981719971,
      "learning_rate": 6.476190476190477e-06,
      "loss": 0.4833441376686096,
      "num_input_tokens_seen": 1032648,
      "step": 1355,
      "train_runtime": 3092.5848,
      "train_tokens_per_second": 333.911
    },
    {
      "epoch": 0.12468965517241379,
      "grad_norm": 6.960275650024414,
      "learning_rate": 6.466165413533835e-06,
      "loss": 0.45002007484436035,
      "num_input_tokens_seen": 1033300,
      "step": 1356,
      "train_runtime": 3094.5647,
      "train_tokens_per_second": 333.908
    },
    {
      "epoch": 0.1247816091954023,
      "grad_norm": 6.904165744781494,
      "learning_rate": 6.456140350877193e-06,
      "loss": 0.46605169773101807,
      "num_input_tokens_seen": 1033900,
      "step": 1357,
      "train_runtime": 3096.4978,
      "train_tokens_per_second": 333.893
    },
    {
      "epoch": 0.1248735632183908,
      "grad_norm": 7.494766712188721,
      "learning_rate": 6.4461152882205525e-06,
      "loss": 0.4720362424850464,
      "num_input_tokens_seen": 1034684,
      "step": 1358,
      "train_runtime": 3098.4558,
      "train_tokens_per_second": 333.935
    },
    {
      "epoch": 0.1249655172413793,
      "grad_norm": 8.161543846130371,
      "learning_rate": 6.43609022556391e-06,
      "loss": 0.5111762881278992,
      "num_input_tokens_seen": 1035332,
      "step": 1359,
      "train_runtime": 3100.4702,
      "train_tokens_per_second": 333.927
    },
    {
      "epoch": 0.1250574712643678,
      "grad_norm": 6.962998390197754,
      "learning_rate": 6.426065162907269e-06,
      "loss": 0.4295594394207001,
      "num_input_tokens_seen": 1035960,
      "step": 1360,
      "train_runtime": 3102.3911,
      "train_tokens_per_second": 333.923
    },
    {
      "epoch": 0.12514942528735631,
      "grad_norm": 6.188365459442139,
      "learning_rate": 6.416040100250627e-06,
      "loss": 0.4945793151855469,
      "num_input_tokens_seen": 1036708,
      "step": 1361,
      "train_runtime": 3104.3236,
      "train_tokens_per_second": 333.956
    },
    {
      "epoch": 0.12524137931034482,
      "grad_norm": 6.810634136199951,
      "learning_rate": 6.406015037593986e-06,
      "loss": 0.3978559374809265,
      "num_input_tokens_seen": 1037272,
      "step": 1362,
      "train_runtime": 3106.2332,
      "train_tokens_per_second": 333.932
    },
    {
      "epoch": 0.12533333333333332,
      "grad_norm": 5.729403495788574,
      "learning_rate": 6.395989974937344e-06,
      "loss": 0.35970085859298706,
      "num_input_tokens_seen": 1037856,
      "step": 1363,
      "train_runtime": 3108.1633,
      "train_tokens_per_second": 333.913
    },
    {
      "epoch": 0.12542528735632183,
      "grad_norm": 6.915588855743408,
      "learning_rate": 6.385964912280702e-06,
      "loss": 0.4961872100830078,
      "num_input_tokens_seen": 1038472,
      "step": 1364,
      "train_runtime": 3110.0953,
      "train_tokens_per_second": 333.904
    },
    {
      "epoch": 0.12551724137931033,
      "grad_norm": 7.976280689239502,
      "learning_rate": 6.375939849624061e-06,
      "loss": 0.4000566303730011,
      "num_input_tokens_seen": 1039248,
      "step": 1365,
      "train_runtime": 3112.0416,
      "train_tokens_per_second": 333.944
    },
    {
      "epoch": 0.12560919540229884,
      "grad_norm": 7.088489532470703,
      "learning_rate": 6.365914786967419e-06,
      "loss": 0.4856930375099182,
      "num_input_tokens_seen": 1039872,
      "step": 1366,
      "train_runtime": 3113.9685,
      "train_tokens_per_second": 333.938
    },
    {
      "epoch": 0.12570114942528735,
      "grad_norm": 5.9422526359558105,
      "learning_rate": 6.355889724310777e-06,
      "loss": 0.5661624670028687,
      "num_input_tokens_seen": 1041652,
      "step": 1367,
      "train_runtime": 3116.1218,
      "train_tokens_per_second": 334.278
    },
    {
      "epoch": 0.12579310344827585,
      "grad_norm": 6.923958778381348,
      "learning_rate": 6.3458646616541366e-06,
      "loss": 0.5581769943237305,
      "num_input_tokens_seen": 1042432,
      "step": 1368,
      "train_runtime": 3118.038,
      "train_tokens_per_second": 334.323
    },
    {
      "epoch": 0.12588505747126436,
      "grad_norm": 5.821872711181641,
      "learning_rate": 6.335839598997494e-06,
      "loss": 0.4299492835998535,
      "num_input_tokens_seen": 1043080,
      "step": 1369,
      "train_runtime": 3120.0305,
      "train_tokens_per_second": 334.317
    },
    {
      "epoch": 0.12597701149425286,
      "grad_norm": 7.4547905921936035,
      "learning_rate": 6.325814536340853e-06,
      "loss": 0.47298097610473633,
      "num_input_tokens_seen": 1043712,
      "step": 1370,
      "train_runtime": 3121.9499,
      "train_tokens_per_second": 334.314
    },
    {
      "epoch": 0.12606896551724137,
      "grad_norm": 7.124066352844238,
      "learning_rate": 6.31578947368421e-06,
      "loss": 0.5197014212608337,
      "num_input_tokens_seen": 1044508,
      "step": 1371,
      "train_runtime": 3123.8831,
      "train_tokens_per_second": 334.362
    },
    {
      "epoch": 0.1261609195402299,
      "grad_norm": 6.706704139709473,
      "learning_rate": 6.30576441102757e-06,
      "loss": 0.41323238611221313,
      "num_input_tokens_seen": 1045168,
      "step": 1372,
      "train_runtime": 3125.8614,
      "train_tokens_per_second": 334.362
    },
    {
      "epoch": 0.1262528735632184,
      "grad_norm": 6.124212265014648,
      "learning_rate": 6.295739348370928e-06,
      "loss": 0.40887323021888733,
      "num_input_tokens_seen": 1045792,
      "step": 1373,
      "train_runtime": 3127.8439,
      "train_tokens_per_second": 334.349
    },
    {
      "epoch": 0.1263448275862069,
      "grad_norm": 6.2818074226379395,
      "learning_rate": 6.285714285714286e-06,
      "loss": 0.3918152451515198,
      "num_input_tokens_seen": 1046396,
      "step": 1374,
      "train_runtime": 3129.7627,
      "train_tokens_per_second": 334.337
    },
    {
      "epoch": 0.12643678160919541,
      "grad_norm": 6.364823341369629,
      "learning_rate": 6.275689223057645e-06,
      "loss": 0.4632605314254761,
      "num_input_tokens_seen": 1047064,
      "step": 1375,
      "train_runtime": 3131.7139,
      "train_tokens_per_second": 334.342
    },
    {
      "epoch": 0.12652873563218392,
      "grad_norm": 6.550987720489502,
      "learning_rate": 6.265664160401003e-06,
      "loss": 0.593927800655365,
      "num_input_tokens_seen": 1047856,
      "step": 1376,
      "train_runtime": 3133.6698,
      "train_tokens_per_second": 334.386
    },
    {
      "epoch": 0.12662068965517242,
      "grad_norm": 6.433656692504883,
      "learning_rate": 6.255639097744361e-06,
      "loss": 0.4245767593383789,
      "num_input_tokens_seen": 1048368,
      "step": 1377,
      "train_runtime": 3135.5827,
      "train_tokens_per_second": 334.346
    },
    {
      "epoch": 0.12671264367816093,
      "grad_norm": 7.409817695617676,
      "learning_rate": 6.245614035087721e-06,
      "loss": 0.5045289397239685,
      "num_input_tokens_seen": 1049156,
      "step": 1378,
      "train_runtime": 3137.5306,
      "train_tokens_per_second": 334.389
    },
    {
      "epoch": 0.12680459770114944,
      "grad_norm": 7.645226001739502,
      "learning_rate": 6.235588972431078e-06,
      "loss": 0.46224796772003174,
      "num_input_tokens_seen": 1049816,
      "step": 1379,
      "train_runtime": 3139.4764,
      "train_tokens_per_second": 334.392
    },
    {
      "epoch": 0.12689655172413794,
      "grad_norm": 6.701642036437988,
      "learning_rate": 6.225563909774437e-06,
      "loss": 0.5383749604225159,
      "num_input_tokens_seen": 1050704,
      "step": 1380,
      "train_runtime": 3141.4453,
      "train_tokens_per_second": 334.465
    },
    {
      "epoch": 0.12698850574712645,
      "grad_norm": 6.478757858276367,
      "learning_rate": 6.215538847117794e-06,
      "loss": 0.3706413507461548,
      "num_input_tokens_seen": 1051476,
      "step": 1381,
      "train_runtime": 3150.0197,
      "train_tokens_per_second": 333.8
    },
    {
      "epoch": 0.12708045977011495,
      "grad_norm": 6.502756595611572,
      "learning_rate": 6.205513784461154e-06,
      "loss": 0.41397708654403687,
      "num_input_tokens_seen": 1052128,
      "step": 1382,
      "train_runtime": 3151.9402,
      "train_tokens_per_second": 333.803
    },
    {
      "epoch": 0.12717241379310346,
      "grad_norm": 8.47773551940918,
      "learning_rate": 6.195488721804512e-06,
      "loss": 0.40870481729507446,
      "num_input_tokens_seen": 1052760,
      "step": 1383,
      "train_runtime": 3153.8462,
      "train_tokens_per_second": 333.802
    },
    {
      "epoch": 0.12726436781609196,
      "grad_norm": 7.772322654724121,
      "learning_rate": 6.18546365914787e-06,
      "loss": 0.4777206480503082,
      "num_input_tokens_seen": 1053512,
      "step": 1384,
      "train_runtime": 3155.8053,
      "train_tokens_per_second": 333.833
    },
    {
      "epoch": 0.12735632183908047,
      "grad_norm": 7.561821460723877,
      "learning_rate": 6.175438596491229e-06,
      "loss": 0.5530266165733337,
      "num_input_tokens_seen": 1054304,
      "step": 1385,
      "train_runtime": 3157.7642,
      "train_tokens_per_second": 333.877
    },
    {
      "epoch": 0.12744827586206897,
      "grad_norm": 7.125812530517578,
      "learning_rate": 6.165413533834587e-06,
      "loss": 0.45317232608795166,
      "num_input_tokens_seen": 1054916,
      "step": 1386,
      "train_runtime": 3159.6935,
      "train_tokens_per_second": 333.867
    },
    {
      "epoch": 0.12754022988505748,
      "grad_norm": 6.858726501464844,
      "learning_rate": 6.155388471177945e-06,
      "loss": 0.4123727083206177,
      "num_input_tokens_seen": 1055588,
      "step": 1387,
      "train_runtime": 3161.6197,
      "train_tokens_per_second": 333.876
    },
    {
      "epoch": 0.12763218390804598,
      "grad_norm": 9.354341506958008,
      "learning_rate": 6.145363408521305e-06,
      "loss": 0.3369084298610687,
      "num_input_tokens_seen": 1056152,
      "step": 1388,
      "train_runtime": 3163.5271,
      "train_tokens_per_second": 333.853
    },
    {
      "epoch": 0.1277241379310345,
      "grad_norm": 8.496859550476074,
      "learning_rate": 6.135338345864662e-06,
      "loss": 0.5529747009277344,
      "num_input_tokens_seen": 1056916,
      "step": 1389,
      "train_runtime": 3165.4334,
      "train_tokens_per_second": 333.893
    },
    {
      "epoch": 0.127816091954023,
      "grad_norm": 7.04681921005249,
      "learning_rate": 6.125313283208021e-06,
      "loss": 0.4281120300292969,
      "num_input_tokens_seen": 1057568,
      "step": 1390,
      "train_runtime": 3167.3365,
      "train_tokens_per_second": 333.898
    },
    {
      "epoch": 0.1279080459770115,
      "grad_norm": 8.527612686157227,
      "learning_rate": 6.115288220551378e-06,
      "loss": 0.534282386302948,
      "num_input_tokens_seen": 1058232,
      "step": 1391,
      "train_runtime": 3169.2875,
      "train_tokens_per_second": 333.902
    },
    {
      "epoch": 0.128,
      "grad_norm": 7.677074909210205,
      "learning_rate": 6.105263157894738e-06,
      "loss": 0.47458934783935547,
      "num_input_tokens_seen": 1058864,
      "step": 1392,
      "train_runtime": 3171.2049,
      "train_tokens_per_second": 333.9
    },
    {
      "epoch": 0.1280919540229885,
      "grad_norm": 5.942317485809326,
      "learning_rate": 6.095238095238096e-06,
      "loss": 0.41658198833465576,
      "num_input_tokens_seen": 1059528,
      "step": 1393,
      "train_runtime": 3173.1436,
      "train_tokens_per_second": 333.905
    },
    {
      "epoch": 0.128183908045977,
      "grad_norm": 8.285112380981445,
      "learning_rate": 6.085213032581454e-06,
      "loss": 0.488636314868927,
      "num_input_tokens_seen": 1060264,
      "step": 1394,
      "train_runtime": 3175.0821,
      "train_tokens_per_second": 333.933
    },
    {
      "epoch": 0.12827586206896552,
      "grad_norm": 6.375500679016113,
      "learning_rate": 6.075187969924813e-06,
      "loss": 0.5229175090789795,
      "num_input_tokens_seen": 1061080,
      "step": 1395,
      "train_runtime": 3177.042,
      "train_tokens_per_second": 333.984
    },
    {
      "epoch": 0.12836781609195402,
      "grad_norm": 7.5325846672058105,
      "learning_rate": 6.065162907268171e-06,
      "loss": 0.5719152688980103,
      "num_input_tokens_seen": 1061732,
      "step": 1396,
      "train_runtime": 3179.0205,
      "train_tokens_per_second": 333.981
    },
    {
      "epoch": 0.12845977011494253,
      "grad_norm": 7.508668899536133,
      "learning_rate": 6.055137844611529e-06,
      "loss": 0.5196084976196289,
      "num_input_tokens_seen": 1062472,
      "step": 1397,
      "train_runtime": 3181.0075,
      "train_tokens_per_second": 334.005
    },
    {
      "epoch": 0.12855172413793103,
      "grad_norm": 6.6803178787231445,
      "learning_rate": 6.045112781954889e-06,
      "loss": 0.5810975432395935,
      "num_input_tokens_seen": 1063344,
      "step": 1398,
      "train_runtime": 3182.9884,
      "train_tokens_per_second": 334.071
    },
    {
      "epoch": 0.12864367816091954,
      "grad_norm": 6.288444995880127,
      "learning_rate": 6.035087719298246e-06,
      "loss": 0.45880621671676636,
      "num_input_tokens_seen": 1064348,
      "step": 1399,
      "train_runtime": 3184.9928,
      "train_tokens_per_second": 334.176
    },
    {
      "epoch": 0.12873563218390804,
      "grad_norm": 6.953741073608398,
      "learning_rate": 6.025062656641605e-06,
      "loss": 0.5626611709594727,
      "num_input_tokens_seen": 1065008,
      "step": 1400,
      "train_runtime": 3186.915,
      "train_tokens_per_second": 334.181
    },
    {
      "epoch": 0.12873563218390804,
      "eval_loss": 1.8145204782485962,
      "eval_runtime": 27.9342,
      "eval_samples_per_second": 35.798,
      "eval_steps_per_second": 8.95,
      "num_input_tokens_seen": 1065008,
      "step": 1400
    },
    {
      "epoch": 0.12882758620689655,
      "grad_norm": 6.97532844543457,
      "learning_rate": 6.015037593984962e-06,
      "loss": 0.41040074825286865,
      "num_input_tokens_seen": 1065768,
      "step": 1401,
      "train_runtime": 3216.7731,
      "train_tokens_per_second": 331.316
    },
    {
      "epoch": 0.12891954022988505,
      "grad_norm": 7.571268081665039,
      "learning_rate": 6.005012531328322e-06,
      "loss": 0.4846341609954834,
      "num_input_tokens_seen": 1066336,
      "step": 1402,
      "train_runtime": 3218.6692,
      "train_tokens_per_second": 331.297
    },
    {
      "epoch": 0.12901149425287356,
      "grad_norm": 5.804952144622803,
      "learning_rate": 5.994987468671679e-06,
      "loss": 0.38645362854003906,
      "num_input_tokens_seen": 1067176,
      "step": 1403,
      "train_runtime": 3220.63,
      "train_tokens_per_second": 331.356
    },
    {
      "epoch": 0.12910344827586206,
      "grad_norm": 6.866107940673828,
      "learning_rate": 5.984962406015038e-06,
      "loss": 0.425374299287796,
      "num_input_tokens_seen": 1067920,
      "step": 1404,
      "train_runtime": 3222.5848,
      "train_tokens_per_second": 331.386
    },
    {
      "epoch": 0.12919540229885057,
      "grad_norm": 7.225115776062012,
      "learning_rate": 5.974937343358397e-06,
      "loss": 0.47367221117019653,
      "num_input_tokens_seen": 1068652,
      "step": 1405,
      "train_runtime": 3224.5057,
      "train_tokens_per_second": 331.416
    },
    {
      "epoch": 0.12928735632183908,
      "grad_norm": 6.78947639465332,
      "learning_rate": 5.964912280701755e-06,
      "loss": 0.44649946689605713,
      "num_input_tokens_seen": 1069596,
      "step": 1406,
      "train_runtime": 3226.4587,
      "train_tokens_per_second": 331.508
    },
    {
      "epoch": 0.12937931034482758,
      "grad_norm": 6.747674465179443,
      "learning_rate": 5.954887218045113e-06,
      "loss": 0.5224787592887878,
      "num_input_tokens_seen": 1071032,
      "step": 1407,
      "train_runtime": 3228.5047,
      "train_tokens_per_second": 331.742
    },
    {
      "epoch": 0.12947126436781609,
      "grad_norm": 6.45105504989624,
      "learning_rate": 5.944862155388471e-06,
      "loss": 0.3473295569419861,
      "num_input_tokens_seen": 1071628,
      "step": 1408,
      "train_runtime": 3230.4364,
      "train_tokens_per_second": 331.729
    },
    {
      "epoch": 0.1295632183908046,
      "grad_norm": 7.346982002258301,
      "learning_rate": 5.93483709273183e-06,
      "loss": 0.6219696402549744,
      "num_input_tokens_seen": 1072356,
      "step": 1409,
      "train_runtime": 3232.3553,
      "train_tokens_per_second": 331.757
    },
    {
      "epoch": 0.1296551724137931,
      "grad_norm": 6.83197546005249,
      "learning_rate": 5.924812030075189e-06,
      "loss": 0.4776724576950073,
      "num_input_tokens_seen": 1073052,
      "step": 1410,
      "train_runtime": 3234.2946,
      "train_tokens_per_second": 331.773
    },
    {
      "epoch": 0.1297471264367816,
      "grad_norm": 6.5279645919799805,
      "learning_rate": 5.9147869674185465e-06,
      "loss": 0.4475386142730713,
      "num_input_tokens_seen": 1073656,
      "step": 1411,
      "train_runtime": 3243.2099,
      "train_tokens_per_second": 331.047
    },
    {
      "epoch": 0.1298390804597701,
      "grad_norm": 7.811605930328369,
      "learning_rate": 5.904761904761905e-06,
      "loss": 0.45823603868484497,
      "num_input_tokens_seen": 1074472,
      "step": 1412,
      "train_runtime": 3245.1291,
      "train_tokens_per_second": 331.103
    },
    {
      "epoch": 0.1299310344827586,
      "grad_norm": 7.085739612579346,
      "learning_rate": 5.8947368421052634e-06,
      "loss": 0.3285777270793915,
      "num_input_tokens_seen": 1075088,
      "step": 1413,
      "train_runtime": 3247.0345,
      "train_tokens_per_second": 331.098
    },
    {
      "epoch": 0.13002298850574712,
      "grad_norm": 6.5943922996521,
      "learning_rate": 5.884711779448622e-06,
      "loss": 0.5506067276000977,
      "num_input_tokens_seen": 1075848,
      "step": 1414,
      "train_runtime": 3248.9497,
      "train_tokens_per_second": 331.137
    },
    {
      "epoch": 0.13011494252873562,
      "grad_norm": 7.135017395019531,
      "learning_rate": 5.87468671679198e-06,
      "loss": 0.5625655055046082,
      "num_input_tokens_seen": 1076852,
      "step": 1415,
      "train_runtime": 3250.9387,
      "train_tokens_per_second": 331.243
    },
    {
      "epoch": 0.13020689655172413,
      "grad_norm": 6.710904598236084,
      "learning_rate": 5.864661654135339e-06,
      "loss": 0.5015382766723633,
      "num_input_tokens_seen": 1077540,
      "step": 1416,
      "train_runtime": 3252.8839,
      "train_tokens_per_second": 331.257
    },
    {
      "epoch": 0.13029885057471263,
      "grad_norm": 7.155866622924805,
      "learning_rate": 5.854636591478697e-06,
      "loss": 0.4337519109249115,
      "num_input_tokens_seen": 1078220,
      "step": 1417,
      "train_runtime": 3254.8314,
      "train_tokens_per_second": 331.268
    },
    {
      "epoch": 0.13039080459770114,
      "grad_norm": 5.431839942932129,
      "learning_rate": 5.844611528822055e-06,
      "loss": 0.33701545000076294,
      "num_input_tokens_seen": 1079056,
      "step": 1418,
      "train_runtime": 3256.7776,
      "train_tokens_per_second": 331.326
    },
    {
      "epoch": 0.13048275862068964,
      "grad_norm": 5.556149005889893,
      "learning_rate": 5.8345864661654135e-06,
      "loss": 0.4353293180465698,
      "num_input_tokens_seen": 1080212,
      "step": 1419,
      "train_runtime": 3258.7947,
      "train_tokens_per_second": 331.476
    },
    {
      "epoch": 0.13057471264367815,
      "grad_norm": 6.294834136962891,
      "learning_rate": 5.824561403508773e-06,
      "loss": 0.4655035138130188,
      "num_input_tokens_seen": 1081108,
      "step": 1420,
      "train_runtime": 3260.7419,
      "train_tokens_per_second": 331.553
    },
    {
      "epoch": 0.13066666666666665,
      "grad_norm": 6.818014621734619,
      "learning_rate": 5.8145363408521305e-06,
      "loss": 0.48076024651527405,
      "num_input_tokens_seen": 1081940,
      "step": 1421,
      "train_runtime": 3262.6997,
      "train_tokens_per_second": 331.609
    },
    {
      "epoch": 0.1307586206896552,
      "grad_norm": 7.0069780349731445,
      "learning_rate": 5.804511278195489e-06,
      "loss": 0.5017577409744263,
      "num_input_tokens_seen": 1082984,
      "step": 1422,
      "train_runtime": 3264.6856,
      "train_tokens_per_second": 331.727
    },
    {
      "epoch": 0.1308505747126437,
      "grad_norm": 5.750135898590088,
      "learning_rate": 5.7944862155388475e-06,
      "loss": 0.4967845678329468,
      "num_input_tokens_seen": 1084196,
      "step": 1423,
      "train_runtime": 3266.7676,
      "train_tokens_per_second": 331.886
    },
    {
      "epoch": 0.1309425287356322,
      "grad_norm": 6.643673419952393,
      "learning_rate": 5.784461152882206e-06,
      "loss": 0.5117744207382202,
      "num_input_tokens_seen": 1084892,
      "step": 1424,
      "train_runtime": 3268.6914,
      "train_tokens_per_second": 331.904
    },
    {
      "epoch": 0.1310344827586207,
      "grad_norm": 6.438128471374512,
      "learning_rate": 5.7744360902255644e-06,
      "loss": 0.4591602385044098,
      "num_input_tokens_seen": 1085880,
      "step": 1425,
      "train_runtime": 3270.6994,
      "train_tokens_per_second": 332.002
    },
    {
      "epoch": 0.1311264367816092,
      "grad_norm": 6.030606269836426,
      "learning_rate": 5.764411027568922e-06,
      "loss": 0.3998528718948364,
      "num_input_tokens_seen": 1086744,
      "step": 1426,
      "train_runtime": 3272.6995,
      "train_tokens_per_second": 332.063
    },
    {
      "epoch": 0.1312183908045977,
      "grad_norm": 6.707212448120117,
      "learning_rate": 5.754385964912281e-06,
      "loss": 0.4361865520477295,
      "num_input_tokens_seen": 1087456,
      "step": 1427,
      "train_runtime": 3274.6613,
      "train_tokens_per_second": 332.082
    },
    {
      "epoch": 0.13131034482758622,
      "grad_norm": 7.165959358215332,
      "learning_rate": 5.744360902255639e-06,
      "loss": 0.4381793737411499,
      "num_input_tokens_seen": 1087996,
      "step": 1428,
      "train_runtime": 3276.6011,
      "train_tokens_per_second": 332.05
    },
    {
      "epoch": 0.13140229885057472,
      "grad_norm": 7.175246715545654,
      "learning_rate": 5.7343358395989975e-06,
      "loss": 0.5065217614173889,
      "num_input_tokens_seen": 1088672,
      "step": 1429,
      "train_runtime": 3278.5122,
      "train_tokens_per_second": 332.063
    },
    {
      "epoch": 0.13149425287356323,
      "grad_norm": 7.289319038391113,
      "learning_rate": 5.724310776942357e-06,
      "loss": 0.4486909508705139,
      "num_input_tokens_seen": 1089216,
      "step": 1430,
      "train_runtime": 3280.4188,
      "train_tokens_per_second": 332.036
    },
    {
      "epoch": 0.13158620689655173,
      "grad_norm": 6.761407375335693,
      "learning_rate": 5.7142857142857145e-06,
      "loss": 0.39181989431381226,
      "num_input_tokens_seen": 1089864,
      "step": 1431,
      "train_runtime": 3282.3515,
      "train_tokens_per_second": 332.038
    },
    {
      "epoch": 0.13167816091954024,
      "grad_norm": 7.691015243530273,
      "learning_rate": 5.704260651629073e-06,
      "loss": 0.5006154775619507,
      "num_input_tokens_seen": 1090452,
      "step": 1432,
      "train_runtime": 3284.2545,
      "train_tokens_per_second": 332.024
    },
    {
      "epoch": 0.13177011494252874,
      "grad_norm": 6.98045539855957,
      "learning_rate": 5.694235588972431e-06,
      "loss": 0.47522953152656555,
      "num_input_tokens_seen": 1091132,
      "step": 1433,
      "train_runtime": 3286.1701,
      "train_tokens_per_second": 332.038
    },
    {
      "epoch": 0.13186206896551725,
      "grad_norm": 6.046480655670166,
      "learning_rate": 5.68421052631579e-06,
      "loss": 0.4338725209236145,
      "num_input_tokens_seen": 1091984,
      "step": 1434,
      "train_runtime": 3288.1176,
      "train_tokens_per_second": 332.1
    },
    {
      "epoch": 0.13195402298850575,
      "grad_norm": 6.246063709259033,
      "learning_rate": 5.6741854636591485e-06,
      "loss": 0.5645648241043091,
      "num_input_tokens_seen": 1093008,
      "step": 1435,
      "train_runtime": 3290.1026,
      "train_tokens_per_second": 332.211
    },
    {
      "epoch": 0.13204597701149426,
      "grad_norm": 6.03352165222168,
      "learning_rate": 5.664160401002506e-06,
      "loss": 0.37914758920669556,
      "num_input_tokens_seen": 1093928,
      "step": 1436,
      "train_runtime": 3292.0846,
      "train_tokens_per_second": 332.29
    },
    {
      "epoch": 0.13213793103448276,
      "grad_norm": 4.073951721191406,
      "learning_rate": 5.6541353383458654e-06,
      "loss": 0.3868598937988281,
      "num_input_tokens_seen": 1095212,
      "step": 1437,
      "train_runtime": 3294.0926,
      "train_tokens_per_second": 332.478
    },
    {
      "epoch": 0.13222988505747127,
      "grad_norm": 6.356463432312012,
      "learning_rate": 5.644110275689223e-06,
      "loss": 0.3895990252494812,
      "num_input_tokens_seen": 1095828,
      "step": 1438,
      "train_runtime": 3296.0003,
      "train_tokens_per_second": 332.472
    },
    {
      "epoch": 0.13232183908045977,
      "grad_norm": 6.292166233062744,
      "learning_rate": 5.6340852130325816e-06,
      "loss": 0.3632347881793976,
      "num_input_tokens_seen": 1096436,
      "step": 1439,
      "train_runtime": 3297.8967,
      "train_tokens_per_second": 332.465
    },
    {
      "epoch": 0.13241379310344828,
      "grad_norm": 10.279585838317871,
      "learning_rate": 5.624060150375941e-06,
      "loss": 0.4475225806236267,
      "num_input_tokens_seen": 1097008,
      "step": 1440,
      "train_runtime": 3299.8007,
      "train_tokens_per_second": 332.447
    },
    {
      "epoch": 0.13250574712643678,
      "grad_norm": 6.838550090789795,
      "learning_rate": 5.6140350877192985e-06,
      "loss": 0.6017982959747314,
      "num_input_tokens_seen": 1098320,
      "step": 1441,
      "train_runtime": 3308.3526,
      "train_tokens_per_second": 331.984
    },
    {
      "epoch": 0.1325977011494253,
      "grad_norm": 6.356078147888184,
      "learning_rate": 5.604010025062657e-06,
      "loss": 0.39076799154281616,
      "num_input_tokens_seen": 1098960,
      "step": 1442,
      "train_runtime": 3310.2583,
      "train_tokens_per_second": 331.986
    },
    {
      "epoch": 0.1326896551724138,
      "grad_norm": 6.0813117027282715,
      "learning_rate": 5.593984962406015e-06,
      "loss": 0.4393022656440735,
      "num_input_tokens_seen": 1099672,
      "step": 1443,
      "train_runtime": 3312.1681,
      "train_tokens_per_second": 332.01
    },
    {
      "epoch": 0.1327816091954023,
      "grad_norm": 7.3292646408081055,
      "learning_rate": 5.583959899749374e-06,
      "loss": 0.48649272322654724,
      "num_input_tokens_seen": 1100352,
      "step": 1444,
      "train_runtime": 3314.0728,
      "train_tokens_per_second": 332.024
    },
    {
      "epoch": 0.1328735632183908,
      "grad_norm": 6.762223243713379,
      "learning_rate": 5.5739348370927325e-06,
      "loss": 0.48952534794807434,
      "num_input_tokens_seen": 1101152,
      "step": 1445,
      "train_runtime": 3316.011,
      "train_tokens_per_second": 332.071
    },
    {
      "epoch": 0.1329655172413793,
      "grad_norm": 6.347455024719238,
      "learning_rate": 5.56390977443609e-06,
      "loss": 0.44791457056999207,
      "num_input_tokens_seen": 1101800,
      "step": 1446,
      "train_runtime": 3317.9347,
      "train_tokens_per_second": 332.074
    },
    {
      "epoch": 0.13305747126436782,
      "grad_norm": 6.447908401489258,
      "learning_rate": 5.5538847117794495e-06,
      "loss": 0.4292008876800537,
      "num_input_tokens_seen": 1102528,
      "step": 1447,
      "train_runtime": 3319.8907,
      "train_tokens_per_second": 332.098
    },
    {
      "epoch": 0.13314942528735632,
      "grad_norm": 6.4803466796875,
      "learning_rate": 5.543859649122807e-06,
      "loss": 0.49495232105255127,
      "num_input_tokens_seen": 1103204,
      "step": 1448,
      "train_runtime": 3321.8047,
      "train_tokens_per_second": 332.11
    },
    {
      "epoch": 0.13324137931034483,
      "grad_norm": 6.167976379394531,
      "learning_rate": 5.533834586466166e-06,
      "loss": 0.415465772151947,
      "num_input_tokens_seen": 1103792,
      "step": 1449,
      "train_runtime": 3323.7181,
      "train_tokens_per_second": 332.096
    },
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 6.421348571777344,
      "learning_rate": 5.523809523809525e-06,
      "loss": 0.47656774520874023,
      "num_input_tokens_seen": 1104664,
      "step": 1450,
      "train_runtime": 3325.7162,
      "train_tokens_per_second": 332.158
    },
    {
      "epoch": 0.13342528735632184,
      "grad_norm": 6.786693572998047,
      "learning_rate": 5.5137844611528826e-06,
      "loss": 0.493723601102829,
      "num_input_tokens_seen": 1105940,
      "step": 1451,
      "train_runtime": 3327.7842,
      "train_tokens_per_second": 332.335
    },
    {
      "epoch": 0.13351724137931034,
      "grad_norm": 7.2511887550354,
      "learning_rate": 5.503759398496241e-06,
      "loss": 0.5468142628669739,
      "num_input_tokens_seen": 1106588,
      "step": 1452,
      "train_runtime": 3329.6941,
      "train_tokens_per_second": 332.339
    },
    {
      "epoch": 0.13360919540229885,
      "grad_norm": 7.076693534851074,
      "learning_rate": 5.493734335839599e-06,
      "loss": 0.4080525040626526,
      "num_input_tokens_seen": 1107224,
      "step": 1453,
      "train_runtime": 3331.6097,
      "train_tokens_per_second": 332.339
    },
    {
      "epoch": 0.13370114942528735,
      "grad_norm": 6.785010814666748,
      "learning_rate": 5.483709273182958e-06,
      "loss": 0.44463402032852173,
      "num_input_tokens_seen": 1108068,
      "step": 1454,
      "train_runtime": 3333.5742,
      "train_tokens_per_second": 332.396
    },
    {
      "epoch": 0.13379310344827586,
      "grad_norm": 8.230337142944336,
      "learning_rate": 5.4736842105263165e-06,
      "loss": 0.4688238203525543,
      "num_input_tokens_seen": 1108708,
      "step": 1455,
      "train_runtime": 3335.4783,
      "train_tokens_per_second": 332.399
    },
    {
      "epoch": 0.13388505747126436,
      "grad_norm": 7.023602485656738,
      "learning_rate": 5.463659147869674e-06,
      "loss": 0.4936160445213318,
      "num_input_tokens_seen": 1109428,
      "step": 1456,
      "train_runtime": 3337.3916,
      "train_tokens_per_second": 332.424
    },
    {
      "epoch": 0.13397701149425287,
      "grad_norm": 6.1114397048950195,
      "learning_rate": 5.4536340852130335e-06,
      "loss": 0.4492369294166565,
      "num_input_tokens_seen": 1110100,
      "step": 1457,
      "train_runtime": 3339.3358,
      "train_tokens_per_second": 332.431
    },
    {
      "epoch": 0.13406896551724137,
      "grad_norm": 6.5584492683410645,
      "learning_rate": 5.443609022556391e-06,
      "loss": 0.4723778963088989,
      "num_input_tokens_seen": 1111172,
      "step": 1458,
      "train_runtime": 3341.312,
      "train_tokens_per_second": 332.556
    },
    {
      "epoch": 0.13416091954022988,
      "grad_norm": 7.448948383331299,
      "learning_rate": 5.43358395989975e-06,
      "loss": 0.5321516990661621,
      "num_input_tokens_seen": 1111804,
      "step": 1459,
      "train_runtime": 3343.2311,
      "train_tokens_per_second": 332.554
    },
    {
      "epoch": 0.13425287356321838,
      "grad_norm": 5.970415115356445,
      "learning_rate": 5.423558897243109e-06,
      "loss": 0.43350765109062195,
      "num_input_tokens_seen": 1112456,
      "step": 1460,
      "train_runtime": 3345.1541,
      "train_tokens_per_second": 332.557
    },
    {
      "epoch": 0.1343448275862069,
      "grad_norm": 7.31071138381958,
      "learning_rate": 5.413533834586467e-06,
      "loss": 0.5078038573265076,
      "num_input_tokens_seen": 1113180,
      "step": 1461,
      "train_runtime": 3347.0855,
      "train_tokens_per_second": 332.582
    },
    {
      "epoch": 0.1344367816091954,
      "grad_norm": 7.171257019042969,
      "learning_rate": 5.403508771929825e-06,
      "loss": 0.43725284934043884,
      "num_input_tokens_seen": 1114100,
      "step": 1462,
      "train_runtime": 3349.0351,
      "train_tokens_per_second": 332.663
    },
    {
      "epoch": 0.1345287356321839,
      "grad_norm": 6.14492130279541,
      "learning_rate": 5.393483709273183e-06,
      "loss": 0.4269268214702606,
      "num_input_tokens_seen": 1114704,
      "step": 1463,
      "train_runtime": 3350.9637,
      "train_tokens_per_second": 332.652
    },
    {
      "epoch": 0.1346206896551724,
      "grad_norm": 8.483597755432129,
      "learning_rate": 5.383458646616542e-06,
      "loss": 0.4732326567173004,
      "num_input_tokens_seen": 1115376,
      "step": 1464,
      "train_runtime": 3352.8884,
      "train_tokens_per_second": 332.661
    },
    {
      "epoch": 0.1347126436781609,
      "grad_norm": 6.515466213226318,
      "learning_rate": 5.3734335839599006e-06,
      "loss": 0.5863065719604492,
      "num_input_tokens_seen": 1116764,
      "step": 1465,
      "train_runtime": 3354.9209,
      "train_tokens_per_second": 332.873
    },
    {
      "epoch": 0.13480459770114941,
      "grad_norm": 6.285518169403076,
      "learning_rate": 5.363408521303258e-06,
      "loss": 0.4715114235877991,
      "num_input_tokens_seen": 1117456,
      "step": 1466,
      "train_runtime": 3356.9176,
      "train_tokens_per_second": 332.882
    },
    {
      "epoch": 0.13489655172413792,
      "grad_norm": 6.65487003326416,
      "learning_rate": 5.3533834586466175e-06,
      "loss": 0.4476355314254761,
      "num_input_tokens_seen": 1118176,
      "step": 1467,
      "train_runtime": 3358.8734,
      "train_tokens_per_second": 332.902
    },
    {
      "epoch": 0.13498850574712642,
      "grad_norm": 6.075029373168945,
      "learning_rate": 5.343358395989975e-06,
      "loss": 0.44196251034736633,
      "num_input_tokens_seen": 1118932,
      "step": 1468,
      "train_runtime": 3360.7859,
      "train_tokens_per_second": 332.938
    },
    {
      "epoch": 0.13508045977011493,
      "grad_norm": 6.945167064666748,
      "learning_rate": 5.333333333333334e-06,
      "loss": 0.47767823934555054,
      "num_input_tokens_seen": 1119636,
      "step": 1469,
      "train_runtime": 3362.738,
      "train_tokens_per_second": 332.954
    },
    {
      "epoch": 0.13517241379310344,
      "grad_norm": 6.25655460357666,
      "learning_rate": 5.323308270676692e-06,
      "loss": 0.4757128655910492,
      "num_input_tokens_seen": 1120972,
      "step": 1470,
      "train_runtime": 3364.7111,
      "train_tokens_per_second": 333.156
    },
    {
      "epoch": 0.13526436781609194,
      "grad_norm": 6.348459720611572,
      "learning_rate": 5.313283208020051e-06,
      "loss": 0.3730282783508301,
      "num_input_tokens_seen": 1121584,
      "step": 1471,
      "train_runtime": 3372.9578,
      "train_tokens_per_second": 332.522
    },
    {
      "epoch": 0.13535632183908047,
      "grad_norm": 7.8312296867370605,
      "learning_rate": 5.303258145363409e-06,
      "loss": 0.5653750896453857,
      "num_input_tokens_seen": 1122280,
      "step": 1472,
      "train_runtime": 3374.8739,
      "train_tokens_per_second": 332.54
    },
    {
      "epoch": 0.13544827586206898,
      "grad_norm": 6.325518608093262,
      "learning_rate": 5.293233082706767e-06,
      "loss": 0.4348008930683136,
      "num_input_tokens_seen": 1122924,
      "step": 1473,
      "train_runtime": 3376.858,
      "train_tokens_per_second": 332.535
    },
    {
      "epoch": 0.13554022988505748,
      "grad_norm": 5.947680473327637,
      "learning_rate": 5.283208020050126e-06,
      "loss": 0.4743146002292633,
      "num_input_tokens_seen": 1124464,
      "step": 1474,
      "train_runtime": 3378.9075,
      "train_tokens_per_second": 332.789
    },
    {
      "epoch": 0.135632183908046,
      "grad_norm": 6.109054088592529,
      "learning_rate": 5.273182957393485e-06,
      "loss": 0.40196356177330017,
      "num_input_tokens_seen": 1124984,
      "step": 1475,
      "train_runtime": 3380.8153,
      "train_tokens_per_second": 332.755
    },
    {
      "epoch": 0.1357241379310345,
      "grad_norm": 6.539097309112549,
      "learning_rate": 5.263157894736842e-06,
      "loss": 0.4703242778778076,
      "num_input_tokens_seen": 1126252,
      "step": 1476,
      "train_runtime": 3382.8059,
      "train_tokens_per_second": 332.934
    },
    {
      "epoch": 0.135816091954023,
      "grad_norm": 6.725227355957031,
      "learning_rate": 5.253132832080201e-06,
      "loss": 0.4537864923477173,
      "num_input_tokens_seen": 1126984,
      "step": 1477,
      "train_runtime": 3384.8574,
      "train_tokens_per_second": 332.949
    },
    {
      "epoch": 0.1359080459770115,
      "grad_norm": 7.197235107421875,
      "learning_rate": 5.243107769423559e-06,
      "loss": 0.5095236301422119,
      "num_input_tokens_seen": 1127828,
      "step": 1478,
      "train_runtime": 3386.8172,
      "train_tokens_per_second": 333.005
    },
    {
      "epoch": 0.136,
      "grad_norm": 7.300252437591553,
      "learning_rate": 5.233082706766918e-06,
      "loss": 0.47199469804763794,
      "num_input_tokens_seen": 1128640,
      "step": 1479,
      "train_runtime": 3388.8044,
      "train_tokens_per_second": 333.05
    },
    {
      "epoch": 0.13609195402298851,
      "grad_norm": 6.9750471115112305,
      "learning_rate": 5.223057644110276e-06,
      "loss": 0.42595043778419495,
      "num_input_tokens_seen": 1129392,
      "step": 1480,
      "train_runtime": 3390.7721,
      "train_tokens_per_second": 333.078
    },
    {
      "epoch": 0.13618390804597702,
      "grad_norm": 5.77484130859375,
      "learning_rate": 5.213032581453634e-06,
      "loss": 0.47980445623397827,
      "num_input_tokens_seen": 1130524,
      "step": 1481,
      "train_runtime": 3392.7646,
      "train_tokens_per_second": 333.216
    },
    {
      "epoch": 0.13627586206896553,
      "grad_norm": 6.600121974945068,
      "learning_rate": 5.203007518796993e-06,
      "loss": 0.48391154408454895,
      "num_input_tokens_seen": 1131376,
      "step": 1482,
      "train_runtime": 3394.7547,
      "train_tokens_per_second": 333.272
    },
    {
      "epoch": 0.13636781609195403,
      "grad_norm": 6.862207412719727,
      "learning_rate": 5.192982456140351e-06,
      "loss": 0.5035004019737244,
      "num_input_tokens_seen": 1132288,
      "step": 1483,
      "train_runtime": 3396.7076,
      "train_tokens_per_second": 333.349
    },
    {
      "epoch": 0.13645977011494254,
      "grad_norm": 6.580876350402832,
      "learning_rate": 5.182957393483709e-06,
      "loss": 0.37613195180892944,
      "num_input_tokens_seen": 1132996,
      "step": 1484,
      "train_runtime": 3398.6231,
      "train_tokens_per_second": 333.369
    },
    {
      "epoch": 0.13655172413793104,
      "grad_norm": 6.477765083312988,
      "learning_rate": 5.172932330827069e-06,
      "loss": 0.42193272709846497,
      "num_input_tokens_seen": 1133580,
      "step": 1485,
      "train_runtime": 3400.5249,
      "train_tokens_per_second": 333.354
    },
    {
      "epoch": 0.13664367816091955,
      "grad_norm": 9.42302417755127,
      "learning_rate": 5.162907268170426e-06,
      "loss": 0.447000116109848,
      "num_input_tokens_seen": 1134152,
      "step": 1486,
      "train_runtime": 3402.432,
      "train_tokens_per_second": 333.336
    },
    {
      "epoch": 0.13673563218390805,
      "grad_norm": 7.284692764282227,
      "learning_rate": 5.152882205513785e-06,
      "loss": 0.5637538433074951,
      "num_input_tokens_seen": 1134840,
      "step": 1487,
      "train_runtime": 3404.3471,
      "train_tokens_per_second": 333.35
    },
    {
      "epoch": 0.13682758620689656,
      "grad_norm": 6.81699275970459,
      "learning_rate": 5.142857142857142e-06,
      "loss": 0.4707770347595215,
      "num_input_tokens_seen": 1135780,
      "step": 1488,
      "train_runtime": 3406.3034,
      "train_tokens_per_second": 333.435
    },
    {
      "epoch": 0.13691954022988506,
      "grad_norm": 7.516037940979004,
      "learning_rate": 5.132832080200502e-06,
      "loss": 0.5666313171386719,
      "num_input_tokens_seen": 1136524,
      "step": 1489,
      "train_runtime": 3408.3556,
      "train_tokens_per_second": 333.452
    },
    {
      "epoch": 0.13701149425287357,
      "grad_norm": 7.408823013305664,
      "learning_rate": 5.12280701754386e-06,
      "loss": 0.5359535217285156,
      "num_input_tokens_seen": 1137472,
      "step": 1490,
      "train_runtime": 3410.3086,
      "train_tokens_per_second": 333.539
    },
    {
      "epoch": 0.13710344827586207,
      "grad_norm": 7.798243045806885,
      "learning_rate": 5.112781954887218e-06,
      "loss": 0.5340937972068787,
      "num_input_tokens_seen": 1138068,
      "step": 1491,
      "train_runtime": 3412.2167,
      "train_tokens_per_second": 333.527
    },
    {
      "epoch": 0.13719540229885058,
      "grad_norm": 8.12541675567627,
      "learning_rate": 5.102756892230577e-06,
      "loss": 0.5569170713424683,
      "num_input_tokens_seen": 1138892,
      "step": 1492,
      "train_runtime": 3414.147,
      "train_tokens_per_second": 333.58
    },
    {
      "epoch": 0.13728735632183908,
      "grad_norm": 6.8189849853515625,
      "learning_rate": 5.092731829573935e-06,
      "loss": 0.5268915295600891,
      "num_input_tokens_seen": 1139696,
      "step": 1493,
      "train_runtime": 3416.1279,
      "train_tokens_per_second": 333.622
    },
    {
      "epoch": 0.1373793103448276,
      "grad_norm": 5.879248142242432,
      "learning_rate": 5.082706766917293e-06,
      "loss": 0.37678250670433044,
      "num_input_tokens_seen": 1140444,
      "step": 1494,
      "train_runtime": 3418.0696,
      "train_tokens_per_second": 333.651
    },
    {
      "epoch": 0.1374712643678161,
      "grad_norm": 6.807672500610352,
      "learning_rate": 5.072681704260653e-06,
      "loss": 0.5095071196556091,
      "num_input_tokens_seen": 1141212,
      "step": 1495,
      "train_runtime": 3419.9883,
      "train_tokens_per_second": 333.689
    },
    {
      "epoch": 0.1375632183908046,
      "grad_norm": 6.6341962814331055,
      "learning_rate": 5.06265664160401e-06,
      "loss": 0.5022022724151611,
      "num_input_tokens_seen": 1141796,
      "step": 1496,
      "train_runtime": 3421.9702,
      "train_tokens_per_second": 333.666
    },
    {
      "epoch": 0.1376551724137931,
      "grad_norm": 6.8143768310546875,
      "learning_rate": 5.052631578947369e-06,
      "loss": 0.41124725341796875,
      "num_input_tokens_seen": 1142496,
      "step": 1497,
      "train_runtime": 3423.8907,
      "train_tokens_per_second": 333.684
    },
    {
      "epoch": 0.1377471264367816,
      "grad_norm": 6.307721138000488,
      "learning_rate": 5.042606516290726e-06,
      "loss": 0.3924211263656616,
      "num_input_tokens_seen": 1143152,
      "step": 1498,
      "train_runtime": 3425.8133,
      "train_tokens_per_second": 333.688
    },
    {
      "epoch": 0.1378390804597701,
      "grad_norm": 6.788408279418945,
      "learning_rate": 5.032581453634086e-06,
      "loss": 0.3497253656387329,
      "num_input_tokens_seen": 1143712,
      "step": 1499,
      "train_runtime": 3427.7682,
      "train_tokens_per_second": 333.661
    },
    {
      "epoch": 0.13793103448275862,
      "grad_norm": 6.1640729904174805,
      "learning_rate": 5.022556390977444e-06,
      "loss": 0.4776795208454132,
      "num_input_tokens_seen": 1144672,
      "step": 1500,
      "train_runtime": 3429.7151,
      "train_tokens_per_second": 333.751
    },
    {
      "epoch": 0.13802298850574712,
      "grad_norm": 5.438903331756592,
      "learning_rate": 5.012531328320802e-06,
      "loss": 0.38507819175720215,
      "num_input_tokens_seen": 1146308,
      "step": 1501,
      "train_runtime": 3438.626,
      "train_tokens_per_second": 333.362
    },
    {
      "epoch": 0.13811494252873563,
      "grad_norm": 6.643701076507568,
      "learning_rate": 5.002506265664161e-06,
      "loss": 0.38595324754714966,
      "num_input_tokens_seen": 1146896,
      "step": 1502,
      "train_runtime": 3440.7643,
      "train_tokens_per_second": 333.326
    },
    {
      "epoch": 0.13820689655172413,
      "grad_norm": 7.118875980377197,
      "learning_rate": 4.992481203007519e-06,
      "loss": 0.6216070652008057,
      "num_input_tokens_seen": 1147648,
      "step": 1503,
      "train_runtime": 3442.6889,
      "train_tokens_per_second": 333.358
    },
    {
      "epoch": 0.13829885057471264,
      "grad_norm": 7.120995044708252,
      "learning_rate": 4.982456140350877e-06,
      "loss": 0.4311849772930145,
      "num_input_tokens_seen": 1148372,
      "step": 1504,
      "train_runtime": 3444.6044,
      "train_tokens_per_second": 333.383
    },
    {
      "epoch": 0.13839080459770114,
      "grad_norm": 5.9141764640808105,
      "learning_rate": 4.972431077694236e-06,
      "loss": 0.44043609499931335,
      "num_input_tokens_seen": 1149256,
      "step": 1505,
      "train_runtime": 3446.554,
      "train_tokens_per_second": 333.451
    },
    {
      "epoch": 0.13848275862068965,
      "grad_norm": 7.327331066131592,
      "learning_rate": 4.962406015037594e-06,
      "loss": 0.49830931425094604,
      "num_input_tokens_seen": 1150156,
      "step": 1506,
      "train_runtime": 3448.5009,
      "train_tokens_per_second": 333.523
    },
    {
      "epoch": 0.13857471264367816,
      "grad_norm": 6.454385757446289,
      "learning_rate": 4.952380952380953e-06,
      "loss": 0.4663582742214203,
      "num_input_tokens_seen": 1150828,
      "step": 1507,
      "train_runtime": 3450.4103,
      "train_tokens_per_second": 333.534
    },
    {
      "epoch": 0.13866666666666666,
      "grad_norm": 11.470450401306152,
      "learning_rate": 4.942355889724311e-06,
      "loss": 0.48851844668388367,
      "num_input_tokens_seen": 1151480,
      "step": 1508,
      "train_runtime": 3452.3403,
      "train_tokens_per_second": 333.536
    },
    {
      "epoch": 0.13875862068965517,
      "grad_norm": 6.839486122131348,
      "learning_rate": 4.93233082706767e-06,
      "loss": 0.493771493434906,
      "num_input_tokens_seen": 1152428,
      "step": 1509,
      "train_runtime": 3454.3209,
      "train_tokens_per_second": 333.619
    },
    {
      "epoch": 0.13885057471264367,
      "grad_norm": 7.617315292358398,
      "learning_rate": 4.922305764411027e-06,
      "loss": 0.5075865387916565,
      "num_input_tokens_seen": 1153192,
      "step": 1510,
      "train_runtime": 3456.2342,
      "train_tokens_per_second": 333.656
    },
    {
      "epoch": 0.13894252873563218,
      "grad_norm": 6.4050517082214355,
      "learning_rate": 4.912280701754386e-06,
      "loss": 0.5061872005462646,
      "num_input_tokens_seen": 1154952,
      "step": 1511,
      "train_runtime": 3458.3129,
      "train_tokens_per_second": 333.964
    },
    {
      "epoch": 0.13903448275862068,
      "grad_norm": 7.359883785247803,
      "learning_rate": 4.902255639097745e-06,
      "loss": 0.5475380420684814,
      "num_input_tokens_seen": 1156260,
      "step": 1512,
      "train_runtime": 3460.3238,
      "train_tokens_per_second": 334.148
    },
    {
      "epoch": 0.1391264367816092,
      "grad_norm": 7.349669933319092,
      "learning_rate": 4.892230576441103e-06,
      "loss": 0.525090217590332,
      "num_input_tokens_seen": 1156964,
      "step": 1513,
      "train_runtime": 3462.2457,
      "train_tokens_per_second": 334.166
    },
    {
      "epoch": 0.1392183908045977,
      "grad_norm": 7.001368045806885,
      "learning_rate": 4.882205513784461e-06,
      "loss": 0.4986392855644226,
      "num_input_tokens_seen": 1157840,
      "step": 1514,
      "train_runtime": 3464.2485,
      "train_tokens_per_second": 334.225
    },
    {
      "epoch": 0.1393103448275862,
      "grad_norm": 5.995524883270264,
      "learning_rate": 4.87218045112782e-06,
      "loss": 0.4127148389816284,
      "num_input_tokens_seen": 1158400,
      "step": 1515,
      "train_runtime": 3466.1629,
      "train_tokens_per_second": 334.202
    },
    {
      "epoch": 0.1394022988505747,
      "grad_norm": 6.9042816162109375,
      "learning_rate": 4.862155388471178e-06,
      "loss": 0.4874632656574249,
      "num_input_tokens_seen": 1159196,
      "step": 1516,
      "train_runtime": 3468.0995,
      "train_tokens_per_second": 334.245
    },
    {
      "epoch": 0.1394942528735632,
      "grad_norm": 7.804703235626221,
      "learning_rate": 4.852130325814537e-06,
      "loss": 0.6120208501815796,
      "num_input_tokens_seen": 1159880,
      "step": 1517,
      "train_runtime": 3470.0093,
      "train_tokens_per_second": 334.258
    },
    {
      "epoch": 0.1395862068965517,
      "grad_norm": 7.079176902770996,
      "learning_rate": 4.842105263157895e-06,
      "loss": 0.4638254940509796,
      "num_input_tokens_seen": 1160568,
      "step": 1518,
      "train_runtime": 3471.9284,
      "train_tokens_per_second": 334.272
    },
    {
      "epoch": 0.13967816091954022,
      "grad_norm": 6.242354393005371,
      "learning_rate": 4.832080200501254e-06,
      "loss": 0.3685902953147888,
      "num_input_tokens_seen": 1161184,
      "step": 1519,
      "train_runtime": 3473.8435,
      "train_tokens_per_second": 334.265
    },
    {
      "epoch": 0.13977011494252872,
      "grad_norm": 6.551645755767822,
      "learning_rate": 4.8220551378446114e-06,
      "loss": 0.55516517162323,
      "num_input_tokens_seen": 1162596,
      "step": 1520,
      "train_runtime": 3475.879,
      "train_tokens_per_second": 334.475
    },
    {
      "epoch": 0.13986206896551723,
      "grad_norm": 7.7210001945495605,
      "learning_rate": 4.81203007518797e-06,
      "loss": 0.4628271758556366,
      "num_input_tokens_seen": 1163356,
      "step": 1521,
      "train_runtime": 3477.8351,
      "train_tokens_per_second": 334.506
    },
    {
      "epoch": 0.13995402298850576,
      "grad_norm": 6.228600025177002,
      "learning_rate": 4.802005012531329e-06,
      "loss": 0.4215913414955139,
      "num_input_tokens_seen": 1164228,
      "step": 1522,
      "train_runtime": 3479.7951,
      "train_tokens_per_second": 334.568
    },
    {
      "epoch": 0.14004597701149427,
      "grad_norm": 7.540292739868164,
      "learning_rate": 4.791979949874687e-06,
      "loss": 0.5465987920761108,
      "num_input_tokens_seen": 1164936,
      "step": 1523,
      "train_runtime": 3481.7335,
      "train_tokens_per_second": 334.585
    },
    {
      "epoch": 0.14013793103448277,
      "grad_norm": 6.092329025268555,
      "learning_rate": 4.781954887218045e-06,
      "loss": 0.49597764015197754,
      "num_input_tokens_seen": 1165520,
      "step": 1524,
      "train_runtime": 3483.6633,
      "train_tokens_per_second": 334.567
    },
    {
      "epoch": 0.14022988505747128,
      "grad_norm": 6.8615007400512695,
      "learning_rate": 4.771929824561404e-06,
      "loss": 0.42225661873817444,
      "num_input_tokens_seen": 1166100,
      "step": 1525,
      "train_runtime": 3485.6299,
      "train_tokens_per_second": 334.545
    },
    {
      "epoch": 0.14032183908045978,
      "grad_norm": 6.486217975616455,
      "learning_rate": 4.761904761904762e-06,
      "loss": 0.5414251685142517,
      "num_input_tokens_seen": 1166908,
      "step": 1526,
      "train_runtime": 3487.5937,
      "train_tokens_per_second": 334.588
    },
    {
      "epoch": 0.1404137931034483,
      "grad_norm": 7.315220355987549,
      "learning_rate": 4.751879699248121e-06,
      "loss": 0.44345003366470337,
      "num_input_tokens_seen": 1167528,
      "step": 1527,
      "train_runtime": 3489.5024,
      "train_tokens_per_second": 334.583
    },
    {
      "epoch": 0.1405057471264368,
      "grad_norm": 6.64310884475708,
      "learning_rate": 4.741854636591479e-06,
      "loss": 0.42874985933303833,
      "num_input_tokens_seen": 1168108,
      "step": 1528,
      "train_runtime": 3491.4161,
      "train_tokens_per_second": 334.566
    },
    {
      "epoch": 0.1405977011494253,
      "grad_norm": 6.159165859222412,
      "learning_rate": 4.731829573934838e-06,
      "loss": 0.47418278455734253,
      "num_input_tokens_seen": 1168956,
      "step": 1529,
      "train_runtime": 3493.3878,
      "train_tokens_per_second": 334.62
    },
    {
      "epoch": 0.1406896551724138,
      "grad_norm": 6.183302879333496,
      "learning_rate": 4.7218045112781955e-06,
      "loss": 0.41782593727111816,
      "num_input_tokens_seen": 1169608,
      "step": 1530,
      "train_runtime": 3495.3093,
      "train_tokens_per_second": 334.622
    },
    {
      "epoch": 0.1407816091954023,
      "grad_norm": 7.021867275238037,
      "learning_rate": 4.711779448621554e-06,
      "loss": 0.5794481039047241,
      "num_input_tokens_seen": 1170272,
      "step": 1531,
      "train_runtime": 3503.8917,
      "train_tokens_per_second": 333.992
    },
    {
      "epoch": 0.1408735632183908,
      "grad_norm": 5.9288249015808105,
      "learning_rate": 4.7017543859649125e-06,
      "loss": 0.3630669414997101,
      "num_input_tokens_seen": 1170860,
      "step": 1532,
      "train_runtime": 3505.8235,
      "train_tokens_per_second": 333.976
    },
    {
      "epoch": 0.14096551724137932,
      "grad_norm": 7.3672075271606445,
      "learning_rate": 4.691729323308271e-06,
      "loss": 0.44522491097450256,
      "num_input_tokens_seen": 1171480,
      "step": 1533,
      "train_runtime": 3507.8212,
      "train_tokens_per_second": 333.962
    },
    {
      "epoch": 0.14105747126436782,
      "grad_norm": 6.494082927703857,
      "learning_rate": 4.6817042606516294e-06,
      "loss": 0.4364702105522156,
      "num_input_tokens_seen": 1172136,
      "step": 1534,
      "train_runtime": 3509.7465,
      "train_tokens_per_second": 333.966
    },
    {
      "epoch": 0.14114942528735633,
      "grad_norm": 6.911060810089111,
      "learning_rate": 4.671679197994988e-06,
      "loss": 0.3699270784854889,
      "num_input_tokens_seen": 1172648,
      "step": 1535,
      "train_runtime": 3511.6599,
      "train_tokens_per_second": 333.93
    },
    {
      "epoch": 0.14124137931034483,
      "grad_norm": 6.150845050811768,
      "learning_rate": 4.661654135338346e-06,
      "loss": 0.4344079792499542,
      "num_input_tokens_seen": 1173408,
      "step": 1536,
      "train_runtime": 3513.6082,
      "train_tokens_per_second": 333.961
    },
    {
      "epoch": 0.14133333333333334,
      "grad_norm": 6.985192775726318,
      "learning_rate": 4.651629072681705e-06,
      "loss": 0.4886764883995056,
      "num_input_tokens_seen": 1174080,
      "step": 1537,
      "train_runtime": 3515.5202,
      "train_tokens_per_second": 333.97
    },
    {
      "epoch": 0.14142528735632184,
      "grad_norm": 6.360936641693115,
      "learning_rate": 4.641604010025063e-06,
      "loss": 0.43858852982521057,
      "num_input_tokens_seen": 1175012,
      "step": 1538,
      "train_runtime": 3517.4758,
      "train_tokens_per_second": 334.05
    },
    {
      "epoch": 0.14151724137931035,
      "grad_norm": 7.028363227844238,
      "learning_rate": 4.631578947368421e-06,
      "loss": 0.4857777953147888,
      "num_input_tokens_seen": 1175724,
      "step": 1539,
      "train_runtime": 3519.4121,
      "train_tokens_per_second": 334.068
    },
    {
      "epoch": 0.14160919540229885,
      "grad_norm": 7.59266471862793,
      "learning_rate": 4.6215538847117795e-06,
      "loss": 0.4337865114212036,
      "num_input_tokens_seen": 1176348,
      "step": 1540,
      "train_runtime": 3521.3568,
      "train_tokens_per_second": 334.061
    },
    {
      "epoch": 0.14170114942528736,
      "grad_norm": 5.869190216064453,
      "learning_rate": 4.611528822055138e-06,
      "loss": 0.39313578605651855,
      "num_input_tokens_seen": 1177224,
      "step": 1541,
      "train_runtime": 3523.3102,
      "train_tokens_per_second": 334.124
    },
    {
      "epoch": 0.14179310344827586,
      "grad_norm": 7.339359283447266,
      "learning_rate": 4.6015037593984965e-06,
      "loss": 0.4942663908004761,
      "num_input_tokens_seen": 1177920,
      "step": 1542,
      "train_runtime": 3525.2401,
      "train_tokens_per_second": 334.139
    },
    {
      "epoch": 0.14188505747126437,
      "grad_norm": 7.112597942352295,
      "learning_rate": 4.591478696741855e-06,
      "loss": 0.5195597410202026,
      "num_input_tokens_seen": 1178672,
      "step": 1543,
      "train_runtime": 3527.1959,
      "train_tokens_per_second": 334.167
    },
    {
      "epoch": 0.14197701149425287,
      "grad_norm": 7.303198337554932,
      "learning_rate": 4.5814536340852135e-06,
      "loss": 0.5885412096977234,
      "num_input_tokens_seen": 1179356,
      "step": 1544,
      "train_runtime": 3529.1157,
      "train_tokens_per_second": 334.179
    },
    {
      "epoch": 0.14206896551724138,
      "grad_norm": 7.587347030639648,
      "learning_rate": 4.571428571428572e-06,
      "loss": 0.525458037853241,
      "num_input_tokens_seen": 1180028,
      "step": 1545,
      "train_runtime": 3531.0347,
      "train_tokens_per_second": 334.188
    },
    {
      "epoch": 0.14216091954022989,
      "grad_norm": 12.200189590454102,
      "learning_rate": 4.56140350877193e-06,
      "loss": 0.5025969743728638,
      "num_input_tokens_seen": 1180812,
      "step": 1546,
      "train_runtime": 3532.9971,
      "train_tokens_per_second": 334.224
    },
    {
      "epoch": 0.1422528735632184,
      "grad_norm": 7.2503204345703125,
      "learning_rate": 4.551378446115289e-06,
      "loss": 0.5418580770492554,
      "num_input_tokens_seen": 1181436,
      "step": 1547,
      "train_runtime": 3534.9114,
      "train_tokens_per_second": 334.219
    },
    {
      "epoch": 0.1423448275862069,
      "grad_norm": 7.234675407409668,
      "learning_rate": 4.541353383458647e-06,
      "loss": 0.5090577006340027,
      "num_input_tokens_seen": 1182200,
      "step": 1548,
      "train_runtime": 3536.8327,
      "train_tokens_per_second": 334.254
    },
    {
      "epoch": 0.1424367816091954,
      "grad_norm": 6.909221172332764,
      "learning_rate": 4.531328320802005e-06,
      "loss": 0.49349039793014526,
      "num_input_tokens_seen": 1182924,
      "step": 1549,
      "train_runtime": 3538.7775,
      "train_tokens_per_second": 334.275
    },
    {
      "epoch": 0.1425287356321839,
      "grad_norm": 6.639964580535889,
      "learning_rate": 4.5213032581453635e-06,
      "loss": 0.4774690270423889,
      "num_input_tokens_seen": 1183804,
      "step": 1550,
      "train_runtime": 3540.7258,
      "train_tokens_per_second": 334.339
    },
    {
      "epoch": 0.1426206896551724,
      "grad_norm": 6.520241737365723,
      "learning_rate": 4.511278195488722e-06,
      "loss": 0.5106967091560364,
      "num_input_tokens_seen": 1184504,
      "step": 1551,
      "train_runtime": 3542.6401,
      "train_tokens_per_second": 334.356
    },
    {
      "epoch": 0.14271264367816092,
      "grad_norm": 7.22974967956543,
      "learning_rate": 4.5012531328320805e-06,
      "loss": 0.4782262444496155,
      "num_input_tokens_seen": 1185184,
      "step": 1552,
      "train_runtime": 3544.56,
      "train_tokens_per_second": 334.367
    },
    {
      "epoch": 0.14280459770114942,
      "grad_norm": 7.263352870941162,
      "learning_rate": 4.491228070175439e-06,
      "loss": 0.4990386664867401,
      "num_input_tokens_seen": 1185924,
      "step": 1553,
      "train_runtime": 3546.4945,
      "train_tokens_per_second": 334.393
    },
    {
      "epoch": 0.14289655172413793,
      "grad_norm": 6.859169960021973,
      "learning_rate": 4.4812030075187975e-06,
      "loss": 0.4410400390625,
      "num_input_tokens_seen": 1186652,
      "step": 1554,
      "train_runtime": 3548.4061,
      "train_tokens_per_second": 334.418
    },
    {
      "epoch": 0.14298850574712643,
      "grad_norm": 7.173482894897461,
      "learning_rate": 4.471177944862156e-06,
      "loss": 0.4671162962913513,
      "num_input_tokens_seen": 1187328,
      "step": 1555,
      "train_runtime": 3550.3659,
      "train_tokens_per_second": 334.424
    },
    {
      "epoch": 0.14308045977011494,
      "grad_norm": 7.271553039550781,
      "learning_rate": 4.461152882205514e-06,
      "loss": 0.43279311060905457,
      "num_input_tokens_seen": 1188452,
      "step": 1556,
      "train_runtime": 3552.3551,
      "train_tokens_per_second": 334.553
    },
    {
      "epoch": 0.14317241379310344,
      "grad_norm": 6.336573123931885,
      "learning_rate": 4.451127819548873e-06,
      "loss": 0.3423529863357544,
      "num_input_tokens_seen": 1189056,
      "step": 1557,
      "train_runtime": 3554.2645,
      "train_tokens_per_second": 334.543
    },
    {
      "epoch": 0.14326436781609195,
      "grad_norm": 6.986931800842285,
      "learning_rate": 4.4411027568922314e-06,
      "loss": 0.5558561086654663,
      "num_input_tokens_seen": 1190140,
      "step": 1558,
      "train_runtime": 3556.2356,
      "train_tokens_per_second": 334.663
    },
    {
      "epoch": 0.14335632183908045,
      "grad_norm": 7.816788673400879,
      "learning_rate": 4.431077694235589e-06,
      "loss": 0.44449564814567566,
      "num_input_tokens_seen": 1190752,
      "step": 1559,
      "train_runtime": 3558.1463,
      "train_tokens_per_second": 334.655
    },
    {
      "epoch": 0.14344827586206896,
      "grad_norm": 8.101500511169434,
      "learning_rate": 4.4210526315789476e-06,
      "loss": 0.5737946033477783,
      "num_input_tokens_seen": 1191564,
      "step": 1560,
      "train_runtime": 3560.1087,
      "train_tokens_per_second": 334.699
    },
    {
      "epoch": 0.14354022988505746,
      "grad_norm": 9.103196144104004,
      "learning_rate": 4.411027568922306e-06,
      "loss": 0.5782645344734192,
      "num_input_tokens_seen": 1192240,
      "step": 1561,
      "train_runtime": 3568.6313,
      "train_tokens_per_second": 334.089
    },
    {
      "epoch": 0.14363218390804597,
      "grad_norm": 6.655757427215576,
      "learning_rate": 4.4010025062656645e-06,
      "loss": 0.467924028635025,
      "num_input_tokens_seen": 1192904,
      "step": 1562,
      "train_runtime": 3570.6484,
      "train_tokens_per_second": 334.086
    },
    {
      "epoch": 0.14372413793103447,
      "grad_norm": 5.992223739624023,
      "learning_rate": 4.390977443609023e-06,
      "loss": 0.45279771089553833,
      "num_input_tokens_seen": 1193492,
      "step": 1563,
      "train_runtime": 3572.5868,
      "train_tokens_per_second": 334.069
    },
    {
      "epoch": 0.14381609195402298,
      "grad_norm": 6.550340175628662,
      "learning_rate": 4.3809523809523815e-06,
      "loss": 0.431035578250885,
      "num_input_tokens_seen": 1194256,
      "step": 1564,
      "train_runtime": 3574.5055,
      "train_tokens_per_second": 334.104
    },
    {
      "epoch": 0.14390804597701148,
      "grad_norm": 7.314089775085449,
      "learning_rate": 4.370927318295739e-06,
      "loss": 0.5454208254814148,
      "num_input_tokens_seen": 1194996,
      "step": 1565,
      "train_runtime": 3576.4281,
      "train_tokens_per_second": 334.131
    },
    {
      "epoch": 0.144,
      "grad_norm": 7.040731430053711,
      "learning_rate": 4.360902255639098e-06,
      "loss": 0.48860985040664673,
      "num_input_tokens_seen": 1195740,
      "step": 1566,
      "train_runtime": 3578.3707,
      "train_tokens_per_second": 334.158
    },
    {
      "epoch": 0.1440919540229885,
      "grad_norm": 5.951720237731934,
      "learning_rate": 4.350877192982457e-06,
      "loss": 0.3874664306640625,
      "num_input_tokens_seen": 1196676,
      "step": 1567,
      "train_runtime": 3580.3276,
      "train_tokens_per_second": 334.236
    },
    {
      "epoch": 0.144183908045977,
      "grad_norm": 6.843580722808838,
      "learning_rate": 4.340852130325815e-06,
      "loss": 0.4062352776527405,
      "num_input_tokens_seen": 1197460,
      "step": 1568,
      "train_runtime": 3582.273,
      "train_tokens_per_second": 334.274
    },
    {
      "epoch": 0.1442758620689655,
      "grad_norm": 6.490946292877197,
      "learning_rate": 4.330827067669173e-06,
      "loss": 0.5126763582229614,
      "num_input_tokens_seen": 1198108,
      "step": 1569,
      "train_runtime": 3584.1927,
      "train_tokens_per_second": 334.276
    },
    {
      "epoch": 0.144367816091954,
      "grad_norm": 6.102653503417969,
      "learning_rate": 4.320802005012532e-06,
      "loss": 0.4109741449356079,
      "num_input_tokens_seen": 1198712,
      "step": 1570,
      "train_runtime": 3586.1048,
      "train_tokens_per_second": 334.266
    },
    {
      "epoch": 0.14445977011494252,
      "grad_norm": 6.025990962982178,
      "learning_rate": 4.31077694235589e-06,
      "loss": 0.3658444285392761,
      "num_input_tokens_seen": 1199516,
      "step": 1571,
      "train_runtime": 3588.0707,
      "train_tokens_per_second": 334.307
    },
    {
      "epoch": 0.14455172413793105,
      "grad_norm": 6.144662857055664,
      "learning_rate": 4.3007518796992486e-06,
      "loss": 0.413577675819397,
      "num_input_tokens_seen": 1200164,
      "step": 1572,
      "train_runtime": 3590.0099,
      "train_tokens_per_second": 334.307
    },
    {
      "epoch": 0.14464367816091955,
      "grad_norm": 7.419083118438721,
      "learning_rate": 4.290726817042607e-06,
      "loss": 0.4642472565174103,
      "num_input_tokens_seen": 1200740,
      "step": 1573,
      "train_runtime": 3591.9381,
      "train_tokens_per_second": 334.287
    },
    {
      "epoch": 0.14473563218390806,
      "grad_norm": 7.012404918670654,
      "learning_rate": 4.2807017543859656e-06,
      "loss": 0.4493465721607208,
      "num_input_tokens_seen": 1201628,
      "step": 1574,
      "train_runtime": 3593.8828,
      "train_tokens_per_second": 334.354
    },
    {
      "epoch": 0.14482758620689656,
      "grad_norm": 6.282637119293213,
      "learning_rate": 4.270676691729323e-06,
      "loss": 0.4151184856891632,
      "num_input_tokens_seen": 1202244,
      "step": 1575,
      "train_runtime": 3595.8076,
      "train_tokens_per_second": 334.346
    },
    {
      "epoch": 0.14491954022988507,
      "grad_norm": 6.75771951675415,
      "learning_rate": 4.260651629072682e-06,
      "loss": 0.4637143611907959,
      "num_input_tokens_seen": 1202884,
      "step": 1576,
      "train_runtime": 3597.7598,
      "train_tokens_per_second": 334.342
    },
    {
      "epoch": 0.14501149425287357,
      "grad_norm": 7.565665245056152,
      "learning_rate": 4.250626566416041e-06,
      "loss": 0.47488293051719666,
      "num_input_tokens_seen": 1203780,
      "step": 1577,
      "train_runtime": 3599.7262,
      "train_tokens_per_second": 334.409
    },
    {
      "epoch": 0.14510344827586208,
      "grad_norm": 8.024337768554688,
      "learning_rate": 4.240601503759399e-06,
      "loss": 0.4857102632522583,
      "num_input_tokens_seen": 1204544,
      "step": 1578,
      "train_runtime": 3601.6822,
      "train_tokens_per_second": 334.439
    },
    {
      "epoch": 0.14519540229885058,
      "grad_norm": 6.690497875213623,
      "learning_rate": 4.230576441102757e-06,
      "loss": 0.46639159321784973,
      "num_input_tokens_seen": 1205300,
      "step": 1579,
      "train_runtime": 3603.6109,
      "train_tokens_per_second": 334.47
    },
    {
      "epoch": 0.1452873563218391,
      "grad_norm": 6.756970405578613,
      "learning_rate": 4.220551378446116e-06,
      "loss": 0.49929511547088623,
      "num_input_tokens_seen": 1205904,
      "step": 1580,
      "train_runtime": 3605.6055,
      "train_tokens_per_second": 334.453
    },
    {
      "epoch": 0.1453793103448276,
      "grad_norm": 6.6091837882995605,
      "learning_rate": 4.210526315789474e-06,
      "loss": 0.4757223129272461,
      "num_input_tokens_seen": 1206556,
      "step": 1581,
      "train_runtime": 3607.5171,
      "train_tokens_per_second": 334.456
    },
    {
      "epoch": 0.1454712643678161,
      "grad_norm": 6.709166526794434,
      "learning_rate": 4.200501253132833e-06,
      "loss": 0.4480632543563843,
      "num_input_tokens_seen": 1207188,
      "step": 1582,
      "train_runtime": 3609.4438,
      "train_tokens_per_second": 334.453
    },
    {
      "epoch": 0.1455632183908046,
      "grad_norm": 6.548619747161865,
      "learning_rate": 4.190476190476191e-06,
      "loss": 0.5035223364830017,
      "num_input_tokens_seen": 1208300,
      "step": 1583,
      "train_runtime": 3611.4282,
      "train_tokens_per_second": 334.577
    },
    {
      "epoch": 0.1456551724137931,
      "grad_norm": 5.75042200088501,
      "learning_rate": 4.18045112781955e-06,
      "loss": 0.4471976161003113,
      "num_input_tokens_seen": 1209076,
      "step": 1584,
      "train_runtime": 3613.3671,
      "train_tokens_per_second": 334.612
    },
    {
      "epoch": 0.14574712643678162,
      "grad_norm": 7.029244899749756,
      "learning_rate": 4.170426065162907e-06,
      "loss": 0.5173816680908203,
      "num_input_tokens_seen": 1209928,
      "step": 1585,
      "train_runtime": 3615.3144,
      "train_tokens_per_second": 334.667
    },
    {
      "epoch": 0.14583908045977012,
      "grad_norm": 6.378706932067871,
      "learning_rate": 4.160401002506266e-06,
      "loss": 0.4809526205062866,
      "num_input_tokens_seen": 1210812,
      "step": 1586,
      "train_runtime": 3617.2616,
      "train_tokens_per_second": 334.732
    },
    {
      "epoch": 0.14593103448275863,
      "grad_norm": 6.754574298858643,
      "learning_rate": 4.150375939849624e-06,
      "loss": 0.4018881320953369,
      "num_input_tokens_seen": 1211568,
      "step": 1587,
      "train_runtime": 3619.1965,
      "train_tokens_per_second": 334.762
    },
    {
      "epoch": 0.14602298850574713,
      "grad_norm": 6.6122145652771,
      "learning_rate": 4.140350877192983e-06,
      "loss": 0.5332188606262207,
      "num_input_tokens_seen": 1214064,
      "step": 1588,
      "train_runtime": 3621.3748,
      "train_tokens_per_second": 335.249
    },
    {
      "epoch": 0.14611494252873564,
      "grad_norm": 6.535598278045654,
      "learning_rate": 4.130325814536341e-06,
      "loss": 0.46375539898872375,
      "num_input_tokens_seen": 1214812,
      "step": 1589,
      "train_runtime": 3623.3229,
      "train_tokens_per_second": 335.276
    },
    {
      "epoch": 0.14620689655172414,
      "grad_norm": 6.416146278381348,
      "learning_rate": 4.1203007518797e-06,
      "loss": 0.4666772186756134,
      "num_input_tokens_seen": 1215524,
      "step": 1590,
      "train_runtime": 3625.2421,
      "train_tokens_per_second": 335.295
    },
    {
      "epoch": 0.14629885057471265,
      "grad_norm": 6.42300271987915,
      "learning_rate": 4.110275689223058e-06,
      "loss": 0.4736384153366089,
      "num_input_tokens_seen": 1216204,
      "step": 1591,
      "train_runtime": 3633.8338,
      "train_tokens_per_second": 334.689
    },
    {
      "epoch": 0.14639080459770115,
      "grad_norm": 6.641760349273682,
      "learning_rate": 4.100250626566417e-06,
      "loss": 0.5206786394119263,
      "num_input_tokens_seen": 1217020,
      "step": 1592,
      "train_runtime": 3635.8077,
      "train_tokens_per_second": 334.732
    },
    {
      "epoch": 0.14648275862068966,
      "grad_norm": 6.356281280517578,
      "learning_rate": 4.090225563909775e-06,
      "loss": 0.47648727893829346,
      "num_input_tokens_seen": 1217760,
      "step": 1593,
      "train_runtime": 3637.7325,
      "train_tokens_per_second": 334.758
    },
    {
      "epoch": 0.14657471264367816,
      "grad_norm": 6.1111297607421875,
      "learning_rate": 4.080200501253133e-06,
      "loss": 0.4094933867454529,
      "num_input_tokens_seen": 1218356,
      "step": 1594,
      "train_runtime": 3639.6469,
      "train_tokens_per_second": 334.746
    },
    {
      "epoch": 0.14666666666666667,
      "grad_norm": 7.068531036376953,
      "learning_rate": 4.070175438596491e-06,
      "loss": 0.4315361976623535,
      "num_input_tokens_seen": 1219076,
      "step": 1595,
      "train_runtime": 3641.6063,
      "train_tokens_per_second": 334.763
    },
    {
      "epoch": 0.14675862068965517,
      "grad_norm": 7.483480930328369,
      "learning_rate": 4.06015037593985e-06,
      "loss": 0.5120396614074707,
      "num_input_tokens_seen": 1219760,
      "step": 1596,
      "train_runtime": 3643.5708,
      "train_tokens_per_second": 334.77
    },
    {
      "epoch": 0.14685057471264368,
      "grad_norm": 6.521261215209961,
      "learning_rate": 4.050125313283208e-06,
      "loss": 0.39161306619644165,
      "num_input_tokens_seen": 1220400,
      "step": 1597,
      "train_runtime": 3645.5029,
      "train_tokens_per_second": 334.769
    },
    {
      "epoch": 0.14694252873563218,
      "grad_norm": 6.332038879394531,
      "learning_rate": 4.040100250626567e-06,
      "loss": 0.4736579656600952,
      "num_input_tokens_seen": 1221100,
      "step": 1598,
      "train_runtime": 3647.4244,
      "train_tokens_per_second": 334.784
    },
    {
      "epoch": 0.1470344827586207,
      "grad_norm": 7.104672908782959,
      "learning_rate": 4.030075187969925e-06,
      "loss": 0.5967186689376831,
      "num_input_tokens_seen": 1221788,
      "step": 1599,
      "train_runtime": 3649.3519,
      "train_tokens_per_second": 334.796
    },
    {
      "epoch": 0.1471264367816092,
      "grad_norm": 6.8572587966918945,
      "learning_rate": 4.020050125313284e-06,
      "loss": 0.46505308151245117,
      "num_input_tokens_seen": 1222464,
      "step": 1600,
      "train_runtime": 3651.267,
      "train_tokens_per_second": 334.805
    },
    {
      "epoch": 0.1471264367816092,
      "eval_loss": 1.8033605813980103,
      "eval_runtime": 27.9632,
      "eval_samples_per_second": 35.761,
      "eval_steps_per_second": 8.94,
      "num_input_tokens_seen": 1222464,
      "step": 1600
    },
    {
      "epoch": 0.1472183908045977,
      "grad_norm": 6.851583480834961,
      "learning_rate": 4.010025062656641e-06,
      "loss": 0.5206998586654663,
      "num_input_tokens_seen": 1223332,
      "step": 1601,
      "train_runtime": 3681.1937,
      "train_tokens_per_second": 332.319
    },
    {
      "epoch": 0.1473103448275862,
      "grad_norm": 6.868015289306641,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.5103731155395508,
      "num_input_tokens_seen": 1224728,
      "step": 1602,
      "train_runtime": 3683.2062,
      "train_tokens_per_second": 332.517
    },
    {
      "epoch": 0.1474022988505747,
      "grad_norm": 6.602429389953613,
      "learning_rate": 3.989974937343359e-06,
      "loss": 0.5429903268814087,
      "num_input_tokens_seen": 1225460,
      "step": 1603,
      "train_runtime": 3685.1252,
      "train_tokens_per_second": 332.542
    },
    {
      "epoch": 0.14749425287356321,
      "grad_norm": 7.1812663078308105,
      "learning_rate": 3.979949874686717e-06,
      "loss": 0.5265251398086548,
      "num_input_tokens_seen": 1226144,
      "step": 1604,
      "train_runtime": 3687.0339,
      "train_tokens_per_second": 332.556
    },
    {
      "epoch": 0.14758620689655172,
      "grad_norm": 6.5239996910095215,
      "learning_rate": 3.969924812030075e-06,
      "loss": 0.3950444161891937,
      "num_input_tokens_seen": 1226792,
      "step": 1605,
      "train_runtime": 3688.9373,
      "train_tokens_per_second": 332.56
    },
    {
      "epoch": 0.14767816091954022,
      "grad_norm": 7.124094009399414,
      "learning_rate": 3.959899749373434e-06,
      "loss": 0.4906897246837616,
      "num_input_tokens_seen": 1227400,
      "step": 1606,
      "train_runtime": 3690.8455,
      "train_tokens_per_second": 332.553
    },
    {
      "epoch": 0.14777011494252873,
      "grad_norm": 6.652856349945068,
      "learning_rate": 3.949874686716792e-06,
      "loss": 0.4999001622200012,
      "num_input_tokens_seen": 1228192,
      "step": 1607,
      "train_runtime": 3692.7547,
      "train_tokens_per_second": 332.595
    },
    {
      "epoch": 0.14786206896551723,
      "grad_norm": 6.2461700439453125,
      "learning_rate": 3.939849624060151e-06,
      "loss": 0.37495970726013184,
      "num_input_tokens_seen": 1228728,
      "step": 1608,
      "train_runtime": 3694.6626,
      "train_tokens_per_second": 332.568
    },
    {
      "epoch": 0.14795402298850574,
      "grad_norm": 6.433096408843994,
      "learning_rate": 3.929824561403509e-06,
      "loss": 0.4584512710571289,
      "num_input_tokens_seen": 1229856,
      "step": 1609,
      "train_runtime": 3696.675,
      "train_tokens_per_second": 332.692
    },
    {
      "epoch": 0.14804597701149425,
      "grad_norm": 5.780365943908691,
      "learning_rate": 3.919799498746868e-06,
      "loss": 0.3628767728805542,
      "num_input_tokens_seen": 1230684,
      "step": 1610,
      "train_runtime": 3698.7181,
      "train_tokens_per_second": 332.733
    },
    {
      "epoch": 0.14813793103448275,
      "grad_norm": 5.993629455566406,
      "learning_rate": 3.909774436090225e-06,
      "loss": 0.47672876715660095,
      "num_input_tokens_seen": 1232252,
      "step": 1611,
      "train_runtime": 3700.7812,
      "train_tokens_per_second": 332.971
    },
    {
      "epoch": 0.14822988505747126,
      "grad_norm": 6.184093952178955,
      "learning_rate": 3.899749373433585e-06,
      "loss": 0.3732033371925354,
      "num_input_tokens_seen": 1233156,
      "step": 1612,
      "train_runtime": 3702.7595,
      "train_tokens_per_second": 333.037
    },
    {
      "epoch": 0.14832183908045976,
      "grad_norm": 6.4238810539245605,
      "learning_rate": 3.889724310776943e-06,
      "loss": 0.4720228910446167,
      "num_input_tokens_seen": 1233728,
      "step": 1613,
      "train_runtime": 3704.6676,
      "train_tokens_per_second": 333.02
    },
    {
      "epoch": 0.14841379310344827,
      "grad_norm": 6.06830358505249,
      "learning_rate": 3.879699248120301e-06,
      "loss": 0.3889967203140259,
      "num_input_tokens_seen": 1234312,
      "step": 1614,
      "train_runtime": 3706.5808,
      "train_tokens_per_second": 333.006
    },
    {
      "epoch": 0.14850574712643677,
      "grad_norm": 7.1693549156188965,
      "learning_rate": 3.869674185463659e-06,
      "loss": 0.6546400785446167,
      "num_input_tokens_seen": 1235508,
      "step": 1615,
      "train_runtime": 3708.565,
      "train_tokens_per_second": 333.15
    },
    {
      "epoch": 0.14859770114942528,
      "grad_norm": 7.182287216186523,
      "learning_rate": 3.859649122807018e-06,
      "loss": 0.5240688323974609,
      "num_input_tokens_seen": 1236284,
      "step": 1616,
      "train_runtime": 3710.4782,
      "train_tokens_per_second": 333.187
    },
    {
      "epoch": 0.14868965517241378,
      "grad_norm": 7.047788619995117,
      "learning_rate": 3.849624060150376e-06,
      "loss": 0.47436654567718506,
      "num_input_tokens_seen": 1236828,
      "step": 1617,
      "train_runtime": 3712.3814,
      "train_tokens_per_second": 333.163
    },
    {
      "epoch": 0.1487816091954023,
      "grad_norm": 6.530257225036621,
      "learning_rate": 3.839598997493735e-06,
      "loss": 0.40288904309272766,
      "num_input_tokens_seen": 1237548,
      "step": 1618,
      "train_runtime": 3714.3282,
      "train_tokens_per_second": 333.182
    },
    {
      "epoch": 0.1488735632183908,
      "grad_norm": 6.0984063148498535,
      "learning_rate": 3.829573934837093e-06,
      "loss": 0.5006040334701538,
      "num_input_tokens_seen": 1238564,
      "step": 1619,
      "train_runtime": 3716.3041,
      "train_tokens_per_second": 333.278
    },
    {
      "epoch": 0.1489655172413793,
      "grad_norm": 7.607024669647217,
      "learning_rate": 3.819548872180452e-06,
      "loss": 0.5643741488456726,
      "num_input_tokens_seen": 1239192,
      "step": 1620,
      "train_runtime": 3718.2229,
      "train_tokens_per_second": 333.275
    },
    {
      "epoch": 0.1490574712643678,
      "grad_norm": 7.06319522857666,
      "learning_rate": 3.80952380952381e-06,
      "loss": 0.5532639026641846,
      "num_input_tokens_seen": 1240556,
      "step": 1621,
      "train_runtime": 3727.1797,
      "train_tokens_per_second": 332.84
    },
    {
      "epoch": 0.14914942528735634,
      "grad_norm": 7.681368827819824,
      "learning_rate": 3.7994987468671683e-06,
      "loss": 0.48061004281044006,
      "num_input_tokens_seen": 1241332,
      "step": 1622,
      "train_runtime": 3729.2641,
      "train_tokens_per_second": 332.862
    },
    {
      "epoch": 0.14924137931034484,
      "grad_norm": 6.306829452514648,
      "learning_rate": 3.789473684210527e-06,
      "loss": 0.41942787170410156,
      "num_input_tokens_seen": 1241944,
      "step": 1623,
      "train_runtime": 3731.184,
      "train_tokens_per_second": 332.855
    },
    {
      "epoch": 0.14933333333333335,
      "grad_norm": 7.706522464752197,
      "learning_rate": 3.779448621553885e-06,
      "loss": 0.5494070053100586,
      "num_input_tokens_seen": 1242636,
      "step": 1624,
      "train_runtime": 3733.0925,
      "train_tokens_per_second": 332.87
    },
    {
      "epoch": 0.14942528735632185,
      "grad_norm": 6.535772323608398,
      "learning_rate": 3.7694235588972433e-06,
      "loss": 0.42592912912368774,
      "num_input_tokens_seen": 1243304,
      "step": 1625,
      "train_runtime": 3735.0027,
      "train_tokens_per_second": 332.879
    },
    {
      "epoch": 0.14951724137931036,
      "grad_norm": 8.192347526550293,
      "learning_rate": 3.7593984962406014e-06,
      "loss": 0.5597870945930481,
      "num_input_tokens_seen": 1244020,
      "step": 1626,
      "train_runtime": 3736.9427,
      "train_tokens_per_second": 332.898
    },
    {
      "epoch": 0.14960919540229886,
      "grad_norm": 7.159975528717041,
      "learning_rate": 3.7493734335839603e-06,
      "loss": 0.4892594814300537,
      "num_input_tokens_seen": 1244620,
      "step": 1627,
      "train_runtime": 3738.8571,
      "train_tokens_per_second": 332.888
    },
    {
      "epoch": 0.14970114942528737,
      "grad_norm": 6.74794864654541,
      "learning_rate": 3.739348370927319e-06,
      "loss": 0.5493437051773071,
      "num_input_tokens_seen": 1245312,
      "step": 1628,
      "train_runtime": 3740.8446,
      "train_tokens_per_second": 332.896
    },
    {
      "epoch": 0.14979310344827587,
      "grad_norm": 6.720531940460205,
      "learning_rate": 3.729323308270677e-06,
      "loss": 0.4282424747943878,
      "num_input_tokens_seen": 1246036,
      "step": 1629,
      "train_runtime": 3742.7615,
      "train_tokens_per_second": 332.919
    },
    {
      "epoch": 0.14988505747126438,
      "grad_norm": 6.940911769866943,
      "learning_rate": 3.7192982456140354e-06,
      "loss": 0.5560833811759949,
      "num_input_tokens_seen": 1246856,
      "step": 1630,
      "train_runtime": 3744.7369,
      "train_tokens_per_second": 332.962
    },
    {
      "epoch": 0.14997701149425288,
      "grad_norm": 6.8108415603637695,
      "learning_rate": 3.7092731829573934e-06,
      "loss": 0.4583231210708618,
      "num_input_tokens_seen": 1247520,
      "step": 1631,
      "train_runtime": 3746.6817,
      "train_tokens_per_second": 332.967
    },
    {
      "epoch": 0.1500689655172414,
      "grad_norm": 6.710882663726807,
      "learning_rate": 3.6992481203007523e-06,
      "loss": 0.458967000246048,
      "num_input_tokens_seen": 1248164,
      "step": 1632,
      "train_runtime": 3748.6007,
      "train_tokens_per_second": 332.968
    },
    {
      "epoch": 0.1501609195402299,
      "grad_norm": 5.969411849975586,
      "learning_rate": 3.689223057644111e-06,
      "loss": 0.42513519525527954,
      "num_input_tokens_seen": 1248828,
      "step": 1633,
      "train_runtime": 3750.5278,
      "train_tokens_per_second": 332.974
    },
    {
      "epoch": 0.1502528735632184,
      "grad_norm": 7.540334224700928,
      "learning_rate": 3.679197994987469e-06,
      "loss": 0.45841267704963684,
      "num_input_tokens_seen": 1249580,
      "step": 1634,
      "train_runtime": 3752.5442,
      "train_tokens_per_second": 332.995
    },
    {
      "epoch": 0.1503448275862069,
      "grad_norm": 7.294579029083252,
      "learning_rate": 3.6691729323308274e-06,
      "loss": 0.5723696947097778,
      "num_input_tokens_seen": 1250416,
      "step": 1635,
      "train_runtime": 3754.5019,
      "train_tokens_per_second": 333.044
    },
    {
      "epoch": 0.1504367816091954,
      "grad_norm": 6.645254135131836,
      "learning_rate": 3.6591478696741854e-06,
      "loss": 0.4086427092552185,
      "num_input_tokens_seen": 1251188,
      "step": 1636,
      "train_runtime": 3756.4271,
      "train_tokens_per_second": 333.079
    },
    {
      "epoch": 0.1505287356321839,
      "grad_norm": 7.079519748687744,
      "learning_rate": 3.6491228070175443e-06,
      "loss": 0.4305681586265564,
      "num_input_tokens_seen": 1251796,
      "step": 1637,
      "train_runtime": 3758.3346,
      "train_tokens_per_second": 333.072
    },
    {
      "epoch": 0.15062068965517242,
      "grad_norm": 6.369967937469482,
      "learning_rate": 3.639097744360903e-06,
      "loss": 0.47447091341018677,
      "num_input_tokens_seen": 1252368,
      "step": 1638,
      "train_runtime": 3760.2385,
      "train_tokens_per_second": 333.055
    },
    {
      "epoch": 0.15071264367816092,
      "grad_norm": 6.839728832244873,
      "learning_rate": 3.629072681704261e-06,
      "loss": 0.5020827651023865,
      "num_input_tokens_seen": 1252948,
      "step": 1639,
      "train_runtime": 3762.1563,
      "train_tokens_per_second": 333.04
    },
    {
      "epoch": 0.15080459770114943,
      "grad_norm": 6.284133434295654,
      "learning_rate": 3.6190476190476194e-06,
      "loss": 0.46222183108329773,
      "num_input_tokens_seen": 1254244,
      "step": 1640,
      "train_runtime": 3764.1381,
      "train_tokens_per_second": 333.209
    },
    {
      "epoch": 0.15089655172413793,
      "grad_norm": 5.810016632080078,
      "learning_rate": 3.6090225563909775e-06,
      "loss": 0.40370118618011475,
      "num_input_tokens_seen": 1255088,
      "step": 1641,
      "train_runtime": 3766.0907,
      "train_tokens_per_second": 333.26
    },
    {
      "epoch": 0.15098850574712644,
      "grad_norm": 7.57272481918335,
      "learning_rate": 3.5989974937343364e-06,
      "loss": 0.5796998143196106,
      "num_input_tokens_seen": 1255824,
      "step": 1642,
      "train_runtime": 3768.037,
      "train_tokens_per_second": 333.283
    },
    {
      "epoch": 0.15108045977011494,
      "grad_norm": 6.646892070770264,
      "learning_rate": 3.588972431077695e-06,
      "loss": 0.39232754707336426,
      "num_input_tokens_seen": 1256416,
      "step": 1643,
      "train_runtime": 3769.9477,
      "train_tokens_per_second": 333.271
    },
    {
      "epoch": 0.15117241379310345,
      "grad_norm": 6.2148308753967285,
      "learning_rate": 3.578947368421053e-06,
      "loss": 0.36920031905174255,
      "num_input_tokens_seen": 1257092,
      "step": 1644,
      "train_runtime": 3771.899,
      "train_tokens_per_second": 333.278
    },
    {
      "epoch": 0.15126436781609195,
      "grad_norm": 7.5221381187438965,
      "learning_rate": 3.5689223057644114e-06,
      "loss": 0.537412166595459,
      "num_input_tokens_seen": 1257668,
      "step": 1645,
      "train_runtime": 3773.8109,
      "train_tokens_per_second": 333.262
    },
    {
      "epoch": 0.15135632183908046,
      "grad_norm": 6.563010215759277,
      "learning_rate": 3.5588972431077695e-06,
      "loss": 0.4267057478427887,
      "num_input_tokens_seen": 1258424,
      "step": 1646,
      "train_runtime": 3775.7502,
      "train_tokens_per_second": 333.291
    },
    {
      "epoch": 0.15144827586206897,
      "grad_norm": 6.13692569732666,
      "learning_rate": 3.5488721804511284e-06,
      "loss": 0.3660619854927063,
      "num_input_tokens_seen": 1259052,
      "step": 1647,
      "train_runtime": 3777.7159,
      "train_tokens_per_second": 333.284
    },
    {
      "epoch": 0.15154022988505747,
      "grad_norm": 6.393582820892334,
      "learning_rate": 3.5388471177944864e-06,
      "loss": 0.4673348069190979,
      "num_input_tokens_seen": 1259824,
      "step": 1648,
      "train_runtime": 3779.6358,
      "train_tokens_per_second": 333.319
    },
    {
      "epoch": 0.15163218390804598,
      "grad_norm": 6.13256311416626,
      "learning_rate": 3.528822055137845e-06,
      "loss": 0.5436832308769226,
      "num_input_tokens_seen": 1260612,
      "step": 1649,
      "train_runtime": 3781.5531,
      "train_tokens_per_second": 333.358
    },
    {
      "epoch": 0.15172413793103448,
      "grad_norm": 6.178951740264893,
      "learning_rate": 3.5187969924812034e-06,
      "loss": 0.3987323045730591,
      "num_input_tokens_seen": 1261272,
      "step": 1650,
      "train_runtime": 3783.4813,
      "train_tokens_per_second": 333.363
    },
    {
      "epoch": 0.15181609195402299,
      "grad_norm": 6.734344482421875,
      "learning_rate": 3.5087719298245615e-06,
      "loss": 0.4014202952384949,
      "num_input_tokens_seen": 1261996,
      "step": 1651,
      "train_runtime": 3791.736,
      "train_tokens_per_second": 332.828
    },
    {
      "epoch": 0.1519080459770115,
      "grad_norm": 7.783940315246582,
      "learning_rate": 3.4987468671679204e-06,
      "loss": 0.4166038930416107,
      "num_input_tokens_seen": 1262660,
      "step": 1652,
      "train_runtime": 3793.6642,
      "train_tokens_per_second": 332.834
    },
    {
      "epoch": 0.152,
      "grad_norm": 7.206593990325928,
      "learning_rate": 3.4887218045112785e-06,
      "loss": 0.44669169187545776,
      "num_input_tokens_seen": 1263660,
      "step": 1653,
      "train_runtime": 3795.6258,
      "train_tokens_per_second": 332.925
    },
    {
      "epoch": 0.1520919540229885,
      "grad_norm": 7.104734897613525,
      "learning_rate": 3.478696741854637e-06,
      "loss": 0.524027407169342,
      "num_input_tokens_seen": 1264284,
      "step": 1654,
      "train_runtime": 3797.5657,
      "train_tokens_per_second": 332.92
    },
    {
      "epoch": 0.152183908045977,
      "grad_norm": 6.854971885681152,
      "learning_rate": 3.468671679197995e-06,
      "loss": 0.5613166093826294,
      "num_input_tokens_seen": 1265392,
      "step": 1655,
      "train_runtime": 3799.5672,
      "train_tokens_per_second": 333.036
    },
    {
      "epoch": 0.1522758620689655,
      "grad_norm": 7.128030300140381,
      "learning_rate": 3.4586466165413535e-06,
      "loss": 0.49357134103775024,
      "num_input_tokens_seen": 1266160,
      "step": 1656,
      "train_runtime": 3801.66,
      "train_tokens_per_second": 333.055
    },
    {
      "epoch": 0.15236781609195402,
      "grad_norm": 6.797114849090576,
      "learning_rate": 3.4486215538847124e-06,
      "loss": 0.49934589862823486,
      "num_input_tokens_seen": 1266788,
      "step": 1657,
      "train_runtime": 3803.5801,
      "train_tokens_per_second": 333.051
    },
    {
      "epoch": 0.15245977011494252,
      "grad_norm": 7.697146892547607,
      "learning_rate": 3.4385964912280705e-06,
      "loss": 0.5518667697906494,
      "num_input_tokens_seen": 1267528,
      "step": 1658,
      "train_runtime": 3805.5016,
      "train_tokens_per_second": 333.078
    },
    {
      "epoch": 0.15255172413793103,
      "grad_norm": 6.477298259735107,
      "learning_rate": 3.428571428571429e-06,
      "loss": 0.5126209855079651,
      "num_input_tokens_seen": 1268336,
      "step": 1659,
      "train_runtime": 3807.4633,
      "train_tokens_per_second": 333.118
    },
    {
      "epoch": 0.15264367816091953,
      "grad_norm": 6.133742809295654,
      "learning_rate": 3.418546365914787e-06,
      "loss": 0.44767650961875916,
      "num_input_tokens_seen": 1269216,
      "step": 1660,
      "train_runtime": 3809.4465,
      "train_tokens_per_second": 333.176
    },
    {
      "epoch": 0.15273563218390804,
      "grad_norm": 6.014475345611572,
      "learning_rate": 3.4085213032581455e-06,
      "loss": 0.43354594707489014,
      "num_input_tokens_seen": 1269892,
      "step": 1661,
      "train_runtime": 3811.3655,
      "train_tokens_per_second": 333.186
    },
    {
      "epoch": 0.15282758620689654,
      "grad_norm": 6.745570182800293,
      "learning_rate": 3.3984962406015044e-06,
      "loss": 0.4856205880641937,
      "num_input_tokens_seen": 1270536,
      "step": 1662,
      "train_runtime": 3813.2774,
      "train_tokens_per_second": 333.187
    },
    {
      "epoch": 0.15291954022988505,
      "grad_norm": 6.704166889190674,
      "learning_rate": 3.3884711779448625e-06,
      "loss": 0.5431050658226013,
      "num_input_tokens_seen": 1272424,
      "step": 1663,
      "train_runtime": 3815.3608,
      "train_tokens_per_second": 333.5
    },
    {
      "epoch": 0.15301149425287355,
      "grad_norm": 7.693567752838135,
      "learning_rate": 3.378446115288221e-06,
      "loss": 0.4961692690849304,
      "num_input_tokens_seen": 1273080,
      "step": 1664,
      "train_runtime": 3817.2807,
      "train_tokens_per_second": 333.504
    },
    {
      "epoch": 0.15310344827586206,
      "grad_norm": 7.393714904785156,
      "learning_rate": 3.368421052631579e-06,
      "loss": 0.5113829374313354,
      "num_input_tokens_seen": 1273956,
      "step": 1665,
      "train_runtime": 3819.2093,
      "train_tokens_per_second": 333.565
    },
    {
      "epoch": 0.15319540229885056,
      "grad_norm": 5.7330322265625,
      "learning_rate": 3.3583959899749375e-06,
      "loss": 0.4245642423629761,
      "num_input_tokens_seen": 1274964,
      "step": 1666,
      "train_runtime": 3821.1812,
      "train_tokens_per_second": 333.657
    },
    {
      "epoch": 0.15328735632183907,
      "grad_norm": 5.6045613288879395,
      "learning_rate": 3.3483709273182964e-06,
      "loss": 0.3708139657974243,
      "num_input_tokens_seen": 1276004,
      "step": 1667,
      "train_runtime": 3823.1514,
      "train_tokens_per_second": 333.757
    },
    {
      "epoch": 0.15337931034482757,
      "grad_norm": 7.3522562980651855,
      "learning_rate": 3.3383458646616545e-06,
      "loss": 0.5081470012664795,
      "num_input_tokens_seen": 1276692,
      "step": 1668,
      "train_runtime": 3825.055,
      "train_tokens_per_second": 333.771
    },
    {
      "epoch": 0.15347126436781608,
      "grad_norm": 6.863851070404053,
      "learning_rate": 3.328320802005013e-06,
      "loss": 0.4304991364479065,
      "num_input_tokens_seen": 1277384,
      "step": 1669,
      "train_runtime": 3826.9881,
      "train_tokens_per_second": 333.783
    },
    {
      "epoch": 0.15356321839080458,
      "grad_norm": 6.657381057739258,
      "learning_rate": 3.318295739348371e-06,
      "loss": 0.4671858847141266,
      "num_input_tokens_seen": 1278112,
      "step": 1670,
      "train_runtime": 3828.9231,
      "train_tokens_per_second": 333.805
    },
    {
      "epoch": 0.1536551724137931,
      "grad_norm": 6.320245265960693,
      "learning_rate": 3.3082706766917295e-06,
      "loss": 0.5130354166030884,
      "num_input_tokens_seen": 1278844,
      "step": 1671,
      "train_runtime": 3830.8603,
      "train_tokens_per_second": 333.827
    },
    {
      "epoch": 0.15374712643678162,
      "grad_norm": 6.49079704284668,
      "learning_rate": 3.2982456140350885e-06,
      "loss": 0.3779211640357971,
      "num_input_tokens_seen": 1279416,
      "step": 1672,
      "train_runtime": 3832.7608,
      "train_tokens_per_second": 333.811
    },
    {
      "epoch": 0.15383908045977013,
      "grad_norm": 7.564385414123535,
      "learning_rate": 3.2882205513784465e-06,
      "loss": 0.4618367552757263,
      "num_input_tokens_seen": 1280068,
      "step": 1673,
      "train_runtime": 3834.6726,
      "train_tokens_per_second": 333.814
    },
    {
      "epoch": 0.15393103448275863,
      "grad_norm": 6.59860897064209,
      "learning_rate": 3.278195488721805e-06,
      "loss": 0.4915279150009155,
      "num_input_tokens_seen": 1280784,
      "step": 1674,
      "train_runtime": 3836.6245,
      "train_tokens_per_second": 333.831
    },
    {
      "epoch": 0.15402298850574714,
      "grad_norm": 6.710583209991455,
      "learning_rate": 3.268170426065163e-06,
      "loss": 0.4566461741924286,
      "num_input_tokens_seen": 1281436,
      "step": 1675,
      "train_runtime": 3838.5341,
      "train_tokens_per_second": 333.835
    },
    {
      "epoch": 0.15411494252873564,
      "grad_norm": 7.710618019104004,
      "learning_rate": 3.2581453634085216e-06,
      "loss": 0.4616769850254059,
      "num_input_tokens_seen": 1282268,
      "step": 1676,
      "train_runtime": 3840.4893,
      "train_tokens_per_second": 333.881
    },
    {
      "epoch": 0.15420689655172415,
      "grad_norm": 8.413847923278809,
      "learning_rate": 3.24812030075188e-06,
      "loss": 0.5942468047142029,
      "num_input_tokens_seen": 1283328,
      "step": 1677,
      "train_runtime": 3842.5017,
      "train_tokens_per_second": 333.982
    },
    {
      "epoch": 0.15429885057471265,
      "grad_norm": 7.268781661987305,
      "learning_rate": 3.2380952380952385e-06,
      "loss": 0.5143240690231323,
      "num_input_tokens_seen": 1284144,
      "step": 1678,
      "train_runtime": 3844.5112,
      "train_tokens_per_second": 334.02
    },
    {
      "epoch": 0.15439080459770116,
      "grad_norm": 5.556809902191162,
      "learning_rate": 3.2280701754385966e-06,
      "loss": 0.46623244881629944,
      "num_input_tokens_seen": 1286216,
      "step": 1679,
      "train_runtime": 3846.6154,
      "train_tokens_per_second": 334.376
    },
    {
      "epoch": 0.15448275862068966,
      "grad_norm": 5.99680233001709,
      "learning_rate": 3.218045112781955e-06,
      "loss": 0.3869587779045105,
      "num_input_tokens_seen": 1287176,
      "step": 1680,
      "train_runtime": 3848.569,
      "train_tokens_per_second": 334.456
    },
    {
      "epoch": 0.15457471264367817,
      "grad_norm": 7.0454511642456055,
      "learning_rate": 3.2080200501253136e-06,
      "loss": 0.4825194180011749,
      "num_input_tokens_seen": 1287908,
      "step": 1681,
      "train_runtime": 3857.0651,
      "train_tokens_per_second": 333.909
    },
    {
      "epoch": 0.15466666666666667,
      "grad_norm": 6.443118095397949,
      "learning_rate": 3.197994987468672e-06,
      "loss": 0.44234657287597656,
      "num_input_tokens_seen": 1288496,
      "step": 1682,
      "train_runtime": 3859.1099,
      "train_tokens_per_second": 333.884
    },
    {
      "epoch": 0.15475862068965518,
      "grad_norm": 6.089779376983643,
      "learning_rate": 3.1879699248120305e-06,
      "loss": 0.42578399181365967,
      "num_input_tokens_seen": 1289044,
      "step": 1683,
      "train_runtime": 3861.0221,
      "train_tokens_per_second": 333.861
    },
    {
      "epoch": 0.15485057471264368,
      "grad_norm": 6.117355823516846,
      "learning_rate": 3.1779448621553886e-06,
      "loss": 0.44038915634155273,
      "num_input_tokens_seen": 1289704,
      "step": 1684,
      "train_runtime": 3862.9692,
      "train_tokens_per_second": 333.863
    },
    {
      "epoch": 0.1549425287356322,
      "grad_norm": 7.234719276428223,
      "learning_rate": 3.167919799498747e-06,
      "loss": 0.5101163387298584,
      "num_input_tokens_seen": 1290360,
      "step": 1685,
      "train_runtime": 3864.8826,
      "train_tokens_per_second": 333.868
    },
    {
      "epoch": 0.1550344827586207,
      "grad_norm": 7.175992012023926,
      "learning_rate": 3.157894736842105e-06,
      "loss": 0.5379934310913086,
      "num_input_tokens_seen": 1291124,
      "step": 1686,
      "train_runtime": 3866.8199,
      "train_tokens_per_second": 333.898
    },
    {
      "epoch": 0.1551264367816092,
      "grad_norm": 6.823253631591797,
      "learning_rate": 3.147869674185464e-06,
      "loss": 0.4339263439178467,
      "num_input_tokens_seen": 1292260,
      "step": 1687,
      "train_runtime": 3868.8297,
      "train_tokens_per_second": 334.018
    },
    {
      "epoch": 0.1552183908045977,
      "grad_norm": 7.68180513381958,
      "learning_rate": 3.1378446115288226e-06,
      "loss": 0.4064546823501587,
      "num_input_tokens_seen": 1293028,
      "step": 1688,
      "train_runtime": 3870.7742,
      "train_tokens_per_second": 334.049
    },
    {
      "epoch": 0.1553103448275862,
      "grad_norm": 6.098221302032471,
      "learning_rate": 3.1278195488721806e-06,
      "loss": 0.36478671431541443,
      "num_input_tokens_seen": 1293676,
      "step": 1689,
      "train_runtime": 3872.6901,
      "train_tokens_per_second": 334.051
    },
    {
      "epoch": 0.15540229885057472,
      "grad_norm": 5.838031768798828,
      "learning_rate": 3.117794486215539e-06,
      "loss": 0.40487241744995117,
      "num_input_tokens_seen": 1294560,
      "step": 1690,
      "train_runtime": 3874.6264,
      "train_tokens_per_second": 334.112
    },
    {
      "epoch": 0.15549425287356322,
      "grad_norm": 7.832051753997803,
      "learning_rate": 3.107769423558897e-06,
      "loss": 0.4227607548236847,
      "num_input_tokens_seen": 1295124,
      "step": 1691,
      "train_runtime": 3876.5207,
      "train_tokens_per_second": 334.094
    },
    {
      "epoch": 0.15558620689655173,
      "grad_norm": 6.668942928314209,
      "learning_rate": 3.097744360902256e-06,
      "loss": 0.4082091450691223,
      "num_input_tokens_seen": 1295636,
      "step": 1692,
      "train_runtime": 3878.4186,
      "train_tokens_per_second": 334.063
    },
    {
      "epoch": 0.15567816091954023,
      "grad_norm": 6.953184604644775,
      "learning_rate": 3.0877192982456146e-06,
      "loss": 0.5256580114364624,
      "num_input_tokens_seen": 1296284,
      "step": 1693,
      "train_runtime": 3880.3384,
      "train_tokens_per_second": 334.065
    },
    {
      "epoch": 0.15577011494252874,
      "grad_norm": 6.644675254821777,
      "learning_rate": 3.0776942355889726e-06,
      "loss": 0.41119199991226196,
      "num_input_tokens_seen": 1297164,
      "step": 1694,
      "train_runtime": 3882.2728,
      "train_tokens_per_second": 334.125
    },
    {
      "epoch": 0.15586206896551724,
      "grad_norm": 7.434882640838623,
      "learning_rate": 3.067669172932331e-06,
      "loss": 0.4734553396701813,
      "num_input_tokens_seen": 1297912,
      "step": 1695,
      "train_runtime": 3884.1855,
      "train_tokens_per_second": 334.153
    },
    {
      "epoch": 0.15595402298850575,
      "grad_norm": 7.21633768081665,
      "learning_rate": 3.057644110275689e-06,
      "loss": 0.48472291231155396,
      "num_input_tokens_seen": 1298624,
      "step": 1696,
      "train_runtime": 3886.0938,
      "train_tokens_per_second": 334.172
    },
    {
      "epoch": 0.15604597701149425,
      "grad_norm": 6.820037364959717,
      "learning_rate": 3.047619047619048e-06,
      "loss": 0.34785187244415283,
      "num_input_tokens_seen": 1299336,
      "step": 1697,
      "train_runtime": 3888.0263,
      "train_tokens_per_second": 334.189
    },
    {
      "epoch": 0.15613793103448276,
      "grad_norm": 6.0627007484436035,
      "learning_rate": 3.0375939849624066e-06,
      "loss": 0.3812011182308197,
      "num_input_tokens_seen": 1299880,
      "step": 1698,
      "train_runtime": 3889.9272,
      "train_tokens_per_second": 334.166
    },
    {
      "epoch": 0.15622988505747126,
      "grad_norm": 5.8618292808532715,
      "learning_rate": 3.0275689223057647e-06,
      "loss": 0.4859369099140167,
      "num_input_tokens_seen": 1300944,
      "step": 1699,
      "train_runtime": 3891.8991,
      "train_tokens_per_second": 334.27
    },
    {
      "epoch": 0.15632183908045977,
      "grad_norm": 6.807727336883545,
      "learning_rate": 3.017543859649123e-06,
      "loss": 0.4760059714317322,
      "num_input_tokens_seen": 1301612,
      "step": 1700,
      "train_runtime": 3893.8193,
      "train_tokens_per_second": 334.276
    },
    {
      "epoch": 0.15641379310344827,
      "grad_norm": 5.982649803161621,
      "learning_rate": 3.007518796992481e-06,
      "loss": 0.4735814929008484,
      "num_input_tokens_seen": 1302580,
      "step": 1701,
      "train_runtime": 3895.7661,
      "train_tokens_per_second": 334.358
    },
    {
      "epoch": 0.15650574712643678,
      "grad_norm": 6.831633567810059,
      "learning_rate": 2.9974937343358397e-06,
      "loss": 0.516548752784729,
      "num_input_tokens_seen": 1303332,
      "step": 1702,
      "train_runtime": 3897.7255,
      "train_tokens_per_second": 334.383
    },
    {
      "epoch": 0.15659770114942528,
      "grad_norm": 7.150908470153809,
      "learning_rate": 2.9874686716791986e-06,
      "loss": 0.5031217932701111,
      "num_input_tokens_seen": 1304004,
      "step": 1703,
      "train_runtime": 3899.6403,
      "train_tokens_per_second": 334.391
    },
    {
      "epoch": 0.1566896551724138,
      "grad_norm": 8.302416801452637,
      "learning_rate": 2.9774436090225567e-06,
      "loss": 0.5160796642303467,
      "num_input_tokens_seen": 1304632,
      "step": 1704,
      "train_runtime": 3901.5764,
      "train_tokens_per_second": 334.386
    },
    {
      "epoch": 0.1567816091954023,
      "grad_norm": 7.03695011138916,
      "learning_rate": 2.967418546365915e-06,
      "loss": 0.5238287448883057,
      "num_input_tokens_seen": 1305564,
      "step": 1705,
      "train_runtime": 3903.6113,
      "train_tokens_per_second": 334.45
    },
    {
      "epoch": 0.1568735632183908,
      "grad_norm": 7.0970258712768555,
      "learning_rate": 2.9573934837092732e-06,
      "loss": 0.48913565278053284,
      "num_input_tokens_seen": 1306148,
      "step": 1706,
      "train_runtime": 3905.5277,
      "train_tokens_per_second": 334.436
    },
    {
      "epoch": 0.1569655172413793,
      "grad_norm": 6.433727264404297,
      "learning_rate": 2.9473684210526317e-06,
      "loss": 0.5272034406661987,
      "num_input_tokens_seen": 1306844,
      "step": 1707,
      "train_runtime": 3907.463,
      "train_tokens_per_second": 334.448
    },
    {
      "epoch": 0.1570574712643678,
      "grad_norm": 5.6281914710998535,
      "learning_rate": 2.93734335839599e-06,
      "loss": 0.4618489146232605,
      "num_input_tokens_seen": 1307480,
      "step": 1708,
      "train_runtime": 3909.4034,
      "train_tokens_per_second": 334.445
    },
    {
      "epoch": 0.15714942528735631,
      "grad_norm": 7.447822570800781,
      "learning_rate": 2.9273182957393487e-06,
      "loss": 0.4627896845340729,
      "num_input_tokens_seen": 1308100,
      "step": 1709,
      "train_runtime": 3911.3177,
      "train_tokens_per_second": 334.44
    },
    {
      "epoch": 0.15724137931034482,
      "grad_norm": 6.788646221160889,
      "learning_rate": 2.9172932330827068e-06,
      "loss": 0.36336851119995117,
      "num_input_tokens_seen": 1308704,
      "step": 1710,
      "train_runtime": 3913.3065,
      "train_tokens_per_second": 334.424
    },
    {
      "epoch": 0.15733333333333333,
      "grad_norm": 6.882571697235107,
      "learning_rate": 2.9072681704260652e-06,
      "loss": 0.4588228762149811,
      "num_input_tokens_seen": 1309428,
      "step": 1711,
      "train_runtime": 3921.5543,
      "train_tokens_per_second": 333.905
    },
    {
      "epoch": 0.15742528735632183,
      "grad_norm": 5.915749549865723,
      "learning_rate": 2.8972431077694237e-06,
      "loss": 0.5249640941619873,
      "num_input_tokens_seen": 1311304,
      "step": 1712,
      "train_runtime": 3923.611,
      "train_tokens_per_second": 334.208
    },
    {
      "epoch": 0.15751724137931034,
      "grad_norm": 6.004611015319824,
      "learning_rate": 2.8872180451127822e-06,
      "loss": 0.37069863080978394,
      "num_input_tokens_seen": 1312204,
      "step": 1713,
      "train_runtime": 3925.5443,
      "train_tokens_per_second": 334.273
    },
    {
      "epoch": 0.15760919540229884,
      "grad_norm": 6.544671535491943,
      "learning_rate": 2.8771929824561407e-06,
      "loss": 0.47094351053237915,
      "num_input_tokens_seen": 1312992,
      "step": 1714,
      "train_runtime": 3927.4819,
      "train_tokens_per_second": 334.309
    },
    {
      "epoch": 0.15770114942528735,
      "grad_norm": 7.051267147064209,
      "learning_rate": 2.8671679197994988e-06,
      "loss": 0.5160407423973083,
      "num_input_tokens_seen": 1313932,
      "step": 1715,
      "train_runtime": 3929.4477,
      "train_tokens_per_second": 334.381
    },
    {
      "epoch": 0.15779310344827585,
      "grad_norm": 6.341228485107422,
      "learning_rate": 2.8571428571428573e-06,
      "loss": 0.4405323565006256,
      "num_input_tokens_seen": 1314560,
      "step": 1716,
      "train_runtime": 3931.3638,
      "train_tokens_per_second": 334.378
    },
    {
      "epoch": 0.15788505747126436,
      "grad_norm": 7.63276481628418,
      "learning_rate": 2.8471177944862153e-06,
      "loss": 0.5746864676475525,
      "num_input_tokens_seen": 1315328,
      "step": 1717,
      "train_runtime": 3933.2805,
      "train_tokens_per_second": 334.41
    },
    {
      "epoch": 0.15797701149425286,
      "grad_norm": 6.935079574584961,
      "learning_rate": 2.8370927318295742e-06,
      "loss": 0.4732591211795807,
      "num_input_tokens_seen": 1315996,
      "step": 1718,
      "train_runtime": 3935.2289,
      "train_tokens_per_second": 334.414
    },
    {
      "epoch": 0.15806896551724137,
      "grad_norm": 6.347714424133301,
      "learning_rate": 2.8270676691729327e-06,
      "loss": 0.47049689292907715,
      "num_input_tokens_seen": 1317108,
      "step": 1719,
      "train_runtime": 3937.247,
      "train_tokens_per_second": 334.525
    },
    {
      "epoch": 0.15816091954022987,
      "grad_norm": 7.859142780303955,
      "learning_rate": 2.8170426065162908e-06,
      "loss": 0.5596014261245728,
      "num_input_tokens_seen": 1317852,
      "step": 1720,
      "train_runtime": 3939.1807,
      "train_tokens_per_second": 334.55
    },
    {
      "epoch": 0.15825287356321838,
      "grad_norm": 6.996805667877197,
      "learning_rate": 2.8070175438596493e-06,
      "loss": 0.5160921812057495,
      "num_input_tokens_seen": 1318620,
      "step": 1721,
      "train_runtime": 3941.1569,
      "train_tokens_per_second": 334.577
    },
    {
      "epoch": 0.1583448275862069,
      "grad_norm": 7.523529529571533,
      "learning_rate": 2.7969924812030073e-06,
      "loss": 0.4985366463661194,
      "num_input_tokens_seen": 1319368,
      "step": 1722,
      "train_runtime": 3943.1009,
      "train_tokens_per_second": 334.602
    },
    {
      "epoch": 0.15843678160919542,
      "grad_norm": 7.76933479309082,
      "learning_rate": 2.7869674185463662e-06,
      "loss": 0.4792497158050537,
      "num_input_tokens_seen": 1320100,
      "step": 1723,
      "train_runtime": 3945.0247,
      "train_tokens_per_second": 334.624
    },
    {
      "epoch": 0.15852873563218392,
      "grad_norm": 7.034414291381836,
      "learning_rate": 2.7769423558897247e-06,
      "loss": 0.41507336497306824,
      "num_input_tokens_seen": 1320980,
      "step": 1724,
      "train_runtime": 3946.9731,
      "train_tokens_per_second": 334.682
    },
    {
      "epoch": 0.15862068965517243,
      "grad_norm": 6.495769500732422,
      "learning_rate": 2.766917293233083e-06,
      "loss": 0.5253562331199646,
      "num_input_tokens_seen": 1322188,
      "step": 1725,
      "train_runtime": 3949.049,
      "train_tokens_per_second": 334.812
    },
    {
      "epoch": 0.15871264367816093,
      "grad_norm": 8.071935653686523,
      "learning_rate": 2.7568922305764413e-06,
      "loss": 0.4798712134361267,
      "num_input_tokens_seen": 1322764,
      "step": 1726,
      "train_runtime": 3950.9752,
      "train_tokens_per_second": 334.794
    },
    {
      "epoch": 0.15880459770114944,
      "grad_norm": 7.1461992263793945,
      "learning_rate": 2.7468671679197994e-06,
      "loss": 0.6246401071548462,
      "num_input_tokens_seen": 1323480,
      "step": 1727,
      "train_runtime": 3952.8856,
      "train_tokens_per_second": 334.814
    },
    {
      "epoch": 0.15889655172413794,
      "grad_norm": 6.823047637939453,
      "learning_rate": 2.7368421052631583e-06,
      "loss": 0.35014140605926514,
      "num_input_tokens_seen": 1324044,
      "step": 1728,
      "train_runtime": 3954.7995,
      "train_tokens_per_second": 334.794
    },
    {
      "epoch": 0.15898850574712645,
      "grad_norm": 6.551754474639893,
      "learning_rate": 2.7268170426065167e-06,
      "loss": 0.5031850337982178,
      "num_input_tokens_seen": 1324724,
      "step": 1729,
      "train_runtime": 3956.7389,
      "train_tokens_per_second": 334.802
    },
    {
      "epoch": 0.15908045977011495,
      "grad_norm": 6.587690830230713,
      "learning_rate": 2.716791979949875e-06,
      "loss": 0.4396337568759918,
      "num_input_tokens_seen": 1325392,
      "step": 1730,
      "train_runtime": 3958.6609,
      "train_tokens_per_second": 334.808
    },
    {
      "epoch": 0.15917241379310346,
      "grad_norm": 6.230381965637207,
      "learning_rate": 2.7067669172932333e-06,
      "loss": 0.48301368951797485,
      "num_input_tokens_seen": 1326004,
      "step": 1731,
      "train_runtime": 3960.5856,
      "train_tokens_per_second": 334.8
    },
    {
      "epoch": 0.15926436781609196,
      "grad_norm": 6.516656398773193,
      "learning_rate": 2.6967418546365914e-06,
      "loss": 0.48590826988220215,
      "num_input_tokens_seen": 1326644,
      "step": 1732,
      "train_runtime": 3962.5059,
      "train_tokens_per_second": 334.799
    },
    {
      "epoch": 0.15935632183908047,
      "grad_norm": 8.143502235412598,
      "learning_rate": 2.6867167919799503e-06,
      "loss": 0.5189724564552307,
      "num_input_tokens_seen": 1327536,
      "step": 1733,
      "train_runtime": 3964.474,
      "train_tokens_per_second": 334.858
    },
    {
      "epoch": 0.15944827586206897,
      "grad_norm": 6.90491247177124,
      "learning_rate": 2.6766917293233088e-06,
      "loss": 0.5360757112503052,
      "num_input_tokens_seen": 1328192,
      "step": 1734,
      "train_runtime": 3966.3916,
      "train_tokens_per_second": 334.862
    },
    {
      "epoch": 0.15954022988505748,
      "grad_norm": 6.260725021362305,
      "learning_rate": 2.666666666666667e-06,
      "loss": 0.42252597212791443,
      "num_input_tokens_seen": 1328852,
      "step": 1735,
      "train_runtime": 3968.3117,
      "train_tokens_per_second": 334.866
    },
    {
      "epoch": 0.15963218390804598,
      "grad_norm": 5.750924110412598,
      "learning_rate": 2.6566416040100253e-06,
      "loss": 0.45929282903671265,
      "num_input_tokens_seen": 1330104,
      "step": 1736,
      "train_runtime": 3970.3284,
      "train_tokens_per_second": 335.011
    },
    {
      "epoch": 0.1597241379310345,
      "grad_norm": 7.40118408203125,
      "learning_rate": 2.6466165413533834e-06,
      "loss": 0.4813888967037201,
      "num_input_tokens_seen": 1330680,
      "step": 1737,
      "train_runtime": 3972.2401,
      "train_tokens_per_second": 334.995
    },
    {
      "epoch": 0.159816091954023,
      "grad_norm": 7.092006206512451,
      "learning_rate": 2.6365914786967423e-06,
      "loss": 0.4561431407928467,
      "num_input_tokens_seen": 1331332,
      "step": 1738,
      "train_runtime": 3974.1563,
      "train_tokens_per_second": 334.997
    },
    {
      "epoch": 0.1599080459770115,
      "grad_norm": 7.546247959136963,
      "learning_rate": 2.6265664160401004e-06,
      "loss": 0.4432491660118103,
      "num_input_tokens_seen": 1332064,
      "step": 1739,
      "train_runtime": 3976.0804,
      "train_tokens_per_second": 335.019
    },
    {
      "epoch": 0.16,
      "grad_norm": 7.441892147064209,
      "learning_rate": 2.616541353383459e-06,
      "loss": 0.46018320322036743,
      "num_input_tokens_seen": 1332656,
      "step": 1740,
      "train_runtime": 3977.9985,
      "train_tokens_per_second": 335.007
    },
    {
      "epoch": 0.1600919540229885,
      "grad_norm": 6.686703681945801,
      "learning_rate": 2.606516290726817e-06,
      "loss": 0.4715593457221985,
      "num_input_tokens_seen": 1333284,
      "step": 1741,
      "train_runtime": 3986.5354,
      "train_tokens_per_second": 334.447
    },
    {
      "epoch": 0.160183908045977,
      "grad_norm": 7.182392120361328,
      "learning_rate": 2.5964912280701754e-06,
      "loss": 0.5060937404632568,
      "num_input_tokens_seen": 1333896,
      "step": 1742,
      "train_runtime": 3988.4518,
      "train_tokens_per_second": 334.44
    },
    {
      "epoch": 0.16027586206896552,
      "grad_norm": 6.546034812927246,
      "learning_rate": 2.5864661654135343e-06,
      "loss": 0.41727566719055176,
      "num_input_tokens_seen": 1334504,
      "step": 1743,
      "train_runtime": 3990.4248,
      "train_tokens_per_second": 334.427
    },
    {
      "epoch": 0.16036781609195402,
      "grad_norm": 6.587306022644043,
      "learning_rate": 2.5764411027568924e-06,
      "loss": 0.43473294377326965,
      "num_input_tokens_seen": 1335188,
      "step": 1744,
      "train_runtime": 3992.3309,
      "train_tokens_per_second": 334.438
    },
    {
      "epoch": 0.16045977011494253,
      "grad_norm": 6.231787204742432,
      "learning_rate": 2.566416040100251e-06,
      "loss": 0.4099838137626648,
      "num_input_tokens_seen": 1335752,
      "step": 1745,
      "train_runtime": 3994.232,
      "train_tokens_per_second": 334.42
    },
    {
      "epoch": 0.16055172413793103,
      "grad_norm": 6.253069877624512,
      "learning_rate": 2.556390977443609e-06,
      "loss": 0.42506536841392517,
      "num_input_tokens_seen": 1336388,
      "step": 1746,
      "train_runtime": 3996.2888,
      "train_tokens_per_second": 334.407
    },
    {
      "epoch": 0.16064367816091954,
      "grad_norm": 9.03331470489502,
      "learning_rate": 2.5463659147869674e-06,
      "loss": 0.5423241853713989,
      "num_input_tokens_seen": 1337136,
      "step": 1747,
      "train_runtime": 3998.2712,
      "train_tokens_per_second": 334.429
    },
    {
      "epoch": 0.16073563218390804,
      "grad_norm": 7.214175224304199,
      "learning_rate": 2.5363408521303263e-06,
      "loss": 0.5407247543334961,
      "num_input_tokens_seen": 1337972,
      "step": 1748,
      "train_runtime": 4000.2375,
      "train_tokens_per_second": 334.473
    },
    {
      "epoch": 0.16082758620689655,
      "grad_norm": 6.139636993408203,
      "learning_rate": 2.5263157894736844e-06,
      "loss": 0.40379995107650757,
      "num_input_tokens_seen": 1338576,
      "step": 1749,
      "train_runtime": 4002.1723,
      "train_tokens_per_second": 334.462
    },
    {
      "epoch": 0.16091954022988506,
      "grad_norm": 6.8690385818481445,
      "learning_rate": 2.516290726817043e-06,
      "loss": 0.47756439447402954,
      "num_input_tokens_seen": 1339544,
      "step": 1750,
      "train_runtime": 4004.1659,
      "train_tokens_per_second": 334.538
    },
    {
      "epoch": 0.16101149425287356,
      "grad_norm": 7.211779594421387,
      "learning_rate": 2.506265664160401e-06,
      "loss": 0.4632889926433563,
      "num_input_tokens_seen": 1340204,
      "step": 1751,
      "train_runtime": 4006.1382,
      "train_tokens_per_second": 334.538
    },
    {
      "epoch": 0.16110344827586207,
      "grad_norm": 6.297443389892578,
      "learning_rate": 2.4962406015037594e-06,
      "loss": 0.45367634296417236,
      "num_input_tokens_seen": 1341040,
      "step": 1752,
      "train_runtime": 4008.128,
      "train_tokens_per_second": 334.58
    },
    {
      "epoch": 0.16119540229885057,
      "grad_norm": 5.939846992492676,
      "learning_rate": 2.486215538847118e-06,
      "loss": 0.372577965259552,
      "num_input_tokens_seen": 1341660,
      "step": 1753,
      "train_runtime": 4010.0567,
      "train_tokens_per_second": 334.574
    },
    {
      "epoch": 0.16128735632183908,
      "grad_norm": 6.601139545440674,
      "learning_rate": 2.4761904761904764e-06,
      "loss": 0.4091433882713318,
      "num_input_tokens_seen": 1342384,
      "step": 1754,
      "train_runtime": 4011.9916,
      "train_tokens_per_second": 334.593
    },
    {
      "epoch": 0.16137931034482758,
      "grad_norm": 6.731122970581055,
      "learning_rate": 2.466165413533835e-06,
      "loss": 0.44458529353141785,
      "num_input_tokens_seen": 1342952,
      "step": 1755,
      "train_runtime": 4013.9086,
      "train_tokens_per_second": 334.575
    },
    {
      "epoch": 0.1614712643678161,
      "grad_norm": 7.471959114074707,
      "learning_rate": 2.456140350877193e-06,
      "loss": 0.5619382858276367,
      "num_input_tokens_seen": 1343872,
      "step": 1756,
      "train_runtime": 4015.8825,
      "train_tokens_per_second": 334.639
    },
    {
      "epoch": 0.1615632183908046,
      "grad_norm": 6.827757835388184,
      "learning_rate": 2.4461152882205514e-06,
      "loss": 0.3854348361492157,
      "num_input_tokens_seen": 1344516,
      "step": 1757,
      "train_runtime": 4017.7989,
      "train_tokens_per_second": 334.64
    },
    {
      "epoch": 0.1616551724137931,
      "grad_norm": 6.161599159240723,
      "learning_rate": 2.43609022556391e-06,
      "loss": 0.49510788917541504,
      "num_input_tokens_seen": 1345892,
      "step": 1758,
      "train_runtime": 4019.8311,
      "train_tokens_per_second": 334.813
    },
    {
      "epoch": 0.1617471264367816,
      "grad_norm": 6.267351150512695,
      "learning_rate": 2.4260651629072684e-06,
      "loss": 0.4766969382762909,
      "num_input_tokens_seen": 1346696,
      "step": 1759,
      "train_runtime": 4021.7709,
      "train_tokens_per_second": 334.851
    },
    {
      "epoch": 0.1618390804597701,
      "grad_norm": 10.214824676513672,
      "learning_rate": 2.416040100250627e-06,
      "loss": 0.5045785307884216,
      "num_input_tokens_seen": 1347484,
      "step": 1760,
      "train_runtime": 4023.73,
      "train_tokens_per_second": 334.884
    },
    {
      "epoch": 0.1619310344827586,
      "grad_norm": 7.075687408447266,
      "learning_rate": 2.406015037593985e-06,
      "loss": 0.3947675824165344,
      "num_input_tokens_seen": 1348020,
      "step": 1761,
      "train_runtime": 4025.6301,
      "train_tokens_per_second": 334.859
    },
    {
      "epoch": 0.16202298850574712,
      "grad_norm": 6.410388946533203,
      "learning_rate": 2.3959899749373435e-06,
      "loss": 0.4875537157058716,
      "num_input_tokens_seen": 1348708,
      "step": 1762,
      "train_runtime": 4027.5379,
      "train_tokens_per_second": 334.872
    },
    {
      "epoch": 0.16211494252873562,
      "grad_norm": 6.277807712554932,
      "learning_rate": 2.385964912280702e-06,
      "loss": 0.4184660315513611,
      "num_input_tokens_seen": 1349424,
      "step": 1763,
      "train_runtime": 4029.4461,
      "train_tokens_per_second": 334.891
    },
    {
      "epoch": 0.16220689655172413,
      "grad_norm": 7.242740154266357,
      "learning_rate": 2.3759398496240604e-06,
      "loss": 0.49717608094215393,
      "num_input_tokens_seen": 1350448,
      "step": 1764,
      "train_runtime": 4031.4138,
      "train_tokens_per_second": 334.981
    },
    {
      "epoch": 0.16229885057471263,
      "grad_norm": 6.5039143562316895,
      "learning_rate": 2.365914786967419e-06,
      "loss": 0.424640953540802,
      "num_input_tokens_seen": 1351032,
      "step": 1765,
      "train_runtime": 4033.3334,
      "train_tokens_per_second": 334.967
    },
    {
      "epoch": 0.16239080459770114,
      "grad_norm": 7.855981349945068,
      "learning_rate": 2.355889724310777e-06,
      "loss": 0.4662795066833496,
      "num_input_tokens_seen": 1352144,
      "step": 1766,
      "train_runtime": 4035.3466,
      "train_tokens_per_second": 335.075
    },
    {
      "epoch": 0.16248275862068964,
      "grad_norm": 6.1420063972473145,
      "learning_rate": 2.3458646616541355e-06,
      "loss": 0.4422057867050171,
      "num_input_tokens_seen": 1352836,
      "step": 1767,
      "train_runtime": 4037.2767,
      "train_tokens_per_second": 335.086
    },
    {
      "epoch": 0.16257471264367815,
      "grad_norm": 7.944182395935059,
      "learning_rate": 2.335839598997494e-06,
      "loss": 0.476218044757843,
      "num_input_tokens_seen": 1353508,
      "step": 1768,
      "train_runtime": 4039.1875,
      "train_tokens_per_second": 335.094
    },
    {
      "epoch": 0.16266666666666665,
      "grad_norm": 6.096160411834717,
      "learning_rate": 2.3258145363408524e-06,
      "loss": 0.40059059858322144,
      "num_input_tokens_seen": 1354336,
      "step": 1769,
      "train_runtime": 4041.1302,
      "train_tokens_per_second": 335.138
    },
    {
      "epoch": 0.16275862068965516,
      "grad_norm": 6.71887731552124,
      "learning_rate": 2.3157894736842105e-06,
      "loss": 0.4041825532913208,
      "num_input_tokens_seen": 1354996,
      "step": 1770,
      "train_runtime": 4043.0408,
      "train_tokens_per_second": 335.143
    },
    {
      "epoch": 0.1628505747126437,
      "grad_norm": 6.223047733306885,
      "learning_rate": 2.305764411027569e-06,
      "loss": 0.4100489020347595,
      "num_input_tokens_seen": 1355584,
      "step": 1771,
      "train_runtime": 4051.5699,
      "train_tokens_per_second": 334.582
    },
    {
      "epoch": 0.1629425287356322,
      "grad_norm": 6.081003189086914,
      "learning_rate": 2.2957393483709275e-06,
      "loss": 0.41684621572494507,
      "num_input_tokens_seen": 1356148,
      "step": 1772,
      "train_runtime": 4053.6839,
      "train_tokens_per_second": 334.547
    },
    {
      "epoch": 0.1630344827586207,
      "grad_norm": 7.356428146362305,
      "learning_rate": 2.285714285714286e-06,
      "loss": 0.4125157594680786,
      "num_input_tokens_seen": 1356816,
      "step": 1773,
      "train_runtime": 4055.6857,
      "train_tokens_per_second": 334.547
    },
    {
      "epoch": 0.1631264367816092,
      "grad_norm": 6.269641399383545,
      "learning_rate": 2.2756892230576445e-06,
      "loss": 0.5003457069396973,
      "num_input_tokens_seen": 1357496,
      "step": 1774,
      "train_runtime": 4057.6035,
      "train_tokens_per_second": 334.556
    },
    {
      "epoch": 0.1632183908045977,
      "grad_norm": 6.636397838592529,
      "learning_rate": 2.2656641604010025e-06,
      "loss": 0.3927608132362366,
      "num_input_tokens_seen": 1358084,
      "step": 1775,
      "train_runtime": 4059.5124,
      "train_tokens_per_second": 334.544
    },
    {
      "epoch": 0.16331034482758622,
      "grad_norm": 6.633103370666504,
      "learning_rate": 2.255639097744361e-06,
      "loss": 0.5156530141830444,
      "num_input_tokens_seen": 1359024,
      "step": 1776,
      "train_runtime": 4061.4703,
      "train_tokens_per_second": 334.614
    },
    {
      "epoch": 0.16340229885057472,
      "grad_norm": 7.025842666625977,
      "learning_rate": 2.2456140350877195e-06,
      "loss": 0.42199695110321045,
      "num_input_tokens_seen": 1359540,
      "step": 1777,
      "train_runtime": 4063.3759,
      "train_tokens_per_second": 334.584
    },
    {
      "epoch": 0.16349425287356323,
      "grad_norm": 7.5587286949157715,
      "learning_rate": 2.235588972431078e-06,
      "loss": 0.4943675398826599,
      "num_input_tokens_seen": 1360224,
      "step": 1778,
      "train_runtime": 4065.3012,
      "train_tokens_per_second": 334.594
    },
    {
      "epoch": 0.16358620689655173,
      "grad_norm": 7.1349568367004395,
      "learning_rate": 2.2255639097744365e-06,
      "loss": 0.6079813241958618,
      "num_input_tokens_seen": 1361088,
      "step": 1779,
      "train_runtime": 4067.2138,
      "train_tokens_per_second": 334.649
    },
    {
      "epoch": 0.16367816091954024,
      "grad_norm": 5.987335205078125,
      "learning_rate": 2.2155388471177945e-06,
      "loss": 0.43679770827293396,
      "num_input_tokens_seen": 1361780,
      "step": 1780,
      "train_runtime": 4069.1329,
      "train_tokens_per_second": 334.661
    },
    {
      "epoch": 0.16377011494252874,
      "grad_norm": 5.992737293243408,
      "learning_rate": 2.205513784461153e-06,
      "loss": 0.367605060338974,
      "num_input_tokens_seen": 1362384,
      "step": 1781,
      "train_runtime": 4071.0538,
      "train_tokens_per_second": 334.651
    },
    {
      "epoch": 0.16386206896551725,
      "grad_norm": 7.0040283203125,
      "learning_rate": 2.1954887218045115e-06,
      "loss": 0.4569029211997986,
      "num_input_tokens_seen": 1363044,
      "step": 1782,
      "train_runtime": 4073.0035,
      "train_tokens_per_second": 334.653
    },
    {
      "epoch": 0.16395402298850575,
      "grad_norm": 6.85443639755249,
      "learning_rate": 2.1854636591478696e-06,
      "loss": 0.5083461403846741,
      "num_input_tokens_seen": 1363688,
      "step": 1783,
      "train_runtime": 4074.9377,
      "train_tokens_per_second": 334.652
    },
    {
      "epoch": 0.16404597701149426,
      "grad_norm": 6.63103723526001,
      "learning_rate": 2.1754385964912285e-06,
      "loss": 0.4410957396030426,
      "num_input_tokens_seen": 1364356,
      "step": 1784,
      "train_runtime": 4076.8597,
      "train_tokens_per_second": 334.659
    },
    {
      "epoch": 0.16413793103448276,
      "grad_norm": 6.289424419403076,
      "learning_rate": 2.1654135338345866e-06,
      "loss": 0.49302607774734497,
      "num_input_tokens_seen": 1365028,
      "step": 1785,
      "train_runtime": 4078.768,
      "train_tokens_per_second": 334.667
    },
    {
      "epoch": 0.16422988505747127,
      "grad_norm": 6.578011512756348,
      "learning_rate": 2.155388471177945e-06,
      "loss": 0.38957542181015015,
      "num_input_tokens_seen": 1365668,
      "step": 1786,
      "train_runtime": 4080.6737,
      "train_tokens_per_second": 334.667
    },
    {
      "epoch": 0.16432183908045978,
      "grad_norm": 7.1062798500061035,
      "learning_rate": 2.1453634085213035e-06,
      "loss": 0.41782549023628235,
      "num_input_tokens_seen": 1366204,
      "step": 1787,
      "train_runtime": 4082.5966,
      "train_tokens_per_second": 334.641
    },
    {
      "epoch": 0.16441379310344828,
      "grad_norm": 7.929337978363037,
      "learning_rate": 2.1353383458646616e-06,
      "loss": 0.4896957278251648,
      "num_input_tokens_seen": 1366752,
      "step": 1788,
      "train_runtime": 4084.5871,
      "train_tokens_per_second": 334.612
    },
    {
      "epoch": 0.16450574712643679,
      "grad_norm": 5.988348007202148,
      "learning_rate": 2.1253132832080205e-06,
      "loss": 0.4074459373950958,
      "num_input_tokens_seen": 1367320,
      "step": 1789,
      "train_runtime": 4086.4942,
      "train_tokens_per_second": 334.595
    },
    {
      "epoch": 0.1645977011494253,
      "grad_norm": 7.091726779937744,
      "learning_rate": 2.1152882205513786e-06,
      "loss": 0.4936490058898926,
      "num_input_tokens_seen": 1368120,
      "step": 1790,
      "train_runtime": 4088.4181,
      "train_tokens_per_second": 334.633
    },
    {
      "epoch": 0.1646896551724138,
      "grad_norm": 6.799236297607422,
      "learning_rate": 2.105263157894737e-06,
      "loss": 0.502659022808075,
      "num_input_tokens_seen": 1368840,
      "step": 1791,
      "train_runtime": 4090.3316,
      "train_tokens_per_second": 334.653
    },
    {
      "epoch": 0.1647816091954023,
      "grad_norm": 6.0580363273620605,
      "learning_rate": 2.0952380952380955e-06,
      "loss": 0.4692020118236542,
      "num_input_tokens_seen": 1369660,
      "step": 1792,
      "train_runtime": 4092.3518,
      "train_tokens_per_second": 334.688
    },
    {
      "epoch": 0.1648735632183908,
      "grad_norm": 6.381335735321045,
      "learning_rate": 2.0852130325814536e-06,
      "loss": 0.38795846700668335,
      "num_input_tokens_seen": 1370264,
      "step": 1793,
      "train_runtime": 4094.2678,
      "train_tokens_per_second": 334.679
    },
    {
      "epoch": 0.1649655172413793,
      "grad_norm": 6.306175708770752,
      "learning_rate": 2.075187969924812e-06,
      "loss": 0.6098124384880066,
      "num_input_tokens_seen": 1371552,
      "step": 1794,
      "train_runtime": 4096.2733,
      "train_tokens_per_second": 334.829
    },
    {
      "epoch": 0.16505747126436782,
      "grad_norm": 7.6732587814331055,
      "learning_rate": 2.0651629072681706e-06,
      "loss": 0.5691096782684326,
      "num_input_tokens_seen": 1372264,
      "step": 1795,
      "train_runtime": 4098.2038,
      "train_tokens_per_second": 334.845
    },
    {
      "epoch": 0.16514942528735632,
      "grad_norm": 6.454239845275879,
      "learning_rate": 2.055137844611529e-06,
      "loss": 0.4893447160720825,
      "num_input_tokens_seen": 1373100,
      "step": 1796,
      "train_runtime": 4100.2212,
      "train_tokens_per_second": 334.884
    },
    {
      "epoch": 0.16524137931034483,
      "grad_norm": 7.020412921905518,
      "learning_rate": 2.0451127819548876e-06,
      "loss": 0.49594277143478394,
      "num_input_tokens_seen": 1373912,
      "step": 1797,
      "train_runtime": 4102.1636,
      "train_tokens_per_second": 334.924
    },
    {
      "epoch": 0.16533333333333333,
      "grad_norm": 7.460563659667969,
      "learning_rate": 2.0350877192982456e-06,
      "loss": 0.4521038234233856,
      "num_input_tokens_seen": 1374544,
      "step": 1798,
      "train_runtime": 4104.0742,
      "train_tokens_per_second": 334.922
    },
    {
      "epoch": 0.16542528735632184,
      "grad_norm": 6.527820110321045,
      "learning_rate": 2.025062656641604e-06,
      "loss": 0.4866308271884918,
      "num_input_tokens_seen": 1375344,
      "step": 1799,
      "train_runtime": 4105.987,
      "train_tokens_per_second": 334.961
    },
    {
      "epoch": 0.16551724137931034,
      "grad_norm": 6.184874534606934,
      "learning_rate": 2.0150375939849626e-06,
      "loss": 0.4070171117782593,
      "num_input_tokens_seen": 1375940,
      "step": 1800,
      "train_runtime": 4107.9003,
      "train_tokens_per_second": 334.95
    },
    {
      "epoch": 0.16551724137931034,
      "eval_loss": 1.7983349561691284,
      "eval_runtime": 27.8808,
      "eval_samples_per_second": 35.867,
      "eval_steps_per_second": 8.967,
      "num_input_tokens_seen": 1375940,
      "step": 1800
    },
    {
      "epoch": 0.16560919540229885,
      "grad_norm": 6.80704402923584,
      "learning_rate": 2.0050125313283207e-06,
      "loss": 0.48958921432495117,
      "num_input_tokens_seen": 1376552,
      "step": 1801,
      "train_runtime": 4144.6166,
      "train_tokens_per_second": 332.13
    },
    {
      "epoch": 0.16570114942528735,
      "grad_norm": 6.7276835441589355,
      "learning_rate": 1.9949874686716796e-06,
      "loss": 0.47152501344680786,
      "num_input_tokens_seen": 1377364,
      "step": 1802,
      "train_runtime": 4146.5338,
      "train_tokens_per_second": 332.172
    },
    {
      "epoch": 0.16579310344827586,
      "grad_norm": 6.990015983581543,
      "learning_rate": 1.9849624060150376e-06,
      "loss": 0.36413344740867615,
      "num_input_tokens_seen": 1377880,
      "step": 1803,
      "train_runtime": 4148.44,
      "train_tokens_per_second": 332.144
    },
    {
      "epoch": 0.16588505747126436,
      "grad_norm": 6.266055583953857,
      "learning_rate": 1.974937343358396e-06,
      "loss": 0.467947393655777,
      "num_input_tokens_seen": 1379668,
      "step": 1804,
      "train_runtime": 4150.5003,
      "train_tokens_per_second": 332.41
    },
    {
      "epoch": 0.16597701149425287,
      "grad_norm": 7.12498664855957,
      "learning_rate": 1.9649122807017546e-06,
      "loss": 0.4401271343231201,
      "num_input_tokens_seen": 1380344,
      "step": 1805,
      "train_runtime": 4152.4249,
      "train_tokens_per_second": 332.419
    },
    {
      "epoch": 0.16606896551724137,
      "grad_norm": 7.541736602783203,
      "learning_rate": 1.9548872180451127e-06,
      "loss": 0.5291580557823181,
      "num_input_tokens_seen": 1381068,
      "step": 1806,
      "train_runtime": 4154.3356,
      "train_tokens_per_second": 332.44
    },
    {
      "epoch": 0.16616091954022988,
      "grad_norm": 6.489492416381836,
      "learning_rate": 1.9448621553884716e-06,
      "loss": 0.4885668158531189,
      "num_input_tokens_seen": 1381812,
      "step": 1807,
      "train_runtime": 4156.257,
      "train_tokens_per_second": 332.465
    },
    {
      "epoch": 0.16625287356321838,
      "grad_norm": 6.762707233428955,
      "learning_rate": 1.9348370927318297e-06,
      "loss": 0.5028106570243835,
      "num_input_tokens_seen": 1382684,
      "step": 1808,
      "train_runtime": 4158.1738,
      "train_tokens_per_second": 332.522
    },
    {
      "epoch": 0.1663448275862069,
      "grad_norm": 6.673324108123779,
      "learning_rate": 1.924812030075188e-06,
      "loss": 0.4497567415237427,
      "num_input_tokens_seen": 1383216,
      "step": 1809,
      "train_runtime": 4160.0875,
      "train_tokens_per_second": 332.497
    },
    {
      "epoch": 0.1664367816091954,
      "grad_norm": 6.737390518188477,
      "learning_rate": 1.9147869674185466e-06,
      "loss": 0.4985470175743103,
      "num_input_tokens_seen": 1384276,
      "step": 1810,
      "train_runtime": 4162.0671,
      "train_tokens_per_second": 332.593
    },
    {
      "epoch": 0.1665287356321839,
      "grad_norm": 6.969824314117432,
      "learning_rate": 1.904761904761905e-06,
      "loss": 0.4589076638221741,
      "num_input_tokens_seen": 1384988,
      "step": 1811,
      "train_runtime": 4163.9821,
      "train_tokens_per_second": 332.611
    },
    {
      "epoch": 0.1666206896551724,
      "grad_norm": 6.854494571685791,
      "learning_rate": 1.8947368421052634e-06,
      "loss": 0.5440043807029724,
      "num_input_tokens_seen": 1385884,
      "step": 1812,
      "train_runtime": 4165.9222,
      "train_tokens_per_second": 332.672
    },
    {
      "epoch": 0.1667126436781609,
      "grad_norm": 6.483313083648682,
      "learning_rate": 1.8847117794486217e-06,
      "loss": 0.39688774943351746,
      "num_input_tokens_seen": 1386504,
      "step": 1813,
      "train_runtime": 4167.8278,
      "train_tokens_per_second": 332.668
    },
    {
      "epoch": 0.16680459770114942,
      "grad_norm": 6.542842864990234,
      "learning_rate": 1.8746867167919802e-06,
      "loss": 0.43845048546791077,
      "num_input_tokens_seen": 1387280,
      "step": 1814,
      "train_runtime": 4169.8246,
      "train_tokens_per_second": 332.695
    },
    {
      "epoch": 0.16689655172413792,
      "grad_norm": 7.002591133117676,
      "learning_rate": 1.8646616541353384e-06,
      "loss": 0.5158287286758423,
      "num_input_tokens_seen": 1387924,
      "step": 1815,
      "train_runtime": 4171.7613,
      "train_tokens_per_second": 332.695
    },
    {
      "epoch": 0.16698850574712643,
      "grad_norm": 7.141075134277344,
      "learning_rate": 1.8546365914786967e-06,
      "loss": 0.5292612314224243,
      "num_input_tokens_seen": 1388844,
      "step": 1816,
      "train_runtime": 4173.7319,
      "train_tokens_per_second": 332.758
    },
    {
      "epoch": 0.16708045977011493,
      "grad_norm": 6.492411136627197,
      "learning_rate": 1.8446115288220554e-06,
      "loss": 0.4482610821723938,
      "num_input_tokens_seen": 1389604,
      "step": 1817,
      "train_runtime": 4175.6523,
      "train_tokens_per_second": 332.787
    },
    {
      "epoch": 0.16717241379310344,
      "grad_norm": 6.330272674560547,
      "learning_rate": 1.8345864661654137e-06,
      "loss": 0.39253392815589905,
      "num_input_tokens_seen": 1390692,
      "step": 1818,
      "train_runtime": 4177.6354,
      "train_tokens_per_second": 332.89
    },
    {
      "epoch": 0.16726436781609194,
      "grad_norm": 6.166111946105957,
      "learning_rate": 1.8245614035087722e-06,
      "loss": 0.46502041816711426,
      "num_input_tokens_seen": 1391364,
      "step": 1819,
      "train_runtime": 4179.5684,
      "train_tokens_per_second": 332.897
    },
    {
      "epoch": 0.16735632183908045,
      "grad_norm": 6.842964172363281,
      "learning_rate": 1.8145363408521305e-06,
      "loss": 0.6573138236999512,
      "num_input_tokens_seen": 1392472,
      "step": 1820,
      "train_runtime": 4181.5446,
      "train_tokens_per_second": 333.004
    },
    {
      "epoch": 0.16744827586206898,
      "grad_norm": 7.465213775634766,
      "learning_rate": 1.8045112781954887e-06,
      "loss": 0.4444107115268707,
      "num_input_tokens_seen": 1393228,
      "step": 1821,
      "train_runtime": 4183.5082,
      "train_tokens_per_second": 333.029
    },
    {
      "epoch": 0.16754022988505748,
      "grad_norm": 6.911365509033203,
      "learning_rate": 1.7944862155388474e-06,
      "loss": 0.36707374453544617,
      "num_input_tokens_seen": 1393876,
      "step": 1822,
      "train_runtime": 4185.4746,
      "train_tokens_per_second": 333.027
    },
    {
      "epoch": 0.167632183908046,
      "grad_norm": 7.916104793548584,
      "learning_rate": 1.7844611528822057e-06,
      "loss": 0.5797393321990967,
      "num_input_tokens_seen": 1394536,
      "step": 1823,
      "train_runtime": 4187.4005,
      "train_tokens_per_second": 333.031
    },
    {
      "epoch": 0.1677241379310345,
      "grad_norm": 6.57176399230957,
      "learning_rate": 1.7744360902255642e-06,
      "loss": 0.3756246268749237,
      "num_input_tokens_seen": 1395016,
      "step": 1824,
      "train_runtime": 4189.3225,
      "train_tokens_per_second": 332.993
    },
    {
      "epoch": 0.167816091954023,
      "grad_norm": 6.847811698913574,
      "learning_rate": 1.7644110275689225e-06,
      "loss": 0.513152003288269,
      "num_input_tokens_seen": 1395924,
      "step": 1825,
      "train_runtime": 4191.2987,
      "train_tokens_per_second": 333.053
    },
    {
      "epoch": 0.1679080459770115,
      "grad_norm": 6.873049259185791,
      "learning_rate": 1.7543859649122807e-06,
      "loss": 0.41355061531066895,
      "num_input_tokens_seen": 1396544,
      "step": 1826,
      "train_runtime": 4193.2885,
      "train_tokens_per_second": 333.043
    },
    {
      "epoch": 0.168,
      "grad_norm": 6.391554355621338,
      "learning_rate": 1.7443609022556392e-06,
      "loss": 0.472981333732605,
      "num_input_tokens_seen": 1397468,
      "step": 1827,
      "train_runtime": 4195.2428,
      "train_tokens_per_second": 333.108
    },
    {
      "epoch": 0.16809195402298852,
      "grad_norm": 6.501170635223389,
      "learning_rate": 1.7343358395989975e-06,
      "loss": 0.45052164793014526,
      "num_input_tokens_seen": 1398316,
      "step": 1828,
      "train_runtime": 4197.2152,
      "train_tokens_per_second": 333.153
    },
    {
      "epoch": 0.16818390804597702,
      "grad_norm": 6.45281982421875,
      "learning_rate": 1.7243107769423562e-06,
      "loss": 0.4184761941432953,
      "num_input_tokens_seen": 1398968,
      "step": 1829,
      "train_runtime": 4199.1346,
      "train_tokens_per_second": 333.156
    },
    {
      "epoch": 0.16827586206896553,
      "grad_norm": 6.822763442993164,
      "learning_rate": 1.7142857142857145e-06,
      "loss": 0.5204285383224487,
      "num_input_tokens_seen": 1399840,
      "step": 1830,
      "train_runtime": 4201.0664,
      "train_tokens_per_second": 333.211
    },
    {
      "epoch": 0.16836781609195403,
      "grad_norm": 9.192536354064941,
      "learning_rate": 1.7042606516290728e-06,
      "loss": 0.4176218807697296,
      "num_input_tokens_seen": 1400404,
      "step": 1831,
      "train_runtime": 4209.5387,
      "train_tokens_per_second": 332.674
    },
    {
      "epoch": 0.16845977011494254,
      "grad_norm": 6.564760208129883,
      "learning_rate": 1.6942355889724312e-06,
      "loss": 0.48810386657714844,
      "num_input_tokens_seen": 1401172,
      "step": 1832,
      "train_runtime": 4211.4808,
      "train_tokens_per_second": 332.703
    },
    {
      "epoch": 0.16855172413793104,
      "grad_norm": 7.919020175933838,
      "learning_rate": 1.6842105263157895e-06,
      "loss": 0.4304644465446472,
      "num_input_tokens_seen": 1402288,
      "step": 1833,
      "train_runtime": 4213.4952,
      "train_tokens_per_second": 332.809
    },
    {
      "epoch": 0.16864367816091955,
      "grad_norm": 6.425091743469238,
      "learning_rate": 1.6741854636591482e-06,
      "loss": 0.5326110124588013,
      "num_input_tokens_seen": 1403204,
      "step": 1834,
      "train_runtime": 4215.4436,
      "train_tokens_per_second": 332.872
    },
    {
      "epoch": 0.16873563218390805,
      "grad_norm": 7.043296813964844,
      "learning_rate": 1.6641604010025065e-06,
      "loss": 0.4600408375263214,
      "num_input_tokens_seen": 1403784,
      "step": 1835,
      "train_runtime": 4217.4279,
      "train_tokens_per_second": 332.853
    },
    {
      "epoch": 0.16882758620689656,
      "grad_norm": 6.777863025665283,
      "learning_rate": 1.6541353383458648e-06,
      "loss": 0.4287620782852173,
      "num_input_tokens_seen": 1404356,
      "step": 1836,
      "train_runtime": 4219.345,
      "train_tokens_per_second": 332.837
    },
    {
      "epoch": 0.16891954022988506,
      "grad_norm": 6.217686653137207,
      "learning_rate": 1.6441102756892233e-06,
      "loss": 0.41356709599494934,
      "num_input_tokens_seen": 1405220,
      "step": 1837,
      "train_runtime": 4221.298,
      "train_tokens_per_second": 332.888
    },
    {
      "epoch": 0.16901149425287357,
      "grad_norm": 6.92685079574585,
      "learning_rate": 1.6340852130325815e-06,
      "loss": 0.478983610868454,
      "num_input_tokens_seen": 1405900,
      "step": 1838,
      "train_runtime": 4223.2438,
      "train_tokens_per_second": 332.896
    },
    {
      "epoch": 0.16910344827586207,
      "grad_norm": 6.960223197937012,
      "learning_rate": 1.62406015037594e-06,
      "loss": 0.4747348725795746,
      "num_input_tokens_seen": 1406604,
      "step": 1839,
      "train_runtime": 4225.159,
      "train_tokens_per_second": 332.911
    },
    {
      "epoch": 0.16919540229885058,
      "grad_norm": 7.196669578552246,
      "learning_rate": 1.6140350877192983e-06,
      "loss": 0.5055457949638367,
      "num_input_tokens_seen": 1407376,
      "step": 1840,
      "train_runtime": 4227.0853,
      "train_tokens_per_second": 332.942
    },
    {
      "epoch": 0.16928735632183908,
      "grad_norm": 9.721369743347168,
      "learning_rate": 1.6040100250626568e-06,
      "loss": 0.4707803428173065,
      "num_input_tokens_seen": 1408240,
      "step": 1841,
      "train_runtime": 4229.0363,
      "train_tokens_per_second": 332.993
    },
    {
      "epoch": 0.1693793103448276,
      "grad_norm": 8.509045600891113,
      "learning_rate": 1.5939849624060153e-06,
      "loss": 0.44657400250434875,
      "num_input_tokens_seen": 1408852,
      "step": 1842,
      "train_runtime": 4230.9479,
      "train_tokens_per_second": 332.987
    },
    {
      "epoch": 0.1694712643678161,
      "grad_norm": 6.2532734870910645,
      "learning_rate": 1.5839598997493736e-06,
      "loss": 0.4338662028312683,
      "num_input_tokens_seen": 1409704,
      "step": 1843,
      "train_runtime": 4232.8889,
      "train_tokens_per_second": 333.036
    },
    {
      "epoch": 0.1695632183908046,
      "grad_norm": 7.739737510681152,
      "learning_rate": 1.573934837092732e-06,
      "loss": 0.47888562083244324,
      "num_input_tokens_seen": 1410332,
      "step": 1844,
      "train_runtime": 4234.7973,
      "train_tokens_per_second": 333.034
    },
    {
      "epoch": 0.1696551724137931,
      "grad_norm": 6.617823600769043,
      "learning_rate": 1.5639097744360903e-06,
      "loss": 0.49698400497436523,
      "num_input_tokens_seen": 1411516,
      "step": 1845,
      "train_runtime": 4236.7844,
      "train_tokens_per_second": 333.157
    },
    {
      "epoch": 0.1697471264367816,
      "grad_norm": 7.113321781158447,
      "learning_rate": 1.5538847117794486e-06,
      "loss": 0.5171873569488525,
      "num_input_tokens_seen": 1412100,
      "step": 1846,
      "train_runtime": 4238.7068,
      "train_tokens_per_second": 333.144
    },
    {
      "epoch": 0.16983908045977011,
      "grad_norm": 8.158944129943848,
      "learning_rate": 1.5438596491228073e-06,
      "loss": 0.42578595876693726,
      "num_input_tokens_seen": 1412776,
      "step": 1847,
      "train_runtime": 4240.6054,
      "train_tokens_per_second": 333.154
    },
    {
      "epoch": 0.16993103448275862,
      "grad_norm": 6.14885139465332,
      "learning_rate": 1.5338345864661656e-06,
      "loss": 0.4274827837944031,
      "num_input_tokens_seen": 1413512,
      "step": 1848,
      "train_runtime": 4242.5489,
      "train_tokens_per_second": 333.175
    },
    {
      "epoch": 0.17002298850574712,
      "grad_norm": 6.187997341156006,
      "learning_rate": 1.523809523809524e-06,
      "loss": 0.4465654194355011,
      "num_input_tokens_seen": 1414096,
      "step": 1849,
      "train_runtime": 4244.4692,
      "train_tokens_per_second": 333.162
    },
    {
      "epoch": 0.17011494252873563,
      "grad_norm": 7.026374816894531,
      "learning_rate": 1.5137844611528823e-06,
      "loss": 0.49688419699668884,
      "num_input_tokens_seen": 1414716,
      "step": 1850,
      "train_runtime": 4246.3862,
      "train_tokens_per_second": 333.158
    },
    {
      "epoch": 0.17020689655172414,
      "grad_norm": 6.23845911026001,
      "learning_rate": 1.5037593984962406e-06,
      "loss": 0.5111281871795654,
      "num_input_tokens_seen": 1415380,
      "step": 1851,
      "train_runtime": 4248.2976,
      "train_tokens_per_second": 333.164
    },
    {
      "epoch": 0.17029885057471264,
      "grad_norm": 6.714358329772949,
      "learning_rate": 1.4937343358395993e-06,
      "loss": 0.43978726863861084,
      "num_input_tokens_seen": 1416104,
      "step": 1852,
      "train_runtime": 4250.2254,
      "train_tokens_per_second": 333.183
    },
    {
      "epoch": 0.17039080459770115,
      "grad_norm": 5.875720500946045,
      "learning_rate": 1.4837092731829576e-06,
      "loss": 0.39047926664352417,
      "num_input_tokens_seen": 1416680,
      "step": 1853,
      "train_runtime": 4252.1336,
      "train_tokens_per_second": 333.169
    },
    {
      "epoch": 0.17048275862068965,
      "grad_norm": 6.088798999786377,
      "learning_rate": 1.4736842105263159e-06,
      "loss": 0.4593173861503601,
      "num_input_tokens_seen": 1417944,
      "step": 1854,
      "train_runtime": 4254.1453,
      "train_tokens_per_second": 333.309
    },
    {
      "epoch": 0.17057471264367816,
      "grad_norm": 6.431163311004639,
      "learning_rate": 1.4636591478696743e-06,
      "loss": 0.4786079227924347,
      "num_input_tokens_seen": 1419072,
      "step": 1855,
      "train_runtime": 4256.1238,
      "train_tokens_per_second": 333.419
    },
    {
      "epoch": 0.17066666666666666,
      "grad_norm": 6.588095664978027,
      "learning_rate": 1.4536340852130326e-06,
      "loss": 0.5038584470748901,
      "num_input_tokens_seen": 1419672,
      "step": 1856,
      "train_runtime": 4258.0316,
      "train_tokens_per_second": 333.41
    },
    {
      "epoch": 0.17075862068965517,
      "grad_norm": 6.476990222930908,
      "learning_rate": 1.4436090225563911e-06,
      "loss": 0.3907082676887512,
      "num_input_tokens_seen": 1420224,
      "step": 1857,
      "train_runtime": 4259.9294,
      "train_tokens_per_second": 333.391
    },
    {
      "epoch": 0.17085057471264367,
      "grad_norm": 6.375645637512207,
      "learning_rate": 1.4335839598997494e-06,
      "loss": 0.4861810505390167,
      "num_input_tokens_seen": 1421028,
      "step": 1858,
      "train_runtime": 4261.8695,
      "train_tokens_per_second": 333.428
    },
    {
      "epoch": 0.17094252873563218,
      "grad_norm": 7.222268104553223,
      "learning_rate": 1.4235588972431077e-06,
      "loss": 0.44973140954971313,
      "num_input_tokens_seen": 1421696,
      "step": 1859,
      "train_runtime": 4263.7807,
      "train_tokens_per_second": 333.436
    },
    {
      "epoch": 0.17103448275862068,
      "grad_norm": 5.898995399475098,
      "learning_rate": 1.4135338345864664e-06,
      "loss": 0.3670354187488556,
      "num_input_tokens_seen": 1422308,
      "step": 1860,
      "train_runtime": 4265.6906,
      "train_tokens_per_second": 333.43
    },
    {
      "epoch": 0.1711264367816092,
      "grad_norm": 6.76981258392334,
      "learning_rate": 1.4035087719298246e-06,
      "loss": 0.46958744525909424,
      "num_input_tokens_seen": 1423004,
      "step": 1861,
      "train_runtime": 4273.9514,
      "train_tokens_per_second": 332.948
    },
    {
      "epoch": 0.1712183908045977,
      "grad_norm": 6.871397495269775,
      "learning_rate": 1.3934837092731831e-06,
      "loss": 0.4450916647911072,
      "num_input_tokens_seen": 1423568,
      "step": 1862,
      "train_runtime": 4275.8633,
      "train_tokens_per_second": 332.931
    },
    {
      "epoch": 0.1713103448275862,
      "grad_norm": 7.244660377502441,
      "learning_rate": 1.3834586466165414e-06,
      "loss": 0.526885449886322,
      "num_input_tokens_seen": 1424176,
      "step": 1863,
      "train_runtime": 4277.7682,
      "train_tokens_per_second": 332.925
    },
    {
      "epoch": 0.1714022988505747,
      "grad_norm": 5.575239658355713,
      "learning_rate": 1.3734335839598997e-06,
      "loss": 0.4366419017314911,
      "num_input_tokens_seen": 1425632,
      "step": 1864,
      "train_runtime": 4279.7742,
      "train_tokens_per_second": 333.109
    },
    {
      "epoch": 0.1714942528735632,
      "grad_norm": 7.087784290313721,
      "learning_rate": 1.3634085213032584e-06,
      "loss": 0.6109887361526489,
      "num_input_tokens_seen": 1426364,
      "step": 1865,
      "train_runtime": 4281.6924,
      "train_tokens_per_second": 333.131
    },
    {
      "epoch": 0.1715862068965517,
      "grad_norm": 5.8095622062683105,
      "learning_rate": 1.3533834586466167e-06,
      "loss": 0.410519003868103,
      "num_input_tokens_seen": 1427044,
      "step": 1866,
      "train_runtime": 4283.6001,
      "train_tokens_per_second": 333.141
    },
    {
      "epoch": 0.17167816091954022,
      "grad_norm": 6.8906049728393555,
      "learning_rate": 1.3433583959899751e-06,
      "loss": 0.4923664331436157,
      "num_input_tokens_seen": 1427816,
      "step": 1867,
      "train_runtime": 4285.5524,
      "train_tokens_per_second": 333.17
    },
    {
      "epoch": 0.17177011494252872,
      "grad_norm": 6.780386924743652,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 0.4736308455467224,
      "num_input_tokens_seen": 1428596,
      "step": 1868,
      "train_runtime": 4287.4774,
      "train_tokens_per_second": 333.202
    },
    {
      "epoch": 0.17186206896551723,
      "grad_norm": 7.154784679412842,
      "learning_rate": 1.3233082706766917e-06,
      "loss": 0.4392606317996979,
      "num_input_tokens_seen": 1429348,
      "step": 1869,
      "train_runtime": 4289.4581,
      "train_tokens_per_second": 333.223
    },
    {
      "epoch": 0.17195402298850573,
      "grad_norm": 8.420307159423828,
      "learning_rate": 1.3132832080200502e-06,
      "loss": 0.6212730407714844,
      "num_input_tokens_seen": 1430352,
      "step": 1870,
      "train_runtime": 4291.463,
      "train_tokens_per_second": 333.302
    },
    {
      "epoch": 0.17204597701149427,
      "grad_norm": 6.159582614898682,
      "learning_rate": 1.3032581453634085e-06,
      "loss": 0.42425036430358887,
      "num_input_tokens_seen": 1431124,
      "step": 1871,
      "train_runtime": 4293.4104,
      "train_tokens_per_second": 333.33
    },
    {
      "epoch": 0.17213793103448277,
      "grad_norm": 6.059912204742432,
      "learning_rate": 1.2932330827067672e-06,
      "loss": 0.4253976345062256,
      "num_input_tokens_seen": 1431800,
      "step": 1872,
      "train_runtime": 4295.3295,
      "train_tokens_per_second": 333.339
    },
    {
      "epoch": 0.17222988505747128,
      "grad_norm": 6.3949151039123535,
      "learning_rate": 1.2832080200501254e-06,
      "loss": 0.4122861623764038,
      "num_input_tokens_seen": 1432364,
      "step": 1873,
      "train_runtime": 4297.2456,
      "train_tokens_per_second": 333.321
    },
    {
      "epoch": 0.17232183908045978,
      "grad_norm": 6.341155529022217,
      "learning_rate": 1.2731829573934837e-06,
      "loss": 0.5156214237213135,
      "num_input_tokens_seen": 1433056,
      "step": 1874,
      "train_runtime": 4299.1578,
      "train_tokens_per_second": 333.334
    },
    {
      "epoch": 0.1724137931034483,
      "grad_norm": 6.808344841003418,
      "learning_rate": 1.2631578947368422e-06,
      "loss": 0.42363232374191284,
      "num_input_tokens_seen": 1433628,
      "step": 1875,
      "train_runtime": 4301.1184,
      "train_tokens_per_second": 333.315
    },
    {
      "epoch": 0.1725057471264368,
      "grad_norm": 6.694194316864014,
      "learning_rate": 1.2531328320802005e-06,
      "loss": 0.5186920166015625,
      "num_input_tokens_seen": 1434696,
      "step": 1876,
      "train_runtime": 4303.1378,
      "train_tokens_per_second": 333.407
    },
    {
      "epoch": 0.1725977011494253,
      "grad_norm": 7.251523971557617,
      "learning_rate": 1.243107769423559e-06,
      "loss": 0.44533511996269226,
      "num_input_tokens_seen": 1435452,
      "step": 1877,
      "train_runtime": 4305.0656,
      "train_tokens_per_second": 333.433
    },
    {
      "epoch": 0.1726896551724138,
      "grad_norm": 7.031600475311279,
      "learning_rate": 1.2330827067669174e-06,
      "loss": 0.4380522072315216,
      "num_input_tokens_seen": 1436260,
      "step": 1878,
      "train_runtime": 4307.008,
      "train_tokens_per_second": 333.47
    },
    {
      "epoch": 0.1727816091954023,
      "grad_norm": 6.173227310180664,
      "learning_rate": 1.2230576441102757e-06,
      "loss": 0.44876721501350403,
      "num_input_tokens_seen": 1437456,
      "step": 1879,
      "train_runtime": 4309.0007,
      "train_tokens_per_second": 333.594
    },
    {
      "epoch": 0.1728735632183908,
      "grad_norm": 6.372288703918457,
      "learning_rate": 1.2130325814536342e-06,
      "loss": 0.40218284726142883,
      "num_input_tokens_seen": 1438212,
      "step": 1880,
      "train_runtime": 4310.9523,
      "train_tokens_per_second": 333.618
    },
    {
      "epoch": 0.17296551724137932,
      "grad_norm": 6.622946739196777,
      "learning_rate": 1.2030075187969925e-06,
      "loss": 0.5180130004882812,
      "num_input_tokens_seen": 1438864,
      "step": 1881,
      "train_runtime": 4312.8761,
      "train_tokens_per_second": 333.621
    },
    {
      "epoch": 0.17305747126436782,
      "grad_norm": 5.832489967346191,
      "learning_rate": 1.192982456140351e-06,
      "loss": 0.40099936723709106,
      "num_input_tokens_seen": 1439488,
      "step": 1882,
      "train_runtime": 4314.7929,
      "train_tokens_per_second": 333.617
    },
    {
      "epoch": 0.17314942528735633,
      "grad_norm": 6.789658546447754,
      "learning_rate": 1.1829573934837095e-06,
      "loss": 0.4777843952178955,
      "num_input_tokens_seen": 1440096,
      "step": 1883,
      "train_runtime": 4316.7567,
      "train_tokens_per_second": 333.606
    },
    {
      "epoch": 0.17324137931034483,
      "grad_norm": 7.081912040710449,
      "learning_rate": 1.1729323308270677e-06,
      "loss": 0.47297799587249756,
      "num_input_tokens_seen": 1440636,
      "step": 1884,
      "train_runtime": 4318.6824,
      "train_tokens_per_second": 333.582
    },
    {
      "epoch": 0.17333333333333334,
      "grad_norm": 6.993490219116211,
      "learning_rate": 1.1629072681704262e-06,
      "loss": 0.45466509461402893,
      "num_input_tokens_seen": 1441284,
      "step": 1885,
      "train_runtime": 4320.637,
      "train_tokens_per_second": 333.581
    },
    {
      "epoch": 0.17342528735632184,
      "grad_norm": 6.9370036125183105,
      "learning_rate": 1.1528822055137845e-06,
      "loss": 0.4803755283355713,
      "num_input_tokens_seen": 1442008,
      "step": 1886,
      "train_runtime": 4322.5544,
      "train_tokens_per_second": 333.601
    },
    {
      "epoch": 0.17351724137931035,
      "grad_norm": 6.025440216064453,
      "learning_rate": 1.142857142857143e-06,
      "loss": 0.4667506217956543,
      "num_input_tokens_seen": 1443288,
      "step": 1887,
      "train_runtime": 4324.5947,
      "train_tokens_per_second": 333.739
    },
    {
      "epoch": 0.17360919540229885,
      "grad_norm": 6.843484878540039,
      "learning_rate": 1.1328320802005013e-06,
      "loss": 0.5556914806365967,
      "num_input_tokens_seen": 1443960,
      "step": 1888,
      "train_runtime": 4326.6468,
      "train_tokens_per_second": 333.737
    },
    {
      "epoch": 0.17370114942528736,
      "grad_norm": 6.687643527984619,
      "learning_rate": 1.1228070175438598e-06,
      "loss": 0.5455694198608398,
      "num_input_tokens_seen": 1444624,
      "step": 1889,
      "train_runtime": 4328.6036,
      "train_tokens_per_second": 333.739
    },
    {
      "epoch": 0.17379310344827587,
      "grad_norm": 7.248600006103516,
      "learning_rate": 1.1127819548872182e-06,
      "loss": 0.5499457716941833,
      "num_input_tokens_seen": 1445324,
      "step": 1890,
      "train_runtime": 4330.5272,
      "train_tokens_per_second": 333.752
    },
    {
      "epoch": 0.17388505747126437,
      "grad_norm": 6.184894561767578,
      "learning_rate": 1.1027568922305765e-06,
      "loss": 0.4898326098918915,
      "num_input_tokens_seen": 1446344,
      "step": 1891,
      "train_runtime": 4338.6872,
      "train_tokens_per_second": 333.36
    },
    {
      "epoch": 0.17397701149425288,
      "grad_norm": 6.2451324462890625,
      "learning_rate": 1.0927318295739348e-06,
      "loss": 0.45967888832092285,
      "num_input_tokens_seen": 1447104,
      "step": 1892,
      "train_runtime": 4340.6204,
      "train_tokens_per_second": 333.386
    },
    {
      "epoch": 0.17406896551724138,
      "grad_norm": 7.12343692779541,
      "learning_rate": 1.0827067669172933e-06,
      "loss": 0.515326976776123,
      "num_input_tokens_seen": 1447860,
      "step": 1893,
      "train_runtime": 4342.5987,
      "train_tokens_per_second": 333.409
    },
    {
      "epoch": 0.17416091954022989,
      "grad_norm": 7.140078544616699,
      "learning_rate": 1.0726817042606518e-06,
      "loss": 0.5173025727272034,
      "num_input_tokens_seen": 1448644,
      "step": 1894,
      "train_runtime": 4344.5177,
      "train_tokens_per_second": 333.442
    },
    {
      "epoch": 0.1742528735632184,
      "grad_norm": 4.653764724731445,
      "learning_rate": 1.0626566416040103e-06,
      "loss": 0.4386482238769531,
      "num_input_tokens_seen": 1450168,
      "step": 1895,
      "train_runtime": 4346.5598,
      "train_tokens_per_second": 333.636
    },
    {
      "epoch": 0.1743448275862069,
      "grad_norm": 7.022873401641846,
      "learning_rate": 1.0526315789473685e-06,
      "loss": 0.5092371702194214,
      "num_input_tokens_seen": 1450812,
      "step": 1896,
      "train_runtime": 4348.5637,
      "train_tokens_per_second": 333.63
    },
    {
      "epoch": 0.1744367816091954,
      "grad_norm": 6.802220344543457,
      "learning_rate": 1.0426065162907268e-06,
      "loss": 0.4958755075931549,
      "num_input_tokens_seen": 1451784,
      "step": 1897,
      "train_runtime": 4350.5302,
      "train_tokens_per_second": 333.703
    },
    {
      "epoch": 0.1745287356321839,
      "grad_norm": 7.149991035461426,
      "learning_rate": 1.0325814536340853e-06,
      "loss": 0.5435976386070251,
      "num_input_tokens_seen": 1452388,
      "step": 1898,
      "train_runtime": 4352.4419,
      "train_tokens_per_second": 333.695
    },
    {
      "epoch": 0.1746206896551724,
      "grad_norm": 6.909001350402832,
      "learning_rate": 1.0225563909774438e-06,
      "loss": 0.5135350227355957,
      "num_input_tokens_seen": 1453408,
      "step": 1899,
      "train_runtime": 4354.4128,
      "train_tokens_per_second": 333.778
    },
    {
      "epoch": 0.17471264367816092,
      "grad_norm": 6.813198089599609,
      "learning_rate": 1.012531328320802e-06,
      "loss": 0.6000266075134277,
      "num_input_tokens_seen": 1454436,
      "step": 1900,
      "train_runtime": 4356.359,
      "train_tokens_per_second": 333.865
    },
    {
      "epoch": 0.17480459770114942,
      "grad_norm": 6.590569019317627,
      "learning_rate": 1.0025062656641603e-06,
      "loss": 0.4507034718990326,
      "num_input_tokens_seen": 1455120,
      "step": 1901,
      "train_runtime": 4358.2723,
      "train_tokens_per_second": 333.875
    },
    {
      "epoch": 0.17489655172413793,
      "grad_norm": 6.074158191680908,
      "learning_rate": 9.924812030075188e-07,
      "loss": 0.447780042886734,
      "num_input_tokens_seen": 1456432,
      "step": 1902,
      "train_runtime": 4360.2807,
      "train_tokens_per_second": 334.023
    },
    {
      "epoch": 0.17498850574712643,
      "grad_norm": 6.893975257873535,
      "learning_rate": 9.824561403508773e-07,
      "loss": 0.5302909016609192,
      "num_input_tokens_seen": 1457548,
      "step": 1903,
      "train_runtime": 4362.2594,
      "train_tokens_per_second": 334.127
    },
    {
      "epoch": 0.17508045977011494,
      "grad_norm": 6.07625150680542,
      "learning_rate": 9.724310776942358e-07,
      "loss": 0.48071980476379395,
      "num_input_tokens_seen": 1458224,
      "step": 1904,
      "train_runtime": 4364.1914,
      "train_tokens_per_second": 334.134
    },
    {
      "epoch": 0.17517241379310344,
      "grad_norm": 6.833479881286621,
      "learning_rate": 9.62406015037594e-07,
      "loss": 0.5385992527008057,
      "num_input_tokens_seen": 1458908,
      "step": 1905,
      "train_runtime": 4366.1587,
      "train_tokens_per_second": 334.14
    },
    {
      "epoch": 0.17526436781609195,
      "grad_norm": 6.9142746925354,
      "learning_rate": 9.523809523809525e-07,
      "loss": 0.45336848497390747,
      "num_input_tokens_seen": 1459576,
      "step": 1906,
      "train_runtime": 4368.0715,
      "train_tokens_per_second": 334.147
    },
    {
      "epoch": 0.17535632183908045,
      "grad_norm": 6.9626078605651855,
      "learning_rate": 9.423558897243108e-07,
      "loss": 0.5101382732391357,
      "num_input_tokens_seen": 1460532,
      "step": 1907,
      "train_runtime": 4370.0204,
      "train_tokens_per_second": 334.216
    },
    {
      "epoch": 0.17544827586206896,
      "grad_norm": 8.040032386779785,
      "learning_rate": 9.323308270676692e-07,
      "loss": 0.475430428981781,
      "num_input_tokens_seen": 1461216,
      "step": 1908,
      "train_runtime": 4372.0137,
      "train_tokens_per_second": 334.22
    },
    {
      "epoch": 0.17554022988505746,
      "grad_norm": 8.253643035888672,
      "learning_rate": 9.223057644110277e-07,
      "loss": 0.5012615919113159,
      "num_input_tokens_seen": 1461920,
      "step": 1909,
      "train_runtime": 4373.9777,
      "train_tokens_per_second": 334.231
    },
    {
      "epoch": 0.17563218390804597,
      "grad_norm": 8.394492149353027,
      "learning_rate": 9.122807017543861e-07,
      "loss": 0.5508334636688232,
      "num_input_tokens_seen": 1462516,
      "step": 1910,
      "train_runtime": 4375.9422,
      "train_tokens_per_second": 334.217
    },
    {
      "epoch": 0.17572413793103447,
      "grad_norm": 6.01444149017334,
      "learning_rate": 9.022556390977444e-07,
      "loss": 0.39604777097702026,
      "num_input_tokens_seen": 1463240,
      "step": 1911,
      "train_runtime": 4377.8615,
      "train_tokens_per_second": 334.236
    },
    {
      "epoch": 0.17581609195402298,
      "grad_norm": 7.359145164489746,
      "learning_rate": 8.922305764411029e-07,
      "loss": 0.4897788166999817,
      "num_input_tokens_seen": 1463912,
      "step": 1912,
      "train_runtime": 4379.7753,
      "train_tokens_per_second": 334.244
    },
    {
      "epoch": 0.17590804597701148,
      "grad_norm": 7.1891069412231445,
      "learning_rate": 8.822055137844612e-07,
      "loss": 0.501335620880127,
      "num_input_tokens_seen": 1464624,
      "step": 1913,
      "train_runtime": 4381.7365,
      "train_tokens_per_second": 334.257
    },
    {
      "epoch": 0.176,
      "grad_norm": 6.721168041229248,
      "learning_rate": 8.721804511278196e-07,
      "loss": 0.4819718599319458,
      "num_input_tokens_seen": 1465268,
      "step": 1914,
      "train_runtime": 4383.6658,
      "train_tokens_per_second": 334.256
    },
    {
      "epoch": 0.1760919540229885,
      "grad_norm": 7.009974956512451,
      "learning_rate": 8.621553884711781e-07,
      "loss": 0.48692968487739563,
      "num_input_tokens_seen": 1465868,
      "step": 1915,
      "train_runtime": 4385.5697,
      "train_tokens_per_second": 334.248
    },
    {
      "epoch": 0.176183908045977,
      "grad_norm": 6.817626953125,
      "learning_rate": 8.521303258145364e-07,
      "loss": 0.4318804144859314,
      "num_input_tokens_seen": 1466640,
      "step": 1916,
      "train_runtime": 4387.4816,
      "train_tokens_per_second": 334.278
    },
    {
      "epoch": 0.1762758620689655,
      "grad_norm": 6.447076320648193,
      "learning_rate": 8.421052631578948e-07,
      "loss": 0.5303728580474854,
      "num_input_tokens_seen": 1467296,
      "step": 1917,
      "train_runtime": 4389.4097,
      "train_tokens_per_second": 334.281
    },
    {
      "epoch": 0.176367816091954,
      "grad_norm": 6.219639778137207,
      "learning_rate": 8.320802005012532e-07,
      "loss": 0.48713910579681396,
      "num_input_tokens_seen": 1468240,
      "step": 1918,
      "train_runtime": 4391.3562,
      "train_tokens_per_second": 334.348
    },
    {
      "epoch": 0.17645977011494252,
      "grad_norm": 6.238115310668945,
      "learning_rate": 8.220551378446116e-07,
      "loss": 0.46840783953666687,
      "num_input_tokens_seen": 1468892,
      "step": 1919,
      "train_runtime": 4393.2898,
      "train_tokens_per_second": 334.349
    },
    {
      "epoch": 0.17655172413793102,
      "grad_norm": 6.222463130950928,
      "learning_rate": 8.1203007518797e-07,
      "loss": 0.44472193717956543,
      "num_input_tokens_seen": 1469560,
      "step": 1920,
      "train_runtime": 4395.194,
      "train_tokens_per_second": 334.356
    },
    {
      "epoch": 0.17664367816091955,
      "grad_norm": 6.146585464477539,
      "learning_rate": 8.020050125313284e-07,
      "loss": 0.4363603889942169,
      "num_input_tokens_seen": 1470256,
      "step": 1921,
      "train_runtime": 4403.8036,
      "train_tokens_per_second": 333.86
    },
    {
      "epoch": 0.17673563218390806,
      "grad_norm": 7.714982986450195,
      "learning_rate": 7.919799498746868e-07,
      "loss": 0.4598209261894226,
      "num_input_tokens_seen": 1470884,
      "step": 1922,
      "train_runtime": 4405.7283,
      "train_tokens_per_second": 333.857
    },
    {
      "epoch": 0.17682758620689656,
      "grad_norm": 6.125604152679443,
      "learning_rate": 7.819548872180452e-07,
      "loss": 0.3589984178543091,
      "num_input_tokens_seen": 1471700,
      "step": 1923,
      "train_runtime": 4407.9266,
      "train_tokens_per_second": 333.876
    },
    {
      "epoch": 0.17691954022988507,
      "grad_norm": 6.287062168121338,
      "learning_rate": 7.719298245614036e-07,
      "loss": 0.4955197274684906,
      "num_input_tokens_seen": 1472448,
      "step": 1924,
      "train_runtime": 4409.8464,
      "train_tokens_per_second": 333.9
    },
    {
      "epoch": 0.17701149425287357,
      "grad_norm": 6.434872150421143,
      "learning_rate": 7.61904761904762e-07,
      "loss": 0.4396565556526184,
      "num_input_tokens_seen": 1473268,
      "step": 1925,
      "train_runtime": 4411.8362,
      "train_tokens_per_second": 333.935
    },
    {
      "epoch": 0.17710344827586208,
      "grad_norm": 7.826362609863281,
      "learning_rate": 7.518796992481203e-07,
      "loss": 0.4749208092689514,
      "num_input_tokens_seen": 1473976,
      "step": 1926,
      "train_runtime": 4413.7727,
      "train_tokens_per_second": 333.949
    },
    {
      "epoch": 0.17719540229885059,
      "grad_norm": 7.021556854248047,
      "learning_rate": 7.418546365914788e-07,
      "loss": 0.43029263615608215,
      "num_input_tokens_seen": 1474644,
      "step": 1927,
      "train_runtime": 4415.6815,
      "train_tokens_per_second": 333.956
    },
    {
      "epoch": 0.1772873563218391,
      "grad_norm": 6.2591047286987305,
      "learning_rate": 7.318295739348372e-07,
      "loss": 0.4811077117919922,
      "num_input_tokens_seen": 1475748,
      "step": 1928,
      "train_runtime": 4417.6511,
      "train_tokens_per_second": 334.057
    },
    {
      "epoch": 0.1773793103448276,
      "grad_norm": 6.473521709442139,
      "learning_rate": 7.218045112781956e-07,
      "loss": 0.4754500091075897,
      "num_input_tokens_seen": 1476600,
      "step": 1929,
      "train_runtime": 4419.6129,
      "train_tokens_per_second": 334.102
    },
    {
      "epoch": 0.1774712643678161,
      "grad_norm": 6.3368611335754395,
      "learning_rate": 7.117794486215538e-07,
      "loss": 0.4027555584907532,
      "num_input_tokens_seen": 1477272,
      "step": 1930,
      "train_runtime": 4421.5345,
      "train_tokens_per_second": 334.108
    },
    {
      "epoch": 0.1775632183908046,
      "grad_norm": 6.811783313751221,
      "learning_rate": 7.017543859649123e-07,
      "loss": 0.48822206258773804,
      "num_input_tokens_seen": 1478012,
      "step": 1931,
      "train_runtime": 4423.4457,
      "train_tokens_per_second": 334.131
    },
    {
      "epoch": 0.1776551724137931,
      "grad_norm": 7.439375400543213,
      "learning_rate": 6.917293233082707e-07,
      "loss": 0.3694459795951843,
      "num_input_tokens_seen": 1478664,
      "step": 1932,
      "train_runtime": 4425.3499,
      "train_tokens_per_second": 334.135
    },
    {
      "epoch": 0.17774712643678162,
      "grad_norm": 6.692470073699951,
      "learning_rate": 6.817042606516292e-07,
      "loss": 0.5403836369514465,
      "num_input_tokens_seen": 1479676,
      "step": 1933,
      "train_runtime": 4427.3233,
      "train_tokens_per_second": 334.215
    },
    {
      "epoch": 0.17783908045977012,
      "grad_norm": 7.127777576446533,
      "learning_rate": 6.716791979949876e-07,
      "loss": 0.48094385862350464,
      "num_input_tokens_seen": 1480384,
      "step": 1934,
      "train_runtime": 4429.25,
      "train_tokens_per_second": 334.229
    },
    {
      "epoch": 0.17793103448275863,
      "grad_norm": 6.633772373199463,
      "learning_rate": 6.616541353383458e-07,
      "loss": 0.4813235104084015,
      "num_input_tokens_seen": 1481416,
      "step": 1935,
      "train_runtime": 4431.2194,
      "train_tokens_per_second": 334.313
    },
    {
      "epoch": 0.17802298850574713,
      "grad_norm": 7.7599406242370605,
      "learning_rate": 6.516290726817042e-07,
      "loss": 0.5024576783180237,
      "num_input_tokens_seen": 1482040,
      "step": 1936,
      "train_runtime": 4433.1256,
      "train_tokens_per_second": 334.31
    },
    {
      "epoch": 0.17811494252873564,
      "grad_norm": 7.687955856323242,
      "learning_rate": 6.416040100250627e-07,
      "loss": 0.49235543608665466,
      "num_input_tokens_seen": 1482880,
      "step": 1937,
      "train_runtime": 4435.0401,
      "train_tokens_per_second": 334.355
    },
    {
      "epoch": 0.17820689655172414,
      "grad_norm": 6.23078727722168,
      "learning_rate": 6.315789473684211e-07,
      "loss": 0.40074771642684937,
      "num_input_tokens_seen": 1483476,
      "step": 1938,
      "train_runtime": 4436.9421,
      "train_tokens_per_second": 334.346
    },
    {
      "epoch": 0.17829885057471265,
      "grad_norm": 6.496601104736328,
      "learning_rate": 6.215538847117795e-07,
      "loss": 0.45595455169677734,
      "num_input_tokens_seen": 1484148,
      "step": 1939,
      "train_runtime": 4438.8462,
      "train_tokens_per_second": 334.354
    },
    {
      "epoch": 0.17839080459770115,
      "grad_norm": 7.491678237915039,
      "learning_rate": 6.115288220551379e-07,
      "loss": 0.6587531566619873,
      "num_input_tokens_seen": 1484968,
      "step": 1940,
      "train_runtime": 4440.8207,
      "train_tokens_per_second": 334.39
    },
    {
      "epoch": 0.17848275862068966,
      "grad_norm": 6.942194938659668,
      "learning_rate": 6.015037593984962e-07,
      "loss": 0.4629093408584595,
      "num_input_tokens_seen": 1485736,
      "step": 1941,
      "train_runtime": 4442.7364,
      "train_tokens_per_second": 334.419
    },
    {
      "epoch": 0.17857471264367816,
      "grad_norm": 7.357885837554932,
      "learning_rate": 5.914786967418547e-07,
      "loss": 0.4690327048301697,
      "num_input_tokens_seen": 1486344,
      "step": 1942,
      "train_runtime": 4444.6525,
      "train_tokens_per_second": 334.412
    },
    {
      "epoch": 0.17866666666666667,
      "grad_norm": 7.242325305938721,
      "learning_rate": 5.814536340852131e-07,
      "loss": 0.5480672121047974,
      "num_input_tokens_seen": 1487076,
      "step": 1943,
      "train_runtime": 4446.6194,
      "train_tokens_per_second": 334.428
    },
    {
      "epoch": 0.17875862068965517,
      "grad_norm": 6.363503932952881,
      "learning_rate": 5.714285714285715e-07,
      "loss": 0.4140477776527405,
      "num_input_tokens_seen": 1487680,
      "step": 1944,
      "train_runtime": 4448.524,
      "train_tokens_per_second": 334.421
    },
    {
      "epoch": 0.17885057471264368,
      "grad_norm": 6.547098636627197,
      "learning_rate": 5.614035087719299e-07,
      "loss": 0.5084044933319092,
      "num_input_tokens_seen": 1488312,
      "step": 1945,
      "train_runtime": 4450.4288,
      "train_tokens_per_second": 334.42
    },
    {
      "epoch": 0.17894252873563218,
      "grad_norm": 6.4081196784973145,
      "learning_rate": 5.513784461152883e-07,
      "loss": 0.517520546913147,
      "num_input_tokens_seen": 1489008,
      "step": 1946,
      "train_runtime": 4452.4153,
      "train_tokens_per_second": 334.427
    },
    {
      "epoch": 0.1790344827586207,
      "grad_norm": 6.37542724609375,
      "learning_rate": 5.413533834586466e-07,
      "loss": 0.5319074988365173,
      "num_input_tokens_seen": 1490272,
      "step": 1947,
      "train_runtime": 4454.3832,
      "train_tokens_per_second": 334.563
    },
    {
      "epoch": 0.1791264367816092,
      "grad_norm": 6.55473518371582,
      "learning_rate": 5.313283208020051e-07,
      "loss": 0.43775203824043274,
      "num_input_tokens_seen": 1490920,
      "step": 1948,
      "train_runtime": 4456.3222,
      "train_tokens_per_second": 334.563
    },
    {
      "epoch": 0.1792183908045977,
      "grad_norm": 6.583804607391357,
      "learning_rate": 5.213032581453634e-07,
      "loss": 0.4879281222820282,
      "num_input_tokens_seen": 1491816,
      "step": 1949,
      "train_runtime": 4458.2926,
      "train_tokens_per_second": 334.616
    },
    {
      "epoch": 0.1793103448275862,
      "grad_norm": 6.798384666442871,
      "learning_rate": 5.112781954887219e-07,
      "loss": 0.43789058923721313,
      "num_input_tokens_seen": 1492492,
      "step": 1950,
      "train_runtime": 4460.2035,
      "train_tokens_per_second": 334.624
    },
    {
      "epoch": 0.1794022988505747,
      "grad_norm": 6.2381815910339355,
      "learning_rate": 5.012531328320802e-07,
      "loss": 0.4565480351448059,
      "num_input_tokens_seen": 1493144,
      "step": 1951,
      "train_runtime": 4468.7669,
      "train_tokens_per_second": 334.129
    },
    {
      "epoch": 0.17949425287356321,
      "grad_norm": 6.779036045074463,
      "learning_rate": 4.912280701754387e-07,
      "loss": 0.49072062969207764,
      "num_input_tokens_seen": 1493928,
      "step": 1952,
      "train_runtime": 4470.6831,
      "train_tokens_per_second": 334.161
    },
    {
      "epoch": 0.17958620689655172,
      "grad_norm": 7.393740653991699,
      "learning_rate": 4.81203007518797e-07,
      "loss": 0.4488428831100464,
      "num_input_tokens_seen": 1494560,
      "step": 1953,
      "train_runtime": 4472.7891,
      "train_tokens_per_second": 334.145
    },
    {
      "epoch": 0.17967816091954023,
      "grad_norm": 5.62514591217041,
      "learning_rate": 4.711779448621554e-07,
      "loss": 0.4903565049171448,
      "num_input_tokens_seen": 1495944,
      "step": 1954,
      "train_runtime": 4474.8408,
      "train_tokens_per_second": 334.301
    },
    {
      "epoch": 0.17977011494252873,
      "grad_norm": 6.749486923217773,
      "learning_rate": 4.6115288220551385e-07,
      "loss": 0.40888097882270813,
      "num_input_tokens_seen": 1496692,
      "step": 1955,
      "train_runtime": 4476.8014,
      "train_tokens_per_second": 334.322
    },
    {
      "epoch": 0.17986206896551724,
      "grad_norm": 6.747584342956543,
      "learning_rate": 4.511278195488722e-07,
      "loss": 0.45244961977005005,
      "num_input_tokens_seen": 1497372,
      "step": 1956,
      "train_runtime": 4478.7444,
      "train_tokens_per_second": 334.329
    },
    {
      "epoch": 0.17995402298850574,
      "grad_norm": 7.246510982513428,
      "learning_rate": 4.411027568922306e-07,
      "loss": 0.48466747999191284,
      "num_input_tokens_seen": 1498084,
      "step": 1957,
      "train_runtime": 4480.6731,
      "train_tokens_per_second": 334.344
    },
    {
      "epoch": 0.18004597701149425,
      "grad_norm": 7.051685333251953,
      "learning_rate": 4.3107769423558905e-07,
      "loss": 0.5010166764259338,
      "num_input_tokens_seen": 1498816,
      "step": 1958,
      "train_runtime": 4482.5995,
      "train_tokens_per_second": 334.363
    },
    {
      "epoch": 0.18013793103448275,
      "grad_norm": 6.835502624511719,
      "learning_rate": 4.210526315789474e-07,
      "loss": 0.47410574555397034,
      "num_input_tokens_seen": 1499420,
      "step": 1959,
      "train_runtime": 4484.5147,
      "train_tokens_per_second": 334.355
    },
    {
      "epoch": 0.18022988505747126,
      "grad_norm": 6.855109691619873,
      "learning_rate": 4.110275689223058e-07,
      "loss": 0.44139620661735535,
      "num_input_tokens_seen": 1500072,
      "step": 1960,
      "train_runtime": 4486.4225,
      "train_tokens_per_second": 334.358
    },
    {
      "epoch": 0.18032183908045976,
      "grad_norm": 7.25023078918457,
      "learning_rate": 4.010025062656642e-07,
      "loss": 0.38036903738975525,
      "num_input_tokens_seen": 1500600,
      "step": 1961,
      "train_runtime": 4488.3333,
      "train_tokens_per_second": 334.333
    },
    {
      "epoch": 0.18041379310344827,
      "grad_norm": 6.056260585784912,
      "learning_rate": 3.909774436090226e-07,
      "loss": 0.39658379554748535,
      "num_input_tokens_seen": 1501240,
      "step": 1962,
      "train_runtime": 4490.2453,
      "train_tokens_per_second": 334.334
    },
    {
      "epoch": 0.18050574712643677,
      "grad_norm": 6.8434247970581055,
      "learning_rate": 3.80952380952381e-07,
      "loss": 0.6031206250190735,
      "num_input_tokens_seen": 1502648,
      "step": 1963,
      "train_runtime": 4492.2897,
      "train_tokens_per_second": 334.495
    },
    {
      "epoch": 0.18059770114942528,
      "grad_norm": 7.027190685272217,
      "learning_rate": 3.709273182957394e-07,
      "loss": 0.37974339723587036,
      "num_input_tokens_seen": 1503312,
      "step": 1964,
      "train_runtime": 4494.2092,
      "train_tokens_per_second": 334.5
    },
    {
      "epoch": 0.18068965517241378,
      "grad_norm": 6.077744483947754,
      "learning_rate": 3.609022556390978e-07,
      "loss": 0.44487810134887695,
      "num_input_tokens_seen": 1503856,
      "step": 1965,
      "train_runtime": 4496.1099,
      "train_tokens_per_second": 334.479
    },
    {
      "epoch": 0.1807816091954023,
      "grad_norm": 6.1048407554626465,
      "learning_rate": 3.5087719298245616e-07,
      "loss": 0.4310495853424072,
      "num_input_tokens_seen": 1504752,
      "step": 1966,
      "train_runtime": 4498.0547,
      "train_tokens_per_second": 334.534
    },
    {
      "epoch": 0.1808735632183908,
      "grad_norm": 6.022492408752441,
      "learning_rate": 3.408521303258146e-07,
      "loss": 0.5145743489265442,
      "num_input_tokens_seen": 1505404,
      "step": 1967,
      "train_runtime": 4499.9853,
      "train_tokens_per_second": 334.535
    },
    {
      "epoch": 0.1809655172413793,
      "grad_norm": 6.707273483276367,
      "learning_rate": 3.308270676691729e-07,
      "loss": 0.4110385477542877,
      "num_input_tokens_seen": 1506108,
      "step": 1968,
      "train_runtime": 4501.9056,
      "train_tokens_per_second": 334.549
    },
    {
      "epoch": 0.1810574712643678,
      "grad_norm": 5.941143035888672,
      "learning_rate": 3.2080200501253136e-07,
      "loss": 0.47294989228248596,
      "num_input_tokens_seen": 1507116,
      "step": 1969,
      "train_runtime": 4503.8798,
      "train_tokens_per_second": 334.626
    },
    {
      "epoch": 0.1811494252873563,
      "grad_norm": 6.705524444580078,
      "learning_rate": 3.1077694235588974e-07,
      "loss": 0.43166616559028625,
      "num_input_tokens_seen": 1507820,
      "step": 1970,
      "train_runtime": 4505.798,
      "train_tokens_per_second": 334.64
    },
    {
      "epoch": 0.18124137931034484,
      "grad_norm": 5.969166278839111,
      "learning_rate": 3.007518796992481e-07,
      "loss": 0.3799688220024109,
      "num_input_tokens_seen": 1508692,
      "step": 1971,
      "train_runtime": 4507.7712,
      "train_tokens_per_second": 334.687
    },
    {
      "epoch": 0.18133333333333335,
      "grad_norm": 6.527568340301514,
      "learning_rate": 2.9072681704260656e-07,
      "loss": 0.45936867594718933,
      "num_input_tokens_seen": 1509332,
      "step": 1972,
      "train_runtime": 4509.6779,
      "train_tokens_per_second": 334.687
    },
    {
      "epoch": 0.18142528735632185,
      "grad_norm": 6.620479583740234,
      "learning_rate": 2.8070175438596494e-07,
      "loss": 0.5485371351242065,
      "num_input_tokens_seen": 1510316,
      "step": 1973,
      "train_runtime": 4511.6165,
      "train_tokens_per_second": 334.762
    },
    {
      "epoch": 0.18151724137931036,
      "grad_norm": 6.678211212158203,
      "learning_rate": 2.706766917293233e-07,
      "loss": 0.4263443052768707,
      "num_input_tokens_seen": 1511044,
      "step": 1974,
      "train_runtime": 4513.5214,
      "train_tokens_per_second": 334.782
    },
    {
      "epoch": 0.18160919540229886,
      "grad_norm": 6.550446510314941,
      "learning_rate": 2.606516290726817e-07,
      "loss": 0.543639063835144,
      "num_input_tokens_seen": 1511860,
      "step": 1975,
      "train_runtime": 4515.4603,
      "train_tokens_per_second": 334.819
    },
    {
      "epoch": 0.18170114942528737,
      "grad_norm": 8.202722549438477,
      "learning_rate": 2.506265664160401e-07,
      "loss": 0.46255484223365784,
      "num_input_tokens_seen": 1512564,
      "step": 1976,
      "train_runtime": 4517.3681,
      "train_tokens_per_second": 334.833
    },
    {
      "epoch": 0.18179310344827587,
      "grad_norm": 6.744688987731934,
      "learning_rate": 2.406015037593985e-07,
      "loss": 0.5105394124984741,
      "num_input_tokens_seen": 1513328,
      "step": 1977,
      "train_runtime": 4519.2798,
      "train_tokens_per_second": 334.86
    },
    {
      "epoch": 0.18188505747126438,
      "grad_norm": 6.711732864379883,
      "learning_rate": 2.3057644110275693e-07,
      "loss": 0.5073504447937012,
      "num_input_tokens_seen": 1514244,
      "step": 1978,
      "train_runtime": 4521.2544,
      "train_tokens_per_second": 334.917
    },
    {
      "epoch": 0.18197701149425288,
      "grad_norm": 6.5366926193237305,
      "learning_rate": 2.205513784461153e-07,
      "loss": 0.48320409655570984,
      "num_input_tokens_seen": 1515112,
      "step": 1979,
      "train_runtime": 4523.237,
      "train_tokens_per_second": 334.962
    },
    {
      "epoch": 0.1820689655172414,
      "grad_norm": 6.033635139465332,
      "learning_rate": 2.105263157894737e-07,
      "loss": 0.4099450409412384,
      "num_input_tokens_seen": 1515796,
      "step": 1980,
      "train_runtime": 4525.1739,
      "train_tokens_per_second": 334.97
    },
    {
      "epoch": 0.1821609195402299,
      "grad_norm": 7.067359924316406,
      "learning_rate": 2.005012531328321e-07,
      "loss": 0.4702427089214325,
      "num_input_tokens_seen": 1516448,
      "step": 1981,
      "train_runtime": 4533.8924,
      "train_tokens_per_second": 334.469
    },
    {
      "epoch": 0.1822528735632184,
      "grad_norm": 6.445496559143066,
      "learning_rate": 1.904761904761905e-07,
      "loss": 0.5270073413848877,
      "num_input_tokens_seen": 1517276,
      "step": 1982,
      "train_runtime": 4535.8493,
      "train_tokens_per_second": 334.508
    },
    {
      "epoch": 0.1823448275862069,
      "grad_norm": 7.172682762145996,
      "learning_rate": 1.804511278195489e-07,
      "loss": 0.4619142413139343,
      "num_input_tokens_seen": 1517912,
      "step": 1983,
      "train_runtime": 4537.7494,
      "train_tokens_per_second": 334.508
    },
    {
      "epoch": 0.1824367816091954,
      "grad_norm": 5.525018692016602,
      "learning_rate": 1.704260651629073e-07,
      "loss": 0.40037378668785095,
      "num_input_tokens_seen": 1518664,
      "step": 1984,
      "train_runtime": 4539.7055,
      "train_tokens_per_second": 334.529
    },
    {
      "epoch": 0.1825287356321839,
      "grad_norm": 6.291981220245361,
      "learning_rate": 1.6040100250626568e-07,
      "loss": 0.4811934232711792,
      "num_input_tokens_seen": 1519408,
      "step": 1985,
      "train_runtime": 4541.6229,
      "train_tokens_per_second": 334.552
    },
    {
      "epoch": 0.18262068965517242,
      "grad_norm": 7.118106842041016,
      "learning_rate": 1.5037593984962406e-07,
      "loss": 0.485521525144577,
      "num_input_tokens_seen": 1520404,
      "step": 1986,
      "train_runtime": 4543.5949,
      "train_tokens_per_second": 334.626
    },
    {
      "epoch": 0.18271264367816092,
      "grad_norm": 8.06576156616211,
      "learning_rate": 1.4035087719298247e-07,
      "loss": 0.5489382147789001,
      "num_input_tokens_seen": 1521064,
      "step": 1987,
      "train_runtime": 4545.5171,
      "train_tokens_per_second": 334.629
    },
    {
      "epoch": 0.18280459770114943,
      "grad_norm": 5.893178939819336,
      "learning_rate": 1.3032581453634085e-07,
      "loss": 0.3916177749633789,
      "num_input_tokens_seen": 1521748,
      "step": 1988,
      "train_runtime": 4547.4392,
      "train_tokens_per_second": 334.638
    },
    {
      "epoch": 0.18289655172413793,
      "grad_norm": 7.13472843170166,
      "learning_rate": 1.2030075187969926e-07,
      "loss": 0.5219954252243042,
      "num_input_tokens_seen": 1522504,
      "step": 1989,
      "train_runtime": 4549.3946,
      "train_tokens_per_second": 334.661
    },
    {
      "epoch": 0.18298850574712644,
      "grad_norm": 7.152637004852295,
      "learning_rate": 1.1027568922305765e-07,
      "loss": 0.5631678700447083,
      "num_input_tokens_seen": 1523188,
      "step": 1990,
      "train_runtime": 4551.3117,
      "train_tokens_per_second": 334.67
    },
    {
      "epoch": 0.18308045977011494,
      "grad_norm": 7.595930576324463,
      "learning_rate": 1.0025062656641605e-07,
      "loss": 0.47467783093452454,
      "num_input_tokens_seen": 1523856,
      "step": 1991,
      "train_runtime": 4553.2606,
      "train_tokens_per_second": 334.674
    },
    {
      "epoch": 0.18317241379310345,
      "grad_norm": 7.325425624847412,
      "learning_rate": 9.022556390977444e-08,
      "loss": 0.5168801546096802,
      "num_input_tokens_seen": 1524544,
      "step": 1992,
      "train_runtime": 4555.1851,
      "train_tokens_per_second": 334.683
    },
    {
      "epoch": 0.18326436781609196,
      "grad_norm": 6.573004722595215,
      "learning_rate": 8.020050125313284e-08,
      "loss": 0.5315032005310059,
      "num_input_tokens_seen": 1525476,
      "step": 1993,
      "train_runtime": 4557.1357,
      "train_tokens_per_second": 334.744
    },
    {
      "epoch": 0.18335632183908046,
      "grad_norm": 8.198920249938965,
      "learning_rate": 7.017543859649123e-08,
      "loss": 0.39272958040237427,
      "num_input_tokens_seen": 1526056,
      "step": 1994,
      "train_runtime": 4559.085,
      "train_tokens_per_second": 334.729
    },
    {
      "epoch": 0.18344827586206897,
      "grad_norm": 6.335601806640625,
      "learning_rate": 6.015037593984963e-08,
      "loss": 0.48748651146888733,
      "num_input_tokens_seen": 1526836,
      "step": 1995,
      "train_runtime": 4561.0,
      "train_tokens_per_second": 334.759
    },
    {
      "epoch": 0.18354022988505747,
      "grad_norm": 6.505175590515137,
      "learning_rate": 5.0125313283208025e-08,
      "loss": 0.4721751809120178,
      "num_input_tokens_seen": 1527440,
      "step": 1996,
      "train_runtime": 4562.9113,
      "train_tokens_per_second": 334.751
    },
    {
      "epoch": 0.18363218390804598,
      "grad_norm": 5.909262180328369,
      "learning_rate": 4.010025062656642e-08,
      "loss": 0.40433812141418457,
      "num_input_tokens_seen": 1528240,
      "step": 1997,
      "train_runtime": 4565.0634,
      "train_tokens_per_second": 334.769
    },
    {
      "epoch": 0.18372413793103448,
      "grad_norm": 6.7515788078308105,
      "learning_rate": 3.0075187969924815e-08,
      "loss": 0.47629979252815247,
      "num_input_tokens_seen": 1528996,
      "step": 1998,
      "train_runtime": 4566.9845,
      "train_tokens_per_second": 334.793
    },
    {
      "epoch": 0.183816091954023,
      "grad_norm": 6.242720603942871,
      "learning_rate": 2.005012531328321e-08,
      "loss": 0.47347041964530945,
      "num_input_tokens_seen": 1529620,
      "step": 1999,
      "train_runtime": 4568.9079,
      "train_tokens_per_second": 334.789
    },
    {
      "epoch": 0.1839080459770115,
      "grad_norm": 8.364867210388184,
      "learning_rate": 1.0025062656641605e-08,
      "loss": 0.46602606773376465,
      "num_input_tokens_seen": 1530340,
      "step": 2000,
      "train_runtime": 4570.8216,
      "train_tokens_per_second": 334.806
    },
    {
      "epoch": 0.1839080459770115,
      "eval_loss": 1.7979987859725952,
      "eval_runtime": 27.7,
      "eval_samples_per_second": 36.101,
      "eval_steps_per_second": 9.025,
      "num_input_tokens_seen": 1530340,
      "step": 2000
    }
  ],
  "logging_steps": 1,
  "max_steps": 2000,
  "num_input_tokens_seen": 1530340,
  "num_train_epochs": 1,
  "save_steps": 30,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5481097281246720.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}